Livro 1 - Morfologia e Diversidade-Compactado-1-251

Download as pdf or txt
Download as pdf or txt
You are on page 1of 251

Functional Morphology

and Diversity

Functional Morphology and Diversity. Les Watling and Martin Thiel.


© Les Watling and Martin Thiel 2013. Published 2013 by Oxford University Press.
The Natural History of the Crustacea Series

SERIES EDITOR:
Martin Thiel

Editor ia l A dvisory Boar d:


Geoff Boxshall, Natural History Museum, London, UK
Emmett Duffy, Virginia Institute of Marine Sciences, Gloucester, USA
Darryl Felder, University of Louisiana, Lafayette, USA
Gary Poore, Victoria Museum, Melbourne, Australia
Bernard Sainte-Marie, Fisheries and Oceans Canada, Mont-Joli, Canada
Gerhard Scholtz, Humboldt University Berlin, Berlin, Germany
Fred Schram, Friday Harbor Marine Laboratory, Seattle, USA
Les Watling, University of Hawaii, Hawaii, USA

Functional Morphology and Diversity (Volume 1)


Edited by Les Watling and Martin Thiel
Functional Morphology
and Diversity
The Natural History of the Crustacea,
Volume 1

EDITED BY LES WATLING AND MARTIN THIEL

1
3
Oxford University Press is a department of the University of Oxford.
It furthers the University’s objective of excellence in research, scholarship,
and education by publishing worldwide.

Oxford New York


Auckland Cape Town Dar es Salaam Hong Kong Karachi
Kuala Lumpur Madrid Melbourne Mexico City Nairobi
New Delhi Shanghai Taipei Toronto

With offices in
Argentina Austria Brazil Chile Czech Republic France Greece
Guatemala Hungary Italy Japan Poland Portugal Singapore
South Korea Switzerland Thailand Turkey Ukraine Vietnam

Oxford is a registered trademark of Oxford University Press in the UK and certain other
countries.

Published in the United States of America by


Oxford University Press
198 Madison Avenue, New York, NY 10016

© Oxford University Press 2013

All rights reserved. No part of this publication may be reproduced, stored in a


retrieval system, or transmitted, in any form or by any means, without the prior
permission in writing of Oxford University Press, or as expressly permitted by law,
by license, or under terms agreed with the appropriate reproduction rights organization.
Inquiries concerning reproduction outside the scope of the above should be sent to the
Rights Department, Oxford University Press, at the address above.

You must not circulate this work in any other form


and you must impose this same condition on any acquirer.

Library of Congress Cataloging-in-Publication Data

Functional Morphology and Diversity


p. cm—(The Natural History of the Crustacea series).
Includes bibliographical references and index.
ISBN 978-0-19-539803-8 (hardcover: alk. paper) 1. Crustacea. I. Watling, Les. II. Thiel,
Martin, 1962–
QL435.N38 2013
595.3—dc23 2012012251

9 8 7 6 5 4 3 2 1
Printed in the United States of America
on acid-free paper
PREFACE

Many years ago, when M.T. came to Maine to work with L.W. on shallow subtidal crustaceans,
we established some amphipod populations in aquaria at the University of Maine’s Darling
Marine Center flowing seawater facility. What ensued was not just a series of remarkable behav-
ioral observations but also a series of long discussions of “how do they do that?” in response to
the burrowing and whip making that we observed. The two of us have maintained a lifelong
interest in understanding how crustacean morphology “determines” what the animals can do
and where they can live.
Crustaceans encompass a bewildering array of body forms that they use to occupy almost
every habitat type on Earth. In fact, the only habitat not occupied by crustaceans is the open
air; that is, they cannot fly. Understanding what modifications of the crustacean body plan have
allowed this diversification is the subject of this book.
In its basic form, the crustacean body consists of a bilaterally symmetrical, segmented, more
or less tubular body, with each segment bearing a pair of appendages. The number of body seg-
ments and the design or even presence of appendages vary considerably, with the most extreme
cases being found in those groups that have invaded the bodies of other animals.
In this book we have asked the chapter authors to explore this diversification of form and to
explain how various parts of the crustacean body work. While many authors have examined the
functional morphology and anatomy of crustaceans in individual publications or book contri-
butions, this has never been done in an integral way in one volume.
We were particularly interested in understanding the design limitations of the crustacean
body, for example, how living in a dense fluid medium might restrict the movement capabilities
of the animal, and how that would vary depending on the animal’s size. But movement isn’t all
that crustaceans do in their daily lives—they have to eat, respire, reproduce, and grow, all of
which needs to be controlled so that the animal functions as a successfully coordinated whole.
To set the stage, Schram (chapter 1) reviews the range of crustacean body plans, and Haug
et al. (chapter 2) review what is known about appendage development in the earliest arthro-
pods as they become what we would recognize now as crustaceans. Then follow four chapters
that examine the exterior of the crustacean body. Williams (chapter 3) investigates the genetic
control of appendage development as the animal develops. Olesen (chapter 4) surveys the
functional constraints of the archetypical crustacean structure, the carapace. Dillaman et al.
(chapter 5) take a detailed look at the crustacean cuticle, and Garm and Watling (chapter 6)
review the structure and function of setae and other cuticular outgrowths.

v
Functional Morphology and Diversity. Les Watling and Martin Thiel.
© Les Watling and Martin Thiel 2013. Published 2013 by Oxford University Press.
vi Preface

Our examination of the integrated functional aspects of the crustacean body starts with
Boxshall and Jaume’s overview of crustacean antennae (chapter 7). Watling (chapter 8) inte-
grates food consumption and digestion, showing that mouthpart appendage structure, foregut
morphology, and digestive enzyme secretion are all related to diet. Belanger (chapter 9), Faulkes
(chapter 10), Yen (chapter 11), and Boudrias (chapter 12) examine modes of locomotion in crus-
taceans and the functional constraints imposed on locomotory appendages by the physics of the
medium in which they live.
We end the book with a series of chapters dealing with system-level integrative functional
anatomy. Bauer (chapter 13) reviews the structure and function of appendages used for groom-
ing and reproduction. Wirkner and Richter (chapter 14) examine the integration of respiratory
structures with internal details of the circulatory system. An overview of the internal anatomy
of the reproductive system is provided by L ópez Greco (chapter 15). And lastly, the coordina-
tion of the whole body is dealt with in Sullivan and Herberholz’s review of crustacean nervous
systems (chapter 16).
We have asked the authors of these chapters to deal with their part of the crustacean body in
isolation. Of course, we recognize that crustaceans are living creatures, and in order to live they
have managed to integrate the functional aspects of their bodies that we have studied so well on
their own. What we have not done is try to develop a fully integrated model of the crustacean
body that accounts for all the physical aspects of the various habitats in which crustaceans live.
We suspect this is possible but may be beyond the reach of any one of us.
ACKNOWLEDGMENTS

Our thanks go foremost to all contributors who attended to all our requests during the prep­
aration of this book. Their expertise and their willingness to invest time into the writing of
their contributions make up the value of this book. We especially thank our editorial assistants,
Ivan Hinojosa and Lucas Eastman, who expertly revamped many of the figures and who care­
fully read and edited all the chapters. The generous contribution from Universidad Catolica
del Norte was essential for this project— we are very grateful for the continuous support which
allowed us to focus on the task. The vision and foresight of the university authorities made this
project possible and we hope that this and the upcoming volumes fulfil their expectations.
We also thank those colleagues who read entire or parts of chapters. The publisher, Oxford
University Press, gave us a lot of freedom, and in particular, we express our appreciation to Peter
Prescott, Tisse Tagaki, Jeremy Lewis, and Hallie Stebbins for their help during the past few
years. L.W. would like to acknowledge the role o f his students in stimulating him to think about
crustaceans as functionally whole units, and his friend C. Nouvian, who listened with interest
to many crustacean natural history stories while we were working on other things. Finally, we
thank our families for their patience and interest in this project— without their tolerance, none
of this would have been possible.

Editing of this book was generously supported by Universidad Catolica del Norte, Chile.
CONTRIBUTORS

E D IT O R S Michel A . Boudrias
Martin Thiel Department of Marine Science and
Facultad Ciencias del Mar Environmental Studies
Universidad Catolica del Norte University of San Diego
Larrondo 1281 Science and Technology 267
Coquimbo 5998 Alcala Park
Chile San Diego, CA 92110
USA
Les Watling
Department of Biology Geoff Boxshall
University of Hawaii at Manoa Department of Zoology
152 Edmondson Hall The Natural History Museum
Honolulu, HI 96822 Cromwell Road
USA London SW7 5BD
UK
AU TH O RS
Raymond T. Bauer Richard M. Dillaman
Department of Biology Department of Biology and Marine Biology
University of Louisiana University of North Carolina at Wilmington
PO Box 42451 601 South College Road
Lafayette, LA 70504 Wilmington, NC 28403-5915
USA USA

Jim Belanger Zen Faulkes


Department of Biology Department of Biology
West Virginia University The University of Texas-Pan American
PO Box 6057 1201W. University Drive
Morgantown, WV 26506 Edinburg, TX 78539
USA USA
Contributors ix

Anders Garm Andreas Maas


Department of Cell and Organism Biology Biosystematic Documentation
Lund University University of Ulm
Helgonavagen 3 Helmholtzstrasse 20
222362 Lund 89081 Ulm
Sweden Germany

Carolin Haug
Shannon Modla
University of Greifswald
Delaware Biotechnology Institute
Zoological Institute and Museum
15 Innovation Way, Suite 117
Department of Cytology and Evolutionary
Newark, DE 19716
Biology
USA
Soldmannstr. 23
17487 Greifswald
Germany Jorgen Olesen
Natural History Museum of Denmark
Joachim T. Haug (Zoological Museum)
University of Greifswald University of Copenhagen
Zoological Institute and Museum Universitetsparken 15
Department of Cytology and Evolutionary DK-2100 Copenhagen 0
Biology Denmark
Soldmannstr. 23
17487 Greifswald Stefan Richter
Germany Universitat Rostock
Institut fur Biowissenschaften
Jens Herberholz Abteilung fur Allgemeine und Spezielle
Department of Psychology and Zoologie
Neuroscience and Cognitive Science Universitatsplatz 2
Program 18055 Rostock
2123H Bio-Psych Building Germany
University of Maryland
College Park, MD 20742
Robert Roer
USA
Department of Biology and Marine
Biology
Damiajaume
University of North Carolina at
Instituto Mediterraneo de Estudios
Wilmington
Avanzados IM EDEA (CSIC-UIB)
601 South College Road
Miquel Marques 21
Wilmington, NC 28403-5915
07190 Esporles (Mallorca, Illes Balears)
USA
Spain

Laura S. Lopez Greco Frederick R. Schram


Facultad de Ciencias Exactas y Naturales Burke Museum
University of Buenos Aires University of Washington at
Pabellon 2, Intendente Giiiraldes 2160 Seattle
Ciudad Universitaria C1428EGA Post Box 1567
Buenos Aires Langley, WA 98260
Argentina USA
X Contributors

Thomas Shafer Terri A . Williams


Department of Biology and Marine Department of Biology
Biology Trinity College
University of North Carolina at 300 Summit Street
Wilmington Hartford, CT 06106
601 South College Road USA
Wilmington, NC 28403-5915
USA Christian S. Wirkner
Universitat Rostock
Jeremy M. Sullivan Abteilung fur Allgemeine und Spezielle
Department ofNeurology Zoologie
Johns Hopkins University Institut fiir Biowissenschaften
855 North Wolfe Street Universitiitsplatz 2
2nd Floor, Rangos Building 18055 Rostock
Baltimore, MD 21205 Germany
USA
Jeannette Yen
Dieter Waloszek School of Biology
Biosystematic Documentation Georgia Institute of Technology
University of Ulm 310 Ferst Drive
Helmholtzstrasse 20 Atlanta, GA 30332-0230
89081 Ulm USA
Germany

Les Watling
Department of Biology
University of Hawaii at Manoa
152 Edmondson Hall
Honolulu, HI 96822
USA
CONTENTS

1. Comments on Crustacean Biodiversity and Disparity of Body Plans • 1


Frederick R. Schram

2. Evolution of Crustacean Appendages • 34


Joachim T. Haug, Andreas Maas, Carolin Haug, and Dieter Waloszek

3. Mechanisms of Limb Patterning in Crustaceans • 74


Terri A. Williams

4. The Crustacean Carapace: Morphology, Function, Development, and Phylogenetic


History • 103
Jorgen Olesen

5. The Crustacean Integument: Structure and Function • 140


Richard M. Dillaman, Robert Roer, Thomas Shafer, and Shannon Modla

6 . The Crustacean Integument: Setae, Setules, and Other Ornamentation • 167


Anders Garm and Les Watling

7. Antennules and Antennae in the Crustacea • 199


Geoff Boxshall and Damia Jaume

8. Feeding and Digestive System • 237


Les Watling

9. Appendage Diversity and Modes of Locomotion: Walking • 261


Jim Belanger
xii Contents

10. Morphological Adaptations for Digging and Burrowing • 276


Zen Faulkes

11. Appendage Diversity and Modes of Locomotion: Swimming at Intermediate Reynolds


Numbers • 296
Jeannette Yen

12. Swimming Fast and Furious: Body and Limb Propulsion at Higher Reynolds
Numbers • 319
Michel A. Boudrias

13. Adaptive Modification of Appendages for Grooming (Cleaning, Antifouling) and


Reproduction in the Crustacea • 337
Raymond T. Bauer

14. Circulatory System and Respiration • 376


Christian S. Wirkner and Stefan Richter

15. Functional Anatomy of the Reproductive System • 413


Laura S. Lopez Greco

16. Structure of the Nervous System: General Design and Gross Anatomy • 451
Jeremy M. Sullivan and Jens Herberholz

Index • 485
Functional Morphology
and Diversity
1
COMMENTS ON CRUSTACEAN BIODIVERSITY
AND DISPARITY OF BODY PLANS

Frederick R. Schram

Abstract
The science of natural history is built on twin pillars: cataloging the species found in nature, and
reflecting on the variety and function of body plans into which these species fit. We often use
two terms, diversity and disparity, in this connection, but these terms are frequently used inter-
changeably and thus repeatedly confused in contemporary discourse about issues of function
and form. Nevertheless, diversity and disparity are distinct issues and must be treated as such;
each influences our views of the evolution and morphology of crustaceans.

CRUSTACEAN DIVERSIT Y

Crustaceans exhibit great disparity in basic body plans (I return to this subject below), but
disparity of crustacean form is different from crustacean biodiversity, that is, the number of
species we have within any particular group. No one knows for certain the exact number of
species within any group of organisms, although the situation might improve with the appear-
ance of online catalogs for particular groups. The people who set up these databases and
maintain them as new species are added and old species are placed in synonymy provide a
much-needed service toward adequately cataloging the tree of life. Nevertheless, as humans
we like numbers—they are easily understood. So I have made my own tally (Table 1.1) and
present a summary of estimates compiled from various authorities as to the total number of
crustacean species.
There is clearly no agreement on numbers among the authors listed in Table 1.1, although
the estimates have gone up through time. With the exception of Minelli (1993) and Brusca and

1
Functional Morphology and Diversity. Les Watling and Martin Thiel.
© Les Watling and Martin Thiel 2013. Published 2013 by Oxford University Press.
2 Functional Morphology and Diversity

Table 1.1. Various estimates of global numbers of species of crustaceans.

Estimated number of species Source


44,950 Bouchet (2006)
32,000 Brusca and Brusca (1990)
68,171 Brusca and Brusca (2003)
40,000 Groombridge and Jenkins (2000)
39,000 May (1988)
75,000 Meglitsch and Schram (1991)
55,364 Minelli (1993)
38,000 Ruppert and Barnes (1994)
49,658 This chapter

Brusca (2003), who appear to have attempted a real count, the other authors obviously provided
rounded off and rough estimates. For example, the number provided by Meglitsch and Schram
(1991) was an estimate of what the highest number might be at some point in time when knowl-
edge of the number of species will have reached a plateau.
Although we know a great deal about groups of invertebrates, our knowledge is not very
good and rather incomplete. I examined the patterns through time in documenting animal
taxon diversity (Schram 2003) and noted several periods during which plateaus of relative inac-
tion followed bursts in activity. It seems clear from these charted patterns that we are currently
in one of those periods of increased activity, but whether we will soon reach a new plateau, or
whether increased use of molecular techniques to identify monophyletic groups might con-
tinue to add new taxa at all levels—from phylum down to species—I cannot say. However,
increasing application of molecular techniques does seem to indicate that we have underesti-
mated the degree of cryptic speciation in nature.
Having stated this, I feel honor bound by the charge given to me by the editors to provide
my own numbers, so I tally here the currently known crustacean species. Table 1.2 is based on a
census of relevant websites, currently available monographic literature, and the best estimates
of authorities active in one or more of these groups. The reader should keep in mind that this
is a tally of species numbers at this point in time, and these figures can only increase as our
knowledge of these taxa evolves. In fact, the survey made by Martin and Davis (2006) seems
to indicate that no asymptotes are yet emerging in the pace at which new species are being
described.
First, the total number of species obtained by this survey, 49,658, is not too far off from the
estimates of Bouchet (2006) and Minelli (1993). Within that number, some things deserve spe-
cial notice. Of the two largest groups on this list, Maxillopoda and Malacostraca, the numbers
are of similar magnitude—almost 19,000 and something more than 29,000, respectively. The
number of maxillopodans can only increase. The 9,500 copepods is only an estimate, although
it may stand close to the actual numbers of currently described species. Nevertheless, copepod
taxonomy is an active discipline, and increasingly sophisticated techniques of study will help
isolate cryptic species. The 8,008 species of ostracodes is only an estimate, and if we factor in
fossil species, we would more than double that number. Furthermore, the application of molecu-
lar methods in Crustacea will likely affect our understanding of species level biodiversity. For
example, I am surprised at the relatively low number for the thecostracans, but parasitism is
rampant in the group, and underestimates of species diversity would prevail in taxa with such
Crustacean Biodiversity and Disparity of Body Plans 3

Table 1.2. Census of species numbers in various crustacean groups.

Taxon Number of species


Branchiura (Argulida) 175
Branchiura (Pentastomida) 100
Mystacocarida 13
Branchiopoda 509
Anostraca 307
Cyclestherida 1
Laevicaudata 36
Notostraca 15
Spinicaudata ≈150
Cladocera 450
Maxillopoda 18,911
Copepoda 9,500
Ostracoda ≈8,008
Myodocopida 1,608 (+500 fossils)
Podocopida 6,400 (+9,500 fossils)
Thecostraca 1,403
Ascothoracica >99
Cirripedia 1,304
Acrothoracica >61
Rhizocephala >255
Thoracica 948
Facetotecta 12
Tantulocarida 28
Remipedia 19
Cephalocarida 10
Malacostraca 29,471
Phyllocarida 39
Stomatopoda 456
Eumalacostraca 28,976
Syncarida >187
Bathynellacea >170
Anaspidacea 17
Peracarida 15,686
Amphipoda 6,950
Cumacea 1,342
Isopoda 5,270
Lophogastrida 56
Mictacea 5
Mysida sensu lato 1,085
Mysida sensu stricto 1,075
Stygiomysida 10
Spelaeogriphacea 4
Tanaidacea 940
4 Functional Morphology and Diversity

Table 1.2. (Continued)

Taxon Number of species


Thermosbaenacea 34
Eucarida 13,103
Amphionidacea 1
Decapoda 13,016
Dendrobranchiata 522
Caridea 2730
Stenopodidea 57
Reptantia 9,707
Euphausiacea 86
Total 49,658

The higher taxonomic grouping of this table accords with the conclusions derived from the discussion of disparity of form
given later in this chapter. Classes are shown in boldface.

highly reduced body forms. For example, rhizocephalans seem poised on the edge of a renais-
sance in interest, and the number of species anticipated will increase.
Malacostraca constitutes a large number of species, but the species distribution is uneven
because some subgroups are very large (amphipods, isopods, reptant decapods), while others
are small (mictaceans, spelaeogriphaceans, and the amphionidacean). In fact, any group associ-
ated with cave or groundwater habitats appears likely at the lower end of species number esti-
mates, but these habitats are difficult to study, and every attempt to sample these communities
turns up new and interesting species, which can only continue into the future. (In this connec-
tion, one need only consider the work on crayfish in North America to see what happens when
intensive systematic interest is focused on a group.)
Some major class- and order-level taxa presently have low species numbers (remipedes
and cephalocarids), but here, too, we have animals living in habitats that are difficult to sam-
ple (anchialine caves and the deep sea). Other groups contain very cryptic creatures living in
places that, although well studied, nevertheless are often overlooked (mystacocarids in inter-
stitial beaches).
Because of the great disparity of body plans exhibited by crustaceans, we have a problem
in comparing the species numbers in one group with another. The taxa in Table 1.2 are organ-
ized around the currently recognized class and order levels, but how does one compare ordi-
nal differences seen in malacostracans with what are called orders within the maxillopodans?
Recognizing a decapod from an amphipod is quite easy (both are orders of Eumalacostraca),
but not many people could easily distinguish a cyclopoid from a calanoid (they are both orders
of Copepoda) without being carefully schooled in the differences. Hence, trying to compare
numbers of species within groups across the major taxonomic (class-level) units of crusta-
ceans is truly like comparing apples to oranges or, in this case, lobsters to zooplankton.
Nevertheless, strange patterns arise when we look within groups. Consider the peracari-
dans, for example. Why are there so many species of amphipods (6,900) compared to ther-
mosbaenaceans (34) or mictaceans (5)—approximately two and three orders of magnitude
difference? Are amphipods truly that much better adapted to their environments, an expla-
nation often assumed to be true? If so, how and why? Or, are some other factors at play that
might augment or possibly even ignore issues of adaptation? Some of these factors might be
Crustacean Biodiversity and Disparity of Body Plans 5

40

A
30

Number of genera
20

10

0
5 10 15 20 25
Number of species in genus

100
B
Number of genera

10

1
1 10 100
Number of species in genus

Fig. 1.1.
Arithmetic hollow curve (A) and log-log (B) plots of size distributions of genera of Stomatopoda (as
number of included species) roughly conforming to power law N(x) = ax–b. For details about the method,
see Minelli et al. (1991).

difficulty of habitat access for study (mentioned above), age of a clade, habitat heterogeneity,
and expressions of chance in nature. The various authors of other volumes in this series will
explore many of these issues.
The element of chance plays an important role in classification. Willis and Yule (1922) and
Minelli et al. (1991) observed that the size of supraspecific taxa as related to the included sub-
taxa (species in genera, genera in families, etc.) follows a power law. They concluded that the
structure of biological classification is naturally fractal. This structure can be expressed as a
hollow curve that, if plotted on a log-log scale, would conform to N(x) = ax–b.
We can illustrate this with one example from Malacostraca, the unipeltate stomatopods
(mantis shrimp). As of this writing, we recognize 456 species in 112 genera of mantis shrimp,
with an additional 123 nominal species currently in synonymy. If we consider only the 456 rec-
ognized species, distribution numbers range from one of the largest genera, Nannosquilla, with
some 26 species, down to 36 genera with but a single species each. Graphing this diversity, we
can see that on an arithmetic scale it forms a hollow curve (Fig. 1.1A), and on a log-log plot a
6 Functional Morphology and Diversity

straight line emerges (Fig. 1.1B). The fractal pattern becomes apparent when examining genera
within families (data not shown), where we would again see a log-log plot that roughly matches
that of species in genera. Whether this pattern appears in other groups of crustaceans remains to
be tested, but I have no doubt that it will hold as it has in other groups of animals and plants.
As humans, we are naturally inclined to seek causative explanations for patterns of biodi-
versity. However, I believe we do not necessarily need to explain why one particular genus,
such as Nannosquilla with 26 species, is somehow better adapted than its confamilial sister
genera, in this case Mexisquilla and Keppelius, each with only a single species. As we chart spe-
cies biodiversity, we should be open to the possibility that the relative number of taxa within
any particular group may represent nothing other than the manifestation of the operations of
a stochastic, fractal universe, to say nothing of the vagaries of individual taxonomic decisions.
Many authorities might reject my pessimism here, but at the very least, a stochastic, fractal
biodiversity has to be one of several alternative hypotheses to consider.

CRUSTACEAN DISPARIT Y OF BODY PLANS

The crustaceans are the most variable of all the arthropod groups; that is, there is a great dis-
parity of body plans throughout their ranks (Fig. 1.2). If we are to assume that Crustacea is a
monophyletic group, then they are not like any other arthropods. This high degree of variability
is a very real problem with some serious implications, because if we take this disparity of form
at face value, then we should seriously question whether all these various groups can constitute
a single monophylum.
When one looks at other major arthropod groups outside of Crustacea, there appears to be
no great disparity of plan within these taxa (see Meglitsch and Schram 1991); members of each
group fit a concise definition. For example, members of Insecta (Hexapoda) have a body divided
into a five-segment head with the first postantennal segment bearing appendages modified as
a labrum, a three-segment thorax with two sets of wings in the pterygote insects borne on the
second and third segments, and an abdomen of 10–12 somites. All insects conform to this defini-
tion with some exceptions, for example, allowing for fusion of segments at the terminus of the
abdomen or modification of wing arrangements. Insects have a unified body plan.
Myriapoda as a whole do vary in some features such as body length but have in common
that their trunk is not divided into a thorax and abdomen and that their gonopores are gener-
ally located on the anterior aspect of the trunk. The individual groups of myriapods conform
to common plans: Symphyla have 12 trunk segments with the gonopores on the fourth somite;
Pauropoda bear 12 trunk segments with the gonopore on the second somite; and Diplopoda,
with several very distinct orders, all exhibit well-developed diplosomites, that is, pairs of seg-
ments fused dorsally but distinct ventrally, and their gonopores are located on the second
trunk segment. The individual orders of diplopods vary only regarding the total number of
trunk somites: pselaphognaths have at least 10–12, but colobognaths can exceed 30. Chilopoda
have variable trunk segment numbers, extending from 15 to more than 180 pairs of legs, depend-
ing on group, but all chilopods without exception have long antennae and modify the first
trunk limb as a fang equipped with a poison gland to facilitate their carnivory. Centipedes also
uniquely bear gonopores on the posterior aspect of the trunk.
The subphylum Cheliceriformes exhibits only a few “head” segments, essentially two, and
these are fused with the anterior, or locomotory, part of the trunk to form a prosoma. The ante-
riormost somite (the one just posterior to the asegmental acron), the homolog of the antennal
segment in other arthropods, does not carry antennae but rather is equipped with a pair of che-
licerae. The second segment, what in other arthropods is referred to as the first postantennal
Crustacean Biodiversity and Disparity of Body Plans 7

Fig. 1.2.
Disparate body types among crustaceans.

segment, typically bears a well-developed set of limbs, albeit variously developed. There are no
exceptions to this basic format.
Within the cheliceriforms, the highly distinctive Pycnogonida appear to be all legs, their
prosoma reduced to a thin cylinder. The mouth is located terminally on a long proboscis. The
small turreted “head” bears chelicerae, a second set of limbs called palps, and a third set of
limbs modified as ovigers in the males. Posterior to these limbs, most pycnogonids utilize four
pairs of legs for locomotion (a few forms with five or six pairs are known). All sea spiders con-
form to this body plan.
Arachnida have a six-segment prosoma, with chelicerae, pedipalps, and four pairs of walk-
ing legs. The trunk bears an additional opisthosoma of diverse form but composed of some 13
somites, with the first segment greatly reduced as a narrow pedicel and the second bearing the
gonopores. Opisthosomal limbs are missing or greatly reduced. Despite a great variety of body
profiles, especially regarding the opisthosoma, all arachnids conform to this single plan.
Merostomata, a small group today, was more extensive (and diverse) in the past. The
prosoma bears six pairs of limbs. The chelicerae are followed by four sets of modest-sized
walking limbs with specialized gnathobases, the first of which in the males is modified for
8 Functional Morphology and Diversity

grasping the female during copulation, and a somewhat larger fifth set effective in groom-
ing the underside of the prosoma. The next somite, the pregenital segment, is reduced but
bears modified limbs, the chilaria. Unique among living cheliceriforms, the opisthosome
of the merostomes bears six pairs of limbs posterior to the genital segment. All merostomes
conform to this plan.
From this short review, we can see that all these groups of arthropods have concise diag-
noses, with distinct sets of apomorphies that characterize all members of the group. Crustacea,
if viewed as a single group, simply does not have this.

CAN WE DIAGNOSE A MONOPHYLETIC CRUSTACEA?

Any invertebrate zoology textbook can provide a set of characters for Crustacea. When I was
asked to provide such a diagnosis 25–30 years ago (Schram 1979, 1982), I certainly did not hesi-
tate. However, we now realize it is not sufficient to simply string together any list of characters.
Ideally, as in the arthropod examples cited above, these characters should be unique derived
features that diagnose all members of the group. To rephrase this in contemporary terms, a diag-
nosis should offer synapomorphies that together uniquely delineate a monophyletic group. We
strive for natural taxonomies, classifications that reflect evolution. It is critical to determine if
this is possible for Crustacea.
A commonly accepted diagnosis of Crustacea consists of the following: (1) head of five
somites, each bearing a set of appendages consisting of two pairs of antennae, a pair of mandi-
bles, and two pairs of maxillae; (2) body consisting of three regions: head, thorax, and “abdo-
men”; (3) trunk appendages primitively multiramous; and (4) development consisting of a
series of discrete larval and/or juvenile stages, initiated by a stage termed a nauplius.
Let us inspect these features one by one in order to determine if these characters provide
that unique set of descriptors we require for a diagnosis of Crustacea. In the discussion below, I
restrict the term Crustacea to mean a monophyletic group and the term crustaceomorph to con-
note the amalgam of arthropod types that we generally and broadly refer to as “crustaceans”
(fossil and recent) but that may or may not be monophyletic. Table 1.3 will assist the reader in
following along the taxa and many of the relevant features discussed below.

“Head of five somites, each bearing a set of appendages consisting of two pairs of
antennae, a pair of mandibles, and two pairs of maxillae”

These are not a unique set of features. A head consisting of five somites is shared with insects
and the myriapods (see above; Meglitsch and Schram 1991), and, as would follow, most of the
head appendages of these somites are shared among the three groups, namely, the first set of
antennae, mandibles, and two sets of maxillae. It is only regarding the so-called second set of
antennae that we might have a distinctive crustaceomorph feature since myriapods and insects
lack a limb in this position.
The second antennae are generally perceived as specialized sensory limbs and as such could
serve as a defining apomorphy. This descriptor arises from the mental image of Crustacea con-
jured up by thinking of a shrimp or a lobster (a malacostracan), and this image without a doubt
presents us with an icon of an arthropod with a set of sensory limbs at this position, albeit with
slight anatomical variations, depending on group (Fig. 1.3A).
Nevertheless, sensory second antennae are not characteristic of all groups of crustaceo-
morphs. Chapter 7 provides additional details concerning antennae, but a short overview
will suffice here to make a point. Remipedes have quite distinctive limbs in this position
(Fig. 1.3B) that I suspect serve equally as a hydrofoils to direct currents of water that f low
Table 1.3. Comparative morphology of various aspects that define body plans among various groups of crustaceomorphs.

Taxa Head Trunk Postthorax Thorax length Segment no. of gonopore Trunk First larval
(no. somites) regions (no. somites) position limb rami segments

Male Female
Oligo-crustacea
Branchiura 5 2 – 4 4 4 2 ?
Mystacocarida 2 a 5 4 4 1 4
Skara 5 2 a 1 ? ? 2 ?
Martinssonia 4 2 a 3 ? ? 2 4
Branchiopoda
Laevicaudata 5 1 – 10–12 11 11 M 3
Notostraca 5 ?2 ?p 11 11 11 M 3
Spinicaudata 5 1 – 16–32 11 11 M 3
Cyclestheria 5 1 – 16–32 11 11 M 3
Cladocera 5 2 – 4–6 3
Anostraca 5 2 a 12 12 M 3
Lepidocaris 5 2 a ? 2, M
Rehbachiella 5 2 a 12 ? 2 3
Waptia 5 2 a 11 ? ?
Eucrustacea
Cephalocarida 4 2 a 8 6 6 M 5
Maxillopoda
Copepoda 5 2 a 7 7 7 2 3
Cirripedia 5 1 a 7 7 1 2 3
Ostracoda ?7 ?7 3
Bredocaris 5 2 a 7 ? 2 4
Malacostraca
Eumalacostraca 5(+) 2 p 8 8 6 2 3
Hoplocarida 5(+) 2 p 8 8 6 2 3
Phyllocarida 5 2 p 8 8 6 M 3
Remipedia 6 1 – – 15 8 2 3

Abbreviations: a, abdomen; M, multiramous; p, pleon. Neither the list of taxa nor the features are exhaustive of all possibilities but cover the major elements discussed in the text.
10 Functional Morphology and Diversity

C C'

anterior spine
palp
antennule F

E
antenna

posterior
spine G

Fig. 1.3.
Various functional types of “antennal” limbs found in crustaceans. (A) Sensory: Apseudes hermaphro-
diticus, tanaid (from Lang 1953). (B) Swimming hydrodynamic plane: Lasionectes entrichoma, a remipede
(from Schram et al. 1986). (C and C′) Locomotory/feeding: Derocheilocaris ingens, a mystacocarid (from
Hessler 1969), antenna (C) and mandible (C′). The arrow indicates gnathal lobes. (D) Part of an attach-
ment complex: Argulus foliaceous, a branchiuran (from Martin 1932). (E) Swimming: Archimisophria dis-
coveryi, a copepod (from Boxshall 1983). (F) Host penetration: Gorgonolaureus muzikae, an ascothoracidan
(from Grygier 1981). (G) Swimming/feeding: Bredocaris admirabilis, a Cambrian maxillopodan (from
Mü ller and Walossek 1988).

around the head and perhaps also to aid in creating some of those currents by f lapping the
hydroplane-like exopods. Although no work has yet been done on functional morphology
of this limb, I believe it safe to say that the remipede antenna is not a purely sensory append-
age. Mystacocarids have a pair of limbs behind the first antennae that are virtually identi-
cal in form to the mandibles (Fig. 1.3C,C′), except these so-called second antennae lack the
gnathal armament at the base of the limb that occurs on the mandibles. The mystacocarid
“antennae” are locomotory limbs. Branchiurans possess broad plates in this position with
basal hooks and terminal, recurved spines—nothing sensory at all but rather serving to assist
with attachment (Fig. 1.3D). Tantulocarids lack head appendages altogether. Copepods have
well-developed limbs in this position that serve for the most part as the primary organs of
swimming (Fig. 1.3E). It is difficult to specify what this limb does in ascothoracicans, where
Crustacean Biodiversity and Disparity of Body Plans 11

all head limbs are highly modified to achieve attachment to a host or to penetrate host tis-
sues (Fig. 1.3F). In cirripedes, the adults lack the antennae, but the nauplius and cypris lar-
vae have limbs in this position to assist in swimming; the second antennae disappear at the
time of attachment prior to metamorphosis to the adult cirripede. Finally, within the wide
array of Cambrian microarthropods that are considered to bear some relationships to modern
groups (see chapter 2), such as Bredocaris (Fig. 1.3G), Martinssonia, Rehbachiella , Skara , and
Walossekia , the so-called second antennae are more often than not locomotory limbs, similar
in structure to the mandibles and maxillae of these fossils.
We can conclude from this brief survey that the only character that Crustacea share at this
position, that is, the first segment posterior to the true antennae, is simply the presence of a pair
of limbs. However, this is to say nothing—the mere presence of limbs on the first postantennal
segment, or any postantennal segment for that matter, is a generalized, primitive, or plesiomor-
phic feature.
As noted above, merostomes, pycnogonids, and arachnids also have a limb in this position,
but that does not make them crustaceomorphs. In arthropods, all segments generally carry
limbs, at least on the head and thorax; it is only when limbs are particularly specialized, or even
missing, that things become more interesting and can serve to help diagnose a group. For exam-
ple, the presence of a limb on this first somite posterior to the antennae in crustaceomorphs
stands in contrast to what occurs in myriapods and insects. In these latter groups, the limb buds
on the first postantennal somite are diverted from forming a limb into producing the special
labrum seen in these groups. We know this is so because, at least for insects, developmental gene
expression studies reveal that the labrum is the “appendage” of the so-called intercalary (first
postantennal) segment (Boyan et al. 2002). This diversion of the first postantennal anlagen into
forming the upper lip rather than a set of limbs clearly is a derived feature. It is the lack of limbs
on the first postantennal segment of insects and myriapods that is a noteworthy and significant
apomorphy, not the mere presence of a limb on that segment as occurs in crustaceomorphs,
cheliceriforms, and many fossil groups such as trilobites.
Crustacea are generally said to have a five-segment head. However, many crustaceomorph
groups include at least one pair of maxillipeds and the associated “thoracic” somite into the
head, and we thus speak of a cephalothorax. Most of the time, it is clear that these maxillipeds are
obviously modified anterior thoracic limbs. Development in the many crustaceomorph groups
that have maxillipeds allows us to document successive stages wherein the maxillipeds become
specialized and their associated somites through successive molts become incorporated into
the cephalon during ontogeny. However, at least one group of crustaceomorphs, the remipedes,
does not exhibit such a transition. Koenemann et al. (2007, 2009) observed no biramous precur-
sor state to the uniramous maxilliped in the earliest larval stages—the remipede maxilliped and
its segment are part of the head in the earliest recognized ontogenetic stages. Consequently, we
could say that the remipedes, for all intents and purposes, have a six-segment head (Koenemann
et al. 2009).
In summary, this first part of the diagnosis of Crustacea (head of five somites, each bearing
a set of appendages consisting of two pairs of antennae, a pair of mandibles, and two pairs of
maxillae) is not informative.

“Body consisting of three regions: head, thorax, and ‘abdomen’”

Body tagmosis is often an important component of defining an arthropod body plan. For exam-
ple, as noted above, among the chelicerates a discrete head is lacking because the anterior seg-
ments associated with feeding and sensation are fused with the segments bearing the walking
limbs to form a solid unit, the prosoma, a very distinctive feature.
12 Functional Morphology and Diversity

The possession of a head, thorax, and abdomen is certainly distinctive, but it is also a feature
shared with insects. Hence, while we might appear to have, with tagmosis, another argument
for seeking some kind of relationship between crustaceomorphs and insects, that is, within a
monophylum Pancrustacea or Tetraconata (Wheeler et al. 2004, Giribet et al. 2005), we do not
have an effective component for a definition that seeks to uniquely define Crustacea.
Furthermore, crustaceomorphs themselves vary considerably in this regard, as we will pur-
sue in more detail below. The number of thoracic segments can be characteristic, but only
for individual crustaceomorphs and not for Crustacea as a whole. Remipedes have a long,
homonomously developed trunk with no differentiation between anterior and posterior sec-
tors. Mystacocarids have five and branchiurans have four thoracomeres. Large-bodied bran-
chiopods often have 11 or 12 thoracomeres. Many maxillopodans and the malacostracans have
seven or eight thoracomeres: maxillopodans sensu stricto have seven, while cephalocarids and
malacostracans have eight.
Moreover, the possession of an abdomen is not a uniting feature. This variability is true
not only regarding external, gross anatomical features such as total numbers of segments and
those with and without paired limbs on the segments, but also for the underlying expression of
Hox (homeobox) family genes as well (Fig. 1.4). In connection with the latter, Abzhanov and
Kaufman (2004) and Schram and Koenemann (2004a) surveyed the available information
concerning Hox gene expression in crustaceans. There are two fundamentally different types
of posterior tagmata: the abdomen, a region without expression of the abd-A (abdominal A)
Hox gene; and the pleon, a region with the expression of abd-A . Species with the latter type,
the malacostracans, possess appendages on the segments and also display a well-differentiated
central nervous system in that body region, whereas species with the former type, which lack
abd-A expression in that body region (branchiopods and maxillopodans), lack appendages on
these segments and do not have a well-differentiated central nervous system in these segments.
It is for this reason that Schram and Koenemann (2004a) concluded that the old term pleon, as
applied to the posterior region of the trunk of malacostracans, is not just an equal and inter-
changeable alternative for the term abdomen; the use of pleon as a descriptor is an absolute neces-
sity. Hox gene expression indicates that the pleon of malacostracans and the abdomen of other
crustaceans exhibit fundamentally different developmental pathways.
Admittedly, the amount of available data is limited. As is the case with developmental work,
researchers focus on the study and manipulation of model organisms. Among malacostracans,
Porcellio scaber and Procambarus clarkii provided the model systems of preference for studies
of Hox patterning, and among entomostracans, Artemia franciscana and Mesocyclops edax have
served as the models, and the latter has been only incompletely investigated. The determina-
tion of Hox gene expression in diverse arthropods was a leading line of research in arthropod
evo-devo studies in the late 1990s and early 2000s, but such investigations have waned, at least
for now. In light of the above phylogenetic usefulness of this line of research, we should look
forward to more animals being investigated in this regard.
Nevertheless, this part of the definition of crustaceans (body consisting of three regions:
head, thorax, and “abdomen”) is not a particularly informative statement.

“Trunk appendages primitively multiramous”

This descriptor is also not very informative. The presence of bi- and/or multiramous limbs is
widely accepted to be a primitive condition in Arthropoda; most authorities would concede
that uniramy is derived. However, here, too, the devil is in the details. Schram and Koenemann
(2001) reviewed the information available concerning early development of crustacean limbs,
and Williams (see chapter 3) delves into this subject more deeply.
Crustacean Biodiversity and Disparity of Body Plans 13

lap pb z2 zen Dfd Scr ftz Antp Ubx abd-A abd-B

Anostraca
Antp (Branchiopoda)
Ubx
abd-A Abd-B

lab
pb Porcellio scaber
Dfd Scr Antp
(Malacostraca)
Ubx abd-A
Abd-B Abd-B

Procambarus clarkii
Scr
Antp (Malacostraca)
Ubx
Abd-B abd-A Abd-B

Mesocyclops edax
Ubx (Maxillopoda)
abd-A Abd-B

Fig. 1.4.
Hox gene expression pattern for various crustaceans. Shaded areas denote thorax or thorax/pleon. Note
the different patterns from an abdomen (no Hox) and a pleon (with abd-A). Modified from Schram and
Koenemann (2004a).

We can summarize here, nevertheless, a few basic patterns of limb development. One,
in which the proximal pedestal of the limb carries distally a tubular, segmented telopod, is
sometimes referred to as the Drosophila model because it was first recognized and studied
in detail using the fruit f ly Drosophila (Cohen 1990). It is the most common pattern of limb
development seen in all biramous crustacean limbs that have been examined, particularly
using Mysidopsis bahia (Panganiban et al. 1995). The limb anlage becomes forked, leading
14 Functional Morphology and Diversity

eventually to the exopodal and endopodal rami. The gene distalless (dll) is expressed at the
tips of the developing rami. A rather different pattern, however, prevails in Branchiopoda,
often referred to as the Artemia model and documented with studies on Artemia and Triops
(Williams and Mü ller 1996, Williams 1998). Rather than a uni- or biramous limb anlage,
limb development begins with a mediolaterally directed ridge upon which eight lobes sub-
sequently appear. The expression of dll occurs in varying patterns on these eight lobes,
which proceed to form the unarticulated, leaf like limb, or corm, characteristic of the bran-
chiopods. Similar gross anatomical sequences of limb development (though without the
related gene expression patterns) have been documented for Cyclestheria (Olesen 1999)
and the cladocerans (Olesen 1998). Hence, the multiramous limb of branchiopods has a
fundamentally different mode of development from that seen in the crustaceans bearing
biramous or uniramous limbs.
Thus, this part of the definition of Crustacea (trunk appendages primitively multiramous)
is not an informative statement. The statement equates all crustaceomorph limbs and ignores
widely divergent, perhaps incompatible, modes of development.

“Development consisting of a series of discrete larval and/or juvenile stages,


initiated by a stage termed a nauplius”

In examining this characteristic sequence, we possibly come upon firmer ground in seeking a
unique set of features to define Crustacea. Many living groups of arthropods exhibit epimorphic
development. The animals essentially hatch with the complete set of segments characteristic of
the adult; the individuals increase in size only with each molt.
Other groups of arthropods (some of the myriapods), although they resemble the adults
in general form, hatch with fewer segments than the adults and add segments with each
molt. Some Crustacea do this; for example, peracarids brood their young, and some of these
are expelled from the marsupium as little “juvenile” forms, called mancas , which eventually
molt and add a segment to achieve the adult condition.
Many crustaceomorphs, however, hatch as larvae, and these larvae not only possess fewer
segments than the adults but also exhibit a distinctive larval form. Successive molts then not
only add segments but also metamorphose the form. Does this constitute an apomorphy for
Crustacea? Other arthropods have larvae. Extensive larval stages are known for the trilobites,
and pycnogonids have a larva; many larvae, both nauplii and other intriguing forms, are known
from the fossil record (see Mü ller and Walossek 1986). However, there are distinctive patterns
of molting and metamorphosis that serve to absolutely unite some crustaceomorphs. Taxa
within Cirripedia are clearly united by the presence of a distinctive nauplius with frontola-
teral horns and a postnaupliar cypris larva in the life cycle. Branchiopods have a characteris-
tic nauplius with a naupliar process on the antennae. Zoeae are diagnostic larvae of decapod
malacostracans.
The nauplius stage is often said to represent a phylotypic stage through which in theory
all Crustacea passed in the course of the evolution of the group. We need to express some
caution here—not all crustaceomorphs begin independent life as a nauplius larva, that is,
exhibiting a larva characterized by possession of only three sets of limbs: the first and sec-
ond antennae and the mandibles. There are crustaceomorphs that do (or did) not begin life
as a nauplius but rather have as the initial stage a metanauplius, that is, a stage with more
than just the three sets of naupliar limbs and/or more than the three naupliar segments.
The issue is confused in the literature with the almost completely interchangeable use of
the terms nauplius and metanauplius . This interchangeability implies that it is almost irrel-
evant as to what the basic structure of the first larva is—if it is tiny, possesses only a small
number of limbs and segments, is given to swimming, and may or may not be filter feeding,
Crustacean Biodiversity and Disparity of Body Plans 15

then it is a “nauplius.” We see here the differences between a structural and a functional
definition.
Which groups have an orthonauplius—a larva with only three pairs of appendages as
seen in Branchiopoda, Maxillopoda, Remipedia, and euphausiacean and dendrobranchiate
Malacostraca? Each of these orthonauplii bears a distinctive form. As noted above, branchio-
pod nauplii possess a naupliar process on the second antenna designed to facilitate feeding.
Variations occur within the Maxillopoda. Among the most distinctive of nauplii, those of
cirripedes bear anterolateral horns, frontal filaments anterior to the first antennae, and a long
caudal process. There are four to six naupliar stages, depending on the group. Copepod nauplii
exhibit a nauplius in almost its complete and pristine state, although the two orthonaupliar stages
are nonfeeding because the gut is not developed until the metanaupliar phase. Ostracodes pass
through a single nauplius stage, but the limbs are not completely developed, and in some species
the early developmental stages (nauplius and the metanauplii) are retained within the mother’s
shell until they are shed near the end of their development. The Cambrian fossil Rehbachiella
had an orthonauplius. Finally, the free nauplii of the euphausiaceans (two) and dendrobranchi-
ates (one) are very simple in form and do not feed, and even the succeeding metanauplii can be
nonfeeding, also true of remipedes. Most of the other eumalacostracans pass through a clear
egg-nauplius phase within the egg (Schram 1986).
The diversity of naupliar form and function led Scholtz (2000) to suggest that we should dis-
tinguish between primary and secondary nauplii, that is, between nauplii that are indeed primi-
tive and an original part of the life cycle, and nauplii that are secondarily reevolved. Scholtz
believes that the primitive stage for malacostracans is the embryonized egg-nauplius and that
the nonfeeding, free nauplii of euphausiaceans and dendrobranchiates actually evolved from
ontogenetic sequences without a free nauplius. One consequence of Scholtz’s observations is
that the nauplius larva would not be a phylotypic stage for all crustaceomorphs.
Other groups of crustaceomorphs exhibit a variety of first stages in their development.
Cephalocarida begin as a metanauplius, the first stage of which has five limbs and a variable
number of limbless segments. Mystacocarida hatch as metanauplii with four sets of limbs and
five additional limbless segments.
The significance of these metanaupliar stages becomes evident when we consider the lar-
val development in certain of the Cambrian Orsten microarthropods. The larval sequences for
many of the Cambrian Orsten taxa are known; Bredocaris, Martinssonia, and Phosphatocopina
all had four sets of limbs in the earliest phases, what Walossek has referred to as a “head larva”
(Walossek and Mü ller 1990). Agnostus and the other trilobites in their earliest stages also bear
four. There seems to be a basis for concluding that the naupliar stage, with its three sets of limbs,
is derived from forms with four (possibly five) sets of limbs.
Larvae are features of aquatic arthropods, but the nauplius larva is undoubtedly a derived
form. Unfortunately, not all crustaceomorphs have a nauplius, which is perhaps a problem
whose full implication remains to be determined; some groups may have lost it, but other groups
probably never had it.
At the beginning of this section I asked the question, What is Crustacea? It appears that
we cannot use an unambiguous set of apomorphic descriptors to diagnose a monophyletic
Crustacea. Developmental patterns and the nauplius larva appear to offer the best chance of
doing so. However, since we have crustaceomorphs that do not exhibit the naupliar stage, we
might conclude that the nauplius has evolved independently several times in the evolution of
crustaceomorphs or has been lost several times; otherwise, if one demands that the nauplius be
treated as diagnostic, Crustacea is not a monophyletic group.
It would appear from the above discussion that we must conclude that crustaceomorphs
are whatever is left over among the arthropods after we have assigned everything else to other
clearly defined monophyletic groups.
16 Functional Morphology and Diversity

WHAT ARE THE CRUSTACEOMORPH BODY PLANS THAT MIGHT


BE MONOPHYLETIC?

We now have a conundrum. If we cannot define a monophyletic Crustacea with a single, consist-
ent set of derived characters, can we perhaps diagnose smaller monophyletic groups within the
current array of crustaceomorphs? I do believe that there are groups within this assemblage that
are monophyletic (Schram and Koenemann 2004a).

Short-Bodied Forms (Oligo-Crustacea)

Branchiura

Two groups of short-bodied forms at first glance would not appear to be at all alike (Fig. 1.5)
but share a similarity regarding gonopore location. The living branchiurans are parasites of
fish with highly modified mouthparts, but their gonopores open on the fourth thoracic somite.
While there appears to be an abdomen, it is not differentiated into segments and is little more
than a single or bilobed sac (Fig. 1.5A–C). Of special note is Pentastomida, the sister group of the
branchiurans. Comparative sperm ultrastructure (Wingstrand 1972) and molecular sequence
studies (Abele et al. 1989) revealed a close link of Branchiura with Pentastomida, odd wormlike
parasites of the respiratory system in higher vertebrates (Fig. 1.5E).
This pairing of branchiurans and pentastomids might appear peculiar, but it is a group
of great age; pentastomid fossils exist from the early Paleozoic Orsten faunas (Walossek and
Mü ller 1994, Walossek et al. 1994). The several species of Cambrian/Ordovician pentastomids
(Fig. 1.5D) can convincingly be compared to living pentastomids (see Walossek and Mü ller 1994,
their fig. 21), although the fossils have trunk limbs but lack the proboscis bearing the mouth.
Walossek and colleagues interpret these fossils as parasites, but there is no direct evidence of
this. These fossils could have been ordinary free-living members of the infauna. Nevertheless,
what the fossils do show without any debate is that, in combination with the sperm and sequence
data above, the ancestry of branchiurans is very ancient.

Mystacocarida

In contrast to the branchiurans, the mystacocarids are microscopic members of the beach meio-
fauna, almost wormlike in form, with a well-developed set of mouthparts, including maxilli-
peds, but with four pairs of rudimentary thoracic limbs (Fig. 1.5F). The gonopores are located
on the fourth thoracic somite. Mystacocarids, too, may be of great age because, in some respects,
they are not unlike Skaracarida, the Cambrian fossil group from the Orsten of Sweden (Mü ller
and Walossek 1985) (Fig. 1.5G).
Schram and Koenemann (2004a), using morphologic analysis tempered by Hox gene expres-
sion, found mystacocarids and branchiurans to be sister taxa. The results of molecular studies
for both of these groups are confusing because long-branch attraction has been a persistent prob-
lem in these analyses; for example, Spears and Abele (1997) encountered this phenomenon when
their results placed mystacocarids, remipedes, and cephalocarids together and in some proxim-
ity to chelicerates (a strange array), and the branchiurans emerged in a clade with podocopan
ostracodes. Giribet et al. (2005) increased both the number of taxa sampled and genes sequenced
but obtained a confusing collection of results depending on variant runs of taxa sampled (with
and without fossils): mystacocarids and branchiurans sometimes appear alongside copepods and
ostracodes; under other circumstances, branchiurans emerge elsewhere. Although the taxon
sampling of Giribet et al. (2005) is impressive for all arthropods (and especially for hexapods), it is
not particularly broad within crustaceomorphs. More recently, Regier et al. (2008) using nuclear
Crustacean Biodiversity and Disparity of Body Plans 17

A
C

D
m E

mp

Fig. 1.5.
Body types of “short-bodied” crustaceans. (A–C) Diverse types of Branchiura (from Schram 1986). (A)
Argulus. Note the highly modified mouthparts for attachment. (B) Dipteropeltis, with highly reduced
body and winglike carapace. (C) Chonopeltis, displaying weak trunk and limb segmentation. (D and E)
Diagrammatic Pentastomida (modified from Walossek and Mü ller 1994, their fig. 21). (D) Diagram similar
to the Cambrian genus Heymonsicambria. Note the reduced trunk limbs (t). (E) Diagram of a general-
ized living pentastomid. m, mouth; g, gonopore; light gray, anterior trunk; dark gray, posterior trunk or
abdomen. (F) Derocheilocaris, a mystacocarid. The arrow indicates approximate location of gonopore on
fourth trunk limb. (G) Skara minuta Mü ller and Walossek, 1985, a Cambrian fossil crustacean that might
represent a mystacocarid stem form. mp, maxilliped.
18 Functional Morphology and Diversity

A
Malacostraca
Maxillopoda
Branchiopoda
Cephalocarida
Remipedia

B
Hexapoda

Xenocarida Remipedia
Leptocarida
Malacostraca
Maxillopoda
Sensus stricto
Branchiopoda
Brachiura
Mandibulata Oligostraca Mystacocarida

Arthropoda Ostracoda
Myriapoda

Chelicerata + Pycnogonida

Onychophora
Tardigrada

Fig. 1.6.
(A) A classic understanding of crustacean phylogenetic relationships, based on morphology. (B) A sum-
mary version of one of the more recent molecular phylogenies. Modified after Regier et al. (2010).

protein coding genes also found branchiurans as a sister taxon to podocopan ostracodes. While the
breadth of their molecular sample was impressive, the taxon sample was again selective; for exam-
ple, no mystacocarid was included. To remedy the situation, Regier et al. (2010) have expanded
the taxon base and increased the number of genes sequenced; their results identified a clear clade
with Mystacocarida, Branchiura, and Pentastomida within a group they termed “Oligostraca”
(Fig. 1.6). The analysis by Koenemann et al. (2010) also placed these short-bodied groups together
these short-bodied groups. Oddly, these clades also contained Ostracoda (see below).
The shortness of the body in these orders imposes definite constraints. The lack of an
elaborated abdomen in branchiurans and pentastomids undoubtedly limits their ability to
move around. One could speculate whether this lack was a factor in both groups adapting para-
sitic lifestyles. So, too, with mystacocarids: the lack of a well-developed abdomen could have
constrained adapting a vermiform, interstitial existence where abilities to swim or otherwise
move around are minimized.

Branchiopoda

Living Branchiopoda

This large and fascinating group is almost exclusively restricted to freshwater, with a few
exceptional cladocerans that are marine. One is tempted to speculate that they might have been
Crustacean Biodiversity and Disparity of Body Plans 19

marine to begin with and then shifted to fresh waters. However, the evidence for that is not
robust, and even most of the fossils, such as they are (mostly conchostracans), are preserved in
freshwater to brackish water situations.
The branchiopods do exhibit a distinctive set of features that outline a body plan for
the group. We have noted above in passing the distinctive mode of limb formation—from
horizontal ridges that subsequently become multilobed, rather than uni- or biramous limb
bud anlagen—and the nauplius larva. Consideration of the large-bodied branchiopods adds
some further depth to our knowledge of the branchiopod bauplan, of which Anostraca serves
as a model.
Traditionally (see Calman 1909, Schram 1986), anostracans were conceived as having
an 11-segment thorax and a 9-segment abdomen. The first two legless, abdominal segments
formed a fused genital complex. However, Hox gene expression studies reveal that the genes
Antennapedia (Antp), Ultrabithorax (Ubx), and abd-A all are expressed in the thorax of Artemia,
with a residual expression of Antp in the genital segments (Fig. 1.4). Abdominal B (abd-B), a
marker for “end of thorax” and the genital segments, occurs in the genital segments (Abzhanov
and Kaufman 2004, Schram and Koenemann 2004a). Hence, the genital segments are better
considered thoracic, rather than abdominal, with the gonopores being carried on the twelfth
segment of the thorax. The abdominal segments posterior to the genital complex do not exhibit
Hox gene expression.
Notostraca and the conchostracans have the gonopores opening on the eleventh or
between the eleventh and twelfth trunk segments. Notostraca (Fig. 1.7A,B) carry a well-
developed pair of limbs on each of the “thoracic” segments (the first two being somewhat
modified from that seen on the others), but posterior to this region the limbs become
increasingly smaller as one moves posterior in the sequence of somites, and there is little
correspondence between the number of limbs and the segment boundaries—there are many
more limbs than apparent segments. Regrettably, as yet no Hox gene expression studies have
been performed on notostracans.
The conchostracans are now generally divided into three monophyletic groups: Laevicaudata
(Fig. 1.7C), Spinicaudata (Fig. 1.7D), and Cyclestherida (Fig. 1.7E) (Martin and Davis 2001), but
they, too, appear to carry the gonopore in a position similar to that of Notostraca. The first
of these, the laevicaudatans, or Lynceidae, have fewer trunk segments than the other two, but
at least the female gonopore opens near the base of the eleventh, or penultimate, appendage
(Linder 1945). The location of the male pore still must be confirmed (Martin et al. 1986). The
other conchostracan groups have many more trunk segments, up to 32, with no differentiation
between segments and limbs posterior to the genital openings, which are said to occur on the
eleventh somite.
Thus, most Branchiopoda feature thin, foliaceous, unjointed limbs, with trunks divided into
an anterior section with well-developed limbs and good Hox expression, and with gonopores at
or near the eleventh or twelfth trunk somite. The four groups of Cladocera (Fig. 1.7F–I), while
they are clearly branchiopods, exhibit extreme forms of body reduction, or oligomery. All these
branchiopods bear either a well-developed carapace or a derivative thereof. Only the anostra-
cans (Fig. 1.7J) lack a carapace, and most authorities place the fairy shrimp as a sister group to all
other branchiopods (Richter et al. 2007).
The restriction of branchiopods to freshwater habitats entices one to wonder why these
groups have become so limited. The unique mode of limb development perhaps precludes the
development of anything other than thin, foliaceous, cormlike appendages. This in turn might
have engendered an overall body habitus that lacks well-sclerotized and/or calcified body
somites. Under these constraints, freshwater habitats, especially transient ones, provide satisfac-
tory refugia.
20 Functional Morphology and Diversity

C
G

Fig. 1.7.
Body types of Branchiopoda. (A and B) Lepidurus arcticus, Notostraca (after Sars 1896): with carapace
removed (A) and with carapace intact (B). (C–E) Various conchostracans. (C) Lynceus gracilicornis,
Laevicaudata (modified from Martin et al. 1986). (D) Limnadia lenticularis, Spinicaudata (after Sars 1896).
(E) Cyclestheria hislopi, Cyclestherida (after Sars 1887). (F–I) Various infraorders of Cladocera (after
Lilljeborg 1901, Birge 1918). (F) Sida crystalline, Ctenopoda. (G) Bosmina longispina, Anomopoda. (H) Podon
intermedius, Onychopoda. (I) Leptodora kindtii, Hoplopoda. (J) Branchinecta lindahli, Anostraca (after
Lynch 1964).

Fossil Stem-Branchiopods

All authorities accept crown group Branchiopoda as a monophyletic group, based on the dis-
tinctive nauplius larva and the form and ontogeny of the trunk limbs. However, the branchio-
pods are also noteworthy in that a number of fossil forms are known that either occupy a stem
position to the branchiopod clade or in some instances actually stand within the group.
Crustacean Biodiversity and Disparity of Body Plans 21

A
B

? E

Fig. 1.8.
Fossil species that might have some relationship to Branchiopoda, either near the base of that group or
as stem forms. Arrows indicate the twelfth thoracomere, the segment at or just posterior to the end of
the thoracic limb series, which might bear the gonopores. (A) Lepidocaris rhyniensis, Devonian (from
Scourfield 1926). (B) Rehbachiella kinnekulensis, Upper Cambrian (from Walossek 1993). (C–E) various
Cambrian waptiids. (C) Chuandianella ovata (from Chen and Zhou 1997). (D) Pauloterminus spinodorsalis
(from Taylor 2002). (E) Waptia fieldensis (from Briggs et al. 1994). (F) Castracollis wilsonae (from Fayers
and Trewin 2003).

The lipostracan Lepidocaris rhyniensis, a unique Devonian fossil (Fig. 1.8A), is preserved in
great detail within nodules of chert. The thoracic limbs are both multiramous (thoracopods
1 and 2) and biramous (thoracopods 3–5). However, the presence of fully developed maxillae
(Schram 1986), rather than the vestigial form characteristic of true branchiopods, indicates
possibly only a sister-group relationship with crown Branchiopoda.
22 Functional Morphology and Diversity

Rehbachiella (Fig. 1.8B) has figured prominently in discussions of branchiopod origins


(Walossek 1993). However, Schram and Koenemann (2001) took exception with this and con-
cluded that Rehbachiella, while possibly a stem form, was not a branchiopod sensu stricto since
not only do they possess biramous thoracopods, but also these limbs arise from biramous anla-
gen and not the multilobed ridge of true branchiopods. Hence, I believe Rehbachiella, at best, is
a stem form.
Another fossil group from the Cambrian could be relevant to understanding stem evolu-
tion of branchiopods, the waptiids (Fig. 1.8C–E). The genus Waptia from the Burgess Shale of
Canada is probably the most famous. However, several genera are known (Briggs et al. 1994,
Chen, and Zhou 1997, Taylor 2002) and all appear to have a subdivided thorax with apparently
four anterior telopodous limbs and six posterior foliaceous limbs. The gonopores have yet to be
identified for waptiids, but I would venture a guess that they probably occurred on the eleventh
trunk segment.
Although the recent large-scale molecular analyses of Giribet et al. (2005) and Wheeler et al.
(2004) typically find Hexapoda as a sister group to all crustaceomorphs, there is an alterna-
tive hypothesis. Schram and Koenemann (2004b), VanHook and Patel (2008), Lartillot and
Philippe (2008), and Dell’Ampio et al. (2009) obtained trees with insects and Branchiopoda
as sister groups. These results were based on developmental gene expression patterns and mol-
ecule sequences, and these trees serve to propose alternative hypotheses concerning branchio-
pod relationships.
Finally, there are fossils such as Castracollis wilsonae (Fig. 1.8F) that exhibit body plans
that are complex but nevertheless place them within branchiopods, in this case 11 large, folia-
ceous, cormlike limbs on the anterior thorax followed by another series of similar limbs but
much reduced in size (Fayers and Trewin 2003). Castracollis might or might not have had a
carapace.

Eucrustacea

What remains of the crustaceomorph taxa after clades of short-bodied and branchiopo-
dan types are isolated is a confederation of diverse forms: Cephalocarida, Malacostraca,
Remipedia, and Maxillopoda. When viewed as a whole, these taxa are divergent in terms of
both habitus and habitat; nevertheless, all these groups bear gonopores on the sixth through
eighth thoracic somites. There are a couple of interesting exceptions to this rule, which I note
below.

Cephalocarida

This group of hermaphrodites is small both in size and in species numbers. It has a thorax of
eight segments and a limbless abdomen of 12 segments (Fig. 1.9A). The form of the maxillae is
very similar to that seen for the thoracopods. The gonopores are located on the sixth thoracom-
ere. Nothing is known of Hox gene expression in cephalocarids.
The body plan of cephalocarids might exhibit the results of the same sorts of constraints
we saw above with mystacocarids. In this case, the elongate, limbless abdomen with extended
terminal caudal rami at best probably functions like the tail on a kite: a stabilizer to mini-
mize drag and the effect of turbulence as the animals swim. The long series of thoracic limbs
developed as swimming paddles provide more locomotory abilities than that seen in the tiny
thoracopods of mystacocarids, but nonetheless, competition from larger and more mobile
forms probably forced the cephalocarids to retreat to f locculent bottom sediments in order
to make a living.
Crustacean Biodiversity and Disparity of Body Plans 23

B
A

F D

Fig. 1.9.
The four major groups among the core bauplan of Crustacea, with gonopore-bearing segments indicated
by arrows. (A) A cephalocarid, Hutchinsoniella macracantha, a hermaphrodite with pores on the sixth tho-
racomere (modified from Schram 1986). (B) A hoplocarid Malacostraca, a male Squilla mantis. The male
pore (long arrow) would be on the eighth thoracomere; the female pore (short arrow) would be on the
sixth thoracomere (modified from Calman 1909). (C and D) Two types of eumalacostracan Malacostraca:
with (C) and without (D) a carapace. (C) A euphausiid, Meganyctiphanes norvegica . The male pore is on
the eighth thoracomere (long arrow); the female pore, on sixth thoracomere (sort arrow) (modified from
Mauchline and Fisher 1969). (D) The syncarid Anaspides tasmaniae. The male pore is on the eighth tho-
racomere (long arrow); the female pore, on sixth thoracomere (short arrow) (modified from Schminke
1978). (E and F) Two types of remipede, hermaphrodites with the male pore on the eighth thoracomere
(long arrow) and the female pore on the fifteenth thoracomere (short arrow). (E) Medium-length body,
Speleonectes gironensis (modified from Yager 1994). (F) Short-length body, Micropacter yagerae (modified
from Emerson and Schram 1991). (G) A typical maxillopodan, Calanus finmarchicus. Male and female
pores open on seventh thoracomere (modified from Calman 1909).
24 Functional Morphology and Diversity

Malacostraca

This most variable of crustacean groups nevertheless has a fundamentally uniform structural
plan. The trunk is divided into an anterior thorax of eight segments and a posterior pleon of six
or seven segments, sometimes fewer. All trunk somites generally bear appendages, but note-
worthy variations can occur, such as one or more thoracopods serving as maxillipeds or pos-
terior thoracopods and/or pleopods being greatly reduced or absent. Hox genes are expressed
throughout the body (Abzhanov and Kaufman 2004, Schram and Koenemann 2004a) with Ubx
characteristic of the thorax and abd-A of the pleon (Fig. 1.4). The female gonopores occur in
association with the sixth thoracic segment, while the male pores are on the eighth.
The malacostracans are typically said to contain three groups: the small nectobenthic lepto-
stracans (not illustrated), the obligate carnivorous hoplocaridans (Fig. 1.9B), and the extremely
diverse caridoid eumalacostracans that have forms both with a carapace (Fig. 1.9C) and with-
out (Fig. 1.9D). The diversity of this group is examined in greater detail in other chapters in this
volume.
We might say that the great versatility imparted by the malacostracan body plan is respon-
sible for its success. The long series of limbs, extending through both the thorax and pleon,
allows a great degree of variation and specialization that undoubtedly has allowed the group
to radiate to the extent it has, with great numbers of species and remarkable variations in
structure.

Maxillopoda

With the problematic Mystacocarida and Branchiura removed from the maxillopodans, where
textbooks often place them, there remains a core set of taxa that appear to conform to a single
body plan. The old formula of 5–6-5 or the newer viewpoint of 5–7-4—five cephalic, seven tho-
racic, and four abdominal somites (see Newman 1987)—has great consistency throughout the
group. The old interpretation was of a thorax with six limb-bearing segments and an abdomen
of five segments always lacking limbs. An alternative interpretation of the gonopore-bearing
segment as actually part of the thorax (Newman 1987) leaves only four abdominal somites; this
interpretation makes more sense not only in terms of what we can see in other groups, for exam-
ple, the free gonopore-bearing segment of the anostracans that occurs just posterior to a set of
trunk limbs mentioned above, but also in terms of what limited information we have concerning
Hox gene expression, with Ubx and Abd-A expression in the thorax and abd-B in the genital seg-
ment (Averof and Patel 1997).
Copepoda (Fig. 1.9G) most clearly present the pattern of 5–7-4. The gonopores of both
sexes occur on the seventh thoracomere. Thecostraca conform to the basic maxillopodan pat-
tern with some variations. Ascothoracica exhibit 5–7-4. Facetotecta appear to manifest 5–7-3,
based on the anatomy of the Y-cypris. Cirripedia exhibit 5–7-0, considering the cypris larva as
a stand-in model for the highly derived adults. While male gonopores in the cirripedes appear
on the seventh thoracic segment, the female pore has shifted forward onto the first thoracic seg-
ment. Furthermore, the cirripedes lack an abdomen and coincidently also lack any expression of
Abd-A (Mouchel-Vielh et al. 1998).
The parasitic Tantulocarida present problems since these microscopic forms have an
extremely aberrant life cycle. However, recent advances in elucidating that life cycle (Huys
et al. 1993) allow us to conclude that the tantulocarids express a 5–7-2 pattern, with the
male gonopore appearing on the seventh thoracic somite and the single median female pore
occurring on the first. This latter feature clearly unites tantulocarids and thecostracans as
sister groups.
Crustacean Biodiversity and Disparity of Body Plans 25

The constraints exerted by a limbless abdomen on lifestyle may explain much of what we
see in maxillopodan evolution. The maxillopodans certainly thrive under unusual conditions.
Parasitism is widespread in the group, especially among thecostracans, and those thecostracans
that are not parasites have lost the abdomen altogether and settled (literally) into the completely
sedentary, highly aberrant body plan seen in the barnacles. Only the copepods possess the kind
of biodiversity and habitat variability we associate with “successful” groups. Even so, the small
sizes of copepods could be related to the limits engendered by an abdomen lacking limbs.

Ostracoda

These animals remain the most vexing of arthropods to place phylogenetically and, if molecular
sequences are to be believed, may not be a monophyletic group. Their extreme reduction of body
plan (oligomery), complete enclosure within a calcareous shell, and specializations directed at
life carried on at a microscale have hindered attempts to link them to other crustaceomorphs.
There are contentious debates about homologies within Ostracoda (Horne et al. 2005), and
ostracodes do not appear to share obvious apomorphies with other crustaceomorphs. Most
textbooks and reference books consign ostracodes to the maxillopodans (see Schram 1986), but
that is more of a default placement.
K. Martens (personal communication, 2004) and R.A. Jenner (personal communication,
2009) expressed an informal view of at least some researchers that Ostracoda might not be a
monophyletic group. This possibility obtains some support from molecular data that some-
times finds Podocopa and Myodocopa in different parts of cladograms (see Spears and Abele
1997, Regier et al. 2008, 2010, Koenemann et al. 2010). However, most of these analyses have a
very limited taxon sample with sequences from only a handful of ostracode species.
There is much variation in form in ostracode limbs, but there is a consensus at least that both
the myodocopes and the podocopes are themselves monophyletic (Horne et al. 2005). However,
debates about the number of somites in each group are not settled. At first glance, one perceives
that only very few thoracic segments bear limbs, but Schulz (1976) presented some evidence that
indicates Cytherella pori, a podocopan, might have 11 trunk somites (a 5–7-4 pattern) and that
the penis appears to be associated with the sixth or seventh of these segments (see Schram 1986,
their fig. 33–1C). Tsukagoshi and Parker (2000) confirmed this in other species of podocopans.
No similar information is available yet for Myodocopa.
From this, it might appear that podocopan ostracodes are possibly maxillopodans. Some
authorities classified Ostracoda as a subclass of Maxillopoda (Schram 1986), but others main-
tain them as an independent class (Martin and Davis 2001). Most recently, Koenemann et al.
(2010) and Regier et al. (2010), on the basis of molecular sequences, obtained both podocopan
and myodocopan ostracodes as a sister group to a clade of mystacocarids, branchiurans, and
pentastomids. This arrangement would then unite all the short-bodied “oligostracans” into a
single clade near the base of the crustaceomorph tree (Fig. 1.6).
However, there existed Paleozoic, especially Cambrian, taxa that may have some bearing on
eventually determining ostracode affinities. Several such groups are under active study, such
as the bradoriids, Phosphatocopida (Maas et al. 2003), and perhaps even the thylacocephalans.
These groups will eventually have to be integrated into any classification of the crustaceo-
morphs, and undoubtedly they will prove very interesting in this regard.

Remipedia

This most recently discovered group of crustaceomorphs is noteworthy for several rea-
sons. The trunk is not differentiated into a thorax and abdomen/pleon. If we consider the
26 Functional Morphology and Diversity

maxilliped-bearing segment as a modified trunk somite (even though it is completely merged


into the cephalon), then the female gonopore occurs on the eighth postmaxillary segment, and
the male gonopore, on the fifteenth. However, as noted above (Koenemann et al. 2007, 2009),
the distinctive maxillipeds, virtually identical in general form to the maxillae, display no devel-
opmental evidence that this limb is modified from a thoracopod format. In addition, the number
of trunk segments is not fixed, either within or between species (Koenemann et al. 2006), with
many long-bodied forms recognized (Fig. 1.9E)—although there appears to be at least a lower
limit of 16 trunk segments in the adults (Fig. 1.9F). The significance of all this variability remains
to be explored.
The remipede body plan ensured that these animals are excellent swimmers, on a par with
anything seen among the malacostracans. Even so, their habitat restrictions are quite profound;
they prefer anchialine cave habitats in low-oxygen conditions.

CLASSIF ICATION

The above review indicates there have to be changes in our concepts of crustaceomorph
classification, but this is not the place to present any new or radical higher taxonomy. In princi-
ple, we want our taxonomies to reflect phylogeny, but that is not always possible. There is much
conflicting evidence from molecular analyses, which along with morphological data often suf-
fers from limited taxon sampling, and the latter often ignores or minimizes input from fossils.
We still need to more effectively integrate data from gross morphology, molecular sequencing,
and paleontology into a coherent whole. Nevertheless, we should extend some effort to rec-
ognize the monophyletic groups about which we are certain (Fig. 1.10); there are patterns that
should be acknowledged.
To these ends, we can make good use of the concept of the plesion, a particular taxon that
does not fit well into another category and that eventually might be assigned to its own higher
category. I believe that, in this instance, we should begin to think of the infraphyla below
as monophyletic groups on a par with other well-established arthropod monophyla such as
Hexapoda, Chelicerata, Trilobita, and Pycnogonida. What fossils and where they will fall
within or between these monophyletic groups will be explored elsewhere. The scheme is not
complete in terms of all possible fossil plesions but does include most of those mentioned in
the text above (Table 1.4).

WHAT MIGHT THE CRUSTACEOMORPH ANCESTOR HAVE LOOKED LIKE?

At one time, there was a fair consensus as to what the ancestor of Crustacea might have looked
like. Hessler and Newman (1975) devised an ancestor with a long, homonomously segmented
body, each segment bearing a set of limbs not unlike a cephalocarid, for which Newman preferred
a form with a carapace, and Hessler one without (Fig. 1.11A). Cisne (1982) believed that crusta-
ceans arose from a trilobite-like ancestor. Schram (1982) concurred with Hessler and Newman
(1975), although he would have preferred a somewhat more foliaceous limb, intermediate between
cephalocarids and branchiopods (Fig. 1.11B). However, Schram’s 1982 paper had been written in
1978 (delayed due to a delay in the publication of the book in which it appeared), and in the inter-
vening years the remipedes had come to light. By 1983, Schram had altered his views as to the
form of an ancestor, which, while still in possession of a long homonomous body, was viewed
as equipped with biramous, paddlelike limbs (Schram 1983). Schram positioned this biramous
theory as an alternative hypothesis to the mixopodial theory of Hessler and Newman, and this
then postulated a remipede-like alternative ancestor as opposed to a cephalocarid-like forebear.
Crustacean Biodiversity and Disparity of Body Plans 27

4
A

Mystacocarida
6-8
B

Maxillopoda

Malacostraca

Remipedia

Cephalocarida
12
C
Branchiopoda

Fig. 1.10.
Major crustaceomorph body plans based on gonopore position. (A) Mystacocarida, an “oligostracan.” (B)
“Eucrustaceans.” (C) Branchiopoda. From Schram and Koenemann (2004a).

The debate outlined above was based on morphology. Some information derived from
molecular sequences now suggests that hexapods could factor into this mix. One fossil that
might have facilitated a visual understanding of how this transition might have occurred is
Wingertschellicus backesi Briggs and Bartels, 2001 (= Devonohexapodus bocksbergensis Haas
et al., 2003). A recent reexamination of all available fossils of this species from the famous
Devonian Hunsr ück Shale (K ü hl and Rust 2009) synonymized the two names, but the origi-
nal reconstruction of Haas et al. (2003) presented a strange chimera—it appears to have a
dragonf ly anterior end and a very long myriapodous posterior end (Fig. 1.11C). Although the
interpretation of Haas et al. (2003) of D. bocksbergensis offered a head (of possibly four segments)
and a short three-segment thorax followed by a long abdomen, the new interpretation presents
a six- or seven-segment head, with the posteriormost three pairs of cephalic limbs as long, pos-
sibly prehensile appendages and followed by a long trunk with biramous limbs. Neither Briggs
and Bartels (2001) nor Kü hl and Rust (2009) offer a reconstruction of W. backesi, but the latter
believe that this species is neither a stem hexapod nor within crown group Malacostraca. Of
these I am not so sure, having once had the opportunity to examine D. bocksbergensis courtesy
of Dieter Walossek. Even though what this Devonian species might represent remains uncer-
tain, nevertheless, it does demonstrate that there is an abundance of long-bodied forms in the
Paleozoic that may have significance for understanding the early evolution and possible origins
of surviving groups of arthropods.
Another source of information that is relevant for understanding crustacean ancestry is
derived from the study of the Cambrian Orsten microfossils (a few of which were mentioned
28 Functional Morphology and Diversity

Table 1.4. A classification of tetraconate arthropods with inclusion of fossil plesions.

Subphylum: Tetraconata (= Crustaceomorpha = Pancrustacea)


Infraphylum: Hexapoda
Infraphylum: unnamed (short-bodied crustaceomorphs—“Oligostraca”)
Class Branchiura
Order: Arguloida
Order: Pentastomida
Class: Mystacocarida
Plesion: Skaracarida
Plesion: Ostracoda (one possible position; includes Myodocopa and Podocopa)
Infraphylum: Branchiopoda
Class: Phyllopoda (= Calmanostraca)
Order: Laevicaudata
Order: Notostraca
Order: Spinicaudata
Order: Cyclestherida
Order: Cladocera
Class: Sarsostraca
Order: Anostraca
Plesion: Lipostraca (= Lepidocaris)
Plesion: Rehbachiellida (= Rehbachiella)
Plesion: Waptiidae
Infraphylum: Crustacea
Class: Cephalocarida
Class: Maxillopoda
Subclass: Copepoda
Subclass: Thecostraca
Infraclass: Ascothoracica
Infraclass: Cirripedia
Infraclass: Facetotecta
Infraclass: Tantulocarida
Plesion: Ostracoda (one possible position; includes Myodocopa and Podocopa)
Class: Malacostraca
Subclass: Eumalacostraca
Subclass: Hoplocarida
Subclass: Phyllocarida
Class: Remipedia

Cephalocarida and Remipedia might constitute a single class, Xenocarida, based on molecular evidence. Ostracoda could
occupy two possible positions: among oligostracans, based on molecular data, or within maxillopodans, based on some
morphological data. Hoplocarida and Eumalacostraca (sensu stricto) could be arranged as a single subclass Eumalacostraca
(sensu lato) with infraclasses Caridoida and Hoplocarida.

above), and these have raised the possibility of alternative hypotheses. Incompletely under-
stood in the 1980s, the depth of knowledge about these animals is now astounding, extend-
ing as it does to even developmental stages for many of these species (see chapter 2). The full
impact of these studies remains to be assessed within the larger framework of the anatomy of
modern forms, molecular sequences, and gene expressions, but much of this work suggests a
Crustacean Biodiversity and Disparity of Body Plans 29

A B

A'

Fig. 1.11.
Crustaceomorph ancestors (see text for details). (A and A′) Without and with a carapace, according to
Hessler and Newman (1975). (B) According to Schram (1982). (C) Devonohexapodus bocksbergensis (from
Haas et al. 2003).

possible alternative hypothesis: a short-bodied ancestor rather than a long-bodied one. This
merits consideration (Schram and Koenemann 2004a), but it is not possible or appropriate to
examine here.

CONCLUSIONS

It would appear that we are little closer to understanding the origin of crustaceomorphs than
we were 30 years ago. While the larger assemblage of the crustaceomorphs (or pancrustaceans,
or tetraconatans, if you prefer) might be in some way monophyletic, just how it can (or even if
it can) be diagnosed with a single set of apomorphies is not clear at this point. There are, how-
ever, good monophyletic groups within this vast array that can be clearly defined. Furthermore,
these body plans appear to be constrained regarding biodiversity, functional morphology, and
habitats they can occupy. In addition, we have a growing array of fascinating fossil taxa scattered
within and between these monophyla, but how these are related to the monophyletic groups for
which they may serve as stem forms remains to be determined.
But take heart! It is a time not to mourn the demise of the monophylum Crustacea but to
embrace what will be a new world order and a better understanding of this whole branch of the
crustaceomorph arthropods.

