Manual of Field GEOLOGY
Manual of Field GEOLOGY
Manual of Field GEOLOGY
Robert R. Compton
Professor of Geology
Stanford University
MANUAL
OF
FIELD
GEOLOGY
JOHN WILEY & SONS, INC.
J
,r
Copyright
Preface
19
18
17
-<11.
vi
Preface
Contents
Chapter 1
'
21
2-1. The Brunton compass. 2-2. Setting the magnetic declination. 2-3. Taking bearings with the compass. 2-4. Magnetic deflections of compass bearings. 2-5. Measuring vertical
angles with the clinometer. 2-6. Using the Brunton compass
as a hand level. 2-7. Measuring strike and dip. 2-8. Where
to take strike and dip. 2-9. Measuring trend and plunge of
linear features. 2-10. Care and adjustment of the Brunton
compass.
Chapter 3
36
49
viii
Contents
Chapter 5
Contents
73
88
6-1. General value of the alidade and plane table. 6-2. The
peep-sight alidade and the principles of alidade surveying.
6-3. The telescopic alidade. 6-4. Adjusting the alidade in the
field. 6-5. Major adjustments of the alidade. 6-6. Care of the
alidade in the field. 6-7. The plane table and tripod. 6-8.
Plane table sheets and their preparation. 6-9. Setting up and
orienting the plane table. 6-10. Measuring vertical angles with
the alidade. 6-11. The stadia method and the Beaman arc.
6-12. Stadia procedure with the Beaman arc. 6-13. The stadia
interval factor and the stadia constant. 6-14. The gradienter
screw. 6-15. Differences in elevation by the stepping method.
6-16. Using the stadia rod.
Chapter 7
Chapter 8
135
113
ix
"
154
170
10-1. General nature of detailed studies. 10-2. Detailed surface maps and sections. 10-3. Cleaning, excavating, and
drilling. 10-4. Underground mapping. 10-5. Sampling.
Chapter 11
185
\
Contents
Chapter 12
208
MANUAL
OF
250
13-1. Volcanic sequences and unconformities. 13-2. Cartographic units of volcanic rocks. 13-3. Naming volcanic rocks.
13-4. Structures of basic lavas. 13-5. Structures of silicic lavas.
13-6. Pyroclastic and closely related deposits. 13-7. Volcanic
feeders and related intrusions.
Chapter 14
FIELD
'GEOLOGY
"
272
296
327
Index
363
lJ
Observing and Collecting
Data and Samples
'
elusions can be reached from a field study. The study may still have
great value, however, if the observations are described fully in a report. In many cases, the geologist can state the relative probability
of several alternate hypotheses. The uncertainties may be resolved
when more data become available or when laboratory or mathematical tests are made.
The basic equipment needed for examining, describing, and collecting rocks is modest in amount and need not be expensive. It consists of a hammer with either a pick or chisel point at one end, a hand
lens, a pocket knife, a notebook or looseleaf clip folder, a 2H or 3H
pencil or a good ballpoint pen, a 6-inch scale, heavy paper or cloth
bags for collecting samples, and a knapsack for carrying lunch and
field gear.
The field notebook must be considered and selected with care because the notes recorded in it will become part of a permanent record.
The paper should be of the highest quality, for thin and inexpensive
papers tend to disintegrate in dry climates, and heavily filled ("slick")
papers wrinkle permanently when dampened. Top-quality engineer's
level-books (standard 4Y:J X 7%, in.) with water-resistant paper and
waterproof covers are an excellent choice. They are small enough to
fit in a trouser pocket and are bound so that they can be opened Hat
or folded back cover-to-cover. The U.S. Geological Survey uses a
somewhat larger ( 5%, X 8Y:J in.) book that is bound at the upper
edge and has perforated pages that can be removed easily and reorganized during the report-writing stage.
The possibility of using standard ( 8Y:J X 11 in.) looseleaf sheets for
notes should also be considered. These sheets give extra space for
sketches, accessory maps, and sections, and they may be carried in
6
0
~Spring
clip
!>{i in.
Masonite
board
hinge in cover
Notepaper
Fig. 1-1.
Clip folder for 9 X 9 in. aerial photographs and 81,6 X 11 in. note sheets.
a clip folder that is large enough to carry B.eld maps or aerial photographs as well. This has the advantage of reducing the number of
items of field gear. The individual sheets, however, must be labeled
and stored with care so that they will not be misplaced. If the climate
in the field area is either very dry or quite wet, the sheets should be
made up from the heaviest ledger stock available.
The clip folder must measure at least 9J,4 X 11 in. in order to hold
9 X 9 in. aerial photographs. An excellent one can be made by using
Ys -inch Masonite for a baseboard and a sheet of heavy plastic or
cardboard for the cover (Fig. 1-1). Some geologists prefer an aluminum folder with a steel spring hinge, and this type of carrier is indeed
good for wet or dirty working conditions because of the tight B.t of the
hinge and cover. For average mapping, however, its extra weight and
rather sharp edges are undesirable. It also tends to become very hot
when left in the summer sun for a few minutes and exceedingly cold
when carried in freezing weather. Moreover, the common 8Y2 X 11
in. size is not large enough for 9 X 9 in. aerial photographs.
For note taking, various drawing pencils and ballpoint pens should
be tested. A good ballpoint pen gives dark, legible copy that will not
smear on good paper, will erase reasonably well, and is more or less
waterproof. It may clog, however, under dusty working conditions,
and even the best pen may run out of ink in the middle of a day's
work. The ideal pencil produces legibly dark copy that will not smear
easily. The optimum hardness of its lead will depend on the paper,
the individual, and the climate (damp paper is gouged by even moderately hard pencils). Generally, something from H to 3H is best.
A cap eraser and a pocket clip should be put on each pencil, and at
least one extra set should be available in camp in case one is lost in
the B.eld. A fountain pen should be used only if the ink is known to
be waterproof and permanent.
Hand lenses used in most B.eld studies should magnify 10 to 15
times, although additional higher or lower-power lenses may be needed
for some determinations. Relatively high-powered lenses have a small
B.eld, a short depth of focus, and must be held very close to the object
being examined. They are therefore more difficult to use, especially
where lighting is poor. Regardless of its magnifying power, the
quality of a lens should be tested by focusing on both smooth and
rough surfaces. The fnagniB.ed image should be sharp throughout,
even near its edges. Three-element lenses _generally give undistorted
images.
An ordinary geologist's hammer is adequate for most rocks, but a
hammer weighing about 2 lb may be needed to colle~ hard rocks
(tough sandstones, lavas, hornfels~s, and so forth). Rectangular
corners and edges on the striking end of a new hammer should be
hammered or B.led to a bevel because they may send off steel chips at
dangerous speeds when struck against a tough rock.
Preparations for wet weather. Wet weather should be anticipated;
it can upset a B.eld schedule seriously. Maps and aerial photographs
can be waterproofed completely by painting or spraying them on both
sides with transparent plastic. Notepaper can be treated in the same
way, or it can be dipped in a solution of paraffin and a volatile solvent.
Paraffin should not be used on maps or photographs because they
cannot then be marked with ink. The most serious damage to maps
and notes is caused by slipping on steep muddy slopes, and this can
be remedied by wearing nailed boots. In order that B.eld work in
rainy areas will be thorough and consistently precise, the geologist
should clothe himself so that he will be comfortable (though not
necessarily dry).
1-4. Taking Geologic Notes in the Field
'
Exactly what data should be recorded
in field notes will vary with
the project. In most geologic surveys, the notes concentrate on descriptions of rocks and structures, esplk:ially those features that indicate the origin of the rocks or their, relative ages. As a field project
gets under way, these descriptions typically cover one outcrop at
a time, in the order in which they are found and examined. As the
geologist becomes more familiar with the rock formations and structures, his notes should record critical descriptions of features that have
been traced through a series of outcrops. This organization is very
important because it keeps notes from becoming repetitive and difficult to use.
Before the geologist leaves the field for the season, he should make
sure that his field notes include full descriptions of rock units and
structures in all parts of the area he has surveyed; otherwise he will
not be able to write an accurate report on them.
In Fig. 1-2, note 1 is an example of a rock description made on
the basis of a single outcrop, while note 8 is an example of a more complete lithologic description relating to a number of outcrops. Lithologic descriptions are more usable if recorded in a fairly systematic
way, as by the outline that follows.
1.
2.
3.
4.
5.
JO
.Tul'f z, 1'153
\T.R. Doan
Fern/if 9uad- Lower Ston'f Crk Area
Mflp sher;f .F-3
I. Ss massive thru 120' of this large ofcp; If qratt 1 hard
calc, local/If wecdhered fan, friable, well sorted, med. fo
coarse-qrained1 plaq.-ZOJ., qreen chert(?) qra/ns abnt.
2. Crags 5 of I, f/ne-9ra/ned
~-~-=--=--=--mdr;f
Ss in I- 3 'beds w/th qradinq ~~:;::-:.-t;:::~~D- A
and lam. as shown
-~:-~,..;::-~_.::;::,:~
Zrock Samples A and 8 show fypical ~:-::_--.:.::-.:<.<:_:::-:~8
samples qrain-s1ze varial/ons (see~) ~~ mdsf
mdst beds i< -J' thick are
' 10 m. 1
silftf, apparan111f wilhoul forams.
3. Recon traverse from pis. I and Z down rldqe shows sed/on:
I' 1
... 2
e~
f te I
NW
qray calc sh
/000 ff
? .:.-.- }$\\\:\1:~"0>0.~
.
\-\~~-'-'\\:-\'<>.'\~'"
?-m7f1l
.
.
..a>"~
Ss
s;;~
~~
')
..,
.. '\' ).( ,
Fig. 1-2.
appqrently
m!ss!n9 on this ridge; check Silor and Mill Crk:; for
.Check major fault.
fossif6. Tn9onia COI/ecfed 10ft. a6ove base massive ss untf. Exc.
preservqf/on /nd/cafes not re-worked.
1:2.: Ston'f. Crk area, Jon .JJ:Ji S-flow/n9 tr/b. above
Jones Mill, 1170' {paced along sfream) a/;;ove confluence
w/fh Sfom; Crk; prominent flufed gray ofcp on !VE bank
crk. Approx: center N}/4 S,ec3j T3S, R4c.
7. Vnconf between massive ss . ana' deepllj weathered qr,
reddish arg qr looks like old soil profile under ss ;
very crude %-bdng in ss suqqest frouqh-sefs (and
fluv/afile onq/n ?). Poor exp.
B. P.M. spent exam/ninq qr over ma/n ridqes SW of 7.
.!f is proh. whaf Smith called Pine Mt. gr in req/on
to 5 W. Areo ls craqglj j covereel blj low brush and
small p/ne groves. Reddish mica-bear/nq so;'/
characfer/sf/c. #o lg met ofcps buf seal/erect /nels of"
q.zf and hio- relc/ granul/te; /nels loca II'! in crudely
planar swarms (see map) buf no other planar
or //near sfrucfures. Stream ofcps show -fresh
pink rock., w/lh comp: 9Z 1 .35; Kfeld1 40;
plagj .ZOJ b/o {much altered to chi ) 1 5 to ~j muse, L/.
Grains Y.3- 12. /n.j local/If w/lh lg euhedra K fe/d.
{/.5 /o of" rock.); fexfure hypid. to al/ofr/omorph /c.
Rock C1ppears veri( sim//iar over /his qrea.
Grqve/s Ridge cresfs have pafche~ of 9ravel wdh well-rounded
rebbles of qz and ap!/fe (prob from qr area); these
support Smiths /dea of' upl/f"fed fluv/af//e surface
on NE !'lank Pine Mf.
Fig. 1-2.
(Continued)
Jl
l~.~.
~11
12
l3
14
J5
16
l7
Each rock or fossil specimen must be marked with a number matching that used in the notes and on a map or aerial photograph. Most
samples can be marked directly with a felt-tip pen. The number may
also be written on a piece of adhesive tape fixed firmly to the sample
in the field. If the sample is wet, the number can be written on a piece
of paper that is secured to the sample with string or a rubber band.
The sample should then be put in a paper or cloth bag on which its
number is marked clearly so that it can be identified without being
unpacked.
J8
19
known to be generalized or obsolete, important rock and fossil localities must be described fully in the notes. The purposes of locality
descriptions are: ( 1) to assist others in finding the locality, ( 2) to assist
the geologist in revisiting the locality, and ( 3) to provide the means of
plotting the localities on a suitable base map when it becomes available.
Locality descriptions are best based on points or features that are
permanent, can be pinpointed on the ground, and are generally shown
accurately on maps. Pe.rhaps the best of th?'e are .triangulation stations and bench marks of the various government surveys or the township-section system of the Bureau of Land Management. These points
generally carry a marker with the name or numbers of the station
stamped in a brass plate, though some section corners are marked
only by a pipe or a roughly marked stone. From fthese points, a locality
may be measured and described by m,eans of a compass bearing and
a paced distance, as described in Chapters 2 and 3. In areas covered
by the township-section system of the Bureau of Land Management,
roads or fences commonly lie along the north-south and east-west
borders of sections and quarter sections, and this gives a handy refer-
R3 w
R1"1R2EL
I .. : .. I .. .. I "-I
I Mile
R2E
2
10
11
I2
I8
I7
I6
I5
I4
I3
I9
20
2I
22
23
24
30
29
28
27
26
25
33
34
35
36
TIN
T2 N
TIS
T2S
Data and specimens that are worth collecting are worth locating
with care; this can be done either by marking the point directly on a
map or aerial photograph or by recording a description of the locality
in the fl.eld notes. When a map has a scale of 1:62,500 or less, or is
= township; R = range
~ r:ornPrL~Pr ~1~
Sec.
corner
NE>.i SE>.i
J___L____,l,
W~SE>.i
20
ence grid from which to pace out locality positions (Fig. 1-3). Other
suitable landmarks for locality descriptions are sharply defined hilltops, stream intersections, road or railroad crossings, solid buildings,
and similarly permanent features. Where localities are so far from
these features that a compass bearing and paced distance cannot be
used to describe them in the field notes, they may be located by taking
bearings on several prominent points ( Section 4-2).
Locality descriptions should begin with the name of the quadrangle
or some large and well-established geographic feature, proceed to
smaller and more local features, and, finally, describe the immediate
landmarks and appearance of the outcrop itself. The field notes shown
in Fig. 1-2 include an example of a locality description (note 6).
References Cited
Bureau of Land Management, 1947, Manual of instructions for the survey of the
public lands of the United States, 1947: Washington, D.C., U.S. Government
Printing Office, 613 pp.
Chamberlin, T. C., 1897, The method of multiple working hypotheses: Journal of
Geology, v. 5, pp. 837-848.
Gilbert, G. K., 1886, The inculcation of scientific method by example, with an
illustration drawn from the Quaternary geology of Utah: American Journal
of Science, v. 31, pp. 284-299.
~
Using the Compass, Clinometer,
and Hand level
..
.,J
Fig. 2-1. The Brunton compass. Insert at lower left shows enlarged section through
needle bearing.
2J
23
The round hull's eye bubble is used to level the compass when a bearing is read, and the tube bubble is used to take readings with the
clinometer. The clinometer is moved by a small lever on the underside of the compass box (not shown in the figure).
A compass should be checked to ascertain that: ( 1) both levels have
bubbles, ( 2) the hinges are tight enough so that the lid, sighting arm,
and peep sights do not fold down under their own weight, and ( 3)
the point of the sighting arm meets the black axial line of the mirror
when the mirror and sighting arm are turned together until they touch.
Other adjustments that may be required are described in Section 2-10.
should be used.
22
1. Open the lid about 135; turn the sigiting arm out and turn up
its hinged point (Fig. 2-2A ).
2. Standing with the feet some~hat apart, hold the compass at
waist height with the box cupped in the left hand.
3. Center the hull's eye bubble,. and, keeping it approximately
centered, adjust the mirror with the right hand until the point sighted
and the end of the sighting arm appear in it.
4. Holding the compass exactly level, rotate the whole compass (on
an imaginary vertical axis ) until the mirror images of the point sighted
and the tip of the sighting arm are superimposed on the black axial
line of the mirror.
5. Read the bearing indicated by the white end of the needle, which
should be nearly at rest.
6. After reading the bearing, check to make sure the line of sight
is correct and the compass is level.
7. Record or plot the bearing at once.
~-
Fig. 2-2., Compass set for taking a bea~ing at waist height (A) and at height of
eye (B).
24
2. Hold the compass in the left hand at eye level, with the sighting
arm pointing toward, and about 1 ft from, the right eye.
3. Level the compass approximately by observing the mirror image
of the hull's eye bubble, and, holding the compass approximately
level, rotate it until the point sighted appears in the small sighting
The compass will give incorrect bearings if there is any local deflection of the earth's magnetic field. Objects containing iron, such
as knives, hammers, and belt buckles, should be kept at a safe distance
while a reading is made. This distance can be determined by placing
the compass on a level surface and bringing the object toward it until
the needle is deflected. A strong pocket magnet must never be carried near a compass. Steel fences, railroad rails: and steel pipelines
should be avoided if possible.
'!If)
Rocks and soils rich in iron, especially those containing the mineral
magnetite, can cause deflections that are difficult to detect. Bodies
of basalt, gabbro, skarn, and ultrabasic rocks are particularly likely
to affect compass readings. Relatively strong effects can be tested by
bringing large pieces of rock close to the compass. If the magnetic
mass is small compared to the distance between two stations, foresights and backsights between the stations will give inconsistent re-
25
t::
0
....2
A
ff'rri:\._
~L- _ _ _
Apparent bearing is N85 E
Apparent beanng
is N75W
=-rf'TVi\
-
-::, -
B~
suits (Fig. 2-3). Where larger masses are involved, the deflection
can be measured by pinpointing .two stations on an accurate map,
measuring the bearing between them 'with a protractor, and comparing
this bearing with a compass bearing taken between the same stations
in the field. This measurement will correct the declination for that
part of the area. The same result may be achieved by taking readings
on Polaris at a number of points within the area to be mapped. If
the magnetic disturbances are moderate and vary systematically over
a given area, a local isogonic map can be constructed from which
corrections of compass readings may be made. If magnetic deflections are large and distributed irregularly, mapping must generally
be done with other instruments, as the peep-sight alidade (Section
6-2) or the sun compass, a nonmagnetic instrument operated on the
basis of the time of day and the direction of the sun's rays. It is also
possible to make a compass traverse in such a way that the deflections
are accounted for (Section 3-5).