ACKNOWLEDGMENTS

I am grateful for the invitation from Profs. Martin Thiel and Les Watling to write this chap-
ter. I also thank Dr. Ronald Jenner (Natural History Museum, London) for reading a draft and
offering some valuable comments and suggestions. Prof. Stefan Koenemann (Hannover) was
involved in producing some of the earlier papers from which parts of this chapter developed.
Any and all faults, however, are my own.
30 Functional Morphology and Diversity

REFERENCES

Abele, L.G., W. Kim, and B.E. Felgenhauer. 1989. Molecular evidence for the inclusion of the phylum
Pentastomida in the Crustacea. Molecular Biology and Evolution 6:685–691.
Abzhanov, A., and T.C. Kaufman. 2004. Hox genes and tagmatization of the higher Crustacea
(Malacostraca). Crustacean Issues 15:43–74.
Averof, M., and N.H. Patel . 1997. Crustacean appendage evolution associated with changes in Hox gene
expression. Nature 388:682–686.
Birge, E.A. 1918. The water fleas (Cladocera). Pages 676–740 in H.B. Ward and G.C. Whipple, editors.
Freshwater biology. Wiley, New York.
Bouchet, P. 2006. The magnitude of marine biodiversity. Pages 33–62 in C.M. Duarte, editor. The explo-
ration of marine biodiversity: Scientific and technical challenges. Fundación BBVA, Bilbao, Spain.
Boxshall, G.A. 1983. Three new genera of misophrioid copepods, with special reference to
Benthomisophria palliata . Philosophical Transactions of the Royal Society of London Series B
297:125–181.
Boyan, G.S., J.L.D. Williams, S. Posser, and P. Bäunig. 2002. Morphological and molecular data argue
for the labrum being non-apical, articulated, and the appendage of the intercalary segment in the
locust. Arthropod Structure and Development 31:65–76.
Briggs, D.E.G., and C. Bartels . 2001. New arthropods from the Lower Devonian Hunsr ück Slate (Lower
Emsian, Rhennish Massif, Western Germany). Palaeontology 44:275–303.
Briggs, D.E.G., D.H. Erwin, and F.J. Collier. 1994 . The fossils of the Burgess Shale. Smithsonian
Institution Press, Washington, DC.
Brusca, R.C., and G.J. Brusca . 1990. Invertebrates. Sinauer, Sunderland, MA.
Brusca, R.C., and G.J. Brusca . 2003. Invertebrates, 2nd ed. Sinauer, Sunderland, MA.
Calman, W.T. 1909. Crustacea. Part VII, 3rd fasc., in E.R. Lankester, editor. A treatise on zoology, Vol. 7.
Adam and Charles Black, London.
Chen, J.-Y., and G.-Q. Zhou. 1997. Biology of the Chengjiang fauna. Bulletin of the National Museum of
Natural Sciences (Taiwan) 10:11–105.
Cisne, J.L. 1982. Origin of Crustacea. Pages 65–92 in L.G. Abele, editor. Biology of the Crustacea, Vol. 1.
Academic Press, New York .
Cohen, S.M. 1990. Specifications of limb development in the Drosophila embryo by positional cues from
segmentation genes. Nature 343:173–177.
Dell’Ampio, E., N.U. Szucsich, A. Carapelli, F. Frati, G. Steiner, A. Steinacher, and G. Pass . 2009. Testing
for misleading effects in the phylogenetic reconstruction of ancient lineages of hexapods: Influence
of character dependence and character choice in analyses of 28S rRNA sequences. Zoologica
Scripta 38:155–170.
Emerson, M.J., and F.R. Schram . 1991. Remipedia. Part 2. Proceedings of the San Diego Society of
Natural History 7:1–52.
Fayers, S.R., and N.H. Trewin. 2003. A new crustacean from the Early Devonian Rhynie chert,
Aberdeenshire, Scotland. Transactions of the Royal Society of Edinburgh, Earth and
Environmental Science 93:355–382.
Giribet, G., S. Richter, G.D. Edgecombe, and W. Wheeler. 2005 . The position of crustaceans within
Arthropoda—evidence from nine molecular loci and morphology. Crustacean Issues 16:307–352.
Groombridge, B., and M.D. Jenkins. 2000. Global diversity: Earth’s living resources in the 21st century.
World Conservation Press, Cambridge.
Grygier, M.J. 1981. Gorgonolaureus muzikae sp. nov. parasitic on a Hawaiian gorgonian, with special refer-
ence to its protandric hermaphroditism. Journal of Natural History 15:1019–1045.
Haas, F., D. Waloszek , and R. Hartenberger. 2003. Devonohexapodus bocksbergensis, a new marine
hexapod from the Lower Devonian Hunsr ück Slates, and the origin of Atelocerata and Hexapoda.
Organic Diversity and Evolution 3:39–54.
Hessler, R.R. 1969. A new species of Mystacocarida from Maine. Vie et Milieu 22:105–116.
Hessler, R.R., and W.A. Newman . 1975 . A trilobitomorph origin for the Crustacea. Fossils and Strata
4:437–459.
Crustacean Biodiversity and Disparity of Body Plans 31

Horne, D.J., I. Schön, R.J. Smith, and K. Martens . 2005 . What are the Ostracoda? A cladistic analysis
of the extant superfamilies of the subclasses Myodocopa and Podocopa (Crustacea: Ostracoda).
Crustacean Issues 16:249–273.
Huys, R., G.A. Boxshall, and R.J. Lincoln. 1993. The tantulocaridan life cycle: The circle closed? Journal
of Crustacean Biology 13:432–442.
Koenemann, S., R.A. Jenner, M. Hoenemann, T. Stemme, and B.M. von Reumont. 2010. Arthropod phy-
logeny revisited, with a focus on crustacean relationships. Arthropod Structure and Development
39:88–110.
Koenemann, S., J. Olesen, F. Alwes, T.M. Iliffe, M. Hoenemann, P. Ungerer, C. Wolff, and G. Schultz.
2009. The post-embryonic development of Remipedia (Crustacea)—additional results and new
insights. Development Genes and Evolution 219:119–129.
Koenemann, S., F.R. Schram, A. Bloechl, T.M. Iliffe, M. Hoenemann, and C. Held . 2007.
Post-embryonic development of remipede crustaceans. Evolution and Development 9:117–121.
Koenemann, S., F.R. Schram, and T.M. Iliffe. 2006. Trunk segmentation patterns in Remipedia.
Crustaceana 79:607–631.
Kü hl, G., and J. Rust . 2009. Devonohexapodis bocksbergensis is a synonym of Wingertshellicus backesi
(Euarthropoda)—no evidence for marine hexapods in the Devonian Hunsr ück Sea. Organisms
Diversity and Evolution 9:215–231.
Lang , K. 1953. Apseudes hermaphroditicus, a hermaphroditic tanaid from Antarctica. Arkiv för Zoologie
(2) 9:341–350.
Lartillot, N., and H. Philippe 2008. Improvement of molecular phylogenetic inference and the phylogeny
of Bilateria. Philosophical Transactions of the Royal Society of London Series B 363:1463–1472.
Lilljeborg , W. 1901. Caldocera Suiciae. Nova Acta Regiae Societatis Scientarium Upsaliensus. Uppsala
3:1–701.
Linder, F. 1945 . Affinities within the Branchiopoda, with some notes on some dubious fossils. Arkiv för
Zoologie 37:1–28.
Lynch, J.E. 1964 . Packa’s and Pearse’s species of Branchinecta: Analysis of nomenclatural involvement.
American Midland Naturalist 71:466–488.
Maas, A., D. Waloszek , and K. Mü ller. 2003. Morphology, ontogeny, and phylogeny of the
Phosphatocopina (Crustacea) from the Upper Cambrian “Orsten” of Sweden. Fossils and Strata
49:1–238.
Martin, M.F. 1932. On the morphology and classification of Argulus. Proceedings of the Zoological
Society of London 1932:771–806.
Martin, J.W., and G.E. Davis. 2001. An updated classification of the recent Crustacea. Science Series,
Natural History Museum of Los Angeles County 39:1–124.
Martin, J.W., and G.E. Davis. 2006. Historical trends in crustacean systematics. Crustaceana
79:1347–1368.
Martin, J.W., B.E. Felgenhauer, and L.G. Abele. 1986. Redescription of the clam shrimp Lynceus gracili-
cornis (Packard) (Branchiopoda, Conchostraca, Lynceidae) from Florida, with notes on its biology.
Zoologica Scripta 15:221–232.
Mauchline, J., and L.R. Fisher. 1969. The Biology of Euphausiids. Advances in Marine Biology 7:1–454.
May, R.M. 1988. How many species are there on Earth? Science 241:1441–1449.
Meglitsch, P.A., and F.R. Schram. 1991. Invertebrate zoology, 3rd ed. Oxford University Press, New York .
Minelli, A. 1993. Biological systematics. Chapman and Hall, London .
Minelli, A., G. Fusco, and S. Sartori. 1991. Self-similarity in biological classifications. BioSystems
26:89–97.
Mouchel-Vielh, E., C. Rigolot, J.-M. Gibert, and J.S. Deutsch . 1998. Molecules and the body plan: The
Hox genes in cirripedes (Crustacea). Molecular Phylogenetics and Evolution 9:382–389.
Mü ller, K.J., and D. Walossek. 1985 . Skaracarida, a new order of Crustacea from the Upper Cambrian of
V ä stergotland, Sweden. Fossils and Strata 17:1–65.
Mü ller, K.J., and D. Walossek. 1986. Arthropod larvae from the Upper Cambrian of Sweden. Transactions
of the Royal Society of Edinburgh, Earth and Environmental Science 77:157–179.
Mü ller, K.J., and D. Walossek. 1988. External morphology and larval development of the Upper Cambrian
maxillopod Bredocaris admirabilis. Fossils and Strata 23:170.
32 Functional Morphology and Diversity

Newman, W.A. 1987. Evolution of cirripedes and their major groups. Crustacean Issues 5:3–42.
Olesen, J. 1998. A phylogenetic analysis of the Conchostraca and Cladocera (Crustacea, Branchiopoda,
Diplostraca). Zoological Journal of the Linnean Society 122:491–536.
Olesen, J. 1999. Larval and post-larval development of the branchiopod clam shrimp Cyclestheria hislopi
(Crustacea, Branchiopoda, Conchostraca, Spinicaudata). Acta Zoologica 80:163–184.
Panganiban, G., A. Sebring , L. Nagy, and S. Caroll . 1995 . The development of crustacean limbs and the
evolution of arthropods. Science 270:1363–1366.
Regier, J.C., J.W. Shultz , A.R.D. Ganley, A. Hussey, D. Shi, B. Ball, A. Zwick, J.E. Stajich, M.P.
Cummings, J.W. Martin, and C.W. Cunningham . 2008. Resolving arthropod phylogeny: Exploring
phylogenetic signal within 41 kb of protein-coding nuclear gene sequence. Systematic Biology
57:920–938.
Regier, J.C., J.W. Shultz , A. Zwick, A. Hussey, B. Ball, R. Wetzer, J.W. Martin, and C.W. Cunningham.
2010. Arthropod relationships revealed by phylogenomic analysis of nuclear protein-coding
sequences. Nature 463:1079–1083.
Richter, S., J. Olesen, and W.C. Wheeler. 2007. Phylogeny of Branchiopoda (Crustacea) based on a
combined analysis of morphological data and six molecular loci. Cladistic 23:1–36.
Ruppert, E.E., and R.D. Barnes. 1994 . Invertebrate zoology, 6th ed. Saunders College Publications, Fort
Worth, TX.
Sars, G.O. 1887. On Cyclestheria hislopi (Baird), a new generic type of bivalved phyllopod raised from
dried Australian mud. Christiania Videnskabernes Selskab Fordhandlinger 1:1–65.
Sars, G.O. 1896. Phyllocarida and Phyllopoda. Fauna Norvegiae, Vol 1. Stock, Christiania, Aschehoug.
Scholtz , G. 2000. Evolution of the nauplius stage in malacostracan crustaceans. Journal of Zoological
Systematics and Evolution Research 38:175–187.
Schminke, H.K. 1978. Die phylogenetische Stellung der Stygocarididae—unter besonderer
Berucksichtigung morphologischer Ahnlichkeiten mit Larvenformen der Eucarida. Zeitschrift f ü r
zoologische Systematik und Evolutionsfoschung 16:225–239
Schram, F.R. 1979. Crustacea. Pages 238–244 in R.W. Fairbridge and D. Jablonski, editors. The encyclo-
pedia of paleontology. Dowden, Hutchinson, and Ross, Stroudsberg, PA .
Schram, F.R. 1982. The fossil record and evolution of Crustacea. Pages 93–147 in L.G. Abele, editor. The
biology of Crustacea, Vol. 1. Academic Press, New York .
Schram, F.R. 1983. Remipedia and crustacean phylogeny. Crustacean Issues 1:23–28.
Schram, F.R. 1986. Crustacea . Oxford University Press, New York .
Schram, F.R. 2003. Our evolving understanding of biodiversity through history and its impact on the recog-
nition of higher taxa in Metazoa. Pages 359–368 in A. Legakis, editor. The new panorama of animal evo-
lution, Proceedings of the 18th International Congress of Zoology. Pensoft Publishers, Sofia, Bulgaria.
Schram, F.R., and S. Koenemann . 2001. Developmental genetics and arthropod evolution: Part I, on legs.
Evolution and Development 3:343–354
Schram, F.R., and S. Koenemann. 2004a . Developmental genetics and arthropod evolution: Part II, on
body regions of crustaceans. Crustacean Issues 15:75–92.
Schram, F.R., and S. Koenemann. 2004b. Are the crustaceans monophyletic? Pages 319–329 in J. Cracraft
and M.J. Donaghue, editors. Assembling the tree of life. Oxford University Press, New York .
Schram, F.R., J. Yager, and M.J. Emerson. 1986. Remipedia. Part 1. Systematics. San Diego Society of
Natural History Memoires 15:1–60.
Schulz , K. 1976. Das Chitinskelett der Podocopida (Ostracoda, Crustacea) und die Frage der Metamerie
dieser Gruppe. PhD dissertation, University of Hamburg.
Scourfield, D.J. 1926. On a new type of crustacean from the Old Red Sandstone—Lepidocaris rhyniensis.
Philosophical Transactions of the Royal Society of London Series B 214:153–187.
Spears, T., and L.G. Abele. 1997. Crustacean phylogeny inferred from 18S rDNA. Pages 169–186 in R.A.
Fortey and R.A. Thomas, editors. Arthropod relationships. Chapman and Hall, London .
Taylor, R.S. 2002. A new bivalved arthropod from the early Cambrian Sirius Passet fauna, North
Greenland. Palaeontology 45:97–123.
Tsukagoshi, A., and A.R. Parker. 2000. Trunk segmentation in some podocopine lineages in Ostracoda.
Hydrobiologia 419:15–30.
Crustacean Biodiversity and Disparity of Body Plans 33

VanHook , A.M., and N.H. Patel. 2008. Crustaceans. Current Biology 18:R547–R550.
Walossek , D. 1993. The Upper Cambrian Rehbachiella and the phylogeny of Branchiopoda and Crustacea.
Fossils and Strata 32:1–202.
Walossek , D., and K. Mü ller. 1990. Stem-lineage crustaceans from the Upper Cambrian of Sweden and
their bearing upon the position of Agnostus. Lethaia 23:409–427.
Walossek , D., and K.J. Mü ller. 1994 . Petastomid parasites from the Lower Palaeozoic of Sweden.
Transactions of the Royal Society of Edinburgh, Earth and Environmental Science 85:1–37.
Walossek , D., J.E. Repetski, and K.J. Mü ller. 1994 . An exceptionally preserved parasitic arthro-
pod, Heymonsicambria taylori n. sp. (Arthropoda incertae sedis: Pentastomida), from
Cambrian-Ordovician boundary beds of Newfoundland, Canada. Canadian Journal of Earth
Sciences 31:1664–1671.
Wheeler, W., G. Giribet, and G.D. Edgecombe. 2004 . Arthropod systematics: the comparative study
of genomic, anatomical, and paleontological information. Pages 281–295 in J. Cracraft and M.
Donoghue, editors. Assembling the tree of life. Oxford University Press, New York .
Williams, T.A. 1998. Distalless expression in crustaceans and the patterning of branched limbs.
Development Genes and Evolution 207:427–434.
Williams, T.A., and G.B. Mü ller. 1996. Limb development in a primitive crustacean, Triops longicauda-
tus: Subdivision of the early limb bud gives rise to multibranched limbs. Development Genes and
Evolution 206:161–168.
Willis, J.C., and G.U. Yule. 1922. Some statistics of evolution and geographical distribution in plants and
animals, and their significance. Nature 109:177–179.
Wingstrand, K.G. 1972. Comparative spermatology of a pentastomid, Raiallietiella hemidactyli, and
a branchiuran crustacean, Argulus foliaceous, with a discussion of pentastomid relationships.
Biologiske Skrifter af det Kongelige Danske Videnskabernes Selskab 19(4):1–72.
Yager, J. 1994 . Speleonectes gironensis, new species (Remipedia: Speleonectidae), from anchialine caves in
Cuba, with remarks on biogeography and ecology. Journal of Crustacean Biology 14:752–762.
2
EVOLUTION OF CRUSTACEAN APPENDAGES

Joachim T. Haug, Andreas Maas, Carolin Haug,


and Dieter Waloszek

Abstract
The evolutionary history of the postantennular appendages of Crustacea is reviewed, includ-
ing information on limb development early in the evolutionary lineage of this taxon. This is
particularly well demonstrated in the exceptional three-dimensionally preserved Cambrian
fossils of the “Orsten” type (~500 million years old). Crustaceans started with serially similar
limbs obtained from euarthropod ancestors, the “euarthropodium” comprising a biramous
limb with a joint membrane and a prominent rigid stem portion, the basipod , carrying two
rami, called endopod and exopod , of different phylogenetic origin. However, as a key inno-
vation, they used the anterior three appendages (the “still” food-gathering antennulae and
two more limbs) to collaborate as a set for feeding and swimming. For this collaboration,
the two postantennular limbs had special outer lateral rami, exopods with fine annulation,
and swimming setae inwardly positioned; a setiferous “proximal endite” developed medio-
proximal to the basipod. Further changes of the appendages and the effects on the feeding and
locomotory system are followed along the evolutionary lineage of the crustaceans into the
crown group, Eucrustacea. Acknowledging these changes is crucial to understand the high
degree of variation of modern crustacean limb morphology and to overcome difficulties in
recognizing their common features in terms of homology and relationships. The high plas-
ticity of crustacean limb morphology is, in fact, not surprising since the major branchings
along the crustacean lineage had already occurred far back in the Cambrian. It also means
that there is no general “crustaceopodium”; instead, each limb must be viewed individually
in terms of its history and its fate. Moreover, restructuring affected virtually all limb parts,
including the possibility of formation of epipod(ite)s/gills (not mentioned in detail herein)
and development of numerous types of surface outgrowths (spines, setae, and subordinate
structures; discussed elsewhere in this volume) for various duties. Also, the rami underwent

34
Functional Morphology and Diversity. Les Watling and Martin Thiel.
© Les Watling and Martin Thiel 2013. Published 2013 by Oxford University Press.
Evolution of Crustacean Appendages 35

significant changes regarding functional adaptations. One example of this change is that the
rami became paddle shaped and/or symmetrical (of same morphology) as an adaptation to
swimming, and sometimes multiannulated or even lost. Other strategies that evolved subse-
quently in eucrustacean ingroups include the arrangement of limbs into functional units and
consequent changes in their morphologies, and the high modification of limbs for very spe-
cific purposes, for example, reproduction/copulation. Lastly, limbs are also lost repeatedly
in various taxa. Reconstructing the evolutionary history of limbs along different crustacean
lineages is still a major task for future research.

HOW DID THE CRUSTACEAN APPENDAGES EVOLVE?

Noncrustacean euarthropod taxa such as myriapods, insects, and the arachnid chelicerates have
comparatively uniform, exclusively uniramous appendages (except for appendages 2–4) that
function mainly for walking. Moreover, they are made of one long, segmented element and are
restricted, at least in arachnids and insects, to either the head region (in arachnids called pro-
soma) or the thorax. The many thousands of species of Crustacea, however, exhibit a remark-
ably large variation regarding the morphology of their appendages, particularly the initially
biramous postantennular appendages, which are the focus of this chapter. At one extreme, crus-
tacean appendages may display high multifunctionality (Swiss Army knife effect); at the other
extreme, they or parts of them may be specialized for a particular function. To highlight the
problem of comparability, limbs that are used exclusively for walking, as in several malacostra-
can taxa, appear, at first sight, strikingly similar to those of the (mostly) terrestrial euarthropod
taxa mentioned above; that is, they are made of one long, segmented element. This superficial
similarity is misleading, however, because even this design is based on a quite different origi-
nal morphology. Yet, it is fairly easy to distinguish crustacean limbs from those of noncrusta-
cean taxa, mainly because of the morphology of their proximal parts, original constitution, and
details of joint morphology—but only when we have knowledge of the morphologies in the sur-
rounding taxa and of the historical traits.
When tracing limb morphology back in time, postantennular limbs in the ground pattern
of the crown arthropods, the Euarthropoda, were indeed all “biramous.” This means that they
consisted of a rigid, large, platelike basal structure, the “basipod,” that carried one rod-shaped
ramus mediodistally, the so-called endopod, and a paddle-shaped structure on its sloping lateral
edge, the exopod (basipod = carrier of the rami; e.g., Waloszek et al. 2005, 2007).
Consequently, uniramy within euarthropods constitutes the apomorphic state.
Development of uniramy in arachnid chelicerates is readily reproducible since postcheliceral
prosomal arachnid appendages possess the basipod as a small proximal element and an elon-
gate endopod made of several articles, possibly more than eight originally. The disappear-
ance of the exopod is evidenced by the presence of this ramus on opisthosomal limbs and on
the last walking limb (called flabellum; see Boxshall 2004) of xiphosurids. Moreover, within
arachnids, the basipod even becomes fixed to the body. This results in a situation where the
endopod-basipod joint is the limb-mover joint, and the main part for walking is the endopod:
arachnids are “endopod” walkers (and head walkers). In all uniramous myriapods, insects, and
crustaceans, the main joint remains the body-limb joint, so they are “whole-limb” walkers.
Uniramy in crustaceans has apparently developed many times convergently, such as in various
taxa of malacostracan crustaceans, for example, amphipods, isopods, and different ingroups
of decapods. In these “thoracic walkers” (rather, walking on limbs of thorax I sensu Walossek
and Mü ller 1998a), the proximal limb portion is a coxa (see below for its origin), while the
36 Functional Morphology and Diversity

(plesiomorphically retained) basipod is the second limb portion and about as small as the coxa.
This basipod then gives rise to the basically five-segmented and astonishingly uniform, elon-
gate endopod. In all cases, the original state as deducible from the related taxa is the possession
of an exopod stemming from the basipod, so there is convergence in all cases. Uniramy also
occurs in Entomostraca, such as in some ostracodes and in the predatory water fleas (onycho-
pods and haplopods). The latter do not really walk on their thoracopods, but use them for many
tasks, including grasping prey. The number of limb portions is also different from that of mala-
costracans because a coxa is lacking from the beginning. The situation is even more complex for
myriapods and insects because we cannot apply, at present, the terms coxa and basipod to the
proximal limb elements due to lack of reference structures, for example, the exopod.
The question therefore arises of how this large variety of appendage morphologies evolved
in crustaceans. Moreover, what was the original condition and morphology that crustaceans
received from their ancestors to start with? With this and a better knowledge of what the
appendages looked like in the ground pattern of Crustacea, we may understand what they were
used for, how they evolved subsequently, and what the probable driving forces were. It is, in this
context, also important to clearly homologize the different limb parts—to use an appropriate
and consistent terminology—in order to trace them from the beginning of limb formation in
arthropods to modern crustacean taxa. Clearly, the basic form was laid down much earlier, and
crustacean postantennular limbs retained (as a “historical burden”) much of the morphology
that was present even in the stem taxa of Arthropoda sensu stricto (= s. str.) and Euarthropoda,
at least initially (see below).
It seems useful, therefore, to start this review with a look at our current understanding of
postantennular limbs before Crustacea. This might help us to better understand the specific
changes in the early evolutionary lineage of crustaceans that led to the conditions developed
within the modern ingroups. This might also allow us to overcome the difficulties with mor-
phologies and terminologies people have had in the past when addressing the question of the
general morphology of crustacean appendages and their origin—particularly those attempts
made before data became available about the early to late Cambrian fossils of the Chengjiang
and “Orsten” fossil deposits (lagerstätten). Examples of such historical studies are Hansen (1925),
Størmer (1939), Heegaard (1945), Snodgrass (1958), and Kaestner (1967). Neglecting the fossil
data even after this information became available is not much better. And sometimes it appears
that there are as many different answers as there are researchers in this field—all researchers
seem to have developed their own ideas and terminologies (examples of more recent attempts:
Boxshall 2004, Williams 2004, Boxshall and Jaume 2009). Accordingly, there is still no consen-
sus in sight today.
Our review cannot cover all of the different morphologies present in Crustacea but instead
seeks to focus on some major issues, such as origin and fate of the crustacean postantennular
appendages and their major components. Also, some examples of the capabilities crustaceans
have achieved are given to demonstrate that crustaceans are the euarthropod taxon that made
the most use of its appendages—and did so very successfully.

THE “ORSTEN” PERSPECTIVE

Our overview is based on data from studies of extant larval and adult Crustacea, including our
own investigations, and particularly on the Cambrian fossil record of the Crustacea, which
is evident exclusively from Orsten crustaceans with an age of 520–495 million years. There
are multiple reasons for considering this fossil evidence. First, Orsten fossils are old, but
superbly three-dimensionally preserved, thus interpretable almost as living forms. Second,
Evolution of Crustacean Appendages 37

because of their ancient age, these fossils exhibit characters or character combinations that
may not, or should no longer, exist in extant crustaceans, therefore helping to identify early
evolutionary pathways. Third, different ontogenetic stages have been uncovered from several
of the Orsten-type fossils, particularly taxa of the crustacean lineage, permitting interpreta-
tions of the sequential morphogenesis of structures. Three-dimensional Orsten preservation
in full detail, down to even submicrometer scale and in topologically correct position, also
facilitates functional-morphological interpretations, which represent another helpful tool in
understanding evolutionary pathways. Fourth, these Cambrian crustacean fossils comprise
species that have been demonstrated to represent different evolutionary levels. Accordingly,
their data can aid in sorting features in the order of evolutionary and ontogenetic appearance.
The resulting phylogenetic tree directly ref lects the appearance of particular features on cer-
tain nodes, with some species already possessing certain features that others (still) lack. This
knowledge can be readily used for reciprocal illumination sensu Hennig (1965). Fifth, these
fossils help us to progressively establish a consistent terminology for crustacean and even
arthropod appendages, including a coloration scheme for homologous parts of limbs based
on the fossil material, which can be followed up from Arthropoda s. str. (the evolutionary
level of sclerotized arthropods; see Maas et al. 2004) into the different crustacean ingroups
(e.g., Walossek 1993; see also Walossek and Mü ller 1990, 1998a, 1998b, Maas et al. 2003, 2004,
Waloszek 2003a, Waloszek et al. 2007, Zhang et al. 2007, Stein et al. 2008, Haug et al. 2009a).
Our approach permitted, for the first time, large-scale comparability between the various
crustacean and euarthropod taxa.
This helped not only to sort in new taxa, including two-dimensionally preserved forms (e.g.,
from Chengjiang and Burgess lagerstätten) but also to identify modifications of known morphol-
ogies and to correct past errors in interpretation. This allowed for various new better-known
taxa to be added to our knowledge base and significantly improved our understanding of the
evolutionary pathways of structures and structural systems in arthropods and Crustacea. It is
particularly the evidence of structures and structural systems in the early fossil record that we
consider useful for understanding phylogeny and evolution—sometimes even more than any
other data, because the once living morphologies are the proof of existing structures and are
always superior to models of any kind. Altogether, research on fossils in Orsten-type preserva-
tion and the huge character data set provided by the other sources has challenged traditional
assumptions on the early evolution of Crustacea and has significantly stimulated this subject.
The result of this is, it is hoped, a straightforward, consistent view of the evolution and phylog-
eny of the Crustacea and their appendages.

Sources for Fossilized Appendages

In studying the early radiation of arthropods in the Cambrian (first geological period for
the fossil record of animals), we made use (mainly) of four lagerstätten: the North American
Burgess Shale, the Sirius Passet in northern Greenland, the Chinese Chengjiang biota, and the
world-wide occurring Orsten. The name Orsten has been applied to limestone nodules, which
are embedded within alum shales. They were first found in Sweden in rocks dated to the mid-
dle to upper Cambrian, but they occur worldwide (North America, Europe, Siberia/Russia,
China, and Australia) from the Terreneuvian series (former lower Cambrian; ca. 520 mya) to
the Lower Ordovician (ca. 490 mya) (Maas et al. 2006). While fossils from Burgess Shale, Sirius
Passet, and Chengjiang provide us with data from the level of Arthropoda sensu lato (= s. l.)
up to Euarthropoda (unassignable taxa, chelicerates, trilobites and allied), only the Orsten has
yielded confirmed crustacean fossils from the Cambrian (besides lobopods, chelicerates, and
agnostid euarthropods).
38 Functional Morphology and Diversity

Several Cambrian species have been claimed to be Crustacea. A few prominent examples of
so-called “crustaceans” from the Burgess Shale are Branchiocaris pretiosa Resser, 1929 (see Briggs
1976), Waptia fieldensis Walcott, 1912 (Heldt 1954), and Canadaspis perfecta Walcott, 1912 (Briggs
1978). There are further examples from the Chengjiang biota, such as Pectocaris spatiosa Hou,
1999 (Hou et al. 2004), Ercaia minuscula Chen et al., 2001 (Chen et al. 2001), and Isoxys auritus
Jiang, 1982 (Shu et al. 1995). Although all these species have been assigned to Crustacea or even
to ingroups, many if not all of them may be euarthropods or, in some cases, not even members
but stem derivatives (e.g., Dahl 1984, Maas and Waloszek 2001a, Taylor 2002, Waloszek et al.
2005, Budd 2008). Indeed, interpretations of the mentioned species as crustaceans suffered
from a rather incomplete knowledge of their morphologies, especially concerning the segmen-
tal composition of the head or the (important) proximal parts of the appendages. Likewise, the
Cambrian record of malacostracans (based on the stem euarthropod Canadaspis perfecta; see
Maas and Waloszek 2001a) and ostracodes (bradoriids and phosphatocopines misidentified as
eucrustacean ostracodes) is based simply on misinterpretations, even though it is mentioned in
modern textbooks. Until now, no unequivocal malacostracan and no ostracode had been uncov-
ered in the Cambrian (Siveter 2008).
Orsten-type preservation, by contrast, provides a rather complete view of historic (crusta-
cean) animals, because Orsten fossils are uncompressed and exhibit the finest details, such as
setae carrying minute setules, eye structures with visible facet patterns, and weakly sclerotized
membranous areas such as the arthrodial membranes of the appendages, with the smallest vis-
ible structures being only 0.2 μm (for a summary, see Maas et al. 2006). The reason for this may
be the fine impregnation of the epicuticular surface by phosphate (apatite), which conserves the
finest denticles and pores. Strangely, and for unknown reasons, the fossils do not exceed a length
of 3 mm, and the best-preserved ones are only 100–200 μm in length. Thus, a large amount of
larval and young developmental stages occurs in the material.
Given the small sizes of the preserved organisms, species from the Orsten are rarely rep-
resented by adults. If preserved, these adults are small in size, such as for the species of Skara
Mü ller, 1983 (e.g., Mü ller and Walossek 1985, Liu and Dong 2007) and Dala peilertae Mü ller,
1982 (Walossek and Mü ller 1998b), or possibly Bredocaris admirabilis Mü ller, 1983 (Mü ller and
Walossek 1988). A larger number of taxa are known only from certain early postembryonic (lar-
val) stages or sets of them (e.g., Mü ller and Walossek 1986a, Walossek and Mü ller 1989). Some
species are even represented by a larger set of ontogenetic stages that allow the reconstruc-
tion of more or less complete ontogenetic sequences. Examples are the euarthropod Agnostus
pisiformis Wahlenberg, 1818 (Mü ller and Walossek 1987) and various crustacean taxa such as
Hesslandona unisulcata Mü ller, 1982 (Maas et al. 2003; see Fig. 2.1E,F), Henningsmoenicaris scu-
tula (Walossek and Mü ller 1990) (Haug et al. 2010a), and Rehbachiella kinnekullensis Mü ller,
1983 (Walossek 1993; see Fig. 2.1I,J).
This large set of exceptionally preserved fossils also provides us with a wealth of morpho-
logical features and allows us to study the ontogeny and, particularly, the morphology and
morphogenesis of appendages in much detail. In a combined approach using data from extant
taxa, fossils, and data from ontogenies of both sources (although size limitation might cause a
slight bias), Orsten-type fossils have yielded key information for understanding early crusta-
cean evolution and might be regarded as the principal source. While Schram (1986) initially
considered them “larval f lotsam,” and Lauterbach (particularly Lauterbach 1986) tried to pro-
mote the view that Orsten fossil taxa are nothing but a bunch of stem mandibulates, intense
study of Orsten species over almost 30 years uncovered them as a set of taxa derived from very
different evolutionary levels. This differentiated view and understanding of their morpholo-
gies in the light of changes along the evolutionary lineage of Crustacea permitted the recog-
nition of the specific fate of particular appendages, their morphology, and their association
Evolution of Crustacean Appendages 39

Fig. 2.1.
Examples of “stem crustaceans” and labrophorans (Phosphatocopina + Eucrustacea) in the Orsten mate-
rial (scanning electron microscopic images and reconstructions). (A and B) The cambropachycopid
Goticaris longispinosa Walossek and Mü ller, 1990. (C and D) The “stem crustacean” Martinssonia elongata
Mü ller and Walossek, 1986. (E and F) The phosphatocopine Hesslandona unisulcata Mü ller, 1982. (G and H)
The entomostracan Yicaris dianensis Zhang et al., 2007. (I and J) The branchiopod Rehbachiella kinnekul-
lensis Mü ller, 1983. In this figure and those that follow, repository numbers and museum names are given
so that the reader can locate these specimens: BM, Natural History Museum, London; MB.A, Museum
f ü r Naturkunde, Berlin, Germany; UB, University of Bonn, Germany; YKLP, Yunnan Key Laboratory for
Paleontology, Yunnan University, Kunming, China. (A) UB 98; (C) UB 752; (E) UB 1570; (G) YKLP 10841
(Zhang et al. 2007, their fig. 1a); (H) from Zhang et al. 2007, their fig. 2; (I) UB 644. All figures used with
permission of the authors.
40 Functional Morphology and Diversity

within the series of limbs along the body. These fossils could also be used to illuminate the
early branchings of the crustacean lineage into lines with living descendants and the evolu-
tion of structural systems such as locomotory and feeding apparatus (e.g., Stein et al. 2005,
Waloszek et al. 2007).
In fact, several species from the Cambrian Orsten fossil sites could thus be assigned to the
crown group of Crustacea, Eucrustacea (Müller 1983, Müller and Walossek 1985, 1988, Walossek
1993, Zhang et al. 2007) and are even deeply nested within extant taxa. Others belong to the
Phosphatocopina, a group of small bivalved forms that could be identified as the sister group of
Eucrustacea (Phosphatocopina + Eucrustacea = Labrophora; see Maas et al. 2003, Siveter et al.
2003; see also Waloszek 2003a, 2003b, Maas and Waloszek 2005). Lastly, a set of taxa turned out
to represent derivatives of earlier branchings (see below). Until now, the Orsten is the only pale-
ontological source providing species that are derivatives of the stem lineage of Labrophora, thus
extending the view of the crustacean stem lineage further back in time phylogenetically (Walossek
and Müller 1990, Stein et al. 2005, 2008, Haug et al. 2009a, 2010a, 2010b; see Fig. 2.1A–D).

Appendage Morphologies

The Basis, 1: Appendages at the Level of Arthropoda sensu lato and sensu stricto

Before discussing the appendage morphologies of crustaceans, we need to know more about the
evolution and functional history of these limbs. This will also ensure that we understand the
respective homologies and apply appropriate terminology to these appendages. Accordingly,
we need to know about the ground pattern status of limbs before the crustacean level, that is,
the postantennular limbs in the ground pattern of Euarthropoda and, even prior to this, of the
Arthropoda s. str. and s. l. (see Maas et al. 2004, Waloszek et al. 2005). The arthropod system as
it is used throughout this chapter has been developed over a period of almost 30 years and laid
down in many of our papers. It is presented in Table 2.1 in a simplified written version, following
the notation proposed by Hennig (1965) and Ax (1995).

Table 2.1 Phylogenetic system of Arthropoda.


Arthropoda sensu lato Maas et al., 2004
(= Aiolopoda Hou and Bergström, 2006 [see Bergström et al. 2008])
Arthropoda sensu stricto Maas et al., 2004
(= Arthropoda sensu Bergström et al. 2008)
“Fuxianhuiidae” Hou and Bergström, 1997
Euarthropoda Walossek, 1999
Chelicerata Heymons, 1901
Myriapoda Latreille, 1796
Insecta Linnaeus, 1758
Crustacea sensu lato Stein et al., 2008
(= Crustacea Brünnich, 1772)
N.N. 2 = unnamed sister taxon
Labrophora Siveter, Waloszek, and Williams, 2003
Phosphatocopina K.J. Müller, 1964
Eucrustacea Kingsley, 1894 (sensu Walossek 1999)
Entomostraca O.F. Müller, 1785
Malacostraca Latreille, 1802
Evolution of Crustacean Appendages 41

A B C D E
mem mem mem

basipod endopod exopod proximal endite / coxa

Fig. 2.2.
Postantennular appendages from Arthropoda sensu stricto to Eucrustacea. mem, membrane. (A)
Arthropoda sensu stricto: Fuxianhuia protensa Hou, 1987. (B and C) Euarthropoda. (B) Leanchoilia
illecebrosa Hou, 1987 (modified after Liu et al. 2007). (C) Agnostus pisiformis Wahlenberg, 1818. (D)
Crustacea sensu lato: Martinssonia elongata Müller and Walossek, 1986. (E) Eucrustacea: Skara anu-
lata Müller, 1983. Color scheme modified from Walossek (1993): black, proximal endite/coxa; light
gray, basipod; medium gray, endopod; dark gray, exopod. All figures used with permission of the
authors.

The crustacean postantennular limbs consist of four major parts, for which we have to iden-
tify the traits in earlier morphologies. The ground pattern of the soft cuticle-bearing arthro-
pods in the broad sense (Arthropoda s. l.) forms the start, because we have some knowledge
from extant onychophorans and about 10 Cambrian so-called lobopodians (Maas et al. 2007).
Legs of such forms (and autapomorphy of the taxon Arthropoda s. l.) were tubular, possibly
ending in a distal pair of claw hooks, and as soft as the body proper. Arthropods in the strict
sense (Arthropoda s. str.) have, among other features, a much more sclerotized dorsoventrally
flattened body, each segment bearing a tergitic dorsal part connected by softer membranous
cuticle (the evolutionary process is called arthrodization). This morphology is known now
from three two-dimensionally but well-preserved, several-centimeters-long lower Cambrian
taxa: Fuxianhuia protensa Hou, 1987, Chengjiangocaris longiformis Hou and Bergström, 1991, and
Shankouia zhengei Chen et al., 2005 (Waloszek et al. 2005; see also Figs. 2.2A, 2.3A), tradition-
ally combined in the taxon Fuxianhuiidae with unclear phylogenetic status. These fossils also
helped to reconstruct the appendages in the ground pattern of Arthropoda s. str.
The main portion of their limbs, or limb stem, is tubular as before, but it is apomorphically
annulated consisting of approximately 20 sclerotic rings with membranes between. The articles
of these “arthropodia” articulate against each other via two opposing so-called pivot joints, that
is, knoblet-against-depression joints (evolutionary process: arthropodization). This construc-
tion might have guaranteed flexibility as before but also provided much more stability for walk-
ing on the limbs with the body lifted up to some degree. Yet the joint angle of bending between
the articles is limited—possibly the high number of articles counterbalanced this. Another new
limb structure is a paddle-shaped structure on all postantennular limbs arising from the outer
proximal edge of the limb stem (Fig. 2.2A, 2.3A), which is missing, however, on the first append-
age, the antennula.
The new postantennular arthropodium (Fig. 2.2A) consists, therefore, of two major ele-
ments and one significant detail: the multiannulated stem (originating from the lobopodium,
but now segmented), the pivot joints (new), and the flaplike outer ramus (new). In a way, this
is a biramy induced at this level, but it is different from the next step (the Euarthropoda level)
42 Functional Morphology and Diversity

A B

C D

E F

G H

?
I basipod
endopod
exopod
proximal endite /
coxa

Fig. 2.3.
Schematic evolution of the arthropod limb apparatus from Arthropoda sensu stricto into Eucrustacea:
ground pattern status of the limb series. (A) Arthropoda sensu stricto based on Shankouia zhenghei Chen
et al., 2005 in Waloszek et al. (2005). (B) Euarthropoda, based on different taxa; number of endopod seg-
ments and exopod morphology mainly based on Leanchoilia illecebrosa Hou, 1987 (Liu et al. 2007, used
with permission). (C) Crustacea sensu lato based on Oelandocaris oelandica Mü ller, 1983 (Stein et al. 2008).
(D) Crustacea sensu lato excluding Oelandocaris oelandica, based partly on Martinssonia elongata Mü ller and
Walossek, 1986; ? indicates that some aspects are still unclear, such as exopod morphology. (E) Labrophora
(Waloszek 2003b, Maas et al. 2003, Waloszek et al. 2007). (F) Phosphatocopina (Maas et al. 2003, Maas and
Waloszek 2005); ? indicates unclear condition of coxa-basipod border in second appendage. (G) Eucrustacea
(Maas et al. 2003, Waloszek 2003b, Waloszek et al. 2007); ? indicates that the exact ground pattern condi-
tion of maxillula is unclear. (H) Entomostraca (Waloszek 2003b). (I) Malacostraca (Waloszek 2003b). Gray
shading scheme is as described in Fig. 2.2. All figures used with permission of the authors.

because only the flap is a ramus, in the sense of subsequent evolutionary steps. The stem is
in fact the main part that may be subdivided into functional portions and not a ramus. It also
demonstrates that the two portions, stem and flap, are of different age phylogenetically and by
no means symmetrical.
It may be, as known from subdivided limb parts of living arthropods and particularly crus-
taceans, that all articles were interconnected by fine muscle strands operating as intrinsic and
extrinsic musculature, with the possible exception of the terminal article (see Waloszek et al.
2007, their fig. 4G). The flap may have served to some degree as a respiratory surface but more
likely as a locomotive aid, permitting the animal to produce some water flow around the body.
Thus, it may have initiated a swimming mode of life for arthropods in addition to their crawling
lifestyle.
It is also important to note the body parts that the early arthropodium lacks: there is, most
likely, no basal limb joint; there are only a few shorter annules; and no setae or spines (immo-
bile stronger outgrowths) developed, either on the stem or along the margin of the flap. It
Evolution of Crustacean Appendages 43

seems that the limbs were exclusively used for locomotion—and were all very similar to each
other (serially similar). The only structure possibly useful for food grasping and transport to
the mouth was the uniramous 15-segmented antennula (Waloszek et al. 2005; see Fig. 2.3A), the
first head appendages associated with the deutocerebrum. And at least in Fuxianhuia protensa
it may have carried one fine spine medially on each of the articles (Hou and Bergström 1997,
Waloszek et al. 2005).

The Basis, 2: Appendages at the Level of Euarthropoda

Various species, such as Canadaspis perfecta (Briggs 1978), have the potential of helping research-
ers document the evolutionary lineage to the next level, Euarthropoda. However, our knowledge
of these species is still too incomplete for a more conclusive discussion. While the antennula
apparently has not changed morphologically or functionally (grasping, food collecting aid) in
the ground pattern of Euarthropoda, two major events toward this level had taken place—most
likely achieved by more than just one single stem species. One event is the development of a
larger head unit, including the fact that now all postantennular limbs, though remaining serially
similar, are grouped into three cephalic and a set of trunk limbs.
The other event is the development of a significantly different postantennular limb, which
in the ground pattern of Euarthropoda possesses three major elements and another important
structure (Figs. 2.2B,C, 2.3B). All postantennular limbs are anteroposteriorly compressed, there-
fore extended in mediolateral direction and not circular in cross section like the limb stem of the
arthropodium. The first is a large subrectangular to triangular basal portion. A rodlike, distally
tapering and possibly nine-segmented structure (articles interconnected by pivot joints) arises
mediodistally from this portion. Another, paddle-shaped structure, marginally adorned with
setae, inserts on the outer sloping margin of the stem portion.
Importantly, the stem portion, as we know from several Cambrian fossil arthropods, shows
more details, demonstrating the significance of this level for the evolution of arthropod and
crustacean limbs. All rigid basal portions arise from the ventral body proper like bricks placed
abaxially on their narrower edges in a regular series on the left and right sides at some distance
between the pair, leaving a median path for food transport. Another morphological detail is that
the basal portions articulate with the body in an ample folded arthrodial or joint membrane.
Such folds are even visible in two-dimensionally preserved fossils (e.g., Leanchoilia illecebrosa
Hou, 1987; see Liu et al. 2007). This joint membrane may have facilitated movability of the limb,
while the rigidity of the basal portion may have permitted attachment of stronger muscles within
the stem in order to enhance operability of the limb. The elongation of the limb in mediolateral
direction limits movability to a largely forward-backward swing, but this was even enhanced by
a second feature: anterior and posterior sides of the limb stems are of different morphology. The
anterior edge is straight proximally, while the posterior edge is excavated. This condition may
have permitted wider backward than anterior swinging (e.g., Liu et al. 2007).
The next noteworthy feature is that the elongate narrow median edge of the stem portion
carries a row of spines, or pairs of spines, from proximal to distal pointing medially into the
interlimb space or food path. It is likely that these structures aided in food transport toward
the mouth, but it is possible that they (and others mentioned here) developed earlier, as such
structures are developed in the putative stem lineage derivate Canadaspis perfecta. It should
be noted that early in the evolution of Arthropoda and Euarthropoda, it is very difficult to dif-
ferentiate between immobile spines and spinelike setae that have a joint at their base. Fossils
are not preserved well enough to verify this. In our Orsten forms, this detail can be observed
more clearly, but it seems that because of size, many more robust structures are simply lacking
a joint, although they have a socket; thus, they should be named spines. By contrast, marginal
outgrowths of exopods used as swimming aids should continue to be named setae, although
44 Functional Morphology and Diversity

they are not necessarily mobile. Outgrowths of small entomostracan crustaceans are not eas-
ily compared with those of Malacostraca (see below). Terminological classification may not
be easily applied across larger taxa because such typification may also be phylogenetically
misleading.
We have called this basal portion of Euarthropoda basipod (e.g., Waloszek et al. 2005).
The basipod not only has a characteristic shape but also carries the two rami (see above). The
endopod arises mediodistally and in line with the spine-bearing inner edge of the basipod. It
is rodlike and roughly circular in cross section. Its portions, called podomeres, are all about
the same size and are slightly humped mediodistally, each hump bearing one or two spines or
setae, very similar to those along the median edge of the basipod. The distal smaller element
bears a tuft of one or few spines or setae. It remains unclear if there were eight, nine, or even
more endopodal podomeres originally, and it is also unclear if and how much the endopod
was proximally partly fused with the outer f lap (Liu et al. 2007, their fig. 5, Stein et al. 2008,
their fig. 7D–F).
The flap arising from the sloping outer margin of the basipod is fringed with marginal setae.
The oblique orientation of this joint permits the flap to be held more laterally, therefore facili-
tating or improving the use of this flap as a locomotory device. Subdivision of the flap into a
triangular proximal part and a distal portion might even enhance locomotory abilities (Liu et al.
2007, Stein et al. 2008, Haug et al. 2010a), though we cannot yet validate this conclusively due
to lack of more evidence. We understand this setose flap, the exopod, as homologous to the
lateral flap of the arthropodium in the ground pattern of Arthropoda s. str. It remains unclear
if the exopod was individually movable by musculature at this stage, as demonstrated in extant
paddle-shaped exopods of crustaceans (e.g., branchiopods).
The entire morphology of the “euarthropodium,” as named herein, characterizing the level
of Euarthropoda, must be understood as derived from prior morphologies. It may hence be spec-
ulated that the folds of the arthropodial membrane and the sets of spines along the inner basipod
margin are indicators of an original subdivision, possibly corresponding to annules of the limb
rod of the “arthropodium.” Consequently, the cuticle of the proximal articles may have become
softer to form the basal joint area, while a set of more distal articles should have elongated in
abaxial aspect and fused to form the brick-shaped basipod. The only intermediate situation
must have been the development of spines, possibly in the course of achievement of a feeding
function of the proximal part of the limb stem. The limbs of the Cambrian species Canadaspis
perfecta (Briggs 1978, his fig. 108) could indeed give a hint to such a pathway (Maas and Waloszek
2001a), but this has to be verified in more detail.
Continuing this line of logic, the remaining part of the original multiannulated stem of the
arthropodium may have evolved into the endopod of the euarthropodium. Again, the limb of
Canadaspis perfecta can help us understand this change, because it has spines mediodistally on all
its distal articles. A kind of functional split seems indicated, which lets us view the evolutionary
path to a complex multifunctional arthropodium, with its proximal part serving for feeding and
the distal for locomotion plus food intake. While the basipod muscles may have become some-
what concentrated, the endopod retained interior muscles running individually from podomere
to podomere, which indeed can still be observed in the endopods of extant crustaceans.
In any case, it becomes quite evident also for this euarthropodium that endopod and exopod
are very different structures in terms of origin and morphology. Likewise biramy is a problem
because the original arthropodium consisted in fact of only a stem and a single extra ramus,
while the euarthropodium has a basipod that carries two real rami (endopod and exopod). Yet,
the endopod is simply not the old stem. In fact, the stem has very likely been split into one piece
forming the membrane, one forming the basipod and one forming the endopod.
The only conservative trait in this scenario is the f lap-shaped exopod, which from its
appearance in the arthropodium did not change its shape and function in locomotion
Evolution of Crustacean Appendages 45

(swimming), although, in addition, setae appear along the margin. The exopod has therefore
almost as deep phylogenetic roots as the stem, but its first appearance can be traced only to
the level of Arthropoda s. str. at present. It may, however, be possible that it had appeared ear-
lier, but informative fossils in these surroundings are still too poorly understood (e.g., Budd
1998, Liu et al. 2007). The stem, on the other hand, can be traced down to the “lobopodium”
of Arthropoda s. l.
The complex limb morphology of the euarthropodium indicates that it had developed into
a multifunctional tool, using the whole set of appendages in a simple heterochronal movement
cycle that aided in locomotion as much as in food intake and manipulation. Moreover, the two
processes were rather neatly coupled. The basipod spines could assist in food transfer to the
mouth, while the endopod aided in locomotion and, with its spines, also aided in food manip-
ulation. The distal spines might even have permitted some scratching and sorting of food.
Though everything was (still) strictly serial, it was a huge innovation for arthropods to have
such a three-part euarthropodium comprising a basipod carrying the endopod and exopod.
Indeed, it was so successful that much of it—for example, the slope for the exopod—is still
present in certain appendages particularly of early larval stages of crustaceans (e.g., cirriped
nauplii), as well as in the Orsten crustaceans (see below). This morphology, particularly the
slope, can be used as a nice reference when searching for homologies of structures in different
taxa and for testing hypotheses about limb evolution in arthropods (see also Fig. 2.3).
The morphology must have been so effective that this tripartite euarthropodium (not
two or four parts! see Fig. 2.2B) and the strict seriality of the postantennular limbs was
transferred into the various descending evolutionary lineages. Various Paleozoic repre-
sentatives of the Euarthropoda such as trilobites and allied taxa exhibit this morphology
and seriality (though some head limbs may be smaller than the rest, as in Emeraldella brocki
Walcott, 1912 or Parapeytoia yunnanensis Hou, Bergström, and Ahlberg, 1995). Parts of this
complex limb morphology may well have developed earlier. Chelicerates are another exam-
ple taxon with extant descendants that retained this morphology during their evolution,
changing little more than the grasping antennula into a short chelicera (e.g., Chen et al.
2004, esp. their fig. 5).
The morphology of the entire apparatus suggests that it was a conjointly feeding and loco-
motion system: stopping locomotion meant stopping feeding. This may have been the reason
that this strategy was not taken over completely by all successive euarthropod lineages, at least
in those reducing the food-intake possibilities of more posterior limbs, as in the modern cheli-
cerates, myriapods, and insects.
In a way, Crustacea are the ones among the taxa with living representatives that retained
most of the morphology of the euarthropodium in their ground pattern—also Agnostus pisi-
formis Wahlenber, 1818, a minute euarthropod (Mü ller and Walossek 1987) with supposed affini-
ties to crustaceans (Walossek and Mü ller 1990, Stein et al. 2005, Haug et al. 2010a). And they
made the most use of it. Initially, however, much of the seriality was retained.

Appendages at the Level of Crustacea sensu lato

The appendages of crustaceans indeed exhibit such various plesiomorphic traits. This is most
evident in several Orsten fossil crustaceans that we refer to as stem derivatives—more precisely,
derivatives of the evolutionary lineage toward the Labrophora, comprising the Phosphatocopina
and the Eucrustacea (e.g., Waloszek et al. 2007). Examples of plesiomorphies are the seriality
and the gross morphology of the postantennular limbs, with the ample joint, the basipod (pos-
teriorly more excavated) with median spines on a rather straight median edge, the mediodistal
endopod with sets of spines, and the paddle-shaped exopod with marginal setae on the sloping
edge of the basipod (Fig. 2.4A).
46 Functional Morphology and Diversity

However, this holds true only for the more posterior limbs because all three anterior append-
ages, that is, the antennulae and the following two limbs, are more specialized in the ground
pattern of Crustacea s. l. than are the corresponding appendages in the euarthropod ground pat-
tern and therefore are more like those in modern crustaceans. Nevertheless, these fossil species
combined as stem derivatives are not stem species themselves but exhibit also autapomorphies
indicative of their own evolutionary lineage (the reason that we usually speak of stem-lineage
derivatives). The following interpretations are therefore based not on single species or speci-
mens but on more than half a dozen known species. Moreover, they apparently represent differ-
ent evolutionary levels, and several of them are preserved as a set of instar stages.
Based on this evidence, we have identified several evolutionary novelties of Crustacea in the
wider sense. Many if not most affected the morphology particularly of the appendages and are
coupled with the locomotory and feeding apparatus, apparently an important issue for this euar-
thropod group. These are, for example,

• the deviation from a strict seriality of all postantennular body limbs (state of Arthropoda
s. str. and Euarthropoda; Fig. 2.3A,B);
• changing tagmosis of the head-trunk system, for example, by incorporation of more trunk
segments into the head, and specializations of limbs;
• changes in the locomotory and feeding system, again with differentiated development in
the head and the trunk, likewise independent of the changing tagmosis;
• the development of associated features with the changes in the functional system; and
• changes in the morphology of the appendages, individually as well as in sets (further
development of tagmata).

The first significant changes of the appendages and their morphologies along the crustacean
evolutionary lineage based on the stem derivatives are interpreted as autapomorphies in the
ground pattern of Crustacea s. l.:

1. Antennulae uniramous, composed of a few tubular portions or articles, though large,


limblike, and not feelerlike, but involved in both feeding and locomotion; several long
setae along the posterior side of the appendage (originally the median side) and on the tip
(see, e.g., Haug et al. 2009a for Cambropachycopidae or Stein et al. 2008 for Oelandocaris
oelandica Mü ller, 1983).

Comments: Plesiomorphies associated with the antennula include the insertion anterola-
terally at the hypostome, and the function of the antennula possibly only for food gathering
(a similar specialization has also been found in Agnostus pisiformis, which, by contrast, has
a 15-segmented antennula with short spines). A sensorial antennula seems to have evolved
convergently in several evolutionary lineages of the Euarthropoda, for example, in mala-
costracan crustaceans and atelocerate taxa (myriapods and insects). In the evolutionary
lineage of chelicerates, the antennula changed into a highly shortened grasping element,
the chelicera (Chen et al. 2004).

2. Anterior three appendages morphologically different from each other and from the more
posterior ones.

Comments: In the ground pattern of Arthropoda s. str., only the antennula differs from
the serial biramous trunk limbs in being uniramous and the major food-raking element
Evolution of Crustacean Appendages 47

Fig. 2.4.
Scanning electron microscopic images of different postantennular limbs of Orsten fossil crustaceans.
(A) Isolated trunk limb of Henningsmoenicaris scutula Walossek and Mü ller, 1990, stage 10 with paddle-shaped
exopod. No proximal endite is present at this ontogenetic stage. (B) Third appendage of Hen. scutula, stage
8, multiannulated exopod with few articles, distal part of endopod broken off. The proximal endite is still
small at this evolutionary level. (C) Second appendage of Goticaris longispinosa Walossek and Mü ller, 1990,
stage 2 with multiannulated exopod. (D) Proximal endite of a trunk limb of a very late ontogenetic stage
of Hen. scutula. At this ontogenetic stage, trunk limbs also possess proximal endites. Scale bar, 10 μm. (E)
Postmandibular appendage of the phosphatocopine Hesslandona unisulcata Mü ller, 1982, growth stage III.
Note the large proximal endite, the enditic protrusions on basipod and endopod drawn out medially, and
the exopod being paddle-shaped proximally and multiannulated distally. (F) Mandible of Hes. unisulcata
with multiannulated exopod, short endopod, small basipod, and large coxa. (G) Cephalic appendages of
Martinssonia elongata Mü ller and Walossek, 1986. Multiannulation of exopod is only weakly developed. (H)
Postmandibular appendage of growth stage V of Hes. unisulcata. The exopod is paddle-shaped proximally
and multiannulated distally as in Fig. 2.4E, but the multiannulated part is smaller in this specimen due to
later ontogenetic stage. (I) Subequal (second) antenna (right) and mandible (left) of Skara anulata Mü ller,
1983, with multiannulated exopods. Abbreviations: bas, basipod; cox, coxa; en, endopod; ex, exopod; pe,
proximal endite. Repository numbers (see Fig. 2.1 for abbreviations): (A) UB W 338; (B) UB W 335; (C) UB
W 124; (D) UB 103; (E) UB 103; (F) UB W 106; (G) UB 780; (H) UB W 156; (I) UB 692.
48 Functional Morphology and Diversity

(Fig. 2.3A). Also in the euarthropods, all postantennal limbs are serial. The second append-
age is slightly shifted anteriorly in its position and may be smaller than the more posterior
limbs but still resembles them (Fig. 2.3B). Here in crustaceans, all three anterior append-
ages move in accord and are somewhat set off from the posterior ones with their hetero-
chronous beat (Fig. 2.3C). This points to a new type of combined locomotion and feeding
in Crustacea s. l., the so-called “sweep-net feeding” (Waloszek 2003b).

3. Exopods of the first and second postantennular limbs comprising a tubular proximal part
and a multiannulated distal part with each annulus carrying a long seta on the median
side, which is directed mediodistally against the endopod (Fig. 2.4C); the proximal socket
element may have served to raise the setae-bearing part slightly away from basipod and
endopod in order to facilitate the swing of the setae.

Comments: Plesiomorphically, the exopod is a simple flap with setae around its entire free
margin (compare Fig. 2.3B,C). In the putatively closest relative to Crustacea s. l., Agnostus
pisiformis (Stein et al. 2005), the exopods of the first and second postantennular limbs are
also multiannulated, but bear setae on their lateral = outer side (Mü ller and Walossek 1987,
their plates 18:1–5, 19:1). These two limbs are generally similar to each other, except that
because of the different position regarding the mouth, the basipod and its median enditic
armature were most likely already different.

However, the most apparent new feature of Crustacea s. l. is the following:

4. Possession of a small setiferous sclerotized humped area, the “proximal endite,” within the
body-basipod membrane of the postantennular limbs and clearly proximomedially of the
basipod (Fig. 2.2D).

Comments: This endite was termed as such by Walossek and Mü ller (1990; see also
Walossek 1993) because of its morphological similarity to median basipodal endites of
extant entomostracan eucrustaceans. Much earlier, Calman (1909, 51) had pointed to its
existence in branchiopods and remarked upon it being most likely a very old feature. In
contrast to Walossek and Mü ller (1990), who assumed its presence on all postantennular
limbs originally, recent investigations of the stem derivatives demonstrated its first appear-
ance only on the third appendage in young larvae, that appendage later called mandible
in Labrophora (e.g., Stein et al. 2005, 2008, Waloszek et al. 2007, Haug et al. 2009a, 2010a,
2010b; Fig. 2.4B).

The proximal endite is regarded as one of the key evolutionary characters of Crustacea
(Walossek and Mü ller 1990, Walossek 1993, Waloszek 2003a, 2003b), possibly appearing only
on the third appendage at first (Stein et al. 2005, 2008, Waloszek et al. 2007; but see Haug et al.
2010a). Further along the evolutionary lineage to the Eucrustacea, the proximal endite occurs
on all postantennular appendages, having become larger and more setiferous (Figs. 2.3D, 2.4D).
Its possible main purpose was food manipulation and transport, as seen still today, for example,
in cephalocarids and mystacocarids (Fig. 2.5G,H).
It is still unclear if the proximal endite was already movable at this early stage, as described,
for example, for the proximomedial endite (and the more distal ones) of Cephalocarida (see
Sanders 1963, Hessler 1964, his fig. 11). It also remains unclear if it served as a food-raking device
from the very beginning. Its functionality as an aid in food transport is, however, clear from
its later shape, position, and use in living taxa, indicative of its continuing significance in the
Evolution of Crustacean Appendages 49

Fig. 2.5.
Scanning electron microscopic images of proximal and basipodal endites and appendage armatures.
(A) Mandible of an undetermined phosphatocopine. Anterior and posterior sets of setae are not yet
differentiated from each other. Numbers indicate different sets of setae: 1, anterior retention setae;
2, median set, usually developed as stronger spines; 3, posterior sieving setae. (B) Trunk limb of
Rehbachiella kinnekullensis Mü ller, 1983; numbers are as in A. While the retention setae (1) are arranged
in parallel rows, the sieving setae (3) form more or less a triangle. (C) Trunk appendage of Yicaris dian-
ensis Zhang et al., 2007 with about the same arrangement of setal sets as R. kinnekullensis; numbers are
as in A. (D) Trunk limbs of fossil R. kinnekullensis . (E) Trunk limbs of extant Cyclestheria hislopi (Baird,
1859) (Olesen 2007). Enditic arrangement very similar to that of R. kinnekullensis (see panel D). (F)
Anterior head region of the phosphatocopine Hesslandona unisulcata Mü ller, 1982. Note the enlarged
coxa of the mandible as an extreme variation of the proximal endite. (G) Postmaxillulary appendage
of the cephalocarid Lightiella monniotae Cals and Delamare-Deboutteville, 1970. (H) Maxillula of the
mystacocarid Derocheilocaris remanei Delamare-Deboutteville and Chappuis, 1951. Abbreviations: cox,
coxa; end, endite; pe, proximal endite; ste, sternum. Repository numbers (see Fig. 2.1 for abbreviations):
(A) UB W 255; (B) UB W 380; (C) Spec. 5/YKLP 10846; (D) UB W 86; (F) UB W 381. All figures used
with permission of the authors.
50 Functional Morphology and Diversity

food-gathering apparatus of head and thorax. Its absence in other euarthropods, including
Agnostus pisiformis (Mü ller and Walossek 1987, Waloszek et al. 2007), is interpreted as primary
lack, while other general appendage features are shared between A. pisiformis and Crustacea s. l.,
that is, multiannulated exopods on appendages 2 and 3, suggestive of closer alliance.
One less obvious feature of Crustacea s. l. concerns the number of endopod podomeres. In
the ground pattern of euarthropods, there may have been 9, possibly up to 11. Not least regarding
its armament with median spines, this may point to the importance of the endopod in both food
gathering and locomotion (possibly walking). The exclusively Paleozoic trilobites had con-
stantly seven endopod podomeres, as also Agnostus pisiformis (Mü ller and Walossek 1987, also
for data on trilobite limbs). In crustaceans the maximum number observed is six (Zhang et al.
2007 for Yicaris dianensis), five or six in Cephalocarida depending on the authors (e.g., Jones
1961), or always five in Malacostraca (specializations not counted). The maximum observed in
Orsten stem derivatives was originally thought to be five, but maybe six or seven (Haug et al.
2010a). Most fossil and extant taxa have fewer endopod podomeres, at times four or even as few
as one portion (which may not be a simple single podomere, however). This may point to a rather
early initiation of different uses of endopods, likely in accordance with differences in lifestyles
of their carriers. Walkers among malacostracans, for example, have five podomeres, but these
are tubular and may have enormous lengths so that in the extreme a limb can be more than a
meter long (for a size comparison of large arthropods, see Rudkin et al. 2003, their fig. 5). We
cannot, however, follow up this interesting question in more detail at present.
Seriality of limbs is also retained in crustaceans principally in all appendages behind the
third limb (Fig. 2.3C,D, level with more proximal endites), but these limbs share the proximal
endite with the preceding two limbs, and they differ from those in that they have paddle-shaped
exopods (Fig. 2.4A). Although partly retained from the euarthropod ground pattern, this indi-
cates the tagmotic break and the different use of anterior and posterior limbs.
We must admit that reconstructing the ground pattern status of exopods at this level remains
problematic because a multiannulated state (although with inwardly pointing setae, a status not
known from other arthropods) is known from a number of Orsten stem derivatives, such as
the cambropachycopid Goticaris longispinosa Walossek and Mü ller, 1990 (Haug et al. 2009a;
Fig. 2.1A,B) and Martinssonia elongata Mü ller and Walossek, 1986 (Fig. 2.1C,D) (unclear for
Cambropachycope clarksoni Walossek and Mü ller, 1990). In M. elongata, the annulation of its
exopods is only weakly developed (Mü ller and Walossek 1986b, their fig. 2.4G; see also Haug
et al. 2010b). As mentioned earlier, a triangular plate may be set off from the exopods basally,
as in Oelandocaris oelandica and Henningsmoenicaris scutula. In both, the distal part forms the
main setae-bearing paddle (Stein et al. 2005, 2008, Haug et al. 2010a). This morphology is
remarkably similar to what has also been described for a derivative of the stem lineage toward
Euchelicerata, Leanchoilia illecebrosa from the Terreneuvian Chengjiang fauna of China (Liu
et al. 2007). Whether these two exopod articles are homologous and how they relate to sub-
divisions developed in other Cambrian euarthropods such as trilobites and naraoiids remain
unclear at present.
Two noteworthy plesiomorphies have been taken over into the crustacean ground pattern.
One is the retention of a fairly rigid basipod with a straight inner rim carrying robust medially
pointing spines. This feature is recognizable in two of the Orsten stem derivatives, Oelandocaris
oelandica and Henningsmoenicaris scutula (Fig. 2.4A,B). All others have shortened this edge
in proximodistal aspect but elongated it in mediolateral aspect to a more humplike or endite-
like shape with a central major spine and some flanking spines or setae (e.g., in Goticaris long-
ispinosa; Fig. 2.4C). It seems that this feature may be useful to discriminate within the set of early
crustacean taxa and to resolve the early evolutionary lineage of Crustacea in more detail. The
other is an often-neglected plesiomorphy retained from Euarthropoda, namely, that the head
comprises only four appendage-bearing segments (Maas et al. 2003, Waloszek 2003b)—as in
Evolution of Crustacean Appendages 51

Agnostus pisiformis. When exactly this change to a larger head unit occurred, and why, remains
unclear because several of our stem derivatives exhibit this condition. Other taxa have five
appendage-bearing head segments (H. scutula and O. oelandica). This cannot be evaluated any
further at present, but other characters indicate that the longer head may have been achieved
in parallel to the evolutionary development in the remaining crustacean taxa—as much as the
number six has been achieved various times in parallel in the different eucrustacean ingroups.
The shield cannot be the reason because H. scutula has a large shield and O. oelandica a smaller
one. There is also no change in the tagmotic pattern on the ventral side.
In summary, crustaceans have retained much of the limb morphology of the euarthropo-
dium, but evolved (1) a new type of exopod that is multiannulated and bears setae along the
inner side and (2) a fourth limb element, the proximal endite, that has no counterpart in any
other arthropod taxon. Up to the next evolutionary level, the described condition remains sta-
ble, except for the occurrence of more proximal endites along the limb series. This also implies
that the fourth and fifth appendages resemble trunk appendages—regardless of whether they
are already included into the head.

Appendages at the Level of Labrophora

Major novelties along the crustacean lineage characterize the evolutionary level of Labrophora
(Maas et al. 2003, Siveter et al. 2003). As before, most if not all of them are allied with significant
changes in the locomotory and feeding apparatus, more specifically the cephalic one (Waloszek
et al. 2007). However, only few changes affected the limb morphology. Examples of innovations
(autapomorphies) not associated with the appendages include the following:

• The appearance of an enormous fleshy “labrum” with slime glands and chemoreceptors on
the posterior side of the hypostome (note that labrum and hypostome are both present and
not synonymous structures!)
• The occurrence of a “sternum” as a fusion product of the sternites of the mandibular and
first postmandibular segments (no sternite belonging to the second appendage visible)
• A pair of humps on the mandibular sternite, the “paragnaths”
• Fine hairs or denticles occurring in rows on the labral flanks and like a carpet on the
sternum and paragnaths (for details, see, e.g., in Maas et al. 2003, Waloszek et al. 2007).

It is very likely that this, again, did not happen in a single stem species, but up to now we have not
discovered more fossil species to be able to split up this set more precisely.
Regarding appendage morphology, labrophoran autapomorphies occur, as far as we could
detect them in our material, only on the anterior two postantennular limbs but in two impor-
tant ways:

• The proximal endite of the second appendage, from this level on called antenna, is enlarged to a
ring-shaped sclerotic structure on the proximal end of the limb, long ago named coxa; its original
enditic surface with setae is prolonged and points, due to the special position of the antenna
laterally, at the labrum posteromedially around the labral flanks toward the mouth; the coxa now
carries the basipod with its median enditic hump and armature, which itself carries the rami.
• The proximal endite of the third appendages, now called “mandible,” is also enlarged to a
ring-shaped sclerotic structure on the proximal end of the limb; also here this coxa is medi-
ally prolonged but points medially because of the position of the mandible lateral to the
paragnaths; different from the antenna, the surface of the coxal endite is slightly flattened
and tilted against the mouth, a morphology often called gnathobase (Figs. 2.3E, 2.5F; see
Fig. 2.6F for an extreme variation).
52 Functional Morphology and Diversity

Fig. 2.6.
Scanning electron microscopic images and schematic drawings of malacostracan and entomostracan trunk
limbs. (A) Pereopod of euphausiid malacostracan, not to scale. In contrast to Entomostraca, the thoracic
limbs of Malacostraca do not have endites. (B–D) Postmaxillulary appendages of different entomostra-
can species. (B) Lightiella monniotae Cals and Delamare-Deboutteville, 1970 (Cephalocarida). Note the
two-parted exopod. (C) Yicaris dianensis Zhang et al., 2007. The proximal part of the basipod is concealed.
(D) Dala peilertae Mü ller, 1982. Endites are bent toward the reader. (E) Schematic cross section of thorax
I (sensu Walossek and Mü ller 1998a) at the level of the first thoracopod of the malacostracan Speonebalia
cannoni Bowman, Yager, and Iliffe, 1985, modified after Bowman et al. (1985, their fig. 2a). Each sternite
has a median keel for food manipulation (Walossek 1993). (F and G) Cross sections at the level of post-
maxillulary appendages of two entomostracan species. Note the large number of endites and the proximal
endite, all used for feeding (Walossek 1993). (F) Y. dianensis, the only entomostracan with the unequivocal
(high) number of six endopod articles. (G) The branchiopod Rehbachiella kinnekullensis Mü ller, 1983 with
the ventral food groove, autapomorph for Branchiopoda and used for feeding (Walossek 1993). Gray shad-
ing scheme is as described in Fig. 2.2. Abbreviations: bas, basipod; cox, coxa; en, endopod; ex, exopod; pe,
proximal endite. Repository numbers (see Fig. 2.1 for abbreviations): (C) YKLP 10859 (Zhang et al. 2007,
their fig. 1i–k); (D) UB W 310. All figures used with permission of the authors.

Often the coxal endite is called a gnathite and is even thought to serve for grinding or cutting.
At this early evolutionary stage, nothing of that kind is even remotely present. Equipped with
fine, setalike outgrowths distally, it is not capable of biting. This may even be a major misunder-
standing of arthropods in general when applying functional terms from other animals, such as
jawed vertebrates. An example is the molar surface of gammarid amphipods, which is made of
Evolution of Crustacean Appendages 53

a large number of extremely soft, densely packed outgrowths. The molars act, for example, to
destroy the cells of diatoms or plants (Watling 1993), and they improve the ability to hold food
(Mayer et al. 2009, their fig. 3E,F; G. Mayer, personal communication). Plesiomorphically, the
antenna and mandible are very similar to each other (see Fig. 2.4I) (except for the tilted endite
surface of the mandibular coxa), and the spines and setae-bearing enditic protrusions of coxae
and basipods served to stuff food into the mouth (Labrophora are in a way “Di-Mandibulata,” as
Waloszek 2003b pointed out). Also, the distal elements are little different from the state before;
both limbs retain the multiannulated exopod for sweeping. This situation has even been retained
in all those eucrustacean taxa with a larval development that includes feeding nauplii/metan-
auplii (e.g., cephalocarids, mystacocarids, copepods, branchiopods). Slight differences between
the two appendages refer to the different positions and orientation along the body axis regard-
ing the mouth opening. While the posteromedially oriented antenna is inserting laterally to the
mouth, the mandible inserts posterolaterally, being orientated medially to anteromedially. Also,
the more posterior appendages appear to have remained largely unaltered in their gross morphol-
ogy; that is, they are serial and composed of four elements: proximal endite, basipod, endopod,
and (here) paddle-shaped exopod.
Even so, all postantennular limbs show another evolutionary novelty, clearly recognizable
in our material:

• A special arrangement of the spines and setae on the median enditic protrusions of the
proximal endite, the basipod, and likewise, all endopodal podomeres. This armature is
clearly different from that developed in all known stem derivatives and also all available
outgroups. It comprises basically one central spine and a set (crescentic row) of setae flank-
ing the spine anteriorly and posteriorly (e.g., Walossek 1993, Maas et al. 2003, Waloszek
2003b; Fig. 2.5A–C; for more details, see below).

Remarkably, at this level of Labrophora, all median structures of the appendages are rather similar
to each other, that is, the proximal endite, the basipod protrusion, and the mediodistal humps on
the endopodal podomeres. Despite these similarities, the proximal endite is often easily detect-
able. It is usually larger and better armed than the other enditic protrusions, and it is slightly ante-
riorly tilted (e.g., Walossek 1993, his plate 2–2). This tilting of the proximal endite against the axis
of the limbs can be almost 90°, as in triopsid branchiopods (Fryer 1988, his fig. 102). The enditic
origin remains visible even if the proximal endite is modified as in the mandibular coxal median
prolongation (e.g., in Figs. 2.4F, 2.5A), the so-called gnathobase—best seen in phosphatocopine
larval stages (see Maas et al. 2003, their fig. 65A for the antenna and fig. 66A for the mandible).
Another example of extreme modification is that of the maxillulae (or first maxillae) and maxil-
lae (or second maxillae) of eubranchiopods, which are nothing but the retained proximal endites
of the otherwise strongly reduced limbs (e.g., Martin and Cash-Clark 1995, their figs. 6, 11A–C,
Olesen et al. 2003, their fig. 9E).
The entire feeding apparatus of Labrophora may be regarded as a posteriorly open
food-path system, with the anterior appendages as much engaged in feeding and locomotion
as the posterior set that brought in food from the posterior along the interlimb space. The
more specialized appendages were located in the vicinity of the mouth. Here food could be
checked (chemosensed at rear of labrum; Waloszek 2003b, his fig. 3B), sorted, and directed
into the mouth by means of the strong enditic spines of antenna and mandible. As before,
the posterior feeding system works while moving, but the anterior system was decoupled
and could operate much more individually. During rotation of the limbs around their basal
joints, the limbs produced moving currents by their exopods as much as inwardly directed
food flow (even more so by the antennae and mandibles). From this, nutrients were raked off
54 Functional Morphology and Diversity

and shoveled against the mouth. It is important to note that the basipod, at this level, possesses
only one major enditic protrusion and that the median edge was not subdivided into soft lobate
endites.
Within the Labrophora, phosphatocopines deviate from this in having significantly modi-
fied their postantennular limbs in at least two aspects (Fig. 2.3F):

• All endopods are basically three segmented (Maas et al. 2003; endopod of antennae and
mandibles may be only two segmented in ingroups); each article of the otherwise serial
postmandibular limbs is drawn out into long endites (Fig. 2.4E,H) (yet with the triplet of
setae as described above).
• Within the taxon, the endopods may be further modified from the ground pattern for spe-
cial feeding purposes as observed from isolated appendages from the middle Cambrian of
Australia. Here the endopod is involved in the formation of a kind of gnathic edge together
with the coxa and the basipod, comprising more or less a “whole-limb jaw” (Walossek et al.
1993, their fig. 3C).
• The mandibular basipod becomes progressively shorter and smaller during ontogeny until
it almost disappears, which gives the superficial impression of a coxa carrying the rami.

The fate of the basipod was documented at least for Vestrogothia spinata Mü ller, 1964 (Maas et al.
2003, their fig. 59A,C and plate 45A,B, Maas and Waloszek 2005, their fig. 2A,B). Eventually, the
basipod can no longer be recognized as a larger subtriangular unit carrying the two rami and is
nothing more than a separate setae-bearing endite below the proximal endopodal endite, while
the exopod seems to stem directly from the coxa (Fig. 2.4F; see Maas et al. 2003). This final mor-
phology is quite misleading because the basipodal endite seems very much like an endopodal
article and gives the appearance of three articles altogether, as in the endopods of all posterior
limbs and in the ground pattern of Phosphatocopina. The developmental sequence can be rec-
ognized only in intermediate (= younger) stages that possess a narrow ring around the exopo-
dal basis with connection to the median protrusion (Maas et al. 2003, their fig. 59C). Thus, it
becomes clear that coxa and basipod do not really fuse.
A number of phosphatocopine species also have multiannulated exopods on the postmandib-
ular appendages, for example, Vestrogothia spinata, Falites fala Mü ller, 1964, and Hesslandona
neocopina Mü ller, 1964 (Maas et al. 2003). In Phosphatocopina, the exopod morphology is
very special, because there are species that change during ontogeny from one morphology,
that is, multiannulated exopods on the posterior appendage series with inner setation, to the
other morphology, that is, paddle-shaped exopods with marginal setation (Maas et al. 2003 for
Hesslandona unisulcata).
Phosphatocopina are also a good model for the understanding of the ontogenetic change
of the proximal endite into a huge coxa with its median gnathic edge. Also, the armature, still
present in the earliest stages, is a nice reference for the recognition of the transition (see Fig.
2.5A). Lastly, Phosphatocopina also demonstrate the retention of the large excavation of the
proximal margin of a limb (Fig. 2.4F here referring to the coxal posterior side), as initiated in the
euarthropodium.

Appendages of the Eucrustacea

The ground pattern of Eucrustacea includes the four-element postantennular limb, with coxae
on the antenna and mandible as well as proximal endites on all posterior limbs (Figs. 2.2E,
2.3G). Some doubt remains regarding the development of the proximal endite into the coxa in
the antenna and mandible. Only a few taxa exhibit the morphogenetic change of a proximal
Evolution of Crustacean Appendages 55

endite to a coxa on the mandible in very early larval stages, such as the Cambrian Rehbachiella
kinnekullensis, while the antenna clearly has a coxa from the beginning (Walossek 1993). It is
also difficult to determine the ground pattern of phosphatocopines, though it could be demon-
strated for one species from the lower Cambrian, Klausmuelleria salopensis Siveter et al., 2003
(Siveter et al. 2003, their text fig. 4). At present, we therefore favor the hypothesis that the mor-
phogenetic switch in the antenna occurred in the labrophoran ground pattern.
The only notable change concerning limb morphology in the ground pattern of Eucrustacea
seems to have affected the third postantennular or fourth cephalic limb, which from this evolu-
tionary level on is termed maxillula and may represent an autapomorphy of Eucrustacea:

• First postmandibular limb is dissimilar in morphology to the preceding and succeeding


limbs, functioning as a feeding aid.

The word seems is used to express our cautiousness, because the morphology of the maxillula
remains unclear. This is because (a) its shape differs between the two sister taxa Malacostraca
and Entomostraca (see below; examples given in Walossek and Mü ller 1998a, their fig. 12.11),
and (b) the maxillulary shape of these taxa differs significantly from that of the fourth limb of
phosphatocopines and that of “stem crustaceans.” Therefore, we cannot reconstruct a ground
pattern state for the morphology of the maxillula. We can only construct a ground pattern
state for its function because function is generally the same in all ingroup eucrustacean taxa.
What seems clear is only that the maxillula was already reduced in size compared to that of the
more posterior limbs. This situation can be seen in all known Cambrian Orsten taxa referred
to Eucrustacea, for example, Bredocaris admirabilis (Mü ller and Walossek 1988), Rehbachiella
kinnekullensis (Walossek 1993), and Yicaris dianensis (Zhang et al. 2007). (The same seems to be
the case in Dala peilertae and Walossekia quinquespinosa Mü ller, 1983, but detailed descriptions
of these are still under way.)
We also have difficulties reconstructing the morphology of the fourth postantennular or
fifth cephalic limb, traditionally called the maxilla. We can only clearly state that it was not a
“mouthpart”—a still repeated misunderstanding of crustacean morphology. It was included in
the head, and it was acting in accord with the trunk limbs, a plesiomorphic trait. So this holds
only for its feeding and locomotory duties in the ground pattern of Eucrustacea. Clearly, the
maxilla at that level did not have the morphology of previous levels, as we learned for other limbs
and for earlier nodes. And it is also not a mixture of the morphology of ingroup maxillae, but the
maxillae have adopted the specific trunk-limb morphology of one of the two sister taxa within
Eucrustacea. As discussed below, these limbs are also very differently developed, in terms of
specific changes in the locomotory and feeding apparatus and tagmotic changes, for example, in
the Malacostraca. For both maxilla and the trunk limbs, we cannot state that the malacostracan
morphology should be plesiomorphic, but we cannot state this for the entomostracan morphol-
ogy either. Likewise, it is difficult to argue that entomostracans are a paraphylum “along the evo-
lutionary lineage toward Malacostraca” because this would intimately violate the data around
limb morphology.
The retention of the maxilla within the posterior limb system appears plesiomorphic, while
the specific similarity in shape, flatness, and softness of the maxillula and maxilla in malacostra-
cans only adds to the large number of other autapomorphies validating this monophylum. The
condition that the maxilla has a trunk-limb morphology is developed in several Cambrian fos-
sil eucrustacean taxa such as Bredocaris admirabilis, Dala peilertae, Rehbachiella kinnekullensis,
Walossekia quinquespinosa, and Yicaris dianensis, and in extant cephalocarids (e.g., Sanders 1957,
1963). It may hence be likewise interpreted as a specific feature of Entomostraca and not as an
ancient trait.
56 Functional Morphology and Diversity

All more posterior appendages remain basically unaltered in their gross morphology. They
are serially homonomous and composed of the proximal endite, the basipod, the endopod, and
the paddle-shaped exopod (= euarthropodium + proximal endite). It appears likely that epipods
on these limbs belong to the ground pattern of Eucrustacea, but their number remains uncer-
tain, being three in Entomostraca and no more than two in Malacostraca. Their exact origin
also remains unclear at the moment (Maas et al. 2009). (For a discussion of the second autapo-
morphy, the ontogeny via a short-segmented [ortho]nauplius, see Maas et al. [2003].)
Lastly, it must be stressed that coxa and proximal endite exclude each other because a coxa,
as far as we can state at present, originates from this endite below the basipod. The joint between
coxa and basipod originates from the fact that the proximal endite lies embedded within the
proximal limb joint membrane. When the coxa develops, the distal portion of the membrane is
simply retained between the final sclerotic stem portions.

Different Evolution of Limb Morphologies in Entomostraca and Malacostraca

The appendages remain more or less unchanged in the ground pattern of Eucrustacea, but
substantial further evolutionary modifications of them occurred in the evolutionary lineages
of the two sister taxa. This is fully understandable since all branchings mentioned so far, and
possibly even a few more already on the lines of the two sister taxa, had already occurred in the
Cambrian. The others had, accordingly, another 500 million years to make changes. Significant
differences in appendage morphology already in the ground patterns of Entomostraca
(Fig. 2.3H) and Malacostraca (Fig. 2.3I) indicate that, as in the maxillulae and maxillae, it is often
difficult to reconstruct well-founded ground pattern states, because the differences between the
taxa to be compared are not traceable down to a common ancestral morphology.
Up to now, only the evolution of Entomostraca has been documented (as we think) by
well-preserved Cambrian fossils in Orsten-type preservation, for example, Yicaris dianen-
sis (Zhang et al. 2007, Fig. 2.1G,H), Bredocaris admirabilis (Mü ller and Walossek 1988), and
Rehbachiella kinnekullensis (Walossek 1993; see Fig. 2.5D,E for comparison with an extant bran-
chiopod). The early phase of malacostracan evolution, in contrast, is still completely unknown.
The earliest reliable malacostracan fossils occur in the Lower Ordovician, and even the status
of these so-called phyllocarids must be considered uncertain at best: living representatives of
phyllocarids, having leaf-shaped limbs in the anterior thoracic portion (thorax I after Walossek
and Mü ller 1998a), are indeed very different from the well-sclerotized large fossil forms possibly
having rather stenopodial limbs.
This bias may have led to the impression of a “well-defined” taxon Malacostraca
(admittedly the two-section thorax region present in all living taxa is a striking feature),
while entomostracans seem to express—another misunderstanding—“morphological insta-
bility as an expression of plesiomorphy,” because their ingroup taxa exhibit very different
morphologies. In fact, this is exactly their “trick.” The high number of autapomorphies of the
Malacostraca in its ground pattern means nothing but the result of a long, though unknown,
evolutionary lineage in which these characters have been accumulated. These are assigned
now to the last common ancestor of all descendants with living representatives. The morpho-
logical conservatism of Malacostraca may then just be the retention of plesiomorphies, not
expression of evolutionary success.
The bias also results, in our opinion, in the superficial view that Entomostraca should have
retained more plesiomorphic traits than the malacostracans (the often-used terms lower and
higher crustaceans reflect this interpretation) and hence should not be monophyletic. Indeed,
some body parts retained plesiomorphic traits, such as the anterior appendages. Antennular and
antennal morphologies of malacostracans are easily identified as their evolutionary novelties,
Evolution of Crustacean Appendages 57

while Entomostraca apparently retained a plesiomorphic appearance of these two append-


ages. However, all further posterior appendages underwent significant evolutionary changes in
both lineages, and in both, the morphology of these limbs deviates from the ground pattern of
Eucrustacea or Labrophora. This can even be stated for the mandibles, which lose their basipod
plus the rami during the ontogeny in entomostracans, and for the maxillulae and maxillae as
well. Accordingly, many features in the stem species of Entomostraca and Malacostraca, as far
as they can be reconstructed, can be well understood as modifications in accordance with adap-
tations to specific life habits achieved in each group separately. This makes any evolutionary
transition from the one to the other morphology unlikely; that is, in these cases, the features
represent autapomorphic states of the two taxa.
Therefore, we view the assumption of a paraphyly of Entomostraca, as commonly proposed
(see Waloszek 2003b for discussion), to be improbable and less parsimonious and uphold the
hypothesis of their monophyly. One area of evidence for this is the specific morphology of
appendages, which is detailed next.

Anterior Cephalic Appendages

The antennula of adult malacostracans comprises a prominent tubular basal part and two distal
flagella (three in some taxa) and has sensory function. Only in the nauplii of Euphausiacea and
Dendrobranchiata (the only malacostracan taxa with free-living nauplius larvae) is the anten-
nula uniramous and limblike as in their ancestors; it even bears a number of long swimming
setae on its distal end (e.g., Hirota et al. 1984a, 1984b, Kidd 1991). Thus, it appears that in these
early larvae the antennula recalls at least part of its original state. The original functions of the
antennulae, swimming and helping in sweeping in food particles, may have been abandoned
early during evolution and ontogeny. Nauplii of these groups only swim and do not feed; having
no feeding aids at all may even point to a very early loss and change to lecithotrophy. This loss
may have been in line with the modification of the other appendages taking these functions over
later during development, that is, from the protozoea onward.
Similarly, a plesiomorphy exposed in larvae is recognizable in the antenna of Euphausiacea
and Dendrobranchiata (e.g., Cockcroft 1985 for the penaeid Macropetasma africanum Balss,
1913, Maas and Waloszek 2001b for the Antarctic krill Euphausia superba Dana, 1852). The mor-
phology is very similar to that of feeding entomostracan nauplii, but those of malacostracan
taxa lack any median feeding structures because they do not feed. Later in ontogeny the anten-
nal endopod is transformed into a sensorial “feeler” with a long multiannulated flagellum, and
the exopod becomes the paddle-shaped “scaphocerite” (e.g., Maas and Waloszek 2001b for E.
superba as an example of euphausiids). Lastly, the naupliar mandible of malacostracans with
early larvae is very similar to those of early larval entomostracans, including coxa, basipod,
and two rami. The retention of the parts distal to the coxa, called palp in later stages and the
adult, may be plesiomorphic, but this uniramous palp is only remotely similar to the original
condition of the Eucrustacea. The adult palp consists of three articles interpreted as basipod
and the bipartite endopod (see, e.g., Olesen and Walossek 2000). The morphology of the palp
is remarkably similar in all known ingroup taxa and clearly represents an autapomorphy of
Malacostraca.
In Cephalocarida, Branchiopoda, and, as the ontogenetic path suggests, most likely in a
number of Orsten species, such as Yicaris dianensis, Rehbachiella kinnekullensis, and Walossekia
quinquespinosa, all parts of the mandible distal to the coxa are lost during ontogeny. This condi-
tion is considered an autapomorphy in the ground pattern of Entomostraca. We thus understand
the presence of a palpus in Maxillopoda as resulting from a pedomorphic event in their evolu-
tionary history (e.g., Newman 1983, Walossek and Mü ller 1998a, Waloszek 2003b).
58 Functional Morphology and Diversity

The situation of the exopods of antennae and mandibles is more complicated. In entomo-
stracans, they may retain their multiannulated morphology (e.g., cephalocarids, some maxil-
lopodans), be modified (some phyllopod branchiopods), or become lost (some maxillopodans,
anostracan branchiopods). Immature stages retain the original morphology in all taxa. In
Malacostraca, the larvae—where developed—have multiannulated antennal exopods, but
in adults the exopod has at most two parts and is flattened, that is, the scalelike scaphocerite.
Furthermore, these appendages can undergo high modification or complete loss in particular
ingroups, but mostly entomostracans. Examples are the tantulocarids (all cephalic appendages),
facetotectans (antenna, mandible), and cirripedes (antenna).
Not only do the maxillulae of Malacostraca and Entomostraca differ significantly in
their morphologies, but they also differ from the equivalent limb in the Phosphatocopina
and stem derivatives. The homotope, that is, the positional homologue, of the maxillula in
Phosphatocopina is still serial to all more posterior appendages and consists of a lobate proximal
endite, a basipod with one median endite, and the rami (Maas et al. 2003). In both Malacostraca
and Entomostraca, the maxillula is developed as a further “mouthpart” and differs in morphol-
ogy from the more posterior appendages. The malacostracan maxillula is uniformly rather thin
and has a slight C-shape (concave anteriorly) and a proximal endite modified into a coxa. Both
coxa and basipod are medially drawn out into one blade-shaped enditic process with a rim of
spines or setae. Both maxillula and maxilla are similar and act as a unit behind the mandible,
closing the food chamber posteriorly. In the ground pattern, the entomostracan maxillula has
a proximal endite and three more setose endites along the median edge of the basipod (not
one as in phosphatocopines or stem derivatives). This state is realized in copepods, in mysta-
cocarids, and in cephalocarids during the larval phase. In the stem-branchiopod Rehbachiella
kinnekullensis it occurs until the latest instar stage known, and it is also developed in, for exam-
ple, the Orsten taxa Yicaris dianensis (Zhang et al. 2007), Walossekia quinquespinosa (unpub-
lished), and Bredocaris admirabilis (Mü ller and Walossek 1988). Crown-group branchiopods
(Eubranchiopoda) have reduced maxillulae and maxillae retaining only their proximal endites
(see Walossek 1993).
In summary, the maxillulae have a different fate from that of the maxillae in entomostracans
but are both fully integrated and acting together in a cephalic feeding system in malacostracans.
There they are rather similar to each other, and both have a coxal portion below the basipod.
Only in cephalocarids among the living entomostracans does the maxillula serve as a single
“mouthpart,” while the maxilla is a trunk limb. In all other taxa, the situation is different. In
the extreme, maxillula and maxilla are almost lost (eubranchiopods) or completely lost (para-
sitic tantulocarids). Remarkably, the maxillae retain a paddle-shaped exopod in all taxa where it
remains developed—and there is no exception throughout.