2-5. Measuring Vertical Angles with the Clinometer
26
ooe---
27
"'
.Fig. 2-4.
3. Look through the window of the lid and find the point to be
sighted, then tilt the compass until the point of the sighting arm, the
axial line of the window, and the point sighted coincide.
4. Move the clinometer by the lever on the back of the compass
box until the tube bubble is centered, as observed in the mirror.
5. Check to make sure the sights are still aligned, then bring the
compass down and read and record the angle.
Computing difference in elevation. The approximate difference in
elevation between the point occupied and the point sighted can be
computed in the field if the slope distance is paced and if a small table
A
SP = QP tan LM
~----
:::l!t-Estimated
.
height of
C~LM
-- -
eyeontree
Station
B
I~
/
---Horizontal distance - - -
Point sought
k~3---ll'
''
(-1
SP= QPtanLM+ WQ
r
__. -
~/u
l'j;
___ ---
==--=-=-~~
""'-:::-~-. ---
i}
I
1:
Height of
- - - - - - - - - - - - - - - ~ instrument
Fig. 2-5. Finding difference in elevation ( SP) from a vertical angle ( M) and
horizontal distance ( QP). (A) Relation used when sighting on a point at height
of eye. (B) Relation used when sighting uphill to point itself.
Fig. 2-6. Measuring the difference in elevation between two stations by using
a hand level and counting eye-level increments.
28
29
two end points. As the first level sight is made, an object such as a
stick, leaf, or stone is marked mentally and kept in sight while walking to it. Standing on this marker, another point at eye level is chosen
farther uphill, and the procedure is repeated until the end point is
reached (Fig. 2-6). The number of moves is tallied and multiplied
by the height of the surveyor's eye, with the last fractional reading
estimated to the nearest even foot or half a foot. If the country is
reasonably open, the traverse can be made both quickly and accurately. The serious error of miscounting the tally can be prevented
by keeping a pencil tally on the cover of the compass or by using a
tally counter.
2-7. Measuring Strike and Dip
plotted.
To measure the dip, the observer opens the lid and sighting arm
of the compass and holds it in the line of sight used to measure the
strike. The compass is then tilted until the upper edge of the box
and lid appear to lie along the bedding plane (Fig. 2--7B). The
clinometer lever is rotated until the tube bubble is centered, and the
dip is then read and recorded to the nearest degree. The intersection
of the strike line and the dip line on the map is customarily taken to
be the point at which the reading was made.
If the bedding line contains no distinctive feature that marks the
point on a level with the observer's eye, it is necessary to mark the
point with a pebble, stick, or some other object; otherwise, the reading
will be approximate only.
A.
B.~
--_['_-~ /<{'R>J eddmg
plane~.;,.
4:<"' .:-- --;f/i f/ ~ ". ,;;-;ifl0.
B
, ...
,/;> . ,
~
:
;
:
~:0
c_J?J
":1: ~
____/"'
-~
[
1 . ..
v---
.
Measuring strike and dip. (A) Sighting a level line in the plane of a
bedding surface. (B) Measuring the dip of a bedding surface.
Fig. 2-7.
30
II. Method for steeply dipping beds. Where beds dip more than
60, a level line of sight to a bedding plane can be found by a somewhat less precise method than that just described. After the observer
is in a position to see a bedding surface as a line, he takes a bearing
by the eye-level method (second method of Section 2-3), being careful to center the hull's eye bubble exactly and to sight on the trace of
the bedding. If he prefers to take the bearing by the chest-level
method, he may move to such a position that the compass will be in
the plane of the bedding surface. By either method, the compass
can be held level enough to define the strike line within a few degrees.
The dip is measured in the same way as described above.
Ill. Method of leveling between two outcrops. This is the best
method for measuring the attitude of gently dipping beds. It can be
used where a well-marked bed crops out on the opposite sicjles of a
small valley, gully, or excavation. The observer stands in front of one
outcrop so that his eye is at the level of the top or base of the bed.
He then uses the Brunton compass as a hand level to find a level line
to the same bedding surface in the opposite outcrop. The bearing
of this line is the strike of the bedding, and the dip is measured by
sighting across at the opposite outcrop, just as described above for
method I.
IV. Method of holding compass against bedding surface. Methods
I and II cannot be used where brush, rocks, or trees make it impossible
to get in a position to see a bedding surface as a line. It may then
be necessary to take a reading by holding the compass against a bedding surface. The surface chosen must be smooth, clean, and representative of the outcrop. The compass is opened and one of the lower
edges of the compass box is held firmly against the bedding surface;
the compass is then rotated until the hull's eye bubble is centered
(Fig. 2-8A). The bearing in this position is the approximate strike.
The dip is read by placing the side of the compass box and lid directly
on the bedding surface and at right angles to the direction of the strike
(Fig. ~B). The clinometer is then turned until the tube bubble is
centered.
The simplicity of this method makes it appealing; however, the
base of the compass is only 2 in. long, and therefore large errors may
result unless judgment is used in selecting the surface. On a fairly
large bedding surface, a clip board or map case (with no steel parts)
may be held firmly against the surface to help average out its irregularities. The compass- is then jleld against the board to make the
reading.
V. Methods for nearly ltorizontpl beds. Where dips are less than
10, small irregularities in the bedding will cause major local variations in strike. Method III will give superior results, but it requires
unusually good exposures. Method I will give reasonably good results
down to dips of about 4 if a large single section of a bedding surface
is exposed. To obtain the strike of beds dipping still more gently,
it is often .easier to determine the direction of maximum slope (dip)
first and then to take the bearing of a line at right angles to it. A
quick way to do this is to pour a little water on the surface, stand
over the streak it forms, and measure the strike as the bearing at
right angles to the streak. Another way to find the maximum slope is
to place the compass on a bedding surface as though measuring dip,
then center the clinometer bubble while rotating the compass slowly
back and forth against the surface. Because of the irregularities on
most bedding surfaces, however, this method is generally unreliable.
VI. Three-point method. The strike and dip of gently dipping beds
can be determined accurately from three points that lie at different
elevations on one bedding surface. The bedding surface must be
identified with certainty at each outcrop, and therefore the top or
base of a distinctive rock unit or bed should be used. The construction requires that the distances and directions between the three points
be known, as well as the gifferences in elevation between them. The
distances and directions can generally be measured from a map or
aerial photograph; they can also be determined by a compass traverse
(Chapter 3). The differences in elevation can be determined from
a contour map or by the methods described in Sections 2-5 and 2-6.
Figure 2-9 illustrates three outcrops (A, B, and C) and the construction used to determine strike and dip from them. The strike is found
by locating a point, as D, which has the same elevation as the inter-
Fig. 2-8. Measuring approXimate strike and dip by holding compass against a
bedding surface.
31
32
7>
I
Fig. 2-9. Determining average strike and dip by making a construction based on
a map of three points on a bedding surface (see text for explanation).
mediate point, B, and is on the line joining the highest and lo)Vest
points, C and A. Point D can be located by solving the relation
AD
= difference in elevation
AE between A
and B
The dip is then read from a table of tangents. The dip may also be
determined graphically by constructing a view at right angles to the
strike, as explained in books on engineering drawing and structural
geology.
VII. Method for beds truncated by nearly level surfaces. The strike
of beds exposed on horizontal or nearly horizontal surfaces of unpaved
roads, trails, stream beds, terraces, or bare ridges may be measured
quickly and accurately. This is done by standing directly over the
outcrop, cupping the compass in both hands at waist heigf{t, and
aligning it with the trace of bedding that passes underfoot. The long
slot in the sighting arm is useful for aiming the compass in this way.
In order to determine the dip, it is generally necessary to hammer out
and clean off a bedding surface, upon which the compass is placed
directly.
2-8. Where to Take Strike and Dip
33
place at all, being large boulders, blocks of float, or segments of landslides. A general survey of the slopes around outcrops will generally
resolve such problems. If there is still some question as to the reliability of a measurement, a question mark may be entered next to the
plotted symbol or the strike line may be broken (Appendix 4).
Downhill creep of soil and mantle commonly bends planar structures that occur on slopes. Where this is likely, readings are best restricted to the bottoms of valleys, to the tops of ridges, and to cuts
, that are deep enough to get beneath the 'zone of deformation. Thick
beds of massive rocks: such as san~one, quartzite, and lava, are not
likely to be bent by downhil}, cr~ep.
Outcrops should also be examinM to make certain that what is
taken for bedding or foliation is not jointing, bands of limonite staining, or some other kind of discoloration. Changes in texture ( especially grain-size) or changes in mineral composition are the best indicators ot bedding. In massive sandstones, bedding may be shown
only by the approximate planar orientation of mica flakes, platy carbonaceous or fossil fragments, shale chips, or platy and elongate pebbles. The possibility that bedding features in sandstones are only
local cross-bedding must be considered. The identification of bedding
in metamorphic rocks may be still more difficult, and there are a
number of planar structures of igneous and metamorphic rocks that
should be identified carefully wherever they are measured and plotted
(see appropriate Sections in Chapters 12, 13, 14, and 15).
2-9. Measuring Trend and Plunge of Linear Features
Trend and plunge are used to define the attitudes of linear features.
The trend of a linear feature is the compass direction of the vertical
plane that includes the feature. If the feature is horizontal, only the
compass direction is needed to define its attitude. If it is not horizontal, the trend is taken as the direction in which the feature points
(plunges) downward. The plunge is the vertical angle between the
feature and a horizontal line.
To measure the trend of a linear feature, the observer stands, if
possible, directly over ~ surface that is parallel to the linear feature
(Fig. 2-10). This surface is sometimes described as "containing" the
feature or as the surface on which its maximum length is seen. The
observer faces in the direction in which the linear feature points
downward. He determines the bearing of this direction (the trend)
by holding the compass at waist height and looking down vertically
on the feature through the slot of the sighting arm. When the slot
34
__ {Trend of lineation
----- ......
---
---
.,._Horizontal line in
vertical plane
of plunge
....
Fig. 2-10. Measuring the trend of linear structures (upright figure) and the amount
of their plunge (kneeling figure). The map symbol (insert) shows how the
lineation arrow can be combined with the strike and dip symbol of the foliation
surfaces.
is parallel to the trend of the feature, the bearing at the white end
of the needle is read. The trend is then plotted on the map as a line
originating at the point occupied by the observer.
To measure the plunge of the feature, the observer moves so that
he is looking at right angles to its trend (Fig. 2-10, right). The
reading is taken on the trace of the linear feature seen from this position, exactly as in taking the, dip of the trace of a bed. An arrow
point is then drawn on the map at the downward plunging end of
the trend line, and the amount of plunge is lettered at the end of this
arrow. For horizontal linear features, an arrow point is drawn at
both ends of the line.
,;
The compass should never be carried open in the hand while walking over rough or rocky ground. If an extra mirror and glass cover
are included in the field gear, these can be replaced in the field, but
if the hinges are bent or the level vials broken, the instrument must
be sent to the manufacturer for repair.
If the compass is used in the rain, or if it is accidentally dropped
in water, it should be opened and dried because the needle will not
function properly when its bearing is wet. The glass cover can be
35
removed by forcing the point of a knife blade under the spring washer
that holds it in place. This should be done at a point opposite the
dove-tail join of the washer, using care not to crack the glass cover
along its edge. With the washer off, the glass cover can be lifted
from the box, and the needle taken off its bearing. The cone-shaped
pit of the je'Yel bearing should be cleaned and dried with a sharpened
toothpick and a bit of soft cloth or soft paper. The needle lift is
then removed and the inside of the compass dried and cleaned. Care
should be taken not to push so hard against the clinometer level vial
that it turns on its aris. After the ne~~lle-lift arm, needle, and glass
cover are reassembled, the spring washer should be placed on the
glass cover by joining its dove!tailed tnds first, placing them firmly
against the compass box and then forcing the rest of the washer
down by moving the fingers in both directions away from the dovetail join. If this cannot be don~ with the fingers, two small pieces
of wood can be used.
The mirror can be removed by tapping in the small retaining pin
and removing a spring washer like that on the box. The new mirror
must be inserted so that its black sighting line is at right angles to
the hinge axis of the lid. This can be done approximately by rotating
the mirror until the sighting line bisects the sighting window in the
lid. The setting should then be checked by closing the lid against the
upturned point of the sighting arm and determining if the point of
the arm meets the sighting line of the mirror.
Before a new or a borrowed compass is used in the field, it should
be checked to make certain the clinometer level is correctly set. To
do this, the clinometer is set at 0, and the compass is placed on a
smooth board that has been leveled exactly with an alidade or a good
carpenter's level (a bull's eye level is not sufficiently accurate for this).
If the tube bubble does not come to center, the compass is opened as
described above and the clinometer level vial moved appropriately.
Ordinarily this can be done without loosening the clinometer set
screw. The new setting is checked by placing the compass on the
board again, and the procedure repeated until the bubble is centered
exactly.
In starting work in a new field area, one may find that the dip of
the earth's magnetic field is so great as to cause the compass needle
to rub against the glas1>~1id when the compass is held level. To correct
this, the glass cover is removed and the copper wire coil on the needle
moved one way or the other until the needle lies level.
In a traverse a series of points are surveyed by measuring the direction and distance from one point to a second, and from the second
to a third, and so on to the last point. The directional course of
this series of measurements is generally irregular; if it is eventually
brought around to the starting point, the traverse is said to be closed.
Each of the points of the traverse is called a station, while the measured "
distance between two stations may be called a leg of the traverse.
The traverse is used as a skeletal map on which geologic data are
plotted along or near the traverse legs. These data may be compiled
with those from other traverses to make. a complete geologic map.
They may also be used to construct a vertical cross section and
columnar section showing the rock units and structures traversed.
The traverse is commonly used to measure thicknesses of rock units,
to compile detailed descriptions of sedimentary or volcanic sequences,
and to study defonnation in complexly faulted or folded rocks. If
a topographic map of suitable scale and accuracy is available, it should
be used as a base on which to plot the traverse. Published topographic
maps, however, have scales of 1:24,000 to 1:62,500, and many geologic
studies require scales larger than 1 : 6,000 ( 1 in. = 500 ft). Most
detailed studies therefore require constructing a map from the traverse
data.
The scale of the traverse is chosen so that the smallest units that
must be shown to scale can be plotted easily. If the purpose of the
traverse requires, for example, that beds 10 ft thick must be shown,.;;
to scale, the best scale for the work is 1 in. = 100 ft. In general, anything that cannot be plotted as a feature Y!o in. wide cannot be mapped
easily to scale, although features that plot lho in. wide can be shown
accurately if great care is used.
The method by which the traverse should be surveyed is determined
by the accuracy required and the time and equipment available. For
many projects, a compass is adequate for determining the bearings
of the legs, and the legs can be measured with a tape or by pacing.
An advantage of pacing is that it can be done by one man; its accuracy
36
37
38
39
ll
"
..
/s I A 2"
5
4! """'
sferke If
z-. 3
Dt'sf. == 147
The procedures of a typical compass traverse include: ( 1) surveying the stations and legs of the traverse, ( 2) measuring a profile
along the traverse course, ( 3) plotting the stations and geological
features on a field sheet, and ( 4) describing the geologic features in
field notes. The equipment, most of which is described in Section l-3,
should include a geologist's hammer, Brunton compass, clip. board or
notebook, pocket knife, medium-hard pencil with clip and eraser for
plotting, ballpoint pen or medium pencil for taking notes, hand lens,
knapsack, specimen bags, protractor, and 6-inch scale. The scale
Fig. 3-1.
..
of
40
Before the firs~ leg is plotted on it, the general layout of the traverse
must be planned so the plot will not go off the sheet on the second or
third leg.
Generally, the scale of this map need be only half as large as that
of the final illustration. Its purpose is to develop a continuous picture of both the geographic features and the geologic structures as
they are traversed. This permits checking the continuity of faults
and contacts that cross the traverse course more than once. Bearings
and structural attitudes that have been misread or transposed can
be detected by orient'ing the map in the field and comparing the
plotted symbols with the out'brops.
......
'
4~2/-60
ZOO,Feef
70
Fig. 3-2.
41
""
"!
42
43
veniently by the number of feet from the last station. If parts of the
traverse are revisited and supplementary notes are taken, these numbers offer a simple means of fitting new data into place.
-\
It
44
,-----------
pen and holder, black waterproof ink, 12-inch scale, and an accurate
protractor with a base of 6 or more inches. Additional items that may
be useful are a contour pen, a drop-circle compass, lettering guides, and
a pen cleaner.
The traverse should be plotted accurately in pencil before any ink
is used. This reduces ink erasures and provides a means of arranging
the figures in a single illustration. The instructions that follow suggest
consecutive steps for this procedure.
_
-----------------------------]
PrOfile of
road
r-------------------------------~tt~o~:~~~:n
Fig. 3-5.
45
Actual profile
of section
the road.
As shown in Fig. 3-5, this problem may be met in three ways. First,
the actual ground profile may be surveyed along the line of the cross
section. This requires that the section line be chosen first, as described
in Section 3-8. The survey of this profile may take considerable time
and effort if the terrain is irregular or covered by trees and brush. The
second possibility is to estimate an approximate profile along the line.
of section by reading vertical angles from the traverse course and
sketching the terrain between. If this is done, the profile on the final
illustration must be dashed and labeled as approximate. The third
possibility is to ignore the profile along the line of section and to''"use
the profile of the traverse on the final illustration, dashing it and
labeling it suitably.