Postmaxillulary Appendages
In Malacostraca the maxilla is much flattened in anteroposterior aspect, in this way looking much
like the maxillula. It is also C-shaped (anteriorly concave), and its proximal endite forms a coxa.
Coxa and basipod are medially drawn out into bladelike spine- to setae-bearing protrusions, but in
contrast to the maxillula, they have a median cleft, so they appear divided in two. Both maxillulae
and maxillae have a very soft and fragile appearance and are considerably shorter than all subsequent
limbs of the trunk. These postcephalic trunk appendages of Malacostraca not only are different
from the maxillae but also cannot easily be discussed because (1) they occur in two very distinctive
sets, and (2) they are rather different in the two sister taxa Phyllocarida and Eumalacostraca. Only
the first eight thoracic limbs (of thorax I according to Walossek and Müller 1998a) have a limb stem
made of a coxal and basipod portion, but the so-called pleopods (limbs of thorax II sensu Walossek
and Müller 1998a) have neither a coxa nor a proximal endite.
Evolution of Crustacean Appendages 59

The anterior eight thoracopods of phyllocarids are flattened and appear very superficially
similar to the filter appendages of Branchiopoda—though virtually all details are differ-
ent, the feeding system is a closed one, phyllocarids do not filter feed in the strict sense (see
Walossek 1993 for detailed comparisons), and they have fairly large coxae and basipods, from
which an elongate endopod and a paddle-shaped exopod arise. The first four pleopods have a
rod-shaped basal part, which carries two rami of equal size immediately on its top (e.g., Olesen
and Walossek 2000, their fig. 7b). Pleopods 5 and 6 are small and consist only of an elongated
lobe. The eight anterior thoracopods of Eumalacostraca have fairly short coxae and basipods
but likewise elongate endopods. The exopods are nowhere prominent except in euphausiids
and may be small paddles (e.g., Maas and Waloszek 2001b, their fig. 11) or multiannulated rods
stemming from a proximal peduncle piece (e.g., Neil et al. 1976, their figs. 3a, 5c, 9)—the slop-
ing articulation area of the exopod recalls the original morphology taken over from the euar-
thropodium. As in Phyllocarida, eumalacostracan pleopods lack a coxa or a proximal endite,
and also their rami rest on top of the stem portion. The variety of morphologies of the rami is,
however, large.
The maxilla in the ground pattern of Entomostraca is, on the other hand, very different
from the morphology in the maxilla or trunk limbs of malacostracans—and it matches that of
the series of posterior limbs (Fig. 2.3H). In size and morphology, it virtually equals the trunk
limbs at this stage. Its proximal endite is prominent, possibly the largest of all postmaxil-
lulary limbs, and in its setation grossly similar to that developed in Phosphatocopina. The
basipod is a large subrectangular element, longer than wide. Its mediodistal extension forms
a kind of socket for the transition to the endopod and a sloping distal outer margin from
which the exopod is articulated. Its straight median edge is drawn out into several lobate
setiferous endites (Waloszek 2003b). Their armature resembles that of the proximal endite
but is progressively less elaborate. The basipod body is fairly f leshy and little sclerotized,
with the exception of the lateral edge proximal to the exopod insertion, which is slightly
better sclerotized but interrupted by two furrows, giving this side a tripartite appearance
(Fig. 2.6F,G). For the extant cephalocarids (see Fig. 2.6B), it is clear that the proximal endite
and the endites along the inner rim of the basipod have muscles internally and can be moved
accordingly (Sanders 1963, Hessler 1964). It is possible that, as the preservation of endites in
Rehbachiella kinnekullensis and other such Orsten eucrustaceans suggests, the movability
was a particular feature of these endites already in the ground pattern of Entomostraca and
an additional autapomorphy. In those taxa where the basipod rims are more or less straight,
more rigid, and adorned with rows of setae, this would be a secondary adaptation in line with
somewhat modified food intake.
The endopod appears to be the direct continuation of the basipod. Not only do its proxi-
mal podomeres match the shape of the distal enditic protrusions of the basipod, but also the
setation pattern is continued, again being less and less developed toward the distal end. The
endopod is maximally six segmented in Eucrustacea and ends in a much smaller distal conelike
to caplike piece with a terminal tuft of setae or spines. The exopod is paddle shaped with a mar-
ginal row of long setae. This entire morphology matches that of the subsequent limbs and is
developed in this way not only in certain Orsten taxa (Yicaris dianensis, Rehbachiella kinnekul-
lensis, Dala peilertae, and Bredocaris admirabilis; see Mü ller 1983, Mü ller and Walossek 1985,
1988, Walossek 1993, Zhang et al. 2007) but also in the extant Cephalocarida (e.g., Carcupino
et al. 2006, their fig. 4A).
Although this type of appendage does not match that of any of the known stem derivatives,
phosphatocopines, or malacostracans (which have a coxa in the maxilla and anterior eight tho-
racopods), we can find here a mixture of plesiomorphies and apomorphies present on a sin-
gle structure. In terms of seriality, the maxilla in the ground pattern of entomostracans (and
60 Functional Morphology and Diversity

ingroups!) is similar, that is, homonomous, to the more posterior limbs. In other words, it is not
included into a specific cephalic feeding and locomotory apparatus, so it can be termed maxilla
in the strict sense.
Superimposed on this plesiomorphy are other, autapomorphic features, such as a subdivi-
sion of the median edge of the basipod into several setae-bearing lobate endites (up to seven
in Yicaris dianensis; see Zhang et al. 2007). The same applies to the more posterior append-
ages, the thoracopods, in Entomostraca (up to eight basipodal endites in Yicaris dianensis; see
Zhang et al. 2007). Furthermore, the otherwise transversely inserting basipods of all post-
maxillulary appendages are elongated in proximodistal and lateral axis and are rather f leshy
and slightly C-shaped curved backward, so the posterior side is concave. In cephalocarids this
special morphology forms narrow chambers between the limbs in the row. These are opened
and closed during the moving cycle (heterochronal beat) of the limbs pressing out water or
sucking it in during opening. The system of pumping chambers is known from these and
other eucrustaceans but is studied in more detail in anostracan branchiopods, which filter
feed (Fryer 1983), a habit not applicable to cephalocarids (Walossek 1993). Suspension feeding
is widespread among the Crustacea and is not discussed here any further (for more details,
see chapter 8).
Yet, for both taxa it is known that the system has two functions: feeding and locomotion.
Moreover, it requires the open system, as noted for the Labrophora level. Maxillae forming a
chamber as in Malacostraca is, however, a very different operational system. Also, maxillae and
a so-called maxilliped operating as a closed feeding system, as in copepods, can be set off, but
since this condition occurs in ingroup entomostracans, this system is derived from the open ento-
mostracan type. A similar system with fleshy basipods has also been reconstructed for Orsten
entomostracans such as Rehbachiella kinnekullensis and Yicaris dianensis, so it may have operated
similarly. The differences in detail (setae, subsetules, endite form, etc.) not only point to differ-
ences in feeding and locomotion but also may indicate different development in different evolu-
tionary lineages.
The basipodal endites of all postantennular appendages are, as mentioned above, equipped
with setae and spines. These occur in a pattern of three different rows of spines/setae, similar
to that of Phosphatocopina (ground pattern of Labrophora). However, while the anterior and
posterior sets of setae on the enditic protrusions are still undifferentiated in Phosphatocopina
(Fig. 2.5A), in Entomostraca the setae are differentiated both in structure and arrangement into
anterior retention setae with rows of more backward oriented setulae, median stronger spines,
and posterior setae, which have rows of anteriorly pointing setulae (Fig. 2.5B,C), each set form-
ing a small basket. The specific morphology and the so-called sucking chambers between the
limbs enable the animals to feed and locomote at the same time using the entire postmaxillulary
apparatus (Walossek 1993), as seen today in cephalocarids and branchiopods. This morphology
was most likely modified again in Maxillopoda.
The tripartite set of setae develops ontogenetically through a two-part set, as exemplified
by Rehbachiella kinnekullensis (Walossek 1993). This ontogenetic change from a two-row to a
three-row system can also be observed in Yicaris dianensis (Zhang et al. 2007). The presence
of a two-row system in Maxillopoda, such as in Bredocaris admirabilis (Mü ller and Walossek
1988), can thus be understood as another effect of pedomorphosis that affected Maxillopoda
(Newman 1983, Walossek 1993).
None of these special changes of the basipods can be found in Malacostraca. There the
maxilla is indeed developed as a specialized mouthpart in the ground pattern of this taxon,
more or less an aid to close the oral chamber. The maxilla is extremely f lat, similar to the max-
illula, and the proximal endite is also enlarged to form a coxa. Coxa and basipod are medially
drawn out into two (not just one) bladelike protrusions each. In both maxillula and maxilla,
Evolution of Crustacean Appendages 61

the endopods are most likely subdivided into at least three articles in the ground pattern of
Malacostraca, as exemplified by the euphausiacean Bentheuphausia amblyops G.O. Sars, 1883
(see Maas and Waloszek 2001b). These limbs differ from those of the Entomostraca also in
that they surround the mandibles like hands held over the mouth. Accordingly, this apparatus
is a closed system, whereas that of entomostracans is an open system.
Basipods of more posterior limbs in Malacostraca may carry setation but do not have lobate
endites (Fig. 2.6A,E). It has to be noted here that the appendages of the Silurian fossil phyl-
locarid Cinerocaris magnifica Briggs et al., 2004 have been interpreted as possibly possessing
endites (Briggs et al. 2004). Unfortunately, the appendages of this species have been presented
as two-dimensional line drawings (Briggs et al. 2004, their fig. 2i) and not, as usual for this pres-
ervational type, as three-dimensional reconstructions (see, e.g., Siveter et al. 2007). It is there-
fore impossible to verify the presence of any enditic subdivision of the basipods of trunk limbs
in Malacostraca. Consequently, based on the existing data, a proximodistally elongated basipod
with lobate setae-bearing enditic subdivisions of the narrow median edge is regarded here as an
autapomorphy of Entomostraca (Fig. 2.6B–D,F,G; Waloszek 2003b).
Malacostraca possess further specializations on their postmaxillary appendages, namely,
in their division into two series, eight belonging to the first part of the thorax (thorax I sensu
Walossek and Mü ller 1998a) and six so-called pleopods belonging to the second part of the tho-
rax (thorax II or pleon) in the ground pattern. Only thoracopods 1–8 appear to bear a true coxa;
that is, the proximal endite is laterally enlarged to form a complete enclosed sclerotized ring
(Waloszek 2003b). The presence or absence of a coxa or proximal endite could not be shown
with certainty in any malacostracan so far. Another evolutionary modification that cannot
be followed up here in more detail, however, is the specialization of one or even more limbs
as mouthparts, so-called maxillipeds. This feature, possibly even convergently developed, is a
gradual process and does not change the general functionality of the closed oral chamber, but
just adds the next limb—for example, as a food grasper or holder—to form a larger unit. The
trunk limbs progressively lose their ability to provide assistance in food intake, a trait that is
still recognizable in phyllocarids and at least distally in the case of the basket-feeding euphausi-
ids (Hamner 1988). It should be noted that similar strategies occur in entomostracans, where a
closed oral feeding chamber including maxillipeds exists, and the remaining thoracopods are
devoid of feeding structures and function.
Summing up regarding limb morphology, the following features are interpreted as autapo-
morphies of Malacostraca:

• Mandible with a typical tripartite palp (basipod and two endopod elements)
• Maxillula with a coxa; coxa and basipod each with a bladelike endite
• Maxilla with a coxa; coxa and basipod with two endites each
• Trunk limb series divided into two special series (thorax I and II of Walossek and Mü ller
1998a) of eight anterior and six posterior pairs of appendages; appendages of the anterior
series with coxae

Limb-related autapomorphies of Entomostraca include the following:

• Mandible loses palp late in ontogeny


• Maxillula with proximal endite and with basipod bearing three endites
• Basipod of “maxilla” (not specialized as a mouthpart) and all trunk limbs elongated in
proximodistal axis of the limb, medially equipped with a series of endites (seven to eight
originally?); basipod laterally subdivided into three major parts; setation on the endites
organized in three sets of setae specialized for sorting of food particles anteriorly and
retention setae posteriorly (orientation of subsetules different)
62 Functional Morphology and Diversity

VARIATIONS ON A THEME: LIMB MODIF ICATIONS

The two rami of an appendage, endopod and exopod, are modified and specialized in many
ways in the various eucrustacean ingroup taxa; in the extreme, they are completely lost. In the
following we give a number of examples for different morphological variations of the two rami.
This is, however, a limited set since the morphological variation is very extensive within such a
large taxon as the Crustacea.

Endopod Variations

The crustacean endopod plesiomorphically has a simple tubelike shape with setae or spines on
mediodistal humps along the inner margin (one such hump per podomere) and a setiferous to
spine-bearing tip, possibly an adaptation for food gathering and locomotion (much retained
from the euarthropod ground pattern; see above). Alterations occur in various ways. On the one
hand, the endopod can be, very rarely, partially reduced to a simple bulbous structure within
Eubranchiopoda (Olesen et al. 2001). (Another example outside Crustacea is the endopod of
the second head appendage of Agnostus pisiformis [see Mü ller and Walossek 1987, their fig. 6B
and plates 12.5, 16.1].) Reduction in size of the endopods and/or of their segmentation in liv-
ing branchiopods led some authors to (among other errors) misidentify the basipodal endites
as endopod podomeres of the according appendage (e.g., Wehner and Gehring 1995, caused by
Preuss 1957; for correction, see Walossek 1993).
On the other hand, the endopod may become elongated by its podomeres, the number of
which stays the same. This occurs, for example, in the thoracopods 1–8 of Eumalacostraca, and
there even the whole limb may become stenopodous by parallel reduction of the coxa and basi-
pod as well as the reduction or even loss of the exopod (walking legs). This is also independent
of the numbers in toto: an endopod made of one or two portions may be as large as others with
more portions. Because of this it is extremely difficult to identify functional or evolutionary
adaptations. As hinted at before, each limb has to be viewed separately and in the context of the
other limbs.
Another specialization is that endopod tips may become subchelate or chelate in various
lineages, for example, in Stomatopoda (Morgan and Goy 1987), Peracarida (tanaidaceans or
amphipods), and Decapoda (Richter and Scholtz 2001). Also within Entomostraca, endopods
with a subchelate tip may be found, such as in the form of claspers of males to attach to the
females—examples are the first trunk limbs of certain diplostracans (e.g., Olesen et al. 1996)
or the antenna of Anostraca (e.g., Dumont and Negrea 2002, their fig. 130I). It is apparently
evolutionarily possible even to change the endopod morphology from one extreme (partially
reduced endopod) to the other (elongated stenopodous endopod), as proven by the raptorial
water fleas within Cladocera. In the ground pattern of Cladocera, the thoracic endopods are,
as in the ground pattern of Eubranchiopoda, simple undivided bulbs (Olesen 2004, 2007).
Within Onychopoda and Leptodora kindtii Focke, 1844, the whole anterior thoracic appendages
are elongated into multisegmented stenopodia to fulfill raptorial functions (Olesen et al. 2001,
2003; Fig. 2.7F). The endopod may also become a paddle-shaped structure for swimming (see
“Symmetries in the Morphology of the Rami,” below).
Although partial reduction is possible, a complete loss of the endopod is quite unusual
within eucrustaceans. Often only the loss of the entire appendage also permits the endopod
to be missing as in the pleopods of various interstitial peracarids, or the loss of the distal part
of the appendage including the basipod such as in branchiopod mandibles. Investigation of the
ontogeny of, for example, the Cambrian branchiopod Rehbachiella kinnekullensis, in which the
mandible is complete in early stages (Walossek 1993, his plate 4–1), reveals that the distal part,
Evolution of Crustacean Appendages 63

Fig. 2.7.
Variations of biramous crustacean limbs. (A and B) The branchiopod Lepidocaris rhyniensis Scourfield,
1926. Images show a maximum intensity projection; that is, images of different focal planes are combined
(for details, see Haug et al. 2009b). Arrows indicate exopods of adjacent limbs. (A) Anterior postmandib-
ular appendage in anterior view. Exopod and endopod differ strongly in their morphology and in their
insertion angle and position on the basipod. (B) Posterior postmandibular appendage in posterior view.
Exopod and endopod are rather symmetric morphologically and in their insertion position and angle on
the basipod (“copepod-like”). (C) Thoracopod of Lepas sp. Linnaeus, 1758 (Cirripedia). Endopod and exo-
pod (cirri) are morphologically indistinguishable (arrows); only their position in the living animal shows
their identity. Image not to scale. (D) Tail fan of the Jurassic stomatopod Sculda pennata Mü nster, 1840.
The uropodal exopod is a simple undivided paddle. (E) Tail fan of the extant stomatopod Squilla mantis
Linnaeus, 1758. The uropodal exopod is divided into a distal paddle and a proximal article. (F) Ventral view
on a late embryo of the phyllopod Leptodora kindtii Focke, 1844. Arrows mark the second and third tho-
racopods, developed as stenopodous limbs with median setation, whereas the ground pattern of phyllopod
branchiopods is characterized by leaf-shaped limbs. (G) Second antenna of the branchiopod Macrothrix
laticornis Jurine, 1820. Exopod and endopod are symmetrically composed of three tubular elements. (H)
Thoracopod of an undetermined copepod. Endopod and exopod are symmetrically composed of three
setose anteroposteriorly flattened elements. Image not to scale. Abbreviations: bas, basipod; cox, coxa;
dex, distal paddle of exopod; en, endopod; ex, exopod; pex, proximal article of exopod. Repository num-
bers (see Fig. 2.1 for abbreviations): (A and B) BM 25698; (D) MB.A.669.
64 Functional Morphology and Diversity

the “palp,” of the appendage becomes progressively smaller. Finally, a small round scar marks
the original insertion of the palp, but the appendage consists solely of the coxa (Walossek 1993,
his plate 17–4). One of the rare examples of complete endopod loss in Eucrustacea is the degen-
eration of the posterior two thoracopods within Euphausiacea. The gill in these two appendages
is the most prominent structure. A coxa-basipod subdivision is not identifiable; a uniramous
structure with marginal setation is identified as exopod, while an endopod is lacking (see, e.g.,
Maas and Waloszek 2001b, their fig. 13).

Exopod Variations

The exopods, often referred to as the swimming branches (e.g., Calman 1909), also exhibit
a wide variety of morphologies in the different eucrustacean taxa. As mentioned above, in
Eumalacostraca, the antennal exopod is a two-part paddle that is reduced even further to a
simple undivided paddle in Caridoida (Richter and Scholtz 2001; caridoids are the sister group
to stomatopods). While exopods are paddle shaped in the maxillulae and maxillae where
developed, they are multiannulated on the first eight thoracopods of various taxa within the
Eumalacostraca. There the exopod is highly movable and is used for locomotion—for exam-
ple, in larvae of lobsters (Neil et al. 1976) or in mysid peracaridans, where it may move at
high speed to keep the animal in position like a helicopter (the German name Schwebegarnele
refers to this). In certain entomostracan taxa, the exopod is subdivided into several articles,
often three, as in the trunk limbs of Copepoda and Remipedia (Itô 1982, 1989), but this is
unlikely to be a hint to relationships but merely a functional enforcement to achieve a more
f lexible paddle. Another example of the appearance of an additional joint in an exopod is the
uropodal exopod of some Eumalacostraca, for example, the Stomatopoda. All extant species
of Stomatopoda have a bipartite uropodal exopod (Fig. 2.7E), but a number of fossil species
demonstrate that this condition is most parsimoniously derived from a simple undivided exo-
pod (Schram 2007; Fig. 2.7D).
The exopod has become more subject to loss than the endopod, both during evolution and
during ontogeny. This is most evident in all land-living bottom-walking crustaceans, as well as
in other euarthropods such as arachnids, myriapods, and insects. But this also holds true for
limbs of benthic walkers in the marine environment such as thoracopods 4–8 of Eureptantia
among the decapod Malacostraca (not the anterior thoracopods 1–3, which are modified into
the maxillipeds, specialized limbs that support feeding) or all isopods, in both cases an autapo-
morphy of these taxa, or outside crustaceans in the pantopod Chelicerata (already observa-
ble in fossils from the Devonian; Bergström et al. 1980). Exopod loss in nonlocomotive limbs
can be recognized, for example, on the mandibles of land-based isopod peracarids among the
Malacostraca (so clearly convergent to Entomostraca) or on the maxillulae and maxillae of
Mystacocarida (species of the taxon Ctenocheilocaris Renaud-Mornant, 1976 also lack the exo-
pod on the maxilliped) and Copepoda (unclear if this hints at a close relationship).
We consider these true losses, not a lack of subdivision of exopod and endopod. The idea of the
loss of exopods through a lack of subdivision of exo- and endopod (and thus the remaining “endo-
pod” being a homologue of former endopod and exopod) has been concluded from the results of a
cell lineage investigation of an amphipod (Wolff and Scholtz 2008). It appears to be unfounded to
expand this putative mechanism, which is interpreted to exist in a highly derived ingroup malacos-
tracan taxon, to all Crustacea, or even Arthropoda. This also contradicts all data accumulated that
depict the evolutionary path from the lobopodium to the arthropodium and euarthropodium and
eventually the crustacean limbs, as we explained it here. Even though the cell lineage data may indi-
cate that the walking limbs of the pereopods are the result of “undivided” branches, the structural
similarities to the endopods of other malacostracans and the endopod origin from the distal part of
Evolution of Crustacean Appendages 65

the limb stem in Arthropoda s. str. clearly reject this assumption (for discussion, see also Boxshall
and Jaume 2009). Other Crustacea also gain uniramous appendages through loss of exopods. In the
mystacocarid species of the taxon Derocheilocaris Pennak and Zinn, 1943, for example, the maxillula
and maxilla appear very similar to the maxilliped—except that the maxilliped possesses an exopod.
Therefore, it appears to be much more parsimonious to assume a simple loss of the exopod in max-
illula and maxilla than to homologize their endopods with both rami of the maxilliped.

Symmetries in the Morphology of the Rami

Despite their clearly different phylogenetic origin, the two rami, endopod and exopod, can
be almost identical or symmetrical in certain eucrustacean taxa. There they even serve the
same functions. Examples of such symmetrization of rami can be found in various taxa and
on different appendages—clearly an example of convergence. Some examples of this are the
“cirri” (endopod and exopod) of the thoracopods of adult barnacles (Fig. 2.7C) and, within the
Branchiopoda, the antennae of Diplostraca. These bear two very symmetric rami, best known
probably from Cladocera (= “branched horns”; i.e., the taxon name established by Pierre André
Latreille in 1829 even refers to this symmetrization), including one of the standard laboratory
organisms in student courses, Daphnia pulex De Geer, 1778. Both endopod and exopod comprise
three articles, with the proximal articles carrying a single median seta each and the terminal
article carrying a number of setae concentrated medially (Fig. 2.7G). All these similar-appearing
setae are long and setulose and mainly serve as swimming devices.
Another example of symmetrization of rami is Lepidocaris rhyniensis Scourfield, 1926, a
putatively branchiopod species from the famous Early Devonian Rhynie Chert lagerstätte in
Scotland (Scourfield 1926). The anterior trunk appendages exhibit clearly differing morpholo-
gies in the endopod and the exopod (Fig. 2.7A). But in the series of posterior trunk appendages,
the so-called copepod-like appendages, both rami appear very similar. They are developed as
simple paddles with seven to nine setae along their whole margins (Fig. 2.7B). Symmetrization
in this case affects not only the shape of the two rami but also their insertion at the basipod—
both rami insert distally. In the anterior trunk appendages, the exopods insert laterodistally
(Scourfield 1926). The morphology of these posterior limbs has been interpreted as an adapta-
tion to swimming.
As the terminology in Lepidocaris rhyniensis (“copepod-like” appendages) already indicates,
copepods also have symmetrical rami on the posterior trunk limbs. Unlike in L. rhyniensis, the
endopod and exopod are not simple paddles but are more elongated and comprise three arti-
cles (Fig. 2.7H). Even so, endopod and exopod have a very similar appearance. A comparable
morphology of the two rami as symmetric elongated paddles with three articles on the trunk
appendages is also found in Remipedia (Itô 1989).
Also, in various malacostracan taxa, appendages with symmetrical rami occur. The pleopods
of Eumalacostraca are one example. In Stomatopoda the endopods and exopods of the pleopods
(limbs of thorax II sensu Walossek and Mü ller 1998a) are developed as symmetrical paddles
bearing a large number of setae along the whole margin. Yet, the symmetry is partly interrupted,
as the exopod carries gills on its anterior side (Morgan and Goy 1987, Maas et al. 2009).

Seriality of the Limbs and Tagmatization

Changes to the appendage morphologies can affect single appendages or complete series and
can be coupled to tagmatization pattern changes, but need not. For example, the inclusion of
further appendages into a specialized feeding apparatus is not necessarily linked to the inclu-
sion of the corresponding segments into the head tagma. There are, on the other hand, examples
66 Functional Morphology and Diversity

of the inclusion of segments into the head tagma with unmodified appendages, that is, append-
ages that look exactly like trunk appendages, as it is reconstructed for the eucrustacean ground
pattern and also exhibited by extant cephalocarids. Also, the reverse case can be found, where
an appendage is distinctly evolved into a specialized feeding apparatus, while the segment is
dorsally clearly set off from the head tagma. This can be seen in the Copepodoida. The autapo-
morphy of this taxon comprising, in our view, the Copepoda, the tiny interstitial Mystacocarida
and the Cambrian Skaracarida is a special cephalothoracic feeding apparatus with one maxil-
liped involved. The maxilliped segment is incorporated in the cephalothorax in copepods; that
is, its dorsal surface is included into the shield. In Skaracarida and Mystacocarida, the segment
is set off dorsally: it is movable against the head (Walossek and Mü ller 1998b).
Stomatopoda possess a feeding apparatus, which involves five postmaxillary maxillipeds.
The posterior four maxillipeds are arranged in a highly condensed area on the ventral surface.
Nevertheless, the fifth maxilliped indeed belongs to an extra segment, which is not included
into the cephalothorax and which possesses an extra tergite (e.g., Ahyong 2001). Research in
developmental genetics has supported this long known morphological phenomenon: dorsal and
ventral segmentation patterns need not necessarily be coupled (see Janssen et al. 2004).
Also, the postcephalic trunk region of labrophorans or eucrustaceans (head includes here
the maxillary = fifth appendage-bearing segment) may be further subdivided into tagmata and
with this exhibit differentiated appendage morphologies. A subdivision of the trunk into an
appendage-bearing thorax and a limbless abdomen is a possible autapomorphy supporting the
taxon Entomostraca (Walossek and Mü ller 1998a, Maas et al. 2003, Waloszek 2003b).
An autapomorphic trunk division of Malacostraca is the differentiation into two compart-
ments, called thorax and pleon or thorax I and II (see Walossek and Mü ller 1998a), both with
differently specialized appendages (Waloszek 2003b). In Eumalacostraca the pleon is further
subdivided as the sixth pair of appendages is transformed in structure and insertion to form a
tail fan together with the telson (see Fig. 2.7D,E for two stomatopod species). Functional subdi-
vision of the limb-bearing thorax is rarely known from Entomostraca, such as in the Devonian
fossil Lepidocaris rhyniensis. In this species, the anterior thoracopods have well-developed
endites and are also used for food transport, while the posterior appendages have symmetric
rami and are mainly used for swimming (Scourfield 1926).
In many Entomostraca, the last thoracopods may be specialized, often highly modified, for
copulation or as an egg carrier (e.g., Sanders 1957 for Cephalocarida; Torrentera and Dodson
1995 for Anostraca). In Malacostraca, the anterior one or two pleopods may be modified as a
sperm-transfer appendage (mostly called petasma, though likely comprising various different
morphologies; e.g., Martin and Abele 1986), or a thoracopod of the first thoracic limb series
bears an appendix for sperm transfer. In summary, many morphological changes occur in con-
junction with copulation and the transfer (in males) or the reception (in females) of sperm. The
same holds for brood care, which also led to the development or modification of specific struc-
tures associated with the thoracopods, such as oostegites in Peracarida (as possible modified
epipodites; see Maas et al. 2009) or specific setae that hold the eggs, such as in diplostracans.
Within various crustacean groups and possibly in the ground pattern character at least of
Eucrustacea, epipodites occur as rather soft outgrowths on the lateral side of the limb stems of
thoracopods. They may serve for a respiratory or osmoregulatory function (see Maas et al. 2009
for an extensive essay of these structures).

FUTURE DIRECTIONS

One important unsolved riddle is the early evolution of Malacostraca. As mentioned above,
entomostracans are well represented in the fossil record as early as the Terreneuvian, while we
Evolution of Crustacean Appendages 67

simply lack malacostracan fossils for this period and even several million years later. All Paleozoic
malacostracans show clear malacostracan features, so they are not derivatives of the stem line-
age and cannot contribute much to the reconstruction of the early evolution of Malacostraca
and their appendages. Here we have to wait for new fossil discoveries, because comparison of
extant malacostracan and entomostracan species cannot tell us anything about the evolution-
ary split of these two taxa. Also, the ground pattern status of Maxillopoda needs further sta-
bilization, because this taxon represents, together with Branchiopoda and Cephalocarida, the
taxa with extant derivatives within the Entomostraca. Therefore, we are confident that a closer
look at some fossil maxillopods, for example, Dala peilertae, will contribute to the solution of
this problem.
Another problem awaiting a solution is the phylogenetic position of Insecta, respectively
Tracheata. Especially for the appendages, it is difficult to draw out a plausible evolutionary sce-
nario that would show the evolution from a crustacean ingroup taxon to Insecta/Tracheata.
Especially difficult would be the derivation via Branchiopoda, as has been suggested quite
recently (Glenner et al. 2006). Insecta or Tracheata may well be Labrophora or at least their
sister group (Zhang et al. 2007), but still an evolutionary scenario remains hard to reconstruct
with the existing data on early insect/tracheate evolution. Leaving the character “appendages”
aside for reconstructing phylogenetic trees is not helpful since the animals evolved as a whole
and not just their single character complexes. Besides studying fossil specimens, a closer look at
the development of several taxa living today will provide additional information of probable use
for this question. While many publications in the modern evo-devo field already show promis-
ing results, the classic morphological studies of arthropod ontogeny also contribute important
data (see, e.g., Liu et al. 2009). Nevertheless, many fossil and extant arthropod species need to
be studied to finally solve this section of the tree of life.

CONCLUSIONS

Postantennular appendages of Crustacea show an extremely large variety of morphologies—in


our view more than in any other taxon within Euarthropoda, and this diversity is considered the
driving force of the crustacean success. In living eucrustaceans the head appendages are specifi-
cally dissimilar to the trunk appendages. This dissimilarity is also evident in land-living euarthro-
pods such as spiders and insects. As we know now from a number of fossil species, crustacean
evolution began, however, with only three specialized head appendages, that is, three anterior
appendages differing from the remaining ones—yet more than in the euarthropod ground pat-
tern! Specialization of the two postantennular appendages occurred successively along the
crustacean evolutionary lineage, most significantly in the ground pattern of Labrophora. Any
coincidence in the morphology of head appendages between Eucrustacea and the insects refers
to rather basal features, which would be explainable only when assuming at most a derivation
from a basal node within Labrophora. Ingroup eucrustacean affinities, as suggested, for example,
by molecular studies (e.g., Regier et al. 2008) suffer from little support by any morphological
structures (e.g., one criticism is that the segments bearing the excretory glands in insects differ
from those in all eucrustacean taxa). Furthermore, a huge variety of tree suggestions may also be
explained as being founded on symplesiomorphies.
Again, tagmatization of the body into head and trunk does not correspond to a functional
tagmatization of the limb series. A head with four appendage-bearing segments in the crustacean
ground pattern has only three specialized limbs originally. The fourth head limb is functionally
a trunk appendage. A larger head including the fifth limb-bearing segment in the eucrustacean
ground pattern is in conjunction with the specialization of the fourth head limb. However, the
fifth head limb is still trunk-limb shaped. Therefore, defining what a trunk limb or thoracopod
68 Functional Morphology and Diversity

is depends heavily on the evolutionary level. We even think that a differentiation is rather inap-
propriate, since further trunk segments become involved in the head, although their appendages
do not necessarily change their morphology.
Lastly, it is important to note that counting (at particular structures and taking the evolu-
tionary level into account) deserves a better reputation. The endopod, for example, originally
having more than eight articles in the ground pattern of Euarthropoda, shortened to comprise
seven articles at most in the sister taxon to Chelicerata (e.g., trilobites, Agnostus pisiformis). The
basal number is unclear for myriapods and hexapods (or Tracheata if monophyly is favored)
because of the unclear subdivision of the entire limb but should be six in the endopods of
Crustacea. Within this group, many taxa evolved lower numbers, individually on different
limbs and groups of limbs, but never more.
Probably most important when dealing with crustacean appendages is to recognize that
there is no special “crustaceopodium.” The morphology, ontogeny, and evolution of each
crustacean appendage need to be studied separately and in combination with the other
appendages.

ACKNOWLEDGMENTS

First, we thank Martin Thiel and Les Watling for their invitation to contribute to this volume
and for their editorial work. Thanks are also due to the Central Facility of Electron Microscopy
of the University of Ulm for their support in using their equipment. Several images were taken
during research stays in Copenhagen and London by J.T.H. and C.H., which were supported
by grants from the European Commission’s (FP 6) Integrated Infrastructure Initiative pro-
gram SYNTHESYS (DK-TAF-2171, DK-TAF-2652, GB-TAF-4733). During these visits, scan-
ning electron microscopic (SEM) images were taken on a JEOL JSM-840 and JEOL SM-31010
(Copenhagen), and light microscopy was performed on a Leica DFC 480 (London). Jørgen
Olesen, Copenhagen, kindly provided access to specimens from his collection (Figs. 2.5G,H,
2.6B, 2.7F,G). We would also like to thank several persons for their permission to use some of
their images: Zhang Xi-guang, Kunming (Figs. 2.1G, 2.5C, 2.6C), Jørgen Olesen, Copenhagen
(Fig. 2.5E), Yu Liu, Munich (Fig. 2.2B), and Verena Kutschera, Ulm (Fig. 2.7E). In Ulm, images
were taken on an SEM Zeiss DSM 962 at the Central Unit for Electron Microscopy and on a
Zeiss Axioskop with a mounted DCM 510 ocular camera. Some images had to be enhanced or
processed with various computer programs. Therefore, we express our sincere thanks to the
people that spent their time in providing open-access and open-source software programs such
as ImageJ, CombineZM, GIMP, and Inkscape. Lastly, we thank Klaus J. Mü ller, discoverer of
the Orsten, and the German Research Foundation (DFG) for its continuous funding of the
Orsten research activities, for which J.T.H. received funding under DFG WA-754/15–1.

REFERENCES

Ahyong , S.T. 2001. Revision of the Australian stomatopod Crustacea. Records of the Australian Museum,
Supplement 26:1–326.
Ax , P. 1995 . Das System der Metazoa I. Ein Lehrbuch der phylogenetischen Systematik. Gustav Fischer,
Stuttgart.
Bergström, J., X. Hou, X. Zhang , and S. Clausen . 2008. A new view of the Cambrian arthropod
Fuxianhuia. Geologiska Föreningens i Stockholm Förhandlingar 130:189–201.
Bergström, J., W. Stürmer, and G. Winter. 1980. Palaeoisopus, Palaeopantopus and Palaeothea, pycnogonid arthro-
pods from the Lower Devonian Hunsrück Slate, West Germany. Paläontologische Zeitschrift 54:7–54.
Evolution of Crustacean Appendages 69

Bowman, T.E., J. Yager, and T.M. Iliffe. 1985 . Speonebalia cannoni n.gen., n.sp., from the Caicos Islands,
the first hypogean leptostracan (Nebaliacea: Nebaliidae). Proceedings of the Biological Society of
Washington 98:439–446.
Boxshall, G.A. 2004 . The evolution of arthropod limbs. Biological Reviews 79:253–300.
Boxshall, G.A., and D. Jaume. 2009. Exopodites, epipodites and gills in Crustaceans. Arthropod
Systematics and Phylogeny 67:229–254.
Briggs, D.E.G., 1976. The arthropod Branchiocaris n. gen., Middle Cambrian, Burgess Shale, British
Columbia. Geological Survey Bulletin 264:1–29.
Briggs, D.E.G., 1978. The morphology, mode of life, and affinities of Canadaspis perfecta (Crustacea,
Phyllocarida), Middle Cambrian, Burgess Shale, British Columbia. Philosophical Transactions of
the Royal Society of London 281:439–487.
Briggs, D.E.G., M.D. Sutton, David J. Siveter, and Derek J. Siveter. 2004 . A new phyllocarid (Crustacea:
Malacostraca) from the Silurian Fossil-Lagerstätte of Herefordshire, UK. Proceedings of the Royal
Society of London Series B 271:131–138.
Budd, G.E. 1998. Stem group arthropods from the Lower Cambrian Sirius Passet fauna of North
Greenland. Pages 125–138 in R.A. Fortey and R.H. Thomas, editors. Arthropod Relationships.
Systematics Association Special Volume Series, Vol. 55. Chapman and Hall, London .
Budd, G.E. 2008. Head structure in upper stem-group euarthropods. Palaeontology 51:561–573.
Calman, W.T. 1909. Part VII. Appendiculata. Pages 1–346 in Lankester, R., editor. A Treatise on Zoology.
Crustacea, Vol. 3. Adam and Charles Black , London .
Carcupino, M., A. Floris, A. Addis, A. Castelli, and M. Curini-Galletti. 2006. A new species of the genus
Lightiella: The first record of Cephalocarida (Crustacea) in Europe. Zoological Journal of the
Linnean Society 148:209–220.
Chen, J.-y., J. Vannier, and D.-y. Huang. 2001. The origin of crustaceans: New evidence from the Early
Cambrian of China. Proceedings of the Royal Society of London Series B 268:2181–2187.
Chen, J.-y., D. Waloszek , and A. Maas . 2004 . A new “great-appendage” arthropod from the Lower
Cambrian of China and homology of chelicerate chelicerae and raptorial antero-ventral append-
ages. Lethaia 37:3–20.
Cockcroft, A.C. 1985 . The larval development of Macropetasma africanum (BALSS, 1913) Decapoda,
Penaeoidea) reared in the laboratory. Crustaceana 49:52–74.
Dahl, E. 1984 . The subclass Phyllocarida (Crustacea) and the status of some early fossils: A neontologist’s
view. Videnskabelige Meddelelser fra Dansk naturhistorik Forening 145:61–76.
Dumont, H.J., and S.V. Negrea. 2002. Introduction to the class Branchiopoda. Backhuys, Leiden .
Fryer, G. 1983. Functional ontogenetic changes in Branchinecta ferox (Milne-Edwards) (Crustacea,
Anostraca). Philosophical Transactions of the Royal Society of London Series B 303:229–343.
Fryer, G. 1988. Studies on the functional morphology and biology of the Notostraca (Crustacea:
Branchiopoda). Philosophical Transactions of the Royal Society of London Series B 321:27–124.
Glenner, H., P.F. Thomsen, M.B. Hebsgaard, M.V. Sørensen, and E. Willerslev . 2006. The origin of
insects. Science 314:1883–1884.
Hamner, W.M. 1988. Biomechanics of filter feeding in the Antarctic krill Euphausia superba: Review of
past work and new observations. Journal of Crustacean Biology 8:149–163.
Hansen, H.J. 1925 . On the comparative morphology of the appendages in the Arthropoda. A. Crustacea.
Studies on Arthropoda, Vol. 2. Gyldendalske Boghandel, Kopenhagen.
Haug , J.T., C. Haug , A. Maas, S.R. Fayers, N.H. Trewin, and D. Waloszek . 2009b. Simple 3D images from
fossil and Recent micromaterial using light microscopy. Journal of Microscopy 233:93–101.
Haug , J.T., Maas, A., and D. Waloszek. 2009a . Ontogeny of two Cambrian stem crustaceans, Goticaris
longispinosa and Cambropachycope clarksoni. Palaeontographica Abteilung A 289:1–43.
Haug , J.T., Maas, A., and D. Waloszek. 2010a . Henningsmoenicaris scutula, Sandtorpia vestrogothiensis
gen. et sp. nov., and heterochronic events in early crustacean evolution. Transactions of the Royal
Society of Edinburgh, Earth and Environmental Science 100:311–350.
Haug , J.T., D. Waloszek , C. Haug , and A. Maas. 2010b. High-level phylogenetic analysis using develop-
mental sequences: The Cambrian Martinssonia elongata, Musacaris gerdgeyeri gen. et sp. nov., and
their position in early crustacean evolution. Arthropod Structure and Development 39:154–173.
70 Functional Morphology and Diversity

Heegaard, P. 1945 . Remarks on the phylogeny of arthropods. Arkiv för Zoologi 37A:1–15.
Heldt, J.H. 1954 . Waptia fieldensis Walcott et les stades larvaires des Pénéides. Bulletin de la Societe des
Sciences Naturelles de Tunisie 6:177–180.
Hennig , W. 1965 . Phylogenetic systematics. Annual Review of Entomology 10:97–116.
Hessler, R.R. 1964 . The Cephalocarida. Comparative skeletomusculature. Memoirs of the Connecticut
Academy of Arts and Sciences 16:1–97.
Hirota, Y., T. Nemoto, and R. Marumo. 1984a . Larval development of Euphausia nana (Crustacea:
Euphausiacea). Marine Biology 81:311–322.
Hirota, Y., T. Nemoto, and R. Marumo. 1984b. Larval development of Euphausia similis (Crustacea,
Euphausiaeea) in Sagami Bay, Central Japan. Journal of the Oceanographical Society of Japan
40:57–66.
Hou, X.-G., and J. Bergström. 1997. Arthropods of the Lower Cambrian Chengjiang fauna, southwest
China. Fossils and Strata 45:1–116.
Hou, X.-G., J. Bergström, and G.-H. Xu. 2004 . The lower Cambrian crustacean Pectocaris from the
Chengjiang biota, Yunnan, China. Journal of Paleontology 78:700–708.
Itô, T. 1982. The origin of “biramous” copepod legs. Journal of Natural History 16:715–726.
Itô, T. 1989. Origin of the basis in copepod limbs, with reference to remipedian and cephalocarid limbs.
Journal of Crustacean Biology 9:85–103.
Janssen, R., N.M. Prpic , and W.G.M. Damen. 2004 . Gene expression suggests decoupled dorsal and
ventral segmentation in the millipede Glomeris marginata (Myriapoda: Diplopoda). Developmental
Biology 268:89–104.
Jones, M.L. 1961. Lightiella serendipita gen. nov., sp. nov., a cephalocarid from San Francisco Bay,
California. Crustaceana 3:31–46.
Kaestner, A. 1967. Lehrbuch der Speziellen Zoologie. Gustav Fischer, Stuttgart .
Kidd , R.J. 1991 . Development of embryonic and naupliar setae and spines and their role in hatching
in the penaeid Sicyonia ingentis: A light and electron microscopic study. Journal of Crustacean
Biology 11:40–55.
Latreille, P. A. 1829. Les crustacés, les arachnides, les insectes. Pages 1–584 in G. Cuvier, editor. Le règne
animal distribué d’après son organisation, pour servir de base à l’histoire naturelle des animaux et
d’introduction à l’anatomie comparée. Déterville, Crochard, Paris.
Lauterbach, K.E. 1986. Zum Grundplan der Crustacea. Verhandlungen des naturwissenschaftlichen
Vereins Hamburg (NF) 28:27–63.
Liu, J., and X.-P. Dong. 2007. Skara hunanensis a new species of Skaracarida (Crustacea) from Upper
Cambrian (Furongian) of Hunan, South China. Progress in Natural Science 17: 934–942.
Liu, Y., X.-G. Hou, and J. Bergström. 2007. Chengjiang arthropod Leanchoilia illecebrosa (Hou, 1987)
reconsidered. Geologiska Föreningens i Stockholm Förhandlingar 129:263–272.
Liu, Y., A. Maas, and D. Waloszek . 2009. Early development of the anterior body region of the grey
widow spider Latrodectus geometricus Koch, 1841 (Theridiidae, Araneae). Arthropod Structure and
Development 38:401–416.
Maas, A., A. Braun, X.-p. Dong , P.C.J. Donoghue, K.J. Mü ller, E. Olempska , J.E. Repetski, D.J. Siveter,
M. Stein, and D. Waloszek . 2006. The “Orsten”—more than a Cambrian Konservat-Lagerst ätte
yielding exceptional preservation. Palaeoworld 15:266–282.
Maas, A., C. Haug , J.T. Haug , J. Olesen, X. Zhang , and D. Waloszek . 2009. Early crustacean evolution and
the appearance of epipodites and gills. Arthropod Systematics and Phylogeny 67:255–273.
Maas, A., G. Mayer, R.M. Kristensen, and D. Waloszek . 2007. A Cambrian micro-lobopodian and the
evolution of arthropod locomotion and reproduction. Chinese Science Bulletin 52:3385–3392.
Maas, A., and D. Waloszek. 2001a . Cambrian derivatives of the early arthropod stem lineage,
pentastomids, tardigrades and lobopodians—an “Orsten” perspective. Zoologischer Anzeiger
240:451–459.
Maas, A., and D. Waloszek. 2001b. Larval development of Euphausia superba Dana, 1852 and a phyloge-
netic analysis of the Euphausiacea. Hydrobiologia 448:143–169.
Maas, A., and D. Waloszek . 2005 . Phosphatocopina—ostracode-like sister group of Eucrustacea.
Hydrobiologia 538:139–152.
Evolution of Crustacean Appendages 71

Maas, A., D. Waloszek , J.-y. Chen, A. Braun, X.-q. Wang , and D.-y. Huang. 2004 . Phylogeny and life habits
of early arthropods—predation in the early Cambrian sea. Progress in Natural Science 14:158–166.
Maas, A., D. Waloszek , and K.J. Mü ller. 2003. Morphology, ontogeny and phylogeny of the
Phosphatocopina (Crustacea) from the Upper Cambrian “Orsten” of Sweden. Fossils and Strata
49:1–238.
Martin, J.W., and L.G. Abele. 1986. Notes on male pleopod morphology in the brachyuran crab family
Panopeidae Ortmann, 1893, sensu Guinot (1978) (Decapoda). Crustaceana 50:182–198.
Martin, J.W., and C.E. Cash-Clark. 1995 . The external morphology of the onychopod “cladoceran” genus
Bythotrephes (Crustacea, Branchiopoda, Onychopoda, Cercopagididae), with notes on the mor-
phology and phylogeny of the order Onychopoda. Zoologica Scripta 24:61–90.
Mayer, G., G. Maier, A. Maas, and S. Waloszek . 2009. Mouthpart morphology of Gammarus roeselii com-
pared to a successful invader, Dikerogammarus villosus (Amphipoda). Journal of Crustacean Biology
29:161–174.
Morgan, S.G., and J.W. Goy. 1987. Reproduction and larval development of the mantis shrimp
Gonodactylus bredini (Crustacea: Stomatopoda) maintained in the laboratory. Journal of
Crustacean Biology 7:595–618.
Mü ller, K.J. 1983. Crustacea with preserved soft parts from the Upper Cambrian of Sweden. Lethaia
16:93–109.
Mü ller, K.J., and D. Walossek. 1985 . Skaracarida, a new order of Crustacea from the Upper Cambrian of
V ä stergötland, Sweden. Fossils and Strata 17:1–65.
Mü ller, K.J., and D. Walossek. 1986a . Arthropod larvae from the Upper Cambrian of Sweden.
Transactions of the Royal Society of Edinburgh, Earth and Environmental Science 77:157–179.
Mü ller, K.J., and D. Walossek. 1986b. Martinssonia elongata gen. et sp. n., a crustacean-like euarthropod
from the Upper Cambrian “Orsten” of Sweden. Zoologica Scripta 15:73–92.
Mü ller, K.J., and D. Walossek . 1987. Morphology, ontogeny and life habit of Agnostus pisiformis from the
Upper Cambrian of Sweden. Fossils and Strata 19:1–124.
Mü ller, K.J., and D. Walossek . 1988. External morphology and larval development of the Upper Cambrian
maxillopod Bredocaris admirabilis. Fossils and Strata 23:1–70.
Neil, D.M., D.L. Macmillan, R.M. Robertson, and M.S. Laverack. 1976. The structure and function of
thoracic exopodites in the larvae of the lobster Homarus gammarus (L.). Philosophical Transactions
of the Royal Society of London 274:53–68.
Newman, W.A. 1983. Origin of the Maxillopoda; urmalacostracan ontogeny and progenesis. Pages 105–
120 in F.R. Schram, editor. Crustacean phylogeny. Crustacean Issues, Vol. 1. Balkema, Rotterdam,
The Netherlands.
Olesen, J. 2004 . On the ontogeny of the Branchiopoda (Crustacea): Contribution of development to phy-
logeny and classification. Pages 217–269 in G. Scholtz , editor. Evolutionary developmental biology
of Crustacea. Crustacean Issues, Vol. 15. Balkema, Lisse, The Netherlands.
Olesen, J. 2007. Monophyly and phylogeny of Branchiopoda, with focus on morphology and homologies
of branchiopod phyllopodous limbs. Journal of Crustacean Biology 27:165–183.
Olesen, J., J.W. Martin, and E.W. Roessler 1996. External morphology of Cyclestheria hislopi (Baird,
1859) (Crustacea, Branchiopoda, Spinicaudata), with a comparison of male claspers among the
Conchostraca and Cladocera and its bearing on phylogeny of the “bivalved” Branchiopoda.
Zooogica Scripta 25:291–316.
Olesen, J., S. Richter, and G. Scholtz . 2001. The evolutionary transformation of phyllopodous to sten-
opodous limbs in the Branchiopoda (Crustacea)—is there a common mechanism for early limb
development in arthropods? International Journal of Developmental Biology 45:869–876.
Olesen, J., S. Richter, and G. Scholtz . 2003. On the ontogeny of Leptodora kindtii (Crustacea,
Branchiopoda, Cladocera), with notes on the phylogeny of the Cladocera. Journal of Morphology
256:235–259.
Olesen, J., and D. Walossek. 2000. Limb ontogeny and trunk segmentation in Nebalia species (Crustacea,
Malacostraca, Leptostraca). Zoomorphology 120:47–64.
Preuss, G. 1957. Die Muskulatur der Gliedma ßen von Phyllopoden und Anostraken. Mitteilungen des
Zoologischen Museums Berlin 33:221–256.
72 Functional Morphology and Diversity

Regier, J.C., J.W. Shultz , A.R. Ganley, A. Hussey, D. Shi, B. Ball, A. Zwick , J.E. Stajich, M.P. Cummings,
J.W. Martin, and C.W. Cunningham . 2008. Resolving arthropod phylogeny: Exploring phylogenetic
signal within 41 kb of protein-coding nuclear gene sequence. Systematic Biology 57:920–938.
Richter, S., and G. Scholtz. 2001. Phylogenetic analysis of the Malacostraca (Crustacea). Journal of
Zoological Systematics and Evolutionary Research 39:113–136.
Rudkin, D.M., G.A. Young , R.J. Elias, and E.P. Dobrzanski. 2003. The world’s biggest trilobite—
Isotelus rex new species from the Upper Ordovician of Northern Manitoba, Canada. Journal of
Paleontology 77:99–112.
Sanders, H.L. 1957. The Cephalocarida and crustacean phylogeny. Systematic Zoology 6:112–128.
Sanders, H.L. 1963. The Cephalocarida: Functional morphology, larval development, comparative exter-
nal anatomy. Memoirs of the Connecticut Academy of Arts and Sciences 15:1–80.
Schram, F.R. 1986. Crustacea . Oxford University Press, New York .
Schram, F.R. 2007. Paleozoic proto-mantis shrimp revisited. Journal of Paleontology 81:895–916.
Scourfield, D.J. 1926. On a new type of crustacean from the Old Red Sandstone (Rhynie Chert Bed,
Aberdeenshire)—Lepidocaris rhyniensis gen. et sp. nov. Philosophical Transactions of the Royal
Society of London 214:153–187.
Shu, D., X.-L. Zhang , and G. Geyer. 1995 . Anatomy and systematic affinities of the Lower Cambrian
bivalved arthropod Isoxys auritus. Alcheringa 19:333–342.
Siveter, David J. 2008. Ostracods in the Palaeozoic? Senckenbergiana lethaea 88(1):1–9.
Siveter, David J., D. Waloszek , and M. Williams. 2003. An early Cambrian phosphatocopid crusta-
cean with three-dimensionally preserved soft parts from Shropshire, England. Special Papers in
Palaeontology 70:9–30.
Siveter, Derek J., R.A. Fortey, M.D. Sutton, D.E.G. Briggs, and David J. Siveter. 2007. A Silurian “marrel-
lomorph” arthropod. Proceedings of the Royal Society of London Series B 274:2223–2229.
Snodgrass, R.E. 1958. Evolution of arthropod mechanisms. Smithsonian Miscellaneous Collections
138:1–77.
Stein, M., D. Waloszek , and A. Maas. 2005 . Oelandocaris oelandica and its significance to resolving the
stem lineage of Crustacea. Pages 55–71 in S. Koenemann and R. Vonck , editors. Crustacea and
arthropod relationships. CRC Press, Boca Raton, FL.
Stein, M., D. Waloszek , A. Maas, J.T. Haug , and K.J. Mü ller. 2008. The stem crustacean Oelandocaris
oelandica re-visited. Acta Palaeontologica Polonica 53:461–484.
Størmer, L. 1939. Studies on trilobite morphology. Part I. The thoracic appendages and their phylogenetic
significance. Norsk Geologisk Tidsskrift 19:143–273.
Taylor, R.S. 2002. A new bivalved arthropod from the Early Cambrian Sirius Passet fauna, North
Greenland. Palaeontology 45:97–123.
Torrentera, L., and S.I. Dodson. 1995 . Morphological diversity of populations of Artemia (Branchiopoda)
in Yucatá n. Journal of Crustacean Biology 15:86–102.
Walossek , D. 1999. On the Cambrian diversity of Crustacea. Pages 3–27 in F.R. Schram and J.C.
von Vaupel Klein, editors. Crustaceans and the biodiversity crisis, Proceedings of the Fourth
International Crustacean Congress, Amsterdam, The Netherlands, July 20–24, 1998, Vol. 1. Brill
Academic Publishers, Leiden, The Netherlands.
Walossek , D. 1993. The Upper Cambrian Rehbachiella and the phylogeny of Branchiopoda and Crustacea.
Fossils and Strata 32:1–202.
Walossek , D., I. Hinz-Schallreuter, J.H. Shergold, and K.J. Mü ller. 1993. Three-dimensional preservation
of arthropod integument from the Middle Cambrian of Australia. Lethaia 26:7–15.
Walossek , D., and K.J. Mü ller. 1998a . Cambrian “Orsten”-type arthropods and the phylogeny of
Crustacea. Pages 67–86 in R.A. Fortey and R.H. Thomas, editors. Arthropod relationships.
Systematics Association Special Volume Series, Vol. 55. Chapman and Hall, London .
Walossek , D., and K.J. Mü ller. 1998b. Early arthropod phylogeny in light of the Cambrian “Orsten” fossils.
Pages 185–231 in G.D. Edgecombe, editor. Arthropod fossils and phylogeny. Columbia University
Press, New York.
Walossek , D., and K.J. Mü ller. 1989. A second type A-nauplius from the Upper Cambrian “Orsten” of
Sweden. Lethaia 22:301–306.
Evolution of Crustacean Appendages 73

Walossek , D., and K.J. Mü ller. 1990. Upper Cambrian stem-lineage crustaceans and their bearing upon
the monophyletic origin of Crustacea and the position of Agnostus. Lethaia 23:409–427.
Waloszek , D. 2003a . The “Orsten” window—a three-dimensionally preserved Upper Cambrian meio-
fauna and its contribution to our understanding of the evolution of Arthropoda. Paleontological
Research 7:71–88.
Waloszek , D. 2003b. Cambrian “Orsten”-type preserved arthropods and the phylogeny of Crustacea.
Pages 66–84 in A. Legakis, S. Sfenthourakis, R. Polymeni, and M. Thessalou-Legaki, editors. The
new panorama of animal evolution. Pensoft Publishers, Sofia, Bulgaria .
Waloszek , D., J. Chen, A. Maas, and X. Wang. 2005 . Early Cambrian arthropods—new insights into
arthropod head and structural evolution. Arthropod Structure and Development 34:189–205.
Waloszek , D., A. Maas, J. Chen, and M. Stein . 2007. Evolution of cephalic feeding structures and the
phylogeny of Arthropoda. Palaeogeography Palaeoclimatology Palaeoecology 254:273–287.
Watling. L. 1993. Functional morphology of the amphipod mandible. Journal of Natural History
27:837–849.
Wehner, R., and W. Gehring. 1995 . Zoologie. Thieme, Stuttgart .
Williams, T.A. 2004 . The evolution and development of crustacean limbs: an analysis of limb homolo-
gies. Pages 169–193 in G. Scholtz , editor. Evolutionary developmental biology of Crustacea.
Crustacean Issues, Vol. 15. Balkema, Lisse, The Netherlands.
Wolff, C., and G. Scholtz . 2008. The clonal composition of biramous and uniramous arthropod limbs.
Proceedings of the Royal Society of London Series B 275:1023–1028.
Zhang , X.-g., D.J. Siveter, D. Waloszek , and A. Maas. 2007. An epipodite-bearing crown-group crusta-
cean from the Lower Cambrian. Nature 449:595–598.
3
MECHANISMS OF LIMB PATTERNING IN CRUSTACEANS

Terri A. Williams
© Watling, Les; Thiel, Martin, Nov 28, 2012, Functional Morphology and Diversity

Abstract
The structural diversity of crustacean limbs is enormous, and their evolution is hotly debated.
Attempts have been made to understand the developmental patterning mechanisms that gen-
erate distinct types of adult limbs by analyzing genes known to regulate limb development in
the arthropod model system, Drosophila. This has led to the discovery of deeply conserved
Oxford University Press, USA, Oxford, ISBN: 9780199875450

features of limb patterning, although it has not clarified how some of the most basic limb
structures—a biramous limb or endites or exites—are patterned. Indeed, based on available
data, one hypothesis is that endites and exites may have varied independently during evolu-
tion. Analyses of patterning during crustacean limb development are further complicated by
the fact that their larval stages can have limbs that are quite distinct from those of adults. This
means that any particular body segment may develop two or more structurally distinct limbs
during the course of the life cycle—a phenomenon yet to be captured by models of limb pat-
terning. Finally, the diversity of limbs is evident not only in their overall plan but also in the
details of setae, joints, and muscles that permit their functional specialization. Like many other
areas of study in crustacean limb development, the analysis of how these structures develop
has barely begun.

INTRODUCTION

The evolutionary radiation of arthropods was driven by limb diversity, and nowhere is this more
evident than in crustaceans. The structural diversity of adult limbs is enormous. Indeed, the
differences are so large that homologies are not fully resolved among some limbs (Williams
2004). The intricacy and functional breadth of crustacean limbs make them a natural choice
for analyzing how development might have been modified to produce such morphological vari-
ation. While we lack a systematic comparison of leg morphogenesis in crustaceans based on

Functional Morphology and Diversity. Edited by Les Watling and Martin Thiel.
74 © 2013 Oxford University Press. Published 2013 by Oxford University Press.
Mechanisms of Limb Patterning in Crustaceans 75

classical descriptive research, more specialized studies—for example, of cellular morphogen-


esis or molecular patterning—are emerging. The purpose of this chapter is to provide an over-
view of the genetic regulation of patterning in crustacean thoracic limb development. The focus
is on the morphological diversity of crustaceans, and the failure of models of comparative limb
development to explain that diversity is noted.
The first section is a brief overview of leg development in crustaceans, emphasizing that most
crustaceans have larval stages that include functional limbs. There are various degrees of meta-
morphosis between stages, and these include transformations of limb morphology. Thus, mod-
els of crustacean limb development and patterning should include not just the initial embryonic
limb but also the transformations of later developmental stages.
Currently, our understanding of which genes function to regulate patterning in crustacean
limbs derives directly from comparisons of development in the fruit fly Drosophila melanogaster.
Therefore, to summarize what is known about crustacean limb patterning, the next section is
an overview of leg patterning in Drosophila and other insects. This is followed by specific cases
where these leg-patterning genes have been examined in crustaceans. No general model of devel-
opmental patterning has emerged that explains adult limb diversity in crustaceans. However, in
this chapter it is hypothesized that limb patterning in crustaceans was controlled ancestrally by
a number of distinct regulatory networks that later became more or less interdependent in the
highly derived and specialized legs of Drosophila. One novel implication of this hypothesis is
that medial and lateral limb structures may have varied independently during evolution. This
hypothesis is followed by an exploration of the fact that crustaceans have different larval stages
with functionally different limbs. Explaining these phenomena requires explaining how radical
transformations of limb structure can precede the development of the adult limb. However, this
© Watling, Les; Thiel, Martin, Nov 28, 2012, Functional Morphology and Diversity

is not addressed in current models of limb development. That is, the question is not simply how
one limb is patterned but how a series of quite distinct limbs is patterned on one segment.
Most models of limb patterning explain only a basic coordinate system of positional infor-
mation, that is, proximal-distal, anterior-posterior, and dorsal-ventral axes, that map out a
generic limb field. In the final section of this chapter, limbs are fleshed out as structures made
of setae, muscle, and nerves. The development of these structures and the potential for inte-
Oxford University Press, USA, Oxford, ISBN: 9780199875450

grating them into current models of limb patterning are considered. Although crustaceans
provide the most diverse taxon for studies of limb development and evolution, that diversity
has just begun to be sampled and underrepresents the actual limb diversity in crustaceans.
This limits our ability to make inferences about which features of limb patterning are ancestral
and which patterning mechanisms are variable versus constrained. The emphasis throughout
this chapter is on the richness of structural limb diversity that remains virtually unexplored by
developmental analysis and yet is the basis for the functional limb diversity that is showcased
in this book.

A GENERAL DESCRIPTION OF LEG DEVELOPMENT IN CRUSTACEANS

Despite their varied life histories, crustaceans tend to develop in a progressive and sequential
manner. That is, body segments form in an anterior-to-posterior sequence regardless of whether
they develop in the embryonic or larval stage. Subsequently, limb buds develop as direct out-
pocketings of the ventral or ventral lateral body wall. In parallel with the sequential segment
development in crustaceans, limb development occurs in a progressive fashion, with limb buds
gaining complexity over time (Fig. 3.1).
Are there common cellular dynamics that accomplish outgrowth of the limb bud from the
ventral body wall? We know very little about this in crustaceans. In Drosophila, the arthropod
76 Functional Morphology and Diversity
© Watling, Les; Thiel, Martin, Nov 28, 2012, Functional Morphology and Diversity

Fig. 3.1.
Progressive leg development in the anostracan larva Thamnocephalus platyurus. T. platyurus has a series
of similar limbs on the trunk. As they develop, limb buds show a gradual elaboration of morphological
features, from less developed posterior limbs to more developed anterior ones (inset). Scale bar, 100 μm.
Oxford University Press, USA, Oxford, ISBN: 9780199875450

model for limb development, initial specification of the limb bud differentiates a cluster of cells
that do not undergo immediate rapid growth relative to flanking cells but are instead set aside
for later proliferation. By contrast, in crustaceans, limb buds emerge directly from the surround-
ing body wall in a manner analogous to vertebrate limbs, where we know cell dynamics play
an important role. In vertebrate limbs, the sites of future limb development are detectable as
regions of cell proliferation in the lateral flank. After the emergence of the limb bud from the
flank, a ridge of stratified epithelium, the apical ectodermal ridge, extends across the distal tip
of the limb bud and functions via signaling to maintain a higher rate of cell division in the distal
limb mesenchyme (Sun et al. 2002). However, these types of cell dynamics typically remain
undescribed for direct developing arthropod limbs. In one case where the cell dynamics of ini-
tial outgrowth was examined, Freeman et al. (1992) found that a combination of cell division
and cell shape change accounts for initial outpocketing of the limb bud from the ventral body
wall in the branchiopod Artemia. Furthermore, they found that pharmacologically arresting cell
proliferation could prevent normal evagination of the limb bud. This is a potentially fruitful but
unexplored area; characterizing the cell dynamics underlying limb outgrowth in more species
might uncover common mechanisms of outgrowth, analogous to the apical ectodermal ridge in
vertebrate limbs.
Once the limb bud differentiates from the body wall, it undergoes growth, elongation, and
segmentation. Again, in vertebrate limbs, the course of this process is well described at the cel-
lular level: differentiation proceeds in a proximal-to-distal sequence, driven by a distal zone of
Mechanisms of Limb Patterning in Crustaceans 77

proliferating cells (Sun et al. 2002). By contrast, in arthropods, the basic sequence of morpholog-
ical differentiation of limb segments is not well known for most species. What is known suggests
that arthropods do not form their leg segments in a simple proximal-to-distal sequence. Instead,
they undergo a process of intercalary growth. In crickets, for example, boundaries formed by
developing joints arise sequentially, sometimes proximal or sometimes distal to the previous
boundary (Inoue et al. 2002). It would be interesting to know whether segmented walking legs in
crustaceans share a developmental sequence in segmentation that might reflect a shared under-
lying patterning mechanism. This has never been examined systematically in crustaceans, nor
have growth and elongation in nonwalking legs, for example, pleopods or phyllopods.
How does adult morphology arise? In crustaceans, there are two rather distinct routes to
generate adult limb morphology. First, adult legs may arise directly from the gradual develop-
ment and refinement of the initial limb bud that emerges from the body wall. This happens in,
for example, peracarids and anostracans (Fig. 3.2A). In these species, limb buds emerge from
the body and gradually increase in size and complexity as they form the adult limb morphol-
ogy. However, most crustacean taxa undergo varying degrees of metamorphosis during their
life cycle (Snodgrass 1956). So, for the majority of species, adult limb morphology develops from
modifications of preexisting larval limbs. In these cases, the first limb bud that emerges from
the body wall becomes a functioning larval limb. This limb may then be modified one or more
times before assuming the adult morphology. Adult limb morphology is produced only after
limbs of quite different morphology have developed on the same segment (Fig. 3.2B). Notably,
these differences can arise from one molt to the next in quite dramatic fashion without a period
of quiescence. The types of metamorphosis are too numerous to catalogue systematically but
include the following transformations of limb morphology (typically observed in a single molt):
© Watling, Les; Thiel, Martin, Nov 28, 2012, Functional Morphology and Diversity

gain or loss of the exopod, gain or loss of limb segments, gain or loss of endites or exites, com-
plete loss of limb and later redevelopment in alternate form, and extreme hypertrophy of one
feature of the limb (for examples, see Gurney 1942, Snodgrass 1956). Although not accounted
for by models of leg development, explaining these diverse routes of development are crucial for
understanding most crustaceans.
Oxford University Press, USA, Oxford, ISBN: 9780199875450

LEG PAT TERNING IN INSECTS BASED ON A DROSOPHILA MODEL OF LEG


DEVELOPMENT

Models of arthropod leg patterning are based on leg development in Drosophila melanogaster
because it has been possible in that species to dissect the function of many genes that provide
positional information to the developing leg. These genes give cells in the developing leg their
spatial fates, for example, proximal or distal and dorsal or ventral. Although a wealth of infor-
mation has been discovered about limb patterning in Drosophila, it has been challenging to
compare this information to other arthropods. In particular, Drosophila’s metamorphic devel-
opment is highly atypical among arthropods: legs do not grow directly out of the body wall, and
indeed, Drosophila has no legs until it reaches the adult stage. However, leg patterning occurs
throughout the life cycle, beginning with specification of the leg primordium in embryogenesis,
followed by elaboration of the leg positional information during the larval stages, and finally
differentiation of leg morphology in the pupal stage.
Leg development in Drosophila begins during embryogenesis when a small group of cells on
the ventral body wall are specified to become leg cells (Cohen 1990, Cohen et al. 1993). These
10–15 cells invaginate to form a saclike structure of epithelial cells lying beneath the larval
epidermis called the leg imaginal disc . Then, during the second and third larval instars, cells
within the leg disc proliferate and form a highly folded epithelium (Fristrom and Fristrom
78 Functional Morphology and Diversity
© Watling, Les; Thiel, Martin, Nov 28, 2012, Functional Morphology and Diversity
Oxford University Press, USA, Oxford, ISBN: 9780199875450

Fig. 3.2.
(A) Progressive growth of legs on a crustacean that develops without radical metamorphosis in the notost-
racan Triops longicaudatus (successive panels represent more than a single larval molt). (B) Transformations
of a leg in a crustacean that has metamorphic development in the brachyuran Perisesarma fasciatum (after
Guerao et al. 2004); first maxilliped in first zoea, megalopa, and first crab larva.

1993). The teardrop-shaped disc contains the leg in compressed fashion, accordioned upon
itself. At metamorphosis, it everts and extends to form the adult leg (Fig. 3.3A; Fristrom and
Fristrom 1993, Taylor and Adler 2008). The constraints of metamorphosis and the consequent
highly modified leg development cause leg patterning to be analyzed in discrete phases: (1)
embryonic positioning of the primordial limb field on the body wall, (2) larval patterning of
leg axes, and (3) the late pupal transition to metamorphosis. That is, there are periods when
it is technically difficult to follow the fate of leg cells directly, for example, after invagination
but before appreciable cell proliferation and growth of the disc. This lack of simple temporal
continuity creates not only technical barriers but also conceptual difficulties when formu-
lating a general model of limb development applicable to more typical arthropods that lack
metamorphosis.
Mechanisms of Limb Patterning in Crustaceans 79
© Watling, Les; Thiel, Martin, Nov 28, 2012, Functional Morphology and Diversity
Oxford University Press, USA, Oxford, ISBN: 9780199875450

Fig. 3.3.
Establishing the proximal-distal axis in Drosophila leg development. (A) Comparison of a larval leg disc to
an adult leg showing the relative positions of proximal (P) and distal (D) leg. During eversion and elonga-
tion, the leg disc extends out of the plane of the paper to form the cylindrical structure of the adult leg (disc
after Schubiger 1971; leg traced from a wild-type adult leg). (B) Successive time points during leg develop-
ment in Drosophila showing the intercalation of additional domains of gene expression subdividing the leg
disc: Distal-less in dark gray, dachshund in medium gray, overlap of Distal-less and dachshund in light gray,
and homothorax (nuclear extradenticle) in black (after photo of expression patterns in Abu-Shaar and Mann
1998).

Early Positioning of Limb Primordia Straddles the Anterior-Posterior Boundary of


the Embryonic Segment

Leg patterning in Drosophila begins with the initial positioning and specification of the append-
age primordia on the ventral body wall of the embryo. Appendage primordia are specified rel-
ative to boundaries along the anterior-posterior (AP) and dorsal-ventral (DV) body axes. In
fact, the information used to pattern the body axes is sufficient to position the limb primordia
without other input and imparts AP positional information to the leg. This can be visualized by
boundaries of gene expression1; cells expressing the engrailed gene mark the posterior portion
of each segment and subdivide the leg disc. In Drosophila, wingless-expressing cells, which form
80 Functional Morphology and Diversity

a stripe just anterior to the engrailed-expressing cells, activate cells along the AP boundary to
form limb primordia that are visible as a cluster of cells that express Distal-less. Mutants lacking
Distal-less develop only proximal leg structures (Cohen and Jü rgens 1989).
Comparative studies show that, while the positioning of the primordia along the AP segmen-
tal boundary appears conserved in other arthropods, the actual genes that pattern the early limb
primordia—specifically, the activation of Distal-less by wingless—are not conserved even in other
insect taxa (reviewed in Angelini and Kaufman 2005). Thus, mechanisms of primordia patterning in
Drosophila are therefore unlikely to be ancestral for insects or, by extension, for the Pancrustacea.

Establishing Proximal/Distal Positional Information in the Leg

In Drosophila, fate mapping has shown that the cells that initially express Distal-less in the
embryo assume multiple appendage fates (McKay et al. 2009)—wing, leg, or Keilin’s organs
(larval sensory structures “homologous” to larval legs). As embryogenesis proceeds, this group
of cells becomes subdivided into subsequent fates: one population of cells forms the distal limb
structures in the adult (the telopod), another population forms proximal leg structures, and a
third population produces the Keilin’s organs. Thus, one of the earliest patterning events in
the leg proper is the polarization of proximal and distal leg domains: cells that form proximal
leg structures lose the initial Distal-less expression, while cells that become distal leg maintain
Distal-less expression (albeit driven by an enhancer different from that of the initial Distal-less
expression). The leg cells at this point do not divide, and only during the larval instars will cell
division resume and the bulk of patterning that defines the proximal-distal (PD) axis occur.
Because of the small number of cells in the disc and their relatively cryptic nature early
© Watling, Les; Thiel, Martin, Nov 28, 2012, Functional Morphology and Diversity

in larval development, limb patterning in Drosophila is most frequently studied from


late-second-instar larvae through third-instar larvae to pupation. During this time, the leg
disc grows by cell proliferation and assumes its characteristic folded shape within the larva.
The initial specification of proximal and distal accomplished in the embryo becomes elabo-
rated as new domains of PD genes are established and refined (reviewed in Kojima 2004). The
elaboration of PD positional information depends on the secreted signaling molecules wing-
Oxford University Press, USA, Oxford, ISBN: 9780199875450

less and decapentaplegic (reviewed in Campbell and Tomlinson 1995, Held 1995, Brook et al.
1996, Williams and Nagy 1996, Blair 1999). wingless and decapentaplegic cooperatively activate
target genes that are expressed in discrete domains along the PD axis of the limb (Distal-less,
dachshund, and indirectly extradenticle through the gene homothorax; Lecuit and Cohen 1997).
These genes are collectively termed the leg gap genes because mutations of these genes form
truncated legs with gaps in PD leg morphology.
In the second-instar imaginal discs, a boundary is established between extradenticle expres-
sion in a proximal domain and Distal-less expression in a distal domain. In the third instar, the
Distal-less domain is subsequently subdivided into a Distal-less domain and dachshund domain
(Fig. 3.3B). The boundaries of dachshund and Distal-less expression are regulated by low and
high levels of wingless and decapentaplegic signaling, respectively, and through repression
proximally by homothorax. These three primary domains—homothorax/extradenticle, dachs-
hund, and Distal-less—are maintained by mutual repression. In addition, wingless and decap-
entaplegic regulate cell proliferation (a role particularly well studied in the wing; see Posakony
et al. 1991; reviewed in Serrano and O’Farrell 1997). Similarly, the gap genes also influence leg
growth. Loss-of-function mutants in leg gap genes cause truncations of normal leg morphology
that span several morphological leg segments.
The expression and function of the leg gap genes appear conserved in other insects. Distal-
less is found in the distal part of legs (Jockusch et al. 2000, 2004, Beermann et al. 2001, Abzhanov
and Kaufman 2000, Inoue et al. 2002, Rogers et al. 2002), with dachshund and homothorax in
Mechanisms of Limb Patterning in Crustaceans 81

intermediate and proximal domains, respectively (Prpic et al. 2001, Inoue et al. 2002, Angelini
and Kaufman 2004).

Notch Signaling Plays a Key Role in Joint Formation in the Drosophila Leg

Joints within the leg disc develop during the pupal stage. They arise by invagination of the leg
epithelium, driven at least in part by cell shape changes (Mirth and Akam 2002). The epithe-
lium that forms joints can be divided into three PD positions based on gene expression and cell
behavior: proximal, mid-distal, and distal. When the joint begins to form, cells of the distal and
mid-distal region undergo apical constriction causing indentation of the leg cylinder. Then dis-
tal cells on the anterior and posterior sides become columnar so that the mid-distal tissue bends
into the leg. Finally, proximal cells extend in a proximodistal direction forming a palisade over
the indented tissue.
The complex cellular dynamics of joint formation are regulated by Notch. Notch is a
transmembrane receptor involved with numerous developmental decisions (reviewed in
Artavanis-Tsakonas et al. 1999). Notch signaling is activated by binding to its ligands, Delta and
Serrate. In the leg, Notch signaling is localized to the joints and promotes both joint formation
and leg growth. The localization of Notch signaling is regulated by the leg gap genes, which act
in a combinatorial fashion to regulate the expression of its ligands as well as the modulator fringe
(Rauskolb 2001).
In early larval stages, Notch is expressed broadly, but by the third larval instar, it is more
concentrated at the joint regions (de Celis et al. 1998). Delta and Serrate are expressed as rings in
each leg segment in the proximal joint region and signal to the more distal cells (de Celis et al.
© Watling, Les; Thiel, Martin, Nov 28, 2012, Functional Morphology and Diversity

1998, Bishop et al. 1999, Rauskolb and Irvine 1999, Rauskolb 2001). Notch is activated at the distal
margin of the joints, and this expression correlates with the distal cells known to invaginate to
form the joint (Mirth and Akam 2002). Clones of cells produced within the pupal epithelium
mutant for either Notch or Delta and Serrate show the same phenotype: if the clones are in the
joint regions, normal joints fail to form and the leg is shorter than wild type. If expressed ectopi-
cally, Notch causes supernumerary joints and leg outgrowths (de Celis et al. 1998, Bishop et al.
Oxford University Press, USA, Oxford, ISBN: 9780199875450

1999, Rauskolb and Irvine 1999). Thus, Notch not only regulates joint formation but, like the
other PD patterning genes, also regulates leg growth.
Reports from insects other than Drosophila corroborate a role for Notch signaling in leg
segmentation. Beermann et al. (2004) showed the expression of Serrate in wild-type beetle
Tribolium legs. Tc-Serrate is expressed in a series of rings corresponding to each segment of the
leg. Similarly, in analyzing the role of fringe in grasshopper body segmentation, Dearden and
Akam (2000) also described fringe expression in the leg. fringe is expressed in rings along the leg
corresponding to regions proximal to the infolding joints.

Summary of Insect Leg Patterning Based on Drosophila

Drosophila legs are initially patterned by the same coordinate system that provides positional
values within the segments. After this positioning on the body wall, the leg axis is patterned.
However, this axial patterning does not directly regulate morphologically defined structures.
For example, genes that control different PD regions of the limb do not map to specific limb
segments or branches. Instead, the coordinate system appears to be generic: it provides cells
with positional values along PD, DV, and AP axes but does not directly specify adult morphol-
ogy. Indeed, this generic feature of many patterning networks allows them to be conserved
in evolution and to pattern a range of morphologies by modifications of downstream regula-
tion. Thus, although taxon sampling is sparse, it appears that the genes that pattern the PD
82 Functional Morphology and Diversity

leg axis—Distal-less, dachshund, extradenticle —are conserved in other insects and used to
pattern legs from flies to beetles to grasshoppers. How far does this conservation extend in
crustaceans?

REVIEW OF PD PAT TERNING IN CRUSTACEAN LIMBS

Some features of the generic patterning network discovered in Drosophila are found in crusta-
ceans. This supports the idea that a generic limb-patterning network is used throughout the
arthropods, within which different limb morphologies can be specified. However, not all fea-
tures of the regulatory network described above are conserved and found in most taxa sampled.
The two main conserved features are the positioning of limb buds at an AP boundary within the
segment and the initiation and early subdivision of the PD axis. The crustacean data are repre-
sented in Table 3.1, which includes all the leg-patterning genes known from crustaceans.

Positioning of Limb Buds at the AP Segment Boundaries

Positioning of the limb bud along the AP boundary within a segment is found across crusta-
ceans (Fig. 3.4; reviewed in Williams and Nagy 2001). As in all arthropods, engrailed expression
marks the posterior part of each segment within crustaceans (e.g., Patel et al. 1989, Scholtz et al.
1993, 1994, Queinnec et al. 1999, Abzhanov and Kaufman 2000). In malacostracans, in which the
segmentation of the body and the early development of the limb bud can be followed via a pre-
dictable cell lineage, the cells that will produce the limb bud straddle the AP boundary within
© Watling, Les; Thiel, Martin, Nov 28, 2012, Functional Morphology and Diversity

the segment. Double labeling of Distal-less and engrailed in biramous limbs of malacostracans
shows that Distal-less is initiated just anterior to the engrailed boundary (Parhyale hawaiiensis,
Browne et al. 2005; Orchestia cavimana and Porcellio scaber pleopods, Hejnol and Scholtz 2004;
Thamnocephalus platyurus, T.A. Williams, unpublished observation). The signaling genes wing-
less and decapentaplegic are virtually unsampled in crustaceans. In the one case that wingless
has been examined, the phyllopodous limbs of the branchiopod Triops longicaudatus, it appears,
Oxford University Press, USA, Oxford, ISBN: 9780199875450

just like in insects, to be positioned just anterior to engrailed expression (L.M. Nagy, personal
communication).

The Initiation and Early Subdivision of the PD Axis

The most extensively sampled leg-patterning genes in crustaceans are the leg gap genes. Of
those, Distal-less has been most widely sampled (Table 3.1) and has been examined in repre-
sentative crustacean limbs of quite distinct structure: uniramous, biramous, and phyllopo-
dous. In uniramous limbs of peracarids, Distal-less is expressed on the body wall in a relatively
small cluster of cells that eventually form the tip of the outgrowing limb bud (thoracic limbs of
Porcellio scaber, Abzhanov and Kaufman 2000, Hejnol and Scholtz 2004; Orchestia cavimana,
Hejnol and Scholtz 2004). In biramous thoracic limbs, the cluster of Distal-less expressing cells
is proportionately somewhat larger but also forms an unbroken zone of expression on the body
wall. This cluster will eventually subdivide to form the endopod and exopod (Mysidopsis bahia,
Panganiban et al. 1995). This also occurs in biramous abdominal limbs: in two other peracarids,
the amphipod Orchestia cavimana and the isopod Porcellio scaber, biramous abdominal pleopods
initially develop Distal-less-expressing limb buds that subsequently subdivide to form endopod
and exopod. That is, both branches are part of the distal leg domain (Hejnol and Scholtz 2004).
Interestingly, this pattern is also found in phyllopodous limbs both in branchiopods (Williams
1998, 2008, Williams et al. 2002) and in phyllocarids (Williams 1998). In both groups, an initial
Table 3.1 Overview of limb-patterning genes examined within crustaceans.
© Watling, Les; Thiel, Martin, Nov 28, 2012, Functional Morphology and Diversity

Class/species Signaling Leg “gap” genes Leg segmentation genes “Wing” genes “Trachea” genes Body segmen-
genes tation gene
engrailed
wg dpp Dll exd hth dac N Dl nub vvl
Branchiopoda
Oxford University Press, USA, Oxford, ISBN: 9780199875450

Artemia franciscana ✓1, 2 ✓2 ✓3 ✓ 4, 5 ✓6


Thamnocephalus ✓7, 8 ✓7
platyurus
Triops longicaudatus ✓9 ✓7, 10 ✓7 ✓11 ✓11 Preliminary
Cyclestheria hislopi ✓12
Leptodora kindtii ✓12
Daphnia magna ✓13
Maxillopoda
Sacculina carcini ✓14 ✓
Malacostraca
Nebalia pugettensis ✓10
Mysdopsis bahia ✓1
Porcellio scaber ✓15, 16 ✓15 ✓15 ✓15 ✓15 ✓16
Orchestia cavimana ✓16 ✓16
Parhyale hawaiensis ✓18 ✓17 ✓17 ✓18
Pacifastacus l ✓3, 5, 19
eniusculus

Gene symbols: wg , wingless; dpp, decapentaplegic; Dll , Distal-less; exd , extradenticle; hth , homothorax; dac , dachshund; N, Notch; Dl , Delta; nub, nubbin; vvl , ventral veinless.
References: 1Panganiban et al. 1995; 2Gonzalez-Crespo and Morata 1996; 3Averof and Cohen 1997; 4Mitchell and Crews 2002; 5Franch-Marro et al. 2005 6Manzanares et al. 1996; 7 Williams et al. 2002; 8Williams
2008; 9Nulsen and Nagy 1999; 10Williams 1998; 11Sewell et al. 2008; 12Olesen et al. 2001; 13Shiga et al. 2002; 14Mouchel-Vielh et al. 1998; 15Abzhanov and Kaufman 2000; 16Hejnol and Scholtz 2004; 17 Prpic and
Telford 2008; 18Browne et al. 2005; 19Damen et al. 2002.
84 Functional Morphology and Diversity
© Watling, Les; Thiel, Martin, Nov 28, 2012, Functional Morphology and Diversity

Fig. 3.4.
The size of the limb primordia is highly variable in crustaceans: schematic views of ventral segments in
different crustaceans illustrating the relative amount of ventral segment fated to become limbs. In all
cases, primordia retain their anterior-posterior positioning regardless of size (see text for examples).
Oxford University Press, USA, Oxford, ISBN: 9780199875450

domain of Distal-less expression forms on the ventrolateral body wall. Like the biramous domain,
it eventually subdivides to form the endopod and exopod. (However, subsequent Distal-less
expression is more widely distributed; see below.) Thus, throughout crustaceans—as through-
out all arthropods sampled—Distal-less expression is consistent with the role of patterning the
distal region of the leg and promoting PD outgrowth.
For the other leg gap genes, extradenticle and dachshund, published data exist only for
uniramous and phyllopodous limbs. Where they have been examined in crustaceans, extraden-
ticle is expressed in a proximal domain complementing and exclusive of the Distal-less domain,
indicating a conserved role for the early stages of limb patterning.2 In uniramous limbs, extraden-
ticle is expressed in the first two limb segments (Porcellio scaber, Abzhanov and Kaufman 2000).
In early phyllopodous limb bud, extradenticle is expressed in the limb bud outside the Distal-
less domain, but later, when Distal-less is expressed more proximally, extradenticle expression is
maintained so that the two are coexpressed (Triops longicaudatus and Thamnocephalus platyurus,
Williams et al. 2002). Like extradenticle, dachshund expression in uniramous crustacean limbs
is similar to that in Drosophila. It arises between the distal Distal-less and proximal extradenticle
domain (Porcellio scaber, Abzhanov and Kaufman 2000). In phyllopodous limbs, one domain
of initial dachshund expression is just proximal to the initial Distal-less expression, as might be
expected (Sewell et al. 2008). However, the expression is dynamic and becomes more complex
Mechanisms of Limb Patterning in Crustaceans 85

as the limbs develop (see below). Taken as a whole, the patterns of expression in leg gap genes in
crustaceans are similar to Drosophila: initially, proximal and distal domains are established, and
subsequently, new domains are intercalated that can provide more precise PD information.

Notch in Crustaceans

Although there are no published reports of expression of Notch pathway genes in crustaceans,
there are preliminary data that Notch protein is expressed in reiterated stripes in developing
endites in branchiopods and functions in proper endite formation (T.A. Williams, unpub-
lished observation). Notch is expressed differentially in the limb, in the medially repeated lobes
(endites) but not the lateral lobes. DAPT, a gamma secretase inhibitor that blocks Notch sign-
aling, can block formation of endites in branchiopod legs. In the most extreme phenotype, all
endites (except the most proximal one) fuse to form a single unbranched lobe. Again, the effect
within the limb is not uniform, fusing medial lobes (endites) but not lateral lobes.

Summary

All crustacean limbs currently sampled show conservation in AP limb bud positioning and
initial PD patterning. In a sense, it is remarkable that a generic PD coordinate system is used
throughout arthropods given their limb diversity. At the same time, the conservation of pattern-
ing mechanisms in the face of great morphological diversity makes PD patterning genes surpris-
ingly unhelpful in explaining how limb morphology may have evolved; known mechanisms do
not specify whether the limbs will be fundamentally uni- or biramous, for example, nor do they
© Watling, Les; Thiel, Martin, Nov 28, 2012, Functional Morphology and Diversity

pattern the medial and lateral lobes. In short, patterning genes known from Drosophila cannot
as yet explain any of the differences so prevalent in crustacean limbs. In particular, the fact that
crustaceans have legs quite distinct from uniramous walking legs gives rise to questions about
how biramous or phyllopodous legs are patterned.
Oxford University Press, USA, Oxford, ISBN: 9780199875450

How Are the Two Primary Limb Branches Patterned?

Patterning the primary limb branches is a fundamental and yet unanswered question in crus-
taceans. Everything we have learned to date points to a single PD patterning axis with a single
domain of Distal-less from which both the endopod and exopod arise. That is, there is no evi-
dence to indicate that the formation of the endopod and exopod within the distal domain is
controlled by genes known from Drosophila leg patterning. More generally, outgrowths from
the limbs are not patterned as new PD growth axes (reviewed in Williams 1998, Nagy and
Williams 2001, Williams and Nagy 2001). However, patterning of branches from the main axis
is particularly important regarding the formation of the endopod and exopod because biramous
limbs are both ancestral and widespread among crustaceans. While loss of Distal-less function
in crustaceans should produce loss of both the endopod and exopod, we do not know which
genes would prevent splitting while permitting outgrowth. In addition, although we know that
limb branches are not patterned as simply reiterated PD growth axes, we do not know to what
degree the bifurcation of the endopod and exopod is distinct from the regulatory mechanisms
that form other limb outgrowths (exites and endites).
Phenomenologically, the formation of the two branches can vary among species. Fig. 3.5
shows some scenarios for morphogenesis in biramous limbs: both limb branches might arise
simultaneously from the body wall, or limb branches might form from a subdivision of an already
elongating limb bud, or a second branch might arise late in morphogenesis as an outgrowth from
the main branch. There is some evidence that all three modes are used in crustaceans.
86 Functional Morphology and Diversity
© Watling, Les; Thiel, Martin, Nov 28, 2012, Functional Morphology and Diversity
Oxford University Press, USA, Oxford, ISBN: 9780199875450

Fig. 3.5.
Schematic of possible types of limb morphogenesis leading to biramous limbs. (A) The limb bud arises
from the body wall bilobed and continues to extend both branches. (B) The limb bud arises unbranched
from the body wall and only after some growth subdivides and forms two branches. (C) The limb bud
arises unbranched from the body wall, grows, and then forms a small bud that subsequently grows as the
second branch.

What controls patterning in biramous crustacean limbs is unknown, and there are no
cross-species comparisons of the development of biramous limbs. However, there have been
comparisons within peracarid species of the development of biramous abdominal limbs to their
uniramous thoracic counterparts. These comparisons depend on the fact that peracarids estab-
lish their segments (and early limb buds) via a repeatable cell lineage. Therefore, it is straight-
forward to compare the fates of cells that formed the exopod in the biramous abdominal limbs
to the same cells on the thorax. In the amphipod Orchestia cavimani, Hejnol and Scholtz (2004)
found the cells that are homologous to abdominal exopod cells indeed express Distal-less early
on but then subsequently lose Distal-less expression. This suggests that these cells still have an
early specification to form an exopod—or at least distal leg structures—but that that specifica-
tion is lost during subsequent development. By contrast, in the isopod Porcellio scaber, cells in
the thorax that are homologous to abdominal exopod cells never express Distal-less (Hejnol and
Scholtz 2004). This suggests that these cells never acquire a distal limb fate at all. Thus, in taxa
Mechanisms of Limb Patterning in Crustaceans 87

that independently evolved uniramous from biramous limbs, the mechanisms controlling sup-
pression of the exopod in the thorax appear distinct.
In related studies, Wolff and Scholtz (2008) marked cells in the early uniramous thoracic
limb bud and the biramous abdominal limb bud in the amphipod Orchestia cavimani, in order
to follow the subsequent fate of those cells as the embryos grew. Before limb outgrowth, body
segments contain rows of aligned cells approximately 10 cells wide. In the abdomen, 6 of 10 cells
contribute to the limb bud and do so without much change of register; for example, the medial
cells form the endopod, and the lateral cells form the exopod. In the thorax, cells also grow out
in register, although fewer cells of the row form the leg. Notably, the cells that form the exopod
in pleopods form the lateral part of the walking thoracic leg. This suggests that, in these amphi-
pods, the formation of an unbranched limb evolved from the suppression of the mechanism that
would split the initial domain into endopod and exopod.
These data are very interesting and, in the absence of genetic manipulation, allow some spec-
ulation on how a single axis forms two axes. Of course, in these cases, one of two branches is lost
during evolution, leaving unanswered the original question—how two branches are patterned
from a single PD domain. It is also unclear how much these results can be generalized. The line-
age results that lead to the conclusion that there is a suppression of the split into a biramous limb
are derived from Orchestia cavimani (Wolff and Scholtz 2008). However, the earlier comparison
of Distal-less expression within Orchestia cavimani and Porcellio scaber suggests that, even within
the peracarids, exopod loss occurs via different mechanisms (Hejnol and Scholtz 2004). Also,
malacostracans are unique in having teloblastic growth and cell-lineage-based formation of the
germband and limb buds. Whether their mechanisms of limb patterning are similarly derived
remains to be seen.
© Watling, Les; Thiel, Martin, Nov 28, 2012, Functional Morphology and Diversity

Common Aspects of Patterning in Phyllopodous Limbs of Branchiopods

Phyllopodous limbs are even more distinct than biramous ones from the kind of uniramous
walking leg represented in most models of arthropod limb patterning. Despite the previously
mentioned conservation of some aspects of limb patterning, phyllopodous limb buds have com-
Oxford University Press, USA, Oxford, ISBN: 9780199875450

plicated temporal and spatial expression of the PD leg-patterning genes. In addition, both bran-
chiopods and phyllocarids have large thoracic limb buds that occupy virtually the entire ventral
to ventrolateral body wall. For example, in Artemia the limb bud occupies six of the eight precur-
sor rows that form the AP segmental anlage, whereas the interleg region is only two of eight rows
(Freeman et al. 1992). This contrasts with peracarids, where the initial limb buds occupy less
than half the segment (e.g., Dohle and Scholtz 1988, Scholtz 1990). These differences in relative
size of the limb anlagen, as well as differences in the geometry of the developing limbs, argue
against hypotheses of straightforward conservation of limb patterning since such differences
would certainly influence the action of, for example, short-range signaling molecules.
Additional arguments arise simply from the complex patterns of gene expression found in
phyllopodous limbs. As mentioned above, expression of the leg signaling gene wingless has been
examined in the branchiopod Triops longicaudatus and is expressed initially in segmental stripes
along the body anterior to the engrailed stripes. However, as the limbs develop on the segments,
each wingless stripe breaks up, eventually coming to occupy only a portion of each lobe of the
phyllopod (Nulsen and Nagy 1999). The wingless expression varies by lobe: on the most medial
lobe (the gnathobase), expression is on the lateral margin; on the other endites as well as the
endopod and exopod, the expression is on the medial margin; and on the epipod, the expression
outlines the entire lobe. Similarly, expression of Distal-less and extradenticle in branchiopods is
initially comparable to Drosophila: Distal-less is expressed in the distal limb, in both the devel-
oping endopod and exopod, exclusive of and surrounded by extradenticle expression. However,
88 Functional Morphology and Diversity

wingless

dachshund

Distal-less

epi
gn

exo E2

E3
endo
E5 E4

Fig. 3.6.
Schematic of gene expression in the early limb bud of the branchiopod Triops longicaudatus. With the
exception of the gnathobase, the endites show identical patterns of expression. All other lobes are unique.
Abbreviations: E2–E5, endites 2–5; endo, endopod; epi, epipod; exo, exopod; gn, gnathobase.
© Watling, Les; Thiel, Martin, Nov 28, 2012, Functional Morphology and Diversity

as development proceeds, Distal-less expression spreads into the more proximal lobes—with
particularly extensive expression in the endites—and this later expression of Distal-less overlaps
with extradenticle. When this later Distal-less expression is traced through limb development, it
Oxford University Press, USA, Oxford, ISBN: 9780199875450

coincides exactly with seta-forming cells of the limb (Williams 2008). Early expression of dachs-
hund in the Triops limb bud is consistent with an intermediate dachshund region between the
Distal-less- only expression domain and the extradenticle domain of the proximal leg. However,
expression occupies only a small region of cells in a large limb bud, and the topology that
would correspond to a uniramous limb can be only roughly inferred. In addition, dachshund is
expressed in a proximal stripe and, later, in a series of medially reiterated stripes in each endite
(Sewell et al. 2008). Indeed, one striking feature of many limb-patterning genes in phyllopodous
limbs is that they show reiterated domains of expression in the endites. In Triops, the branchio-
pod that has been examined most extensively, the endites between the gnathobase and endopod
show identical expression patterns for wingless, Distal-less, extradenticle, dachshund, and Notch
(Fig. 3.6). This is true in another branchiopod, Thamnocephalus platyurus, although the dachs-
hund and Notch data are preliminary. Below I argue that a medial patterning system controls
medially reiterated structures in crustacean limbs.

Summary

In Drosophila a generic coordinate system provides positional information for developing limbs,
and aspects of this generic system appear conserved even in limbs of highly different morphol-
ogy, that is, initial PD patterning in phyllopodous limbs. However, this fact alone highlights how
little these positional coordinates inform us about morphology later in development: phyllopo-
dous limbs are so structurally different from uniramous limbs that sharing the same axial PD
Mechanisms of Limb Patterning in Crustaceans 89

patterning must relegate that patterning to only the most indirect control of adult morphology.
Within this context, I propose a modified view of limb patterning in crustaceans and describe
how that view accounts for some patterns of diversity.

LIMITS OF PD PAT TERNING MODEL IN EXPLAINING THE DIVERSIT Y OF LIMB


MORPHOLOGIES AND AN ALTERNATIVE VIEW OF LIMB PAT TERNING

One point emerges clearly from the analysis of Drosophila leg-patterning genes in crustaceans:
interactions of those genes alone cannot provide a satisfactory model to explain the diversity of
adult limb morphologies in crustaceans. One particular point is worth reiterating since it is cen-
tral to crustacean limb morphology. We do not know how the two main branches of a biramous
limb are patterned. This is important because the biramous limb is most likely ancestral (see
chapter 2). Beyond that, biramous thoracic limbs are very common among crustaceans, although
they happen not to occur in the few crustaceans typically used for studies of limb patterning.
Beyond the question of biramous limbs, crustaceans have a variety of endites and exites not
accounted for by the model based on patterning in Drosophila. For example, Boxshall (2004)
distinguishes as many as 9–10 distinct nonhomologous exites among crustaceans. In addition,
on medial limb margins, most anterior feeding thoracic limbs have well-developed endites used
for sorting and breaking up food. In branchiopods, the phyllopodous limbs are defined by their
well-developed endites and exites. These patterns among crustacean limbs lead to an alternative
view of crustacean limb patterning based on certain contrasts between Drosophila and crusta-
ceans (Fig. 3.7).
© Watling, Les; Thiel, Martin, Nov 28, 2012, Functional Morphology and Diversity

Model for Ancestral Limb Patterning

First, the amount of ventral and ventrolateral body wall devoted to the limb bud and the timing
of adult limb development vary. In Drosophila, the leg develops from a very small set of cells
patterned in the embryo that then invaginate and develop segregated from the larval epidermis
Oxford University Press, USA, Oxford, ISBN: 9780199875450

(see above). Through further proliferation and patterning, these cells produce the adult leg mor-
phology. By contrast, no crustaceans set aside part of the body wall epidermis for limb forma-
tion. Limbs develop directly from the body wall in coordination with adjacent tissue. In some
cases the amount of body wall devoted to the adult limb is relatively small, but in some cases it
is not. In branchiopods, virtually the entire ventrolateral body wall becomes limb. This is likely
to represent the ancestral case for pancrustaceans since stem lineage larval forms appear to have
developing limbs that are large relative to the body wall. In parallel with these differences in the
timing and extent of ventral body wall involvement in leg development, crustaceans show an
enormous variety of limb morphologies compared to the single unbranched leg in Drosophila.
Laterally, exite morphology is highly variable. Medial limb morphology is diverse as well but
often is organized around a series of medially reiterated structures: endites, leg segments, or
some combination of those.
These considerations lead me to develop a hypothetical model for how limb-patterning
genes were deployed ancestrally in Pancrustacea and modified in extant taxa (Fig. 3.7).
I hypothesize that, ancestrally, the limb-patterning field was broad and occupied a large area
of the ventral to lateral body wall. Within that patterning field, a number of patterning net-
works operated but were only loosely coupled with one another. An overall PD patterning axis
promoted limb outgrowth and controlled the formation of both the endopod and exopod. This
is supported by the fact that all crustaceans have limb buds with a region of exclusive Distal-
less expression that gives rise to the exopod and endopod. In concert, a medial patterning
90 Functional Morphology and Diversity

Drosophila crustaceans

field of limb patterining


on body wall
segregation early in development continuous functional integrity
from the rest of the ventral of the ventral body wall
body wall

adult limb morphology unbranched leg high diversity of morphologies,


e.g., biramous, phyllopodous

HYPOTHESIS

P
L P N L N

limb pattering D
systems D
PD and N patterning are strictly Patterning axes are only loosely
superimposed; lateral genes (used coupled. PD patterns main axis of leg
to specify the wing) maintain the (1 or 2 branches); N patterns medially
same topological relationship but reiterated structures (lobes or
are not in the limb field. segments); L patterns lateral limb lobes.
© Watling, Les; Thiel, Martin, Nov 28, 2012, Functional Morphology and Diversity

Fits rapid life cycle highly Flexibility of patterning permits high


specialized for ecological niche limb diversity

consequences
Oxford University Press, USA, Oxford, ISBN: 9780199875450

DV homologies are comparable Medial/lateral homologies can vary


along PD axis independently along PD axis

Fig. 3.7.
Comparison of a number of features of Drosophila leg development and crustacean leg development illus-
trating the hypothesis that patterning systems now coupled in Drosophila may have been only loosely
coupled ancestrally. The rectangles represent a half segment, for example, from the midline outward. Gray
shading represents the size of the early limb primordia relative to that half segment. N, L, and PD repre-
sent gene patterning networks: Notch, lateral, and proximodistal, respectively. DV, dorsal-ventral axis.

system based on Notch signaling produced repeated structures. Those structures could be
either endites or segments. This is supported by preliminary data of Notch expression and
function in branchiopods. Laterally, another set of genes (nubbin, apterous) operated to pro-
duce exites (Averof and Cohen 1997). These sets of loosely coupled patterning networks would
have high f lexibility to produce the various limbs found in extant groups. In Drosophila , those
loosely coupled networks are now tightly coupled: the PD and Notch networks are linked, and
Mechanisms of Limb Patterning in Crustaceans 91

the lateral genes are expressed in cells that undergo an early migration away from the rest of
the limb anlagen. In uniramous limbs, Notch regulates leg segments; in the ancestral case, it
regulated any medially repeated structures. In the derived, uniramous limb of Drosophila ,
Notch and PD patterning are linked, the limb bud is a reduced part of the body wall, and much
more of the limb bud is restricted to the initial distal region demarcated by the PD patterning
system. This contrasts with the much more extensive limb buds found in branchiopods and
in stem group crustaceans.

Consequences and Predictions

This hypothesis suggests a rethinking of how development of limbs may have evolved and forces
a somewhat atypical view of the developmental basis of limb homologies. In particular, it pre-
dicts that homologs can vary independently on the medial versus lateral margins along the PD
axis; that is, exite morphology varies independently from endite morphology. This is consistent
with the fact that lateral limb branches are highly variable (Boxshall 2004). This hypothesis
implies that any particular slice along the PD leg axis cannot be considered as an integrated
homolog the way we conceive of a PD series of limb elements in the vertebrate limb. It also
implies that Notch regulates all medially repeated structures, that within any limb medial struc-
tures are serial homologs, and that between limbs certain dissimilar medial structures, lobes,
and segments are direct homologs. (Despite these elements being direct homologs, it may not be
possible to draw one-to-one homologies between elements, a common problem with reiterated
serial homologs [Bateson 1894, Wagner 1989, Van Valen 1993].) This hypothesis also leads to
specific predictions in taxa outside of crustaceans. For example, in insects, the gnathal append-
© Watling, Les; Thiel, Martin, Nov 28, 2012, Functional Morphology and Diversity

ages have repeated medial lobes (endites). I predict that these lobes will be regulated by Notch
signaling as are leg joints. This has not been examined in Drosophila because of their highly
reduced mouthparts.

THE IMPACT OF DIVERSE LARVAL STAGES ON MODEL S OF LIMB DEVELOPMENT


Oxford University Press, USA, Oxford, ISBN: 9780199875450

In the preceding sections, the focus was on adult thoracic limb diversity from a standard per-
spective in evolutionary developmental biology: how is development modified to produce one
adult leg morphology versus another? Although this is normally how comparisons between
species are framed, this narrow focus is justified only in the case of crustaceans that develop
their adult limbs directly. However, the majority of crustacean taxa undergo some metamor-
phosis. In these cases, a particular thoracic segment can, for example, produce a series of limbs
with distinct morphologies during successive molts. As described above, adult limb morphol-
ogy may be achieved only after limbs of quite different morphology have developed on the
same segment.
Why is this significant? Consider the differences in limb development in direct versus indi-
rect developers that are illustrated schematically in Fig. 3.8A. For direct developers, each species
produces its own particular adult morphology in a single developmental pathway from primor-
dium to adult limb. This type of life history lends itself to the question commonly posed by cur-
rent research: How do different morphologies develop? By contrast, crustaceans with distinct
larval stages develop the adult limb morphology via a sequence of potentially distinct develop-
mental pathways. In these cases, patterning of the adult limb is built upon prior functional limbs
that are transformed in a variety of ways. For example, development to the adult limb may occur
gradually or as a marked modification of a preexisting larval limb, or even as a wholesale replace-
ment after a larval limb is lost.
92 Functional Morphology and Diversity
© Watling, Les; Thiel, Martin, Nov 28, 2012, Functional Morphology and Diversity

Fig. 3.8.
Oxford University Press, USA, Oxford, ISBN: 9780199875450

Conceptual differences in modeling limb development in direct developers versus those with multiple
functional larval stages, as are found in many crustaceans. (A) For direct developers, rows represent a
single thoracic segment in each of three different species. The adult thoracic limb morphology is differ-
ent in each species (as represented by different shading patterns). The typical question posed by current
research is how these different morphologies develop. However, in crustaceans with functional larvae, the
three rows of different species have more complex pathways to reach the adult thoracic limb morphology.
Larval stages are represented in which limbs may be only slightly different from the adult (row 1), quite
distinct from the adult (row 2), and even be lost (no color) and regained (row 3). (B) The impact of multiple
functional larval stages is similar for considerations of patterning segment identity along the body axis. For
direct developers, rows 1–3 represent three different species with different patterns of functional clusters
of limbs (represented by different shading patterns along a series of segments). The typical question posed
by current research is how these different patterns of segment identity develop. However, in crustaceans
with functional larvae, the developmental pathway leading to the adult pattern is more complex since func-
tional larvae can have distinct clusters of limb identities. In both cases, these changes in morphology over
time need to be accounted for in models of limb development.

The series of regulatory control mechanisms that might pattern such diverse routes to
adult limb morphology is wholly unaccounted for in models of limb patterning. At the least,
analysis of adult limb patterning should include every larval stage and the expression and
function of genes in each stage. As yet, patterning of different limb morphologies found in a
complex larval sequence remains unexplored. An important and related question is whether,
on any one particular segment, the diversity in limb structure is somehow constrained during
Mechanisms of Limb Patterning in Crustaceans 93

development. That is, are only some morphological transformations possible from one molt
to the next?
These differences between direct and indirect developers also influence our models of
how segment identity develops. Comparative data exist for the patterning of segment identi-
ties that define body tagmata (reviewed in Deutsch and Mouchel-Vielh 2003). Genetic regula-
tion of different body regions is provided by homeotic (Hox) genes, which show differential
expression and control segment identity along the AP body axis. Hox genes are found in all
the major crustacean taxa, and differences in their expression control the fate of anterior tho-
racic segments, for example, determining what segments bear maxillipeds (Averof and Patel
1997, Pavlopoulos et al. 2009). However, the same phenomena that complicate the patterning of
adult limb morphology on any one segment can obviously take place anywhere along the body
axis. Therefore, the patterns of segment identity within tagmata can change during develop-
ment (schematized in Fig. 3.8B). Whereas the standard question is to ask how the tagmata in
species A are patterned differently from the tagmata in species B, for most crustaceans larval
stages with different groupings of limb morphologies present a more complex sequence of tag-
mata that presumably would require differential regulation at each stage. The presence of larval
forms implies that groupings of similar segments seen in the adult may not have been stable
throughout the life cycle. In larval forms, patterns of segment identity are much more complex
since larval limbs can be structurally distinct from adults. I do not suggest that the underlying
tagmata are not stable, merely that the clustering of similar limbs—that is, the apparent seg-
ment identity—can change during development thus complicating our models of how segment
identity is specified.
Despite the absence of patterning data that bear on these questions, I discuss and illustrate
© Watling, Les; Thiel, Martin, Nov 28, 2012, Functional Morphology and Diversity

them at length because the sheer variability in limb morphology and life history patterns within
crustaceans is generally not incorporated into models of arthropod limb patterning. This diver-
sity is key to understanding the radiation of crustaceans, but our models do not account for it.

ANALYZING THE DEVELOPMENT AND DIFFERENTIATION OF LIMB STRUCTURES


Oxford University Press, USA, Oxford, ISBN: 9780199875450

Models of limb patterning mainly explain how spatial pattern arises during development,
typically through the formation of Cartesian axes: proximal-distal, anterior-posterior, and
dorsal-ventral. In these models, actual limb structures are a proxy for position; for example,
claws at the limb tip become most distal positional values. Although this has been a very use-
ful perspective in illuminating how certain generic regulatory pathways can be used in differ-
ent situations, it tends to obscure morphological structures. After all, it is specific structures
that function in the particular habits and life history of each species. Considered in their own
right, the structures that compose the leg display a number of intriguing developmental fea-
tures. For example, some setae show regularities in positioning that transcend species, such
as their position adjacent to joints. Also, many of the genes that provide axial positional infor-
mation in the leg also participate directly in patterning morphological structures. It seems
quite plausible that patterning of leg axes and morphological structures is more integrated
than expected; we are slightly misled by Drosophila , where the timing of patterning and dif-
ferentiation is atypical. In most crustaceans, patterning and differentiation are, at least phe-
nomenologically, quite closely integrated—giving rise to the question of whether they are
mechanistically integrated as well. If this is accurate, understanding the development of limb
structures could elucidate novel morphological subunits within limbs. For example, fields of
setae or setal joint complexes that are linked developmentally would constrain the possibili-
ties of limb evolution.
94 Functional Morphology and Diversity

In addition to exploring this hypothesis of novel morphological subunits, a more straight-


forward and pragmatic reason exists for studying descriptive limb morphogenesis. To interpret
patterns of gene expression or phenotypes produced by perturbations of gene function, we must
know the details of tissue morphogenesis. As the model system, Drosophila has had much of
the descriptive groundwork done: the timing of differentiation is known, and cell types are
often well characterized. By contrast, that groundwork is lacking in crustaceans. However, to
correctly interpret details of gene expression and function, we need to know what tissues these
differentiating cells will become. This is particularly important because many patterning genes
have multiple functions, and dissecting out those functions requires knowing the fate of cells
expressing the gene.
To consider whether developmentally novel subunits exist in crustaceans, I discuss devel-
opment and patterning of muscles and setae within crustacean limbs. I again use Drosophila
as a point of reference, although it is important to note that the presence of the pupal stage has
obscured analysis that might integrate aspects of limb patterning with setal and muscle develop-
ment in that species. Various aspects of limb development in Drosophila have been compressed
to fit the demands of radical metamorphosis. In particular, the timing of differentiation into
the adult relative to patterning of the leg is not characteristic of most crustaceans. This is best
grasped schematically, as in Fig. 3.9, where, if we compare events along a normalized timeline,
it is obvious that the lack of larval limbs and the radical metamorphosis in Drosophila create
an atypical distribution of the timing of patterning versus differentiation compared to most
crustaceans—indeed, most arthropods. Cell differentiation occurs along with ongoing posi-
tional patterning in crustaceans, as illustrated by the development of the limbs in branchiopod
crustaceans. In branchiopods, the earliest differentiation of the limb bud from the body wall
© Watling, Les; Thiel, Martin, Nov 28, 2012, Functional Morphology and Diversity

is followed by differentiation of both setae and muscles (Williams and Muller 1996, Williams
2007a). The possible integration of patterning and differentiation remains an open question in
crustaceans.

LIMB DEVELOPMENT
Oxford University Press, USA, Oxford, ISBN: 9780199875450

Drosophila embryo larva pupa adult

PATTERNING

DIFFERENTIATION

crustacean embryo larva larva juvenile adult

PATTERNING

DIFFERENTIATION

Fig. 3.9.
A schematic timeline of limb development showing the relative timing of patterning and differentiation in
Drosophila and crustaceans. In crustaceans, as is likely ancestral, patterning and differentiation are highly
overlapping during development. In Drosophila, the constraints of a radical metamorphosis segregate the
two events to a marked degree.
Mechanisms of Limb Patterning in Crustaceans 95

Development of Muscle in Crustacean Limbs

Muscle in Drosophila and other insects arise from precursor cells that differentiate into
muscle founders. These founder cells act as positional templates, attracting other myotubes
to fuse with them to form functional muscle. However, in Drosophila , a radical metamor-
phosis divides muscle formation into two phases: larval and adult. Larval muscles develop
directly during embryogenesis. Some adult muscle precursors are specified during embryo-
genesis, but they become fully developed only during the pupal metamorphosis (Roy and
VijayRaghavan 1999).
In Drosophila legs, muscles originate from a population of precursors (5–10 myoblasts)
associated with the embryonic leg disc primordia. These cells proliferate during the larval
period to form about 500 myoblasts associated with the disc epithelium. During late larval
and early pupal development, some of these myoblasts begin expressing dumbfounded , an
immunoglobulin that attracts other myotubes to fuse with the founder cells and thus serves
as a marker of founders. As pupation proceeds, internal tendons form closely associated with
founder cells, myotubes fuse with the founders, and attachment sites form on the leg epithe-
lium (Soler et al. 2004).
How do crustacean muscles develop? This question is largely unexplored, although one
recent paper offers the first evidence that founder cells are present in crustaceans as well as
insects. Kreissl et al. (2008) generated a monoclonal antibody against heavy-chain myosin in
isopod crustaceans and then traced myosin-expressing cells during development. They found
that muscle development proceeded in a fashion similar to that in insects. Initially, only single
cells expressed the antigen, and these were arranged in a pattern that was a precursor to the adult
© Watling, Les; Thiel, Martin, Nov 28, 2012, Functional Morphology and Diversity

pattern. Subsequently, these cells became multinucleate syncytia, although it was not possible
to distinguish whether this occurred via fusion (Kreissl et al. 2008).

Development of Setae in Crustacean Limbs


Oxford University Press, USA, Oxford, ISBN: 9780199875450

Crustacean setae are enormously diverse (as described in chapter 6). If external morphology,
accessible through light and scanning electron microscopy studies, remains underdescribed,
the internal morphology and development of setae are even less explored. Although setae are
assumed to be homologous among arthropods, we have no general model for noninsect setae
for comparison. In insects, sensilla that form bristles (setae) and other sense organs develop
through a well-described lineage mechanism from a single precursor epidermal cell (reviewed
in Hartenstein 2005). Stereotyped divisions of the precursor cell produce both sensory neurons
and accessory cells that ensheath the neuron or form cuticular outgrowths. Modifications of
this lineage via apoptosis and/or repeated divisions in certain lineages can produce all the sen-
silla types found in insects (Lai and Orgogozo 2004), including setae without sensory function
(via apoptosis of the lineage leading to the neuron). Furthermore, whereas insect sensilla can
be grouped into two or three main categories based on adult morphology (Hartenstein 2005),
crustacean sensilla appear to have a much broader range of underlying cell numbers (Hallberg
and Hansson 1999).
How do crustacean setae develop? To my knowledge, only two studies describe events of the
cellular morphogenesis of crustacean setae. Based on transmission electron microscopy, Guse
(1983) described the development of the aesthetascs, tubular sensory setae, on the antennule of
the mysid Neomysis integer. The aesthetascs have 60–80 sensory cells and eight accessory cells
that ensheath the neurons. Sensory cells are differentiated before the molt, during which the
shaft is formed. The shaft begins to develop during apolysis with a retraction of the epidermis
from the old cuticle. The shaft develops by forming a cylindrical invagination of the epidermis
96 Functional Morphology and Diversity

with the midpoint of the shaft being the deepest part of the invagination. The eight ensheathing
cells that secrete the shaft cuticle are arranged in a telescoping manner from proximal to distal
such that the cuticle of the distal shaft is laid down by one ensheathing cells. with more proximal
ensheathing cells laying down more proximal cuticle.
In the second study, setal morphogenesis was described for the anostracan Thamnocephalus
platyurus using light microscopy: differentiating setal cells could be seen within the develop-
ing thoracic limb bud (Williams 2007b). The thoracic limbs in T. platyurus are highly setose,
although almost none of the setae are sensory. The only sensory setae present are mechanore-
ceptors on each endite. In parallel to the anamorphic development of the larva in anostracans,
limb buds and setae on the limb buds develop gradually. Even small limb buds just emerging
from the body wall form the lobes of the adult limb and begin to develop setae with very small
shafts. The cellular composition of the nonsensory setae is uniform: all have six accessory cells,
discernable as three pairs whose size and positions are distinct. These cells are clearly identified
in the developing limb bud by the hypertrophy and alignment of their nuclei. Although the pairs
of accessory cells have distinct nuclei and positions relative to the setae, their functions remain
unknown.
Even these two studies show that very different numbers of cells are involved in setal produc-
tion. Neither illuminate the lineage of setal forming cells. Indeed, there are no lineage-tracing
studies of cells that generate crustacean setae, and we do not know whether a lineage-based
development comparable to insects occurs. Comparisons of setal development within crusta-
ceans would be very useful in their own right, and they would help address some of the bigger
questions of the evolution of setae in arthropods. For example, although nonsensory bristles in
insects can be derived from sensory ones by apoptosis of the neuronal lineage, is there reason to
© Watling, Les; Thiel, Martin, Nov 28, 2012, Functional Morphology and Diversity

suppose that sensory setae actually evolved first? That is, might not there have been a selective
pressure for structural setae just as there was for sensory accessory cells? In general, the origins
of arthropod setae remain unexplored.

Genes Important in Limb Patterning Also Pattern Setae


Oxford University Press, USA, Oxford, ISBN: 9780199875450

Another reason to consider the development of crustacean setae is that a number of leg-patterning
genes contribute to patterning bristles or sense organs. This is not surprising since many pat-
terning genes have multiple functions. Nonetheless, it is interesting to consider whether dual
patterning might be a consequence, at least ancestrally, of coordination within the leg of axial
patterning and setae. For example, two of the main leg “gap” genes in Drosophila—Distal-less
and dachshund—additionally play a role of regulating sensory structures. Distal-less plays a
role in the development of certain sensory structures in Drosophila. Distal-less-negative clones
induced late in development during the third larval period in Drosophila are often associated
with anomalous bristle morphology (although it is not reported that cells specifically express
Distal-less; Campbell and Tomlinson 1998). Distal-less is required in cells that form bracts, which
are cuticular elaborations adjacent to the mechanosensory bristles of the leg (Held 2002). In
addition, Distal-less specifies Keilin’s organs, the larval leg sense organs that consist of three
bristlelike external sensilla (Cohen and Jü rgens 1989). Indeed, the earliest ideas of the multi-
functionality of Distal-less involved co-opting an ancestral function in the nervous system for
use in body wall extensions (Panganiban et al. 1997, Panganiban 2000, Mittmann and Scholtz
2001, Williams et al. 2002).
Mutations in dachshund, another leg “gap” gene, are known to cause fusions of leg seg-
ments. However, they also change the number and distribution of bristles (Mardon et al. 1994).
Similarly, Notch signaling, which helps position joints along the PD leg axis, is also used in the
early dete mination of sense organs (reviewed in Hartenstein 2005). In addition, pox-neuro
Mechanisms of Limb Patterning in Crustaceans 97

and BarH, two genes that play a role in forming tarsal joints, can also transform sense organs
(Awasaki and Kimura 2001). Thus, numerous genes that play a role in limb patterning also play
a role in forming sensory structures.

Are Evolutionarily Relevant Substructures within Limbs Formed by


Developmentally Integrating Multiple Morphological Elements?

Given this kind of dual function in genes that pattern leg axes and sensory structures, it seems
possible that these two types of patterning were more closely linked ancestrally than they are in
Drosophila, with its highly specialized metamorphic development. Such linkage might be found
in arthropods like crustaceans that develop limbs directly. Certain patterns in crustacean leg
morphology could support this idea. For example, one notable regularity within limbs is the
position of setae at the distal margin of the limb podomere. There is a strong functional reason
for this position given that such setae can transmit information about the relative position of the
two limb joints. This function might be ancestral and would provide an apt selective scenario
for joints and sensory structures to be developmentally linked. Another morphological regular-
ity often found in crustacean limbs is the one-to-one correspondence between the annulation
and projecting setae in limbs that are multiannulate. Given that mutations in genes that regulate
joint formation can also change setal patterns, it seems plausible that the regulation of joints
and adjacent setae is mechanistically linked. Robust models of setal patterning in crustaceans
are a first step toward evaluating this hypothesis. Although the examples of structural linkage
I describe are hypothetical, they point to the fact that understanding such developmental link-
ages could provide new insights into homologies within crustacean legs, as well as the pathways
© Watling, Les; Thiel, Martin, Nov 28, 2012, Functional Morphology and Diversity

of limb diversification.

FUTURE DIRECTIONS
Oxford University Press, USA, Oxford, ISBN: 9780199875450

Comparative limb development is a wide open field in crustacean evolutionary morphology.


As shown at many points in this chapter, crustacean limbs are highly diverse both in their
basic morphology and in the transformations they undergo during the life cycle. This diversity
distinguishes them from most other arthropod groups and is part of what makes their study
so fascinating. Our developmental models still cannot explain variation among limbs of differ-
ent species. This is critical because crustacean limbs have diversified by varying their branches
and lobes, including the biramous branching of endopod and exopod and the medial and lat-
eral lobes that serve to specialize and differentiate limbs functionally. While it is remarkable
that crustacean (and arthropod) limbs have a conserved PD patterning module, we still need
to understand how the branches and lobes of limbs are patterned. We will be aided in this pur-
suit by techniques that are becoming easier to use in nonmodel organisms, for example, gene
sequencing and RNA interference. In addition, the search for other genes, not used in pattern-
ing Drosophila legs, is facilitated by the prevalence of genomic techniques. These advances will
allow both functional studies and much broader taxon sampling. Beyond the question of how
different limbs are patterned remains the question of how series of limbs are patterned, par-
ticularly since the metamorphic lifestyle of so many crustaceans means that a given segment
can develop morphologically diverse limbs during ontogeny. None of our models address such
metamorphic change, and yet this ability has permitted the specialization of larval lifestyles
radically distinct from the adult.
Finally, the field of comparative limb patterning has mainly focused on the generic Cartesian
axes used to pattern an outgrowth from the body wall. While this has shown some surprising
98 Functional Morphology and Diversity

conservation, we have not yet begun linking that patterning to the details of the morphology
that make up the limbs. Indeed, for crustaceans, we lack even basic models as to how fundamen-
tal limb structures, such as muscles and setae, develop. Furthermore, exploring this interface
between patterning limb axes and patterning limb morphology may uncover functional links,
revealing biologically relevant units of development. If so, these developmentally operational
units in the limb could help resolve some of still tangled issues of homology between highly
divergent limbs.

ACKNOWLEDGMENTS

My thanks to M. Thiel and L. Watling for their comments and editing of this chapter and to
L. Nagy and K. Dunlap for critical readings. Ívan Hinojosa generously made the final Fig. 3.8
from my color original.

NOTES

1 Typical nomenclature is to write gene names and their abbreviations in italics and to write the protein
product of a gene in all capitals. However, to simplify the presentation, I have avoided abbreviations and
distinguishing between RNA and protein expression.
2 The function of extradenticle depends on its being transported to the nucleus. This occurs in the presence
of homothorax. Strictly speaking, when I refer to extradenticle, I am referring to this nuclear extradenticle,
in contrast to other extradenticle expression that may be present but is nonnuclear.
© Watling, Les; Thiel, Martin, Nov 28, 2012, Functional Morphology and Diversity

REFERENCES

Abu-Shaar, M., and R.S. Mann. 1998. Generation of multiple antagonistic domains along the proximodis-
tal axis during Drosophila leg development. Development 125:3821–3830.
Abzhanov, A., and T.C. Kaufman. 2000. Homologs of Drosophila appendage genes in the patterning of
Oxford University Press, USA, Oxford, ISBN: 9780199875450

arthropod limbs. Developmental Biology 227:673–689.


Angelini, D.R., and T.C. Kaufman. 2004 . Functional analyses in the hemipteran Oncopeltus fasciatus
reveal conserved and derived aspects of appendage patterning in insects. Developmental Biology
271:306–321.
Angelini, D.R., and T.C. Kaufman. 2005 . Insect appendages and comparative ontogenetics.
Developmental Biology 286:57–77.
Artavanis-Tsakonas, S., M.D. Rand, and R.J. Lake. 1999. Notch signaling: Cell fate control and signal
integration in development. Science 284:770–776.
Averof, M., and S. Cohen. 1997. Evolutionary origin of insect wings from ancestral gills. Nature
385:627–630.
Averof, M., and N.H. Patel. 1997. Crustacean appendage evolution associated with changes in Hox gene
expression. Nature 388:682–686.
Awasaki, T., and K. Kimura. 2001. Multiple function of poxn gene in larval PNS development and in adult
appendage formation of Drosophila. Development Genes and Evolution 211:20–29
Bateson, W. 1894 . Materials for the study of variation. Macmillan, London .
Beermann, A., M. Aranda, and R. Schroder. 2004 . The Sp8 zinc-finger transcription factor is involved in
allometric growth of the limbs in the beetle Tribolium castaneum. Development 131:733–742.
Beermann, A., D.G. Jay, R.W. Beeman, M. Hulskamp, D. Tautz , and G. Jurgens. 2001. The Short antennae
gene of Tribolium is required for limb development and encodes the orthologue of the Drosophila
Distal-less protein. Development 128:287–297.
Bishop, S.A., T. Klein, A.M. Arias, and J.P. Couso. 1999. Composite signaling from Serrate and Delta
establishes leg segments in Drosophila through Notch. Development 126:2993–3003.
Mechanisms of Limb Patterning in Crustaceans 99

Blair, S.S. 1999. Drosophila imaginal disc development: Patterning the adult fly. in V. E.A. Russo, D.
Cove, L. Edgar, R. Jaenisch, and F. Salamini, editors. Development, genetics, epigenetics and envi-
ronmental regulation. Springer, Heidelberg.
Boxshall, G. A. 2004 . The evolution of arthropod limbs. Biological Reviews 79:253–300.
Brook , W.J., F.J. Diza-Benjumea, and S.M. Cohen. 1996. Organizing spatial pattern in limb development.
Annual Review of Cell and Developmental Biology 12:161–180.
Browne, W.E., A.L. Price, M. Gerberding , and N.H. Patel. 2005 . Stages of embryonic development in the
amphipod crustacean, Parhyale hawaiensis. Genesis 42:124–149.
Campbell, G., and A. Tomlinson. 1995 . Initiation of proximodistal axis in insect legs. Development
121:619–628.
Campbell, G., and A. Tomlinson. 1998. The roles of the homeobox genes aristaless and Distal-less in pat-
terning the legs and wings of Drosophila. Development 125:4483–4493.
Cohen, B., A.A. Simcox , and S.M. Cohen. 1993. Allocation of the imaginal primordia in the Drosophila
embryo. Development 117:597–608.
Cohen, S.M. 1990. Specification of limb development in the Drosophila embryo by positional cues from
the segmentation genes. Nature 343:173–177.
Cohen, S. M., and G. Jü rgens. 1989. Proximal-distal pattern formation in Drosophila: Graded require-
ment for Distal-less gene activity during limb development. Roux’s Archive for Developmental
Biology 198:157–169.
Damen, W.G., T. Saridaki, and M. Averof. 2002. Diverse adaptations of an ancestral gill: A common evo-
lutionary origin for wings, breathing organs, and spinnerets. Current Biology 12:1711–1716.
Dearden, P., and M. Akam. 2000. A role for Fringe in segment morphogenesis but not segment formation
in the grasshopper, Schistocerca gregaria. Development Genes and Evolution 210:329–336.
de Celis, J.F., D.M. Tyler, J. de Celis, and S.J. Bray. 1998. Notch signaling mediates segmentation of the
Drosophila leg. Development 125:4617–4626.
© Watling, Les; Thiel, Martin, Nov 28, 2012, Functional Morphology and Diversity

Dohle, W., and G. Scholtz. 1988. Clonal analysis of the Crustacean segment—the discordance between
genealogical and segmental borders. Development 104:147–160.
Deutsch, J.S., and E. Mouchel-Vielh. 2003. Hox genes and the crustacean body plan. BioEssays 25:878–887.
Franch-Marro, X., N. Mart í n, M. Averof, and J. Casanova. 2005 . Association of tracheal placodes with
leg primordia in Drosophila and implications for the origin of insect tracheal systems. Development
133:785–790.
Oxford University Press, USA, Oxford, ISBN: 9780199875450

Freeman, J.A., L.B. Cheshire, and T.H. MacRae. 1992. Epithelial morphogenesis in developing Artemia: The
role of cell replication, cell shape change, and the cytoskeleton. Developmental Biology 152:279–292.
Fristrom, D., and J.W. Fristrom. 1993. The metamorphic development of the adult epidermis. Pages
843–897 in M. Bate and A. Mart ı nez-Arias, editors. The development of Drosophila melanogaster.
Cold Spring Harbor Laboratory Press, New York .
Gonzalez-Crespo, S., and G. Morata. 1996. Genetic evidence for the subdivision of the arthropod limb
into coxopodite and telopodite. Development 122:3921–3928.
Guerao, G., K. Anger, U. Nettelmann, and C.D. Schubart. 2004 . Complete larval and early juvenile
development of the mangrove crab Perisesarma fasciatum (Crustacea: Brachyura: Sesarmidae)
from Singapore, with a larval comparison of Parasesarma and Perisesarma . Journal of Plankton
Research 26:1389–1408.
Gurney, R. 1942. Larvae of decapod Crustacea. Ray Society, London .
Guse. G.W. 1983. Ultrastructure, development, and moulting of the aesthetascs of Neomysis integer and
Idotea baltica (Crustacea, Malacostraca). Zoomorphology 103:121–133.
Hallberg , E., and B.S. Hansson. 1999. Arthropod sensilla: Morphology and phylogenetic considerations.
Microscopy Research and Technique 47:428–439.
Hartenstein, V. 2005 . Development of insect sensilla. Pages 379–419 in L. Gilbert, K. Iatrou, and S. Gill,
editors. Comprehensive molecular insect science, Vol. 1, Reproduction and development. Elsevier,
Oxford .
Hejnol, A., and G. Scholtz. 2004 . Clonal analysis of Distal-less and engrailed expression patterns dur-
ing early morphogenesis of uniramous and biramous crustacean limbs. Development Genes and
Evolution 214:473–485.
100 Functional Morphology and Diversity

Held, L.I. 1995 . Axes, boundaries and coordinates: The ABCs of fly leg development. BioEssays
17:721–732.
Held, L.I. 2002. Bristles induce bracts via the EGFR pathway on Drosophila legs. Mechanisms of
Development 117:225–234.
Inoue, Y., T. Mito, K. Miyawaki, K. Matsushima, Y. Shinmyo, T.A. Heanue, G. Mardon, H. Ohuchi, and
S. Noji. 2002. Correlation of expression patterns of homothorax, dachshund, and Distal-less with the
proximodistal segmentation of the cricket leg bud. Mechanisms of Development 113:141–148.
Jockusch, E.L., C. Nulsen, S.J. Newfeld, and L.M. Nagy. 2000. Leg development in flies versus grasshop-
pers: Differences in dpp expression do not lead to differences in the expression of downstream
components of the leg patterning pathway. Development 127:1617–1626.
Jockusch, E.L., T.A. Williams, and L.M. Nagy. 2004 . The evolution of patterning of serially homologous
appendages in insects. Development Genes and Evolution 214:324–338.
Kojima, T. 2004 . The mechanism of Drosophila leg development along the proximodistal axis.
Development Growth and Differentiation 46:115–129.
Kreissl, S., A. Uber, and S. Harzsch. 2008. Muscle precursor cells in the developing limbs of two isopods
(Crustacea, Peracarida): An immunohistochemical study using a novel monoclonal antibody
against myosin heavy chain. Development Genes and Evolution 218:253–265.
Lai, E.C., and V. Orgogozo. 2004 . A hidden program in Drosophila peripheral neurogenesis revealed:
Fundamental principles underlying sensory organ diversity. Developmental Biology 269:1–17.
Lecuit, T., and S.M. Cohen. 1997. Proximal-distal axis formation in the Drosophila leg. Nature
388:139–145.
Manzanares, M., T.A. Williams, R. Marco, and R. Garesse. 1996. Segmentation in the crustacean
Artemia as revealed by engrailed protein distribution. Roux’s Archive for Developmental Biology
205:424–431.
Mardon, G., N.M. Solomon, and G.M. Rubin. 1994 . dachshund encodes a nuclear protein required for
© Watling, Les; Thiel, Martin, Nov 28, 2012, Functional Morphology and Diversity

normal eye and leg development in Drosophila. Development 120:3473–3486.


McKay, D.J., C. Estella, and R.S. Mann. 2009. The origins of the Drosophila leg revealed by the
cis-regulatory architecture of the Distalless gene. Development 136:61–71.
Mirth, C., and M. Akam. 2002. Joint development in the Drosophila leg: Cell movements and cell popula-
tions. Developmental Biology 246:391–406.
Mitchell, B., and S.T. Crews. 2002. Expression of the Artemia trachealess gene in the salt gland and epi-
Oxford University Press, USA, Oxford, ISBN: 9780199875450

pod. Evolution and Development 4:344–353.


Mittmann, B., and G. Scholtz. 2001. Distal-less expression in embryos of Limulus polyphemus Chelicerata,
Xiphosura and Lepisma saccharina Insecta, Zygentoma suggests a role in the development of mech-
anoreceptors, chemoreceptors, and the CNS. Development Genes and Evolution 211:232–243.
Mouchel-Vielh, E., C. Rigolot, J. Gibert, and J.S. Deutsch. 1998. Molecules and the body plan: The Hox
genes of cirripedes Crustacea. Molecular Phylogenetics and Evolution 9:382–389.
Nagy, L.M., and T.A. Williams. 2001. Comparative limb development as a tool for understanding the
evolutionary diversification of limbs in arthropods: Challenging the modularity paradigm . Pages in
G.P. Wagner, editor. The character concept in evolutionary biology. Academic Press, New York.
Nulsen, C., and L.M. Nagy. 1999. The role of wingless in the development of multi-branched crustacean
limbs. Development Genes and Evolution 209:340–348.
Olesen, J., S. Richter, and G. Scholtz. 2001. The evolutionary transformation of phyllopodous to stenopo-
dous limbs in the Branchiopoda Crustacea—is there a common mechanism for early limb develop-
ment in arthropods? International Journal of Developmental Biology 45:869 –876.
Panganiban, G. 2000. Distal-less function during Drosophila appendage and sense organ development.
Developmental Dynamics 218:554–562.
Panganiban, G., S.M. Irvine, C. Lowe, H. Roehl, L.S. Corley, B. Sherbon, J.K. Grenier, J.F. Fallon, J.
Kimble, M. Walker, G.A. Wray, B.J. Swalla, M.Q. Martindale, and S.B. Carroll. 1997. The origin
and evolution of animal appendages. Proceedings of the National Academy of Sciences of the USA
94:5162–5166.
Panganiban, G., A. Sebring , L. Nagy, and S. Carroll. 1995 . The development of crustacean limbs and the
evolution of arthropods. Science 270:1363–1366.
Mechanisms of Limb Patterning in Crustaceans 101

Patel, N.H., T.B. Kornberg, and C.S. Goodman. 1989. Expression of engrailed during segmentation in
grasshopper and crayfish. Development 107:2 01–212.
Pavlopoulos, A., Z. Kontarakis, D.M. Liubicich, J.M. Serano, M. Akam, N.H. Patel, and M. Averof. 2009.
Probing the evolution of appendage specialization by Hox gene misexpression in an emerging
model crustacean. Proceedings of the National Academy of Sciences of the USA 106:13897–13902.
Posakony, L.G., L.A. Raftery, and W.M. Gelbart. 1991. Wing formation in Drosophila melanogaster
requires decapentaplegic gene function along the anterior–posterior compartment boundary.
Mechanisms of Development 33:69 –82.
Prpic , N.-M., and M.J. Telford. 2008. Expression of homothorax and in the legs of the crustacean Parhyale
evidence for a reversal of gene in the pancrustacean lineage. Development Genes and Evolution
218:333–339.
Prpic , N.-M., B. Wigand, W.G.M. Damen, and M. Klinger. 2001. Expression of dachshund in wild-type
and Distal-less mutant Tribolium corroborates serial homologies in insect appendages. Development
Genes and Evolution 211:467–477.
Queinnec , E., E. Mouchel-Vielh, M. Guimonneau, J.-M. Gibert, Y. Turquier, and J.S. Deutsch. 1999.
Cloning and expression of the engrailed.a gene of the barnacle Sacculina carcini. Development Genes
and Evolution 209:180–205.
Rauskolb, C. 2001. The establishment of segmentation in the Drosophila leg. Development 128:4511–4521.
Rauskolb, C., and K.D. Irvine. 1999. Notch-mediated segmentation and growth control of the Drosophila
leg. Developmental Biology 210:339 –350.
Rogers, B.T., M.D. Peterson, and T.C. Kaufman. 2002. The development and evolution of insect mouth-
parts as revealed by the expression patterns of gnathocephalic genes. Evolution and Development
4:96 –110.
Roy, S., and K. VijayRaghavan. 1999. Muscle pattern diversification in Drosophila: The story of imaginal
myogenesis. BioEssays 21:486 –498.
© Watling, Les; Thiel, Martin, Nov 28, 2012, Functional Morphology and Diversity

Scholtz , G. 1990. The formation, differentiation and segmentation of the post-naupliar germ band of the
amphipod Gammarus pulex L. (Crustacea, Malacostraca, Peracarida). Proceedings of the Royal
Society of London Series B 239:163–211.
Scholtz , G., W. Dohle, R.E. Sandeman, and S. Richter. 1993. Expression of engrailed can be lost and
regained in cells of one clone in crustacean embryos. International Journal of Developmental
Biology 37:299 –304.
Oxford University Press, USA, Oxford, ISBN: 9780199875450

Scholtz , G., N.H. Patel, and W. Dohle. 1994 . Serially homologous engrailed stripes are generated via
different cell lineages in the germ band of amphipod crustaceans (Malacostraca, Peracarida).
International Journal of Developmental Biology 38:471–478.
Schubiger, G. 1971. Regeneration, duplication, and transdetermination in fragments of the leg disc of
Drosophila melanogaster. Developmental Biology 26:277–295.
Serrano, N., and P.H. O’Farrell. 1997 Limb morphogenesis: Connections between patterning and
growth. Current Biology 7:R186–R195.
Sewell, W., T.A. Williams, J. Cooley, M. Terry, R. Ho, and L.M. Nagy. 2008. Evidence for a novel role for
dachshund in patterning the proximal leg. Development Genes and Evolution 218:293–305.
Shiga, Y., R. Yasumoto, H. Yamagata, and S. Hayashi. 2002. Evolving role of Antennapedia protein in
arthropod limb patterning. Development 129:3555–3561.
Soler, C., M. Daczewska, J.P. Da Ponte, B. Dastugue, and K. Jagla. 2004 . Coordinated development of
muscles and tendons of the Drosophila leg. Development 131:6041–6051.
Snodgrass, R.E. 1956. Crustacean metamorphoses. Smithsonian Miscellaneous Collections 131:1–78.
Sun, X., F.V. Mariani, and G.R. Martin. 2002. Functions of FGF signaling from the apical ectodermal
ridge in limb development. Nature 418:501–508
Taylor, J., and P.N. Adler. 2008. Cell rearrangement and cell division during the tissue level morphogen-
esis of evaginating Drosophila imaginal discs. Developmental Biology 313:739 –751.
Van Valen, L.M. 1993. Serial homology: The crests and cusps of mammalian teeth. Acta Palaeontologica
Polonica 38:145–158.
Wagner, G.P. 1989. The origin of morphological characters and the biological basis of homology.
Evolution 43:1157–1171.
102 Functional Morphology and Diversity

Williams, T.A. 1998. Distalless expression in crustaceans and the patterning of branched limbs.
Development Genes and Evolution 207:427–434.
Williams, T.A. 2004 . The evolution and development of crustacean limbs: An analysis of limb homolo-
gies. Pages 169–193 in G. Scholtz , editor. Evolutionary developmental biology of Crustacea.
Crustacean Issues, Vol. 15. Balkema, Lisse, The Netherlands.
Williams, T.A. 2007a . Limb morphogenesis in the branchiopod crustacean, Thamnocephalus platyurus,
and the evolution of proximal limb lobes within Anostraca. Journal of Zoological Systematics and
Evolutionary Research 45:191–201.
Williams, T.A. 2007b. Development of setal cells in the crustacean, Thamnocephalus platyurus.
Arthropod Structure and Development 36:63–76.
Williams, T.A. 2008. Early Distal-less expression in a developing crustacean limb bud becomes restricted
to setal forming cells. Evolution and Development 10:114–120.
Williams, T.A., and G.B. Muller. 1996. Limb development in a primitive crustacean, Triops longicauda-
tus: Subdivision of the early limb bud gives rise to multibranched limbs. Development Genes and
Evolution 206:161–168.
Williams, T.A., and L.M. Nagy. 1996. Comparative limb development in insects and crustaceans.
Seminars in Cell and Developmental Biology 7:615–628.
Williams, T.A., and L.M. Nagy. 2001. Developmental modularity and the evolutionary diversification
of arthropod limbs. Journal of Experimental Zoology, Molecular and Developmental Evolution
291:241–257.
Williams, T.A., C. Nulsen, and L.M. Nagy. 2002. A complex role for Distal-less in crustacean appendage
development. Developmental Biology 241:302–312.
Wolff, C., and G. Scholtz. 2008. The clonal composition of biramous and uniramous arthropod limbs.
Proceedings of the Royal Society of London Series B 275:1023–1028.
© Watling, Les; Thiel, Martin, Nov 28, 2012, Functional Morphology and Diversity
Oxford University Press, USA, Oxford, ISBN: 9780199875450
4
THE CRUSTACEAN CARAPACE: MORPHOLOGY, FUNCTION,
DEVELOPMENT, AND PHYLOGENETIC HISTORY

Jørgen Olesen

Abstract
A carapace (a shield extending from the head region and enveloping a smaller or larger part
of the body) is a characteristic feature of many crustaceans. This chapter reviews functional,
ontogenetic, and evolutionary aspects of the crustacean carapace. Carapace morphology in
Crustacea shows much variation, which is reflected in the many functions present in the vari-
ous subgroups. Among the more widespread functions of carapaces are that in many taxa, they
provide hydrodynamic advantages, offer protection, or form a feeding chamber, a respira-
tion chamber, or a brooding chamber. Special attention is devoted to the Branchiopoda and
Malacostraca, which both show a large variation in carapace morphology and ontogeny. The
influential textbook by Calman (1909) on crustacean morphology and systematics suggested
that a carapace was present primitively in both Malacostraca and Crustacea. This assumption
was long unchallenged, but a few decades ago attempts were made to invalidate/reject Calman’s
carapace hypothesis. Here it is argued that the best starting point may still be to assume homol-
ogy, at least within Malacostraca. Whether a carapace is homologous between major crustacean
taxa is more uncertain due to a general large morphological disparity, but most major taxa have
members with a “classical Calman type” of carapace extending from the rear of the head region
either as adults or as larvae. The information from the Cambrian “Orsten” crustaceans is ambig-
uous on this question. Some taxa, such as Rehbachiella and Walossekia, have a classical type of
carapace, while others, such as Skara and the even older Yicaris, lack a carapace entirely.

INTRODUCTION

A carapace, broadly defined as a shield extending from the head region and enveloping a
smaller or larger part of the body, is a characteristic feature of many crustaceans, for example,

103
Functional Morphology and Diversity. Les Watling and Martin Thiel.
© Les Watling and Martin Thiel 2013. Published 2013 by Oxford University Press.
104 Functional Morphology and Diversity

B D

1) 2) 3)

Fig. 4.1.
Examples of carapace-bearing nonmalacostracan Crustacea. (A) Triops cancriformis (Branchiopoda:
Notostraca) (from Gruner 1993). (B) Cyzicus sp. (Branchiopoda: Spinicaudata) mating (from Gravier
and Mathias 1930). (C) Bythotrephes longimanus (Branchiopoda: Cladocera) (from Lilljeborg 1901). (D)
Chonopeltis inermis (Branchiura) (from Fryer 1956). (E) Euphilomedes aspera (Ostracoda) (from Müller
1894). (A–E all taken from Gruner 1993, with permission from Gustav Fischer Verlag). (F) Metamorphosis
of Lepas (Cirripedia): 1, cypris larva; 2, attached larva; 3, young Lepas still surrounded by loosened cyprid
carapace (modified from Korschelt and Heider 1890). Arrowheads indicate the carapace.

branchiopods, ostracods, and decapods (Figs. 4.1, 4.2). The term carapace is traditionally used
for various types of shields or outgrowths covering the body of many crustaceans, but its use is
not restricted to crustaceans. The word is also used for the shields of a variety of noncrustacean
arthropod fossils and even for the dorsum of the fused head and thorax in spiders or horseshoe
crabs, as well as the well-known shell of a turtle.
Crustacea is a large taxon (>60,000 species; see Martin and Davis 2006) with an almost
unchallenged variation in morphological appearance spanning from the cave-dwelling, worm-
like remipedians to the large claw-bearing lobsters. An equal variation is seen in the ways
Crustacea have developed shields (carapaces) to cover the body. Sometimes the shield is fused
with the body, as in crabs and lobsters (Fig. 4.2); in other species it is a free shield attached
only anteriorly, as in leptostracans (Fig. 4.2A) or clam shrimps (Spinicaudata, Laevicaudata)
(Fig. 4.1B); in others it is modified to a dorsal brood pouch, as in raptorial cladocerans
(Fig. 4.1C). The large variation in carapace morphology in Crustacea is reflected in the many
known (or assumed) functions of carapaces in the various crustacean subgroups. Among the
most important functions, not all of which apply to all taxa, are protection, respiration cham-
ber, filtration chamber, brood chamber, and hydrodynamic streamlining.
The Crustacean Carapace 105

A
B

C D

E F G

Fig. 4.2.
Examples of carapace-bearing malacostracans. (A) Nebalia geoffroyi (Leptostraca) (from Claus
1888). (B) Squilla mantis (Stomatopoda) (from Giesbrecht 1916). (C) Spelaeogriphus lepidops
(Spelaeogriphacea) (from Gordon 1960). (D) Euphausia superba (Euphausiacea) (from Gruner 1993).
(E) Mysis relicta (Mysida) (Tattersal and Tattersal 1951). (F) Callinectes sapidus (Brachyura) (from Rathbun
1930). (G) Nannastacus unguiculatus (Cumacea) (from Gruner 1993). (H) Penaeus setiferus (Penaeidae)
(from Pérez Farfante 1988). (B-H all taken from Gruner 1993, with permission from Gustav Fischer
Verlag). Arrowheads indicate the carapace.

Calman (1909), in his classical and influential treatment of crustacean morphology and
systematics, suggested that the carapace is a primitive character in Crustacea. In the 1980s
and 1990s, the homologies and evolution of the carapace were discussed intensively for vari-
ous crustacean groups (Dahl 1983, 1991, Newman and Knight 1984, Walossek 1993, Fryer 1996,
Watling 1999), but since then this subject has rarely been touched. Considering the variation in
morphology, function, and development of carapace structures within Crustacea, it is not sur-
prising that the evolution of this structure has generated much discussion. One of the key ques-
tions in the earlier arguments by Dahl (1991) and Newman and Knight (1984) was whether all
crustacean carapaces are homologous or, phrased differently, whether a carapace is ancestral in
Crustacea. This question is still difficult to answer, but at least the discovery of well-preserved,
fossil microcrustaceans from the Cambrian, some of which have classical Calman-type cara-
paces extending from the rear of the cephalic region (e.g., Mü ller and Walossek 1988, Walossek
1993), provides some hints of early evolution of the carapace.
The wide occurrence of carapace structures in representatives of nearly all higher crusta-
cean taxa (see Figs. 4.1, 4.2), and the fact that so many different functions can be attributed
to the carapace in different taxa (Table 4.1) suggests that a carapace has been important in
Table 4.1. Overview of carapace morphology and function in recent Crustacea.

General morphology in adults Important functions in adults


Branchiopoda Much variation. Notostraca: carapace dorsal shield continu- General protection of body and limbs (most taxa); streamlining
(Figs. 4.1A–C, ous with head; Spinicaudata, Laevicaudata, and Cyclestherida: during swimming (some cladocerans and Notostraca); facilitates
4.4, 4.5A, 4.7A–E, carapace bivalved, capable of enclosing entire body (or most); burrowing (Notostraca); part of respiration/feeding current sys-
4.8–4.14) Cladocera: most taxa with bivalve carapace covering body but not tem (“Conchostraca” and Cladocera); protection of eggs/embryos
head; raptorial cladocerans (Onychopoda, Haplopoda): carapace (Notostraca) or as dorsal brood chamber (“Conchostraca” and
as dorsal brood pouch without free valves covering body. Cladocera); housing long, curled tubules of maxillary glands.
Branchiura All branchiurans with flat, shieldlike carapace (e.g., Møller Smooth, flattened carapace together with generally low pro-
(Fig. 4.1D) 2009) with head integrated into anterior part and with free pos- file of all branchiurans reduces frictional drag when attached
terior lobes overhanging varying parts of body; carapace evenly to hosts; general protection of body and limbs; houses widely
rounded anteriorly in some taxa (Dolops and some Argulus) and branched digestive system; carapace lobes ventral surface with
in other taxa divided into lobes (Chonopeltis). so-called respiratory areas involved in osmoregulation (Haase
1975, Boxshall and Jaume 2009); houses spermatophore glands in
Dolops (Fryer 1960).
Thecostraca Carapace of head shield type in cyprid or cypridiform stages of Hydrodynamics (e.g., Crisp 1955, Walker and Lester 2000, Walker
(Fig. 4.1F) all thecostracans (e.g., Høeg et al. 2004); Facetotecta: carapace 2004); general protection of body and limbs; surface with sensory
more or less maintains naupliar shape (not bivalved) overhanging organs such as “lattice organs” (Jensen et al. 1994, Rybakov et al.
body (Høeg et al. 2004); Ascothoracida and Cirripedia: carapace 2003).
bivalved with dorsal hinge line (Høeg et al. 2009); adult cirri-
pedes (barnacles, etc.): no carapace, but carapace valves of cyprid
larvae are ontogenetic precursors to fleshy mantle with calcified
shell plates (e.g., Calman 1909, Glenner and Høeg 1993).
Ostracoda Carapace most often calciferous and bivalved with valves held General protection of limbs and body; respiration (e.g., Abe and
(Figs. 4.1E, 4.7F) together dorsally by ligament, often also by hinge (Maddocks Vannier 1995); spaces between internal lamellae of carapace valves
1992); a few recent taxa (Swanson 1989) have a univalved are continuations of body cavity and may house organs related to
carapace. reproduction, digestion, nervous system (Maddocks 1992).
Leptostraca Large, bivalved, but with no dorsal hinge; dorsum of carapace General protection of body and limbs; form sides of filtra-
(Figs. 4.2A, 4.16A,B) continues anteriorly into small articulated rostral plate; in most tion chamber (see Cannon 1927); respiration chamber, but gas
taxa carapace covers phyllopodous thoracopods laterally (e.g., exchange over inner surface uncertain (see chapter 14 in this vol-
Nebalia), but in some taxa (e.g., Paranebalia and Nebaliopsis) tho- ume); hydrodynamics; rostral plate acts as ram during burrowing
racopods extend between valves ventrally. (Vannier et al. 1997); embryos carried ventrally by thoracopods
between carapace valves.
Euphausiacea Carapace fused with thorax dorsally so that cephalon and thorax Hydrodynamics; inner side possesses very thin cuticle so proba-
(Fig. 4.2D) form one piece (cephalothorax); a short spine extends in the mid- bly respiratory (Zimmer and Gruner 1956) (see chapter 14); carry-
line anteriorly; laterally carapace folds relatively short not cover- ing compound organs dorsally (function unknown) (Mauchline
ing gills of thoracopods. and Nemoto 1977).
Decapoda Carapace fused with thorax dorsally so that cephalon and thorax General protection due to heavy calcification; hydrodynamics;
(Figs. 4.2F,H, 4.3, form one piece (cephalothorax); lateral folds of carapace form provides robust attachment sites for limb musculature; more
4.5B–D, 4.15C, gill chambers; much taxon-specific variation in general carapace room for internal organs; carapace fused with thorax (cephalot-
4.16E,F) morphology. horax) may facilitate backward swimming as performed during
“escape reaction”; sides of carapace (branchiostegites) form gill
chambers closed to varying degree in various taxa; branchioste-
gites involved in respiration known only for few species (see chap-
ter 14); sound production (Henninger and Watson 2005).
Stomatopoda Carapace of cephalothorax type incorporating head and one or Sound production by carapace vibration (Patek and Caldwell
(Figs. 4.2B, 4.15B, two thorax segments (number uncertain since this body region is 2006); general protection of mouth parts and bases of maxilli-
4.16C,D) compressed); cephalothorax has free lateral margins; anteriorly is peds; female carapace grabbed by male during copulation (e.g.,
median rostral plate articulated to cephalothorax; some taxa with Dingle and Caldwell 1972); carapace length shorter than bur-
distinct longitudinal keel in the midline of the cephalothorax and row width (probably to enable possibility of turning around)
with additional longitudinal lateral ridges. (Atkinson et al. 1997).
Mysidacea Carapace typically large and with free folds posteriorly and ven- Mysida: carapace forms a respiratory chamber with gas exchange
(Figs. 4.2E, 4.6A,B) trally covering most of thorax and the basal parts of the thoraco- taking place on the inner surface (Mayrat et al. 2006, Wirkner and
pods; carapace in Mysida dorsally fused with first three thoracic Richter 2007); respiratory current is produced by lamellar epipod
segments (in Stygiomysis with first four) (Gruner 1993). of first thoracopod (Cannon and Manton 1927); Lophogastrida:
carapace previously considered nonrespiratory (but see Wirkner
and Richter 2007).

(Continued)
Table 4.1. (Continued)

General morphology in adults Important functions in adults


Cumacea Carapace of cephalothorax type incorporating dorsally anterior Respiration/feeding chambers; water current enters ventrally
(Figs. 4.2G, 4.6D–F) three or four thoracic segments (sometimes also segments 5 and and leaves anteriorly in Diastylis (but see text) generated by mov-
6). The lateral lobes of carapace form chambers closed posteriorly able and ventilatory epipod of maxilliped (Calman 1909, Oelze
(comparable to gill chambers in brachyurans) (Zimmer 1941). 1931, Zimmer 1932; Dennell 1937).
Tanaidacea Carapace of cephalothorax type incorporating dorsally two Respiration chambers with inner side being involved in gas
(Fig. 4.6G,H) (rarely three) thoracic segments; lateral carapace folds cover exchange (e.g., Tanais cavolinii) (Lauterbach 1970, Johnson and
mouth appendages laterally, forming semiclosed lateral respira- Attramadal 1982).
tory chambers; posteriorly with a pair of respiratory pores at edge
of carapace.
Thermosbaenacea In Thermosbaena mirabilis short carapace extends from head to Respiratory chamber; inner side of carapace with epithelium
(Figs. 4.6C, 4.7G) posterior border of fourth thoracic segment; dorsally fused with containing system of vascular lacunae connected to perivisceral
first thoracic segment (Barker 1962); lateral folds of carapace and pericardial sinuses (Siewing 1958); epipodite of the maxilli-
extend anteriorly and cover lateral bases of mouth parts and first ped beat within branchial chambers to draw inhalant respiratory
and second antennae. current (Barker 1962); ovigerous females carry embryos under
enlarged carapace.
Speleogriphacea In Spelaeogriphus lepidops short carapace, which extends from At inside is “oval patch” that may be respiratory (Gordon 1957,
(Fig. 4.2C) head posteriorly, consists of pair of lateral lobes overhanging Grindley and Hessler 1971).
anterior somites.
Mictacea Carapace not developed posteriorly but small lateral carapace Lateral carapace lobes thought to be respiratory but maxillipedal
folds cover bases of maxillae mx1, mx2, and mxp (head fused to epipod associated with respiratory function, as in other taxa, lack-
first thoracic somite) (Bowman et al. 1985). ing carapace (Bowman et al. 1985).
The Crustacean Carapace 109

crustacean evolution. The evolutionary success of crustaceans (or arthropods in general) is


probably very much linked to the plasticity of segmented limbs and the possibilities this has
given (e.g., Walossek 1993, Boxshall 2004; see also chapters 2 and 7 in this volume), but the car-
apace, with all the functional possibilities such a structure gives, seems to have been equally
important.
This chapter highlights the evolutionary significance of carapace structures within
Crustacea, both from a functional perspective and from a comparative morphological per-
spective. The following aspects are reviewed: (1) morphology and function of the carapace
in selected crustaceans; (2) case studies on the ontogeny of the carapace in two crustacean
key taxa, Branchiopoda and Malacostraca, both of which exhibit much variation in carapace
morphology and function of both larvae and adults, and both of which played a central role
when Calman formulated his crustacean carapace hypothesis; (3) critical summary of recent
discussions of Calman’s (1909) inf luential carapace hypothesis; and (4) Cambrian (“Orsten”)
evidence of carapace structures.

MORPHOLOGY AND FUNCTION OF THE CRUSTACEAN CARAPACE

The following is a brief overview of some of the main functions of the crustacean carapace, but
there are other functions not mentioned below (but see Table 4.1). The literature is vast, so this
summary is by no means exhaustive.

Hydrodynamics

Carapace morphology and hydrodynamics are linked in many ways in Crustacea. Most obvi-
ously, a carapace often provides general streamlining of the body surface and reduces drag
since it covers limbs and other projecting body parts. The carapace in Cambrian crustaceans
such as Rehbachiella (see Fig. 4.8, below) and Bredocaris probably had such a streamlining
function, since in these taxa the carapace is a simple posterior extension of the naupliar shield.
Indeed, hydrodynamic advantages may have been driving early carapace evolution (but were
probably also coupled with feeding advantages). Multiple other functions have evolved (feed-
ing, respiration, brooding), but hydrodynamic aspects remain important in many taxa.
Streamlining (reduction of drag) is important in two ways: (1) it allows for association
(attached or moving) with the substratum without being pulled off by water currents, and (2) it
increases swimming speed or makes it less costly. Here I present a few examples of crustacean
carapaces where hydrodynamic properties clearly play a significant role.
In the parasitic Branchiura (carp lice), streamlining as an adaptation to ectoparasitism on
fish has been taken to an extreme (Fig. 4.1D). The flattened carapace and the generally low body
profile effectively reduce drag when attached to the fish host. A general flattening of the body is
seen in many other crustacean taxa as a modification to their parasitic lifestyle (e.g., Copepoda,
Isopoda, Amphipoda) but without involving a carapace. Another example of a carapace with
clear hydrodynamic advantages is that of the cyprid larvae in cirripede crustaceans. The cyprid
is a larval type found only in cirripedes. It is a nonfeeding larva with a suite of specializations for
the purpose of locating a suitable substratum before irreversibly settling and molting to a juve-
nile cirripede (barnacles, gooseneck barnacles, or parasitic barnacles) (e.g., Høeg et al. 2004,
Høeg and Møller 2006) (Fig. 4.1F). Among the specializations of the cyprid larva is a bivalved
carapace capable of enclosing the body almost entirely between the valves. The hydrodynamic
advantages of this carapace shape for swimming cyprid larvae were explored by Walker (2004),
who found that cyprids of Heterosaccus lunatus were significantly faster than the naupliar
stages earlier in development, which reflects the efficiency of the fusiform shape of the cyprid.
110 Functional Morphology and Diversity

In general, cyprid larvae swim faster than other similarly sized invertebrate larvae (Walker
et al. 1987). During the phase where the cyprid explores the substratum and where eventually
attachment takes place, a streamlined carapace is clearly important since it reduces drag from
strong currents and tides and allows the cyprids to rapidly scan multiple sites. Also in bran-
chiopods, such as notostracans, where the relatively flattened carapace forms one piece with the
head, streamlining during swimming is important (Fig. 4.1A), but this morphology may also be
related to its occasional burrowing behavior (Fryer 1988).
A well-known locomotory adaptation in malacostracan crustaceans is the “escape reaction,”
which involves repeated tail-flips performed by the often heavily muscularized pleon. In some
taxa a tail-flip results in a 180° turn along the body axis (e.g., stomatopods or thermosbaen-
aceans; see Olesen et al. 2006), while in heavily calcified decapods such as crayfish, a series
of repeated flips results in backward swimming. Jacklyn and Ritz (1986) compared the swim-
ming of scyllarid lobsters with that of a panulirid lobster with respect to how much the shape
of the carapace contributes to maneuverability during backward swimming. They found that
the flattened carapace in scyllarid lobsters, in contrast to panulirid lobsters, is formed in such
a way that it produces a hydrodynamic lift when the large flattened scales of the antennae are
lowered (Fig. 4.3). They also saw that the independent movement of the right and left antennal
scales, which are at the rear during backward swimming, may alter the water flow and change
the distribution of the lift and thereby control rolling or “pitching” during swimming. In the
panulirid lobster also examined by Jacklyn and Ritz (1986), the tail-flips were used only to travel
in straight lines and over a short distance during the escape response, which they explained as
the result of the production of a negligible amount of lift during each tail-flip and by the lack of
antennae shaped or positioned to control lift.
Another type of swimming among decapods is that found in portunid crabs, which
use their fifth pair of pereopods as swimming paddles (Fig. 4.2F). Portunids can swim in
all directions, slowly when going backward or forward but rapidly when moving sideways
(Lockhead 1961). Blake (1985) examined the hydrodynamic properties of the carapace of
Callinectes sapidus. He found that f luid resistance (drag) was least when the side of the cara-
pace was oriented at right angles to the f low, which was interpreted as an adaptation to side-
ways high-speed swimming. Blake (1985) also found that the carapace of Callinectes in general
is adapted for minimum resistance and to generate hydrodynamic lift at low speeds during
forward swimming.

Feeding

In a number of well-known examples among crustaceans, the carapace plays a direct role in the
feeding process. In branchiopods (except some raptorial cladocerans), the beating of the trunk
limbs sets up a powerful current of water that is drawn in medially between the limbs and then
expelled laterally, a current that is an important part of the feeding process and sometimes also
in locomotion (e.g., Anostraca and Notostraca). In diplostracan branchiopods, where a large and
often bivalved carapace is present, this current is mostly used for feeding only. In laevicauda-
tan branchiopods, which have a large and globular carapace (see Figs. 4.7A, 4.12C, below), the
carapace plays a role in setting up the feeding current. The exopods of the trunk limbs fit neatly
against the inner wall of the carapace, thereby sealing the interlimb spaces laterally and creating
a vacuum in these spaces, with the result that water is drawn medially into the food groove when
the limbs are beating metachronically (Fryer and Boxshall 2009). Laevicaudatan branchiopods
are not true filtrators, but the basic principles of how the feeding current is set up are the same as
in most other branchiopods. In cladocerans such as Daphnia, a feeding current is generated in a
largely similar way (see Fryer 1991), and also here the carapace plays a role in guiding the currents
The Crustacean Carapace 111

1)

2)

3)

B L-
L+

Fig. 4.3.
Hydrodynamics of the scyllarid Ibacus peronii. (A) Three different types of backward swimming (1–3).
The propulsion is generated with repeated tail-flips. The carapace and the antennal scales together form
an effective aerofoil profile resulting in lift. The swimming direction is controlled by changing the angles
of the antennal scales. (B) By changing the articulation angle of the flattened antennal scales, the distri-
bution of the lift can be altered so that pitching and rolling movements are created. L– and L+ indicate a
downward- and upward-directed lift, respectively, when one antennae (right) is raised and the other (left)
is lowered; d indicates the perpendicular distance from rolling axis to the point of application of the result-
ant lift forces. From Jacklyn and Ritz (1986), with permission from Elsevier.

(Fig. 4.4A,B). A number of cladoceran branchiopods have exploited niches where carapace spe-
cializations and function go hand in hand. For example, the daphniid genera Scapholeberis and
Megafenestra have a modified ventral carapace in such a way that they are capable of suspending
themselves in an inverted position from the surface film of water (e.g., Fryer 1991) (Fig. 4.4C,D).
In general, modifications of the ventral carapace margins have been important within Cladocera
112 Functional Morphology and Diversity

C
A

D
water entering
carapace chamber
G

Exhalant current

F H
E

Fig. 4.4.
The carapace as a feeding chamber and modifications of ventral carapace margins in cladoceran
Branchiopoda. (A and B) Daphnia galeata (Daphniidae), with carapace drawn transparent to show feeding
appendages (A) and schematic drawing showing where feeding/respiration currents enter and leave the
carapace chamber (B). (C and D) Scapholeberis mucronata (Daphniidae) using modifications of ventral
carapace margins to suspend itself in an inverted position from the surface film of water (C) and showing
ventral carapace modifications (D). (E and F) Peracantha truncata (Chydoridae) crawling over the sur-
face when balancing on the ventral margins. (G and H) Graptoleberis testudinaria (Chydoridae) seen from
below (G) and behind (H) as it glides over a surface. Note the row of long setae on the ventral carapace
margins, which seals the carapace chamber when gliding. A–D from Fryer (1991), and E–H from Fryer
(1968), with permission from the Royal Society of London.

(e.g., Fryer 1968, 1991, Smirnov and Kotov 2009), but especially within Chydoridae this has been
an evolutionary trend. Fryer (1968) described a number of cases where the morphology of the
ventral carapace margins are important for crawling while food is collected in various ways from
the substratum by the trunk limbs (e.g., Fig. 4.4E–H). In the least specialized forms, such as
Alonopsis elongata, the setae on the ventral carapace margin are in contact with the substratum
on which it balances when it crawls forward by means of the first trunk limbs. Specializations
for crawling have been further exploited by Alonella exigua, which has more ventral modifica-
tions (e.g., more ventral setae, wider ventral flange). It is capable of largely sealing the carapace
chamber and then pumping water from it, which maintains a pressure difference between the
water inside and outside, enabling it to cling to and crawl over surfaces like a fly on a ceiling
(Fryer 1968). Graptoleberis testudinaria has established an almost entirely water-tight seal with a
very wide ventral flange and specialized setae (Fryer 1968) (Fig. 4.4G,H), enabling it to slide over
surfaces like a gastropod mollusc by setting up a pressure difference between the water inside and
outside the carapace chamber.
Species from the genus Nebalia (Leptostraca) spend much of their life burrowed in mud.
They have a large, bivalved carapace that constitutes the lateral sides of a large filtration cham-
ber. Cannon (1927) reported that the phyllopodous trunk limbs produce a water current, which
The Crustacean Carapace 113

enters the filtration chamber anteriorly and leaves it posteroventrally, from which particles
are being filtered out by the setae on the limbs. According to Cannon (1927), the anterior free-
moving part of the carapace, the rostrum, can be depressed and thereby appears to control the
current entering the filtration chamber. Vannier et al. (1997) suggested that the rostrum acts as
a ram during burrowing, preventing large particles from entering the filtration chamber from
anterior (see Fig. 10.1D in chapter 10 in this volume).

Respiration

In a number of examples within the Crustacea, the carapace forms a respiration chamber in
which gas exchange takes place, commonly over gills situated within the chamber (decapods),
but sometimes directly over the inner side of the carapace (branchiopods, ostracods, peracarid
malacostracans). There are even cases where the carapace chamber functions partially as a lung,
as it contains air, which is used for respiration while the animal is on land (e.g., grapsid or ocy-
podid crabs).
For branchiopods, it was previously assumed that lateral, saclike limb structures, the epipods
(or “gills”), were the main organs for respiration (e.g., Gicklhorn 1925), but later studies of the
epithelia (e.g., Kikuchi 1983) have shown that osmoregulation seems to be the main function of
the epipods (see Maas et al. 2009). It remains open to discussion whether branchiopod epipods
also have a respiratory function (see chapter 14 in this volume). Instead, the inner side of the
carapace is important for gas exchange. For Daphnia magna, recent studies show that the feed-
ing current is important for uptake of oxygen from the ambient medium and that gas exchange
occurs mainly within the filtering chamber (Pirow et al. 1999a) (Fig. 4.5A). A follow-up study by
the same authors showed that the inner wall of the carapace is the major site of respiratory gas
exchange (Pirow et al. 1999b). In contrast to earlier studies, where assumptions about respiratory
functions were mainly based on studies of tissue composition, Pirow et al. (1999b) developed an
elegant method to directly image hemoglobin oxygen saturation in transparent animals, which
allowed them to localize the specific areas on the body surface where oxygen uptake and release
take place. In Daphnia magna the highest values of hemoglobin oxygen saturation occurred near
the posterior margin of the carapace and in the rostral part of the head (which is outside the cara-
pace), suggesting that these areas are the most important for oxygen uptake (Pirow et al. 1999b).
In small animals living in the range of low Reynolds numbers, where viscous forces are domi-
nant (Koehl and Strickler 1981), the boundary layer is an obstacle for diffusive gas transport.
The thickness of the boundary layer depends on the velocity of the surrounding water relative
to the surface of the organism, and Pirow et al. (1999b) pointed out that reduced boundary layers
should occur inside the filtering chamber, where water is being pumped by the beating phyllopo-
dous limbs. They found that the enlarged carapace valves not only form the lateral boundaries of
the filtering chamber but are also employed for oxygen uptake due to the thin nature of the inner
wall of the carapace. In Daphnia magna, oxygen uptake does not only occur through the inner
wall of the carapace—the rostral part of the head had high oxygen values, which may provide
additional oxygen for sensory and central nervous system structures located in the head. This
additional respiratory surface may be of advantage during hypoxia if the carapace-based respira-
tory system fails to supply enough oxygen to the head regions (Pirow et al. 1999b).
In ostracods, as with many other smaller crustaceans, it is generally assumed that gas exchange
takes place over the general surface of the body. But notable exceptions are found in larger ostra-
cods, for example, in Vargula hilgendorfii, where oxygen uptake is assumed to occur preferen-
tially through the inner (posterior) surface of the carapace where the hemolymph sinuses are
best developed and are in direct contact with seawater (Abe and Vannier 1995). In Branchiura,
the so-called respiratory areas on the ventral sides of the carapace lobes have traditionally been
114 Functional Morphology and Diversity

B Midgut
A Heart Pericarical sinus
gland
Carapace
Carapace
Gills
Gill
chamber

Artery to leg Ventral nerve cord

C D
–∼ 60˚
AIR

AIR

5 4 3
2
1

Fig. 4.5.
The carapace as a feeding and respiration chamber in various Crustacea. (A) Respiratory function of cara-
pace in Daphnia magna (Cladocera). Arrows left of specimen show feeding/respiratory current. Arrows
inside specimen show main routes of blood flow. Fluorescence microscopy has revealed that the main sites
of gas exchange are at the inner wall of the carapace near the posterior margin and at the rostrum (from
Pirow et al. 1999b, with permission from the Company of Biologists, Ltd.). (B) Cross section of general-
ized decapod showing position of gill chambers, which are formed by lateral flanges of the carapace (gill
covers). Drawing combined from various sources. (C) Procambarus clarkii (Astacidea) with the gill covers
removed showing outer layer of gills. 1–5 represent pereopods 1–5 (from Bauer 1998, with permission from
Wiley and Sons). (D) Branchial chambers modified for air breathing and water circulation in the sema-
phore crab Heloecius cordiformis (Ocypodidae). The chambers are divided in an upper part containing air
and a lower part containing water (where the gills are). Forward tilting allows water to be tipped forward
and pumped out of the branchial chambers anteriorly and air simultaneously to be sucked in posteriorly
(from Maitland 1990b, with permission from Springer).

assumed to be involved in gas exchange, but physiological/anatomical experiments by Haase


(1975) suggested that these areas have an osmoregulatory function.
Among malacostracans, there are multiple examples of the carapace forming a respiratory
chamber, often with water being actively pulled into it by beating limbs specialized for this
purpose. The respiratory system of decapod crustaceans is well known—the lateral lobes of
the carapace (branchiostegites) form a chamber on each side of the body, housing gills of vary-
ing morphology attached to the thoracopods basally or to the body laterally (see Hong 1988 for
details and overview) (Fig. 4.5B,C). A specialized part of maxilla 2, the scaphognathite or gill
bailer, beats and performs rapid movements independent from the movements of the remaining
part of the limb, thereby dragging a ventilatory water current through the carapace chamber.
The gills are normally irrigated with a posterior-anterior flow, and the water is expelled from
the gill chambers through openings near the mouth. The precise site where water enters the
chambers varies among taxa, but in brachyurans it is generally through discrete openings near
The Crustacean Carapace 115

the bases of the pereopods, while in other taxa it may enter at varying positions along the ventral
carapace margin (e.g., Dyer and Uglow 1978, Batang and Suzuki 1999). Reversal of flow direc-
tion within the gill chambers has been recognized as a gill-cleaning strategy in certain decapods
(Bauer 1989, 1998) and may be particularly important for removing sediment from the gills in
burrowing species. In one example, Metapenaeus macleayi, the respiratory current enters a tube
in the sediment formed by the antennal scales and the antennules before it flows into the gill
chambers (Ruello 1973).
It is well known that many terrestrial or semiterrestrial brachyuran decapods retain water
in the branchial chambers of their carapace while active on land and, at the same time, have a
part of the chamber modified as a lung, relying on oxygen uptake directly from the air (Farrelly
and Greenaway 1994). Aspects of water retaining behavior are particularly well studied in the
semaphore crab, Heloecius cordiformis (Ocypodidae). A series of papers by Maitland (1990a,
1990b, 1992a, 1992b) showed that the water-retaining capacity of the carapace is intimately linked
to both respiration and feeding (Fig. 4.5D). One of the key findings is that H. cordiformis is an
obligate air breather while active on land. Air-based gas exchange takes place above the gills in
air-filled cavities lined with a vascular epithelium that function as lungs. While active on land,
the gill chambers are still partly filled with water thus irrigating the gills, but in experiments
where the branchial water has been removed, activity levels and oxygen percentages were unaf-
fected, suggesting that gill respiration is of less importance (for more details, see Maitland 1990a,
1990b, 1992a, 1992b). While on land, H. cordiformis sequentially depresses and elevates its cara-
pace in a pumplike manner (as in many other terrestrial brachyurans). The specific function of
this behavior is complicated, but Maitland (1992a) suggests that the carapace pump may enable
Heloecius to functionally partition lung ventilation from water circulation, thereby alleviating
the potential conflict between air breathing and water circulation. Another important func-
tion of the branchial water in H. cordiformis is for feeding. Water from the branchial chambers is
pumped out onto the mouthparts and is used in the separation of edible material from mud and
sand (Maitland 1990b).
The carapace also serves as a respiratory chamber in some peracarids. One example is the
Mysida, where no epipods acting as gills are present, but respiration takes place over the inner
wall of a relatively long carapace (e.g., Mayrat et al. 2006, Wirkner and Richter 2007). Cannon
and Manton (1927) studied the feeding and swimming of Hemimysis lamornae and found
that alongside with feeding and swimming currents, a special respiratory current is drawn in
under the posterior edge of the carapace and pushed out anteriorly at the sides of the maxillae
by the beating, lamellar epipods of the first thoracic limbs (see also Laverack et al. 1977) (Fig.
4.6A). Lophogastrids, which are sometimes grouped with Mysida in Mysidacea (e.g., Richter
and Scholtz 2001), are in many ways similar to mysids, but the carapace has been considered
nonrespiratory (e.g., by Dahl 1991). This notion has recently been challenged by Wirkner and
Richter (2007). As in mysids, a respiration current inside the carapace is generated by a beat-
ing epipodite of the first pair of thoracopods and apparently also by the exopods of the second
maxillae (Childress 1971), but in contrast to mysids, gills (epipods) are present on thoracopods
2–7 (see Boxshall and Jaume 2009), which probably takes care of most of the gas exchange
(Fig. 4.6B).
In the Thermosbaenacea, Tanaidacea, Cumacea, and Speleogriphacea, a short carapace
forms a small respiratory chamber (Grindley and Hessler 1971), inside of which a branchial/
ventilatory first thoracopod (maxilliped) epipod generates a water current. The thermosbaen-
acean Thermosbaena mirabilis was studied in detail by Barker (1962), and notes were also made
on the respiratory system based on the study of living specimens. As in mysids, a respiratory
current is drawn through the posterior opening of the short carapace (Fig. 4.6C) by the beat-
ing of the maxilliped epipod. Barker (1962) suggested that the epipod functions as a gill since it
116 Functional Morphology and Diversity

Inhalant
respiratory MYSIDA LOPHOGASTRIDA
Carapace Carapace
current B SCAPHOGNATHITE # 2
A
LATERAL
GILL

LEG
BASE SCAPHOGNATHITE # 1
VENTRAL GILL
Carapace
THERMOSBAENACEA
Inhalant D Carapace CUMACEA Carapace
respiratory E
C current
Exhalant th 2 th 3 th 4 th 5
respiratory mxp
ep
current Exhalant
II III IV current
Antennule endopod

Mandible Inhalant current


F epipod
Maxilliped

Inhalant
G TANAIDACEA respiratory
Respiratory H
current
pores

Fig. 4.6.
Respiratory carapaces and ventilation currents in peracarid malacostracans. (A) Lateral view of thorax
of Hemimysis lamornae (Mysida) showing the course of the respiratory currents (arrows). The current
is drawn in under the hind edge of the carapace, the inner wall of which is respiratory (no gills) (from
Cannon and Manton 1927, with permission from the Royal Society of Edinburgh). (B) Schematic lateral
view of Gnathophausia ingens (Lophogastrida) showing the course of the respiratory current under the
carapace, where gills take care of gas exchange. The respiratory current is generated by two different
“scaphognathites”: one is the exopod of the second maxilla; the other is the epipod of the first trunk limb
(from Childress 1971, with permission from the Biological Bulletin/Marine Biological Laboratory, Woods
Hole, MA). (C) Respiratory carapace in Thermosbaena mirabilis (Thermosbaenacea), carapace partly
removed (outline indicated by dashed line). Arrows show direction of respiration current (from Barker
1962, with permission from the Company of Biologists, Ltd.). (D) Diastylis species (Cumacea) buried in
sand, showing the inhalant and exhalant feeding and respiratory/feeding currents (arrows) in buried ani-
mals (from Dennell 1937, based on Zimmer, with permission from the Royal Society of Edinburgh). (E)
Anterior part of body of Diastylis from lateral showing carapace with large right respiratory maxillipedal
epipods inside. Water current enters ventrally (from Oelze 1931). (F) Maxilliped of Diastylis with respira-
tory epipod (from Calman 1909). (G and H) Tanais cavolinii (Tanaidacea) (from Johnson and Attramadal
1982, with permission from Taylor and Francis, Ltd.). (G) Schematic drawing showing water flow around
the body (arrows). Respiratory water enters through dorsal respiratory pores into respiratory carapace.
(H) Dorsal view of carapace showing position of respiratory pores (arrows).

contains vascular lacunae and offers a relatively large surface. However, the main site of respira-
tory exchange is the inner side of the carapace wall, where the cuticle covers a system of vascular
lacunae in the carapace epithelium (Siewing 1958, Barker 1962).
In the Cumacea, the epipods of the first maxilliped generate a respiratory current as in the
other mentioned taxa, but by contrast, this current also serves as a feeding/filtration current.
Zimmer (1932) found, in nonburied specimens of Diastylis rathkei , that the inhalant current
The Crustacean Carapace 117

enters the respiratory chambers between the bases of the third maxillipeds, which seem in
accordance with Oelze (1931) (Fig. 4.6E). In buried specimens, which had been undisturbed
for a period of time, Zimmer (1932) found that the inhalant current forms a small funnel in
the sediment between the bases of the third maxillipeds and the body. The exhalant current
leaves anterior to the site of intake, guided by the pseudorostrum (Zimmer 1932). Dennell
(1937), based on Zimmer (1932), provided an illustration of a buried specimen of Diastylis spe-
cies with arrows close to the anterior margin of the carapace indicating the respiratory cur-
rents (Fig. 4.6D). This illustration has since been reproduced in many text books. However,
the anterior position of the inhalant current indicated by Dennell is rather different from
that shown in a figure by Oelze (1931), where the current was found to enter more ventrally
through small slits (compare Fig. 4.6D,E). Possibly the two different representations can
be reconciled. The ventral current entrance shown by Oelze (1931) (Fig. 4.6E) may be the
anatomically true one, while the entrance shown in the overview figure by Dennell (1937)
shows where the inhalant current approximately enters when the animal is buried and the
ventral side is covered by sediment. The ventilatory epipod of the first maxilliped in Diastylis
is rather large and has a region that is folded into branchial lamellae (Calman 1909, Oelze 1931)
(Fig. 4.6E,F), which must function in gas exchange. The inner side of the carapace must be
involved in gas exchange, as in a number of other peracarids, since both valves are filled with
hemolymph channels (Oelze 1931, Siewing 1952).
In Tanaidacea the carapace forms a respiratory chamber on each side of the body as well.
Calman (1909) reported that the respiratory system of Apseudidae resembled that of cumaceans,
for example, with respect to the presence of a ventilatory maxillipedal epipod. Calman (1909)
also mentioned that the lateral folds of the carapace in tanaidaceans are traversed by a net-
work of blood channels and suggested that these form the chief organs of respiration, possibly
assisted by the epipodites of the maxillipeds, all of which was later confirmed by Lauterbach
(1970) for Tanais cavolinii. Johnson and Attramadal (1982) made detailed observations on the
mechanics of the carapace-based respiratory system in T. cavolinii. The tube-building lifestyle
of T. cavolinii prevents simple observation of its behavior, so to overcome this problem, the
animals were offered capillary glass tubes of varying diameters, which were readily accepted
(Fig. 4.6G). Observations showed that the basics of the respiratory system of T. cavolinii is
comparable to that of thermosbaenaceans and cumaceans (see above) but that certain char-
acteristics can be explained as adaptations to tube-dwelling behavior. According to Johnson
and Attramadal (1982), the ventilatory epipods of the maxillipeds, under normal conditions,
generate a respiratory current in each branchial chamber, which enters the chamber through
two respiratory pores at the posterior edge of the carapace valves (see Fig. 4.6H) and leaves
through ventral pores close to the base of the maxillipeds. Occasionally, the respiration current
is reversed (linked with reversal of pleopod pumping) so that the water enters the branchial
chamber ventrally and is expelled through the dorsal pores. Johnson and Attramadal (1982)
assume that this is to protect the dorsal respiratory pores and the branchial chamber from being
clogged by foreign particles in the suspension.
Spelaeogriphus lepidops, the first discovered species of the Spelaeogriphacea (Gordon 1957)
(Fig. 4.2C), has a respiratory carapace, but the respiratory chamber is less closed than that of
tanaidaceans and cumaceans and consists basically of lateral carapace lobes overhanging the
anterior somites. Based on Gordon (1957, 1960) and Grindley and Hessler (1971), who contrary
to Gordon had the opportunity to study live material, it can be concluded that S. lepidops appar-
ently has two different respiratory systems, which is unusual for Crustacea: one based on the
thoracopodal exopods (which will not be treated here) and one involving the carapace. Both
Gordon (1957) and Grindley and Hessler (1971) assumed that the cup-shaped maxillipedal epi-
pod probably is respiratory and that the “oval patch” (Gordon 1957) on the inside of the carapace
118 Functional Morphology and Diversity

might well be a respiratory surface, but unfortunately it was not possible for Grindley and
Hessler (1971) to observe the specific functioning of the epipodal gill of the maxilliped in the
living specimens studied by them.
Within Mictacea, which is another small peracarid order with only few species and which
show much resemblance to thermosbaenaceans and spelaeogriphaceans, it has been reported
that Mictocaris halope has an inflated, thin-walled elliptical area dorsal to the carapace fold
apparently functioning in respiratory exchange, but no respiratory maxillipedal epipod is
present (Bowman et al. 1985).