'1
!f
I
/"
~
///.-('"
Fig. 3-6.
I-
I
1I
II
_,(
_.-
,-,
Line ol section
46
47
....
~ ~ rJ: _/~'E-
II
..v,.:-'i,''':.<::-.....
.... --.-.,..
Map
\1 I
~~-~
:
:MI
Accessory section
1.
/
Fig. 3-7. Using aT-square to project features from the line of section AA' to the
cross section.
6. About % in. above the base line of the section, plot a profile line
that represents the datum at which all or most of the structure readings were taken in the field.
7. Tape the sheet to a drawing board -so th~t the section line is
parallel to an edge of the board. Using a T-square on this edge,
project all structures from the section line of the traverse to the profile
of the cross section or to whatever elevation was used for each structure
reading (Fig. 3-7).
8. Draw bedding lines where structure readings have been projected
to the profile, making each about 7i in. long. Where the. line of section crosses the strike obliquely, the bedding shown on the cross section must dip less steeply than the dip measured in the field. This
dip, called the apparent dip, can be determined from the true dip by
using the diagram of Appendix 9.
9. Complete structures under the profile line, projecting them down
to whatever depth the geologic data will permit. In many cases this
will require shifting the base of the section that was drawn provisionally in step 5. Folded beds must be constructed accurately, as noted
in step 4. Examine the structures mapped on either side of the line
1
of section to determine if some may cut the section below the ground
profile. Folded units, for example, may plunge into the section, and
faults that strike parallel to the section may dip into it below the surface (Fig. 3-8). An example of a cross section is shown in Fig. 11-5.
10. Determine the true thicknesses of the rock units. They may be
scaled from the cross section if the section cuts the strike of the units
at about 90. If this is not the case, calculate the thicknesses trigonometrically, as described in Section 12-8.
Age
''II
Formation
~
"'c
~
Thickness
Description
Lithology
Eroded
Linder
Format1on
-:-:::-(ofr(y...:;:
800' +
~;;_;;:_~:::_~
._.
lg
~
-.,
c::>
exposed
~: ~o_
. :.... -.:-:
Ro]'as
Sandstone
1050'
..: : ......: .
... .::.o<?~
oo.. .. .
..
:::...<-~ ~-,'.': o.
-n-~=--r=~ll~:~!--.~~---~~-;.?;: .1....;
o o o
o o
.<>
-=:
Miler
Shale
1200' +
Unconformity
-most
30-100shale
It thick
Brownbeds
calcareous
intercalated with 12 in. beds
of fine-grained sandstone
Faulted
Inoceramus bed
-Incomplete
Fig. 3-9. Columnar section, showing a sequence that has been divided into three
rock units.
48
.,
!J
Plotting Geologic Features
on a Base Map
Maps used for plotting geologic features and note numbers in the
field are called base maps. Planimetric base maps show drainage, culture (man-made features), and perhaps scattered elevations; topographic base maps show contours as well. Accurate topographic maps
are ideal base maps, for cross sections can be made from them in any
direction, and their contours provide several means of plotting outcrops
accurately.
To be useful, a given map must have a suitable scale, must have
been made recently enough to show existing culture and drainage, and
must have contours that delineate the topography accurately. The
7Yz-minute (1:24,000) topographic maps of the U.S. Geological
Survey are excellent base maps. Their scale is such that features 200
ft wide ( 0.1 in. on the map) can be plotted easily, and features 100 ft
wide can be shown to scale if drawn with care ( an average pencil line
,~ covers 5 to 10 ft of a feature at this scale). Because most of these
maps have been made from aerial photographs, however, trails, minor
roads, and buildings that lie under trees may be shown inaccurately.
Many quadrangle maps with a scale of 1: 62,500 are useful base
maps, but some older maps having this scale may be generalized or
obsolete. Small features must be plotted carefully on them because
features 500 ft across on the ground are only 0.1 in. wide on the map.
Moreover, it make take considerable time in the field to locate individual points on these maps, for many topographic and man-made
features are not shown on them. It is often preferable to map on
aerial photographs with a larger scale and then transfer the geologic
features to a topographic map (Section 5-11).
It may be desirable to have a base map enlarged two or three times
to give more space for plotting symbols and note numbers. Enlarging
does not increase the accuracy of the base map, but it may improve
the geological mapping because of the greater ease of plotting. Enlargements must be made so as not to distort the scale of the map
and must be printed on paper that will take ink and pencil marks.
Ordinary photostat prints are not suitable for most mapping .
1
49
50
51
~ i
52
53
on the map will locate the point. The elevation of the point occupied
can be found with a Brunton compass as follows:
.,.,.
Fig. 4-1.
hilltop.
54
55
5. If the two tentative points do not coincide, re-estimate the distance to the feature and calculate the difference in elevation again. A
few approximations like this should bring the two points to coincidence,
provided the contours are spaced fairly closely.
4-3. Locating Geologic Features by Traversing
Fig. 4-2.
56
57
Contacts and faults are the most important geologic features plotted
on the map; they will be considered in detail in Sections 4-6 to 4-12.
Folds are generally depicted on the map as a line showing the trace
on the ground of the axial plane (Appendix 4). If the crest or trough
is plotted, rather than the trace of the axial plane, this must be noted
in the explanation of the map. Although these trace lines can be
located and plotted directly in some places, symbols for most large
folds can be drawn only after rock units and bedding attitudes have
been plotted over large areas. Where folds are well exposed, it is
important to observe and record: ( 1) the trend and plunge of the
axis, ( 2) the strike and dip of the axial plane, ( 3) the plunge of smallscale folds associated with the axial region and limbs, and ( 4) the
strike and dip of secondary cleavages, or the plunge of the intersections between these cleavages and the bedding. These data will be
especially valuable if drawings can be entered in the field notes showing the actual shapes and interrelations of small-scale folds, cleavages,
or other secondary structures. If possible, these drawings should show
the fold in profile-that is, in a cross section drawn at right angles to
the fold axis. Small-scale folds and related structures are described
in Chapter 15.
A large number of planar structures can be plotted as symbols that
show their strike and dip; the more common of these are bedding, compositional layering ("banding") in igneous and metamorphic rocks,
various cleavages and mineral foliations (including How structures in
58
59
:I
Manual of Field Geology
given unit may vary laterally; thus fossils from different areas might
seem to disprove continuity of the unit. Second, some fossils do not
permit sufficiently limited age determinations to prove or disprove
lithic correlations. Two outcrops of shale, for example, are not necessarily correlative simply because each bears Early Cambrian fossils.
Further suggestions regarding units based on time rather than on
lithologic characteristics are given in Section 12-2.
Selecting new units. It is often impossible to identify a complete
suite of established units in a new area, even after considerable reconnaissance. New units must then be selected and defined, or perhaps established units should be subdivided or redefined. First, however, the established units should be evaluated fairly and completely,
for redefined names tend to confuse the published stratigraphic record.
New units must be of practical value; thus they should be as distinctive and genetically meaningful as possible. Some units cannot be
defined exactly until adjacent units have been mapped. Mapping
should, consequently, start with the more clearly delimited of the
tentative units.
fact, inking has the unfortunate effect of making sketchy lines appear
exact.
Many contacts are exposed at only a few places; some are not exposed at all in natural outcrops. Such contacts can be mapped by
walking a zigzag course between outcrops of the two rock units that
lie on either side and by plotting a line that passes between the limits
thereby established. These limits are often so narrow that the line
will be located as exactly as a well-exposed contact (at a scale of
1: 24,000, for example, within about 100 ft). More approximate contact lines should be drawn with dashes, and although the limits for
these symbols are not standard (varying with the project and the
personal feelings of the geologist), it is desirable to set limits for each
project. Dashes about Ys in. long, for example, might be used for
contacts located within 7{ 0 in. on the map, while dashes about ~ 0 in.
long might be used for less accurate contacts. Question marks can be
added to dashed lines when the existence of the contact itself is uncertain. Such admissions of inexactitude do not commonly appear
on published maps, but they are an important part of the field record.
There are several means for locating traces of poorly exposed contacts, and the success of a field project may depend on how skillfully
they are used. Rock fragments in the soil or mantle often delimit
the area underlain by a given unit. Even where downslope creep has
displaced and mixed the float, the upslope limit of fragments from a
unit can be used to locate the upslope contact of that unit (Fig. 4-3).
When rocks such as shales, tuffs, and soft siltstones are covered by a
soil that carries no rock fragments, burrowing animals commonly
bring up fragments of rock from the subsoil. Units of soft clastic
rocks can often be identified by scattered concretions or hard fossils
in clayey soils. Where no large residual fragments can be found, the
composition, color, and texture of the soil itself may be used to trace
60
Tracing and plotting contacts between rock units are basic procedures of geologic mapping. This is the most efficient way to map
units at small and intermediate scales ( 1:24,000 or less), especially
where rocks are well exposed. The method of plotting each exposure
to scale ( Section 4-8) should be considered for larger-scale mapping.
Mapping is best started along a sharp contact between two distinctive rock units. The contact should be mapped by walking along its
trace and plotting points on the map where the contact can be seen or
where'its position can be inferred closely, as within ~ 5 in. on the map.
The number of points that must be located accurately will vary with
the degree of irregularity of the contact. If the points are spaced ~
to ~ in. apart on the map, the contact can generally be plotted between them with little error. A sharp, well-exposed contact is drawn
as a solid thin line. This line should show the actual location of the
contact, except where minor irregularities must be generalized because
of the scale of the map. The strike and dip of the contact should be
measured and plotted where possible, and all data that may be pertinent to its origin should be described in the notes. Lines and symbols
must be plotted neatly and exactly. Inking in the office will improve
the appearance of the lines but it cannot improve their accuracy; in
61
63
62
Tracing contacts is an efficient method of mapping at small and intermediate scales, but if the map scale is large ( 1: 12,000 or more) the
outcrop or exposure method is often preferable. In this method, each
exposure is plotted to scale by drawing its contact with the surrounding
surficial materials (Fig. 4-4). Although small outcrops must be generalized, the contact lines should show the shapes and sizes of the
outcrops as exactly as possible. Contacts between bedrock units are
drawn as solid lines within outcrop areas and as dotted lines across
covered areas. The outcrops should be mapped by working systematically across an area; otherwise the continuity and significance of bedrock structures can be missed.
Letter symbols or colors can be used to designate the units within
the outcrop areas. Some geologists use colored ink for the lines around
exposed areas to distinguish them from other contacts. Others draw
the lines in black ink, but then add a colored line on the side away from
the exposure; this clarifies the positions of large, irregular exposures.
One advantage of the outcrop method is that observed facts are
separated clearly from inferences. Another is that other geologists
can find isolated or hidden outcrops easily, and can themselves evaluate
the evidence on which concealed contacts have been drawn. A further
description of this method has been given by Greenly and Williams
( 1930, especially pp. 189-208).
Fig. 4-4.
symbols.
64
Where two rock units grade into one another, criteria for mapping
a contact must be established with care and used consistently. The
gradation should be examined at several places where it is well exposed so that its genetic significance can be considered. The criteria
for drawing a line should have genetic meaning and be such as to
persist laterally. If there is no clear genetic basis for drawing a contact, an arbitrary physical criterion can be used for a particular area
and project. Criteria must be evaluated as to the amount of exposure,
the scale of the base map, and the schedule of the project, for mapping
gradational contacts may be difficult or time consuming.
In one type of gradation, rock units grade into one another through
a gradual change of texture or composition. Such gradations probably
form when conditions of genesis change slowly and continuously. Examples of these gradations are sandstone units that grade through silty
sandstones to argillaceous siltstones, mafic granitic rocks that grade
to light-colored granitic rocks within a single pluton, and slates that
grade through phyllites to increasingly coarser-grained schists. Contacts bet~een these units may be drawn at the center of the zone of
gradation, or at the place where a particular mineral or textural feature (average grain size, pebbles, phenocrysts, etc.) is first noted.
If the zone of gradation is narrow relative to the scale of the map, and
the criteria chosen can locate it quite exactly, a solid line can be used
on the map, as described in the last section. Because gradational contacts may pass laterally into sharp contacts, however, the line will be
more meaningful if plotted with some special symboL The hachured
symbol suggested in Appendix 4 and Fig. 4-5 is not standard, but it
expresses the sense of gradation as well as the indefiniteness of the
contact. A line consisting of short ( 112 0 in.) dashes has been used
more widely for gradational contacts; however, this symbol has also
been used for inferred contacts and for poorly located sharp contacts.
In a second type of gradation, units grade into one another through
a zone in which rock types are intercalated or mixed. Such relations
indicate that genetic conditions oscillated between extremes as they
changed or possibly that materials were mixed grossly, as by intrusion.
Examples are sandstone that grades to shale through a zone of interbedded sandstone and shale, granite that grades to schist through a
zone of schist that is impregnated with granite veins, or peridotite
that grades to gabbro through a zone of interstratified basic and ultrabasic layers. These contacts may be mapped by locating midpoints
65
~Ofeet
_ _ _ _ _ _ _N
~0
-------=-=========.:~::~
=::::::______
0
------=--------
,.
66
color. Colored pencils may also be used for faults or fold axes where
structural data are crowded on a map.
Pencil marks must be moderately erasable and waterproof. Since
pencils are lost easily in the field, a piece about 2 in. long may be cut
from each and carried in a pocket. A color can be selected quickly
from these stubs, and if one is lost it can be replaced from the supply
in camp.
67
Faults must be sought out and mapped with care. Supposed faults
have been plotted on the basis of a few slickensides or a valley that
lies suspiciously across the strike of beds; whereas some actual faults
have gone unmapped because they could not be seen in many outcrops. This occurs partly because most faults form zones of crushed or
altered rock that erode easily and therefore are rarely seen in natural
exposures. Furthermore, where faults are exposed, minor ones may
appear as impressive as major ones. Superficial shears formed by
landslides, for example, may be prominent in shallow cuts.
The geologic map provides the most important evidences of faulting:
( 1) offsets of contacts between rock units, ( 2) repetition of rock units,
or ( 3) cutting out of parts of rock units. The mapping of contacts and
the determination of detailed stratigraphic sequences will therefore
show the positions of most faults.
During the early stages of mapping, when the stratigraphic sequence
is known only in part, unexposed faults (or possible faults) can' be
recognized by: ( 1 ) nearly straight scarps that trend somewhat uphill
and downhill, ( 2) nearly straight valleys that cross structures obliquely
or that appear to offset other valleys, ( 3) lines of ponds, springs, or
water seeking plants, ( 4) linear zones of alteration, strong cementation,
or discoloration, ( 5) abrupt changes in attitude of bedding or foliations, ( 6) abrupt termination of folds, dikes, or faults, and ( 7) blocks
of float that are foreign to the underlying rock units. When such
features are seen clearly from a distance or on aerial photographs,
they should be sketched tentatively on the map. These possible faults
are then examined and either proved by mapping of adjoining rock
units or disproved by tracing contacts or beds across them. The trace
of a fault should not be projected as a straight dashed line without good
evidence, for some faults turn abruptly, and many end suddenly in a
flexure or a maze of minor faults and joints.
When a fault has been detected by the mapping of rock units, it
1
Manual of Field Geology
68
II
l
69
,....---,
/~ ,/ C.il/c .:>h
~
I
Q.ar
\......._
'P~f
b , r ,"'f{j)
I !52.0
I
I
I
\
I
Tva
/ ._.,.
F f
,.,-""'
\ ..!--~r
--- u
Q"r
,..-//
"
, _ _ _ L___
Fig. 4-7.
\.'.
2 lhi
Rl
.' '
z,.s, c...
1
'-
1
1 sh (
Pbs,
'.i~
60
1 7 I\
r/ K~
I
'\
qz v~jns,
I
py 1 ZnS /~
:~--
sh
I
~
1..
1',
' ,
\~
so-\
---_,/',
Nortn
(\
\
'
., .
-.,
~-,
\ \
Qm1
rOCk. Slides
;I .
., \ -.
-~
-~~
-\ -~~-;...
\ ,
'
,..----...-""
(
-'--'
I/
stock:
.~
.. ~-'
r
I-
....,\
/,....l'l~sJ./
j
I
'i
Q.,
'------:....~.
hflc
;;-"...,..
Srt\1fh P'Osp.
- -~
.........., --~/ -?~
...!:_ale sh _ _.,r //
~ \. . . . . --,
......,
'-. .,...
6rctctts, et;
~ l<qt.t\
'
'
/1"11
10 ....
Reconnaissance sketch map, showing the use of labels and brief notes.
gradational contacts. Unconformities are ideal contacts for reconnaissance mapping because they are generally sharp and meaningful.
Faults may be plotted readily where they are expressed clearly in the
topography, but only those faults that cause major offsets of rock
units can be detected by stratigraphic mapping. Many faults must be
sketched approximately or placed in the category of questioned or
possible faults. Major folds will be shown by the mapping of rock
units, but in metamorphic terrains it may be necessary to use smallscale folds and cleavage-bedding relations to detect major folding.
Whether other small-scale features should be observed and plotted
will depend on the purpose of the project. If, for example, one objective of a reconnaissance is to determine the approximate distribution of small-scale faults, a distinctive contact or key bed can be
walked out and small offsets sketched in a qualitative way.
In contrast to exact studies, few contacts can be walked out completely in reconnaissance mapping. Instead, traverses are generally
made along roads, streams, ridges, or trails that cross the strike, and
contacts are plotted as they are crossed. The approximate trace of
a given contact can often be seen from a ridge or road; thus it can be
dashed in for a considerable distance. When there is doubt as to its
continuity, question marks can be inserted in the contact line or descriptive queries can be lettered directly on the map. If contacts are
completely concealed between traverses, the points must be joined by
dashed lines that are drawn to conform to the bedding and the ,topography. This connecting of points should be done in the field, where
changes in soil color, vegetation, or slope can be used to locate the
trace of the contact. Field glasses will assist this work greatly, and
aerial photographs will commonly show indications of contacts and
faults that cannot be seen easily on the ground. Because of the scale
of the work, structural attitudes must be averaged over a number of
outcrops. A given strike and dip symbol may show the average attitude for an area of one quarter of a square mile or more.
I
.,
70
71
72
and cross sections should be saved for use during the preparation of
the final report.
References Cited
Billings, M.P., 1954, Structural geology, 2nd ed.: Englewood Cliffs, N.J., PrenticeHall, 514 pp.
Greenly, Edward, and Howe! Williams, 1930, Methods in geological surveying:
London, Thomas Murby and Co., 420 pp.
Hodgson, R. A., 1957, Precision altimeter survey procedures: Los Angeles, American Paulin System, 59 pp.
Nevin, C. M., 1949, Principles of structural geology, 4th ed.: New York, John
Wiley and Sons, 410 pp.
~
Mapping Geologic Features
on Aerial Photographs
Three kinds of aerial photographs can be used by geologists. Vertical photographs are taken by a camera aimed vertically at the earth's
surface. They are used widely as a base for geological mapping because they are similar to planimetric maps. Low oblique photographs
are taken by a camera aimed at an angle to the vertical, but not so as
to include the horizon. Aerial photographs showing the horizon are
called high oblique photographs. Oblique photographs cover much
larger areas than vertical photographs, but they are not generally used
for geologic mapping because their scales change greatly from foreground to background. They provide overall views of large areas,
however, and are therefore valuable in geomorphic and structural
studies.
The great value of aerial photographs lies in their detailed picture
of the earth's surface. The most detailed maps lack their exact rendering of trees, clearings, trails, gullies, and countless other small features
that can be used to locate the positions of geologic features in the
field. Moreover, adjoining vertical photographs overlap, making it
possible to obtain a three-dimensional or stereoscopic image of the terrain covered by the overlap area. This image is so unmistakable that
it can be used quickly to make approximate locations in hilly or mountainous country. Only two kinds of areas cannot be mapped readily
on aerial photographs: areas of low relief (slopes less than about 5)
where the ground is covered monotonously with grass or brush, and
areas where there is a heavy forest cover.
Besides serving as a base for plotting geologic features in the field,
aerial photographs can be used before the field season to gain an idea
of the geologic features and accessibility of an area and, after the
field season, to aid in compiling geologic and geographic data on maps.
Photographs should be examined each evening to plan the next day's
mapping, even though the mapping itself may be done on a topographic
map. This will locate routes that develop the geologic features effectively and give access through areas of heavy brush, cliffs, deadfall, or
thick woods.
73
74
I.
5-3. Differences between Vertical Aerial Photographs and Maps
Angle
of tilt
-...r--==+
/
Flight
I I~
I I I I II
End lap
-~-
--1-
I I I
1
h}L~
-+---<---4- - - - - - - -
I
I I
\
\
--
l \ \\
I
~\
--
''
'
'
Position of B on
vertical photo
I'
I
...,-Position of B on
I
,
tilted photo
I
I
I
'
I
I
!
I
_....--Position of A on
tilted photo : 1
~I I'- Position
of A on 1
I
vertica! photo :I
I
I'
cone
'
'
I
I
--- -
I Camera \ \
/axis\
I AdL
I
Fig. 5-1. Parts of two flight strips of aerial photographs, superimposed to show
characteristic overlaps.
I
1.._,_
Side
lap
"'-...
/ I1 \ ,_Photo
Photograph
I I \
- +--;--_,. __ ,__ --
75
Fig. 5-2. Tilt of the aircraft (left) produces a distortion of the rectangular grid,
as shown in the aerial photograph on the right.
76
-Aircraft
/1 I
1I I
-----------
,....-
I \ \"
the relief in this figure is greater than average, the geometric relations
are not exaggerated. Even moderate distortion of this kind must be
considered when plotting geologic features on aerial photographs.
In addition to the distortions caused by variations in scale, further
distortions are caused by the angular relations between photo rays and
the slope of the ground. In Fig. 5-3, for example, this has resulted in
a crowding together of the north-south lines on each side of the ridge.
Because this effect is especially bothersome in the outer parts of a
photograph, only the central part is used for mapping (Fig. 5-4).
r------r----
1---
,.........
I \ \\
//1 I \ \ "
/l / _j___ \ \ \
/ ~\
-........
Photograph
------...........____
77
---~v
Fig. S-3.
Distortions due to relief, which are far more common and more .extreme than those due to tilt, are caused in part by variations in scale.
The scale for each part ora photograph varies directly with the distance
between the camera and the corresponding points on the ground.
Figure 5-3 shows how a rectangular grid superimposed on a high
rounded ridge is distorted in a vertical aerial photograph. Although
~
u.
.'
0'-
........ -
....,.,
:.:::~:.:,::~(.,
.
Photographs and
stereoscope
aligned parallel
to flight line
""
Effect of the angular relation between slopes and photo rays. The outcrop widths of the sandstone unit are markedly different on three successive
photographs (bottom) because of the position of the aircraft and the facing of
the slope. The true map width of the unit is shown in photograph 2.
~~i.fur~:.{:.~ili G~~r~if.,~l
Fig. 5-4.
Fig. 5-5.
78
79
Fig. 5-6. Mirror stereoscope, with sets of heavy lines showing image cone from
photographs to eyepieces.
exaggerate the relief about two or three times, and larger mirror models
about two times.
Seeing a stereoscopic image without a stereoscope. If the eyes
look in parallel directions (as when looking at a distant object), an
overlap area can be examined stereoscopically without using a stereoscope. Perhaps the easiest way to do this is to look for several moments at a distant object, then insert the photographs into the line of
view, holding them in position for stereoscopic viewing as described
above. The three-dimensional image will be blurred at first, but after
some moments of staring at the pictures, it will tend to become sharp.
Another way to learn to see a stereoscopic image without a stereoscope is to practice with simple images like those of Fig. 5-7. By
staring hard at the two dots for several moments and then relaxing the
eyes as though looking at nothing in particular, the eyes will tend to
assume their p:rallel position of rest, and the dots will drift together,
~nally merging into one image. This image will be blurred at first, but
should come to focus shortly. The other figures should appear as
three-dimensional forms. To apply this method to aerial photographs,
it is necessary to choose points on the pictures that contrast with their
surroundings.
@
Fig. 5-7.
@ ~
[2J
80
These methods should not be damaging to normal eyes, but a stereoscope must be used for systematic work. Some stereo viewers look
cross-eyed when they force their eyes to make the images merge. The
relief is then reversed ( pseudoscopic), just as it is when two photographs of a stereo pair are reversed from their normal consecutive
position in a flight strip.
5-5. Materials for Mapping on Aerial Photographs
Prints used for geologic mapping should be double weight and have
a semimatte surface. Flight lines should be ordered complete so
that stereoscopic images can be obtained for the entire area. The
prints must be carried in a map folder large enough to protect them
adequately (Section 1-3). It is advisable to carry a pocket stereoscope
in the field. Only a very soft pencil (for example, 3B) and soft eraser
should be used on photographs because the emulsion breaks and
peels quite readily, especially if slightly damp. If the air is very dry,
a harder pencil can be used. A fine needle can be used to prick a
small hole through the print where location marks must be especially
precise and permanent. The location can then be labeled on the back
of the print. The needle may be mounted in a small wooden handle,
with a cork to cover the pointed end. Many colored pencils are too
hard to use on photographs; their marks are almost invisible, and
they tend to cut the emulsion. Soft wax-base pencils make clear marks,
but these rub off easily. Water-soluble pencils are undesirable because their marks spread and dye the emulsion deeply if the photograph is moistened. If a good deal of color work is required, it is best
to use transparent overlays, preferably plastic drawing sheets that are
frosted on one side so they will take pencil and ink marks. Tracing
paper may also be used, but it is typically less transparent, less waterproof, and likely to crack in dry climates.
5-6. Determining Scale~ of Aerial Photographs
Bl
'
82
83
After a north arrow has been drawn on a photograph and the scale
84
.
ch ange m sea1e
= 12
-
For photographs taken with a 6-inch lens (most large-scale photographs), this would give a scale change of 167 ft per inch for each
elevation change of 1,000 ft. If the scale of a photograph has been
determined to be 1 in. = 1,500 ft ( 1: 18,000) at a certain elevation,
the scale on a ridge lying 1,000 ft higher would be 1 in. = 1,333 ft,
and the .scale in a valley lying 1,000 ft lower would be 1 in. = 1,667 ft.
For distances measured over sloping ground, the average elevation
of the paced course should be used. These differences in scale are
appreciable; thus differences in elevation must be estimated carefully
or be determined with a barometer.
85
ered rapidly and only brief notes are being taken, as in reconnaissance
mapping, the notes may be recorded on the backs of the prints-where
they will never be separated from the features to which they refer.
If the north arrow of a photograph is not reliable, the strike or trend
of each structure symbol should be recorded on the back of the photograph or in the notes. Attitudes can then be plotted correctly when
compiled on a map.
Some geologists find it easier to remember the layout of a large number of photographs if they are renumbered in a simple digit sequence.
The new number should be inked permanently on the front and back
of the photographs, but not so as to obscure the original number. If
a field party 'is using duplicate prints, one set should be marked with
a capital letter or similar device. Perhaps the simplest system for
note numbers is to start with 1 on each print. Each page of field notes
must then show the photograph number, and specimens should be
labeled with a complete photograph number. A typical specimen
number would be 3F-16-5, meaning location number 5 on photograph 16 of flight number 3F.
Photographs must be kept dry because the emulsion is softened
and easily rubbed off when even slightly moistened. Furthermore,
photographs will warp severely if unevenly dampened, as by contact
with perspiring hands. If photographs become wet, they should be
interlayered with blotting paper and dried overnight under weights.
Because soft pencil marks on photographs rub off easily, location
nm~bers and structure symbols should be inked each night, and contaats and faults should be inked as soon as their positions are as well
known as they are likely to be. Because of the dark gray tones of
many photographs, colored .inks may be used to advantage to show
up various kinds of structural features; there may also be an advantage in using a colored ink for note numbers. Waterproof colored
inks that are stains rather than pigment-bearing fluids do not show up
weli on dark areas of photographs and should therefore be mixed with
a ~mall amount of postercard paint.
5-11. Transferring Geologic Features from Photographs to a Map
When areas larger than a few square miles are being mapped on
aeri:ll photographs, geologic features should be transferred to an
office map as the work progresses. This is done to check structural
continuity (or critical discontinuities) over areas too large to be seen
on oite print. It also insures that all necessary ground will be covered.
The reasons the photographs themselves cannot be assembled to give
series of triangles may be drawn between points that have been located accurately on both the photograph and the map. Features are
then transferred on the basis of proportional distances from corners,
bisectrices, or other elements of the triangles (Fig. 5-8).
Generally, the map made at this stage need not be as accurate as
the one that will be made when the field work is completed. Nonetheless, details should be added in at least a qualitative way, for it is
important to see the general grain and character of geologic features
as mapping develops them.
86
.."
d~.
~/}~
~r~.;:?t'~J
/
,.
Fig. 5-8. Using a system of triangles and accessory lines to transfer geologic features from a photograph (right) to a map.
87
'
References Cited
American Society of Photogrammetry, 1952, Manual of photogrammetry, 2nd ed.:
Washington, D.C., 876 pp.
---, 1960, Manual of photographic interpretatiOn: Washington, D.C., 868 pp.
Eardley, A. J., 1942, Aerial photographs: their use and interpretation: New York,
Harper and Brothers, 203 pp.
Lueder, D. R., 1959, Aerial photographic interpretation; principles and applications: New York, McGraw-Hill Book Co., 462 pp.
Moffitt, F. H., 1959, Photogrammetry: Scranton, Pa., International Textbook Co.,
455 pp.
Ray,'R. G., 1960, Aerial photographs in geologic interpretation and mapping: U.S.
Geological Survey Professional Paper 373, 230 pp.
Smith, H. T. U., 1943, Aerial photographs and their applications: New York,
D. Appleton-Century Co., 372 pp.
Swanson, L. W., 1949, Photogrammetry (Part 2 of the Topographic Manual):
U.S. Coast and Geodetic Survey, Special Publication 249, 520 pp.
'
89
A
Line of sight
A3
_t_
AI
Az
(A) The peep-sight alidade and traverse board. (B) Map mounted on
traverse board, showing points located with a peep-sight alidade.
Fig. 6-1.
90
~
A
m
C
BC=AClanLm
The two points A and B mark two intervisible points that have been
located on the map and can be occupied with the traverse board and
tripod. The board is set up over A and oriented by sighting on B
with the alidade, as described above. The problem is to map a nearby
contact, which has been marked by bright cloth strips at a number
of points. To do this, the alidade is sighted on each cloth marker in
turn, a pencil ray is drawn toward each, and the rays are carefully
numbered in a clockwise direction ( A1, A2, etc. ) . The board is then
moved to B, oriented by sighting on A, and each of the flagged points
is again sighted. As each pencil ray is drawn from B, the intersection
with the corresponding ray from A marks the position of that point
on the map ( 1, 2, etc.). With the contact located at these points, its
details can be filled in quickly by removing the board from the tripod
and walking along the contact. Nor is it necessary for all features to
be flagged before they can be located by intersection; any points that
are distinctive enough to be identified unmistakably from two instrument positions can be located by this method. It must be noted,
however, that the locations will be unreliable if the angles of intersection are less than 20 or 30.
Triangulation from a base line. A base map is typically used for
mapping geologic features with the peep-sight alidade; however, it
is also possible to construct a map from a blank sheet of paper with
this instrument. Such a map can be started by measuring the distance
between two points' on the ground and plotting this distance on the
paper, using the scale chosen for the survey. In Fig. 6-1B, for example, points A and B might have been plotted in this way. This plotted line
is called a base line, and the survey is extended by measurements from
its two ends. In the figure, instrument stations such as C and D can
be located by intersection from A and B, and checked by sights from
C or D. This general procedure of locating stations, called triangulation, is a basic method of starting a map; it will be considered more
thoroughly in Chapter 7. Elevations can be carried to the points by
measuring a vertical angle with the Brunton clinometer and computing the difference in elevation from this angle and the distance scaled
91
from the map (Fig. 6-2). The map is completed by moving the board
to the various intersected stations and mapping the geologic features
around them by either method described heretofore.
6-3. The Telescopic Alidade
20
(I.
19
I
~
10
Fig. 6-3.
ing the tangent screw ( 7). The striding level ( 8) lies loosely on bearing surfaces that bring its axis exactly parallel to the line of sight of
the telescope. The striding level is readily removed by either a screw
or spring clamp, and it is left mounted on the telescope only during
use.
The blade has a straight beveled edge ( 10), called the fiducial edge,
which is parallel to the sighting plane of the telescope. The bull's eye
level ( 11) (or, in some models, two tube levels set at right angles to
each other) is used to level the plane table when it is being set up.
The two knobs ( 12) provide a means of moving the alidade on the
plane table without touching the telescope or other moving parts.
The compass box ( 13) lies parallel to the fiducial edge, so the instrument can be aligned to magnetic north with the compass needle. The
needle must be lifted off its jewel bearing by the lifting lever (14)
when the compass is not in use.
The telescope may be fitted with either an ordinary eyepiece or a
prism eyepiece ( 15), which turns the line of sight up at a right angle
and permits the instrumentman to look downward to sight through
the telescope. Cross hairs and stadia hairs are mounted just in front
of the eyepiece ( 16), and they appear as in Fig. ~ when they have
been properly focused by turning the knurled ring ( 17) of the eyepiece. The telescope is focused by turning a knob on its right side
(this is hidden in the drawing). A sunshade ( 19) may be slipped
over the end of the telescope, and this is replaced by a protective lens
cap ( 20) when the instrument is not in use. When the knurled retaining ring ( 21) is released, the telescope can be rotated on its sighting axis through 180; during ordinary use the telescope should be
turned firmly against its stop (so that the focusing knob is on the
right side) and then secured by the ring.
On the left side of the alidade is a vertical angle arc and accessories
used for reading vertical angles and stadia correction factors. It inchides the vernier level ( 22), a frame ( 23), a tangent screw ( 24)
that moves the vernier scale and index lines ( 25)' and a calibrated arc
( 26) used in stadia and vertical angle measurements. The arc is
calibrated with both a vertical angle scale and, in most alidades, the
useful Beaman arc scale. Some Beaman arcs (as in the figure) are
read from the side, some from the rear of the instrument, and others
through an optical eyepiece.
Each alidade has a strong wooden carrying case in which it is kept
during transport and in which it should be placed when not in use.
The alidade shown in the figure is commonly called a high-standard
or topographic model. The explorer's ali dade is 2 to 3 in. shorter than
the model illustrated, and its stand is attached directly to the blade,
making it a low and compact instrument. Its carrying case measures
about 3 X 3 X 10 in. as compared to 3 X 8 X 12 in. for the standard
model, and this makes it handier to transport. Its telescope, however,
can be elevated and depressed only about 8 from level, and its blade
may be too short for some sights. Otherwise its construction and
operation are ip every way the same as for the standard alidade. A
third type of ali dade has a blade 20 to 30 in. long and is used where
long rays must be drawn for control surveys or for exacting large-scale
mapping on large plane table boards. The self-indexing alidade is an
instrument that automatically adjusts for minor tilting of the plane
table and th,at has optical devices to increase the accuracy and speed
of readings. Other alidades developed recently have rapid-reading
optical devic~s and are more compact than the explorer's alidade.
Detailed des,criptions of these new instruments can be obtained readily
from manufacturers. . In principle, they do not differ from ordinary
alidades.
92
Stadia
'"''"''
~
\
- -"
Fig. 6-4.
,--
Stadia hair
l- """""' "~ . ,
; -Stadia hair
93
94
95
when the telescope is tilted slowly up and down with the tangent
screw. If it does not, it is necessary to loosen the four capstan screws
of the cross hair mount and rotate them by very light taps in the
desired direction. After checking and resetting as necessary, the
following adjustment should be made.