Brooding Chamber

A variety of crustaceans have brood care, and in some taxa the offspring are protected by the
carapace during part of their development. Branchiopod crustaceans (clam shrimps and water
fleas) are probably best known for this habit, but also in certain ostracods, ascothoracidan
thecostracans, and thermosbaenacean malacostracans, the carapace offers protection during
development.
In spinicaudatan and laevicaudatan (Fig. 4.7A) branchiopods (clam shrimps), the eggs
(embryos) are retained in a pair of clusters under the carapace valves, kept in position by vari-
ous supporting structures of the female’s appendages. In both these taxa the eggs are kept
under the carapace but only during a short part of their development, and they are released as
free-swimming nauplii or nauplia-like larvae (Olesen and Grygier 2003, 2004, Olesen 2005).
Cyclestheria hislopi, a clam shrimp with many similarities to Spinicaudata but now recog-
nized as a sister group to the Cladocera (water f leas) (e.g., Martin and Davis 2001, Richter
et al. 2007), has taken it a step further and retains the offspring under the carapace attached to
dorsal filaments of the trunk limb exopods until they are released as small adultlike juveniles
(“direct development”; Olesen 1999) (Fig. 4.7B; see also Fig. 4.11C, below). In the taxonomi-
cally diverse Cladocera, development is also direct and takes place dorsally under the cara-
pace (a single exception are the winter eggs of Leptodora kindtii, from which free-swimming
metanauplii hatch), but different from C. hislopi, the embryos are not attached to the female
via filaments. In most cladocerans, the inside of the brood chamber is directly connected
with the surrounding media, but in a few taxa (Moinidae, Onychopoda, Penilia), the develop-
ing embryos are nourished by a placenta-like organ, a so-called nä hrboden, inside the brood
chamber (e.g., Claus 1877, Potts and Durning 1980, Egloff et al. 1997, Dumont and Negrea
2002). Branchiopods are well known for producing diapause eggs, or resting eggs, as a part of
their life cycle, which appear when conditions are unfavorable (winter or drought), and in the
diverse Anomopoda the carapace is used in various ways as a container for the shed resting
eggs (ephippia), a habit that is unique among branchiopods (Fryer 1996). Probably familiar
to most students of zoology is the characteristic ephippium of Daphnia (Fig. 4.7C), which is
a modified part of the carapace housing two resting eggs and is the most specialized of such
structures among anomopods. In many anomopods, such as macrothricids and chydorids, the
ephippium is less elaborate than that of Daphnia and basically consists of the entire, unmodi-
fied carapace that after molting separates from the individual and comes to persist as an inde-
pendent envelope for a number of eggs. Even the simplest ephippia can withstand drought
(Fryer 1996). The ephippia of Daphnia with its resting eggs inside probably ensures that the
population can be continued after the winter or after drought but probably also plays a role
in dispersal. In some taxa, such as Streblocerus serricaudatus (Fig. 4.7D,E), the ephippium is
attached to vegetation and so reduces the chance of dispersal, which suggests that it is often
advantageous to ensure persistence of a population instead of undertaking the risks involved
in dispersing (Fryer 1996).
The Crustacean Carapace 119

A C D
Eggs

sexual cycle mating

Ephippium
sexual egg
Ephippium (carapace)
hatching
Carapace after parthenogenetic
diapause cycle
Embryonised
B parthenogenetic
larvae haploid E
daughter
egg
formation

parthenogenetic
son

Carapace
Carapace Embryo

Carapace
F Carapace G H
Eye
egg
Antennule

Pleopod 1

Fig. 4.7.
Carapace as brood chamber in various crustaceans. (A) Lynceus brachyurus (Branchiopoda,
Laevicaudata) with a cluster of eggs in a dorsal brood chamber between body and carapace valves (from
Sars 1896). (B) Cyclestheria hislopi (Branchiopoda, Cyclestherida) with well-developed embryos in dor-
sal brood chamber attached to limb filaments (from Sars 1887). (C) Life cycle of Daphnia (Branchiopoda,
Cladocera), where a part of the carapace is modified to a characteristic ephippium housing the rest-
ing eggs (from Ebert 2005, used with permission from the author). (D) Ephippium of Streblocerus ser-
ricaudatus (Branchiopoda, Cladocera) consisting of carapace attached to leaf (from Fryer 1972, with
permission from Wiley and Sons). (E) Embryos of S. serricaudatus near the point of hatching from a rest-
ing egg (from Fryer 1972, with permission from Wiley and Sons). (F) Vestalenula cornelia (Ostracoda)
with right valve removed (from Smith et al. 2006, with permission from the Royal Society of London).
(G) Thermosbaena mirabilis (Thermosbaenacea) with dorsal brood chamber containing embryos
(from Barker 1962, with permission from the Company of Biologists, Ltd.). (H) Amphionides reynaudii
(Amphionidacea), which has been proposed to have a ventral brood chamber between carapace valves
with elongate first pleopods forming the roof of the chamber (but brooding never directly observed)
(from Williamson 1973, with permission from Brill).

For Ostracoda, Maddocks (1992) reports that many burrowing and swimming forms brood
their eggs and young instars within the posterior part of the carapace (Fig. 4.7F). In many taxa,
however, eggs are laid individually or in clutches on plants or other substrata before develop-
ment (Maddocks 1992). Brooding was recently also reported for a Silurian ostracod where a
specimen with eggs, perhaps even juveniles, has been described, demonstrating that brooding is
a very conserved strategy within ostracods (Siveter et al. 2007).
Within Thecostraca, many species of Ascothoracida brood their offspring inside the cara-
pace that is sometimes much enlarged (Kolbasov et al. 2008). Grygier and Fratt (1984) reported
that, for Ascothorax gigas, a parasite in the bursae of the brittlestar Ophionotus victoriae, the
number of offspring varies between a few hundred and well more than a thousand per female.
120 Functional Morphology and Diversity

They identified five distinct larval stages, which, according to the authors, may not represent all
the instars that A. gigas passes through before release.
Within the Malacostraca, only the Thermosbaenacea brood the embryos dorsally under
a swollen carapace. Barker (1962) examined Thermosbaena mirabilis and found that the cara-
pace in breeding females is greatly enlarged by successive molts to form a brood pouch, which
carries an average of 10 embryos constantly agitated by the inhalant respiratory current (Fig.
4.7G). Barker (1962) suggested that this current may also play a role in sucking the embryos into
the brood pouch after they have appeared from the posteriorly directed vaginae. The embryos
emerge from the brood pouch as miniature adults.
For the aberrant Amphionides reynaudii (Malacostraca, Amphionidacea), Williamson (1973)
suggested that the enlarged female carapace forms a brood chamber where the long, anteriorly
directed first pleopods form the roof of the chamber (Fig. 4.7H). However, specimens with eggs
or embryos within this putative brood pouch are yet to be discovered.

ONTOGENY, EVOLUTION, AND MORPHOLOGICAL DIVERSIT Y OF THE CARAPACE


IN SELECTED CRUSTACEA

The Branchiopod Carapace

The Branchiopoda is an ideal case study on carapace evolution since this taxon is fairly well
defined but still offers some variation in the morphology and function of the carapace.
Branchiopod carapaces include a broad range of types: a more or less flattened dorsal shield
(Notostraca, Fig. 4.1A; see also Fig. 4.10C, below), bivalved carapaces sometimes capable of
enclosing the whole body (clam shrimps and cladocerans, Figs. 4.1B, 4.4, 4.5A, 4.7A–E; see also
Figs. 4.9, 4.11C, 4.12C, below), and dorsally attached brood pouches showing only little resem-
blance to the carapace in other branchiopods (raptorial cladocerans, Fig. 4.1C; see also Figs.
4.13C, 4.14C, below).
As suggested by Walossek (1993) and appreciated by Fryer (1996) in his review of the
branchiopod carapace, the well-preserved Cambrian microfossil Rehbachiella kinnekullensis
Mü ller, 1983 provides important information on early carapace development in Branchiopoda.
Rehbachiella has been suggested by Walossek (1993) to be an early branchiopod, and while there
is room for discussion of its precise phylogenetic position, a branchiopod affinity seems most
convincing. The carapace of Rehbachiella and its growth mode are taken here as an indication
of how the ancestral branchiopod carapace looked and are therefore summarized in the follow-
ing. Walossek (1993) showed very clearly that the carapace of Rehbachiella originates as a simple
extension of the original naupliar shield. In the earliest stage, the naupliar shield covers an area
dorsally that corresponds to the naupliar appendages: the first and second antennae and the
mandibles (Fig. 4.8A). A few stages later, the naupliar shield includes also the first maxilla seg-
ment (Fig. 4.8B) but not the second maxilla segment. Later the shield starts to grow into what
can now be termed a carapace since it has a free posterior fold starting to overgrow the second
maxilla segment and more segments posteriorly (Fig. 4.8C). Free folds also develop laterally
and start overgrowing the limbs. At a later stage, when the posterior margin of the carapace has
become freed, the dorsum of the second maxilla has become fused with the carapace (Fig. 4.8D),
so that the carapace fold at this and later stages appears topographically to be an extension of the
posterior margin of the second maxilla segment, while it can be argued that it is more correct to
consider it as developing from the first maxilla segment. At the latest known stage, the carapace
of Rehbachiella is attached in the “cephalic” region (which includes the maxilla 2 segment) and
has developed free folds posteriorly and laterally overhanging about seven somites posteriorly
and the proximal parts of the limbs (including cephalic appendages).
The Crustacean Carapace 121

Maxilla 2

Antennule Maxilla 1

Mandible Carapace
A
C
Antenna 2
Antennule

Antenna 2 Mandible Antennule


Maxilla 2
D Maxilla 1
Mandible
Antenna 2
Carapace

Mandible
Antennule
Antenna 2
Maxilla 1
Maxilla 2

Fig. 4.8.
Carapace development in Rehbachiella kinnekullensis, a Cambrian crustacean that most likely is a close
relative to the Branchiopoda, from Walossek’s (1993) series A (A–C) and series B (D), not to same scale.
Figures used with permission from Blackwell Publishing.

Given the presumed close relationship between Rehbachiella and other branchiopods,
it seems safe to assume that the carapace types in other branchiopods have evolved from a
Rehbachiella type of carapace. The phylogeny of Branchiopoda is relatively well understood
(Richter et al. 2007, Olesen 2009) (Fig. 4.9), from which a number of conclusions can be drawn.
The first off-split within Branchiopoda is the anostracan lineage (Sarsostraca), which consists
of the recent Anostraca and the Devonian fossil Lepidocaris rhyniensis, both of which lack a
carapace. Based on parsimony, a carapace may have been lost in this lineage. It is not yet cer-
tain which taxon constitutes the next branch within Branchiopoda. Traditionally (based on
morphology), Notostraca has been considered the sister group to a monophyletic Diplostraca
(= clam shrimps and water fleas), but most molecular work has suggested the Notostraca as an
ingroup of Diplostraca (Stenderup et al. 2006, Richter et al. 2007) (Fig. 4.9, dashed line with
question mark). This uncertainty holds importance for the idea of carapace evolution in the
Notostraca. If Notostraca really is a diplostracan ingroup, then it may be assumed that the noto-
stracan ancestor had some kind of bivalved carapace as seen in Spinicaudata, Cyclestherida,
and Laevicaudata. In this context, it is interesting that the carapace of Triops cancriformis has a
clear paired anlage in early larvae (Møller et al. 2003) (Fig. 4.10A), very similar to that seen in
the larvae or embryos of Spinicaudata, Cyclestherida, and Haplopoda (Olesen 1999, Olesen and
122 Functional Morphology and Diversity

(6)

(4) (10)
(12)
(5) (8)

(3)
(7) (13)
(9)
(11)
(1) (2)

)
 1)

a 6

3)
)
)

4)

7)

8)

a 9)

a 11
s 3

da 1
)
a 10

2)
i s

ata
)

ata
lla

hr
ca 5

eri

ra 1
od
)
ca 2

ari

po
hie

ud
art

ud
oll

od
sth

op
tra

do
oc

ho
ica

ca
ac

ac

kh

op
tra

om
cle
tos
pid

pto
ini
hb

str

yc
za

ev

en
os

Cy

On
No

An
Ka

Sp
Ca
Re

La
Le

Ct

Le
An

Gymnomera
? Cladocera
Cladoceromorpha
Sarsostraca Calmanostraca

Diplostraca

Phyllopoda

Branchiopoda (s. str.)


Branchiopoda (s. lat.)

Fig. 4.9.
Phylogeny of the Branchiopoda showing variation in the morphology of the branchiopod carapace
(in gray). Dashed line with question mark indicates alternative position of the notostracan clade
(Calmanostraca). From Olesen (2007), with permission from the Crustacean Society.

Grygier 2003, 2004, Olesen et al. 2003). Despite the similar developmental origin in the men-
tioned taxa, the subsequent development leads to adult carapace types that are quite different:
in Notostraca, the paired larval anlagen are precursors of a large univalved dorsal plate continu-
ous with the head in the adults (Fig. 4.1A, 4.10C), whereas in the Spinicaudata and Cyclestherida
the paired anlagen develop into the bivalved carapace, where each anlage corresponds to the
left or right valve (Fig. 4.11). The paired anlagen of the univalved carapace in Notostraca may
be an indication of a bivalved origin independently of whether or not Notostraca is a diplost-
racan ingroup. The longitudinal dorsal keel seen on the carapace of Triops, for example (e.g.,
Fig. 4.1A), may be a reminiscence of an earlier simple articulation like the ones seen in recent
Spinicaudata. The appearance of a univalved carapace in Notostraca from a bivalved carapace
of some type may have evolved alongside a shift in feeding strategy from filtration, as seen in
Anostraca and Spinicaudata, to a benthic lifestyle as omnivorous scavengers and opportunistic
predators. The Kazacharthra, a Jurassic extinct sister taxon to Notostraca (McKenzie and Chen
1999, Olesen 2009), also has a carapace that apparently is a dorsal plate, but not enough details
are known to be considered in this context.
Walossek (1993) noted that the carapace in diplostracan branchiopods (clam shrimps and
cladocerans) is “disconnected” from the head in the sense that the carapace in these taxa
does not appear as a simple posterior and lateral growth of the margins of a head shield as it
does in Rehbachiella. Consequently, he termed the diplostracan carapace a secondary shield
The Crustacean Carapace 123

Carapace anlagen Carapace

Mandible
Antenna 1
Antenna 1 Antenna 2

A 100 μm B 100 μm
Antenna 2

Carapace

1 mm
C

Fig. 4.10.
Carapace development in Triops (Notostraca): (A and B) stages I and II (from Møller et al. 2003, with per-
mission from Wiley and Sons, Ltd.) and (C) adult (from Olesen 2009, with permission from Senckenberg
Gesellschaft f ü r Naturforschung).

Carapace
Carapace

A 50 μm

Carapace

Embryonised larva
B 50 μm C 500 μm

Fig. 4.11.
Carapace development in Cyclestheria hislopi (Cyclestherida): lateral (A) and dorsal (B) view of embry-
onized larva and (C) adult female with left carapace valve removed. From Olesen (1999), with permission
from Wiley and Sons, Ltd.
124 Functional Morphology and Diversity

(e.g., Cyclestheria embryos in Fig. 4.11A,B). I agree with Walossek (1993) that this is a signifi-
cant evolutionary novelty in branchiopod carapace formation and therefore a synapomorphy at
some level. As explored elsewhere (Richter et al. 2007, Olesen 2009), there is some evidence that
Notostraca qualifies to be included in this group, since the carapace anlage in early larvae of
Triops (Fig. 4.10A) is disconnected from the head region in a way similar to that of Cyclestheria
hislopi, for example (Fig. 4.11A). But setting this aside, the lack of a (probably lost) carapace in
the anostracan lineage leaves uncertainty as to how early in branchiopod evolution this discon-
nection evolved between an anterior head shield and a posterior free carapace fold. It could
have been as early as in the branchiopod ancestor. In any case, the disconnection has taken
place, and it is therefore relevant to consider how. The ontogeny of Lynceus may provide a clue
since it exhibits both types of carapaces but in two different parts of its life cycle. In the larvae,
the carapace is an extension of the head shield as in Rehbachiella but with a unique morphology
(Olesen 2005) (Fig. 4.12A,B). In juveniles and adults, the carapace is large, bivalved, noncon-
tinuous with the “head,” and clearly of the “secondary shield” type as outlined by Walossek
(1993) (Figs. 4.7A, 4.12C). The shift in carapace morphology takes place from one instar to the
next. This abrupt change in morphology between two stages may in some way reflect what took
place in evolution. Hence, Lynceus exhibits at the same time, but in different stages, a plesio-
morphic head shield type of carapace (in larvae) and an apomorphic “disconnected” secondary
shield type of carapace (juveniles and adults).

Carapace Carapace

A 50 μm

Carapace

Antenna 1
B 50 μm C 500 μm
Labrum

Fig. 4.12.
Carapace development in Lynceus (Laevicaudata): (A and B) dorsal (A) and lateral (B) view of larva of L.
brachyurus (from Olesen 2005, with permission from Wiley and Sons, Ltd.) and (C) frontal view of adult of
L. tatei (from Olesen 2009, with permission from Senckenberg Gesellschaft f ü r Naturforschung).
The Crustacean Carapace 125

Cladoceran carapace origin and evolution pose other interesting questions concerning
carapace evolution. Cladocerans are often thought to have evolved neotenically from clam
shrimp ancestors, with free-living clam shrimp larvae as the starting point (e.g., Schminke
1981). However, since it is now known that Cyclestheria hislopi (Cyclestherida), which has
embryonized larvae (see Olesen 1999), is the sister taxon to Cladocera, free-living clam shrimp
larvae cannot have been the starting point for an eventual neotenic origin of cladocerans. If
neoteny (more generally, heterochrony) has been involved, then the embryos of Cyclestheria
hislopi would have instead been the starting point. It is striking to note that all it takes to
“make” a cladoceran carapace from an Cyclestheria-like ancestor is to stop the carapace devel-
opment of Cyclestheria before it starts overgrowing the head region (e.g., at stage VII; see
Olesen 1999).
Most cladocerans have a bivalved carapace where left and right side valves cover the body
and the trunk limbs (the head is free) (e.g., Figs. 4.4, 4.5A, 4.7C). But in the raptorial/preda-
tory cladocerans, Onychopoda and Haplopoda, the carapace has been modified further and
constitutes only a dorsal brood pouch, with no free valves covering the body and trunk limbs.
This has been taken to an extreme in the large, predatory Leptodora kindtii (Haplopoda), where
the carapace in adults is a saclike structure placed very far behind on the body (Fig. 4.13C). In
this respect, L. kindtii is different from other Crustacea, where the carapace in adults often is a
posterior extension of the cephalic region or at least is placed very close to this. Interestingly,
in the early ontogeny of Leptodora, the carapace appears as a narrow dorsal swelling behind the
mandibular region (Fig. 4.14A), which is far more anterior than the position of the carapace/
brood pouch in the adult female would indicate. During development, a posterior displacement
of the carapace takes place that can be followed step by step by studying the embryonized lar-
vae in the female’s brood pouch. After having appeared as a narrow, dorsal anterior fold, the
carapace becomes a dorsal flap (Fig. 4.14B,C) that gradually grows in size while it apparently
moves posteriorly until it constitutes a large, dorsal brood pouch as found in adults (Samter 1895,
Olesen et al. 2003) (Fig. 4.13). Olesen et al. (2003) suggested the anterior part of the carapace has
gradually fused to the dorsal part of the thorax during ontogeny, leaving only the posterior parts
of the valves free. The posterior displacement of the carapace in the ontogeny probably reflects
what took place during evolution of the characteristic brood pouch in Leptodora. A comparable
ontogeny has been described for certain malacostracans where the carapace also “fuses” with
the thoracomeres (see “Ontogeny of the Carapace in Some Malacostracans with Free Larvae,”
below).

The Malacostracan Carapace

Morphological Diversity of the Malacostracan Carapace

A carapace is present in many malacostracans and may have played a major role in the success
of this highly diverse taxon in combination with other features of the caridoid facies, such as
the muscular pleon (Hessler 1983, Hessler and Watling 1999). Not all taxa have a carapace, but
those that do exhibit a variation (Fig. 4.2) that is fascinating not only from a functional perspec-
tive (see summary of carapace functions above) but also from an evolutionary-morphological
perspective. The diverse Malacostraca with their variation in carapace sizes, cephalothorax
sizes, and development modes have provided much room for discussing carapace homolo-
gies, challenged in complexity perhaps only by an even more convoluted discussion on
limb homologies. However, the carapace/cephalothorax structures exhibited by malacost-
racans indeed provide a toolbox for considering very important aspects of malacostracan
evolution.
126 Functional Morphology and Diversity

A
B

Fig. 4.13.
Carapace development in Leptodora kindtii (Cladocera): three different developmental stages (A–C)
showing that the free part of the carapace (brood pouch, arrows) are migrating in posterior direction dur-
ing development. From Samter (1895), with permission from Elsevier.

Carapace anlage Carapace

Antenna 1
Antenna 2
Mandible
Trunk limb 1

A 50 μm B 100 μm

Carapace

Antenna 2

Antenna 1

Labrum
Trunk limb 1
C 100 μm

Fig. 4.14.
Carapace development in Leptodora kindtii (Cladocera). (A) Lateral view of head with early carapace lobe
behind the head. (B) Lateral view of head and thorax showing the early carapace as a small lobe having
migrated slightly more posterior than in A. (C) Juvenile with a small carapace lobe. From Olesen et al.
(2003), with permission from Wiley and Sons, Ltd.
The Crustacean Carapace 127

Taxa such as amphipods and isopods lack a carapace entirely. Only a few taxa, Leptostraca
and Mysidacea, have as adults what could be characterized a classical Calman type of carapace
enveloping a larger part of the thoracic region (Figs. 4.2A,E, 4.6A,B; see also Fig. 4.16A,B, below).
In both decapods and euphausiids, the head and thorax regions are combined into an unseg-
mented cephalothorax (Fig. 4.2D,F,H), which is also the case for stomatopods (Fig. 4.2B) but
involving a smaller part of the thorax. Other malacostracans, such as cumaceans, have a shorter
carapace fused with the first three thoracic segments but with free lateral lobes enclosing the
appendages that serve for respiration and feeding (Figs. 4.2G, 4.6D–F). Thermosbaenaceans (at
least Thermosbaena mirabilis) are unique among malacostracans since females carry the devel-
oping embryos dorsally beneath the carapace (Fig. 4.7G).

Evolution of the Malacostracan Carapace

One of the key questions concerning carapace evolution in the Malacostraca is whether a
Calman type of carapace was present in the common malacostracan ancestor, and the lack of it,
or various modifications, is therefore secondary. Calman suggested that a free carapace envel-
oping the thoracic region was a primitive attribute of the Malacostraca along with a number of
other shrimplike characteristics (his “caridoid facies”) (Fig. 4.15A). This suggestion was based
on the straightforward observation that a carapace of that type occurs in what he called the more
“primitive members” of the main taxa into which he divided the Malacostraca (e.g., Mysidacea
within Peracarida). This view has been adopted by later authors such as Hessler (1983) and
Newman and Knight (1984). Richter and Scholtz (2001) also interpreted the lack of a carapace as
a derived character within the Malacostraca. On the contrary, Dahl (1983, 1991) preferred a cara-
paceless Anaspides-like type of “caridoid facies” (= morphology of common ancestor) for the
Malacostraca. Also, Watling (1999), based on an alternative phylogenetic hypothesis, suggested
that the “caridoid facies” with its Calman-type carapace was not a part of the ground pattern
for the Malacostraca but rather a specialization in a caridoid clade consisting of Lophogastrida,
Mysida, Euphausiacea, and Decapoda.
Any considerations of character evolution, such as carapace evolution, should preferably
be done on the background of a well-supported phylogeny. For the Malacostraca, despite the
attempts mentioned above, it has been surprisingly difficult to obtain a robust result, and the
existing molecular and morphological evidence seems insufficient (Jenner et al. 2009, Richter
et al. 2009). However, if Leptostraca, which also have a large carapace enveloping the thoracic
region (Figs. 4.2A, 4.16A,B) is indeed the sister group to the remaining malacostracans, as per-
ceived by many authors (e.g., Siewing 1963, Richter and Scholtz 2001, Meland and Willassen
2007, Wirkner and Richter 2009, Regier et al. 2010), then it seems simplest to assume its pres-
ence in the malacostracan ancestor, retained largely unchanged in the Mysidacea, modified or
even lost in other taxa, exactly as was implied by Calman (1909).

Ontogeny of the Carapace in Some Malacostracans with Free Larvae

Within the Malacostraca, it seems that dendrobranchiate decapods, euphausiids, and stomato-
pods are particularly useful for considering early carapace evolution since, in these taxa, the
carapace during early larval development is a free fold attached to the posterior margin of the
head region (e.g., Figs. 4.15B,C, 4.16C–F). In adults of these taxa, the free carapace has been
transformed into a cephalothorax involving a varying number of thoracic somites (e.g., Fig.
4.2B,D,F,H). The presence of a naupliar sequence in the development of euphausiids and den-
drobranchiate decapods is most often considered primitive for Malacostraca. Scholtz (2002),
128 Functional Morphology and Diversity

Carapace
A

Carapace
B

Carapace

Fig. 4.15.
Calman’s (1909) hypothetical malacostracan ancestor and two malacostracan free-living larvae with the
carapace as posterior extension of the head shield not involving thorax segments. (A) Calman’s (1909)
illustration of a generalized type of Malacostraca showing a morphology that he called “caridoid fascies”
(shrimplike). He suggested that a large carapace overhanging the thorax was ancestral to Malacostraca.
(B) Antizoea of Lysiosquilla eusebia (Stomatopoda) (from Gurney 1942, with permission from the Ray
Society/Natural History Museum, London). (C) Protozoea of Penaeopsis species (from Gurney 1942, with
permission from the Ray Society/Natural History Museum, London).

on the contrary, suggested secondary reappearance of free-living nauplii within Malacostraca,


based partly on parsimony using the phylogeny of Richter and Scholtz (2001). However, since
the phylogeny of Malacostraca is still highly uncertain ( Jenner et al. 2009, Richter et al. 2009),
which weakens parsimony arguments, and since the early larvae (naupliar sequences) of for
example, both euphausiids and dendrobranchiates in many respects are indeed very similar to
larvae of nonmalacostracans, it is here assumed that an anamorphic development, probably
in many respects similar to that of dendrobranchiate decapods, is ancestral for Malacostraca.
Considering dendrobranchiate development, it is really striking that the early carapace of pro-
tozoea larvae of Penaeopsis (Gurney 1942, Fig. 4.15C) or Penaeus monodon (Fig. 4.16E,F), for
example, is very similar to that of the “Orsten” crustaceans Rehbachiella (Fig. 4.8) or Bredocaris.
The carapace in all of these taxa is an extension of the posterior margin of the naupliar shield.
And, as in Rehbachiella, it seems in Penaeopsis as though the segment of the second maxilla is not
The Crustacean Carapace 129

Carapace

A 100 μm B 100 μm

Carapace

Carapace

C 200 μm D 200 μm

Carapace Carapace
E F 50 μm

50 μm

Fig. 4.16.
Carapace development in various malacostracans. (A and B) Nebalia longicornis (Leptostraca) (from
Olesen and Walossek 2000, with permission from Springer): lateral view of late juvenile (A) and posterior
view showing dorsal attachment of carapace (B). (C and D) Lateral (C) and posterior (D) view of antizoea
of Squilla species (Stomatopoda). (E and F) Protozoea of Penaeus monodon (Penaeidae) (material provided
by G. Scholtz; data in Biffis et al. 2009): early protozoea with free posterior carapace fold (E) and later
protozoea where carapace has started to fuse to thorax (F). The arrow points at the cuticle between the
carapace and the thorax, in the process of shifting backward.

incorporated into the carapace in early stages (Fig. 4.15C), highlighting that a carapace “origi-
nating” from a second maxilla segment is not crucial when considering carapace homologies
among different taxa (see discussion below).
It has long been known that the free carapace in larvae of dendrobranchiate decapods devel-
ops into the well-known decapod type of cephalothorax, but the exact way this happens ontoge-
netically has been difficult to elucidate. According to Newman and Knight (1984), who looked
at the development of several dendrobranchiates and one euphausiid, it takes place as follows.
In the first protozoea, the carapace is represented by a backward and somewhat lateral extension
of the shield, which covers a portion of the thorax. Then, during ontogeny, which is achieved by
successive ecdyses, the thoracic and carapace cuticles are shed, and the underlying new carapace
cuticle shifts farther back on the thorax as molting progresses, until the posterior margin of the
130 Functional Morphology and Diversity

last thoracic segment is reached. Also, as stated by Newman and Knight (1984), in effect the cuti-
cle underlying the carapace and overlying the thorax is withdrawn, and in the process, the tissue
of the two structures melds or fuses together. Thus, the dorsal surface of the cephalothorax in the
adult is formed by the previous surface of the carapace in the larva. Figure 4.16F shows a proto-
zoea stage of Penaeus monodon caught in the middle of transformation of the carapace from a free
shield to a cephalothorax (the arrow points at the cuticle between the carapace and the thorax, in
the process of shifting backward). Casanova (1991, 1993) and Casanova et al. (2002) use another
terminology for describing carapace formation in a number of malacostracans, but there seems to
be a basic congruence with the description of Newman and Knight (1984). Casanova et al. (2002)
described the fusion between the carapace and the thorax in Penaeus indicus as involving a “split-
ting” of the thorax tergites.
It is well known that there is a fundamental connection between development and evo-
lution of living organisms. Any change in morphology must necessarily involve a change
of something in development, for example, a change in the rate or timing of the ontogeny of
certain structures (= heterochrony). With respect to the decapod carapace, it is tempting to
suggest that the way in which the cephalothorax is formed ontogenetically in, for example,
dendrobranchiate shrimps, as described by Newman and Knight (1984) and Casanova et al.
(2002), roughly ref lects the way in which the decapod cephalothorax approximately appeared
originally during evolution, that is, as a gradual withdrawal of cuticle between the carapace
and thorax.

ON CARAPACE HOMOLOGIES AND EARLY CARAPACE EVOLUTION

Calman’s Crustacean Carapace Hypothesis: Critical Survey of Recent Discussions

The term carapace has been used for Crustacea prior to Calman (1909), but since he gave a
brief and precise summary of the occurrence of the carapace within the Crustacea, his work has
often been used as a starting point when discussing its evolution. Because his summary has been
referred to so often in later works and has occasionally been rebutted (e.g., Dahl 1991), a brief
outline of what Calman actually wrote is given here. Calman (1909) suggested that since a dorsal
shield or carapace (he used both terms) occurs in the most diverse groups of Crustacea, it is prob-
ably a primitive attribute of the taxon. He also said that it originates as a fold of the integument
from the posterior margin of the cephalic region. Then he outlined some of the various forms a
carapace can take, such as in notostracan branchiopods, where the carapace loosely envelopes
more or less of the trunk, or as in other branchiopods (clams shrimps) and ostracods (mussel
shrimps), where it forms a bivalved shell completely enclosing body and limbs. Finally, he men-
tioned some of the extreme cases, such as adult cirripedes, where it forms a fleshy mantle usually
strengthened by shelly plates, and some malacostracans, where the carapace has coalesced with
the tergites of some or all thoracic somites, though it may project freely at the sides, overhanging,
as in Decapoda, the branchial chambers.
This short summary of Calman was excellent for the time and encapsulates a number of the
questions still discussed today. For example, the question of whether or not Crustacea originally
had a carapace was treated by Hessler and Newman (1975), who depicted two different possible
“urcrustaceans,” one with a carapace and one without. We have still not reached an entirely
convincing solution to this question, which, given the age, diversity, and morphological dispar-
ity of Crustacea, is not too surprising (see chapter 1 in this volume). Even Hexapoda (insects
and allies) may be an ingroup of Crustacea (e.g., Regier et al. 2010), which highlights the great
variation in crustacean body plans. Since a crustacean/arthropod carapace is an adaptation
The Crustacean Carapace 131

to an aquatic lifestyle, no such structure is needed in the terrestrial/aerial hexapods and must
have been reduced if hexapods eventually turn out to have originated from carapace-bearing
crustaceans, or it may never have been present in the lineage leading to Hexapoda.
Dahl (1983, 1991) spent much effort attempting to weaken/invalidate Calman’s idea of the
carapace as a primitive (and therefore homologous) attribute of the Crustacea. The evolution-
ary status of the carapace for Crustacea as a whole is still uncertain, but at least some of the
main arguments used by Dahl were problematic. The essence of Dahl’s view seems to be that
if structures (e.g., a carapace) have dissimilar ontogenesis in different taxa, then they cannot
be homologous. He argued that since carapace structures originate ontogenetically in so many
different ways in Crustacea and, in Dahl’s (1991) view, seemingly never as a dorsal fold derived
from the cephalon, then this invalidates the “carapace hypothesis” of Calman (1909). Indeed,
the early ontogeny of the carapace in Crustacea is very diverse, and certainly a straightforward
ontogeny from the rear of the cephalon is rare, but is nevertheless seen in various malacostracan
free-living larvae (see above and Figs. 4.15B,C, 4.16C–F), in branchiurans, in various thecostra-
cans, in “Orsten” crustaceans such as Rehbachiella (Fig. 4.8), and in branchiopods such as Triops
(Fig. 4.10).
Dahl (1991) distinguished a number of different carapace types named after the part of the body
that gives rise to the early carapace fold, and even though he said that the terms used for describing
these were “purely descriptive and without evolutionary or phylogenetic implications,” this was
nevertheless probably his most important argument for considering Calman’s “carapace hypoth-
esis” as invalid. He named six types of early carapace ontogenies that differed with respect to the
specific location of the early carapace fold(s) (dorsal, lateral, etc.) and with respect to how much
of the body is involved in forming the early carapace lobe (e.g., how many segments) (Dahl 1991).
Watling (1999) summarized Dahl’s arguments and highlighted that early carapace formation in the
short-carapace peracarids, such as Cumacea, takes place as lateral outpouchings (so-called branchi-
ostegal folds). However, while it obviously speaks in favor of homology if the carapace in various
taxa can be shown ontogenetically to have the same origin, the opposite is not necessarily the case:
structures, such as a carapace, may still be homologous despite having dissimilar ontogenetic ori-
gins (for a summary of the connection between ontogeny and evolution, see de Beer 1958, 149–153).
It is well known that a given ontogenetic sequence in itself is also subject to evolutionary modifica-
tions, so the dissimilar ontogeny of various carapaces in Crustacea does not necessarily indicate
convergent evolution; they could with equal right be explained as modifications of the ontogenetic
sequence.
Much attention has been devoted to whether or not the “origin” of the carapace is maxil-
lary. Calman (1909) was cautious enough to specify the “origin” of the carapace as being in
the “cephalic region ,” while later authors, including Dahl (1991), have taken this as Calman
(1909) specifically suggesting that the carapace folds originated from the rear of the second
maxilla segment. It is difficult to trace at what stage Calman’s carapace hypothesis became
modified from Calman’s broad statement to a more specific definition as the one referred
to by Dahl (1983, 198), who writes that “according to the classical concept the carapace is a
fold growing out from the maxillary segment.” But it seems that much of the discussion and
confusion of carapace homologies within Crustacea could have been avoided if the focus had
not been so much on whether the carapace is maxillary. As a matter of fact, there is no reason
for considering an ontogenetic origin of the carapace from the second maxilla segment as the
holy grail when it comes to carapace homologies within Crustacea. As outlined above, a dis-
similar ontogeny does not necessarily invalidate homology. The development of the carapace
in the Cambrian Rehbachiella kinnekullensis is here of special interest since, as outlined above,
the carapace in different phases of the development integrates a varying proportion of the
cephalic region. For example, at a certain stage, the carapace extends from the rear of the first
132 Functional Morphology and Diversity

maxilla region (Fig. 4.8C) and later from the rear of the second maxilla (Fig. 4.8D) (Walossek
1993). In the same line, as highlighted by Dahl (1991), taxa such as Notostraca and Leptostraca
(see Fig. 4.16A,B) have carapaces that do not extend entirely freely from the cephalic region
but actually are fused to the dorsum of one or more thoracic segments. However, an integra-
tion of a few extra segments into the carapace does not necessarily indicate nonhomology to
the carapace of taxa where this is not the case.

Contribution to Carapace Discussion by Well-Preserved Cambrian “Orsten” Fossils

Important evidence on early carapace evolution within the Crustacea is provided by the stud-
ies on Cambrian microfossils such as Rehbachiella kinnekullensis and Bredocaris admirabilis
(see Mü ller and Walossek 1988, Walossek 1993). The early carapace development in both taxa
is rather similar and can be described as a simple growth of the posterior margin of the nau-
pliar shield present earlier in development, resulting in free posterior shield margins covering
a smaller or larger part of the trunk (Fig. 4.8). Actually, as already recognized by Fryer (1996),
such a simple type of carapace growing from the rear of the head region as seen in Rehbachiella
is indistinguishable from what was suggested primitively for Crustacea by Calman (1909).
Walossek (1993, 198) used the term “cephalic shield” for Rehbachiella but was aware of the
similarity to the “carapace” as this structure is understood by other authors since, for exam-
ple, he stated that the ‘free carapace’ of Newman and Knight (1984) is more or less synony-
mous with the cephalic shield of Rehbachiella and other taxa. As mentioned above, one of
the much-discussed dogmas concerning carapace homologies has been the assumed origin
from the second maxilla segment (“cephalic region” in Calman’s terms). Walossek (1993) has
clarified that this is a simplified discussion since in both Rehbachiella and Bredocaris the exact
segment from which the free carapace margins extend depends on which developmental stage
is considered (Fig. 4.8).
Of course, one should be careful assuming that the mode of carapace development in
Rehbachiella is ancestral just because Rehbachiella is old. Regardless, from an evolutionary point
of view, the carapace seen in Rehbachiella would seem as an ideal starting point for carapaces in
crustaceans such as malacostracans, branchiopods, branchiurans, and thecostracans. However,
the information provided by Cambrian “Orsten” fossils is ambiguous since some taxa lack a
carapace. Skara (see Mü ller and Walossek 1985), for example, is lacking a carapace that could be
an adaptation to an interstitial lifestyle. The even older Yicaris also lacks a carapace and has a
shield restricted to only the head region (Zhang et al. 2007).

CONCLUSIONS

The Crustacea exhibit a large variation in carapace structures with a wealth of functions inti-
mately linked to the lifestyle of the taxon. Some well-demonstrated functions are reduction of
drag during swimming, crawling, or during temporarily resting on the substratum; as a feeding
chamber; as a respiration chamber; and as a brooding chamber. It is clear that the plasticity of
the crustacean carapace has been an important component in crustacean evolution. It has often
been discussed whether a carapace extending from the rear of the cephalon and enveloping a
smaller or larger part of the body was present in the common ancestor to the Malacostraca or
even in the common ancestor to the Crustacea. The occurrence of a carapace of this type within
the Malacostraca, either as larvae or as adults, seems to suggest, in accordance with Calman
(1909), that such a structure was present in the Malacostraca originally, from which it follows
that it has been modified or lost in a number of malacostracan subtaxa.
The Crustacean Carapace 133

Whether a carapace was present in the crustacean ancestor is more uncertain because of
a larger morphological gap between these taxa, but also similarities in carapace ontogeny in
certain taxa (malacostracans, branchiurans, branchiopods, and “Orsten” fossils) may suggest
a common origin. The fossil record is ambiguous on this question since some taxa have a cara-
pace (Rehbachiella, Bredocaris, Walossekia) and others lack it (Skara, Yicaris). The carapace in
the Cambrian fossil Rehbachiella kinnekullensis probably was close in morphology and ontogeny
to such an eventual crustacean ancestor. Differences in carapace ontogeny between different
crustacean taxa have sometimes been considered as evidence for parallel evolution of different
crustacean carapace types (e.g., Dahl 1991). However, in this chapter it is argued that differences
in ontogenetic origin of the carapace do not necessarily indicate nonhomology since an ontoge-
netic sequence in itself is also subject to evolution.

ACKNOWLEDGMENTS

I thank the editors of this book, Martin Thiel and Les Watling, for inviting me to contribute this
chapter. Gerhard Scholtz kindly provided specimens of Penaeus monodon used in Fig. 4.16. This
treatment greatly benefited from discussions with Jens Høeg, Stefan Richter, Dieter Waloszek,
and Les Watling at various occasions over the last years. Martin Thiel, Les Watling, and Stefan
Richter all read the chapter and gave useful comments. This work was supported by the Danish
Research Council (grant 09–066003).

REFERENCES

Abe, K., and J. Vannier. 1995 . Functional morphology and significance of the circulatory system of
Ostracoda, exemplified by Vargula hilgendorfii (Myodocopida). Marine Biology 124:51–58.
Atkinson, R.J.A., C. Froglia, E. Arneri, and B. Antolini. 1997. Observations on the burrows and burrow-
ing behaviour of Squilla mantis (L.) (Crustacea: Stomatopoda). Marine Ecology 18:337–359.
Barker, D. 1962. A study of Thermosbaena mirabilis (Malacostraca, Peracarida) and its reproduction.
Quarterly Journal of Microscopical Science 103:261–286.
Batang , Z.B., and H. Suzuki. 1999. Gill-cleaning mechanisms of the mud lobster Thalassina anomala
(Decapoda: Thalassinidea: Thalassinidae). Journal of Crustacean Biology 19:671–683.
Bauer, R.T. 1989. Decapod crustacean grooming: Functional morphology, adaptive value, and phy-
logenetic significance. Pages 49–73 in B.E. Felgenhauer , L. Watling , and A.B. Thistle, edi-
tors. Functional morphology and grooming in Crustacea. Crustacean Issues, Vol. 6. Balkema,
Rotterdam.
Bauer, R.T. 1998. Gill-cleaning mechanisms of the crayfish Procambarus clarkii (Astacidea: Cambaridae):
experimental testing of setobranch function. Invertebrate Biology 117:129 –143.
Biffis, C., F. Alwes, and G. Scholtz. 2009. Cleavage and gastrulation of the dendrobranchiate shrimp
Penaeus monodon (Crustacea, Malacostraca, Decapoda). Arthropod Structure and Development
38:527–540.
Blake, R.W. 1985 . Crab carapace hydrodynamics. Journal of Zoology (London) 207:407–423.
Bowman, T.E., S.P. Garner, R.R. Hessler, T.M. Iliffe, and H.L. Sanders. 1985 . Mictacea, a new order of
Crustacea Peracarida. Journal of Crustacean Biology 5:74–78.
Boxshall, G.A. 2004 . The evolution of arthropod limbs. Biological Reviews. 79:253–300.
Boxshall, G.A., and D. Jaume. 2009. Exopodites, epipodites and gills in crustaceans. Arthropod
Systematics and Phylogeny 67:229 –254.
Calman, W.T. 1909. Crustacea. Pages 1–346 in E.R. Lankester, editor. A treatise on Zoology, Part 7,
Fascicle 3. Adam and Charles Black , London .
Cannon, H.G. 1927. On the feeding mechanism of Nebalia bipes. Transactions of the Royal Society of
Edinburgh 55(2):355–369.
134 Functional Morphology and Diversity

Cannon, H.G., and S.M. Manton. 1927. On the feeding mechanism of a mysid crustacean, Hemimysis
lamornae. Transactions of the Royal Society of Edinburgh 55:219 –253.
Casanova, B. 1991. Origine protocéphalique antennaire de la carapace chez les Leptostracés, Mysidacés
et Eucaridés (Crustacés). Comptes Rendus hebdomadaires des Séances de l’Académie des Sciences
Paris, Série III 312:461–468.
Casanova, B. 1993. L’origin protocéphalique de la carapace chez les thermosbaenacés, tanaidacés,
cumacés et stomatopodes. The protocephalic origin of the carapace in thermosbaenaceans, tanaid-
aceans, cumaceans, and stomatopods. Crustaceana 65:144–150.
Casanova, B., L. De Jong , and X. Moreau. 2002. Carapace and mandibles ontogeny in the
Dendrobranchiata (Decapoda), Euphausiacea, and Mysidacea (Crustacea): A phylogenetic interest.
Canadian Journal of Zoology 80:296 –306.
Childress, J.J. 1971. Respiratory adaptations to the oxygen minimum layer in the bathypelagic mysid
Gnathophausia ingens. Biological Bulletin 141:109–121.
Claus, C. 1877. Zur Kenntnis des Baues und der Organisation der Polyphemiden. Denkschrift
Kaiserlichen Academie der Wissenschaften, Vienna 37:137–160.
Claus, C. 1888. Ü ber den Organismus der Nebaliiden und die systematische Stellung der Leptostraken.
Arbeiten aus dem Zoologischen Institut der Universität Wien 8:1–149.
Crisp, D.J. 1955 . The behaviour of barnacle cyprids in relation to water movement over a surface. Journal
of Experimental Biology 32:569 –590.
Dahl, E. 1983. Malacostracan phylogeny and evolution. Pages 189–212 in F.R. Schram, editor. Crustacean
phylogeny. Crustacean Issues, Vol. 1. Balkema, Rotterdam.
Dahl, E. 1991. Crustacea Phyllopoda and Malacostraca: A reappraisal of cephalic and thoracic shield and
fold systems and their evolutionary significance. Philosophical Transactions of the Royal Society
London Series B 334:1–26.
de Beer, G.R. 1958. Embryos and ancestors. Clarenden Press, Oxford .
Dennell, R. 1937. On the feeding mechanism of Apseudes talpa, and the evolution of the Peracaridan feed-
ing mechanisms. Transactions of the Royal Society of Edinburgh 59:57–78.
Dingle, H., and R.L. Caldwell. 1972. Reproductive and maternal behavior of the mantis shrimp
Gonodactylus bredini Manning (Crustacea: Stomatopoda). Biological Bulletin 142:417–426.
Dumont, H.J., and S.V. Negrea 2002. Introduction to the class Branchiopoda. Guides to the Identification
of the Microinvertebrates of the Continental Waters of the World, Vol. 19. Backhuys, Leiden .
Dyer, M.F., and R.F. Uglow. 1978. Gill chamber ventilation and scaphognathite movements in Crangon
crangon (L.). Journal of Experimental Marine Biology and Ecology 31:195–207.
Ebert, D. 2005 . Ecology, epidemiology, and evolution of parasitism in Daphnia [excerpt]. U.S. National
Library of Medicine, National Center for Biotechnology Information, Bethesda, MD. Available:
http://www.ncbi.nlm.nih.gov/books/NBK2036/ (accessed May 16, 2012).
Egloff, D.A., P.W. Fofonoff, and T. Onbé 1997. Reproductive biology of marine cladocerans. Advances in
Marine Biology 31:79 –167.
Farrelly, C.A., and P. Greenaway. 1994 . Gas exchange through the lungs and gills in air-breathing crabs.
Journal of Experimental Biology 187:113–130.
Fryer, G. 1956. A report of the parasitic Copepoda and Branchiura of the fishes of Lake Nyasa.
Proceedings of the Zoological Society 127:293–344.
Fryer, G. 1960. The spermatophores of Dolops ranarum (Crustacea, Branchiura): Their structure, forma-
tion, and transfer. Quarterly Journal of Microscopical Science 101:407–432.
Fryer, G. 1968. Evolution and adaptive radiation in the Chydoridae (Crustacea, Cladocera): A study in
comparative functional morphology and ecology. Philosophical Transactions of the Royal Society
of London Series B 254:221–385.
Fryer, G. 1972. Observations on the ephippia of certain macrothricid cladocerans. Zoological Journal of
the Linnean Society 51:79 –96.
Fryer, G. 1988. Studies on the functional morphology and biology of the Notostraca (Crustacea:
Branchiopoda). Philosophical Transactions of the Royal Society London Series B 321:27–124.
Fryer, G. 1991. Functional morphology and the adaptive radiation of the Daphniidae (Branchiopoda:
Anomopoda). Philosophical Transactions of the Royal Society of London Series B 331:1–99.
The Crustacean Carapace 135

Fryer, G. 1996. The carapace of the branchiopod Crustacea. Philosophical Transactions of the Royal
Society of London Series B 351:1703–1712.
Fryer, G., and G.A. Boxshall. 2009. The feeding mechanisms of Lynceus (Crustacea: Branchiopoda:
Laevicaudata), with special reference to L. simiaefacies Harding. Zoological Journal of the Linnean
Society 155:513–541.
Gicklhorn, J. 1925 . Ü ber spezifische und lokale Reduktion von Silber- und anderen Metallsalzen in den
Kiemensäckchen von Daphnia. Lotos 73:83–96.
Giesbrecht, W. 1910. Stomatopoden. Fauna Flora Golf Neapel 33:1–239.
Glenner, H., and J.T. Høeg. 1993. Scanning electron microscopy of metamorphosis in four species of
barnacles (Cirripedia Thoracica Balanomorpha). Marine Biology 117:431–439.
Gordon, I. 1957. On Spelaeogriphus, a new cavernicolous crustacean from South Africa. Bulletin of the
British Museum (Natural History) 5:31–47.
Gordon, I. 1960. On a Stygiomysis from the West Indies, with a note on Spelaeogriphus (Crustacea,
Peracarida). Bulletin of the British Museum (Natural History) Zoology 6:285–324.
Gravier, C., and P. Mathias. 1930. Sur la reproduction d’un Crustacé Phyllopode du groupe des
Conchostracés (Cyzicus cycladoides (Joly)). Comptes rendus de l’Académie des sciences 191:183–185.
Grindley, J.P., and R.R. Hessler. 1971. The respiratory mechanism of Spelaeogriphus and its phylogenetic
significance. Crustaceana 20:141–144.
Gruner, H.E. 1993. Arthropoda (ohne Insecta). A. Kaestner, editor. Lehrbuch der speziellen Zoologie.
Gustav Fischer, Stuttgart .
Grygier, M.J., and D.B. Fratt. 1984 . The ascothoracid crustacean Ascothorax gigas: Redescription, larval
development, and notes on its infestation of the Antarctic ophiuroid Ophionotus victoriae. Biology
of the Anarctic Seas XVI, Antarctic Research Series 41:43–58.
Gurney, R. 1942. The larvae of the decapod Crustacea. Ray Society, London .
Haase, W. 1975 . Ultrastruktur und Funktion der Carapaxfelder von Argulus foliaceus L. (Crustacea:
Branchiura). Zeitschrift f ü r Morphologie der Tiere 81:161–189.
Henninger, H.P., and W.H.I. Watson. 2005 . Mechanisms underlying the production of carapace vibra-
tions and associated waterborne sounds in the American lobster, Homarus americanus. Journal of
Experimental Biology 208:3421–3429.
Hessler, R.R. 1983. A defense of the caridoid facies; wherein the early evolution of the Eumalacostraca is
discussed. Pages 145–164 in F.R. Schram, editor. Crustacean phylogeny. Crustacean Issues, Vol. 1.
Balkema, Rotterdam.
Hessler, R.R., and W.A. Newman. 1975 . A trilobitomorph origin for the Crustacea. Fossils and Strata
4:437–459.
Hessler, R.R., and L. Watling. 1999. Les Péracarides: un groupe controversé. Pages 1–11 in J. Forest, edi-
tor. Traité de Zoologie. Anatomie, Systématique, Biologie, Tome VII, Fascicule IIIA. Crustacés
Péracarides, Vol. 19. Memoires de l’Institut Oceanographique Fondation Albert I, Monaco.
Høeg , J.T., N.C. Lagersson, and H. Glenner. 2004 . The complete cypris larva and its significance in
thecostracan phylogeny. Pages 197–215 in G. Scholtz , editor. Evolutionary developmental biology.
Balkema, Lisse, The Netherlands.
Høeg , J.T., and O.S. Møller. 2006. When similar beginnings leads to different ends: Constraints and
diversity in cirripede larval development. Invertebrate Reproduction and Development 49:125–142.
Høeg , J.T., M. Pérez-Losada, H. Glenner, G.A. Kolbasov, and K.A. Crandall. 2009. Evolution of mor-
phology, ontogeny and life cycles within the Crustacea Thecostraca. Arthropod Systematics and
Phylogeny 67:199 –217.
Hong , S.Y. 1988. Development of epipods and gills in some pagurids and brachyurans. Journal of Natural
History 22:1005–1040.
Jacklyn, P.M., and D.A. Ritz. 1986. Hydrodynamics of swimming in scyllarid lobsters. Journal of
Experimental Marine Biology and Ecology 101:85–99.
Jenner, R.A., C.N. Dhubhghaill, M.P. Ferla, and M.A. Wills. 2009. Eumalacostracan phylogeny and total
evidence: Limitations of the usual suspects. BMC Evolutionary Biology 9:21.
Jensen, P.G., J. Moyse, J.T. Høeg , and H. Al-Yahya . 1994 . Comparative SEM studies of lattice organs:
Putative sensory structures on the carapace of larvae from Ascothoracica and Cirripedia (Crustacea
Maxillopoda Thecostraca). Acta Zoologica 75:125–142.
136 Functional Morphology and Diversity

Johnson, S.B., and Y.G. Attramadal. 1982. A functional-morphological model of Tanais cavolinii
Milne-Edwards (Crustacea, Tanaidacea) adapted to a tubicolous life-strategy. Sarsia 61:29 –42.
Kikuchi, S. 1983. The fine structure of the gill epithelium of a fresh-water flea, Daphnia magna
(Crustacea: Phyllopoda) and changes associated with acclimation to various salinities. Cell and
Tissue Research 229:253–268.
Koehl, M.A.R., and J.R. Strickler. 1981. Copepod feeding currents: Food capture at low Reynolds
number. Limnology and Oceanography 26:1062–1073.
Kolbasov, G.A., M.J. Grygier, J.T. Høeg , and W. Klepal. 2008. External morphology of the two cypridi-
form ascothoracid-larva instars of Dendrogaster: The evolutionary significance of the two-step
metamorphosis and comparison of lattice organs between larvae and adult males (Crustacea,
Thecostraca, Ascothoracida). Zoologischer Anzeiger 247:159 –183.
Korschelt, E., and K. Heider. 1890. Lehrbuch der Vergleichenden Entwicklungsgeschichte der
Wirbellosen Thiere. Gustav Fischer, Jena .
Lauterbach, K.-E. 1970. Der Cephalothorax von Tanais cavolinii Milne Edwards (Crustacea—
Malacostraca). Ein Beitrag zur vergleichenden Anatomie und Phylogenie der Tanaidacea.
Zoologische Jahrbücher Abteilung f ü r Anatomie und Ontogenie der Tiere 87:94–204.
Laverack , M.S., D.M. Neil, and R.M. Robertson. 1977. Metachronal exopodite beating in the mysid
Praunus flexuosus: A quantitative analysis. Proceedings of the Royal Society of London Series B
198:139 –154.
Lilljeborg , W. 1901. Cladocera Sueciae. Nova acta Regiae Societatis Scientiarum Upsaliensis Series 3,
19:1–701.
Lockhead, J.H. 1961. Locomotion. Pages 313–364 in T.H. Waterman, editor. The physiology of Crustacea:
Sense organs, integration, and behavior. Academic Press, New York.
Maas, A., C. Haug , J.T. Haug, J. Olesen, X. Zhang , and D. Waloszek. 2009. Early crustacean evolution and
the appearance of epipodites and gills. Arthropod Systematics and Phylogeny 67:255–273.
Maddocks, R.F. 1992. Ostracoda. Pages 415–441 in F.W. Harrison and A.G. Humes, editors. Microscopic
anatomy of invertebrates, Vol. 9, Crustacea. Wiley-Liss, New York.
Maitland, D.P. 1990a . Aerial respiration in the semaphore crab, Heloecius cordiformis, with or without
branchial water. Comparative Biochemistry and Physiology 95A:267–274.
Maitland, D.P. 1990b. Carapace and branchial water circulation, and water-related behaviours in the
semaphore crab Heloecius cordiformis (Decapoda: Brachyura: Ocypodidae). Marine Biology
105:275–286.
Maitland, D.P. 1992a . Carapace movements aid air breathing in the semaphore crab, Heloecius
cordiformis (Decapoda: Brachyura: Ocypodidae). Journal of Comparative Physiology B
162:375–382.
Maitland, D.P. 1992b. Carapace movements associated with ventilation and irrigation of the branchial
chambers in the semaphore crab, Heloecius cordiformis (Decapoda: Brachyura: Ocypodidae).
Journal of Comparative Physiology B 162:365–374.
Martin, J.W., and G.E. Davis. 2001. An updated classification of the recent Crustacea. Natural History
Museum of Los Angeles County, Science Series 39:1–124.
Martin, J.W., and G.E. Davis. 2006. Historical trends in crustacean systematics. Crustaceana
79:1347–1368.
Mauchline, J., and T. Nemoto. 1977. Integumental sensilla of diagnostic value in euphausiids. Journal of
the Oceanographical Society of Japan 33:283–289.
Mayrat , A., B.R. McMahon, and K. Tanaka. 2006. The circulatory system. Pages 3–84 in J. Forest ,
J.C. von Vaupel Klein, and F.R. Schram, editors. Treatise on zoology—anatomy, taxonomy,
biology The Crustacea revised and updated from the Traité de Zoologie, Vol. 2. Koninklijke Brill,
Leiden .
McKenzie, K.G., and P.-J. Chen. 1999. Kazacharthra. Pages 443–458 in E. Savazzi, editor. Functional
morphology of the invertebrate skeleton. John Wiley and Sons, Chichester.
Meland, K., and E. Willassen. 2007. The disunity of “Mysidacea” (Crustacea). Molecular Phylogenetics
and Evolution 44:1083–1104.
Møller, O.S. 2009. Branchiura (Crustacea)—survey of historical literature and taxonomy. Arthropod
Systematics and Phylogeny 67:41–55.
The Crustacean Carapace 137

Møller, O.S., J. Olesen, and J.T. Høeg. 2003. SEM studies on the early development of Triops cancriformis
(Bosc) (Crustacea: Branchiopoda: Notostraca). Acta Zoologica 84:267–284.
Mü ller, G.W. 1894 . Ostracoda . Fauna und Flora Golf Neapel 21:1–404.
Mü ller, K.J., and D. Walossek. 1985 . Skaracarida, a new order of Crustacea from the upper Cambrian of
V ä stergötland, Sweden. Fossils and Strata 17:1–65.
Mü ller, K.J., and D. Walossek. 1988. External morphology of the Upper Cambrian maxillopod Bredocaris
admirabilis. Fossils and Strata 23:1–70.
Newman, W.D., and M.D. Knight. 1984 . The carapace and crustacean evolution—a rebuttal. Journal of
Crustacean Biology 4:682–687.
Oelze, A. 1931. Beiträge zur Anatomie von Diastylis rathkei Kr. Zoologische Jahrbücher. Abteilung f ü r
Anatomie und Ontogenie der Tiere 54:235–294.
Olesen, J. 1999. Larval and post-larval development of the branchiopod clam shrimp Cyclestheria hislopi
(Baird, 1859) (Crustacea, Branchiopoda, Conchostraca, Spinicaudata). Acta Zoologica 80:163–184.
Olesen, J. 2005 . Larval development of Lynceus brachyurus (Crustacea, Branchiopoda, Laevicaudata):
Redescription of unusual crustacean nauplii, with special attention to the moult between last nau-
plius and first juvenile. Journal of Morphology 264:131–148.
Olesen, J. 2007. Monophyly and phylogeny of Branchiopoda, with focus on morphology and homologies
of branchiopod phyllopodous limbs. Journal of Crustacean Biology 27:165–183.
Olesen, J. 2009. Phylogeny of Branchiopoda (Crustacea)—character evolution and contribution of
uniquely preserved fossils. Arthropod Systematics and Phylogeny 67:3–39.
Olesen, J., and M.J. Grygier. 2003. Larval development of Japanese “conchostracans”: Part 1, larval
development of Eulimnadia braueriana (Crustacea, Branchiopoda, Spinicaudata, Limnadiidae)
compared to that of other limnadiids. Acta Zoologica 84:41–61.
Olesen, J., and M.J. Grygier. 2004 . Larval development of Japanese “conchostracans”: Part 2, larval devel-
opment of Caenestheriella gifuensis (Crustacea, Branchiopoda, Spinicaudata, Cyzicidae), with notes
on homologies and evolution of certain naupliar appendages within the Branchiopoda. Arthropod
Structure and Development 33:453–469.
Olesen, J., S.T. Parnas, and J.F. Petersen. 2006. Tail flip and escape response of Tethysbaena argentarii
(Crustacea: Thermosbaenacea). Journal of Crustacean Biology 26:429 –432.
Olesen, J., S. Richter, and G. Scholtz. 2003. On the ontogeny of Leptodora kindtii (Crustacea,
Branchiopoda, Cladocera), with notes on the phylogeny of the Cladocera. Journal of Morphology
256:235–259.
Olesen, J., and D. Walossek. 2000. Limb ontogeny and trunk segmentation in Nebalia species.
Zoomorphology 120:47–64.
Patek , S.N., and R.L. Caldwell. 2006. The stomatopod rumble: Low frequency sound production in
Hemisquilla californiensis. Marine and Freshwater Behaviour and Physiology 39:99 –111.
Pérez Farfante, I. 1988. Illustrated key to penaeoid shrimps of commerce in the Americas. NOAA
Technical Report NMFS 64. U.S. Department of Commerce, Washington, DC.
Pirow, R., F. Wollinger, and R.J. Paul. 1999a . The importance of the feeding current for oxygen uptake in
the water flea Daphnia magna. Journal of Experimental Biology 202:553–562.
Pirow, R., F. Wollinger, and R.J. Paul. 1999b. The sites of respiratory gas exchange in the planktonic
crustacean Daphnia magna: An in vivo study employing blood haemoglobin as an internal oxygen
probe. Journal of Experimental Biology 202:3089 –3099.
Potts, W.T.W., and C.T. Durning. 1980. Physiological evolution in the branchiopods. Comparative
Biochemistry and Physiology 67B:475–484.
Rathbun, M.J. 1930. The cancroid crabs of America of families Euryliidae, Portunidae, Atelecyclidae,
Cancridae, and Xanthidae. Bulletin of the American Museum of Natural History 152:1–593.
Regier, J.C., J.W. Shultz , A. Zwick, A. Hussey, B. Ball, R. Wetzer, J.W. Martin, and C.W. Cunningham
2010. Arthropod relationships revealed by phylogenomic analysis of nuclear protein-coding
sequences. Nature 463:1079 –1084.
Richter, S., O.S. Møller, and C.S. Wirkner. 2009. Advances in crustacean phylogenetics. Arthropod
Systematics and Phylogeny 67:275–286.
Richter, S., J. Olesen, and W.C. Wheeler. 2007. Phylogeny of Branchiopoda (Crustacea) based on a com-
bined analysis of morphological data and six molecular loci. Cladistics 23:301–336.
138 Functional Morphology and Diversity

Richter, S., and G. Scholtz. 2001. Phylogenetic analysis of the Malacostraca (Crustacea). Journal of
Zoological Systematics and Evolutionary Research 39:113–36.
Ruello, N.V. 1973. Burrowing, feeding, and spatial distribution of the school prawn Metapenaeus macleayz
(Haswell) in the hunter river region, Australia. Journal of Experimental Marine Biology and
Ecology 13:189 –206.
Rybakov, A.V., J.T. Høeg , P.G. Jensen, and G.A. Kolbasov. 2003. The chemoreceptive lattice organs
in cypris larvae develop from naupliar setae (Thecostraca: Cirripedia, Ascothoracida and
Facetotecta). Zoologischer Anzeiger 242:1–20.
Samter, M. 1895 . Die Verä nderung der Form und Lage der Schale von Leptodora hyalina Lillj. wä hrend der
Entwicklung. Zoologischer Anzeiger 483:334–338, 484:341–344.
Sars, G.O. 1887. On Cyclestheria hislopi (Baird), a new generic type of bivalve Phyllopoda; raised from
dried Australian mud. Christiania Videnskabs-Selskabs Forhandlinger 1:1–65.
Sars, G.O. 1896. Fauna Norvegiæ. Bd. I. Beskrivelse af de hidtil kjendte norske Arter af Underordnerne
Phyllocarida og Phyllopoda [Descriptions of the Norwegian species at present known belonging to
the suborders Phyllocarida and Phyllopoda]. Joint-Stock, Christiania, Norway.
Schminke, H.K. 1981. Adaptation of Bathynellacea (Crustacea, Syncarida) to life in the interstitial (“Zoea
theory”). Internationale Revue der gesamten Hydrobiologie 66:575–637.
Scholtz , G. 2002. Evolution of the nauplius stage in malacostracan crustaceans. Journal of Zoological
Systematics and Evolutionary Research 38:175–187.
Siewing , R. 1952. Morphologische Untersuchungen an Cumaceen. Zoologische Jahrbücher. Abteilung f ü r
Anatomie und Ontogenie der Tiere 72:552–559.
Siewing , R. 1958. Anatomie und Histologie von Thermosbaena mirabilis. Ein beitrag zur Phylogenie der
Reihe Pancarida (Thermosbaenacea). Abhandlungen der Mathematisch Naturwissenschaftlichen
Klasse/Akademie der Wissenschaften und der Literatur in Mainz 7:195–270.
Siewing , R. 1963. Studies in malacostracan morphology: Results and problems. Pages 85–110 in H.B.
Whittington and W.D.I. Rolfe, editors. Phylogeny and evolution of Crustacea. Museum of
Comparative Zoology, Cambridge, MA .
Siveter, D.J., D.J. Siveter, M.D. Sutton, and D.E.G. Briggs. 2007. Brood care in a Silurian ostracod.
Proceedings of the Royal Society 274:465–469.
Smirnov, N.N., and A.A. Kotov. 2009. Morphological radiation with reference to the carapace valves of
the Anomopoda (Crustacea: Cladocera). International Review of Hydrobiology 94:580–594.
Smith, R.J., T. Kamiya, and D.J. Horne. 2006. Living males of the “ancient asexual” Darwinulidae
(Ostracoda: Crustacea). Proceedings of the Royal Society of London Series B 273:1569 –1578.
Stenderup, J.T., J. Olesen, and H. Glenner. 2006. Molecular phylogeny of the Branchiopoda
(Crustacea)—multiple approaches suggest a “diplostracan” ancestry of the Notostraca. Molecular
Phylogenetics and Evolution 41:182–194.
Swanson, K.M. 1989. Manawa staceyi n.sp. (Punciidae, Ostracoda): Soft anatomy and ontogeny. Courier
des Forschunginstitutes Senckenberg 113:235–249.
Tattersall, W.M., and O. Tattersall. 1951. The British Mysidacea. Ray Society, London .
Vannier, J., P. Boissy, and P. Racheboeuf. 1997. Locomotion in Nebalia bipes: A possible model for
Palaeozoic phyllocarid crustaceans. Lethaia 30:89 –104.
Walker, G. 2004 . Swimming speeds of the larval stages of the parasitic barnacle, Heterosaccus lunatus
(Crustacea: Cirripedia: Rhizocephala). Journal of the Marine Biological Association of the United
Kingdom 84:737–742.
Walker, G., and R.J.G. Lester. 2000. The cypris larvae of the parasitic barnacle Heterosaccus lunatus
(Crustacea, Cirripedia, Rhizocephala): Some laboratory observations. Journal of Experimental
Marine Biology and Ecology 254:249 –257.
Walker, G., A.B. Yule, and J.A. Nott. 1987. Structure and function in balanomorph larvae. Pages 307–327
in A.J. Southward, editor. Barnacle biology. Crustacean Issues, Vol. 5. Balkema, Rotterdam.
Walossek , D. 1993. The Upper Cambrian Rehbachiella and the phylogeny of Branchiopoda and Crustacea.
Fossils and Strata 32:1–202.
Watling , L. 1999. Toward understanding the relationship of the peracaridan orders: The necessity of
determining exact homologies. Pages 73–89 in F.R. Schram and J.C. von Vaupel Klein, editors.
Crustaceans and the biodiversity crisis. Brill, Leiden .
The Crustacean Carapace 139

Williamson, D.I. 1973. Amphionides reynaudii (H. Milne Edwards), representative of a proposed new
order of Eucaridian Malacostraca. Crustaceana 25:35–50.
Wirkner, C.S., and S. Richter. 2007. The circulatory system in Mysidacea—implications for the phylo-
genetic position of Lophogastrida and Mysida (Malacostraca, Crustacea). Journal of Morphology
268:311–328.
Wirkner, C.S., and S. Richter. 2009. Evolutionary morphology of the circulatory system in Peracarida
(Malacostraca; Crustacea). Cladistics 25:1–25.
Zhang , X. -g., D.J. Siveter, D. Waloszek , and A. Maas. 2007. An epipode-bearing crown-group crustacean
from the Lower Cambrian. Nature 449:595–598.
Zimmer, C. 1932. Beobachtungen an lebenden Mysidaceen und Cumaceen. Sitzungsberichte der
Gesellschaft Naturforschender Freunde zu Berlin 18:326 –347.
Zimmer, C. 1941. Cumacea. Bronns Klassen und Ordnungen des Tierreichs, Band 5, Abteilung 1, Buch IV,
Teil 5. Akademische Verlagsgesellschaft, Leipzig.
Zimmer, C., and H.E. Gruner. 1956. Euphausiacea. Bronns Klassen und Ordnungen des Tierreichs, Band
5, Abteilung 1, Buch VI, Teil 3. Akademische Verlagsgesellschaft, Leipzig.
5
THE CRUSTACEAN INTEGUMENT: STRUCTURE AND FUNCTION

Richard M. Dillaman, Robert Roer, Thomas Shafer, and


Shannon Modla

Abstract
The dorsobranchial exoskeleton of decapods has served as the archetype for studies of the struc-
ture and formation of crustacean cuticle. This cuticle consists of four layers: the epi-, exo-, and
endocuticles, which are mineralized with calcium carbonate, and the inner membranous layer.
The inner three layers are formed from chitin-protein fibrils arranged in parallel lamellae that
have a constantly changing orientation from layer to layer. This results in a plywoodlike com-
posite, with the mineral aligning with the orientation of the fibrils. The exoskeleton is thus a
composite structure with remarkable biomechanical resistance to fracture propagation. In dif-
ferent species and within different regions of the body of individuals, the cuticle shows many
variations on this basic pattern. Both the numbers of layers and presence or degree of miner-
alization are highly variable. Examples of thin, pliable, uncalcified cuticles include those of the
arthrodial, gill, and branchial chamber, contrasted by the heavily mineralized cuticle of the tips
of the chelipeds, which may be reinforced with other minerals. The cuticle is cyclically shed and
reformed to permit growth. The hypodermis first separates from the old cuticle and, in general,
begins to secrete the components of the new epi- and exocuticle. When the animal emerges from
the old exoskeleton, the endocuticle and membranous layer are deposited. Again, modifications
of this scheme are seen in different cuticle types. The cyclical nature of cuticle formation and
the temporal and spatial separation of the events of matrix deposition and calcification render
the crustacean cuticle an excellent model for the study of the control of biomineralization.

INTRODUCTION

The crustacean cuticle shares many structural and functional features with the cuticle of other
arthropods, especially insects. One of the most obvious constraints of a rigid exoskeleton is that

140
Functional Morphology and Diversity. Les Watling and Martin Thiel.
© Les Watling and Martin Thiel 2013. Published 2013 by Oxford University Press.
The Crustacean Integument: Structure and Function 141

it must be periodically shed so that the organism can increase in size. This means that cuticle
deposition is not a singular event but, over a lifetime, is a succession of depositions interspersed
with the shedding or loss of the old cuticle. Any review of the crustacean cuticle must recog-
nize the outstanding work of early anatomists and physiologists who described the ultrastruc-
ture of the arthropod cuticle and who put it in both an anatomical and developmental context.
Much of this early work was done on insects (Wigglesworth 1933 from Travis 1963) and served
as the standard, or frame of reference, for investigations on the crustacean cuticle. For exam-
ple, Neville (1975), in his volume Biology of the Arthropod Cuticle, describes a basic plan for the
arthropod cuticle but one that recognizes that there is not a single type of cuticle. He therefore
describes four different plans: tanned (sclerotized, cross-linked) solid cuticle, untanned solid
cuticle, rubberlike cuticle, and arthrodial membrane.
Revealing the plan for the crustacean cuticle is even more complex because major portions of
the cuticle are calcified, thereby requiring a mechanism for deposition of the mineral as well as
a means for selective removal of the mineral prior to shedding the old cuticle. The extracellular
nature of the cuticle means that it accumulates over time. One consequence of this pattern is that
any changes in the composition of the cuticle are a direct reflection of changes in the activity of
the epithelium underlying the cuticle at a specific time. The changes in the layers can be either
structural (e.g., a change in the composition or organization of the components of the cuticle)
or functional (e.g., the ability or inability of the cuticle to mineralize). Furthermore, in the latter
case function may not be expressed immediately upon deposition but may be expressed at a later
time (e.g., the ability to calcify may be determined early in premolt but not expressed until post-
molt). This type of expression pattern makes it particularly important that one be able to identify
candidate proteins and glycoproteins that are associated with various aspects of cuticle deposi-
tion and also to determine the pattern of gene expression for those proteins and glycoproteins.
Another factor that must be considered when characterizing cuticle formation is that differ-
ent regions of the cuticle vary considerably in structure and timing of their deposition. Structural
differences, of course, lead to functional differences. For example, the arthrodial membranes
have a flexible, noncalcified cuticle, thereby allowing movement of the appendages, but they
are deposited at the same time as the calcified dorsal carapace (Williams et al. 2003). Likewise,
differences in timing have functional consequences. For example, the thin, uncalcified cuticle
of the gills, which presents a minimal barrier to diffusion, is not replaced until all of the other
regions of the cuticle have been synthesized (Andrews and Dillaman 1993).

STRUCTURE AND COMPOSITION OF THE CUTICLE

The dorsal carapace of decapod brachyuran crustaceans (e.g., Carcinus maenas, Callinectes sap-
idus, Scylla serrata) has been the most widely studied and characterized region of the exoskel-
eton. It will thus serve as the prototype for the following description of the structure and
composition of the cuticle and for comparison with other taxonomic groups and regions of the
skeleton. An understanding of the function, development, and dynamics of the dorsal carapace
requires knowledge of its structure when fully elaborated, at the intermolt period. While current
descriptions of the crustacean exoskeleton have been refined since our previous reviews (Roer
and Dillaman 1984, 1993), the basic organization remains remarkably similar to that portrayed
in early studies of the insect cuticle by such icons as Locke (1959, 1960, 1961) and Neville (1975).
In common, current terminology, the four layers of the crustacean cuticle are (from distal to
medial) the epicuticle, exocuticle, endocuticle, and membranous layer. The organization and
structure of these layers are shown in Figs. 5.1 and 5.2. The epicuticle itself is approximately 5 μm
in thickness (Roer and Dillaman 1984) and comprises three layers: the outer surface coat, a cuti-
culin layer that has five sublayers, and a thick inner epicuticle that contains amorphous material
142 Functional Morphology and Diversity

A B C G
epicuticle
exocuticle epicuticle
exocuticle

endocuticle

membranous layer
hypodermis endocuticle
F

E
D
membranous layer
hypodermis

Fig. 5.1.
(A–F) Schematic of the molt cycle in Callinectes sapidus showing the layers present at intermolt (A; stage C4),
apolysis (B; stage D0), late premolt (C; stage D3–D4), early postmolt (D; stage A 1), middle postmolt (E;
stage B), and late postmolt (F; stage C3). (G) Section through an intermolt cuticle stained with acridine
orange.

and fibers (~6 nm in diameter) perpendicular to the surface (Compére 1995). Intercalated
within this set of fibers are others that extend down into the outer margin of the exocuticle
below (Modla 2006; Fig. 5.3A).
The exo- and endocuticles comprise a highly organized, organic framework of chitin-protein
fibrils and, in many regions, are hardened by sclerotization and/or impregnation with mineral
salts. The basis for the laminate structure of these fibrils has been a matter of study and debate
for many years (Bouligand 1972). Recent reports employing more sophisticated analytical tech-
niques have revealed a continuum of organization from the molecular to the tissue levels (Raabe
et al. 2005a, 2005b, 2007, Fabritius et al. 2009). These analyses explain not only the underlying
structure but also many of the composite biomechanical properties of the cuticle.
At the molecular level, the chitin fibrils are composed of β -1,4-linked N-acetylglucosamine
residues arranged in antiparallel chains of α-chitin. Groups of 18–25 chitin fibrils are wrapped
by proteins to form nanofibrils that are 2–5 nm in diameter and approximately 300 nm in length
(Raabe et al. 2005a, 2005b, 2007, Sachs et al. 2006, Romano et al. 2007, Fabritius et al. 2009).
The nanofibrils, in turn, cluster to form chitin-protein fibers that are 50–250 nm in diameter. As
the fibers aggregate into flat sheets parallel to the apical surfaces of the epithelial cells, they are
arranged around the microvilli forming pore canals. This gives the chitin-protein sheets a fenes-
trated appearance when viewed in tangential planes. The chitin-protein sheets are stacked, one
upon the next, with a slight rotation in the axis of the fiber orientation relative to the previous
layer (Fig. 5.2A). This imparts a plywoodlike structure to the cuticle, originally described for the
crustacean exoskeleton by Bouligand (1972). Each lamella in the exo- or endocuticle represents a
180° rotation in the orientation of the chitin-protein fibers within the sheets. The thickness of a
lamella is greater in the endocuticle (~8 μm) than in the exocuticle (~2 μm; see Fig. 5.2), which is
thought to be due to the relative angle between adjacent sheets of fibers, that is, smaller angles of
rotation in the endocuticle resulting in thicker lamellae since more layers are required to effect a
The Crustacean Integument: Structure and Function 143

Fig. 5.2.
Microstructure of lobster cuticle. (A) Schematic representation of the different hierarchical levels in the
microstructure of lobster cuticle starting with the N-acetyl-glucosamine molecules (I) forming antiparal-
lel α-chitin chains (II). Between 18 and 25 of these molecules wrapped with proteins form nanofibrils (III),
which cluster to form chitin protein fibers (IV) that are arranged in horizontal planes in which the long
axes of the fibers are all oriented in the same direction. The fibers are arranged around the cavities origi-
nating from the extremely well-developed pore canal system that gives the structure a honeycomb-like
appearance (V). These chitin protein planes are stacked with the orientation of the fibers in superimposed
layers rotating gradually around the normal axis of the cuticle, thus creating a typical twisted plywood
structure (VI). (B) Scanning electron microscropic (SEM) micrograph showing a cross section through
the three-layered cuticle. The different stacking density of the twisted plywood layers (tp) in the exo-
and endocuticle can be clearly seen. (C) SEM micrograph of obliquely fractured endocuticle displaying
two superimposed twisted plywood layers (tp) and showing their typical honeycomb-like structure. The
arrows indicate the pore canals. From Romano et al. (2007, fig. 1), with permission from Elsevier.

180° rotation. Presumably, the stacking angles are determined by the proteins that wrap the chi-
tin fibrils. In fact, differences in the exo- and endocuticular proteins have been well documented
(Skinner et al. 1992), as have differences in the sugar residues associated with cuticular glycopro-
teins (Marlowe et al. 1994; Compére et al. 2002). These differences also account, at least in part,
for the differences in tanning or sclerotization that occur between these layers. The exocuticle
is stabilized and hardened by quinone cross-linking effected by phenoloxidase, whereas such
tanning does not occur in the endocuticle.
The parallel, planar orientation of the chitin-protein lamellae and associated min-
eral is reflected in the predominant chitin and mineral crystallographic axes as revealed by
144 Functional Morphology and Diversity

Fig. 5.3.
Transmission electron microscopic images of 1-hour postmolt cuticle of Callinectes sapidus fixed with 2.5%
glutaraldehyde. (A) Cross section through epicuticle. Note the dense vertical fibers (dvf), epicuticular
fibers (ef), epicuticular canals (ec), epicuticular roots (er), and inner (ie) and outer (oe) epicuticle. (B)
Cross section through single lamella of exocuticle. Note the vertical fibers (vf) and horizontal fibers
(arrowheads). (C) Tangential section through the proximal exocuticle. Note the anchoring fibers (af),
horizontal fibers (hf), pore canals (pc), pore canal fibers (pcf), and pore canal sheaths (pcs). (D) Cross
section through proximal exocuticle and tendinous epidermal cell. Note the insertion of the tonofibers (tf)
into the cell as well as the dense structures (d) and microtubules (mt) in the cell. (E) Tangential section
through the proximal exocuticle above a tendinous epidermal cell. Note the electron-dense rods (r) and
tonofibers (tf). (F) Tangential section through the exocuticle. Note the region of the interprismatic septa
(IPS). From Modla (2006, figs. 1a, 2d, 5e, 6d, 7d, and 9b), used with permission from the author.

synchrotron Bragg diffraction (Raabe et al. 2005a, 2007) and X-ray diffraction of cuticular
samples (Raabe et al. 2006, 2007, Krywka et al. 2007, Al-Sawalmih et al. 2008). When the dif-
fraction patterns are projected upon the {020} crystallographic plane, which runs normal to
the surface in the transverse direction (90° to the long axis of the body of the lobster Homarus
The Crustacean Integument: Structure and Function 145

Fig. 5.4.
Survey of the {020} synchrotron pole figures of the orthorhombic α-chitin taken from different parts of
the cuticle. LD, longitudinal reference direction; ND, reference direction normal to the local surface; TD,
transverse direction. Specimens were taken from a highly mineralized part of the cuticle on pincher claw
(left cheliped), crusher claw (right cheliped), cephalothorax, and abdomen. Specimens were also taken
from poorly mineralized positions on the telson or abdomen (dashed line). From Raabe et al. (2007, fig. 5),
with permission from Elsevier.

americanus), pole figures (as shown in Fig. 5.4) demonstrate a strong orientation along the long
axis of the body (Raabe et al. 2007). A pole figure is a crystal orientation measurement and is so
named because it is often plotted in polar coordinates consisting of the tilt and rotation angles
with respect to a given crystallographic orientation. These data are interpreted to reflect the
orientation of the chitin-protein molecules within the lamellae that are parallel to the cuticle
surface. The pole figures, however, also reveal a secondary crystallographic axis normal to the
surface of the cuticle. It was concluded that this represents chitin and mineral associated with
the vertical pore canals perpendicular to the lamellar elements. As discussed below, numerous
vertical elements are found in the exoskeleton of Callinectes that could also contribute to the
observed orientation in the normal dimension.
146 Functional Morphology and Diversity

Investigations in our laboratory on the ultrastructure of late premolt and early postmolt blue
crabs have demonstrated that numerous fibrous structures run parallel to the twisted-ribbon
shaped pore canals and therefore perpendicular to the horizontal chitin protein fibers forming
the Bouligand or twisted plywood layers. Modla (2006) used transmission electron microscopy
after conventional glutaraldehyde fixation or uranyl acetate fixation and described the morphol-
ogy and distribution of a variety of fiber types. The epicuticle consists of a thin, three- to five-
layer outer epicuticle and a thicker inner epicuticle that is composed of epicuticular roots (er,
Fig. 5.3A) separated from one another by epicuticular canals (ec, Fig. 5.3A). The canals often
contain dense epicuticular fibers (ef, Fig. 5.3A). The predominant fibers in the exocuticle are
the chitin-protein horizontal fibers (hf, Fig. 5.3C), which run parallel to the cuticle surface and
rotate in successive planes. All other fiber types are oriented perpendicular to the cuticle sur-
face. Within the exocuticle, dense vertical fibers (dvf, Fig. 5.3A) extend from the epicuticle to
distal regions of the exocuticle. Vertical fibers (vf, Fig. 5.3B) are present in the distal and medial
exocuticle and are most likely contiguous with pore canal sheaths (pcs, Fig. 5.3C), which are
fibers associated with pore canal membranes in the medial and proximal exocuticle. Anchoring
fibers (af, Fig. 5.3C) are located in the medial and proximal exocuticle and traverse the cuticle by
intersecting bundles of horizontal fibers. Pore canal fibers (pcf, Fig. 5.3C) are proximally distrib-
uted and are associated with both the outer pore canal membrane and microtubules in the hypo-
dermis. Tonofibers (tf, Fig. 5.3D,E) and electron-dense rods (r, Fig. 5.3E) are specialized fibers
occurring in regions of muscle attachment. Uranyl acetate fixation differed from conventional
fixation in that it did not preserve cellular components, but it greatly increased the contrast of
all fiber types. Uranyl acetate fixation also permitted the visualization of the calcification initia-
tion sites along the epicuticle-exocuticle interface and interprismatic septa (IPS, Fig. 5.3F). The
interprismatic septa delineate an array of roughly hexagonal columns in the exocuticle that have
similar proportions and orientation as the lateral margins of the underlying hypodermal cells
that elaborate the cuticle (Giraud-Guille 1984, Compére 1995). Fig. 5.5 summarizes the distribu-
tion of the vertical and horizontal fibers in the epi- and exocuticle.
All three outer layers (epi-, exo- and endocuticles) are mineralized. The epicuticle is par-
tially calcified, with crystals nucleated at the epi-/exocuticular margin growing up between the
vertical fibers of the inner epicuticle (Hegdahl et al. 1977, Compére 1995, Dillaman et al. 2005).
Electron-dense, amorphous deposits are sometimes seen to be associated with the surface
layer, but it is unclear whether they are synthesized by the crustacean or are a matrix associ-
ated with microorganisms on the outer surface (Read and Williams 1991). Terminations of the
pore canals are also seen on the surface of the epicuticle in some taxa (Halcrow and Bousfield
1987). Within the exo- and endocuticles, mineral is generally in the form of fused spherulites
that align with the chitin-protein fibers (Roer and Dillaman 1984, Romano et al. 2007) (Fig. 5.6,
inset).
Originally, the form of the mineral was thought to be calcium carbonate in the form of cal-
cite crystals. While substantial amounts of crystalline magnesian calcite have been confirmed
in the carapace of lobsters and crabs (Boselmann et al. 2007), it is now clear that more than one
calcium salt may be involved (depending upon the location and the species), as well as noncrys-
talline, amorphous forms of mineral. Pratoomchat et al. (2002) demonstrated the presence of
calcium phosphate (as dicalcium phosphate dehydrate and octacalcium phosphate) as a pre-
cursor to the formation of calcium carbonate during postmolt mineralization in the carapace
of the crab Scylla. Soejoko and Tjia (2003) observed that calcium phosphate minerals persist
throughout the postmolt and intermolt stages in the carapace of the giant prawn Macrobrachium
rosenbergii. The calcium phosphate is present both in crystalline and amorphous forms and
coexists with calcium carbonate in nearly equal proportions. In the terrestrial isopods Porcellio
scaber and Armadillidium vulgare, Becker et al. (2005) found the cuticle to contain crystalline
The Crustacean Integument: Structure and Function 147

epi dvf vf
dvf vf
pc vf
pcs
pc
pcs
pcs
H

A B

af
af

pcf
pcf

C D

Fig. 5.5.
Diagrams illustrating the various vertical fiber types and features within the cuticle as well as their distri-
bution in tangential sections through the distal, medial, and proximal exocuticle. Each hexagonal prism
represents the cuticle overlying a single hypodermal cell. (A) Diagram showing the pore canals (pc)
extending from hypodermal cells (H) and dense vertical fibers (dvf) extending down from the epicuticle
(epi). (B) The distribution of vertical fibers (vf) and the pore canal sheath (pcs). Note that vertical fibers
are distal extensions of the more proximally located pore canal sheaths. (C) The distribution of anchoring
fibers (af). (D) The distribution of pore canal fibers (pcf). Adapted from Modla (2006, fig. 27d), used with
permission from the author.

magnesium calcite, amorphous calcium carbonate (ACC), and amorphous calcium phosphate.
These analyses were performed by X-ray diffraction and Fourier-transform infrared spectros-
copy (FTIR) of bulk specimens, so no precise distribution could be assigned to each of the
minerals and their morphs. However, the application of confocal-Raman spectroscopic imag-
ing has permitted mapping of the different minerals and their forms to the various cuticular
layers. In the isopods Porcellio and Armadillidium, Hild et al. (2008) clearly demonstrated that

Fig. 5.6.
Microstructure of purified chitin heat-treated at 220°C: detail images of the fibers near a pore canal.
Adjacent fibers seem to be connected by small fibrillar structures (circled areas, arrows in inset). From
Romano et al. (2007, fig. 7), with permission from Elsevier.
148 Functional Morphology and Diversity

Epi 1 2 3 4 5 6
A

100
Peak area/counts a.u.

80
60
40
20 c
a
0 b
0 5 10 15 20 25 30 35 40 45
Distance (μm)

Fig. 5.7.
Raman spectroscopic images (A–C) and line scans (D) recorded from a sagittally cleaved and microtome
polished surface of the mineralized tergite cuticles (shown with the epicuticle to the left) of the isopod
Armadillidium vulgare show the local distribution of the various components. Calcium carbonate (A)
occurs within the whole exo- and endocuticle, whereas calcite (B) is located within the exocuticle only.
Pore canals (arrowheads) appear devoid of mineral. The amount of organic material (C) increases from
the distal to the proximal region of the cuticle. The membranous layer (position 6) is devoid of calcium
carbonate. Horizontal dotted lines in the Raman images (A–C) indicate the position where the line scans
(Epi, 1–6) were recorded for carbonate (a), calcite (b), and organic material (c) to determine material dis-
tribution. From Hild et al. (2008, fig. 5), with permission from Elsevier.

calcite was entirely restricted to the exocuticle, while the mineral in the endocuticle was uni-
formly ACC (Fig. 5.7).
The key characteristics that the dorsal carapace should exhibit in order to protect the
organism from predation are hardness and fracture resistance. The helicoidally arranged
lamellae of chitin-protein fibers and associated minerals constitute a composite material
with exceptional biomechanical properties in this regard. The force required to puncture
the dorsal carapace of lobsters (measured with a punch test) is dependent upon the thickness
of the cuticle and ranges from 6.7 kg in Homarus to 27.8 kg in the slipper lobster Scyllarides
latus (Tarsitano et al. 2006).
As would be expected, the cuticle is far more resistant to compressive forces than to ten-
sile forces. In a compressive test in the normal dimension (i.e., perpendicular to the surface),
the walking leg cuticle of the sheep crab Loxorhynchus grandis exhibited a stress to fracture
of 101 ± 11 MPa, a value comparable to aluminum and in excess of limestone (60 MPa) (Chen
et al. 2008). In contrast, the cuticle failed in a tensile test in the normal dimension at only
9.8 ± 2.6 MPa. Interestingly, the cuticle was substantially more resistant to tensile stresses
in the longitudinal dimension (shear), with a stress to fracture of 31.5 ± 5.4 MPa (Chen
et al. 2008).
The Crustacean Integument: Structure and Function 149

HYPODERMIS AND THE MOLT CYCLE

The cuticle is underlain and periodically elaborated by the hypodermis in a cycle of deposition
and resorption referred to as the molt cycle (Drach 1939, Travis 1955, 1957, 1965, Skinner 1962,
Drach and Tchernigovtzeff 1967). These authors and others (as reviewed in Roer and Dillaman
1993) have attempted to standardize the terminology and the sequence of events of the molt
cycle by focusing on dynamics of the dorsal carapace. Recent observations have highlighted,
however, that this scheme may need to be modified for other tissues, as is detailed below.
When the cuticle of the dorsal carapace is fully elaborated, the animal is referred to as being
in intermolt (or stage C4). During this stage, the hypodermis is a simple epithelium that tends to
be squamous. While growth of the internal organs of a crustacean can occur during intermolt,
an increase in the external dimensions requires the shedding of the existing rigid exoskeleton.
A schematic representation of the molt cycle and associated cuticular changes is presented in
Fig. 5.1.
The first step in this process (referred to as early premolt or stage D0) is the separation of the
hypodermis from the cuticle, termed apolysis (Compére et al. 1998). The hypodermis is more
active during this period and becomes cuboidal and then columnar. The “old” cuticle is par-
tially degraded, and its components are resorbed during premolt stages D1 and D2 (Roer 1980,
Compére et al. 1998). Before the cuticle can enlarge, there must be an increase in the surface
area of the epithelium that secretes it and to which it is anchored by the pore canals. Stage D1 is
therefore the period of increased hypodermal mitotic activity that will result in the formation of
a larger new cuticle (Skinner 1965). The outer two layers of the new cuticle (the epi- and exocu-
ticles) begin to be deposited beneath the old cuticle during late stage D1 and stage D2. These are
often referred to as the preexuvial layers, and the organic components of these layers are thought
to be completely formed prior to the shedding of the old cuticle (termed ecdysis). It is important
to note (as discussed below) that mineralization of these layers cannot occur until the animal
emerges from the old exoskeleton and is fully expanded.
During this period of active cuticle synthesis, the individual columnar epithelial cells assume
a packing array that results in roughly hexagonal margins (Fig. 5.8A) and leads to the formation
of the interprismatic septa of the exocuticle. Furthermore, there are surface modifications of the
epithelial cells (Fig. 5.8A). The most obvious are cytoplasmic extensions resembling microvilli
that extend from the apical cell surface throughout the extracellular cuticular layers, referred
to as pore canals. Compére et al. (1998) have extensively documented the formation, extent, and
fate of the pore canals throughout the entire molt cycle. Another modification is the array of
short microvilli that constitute the “plaques” (Fig. 5.8B). These were first described in insects
(Locke 1961) and appear to be the sites of the initial polymerization and organization of the
developing cuticular components (Compére 1995, Greenaway et al. 1995, Elliott and Dillaman
1999). The plaques may provide the template for the orientation of the chitin-protein fibers that
permits them to interact with previously deposited layers. The fibers are then thought to self-
assemble in such a way that their plane of orientation is offset by a fixed angle relative to the
previous layer, thereby forming the Bouligand pattern of lamellae.
Once the organic lamellae of the preexuvial cuticle are fully formed, the next event is the
emergence of the crustacean from the old exoskeleton (exuviae). This extraordinary process of
ecdysis is initiated by the uptake of water. The resultant hydrostatic pressure causes the exuviae
to rupture at predetermined sites referred to as ecdysial sutures. Prior to the onset of postecdy-
sial tanning and mineralization, the mobility of the newly molted crab is possible because of a
hydrostatic support system. Taylor and coworkers (Taylor and Kier 2003, Taylor et al. 2007)
have demonstrated the transition from a hydrostatic skeleton to a rigid skeleton in the very early
postmolt period.
150 Functional Morphology and Diversity

Fig. 5.8.
(A) Scanning electron micrograph of early postmolt cuticle from Callinectes sapidus. Note the fracture
along the interface between the hypodermis (h) and the exocuticle (exo) revealing the roughly hexagonal
margins of the individual hypodermal cells (arrows). In the inset, note the pore canals extending from
the surface of the hypodermal cells (arrowheads). (B) Transmission electronic micrograph of dorsal cara-
pace from premolt cuticle of Callinectes sapidus: high magnification of cross section through the proximal
exocuticle and hypodermis. Note the assembly zone (AZ) and transition zone (TZ) above the microvilli
(mv) of a tendinous epidermal cell and the electron-dense apical plaques at the microvilli (arrowheads)
and abundant microtubules (mt) in the cytoplasm (from Modla 2006, fig. 8b, used with permission from
the author). (C) Backscattered electron micrograph of 24-hour postmolt cuticle from Callinectes sapidus.
Mineralized areas appear white. Note the calcified epicuticle (arrows) and interprismatic septa of the exo-
cuticle (arrowheads) as well as the more fully calcified endocuticle (endo) (from Dillaman et al. 2005, fig.
4d2, used with permission).

The onset of cuticular hardening may be brought on by the release of the hormone bursi-
con, which has been characterized as the agent responsible for the tanning of postmolt insect
cuticle. Bursicon transcripts have recently been found in the green crab Carcinus maenas
(Wilcockson and Webster 2008). While the precise mechanisms by which tanning and min-
eralization are prevented during premolt and initiated during postmolt are not fully under-
stood (and are addressed in more detail below), the patterns of postecdysial mineralization
are apparent.
Within 3 hours of ecdysis (stages A 1 and A 2) in the blue crab Callinectes sapidus, evidence of
calcium carbonate precipitation can be seen at the epicuticle/exocuticle boundary (Dillaman
et al. 2005). Fronts of mineralization then extend along the interprismatic septa, both dis-
tally and proximally, until they meet in the center of the exocuticle, forming calcified margins
delineating the hexagonal prisms (Fig. 5.8C). This general pattern of calcification has been
described in other species (Bouligand 1972, Giraud 1977, Giraud-Guille and Quintana 1982,
Giraud-Guille 1984, Sakamoto et al. 2009). In addition to a defined sequence characteriz-
ing the pattern of mineralization, the form of the mineral also follows a prescribed sequence.
Initially, the mineral at the epi-/exocuticular boundary is presumed to be ACC based on its
The Crustacean Integument: Structure and Function 151

solubility. Subsequently, the ACC is transformed into crystalline calcium carbonate in the
form of calcite as ACC is deposited along the interprismatic septa. This mineral also transi-
tions to calcite, completely encasing the prisms in crystalline material. The prisms themselves
then begin to calcify from the distal toward the proximal portion of the exocuticle; there is
some question as to the final mineral morph in these regions.
At the same time that the preexuvial layers are hardening (stages A 2, B1, and B2), the endo-
cuticle organic matrix is being deposited and, within a short time, calcified. The mineral
morph deposited in the blue crab has not been characterized, but that within the endocuti-
cle of isopods (Hild et al. 2008) and the lobster (Al-Sawalmih et al. 2008) is in the form of
ACC. Another component of the mineralized cuticle appears to be calcium phosphate, based
on elemental analysis (Pratoomchat et al. 2002, Soejoko and Tjia 2003), but X-ray diffraction
and FTIR spectroscopy fail to detect apatite, suggesting that this mineral is amorphous as
well. The deposition and mineralization of the endocuticle continues through postmolt stages
(C1–C3). The transition to intermolt is defined by the completion of the deposition of a non-
calcified membranous layer.

VARIATIONS IN CUTICLE T YPE AND STRUCTURE

Arthrodial Cuticle

Much of our knowledge regarding the control of tanning and mineralization is based upon
observed differences in the composition and the degree of mineralization in different cuticle
types from different regions of the body. Perhaps the most striking difference is between cuti-
cles that mineralize and those that do not.
The arthrodial cuticle is found, as the name implies, in the joints of crustacean appendages
and must remain flexible to allow for locomotion. The basic structure of arthrodial cuticle
and the timing of its deposition are similar to those of the dorsal carapace and the mineralized
cuticle with which it is contiguous. It has an epicuticle that is very similar to that of the dorsal
carapace and a Bouligand pattern of chitin-protein fibers that extend from the epicuticle to the
hypodermis. While the arthrodial cuticle of the blue crab has pore canals (the cuticle-encased
cytoplasmic extensions of the epithelial cells; Fig. 5.9A,B), some authors have stated that in
other species arthrodial cuticle lacks such structures (Raabe et al. 2007). The lamellar portion
of the arthrodial cuticle appears homogeneous, but staining of the carbohydrate moieties of the
cuticular glycoproteins clearly reveals that this lamellar portion can be divided into two lay-
ers (Fig. 5.9C). These two layers correspond spatially to the adjacent exo- and endocuticle of
the calcified cuticle, indicating that they are deposited at the same time. Indeed, the change in
thickness of the arthrodial cuticle mirrors that of the calcified cuticle in both pre- and postmolt
(Williams et al. 2003). Interestingly, the boundary between the two cuticle types is not perpen-
dicular to the cuticular surface; rather, the boundary is oblique to the surface (Fig. 5.9D). This
implies that a given patch of hypodermis must begin synthesizing arthrodial cuticle and change
to synthesizing calcified cuticle during the pre- and postmolt depositional periods. The advan-
tage of this phenomenon is that the interface between the calcified and noncalcified regions has
much greater surface area, reinforcing the junction between the two.

Branchial Chamber Cuticle

The cuticle that lines the branchial chamber of the crab is also noncalcified. In contrast to
the arthrodial cuticle, this structure is only about 6.5 μm thick in the blue crab. It is lamellar
152 Functional Morphology and Diversity

Fig. 5.9.
(A) Transmission electronic micrograph of arthrodial membrane of stage D4 Callinectes sapidus: cuti-
cle showing the numerous lamellae and cytoplasmic extensions (arrows) roughly perpendicular to the
surface of the cuticle. (B) Note the cytoplasmic extensions (arrows) extending into the newly deposited
arthrodial membrane and the abundant microtubules (arrowheads) in the cytoplasm of the hypodermal
cells. (C) Intermolt, stage C 4, arthrodial membrane of Callinectes sapidus stained with periodic acid-Schiff
and hematoxylin. Note the boundary between the preexuvial and postexuvial cuticle (arrowhead). (D)
Section of intermolt, stage C 4, cuticle of Callinectes sapidus at the boundary between the calcified cuticle
and arthrodial membrane (a) and stained with hematoxylin and eosin. Note the diagonal margin for both
the exocuticle (exo) and endocuticle (endo). From Williams et al. (2003, figs. 2c, 2d, 5f, and 5c), used with
permission.

but possesses only approximately 30 lamellae. A feature of such thin, noncalcified cuticles in
crustaceans is a modification of the timing of deposition during the molt cycle. The branchial
chamber cuticle is secreted in its entirety during premolt, being completed by the late premolt
stage, D3 (Elliott and Dillaman 1999). Despite the fact that this cuticle is completely elaborated
at the time of the molt, it is still subject to extreme tensile forces during extraction from the old
cuticle at ecdysis. To resist these tensile forces, the hypodermal cells underlying the branchial
chamber cuticle, which were cuboidal to columnar during their secretory phase, become filled
with microtubules along their long axis and assume the structure and function of tendon cells,
anchoring the new cuticle to the connective tissue below. These microtubule-filled cells persist
through ecdysis but during early postmolt revert to the intermolt/early premolt morphology.

Gill Cuticle

The cuticle that lines the outer surface of the gills not only protects these organs from dam-
age but also represents a potential barrier to the free diffusion of respiratory gases and the
transport of ions across the gill surfaces. Consequently, the gill cuticle is extremely thin. In
The Crustacean Integument: Structure and Function 153

the crayfish Procambarus clarkii , the cuticle surrounding the gill filaments is between 0.3 and
2 μm thick. Like the dorsal carapace, the gill cuticle usually consists of an epi-, exo-, and endo-
cuticle, albeit much thinner. In the crayfish, we observed differences among the cuticles of
the transporting filaments and the afferent and efferent channels of the respiratory filaments
(Dickson et al. 1991). The cuticle of the transporting filaments differs from the dorsal carapace
model not only in lack of mineralization and dimension (~2 μm thick) but also in the relative
proportions of the exo- and endocuticles. Whereas the endocuticle is the predominant layer
in the carapace, the composition of the cuticle of the transporting filament is approximately
two-thirds exocuticle.
The cuticle of the afferent channel of the respiratory filaments is approximately the same thick-
ness as the transport filament cuticle and is also composed of lamellate exo- and endocuticles, but
the proportions are more similar to the dorsal carapace, with the endocuticle representing more
than two-thirds of the thickness. The cuticle of the efferent channel of the respiratory filaments
is only 0.3–0.5 μm thick. It appears to possess an outer epicuticle, but the cuticle below does not
appear to be lamellate and has no obvious delineation into an exo- and endocuticle.
Like the branchial chamber cuticle, the gill cuticle is fully elaborated during premolt, with no
postmolt deposition (Andrews and Dillaman 1993). In fact, the formation of the respiratory cuticle
is delayed until late stage D2 to early D3, and deposition in the transporting gills is delayed until
early in the short, final premolt stage, D4. The net effect of these delays is that transport of ions and
diffusion of respiratory gases are subjected to a double barrier (old and new cuticular layers) for
only a very brief period of time. This principle may apply to the branchial chamber cuticle as well,
since it has been well documented that gas exchange may occur across this surface, particularly
in the case of semiterrestrial and terrestrial crabs (Greenaway and Farrelly 1984, 1990, Taylor and
Greenaway 1984).

Cheliped Cuticle

Whereas the dorsal carapace is adapted for resistance to compression and fracture propaga-
tion to protect the organism from predation, the claws must resist wear and be hard enough to
allow crushing of other mineralized structures. As mentioned in the discussion of biomechan-
ics above, one strategy for increasing hardness is simply increasing the thickness of the cuticle
(Tarsitano et al. 2006). Indeed, the cuticular thickness, even within the calcified cuticle, can
vary markedly from one region to another. Additionally, the relative proportions of exo- and
endocuticle is variable (Fig. 5.10). However, additional structural modifications of the cheliped
cuticle provide wear resistance and impart hardness. For example, the propodus of the cheliped
is some four times harder than that of the pereopods, within both the endocuticle (471 ± 50 vs.
142 ± 17 MPa) and exocuticle (947 ± 74 vs. 247 ± 19 MPa), as revealed by microindentation tests
on the sheep crab Loxorhynchus (Chen et al. 2008). Local differences in hardness have been
described within the claw of the lobster Homarus, the exocuticle of the lateral surface of the claw
being ~325 MPa and the tooth of the claw being 590 MPa. Similar differences were noted in the
stiffness (Young’s modulus) in these regions. The endocuticle of the lateral surface exhibited
a value of 6.9 GPa, compared to 11 GPa for the exocuticle (roughly equivalent to bone). The
endocuticle of the tooth had a stiffness value of 3.7 (comparable to polystyrene), while the tooth
exocuticle was measured at 25.7 (close to that for concrete) (Chen et al. 2008).
Melnick et al. (1996) compared the mechanical properties of the pigmented versus white
regions of the cheliped of the stone crab Menippe mercenaria and found that the black areas had a
higher density, elastic modulus, hardness, and fracture toughness compared to light areas. These
differences were attributed to a decreased porosity in the dark areas and perhaps a greater degree
of tanning. However, hardness may not be the desired characteristic in all chelipeds. Some
crabs employ the claws for rasping and grasping, rather than crushing, and these differences in
154 Functional Morphology and Diversity

Fig. 5.10.
Scanning electron micrographs of transversally fractured lobster cuticle from different body parts, expos-
ing the cross sections. (A) Cuticle from the claws; the exocuticle (exo) is thin in relation to the massive
endocuticle (endo). The detail image shows fibers oriented perpendicular to the fiber planes forming the
twisted plywood layers in the pore canals (pc) and lining them (arrows). (B) Cuticle from the carapace;
exo- and endocuticle have nearly the same thickness. In the detail image, fibers oriented in the normal
direction (arrows) are also present in the pore canals (pc). (C) Cuticle from the tergites; the exocuticle is
about twice as thick as the endocuticle. At a higher magnification, fibers oriented in the normal direction
(arrows) are again visible in the pore canals (pc). (D) Cuticle from the uropods; the exocuticle is very thick
in relation to the very thin endocuticle. Fibers oriented in the normal direction (arrows) are present in the
pore canals (pc), too, but their volume fraction is smaller than in the other mineralized parts of the lobster.
(E and F) Cross section of cuticle from the joint membranes (E), with detail image (F). No pore canals are
present in these unmineralized parts of the lobster; fibers oriented in directions other than in plane with
the cuticle surface cannot be observed. From Raabe et al. (2007, fig. 10), used with permission.

function are reflected in the structure of the cuticle. Cribb et al. (2009) studied the cheliped tips
in the grapsid crab Metopograpsus frontalis and found that this region was poorly mineralized but
contained high levels of halogens. The outer exocuticle was enriched with chlorine, while the
inner exocuticle had elevated bromine. The inner endocuticle also contained abundant chlorine
(Fig. 5.11). The cuticle tips were less hard and less stiff than the carapace but had values equiva-
lent to those found for insect cuticle lacking metals. It was hypothesized that the high levels of
The Crustacean Integument: Structure and Function 155

Fig. 5.11.
Scanning electron microscopic images of the cheliped tip of the crab, Metopograpsus frontalis.
(A) Backscattered electron image of M. frontalis cheliped tip: darker contrast indicates lower average
atomic number. Outer surface is to the left. (B–D) X-ray intensity maps for chlorine, bromine, and chlo-
rine plus bromine, respectively, across a cheliped tip from an area similar to that shown in A. X-ray inten-
sity maps use a thermal color scale; here, gray indicates the higher X-ray intensity for an element, white
lower, and black the lowest. Scale bars, 30 μm. From Cribb et al. (2009, fig. 1f,h,i,j) with permission from
Elsevier.

halogens imparted an increased hardness and low elastic modulus to the chitin-protein matrix.
The cheliped tips of Metopograpsus also have an unusual structural feature. Horizontal, rodlike
structures were observed in the fractured surfaces of the exocuticle (Fig. 5.12). The obvious anal-
ogy is that they resemble reinforcing rods in concrete.

Ecdysial Suture

Variations in cuticular structure may include local modifications of the cuticle. Such is the case
with the ecdysial suture, which is the narrow region of the ventral branchial carapace in crabs
that ruptures during ecdysis (Priester et al. 2005). Modifications include a significant decrease
in thickness, such that the location of the suture is visible to the naked eye in the intermolt crab.
Others include a decrease in mineral density and alteration in matrix composition. The decrease
in mineral density in the cuticle is achieved in a number of ways. First, the exocuticle mineralizes
only along the epicuticle/exocuticle boundary and the interprismatic septa. The prisms of the
exocuticle never fill with mineral. Second, the wedge of endocuticle immediately beneath the
less mineralized exocuticle has a demonstrably lower concentration of mineral than the adjacent
endocuticle (Fig. 5.13A). Third, while it is uncertain if the mineralogy of the suture differs from
the surrounding cuticle, elemental analysis conducted on the exo- and endocuticle components
showed that the suture contained significantly less magnesium (0.86 ± 0.29 vs. 1.75 ± 0.24 for
156 Functional Morphology and Diversity

Fig. 5.12.
Scanning electron micrographs of noncalcified cheliped and leg tips from Metopograpsus frontalis, in trans-
verse section. (A) Fractured surface across a cheliped tip showing multiple layers (2–5). Outer surface is to
the left. Scale bar, 20 mm. (B) Fractured surface of a cheliped tip showing rods (r) filling and protruding
from holes in the layer 3 region. Scale bar, 20 mm. From Cribb et al. (2009, fig. 2a,c), with permission from
Elsevier.

the exocuticle; 0.81 ± 0.30 vs. 1.31 ± 0.14 for the endocuticle). High magnesian calcite has a lower
solubility, so it is possible that this difference renders the mineral of the suture more soluble and
labile. Assuming that the composition and location of the mineral are affected by the associated
organic matrix, one would expect to be able to detect such differences between the suture and
the adjacent cuticle. In fact, lectin binding studies revealed such differences (Fig. 5.13B). The net
effect of these alterations is that during premolt cuticle resorption, the suture region is preferen-
tially removed and weakened (Fig. 5.13C–F) so that the increase in hydrostatic pressure in stages
D4 and E is sufficient to allow the old carapace to break open along the line of the suture. The
resorption is more intense at the posterior region of the suture line, thus allowing the carapace to
open from the posterior margin, making it possible for the crab to back out of the exuviae.

TEMPORAL VARIATIONS IN CUTICULAR SYNTHESIS

The timeline for deposition of some of the aforementioned types of cuticle is summarized in
Fig. 5.14. Considering the wide diversity in the timing of deposition among these few cuticle
types that have been examined, additional variations in depositional patterns will likely be
found as other regions and types are studied.
The temporal disparities in the onset and completion of cuticle synthesis among tissues within
a single organism certainly pose interesting questions regarding the control of the cellular events
involved in the molting process. The complexity represented by these differences speaks against
a regulatory system based upon universal changes in the titers of a small number of hormones.
Nowhere is this more evident than in the biphasic molting of isopod crustaceans, in which the min-
eral of the posterior cuticle is mobilized and stored as ACC in anterior sternal deposits, followed by
the molting of the posterior cuticle, remobilization of the mineral for calcification of the posterior
cuticle, and finally the molting of the anterior cuticle (Steel 1980, 1982, Ziegler et al. 2005, 2007). In
the sea roach Ligia exotica, the temporal separation of posterior and anterior molts is approximately
24 hours, despite having an open circulatory system presumably carrying hormones throughout
both halves of the body. In fact, injections of the molting hormone ecdysterone failed to alter the
delay in molting between the two halves of the animal (Montane 1988).
The Crustacean Integument: Structure and Function 157

Fig. 5.13.
(A) X-ray map of the calcium (Ca) distribution in an embedded and polished posterior piece of dorsal
carapace of Callinectes sapidus that includes the suture. Note the region of the suture in the exocuticle
(ex) as indicated by the arrowhead and in the endocuticle (en) as bounded by the arrows (from Priester
et al. 2005, fig. 7b, used with permission). (B) Intermolt cuticle of Callinectes sapidus in the region of the
suture stained with the fluorescein isothiocyanate–labeled lectin Lens culinaris agglutinin. Note the lack
of staining in the exocuticle (ex), the heavy staining in the suture (arrowhead), and the moderate staining
of the endocuticle (en) except in the region of the suture (arrow) (from Priester et al. 2005, fig. 1b). (C–F)
Backscattered electron micrographs of embedded and cut samples of Callinectes sapidus cuticle in early
(C) and late D2 (D–F) stages in the region containing the suture. Arrows indicate the suture region of the
endocuticle (en); arrowheads, the suture region of the exocuticle (ex). Note the different rate of etching in
the three regions of the cuticle: anterior (D), middle (E), and posterior (F) (from Priester et al. 2005, fig.
5). All figures used with permission.
158 Functional Morphology and Diversity

Apolysis Ecdysis
Intermolt Premolt Postmolt

Calcified
cuticle
Dorsal
C4 D0 D1’ D1’’ D1’’’ D2 D3 D4 A1 A2 B1 B2 C1-C3
Carapace

Arthrodial C4 D0 D1’ D1’’ D1’’’ D2 D3 D4 A1 A2 B1 B2 C1-C3


Membrane

Inner
Uncalcified
Cuticle
Cuticle

Respiratory
Gill

Transport
Gill

No deposition Postmolt deposition

Premolt deposition

Fig. 5.14.
Timeline for cuticle deposition in various tissues of Callinectes sapidus. From Williams et al. (2003, fig. 6),
used with permission.

CONTROL OF MINERALIZATION

As mentioned above, in order for animals to enlarge their exoskeletons and grow as they molt,
they must take up water soon after ecdysis and expand the new layers of the cuticle before they
begin to harden (Drach 1939, Mykles 1980). Only after this postecdysial expansion during stage
A1 do the new epi- and exocuticle undergo quinone tanning (Travis 1957) and mineralization
by deposition of calcium carbonate (Drach 1939, Bouligand 1970, Giraud-Guille and Quintana
1982). Thereafter, the endocuticle is synthesized and mineralizes as it is deposited (postecdysial
stages A 2 through C3). Obviously, the timing of the onset of tanning and calcium carbonate
deposition is of critical importance.
The deposition of mineral within a biological system requires a number of criteria to be met.
The prime criterion is the establishment of concentrations of the composite ions that exceed the
solubility product for precipitation. This generally occurs in the context of a microenvironment
with identifiable boundaries, one of which is commonly an epithelium that can transport ions
into and out of this compartment. Second, there is an organic component that serves to hetero-
geneously nucleate the mineral by lowering the superficial energy of precipitation. The organic
matrix also is responsible for determining the mineral morph and directing the second-order
pattern of mineralization.

Permeability Changes

The microenvironment within the calcifying cuticle is bounded on the inner surface by the
hypodermis, while the outer boundary is the epicuticle. Calcium is transported into this
space by the hypodermis via the action of a calcium-ATPase and sodium-calcium exchanger
(Roer 1980, Greenaway et al. 1995). The transport of bicarbonate is mediated by carbonic anhy-
drase and likely involves the activities of HCO3–-ATPase, a Cl–/HCO3– exchanger, and a Na+/
H+ exchanger (Giraud 1981, Roer and Dillaman 1993). However, during premolt, the preexu-
vial cuticle (including the epicuticle; Fig. 5.15A,B) is freely permeable to ions and breakdown
The Crustacean Integument: Structure and Function 159

Fig. 5.15.
(A–C) Transmission electronic micrograph of dorsal carapace from a 1-hour postmolt crab freshly fixed by
uranyl acetate: low (A) and high (C) magnification of cross sections through the epicuticle. (B) Tangential
sections through the inner epicuticle and distal exocuticle. Note the dense vertical fiber (dvf), epicu-
ticular canals (ec), epicuticular fibers (ef), epicuticular roots (er), horizontal fibers (hf), inner epicuticle
(ie), and outer epicuticle (oe) (from Modla 2006, fig. 10a–c, used with permission from the author). (D)
Scanning electron micrograph of the epicuticle in early postmolt cuticle in Callinectes sapidus. Note the
spaces between the perpendicular fibers of the epicuticle (arrowheads) and the continuous outer layer of
the epicuticle (arrows).

products of the organic matrix that pass through it during the resorption of the old exoskeleton
(Roer 1980, Compére et al. 1998). Following ecdysis, mineralization cannot commence until
the outer boundary of the cuticular microenvironment, the newly exposed epicuticle, becomes
impermeable to calcium and bicarbonate (see Fig. 5.15C,D). This transition occurs within 15 min-
utes postecdysis (Williams et al. 2009).

Matrix Proteins and Glycoproteins

Evidence suggests that the control of calcium carbonate deposition in crustacean cuticle resides
in the organic matrix of the cuticle itself, rather than in, for example, the ion-pumping activ-
ity of the underlying cellular layer (Roer and Dillaman 1984, Roer et al. 1988, Roer and Towle
2005). Using an in vitro nucleation assay, it was shown that pieces of cuticle isolated as early as
3 hours after ecdysis, stripped of all underlying cellular material, and decalcified with EDTA
160 Functional Morphology and Diversity

can nucleate calcium carbonate crystals, whereas pieces containing the same cuticle layers
removed from a preecdysial or an ecdysial animal and treated in the same way cannot nucleate
calcite (Roer et al. 1988, Shafer et al. 1995). That is, during preecdysis and early postecdysis,
the new cuticle is incapable of nucleating calcite. Soon after ecdysis, however, mineralization
commences.
Thus, in the preexuvial layers of the crustacean cuticle, we are presented with an organic
matrix that at one time (preecdysis and early postecdysis) cannot mineralize but that over a
period of only a few hours after ecdysis is transformed into a matrix that can and does mineral-
ize. This synchronized temporal separation of the definitive control events potentially allows
for isolation and identification of the regulatory proteins involved. Moreover, the cuticle con-
tains regions that calcify (e.g., the dorsal carapace) and regions that neither calcify nor tan (the
arthrodial cuticle). Comparisons of these two types of cuticle can provide an excellent means
to elucidate those cuticular proteins important for mineralization. Comparisons of cuticular
proteins across arthropod groups can also be instructive since insects do not mineralize their
exoskeletons. Finally, a search for matrix proteins that stabilize ACC is also instructive. Control
is necessary for the persistence of ACC in the cuticle since crystalline forms of calcium carbon-
ate (calcite and aragonite) are thermodynamically favored and more stable. Evidence for such a
matrix protein has recently been described by Shechter et al. (2008). A 65-kDa protein, GAP-65,
was found in the gastrolith (the storage organ for calcium carbonate in the crayfish). This pro-
tein binds ACC and inhibits the formation of calcite.
Dramatic postecdysial changes in crab cuticular glycoproteins synchronous with initial min-
eralization were documented at the histochemical level (Marlowe et al. 1994), by lectin blotting
following SDS-PAGE (Shafer et al. 1994, 1995) and by their ability to affect calcite nucleation
(Coblentz et al. 1998). Based on these data, it was hypothesized that acid-soluble proteins form
nucleating sites in the cuticle and that glycoproteins active only in preecdysis and for the first
few hours postecdysis shield nucleating proteins from calcium and carbonate ions or otherwise
inhibit their action (Coblentz et al. 1998). A cuticular glycoprotein that disappears from lectin
blots postecdysis (Shafer et al. 1995) was assumed to be the inhibitor. Its purification from rela-
tively large amounts of ecdysial cuticle provided strong support for this hypothesis (Tweedie
et al. 2004). This glycoprotein is mucinlike, being 55% carbohydrate and having both N- and
O -linkages. Immunoblot analysis suggests that several changes occur in its glycosylation pat-
tern during the first few hours after ecdysis. Immunohistochemical staining decreases in the
interprismatic septa as early as 2 hours after ecdysis, coincident with the first appearance of
mineralization. This temporal and fine-scale spatial correlation between the loss of a possible
inhibitor and the initiation of calcium carbonate deposition, and the fact that the protein does
not exist in the arthrodial membrane pre- or postecdysis, strongly suggests a role in control of
nucleation.
The means by which cuticular glycoproteins can be modified in situ after the molt is still a
matter of active investigation. However, evidence suggests that N-acetylhexosaminidase treat-
ment can alter cuticular glycans and change the ability of cuticle explants to calcify in a way that
mimics in vivo postecdysis processes (Pierce et al. 2001). A glycosidase with this activity appears
in the cuticle during the early postecdysial hours coincident with the changes in glycoprotein
profiles observed in vivo (Roer et al. 2001, Roer and Towle 2004).
Cuticular matrix proteins have been sequenced from several decapod crustaceans, either
directly or by virtual translation of cDNA sequences. They include multiple sequences from
Homarus (Kragh et al. 1997, Andersen 1998, Nousiainen et al. 1998), Cancer pagurus (Andersen
1999), Marsupenaeus japonicus (Endo et al. 2000, Watanabe et al. 2000), Callinectes (Wynn and
Shafer 2005, Faircloth and Shafer 2007), and Portunus pelagicus (Kuballa et al. 2007). Functions
for these proteins are at present only matters of speculation. However, a group of proteins from
The Crustacean Integument: Structure and Function 161

the crayfish Procambarus have been produced in recombinant systems and actually evaluated
for their ability to bind calcium carbonate. Calcification-associated peptides 1 and 2 (CAP-1 and
CAP-2), expressed in the crayfish tail fan during postecdysis, have been shown to affect in vitro
calcium carbonate formation and have been implicated as components that control the initiation
of mineralization (Inoue et al. 2001, 2003, 2004). CAP-1 can alter the structure of nanocrystals
deposited on chitin-coated surfaces (Sugawara et al. 2006). Additionally, Casp-2, a more soluble
Procambarus cuticular protein with an unrelated structure, also regulates calcium carbonate
formation when the recombinant protein is added to in vitro systems (Inoue et al. 2008).

CONCLUSION AND FUTURE DIRECTIONS

While there is considerable conservation in the basic architecture of the arthropod exoskeleton
across a wide array of taxa, there are also remarkable differences in cuticle characteristics. These
differences manifest across the crustacean taxa and even among different species of decapods.
Differences are also evident within an individual, depending upon the mechanical and physi-
ological function that the cuticle serves in different regions of the body. Cuticles in the gills of
aquatic crabs and the branchial chambers of terrestrial crabs are thin, nonmineralized, soft, and
highly permeable, while that of the cheliped is heavily mineralized, hard as concrete, and com-
pletely impermeable. Moreover, cuticles in different regions of the body of an individual crusta-
cean undergo synthesis and degradation during different times within the molt cycle, perhaps
most strikingly in the biphasic molting of isopods. These variations reflect temporal delays or
differential sensitivities of the underlying hypodermis to the hormonal cues that initiate these
events in preparation for the molt.
Clearly, the hypodermis holds the key to understanding the differences in cuticle structure,
function, and formation within the molt cycle. Recent work in our laboratory and others has
begun to elucidate the different expression of proteins and glycoproteins both spatially and tem-
porally and may explain the differences in cuticular structure and thus function. This approach
has perhaps been most successful in providing insight into the ability of cuticle to calcify (Inoue
et al. 2001, 2003, 2004, 2008, Wynn and Shafer 2005, Sugawara et al. 2006, Faircloth and Shafer
2007). The establishment and expansion of expression libraries for crustaceans hold great prom-
ise for discovering the cuticular components that are crucial for defining the differences in cuti-
cle morphology and function.

ACKNOWLEDGMENTS

We acknowledge the invaluable assistance of Mr. D. Mark Gay with the preparation of the fig-
ures. We thank the Drs. Raabe, Romano, Cribb, and Hild for supplying high-resolution images
of their figures and their publishers for permission to use them.

REFERENCES

Al-Sawalmih, A., C. Li, S. Siegel, H. Fabritius, S. Yi, D. Raabe, P. Fratzl, and O. Paris. 2008. Microtexture
and chitin/calcite orientation relationship in the mineralized exoskeleton of the American lobster.
Advanced Functional Materials 18:3307–3314.
Andersen, S.O. 1998. Characterization of proteins from arthrodial membranes of the lobster Homarus
americanus. Comparative Biochemistry and Physiology Part A 121:375–383.
Andersen, S.O. 1999. Exoskeletal proteins from the crab, Cancer pagurus. Comparative Biochemistry and
Physiology Part A 123:203–211.
162 Functional Morphology and Diversity

Andrews, S.C., and R. M. Dillaman . 1993. Ultrastructure of the gill epithelia in the crayfish Procambarus
clarkii at different stages of the molt cycle. Journal of Crustacean Biology 13:77–86.
Becker, A., A. Ziegler, and M. Epple. 2005 . The mineral phase in the cuticles of two species of Crustacea
consists of magnesium calcite, amorphous calcium carbonate, and amorphous calcium phosphate.
Dalton Transactions 2005:1814–1820.
Boselmann, F., P. Romano, H. Fabritius, D. Raabe, and M. Epple. 2007. The composition of the exoskel-
eton of two Crustacea: The American lobster Homarus americanus and the edible crab Cancer
pagurus. Thermochimica Acta 463:65–68.
Bouligand, Y. 1970. Aspects ultrastructuraux de la calcification chez les crabes. Septieme Congres
Internationale de Microscopie Electronique (Grenoble) 3:105–106.
Bouligand, Y. 1972. Twisted fibrous arrangements in biological materials and cholesteric mesophases.
Tissue and Cell 4:189 –190, 192–217.
Chen, P.-Y., A.Y.-M. Lin, J. McKittrick , and M.A. Meyers. 2008. Structure and mechanical properties of
crab exoskeletons. Acta Biomaterialia 4:587–596.
Coblentz , F.E., T.H. Shafer, and R.D. Roer. 1998. Cuticular proteins from the blue crab alter in vitro cal-
cium carbonate mineralization. Comparative Biochemistry and Physiology Part B 121:349 –360.
Compére, P. 1995 . Fine structure and morphogenesis of the sclerite epicuticle in the Atlantic shore crab
Carcinus maenas. Tissue and Cell 27:525–538.
Compére, P., M.F. Jaspar-Versali, and G. Goffinet. 2002. Glycoproteins from the cuticle of the Atlantic
shore crab Carcinus maenas: I. Electrophoresis and Western-blot analysis by use of lectins.
Biological Bulletin 202:61–73.
Compére, P., A. Thorez , and G. Goffinet. 1998. Fine structural survey of old cuticle degradation during
pre-ecdysis in two European Atlantic crabs. Tissue and Cell 30:41–56.
Cribb, B.W., A. Rathmell, R. Charters, R. Rasch, H. Huang , and I.R. Tibbetts. 2009. Structure, composi-
tion and properties of naturally occurring non-calcified crustacean cuticle. Arthropod Structure
and Development 38:173–178.
Dickson, J.S., R.M. Dillaman, R.D. Roer, and D.B. Roye. 1991. Distribution and characterization of
ion transporting and respiratory filaments in the gills of Procambarus clarkii. Biological Bulletin
180:154–166.
Dillaman, R., S. Hequembourg , and M. Gay. 2005 . Early pattern of calcification in the dorsal carapace of
the blue crab, Callinectes sapidus. Journal of Morphology 263:356 –374.
Drach, P. 1939. Mue et cycle d’intermue chez les crustacés décapodes. Annales d’Institut Oceanographie,
Paris 19:103–391.
Drach, P., and C. Tchernigovtzeff. 1967. Sur la méthode de détermination des stades d’intermue et son
application générale aux crustaces. Vie et Milieu 18:595–610.
Elliott, E.A., and R.M. Dillaman. 1999. Formation of the inner branchiostegal cuticle of the blue crab,
Callinectes sapidus. Journal of Morphology 242:47–56.
Endo, H., P. Persson , and T. Watanabe. 2000. Molecular cloning of the crustacean DD4 cDNA
encoding a Ca 2+-binding protein. Biochemical and Biophysical Research Communications
276:286 –291.
Fabritius, H.-O., C. Sachs, P.R. Triguero, and D. Raabe. 2009. Influence of structural principles on the
mechanics of a biological fiber-based composite material with hierarchical organization: The
exoskeleton of the lobster Homarus americanus. Advanced Materials 21:391–400.
Faircloth, L.M., and T.H. Shafer. 2007. Differential expression of eight transcripts and their roles in
the cuticle of the blue crab, Callinectes sapidus. Comparative Biochemistry and Physiology Part B
146:370 –383.
Giraud, M.M. 1977. Histochemistry of 1st stages of mineralization of Carcinus maenas cuticle—attempt
of localization of carbonic anhydrase. Comptes Rendus Hebdomadaires des Seances de l’Academie
des Sciences Serie D 284:1541–1544.
Giraud, M.M. 1981. Carbonic anhydrase activity in the integument of the crab Carcinus maenas during
the intermolt cycle. Comparative Biochemistry and Physiology Part A 69:381–387.
Giraud-Guille, M.M. 1984 . Calcification initiation sites in the crab cuticle: The interprismatic septa. An
ultrastructural cytochemical study. Cell and Tissue Research 236:413–420.
The Crustacean Integument: Structure and Function 163

Giraud-Guille, M.M., and C. Quintana. 1982. Secondary ion microanalysis of the crab calcified cuticle:
Distribution of mineral elements and interactions with the cholesteric organic matrix. Biologie
Cellulaire 44:57–67.
Greenaway, P., R.M. Dillaman, and R.D. Roer. 1995 . Quercitin-dependent ATPase activity in the
hypodermal tissue of Callinectes sapidus during the moult cycle. Comparative and Biochemical
Physiology Part A 111:303–312.
Greenaway, P., and C.A. Farrelly. 1984 . The venous system of the terrestrial crab Ocypode cordimanus
(Desmarest 1825) with particular reference to the vasculature of the lungs. Journal of Morphology
181:133–142.
Greenaway, P., and C. Farrelly. 1990. Vasculature of the gas-exchange organs in air-breathing brachy-
urans. Physiological Zoology 63:117–139.
Halcrow, K., and E.L. Bousfield. 1987. Scanning electron microscopy of surface microstructures of some
gammaridean amphipod crustaceans. Journal of Crustacean Biology 7:274–287.
Hegdahl, T., F. Gustavsen, and J. Silness. 1977. The structure and mineralization of the carapace of the
crab (Cancer pagurus L.): The exocuticle. Zoologica Scripta 6:101–105.
Hild, S., O. Marti, and A. Ziegler. 2008. Spatial distribution of calcite and amorphous calcium carbonate
in the cuticle of the terrestrial crustaceans Porcellio scaber and Armadillidium vulgare. Journal of
Structural Biology 163:100 –108.
Inoue, H., T. Ohira, N. Ozaki, and H. Nagasawa. 2003. Cloning and expression of a cDNA encoding
a matrix peptide associated with calcification in the exoskeleton of the crayfish. Comparative
Biochemistry and Physiology Part B 136:755–765.
Inoue, H., T. Ohira, N. Ozaki, and H. Nagasawa. 2004 . A novel calcium-binding peptide from the cuticle
of the crayfish, Procambarus clarkii. Biochemistry and Biophysics Research Communications
318:649 –654.
Inoue, H., N. Ozaki, and H. Nagasawa. 2001. Purification and structural determination of a phosphor-
ylated peptide with anti-calcification and chitin-binding activities in the exoskeleton of the cray-
fish, Procambarus clarkii. Bioscience Biotechnology and Biochemistry 65:1840–1848.
Inoue, H., N. Yuasa-Hashimoto, M. Suzuki, and H. Nagasawa. 2008. Structural determination and func-
tional analysis of a soluble protein associated with calcification of the exoskeleton of the crayfish,
Procambarus clarkii. Bioscience Biotechnology and Biochemistry 72:2697–2707.
Kragh, M., L. Mølbak , and S.O. Andersen. 1997. Cuticular proteins from the lobster, Homarus americanus.
Comparative Biochemistry and Physiology Part B 118:147–154.
Krywka, C., C. Sternemann, M. Paulus, N. Javid, R. Winter, A. Al-Sawalmih, S. Yi, D. Raabe, and M.
Tolan. 2007. The small-angle and wide-angle X-ray scattering set-up at beamline BL9 of DELTA.
Journal of Synchrotron Radiation 14:244–251.
Kuballa, A.V., D.J. Merritt, and A. Elizur. 2007. Gene expression profiling of cuticular proteins across the
moult cycle of the crab Portunus pelagicus. BMC Biology 5:45 .
Locke, M. 1959. The cuticular pattern in an insect, Rhodnius prolixus Stal. Journal of Experimental
Biology 36:459 –476.
Locke, M. 1960. The cuticular pattern in an insect—the intersegmental membranes. Journal of
Experimental Biology 37:398 –407.
Locke, M. 1961. Pore canals and related structures in insect cuticle. Journal of Biophysical and
Biochemical Cytology 10:589 –618.
Marlowe, R.L., R.M. Dillaman, and R.D. Roer. 1994 . Lectin binding by the crustacean cuticle: The cuti-
cle of Callinectes sapidus throughout the molt cycle, and the intermolt cuticle of Procambarus clarkii
and Ocypode quadrata. Journal of Crustacean Biology 14:231–246.
Melnick , C.A., Z. Chen, and J.J. Mecholsky. 1996. Hardness and toughness of exoskeleton material in the
stone crab, Menippe mercenaria. Journal of Materials Research 11:2903–2907.
Modla, S. 2006. The effect of fixation on the morphology of the late premolt and early postmolt cuticle of
the blue crab, Callinectes sapidus . MS thesis, University of North Carolina, Wilmington .
Montane, M.M. 1988. A study of the biphasic molt mechanism of the supralitoral isopod, Ligia exotica .
MS thesis, University of North Carolina, Wilmington .
Mykles, D.L. 1980. The mechanism of fluid absorption at ecdysis in the American lobster, Homarus
americanus. Journal of Experimental Biology 84:89 –101.
164 Functional Morphology and Diversity

Neville, A.C. 1975 . Biology of the arthropod cuticle. Springer, New York.
Nousiainen, M., K. Rafn, L. Skou, P. Roepstorff, and S.O. Andersen. 1998. Characterization of exoskel-
etal proteins from the American lobster, Homarus americanus. Comparative Biochemistry and
Physiology Part B 119:189 –199.
Pierce, D.C., K.D. Butler, and R.D. Roer. 2001. Effects of exogenous N-acetylhexosaminidase on the
structure and mineralization of the post-ecdysial exoskeleton of the blue crab, Callinectes sapidus.
Comparative Biochemistry and Physiology Part B 128:691–700.
Pratoomchat, B., P. Sawangwong , R. Guedes, M. de Lurdes Reis, and J. Machado. 2002. Cuticle
ultrastructure changes in the crab Scylla serrata over the molt cycle. Journal of Experimental
Zoology 293:414–426.
Priester, C., R.M. Dillaman, and D.M. Gay. 2005 . Ultrastructure, histochemistry, and mineralization
patterns in the ecdysial suture of the blue crab, Callinectes sapidus. Microscopy and Microanalysis
11:479 –499.
Raabe, D., A. Al-Sawalmih, S.-B. Yi, and H. Fabritius. 2007. Preferred crystallographic texture of а-chitin
as a microscopic and macroscopic design principle of the exoskeleton of the lobster Homarus ameri-
canus. Acta Biomaterialia 3:882–895.
Raabe, D., P. Romano, C. Sachs, A. Al-Sawalmih, H.-G. Brokmeier, S.-B. Yi, G. Servos, and H.G.
Hartwig. 2005a . Discovery of a honeycomb structure in the twisted plywood patterns of fibrous
biological nanocomposite tissue. Journal of Crystal Growth 283:1–7.
Raabe, D., P. Romano, C. Sachs, H. Fabritius, A. Al-Sawalmih, S.-B. Yi, G. Servos, and H.G. Hartwig.
2006. Microstructure and crystallographic texture of the chitin-protein network in the biological
composite material of the exoskeleton of the lobster Homarus americanus. Materials Science and
Engineering A 421:143–153.
Raabe, D., C. Sachs, and P. Romano. 2005b. The crustacean exoskeleton as an example of a structurally
and mechanically graded biological nanocomposite material. Acta Materialia 53:4281–4292.
Read, A.T., and D.D. Williams. 1991. The distribution, external morphology and presumptive function of
the surface microstructures of Gammarus pseudolimnaeus (Crustacea: Amphipoda) with emphasis
on the calceolus. Canadian Journal of Zoology 69:853–865.
Roer, R.D. 1980. Mechanisms of resorption and deposition of calcium in the carapace of the crab,
Carcinus maenas. Journal of Experimental Biology 88:205–218.
Roer, R.D., S.K. Burgess , C.G. Miller, and M.B. Dail. 1988. Control of calcium carbonate nucleation in
pre- and postecdysial crab cuticle. Pages 21–24 in C.S. Sikes and A.P. Wheeler, editors. Chemical
aspects of regulation of mineralization. University of South Alabama Publication Service, Mobile.
Roer, R.D., and R.M. Dillaman. 1984 . The structure and calcification of the crustacean cuticle. American
Zoologist 24:893–909.
Roer, R.D., and R.M. Dillaman. 1993. Molt-related change in integumental structure and function.
Pages 1–37 in M. Horst and J.A. Freeman, editors. The crustacean integument—morphology and
biochemistry. CRC Press, Boca Raton, FL.
Roer, R.D., K.E. Halbrook , and T.H. Shafer. 2001. Glycosidase activity in the post-ecdysial cuticle of the
blue crab, Callinectes sapidus. Comparative Biochemistry and Physiology Part B 128:683–690.
Roer, R.D., and D.W. Towle. 2004 . Partial nucleotide sequence of a putative cuticular hexosaminidase
from the blue crab, Callinectes sapidus. Bulletin of the Mount Desert Island Biological Laboratory
43:40–42.
Roer, R.D., and D.W. Towle, 2005 . Partial nucleotide sequence and expression of plasma membrane
Ca-ATPase in the hypodermis of the blue crab, Callinectes sapidus. Bulletin of the Mount Desert
Island Biological Laboratory 44:40 –42.
Romano, P., H. Fabritius, and D. Raabe. 2007. The exoskeleton of the lobster Homarus americanus as an
example of a smart anisotropic biological material. Acta Biomaterialia 3:301–309.
Sachs, C., H. Fabritius, and D. Raabe. 2006. Hardness and elastic properties of dehydrated cuticle from
the lobster Homarus americanus obtained by nanoindentation. Journal of Materials Research
21:1987–1995.
Sakamoto, K., W. Honto, M. Iguchi, N. Ogawa, K. Ura, and Y. Takagi. 2009. Post-molt processes of
cuticle formation and calcification in the Japanese mitten crab Eriocheir japonicus. Fisheries Science
75:91–98.
The Crustacean Integument: Structure and Function 165

Shafer, T.H., R.D. Roer, C. Midgette-Luther, and T.A. Brookins. 1995 . Postecdysial cuticle alteration in
the blue crab, Callinectes sapidus: Synchronous changes in glycoproteins and mineral nucleation.
Journal of Experimental Zoology 271:171–182.
Shafer, T.H., R.D. Roer, C.G. Miller, and R.M. Dillaman. 1994 . Postecdysial changes in the protein and
glycoprotein composition of the cuticle of the blue crab Callinectes sapidus. Journal of Crustacean
Biology 14:210 –219.
Shechter, A., L. Glazer, S. Cheled, E. Mor, S. Weil, A. Berman, S. Bentov, E.D. Aflalo, I. Khalaila, and A.
Sagi. 2008. A gastrolith protein serving a dual role in the formation of an amorphous mineral con-
taining extracellular matrix. Proceedings of the National Academy of Science USA 105:7129 –7134.
Skinner, D.M. 1962. The structure and metabolism of a crustacean integumentary tissue during a molt
cycle. Biological Bulletin 123:635–647.
Skinner, D.M. 1965 . Amino acid incorporation into protein during the molt cycle of the land crab,
Gecarcinus lateralis. Journal of Experimental Zoology 160:225–233.
Skinner, D.M., S.S. Kumari, and J.J. O’Brien . 1992. Proteins of the crustacean exoskeleton. American
Zoologist 32:470–484.
Soejoko, D.S., and M.O. Tjia. 2003. Infrared spectroscopy and X ray diffraction study on the morphologi-
cal variations of carbonate and phosphate compounds in giant prawn (Macrobrachium rosenbergii)
skeletons during its moulting period. Journal of Materials Science 38:2087–2093.
Steel, C.G.H. 1980. Mechanisms of coordination between moulting and reproduction in terrestrial iso-
pod Crustacea. Biological Bulletin 159:206 –218.
Steel, C.G.H. 1982. Stages of the intermoult cycle in the terrestrial isopod Oniscus asellus and their rela-
tion to biphasic cuticle secretion. Canadian Journal of Zoology 60:429 –437.
Sugawara, A., T. Nishimura, Y. Yamamoto, H. Inoue, H. Nagasawa, and T. Kato. 2006. Self-organization
of oriented calcium carbonate/polymer composites: Effects of a matrix peptide isolated from the
exoskeleton of a crayfish. Angewandte Chemie International Edition 45:2876 –2879.
Tarsitano, S.F., K.L. Lavalli, F. Horne, and E. Spanier. 2006. The constructional properties of the exoskel-
eton of homarid, palinurid, and scyllarid lobsters. Hydrobiologia 557:9 –20.
Taylor, H.H., and P. Greenaway. 1984 . The role of the gills and branchiostegites in gas exchange in a
bimodally breathing crab, Hothuisana transversa: Evidence for a facultative change in the distribu-
tion of the respiratory circulation. Journal of Experimental Biology 111:103–121.
Taylor, J.R.A., J. Hebrank , and W.M. Kier. 2007. Mechanical properties of the rigid and hydrostatic
skeletons of molting blue crabs, Callinectes sapidus Rathbun. Journal of Experimental Biology
210:4272–4278.
Taylor, J.R.A., and W.M. Kier. 2003. Switching skeletons: Hydrostatic support in molting crabs. Science
301:209 –210.
Travis, D.F. 1955 . The molting cycle of the spiny lobster, Panulirus argus Latreille. II. Pre-ecdysial his-
tological and histochemical changes in the hepatopancreas and integumental tissues. Biological
Bulletin 108:88 –112.
Travis, D.F. 1957. The molting cycle of the spiny lobster, Panulirus argus Latreille. IV. Post-ecdysial
histological and histochemical changes in the hepatopancreas and integumental tissues. Biological
Bulletin 113:451–479.
Travis, D.F. 1963. Structural features of mineralization from tissue to macromolecular levels of organiza-
tion in the decapod Crustacea. Annals of the New York Academy of Science 109:177–245.
Travis, D.F. 1965 . The deposition of skeletal structures in the Crustacea. 5. The histomorphological and
histochemical changes associated with the development and calcification of the branchial exoskel-
eton in the crayfish, Orconectes virilis Hagen. Acta Histochemica 20:193–222.
Tweedie, E.P., F.E. Coblentz , and T.H. Shafer. 2004 . Purification of a soluble glycoprotein from the
uncalcified ecdysial cuticle of the blue crab Callinectes sapidus and its possible role in initial miner-
alization. Journal of Experimental Biology 207:2589 –2598.
Watanabe, T., P. Persson, H. Endo, and M. Kono. 2000. Molecular analysis of two genes, DD9A and
B, which are expressed during the post molt stage in the decapod crustacean Penaeus japonicus.
Comparative Biochemistry and Physiology Part B 125:127–136.
166 Functional Morphology and Diversity

Wigglesworth, V.B. 1933. The physiology of the cuticle and ecdysis in Rhodnius prolixus (Triatomidae,
Hemiptara); with special reference to the function of the oenocytes and of the dermal glands.
Quarterly Journal of Microscopical Science 76:269 –318.
Wilcockson, D.C., and S.G. Webster. 2008. Identification and developmental expression of mRNAs
encoding putative insect cuticle hardening hormone, bursicon in the green shore crab Carcinus mae-
nas. General and Comparative Endocrinology 156:113–125.
Williams, C.L., R.M. Dillaman, E.A. Elliott, and D.M. Gay. 2003. Formation of the arthrodial membrane
in the blue crab, Callinectes sapidus. Journal of Morphology 256:260–269.
Williams, D.L., S. Modla , R.D. Roer, and R.M. Dillaman. 2009. Post-ecdysial change in the
permeability of the exoskeleton of the blue crab, Callinectes sapidus. Journal of Crustacean
Biology 29:550–555.
Wynn, A., and T.H. Shafer. 2005 . Four differentially expressed cDNAs in Callinectes sapidus contain-
ing the Rebers-Riddiford consensus sequence. Comparative Biochemistry and Physiology Part B
141:294–306.
Ziegler, A., H. Fabritius, and M. Hagedorn. 2005 . Microscopical and functional aspects of
calcium-transport and deposition in terrestrial isopods. Micron 36:137–153.
Ziegler, A., M. Hagedorn, G.A. Ahearn, and T.H. Carefoot. 2007. Calcium translocations during the
moulting cycle of the semiterrestrial isopod Ligia hawaiiensis (Oniscidea, Crustacea). Journal of
Comparative Physiology B 177:99 –108.
6
THE CRUSTACEAN INTEGUMENT: SETAE,
SETULES, AND OTHER ORNAMENTATION

Anders Garm and Les Watling

Abstract
The cuticle plays an important role in many aspects of crustacean biology, since it is the interface to
the surrounding world. Thus, the cuticle displays many structural specializations all over the body.
The structures considered here are setae, setules, denticles, and spines. We provide definitions for
them and discuss their functional morphology and development, with the main focus on setae.
We recognize seven types of setae based on their detailed external morphology: plumose, pap-
pose, composite, serrate, papposerrate, simple, and cuspidate. In support of the categorization of
these setae, each seems to correlate with a specific functional outcome such as feeding, grooming,
and locomotion. Setae are also important sensory organs, and in crustaceans they are normally
bimodal chemo- and mechanoreceptors, but there are also indications of thermo-, osmo-, and
hygrosensitivity. Little can be learned about the sensory functions from the external morphology
of setae, but their ultrastructure seems to provide better cues. In particular, mechanoreceptors
display structures related to transduction mechanisms, with the scolopale as a good example. Still,
too few data are available outside malacostracans to draw general conclusions for all crustaceans,
underlining the need for multidisciplinary and broad intertaxon studies. Less is known about the
functional morphology and development of setules and denticles in the general cuticle, but they
seem to be homologous with similar structures on the setae. Arthropods outside Crustacea also
have setae in their cuticle, and many shared features can be found. They are especially well studied
in insects, where many correlations between structure and function have been shown.

INTRODUCTION TO THE STRUCTURES OF THE CRUSTACEAN CUTICLE:


DEFINITIONS/CLASSIFICATION

One of the defining characters of crustaceans as well as other arthropods is their external skel-
eton, the cuticle (see also chapter 5). The cuticle plays a major role in most aspects of crustacean

Functional Morphology and Diversity. Edited by Les Watling and Martin Thiel.
167
© 2013 Oxford University Press. Published 2013 by Oxford University Press.
168 Functional Morphology and Diversity

biology, and this has led to a vast number of structural and functional specializations. Many
of these specializations lie within the detailed surface structures, and they are the topic of this
chapter. First, we provide an overview of the diversity of these structures and their functions
and use this to suggest a classification system. The main part of the review then focuses in detail
on the major group of cuticular specializations, the setae, since they are by far the most studied
and have the greatest functional diversity and importance. We end by comparing with data from
other arthropod groups and listing suggestions for where future research in this field is most
needed and will be most fruitful.
When observing crustaceans with the naked eye, many of the cuticular specializations are
visible in a large number of species (Fig. 6.1A). Some body parts and especially the appendages
appear furry (Fig. 6.1B), and the hairlike structures found in these areas are outgrowths of the
general cuticle, normally with a distinct articulation at the base, making them flexible (Figs.
6.1D, 6.2). There is a general consensus that these structures are homologous within Crustacea
and are also probably homologous with similar structures in other arthropods. Many terms have
been used for these structures, such as setae, sensilla, bristles, or even “hairs.” For crustaceans,
the most often used term is setae, and it will therefore be used here. Even though setae are in
general considered homologous, it is difficult to decide which cuticular projections to include
in this term. A number of authors have addressed this problem and provided definitions of what
they considered setae. Thomas (1970) was one of the first to do so, and he proposed that all
elongate outgrowths with distal pores were setae. His work was based on light microscopy, and
electron microscopy work has since shown his definition to be far too narrow. Fish (1972) con-
sidered elongate outgrowths filled with “cytoplasm” as setae, but this very broad definition will
include many other structures, such as spines (see below), and exclude setae with no cells in the
lumen. Some authors have used the size of the cuticular structures as a basis for classification.
This has led to such terms as microsetae (Jacques 1989) and microtrichs (Cuadras 1982, Steele
and Steele 1997, 1999), but we do not approve of this approach. If a structure complies with our
given definition (see below), we will consider it a seta no matter the size, and we see no reason to
believe that they cannot be small. In fact, we believe that in small crustaceans, such as nauplius
larvae, there has been strong selection pressure for miniaturizing the setae.
An evolutionary perspective was taken by Watling (1989), who stressed the need for a defini-
tion based on homologies. He suggested that the articulation with the general cuticle is such a
homology and used this structure to define setae from other cuticular outgrowths. This defini-
tion has been widely accepted as it seems to hold true for the vast majority of setae, and the “stem
seta” probably also had such an articulation. When considering the diversity of present-day
crustaceans, though, Watling’s definition runs into some problems, which were first addressed
in an earlier review (Garm 2004b).
Some of the articulated outgrowths have an external and internal morphology so similar to
long setules found on some of the setae that there are no structural arguments to consider them
as being different. They are commonly found on the mouthparts of decapods and peracarids
(Fig. 6.1C), and we suggest that they should be included in the term setules (see below). The
other problem concerns a loss of the articulation between the setae and the general cuticle.
This has probably happened a number of times in several crustacean lineages to encompass
mechanical functions requiring a very sturdy seta (Garm and Høeg 2001, Garm 2004a). Clear
examples of such loss are seen for the spinelike projection found on the basis of maxilla 1 of the
squat lobster Munida sarsi (Fig. 6.2B). These unarticulated projections are innervated, have a
continuous lumen, and have a cellular arrangement very similar to other setae. Further, struc-
tures undoubtedly homologous with the spinelike projections (they are situated in the same
place and arranged in the same two parallel rows in other decapod species) are typical setae
with clear articulation (Garm 2004b). The same situation is seen for unarticulated spinelike
Setae, Setules, and Other Ornamentation 169

A B

C D

E F G

Fig. 6.1.
Structures in the crustacean cuticle. (A) At the macroscopic level, many crustaceans, such as the hermit
crab Parapagurus sulcata, appear furry because of very heavy setation (picture courtesy of Dr. Jens T.
Høeg). (B) Maxilliped 1 of the hermit crab Pagurus bernhardus displaying heavy setation, especially on the
medial edges of the coxa and basis. Several types of setae are present. (C) Setules from paragnath of P. bern-
hardus are clearly articulated with the general cuticle (inset). (D) Between the setae (S) on the mouthpart
of Panulirus argus, the cuticle is filled with teethlike structures (denticles). (E) Ultrastructure of setules
from the paragnath of Penaeus monodon shows that they are made entirely of cuticle and lack a lumen and
innervation. (F) Ultrastructure of setae show a round, hollow base filled with sheath cells (ShC). Cu, setal
cuticle. (G) Close-up of the central part of the setal lumen showing that the semicircular sheath cells (ShC)
encircle the outer dendritic segments (ODS) of a number of sensory cells.
170 Functional Morphology and Diversity

Fig. 6.2.
Details of the external morphology of setae. (A) Spinelike setae from maxilla 1 of Munida sarsi, with no
apparent articulation at the base (arrows). (B) Plumose setae displaying a supracuticular articulation
(arrows) with the general cuticle, making them very flexible. (C) Most setae have an infracuticular articu-
lation (arrows) with the general cuticle, reducing their flexibility. (D) Setules from a composite seta show-
ing an articulation (arrows) with the setal shaft. (E) Some setae display unarticulated teethlike structures
arranged parallel to the setal shaft. These denticles (De) are found in two rows on the distal half of the
setae often together with small setules (S). (F) On some setae, there is a graduated transformation between
setules (S) and denticles (De). (G) Many setae display a terminal pore (TP) at the tip, often associated with
small scalelike setules. (H) The tip of a seta used for grooming the gills. Such setae often have a specialized
tip. (I) A newly molted composite seta displaying a very distinct annulus (ringed) as a by-product from the
invagination during development.
Setae, Setules, and Other Ornamentation 171

setae on the dactylus of maxilliped 3 of the shrimp Palaemon adspersus. The definition we will
follow here was put forward by Garm (2004b, 1): “A seta is an elongate projection with a more
or less circular base and a continuous lumen. The lumen has a semicircular arrangement of
sheath cells basally.”
The available data on the ultrastructure of setae provide good support for the internal charac-
teristics—the continuous lumen and the semicircular sheath cells (Fig. 6.1F,G), also called envel-
oping cells (Alexander et al. 1980, Hallberg et al. 1992, Crouau 1997, Paffenhöfer and Loyd 2000).
That the sheath cells seem to be a unifying character for setae indicates that they play important
functional roles. They are involved in setal development, and this complex process could possibly
provide a more detailed definition. The continuous lumen is also functionally significant since it is
closely connected with the sensory properties of setae. Both of these topics are discussed in detail
in later sections.
It is often problematic to use internal characters because categorization is normally based
on light or scanning electron microscopy. The round shape of the basal part of a seta is there-
fore an important character, and it seems to be very consistent for setae found on all body parts
of many groups of crustaceans (see Garm 2004b for review). Still, using this character alone
will not suffice since it will not separate unarticulated setae and spines dealt with below.
Besides setae, there are other surface structures of the cuticle that we will briefly consider.
One group has already been mentioned—the setules. As said above, this is a term widely used
for certain outgrowths on setae, but we believe them to be a general feature of the cuticle. They
are elongate structures, 10–150 μm long, often inserted into the cuticle in a socket, making
them flexible (Fig. 6.1C). They are flattened in cross section and made entirely of cuticle, so
they are never innervated and do not contain semicircular sheath cells basally (Fig. 6.1E). Most
often they have a serrated edge distally. Such setules are commonly found in the general cuticle
throughout the Crustacea, especially on the mouth apparatus and in the foregut, but they have
typically been referred to as setae (Halcrow and Bousfield 1987, Holmquist 1989, Martin 1989,
Olesen 2001).
Another expression often used when describing the cuticle of crustacean is denticles. Like set-
ules, denticles are commonly found on setae but also in the general cuticle. This again stresses that
some of the structures generally considered special features of setae are in fact general cuticular
characteristics. Denticles are relatively small structures (normally < 30 μm long) and, as the name
implies, more or less tooth shaped (Fig. 6.1D). They are unarticulated, made entirely of cuticle,
and never innervated. There is some evidence that they are in fact evolutionarily related to setules
(Garm 2004b).
A common cuticular outgrowth is the spine. This term should be used with care since it
can be very hard to tell a true spine from an unarticulated seta. We consider a spine to be an
unarticulated extension of the general cuticle. It is hollow, and the lumen is lined with normal
epithelial cells; no innervation is present unless the spine carries setae (see, e.g., Martin and
Cash-Clark 1995, their fig. 19A,C). If ultrastructural data are not available, then comparison with
closely related species should be used to verify that they do not have setae in the same position.
While the other types of structures are probably homologies, we find it very likely that spines
have arisen several times and represent convergent evolution.
Scales, like spines, are cuticular outgrowths, but they are generally wider than long and are
not usually hollow (Klepal and Kastner 1980). Most often, scales follow one side of the outline
of the polygons often visible on crustacean cuticles when observed with the scanning electron
microscope.
While it was long known that the crustacean cuticle was often sculptured, the exact
details could not be seen until the invention of the scanning electron microscope. Some of
the main features are summarized by Meyer-Rochow (1980), Holdich (1984), and Halcrow
172 Functional Morphology and Diversity

and Bousfield (1987). A terminology of surface sculpturing was proposed by Harris (1979)
for insects, but it seems equally applicable to crustaceans. The basic unit of sculpturing
seems to be a more or less well-defined polygon, which Hinton (1970) and Duncan (1985)
assert represents the surface mani festation of underlying epidermal cells (see also chap-
ter 5). Scales, microspines, micropores, and a large variety of other structures can be found
within and along the boundaries of polygons (e.g., Klepal and Kastner 1980, Halcrow and
Bousfield 1987). In many other crustaceans and insects, however, the polygon is obliter-
ated by cuticular secretions that form more elaborate sculpture (e.g., Hinton 1970, Meyer-
Rochow 1980).

THE EXTERNAL STRUCTURE OF SETAE

As discussed in the preceding section, setae constitute the largest and most diverse group of
structures, which is also why providing a general definition is not a trivial task. This diversity
is seen between species, but sometimes a single species carries close to the full diversity of seta
types. Most of the setae are found on the appendages, and especially the mouthparts are heav-
ily ornamented with setae, and a single segment (= article) of, for example, a maxilliped can
display quite a number of setal types (Fig. 6.1B). In the following we will try to deconstruct this
diversity and pinpoint some of the important structures that cause the diversity. Structures that
share some kind of similarities can be a product of the evolutionary history of setae and thereby
be considered homologies, and/or they can be products of shared functionality. As we discuss
further below, most of the similarities of setae stem from shared functions, and this will be used
to suggest a classification system.
First, it is important to recognize that all setae can be seen as having a more or less elongate
and round (at least at the base) central part, the shaft, which may or may not have specializations,
including different types of outgrowths. The length of the shaft varies from just a few microm-
eters to several millimeters in large decapods. They are found on all body parts, including inter-
nally in the foregut (Altner et al. 1986, Johnston 1999), and serve many different functions. This
diversity of function has undoubtedly added to the wide range of external morphology. Some
setae are long and slender with no apparent specializations along the setal shaft (Fig. 6.2F,G),
whereas others have many types of outgrowths, resembling feathers or pine trees (Fig. 6.2A,B).
Still others are thorn shaped or bent and appear as hooks. Despite the diversity, several substruc-
tures can be recognized in many setae and can be used to group the setae into different types. If
a classification includes too many details, there is a high risk that the designated setal types will
be highly specialized and appear only in a very limited number of taxa. Here, we try to avoid this
problem and consider only overall structural similarities found in most major crustacean taxa,
since this will have the broadest application and interest.
One of the prominent substructures concerns how the setae attach to the general cuticle.
Three types of attachments are seen: (1) an articulation in the form of a socket, which is drawn
into the general cuticle and gives the seta an infracuticular articulation (Fig. 6.3C)—this is by
far the most common type of attachment; (2) the socket is extended from the general cuticle,
giving the seta a supracuticular articulation (Fig. 6.3B)—this gives the seta great flexibility and
is often seen in setae experiencing large drag forces; (3) no articulation is seen, and the general
cuticle has a direct transition into the cuticle of the seta (Fig. 6.3A)—as mentioned earlier, the
articulation is probably reduced to obtain sturdiness.
Another feature concerning the sturdiness is the length:width (L:W) ratio of the shaft, where
the width is measured at the base of the seta. The vast majority of setae are slim, with a L:W ratio
of >15 (Fig. 6.2), but some setae are more stout and robust, with a L:W ratio <8 (Fig. 6.2G). Surely
there are setae with intermediate L:W ratios, but they seem to be rare.
Setae, Setules, and Other Ornamentation 173

Fig. 6.3.
Setal types. (A) Pappose setae, with long setules scattered along the entire shaft, from the mandibular
palp of Cherax quadrocarinatus. (B) Plumose setae, with two straight rows of long setules, from maxil-
liped 1 of Munida sarsi. (C) Composite setae, with small setules distally, from maxilliped 1 of Panulirus
argus. (D) Serrate setae, with two very distinct row of denticles, from maxilla 1 of C. quadrocarinatus. (E)
Papposerrate setae, with denticles and small setules distally and long setules proximally, from maxilla 2
of M. sarsi. (F) Simple setae, with an almost complete lack of surface structures, from maxilliped 1 of P.
argus. (G) Cuspidate setae, from maxilliped 2 of P. argus. Some cuspidate setae can have small setules on
the middle part.

An annulus situated on the proximal half of the shaft is found on most if not all setae (Fig. 6.3I)
(Garm 2004b). The annulus seems to have no function as such but is a by-product of the onto-
geny, when the setae develop in an invaginated state (for more details, see “Setal Development
and Ontogeny,” below). The annulus is most easily detected in newly molted animals since it
often diminishes as the cuticle thickens, stretches, and wears down during the intermolt stage
(Garm 2004b).
174 Functional Morphology and Diversity

The tip of the seta also varies (Fig. 6.3F–H) but is always more or less pointed. In cases of
a seta with outgrowths, the latter often form the very tip. A specialization of the tip is seen in
peracarids, where some setae have a bifurcate tip (Fish 1972, Brandt 1988). Interestingly, the
thin “additional” tip holds all the sensory cells (Brandt 1988), and it might be a way of sepa-
rating mechanical and sensory functions. Another functionally interesting feature of the tip
is the presence or absence of a pore. The pore-bearing setae may have the pore in one of two
different positions: terminal or subterminal. Most common is a terminal pore situated at the
very tip of the seta, often bending to make the pore point to the side of the seta (Fig. 6.3G).
Less common is a subterminal pore situated on the side of the seta a little proximal to the
tip. In at least some cases, the pore is associated with chemoreception (see later section for
detailed discussion).
The main substructures causing the diversity of setae are the presence or absence of out-
growths on the setal shaft and, when present, their detailed structure and arrangement. We
divide the outgrowths into two types: denticles and setules. Denticles as defined above are
rather small (normally <30 μm long), flat, and pointed outgrowths with smooth edges and no
articulation with the setal shaft. They are solid cuticle; that is, they lack a lumen. On the setae
they occur in two parallel rows, are always oriented with their broad axis parallel to the setal
shaft, and point distally (Fig. 6.3E). Denticles arranged in this way are found distal to the
annulus.
Setules have a wide size range (2–150 μm long), but they share common features. They all
have an articulation with the setal shaft, although, especially in smaller setules, the articulation
is often weak. They are all flattened, with their broad axis perpendicular to the setal shaft at the
point of attachment. Setules taper distally and often have a serrate edge, with most of the minute
tooth-shaped extensions distally (Fig. 6.3D). Like denticles, setules are made of solid cuticle.
Setules always point toward the tip of the seta, but the angle changes with size. Long setules may
be almost perpendicular with the setal shaft, whereas small setules often lie almost flat against
the shaft (often referred to as scales). Long setules can be found along the whole length of a seta,
whereas small setules are normally found only distal to the annulus. They can form straight rows,
but most often they appear to be randomly arranged on the shaft. Even though denticles and set-
ules are normally very distinct and easy to separate, there are intermediate forms. In some cases,
rows of outgrowths gradually changing from setules to denticles (from base to tip) can be found
on the same seta (Fig. 6.3F).

TOWARD A GENERAL CLASSIF ICATION SYSTEM OF CRUSTACEAN SETAE

Setae are very diverse both in function and in morphology, thus necessitating a general clas-
sification system that will enable carcinologists to compare results obtained from a broad range
of taxa within Crustacea and possibly also with other arthropods. Many studies have provided
a setal classification system, but unfortunately, they are often based on a single or a few closely
related species and make use of a large number of structural details (e.g., Garm and Høeg 2000,
Coelho and Rodrigues 2001). While that approach provides good insight into the setal diversity
of the species studied, it has little or no value when setae in general are considered. Here we
try to set the basis for a more general classification system that can be used for most (if not all)
crustacean taxa.
We suggest that the vast majority of crustacean setae can be subdivided into seven catego-
ries or types based on the external morphological characters listed earlier and, to some extent,
their mechanical functions: plumose, pappose, composite, serrate, papposerrate, simple, and
cuspidate (see also Table 6.1). Note that the term serrulate used in earlier work (Watling 1989,
Setae, Setules, and Other Ornamentation 175

Table 6.1. Summary of the characteristics of the seven types of setae.

Seta type Annulus Articulation Long Short Denticles Length: Pore


setules setules width ratio
Pappose + Infra + +/– – >15 –
Plumose + Supra + – – >15 –
Composite + Infra – + – >15 Ter/–
Serrate + Infra – +/– + >15 Ter/
Sub/–
Papposerrate + Infra + +/– + >15 Ter/–
Simple + Infra – – – >15 Ter/–
Cuspidate + Infra/absent – +/– – <8 Sub/–

Abbreviations: Infra, infracuticular articulation; Supra, supracuticular articulation; Ter, terminal; Sub, subterminal; +,
present; –, absent.

Garm 2004b) has been changed to composite. The reason is that serrulate, meaning “small ser-
rate,” indicates the presence of denticles, which is never the case in these setae and why we
found it inappropriate. The reason for excluding the internal structures and thereby almost
all aspects of their sensory functions from this morphological description are that (1) we find
it unrealistic to expect that investigators in general can include these data since the quality of
the material (e.g., ethanol-fixed museum specimens) and time often do not allow this, and (2)
at present there are too few ultrastructural data at hand outside decapods to draw trustworthy
general conclusions on the functional morphology and/or evolutionary history of the internal
structures. However, internal structure is very important for inferring function, as we illus-
trate below.
The morphological characters used to group the setae are not put forward as suggestions of
homologies, even though some of them might be, and thus, the seven types of setae are not to
be considered as separate evolutionary lines. It is in fact very likely that some of the types (e.g.,
the simple setae) are based on shared morphological characters that are convergently derived.
There is little doubt that strong selection pressures have acted on the external morphology of
setae, which comes from the many mechanical functions they serve. The outgrowths of setae
directly involved in food handling, for example, should be closely correlated with the nature of
the food items. Such a situation is bound to result in convergences, and this may to a large extent
impede a homology-based classification system.
Pappose setae. The shaft of pappose setae is long and slender, and they never display a pore.
They have long (50–150 μm) setules with clear articulations scattered randomly along the
entire length of the shaft (Fig. 6.2A). The setules have a serrated edge, with most teeth situated
distally, and they normally project with an angle between 45° and 90° to the setal shaft. The
socket of pappose setae is infracuticular.
Plumose setae. Like pappose setae, plumose setae have long setules along the entire shaft,
but they are arranged in two strict rows on opposite sides of the shaft, giving them a feather-
like appearance (Fig. 6.2B). The setules are never serrated, have a weak articulation with the
setal shaft, and are often situated in a groove. Plumose setae are the only setae that may have a
supracuticular articulation, and they never display a pore.
Composite setae. Composite setae are slim, with a naked proximal part, but have small setules
(<15 μm long) distal to the annulus (Fig. 6.2C). The setules can be arranged in rows or occur
176 Functional Morphology and Diversity

randomly along the shaft. The setules are usually smaller in size toward the tip and, for the most
part, point toward the tip of the seta with an angle of <45°. The setules on composite setae can
be leaf shaped, digitate, or palmate. Composite setae may or may not have a terminal pore, and
the socket has an infracuticular articulation.
Serrate setae. Serrate setae have a naked proximal half, but distal to the annulus they have two
rows of denticles pointing toward the setal tip, with 120–180° between them (Fig. 6.2D). The dis-
tal half may also have setules on the opposite side of the shaft from the denticles. When present,
the setules are small, have teeth along their edge, and lay almost flat against the shaft. Serrate
setae can have a terminal or a subterminal pore, and the articulation with the general cuticle is
infracuticular and often narrow.
Papposerrate setae. Like pappose and plumose setae, papposerrate setae are long and slender
(Fig. 6.2E). On their proximal half to two-thirds (proximal to the annulus), they have long, ran-
domly arranged setules like pappose setae, but on the distal part they have two rows of denticles
like serrate setae. In the area with the denticles, there may be additional small setules on the
opposite side of the denticles. The articulation of papposerrate setae with the general cuticle is
infracuticular and normally broad.
Simple setae. Simple setae are long and slender, and as the name implies, they completely
lack outgrowths on the setal shaft (Fig. 6.2F). They have an annulus typically one-third of the
way up the shaft and a pointed tip, which may or may not have a terminal pore. They have an
infracuticular articulation.
Cuspidate setae. Cuspidate setae are very robust, with a L:W ratio <8 when width is meas-
ured at the base of the seta (Fig. 6.2G). They have a broad base and taper gradually toward the
somewhat rounded tip. They may or may not have a subterminal pore, and in most cases they
have no outgrowths. They can have small setules in the midregion, often combined with a
weak curvature of the setal shaft with the outgrowths on the concave side. The distal part is
almost always naked but may be microtuberculate. In most cases they have an infracuticular
articulation with a very reduced membranous area, but sometimes the articulation is lacking
completely.
In our opinion, the vast majority of crustacean setae are easily put in one of the seven
described categories, but there is morphological variation within the groups, and this some-
times results in intermediate types of setae being described. Intermediate forms are occasion-
ally found between the following types. (1) Pappose and plumose, where the proximal part has
two distinct rows of setules, but on the distal part they are randomly spaced. (2) Pappose and
papposerrate, where the proximal part has long, randomly arranged setules, but the distal part
has short setules arranged in rows. (3) Pappose and composite, where the proximal part has a
few long setules, and the distal part has small or medium sized setules. (4) Composite and ser-
rate, where the proximalmost outgrowths are setules, but they change gradually into denticles.
(5) Composite and simple, where the seta has a few small setules on the distal part. (6) Serrate
and cuspidate, where the outgrowths are denticle-like, but the distal part is naked, and the seta
has a L:W ratio between 10 and 15. (7) Cuspidate and simple, where the seta has no outgrowths
and a L:W ratio between 8 and 15.
It should again be noted that by far the most available data stem from malacostracans, espe-
cially decapods and peracarids. This can, of course, turn out to be fatal for the suggested clas-
sification system, but when going through the descriptions of nonmalacostracan crustaceans, it
seems that it will apply to those groups also (see Garm 2004b for more details). The problem is
that none of the papers directly deal with setae, and therefore the pictures they present often do
not show the needed details. Further, since the authors did not make a survey of the seta, there
may be many other setae that are not shown. Hence, we might be able to say something about
which of these seven types are present but very little about whether additional types should be
named.
Setae, Setules, and Other Ornamentation 177

SETAL DEVELOPMENT AND ONTOGENY

The development of setae can be separated in two different events: (1) the development of a
new seta, and (2) the reconstruction of a seta during a molt. The first case is the more com-
plex and unfortunately only rarely studied. What little is known about formation of new setae
stems almost entirely from malacostracan olfactory setae, called aesthetascs. In decapods, the
aesthetascs are situated in discrete rows on the lateral flagellum of antenna 1. The flagellum
is annulated, and each annulus has one or more rows of setae. With each molt, the distalmost
annuli with their aesthetasc are shed, and new annuli with aesthetascs are added proximally on
the flagellum (Derby et al. 2002, Ekerholm and Hallberg 2002). In the spiny lobster Panulirus
argus, in antenna 1 a cell proliferation zone is found in the proximal part corresponding to the
area where the new aesthetascs will be situated (Steullet et al. 2000a, Derby et al. 2002). Here
new olfactory receptor neurons (ORNs) are formed continuously but with peaks in the prolifer-
ation rate occurring in the premolt state. The ORNs form in discrete clusters of about 300 cells
that will innervate a single aesthetasc (see further below for more details). In these studies, little
attention has been paid to the development of the rest of the seta, however. Earlier studies on
the development of the aesthetascs in peracarids included observations on the role of the sheath
cells and cuticle formation (Guse 1983). Here it was also shown that already in the intermolt
stage the new ORNs are formed and that the sheath cells are also present to encircle them in a
cavity filled with receptor lymph. The number of sheath cells seems to vary, depending on the
size of the seta. Large and thick decapod setae can have as many as 50 sheath cells (Fig. 6.1F,G).
In the small setae of the isopod Idotea baltica, eight sheath cells were most commonly found
(Guse 1983). During the premolt, the sheath cells protect the ORNs against the enzymes in the
exuvial fluid and start making the cuticle of the future seta. This process happens in an invagi-
nation of the epithelium, and the proximal part of the setal shaft is therefore also invaginated
(Fig. 6.4A). The sheath cells seem to work in pairs, and in the isopod aesthetascs, the two inner-
most ones form the distal part of the setal shaft (Guse 1983). The next two pairs form the middle
part, and the last pair forms the basis of the shaft and the socket. When the old cuticle is shed,
hydrostatic pressure everts the seta with its new soft cuticle. The annulus is left as a “scar” in the
new cuticle, indicating where the cuticle was folded during its formation.
The other type of setal development happens when the setae reform during a molt. Again,
little is known about the cellular processes, and the only ultrastructural data available stem from
malacostracan aesthetascs (Guse 1980, 1983) and for bimodal chemo- and mechanosensitive
composite setae of the shrimp Palaemon adspersus (Fig. 6.4C,D). In the premolt stage, the epi-
thelium starts to pull back from the cuticle, and in the isopod Idotea baltica, the sheath cells also
pull back. This means that the outer segments of the ORNs are exposed to the exuvial fluid and
are slowly digested, leaving the ORN insensitive for a period. The sheath cells pull back into an
invagination and start forming the cuticle of the new seta as described above. During this time,
the ORNs rebuild their outer segments, which is done by the time the old cuticle is shed and the
new seta is everted (Guse 1983). In other species such as the mysid Neomysis, not all the sheath
cells pull back; some stay in contact with the old cuticle and protect the outer segments during
most of the premolt stage. The way the tissue pulls back from the old cuticle to form the new one
is very standardized within species and is often used to stage the animal through the molt cycle
(e.g., Reaka 1975, Longmuir 1983, Hunter and Uglow 1998).
What is seen from the descriptions above is that both new setae and molting setae are formed
in an invagination, leaving them with an annulus. This means that the annulus is a by-product
of the complex setal development and that it could be another important character defining a
seta. Still, as stated earlier, it can be hard to detect in the intermolt, and there are indications that
some setae are not developed in this way (Fig. 6.4B; Watling 1989). This difference in ontogeny
suggests the question of whether these structures are true setae, which should be addressed
178 Functional Morphology and Diversity

Fig. 6.4.
Setal development. (A) Crustacean setae form in an invagination during the premolt stage. After the molt,
the seta unfolds, due to hydrostatic pressure. The point of folding (arrowheads) becomes the annulus in
the new seta (see Fig. 6.3). (B) In some cases the new setae form without an invagination, as shown here
from maxilla 1 of the amphipod Uschakoviella echinophora. Asterisks indicate new setae inside the exuvium
(Ex). (C) Cross section of a composite seta during its formation in the premolt stage. Both the proximal
part of the cuticle (PC) and the distal part (DC) are lined with sheath cells (ShC). Outer dendritic seg-
ments (ODS) can be seen in the distal part. The lumen created by the invagination is partly filled with
extra-cellular matrix (ECM) closely encircling the setules (S). (D) Close-up of the tip of a forming seta
showing a thin cuticle (Cu) surrounding sheath cells and outer dendritic segments (ODS) with a dendritic
sheath (DS) around them.

with transmission electron microscopy. Even though the number of sheath cells seems to vary
greatly among different setae, their presence and similar arrangement leave the possibility that
the development might have many identical elements across crustacean taxa and setal types.
That said, far too little data are available from crustaceans at present to verify this. In insects, the
development of their sensilla (believed to be homologous with crustacean setae) follows very
strict patterns when it comes to the cells involved (Keil 1997).