Line of collimation. The line of collimation is the axial line of the
telescope. The cross hairs must coincide with this line. Their coincidence can be tested by ( 1 ) placing the alidade on a smooth solid
surface, ( 2) releasing the knurled telescope ring (number 21 in Fig.
6-3), ( 3) 'sighting on a well-defined point, ( 4) adjusting for the parallax effect, and ( 5) rotating the telescope through 180. If the cross
hairs move off the point, they should be brought halfway back by
screwing the appropriate pair of capstan screws on the cross-hair
mount. They should then be tested again and readjusted until the
cross hairs remain on the point.
6-6. Care of the Alidade in the Field
l
tt
96
The plane table is a well-made drawing board that has a brass base
plate by which it can be attached to the tripod. Eight screw fasteners
are used to fasten a plane table sheet to its upper surface. Most
boards measure 18 X 24 in. or 24 X 31 in. The smaller is preferred
for most work because it is transported easily, but the larger is useful
in surveying control networks or in very large-scale mapping, where
there is an advantage in fitting a large area on one plane table sheet.
The wood used in the boards should be both light and strong, and it
must be seasoned and joined so that the board will not warp. Occasionally, however, new boards warp appreciably when the weather
is hot and dry. Most boards will shrink so much in a dry climate
that open spaces will form at each join. This need cause no particular
concern because a plane table sheet is strong enough to lie rigidly
over the gaps.
Most tripods used with plane tables have an ingenious head, called
the Johnson head, which permits the board to be tilted into a level
position. The board can be partially tightened in this position and
then rotated into suitable orientation, where it can be completely
tightened. Figure 6-5 shows a diagrammatic view of this head. The
board is screwed down over the plane table screw ( 1) until it is
seated firmly against the head's upper surface ( 2). When both screws
( 3 and 4) are loosened (by the usual counterclockwise turn), the head
moves freely on both its central axis ( 5) and its cupped bearing surfaces ( 6). When the upper screw ( 3) is set tightly, the cupped
bearing is locked, and the board moves only on the central axis.
When the lower screw is set tightly, the central axis is also locked,
and the board cannot be moved.
There are two kinds of tripods, those with rigid split legs and those
97
98
punch, and then screw the sheet to the board loosely, allowing freedom for shrinkage or expansion.
3. Using a sharp 9H pencil and a steel straightedge, draw two lines
across the sheet, one parallel to each set of edges.
4. Make two fine needle holes on each line, 12 to 18 in. apart; scale
these distances carefully and record their lengths near the border of
the sheet.
5. Set the board and sheet out to season under conditions as similar
as possible to those of the area to be mapped, and check the lengths
of the lines (using the same scale) to determine when the paper has
reached a reasonably consistent regimen of daily change. A record of
the progress of change should be kept, and the final lengths of the
lines recorded permanently near the edge of the sheet so that the map
scale can be corrected accurately at any time.
grinding the enamel smooth with a very fine abrasive. The enamel
should be colored because the glare from a white or metallic sheet
makes surveying difficult.
Seasoning the sheets in this way will usually take a few days. It is
important that the thumb screws are not set tightly during the seasoning, and that they are loosened each night during the period of actual
mapping; otherwise the map may be strained and distorted.
To keep the sheet clean, a piece of heavy wrapping paper may be
fastened over it before field work is begun and pieces can then be
cut out as the work progresses. The paper cover will help protect the
sheet from moisture, but as soon as rain begins to fall, the canvas case
for the board should be laid over the sheet because dampened sheets
buckle severely. If showers are expected, a piece of plastic or oilcloth
should be carried to cover the board.
99
~bearing
IW'\
level
Map position
~ofpoint
occupied
_..,
~
Fig. 6-6.
100
4. Adjust the position or length of the tripod legs until the board
looks level and the legs are inclined at least 30 to the vertical. Two
legs should be downhill and one uphill.
5. Force the tripod legs into the ground with the heel, or brace
them with stones until they are set firmly. Tighten the wing nuts on
each tripod leg.
6. Place the alidade (without striding level) in the center of the
board and parallel to its length.
7. Place the left arm over the blade, grip the far edge of the board
with the left hand, and press the board against the waist as the left
arm pins the blade to the table (Fig. 6-6C). With the alidade thus
held safely, loosen the screws of the Johnson head with the right hand
and carefully tilt the board until the blade level bubble is centered.
Tighten the upper Johnson head screw firmly.
If the plane table has been set up at the first station of a new survey,
with no points marked on the sheet, it must first be oriented suitably
relative to the area that will be covered. After it has been clamped
in this position, the initial station is marked on it and a magnetic
north arrow is drawn on the sheet by using the magnetic needle of the
alidade. The arrow should be labeled with the station number and
the date, as magnetic declinations at other stations may be somewhat
different.
If the map has already been started, so that two or more stations
are plotted on the sheet, the board can be oriented by using the magnetic arrow drawn at the nearest station. If this method is used, the
reliability of the magnetic north arrow should be considered. In
general, it is preferable to orient the board by sighting on another
point, as by the instructions that follow.
1. Place the alidade so that the fiducial edge just bisects the pinpoint that marks the station occupied and also the pinpoint of another
well-located station whose signal (generally a flag on a vertical pole)
can be seen. The points on the board should be at least 4 in. apart.
A reading glass or hand lens should be used to check this setting.
2. Turn the board slowly and carefully until the signal is bisected
by the vertical cross hair. Focus and refocus the telescope to make
sure that parallax has not affected the setting.
3. Tighten the lower screw of the Johnson head while watching the
signal through the telescope. Give the board a budge or two to make
sure it will not be disoriented readily.
101
1. With the plane table stlt up and oriented as described in the last
section, measure the vertical distance from the station marker to the
horizontal axis of the alidade, using a tape or stadia rod. Record
this as the H.I. (height of instrument) in the notebook (Fig. '6-7).
If a signal flag or cairn marks the point sighted, record its height
( H.F.) above the ground at the station sighted. These values must
be used when computing the difference in elevation.
2. Regardless of where the station occupied and the station sighted
occur on the sheet, place the alidade near the center of the board to
improve the stability of the setup.
3. Loosen the axis clamp screw (no. 6 in Fig. 6-3) and bring the
signal into view in the eyepiece by moving the telescope manually.
Tighten the' axis clamp screw and use the tangent screw (no. 7 in
Fig. 6-3) to, brihg the horizontal cross hair (not a stadia hair) onto
the top of the station signal. The final movement of the tangent
'
....
Sfqf/on
Stat/on
OCCUJ:)/ed
s/ahfed
cc
H./.= 4.1
H. F.= 6;o
Level
read/nas
'Readin9 fo
slana/
Verfica/
anales
+3 02'
zo"
Fig. 6-7. Notebook page for vertical angle measurements, showing both the entries
and the computations of angles averaged from two sets of readings.
102
~
I
_.L
103
AC = ABcos Lm
BC = ABsinLm
rod is held, the amount of the stadia rod included between the hairs
-called the stadia. intercept-is multiplied by 100. Stadia rods are
commonly marked so the stadia intercept can be re'ad directly to the
nearest tenth of a foot, while the nearest hundredth of a foot can be
estimated accurately if the rod is less than 300 ft away (this varies
with the instrument, the rod, and the lighting). These readings
therefore give a distance that is within a foot or two of being correct.
Except for level sights, the distance from alidade to rod is a slope
distance, which must be converted to a horizontal distance before
. being pl6tted on a map. Moreover, if a contour map is being made,
it is also necessary to determine the difference in elevation between
the instrument and the point where the rod is held. These corrections can be made by measuring the vertical angle to the rod and then
computing both the horizontal and vertical distances trigonometrically
(Fig. 6-8). The second computation, however, gives the difference
in elevation from the axis of the instrument to the point where the
cross hair cuts the rod. This difference must be corrected so as to
obtain the difference in elevation between the point occupied and
~
Distance
computed
from stadia
Rod
Cross-hair
reading
Distance sought
H. I.
Most of the horizontal and vertical distances that must be determined to make a map with an alidade are measured by the stadia
method. The principle of this method is simple. The two stadia
hairs of the eyepiece are mounted just such a distance apart that they
intercept 1 ft on a graduated stadia rod held 100 ft away. Therefore,
to find the distance in feet from the alidade to the point where the
H. I.
Cross-hair
reading
Fig. 6-9. Correcting vertical distances for the height of instrument ( H.l.) and
the cross hair reading.
the point where the rod is held (Fig. 6-9). This is done by giving
the H.I. a + sign, the cross-hair reading a - sign, and adding them
algebraically to the difference first computed ( BC in Fig. 6-8). If,
for example, the H.I. is 4.1 ft, the cross-hair reading is 8.3 ft, and the
difference in elevation computed from the vertical angle is -34.5 ft,
the difference in elevation between the station occupied and the point
on which the rod is held is -34.5 + 4.1 - 8.3, or -38.7 ft.
Stadia reduction tables or a stadia slide rule can also be used to
convert the stadia slope distance to a horizontal distance and a difference in elevation; however, these values can be obtained more quickly
with a Beaman arc. As shown in Fig. 6-10, the Beaman arc consists
of two graduated scales scribed on the same arc as the vertical angle
scale but read by a separate index. The arc shown in the figure is
an edge-mounted arc, read from the rear of the instrument, while the
arc shown in Fig. 6-3 is a side-mounted arc. The scale graduations
are the same on each and are spaced to give proportionately increasing
correction factors as the telescope is tilted to steeper and steeper
sights. To correct the stadia distance to a horizontal (map) distance,
the horizontal or H scale is read (although the reading is made by
visually projecting the index across the V scale, this can be done precisely enough for the measurements needed). The H reading is then
multiplied by the st-adia intercept, and the product is subtracted from
the stadia distance. If, for example, the H reading is 2 and the stadia
intercept has been determined to be 2.50, the horizontal correction
is 2 X 2.50 or 5 ft, and the correct horizontal distance is 250 - 5 or
245ft.With regard' to the vertical correction, it must be noted that the V
scale reads 50 when the telescope is level (Fig. 6-10). This is to
eliminate errors caused by transposing signs when sights are close to
level. The number 50 is subtracted from the V reading to obtain the
vertical correction factor, which will therefore have a + sign for
inclined sights and a - sign for depressed sights. The sense of sign
must be carried throughout the computations. The vertical correction, factor is multiplied by the stadia intercept to give the difference
in elevation between the instrument and the point where the cross
hair inte:rsects the rod. If, for example, the V reading is 36, the correction factor would be 36 - 50 or -14; if the stadia intercept were
2.50, this would give a vertical correction of -14 X 2.50 or -35 ft.
This number must then be modified to give the difference in elevation between the instrument station and the point where the rod is
held on the ground, as described earlier (Fig. 6-9). The resulting
number is added algebraically to the elevation of the instrument
slation to give the elevation at the point where the rod is held.
J04
Beamanr--------.---,
a~
2e--=::
1~
-
Vernier for
reading
vertical
angles
5-
(}-
51~
-
2e-= 10 I
105
106
IJ39.
0
20
0
so
0
5.0
0
.3ZI
202
189
+6.4
-100.6
0
+-2
-40
0
-3.2
-4.9
-;o.3
+3.2
107
Remarks
'
I
.
...
Fig. 6-11. Two (facing) pages of notebook, set up for stadia surveying. The column headings can be abbreviated much more than those shown.
A~'
------
Fig. 6-12.
1 - -...
(A) Aligning the alidade on the rod by sighting over the telescope.
lOB
Computations should be carried out point by point as instrumentation progresses. Where points are closely spaced and geologic features
fairly simple, the instrumeptman will need a recorder to keep up with
the rodman. The instrumentman should glance up occasionally while
computations are in progress to see where the rodman is; otherwise
it may- be difficult to find him when he is at a new point.
109
llO
Polnf
2)
0.43
0.23
0. 20
12
I) :.._]_
5
L__L_
fig. 6-13.
---
---
Stadia
distance
5
where lines of sight are inclined at low to moderate angles. The correspondence of this method to the Beaman arc method can be under,stood by noting that the Beaman index moves across one full division
on the V scale when the telescope is moved through one full stadia
intercept on the rod. The reader can demonstrate for himself that the
telescope can be moved through a stadia intercept more accurately
by observing the stadia hairs than by observing the V scale.
The method may be carried out as follows:
!fon'z.
II
rctl'ng_ d/sf
Jll
2475
'
The vertical angle arcs of alidades can be read only to the nearest
1 or Yz minute. It may be desirable to use more precise methods to
measure differences in elevation, and the stepping method is useful
ll2
~
Control for Geologic Maps
J J5
geodetic data. Control data may be obtained from the following agencies (regional-offices should be contacted if their addresses are known) :
114
Regardless of whether a control survey will be made by triangulation or traverse methods, a good deal of time can be saved if it is
based on existing survey data rather than on a new base line. In
many areas, for example, there are two or more established triangulation stations that can be plotted on a plane table sheet and then used
to intersect any number of other points. Search for control data must
be started long before the fielq season begins since it may take several months to sound out all sources and to compute distances from
ll6
The base line and the triangulation stations can be selected by first
making a stereoscopic study of aerial photographs and then reconnoitering the possibilities in the field. The optimum spacing of control
stations depends in part on the scale and nature of the final map. If
the map will be compiled from aerial photographs or from several
plane table sheets, there should be at least two or three stations on
each sheet or each photograph. If map or photograph scales are between 1: 12,000 and 1: 24,000, the sides of the triangles will therefore
be about 1 to 3 mi long.
The size and arrangement of the triangles and the base line also
affect the precision that can be obtained in a given amount of time.
The probability for errors increases as measurement:; are carried
farther and farther from the base line. The examples that follow illustrate four general arrangements. Networks like that shown in Fig.
7-1A are used where the base line must be short and visibility is limited
Fig. 7-1.
base line.
lll
JlB
After the probable station sites have been selected on aerial photographs or by reconnaissance with field glasses, each site must be
visited, checked for suitability, and marked with a signal. This selection must be made with care because the stations should not be moved
once instrumentation starts. Signals of adjoining stations must be
visible at instrument height. The signal should be as completely in
the open as possible so that accessory instrument stations can be
located from it easily. There must be ample space for the instrumentman to move safely around the instrument, especially if the station will
be occupied with a plane table.
Sketch map of network. As the stations are selected and marked in
the field, a sketch map should be made of the network by reading
angles to adjoining stations with a compass. This map is valuable for
estimating the strength of the network and for planning and executing
the instrumentation. Notes regarding visibility and routes of access
should be made directly on it; it may also be used to make a reconnaissance map of the main geologic features.
be plumbed exactly over the station hub, and can be taken down and
set up ,easily during instrumentation. The Hag color is chosen to
contrast with background colors. In most areas, a brightly colored
Hag on a pole 6 to 8 ft high can be seen with an instrument for distances
of at lea'st 5 mi. There will be an advantage in later computations
if all poles are the same height. In addition to the materials shown in
Fig. 7-2, the following will be needed: a small axe or hatchet, pliers
with wire cutters, a Brunton compass for plumbing the pole, materials
, for making a sketch
- map, and, if possible, a pair of field glasses
for observing other stations.
'
Flag
Flag edge
rolled and
tacked
Ir
Baling wire
J
~res
Heavy
stones
~Wire
attached
to stick
Fig. 7-2.
119
I
i
120
Fig. 7-3.
by sighting over him to far station signal. Recorder stands near head
tapeman and, using hand level, instructs him to raise or lower tape
within a few inches of level.
3. Rear tapeman holds 100-foot mark or other whole foot mark at
the station point, while head tapeman holds plumb bob string against
0 mark of tape, with plumb bob point just above ground (Figure 7-3).
4. Recorder clears ground around point of plumb bob, and head
tapeman stills plumb bob by dipping it gently to the ground.
5. Rear tapeman calls out the whole foot he is holding, and recorder drives in a taping pin at the point of the plumb bob.
6. After the distance of the measure is recorded, the last step is
repeated as a check.
7. Rear tapeman examines marks on both sides of mark used to be
sure he read it correctly, then he records the distance.
8. The tape is moved ahead, and when the next measure is
pleted, the rear tapeman pulls out the pin and carries it with him
an additional tally of the number of measures taken.
co,..
121
sag correction
p2
= w2l8
24
where w is the weight of the tape in pounds per foot, l is the distance
between supports, and P is the pull applied in pounds.
Taping on slope. If maximum precision is needed, as when a short
base line will be expanded into a large quadrilateral, the line should
be taped by setting heavy stakes on slope at each 100-foot mark, taping each measure on slope, and then correcting each measure to a
horizontal distance.
In order to obtain suitable precision, the tape should be sent to the
U.S. Bureau of Standards at Washington, D.C., where it can be standardized for a moderate fee. The bureau should be instructed to
determine its length when fully supported and also when supported at
both ends, using a 10-pound pull. The tape will be marked as standardized only if the graduations are etched or engraved in the steel
ribbon, not on separate soldered patches.
Gullied or boulder-strewn ground should be avoided because the
exact position of each stake cannot be determined at the outset. The
stakes, preferably of 2 X 4 in. lumber, should be set by placing a
transit over one end of the base line and lining in a taping party to
each 100-foot mark. This is done so that an exact 100-foot measure
falls somewhere on each stake. The exact measures are then taken by
a taping party, and they should be scratched precisely on tin can strips
nailed to the tops of the stakes. A transit or engineer's level and
Philadelphia rod can be used to determine the difference in elevation
between each pair of stakes. Each slope measure (the standardized
122
length with tape supported at two ends) is then corrected for temperature, as described for taping with tape held level. The exact horizontal
distance of each measure is found by the relation
horizontal distance
= corrected
tape length -
2~
Triangulation with the transit consists of occupying the control stations and measuring the various interior angles of the network as
well as the vertical angles to adjacent stations. To do this efficiently,
the network should first be studied and the angles listed in the notebook in the order they will be measured. Generally, only the angles
between adjacent stations are measured; otherwise the computations
become too involved. All angles within quadrilaterals are measured,
however, because both the triangles and the quadrilateral afford checks
of the sums of angles in each figure. Wherever there is a choice between lines of sight to outlying stations, as in the system of Fig. 7-1C,
the sights used are those that intersect most nearly to 90 at that station.