SETAL MORPHOLOGY AND SENSORY MODALIT Y

Most but not all examined setae are sensory organs of some kind, and they are often referred to as
sensilla. There is experimental evidence for setae being mechano- and chemosensitive but also
indications of thermo-, osmo-, and hygrosensitivity. In most cases they appear to be bimodal
mechano- and chemoreceptors (see also chapter 7). The number of sensory cells that innervate
Setae, Setules, and Other Ornamentation 179

crustacean setae can vary from one to several hundred. Despite this diversity, there are shared
features in the organization and ultrastructure of sensory setae. In this section, we provide an
overview of the sensory properties of setae and discuss whether they can be predicted by some
of the external or internal structures of setae.
One feature common to all sensory setae is having sensory cells with cell bodies arranged in
clusters outside of the seta they innervate and a single dendrite innervating the seta. The den-
drite is composed of two parts, an inner dendritic segment (IDS) and an outer dendritic segment
(ODS). The IDS normally terminates at the setal base, but in some cases it continues into the
proximal part of the setal shaft. The ODS is a sensory cilium, and normally only a single ODS
protrudes from each IDS, but in the case of aesthetascs, the sensory cells often have two ODSs
(Gr ü nert and Ache 1988). In most cases, the ODS extends a long way up the lumen of the setal
shaft, but for some mechanoreceptor cells, the ODS terminates at the base of the seta, where it
contacts the cuticular part of the socket. Another shared feature of sensory setae is the presence
of sheath cells encircling the sensory cells. Besides the described role during setal formation,
little is known about the function of sheath cells. They enclose the dendrites in a fluid-filled
cavity, and even though not much is known about this fluid and what the sheath cells add to its
content, the sheath is known to protect the ODS against stress factors from the external envi-
ronment, such as low salinity (Gleeson et al. 2000). The innermost sheath cell often has special
functions. In mecha noreceptors it contains a scolopale, which is probably a prerequisite for the
signal transduction (see below for more details), and in many but far from all setae, the distal
part of the innermost sheath cell comprises the dendritic sheath, an extracellular structure that
encircles the ODS as it proceeds up through the setal shaft.
Chemosensation has been associated with a broad spectrum of setal types through elec-
trophysiological and morphological studies. In fact, most of the examined setae appear to be
chemoreceptive. Thus, there seems to be little if any evidence for external setal structures being
associated with chemosensitivity. One character that has been debated is the presence or absence
of a terminal or subterminal pore. There is some evidence that, at least for bimodal mechano-
and chemosensitive setae, a pore is needed for the ODS to contact the substratum (Fig. 6.5C)
(Alexander 1977, Garm et al. 2003, Schmidt and Derby 2005). But in other studies the pore has
been shown to be blocked by electron-dense material and it has been suggested that the pore is
a holdover from molting, a so-called molting pore (Haug and Altner 1984a). Further, the olfac-
tory aesthetascs almost always lack a terminal pore, showing that it is certainly not a necessity
for chemosensation in setae. Instead, aesthetascs have a thin and porous cuticle, letting through
the small molecules they detect, such as amino acids, ammonia, and adenosine monophosphate
(Derby et al. 1997, Steullet and Derby 1997, Derby 2000, Steullet et al. 2000b).
Interestingly, there might be external structures that indicate the absence of chemosensation.
To our knowledge, setae with long setules along the entire shaft (plumose and pappose setae)
have never been shown, including from ultrastructure, to be chemosensitive. In some cases these
setae are not innervated (Fig. 6.5A), which suggests that they have no sensory function. This can
be hard to tell, however, since mechanosensitive cells can be situated at a distance from a seta
and still detect displacement of the seta (Bender et al. 1984). In the few cases where they have
been shown to be sensory organs, plumose and pappose setae are mechanoreceptors detecting
waterborne vibrations (Vedel and Clarac 1976, Wiese 1976, Vedel 1985). This makes good sense
since the long setules make the setae more sensitive to water movements, and crustacean mech-
anoreceptors have been shown to respond to even the smallest setal displacements (Wiese 1976,
Fields et al. 2002). Arranging the setules in two rows, as for plumose setae, adds directionality
to the sensitivity. Long setules are by no means necessary for mechanoreception, though, and
mechanosensation has been either proven or at least indicated from ultrastructure for all major
types of setae (Tautz et al. 1981, Derby 1989, Garm et al. 2004, Garm 2005). Regarding osmo-,
180 Functional Morphology and Diversity

Fig. 6.5.
Sensory properties of setae. (A) Cross section close to the base of a plumose seta showing several sheath
cells (ShC) but no sensory cells. S, setules. (B) Cross section of a serrate seta with both denticles (D) and
setules (S). The small lumen holds one outer dendritic segment (ODS). (C) Section through the terminal
pore of a serrate seta showing the outer dendritic segment protruding through the pore. (D) Most bimodal
chemo- and mechanosensory setae have their outer dendritic segments (ODS) enclosed by a dendritic
sheath (DS) until close to the tip. (E) The ciliary region of chemoreceptors (top) and mechanoreceptors
(bottom). In mechanoreceptors, the a-strands of the microtubules are electron dense (arrowhead) and
have dynein-like arms (arrow). (F) Almost all mechanosensory setae in crustaceans are of the scolopedial
type. The scolopale (Sc) is made of single strands of microtubules (arrowheads) connected by a dense
matrix of accessory proteins. (G) The inner dendritic segment (IDS) of mechanosensory cells displays a
large ciliary rootlet (CR), which has desmosomal connections (arrowheads) to the scolopale (Sc). (H) The
outer dendritic segments (ODS) of most mechanosensory cells have very densely packed single strands of
microtubules, which are thought to enhance their sensitivity to distortions.
Setae, Setules, and Other Ornamentation 181

thermo-, and hygroreceptors, far too little data are available to evaluate whether these modali-
ties are associated with any particular external structures (Tazaki 1975, Ache 1982, Ziegler and
Altner 1995, Garm et al. 2004). In conclusion, we find it highly questionable to address the sen-
sory functions of a seta from external characters alone even on the modality level.
From combined studies of sensory physiology and internal structure of setae, some structural
features, especially from the sensory cells, can be used to identify sensory modality (Altner et al.
1983, 1986, Schmidt and Gnatzy 1984, Hallberg et al. 1992, 1997, Derby et al. 2002). Interestingly,
information can be obtained at the histological level. The number of sensory cells can be deter-
mined in this way and in some cases can be used to identify chemosensitivity. From the existing
data, mechanosensation, and probably also osmo- and hygrosensation, is based on a maximum
of four sensory cells per seta (Mellon 1963, Schmidt and Gnatzy 1984, Ziegler and Altner 1995,
Garm et al. 2004). Chemosensation, on the other hand, is often based on many sensory cells,
with the aesthetascs of large decapods being the extreme, containing several hundred sensory
cells (Hallberg et al. 1992, Schmidt and Derby 2005, Hallberg and Skog 2011). The functional
explanation is probably that chemosensory setae need to detect a diversity of functionally rel-
evant chemical stimuli, which are often mixtures, and thus require a large number of receptor
molecules that are distributed nonhomogeneously across a population of sensory cells to iden-
tify the nature of those stimuli (Derby et al. 1994, 2001, Derby 2000). Since the functional unit
appears to be a single seta, each seta needs to contain this entire population of sensory cells
(Steullet et al. 2000b). It is important to keep in mind, though, that many sensory cells are not
a necessity for chemosensation in setae. Especially in small setae, there is strong evidence that
chemosensation can be based on very few cells (Altner et al. 1986, Elofsson and Hessler 1994,
Lagersson et al. 2003), but a small number of chemosensitive cells are also seen in some large
setae (Fig. 6.5B) (Altner et al. 1983). How the low number of sensory cells in these setae influ-
ences source identification is unknown.
On the ultrastructural level, there is better evidence for modality-specific structures. As
stated earlier, the ODS is a modified cilium and is the site of sensory transduction. For mech-
anoreceptors, it appears important that the ODS is anchored, possibly ensuring that the sensory
cells do not just follow the movements of the setae but are distorted when a seta moves. This
anchorage is located in the dendrite just proximal to the ODS, where the scolopale is found in
the innermost sheath cell (Fig. 6.5F,G). The scolopale is a very prominent structure made of
single strands of microtubules tied together by a large amount of associated protein (Fig. 6.5F)
(Schmidt and Gnatzy 1984). The ciliary rootlets of the mechanosensory cells contact the scol-
opale through desmosomes, and this connection provides the anchoring (Fig. 6.5G). Further
evidence for the role of the scolopale in mechanoreception comes from the fact that it is never
found in unimodal chemoreceptive setae. Still, there are some indications that putative osmore-
ceptors are also connected to the scolopale, but these cells were bimodal mechano- and osmore-
ceptors (Fig. 6.6) (Garm et al. 2004).
Within the sensory cells, the ciliary structures seem also to provide information on modality.
The ciliary rootlet of the mechanoreceptors is often very large, perhaps to enhance their stabil-
ity. The transition zone from IDS to ODS contains the ciliary region of the sensory cilium and
displays the classical 9 × 2 arrangement of the microtubules in a ring. In mechanosensory, but
not chemosensory, cells the a-strand is electron dense and carries arms resembling the motor
protein dynein (Fig. 6.5E). Further, the ciliary region of mechanoreceptors is several microm-
eters long, and it has been suggested that the stretch-sensitive ion-channels responsible for the
transduction from mechanical displacement to receptor potential are situated here (Crouau
1989). This is possibly true for some mechanoreceptive cells, but in other cases the transduction
occurs farther up the cilium (Calvet 2003, Praetorius et al. 2003, Garm et al. 2004, Garm and
Høeg 2006). In insects, most of the mechanoreceptors are not associated with a scolopale, but
182 Functional Morphology and Diversity

A Bend-sensitive cell

Stimulus
Bend
B Bend-sensitive cell

Stimulus Dis Dis Dis


0s 2s 4s 6s 8s

C Stimulation on-set

0 sec 2 sec 4 sec 6 sec

DM
D E DS

Se Cu
Cu

MC
S
CR
CR
SD WD

SCR

LCR
ShC
0.5 μm

Type 1 Type 2 Type 3

Fig. 6.6.
Bimodal osmo- and mechanosensitive cells. (A) Electrophysiological recordings from mechanosensi-
tive cells in simple setae of Panulirus argus responding only to bending. (B) The same cell as in A did not
respond to displacements in the socket. (C) The bend-sensitive cells also responded to changes in osmo-
larity. Arrowheads indicate bend-sensitive cell; arrows indicates additional smaller units in the recording.
(D) The seta containing the bimodal osmo- and mechanosensitive cell displays an isolated outer dendritic
segment in the distal part (arrowhead), which is suggested to be the bimodal cell. (E) Schematic drawing
of the three types of sensory cells from simple setae in P. argus, identified by their ultrastructure. Type 3
is suggested as the bimodal cell. Abbreviations: CR, ciliary region; Cu, cuticle; DM, dense microtubules;
DS, dendritic sheath; LCR, large ciliary rootlet; MC, membranous cuticle; S, scolopale; SCR, small ciliary
rootlet; SD, strong desmosomal connection; Se Cu, setal cuticle; ShC, sheath cell; WD, weak desmosomal
connection.

instead, the dendrite ends in a tubular body composed of densely packed microtubules where
transduction is believed to take place (Keil 1998). Structurally similar sensory cells have been
found in association with setae of the terrestrial isopod Titanethes alba (Crouau 1995).
There are also indications that the distal part of the sensory cilium of mechanoreceptors has
special features. Their microtubules are often densely packed in most of the modified cilium
above the ciliary region, where the microtubules diverge in single strands (Fig. 6.5H), perhaps
enhancing the sensitivity of the cell (French 1988). Interestingly, sensory cilia can be activated
by bending without disturbing the proximal part (Praetorius et al. 2003). This has also been
found in mouthpart setae of the spiny lobster Panulirus argus, and in accordance with this
Setae, Setules, and Other Ornamentation 183

different way of stimulation, it was found that the cells have a short ciliary region, a weak con-
nection to the scolopale, and a small ciliary rootlet (Fig. 6.6). The same cells also respond to
osmotic changes, and it is not known if any of the modifications of the structures associated
with mechanoreception are due to this bimodality.
Transduction in chemoreceptor cells is mediated through G-coupled receptors situ-
ated in the ODS (Couto et al. 2005). These cells seem to be in no need for anchorage, as they
are never connected to the scolopale, and their ciliary rootlet is small and often fragmented
(e.g., Snow 1973). Most likely, as a further consequence of their difference in transduction, their
ciliary region is short, often <1 μm, and the ODS has few loosely packed single strands of micro-
tubules. The ODS of chemosensory cells continues to the very tip of the seta they innervate
(Fig. 6.5B,C), but so far it has not been proven where the stimulation of the receptor molecules
takes place. The ultrastructural data indicate that there might be differences here. In bimodal
mechano- and chemosensory setae, the ODSs are wrapped in a dendritic sheath up close to the
tip and sometimes also by projections of the sheath cells (Fig. 6.5D) (Garm et al. 2003, Garm
and Høeg 2006). Even though nothing is known about the functionality of this arrangement,
it indicates that the ODSs are not in contact with the external environment except for the last
few micrometers. This is supported by the setal shaft having a thick and dense cuticle up to
the tip (Fig. 6.5B). The situation is different in aesthetascs, which have no dendritic sheath,
a thin cuticle, and ODSs branching shortly after the ciliary region, resulting in a larger mem-
brane surface area (Gr ü nert and Ache 1988). Chemical stimuli can probably contact the ODS
along its entire length in aesthetascs (Blaustein et al. 1993), and these structural differences
may partly account for the reported threshold differences between the unimodal aesthetascs
and the bimodal mechano- and chemosensitive setae (Thompson and Ache 1980, Voigt and
Atema 1992, Garm et al. 2005).
A highly interesting and intriguing finding in sensory setae is the presence of bimodal
sensory cells. They can either be mechano- and chemosensors (Hatt 1986) or mechano- and
osmosensors (Tazaki 1975, Garm et al. 2004). Nothing is known about the ultrastructural
organization of the former, but in the latter case there are indications that it is intermediate
between mechano- and chemosensors (Garm and Høeg 2006). Bimodal cells have rarely been
identified in crustaceans, but a typical experimental setup applies either mechanical or chemi-
cal stimuli, and therefore, bimodal cells could easily have been missed because of lack of the
proper stimulus. Bimodal cells are also rare outside crustaceans but have been found in some
molluscs (Audesirk and Audesirk 1980).
Hygrosensitivity has never been proven by physiological experiments in crustaceans but has
been suggested from ultrastructure (Haug and Altner 1984b, Ziegler and Altner 1995). Studies
on insect hygroreceptors resulted in five characters suggesting hygroreception: (1) the ODS
proceeds into the seta, (2) the setal shaft has no apparent pores, (3) the ODS borders the setal
wall distally in the seta, (4) the ODSs are organized symmetrically in the setal lumen, and (5)
the setal wall is layered. From our experience, at least the first, second, fifth, and to some extent
the third characters are found in most of the examined bimodal chemo- and mechanoreceptive
setae (e.g., Altner et al. 1983, Crouau 1994).
Thermoreceptors are common in insect sensilla (Steinbrecht 1998), and several crustaceans
have been shown to respond to thermal stimulations and orient to a temperature gradient
(Barber 1961). The receptors responsible for these behaviors have not yet been identified, how-
ever, and might be other than setae.
To conclude, ultrastructural data suggest that crustacean sensory neurons have modal-
ity-specific structures at least for mechanoreceptors and chemoreceptors. Still, the largely
unknown bimodal cells and the possible presence of osmo-, hygro-, and thermoreceptors add a
question mark to what can be deduced about function of setae from their internal morphology.
184 Functional Morphology and Diversity

Until these modalities have been examined in greater detail, no final conclusions can be drawn.
Furthermore, within each sensory modality, ultrastructure alone fails to reveal much about the
sensitivity of cells, such as what compounds and what concentrations stimulate a chemorecep-
tor cell.

SETAL MORPHOLOGY AND NONSENSORY FUNCTIONS

As shown in the preceding section, most setae are sensory organs, but setae also have other,
equally important functions. Especially important are mechanical functions during behav-
iors such as locomotion, digging, grooming, and feeding that mainly involve the setae on the
appendages. Combining macro-video recordings (Fig. 6.7) with scanning electron microscopy
has provided evidence that the mechanical functions are largely correlated with the size, shape,
and location of the setae and with the ultrastructure of the cuticle. In the following, we describe
some of the known mechanical functions of the setal types and how they relate to the setal mor-
phology (summarized in Table 6.2).
The cuticular outgrowths of the setae seem to be the character that offers most insights
about their mechanical function. Plumose setae with long fragile setules in two rows along their
entire length can be mechanoreceptors, but when situated on the pereopods or lateral mouth
appendages, they often serve as surface extensions used for water pumping or swimming (e.g.,
Kohlhage and Yager 1994). The two rows of long setules along the entire length of the setal shaft
operating at low Reynolds numbers ensure a large surface and thereby a large drag. They are also
flexible, which in the case of the water pump on the maxillipeds of decapods is enhanced by an
annulated setal shaft (Garm and Høeg 2000). Further, they are often arranged on the append-
ages such that they expand during the power stroke and collapse during the recovery stroke
(Burrows 2009). Pappose setae appear somewhat similar to plumose setae, but their setules are
randomly arranged along the shaft. Only in a few cases have their functions been studied, but
they have been shown to be involved in feeding in a number of decapods. In filter feeding deca-
pods such as galatheids and thalassinideans, they are found in high densities on the maxillipeds,
where they work as filters holding back small suspended particles (Nicol 1932, Nickell et al. 1998,
Stamhuis et al. 1998). In other crustaceans they are situated on the lateral mouthparts, perform
gentle prey handling, and in general create a setal barrier ensuring that small food particles do
not escape the mouth apparatus (Garm 2004a).
Many setae have small setules situated on the distal half, as in composite setae, and they
are also often involved in feeding (Schembri 1982, Lavalli and Factor 1995). They are typically
involved in gentle prey handling (Garm 2004a), and the small and often scalelike setules on the
distal part probably do not withstand rough handling but will provide a good grip on small par-
ticles. This, and their slender form, indicates that they do not apply much force to the prey items.
When not situated on the medial edge of the mouthparts, they are often involved in grooming
the neighboring mouthparts and head region, and their role during gill cleaning is well docu-
mented (Bauer 1989, 1998, 1999).
Serrate setae with the two rows of denticles on the distal part serve a variety of mechanical
functions. They are found on the mouthparts of almost all examined crustaceans, but again,
their function is best studied in decapods. Most are situated on the food-handling medial edges
and perform rough prey handling, such as holding and shredding (Hunt et al. 1992, Garm 2004a).
Another very common mechanical function of serrate setae is grooming of the head region, and
especially the antennae are groomed frequently. In decapods, specialized clusters on the pro-
podus and carpus of the endopod of maxilliped 3 are used in this behavior (Fig. 6.7E–H). Here
Setae, Setules, and Other Ornamentation 185

Fig. 6.7.
Mechanical functions of setae revealed by macro-video recordings. (A) Overview of the mouth appa-
ratus of Astacus astacus feeding on a mussel. (B) Close-up of Pagurus bernhardus eating a piece of fish.
The setae involved are found on the medial rim of maxilliped 1 and 2 (Mxp1–2), maxilla 2 (Mx2), and the
mandibular palp (MP). The incisor process of the mandibles (IP) and the labrum (La) are also shown.
(C) Endoscope recordings can reveal actions of otherwise hidden setae. Here large cuspidate setae from
the basis of maxilla 1 (Mx1) of Panulirus argus are seen between the incisor process (IP) and medial lobe
(ML). (D) A piece of mussel is held by setae on maxilliped 2 (Mxp2) of Penaeus monodon while setae
on the medial rim of maxilliped 1 (Mxp1) probe its surface. (E–H) Time series of serrate setae on the
endopod (endo) of maxilliped 3 (Mxp3) of P. bernhardus grooming the f lagellum of antenna 1 (Ant1).
Arrows indicate movements.
186 Functional Morphology and Diversity

Table 6.2. Summary of the mechanical functions of the seven types of seta.

Seta type Mechanical function


Pappose Setal barriers, current direction, filtering

Plumose Surface extension of water pumps, setal barriers

Composite Gentle prey handling (reorientation and relocation of small prey


items), gentle grooming
Serrate Rough prey handling (collecting and holding prey and shredding soft
prey), rough grooming, filtering
Papposerrate Setal barriers, gentle prey handling

Simple Rough and gentle prey handling

Cuspidate Very rough prey handling (holding, shredding, and tearing large prey
items), restrict mouthpart movement, filtering

the denticles seemingly make the setae very efficient in scraping of debris, and their size can be
correlated with the robustness of the structure they groom (Garm and Høeg 2001). Grooming
by serrate setae is well documented in the literature (e.g., Bauer 1989, Pohle 1989, Fleisher et al.
1992). This suggests that when the serrate setae are found laterally on the mouthparts, such as
the endopods of the maxillae, they may function in grooming the neighboring limbs.
Papposerrate setae combine the long setules on the proximal part with denticles on the
distal part. They have been reported from a number of species (often as plumodenticulate
setae), but their mechanical functions have been studied only in a few species. In the crayfish
Austropotamobius pallipes and Cherax quadricarinatus, they were observed to perform gentle prey
manipulations such as pushing pieces of prey into the mouth (Thomas 1970, Garm 2004a).
A large number of setae have no outgrowths and appear more or less smooth, and such setae
are found in all crustaceans. Most of them are slender, and their length can vary from a few
micrometers to several millimeters. No mechanical functions have been documented for the
smaller of these simple setae, and the evidence points to them being unimodal chemorecep-
tors (Hipeau-Jacquotte 1986, Elofsson and Hessler 1994). The mechanical functions of the larger
simple setae seem to depend on their location. The aesthetascs on antenna 1 belong to this group,
and they are again unimodal chemoreceptors with no apparent mechanical functions (Hallberg
et al. 1992). When situated on the mouthparts of decapods, simple setae have been found to
perform a broad range of mechanical function, which is species dependent (Garm 2004a). In
some species, they appear mainly gustatory with reduced mechanical functions, but in P. argus,
they are the dominant setal type performing rough prey handling. In the latter case, the lack of
outgrowths is probably to avoid damage during feeding.
Cuspidate setae are also often without outgrowths, and they seem to perform the most rough
mechanical functions. This is clearly seen on the mouth apparatus, where they are involved in
shredding and tearing of the prey. Still, in the hermit crab Pagurus rubricatus, cuspidate setae on
the medial rim of maxilla 1 can act as a filter (Schembri 1982). This is a surprising function for
cuspidate setae and stresses that one should be careful not to assume the functions of setae from
their morphology alone.
Only a few nonsensory functions besides the mechanical functions have been documented
for setae. In a recent study, it was shown from ultrastructural data that some simple setae from
Setae, Setules, and Other Ornamentation 187

the mouthparts of the remipede Speleonectes tanumekes are associated with glands and might
therefore be involved in secretion (van der Ham and Felgenhauer 2008). Setae are also bound
to have an impact on the hydrodynamics of crustaceans, which will be most important for small
swimming species (see chapter 11).

FOSSIL SETAE

The cuticle is advantageous when you want to study the evolutionary history of crustaceans
since it fossilizes easily. Unfortunately, when it comes to fine structural details such as the setae,
information is often lacking (e.g., Collins et al. 2003). A uniquely well-preserved arthropod
fauna, called the “Orsten” fauna (see chapter 2), has been found in which the miniature ani-
mals from the Cambrian still display beautiful setae (Fig. 6.8). Surprisingly, these setae display
a great diversity, and several of the seven types we have suggested can be recognized. On the
mouthparts of the fossil crustaceans Rehbachiella, Bredocaris, and members of Skaracarida, at
least cuspidate, composite, and serrate setae were present (Fig. 6.8A–C) (Mü ller and Walossek
1985, 1988, Walossek 1993). Plumose and pappose setae were probably also present, but even
though well preserved, most of the setules are broken, impeding certainty about their original

Fig. 6.8.
Fossil setae. (A–C) Orsten fossils of Rehbachiella from the Cambrian show very well-preserved setae. The
mouthparts display many composite and serrate setae. Arrows indicate setules; the arrowhead indicates
denticles. (D) Agnostus pisiformis, also from the Orsten fauna, displays advanced setal types such as plu-
mose setae shown here. Arrows indicate setules. Pictures courtesy of Prof. Dieter Walossek.
188 Functional Morphology and Diversity

length. One important aspect of these findings is that they strongly suggest that at least most of
the setal types we have put forward for crustaceans date back to their very early evolutionary
history. If the setal types represent homologies, this was to be expected since all major recent
subgroups of Crustacea seem to display all the seven types of setae, and setal differentiation
therefore predates their last common ancestor. There is even evidence in the fossil record that
some differentiation happened very early in arthropod history. Another member of the Orsten
fauna, Agnostus pisiformis, which is closely related to trilobites, also displays several of these
advanced setal types (Fig. 6.8D) (Mü ller and Walossek 1987).
Only a few authors have addressed the internal relationship and evolution of setae, but it
has been suggested that the most fundamental setal type is the simple seta and that the other
types have evolved from this type by adding more complex features to the setal shaft (Farmer
1974). If so, one would expect to find many simple setae in the early fossil record. Interestingly,
the animals from the Orsten fauna seem to have very few simple setae (Mü ller and Walossek
1985, 1987, 1988, Walossek 1993). The alternative hypothesis is that simple setae have arisen
from other setal types by reduction of the outgrowths. This is supported by setules and den-
ticles being general features of the cuticle and not structures that have evolved exclusively
on the setal shaft. Setules and denticles in the general cuticle appeared early in arthropod
history, which is also seen in the Orsten fauna, where these structures are numerous on the
appendages (Fig. 6.8C) (Walossek 1993). The general lack of simple setae on the mouthparts
of Orsten animals can perhaps be explained by their small size, since small present-day crus-
taceans in general have few simple setae on their mouth appendages (e.g., Høeg et al. 1994,
Olesen 2001). Further, simple setae have limited functionality, and more complex setae would
be needed to support the likely suspension feeding habit of these small animals. The little
change in the external morphology of setae during the last 550 million years also indicates
that the mechanical functions are preserved.

NONSETAL STRUCTURES OF THE CUTICLE

Compared to setae, much less is known about development and functions of other cuticular struc-
tures. Setules in the general cuticle seem to be found throughout Crustacea but almost exclusively
around the mouth apparatus. They are never innervated, and thus they have no sensory func-
tions. Their mechanical function has been studied only in a few decapods and only indirectly
(Garm 2004a). They are found on the paragnaths of some decapods, where they fill the space
between the paragnath and maxilla 1. In this way, they prevent small food particles from escaping
and groom the oral surface of maxilla 1. From their arrangement in other crustaceans, it is likely
that they serve similar functions (e.g., Alexander 1988, Olesen 2001), and it is noteworthy that this
coincides with the functions they serve when situated on setae. Whether setae have adopted set-
ules along with some of their functions from the general cuticle or whether cuticular setules repre-
sent heavily reduced setae are unknown. The fossil record does not shed any light on this question
since even Agnostus filiformis had both composite setae and setules on the cuticle (Mü ller and
Walossek 1987).
Denticles are found in the general cuticle of many crustaceans and can be found on most
body parts. Like setules, they are not sensory, but no other function has been associated with
them. They could play an important role in hydrodynamics, as has been shown for the dermal
teeth of sharks (Lang et al. 2008). Still, denticles are often found on the mouthparts and between
fields of setae (Fig. 6.1D), where they will have no contact with prey items or have effect on the
hydrodynamics, and their function here remains enigmatic. Spines, on the other hand, have
well-documented functions and serve mainly in rough prey handling and predator avoidance
Setae, Setules, and Other Ornamentation 189

(Briones-Fourzan et al. 2006). Spines are mostly found on the appendages, such as antenna 2 of
spiny lobsters, but can be found in many other places.
There are many other cuticular structures that are nonarticulated. Klepal and Kastner (1980)
arranged such features into the following categories: hump, protuberance, scale, tooth, fringe,
spine, and comb. Humps and protuberances are round or irregularly shaped areas, respectively,
that push outward from the surrounding cuticle. They have been documented from tanaids
and cumaceans. Scales are flat structures, usually curved, and as wide as or wider than long,
although these dimensions may vary (see Klepal and Kastner 1980, their figs. 6–13). Teeth are
solid, cone-shaped structures of varying length, and when together in a row, they form combs.
Fringes are similar to combs, but the cuticular structures are longer and may be flexible. The
one feature that Klepal and Kastner (1980) include in their list of nonsensory structures that
may need further investigation is what they call a “spine.” They define a spine as a large, strong,
hollow, cone-shaped structure, and their images clearly show an articulated base. It is likely that
these structures are, in fact, cuspidate setae.
When it comes to the development of the nonsetal structures, both when first appearing and
during a molt, little is known about the processes involved. To our knowledge, nobody has stud-
ied the development of setules and denticles in the cuticle. Transmission electron microscopic
data from both early and late intermolt indicates that they are not associated with sheath cells
(Fig. 6.1E), suggesting that their development differs from that of denticles and setules on setae
(see earlier section). This does not support the supposed homology between these structures,
but it is not at all impossible that sheath cells are involved during the formation of the new cuti-
cle and then later degenerate or transform.

COMPARISON WITH OTHER ARTHROPOD GROUPS

When observing the cuticle of other arthropods, it immediately becomes obvious that some of
the structures are homologous to what we have described from crustaceans. Again, the most
notable and best understood is setae, which are found in all examined arthropod subphyla. In
the following, we compare crustacean setae with those from insects and spiders, which, espe-
cially in the former, are well understood (for detailed reviews, see Steinbrecht 1997, Keil 1997,
1998).
Insects have setae, or sensilla, as they are most often called in this group, scattered on all body
parts, but just as for crustaceans, they are most commonly found on the appendages and in the
head region. When the detailed external morphology is examined, however, insect setae differ
from crustacean setae, probably at least partly due to their terrestrial habitat. First of all, as a
general trend, they have many fewer and smaller outgrowths; long setules would likely collapse
without the support of the water. Instead, insect mechanoreceptors are long and normally slen-
der setae (Keil 1998). They also have large leaf- or club-shaped mechanosensory setae special-
ized for gravity sensing. The ability to sense gravity directly in this way is again due to the lack
of surrounding water. The cuticle of insect chemosensory setae always seems to have detectable
pores. In the case of olfactory setae, two systems are present, double- and single-walled setae,
both with numerous pores in most of the setal shaft allowing the odors to contact the ODSs
(Steinbrecht 1997). The gustatory setae of insects are normally bimodal receptors as in crusta-
ceans, and they also display a prominent terminal pore. The olfactory setae have other speciali-
zations not seen in crustaceans because of the physical differences of their habitats. Since they
detect airborne odors, they need an interface to get them in contact with the ODS in the recep-
tor lymph. This is seen as a complex canal system in the cuticle, where the odors are collected
and transported by so-called odor-binding proteins (R ützler and Zweibel 2005).
190 Functional Morphology and Diversity

The ontogeny of insect setae has been studied in detail and is well understood (Keil 1997).
The pattern of sheath cells (enveloping cells) and their function are much stricter than in crus-
taceans. Normally, three (sometimes two or four) sheath cells are involved in the development
of the setae: the trichogen, tormogen, and thecogen cell. Their functions are similar to what is
known from crustaceans, and they form the cuticle of the seta, protect the sensory cells, and
produce the receptor lymph. They encircle the receptor cells, which are normally fewer than
in crustaceans, although insect olfactory setae can hold up to 50 receptor cells (Hallberg and
Hansson 1999).
All insect setae are considered to be sensory and have been found to be mechano-, chemo-,
hygro-, thermo-, and carbon dioxide receptors. As for crustaceans, the most common are
chemo- or mechanoreceptors, but there seems to be a more strict division, with bimodal chemo-
and mechanosensitive setae being less common. This is partly because insects have many more
types of olfactory setae, which all seem to be unimodal as in crustacean aesthetascs (Hallberg
and Hansson 1999). In insects, the transduction mechanisms are known for most of the modali-
ties, not least because of the powerful molecular tools available to study Drosophila and some
other species (e.g., Caldwell and Eberl 2002). Insects do have scolopedial mechanoreceptors
similar to those of crustaceans, but most insect mechanoreceptors differ in that they have no
scolopale but instead a tubular body distally in the ODS, where both stimulation and trans-
duction takes place (Keil 1998). Here the mechanosensitive ODS stops at the setal base, where
movements of the seta distort the tubular body. This is again likely a terrestrial adaptation,
since structurally similar mechanoreceptors are found in terrestrial isopods (Crouau 1994).
From work on insects, it is also known that thermoreceptors are modified mechanoreceptors
(Mü ller et al. 2008), which is likely also true for hygroreceptors and carbon dioxide receptors
(Stange and Stowe 1999).
Setae in spiders are at least as diverse as in crustaceans and insects, and many of them appear
so similar to crustacean setae that it is difficult to tell them apart (Figs. 6.3, 6.9). Still, as in insects,
most of them have few and short outgrowths (Talarico et al. 2006), probably because of their ter-
restrial habitat. Unfortunately, rather little experimental work has been performed on their func-
tional morphology. They probably have many mechanical functions, and they are all putative
receptors. One of the mechanical functions that has been studied is how setae aid when adher-
ing to surfaces (Niederegger and Stanislav 2006). As for crustaceans, there seems to be a good
correlation between the detailed appearance of the setule-like outgrowths and their mechanical
function. Very little is known about the chemosensors, but some spider mechanosensors, the
trichobothria, are understood in great detail (Barth and Höller 1999). In trichobothria setae
(Fig. 6.9B), which are always unimodal mechanoreceptors, there is a strict correlation between
the structure and arrangement of the setal shaft and the sensory properties.
On the ultrastructural level, the existing data from spider setae indicate that the sensory cells
are similar to both insect and crustacean sensory cells and arranged in similar ways. The sensory
cilia of the chemoreceptors proceed a long way up the setal shaft, and in putative gustatory setae
they are enclosed by a dendritic sheath (Talarico et al. 2006). In olfactory setae, the dendritic
sheath is missing, and the ODSs lay against the setal cuticle and are in contact with the odors
through a pore system as in insects. The ultrastructure of spider mechanoreceptors appears very
similar to insect mechanoreceptors of the tubular body type (Talarico et al. 2006).

WHAT’S NEXT: INTERTAXON AND INTERMETHODOLOGICAL RESEARCH

What we hope to have illustrated in this review is that a fair amount of knowledge is availa-
ble about surface structures of the crustacean cuticle. What is also evident is that this knowl-
edge stems almost entirely from malacostracans and especially from decapod setae. This is
Setae, Setules, and Other Ornamentation 191

Fig. 6.9.
Setae from spiders. (A) Setae associated with the claws of Nicodamus mainae. Arrows indicate setae very
similar to crustacean serrate setae (compare with Fig. 6.3D). (B) Trichobothria from the palp of Mimetus
hesperus. These mechanosensitive setae closely resemble simple setae of crustaceans and are the most
studied and best-understood spider setae. The open socket (arrows) makes them very flexible, and in most
case they respond to tiny distortions. (C) Setae from the scopula of Copa flavoplumosa with outgrowths
that resemble setules. One type is similar to crustacean plumose setae (arrows; compare with Fig. 6.3B).
(D) A common type of spider seta from the abdomen of Pikelinia tam. The function is unknown, and it has
no obvious counterpart in crustaceans. (E) Setae from the metatarsa of a leg of Pseudolampona emmett.
These setae are so similar to many crustacean composite setae that they can hardly be distinguished from
them (compare with Fig. 6.3C). Pictures courtesy of Dr Mart í n J. Ram í rez.

unfortunate when the goal is to have a broad understanding of these structures functionally,
ontogenetically, and evolutionarily.
One of the issues we find of great importance is better understanding of how the different
structures in the cuticle group together and how these groups differ from each other. This should
be studied in an evolutionary context by finding homologies to define the groups. The develop-
mental processes seem to be the most promising place to start. Only a few malacostracan species
have been examined, and a study comparing setal development from a broad range of crustacean
groups is therefore highly desirable. A study of the development and formation of the setules and
192 Functional Morphology and Diversity

denticles on the general cuticle is also wanted for comparison with the formation of the similar
structures on the setae. This is vital to test their possible homology.
Another of the very interesting aspects for crustacean setae is their functional morphology,
which for a large part can be separated into mechanical functions and sensory functions, as we
have described here. We have put forward a classification system based on the external struc-
tures and indicated that this system also follows their mechanical functions. Whether this is
a general rule within crustaceans is too early to decide, since almost all the data come from
large decapods. What is badly needed here are studies combining behavioral observations and
scanning electron microscopy from several distantly related taxa. It is necessary to make the
behavioral observations such that the actions of identifiable setae can be followed during feed-
ing, grooming, walking, and so forth. This is certainly a challenge, but with modern high-speed
and high-resolution video equipment, it is also possible.
The last point we want to make concerns the sensory functions of setae. The bulk of the
available evidence suggests that the most common type of sensory seta in crustaceans is a bimo-
dal chemo- and mechanoreceptor. This needs solid confirmation from more setal types through
electrophysiological experiments. By combining such studies with high-quality transmission
electron microscopy, the structural basis for these modalities can also be finally established. It is
also of the highest interest to get better confirmation on the possible presence of osmo-, hygro-,
and thermosensitive setae. Since there is evidence from insects that all of these modalities are
closely linked to mechanoreception, it is likely that they share many ultrastructural features.
This can turn out to cause serious problems to ultrastructural identification of sensory modal-
ity. The bimodal cells are somewhat of a mystery, and finding out how the combined trans-
duction works, along with how the nervous system separates the information, is of the greatest
interest, not only for crustacean sensory biology but also for neuroscience in general.

ACKNOWLEDGMENTS

We thank Associate Professor Nikolaj Scharff (Natural History Museum of Copenhagen) and
Mart í n J. Ram í rez and Mat ías A. Izquierdo (Museo Argentino de Ciencias Naturales) for pro-
viding pictures of arachnids (all parts of the Assembling the Tree of Life: Phylogeny of Spiders
project). Further, we are grateful for the pictures provided by Associate Professor Jens Høeg
(University of Copenhagen) and Professor Dieter Waloszek (University of Ulm) of Parapagurus
sulcata and fossil arthropods, respectively. We also acknowledge the financial support from the
Danish Science Council to A.G. (grant 272–07–0163).

REFERENCES

Ache, B.W. 1982. Chemoreception and thermoreception. Pages 369–398 in H.L. Atwood and D.C.
Sandeman, editors. The biology of Crustacea, Vol. 3. Academic Press, New York.
Alexander, C.G. 1977. Antennal sense organs in the isopod Ligia oceanica. Marine Behaviour and
Physiology 5:61–77.
Alexander, C.G. 1988. The paragnaths of some intertidal crustaceans. Journal of the Marine Biological
Association 68:581–590.
Alexander, C.G., J.P.R. Hindley, and S.G. Jones. 1980. Structure and function of the third maxillipeds of
the banana prawn, Penaeus merguiensis. Marine Biology 58:245–249.
Altner, I., H. Hatt, and H. Altner. 1983. Structural properties of bimodal chemo- and mechanosensi-
tive setae on the pereiopod chelae of the crayfish, Austropotamobius torrentium . Cell and Tissue
Research 228:357–374.
Setae, Setules, and Other Ornamentation 193

Altner, H., H. Hatt, and I. Altner. 1986. Structural and functional properties of the mechanoreceptors and
chemoreceptors in the anterior oesophageal sensilla of the crayfish, Astacus astacus. Cell and Tissue
Research 244:537–547.
Audesirk , G., and T. Audesirk. 1980. Complex mechanoreceptors in Tritonia diomedea 1. Responses to
mechanical and chemical stimuli. Journal of Comparative Physiology A 141:101–110.
Barber, S.B. 1961. Chemoreception and thermoreception. Pages 109–131 in T.H. Waterman, editor. The
physiology of Crustacea, Vol. 2. Academic Press, New York.
Barth, F.G., and A. Höller. 1999. Dynamics of arthropod filiform hairs. V. The response of spider tri-
chobothria to natural stimuli. Proceedings of the Royal Society of London Series B 354:183–192.
Bauer, R.T. 1989. Decapod crustacean grooming: Functional morphology, adaptive value, and phyloge-
netic significance. Pages 49–74 in B. Felgenhauer, L. Watling , and A.B. Thistle, editors. Functional
morphology of feeding and grooming in Crustacea. Balkema, Rotterdam.
Bauer, R.T. 1998. Gill cleaning mechanisms of the crayfish, Procambarus clarkii (Astacidea: Cambaridae):
Experimental testing of setobranch function. Invertebrate Biology 117:129 –143.
Bauer, R.T. 1999. Gill-cleaning mechanisms of a dendrobranchiate shrimp, Rimapenaeus similis (Decapoda,
Penaeidae): Description and experimental testing of function. Journal of Morphology 242:125–139.
Bender, M., W. Gnatzy, and J. Tautz. 1984 . The antennal feathered hairs in the crayfish: A non-innervated
stimulus transmitting system. Journal of Comparative Physiology A 154:45–47.
Blaustein, D.N., R.B. Simmons, M.F. Burgess, C.D. Derby, M. Nishikawa, and K.S. Olson. 1993.
Ultrastructural localization of 5’AMP odorant receptor sites on the dendrites of olfactory receptor
neurons of the spiny lobster. Journal of Neuroscience 13:2821–2828.
Brandt, A. 1988. Morphology and ultrastructure of the sensory spine, a presumed mechanoreceptor of
Sphaeroma hookeri (Crustacea, Isopoda), and remarks on similar spines in other peracarids. Journal
of Morphology 198:219 –229.
Briones-Fourzan, P., M. Perez-Ortiz , and E. Lozano-Alvarez . 2006. Defense mechanisms and antipreda-
tor behavior in two sympatric species of spiny lobsters, Panulirus argus and P. guttatus. Marine
Biology 149:227–239.
Burrows, M. 2009. A single muscle moves a crustacean limb joint rhythmically by acting against a spring
containing resilin. BMC Biology 27:1–17.
Caldwell, J.C., and D.F. Eberl. 2002. Towards a molecular understanding of Drosophila hearing. Journal
of Neurobiology 53:172–189.
Calvet, J.P. 2003. Ciliary signaling goes down the tubes. Nature Genetics 33:113–114.
Coelho, V.R., and S.A. Rodrigues. 2001. Setal diversity, trophic modes and functional morphology of
feeding appendages of two callianassid shrimps, Callichirus major and Sergio mirim (Decapoda:
Thalassinidea: Callianassidae). Journal of Natural History 35:1447–1483.
Collins, J.S.H., C. Lee, and J. Noad. 2003. Miocene and Pleistocene crabs (Crustacea, Decapoda) from
Sabah and Sarawak. Journal of Systematic Palaeontology 1:187–226.
Couto, A., M. Alenius, and J.B. Dickson. 2005 . Molecular, anatomical, and functional organization of the
Drosophila olfactory system. Current Biology 15:1537–1545.
Crouau, Y. 1989. Feeding mechanisms of the Mysidacea. Pages 153–172 in B.E. Felgenhauer , L. Watling ,
and A.B. Thistle, editors. Functional morphology of feeding and grooming in Crustacea. Balkema,
Rotterdam.
Crouau, Y. 1994 . Ultrastructure d’un organe sensoriel antennaire de l’isopode terrestre cavernicole,
Titanethes alba. Canadian Journal of Zoology 72:2199 –2204.
Crouau, Y. 1995 . Association in a crustacean sensory organ of two usually mutually exclusive mechano-
sensory cells. Biology of the Cell 85:191–195.
Crouau, Y. 1997. Comparison of crustacean and insect mechanoreceptive setae. International Journal of
Insect Morphology and Embryology 26:181–190.
Cuadras, J. 1982. Microtrichs of amphipod Crustacea. Morphology and distribution. Marine Behaviour
and Physiology 8:333–343.
Derby, C.D. 1989. Physiology of sensory neurons in morphologically identified cuticular sensilla of
crustaceans. Pages 27–48 in B.E. Felgenhauer, L. Watling , and A.B. Thistle, editors. Functional
morphology of feeding and grooming in Crustacea. Balkema, Rotterdam.
194 Functional Morphology and Diversity

Derby, C.D. 2000. Learning from spiny lobsters about chemosensory coding of mixtures. Physiology and
Behavior 69:203–209.
Derby, C.D., M.F. Burgess, K. Olson, T. Simon, and A. Livermore. 1994 . Mechanisms of detection and
discrimination of mixtures in the olfactory system of spiny lobster. Pages 775–777 in K. Kurihara ,
N. Suzuki, and H. Ogawa , editors. Olfaction and taste, Vol. 9. Springer, Tokyo.
Derby, C.D., H.S. Cate, and L.R. Gentilcore. 1997. Perireception in olfaction: Molecular mass sieving
by aesthetasc sensillar cuticle determines odorant access to receptor sites in the Caribbean spiny
lobster, Panulirus argus. Journal of Experimental Biology 200:2 073–2081.
Derby, C.D., P. Steullet, H.S. Cate, and P.J.H. Harrison. 2002. A compound nose: Functional organiza-
tion and development of aesthetasc sensilla. Pages 346–358 in K. Wiese, editor. The crustacean
nervous system. Springer, Berlin .
Derby, C.D., P. Steullet, A.J. Horner, and H.S. Cate. 2001. The sensory basis of feeding behaviour in the
Caribbean spiny lobster, Panulirus argus. Marine and Freshwater Research 52:1339 –1350.
Duncan, K.W. 1985 . Cuticular microstructures of terrestrial Amphipoda (Crustacea, Family Talitridae).
Zoologischer Anzeiger 215:140 –146.
Ekerholm, M., and E. Hallberg. 2002. Development and growth patterns of olfactory sensilla in malacos-
tracan crustaceans. Pages 376–385 in K. Wiese, editor. The crustacean nervous system. Springer,
Berlin .
Elofsson, R., and R.R. Hessler. 1994 . Sensory structures associated with the body cuticle of
Hutchinsoniella macracantha (Cephalocarida). Journal of Crustacean Biology 14:454–462.
Farmer, A.S. 1974 . The functional morphology of the mouthparts and pereiopods of Nephrops norvegicus
(L.) (Decapoda: Nephropidae). Journal of Natural History 8:121–142.
Fields , D.M., D.S. Shaeffer, and M.J. Weissburg. 2002 . Mechanical and neural response from the
mechanosensory hairs on the antennule of Gaussia princeps . Marine Ecology Progress Series
227:173–186.
Fish, S. 1972. The setae of Eurydice pulchra (Crustacea: Isopoda). Journal of Zoology 166:163–177.
Fleisher, J., M. Grell, J.T. Hoeg , and J. Olesen. 1992. Morphology of grooming limbs in species of
Petrolisthes and Pachycheles (Crustacea: Decapoda: Anomura: Porcellanidae) a scanning electron
microscopy study. Marine Biology 113:425–435.
French, A.S. 1988. Transduction mechanisms of mechanosensilla. Annual Review of Entomology 33:39 –58.
Garm, A. 2004a . Mechanical functions of setae from the mouth apparatus of seven species of decapod
crustaceans. Journal of Morphology 260:85–100.
Garm, A. 2004b. Revising the definition of the crustacean seta and setal classification systems based
on examinations of the mouthpart setae of seven species of decapods. Zoological Journal of the
Linnaean Society 142:233–252.
Garm, A. 2005 . Mechanosensory properties of the mouthpart setae of the European shore crab Carcinus
maenas. Marine Biology 147:1179 –1190.
Garm, A., C.D. Derby, and J.T. Høeg. 2004 . Mechanosensory neurons with bend- and osmo-sensitivity in
mouthpart setae from the spiny lobster Panulirus argus. Biological Bulletin 207:195–208.
Garm, A., C.D. Derby, and J.T. Høeg. 2005 . Chemosensory neurons in the mouthparts of the spiny
lobsters Panulirus argus and P. interruptus (Crustacea: Decapoda). Journal of Experimental Marine
Biology and Ecology 314:175–186.
Garm, A., E. Hallberg , and J.T. Høeg. 2003. Role of maxilla 2 and its setae during feeding in the shrimp
Palaemon adspersus (Crustacea: Decapoda). Biological Bulletin 204:126 –137.
Garm, A., and J.T. Høeg. 2000. Functional mouthpart morphology of the squat lobster Munida sarsi, with
comparison to other anomurans. Marine Biology 137:123–138.
Garm, A., and J.T. Høeg. 2001. Function and functional groupings of the complex mouth apparatus of the
squat lobsters Munida sarsi Huus and M. tenuimana G.O. Sars (Crustacea: Decapoda). Biological
Bulletin 200:281–297.
Garm, A., and J.T. Høeg. 2006. Ultrastructure and functional organization of mouthpart sensory setae of
the spiny lobster Panulirus argus. Journal of Morphology 267:464–476.
Gleeson, R.A., L.M. McDowell, H.C. Aldrich, K. Hammar, and P.J.S. Smith. 2000. Sustaining olfaction at
low salinities: Evidence for a paracellular route of ion movement from the hemolymph to the sensillar
lymph in the olfactory sensilla of the blue crab Callinectes sapidus. Cell and Tissue Research 301:423–431.
Setae, Setules, and Other Ornamentation 195

Gr ü nert, U., and B.W. Ache. 1988. Ultrastructure of the aesthetasc (olfactory) sensilla of the spiny lobster,
Panulirus argus. Cell and Tissue Research 251:95–103.
Guse, G.W. 1980. Development of antennal sensilla during moulting in Neomysis integer (Leach)
(Crustacea, Mysidacea). Protoplasma 105:53–67.
Guse, G.W. 1983. Ultrastructure, development, and moulting of the aesthetascs of Neomysis integer and
Idotea baltica (Crustacea, Malacostraca). Zoomorphology 103:121–133.
Halcrow, K., and E.L. Bousfield. 1987. Scanning electron microscopy of surface microstructures of some
gammaridean amphipod crustaceans. Journal of Crustacean Biology 7:274–287.
Hallberg , E., and B.S. Hansson. 1999. Arthropod sensilla: Morphology and phylogenetic considerations.
Microscopy Research and Technique 47:428 –439.
Hallberg , E., K.U.I. Johansson, and R. Elofsson. 1992. The aesthetasc concept: Structural variations of puta-
tive olfactory receptor cell complexes in Crustacea. Microscopy Research and Technique 22:325–335.
Hallberg , E., K.U.I. Johansson, and R. Wallén. 1997. Olfactory sensilla in crustaceans: Morphology,
sexual dimorphism, and distribution patterns. International Journal of Insect Morphology and
Embryology 26:173–180.
Hallberg , E., and M. Skog. 2011. Chemosensory sensilla in crustaceans. Pages 103–121 in T. Breithaupt and
M. Thiel, editors. Chemical communication in crustaceans. Springer, New York.
Harris, R.A. 1979. A glossary of surface sculpturing. Entomology Occasional Papers No. 28. California
Department of Food and Agriculture Laboratory Services, Sacramento.
Hatt, H. 1986. Responses of a bimodal neuron (chemo- and vibration-sensitive) on the walking legs of the
crayfish. Journal of Comparative Physiology A 159:611–617.
Haug , T., and H. Altner. 1984a. A cryofixation study of a subcuticular receptor organ in the antennular
tip of the terrestrial isopod, Porcellio scaber Latr. (Crustacea). Journal of Ultrastructure Research
87:62–74.
Haug , T., and H. Altner. 1984b. A cryofixation study of presumptive hygroreceptors on the antennule of a
terrestrial isopod. Tissue and Cell 16:377–391.
Hinton, H.E. 1970. Some little known surface structures. Pages 41–58 in A.C. Neville, editor. Insect
ultrastructure. Royal Entomological Society, London .
Hipeau-Jacquotte, R. 1986. A new cephalic type of presumed sense organ with naked dendritic ends in
the atypical male of the parasitic copepod Pachypygus gibber (Crustacea). Cell and Tissue Research
245:29 –35.
Høeg , J.T., E.S. Karnick , and A. Frølander. 1994 . Scanning electron microscopy of mouth appendages in
six species of barnacles (Crustacea: Cirripedia: Thoracica). Acta Zoologica 75:337–357.
Holdich, D.M. 1984 . The cuticular surface of woodlice: A search for receptors. Symposia of the
Zoological Society of London 53:9 –48.
Holmquist, J.G. 1989. Grooming structure and function in some terrestrial Crustacea. Pages 95–114 in B.
Felgenhauer , L. Watling , and A.B. Thistle, editors. Functional morphology of feeding and groom-
ing in Crustacea. Balkema, Rotterdam.
Hunt, M.J., H. Winsor, and C.G. Alexander. 1992. Feeding by the penaeid prawns: The role of the anterior
mouthparts. Journal of Experimental Marine Biology and Ecology 160:33–46.
Hunter, D.A., and R.F. Uglow. 1998. Setal development at moult staging in the shrimp Crangon crangon
(L.) (Crustacea: Decapoda: Crangonidae). Ophelia 49:195–209.
Jacques, F. 1989. The setal system of crustaceans: Types of setae, groupings, and morphology. Pages 1–13
in B. Felgenhauer, L. Watling , and A.B. Thistle, editors. Functional morphology of feeding and
grooming in the Crustacea. Crustacean Issues 6. A.A. Balkema, Rotterdam.
Johnston, D.J. 1999. Functional morphology of the mouthparts and alimentary tract of the slipper lobster
Thenus orientalis (Decapoda: Scyllaridae). Marine and Freshwater Research 50:213–223.
Keil, T.A. 1997. Comparative morphogenesis of sensilla: A review. International Journal of Insect
Morphology and Embryology 26:151–160.
Keil, T.A. 1998. The structure of integumental mechanoreceptors. Pages 385–404 in F.W. Harrison and
M. Locket, editors. Insecta, Vol. 2B. Wiley-Liss, New York.
Klepal, W., and R.T. Kastner. 1980. Morphology and differentiation of non-sensory cuticular structures
in Mysidacea, Cumacea, and Tanaidacea (Crustacea, Peracarida). Zoologica Scripta 9:271–281.
196 Functional Morphology and Diversity

Kohlhage, K., and J. Yager. 1994 . An analysis of swimming in remipede crustaceans. Philosophical
Transactions of the Royal Society of London Series B 346:213–221.
Lagersson, N.C., A. Garm, and J.T. Høeg. 2003. Notes on the ultrastructure of the setae on the fourth
antennulary segment of the Balanus amphitrite cyprid (Crustacea: Cirripedia: Thoracica). Journal
of the Marine Biological Association 83:361–365.
Lang , A.W., P. Motta, P. Hidalgo, and M. Westcott. 2008. Bristled shark skin: A microgeometry for
boundary layer control. Bioinspiration and Biomimetics 3:1–9.
Lavalli, K., and J. Factor. 1995 . The feeding appendages. Pages 349–393 in J. Factor, editor. Biology of the
lobster. Academic Press, New York.
Longmuir, E. 1983. Setal development, moult-staging and ecdysis in the banana prawn Penaeus merguien-
sis. Marine Biology 77:183–190.
Martin, J.W. 1989. Morphology of the feeding structures in the Conchostraca with special reference to
Lynceus . Pages 123–136 in B. Felgenhauer, L. Watling , and A.B. Thistle, editors. Functional morphol-
ogy of feeding and grooming in Crustacea. Balkema, Rotterdam.
Martin, J.W., and C.E. Cash-Clark 1995 . The external morphology of the onychopod “cladoceran” genus
Bythotrephes (Crustacea, Branchiopoda, Onychopoda, Cercopagidae) with notes on the morphol-
ogy and phylogeny of the order Onychopoda. Zoologica Scripta 24:61–90
Mellon, D., Jr. 1963. Electrical responses from dually innervated tactile receptors on the thorax of the
crayfish. Journal of Experimental Biology 40:137–148.
Meyer-Rochow, V.B. 1980. Cuticular surface structures in Glyptonotus antarcticus—a marine isopod from
the Ross Sea (Antarctica). Zoomorphologie 94:209 –216.
Mü ller, K.J., and D. Walossek. 1985 . Skaracarida, a new order of Crustacea from the Upper Cambrian of
V ä stergötland, Sweden. Fossils and Strata 17:1–65.
Mü ller, K.J., and D. Walossek. 1987. Morphology, ontogeny, and life habit of Agnostus pisiformis from the
upper Cambrian of Sweden. Fossils and Strata 19:1–124.
Mü ller, K.J., and D. Walossek. 1988. External morphology and larval development of the Upper Cambrian
maxillopod Bredocaris admirabilis. Fossils and Strata 23:1–70.
Mü ller, M., M. Olek , M. Giersig , and H. Schmitz. 2008. Micromechanical properties of consecutive lay-
ers in specialized insect cuticle: The gula of Pachnoda marginata (Coleoptera, Scarabaeidae) and
the infrared sensilla of Melanophila acuminata (Coleoptera, Buprestidae). Journal of Experimental
Biology 211:2576 –2583.
Nickell, L.A., J.A. Atkinson, and E.H. Pinn. 1998. Morphology of thalassinidean (Crustacea: Decapoda)
mouthparts and pereiopods in relation to feeding, ecology and grooming. Journal of Natural
History 32:733–761.
Nicol, E.A.T. 1932. The feeding habits of the Galatheidea. Journal of the Marine Biological Association
18:87–106.
Niederegger, S., and N.G. Stanislav. 2006. Friction and adhesion in the tarsal and metatarsal scopulae of
spiders. Journal of Comparative Physiology A 192:1223–1232.
Olesen, J. 2001. External morphology and larval development of Derocheilocaris remanei
Delamare-Deboutteville and Chappuis 1951 (Crustacea, Mystacocarida), with a comparison of
crustacean segmentation and tagmosis patterns. Biologiske Skrifter 53:1–59.
Paffenhöfer, G.A., and P.A. Loyd. 2000. Ultrastructure of cephalic appendage setae of marine planktonic
copepods. Marine Ecology Progress Series 203:171–180.
Pohle, G. 1989. Gill and embryo grooming in lithodid crabs: Comparative functional morphology based
on Lithodes maja . Pages 75–94 in B. Felgenhauer, L. Watling , and A.B. Thistle, editors. Functional
morphology of feeding and grooming in Crustacea. Balkema, Rotterdam.
Praetorius, H., J. Frokiaer, S. Nielsen, and K. Spring. 2003. Bending of the primary cilium opens
Ca2+-sensitive intermediate-conductance K+ channels in MDCK cells. Journal of Membrane
Biology 191:193–200.
Reaka, M.L. 1975 . Molting in stomatopod crustaceans. 1. Stages of the molt cycle, setagenesis, and mor-
phology. Journal of Morphology 146:55–80.
R ützler, M., and L.J. Zweibel. 2005 . Molecular biology of insect olfaction recent progress and conceptual
models. Journal of Comparative Physiology A 191:777–790.
Setae, Setules, and Other Ornamentation 197

Schembri, P.J. 1982 . Functional morphology of the mouthparts and associated structures of Pagurus
rubricatus (Crustacea: Decapoda: Anomura) with special reference to feeding and grooming.
Zoomorphology 101:17–38.
Schmidt, M., and C.D. Derby. 2005 . Non-olfactory chemoreceptors in asymmetric setae activate anten-
nular grooming behavior in the Caribbean spiny lobster Panulirus argus. Journal of Experimental
Biology 208:233–248.
Schmidt, M., and W. Gnatzy. 1984 . Are the funnel-canal organs the “campaniform sensilla” of the
shore crab Carcinus maenas (Decapoda: Crustacea)? II. Ultrastructure. Cell and Tissue Research
237:81–97.
Snow, P.J. 1973. Ultrastructure of the aesthetasc hairs of the littoral decapod, Paragrapsus gaimardii .
Zeitschrift fuer Zellforschung und Mikroskopische Anatomie 138:489 –502.
Stamhuis, E.J., B. Dauwe, and J.J. Videler. 1998. How to bite the dust: Morphology, motion pattern and
function of the feeding appendages of the deposit-feeding thalassinid shrimp Callianassa subterra-
nea. Marine Biology 132:43–58.
Stange, G., and S. Stowe. 1999. Carbon-dioxide sensing structures in terrestrial arthropods. Microscopy
Research and Technique 47:416 –427.
Steele, V.J., and D.H. Steele. 1999. Cellular organization and fine structure of type II microtrich sensilla
in gammaridean amphipods (Crustacea). Canadian Journal of Zoology 77:88 –107.
Steele, V.J., and D.H. Steele. 1997. Type II microtrich sensilla of amphipods: Variations in external
morphology and distribution patterns. Canadian Journal of Zoology 75:1155–1165.
Steinbrecht, R.A. 1997. Pore structure in insect olfactory sensilla: A review of data and concepts.
International Journal of Insect Morphology and Embryology 26:229 –245.
Steinbrecht, R.A. 1998. Bimodal thermo- and hygrosensitive sensilla. Pages 405–422 in F.W. Harrison and
M. Locket, editors, Insecta, Vol. 2B. Wiley-Liss, New York.
Steullet, P., H.S. Cate, and C.D. Derby. 2000a . A spatiotemporal wave of turnover and functional matura-
tion of olfactory receptor neurons in the spiny lobster Panulirus argus. Journal of Neuroscience
20:3282–3294.
Steullet, P., H.S. Cate, W.C. Michel, and C.D. Derby. 2000b. Functional units of a compound nose:
Aesthetasc sensilla house similar populations of olfactory receptor neurons on the crustacean
antennule. Journal of Comparative Neurology 418:270–280.
Steullet, P., and C.D. Derby. 1997. Coding of blend ratios of binary mixtures by olfactory neurons in the
Florida spiny lobster, Panulirus argus. Journal of Comparative Physiology A 180:123–135.
Talarico, G., J.G. Palacios-Vargas, M.F. Silva, and G. Alberti. 2006. Ultrastructure of tarsal sensilla
and other integumental structures of two Pseudocellus species (Ricinulei, Arachnida). Journal of
Morphology 267:441–463.
Tautz , J., M. Masters, B. Aicher, and H. Mark. 1981. A new type of water vibration receptor on the crayfish
antenna. Journal of Comparative Physiology A 144:533–541.
Tazaki, K. 1975 . Sensory units respond to osmotic stimuli in the antennae of the spiny lobster Panulirus
japonicus. Comparative Biochemistry and Physiology A 51:647–653.
Thomas, W.J. 1970. The setae of Austropotamobius pallipes (Crustacea: Astacidae). Journal of Zoology
160:91–142.
Thompson, H., and B.W. Ache. 1980. Threshold determination for olfactory receptors of the spiny lob-
ster. Marine Behaviour and Physiology 7:249 –260.
van der Ham, J.L., and B.E. Felgenhauer. 2008. Ultrastructure and functional morphology of glandular
setae and distal claws of cephalic appendages of Speleonectes tanumekes (Crustacea: Remipedia).
Arthropod Structure and Development 37:235–247.
Vedel, J.P. 1985 . Cuticular mechanoreception in the antennal flagellum of the rock lobster Panulirus
vulgaris. Comparative Biochemistry and Physiology A 80:151–158.
Vedel, J.P., and F. Clarac. 1976. Hydrodynamic sensitivity by cuticular organs in the rock lobster Panulirus
vulgaris. Morphological and physiological aspects. Marine Behaviour and Physiology 3:235–251.
Voigt, R., and J. Atema. 1992. Tuning of chemoreceptor cells of the second antenna of American lobster
(Homarus americanus) with a comparison of four of its other chemoreceptor organs. Journal of
Comparative Physiology A 171:673–683.
198 Functional Morphology and Diversity

Walossek , D. 1993. The upper Cambrian Rehbachiella and the phylogeny of Branchiopoda and Crustacea.
Fossils Strata 32:1–202.
Watling , L. 1989. A classification system for crustacean setae based on the homology concept. Pages 15–26
in B. Felgenhauer, L. Watling , and A.B. Thistle, editors. Functional morphology of feeding and
grooming. Balkema, Rotterdam.
Wiese, K. 1976. Mechanoreceptors for near-field water displacement in crayfish. Journal of
Neurophysiology 39:816 –833.
Ziegler, A., and H. Altner. 1995 . Are the numerous sensilla of terrestrial isopods hygroreceptors?
Ultrastructure of the dorsal tricorn sensilla of Porcellio scaber. Cell and Tissue Research 282:135–145.

The author has requested enhancement of the downloaded file. All in-text references underlined in blue are linked to publications
7
ANTENNULES AND ANTENNAE IN THE CRUSTACEA

Geoff Boxshall and Damià Jaume

Abstract
The remarkable diversity in form and function of the antennules and antennae of crustaceans
is reviewed. The basic form of the crustacean antennule is a single-axis segmented limb, but in
malacostracans it is typically biflagellate, and in remipedes it comprises a dorsal segmented axis
and ventral flagellum. The form of the antenna is also compared across the major crustacean
taxa. It is plesiomorphically biramous, with a multisegmented exopod and up to four-segmented
endopod; the segmentation of the protopodal part is variable.
The antennules are primarily sensory and form a major sensory interface with the envi-
ronment. The distribution patterns of chemosensors, mechanosensors, and bimodal ele-
ments along the antennules are compared across major taxa. The structure of selected
sensors is brief ly described, and their roles in feeding, mating, and predator detection
behavior are discussed. Crustaceans also use their antennules for such purposes as holding
food, grasping a mate, and attaching to a host or other substrate. Morphological adaptations
exhibited by these multifunctional limbs are described. Antennae are used for a wide range
of tasks. They often perform a sensory role but are used in swimming, attachment, feeding,
mating, brooding, and respiration. The manner in which this multifunctionality is ref lected
in morphology is outlined.
The balance among the different functional roles performed by antennules and anten-
nae changes during development. In crustaceans with planktotrophic nauplii, for example,
the antennal coxa bears one or two feeding endites that are subsequently lost during devel-
opment as the mandibular gnathobase takes over their function. The ontogenetic develop-
ment of both limbs is brief ly reviewed. The enhancement of the chemosensory array on the
antennules of males as they attain sexual maturity is found to be common to many different
crustacean groups.

199
Functional Morphology and Diversity. Les Watling and Martin Thiel.
© Les Watling and Martin Thiel 2013. Published 2013 by Oxford University Press.
200 Functional Morphology and Diversity

ANTENNULES AND ANTENNAE IN THE CRUSTACEA

The spectacular adaptive radiation of crustaceans is reflected in the diversity of morphology,


behavior, and mode of life they exhibit. The traditional definition of a crustacean usually refers
to the possession of a mandible in combination with two pairs of antennae. This serves to dis-
tinguish them from chelicerates and fossil taxa such as trilobites, megacheirans, and marrel-
lomorphs, which lack mandibles, and from hexapods and myriapods, which have mandibles
but only a single pair of antennae. While in many different crustaceans the antennae provide
the primary sensory interface with the environment, they can also be involved in locomotion,
attachment, feeding, and mating behavior. The morphology of the antennae differs from group
to group, reflecting their different functional roles, and it can vary during development, as the
balance of functional roles changes from larva to adult. The patterns of antennal morphology
and their relationships are explored here.

ANTENNULES

The anteriormost paired limb on the head of arthropods goes by a variety of names: in crus-
taceans either antennule or first antenna is used, according to group or personal preference,
whereas in the other mandibulatan groups (hexapods and myriapods) and in the trilobites it
is most commonly referred to as the antenna. In modern chelicerates, the anteriormost limbs
on the prosoma are the paired chelicerae, which, on the basis of Hox gene expression patterns
(Telford and Thomas 1998, Damen et al. 1998), can be interpreted as positional homologues of
the antennules in these other arthropods (Akam 2000), but they are not considered here. In
some Paleozoic fossil arthropods of uncertain affinity, such as Leanchoilia superlata, the first
limb is simply termed the “great appendage” (Bruton and Whittington 1983). We use antennule
here for the anteriormost limb pair arising from the deuterocerebral somite of the head of crus-
taceans and their mandibulatan relatives (see Scholtz and Edgecombe 2005).

Structure

The typical arthropodan antennule comprises a single proximodistal axis composed of few to
multiple subdivisions, which have been referred to variously as antennulomeres, podomeres,
f lagellomeres, segments, articles, annuli, or annulations. The terminology for the subdivi-
sions is dependent upon their anatomy. An antennule is segmented if it consists of a linear
series of true segments, as defined by the presence of intrinsic muscles that insert or attach
at each intersegmental articulation. Each articulation is typically provided with a hoop of
arthrodial membrane that allows slight telescoping of the proximal rim of the more distal seg-
ment within the distal rim of the more proximal segment. Segmented antennules, therefore,
have segmentally arranged intrinsic musculature, typically along the entire length of the limb
(see Boxshall 2004).
The other type of single-axis antennule is f lagellate. At least two different types of f lagel-
late antennules are found within recent arthropods: those with terminal f lagella formed by
annulation of the distal segment, as in some collembolans, ectognathan insects, and mala-
costracan crustaceans, and those with nonterminal or intercalary annulations, such as scu-
tigeromorph centipedes and some crustacean larvae (see Boxshall 2004). Terminal f lagella
are by far the most common. Flagella are also composed of few to multiple external subdivi-
sions of the axis, but they lack intrinsic musculature. These subdivisions are referred to here
as annuli .
Antennules and Antennae in the Crustacea 201

Interpreting the nature of the antennules of fossil arthropods can be problematic since infor-
mation on intrinsic musculature is usually lacking. Without evidence derived from knowledge
of the internal anatomy, the subdivisions of the limb cannot be identified as either segments
or annuli, and descriptive terms such as antennulomeres should be used. Whether segmented
or flagellate, antennules are essentially modular in construction, and this modularity confers
important functional attributes, permitting, for example, the enhancement of a sensory array by
the addition of extra modules or by the specialization of individual modules.

Single-Axis Antennules

Single-axis antennules are found in the great majority of fossil arthropod taxa from the early
Paleozoic, as exemplified by the Cambrian arthropods of the Burgess and Maotianshan Shale
faunas. Single-axis, segmented antennules are found in the Myriapoda (with the exception of
the Pauropoda), in some basal Hexapoda (Diplura and Collembola only) (Imms 1939), and in
the Crustacea, with the exception of the Remipedia and the Malacostraca only (see Boxshall
et al. 2010).
In the Crustacea, the maximum number of segments expressed varies with taxon. In recent
Branchiopoda, the antennules are either one- or two-segmented, although the spinicauda-
tan Caenestheriella australis has a distal flagellar section comprising a linear series of lobes
that resemble segments (Boxshall 2004). Fossil branchiopods have more antennulary segments:
the Devonian lipostracan Lepidocaris has short, three-segmented antennules (Scourfield 1926,
1940), but in the Kazacharthra (fossil relatives of the Notostraca) up to 15 antennulomeres have
been reported in Almatium gusevi (McKenzie et al. 1991). In Cambrian branchiopods such as
Rehbachiella and Bredocaris, the antennules comprise a proximal zone of incomplete annuli and
a distal zone of cylindrical segments (Walossek 1993). The presence of proximal annuli is pos-
sibly indicative of intercalary annulation (sensu Boxshall 2004), but the absence of data on mus-
culature makes it impossible to interpret these fossils unequivocally.
In the Cephalocarida the single-axis antennule comprises only six segments in
Hutchinsoniella macracantha (Hessler 1964), and no segments are added during development
(Sanders 1963). In the Mystacocarida the maximum number of segments is eight (Hessler and
Sanders 1966), and this is constant throughout larval development in Derocheilocaris rema-
nei (Olesen 2001). In any extant copepod species, a maximum of 27 antennulary segments is
found, although Huys and Boxshall (1991) demonstrated that a total of 28 different segments
are expressed within the Copepoda as a whole. These multisegmented antennules are not flag-
ellate—all segments are primitively provided with intrinsic musculature (Fig. 7.1). Reductions
in antennulary segmentation are common across the Copepoda and are primarily a result of
failure of expression of intersegmental articulations during development rather than secondary
fusion of segments that were expressed earlier in ontogeny (Boxshall and Huys 1998).
The antennule of Myodocopa is a single axis with up to eight segments, as found in extant
forms and in fossil taxa from the Silurian (Siveter et al. 2003, 2007). Podocopan ostracods have
single-axis antennules, and a maximum of eight segments is expressed, although most recent
taxa have fewer (Maddox 2000). Developmental data indicate that reductions in segmentation
in adults result primarily from failures of expression of intersegmental articulations rather than
secondary fusions (Smith and Tsukagoshi 2005, Smith and Kamiya 2008). It has been suggested
that a reduced exopod, represented by a setose marginal expansion, is present in some podo-
copans (e.g., Kesling 1951, Rossetti and Martens 1996, Karanovic 2005), but a recent review of
developmental and other studies found no evidence to support this interpretation (Boxshall
et al. 2010).
202 Functional Morphology and Diversity

Fig. 7.1.
Proximal part of the antennule of a female Euaugaptilus (Copepoda: Calanoida) showing intrinsic muscles
making intermediate attachments on the proximal rim of each segment. Modified from Boxshall (1985),
with permission from the Royal Society of London.

In the bivalved Cambrian phosphatocopines, the antennules are small and comprise a single
axis with a maximum of 13 irregular subdivisions (in Waldoria), referred to by Maas et al. (2003)
as annuli. Setae are present on the two or three distal annuli only, and the setation pattern does
not change during development.
Antennules of Thecostraca consist of a maximum of six expressed segments, as found in some
Ascothoracida. In adults the prehensile antennule terminates in a clawlike subchela mechanism
that secures attachment to the host. Developmental studies show that the ascothoracidan claw
originates on the fourth segment of the six-segmented naupliar antennule, with the distal two
segments becoming incorporated into the compound apical segment of the adult (Grygier 1984).
A total of eight different segments are therefore expressed during development. In the Cirripedia,
including the parasitic Rhizocephala, the antennules (Fig. 7.2) are short and four-segmented, but
the apical segment is offset at an angle to the main axis that terminates in the attachment disc
carried on the third segment. This disc is responsible for securing the larva to the substrate dur-
ing settlement. A standardized terminology for the segments and their setae was proposed by
Bielecki et al. (2009).
The antennule of the branchiuran fish lice comprises a two-segmented basal part armed with
curved, hooklike outgrowths and a two-segmented distal part, bearing setae (Rushton-Mellor
and Boxshall 1994). The Tantulocarida lack antennules in the tantulus larva stage (indeed, this
larva lacks all cephalic limbs), and in adult males the antennules are represented only by paired
clusters of aesthetascs arising anteriorly on the ventral surface of the cephalothorax (Boxshall
and Lincoln 1987). In the adult sexual female, a pair of unsegmented antennules is present (Huys
et al. 1993).

Double-Axis Antennules

Antennules with two proximodistal axes are found in adult malacostracans and remipedes.
The malacostracan antennule typically comprises a robust basal part, the peduncle, bearing two
multiannulate flagella. The antennulary peduncle in malacostracans is not homologous with
the protopod of the postantennulary biramous limb—there is no evidence that the crustacean
antennule is derived from a fundamentally biramous structure (Boxshall 2004). The peduncle
Antennules and Antennae in the Crustacea 203

Fig. 7.2.
Scanning electron micrograph of distal segments (s2, s4) of antennule of cypris larva of Megabalanus rosea
(Thecostraca: Cirripedia) showing offset apical segment (s4) with complex setal armature, and attach-
ment disc (ad) on third segment. Photo courtesy of J. Høeg.

is the proximal segmented zone that contains intrinsic musculature and carries the flagella. It
can be two-, three-, or four-segmented and houses the flagellar flexor and extensor muscles that
insert inside the proximal rim of the basal annulus of each flagellum (Fig. 7.3). All malacostracan
antennules thus far investigated lack intrinsic musculature in the annulated flagellar parts. In
the terrestrial anomuran Birgus latro, the peduncle segments are long and slender, and the distal
flagella are relatively short.
This biflagellate antennule is an autapomorphy of the Malacostraca (Richter and Scholtz
2001, Jenner et al. 2009) and is present in the earliest unequivocal fossil malacostracan,
Cinerocaris magnifica, an archaeostracan phyllocarid from the Silurian (Briggs et al. 2003). The
two flagella are sometimes referred to as inner and outer rami (Richter and Scholtz 2001) or,
specifically, as exopod and endopod (e.g., Bruce 1986 for isopods, Clark et al. 1998 for decapods,
and Perrier et al. 2006 for palaeocarid syncarids), but this terminology is inappropriate (see
review in Boxshall et al. 2010). The flagella are also referred to as outer and inner (Jaume and
Bréhier 2005), as medial and lateral (e.g., Goldman and Patek 2002, Reidenbach et al. 2008),
or as primary and accessory (Jaume et al. 2006). The terms primary and accessory flagella are
used here.
The biflagellate antennule comprises a laterally located primary flagellum, the annuli of
which each typically carry one or two rows of aesthetascs, and an accessory flagellum. The
more medially positioned accessory flagellum can be as long as or longer than the primary
flagellum, but it never carries aesthetascs and typically appears later in development than the
204 Functional Morphology and Diversity

pd1

pd2

pd3

pf

af

Fig. 7.3.
Antennule of the male of Montucaris (Peracarida: Bochusacea) showing intrinsic musculature within
three-segmented peduncle (pd1–pd3) and inserting on the proximal rim of an accessory flagellum (af). No
musculature is observed going to the base of the primary flagellum (pf) in the male. Modified from Jaume
et al. (2006), with permission from Blackwell Publishing.

primary flagellum. Given the presence of aesthetascs only on the primary flagellum, we infer
that it is homologous with the dorsal axis in remipedes and with the sole axis in other nonmala-
costracans. Most malacostracans, including the anaspidacean and palaeocaridacean Syncarida,
Euphausiacea, Amphionidacea, Decapoda, and members of the peracaridan groups Mysidacea,
Lophogastrida, Cumacea, and Spelaeogriphacea, primitively have two multiannulate flagella.
Both flagella can be reduced; in the Mictacea and Bochusacea, for example, both consist of just
a few annuli. The accessory flagellum is reduced and often absent in isopods, amphipods, and
tanaidaceans. In oniscoidean isopods, perhaps as an adaptation to a terrestrial mode of life, the
entire antennule can be reduced to one to three subdivisions, with subsequent loss of clear sepa-
ration between the peduncle and flagellum (Vandel 1960). In the cirolanid isopod Bathynomus
giganteus, the “scale” carried on the third antennulary segment has been interpreted as repre-
senting the reduced accessory flagellum, as indicated by the retained cluster of setal elements at
its tip (Brusca and Wilson 1991).
In the Leptostraca, the antennule consists of a three-segmented peduncle bearing a flagel-
lum and an articulated scale. The externally located scale of the Leptostraca was interpreted
by Richter and Scholtz (2001) as the “outer ramus” (presumably the homologue of the later-
ally located primary flagellum). It is probably independently derived within the Phyllocarida,
since the archaeostracan phyllocarid Cinerocaris magnifica carries two long antennulary flagella
Antennules and Antennae in the Crustacea 205

dx

vf

Fig. 7.4.
Antennule of Speleonectes (Remipedia) with details showing musculature within protopodal part extend-
ing into segmented dorsal axis (dx) and inserting on proximal rim of annulate ventral flagella (vf).
Modified from Boxshall (2004), with permission from Wiley and Sons, Ltd.

(Briggs et al. 2003). We consider that the leptostracan outer scale is derived from the accessory
flagellum despite its lateral origin. The evidence supporting this inference is the relatively late
appearance of the scale after the main axis (the primary flagellum) during development (Olesen
and Waloßek 2000) and the absence of aesthetascs, which are found only on the primary flagel-
lum. In the Bathynellacea there is an antennulary scale carried internally, and it has been inter-
preted as a rudiment of the accessory flagellum (Siewing 1959, Richter and Scholtz 2001, as the
“inner ramus”).
When first discovered, the Remipedia were described as having biramous antennules
(Yager 1981), but Boxshall (2004) revealed that the dorsal branch is the primary axis consist-
ing of segments defined by intrinsic muscles, and the ventral branch lacks intrinsic muscula-
ture and is weakly annulated (Fig. 7.4). The ventral branch is not segmented and should not
have been interpreted as a ramus: it is a flagellum. During development the primary axis only
is present in the metanaupliar phase, and the ventral flagellum first appears in the postlarva,
initially as a single “segment” lacking annulations (Koenemann et al. 2007). The remipede
antennule therefore comprises the segmented primary axis typical of other Crustacea such
as the Cephalocarida and Copepoda plus a ventral flagellum and differs from the biflagellate
antennule of malacostracans. The common basal part of the remipede antennule is referred to
as the peduncle. It is not homologous with the protopod of a biramous limb because, as for the
Malacostraca, there is no evidence that the crustacean antennule is derived from a biramous
limb (Boxshall 2004).

Multiple-Axis Antennules

A triflagellate antennule occurs in the Stomatopoda and in their Paleozoic relatives belong-
ing to the hoplocaridan orders Aeschronectida and Palaeostomatopoda (Schram 1986). The
triflagellate condition of the stomatopod antennule is an autapomorphy (Richter and Scholtz
2001). In stomatopods the ventrally located accessory flagellum is subdivided into two subflag-
ella (Kunze 1983), but only the primary flagellum carries aesthetascs (Mead and Koehl 2000),
as in other malacostracans.
206 Functional Morphology and Diversity

A B C
af

pf
pf
pf

pd pd pd

Fig. 7.5.
Development of antennule of Pleoticus muelleri (Decapoda) showing single-axis state in protozea I stage
(A) and protozoea III stage (B), consisting of peduncle (pd) and primary flagellum (pf), and biflagellate
condition of mysis II stage (C), after first appearance of accessory flagellum (af) at mysis I stage. Redrawn
from de Calazans (1992).

Within the Caridea (Decapoda), some representatives of at least three families (Palaemonidae,
Alpheidae, and Hippolytidae) also have the accessory flagellum divided to form two subflagella
(see Bacescu 1967). Given current estimates of relationships, we infer that this condition has
arisen independently in the Caridea, probably more than once.

Development

Studies on postembryonic development of crustaceans provide additional evidence support-


ing the secondary derivation of the bif lagellate antennule (e.g., Scholtz 2000, Gruner and
Scholtz 2004, Koenemann et al. 2007). The orthonauplius represents a primordial phylo-
typic stage for the Crustacea, and the naupliar antennule always comprises a single primary
axis (Fig. 7.5A,B). Any additional f lagellar structures present in malacostracans (e.g., the
accessory f lagellum) and remipedes (the ventral f lagellum) develop at a subsequent, post-
naupliar stage (Fig. 7.5C).
In segmented antennules, development follows a distal-to-proximal pattern. Articulations
separating the more distal segments are typically expressed earlier than those separating the
more proximal ones, and segmental setation is typically complete earlier on more distally
located segments than on proximal ones (Boxshall and Huys 1998, Smith and Tsukagoshi
2005). The ground-plan development pattern from nauplius to adult was modeled by Boxshall
and Huys (1998) for a hypothetical ancestral copepod (Fig. 7.6). The metamorphic molt from
Antennules and Antennae in the Crustacea 207

I V X XV XX XXV

N VI

CoI

CoII

CoIII

CoIV

CoV

Fig. 7.6.
Schematic showing hypothetical development of segmentation and setation through the copepodid stages
of Copepoda. The scale at the top indicates the presumed 28 segments of the ancestral copepod. Setae are
shown as lines, and aesthetascs, as solid elements. A segment carrying a seta not present at the preceding
stage is shaded with vertical stripes if the newly added seta is anterodistal or with stippling if the new seta
is anteroproximal. The basal segment is shown with horizontal stripes when its second and third setae
appear. Aesthetascs are shown stippled when they first appear, and then in black. Abbreviations: CoI–
CoV, copepodid stages I–V; F, female; NVI, nauplius VI stage. Adapted from Boxshall and Huys (1998)
with permission from the Royal Society of London.

the sixth and final nauplius stage to the first copepodid stage was marked by the subdivision of
the apical segment of the nauplius to form the distal eight segments of the adult antennule. No
further divisions occur in this distal section of the antennule throughout the five copepodid-
phase molts to adult. During the copepodid phase, the two proximal antennulary segments of
the nauplius undergo a sequence of subdivisions to form segments 1–20 of the adult. Antennules
with fewer expressed segments are envisaged as being generated by early cessation of the proc-
ess of subdivision (Boxshall and Huys 1998, Schutze et al. 2000), a pedomorphic process.
Proximal annulation is expressed transiently during the naupliar phase of development of
some podoplean copepods but is lost by the first copepodid stage (e.g., Dahms 1992). The pro-
tozoeal larvae of penaeid decapods similarly exhibit transient annuli in the proximal part of
the antennule (Fig. 7.5A), which are lost by the end of the zoeal phase. The proximal annulated
part of the antennule of Rehbachiella and Bredocaris may be interpreted as additional evidence
of their larval status but may also indicate that a proximal annulated zone is plesiomorphic for
the Crustacea.
In decapod malacostracans such as Panulirus argus and Cherax destructor, the primary
antennulary flagellum develops by the production of new annuli at the base of the flagellum,
in a meristematic zone (Sandeman and Sandeman 1996, Steullet et al. 2000). In both species
208 Functional Morphology and Diversity

subdivision takes place in annuli distal to the basal meristematic annulus. Few details are avail-
able but the basic process seems similar to that described for the antennal flagellum in isopods
(see “Development,” below).

ANTENNAE

The second limb in arthropods, the one immediately posterior to the antennule, has two main
names: in Crustacea, it is the second antenna or just antenna; in Chelicerata, it is the pedipalp. We
use the term antenna here. In Hexapoda and Myriapoda this limb is missing; its absence serv-
ing as a diagnostic character of the taxon Atelocerata (Heymons 1901). However, the antennal
segment is represented by the limbless intercalary segment. The expression pattern of the gene
engrailed in the epidermal and neural cells of hexapods unequivocally marks the intercalary seg-
ment, and its existence is further supported by the expression pattern of the Hox gene labial in
insects (Diederich et al. 1989, Peterson et al. 1999).
The Silurian arthropod Tanazios was described by Siveter et al. (2007) as a possible
stem-lineage crustacean. Boxshall (2007) reinterpreted the antenna of Tanazios as missing in
the adult—its absence is indicated by the marked gap in the limb series between the antennule
and the mandible. He concluded that the lack of differentiation in the postmandibular limbs of
Tanazios excludes it from the Eucrustacea. Boxshall (2007, 322) considered the lack of anten-
nae in a Silurian marine arthropod to be “of immense significance both to our understanding
of deep mandibulate phylogeny, and to the emerging, but as yet unstable, picture of hexapod
origins within the Pancrustacea.”