The work can be done by one man, but a second man can help carry
gear and assist at the instrument. In addition to the transit and the
plumb bob, the equipment must include a reading glass (2X to 3X ),
notebook and pencil, and tape for measuring height of instrument; a
pair of field glasses is useful for finding signals.
The height of the instrument and the height of the flag should be
recorded before the measurement is made. The instrument is set up
firmly and leveled with care, for if the plates are not level, steep sights
to distant stations will be appreciably in error. The various horizontal
angles should be measured by repetition, as described in surveying
texts. A reading should also be taken on magnetic north at each station to check the conformity of the magnetic declination within the
area; this data will be very useful in later mapping (Sections 8-11 and
8-12).
Measuring vertical angles with the transit. Vertical angles can be
measured more precisely with the transit than with the alidade, but
the procedures are essentially the same (see Section 6-10). The angle
is always found as the difference between the vernier reading to the
station (the top of the flag) and the reading with the telescope leveled.
The + or - sense of the small correction from 0 must be noted with
care. To eliminate errors caused by a poorly adjusted telescope
123
level, one reading is taken in the normal position and one with the
telescope reversed. The cross hairs and objective must be focused
carefully to eliminate parallax. It cannot be emphasized enough that
,t vertical angle readings must be free of mistakes, because the method
can introduce errors that are in themselves appreciable. Observations
should be made when the air temperature gradient between the in. sttument and signal is as small as possible, as on cloudy, windless days.
Refraction by heterogeneous air masses can be compensated for precisely by observing vertical angles simultaneously in both directions
with two instruments. lf this cannot be done, the effects of irregular
11efraction can be compensated for approximately by making a reading
in the reverse direction when conditions are the same as when the
\first reading was made.
For relatively low-angle sights, the stepping method (Section 6-15)
may give greater precision than a vertical angle measurement.
7-8. Surveying an Elevation to the Control Network
'
124
_ ~~-- _c~s~a~~~")~_P~~ __
Path of Polaris---~
1.
\
...,._
r_
':..
"'
Cassiopeia
~~'b\\'>
Polaris
.f.
-1(, 'y'\r:.~<:j
'-.._
125
7. Next morning, set the transit up over the same point, align it on
the midpoint between the two nails, and measure the horizontal angle
to an adjacent control signal.
8. Determine the bearing of the line of sight to the nails by using
t;he tables in the ephemeris.
Apparent
-.........._~
Telescopic view
. 7-10.
,
Manual of Field Geology
J26
127
LJ ABC
A<==]c
AC = 9751.5
L8AC =19.30'10"
L.BCA =77"1!'10"
9751.5
L.ABC = 83 18'40"
log
AC = 3. 989 0581
= 9. 989 0472-10
- !oq sin
3. 978 1053
ABC
log AB
L.
9. 997 0338-10
= 3.981 0715
AB =
9573.5
Fig. 7-5.
using the notation shown for the triangle at the top of Fig. 7-f>, the
unknown sides are solved by the following equations:
AB = AC sin LACB
sin LABC
and
BC
= AC
sin L BAG
sin LABC.
= MB 2
+ NB
Fig. 7-6.
N
D
128
129
"Level" sight
'
Deferm of e/ev. af
c,
A~
us/n1 L1 ABC
15633.8
Elev. af A = 2132.4
flortz. 0/st Ac = 9751.5
Vert.
+ Loq
L.
A to c
= 3 11'30 ''
to
7-8. Relations resulting from curvature and refraction of lines of sight between
two stations. The vertical scale of the drawing is exaggerated greatly, but the
n1tmbers are approximately representative.
- 3 IO' oo"
= 3. 989 0581
C-A = 8.742 9222-!0n
LO'J = 2. 731 9803 n
'
Anfiloq = -539.5
+ H.I.-H.F. = -1.5
+ H. I. -H. F. = -2. I
= 540.8
AC
rLoq fan L
L.
Loq
~ Fig.
C to A =
Vert.
3l6 miles
= /631.3
I
J/ert .c. c to 8 = 4 .32 '40"
I
(Etc. as above)
E/ev. crf C = E/ev. deferm. from A + E/ev. deferm from 8
2
-----
--
--
,~
Bearinqs: L1 ABC
Computation of vertical angles.
of negative numbers.
Fig. 7-7.
'
algebraically to the difference in elevation as measured by a given
sight between two stations (Fig. 7-8). Study of the figure will show
that if sights are made in both directions, they can be averaged directly to give the correct result. The corrections are generally made,
however, to detect errors made in one sight or the other.
Computations of bearings. Bearings of triangle sides are computed
before coordinates for control stations. This is done as shown in Fig.
7-9, starting with the side used for the observation of Polaris and proceeding systematically through the network, one triangle at a time.
The computation is made counterclockwise around each triangle, and
the final back-bearing serves to check the arithmetic.
c
Brq. AB
=S 79"09'30"
Br9.
=N
77 II' !0"
I" 48' 20" W
CA
- .L A C 8
Brq. C8
=N
{ Efc.)
Fig.
7.JJ.
Computation of bearings.
L1
Side
AB
130
--- --l
Ocr fa
Oeparfure Comp-
Oisfcmce=
Coords. A=
Br~
1000, !000
Coords. of' 8
Coast and Geodetic Survey, which can be obtained from the Director,
U.S. Coast and Geodetic Survey, Washington 25, D.C.).
!..afi!ude comp.
/o~
= 1000
+- Laf.,
;ooo +- oepf
--
Fig. 7-10.
Computation of coordinates.
131
1. Draw a vertical line near the center of the sheet and construct a
perpendicular atE by first striking arcs EA and EB, then arcs AC, BC,
AD, and BD.
2. Using the scale, set the beam compass exactly at a whole number
of grid intervals and strike arcs 1, 2, 3, 4, 5, and 6 from points C, E,
'and D.
3. Lay the straightedge next to these arcs and draw east-west lines
that are tangent to them.
4. Now set the beam compass for some whole number of grid intervals (for example, to give the approximate proportions shown in
c
2
Fig. 7-11.
line, it must be plotted on the sheet by means of coordinate lines ( Sections 7-11 and 8-5). The notebook should be set up ahead of time
and the order of sights from each station listed.
Starting with the base line, each station is occupied and rays are
drawn to adjacent stations and to those outlying stations that would
otherwise not be intersected with three pencil rays. The pencil lines
shou~, be made the full length of the fiducial edge so that the board
can be oriented accurately by backsights. Each pencil ray must be
labeled clearly. VerticaJ angles can be measured (Section 6-10), or
the stepping method can be used ( Section 6-15). Special attention
must be given to the level and stability of the plane table; if the wind
causesh to Hutter, for example, the instrumentman must wait for a
lull. If the telescope is the type that can be released and rotated in
an axial sense, as the one shown in Fig. 6-3, the vertical angle may
be repeated with the telescope in reversed position. An average of
the two readings will compensate for incorrect alignment of the cross
hairs with the line of collimation.
Precision. Distances can be scaled along thin pencil lines with considerable precision, for a needle hole need be no more than 0.003
in. across (at a scale of 1 in. = 1,000 ft, this is 3 ft on the ground).
The pencil ray itself, however, cannot be drawn along the fiducial
with tnore than half this precision. The suggestions for drawing
straight lines given in Section 7-11 may be helpful in this regard.
Plotting errors can, with care, be held to less than 0.01 in. Because
the greatest probable plotting error is at right angles to a pencil ray, it
is essential in plane table triangulation that intersection angles be
greater than 30.
132
The scale of the sheet must remain reasonably constant during the
plotting, and the following instructions will help in this regard: ( 1)
season all materials in advance, ( 2) use one master scale or meter bar
only, ( 3) keep materials at the same temperature and humidity, and
( 4) work carefully but steadily from the beginning to the end of the
construction.
Control points can be plotted on the grid by the procedure given
in Section 8-5.
Polyconic grids. Polyconic grids are customarily used as a base for
geodetic coordinate data, especially where a map covers a large area.
They are constructed in the same general way as rectangular grids,
except that tables are needed to determine the linear distances subtended by various arcs of latitude and longitude. The tables of U.S.
Geological Survey Bulletin 809 (Birdseye, 1929) are used to make the
construction in feet and inches, whereas those in U.S. Coast and Geodetic Survey Special Publication 5 ( 1935) are used to make it in meters.
Instructions accompany each of these sets of tables.
7-12. Triangulation with the Alidade and Plane Table
The alidade and plane table are less suited for triangulation surveys
than the transit; however, wind and weather allowing, plane table work
is faster and is precise enough for many projects. Horizontal locations
are made by intersecting lines of sight on the plane table sheet, while
differences in elevation are based on vertical angle measurements.
Top-quality waterproof plane table sheets should be used; if linenfilled sheets must be used, they should be seasoned thoroughly (Section 6-8). The base line should be plotted on the sheet with a steel
straightedge and the end points pricked with small needle holes. If
geodetic or other established control will be used instead of a base
133
Where trees or other obstructions make it impossible to locate stations by triangulation methods, tape and transit traverses can be used
to establish control. The traverses must be run in closed circuits or
must be closed on points that can be located at least as precisely as
the points of the traverse. Because of the precision required, the
courses should be reconnoitered so they can be directed along open,
evenly sloping stretches. Distances may be taped with the tape held
level, and a tape 200 or 300 ft long can be used efficiently in country
with little relief.
The first leg of the traverse should be turned from a line of known
bearing or azimuth. If it must be started from an isolated triangulation
134
station, the bearing of the first leg can be determined by an observation on Polaris. This accurate bearing will not be essential until the
traverse data are computed to coordinates; however, it is useful to
compute bearings of lines as the traverse proceeds. Vertical control
should be run at the same time by reading vertical angles or by stepping vertical distances between stations.
Traverses that close on their starting points may be made efficiently
by measuring the interior angles of the circuits, while traverses that
start at one known point and end at another are best run as deflectionangle traverses. The procedures for both types of traverses are described fully in most surveying texts (as Bouchard and Moffitt, 1959,
ch. 8). They can be executed by two men, but at least one other man
is helpful for taping and recording. Computations, which should be
kept up-to-date in camp, will generally include checks on bearings
computed in the field, correction of taped distances ( Section 7-6),
computations of latitudes and departures, computations of plane coordinates of stations, and computations of differences in elevation
from vertical angles and corrected distances. For clos!'ld traverses, it
will be helpful in the computations of latitudes and departures if any
small closing error in the bearings is first distributed equally among the
angles of the traverse. The latitudes and departures should then
close evenly.
References Cited
Birdseye, C. H., 1929, Formulas and tables for the construction of polyconic projf?Ctions: U.S. Geological Survey, Bulletin 809, 126 pp.
Bouchard, Harry, and F. H. Moffitt, 1959, Surveying: Scranton, Pa., International
4
Textbook Co., 664 pp.
Gannett, S. S., 1918, Geographic tables and formulas: U.S. Geological Survey, BUlletin 650, 424 pp.
Hodgson, R. A., 1957, Precision altimeter survey procedures: Los Angeles, American Paulin System, 59 pp.
Reynolds, W. F., 1928, Manual of triangulation computation and adjustment: U.S.
Coast and Geodetic Survey, Special Publication 138, 242 pp.
.
U.S. Coast and Geodetic Survey, 1935, Tables for a polyconic projection of rn.cqJs
and lengths of terrestrial arcs of meridian and parallels based upon Clark's reference spheroid of 1896: 6th ed., Special Publication 8, 101 pp.
---, 1933, Formulas and tables for the computation of geodetic positions: 7th ed.,
Special Publication 8, 101 pp.
~
Geologic Mapping with the
Ali~ade and Plane Table
11'
lfhe alidade and plane'table are esssential for most projects requiring
an accurate geologic and topographic map with a scale larger than
lf:6,000 (1 in. = 500 ft). Few topographic base maps with such
scales exist, and large-scale aerial photographs do not give the exact
vertical control needed for many structural and quantitative studies.
Accurate topographic and geologic maps are especially valuable in
resolving such problems as offsets on faults, continuity of beds or ore
bodies in complex folds, stratigraphic details in intertonguing sequences, and relations of various alterations to faults or intrusive
bodies. Maps used in these studies typically cover 1 to 25 sq mi at
scales of 1 in. = 200 ft to 1 in. = 500 ft, and the procedures described
in this chapter are appropriate for such projects. Mapping at scales
J.arger than 1 in. = 200 ft may be necessary where measurements and
sampling must be particularly precise, as in some economic studies;
this type of project is considered in Chapter 10.
Geologic and topographic mapping at a scale of 1 in. = 1,000 ft can
be done readily by methods described in this chapter; however, a controlled aerial photograph survey should also be considered because
it takes less time than a plane table survey (Chapter 9). Complete
topographic and geologic mapping at scales smaller than 1 in. =
1,000 ft is not usually undertaken by geologists because it is more
efficient to map on aerial photographs or to have a base prepared by
an engineering or aerial survey company. It may be necessary, how. ever, to map certain limited features or structures at small scales. In
unsurveyed regions, for example, petroleum exploration may require
accurate positions and elevations on a given unit over a large area.
Planning the control for such a survey requires ingenuity and judgment, and plane table procedures must often be improvised to meet
particular needs. Methods that will be especially useful are the
gradienter screw and stepping methods (Sections 6-14 and 6-15) .
and the resection and three-point methods (Sections 8-10 and 8-11).
The suggestions given by Low ( 1952, chs. 4, 6) will also be helpful.
135
136
t'
137
The flrst step in the fleld is to reconnoiter the area in order to get
a general view of the geologic features, plan the scale of the work,
and lay out the control network or plan the control traverses. Aspects
of planning control surveys and selecting stations are given in Chapter 7. A compass sketch map of the control system should be used
as a base for a geologic sketch map (Section 7-4). The scale of this
map will typically be one-fourth or one-fifth that of the final sheets,
and its accuracy need be adequate for planning only. More detailed
sketch maps and sections showing areas of special interest can be made
in field' notes. In many cases, aerial photographs make an ideal base
on which to sketch features that will later be mapped in detail bepause
s'ome structures and rock units can be recognized on them at once.
8-4. Choosing the Layout of Plane Table Sheets
After the control system has been surveyed and computed (Chapter 7), the plane table sheets are prepared for stadia mapping. The
flrst step is to determine the optimum layout of plane table sheets relative to the area to be mapped. Each sheet must ( 1) include at least
J38
z
X
YH--;{'?
1:
xe-----~---~
Yl-----
""s
_v
c--.,y
IAI
After the plane table sheets have been prepared, the primary control points can be plotted on them by using coordinates or by intersecting with a beam compass. To plot them by coordinates, the coordinate grid should first be constructed with moderate precision on
the small-scale layout described in Section 8-4. The approximate positions of two coordinate lines are then found by transferring distances
from the small-scale plot, as distances MN and MP in Fig. 8-1. Start-
s+
-J
139
Fig. 8-2.
This procedure is simple if the ground is reasonably open, but if appreciable parts of the area cannot be seen from certain nearby control
stations, the sheets must be arranged to cover these blind spots. The
scale of the work and the minimum spacing for stadia instrument stations should be considered ( Section 8-6).
Nl
p
Fig. 8-1. Transferring coordinate grid from small-scale sheet (left) to plane table
sheet (right) and locating a control point (A).
I'
..
The primary control points cannot be spaced closely enough to provide instrument stations for all the stadia mapping, and additional
stations must therefore be located. How this will be done depends on
the scale of the project and the nature of the geologic features and
I
140
terrain. If the scale is large, the geologic features complex, and the
terrain reasonably accessible, instrument stations should be selected
during a preliminary and rather detailed examination of the entire
area, as described in Section 8-7. The new stations are marked with
signals and surveyed by intersection from primary stations. These
instrument stations are sometimes called the secondary control because
they are located as a subsidiary network of points, prior to the stadia
mapping. Where mapping scales are rather small, geologic features
are simple, or the terrain is so rough that it is not efficient to work in -.
any part of it more than once, instrument stations should be selected
and surveyed as the mapping progresses. These stations must be
located by resection, three-point methods, and traverses (Sections 8-10,
11, and 12).
Although there should be as few instrument stations as possible, they
must be spaced closely enough so that stadia measurements will be
adequately precise. The stadia intercept can usually be estimated
to the nearest 0.02 ft at distances less than 600 ft. The precision of
readings and Beaman corrections falls off rapidly for sights of more
than 1,000 ft. Instrument stations should therefore be' spaced about
2,000 ft apart for detailed mapping. This distance may be increased
moderately where the map scale is small or where errors may be larger
than 8 or 10 ft on the ground. The limiting distance should be tested
by each party, because stadia accuracy depends not only on the distance to the rod but also on the qualities of the instrument, the rod,
and the lighting.
Instrument stations can be selected systematically and quickly in
open country of moderate relief, but obstacles often limit this ~lection.
Each station must give a clear view of the ground to be mapped and
must be visible from at least two primary stations (three prirfiary stations if three-point methods will be used). Ideal stations are in open
valleys or on the tops of low knolls from which there is a view up and
down a valley as well as to the valley sides. Gently sloping ridges
and saddles are often good locations. The tops of steep hills or ri~ges
are typically unsuitable because the instrumentman will not be able
to see their slopes.
A stake or solid rock can be used to mark each station.. Stations
that will be located by intersection must be marked with a distinctive
signal, as a cairn or a small flag and pole similar to the one described in
Section 7-5.
141
142
After the primary stations have been plotted on each plane table
sheet (Section 8-5), each primary station is occupied and rays are
drawn to all visible instrument station signals. The procedure is the
same as that for primary triangulation (Section 7-12). Stations located by three-line intersections should be pricked carefully with a
needle and labeled with ink. Computed elevations should be lettered
It
...,.
in pencil beside the station, but they must not be inked until the in- ,~
strument stations are occupied for stadia work and the vertical angle
to each primary signal is checked by a reverse reading.
Instrument stations that can be seen from only one or two primary
stations must be located or checked by accessory rays from other secondary stations. In some cases, a ray can be drawn to them by offsetting a short distance from a primary station. This is done by setting
up at the primary station, drawing a ray toward a point from which
the instrument station can be seen, measuring the distance by taping or
stadia, and plotting the new accessory station. When set up at this
new point, the plane table must be oriented carefully on the primary
point used to locate it and then checked by sighting on other points.
8-10. Locating Instrument Stations by Resection
.-----------------~
.,.p
143
Br----------~
.. p
S5
"'N
a.M
85
.. N
Fig. 8-3. Locating a new iJlstrument station ( S6) from an established one ( S5)
by resection.
...
and then other primary signals are sighted and other vertical angles
read to serve as checks. Stadia mapping may begin at 56 as soon
as its elevation is computed from the vertical angle and the horizontal
distance scaled from the map.
8-11. Three-Point Locations
When the plane table must be set up at a new station toward which
no rays have been drawn previously, it is necessary to orient the board
and intersect its position by three-point methods. These methods are
especially useful for intermediate-scale and small-scale work or where
terrain is so rough it is inefficient to set up all station signals before
the stadia mapping. The control for these projects Is typically a network of very large triangles, and therefore the primary signals should
be placed so that they can be seen over large areas.
Tracing sheet method. This quick and alr:~ost foolproof method of
three-point location requires several sheets of tracing paper and some
drafting tape. The procedure is as follows:
1. Set up the plane table at the new station, in view of at least three,
and preferably four signals that are plotted on the sheet.
2. Tape a piece of tradng paper anywhere on the sheet and prick a
small needle hole in it.
3. Tighten the Johnson head, lay the fiducial edge against the needle
hole, and sight each of the three or four signals through the telescope.
Draw rays toward each signal from the needle hole, but stop each ray
just short of the hole.
4. Remove the tape and shift the tracing paper until each of the
pencil rays passes exactly through its corresponding plotted point on
the plane table sheet (Fig. 8-4).
144
Plane table sheet
~V
edle
Tracing
paper I
_I
/~
~~~~
A'J
,.
Fig. 8-5. Relation between control stations (A, B, and C), the great circle that
passes through them, and the small triangles of error fo~ed by three-line intersections on an imperfectly oriented plane table. In each case the small circle
shows the correct position of the point sought. After Birdseye ( 1928, p. 203).
l45
estimation. Three or more points are then sighted, and rays from
each of them are drawn back along the fiducial edge. If the board is
oriented correctly, these rays will interstt.ct in a point, and the location is completed unless all the points lie on a circle, as noted above.
- If the rays form a small triangle, the orientation must be corrected,
and this is done essentially by trial and error. The following rules,
based on those given by Beaman in U.S. Geological Survey Bulletin
788 (Birdseye, 1928, p. 203) will help to determine the approximate
position of the true po~nt relative to the small triangle. The rules
and the terms used are illustrated by Fig. 8-5.
1. If the small triangle is within the great triangle, the true position of the point is within the small triangle.
2. If the small triangle falls between the great triangle and the
great circle, the true position of the point lies outside the small triangle,
opposite the side formed by the ray from the middle station sighted.
3. If the small triangle lies outside the great circle, the point sought
lies outside the small triangle and on the same side of the ray from
the most distant station sighted as does the intersection of the other
two rays.
The actual position of the point sought can be estimated by considering the distances to the three points sighted. The triangle was
caused by rotation relative to these three stations; the true position of
'the point is therefore closest to the ray from the nearest station and
farthest from the ray from the farthest station. Moreover, the distances between the true position of the point and the three rays are
proportional to the distances from the point occupied to the three
points sighted.
After a new point is marked, the alidade is placed alongside it and
also alongside the point representing_ the farthest station. The plane
table is reoriented by sighting on the farthest station, and rays are
drawn back from each station, as before. The procedure is repeated
until any remaining small triangle can be resolved into a point. With
moderate experience, this can be done in about three trials; however,
the final point should not be chosen too quickly (a needle hole is very
small).
Regardless of whether this method or the tracing paper method is
used, the reliability of the locations will depend largely on the trigonometric strength of the intersections. Points within the great triangle
. can be located most exactly, while those outside it become less and
less reliable as intersections of rays become more acute. In Fig. 8-5,
I
I
I.I
146
a:_ 13.
147
l48
7129
Inst:- af Mti
u~.
IR
N"'
</<'
7:: -~E
f-;vJ.i......__u.~d th
.~h-Jr-
<.in
;,,.,.,,
'"
'f..'ti~I!..P.
tce.a;.6
6'1
l7o.
~ '~
.;;:
'"~/'~'
67.
.,;:....n~
.~e./
~
2o
op:~;~ ~~
NE
73
I ?4
<'!:~
n.
<
,f f,.-.
IV
.A. ,-t+
nos/~.
;;..
;;.
' / ; . ,_j
Nnf.
N ol
c.-w
..
...
n
~
.t... "r.;
Jl.. ~ ~ .!
N,
w/
'
a.
.sr:!.
~,of"
':Z~
;t.r
Y'<,
PJ~r..
.
\~'
..
)-
soy~.
7Z
.t:f!-rz.
4S:
71#
:5amn/e
..
"
6l'.
ht.
wht'Ob {_ol!s:J.~&~
s,.,.
-~:
>l'l"...
. ~.-.r N5oE
'>W
Fig. 8-7. Part of a page from a rodman's notebook, showing how a large-scale
sketch map is used to record features that are based on several stadia points.
l49
....g
:i5
....
403.5
67
Fig. 8-8. Interpolating 20-foot contours between two stadia points; profile at left,
map at right. .
t-
Most points marked in these ways can be found easily by anyone carrying the plane table sheet, even several years after the survey.
150
~Ji'(
151
The procedure that follows may be preferred for mapping at intermediate scales. Its only special requirement is that the magnetic declination must be nearly constant over the- area or the magnetic meridian
must be determined for each control station.
1. The rod is placed on a station marked on the plane table sheet.
2. The geologist sets up the plane table at a feature he wishes to plot
and then orients the board by the magnetic meridian.
3. To locate his position, he draws a ray toward the rod and makes
the u~ual stadia readings and computations.
4. After plotting the point and sketching the features around it, he
moves the table to another point.
--~--
'?
~Transferred
point'
16=1-
T<dophod
After each day's mapping, the stadia points located during the day
should be marked with waterproof ink by placing a small dot in the
needle hole that marks the point. The point number should be lettered
about 71_ 0 in. high beside the point and facing so as to read from the
south edge of the sheet. Elevations should be left in pencil. Structure
symbols may be inked in from day to day, but other geologic features
are best left in pencil until the final field check of the map, unless the
features are_simple and are located by many stadia points. If possible,
contours should be left in pencil until the map is completed. Pencil
152
work in completed areas can be protected against smudging by overlays of brown paper.
The instrumentman or recorder should also check through the stadia
computations for arithmetical mistakes, while the rodman-geologist examines the notes and sketch maps of the preliminary geologic examination to determine what should be covered in the next day's mapping.
8-17. Vertical Cross Sections from Plane Table Maps
Almost any project requiring a plane table map will also require detailed vertical cross sections. Section lines should be chosen as the
mapping approaches completion. They can then be walked out during
the final stages of the mapping, when geologic and geographic details
can be added easily (Section 8-18). The sections should be drawn in
pencil before the party leaves the field.
The section lines should be chosen to intersect as many geologic
features as possible at about right angles to their structural trend. The
lines should be drawn with a sharp hard pencil and a steel straightedge.
8-18. Completing Plane Table Maps
After the stadia work on a plane table sheet has been completed
and cross-section lines penciled on it, the sheet should be carried into '
the field to check the shapes and positions of the various features.
Details overlooked in the stadia mapping should be added, especially '
along the cross-section lines. At this stage of the work, the value of
the stadia point markers described in Section 8-13 becomes apparent.
The markers permit exact recovery of most stadia points, so that data
can be added accurately.
The map is generally carried on a plane table board, and the changes
and additions are made by compass and pace (or tape) measurements
from the stadia points and instrument stations. Some of the following
are commonly added at this stage: structure symbols to complete coverage (especially those minor structures that were overlooked until
the mapping was well along), thin or partially hidden key beds, lithologic variants with gradational boundaries (especially in igneous and
metamorphic rocks), areas of hydrothermal alteration, and contacts of
surficial materials. This is also the best time to make systematic collections of rocks, minerals, or fossils, particularly where there are specific problems to be solved.
After the field work is completed, the map should be inked, labeled,
and perhaps colored. Geologic features should be inked first, either
153
all in black, or by using green for faults arul red for veins, ore bodies,
or other features. that require emphasis. The culture should then be
~ed in black, followed by the drainage in blue. Finally, contours
may be inked in brown. The entire map should then be cleaned with
a soft eraser, after which it may be colored in pale erasable tints. Symbols for rock units may be added in black ink if space permits.
Each plane table sheet should then be given an inked title, including
project or geographic name, sheet number (or an index map showing
its location),, names of organization and party members, and dates of
the survey. A north arrow and bar scale must be added, even if the
map has a coordinate grid. Finally, an explanation should be drawn
that includes the rock colors, symbols, structures, and culture symbols.
If there is not room for all these items on the front of the map, they
can be put on the back. Although they may seem unimportant in the
field, the map would eventually become worthless without them.
Before packing gear to leave the field area, all notebooks, sketch
maps, computations, aerial photographs, and other items bearing information on or about the survey should be accumulated, checked to
see that they are suitably labeled, put in correct sequence, and packed
for transport. Plane table maps must always be packed flat. The instruments ~nd their carrying boxes should be dusted with a soft brush
or cloth and then packed for transport.
References .Cited
Birdseye, C. H., 1928, ,Topographic instructions of the United States Geological
Survey: U.S. Geological Survey, Bulletin 788, 432 pp.
Low, J. W., i952, Plane table mapping: New York, Harper and Brothers, 365 pp.
Moffitt, F. H;, 1959, Photogrammetry: Scranton, Pa., International Textbook Co.,
455 pp.
~
Making a Geologic Map
from Aerial Photographs
Geologic features plotted on aerial photographs are generally transferred to a base map (Section 5-11), but a map can also be compiled
directly from the photographs. This method of making a geologic map
is often superior to the plane table method, especially if the scale is
small or intermediate. The control network can usually be planned
and surveyed more quickly than that for a plane table project, and
there is no need for time-consuming stadia measurements. Moreover,
unless a base line must be taped, the entire photograph survey can be
made by one man. The -eompilation of the final map is not nearly so
simple as drafting a map from plane table sheets; however, this compilation does not take up any field time.
The plane table method is generally superior for mapping at large
scales. Photographs with scales larger than 1: 18,000 are usually not '
available, and it is difficult to compile a map from photographs en-
larged more than two times. Moreover, the detailed relations shown ....
on such maps are, in a sense, misleading, for the compilation can be ,
no more accurate than the original photographs from which the enlargements were made.
The methods described in this chapter can be used to make a plani~
metric map from an overlapping set of aerial photographs. No special
equipment is required; however, photogrammetric instruments and
office equipment will aid the compilation considerably. Compiling
contour maps from aerial photographs requires special instruments
and involved procedures, as those described by the American, Society
of Photogrammetry ( 1952) and Moffitt ( 1959). The third dimension
on the planimetric compilation can be shown locally by elevations
measured during the control surveys. It can also be shown by a series
of cross sections if section profiles are surveyed during the field season.
9-2. Preparations for an Aerial Photograph Project
155
5-1). These photographs must have more than 50 percent end lap.
Hittle side lap is necessary if the compilation can be controlled adequately; otherwise, 30 percent side lap is desirable. Ideally, the photogr~)ls should be tilted little if at all, but this cannot be determined
undl points are surveyed on the ground. In any case, the compilation
compensates for tilted pictures if enough control points are surveyed.
Data on established ground control should be obtained well in advance, and the photographs that have established control points on
them should be listed. It will also be useful to determine the average
scale of the photograph~ and to make a thorough photogeologic study
of them before the field season. The control network can often be
planned in detail when this is done.
Equipment and supplies are listed in Appendix 1, and additional suggestions will be found in Chapters 5 and 7. The materials described
in Section 9-6 should also be included in field gear because the photographs can often be prepared for the compilation while in the field.
9-3. Mapping Geologic Features on Aerial Photographs
156
"A
Miner
Fig. 9-1. Ground contror based on established triangulation stations (double triangles).
EA
"'D
157
Section-line surveys are made to measure accurate topographic proBles and te make thorough ebservations and checks of the geologic
features along them. If outcrops permit, these data should be about
twice as detailed and precise as that plotted throughout the area.
Cross sections are especially important in aerial photograph projects,
for the compiled planimetric map will not show the vertical dimension
of geologic features. The exceptions are areas where there is little or
no relief or where rock units and structures are nearly vertical. The
section lines must be chosen before the end of the field season because
the section proflles must be surveyed in the field. If their approximate
positions can be selected before the control survey, the section lines
can be controlled by this survey.
If slopes are even for considerable distances and geologic features
reasonably simple, a proflle can be based on a few accurate locations
..
..
~
~
photographs, the cross section can be plotted directly in the field ( preferably at about t~e the scale of the photographs).
The alidade and plane table should be used when an especially
thorough survey is needed. If signals are set up at each end of the
section line, instrument stations along the line can be located by ( 1)
setting the plane table 11P on the line, ( 2) orienting it by sighting to
a signal on either end of the line, and ( 3) resecting the point by sighting on any other signal (Fig. 9-3). Three-point methods and accessory traverses can be used where the view along the section line is
obstructed. Details between plane table stations can be added by
stadia methods, and notes and geologic features can be plotted directly
on the plane table sheet. Each instrument station should be marked on
a photograph when it is occupied.
158
'
II
Ir---
I1
II
I
II
at breaks in slope. In Fig. 9-2, for example, the elevations and positions of five key points on the profile were determined by intersection
during the control survey for the map. After the main geologic features had been mapped, the section lines were walked out and structural details were added.
When the final map is compiled from data such as these, it may be
found that the key points do not lie quite on the final section line (because of distortion within the photograph image). They should be
close enough to it, however, so that elevations along the line can be
estimated closely by stereoscopic examination of the photographs.
More involved section-line surveys must be made if slopes are un- .
even and geologic features so complex that elevations and positions are _
required at each contact or change in dip. These l:l-Ccessory elevations
can commonly be measured by hand level traverses along the section
line (Section 2-6) or by using a barometer (Section 4-4). The data __
may be plotted on an acetate sheet placed over the photograph. If
relief is so moderate that distances can be scaled directly from the
..
----r---~
I'
---4'---+
,I
jl
jl
II
1\-
--I
line of
/section
' \ Instrument station
A'~
l59'
II
\_
I
\
''tbC
--., I
:~
f'B'
Overlay
Fli~ht lin;
/- A
B'
---~
--.~~
J;'c
Fig. 9-4. Two overlapping photographs, showing the vertical planes that include
the camera, the centers of the prints, and the images of a point, C. In the drawing on the right, the position of C has been intersected by placing a transparent
overlay over each of the prints and drawing. the radial rays A'C and B'C.
160
,.
camera and the point at the center of that photograph (points A' and
B'). It must follow that C will lie on a line (a radial line) that passes
from the center of each print out through the image of C. This will
be true regardless of the relative elevation of points A', B', and C. If
the center points of two overlapping photographs can be marked accurately on a transparent overlay, any point such as C can be located
by placing the overlay on the photographs and drawing the two rays
A'C and B'C. This corrects photograph distances that would otherwise be modified by the effects of relief. The location of centers on the
overlay and the orientation of each print are achieved in part by the
use of ground control and in part by the compilation methods described in the remaining sections of this chapter. Briefly, the steps in
the compilation are as follows:
1. The center points and the points to be intersected are marked
on the prints.
2. The ground control points are plotted on a transparent overlay
at a scale close to the average scale of the prints.
3. The center points of the photographs are located on the overlay
by resecting from the plotted control points.
4. The photograph points are intersected on the overlay by drawing
radial lines from photograph centers.
5. Geologic features, drainage, and culture are penciled on the over-
lay by referring to the locations of photograph centers, intersected
points, and control points.
Materials for the compilation. In addition to the instruments
needed for plotting the control points on the overlay (Section 9-8),
the following materials are needed: several bright-colored opaque inks,
drop-circle compass, pencils ranging from 3B to 6H, file for pointing
pencils, 10-inch triangle, soft eraser, transparent sheet for overlay, tracing paper, black waterproof ink, and crow quill or other fine pens.:
The opaque inks can be prepared by mixing brightest red, green, or'
orange transparent inks with a tempera paint of the same color. They
must be mixed thinly enough so that they will flow readily from the
point of the compass, and they must also be opaque enough when dry
to show clearly against either light or dark areas of a photograph.
The transparent sheet must be large enough for the whole map and
must be as strong and transparent as possible. Frosted plastic drawing
stock and top quality water-resistant linen are suitable. The tracing
paper should be exceedingly transparent.
161
000
Fig. 9-5.
0 0 0~
9-7. Marking and Transferring Photograph Points
A-----.----. B---....----.
Collimation
/
marks~
+-"Center
w
Fig. 9-6.
<D
<D
Control
point
Transferred centers
Pass
point--...
\.
9
<D
c--,-----
...
'
i.
..
16.3
162
The radial line method can be used quickly and precisely to compile
a ma,p from photographs that include three or more intersected control
points.-._. The photographs are first marked with radial rays that pass
from the center out through the control points (Fig. 9-7A). This can
be done either with a transparent triangle and a sharp 3B pencil, or,
if the pripts are so dark that the pencil lines cannot be seen through the
overlay, with a ruling pen and par,tly opaque yellow ink.