Structure

The basic postantennulary limb of arthropods is widely regarded to be biramous (see Walossek
1993, Hou and Bergström 1997, Boxshall 2004), comprising a proximal stem, the protopod, bear-
ing an inner endopod and outer exopod. These rami are major limb axes and are carried dis-
tally on the protopod. They are plesiomorphically supplied with intrinsic muscles originating
within the protopod and are commonly segmented. The antenna of crustaceans is plesiomor-
phically biramous: the protopod is two-segmented, comprising coxa and basis; the exopod is
usually multisegmented, and the endopod typically comprises only four expressed segments. In
malacostracans the protopod may be three-segmented (Fig. 7.7), but there is no evidence for the
existence of a precoxa in the antenna of cephalocaridans, remipedes, branchiopods, copepods,
mystacocaridans, podocopan ostracods, myodocopans, or branchiurans. The Tantulocarida
lack antennae at all stages of the described life cycle (Huys et al. 1993). The nonfeeding plank-
tonic adults of monstrilloid copepods lack all cephalothoracic limbs from antennae to maxil-
lipeds, inclusive. The infective nauplius phase, however, has biramous antennae with a setose,
natatory exopod and an endopod that carries a terminal clawlike seta (Grygier and Ohtsuka
1995). The antennal endopod together with the clawed mandible are used to grasp the host sur-
face during infection.
The antenna of Cambrian phosphatocopines is biramous: the protopod is unsegmented or
subdivided into coxa and basis, the endopod is two- or three-segmented, and the exopod is
described as multiannulate (Maas et al. 2003). When undivided, the protopod forms an entire
gnathobase; when divided, both coxa and basis are drawn out medially into a spine-bearing
endite (Siveter at al. 2001). The number of “annuli” in the exopod ranges from about 8 up to 24,
and these typically carry a single seta each, with two at the apex (Maas et al. 2003).
In the Anostraca, the antennae exhibit extreme sexual dimorphism. In adult males they may
be modified as enormous grasping structures, often with species-specific basal outgrowths
Antennules and Antennae in the Crustacea 209

A B

en3

en2

ex
en1 en3

pr3
pr2

pr1
en2

pr1
en1

pr3
pr2

Fig. 7.7.
(A) Biramous antenna of Hemimysis (Peracarida: Mysidacea) showing intrinsic musculature within
three-segmented protopod (pr1–pr3), within the proximal three endopodal segments (en1–en3) insert-
ing on the proximal rim of the endopodal flagellum and within the exopod (ex). (B) Uniramous antenna
of Typhlocirolana (Peracarida: Isopoda) showing intrinsic musculature within three-segmented proto-
pod (pr1–pr3) and proximal three endopodal segments (en1–en3), inserting on the proximal rim of the
endopodal flagellum. The exopod is absent.

and elaborate ornamentation, whereas in females they are typically smaller and simpler in
construction (Linder 1941). In conchostracan branchiopods the antenna is biramous, and
both of the large rami appear to be f lagellate, but both exopod and endopod comprise numer-
ous segments as defined by the presence of intrinsic muscles (Shakori 1968, Boxshall 2004).
The multisegmented nature of the endopod was interpreted as secondarily derived within
the Conchostraca by Boxshall (2004). In cladocerans the number of exopodal and endopo-
dal segments does not exceed three in Anomopoda, Onychopoda, and Ctenopoda or four in
Haplopoda and the fossil order Cryptopoda. In Holopedium , the antenna is uniramous due to
loss of the endopod.
The cephalocaridan antenna is biramous, comprising a protopod divided into coxa and basis,
a short two-segmented endopod, and a well-developed, multisegmented exopod (Sanders 1963).
The number of exopodal segments and associated setal elements increases during ontogeny;
Hutchinsoniella, for example, has 13 segments bearing 15 setae at the metanauplius stage, rising
to 19 and about 70 setae in the adult (Elofsson and Hessler 1991). The setae are structurally simi-
lar, with the lumen of each containing 6–12 unbranched outer dendritic segments surrounded
by two layers of sheath cells. All the setae were interpreted as chemosensors by Elofsson and
Hessler (1991), who remarked on the lack of mechanoreceptors, not having found scolopale bod-
ies associated with any of the setal sensory cells (see chapter 6).
210 Functional Morphology and Diversity

In mystacocarids, cirripedes, and copepods, the antenna is typically biramous, comprising


coxa and basis, which bears the exopod and endopod. The endopod is apparently six-segmented
in mystacocarids, but in copepods it is typically only three-segmented, although traces of an
ancestral four-segmented condition are retained in some basal calanoids (Huys and Boxshall
1991). It is only two-segmented in cirripede nauplii. The exopod has at most nine segments
in mystacocarids and cirripede nauplii and ten segments in copepods. However, within the
Thecostraca, ascothoracid metanauplii have antennal exopods of 12 or more segments (Boxshall
and Böttger-Schnack 1988).
In malacostracans the protopod has typically been regarded as two-segmented, comprising
coxa and basis, but we find in Tulumella (Thermosbaenacea), Hemimysis (Fig. 7.7A), Stygiomysis
(Mysidacea), and Typhlocirolana (Isopoda) (Fig. 7.7B) that the protopod is three-segmented.
The coxa articulates proximally with a precoxa that may have been interpreted as a pedestal, but
the presence of muscles inserting at this joint indicates that it is a segment. The antennal endo-
pod in malacostracans is well developed, consisting of up to three defined segments proximally,
plus a terminal flagellum representing a subdivided fourth segment. The proximal endopodal
segments are usually referred to as peduncle segments, so the six-segmented peduncle of some
peracaridans comprises a combination of three protopodal and three endopodal segments (Fig.
7.7), whereas in other peracaridans and euphausiaceans the peduncle is five-segmented, consist-
ing of two protopodal and three endopodal segments. Muscles are present only in the peduncle
segments (Fig. 7.7). In the syncarid Anaspides the peduncle is only four-segmented, comprising
coxa, basis, and two defined endopodal segments plus the flagellum.
In leptostracans the exopodal scale is absent, and the antenna is uniramous, but in the
archaeostracan phyllocarid Cinerocaris the antenna is biramous with two slender flagellate rami
(Briggs et al. 2003). In eumalacostracans the exopod is reduced to a scale, although in stomato-
pods the scale has a well-defined proximal segment (Schram and Hof 1998). The exopodal scale,
also referred to as the scaphocerite, is greatly enlarged in the Amphionidacea and in decapod
shrimps but can be reduced or lost in anomurans and brachyurans. In asellote isopods and in
the thermosbaenacean Tulumella, there is also a scalelike exopod, but other members of these
orders, as well as members of the Cumacea and Amphipoda, have uniramous antennae lack-
ing any trace of the exopod (Fig. 7.7B). The flattened, platelike antenna of scyllarid lobsters is
derived from the flagellum, not from the exopodal scale.
Extreme development of the antennal flagellum is exhibited in penaeidean decapods, in which
it can attain a total length more than three times that of the body. In these penaeideans, the flagel-
lum comprises a relatively rigid proximal section and a long whiplike distal section, separated by
a flexure zone. During normal swimming, the proximal section is held out from the body, and the
flexible distal part trails out parallel to and at a distance from the animal (Foxton 1969). Distal
to the flexure zone, each annulus typically carries a pair of arched, plumose setae that together
effectively form a tube running the entire length of the distal section. Within this “tube” a linear
array of finely plumose setae can be present, although there is not necessarily one per annulus.
This system allows the detection of near-field vibrating sources and can indicate the direction of
the source because the near-field effect attenuates rapidly (i.e., along the length of the flagellum)
(Denton and Gray 1986).
The stiffness of the flagellum is nonuniform. In Cherax destructor, the flagellum is tapered
and consists of 220–250 annuli that increase in length and decrease in diameter toward the tip
(Sandeman 1989). The flagellum is oval in cross section and exerts the greatest mechanical
resistance against medial deflection and the lowest against dorsal deflection. Medial mechani-
cal resistance decreases markedly toward the tip. When bent passively, curvature is confined to
a small part of the flagellum, but with increasing force, the apex of the bend moves toward the
base. Sensing curvature at several locations along the flagellum is likely to provide sufficient
Antennules and Antennae in the Crustacea 211

information to enable the crayfish to distinguish whether the bending is caused by the flow
fields in the surrounding medium or by physical contact with an obstacle (Barnes et al. 2001).

Development

During the early naupliar phase, the antenna bears one or two feeding endites, referred to as
the naupliar or antennal processes , on the proximal segment (the coxa) of the two-segmented
protopod. These naupliar processes are found in copepods, mystacocarids (Fig. 7.8A),
cephalocarids (Fig. 7.8B), and branchiopods—taxa that retain nonlecithotrophic nauplii.
They decrease in size during later development and are lost at or before the molt to the post-
naupliar phase.
In dendrobranchiate decapods, the antenna of the naupliar and protozoeal phases has a mul-
tisegmented exopod that gradually ceases to express segmentation during development until,
by the megalope phase (= mysis I stage), the exopod has transformed into the unsegmented
antennal scale (Fig. 7.9) so characteristic of the adult caridoid facies (Hessler 1983). This transi-
tion from segmented naupliar ramus to unsegmented scale is unique to these malacostracans. It
is accompanied by a change in form of the endopod, from a two-segmented ramus (Fig. 7.9A) to
an annulate flagellum (Fig. 7.9C).

A B

np

np

Fig. 7.8.
(A) Antenna of first stage metanauplius of Derocheilocaris (Mystacocarida) showing large feeding nau-
pliar processes (np) on coxa (redrawn from Hessler and Sanders 1966). (B) Antenna of first stage nau-
plius of Hutchinsoniella (Cephalocarida) showing developing naupliar process on coxa (redrawn from
Sanders 1963).
212 Functional Morphology and Diversity

A B C

en en en

ex ex ex

Fig. 7.9.
Developmental of the antenna of Pleoticus muelleri (Decapoda) showing transition in form of exopod (ex)
and endopod (en) from protozoea III stage (A) to mysis I stage (B), and megalopa stage (C). Redrawn from
de Calazans (1992).

The isopod Asellus aquaticus is a useful model for the development of the endopodal flagel-
lum, which consists of a single segment divided into annuli that are devoid of intrinsic muscu-
lature (Wege 1911). The antennal flagellum comprises a proximal meristematic region, a central
region composed of quartets (sets of four annuli, each having a specific arrangement of setae),
and an apical complex consisting of the apical annulus plus the four preceding annuli with spe-
cific setal patterns. The number of quartets in the central region is variable in Asellus aquaticus
since this species never ceases molting and continues to add annuli throughout life (Maruzzo
et al. 2007). The proximal meristematic annulus divides into a copy of itself (the meristem) and
a distal annulus that is effectively an incomplete quartet and divides following a set pattern to
produce the complete quartet.

SENSORS

In most crustaceans, the antennules are primarily sensory in function and carry rich arrays
of sensory receptors that play a role in almost every aspect of their behavior. The main setal
receptor types can be defined by their sensory modality, and although the sensitivity and func-
tional responses of most setal elements have not been experimentally investigated, compara-
tive studies allow for putative functions to be inferred. Setal elements with a sensory function
are commonly referred to as sensilla (see Hallberg and Skog 2011), but the terminology has not
yet stabilized (see chapter 6). In some taxa, such as copepods, aesthetascs are regarded as dis-
tinct from setae rather than as a setal subtype.
Various terminologies are employed to discuss crustacean sensory modalities, and this can
generate confusion. Chemosensors that mediate perception of a remote chemical are referred
to as olfactory. Olfaction (sense of smell) is distance chemoreception and usually involves the
detection of a gradient of chemical entrained within a plume originating at its source, such
as a pheromone produced by congeneric females or amino acids originating from food items.
Chemosensors that mediate perception of a chemical stimulus at extremely close range, or
when in direct physical contact, are often referred to as gustatory or taste receptors. This is con-
tact chemoreception (Atema 1977). Distance and contact chemoreceptors represent extremes
Antennules and Antennae in the Crustacea 213

in sensitivity along a continuous spectrum, so the separation is somewhat artificial although


widely adopted (see Hallberg and Skog 2011).
There is a similar functional division in mechanoreceptors. Mechanosensors that mediate
remote perception of hydrodynamic signals, such as disturbances in flow fields caused by par-
ticles suspended in the water column or by the swimming motions of a live prey, are distance
mechanosensors. Sensors that detect the presence of objects only when they come into direct
physical contact with a surface are tactile sensors. Again, distance and tactile sensors repre-
sent extremes in sensitivity along a continuous spectrum of mechanosensors. Displacement of a
mechanosensory setal shaft in its socket or the bending of a seta in direct contact with an object
can both constitute a stimulus.

The Aesthetasc

Aesthetascs are the most studied crustacean chemosensor. They are typically elongate, thin-
walled, cylindrical elements (Fig. 7.10), which lack internal enveloping cells in their outer part
(Hallberg and Skog 2011). They are innervated by sensory cells that vary in number from a few to
hundreds, for example, 400 in Pagurus hirsutiusculus (Ghiradella et al. 1968a). The sensory cells
are surrounded by enveloping cells basally. They have dendrites from which cilia emerge, usu-
ally two cilia in malacostracans, sometimes only one, sometimes more. The cilia typically, but
not always, branch dichotomously and repeatedly to form a dense mass of outer branches called
outer dendritic segments (Laverack 1964, Hallberg et al. 1992). Gr ü nert and Ache (1988) reported
8,000–10,000 such outer dendritic segments per aesthetasc in Panulirus argus.
Aesthetascs’ function as a chemoreceptor has been confirmed experimentally by direct neu-
rophysiological studies in various decapod malacostracans (Anderson and Ache 1985, Michel
et al. 1993, Steullet et al. 2000). The permeability of the thin cuticle allows water-borne com-
pounds to enter over the entire surface area, but the cuticle remains sufficiently impermeable to
prevent loss of physiologically important molecules (Derby et al. 1997).
The aesthetasc can vary in external form: curved aesthetascs can be found, for example, in
some isopods (Ka ï m-Malka et al. 1999) and some cavernicolous copepods (Boxshall and Jaume
2000). They vary in size with taxa: body size is a factor influencing aesthetasc length. Aesthetascs
in Cancer productus can attain a length of >1700 μ m, one of the longest in the Crustacea, whereas
those of small cladocerans or interstitial copepods may reach only 10–12 μ m in length (Hallberg
et al. 1992, Boxshall et al. 1993). Aesthetascs increase in size during development in the stomat-
opod Gonodactylus mutatus (Mead 2002). As well as varying in size with species, aesthetascs
can be sexually dimorphic, with males carrying larger aesthetascs than conspecific females
(Boxshall and Huys 1998).
Terrestrial crustaceans retain functional aesthetascs (Ghiradella et al. 1968b). In the robber
crab Birgus latro, the antennules consist of an elongate, three-segmented peduncle bearing two
short distal flagella, and the aesthetascs are arrayed in rows across the central groove of the pri-
mary flagellum. They differ from those of aquatic relatives in being relatively short and blunt.
In addition, they have an asymmetric profile, with the side nearest the surface of the annulus
being thickened and the exposed surface being thin and wrinkled (Stensmyr et al. 2005). The
thickened surface and flattened profile presumably enhance the rigidity of the aesthetasc out
of water.
The arrangement of aesthetascs on the antennule varies but is normally taxon specific and
relatively highly conserved within a higher taxon. Cephalocarids, for example, carry just a sin-
gle terminal aesthetasc. In adult Hutchinsoniella the aesthetasc is located in a terminal tuft,
together with 11 other setae, and a further 11 setae are distributed over the more proximal seg-
ments (Elofsson and Hessler 1991). In anostracan and cladoceran branchiopods the aesthetascs
214 Functional Morphology and Diversity

ae

ods

cut
isc

osc

ods

bb

osc
ids

isc isc

Fig. 7.10.
Fine structure of typical malacostracan aesthetasc. The aesthetasc (ae) has a short stalk attached to the
cuticle (cut), and its lumen is filled with a mass of branching outer dendritic segments (ods), which arise
from inner dendritic segments (ids) within the antennule. Just proximal to the origin of the outer dendritic
segments are the 9 × 2 + 0 ciliary basal bodies (bb). The array of dendritic segments is sheathed with mul-
tiple layers formed by inner (isc) and outer (osc) sheath cells. Schematic shown with reduced numbers of
components, there are multiple dendrites and a variable number of sheath cells. Adapted from Heimann
(1984), with permission from Springer.

are typically found in a tuft on the apex of the antennule, whereas in the Notostraca and most
Spinicaudata they are arrayed along the margin (Martin 1992). In mystacocarids, there is a
single aesthetasc located subterminally on the apical segment (Olesen 2001). The sensory
tubes described from the antennules of the myodocopan Conchoecia spinirostris are probably
aesthetascs, although they differ from the typical decapod aesthetasc (Heimann 1979). In the
remipede Speleonectes tanumekes the antennule carries several dozen aesthetascs arranged in
parallel rows on the peduncle (van der Ham and Felgenhauer 2006).
In planktonic copepods, Giesbrecht (1893) observed that, with the exception of the basal and
a few distal segments, all antennule segments carry a trithek of setal elements (one aesthetasc
and two setae). Traces of this pattern can be identified in all copepods, whether planktonic, ben-
thic, or parasitic (Huys and Boxshall 1991). Giesbrecht also noted that males in some calanoid
Antennules and Antennae in the Crustacea 215

families possess two aesthetascs per segment, giving a quadrithek arrangement (two setae plus
two aesthetascs) compared with the trithek of the female. Quadritheks occur on one or more
segments of the male antennules in many calanoids (Huys and Boxshall 1991). In its extreme
form, double aesthetascs are present along almost the entire length of the male antennules
(from segments II to XXIV in Eucalanus attenuatus), whereas in other calanoid families they are
restricted to the proximal part of the antennule.
In malacostracans aesthetascs are found only on the primary (= lateral) flagellum. They are
typically aligned in regular rows, either one or two rows per annulus, and the number of aes-
thetascs per row varies with taxon (Table 7.1). The rows may be distributed along the length of
the flagellum or restricted to the more proximal part. In some amphipods, the proximal annuli
are completely or partially fused and bear multiple transverse rows of aesthetascs, forming an
organ referred to as the callynophore by Lowry (1986). Similarly, in the isopod Idotea baltica,
aesthetascs are carried in six or seven rows on the distal antennulary “segment” (Guse 1983), but
this is undoubtedly a compound unit representing several annuli that failed to separate during

Table 7.1. Arrangement of aesthetascs on the primary flagellum of malacostracans.

Higher taxon Species No. rows per No. aes- Reference


annulus thetascs per
row
Stomatopoda Gonodactylus 1 3 Mead 2002
mutatus
Decapoda Panulirus argus 2 8–12 Daniel et al.
2008
Decapoda Orconectes virilis 2 3–4 Mead 2008
Decapoda Homarus 2 ~12 Derby 1982
americanus
Decapoda Typhlatya arfeae 1 1–3 Jaume and
Bréhier 2005
Decapoda Lysmata boggessi 2 4–5 Zhang et al.
2008
Decapoda Birgus latro 2 ~12–14 Stensmyr
et al. 2005
Lophogastrida Lophogaster 2 1–2 (30–40 in Johansson
typicus males) et al. 1996
Isopoda Bathynomus pelor 1 4–5 Thomson
et al. 2009
Isopoda Natatolana 1 10 Ka ï m-Malka
borealis et al. 1999
Isopoda Asellus aquaticus 1 1 Heimann 1984
Isopoda Idotea baltica — 2 Guse 1983
Amphipoda Metacrangonyx 1 1–2 Jaume and
dominicanus Christenson
2001
Amphipoda Bathymedon 2 4–5 Jaume et al.
longirostris 1998
216 Functional Morphology and Diversity

development. Fusion of proximal annuli and subsequent concentration of aesthetasc rows occur
elsewhere in eucaridan and peracaridan malacostracans. In the Brachyura, the aesthetascs are
arrayed in a dense tuft on the ventral surface at the tip of the flagellum.
The fine scale arrangement of aesthetascs, such as the precise angle at which they are carried
relative to the segment, can have functional significance (Mead and Weatherby 2002, Thomson
et al. 2009). The guard setae located on either side of the aesthetasc rows in Panulirus argus have
a dual function: protecting the array from damage and conditioning the flow to allow for the
correct leakiness during a flick and return sequence (Reidenbach et al. 2008).

Unimodal Nonaesthetasc Chemosensors

Several different morphologies of nonaesthetasc chemosensory setae are found in the


Crustacea (Hallberg and Skog 2011). The presence of a terminal pore on a setal element is
often taken as an indication of a chemoreceptive function. Such setae often have their outer
dendritic segments shielded by a dendritic sheath of enveloping cells, so their sensory surfaces
are exposed only near the tip. This setal type probably functions as a contact chemoreceptor.
Several are present on the antennule of the cirripede cyprid larva (see Bielecki et al. 2009) that
carry a single terminal setal element (TS-D in terminology of Bielecki et al. 2009) identified
as an aesthetasc on the basis of external morphology. However, at least in Megabalanus rosa ,
seta TS-D has a terminal pore, which is typically absent from true aesthetascs, and Bielecki
et al. (2009) preferred to call it “aesthetasc-like.” Its inferred function was detection of water-
borne chemicals along its entire length.
The setae on the cephalocaridan antennule all share a similar structure: each contains an
outer dendritic segment that arises from a single ciliary body (9 × 2 + 0 microtubule pattern),
which in turn arises from a basal body connected to the sensory cell via the inner dendritic
segment. All the setae lack scolopale bodies and display features suggesting chemoreception
(Elofsson and Hessler 1991).
In two superfamilies of Podocopa (Cypridoidea and Pontocypridoidea), a modified seta
known as “aesthetasc Y” is carried on the first endopodal segment of the antenna. Its internal
structure was studied in detail by Andersson (1975), who inferred that it is a chemoreceptor.
Kaji and Tsukagoshi (2008) compared the development and structure of “aesthetasc Y” and the
linear array of so-called grouped setae located in a similar position on the first endopodal seg-
ment on five other ostracod superfamilies. They suggested that “aesthetasc Y” is derived from
grouped setae.
In cypridoidean ostracods, the Wouters organ is found on the dorsal surface of the first
antennulary segment, and the Rome organ is found on the ventral side of the second moveable
segment of the antennule in cypridoideans (Smith and Matzke-Karasz 2008). Both the Rome
and Wouters organs vary in form from buttonlike to tubular and may be derived from setae.
They were presumed to be chemosensors by Smith and Matzke-Karasz (2008).

Unimodal Mechanosensors

Mechanosensory setae are rigid structures with relatively thick cuticles, and in copepods each
is usually innervated by two dendrites, sometimes only one (Gresty et al. 1993a). The struc-
ture and innervation of the mechanosensory seta of Pleuromamma xiphias (Fig. 7.11), a calanoid
copepod, was described in detail by Weatherby and Lenz (2000). A pair of long dendritic proc-
esses extends from the paired sensory cell bodies (the somata), located in the antennule itself,
to the seta. Distal to each soma, the basal body is located within the dendrite about at the level
where the scolopale begins. The scolopale tube extends almost to the socket at the base of the
Antennules and Antennae in the Crustacea 217

se cut

ods

isc
osc

tsc

mt tsc

bb
ac
scc

ids

scb

ax

Fig. 7.11.
Schematic of typical copepod mechanosensory seta. The seta (se) articulates basally with the cuticle (cut).
It is innervated by a pair of mechanosensory dendrites which pass from the cell bodies, through the tubular
scolopale (tsc) contained within the scolopale cell (scc) sensory cell body (scb) before entering the base
of the seta and inserting on its inner wall. The whole structure is enclosed by inner (isc) and outer (osc)
sheath cells and is anchored to the cuticle by microtubules (mt) within the anchor cell (ac). At the level
where the dendrite enters the scolopale are the 9 × 2 + 0 ciliary basal bodies (bb), marking the division
of the dendrite into inner (ids) and outer segments (ods). The nerve cell connects to the central nervous
system via an axon (ax). Adapted from Weatherby and Lenz (2000), with permission from Elsevier.

setal shaft. More distal dendrites become densely packed with a regular array of proliferating
microtubules. Distally the microtubule array and surrounding dendrites bend and enter the
lumen of the seta, inside which the dendrites are enveloped by inner and outer sheath cells. The
dendrites and their microtubules terminate at attachments to the setal cuticle.
The scolopale cell is typical of arthropodan mechanosensors, and many of the cuticular
setae in decapods are scolopidial in structure (Ball and Cowan 1977). The copepod mech-
anoreceptor described above (from Weatherby and Lenz 2000) differs from a typical scolo-
pidial organ, which is characterized by the possession of a ciliary body with very large rootlets.
These are in close contact with the dendritic membrane that itself makes desmosomal con-
nections to the scolopale, firmly anchoring it to the scolopale (Schmidt and Gnatzy 1984). In
Pleuromamma , dendrite anchoring appears to occur by the tight fit of the microtubule-packed
dendrites within the tubular scolopale. In many crustaceans there is an extracellular den-
dritic sheath or cap that ensheathes the distal dendrites (Guse 1978), but none is present in
Pleuromamma .
Transduction in a mechanosensor is the process by which a mechanical stimulus—a force
acting to displace the seta—is transformed into a cellular physiological response. Weatherby and
218 Functional Morphology and Diversity

Lenz (2000) suggest a likely mechanism for mechanotransduction involving linkages between
individual microtubules and mechanogated channels in the dendritic membrane. The scolopale
increases the rigidity of the system, and this rigidity probably contributes to high-frequency
sensitivity. The attachment of the distal dendrites within the setal lumen instead of at the base
probably enhances the physical displacement at the transduction site, promoting high sensitiv-
ity. The highly specialized receptor allows this planktonic copepod to detect the hydrodynamic
disturbances generated by an incoming predator.
It is almost always the movement of the shaft in its socket that serves as the stimulus, but
during molting in Panulirus argus the scolopale body is briefly located in the basal region of the
setal shaft, and Schmidt and Derby (2005) suggested that, in this configuration, bending of the
setal shaft itself could serve as an adequate stimulus.

Bimodal Chemo- and Mechanosensors

Antennulary setal elements that are bimodal—with both mechanosensory and chemosensory
function—are well known in decapods and other malacostracans (Schmidt and Ache 1996,
Ka ï m-Malka et al. 1999). They have also been reported from cirripede cyprid larvae (Lagersson
et al. 2003), in copepods (Laverack and Hull 1993, Weatherby et al. 1994), and in anostracan
Branchiopoda (Tyson 1980), and they are probably widespread in the Crustacea.
In the model decapod Panulirus argus, the aesthetasc rows on the antennulary annuli are
accompanied by setal elements, including guard setae, companion setae, and asymmetric setae.
Guard setae flank the distal row of aesthetascs, and one or two companion setae are located
lateral to each guard seta. The asymmetric seta is located between the end of the aesthetasc
row and the guard seta. Only the asymmetric seta has been studied both ultrastructurally and
electrophysiologically (Schmidt and Derby 2005). The asymmetric seta of Panulirus argus is
bimodal. It has a terminal pore at the tip of the setal shaft, indicative of a chemosensory func-
tion, plus an associated scolopale body located in the annulus below its base, indicative of a
mechanosensory function. Its position on the annulus provides for a loose mechanical coupling
with the aesthetasc row.
The hooded sensillum of decapods is another bimodal element. In Panulirus argus the
hooded sensillum is the second most abundant setal type on the antennulary f lagella
after the aesthetascs. Each has a porous cuticle and is innervated by nine or ten chemo-
sensory and three mechanosensory neurons whose dendrites project to the tip (Cate and
Derby 2002). The chemosensory neurons respond to waterborne chemical stimuli, while
the mechanosensory neurons respond to direct tactile stimulation, not to distant hydro-
dynamic signals. Similarly, the pocilliform seta on the antenna of the scavenging isopod
Natatolana borealis has a subapical pore and is probably a chemosensor (Ka ï m-Malka
et al. 1999).

Statocysts

Statocysts are paired equilibrium organs that contain mechanosensory setae responsive to
changes in spatial orientation. In dendrobranchiate, caridean, and reptantian Decapoda,
they are carried in the basal segment of the antennulary peduncle. They are also present
in Anaspides and in the mysid Hansenomysis , according to Siewing (1956). In decapods,
each statocyst consists of an invaginated cavity containing a statolith formed from sand
grains cemented together by a secretion from integumental glands in the f loor of the cyst
(Prentiss 1901). Cysts are closed to the external environment and house two circular canals,
one vertical and one horizontal, formed by compression of the cyst walls. Three groups of
Antennules and Antennae in the Crustacea 219

Fig. 7.12.
Scanning electron micrograph of calceolus from antenna of Eusirus perdentatus (Amphipoda). Photo
courtesy of R. Lincoln.

mechanosensory setae are located in the f loor of the statocyst and support the statolith. The
setae, known as “statolith hairs,” form concentric rows and are stimulated by shearing force
created by movement of the statolith or by f luid movement within the circular canals. As the
animal changes its orientation relative to gravity, the statolith is displaced creating a shear-
ing force on the “statolith hairs.” Responses of receptors within the decapod statocyst con-
trol ref lexive compensatory and righting movements of the animal (see references in Cate
and Roye 1997).

The Calceolus

Calceoli are enigmatic setal elements found in gammaridean amphipods (Lincoln and Hurley
1981). Each calceolus (Fig. 7.12) typically consists of a cup-shaped proximal element and a
transversely banded distal element, both supported by a main receptacle and attached to the
antenna by a short stalk. They can occur singly or in small groups on the f lagellar annuli, and
occasionally also on the peduncle. They frequently occur in males only, sometimes in both
sexes, but never in females only. They can occur on the peduncle and f lagellum of the anten-
nule and antenna but are often confined to the antenna only. Dahl et al. (1970) inferred that
calceoli are chemoreceptors, functioning as pheromone detectors, but Lincoln and Hurley
(1981) concluded that they were mechanosensors despite being unable to confirm the pres-
ence of a nerve supply.

FUNCTIONAL MORPHOLOGY AND BEHAVIOR

The antennules of recent crustaceans are primarily sensory in function and form an important
sensory interface with the environment. However, the hallmark of the arthropodan limb is its
multifunctional potential, and antennules may play a role in many other behaviors, including
locomotion, attachment, feeding, and mating. Performance of multiple functions is usually
reflected in morphological adaptations, but in addition, comparative studies have revealed
220 Functional Morphology and Diversity

numerous examples of adaptive fine-tuning of the sensory array, by changes in the number of
modular sets of sensory elements, by increasing or decreasing annulation, or by changes in the
size or density of receptors. The antennae often perform a sensory role but can be similarly
involved in a variety of other behaviors, including swimming, attachment, respiration, feeding,
and brooding.

Swimming

Nauplii have only three pairs of limbs that perform a range of functions. The antennules already
fulfill the sensory role that continues through to the adult, but in addition they may be involved
in swimming and feeding in certain taxa. In cirripede nauplii, for example, the antennules are
typically held directed forward while the paired antennae and mandibles generate the swim-
ming motions (Rainbow and Walker 1976). In such cases, the role of antennules is presumably
mainly sensory, whereas in calanoid copepods the antennules are often paddle shaped and setose
and can contribute significantly to swimming (Bjørnberg 1986). The antennae are important
locomotory limbs in all swimming nauplii.
Pelagic myodocopans such as Conchoecia swim almost continuously in the laboratory,
using their antennal exopods, and begin sinking as soon as swimming motions cease dur-
ing feeding episodes (Lochhead 1968). In podocopan Ostracoda, the antennules often have
a locomotory function. This includes swimming in the Cypridoidea and crawling or digging
in many Cytheroidea and in Darwinuloidea. Podocopans, such as Cypridopsis vidua, use both
antennules and antennae in swimming. The motions of thrust-producing limbs are uniquely
coordinated with the power strokes synchronized diagonally so that the left antennule and
right antenna start together and end just as the right antennule and left antenna commence
their power stroke (Hunt et al. 2007). During the power stroke, the long setae located on the
four distal segments of the antennule and the array of setae on segment 3 of the antenna splay
out into a fanlike configuration that accounts for >75% and about 50%, respectively, of the
surface area in contact with the medium. During the recovery stroke, these natatory setae col-
lapse back against the limb shaft. Swimming in this species is smoothly continuous because
power strokes occur throughout the stroke cycle, in contrast to other small crustaceans, such
as Daphnia, which exhibit rapid accelerations and decelerations due to the alternate power and
recovery strokes of the antennae (the swimming limbs).
Cladocerans propel themselves through the water by the swimming motions of the paired
antennae that have robust protopods housing powerful muscles. The zone of conspicuously
folded cuticle at the base of each antenna (Fig. 7.13) functions to increase the angle through
which the limb can swing during swimming. They carry arrays of long plumose natatory setae
that are typically splayed out to maximize their surface area in contact with the water during the
swimming power stroke. The antennae are folded toward the body to minimize their contribu-
tion to drag during the recovery stroke (Zaret and Kerfoot 1980). The predatory haplopodan
branchiopod Leptodora kindti swims continuously in the water column using its large anten-
nae to attain speeds up to 28 mm/s (Browman et al. 1989). The large, powerful antennae are
also important swimming limbs in the Laevicaudata and Spinicaudata. The long antennules of
planktonic copepods are folded back against the body during swimming, but evidence that they
contribute to swimming itself is lacking.
Many malacostracans use their abdomen to perform a powerful tail-flip swimming action.
This is an energetically expensive behavior and is usually employed as a startle escape response
from an attacking predator. Tail-flip swimming in shrimps such as Crangon involves motions
of a head fan, formed by the antennal scales, as well as the tail fan, formed from telson and uro-
pods. The antennal scales pivot laterally during body flexion to form an expanded propulsive
Antennules and Antennae in the Crustacea 221

Fig. 7.13.
Scanning electron micrograph of male Moina (Branchiopoda) in lateral view, showing the curved form
of antennules (a1) and a large biramous antenna (a2) with a robust protopod and proximal zone of highly
folded cuticle (arrow).

surface, the head fan, which assists in the generation of thrust during tail-flips (Arnott et al.
1998). In the heavily armored scyllarid lobsters, lift is created during tail-flips, and changes in
orientation of the transformed, platelike antennal flagella are used to alter the distribution of
this lift to create pitching and rolling movements during maneuvers.

Attachment

Attachment to Substratum

The cyprid larva of cirripedes is a transitional stage marking the end of the planktonic dispersal
phase of the life cycle and the beginning of the sessile adult phase. The highly mobile cyprid
is specialized for settlement: the task of selecting a site for permanent attachment and meta-
morphosis, mediated by specific environmental cues (Crisp and Meadows 1962). The paired
antennules play a key role in settlement behavior in free-living cirripedes. Cyprid larvae explore
surfaces by walking across the substratum in a bipedal fashion, using alternate movements of
their antennules (Nott and Foster 1969). The antennules are equipped with an attachment disc
and cement gland for temporary adhesion to the substratum during walking, and they also carry
an array of sensory elements (Nott and Foster 1969, Walker 1971), almost all of which were con-
sidered to be bimodal by Lagersson et al. (2003).
The attachment disc, carried on the third segment of the four-segmented antennule, is a
disc-shaped pad, the surface of which is densely covered with cuticular villi (Fig. 7.14). The pad
is surrounded by a membranous velum and also contains the opening of the cement gland duct.
The cement gland is located in the head, close to the brain, and a long duct with a muscular sac
located proximally carries the glycoproteinaceous cement down the antennule into the disc.
The mechanism that allows rapidly reversible, temporary attachment via the discs is poorly
understood, but a preliminary study of the footprints left by a walking cyprid led Phang et al.
(2007) to suggest that a solely viscoadhesive mechanism, with the cement functioning as an
adhesive, could not generate sufficient adhesive force to explain the real attachment tenacity of
the cyprids.
222 Functional Morphology and Diversity

Fig. 7.14.
Scanning electron micrograph of attachment disc on antennule of cypris larva of Megabalanus (Cirripedia),
showing cuticular villi covering the surface. Photo courtesy of J. Høeg.

In parasitic cirripedes, the Rhizocephala, the antennules play a major role in the unique host
invasion process. As with other cirripedes, rhizocephalans typically hatch as a nauplius larva
that, after four or five stages, metamorphoses into the cyprid larva. In the Kentrogonida, female
cyprids attach to the surface of their crustacean hosts using their antennular cement glands and
undergo a postsettlement molt into a flask-shaped, infective kentrogon stage that becomes firmly
attached to the host. Internally, the kentrogon forms a stylet that it uses to penetrate the host cuti-
cle and inoculate the next stage, the vermigon larva, into the host. This vermigon develops to form
the adult body comprising the interna (the rootlet system within the host) and the externa (the
external reproductive body). In the Lernaeodiscidae and Peltogastridae, the release of supplemen-
tary cement from the cement gland secures the kentrogon larva onto the host, and the stylet is
evaginated through its midventral surface. In contrast, in the Sacculinidae, the kentrogon remains
attached only by the cyprid antennules, and the stylet penetrates the host cuticle either via one
of the antennules or between the antennules (see Glenner 2001). Again, the host is inoculated
with the vermigon larva via the stylet. As do females, male cyprids attach using their antennules,
but their postsettlement metamorphosis produces a motile larva, the trichogon, which escapes
from the cyprid via one of its antennules at the distal break zone (Høeg 1987) and enters the female
externa via its mantle aperture.
Penetration has been observed in only three akentrogonid rhizocephalans. The “female”
cyprid of Clistosaccus uses its long antennules to permanently attach to the host (Høeg 1990),
and then one antennule penetrates the host cuticle. The method of penetration is unknown. The
distal region of the penetrating antennule breaks off, allowing the embryonic cells within the
cyprid to pass into the host and commence interna formation. The male cyprid uses its anten-
nules in a similar way to inject male generative cells either into the host adjacent to a developing
female interna or directly into a newly emerged externa (Høeg 1985).
Antennules and Antennae in the Crustacea 223

Fig. 7.15.
Scanning electron micrograph of Argulus (Branchiura) showing base of antennule (a1) with hooked claw
(hc) and spinous processes (sp), and base of antenna (a2) with spinous process (sp). Both limbs have slen-
der, cylindrical distal segments carrying sensory elements.

Attachment to Host

Many crustaceans are parasitic, and several others, in addition to the rhizocephalans, attach to
the host using modified antennules. In the Ascothoracida, for example, the antennule is prehen-
sile and used for attachment to the invertebrate host (Grygier 1984). In the majority of parasitic
copepods, it is the antennae that form the grasping mechanism by which attachment to the host
surface is secured (Huys and Boxshall 1991). They are typically subchelate structures, with a
robust protopodal part housing the main musculature, and a clawlike distal subchela that grasps
the host. The exopod is typically lost.
In the branchiuran fish lice, the basal part of the antennule is robust, heavily sclerotized, and
armed with curved hooklike processes on the segmental margins (Fig. 7.15). In Argulus the first
segment is drawn out into a hooklike process medially, and the more elongate second segment
carries spinous processes anteriorly and posteriorly and terminates in a robust ventral hook.
Adduction by extrinsic muscles originating on the dorsal cephalic wall brings these hooks into
contact with the host surface (Gresty et al. 1993b). These hooks form part of the multilimb mech-
anism employed by these ectoparasites to attach to their fish hosts. In branchiurans, the coxa of
the five-segmented antenna is heavily sclerotized and carries a stout hooked process proximally.
However, since no adduction-abduction motions of the coxa are possible, this process probably
only assists in securing the attachment of the parasite by preventing it from being dislodged
(Gresty et al. 1993b).

Mating

Sexual dimorphism in antennule morphology reflects differing behavioral roles of the sexes in
mating, although the functional significance of structures is not always well understood. For
example, in the male of the mysid Mesopodopsis slabberi, the third segment of the antennulary
peduncle carries, in addition to the two flagella, a large setose lobe referred to as the appendix
224 Functional Morphology and Diversity

B A

ae

pf

af

pd3

pd2

pd1

Fig. 7.16.
(A) Antennule of male Leptostylis longimana showing three-segmented peduncle (pd1–pd3) bearing pri-
mary (pf) and accessory (af) flagella. The proximal annulus of the primary flagellum is flared at its base
and carries a dense array of aesthetascs; this flagellum also carries a pair of aesthetascs (ae) distally, as in
the female. (B) Isolated aesthetasc from the basal array.

masculine plus an accessory appendix, which is an elongate lobe with a single long apical seta
(Tattersall and Tattersall 1951). The function of these lobes is unknown.

Mate Location

In many groups of crustaceans, males actively locate sexually receptive females using chem-
osensory-mediated behavior based on the detection and localization of pheromone plumes
produced by the female (e.g., Doall et al. 1998, Yen et al. 1998, Johnson and Atema 2005).
Ref lecting this sexual difference in behavioral roles, males commonly carry an enhanced
array of aesthetascs compared with conspecific females (Hallberg et al. 1997). For example,
the male of the mysid Neomysis integer has additional specialized setae on the antennulary
peduncle (Guse 1983, Johansson and Hallberg 1992). In some male cumaceans, the proximal
annulus of the primary f lagellum carries a spectacular array of long aesthetascs (Fig. 7.16)
on a f lared basal swelling. In the male of Lophogaster typicus , the primary f lagellum carries
several hundred additional slender “male-specific sensilla” in regular arrays on each annulus
in the middle and distal parts of the f lagellum. Although considerably smaller, these male-
specific sensilla do not differ significantly from aesthetascs in fine structure. In decapods,
only aesthetascs, not the nonaesthetasc chemosensors, are involved in processing pherom-
ones (Johnson and Atema 2005).
In copepods, the array of chemosensors on the antennules of adult males is often enhanced
relative to that of the female (Boxshall and Huys 1998). Enlargement of aesthetascs in the male,
rather than higher numbers of aesthetascs, is found in some planktonic copepods such as
Antennules and Antennae in the Crustacea 225

Fig. 7.17.
Dorsal view of head of male Eubranchipus grubii (Anostraca) showing species-specific form of antenna that
is used for grasping the female during mating. Redrawn from Daday (1910).

Pontoeciella abyssicola. The female carries two aesthetascs; the male has a third located proxi-
mally, but the middle aesthetasc is grossly enlarged (1.12 mm long) in an animal with a body
length of only 1.10 mm. Outspread, this pair of aesthetascs spans more than 2.2 mm, providing
an enormous surface area and enhanced ability to scan the water passing across the antennules
for pheromonal signals.

Copulation

Male anostracans clasp onto females during mating using their antennae. In all anostracan fam-
ilies except the Streptocephalidae, the basal and apical parts of the male antennae flex against
each other to form a prehensile mechanism that enables them to grasp the female around the
genital segment (Linder 1941). The basal part of the antenna carries an array of spinous or digi-
tiform processes and smaller warts, and in some anostracans large elaborate frontal processes
form as outgrowths of the basal part (Fig. 7.17). This ornamentation and the frontal processes
may function as part of a species-specific mate recognition system (Belk 1984).
All neocopepodan orders primitively have geniculate antennules in adult males that
are used for grasping the female during precopulatory and postcopulatory mate guarding,
as well as during mating (Boxshall 1990). The specialized geniculation lies between ances-
tral segments XX and XXI, either side of which there is usually segmental fusion and setal
modifications.
Siphonostomatoid copepods exhibit precopulatory mate guarding behavior in which adult
males grasp developing females using modified antennae (Boxshall 1990). The dimorphism
in antennal form and structure reflects the different behavior patterns during mating. Sexual
dimorphism in antennule morphology can also reflect gender differences in feeding biology in
copepods with dwarf parasitic males or with nonfeeding adult males. These are reflected in the
number and size of sensors on the antennules (Boxshall et al. 1997, Boxshall and Huys 1998).

Offspring Brooding

Some arcturid isopods of the genera Astacilla and Arcturus provide extended parental care for
their developing young by carrying their first stage juveniles on their long antennae (Sars 1899).
Similar behavior has been reported for the caprellid amphipod Pseudoprotella phasma, in which
226 Functional Morphology and Diversity

the mother lifts the young out of the marsupium and places them on her antennae (and gnatho-
pods), where they remain for up to three weeks, until the third instar (Harrison 1940).

Feeding

As a major sensory interface, the antennules function to identify and locate potential food items
for many crustaceans. Chemosensors along the antennules allow the animal to detect odorant
molecules (often amino acids) emanating from their food, and mechanosensors allow the ani-
mal to detect the presence of particles nearby.
Decapods and stomatopods use their antennules to track the chemical signatures of their
food (Goldman and Patek 2002). The chemosensors are carried on the primary f lagellum
(aesthetascs) and on the accessory f lagellum (nonaesthetasc chemosensors), and both are
involved in detecting food odors (Keller et al. 2003). Antennulary f licking is a well-studied
behavior pattern (Schmitt and Ache 1979). The primary f lagellum is f licked through the
water, a process of sampling the surrounding medium that has been likened to sniffing. In
Panulirus argus , water f lows through the aesthetasc array during the rapid f lick downstroke
and is retained there during the slower return stroke (Koehl et al. 2001). Intermittent f licking
possibly enables the lobsters to take discrete samples of odor-containing f luid (Reidenbach
et al. 2008).
Antennae can be used mechanically in deposit and suspension feeding. The amphipod
Corophium, for example, is a deposit feeder that uses its long antennae to rake surface sediment
into the entrance of its burrow, from where it is drawn down to the setose gnathopods by a water
current generated by the beating pleopods. Some Caprella also use their antennae for scraping
food particles from the substrate. The mole crab Emerita is a swash rider, burrowing into the
sediment repeatedly as it follows the tide up and down the beach. It uses its large setose anten-
nae for suspension feeding on phytoplankton during each wave cycle. Podocerid amphipods are
also known to sieve seston from the water using their setose antennae.
The planktonic copepod Euchaeta is a specialist predator that detects the presence of prey
by means of a three-dimensional array of long setae on the antennules (Yen and Nicoll 1990).
It feeds only on live prey, typically other copepods, which it detects by responding to distur-
bance (shear) in the water column generated by the swimming motions of the prey. The setal
array develops through the copepodid stages and reaches the peak of development in the adult
female but is atrophied in the adult male, which is a nonfeeding stage with vestigial mouthparts
(Boxshall et al. 1997). Small-particle-feeding copepods beat their antennae in opposition to the
maxillipeds to create the flow fields that bring food particles within range, so they can be cap-
tured by the maxillae (Price et al. 1983).

Excretion

The antennal gland is the primary excretory gland in early crustacean larval stages, but in
the adults of most taxa the paired maxillary glands take over this role during development.
However, in the Mystacocarida, Eucarida, Amphipoda, Mysidacea, some Myodocopa, and miso-
phrioid copepods, the antennal glands are retained in the adults as the sole functional excretory
gland (Hessler and Elofsson 2007, Boxshall 1982). The gland is typically located within the head,
and only the excretory duct passes into the antenna. In amphipods, the exit pore of the anten-
nal gland is typically carried at the tip of a conical process located at the base of the antenna. In
Derocheilocaris typica, the antennal excretory gland is, uniquely, located entirely within the coxa
of the limb. It comprises only eight cells and has an excretory duct that opens via a pore on the
posterior surface of the coxa (Hessler and Elofsson 2007).
Antennules and Antennae in the Crustacea 227

a1

Fig. 7.18.
Lateral view of Bosmina longispina (Branchiopoda) showing elongate spinous form of antennules (a1).
Redrawn from Lilljeborg (1900).

Defense

Predator Deterrence

The antennules of some cladocerans take the form of curved spinous processes (Fig. 7.18) that,
together with carapace spines and crests, help to protect them against predators. Long features,
including long antennules, may be selected to reduce predation risk from grasping predators
such as copepods and Leptodora (Kerfoot 1978). Such long-featured morphs may be replaced by
short-featured morphs when predation pressure declines (Kerfoot and Peterson 1980, Tollrian
and Dodson 1999). In Bosmina coregoni gibbera, feature length seems to be determined by fit-
ness trade-offs between increased protection afforded by the better developed antennules
and spines and negative hydrodynamic effects of these structures on swimming performance
(Lord et al. 2006).
Migrating spiny lobsters (Panulirus argus) exhibit collective defensive behavior when
attacked by predatory fish. Under threat, they cluster into outward-facing, rosettelike groups
and defend themselves using their spinous antennae. This behavior is effective, with per cap-
ita mortality declining with increasing lobster group size. Panulirus also produces sound by
stridulation. The antenna has an area of ridged membrane, the stridulatory membrane, on a
medial process on the proximal antennal segment. When the antennae are raised, the stridula-
tory membrane moves over the toothed ridge on the adjacent margin of the carapace, generat-
ing a rasping sound or a slow rattle. Bouwma and Herrnkind (2009) showed that stridulation
in P. argus improved the chances of escape from attacking predators such as Octopus.

Detection of Incoming Predators

The array of setae on the eight distal (ancestral) segments of copepod antennules is highly
conservative (Boxshall and Huys 1998) throughout the copepodid phase of development.
This underlying conservatism of the distal array of sensors through ontogeny indicates
228 Functional Morphology and Diversity

a requirement for functional continuity. Electrophysiological studies by Yen et al. (1992)


demonstrated that these antennulary setae are mechanosensors and that ablation of the
distal tip of the antennule in Pleuromamma and Euchaeta deprives the copepod of its rapid
escape response to mechanical signals (Lenz and Yen 1993). The distal array of setae forms
the mechanosensory early warning system and is involved in detecting incoming predators.
In the water column, the copepod represents a potential prey item for many predators, and
the possession of a functional approaching-predator warning system is highly advantageous.
The setae of the distal array are already present in late nauplii of planktonic calanoids. This
mechanosensory system is operational very early in ontogeny, and functional continuity is
maintained throughout the copepodid phase.

Grooming

Antennular grooming behavior is a distinctive and stereotyped behavior in decapods in which


the antennulary flagella are repeatedly brought down to the third maxillipeds and drawn
through pads of densely packed specialized setae on the endopods (Bauer 1989). In Panulirus
argus, this behavior is elicited by stimulation with L-glutamate (Barbato and Daniel 1997), and
Schmidt and Derby (2005) concluded that it is mediated through the chemoreceptors within
the asymmetric setae on the flagella. The long antennal flagellum of most dendrobranchiate,
stenopodidean, and caridean shrimps is groomed by specialized brushes of setae located either
side of the carpus-propodus articulation of the first pereopod, whereas in most other decapods
antennal grooming is performed by the third maxillipeds (Bauer 1989).

Respiration

In the burrowing crab Corystes, the long antennae carry linear arrays of setae that interlock
when the antennae are closely apposed, thus forming a breathing tube, the tip of which extends
to the surface. Inhalant water is filtered and drawn down this tube into the branchial chamber
(Hartnoll 1972).

WIDER COMPARISONS

The variation in form and function of the antennules and antennae within the Crustacea is
remarkable. The other mandibulatans, hexapods, and myriapods, all lack antennae, but their
antennules can also perform a range of roles. With the exception of the Protura, which lack
them, the antennules of hexapods and myriapods function primarily as sense organs. The
flagellate antennules of insects (winged hexapods) typically carry an array of sensilla that can
serve as olfactory receptors, mechanoreceptors, or thermohygroreceptors (Zacharuk 1985).
As in the Malacostraca, the modular construction of the antennule provides a basic flexibility
since the numbers of modules expressed during development can be adapted to suit changing
requirements. Very long antennules, as found in cockroaches, for example, may have a tactile
role analogous to that of the long antennal flagellum in Cherax.
Even in insects, the antennules can perform additional functions in mating, respiration,
or feeding behavior. The terminal annuli of the flagellum in the water beetle Hydrophilus are
covered in hydrophobic hairs and are involved in the formation of a funnel that carries air
down to the layer of trapped air on the ventral surface. Male fleas grasp their mates using their
antennules, which are provided with stalked discs, and many male collembolans exhibit simi-
lar clasping behavior during mating. The plasticity of form, as it reflects different functional
Antennules and Antennae in the Crustacea 229

priorities, is apparent in all arthropod limbs, but the sheer structural diversity of the antennules
and antennae of the Crustacea is more spectacular than in any other arthropod group.

ACKNOWLEDGMENTS

We thank the editors for their constructive suggestions and improvements to the manuscript.
Paul Clark pointed us to important references on decapod morphology. We are grateful to Jens
Høeg and Roger Lincoln for providing photographic illustrations.

REFERENCES

Akam, M. 2000. Arthropods: Developmental diversity within a (super) phylum. Proceedings of the
National Academy of Sciences of the USA 97:4438 –4441.
Anderson, P.A., and B.W. Ache. 1985 . Voltage- and current-clamp recordings of the receptor potential in
olfactory receptor cells in situ. Brain Research 338:273–280.
Andersson, A. 1975 . The ultrastructure of the presumed chemoreceptor aesthetasc “Y” of a cypridid
ostracode. Zoologica Scripta 4:151–158.
Arnott, S.A., D.M. Neil, and A.D. Ansell. 1998. Tail-flip mechanism and size-dependent kinemat-
ics of escape swimming in the brown shrimp Crangon crangon. Journal of Experimental Biology
201:1771–1784.
Atema, J. 1977. Functional separation of smell and taste in fish and Crustacea. Pages 165–174 in
J. Le Magnen and P. MacLeod, editors. Olfaction and taste, Vol. 6. Information Retrieval,
London .
Bacescu, M. 1967. Decapoda. Fauna Republicii Socialiste Romania (Crustacea), Vol. 4, Fasicle 9. Editura
Academiei R.S.R., Bucharest .
Ball, E.E., and A.N. Cowan. 1977. Ultrastructure of the antennal sensilla of Acetes (Crustacea, Decapoda,
Natantia, Sergestidae). Philosophical Transactions of the Royal Society of London Series B
277:429 –457.
Barbato, J.C., and P.C. Daniel. 1997. Chemosensory activation of an antennular grooming behav-
ior in the spiny lobster, Panulirus argus, is tuned narrowly to L-glutamate. Biological Bulletin
293:107–115.
Barnes, T.G., T.Q. Truong , G.G. Adams, and N.E. McGruer. 2001. Large deflection analysis of a biomi-
metic lobster antenna due to contact and flow. ASME Journal of Applied Mechanics 68:948 –951.
Bauer, R.T. 1989. Decapod crustacean grooming: Functional morphology, adaptive value and phy-
logenetic significance. Pages 49–73 in B.E. Felgenhauer , L. Watling , and A.B. Thistle, editors.
Functional morphology of feeding and grooming in Crustacea. Balkema, Rotterdam.
Belk , D. 1984 . Antennal appendages and reproductive success in the Anostraca. Journal of Crustacean
Biology 4:66 –71.
Bielecki, J., B.K.K. Chan, J.T. Høeg , and A. Sari. 2009. Antennular sensory organs in cyprids of bal-
anomorph cirripedes: Standardizing terminology using Megabalanus rosea. Biofouling 25:203–214.
Bjørnberg , T.S. 1986. The rejected nauplius: A commentary. Syllogeus 58:232–236.
Bouwma, P.E., and W.F. Herrnkind. 2009. Sound production in Caribbean spiny lobster Panulirus argus
and its role in escape during predatory attack by Octopus briareus. New Zealand Journal of Marine
and Freshwater Research 43:1–13.
Boxshall, G.A. 1982. On the anatomy of the misophrioid copepods, with special reference to
Benthomisophria palliata Sars. Philosophical Transactions of the Royal Society of London Series B
297:125–181.
Boxshall, G.A. 1985 . The comparative anatomy of two copepods, a predatory calanoid and a particle feed-
ing mormonilloid. Philosophical Transactions of the Royal Society of London Series B 311:303–377.
Boxshall, G.A. 1990. Precopulatory mate guarding in copepods. Bijdragen tot de Dierkunde 60:209 –213.
Boxshall, G.A. 2004 . The evolution of arthropod limbs. Biological Reviews 79:253–300.
230 Functional Morphology and Diversity

Boxshall, G.A. 2007. Crustacean classification: On-going controversies and unresolved problems.
Zootaxa 1668:313–325.
Boxshall, G.A., and R. Böttger-Schnack. 1988. Unusual ascothoracid nauplii from the Red Sea. Bulletin of
the British Museum (Natural History), Zoology Series 54:275–283.
Boxshall, G.A., D.L. Danielopol, D.J. Horne, R.J. Smith, and I. Tabacaru. 2010. A critique of biramous
interpretations of the crustacean antennule. Crustaceana 83:153–167.
Boxshall, G.A., T.D. Evstigneeva, and P.F. Clark. 1993. A new interstitial cyclopoid copepod from a sandy
beach on the western shore of Lake Baikal, Siberia. Hydrobiologia 268:99 –107.
Boxshall, G.A., and R. Huys, 1998. The ontogeny and phylogeny of copepod antennules. Philosophical
Transactions of the Royal Society of London Series B 353:765–786.
Boxshall, G.A., and D. Jaume. 2000. Discoveries of cave misophrioids (Crustacea: Copepoda) shed new
light on the origin of anchialine faunas. Zoologischer Anzeiger 239:1–19.
Boxshall, G.A., and R.J. Lincoln. 1987. The life cycle of the Tantulocarida (Crustacea). Philosophical
Transactions of the Royal Society of London Series B 315:267–303.
Boxshall, G.A., J. Yen, and J.R. Strickler. 1997. Functional significance of the sexual dimorphism in the
cephalic appendages of Euchaeta rimana Bradford. Bulletin of Marine Science 61:387–398.
Briggs, D.E.G., M.D. Sutton, D.J. Siveter, and D.J. Siveter. 2003. A new phyllocarid (Crustacea:
Malacostraca) from the Silurian Fossil-Lagerstätte of Herefordshire, UK. Proceedings of the Royal
Society of London Series B 271:131–138.
Browman, H.I., S. Kruse, and W.J. O’Brien. 1989. Foraging behavior of the predaceous cladoceran,
Leptodora kindti, and escape responses of their prey. Journal of Plankton Research 11:1075–1088.
Bruce, N. 1986. Cirolanidae (Crustacea: Isopoda) of Australia. Records of the Australian Museum,
Supplement 6:1–239.
Brusca, R.C., and G.D.F. Wilson. 1991. A phylogenetic analysis of the Isopoda with some classificatory
recommendations. Memoirs of the Queensland Museum 31:143–204.
Bruton, D.L., and H.B. Whittington. 1983. Emeraldella and Leanchoilia, two arthropods from the Burgess
Shale, Middle Cambrian, British Columbia. Proceedings of the Royal Society of London Series B
300:553–585.
Cate, H.S., and C.D. Derby. 2002. Ultrastructure and physiology of the hooded sensillum, a bimodal
chemo-mechanosensillum of lobsters. Journal of Comparative Neurology 442:293–307.
Cate, H.S., and D.B. Roye. 1997. Ultrastructure and physiology of the outer row statolith sensilla of the
blue crab Callinectes sapidus. Journal of Crustacean Biology 17:398 –411.
Clark , P.F., D.K. Calazans, and G.W. Pohle. 1998. Accuracy and standardisation of brachyuran larval
descriptions. Invertebrate Reproduction and Development 33:127–144.
Crisp, D.J., and P.S. Meadows. 1962. The chemical basis of gregariousness in cirripedes. Philosophical
Transactions of the Royal Society of London Series B 156:500 –520.
Daday, E. 1910. Monographie systématique des Phyllopodes Anostracés. Annales des Sciences Naturelle,
Zoologie, Series 9, 9:91–489.
Dahl, E., H. Emanuelsson, and C.V. Mecklenburg. 1970. Pheromone reception in the males of the amphi-
pod Gammarus duebeni Lilljeborg. Oikos 21:42–47.
Dahms, H.-U. 1992. Metamorphosis between naupliar and copepodid phases in the Harpacticoida.
Philosophical Transactions of the Royal Society of London Series B 335:221–236.
Damen, W.G.M., M. Hausdorf, E.A. Seyfarth, and D. Tautz. 1998. A conserved mode of head segmenta-
tion in arthropods revealed by the expression patterns of Hox genes in a spider. Proceedings of the
National Academy of Sciences of the USA 95:10665–10670.
Daniel, P.C., M. Fox , and S. Mehta. 2008. Identification of chemosensory sensilla mediating antennular
flicking behavior in Panulirus argus, the Caribbean spiny lobster. Biological Bulletin 215:24–33.
De Calazans, D.K. 1992. Taxonomy, distribution and abundance of protozoea, mysis and megalopa stages
of Penaeidean decapods from the southern Brazilian coast. PhD dissertation, University of London.
Denton, E., and J.A.B. Gray. 1986. Lateral-line-like antennae of certain of the Penaeidea (Crustacea,
Decapoda, Natantia). Proceedings of the Royal Society of London Series B 226:249 –261.
Derby, C.D. 1982. Structure and function of cuticular sensilla of the lobster Homarus americanus. Journal
of Crustacean Biology 2:1–21.
Antennules and Antennae in the Crustacea 231

Derby, C.D., H.S. Cate, and L.R. Gentilcore. 1997. Perireception in olfaction: Molecular mass sieving
by aesthetasc sensillar cuticle determines odorant access to receptor sites in the Caribbean spiny
lobster Panulirus argus. Journal of Experimental Biology 200:2073–2081.
Diederich, R.J., V.K.L. Merrill, M.A. Pultz , and T.C. Kaufman. 1989. Isolation, structure and expression
of labial, a homeotic gene of the Antennapedia complex involved in Drosophila head. Genes and
Development 3:399 –414.
Doall, M.H., S.P. Colin, J.R. Strickler, and J. Yen. 1998. Locating a mate in 3D: The case of Temora longi-
cornis. Philosophical Transactions of the Royal Society of London Series B 353:681–689.
Elofsson, R., and R.R. Hessler. 1991. Sensory morphology in the antennae of the cephalocarid
Hutchinsoniella macracantha. Journal of Crustacean Biology 11:345–355.
Foxton, P. 1969. The morphology of the antennal flagellum of certain of the Peneidea (Decapoda,
Natantia). Crustaceana 16:33–43.
Ghiradella, H.T., J. Case, and J. Cronshaw. 1968a . Fine structure of the aesthetasc hairs of Pagurus hirsu-
tiusculus Dana. Protoplasma 66:1–20.
Ghiradella, H.T., J. Case, and J. Cronshaw. 1968b. Structure of aesthetascs in selected marine and terres-
trial decapods: Chemoreceptor morphology and environment. American Zoologist 8:603–621.
Giesbrecht, W. 1893. Systematik und Faunistik der pelagischen Copepoden des Golfes von Neapel und
der angrenzenden Meeres-Abschnitte. Fauna und Flora des Golfes von Neapel 19:1–831 (text), pls.
1–54 (atlas).
Glenner, H. 2001. Cypris metamorphosis, infection and earliest internal development of the rhizocepha-
lan Loxothylacus panopaei (Gissler). Crustacea: Cirripedia: Rhizocephala: Sacculinidae. Journal of
Morphology 249:43–75.
Goldman, J.A., and S.N. Patek. 2002. Two sniffing strategies in panulirid lobsters. Journal of
Experimental Biology 205:3891–3902.
Gresty, K.A., G.A. Boxshall, and K. Nagasawa. 1993a . Antennulary sensors of the infective copepodid
larva of sea lice (Copepoda: Caligidae). Pages 83–98 in G.A. Boxshall and D. Defaye, editors.
Pathogens of wild and farmed fish: Sea lice. Ellis Horwood, Chichester, UK .
Gresty, K.A., G.A. Boxshall, and K. Nagasawa. 1993b. The fine structure and function of the cephalic
appendages of the branchiuran parasite, Argulus japonicus Thiele. Philosophical Transactions of the
Royal Society of London Series B 339:119 –135.
Gruner, H.-E., and G. Scholtz , 2004 . Segmentation, tagmata, and appendages. Pages 13–57 in J. Forest,
J.C. von Vaupel Klein and F.R. Schram, editors. The Crustacea. Brill, Leiden, The Netherlands.
Gr ü nert, U., and B.W. Ache. 1988. Ultrastructure of the aesthetasc (olfactory) sensilla of the spiny lobster,
Panulirus argus. Cell and Tissue Research 251:95–103.
Grygier, M.J. 1984 . Comparative morphology and ontogeny of the Ascothoracida, a step towards a phyl-
ogeny of the Maxillopoda. PhD dissertation, University of California, San Diego.
Grygier, M.J., and S. Ohtsuka. 1995 . SEM observation of the nauplius of Monstrilla hamatapex, new
species, from Japan and an example of upgraded descriptive standards for monstrilloid copepods.
Journal of Crustacean Biology 15:703–719.
Guse, G.-W. 1978. Antennal sensilla of Neomysis integer (Leach). Protoplasma 95:145–161.
Guse, G.-W. 1983. Ultrastructure, development and moulting of the aesthetascs of Neomysis integer and
Idotea baltica (Crustacea, Malacostraca). Zoomorphology 103:121–133.
Hallberg , E., K.U.I. Johansson, and R. Elofsson. 1992. The aesthetasc concept: Structural variations of puta-
tive olfactory receptor cell complexes in Crustacea. Microscopy Research and Technique 22:325–335.
Hallberg , E., K.U.I. Johansson, and R. Wallén . 1997. Olfactory sensilla in crustaceans: Morphology,
sexual dimorphism, and distribution patterns. International Journal of Insect Morphology and
Embryology 26:173–180.
Hallberg , E., and M. Skog. 2011. Chemosensory sensilla in crustaceans. Pages 103–121 in T. Breithaupt and
M. Thiel, editors. Chemical communication in crustaceans. Springer, New York.
Harrison, R.J. 1940. On the biology of the Caprellidae. Growth and moulting in Pseudoprotella phasma
Montagu. Journal of the Marine Biological Association of the UK 24:483–493.
Hartnoll, R.G. 1972. The biology of the burrowing crab, Corystes cassivelaunus. Bijdragen tot de
Dierkunde 42:139 –155.
232 Functional Morphology and Diversity

Heimann, P. 1979. Fine structure of sensory tubes on the antennule of Conchoecia spinirostris (Ostracoda,
Crustacea). Cell and Tissue Research 202:461–477.
Heimann, P. 1984 . Fine structure and molting of aesthetasc sense organs on the antennules of the isopod,
Asellus aquaticus (Crustacea). Cell and Tissue Research 235:117–128.
Hessler, R.R. 1964 . The Cephalocarida. Comparative skeletomusculature. Memoirs of the Connecticut
Academy of Arts and Sciences 16:1–97.
Hessler, R.R. 1983. A defense of the caridoid facies, wherein the early evolution of the Eumalacostraca.
Pages 145–164 in F.R. Schram, editor. Crustacean phylogeny. Balkema, Rotterdam.
Hessler, R.R., and R. Elofsson. 2007. An excretory organ in the Mystacocarida (Crustacea). Arthropod
Structure and Development 36:171–181.
Hessler, R.R., and H.L. Sanders. 1966. Derocheilocaris typicus Pennak and Zinn (Mystacocarida) revis-
ited. Crustaceana 11:141–155.
Heymons, R. 1901. Die Entwicklungsgeschichte der Scolopender. Zoologica 33:1–33, pls. I–VIII.
Høeg , J.T. 1985 . Male cypris settlement in Clistosaccus paguri Lilljeborg (Crustacea: Cirripedia:
Rhizocephala). Journal of Experimental Marine Biology and Ecology 89:221–235.
Høeg , J.T. 1987. Male cypris metamorphosis and a new male larval form, the trichogon, in the parasitic
barnacle Sacculina carcini (Crustacea: Cirripedia: Rhizocephala). Philosophical Transactions of the
Royal Society of London Series B 317: 47–63.
Høeg , J.T. 1990. “Akentrogonid” host invasion and an entirely new life cycle in the rhizocephalan parasite
Clistosaccus paguri (Thecostraca: Cirripedia). Journal of Crustacean Biology 10:37–52.
Hou, X., and J. Bergström. 1997. Arthropods of the Lower Cambrian Chengjiang fauna, southwest China.
Fossils and Strata 45:1–116.
Hunt, G., L.E. Park , and M. Labarbera. 2007. A novel crustacean swimming stroke: coordinated
four-paddled locomotion in the Cypridoidean ostracode Cypridopsis vidua (Mü ller). Biological
Bulletin 212:67–73.
Huys, R., and G.A. Boxshall. 1991. Copepod evolution. Ray Society, London .
Huys, R., G.A. Boxshall, and R.J. Lincoln. 1993. The tantulocaridan life cycle: The circle closed? Journal
of Crustacean Biology 13:432–442.
Imms, A.D. 1939. On the antennal musculature in insects and other arthropods. Quarterly Journal of the
Microscopical Society 81:273–320.
Jaume, D., G.A. Boxshall, and R.N. Bamber. 2006. A new genus of Hirsutiidae from the continental slope
off Brazil and the discovery of the first males in the Bochusacea (Crustacea: Peracarida). Zoological
Journal of the Linnean Society 148:169 –208.
Jaume, D., and F. Bréhier. 2005 . A new species of Typhlatya (Crustacea: Decapoda: Artyidae) from
anchialine caves on the French Mediterranean coast. Zoological Journal of the Linnean Society
144:387–414.
Jaume, D., J.E. Cartes, and J.-C. Sorbe. 1998. A new species of Bathymedon Sars, 1892 (Amphipoda:
Oedicerotidae) from the western Mediterranean bathyal floor. Scientia Marina 62:341–356.
Jaume, D., and K. Christenson. 2001. Amphi-Atlantic distribution of the subterranean amphipod family
Metacrangonyctidae (Gammaridea). Contributions to Zoology 70:99 –125.
Jenner, R. A., C. Ní Dhubhghaill, M.P. Ferla, and M.A. Wills. 2009. Eumalacostracan phylogeny and total
evidence: Limitations of the usual suspects. BMC Evolutionary Biology 9:21.
Johansson, K.U.I., L. Gefors, R. Wallén, and E. Hallberg. 1996. Structure and distribution patterns of
aesthetascs and male-specific sensilla in Lophogaster typicus (Mysidacea). Journal of Crustacean
Biology 16:45–53.
Johansson, K.U.I., and E. Hallberg. 1992. Male-specific structures in the olfactory system of mysids
(Mysidacea: Crustacea). Cell and Tissue Research 268:369 –368.
Johnson, M.E., and J. Atema. 2005 . The olfactory pathway for individual recognition in the American
lobster Homarus americanus. Journal of Experimental Biology 208:2865–2872.
Ka ï m-Malka, R.A., S. Maebe, C. Macquart-Moulin, and C. Bezac. 1999. Antennal sense organs of
Natatolana borealis (Lilljeborg, 1851) (Crustacea: Isopoda). Journal of Natural History 33:65–88.
Kaji, T., and A. Tsukagoshi. 2008. Origin of the novel chemoreceptor Aesthetasc “Y” in Ostracoda:
Morphological thresholds and evolutionary innovation. Evolution and Development 10:228 –240.
Antennules and Antennae in the Crustacea 233

Karanovic , I. 2005 . Comparative morphology of the Candoninae antennula, with remarks on the ances-
tral state in ostracods and a proposed new terminology. Spixiana 28:141–160.
Keller, T.A., I. Powell, and M.J. Weissburg. 2003. Role of olfactory appendages in chemically mediated
orientation of blue crabs. Marine Ecology Progress Series 261:217–231.
Kerfoot, W.C. 1978. Combat between predatory copepods and their prey: Cyclops, Epischura and Bosmina.
Limnology and Oceanography 23:1089 –1102.
Kerfoot, W.C., and C. Peterson. 1980. Predatory copepods and Bosmina: Replacement cycles and further
influences of predation upon prey reproduction. Ecology 61:417–431.
Kesling , R.V. 1951. The morphology of ostracod molt stages. III. Biological Monographs 21:1–324.
Koehl, M.A.R., J.R. Koseff, J.P. Crimaldi, M.G. McCay, T. Cooper, M.B. Wiley, and P.A. Moore. 2001.
Lobster sniffing: Antennule design and hydrodynamic filtering of information in an odor plume.
Science 294:1948 –1951.
Koenemann, S., F.R. Schram, A. Bloechl, T.M. Iliffe, M. Hoenemann, and C. Held, 2007. Post-embryonic
development of remipede crustaceans. Evolution and Development 9:117–121.
Kunze, J. K., 1983. Stomatopoda and the evolution of the Hoplocarida. Pages 165–188 in F.R. Schram, edi-
tor. Crustacean phylogeny. Crustacean Issues, Vol. 1. Balkema, Rotterdam.
Lagersson, N.C., A. Garm, and J.T. Høeg. 2003. Notes on the ultrastructure of the setae on the fourth
antennulary segment of the Balanus amphitrite cyprid (Crustacea: Cirripedia: Thoracica). Journal
of the Marine Biological Association of the UK 83:361–365.
Laverack , M.S. 1964 . The antennulary sense organs of Panulirus argus. Comparative Biochemistry and
Physiology 13:301–321.
Laverack , M., and M.Q. Hull. 1993. Sensory innervation of the antennule of the preadult male Caligus
elongatus . Pages 114–122 in G.A. Boxshall and D. Defaye, editors. Pathogens of wild and farmed fish:
Sea lice. Ellis Horwood, Chichester, UK .
Lenz , P.H., and J. Yen. 1993. Distal setal mechanoreceptors of the first antennae of marine copepods.
Bulletin of Marine Science 53:170–179.
Lilljeborg , W. 1900. Cladocera Sueciae oder Beträge zur Kenntniss der in Schweden lebenden
Krebsthiere von der Ordnung der Branchiopoden und der Unterordnung der Cladoceren.
Akademischen Buchdruckerei, Upsala.
Lincoln, R.J., and D.E. Hurley. 1981. The calceolus, a sensory structure of gammaridean amphipods
(Amphipoda: Gammaridea). Bulletin of the British Museum (Natural History), Zoology Series
40:103–116.
Linder, F. 1941. Contributions to the morphology and the taxonomy of the Branchiopoda Anostraca.
Zoologiska Bidrag frå n Uppsala 20:101–302.
Lochhead, J.H. 1968. The feeding and swimming of Conchoecia (Crustacea, Ostracoda). Biological
Bulletin 134:456 –464.
Lord, H., R. Lagergren, J.-E. Svensson, and N. Lindqvist. 2006. Sexual dimorphism in Bosmina: The role
of morphology, drag, and swimming. Ecology 87:788 –795.
Lowry, J.K. 1986. The callynophore, a eucaridan/peracaridan sensory organ prevalent among the
Amphipoda (Crustacea). Zoologica Scripta 15:333–349.
Maas, A., D. Waloszek , and K.J. Mü ller. 2003. Morphology, ontogeny and phylogeny of the
Phosphatocopina (Crustacea) from the Upper Cambrian “Orsten” of Sweden. Fossils and Strata
49:1–238.
Maddox , R.F. 2000. The antennule in podocopid Ostracoda: Chaetotaxy, ontogeny and morphometrics.
Micropaleontology 4(suppl 2):1–72.
Martin, J. 1992. Branchiopoda. Pages 25–224 in F.W. Harrison, editor. Microscopic anatomy of inverte-
brates, Vol. 9, Crustacea. John Wiley and Sons, New York.
Maruzzo, D., A. Minelli, M. Ronco, and G. Fusco. 2007. Growth and regeneration of the second anten-
nae of Asellus aquaticus (Isopoda) in the context of arthropod antennal segmentation. Journal of
Crustacean Biology 27:184–196.
McKenzie, K.G., P.-J. Chen, and S. Majoran. 1991. Almatium gusevi (Chernyshev 1940): Redescription,
shield-shapes, and speculations on the reproductive mode (Branchiopoda, Kazacharthra).
Paläontologische Zeitschrift 65:305–317.
234 Functional Morphology and Diversity

Mead, K.S. 2002. From odor molecules to plume tracking: An interdisciplinary, multilevel approach to
olfaction in stomatopods. Integrative and Comparative Biology 42:258 –264.
Mead, K.S. 2008. Do antennule and aesthetasc structure in the crayfish Orconectes virilis correlate with
flow habitat? Integrative and Comparative Biology 48:823–833.
Mead, K.S., and M.A.R. Koehl. 2000. Stomatopod antennule design: The asymmetry, sampling effi-
ciency and ontogeny of olfactory flicking. Journal of Experimental Biology 203:3795–3808.
Mead, K.S., and T.M. Weatherby. 2002. Morphology of stomatopod chemosensory sensilla facilitates
fluid sampling. Invertebrate Zoology 121:148 –157.
Michel, W.C., H.G. Trapido-Rosenthal, E.T. Chao, and M. Wachowiak. 1993. Stereoselective detec-
tion of amino acids by lobster olfactory receptor neurons. Journal of Comparative Physiology A
171:705–712.
Nott, J.A., and B.A. Foster. 1969. On the structure of the antennular attachment organ of the cypris larva
of Balanus balanoides (L.). Philosophical Transactions of the Royal Society of London Series B
256:115–134.
Olesen, J. 2001. External morphology and larval development of Derocheilocaris remanaei
Delamare-Deboutteville and Chappuis, 1951 (Crustacea, Mystacocarida), with a comparison of
crustacean segmentation and tagmosis patterns. Biologiske Skrifter 53:1–59.
Olesen, J., and D. Waloßek. 2000. Limb ontogeny and trunk segmentation in Nebalia species (Crustacea,
Malacostraca, Leptostraca). Zoomorphology 120:47–64.
Perrier, V., J. Vannier, P.R. Racheboeuf, S. Charbonnier, D. Chabard, and D. Sotty. 2006. Syncarid crusta-
ceans from the Montceau Lagerstätte (Upper Carboniferous; France). Palaeontology 49:647–672.
Peterson, M., B.T. Rogers, A. Popadić , and T.C. Kaufman. 1999. The embryonic expression pattern
of labial, posterior homeotic complex genes and the teashirt homologue in an apterygote insect.
Development, Genes and Evolution 209:77–90.
Phang , I.Y., N. Aldred, A.S. Clare, and G.J. Vancso. 2007. Towards a nanomechanical basis for temporary
adhesion in barnacle cyprids (Semibalanus balanoides). Interface 5:397–402.
Prentiss, C.W. 1901. The otocyst of decapod Crustacea. Bulletin of the Museum of Comparative Zoology
36:167–251.
Price, H.J., G.-A. Paffenhöfer, and J.R. Strickler. 1983. Modes of cell capture in calanoid copepods.
Limnology and Oceanography 28:116 –123.
Rainbow, P.S., and G. Walker. 1976. The feeding apparatus of the barnacle nauplius larva: A scanning
electron microscope study. Journal of the Marine Biological Association of the UK 56:321–326.
Reidenbach, M.A., N. George, and M.A.R. Koehl. 2008. Antennule morphology and flicking kinemat-
ics facilitate odor sampling by the spiny lobster, Panulirus argus. Journal of Experimental Biology
211:2849 –2858.
Richter, S., and G. Scholtz. 2001. Phylogenetic analysis of the Malacostraca (Crustacea). Journal of
Zoological Systematics and Evolutionary Research 39:113–136.
Rossetti, G., and K. Martens. 1996. Redescription and morphological variability of Darwinula stevensoni
(Brady and Robertson, 1870) (Crustacea, Ostracoda). Bulletin de l’Institute Royal des Sciences
Naturelles de Belgique, Biologie 66:73–92.
Rushton-Mellor, S., and G.A. Boxshall. 1994 . The developmental sequence of Argulus foliaceus. Journal of
Natural History 28:763–785.
Sandeman, D.C. 1989. Physical properties sensory receptors and tactile reflexes of the antenna of the
Australian freshwater crayfish Cherax destructor. Journal of Experimental Biology 141:197–218.
Sandeman, R.E., and D.C. Sandeman. 1996. Pre- and postembryonic development, growth and
turnover of olfactory receptor neurones in crayfish antennules. Journal of Experimental Biology
199:2409 –2418.
Sanders, H.L. 1963. The Cephalocarida. Functional morphology, larval development, comparative exter-
nal anatomy. Memoirs of the Connecticut Academy of Arts and Sciences 15:1–80.
Sars, G.O. 1899. Isopoda. An account of the Crustacea of Norway. Bergen Museum 2:1–270.
Schmidt, M., and B.W. Ache. 1996. Processing of antennular input in the brain of the spiny lobster,
Panulirus argus. I. Non-olfactory chemosensory and mechanosensory pathway of the lateral and
median antennular neuropils. Journal of Comparative Physiology A 178:579 –604.
Antennules and Antennae in the Crustacea 235

Schmidt, M., and C.D. Derby. 2005 . Non-olfactory chemoreceptors in asymmetric setae activate anten-
nulary grooming behaviour in the Caribbean spiny lobster Panulirus argus. Journal of Experimental
Biology 208:233–248.
Schmidt, M., and W. Gnatzy. 1984 . Are the funnel-canal organs the campaniform sensilla of the shore
crab Carcinus maenas (Crustacea, Decapoda). 2. Ultrastructure. Cell and Tissue Research
237:81–93.
Schmitt, B.C., and B.W. Ache. 1979. Olfaction: Responses of a decapod crustacean are enhanced by flick-
ing. Science 205:204–206.
Scholtz , G. 2000. Evolution of the nauplius stage in malacostracan crustaceans. Journal of Zoological
Systematics and Evolutionary Research 38:175–187.
Scholtz , G., and G.D. Edgecombe. 2005 . Heads, Hox and the phylogenetic position of trilobites. Pages
139–165 in S. Koenemann and R.A. Jenner, editors. Crustacea and arthropod relationships.
Crustacean Issues, Vol. 16. Taylor and Francis, Boca Raton, FL.
Schram, F.R. 1986. Crustacea. Oxford University Press, New York.
Schram, F.R., and C. Hof. 1998. Fossils and the interrelationships of major crustacean groups. Pages
233–302 in G.D. Edgecombe, editor. Arthropod fossils and phylogeny. Columbia University Press,
New York.
Schutze, M.L.M., C.E.F. da Rocha, and G.A. Boxshall. 2000. Antennular development during the cope-
podid phase in the family Cyclopidae (Copepoda, Cyclopoida). Zoosystema 22:749 –806.
Scourfield, D.J. 1926. On a new type of crustacean from the Old Red Sandstone (Rhynie Chert Bed,
Aberdeenshire)—Lepidocaris rhyniensis, gen, et sp. nov. Philosophical Transactions of the Royal
Society of London Series B 214:153–186.
Scourfield, D.J. 1940. Two new complete specimens of young stages of the Devonian fossil crustacean
Lepidocaris rhyniensis. Proceedings of the Linnean Society 152:290 –298.
Shakori, A.R. 1968. Morphology and skeletomusculature of Caenestheria propinqua (Sars)
(Conchostraca; Branchiopoda; Crustacea). Bulletin of the Department of Zoology University of
the Panjab 2:1–48.
Siewing , R. 1956. Untersuchungen zur Morphologie der Malacostraca. Zoologische Jahrbücher,
Abteilung f ü r Anatomie 75:39 –176.
Siewing , R. 1959. Syncarida. Pages 1–121 in H.-E. Gruner, editor. Dr. H. G. Bronns Klassen und
Ordnungen des Tierreiches, Vol. 5, Band 1. Akademische-Verlagsgesellschaft Geest, Leipzig.
Siveter, David J., Derek J. Siveter, M.D. Sutton, and D.E.G. Briggs. 2007. Brood care in a Silurian ostra-
cod. Proceedings of the Royal Society of London Series B 274:465–469.
Siveter, David J., M.D. Sutton, D.E.G. Briggs, and Derek J. Siveter. 2003. An ostracode crustacean with
soft parts from the Lower Silurian. Science 302:1749 –1751.
Siveter, David J., M. Williams, and D. Waloszek. 2001. A phosphatocopid Crustacean with appendages
from the Lower Cambrian. Science 293:479 –481.
Smith, R.J., and T. Kamiya. 2008. The ontogeny of two species of Darwinuloidea (Ostracoda, Crustacea).
Zoologischer Anzeiger 247:275–302.
Smith, R.J., and R. Matzke-Karasz. 2008. The organ on the first segment of the cypridoidean (Ostracoda,
Crustacea) antennule: Morphology and phylogenetic significance. Senckenbergiana Lethaea
88:127–140.
Smith, R.J., and A. Tsukagoshi. 2005 . The chaetotaxy, ontogeny and musculature of the antennule of
podocopan ostracods (Crustacea). Journal of Zoology 265:157–177.
Stensmyr, M.C., S. Erland, E. Hallberg, R. Wallén, P. Greenaway, and B.S. Hansson. 2005 . Insect-like
olfactory adaptations in the terrestrial giant robber crab. Current Biology 15:116 –121.
Steullet, P., H.S. Cate, W.C. Michel, and C.D. Derby. 2000. Functional units of a compound nose:
Aesthetasc sensilla house similar populations of olfactory receptor neurons on the crustacean
antennule. Journal of Comparative Neurology 418:270–280.
Tattersall, W.M., and O.S. Tattersall. 1951. The British Mysidacea. Ray Society, London .
Telford, M.J., and R.H. Thomas. 1998. Expression of homeobox genes shows chelicerate arthropods
retain their deutocerebral segment. Proceedings of the National Academy of Sciences of the USA
95:10671–10675.
236 Functional Morphology and Diversity

Thomson, M., K. Robertson, and A. Pile. 2009. Microscopic structure of the antennulae and antennae on
the deep-sea isopod Bathynomus pelor. Journal of Crustacean Biology 29:302–316.
Tollrian, R., and S.I. Dodson. 1999. Inducible defenses in Cladocera: Constraints, costs, and multipreda-
tor environments. Pages 177–202 in R. Tollrian and C.D. Harvell, editors. The ecology and evolu-
tion of inducible defenses. Princeton University Press, Princeton, NJ.
Tyson, G.E. 1980. Fine structure of the type 2 antennular sensilla of the brine shrimp, Artemia salina.
American Zoologist 20:816 –000.
Vandel, A. 1960. Isopodes terrestres (Première partie). Faune de France 64:1–416.
Van der Ham, J.L., and B.E. Felgenhauer. 2006. Ultrastructure of the peduncular aesthetascs of
Speleonectes tanumekes (Crustacea, Remipedia) in comparison with crustaceans from stressful
environments. Invertebrate Biology 125:256 –264.
Walker, G. 1971. A study of the cement apparatus of the cypris larva of the barnacle Balanus balanoides.
Marine Biology 9:205–212.
Walossek , D. 1993. The Upper Cambrian Rehbachiella and the phylogeny of Branchiopoda and Crustacea.
Fossils and Strata 32:1–202.
Weatherby, T.M., and Lenz , P.H. 2000. Mechanoreceptors in calanoid copepods: Designed for high
sensitivity. Arthropod Structure and Development 29: 275–288.
Weatherby, T.M., K.K. Wong , and P.H. Lenz. 1994 . Fine structure of the distal sensory setae on the
first antenna of Pleuromamma xiphias Giesbrecht (Copepoda). Journal of Crustacean Biology
14:670–685.
Wege, W. 1911. Morphologische und experimentelle Studien an Asellus aquaticus. Zoologische
Jahrbücher, Abteilung f ü r Allgemeine Zoologie und Physiologie der Tiere 30:217–320.
Yager, J. 1981. Remipedia, a new class of Crustacea from a marine cave in the Bahamas. Journal of
Crustacean Biology 1:328 –333.
Yen, J., P.H. Lenz , D.V. Gassie, and D.K. Hartline. 1992. Mechanoreception in marine copepods:
Electrophysiological studies on the first antennae. Journal of Plankton Research 14:495–512.
Yen, J., and N.T. Nicoll. 1990. Setal array on the first antennae of a carnivorous marine copepod, Euchaeta
norvegica. Journal of Crustacean Biology 10:218 –224.
Yen, J., M.J. Weissberg , and M.H. Doall. 1998. The fluid physics of signal perception by mate-tracking
copepods. Philosophical Transactions of the Royal Society of London Series B 353:787–804.
Zacharuk , R.Y. 1985 . Antennae and sensilla. Pages 1–69 in G.A. Kerkut and L.I. Gilbert, editors.
Comprehensive insect physiology, biochemistry and pharmacology, Vol. 6. Pergamon Press,
Oxford.
Zaret, R.E., and W.C. Kerfoot. 1980. The shape and swimming technique of Bosmina longirostris.
Limnology and Oceanography 25:126 –133.
Zhang , D., S. Cai, H. Liu, and J. Lin. 2008. Antennal sensilla in the genus Lysmata (Caridea). Journal of
Crustacean Biology 28:433–438.

You might also like

pFad - Phonifier reborn

Pfad - The Proxy pFad of © 2024 Garber Painting. All rights reserved.

Note: This service is not intended for secure transactions such as banking, social media, email, or purchasing. Use at your own risk. We assume no liability whatsoever for broken pages.


Alternative Proxies:

Alternative Proxy

pFad Proxy

pFad v3 Proxy

pFad v4 Proxy