The compilation should be started at one end of a flight line where
relief is moderate. The steps are as follows:
1. Fasten the first photograph to a smooth surface with drafting tape
and orient the overlay so each control point on the overlay lies exactly
over the corresponding radial line on the print (Fig. 9-7B). This will
orient the photograph correctly unless the center falls on the circle that
passes through the control points used in the location (see Section
&-11).
2. Using a sharp hard pencil and a triangle, draw a small cross over
the center of the photograph, then draw a line through the center of
the cross and over the circle marking the center transferred from the
next photograph (Fig. 9-7C).
3. With the overlay still in place, draw radial lines across each pass
point of the underlying photograph (Fig. 9-7D).
4. Tape down the next photograph and shift the overlay until the
control points lie exactly over the radial lines drawn on the print, just
as in step 1. Draw a center cross and radial lines as in steps 2 and 3
(Fig. 9-7E). Unless one of the photographs is severely tilted, the sec-
164
A
"'o /o
:' ~ I
I
'
--1
I
~~
I
I
I
I
I
"'c ",__,..,,I
~------------'::1
A /:
,..
----------- _}j
r-------------l
F
~
A{rM )\ /
I
4o
+---f+-+
~~
Y- _____
\ A~~.."'J
l~:r
___ ----
ond photograph center should lie exactly on the center line drawn for
the first photograph.
5. Tape down the third photograph and orient and mark the overlay
as before. The radial lines to pass points that lie in the overlap area
of the first and second prints (as M and N in Fig. 9-7F) should now
make three-line intersections.
6. Plot the other photographs of the Hight strip by the same procedure, and then plot adjoining Hight strips.
When completed, the overlay will show a series of Hight lines with
small crosses marking photograph centers and a large number of intersecting rays that represent pass points. Most detectable errors are
~~
I+---<-.-~
"'I
.. "'I
E---------------------r------------1
c~
-i
+-I
I
II
--,
a.
<::l-
i /I
I
,.
a.
I
\
< I
'
\ '$:
Nl
L------- _____
.~ 711
":.J
"'
c ----.:)
'
~~
',
I - - - - - - - - - - --,
165
"
J66
(Fig. 9-8C). If the overlay cannot be oriented so the points and lines
coincide, check the previous constructions carefully. Any remaining
error is probably caused by tilt, and the points can usually be brought
closer to coincidence by shifting the overlay slightly off the center
points of the photograph or by twisting it slightly from the flight line
orientation. Any residual errors should be adjusted after the adjacent
flight lines are plotted.
8. Draw the center line to the next photograph and all radial lines to
pass points (Fig. 9-8D).
9. Prepare and use the next uncontrolled photograph as in steps 6,
7, and 8, and continue until the compilation has gone halfway to the
next controlled photograph; then begin the same procedure at this
photograph, working back in the opposite direction.
ABC
8~
-,
II"',/"
~~
'
_,/I
___./'
I
I
t:
fl.""
,.
_ _ .J
w~-----I
I
',
I
I
I
&
I
I
:;f(s
1...:--------
}1
'{s
"! i
- - - - - - - - - - . -- j_l
I
___./'
I ~-~
"\_
o
'!!___________ _
i{____{R
i~----1--/-l
I
I
I
tr
"'!
------------- _ _ __j
fig. 9-8. Steps in compiling points where all photographs do not include ground
control points.
HI
"'
~
1'- /f.
A
.!.
A
Fig. 9-9. Using controlled intersections in one flight line to improve the plot of
an uncontrolled line.
~-
ij'\____7__
DE
.!.
..
167
The control points, center points, and pass point~ plotted on the overlay can now be used as a basis for transferring geologic features, drainage, and culture from the photographs to the overlay. The pencil rays
and photograph numbers on the overlay should be erased and the
points marked permanently in ink, or the positio~s of the points 'should
be traced on a fresh overlay. The compilation will be facilitated if
small ink circles are drawn around pass points, small triangles drawn
around control points, and crosses drawn at center points. This should
btl done with a blue transparent ink that will not be photographed or
printed when the final map is reproduced. The final overlay sheet must
be strong enough to withstand considerable erasing, and pencils used
on it must be soft enough (typically lH to 3H) so they do not crease it.
Some photographs have such weak contrast or such strong vegetation patterns that it is necessary to use a stereoscope to see their drainage lines and ridge lines. These lines must be delineated on the photographs before they can be transferred, as by marking them with yellow
or orange ink. It is easiest to delineate streams by placing photographs
in reverse order under the stereoscope, because the peculiar appearance
of the pseudoscopic image makes streams very apparent.
168
Fig. 9-10.
J69
iii:.
The principal jpb in completing the map is to check the first pencil
compilation careflilly against the photographs; this should be done
before the cross sections are compiled. It is best to do the checking
with a stereoscope, working systematically through one flight line at
a time. The m(jor lines and features will already be drawn accurately,
but much detail can generally be added. This is an important step
since details of the drainage and culture will give other geologists their
only means of locating geologic features in the field. The pencil draft
and cross sections shoulcj. be completed before either are inked. Suggestions for the final drafting are given in Section 11-10.
References Cited
American Society of Photogrammetry, 1952, Manual of photoirammetry, 2nd ed.:
Washington, D.C., 876 pp.
Moffitt, F. H., 1959, Photogrammetry: Scranton, ~a., International Textbook Co.,
155 pp.
!QJ
l7l_
~
Detailed geologic studies usually require one or more of the following operations: topographic and geologic mapping at scales larger than
1 in. = 200 ft; cleaning or excavating of exposures; drilling; mapping
of underground workings; and extensive sampling. Generally, these
operations can be used to advantage wherever geologic features must
be studied in a thorough, quantitative way. They are commonly
used in economic studies, for example, to improve the quarrying of a
gypsum body or the underground mining of a coal seam or copper deposit. In addition to their economic uses, however, they may also
solve a great variety of noneconomic problems, such as the origin of
a complex igneous body, the degree of alteration around an intrusion,
or the effects of deformation on a sandstone. Because they are costly
and time consuming, these operations are done only after preliminary
studies have shown they are justified.
The regional geologic setting of local studies should be worked out
thoroughly; otherwise the significance of many rock units or structures might be overlooked. The magnitude and age of most faults,
for instance, cannot be determined from maps of small areas alone.
Two kinds of preliminary studies should be considered: ( 1) regional
reconnaissance of rock units and structures and ( 2) mapping at intermediate scales ( 1:24,000 to 1: 62,500) around the area of specific interest. The first study defines the areas or features that should be
covered by the second; together the two indicate which operations
should be used in the final local study.
The local study itself should be started with mapping, for drilling,
excavating, and sampling cannot be planned effectively until detailed
surface and underground maps are well underway. Excavating and
drilling should be overseen directly by the geologist or geologist-engineer lest critical relations are destroyed before he sees them. Besides insuring that he receives data promptly, this can increase the
effectiveness, or reduce the cost, of a given program accordingly. The
geologist or engineer should also supervise the sampling to make sure
the samples will be as meaningful as possible.
170
172
173
'IIi,
Scaled
vertical
distances
paper
____.-/
",!'-
..
Instrument
station
Plane table sheet
Fig. 10-1. Plotting the geologic features of a steep cut (left) on a vertical projection. The projection can be constructed directly on a plane table if a strip
of cross section paper is fastened to the plane table sheet (right).
and excavating are usually done with a bulldozer or mechanized ditchdigger. Thin soil and overburden can be hosed off if water under high
pressure is available. When funds do not permit these large-scale
methods, trenching with a pick an9- shovel may be very effective. Even
in the most modest projects, contacts covered by thin soil can be
exposed quickly with a folding entrenching tool (Fig. 10-2).
Drilling. Drills can be used to determine the position of geologic
structures, to measure subsurface units, and to sample rocks and ores.
Core drills, especially diamond drills, are particularly ,useful for drilling
long distances in hard rdcks. The diamond drill consists of a tube with
small diamonds mounted in the bit end. As the drillis rotated, it cuts
a cylindrical core of rock, unless the rock is so fractured or friable that
it crumbles. Some drl].ls are pulled from the hole after each 5 or 10 ft
have been drilled, and the core is removed and placed in a core box.
Other kinds of drills permit longer intervals to be cored. As the core
is examined, all rocks, ores, and geologic structures should be described
in a log of the hok (Section 12-10). Where rocks are so friable or
fractured that no core is recovered, the log can be augmented in part
by records of the rate of drilling and water loss in the uncored interval.
The finely ground sludge from the drill hole should be sieved, cleaned,
and examined microscopically to identify the uncored rocks. The
s-ludge is also collected for assays.
The diamond drill can be pointed in almost any direction and used
either underground or on the surface. Occasionally, however, long
holes are deflected appreciably from their initial direction, especially
where they cross bedding or other structures obliquely. It is then
necessary to survey the course of the hole with especially designed
equipmel'tt (Collins, 1946; McKinstry, 1Q48).
Other drills may be used where materials are soft, where drilling
distances are short, or where geological targets are simple. Hand- or
If the outcrop map shows that exposures are not adequate to delineate critical features or to sample efficiently, excavating or drilling
may be necessary. Trenching at right angles to the strike is particularly effective for uncovering steeply dipping planar structur_es, as a
contact, a thin bed, or a vein. Pits can generally be carried to greater
depths and are better suited for measuring or sampling gently dipping
beds or veins, or unstratified igneous bodies. Large-scale cleaning
Fig. 10-2.
6 in,
i'
174
175
'
R..a Rot:lc. Mine
400 Level
I"= 40'
R. .Tones 1
1-21-61
......
d
I
--
1 _,,/
_.... _. ./ . . . __ . . .
4oo S'ft>pe
I
I.
Fig. lll-3.
labeled.
one sheet to the next so that sheets can be connected readily without
the office map. Transit survey points will be used for measurements
underground and must therefore be traced precisely. Finally, each
sheet should show the name of the workings, the scale, and the names
of the geologists. A typical work sheet is shown in Fig. 10-3; the symbols used are explained in Appendix 4.
The following terms, which are often used on work sheets, should
be understood before going underground:
SQ,!lft-relatively narrow working excavated from the surface; it may
be vertical or inclined.
Raise-narrow vertical or inclined working excavated upward from
underground.
Winze-narrow vertical or inclined working excavated downward
from underground.
Level-groups of tunnel-like workings that lie approximately the
same distance below the surface in a given mine.
Adit-tunnel-like entrance to a mine.
Drift-tunnel-like working along a vein.
Crosscut-tunnel-like working that intersects a vein.
Stope-an underground working from which ore has been extracted,
either above (overhand) or below (underhand) a level.
Back-ceiling of a working.
Face or breast-wall at the end of a working.
176
Before mapping starts, the workings should be reconnoitered to determine ( 1) the general nature of the rock units, the ore, and the
principal structures, ( 2) how much the walls must be cleaned, and
( 3) the accessibility of old workings. Survey data applying to stopes
and other irregular or partly filled openings should be drawn on the
work sheets or on separate vertical sections. Assay maps should also
be studied.
It is advantageous to start mapping where rocks are fresh and
structurally uncomplicated. A hundred feet or so of walls are examined first. If they are dirty, it is necessary to scrub them with a
brush and water or to chip them at closely spaced intervals with a
hammer. The types of rocks, ores, and alterations are plotted on the
work sheet, either with pencil patterns or colors (Fig. 10-4). Contacts, faults, and other structures are shown by the usual symbols.
Mining geologists often use green for faults and red for ore. Where
planar structures can be traced from one side of a working to the
other, their strike can be found by standing with one's back against
the plane in one wall and sighting a level compass line to the same
plane on the other wall. Many strikes and dips can also be measured
by method I of Section 2-7.
Pori' I
oJbvnd. pyr.,
t:lllcopyr.
Fig. 10--4. Work sheet, showing plotted geologic features and brief notes. A flatlying dike has been plotted in an accessory vertical section at upper right.
fig. 1~5.
177
178
The kinds and numbers of samples needed will vary greatly with .
the purpose of a project and its geological situation. Plans for sampling
should be based as much as possible on knowledge of sampling plans
used elsewhere on similar deposits. Examples of such sampling
plans are given by Parks ( 1949), Jackson and Knaebel ( 1932), Pitcher
and Sinha ( 1958), Grout ( 1932), and Shaw ( 1956). Alternate
sampling plans can be evaluated statistically to determine which is
likely to give the information needed with least effort and expense.
Statistical methods should also be used when determining the significance of sample data. Methods of statistical evaluation are described by Dixon and Massey (1951), Hoel (1960), and Wallis and
Roberts (1956).
179
180
181
. _ 1 . 3 _ . 1 _ . 3 _ . L _ J__ .J
;{\;{\;{\;{\;{\;{\;{\2
Small circles
1:\
;{\ ;(\ ;(\ ;(\ ;;"\ ;(\ ;;\r
t--------
1\.1\j\/z\j \/z\J\/z\.i\2
1\i\1\i\1\i\1\A\~
.\ ;(\ j"\ ;{.\;(\ ;;"\ i.\ ;(\ ~z
-------
Fig. 10-7. Sampling plan for determining the bulk composition of the outlined
body," using four groups of equi-spaced samples.
of many bodies can be determined only within broad limits; thus the
limits allowable for a given study should be considered at the outset.
The precision of the.-analytical methods that will be used should also
be considered (see Ahrens, 1950; Fairbairn et al., 1951).
To keep to a minimum the number of samples analyzed, the body
should first be evaluated by comparisons with other sampled bodies.
Samples must be distributed so as to provide a statistically valid
(random) measure of the entire body; they could, for example, be
spaced evenly on the horizontal coordinates and levels of a mine or
at intersections of a rectangular grid on the surface. Materials must
be collected as flosely as possible to the prescribed points. If they
must be distributed unevenly, it is necessary to weight each sample
according to the distances from adjacent samples (Parks, 1949, ch. 4).
The number ~f samples analyzed can sometimes be minimized by
collecting the samples in several groups, each distributed evenly over
the body (Fig. 10-7). One group of samples is analyzed first, and
their average composition (mean) and some measure of the degree
to which they differ (as their standard deviation) are computed. The
standard-deviation will show the approximate reliability of the average
composition. If the average is too indefinite, a second group of samples
is analyzed and a new total mean and standard deviation computed.
This procedure is repeated until the overall mean can be shown to
lie within limits that are small enough to satisfy the purpose of the
~d
.
Sampling for dfstribution of compositions. When a large number
of samples can be used, it may be possible to determine the distribu-
'
182
:: ~. ::.
f
I
tl
183
Unit
to be
sampled
Fig. IG-9. Diagrammatic section through a working with uneven walls, showing
why some parts of a channel sample have to be weighted relative to other parts
(compare segments a and b ) .
184
_!lj
If only fresh rocks are collected, the bulk will be biased against
rocks containing unstable minerals, and if only outcrops are sampled,
the bulk will be biased relative to weak rocks. A particularly large
error can occur when diamond drill holes are sampled only from the
core rather than from the core, fragmental materials, and sludge.
Bias can also be introduced by modifying samples after they are
collected. Possibilities for fraud in mine sampling are manifold; some
examples are described by Hoover ( 1948).
References Cited
Ahrens, L. H., 1950, What to expect from a standard spectrochemical analysis of
common silicate rock types: American Journal of Science, v. 248, pp. 142-145.
Collins, J. J., 1946, Some problems involved in the interpretation of diamond-drillhole sampling and surveying: Mining Technalogy, American Institute of Mining
and Metallurgical Engineers, Technical Publication 1842, pp. 1-28.
Dixon, W. ]., and F. J. Massey, 1951, Introduction to statistical analysis: New
York, McGraw-Hill Book Co., 370 pp.
Fairbairn, H. W., et al., 1951, A cooperative investigation of precision and accuracy
in chemical, spectrochemical and modal analysis of silicate rocks: U.S. Geological Survey, Bulletin 980, 71 pp.
Flinn, Derek, 1959, An application of statistical analysis to petrochemical data:
Geochimica et Cosmochimica Acta, v. 17, pp. 161-175.
Grout, F. F., 1932, Rock sampling for chemical analysis: American Journal of
Science, v. 24, pp. 394-404.
Hoel, P. G., 1960, Elementary statistics: New York, John Wiley and Sons, 261 pp.
Hoover, T. J., 1948, The economics of mining: Stanford, Stanford University Press,
~~
Jackson, C. F., and J. B. Knaebel, 1932, Sampling and estimation of ore deposits:
U.S. Bureau of Mines, Bulletin 356, 15.5 pp.
Krumbein, W. C., and J. W. Tukey, 1956, Multivariate analysis of mineralogic,.
lithologic, and chemical composition of rock bodies: Journal of Sedimentary
Petrology, v. 26, pp. 322-337.
McKinstry, H. E., 1948, Mining geology: New York, Prentice-Hall, 680 pp.
Moffitt, F. H., 1959, Photogrammetry: Scranton, Pa., International Textbook Co.,
455 pp.
Parks, R. D., 1949, Examination and valuation of mineral property, 3rd ed.: Cambridge, Mass., Addison-Wesley, 504 pp.
Pitcher, W. S., and R. C. Sinha, 1958, The petrochemistry of the Ardara aureole:
Quarterly Journal of the Geological Society of London, v. 113, pp. 398-408.
Shaw, D. M., 1956, Geochemistry of pelitic rocks (Part III: Major elements and ,
general geochemistry): Bulletin of the Geological Society of America, v. 67,
pp. 919-934.
Wallis, W. A., and H. V. Roberts, 1956, Statistics, a new approach: Glencoe,
Ill., The Free Press, 646 pp.
..
187
fore writing the report. Field notes should be paginated in chronological order and then reviewed systematically. Some geologists find it
useful to reshuffie or cut note pages so that all items pertaining to a
certain part of the report are brought together; others prefer to leave
their notes intact and to make lists or card files of items pertaining to
particular sections of the report.
The main divisions of the report should be set down first, and then
its subdivisions and details filled in in outline form. The outline that
follows presents the main parts and principal subdivisions of a typical
geologic report.
186
I. Introduction.
a. Purpose of project, including review or explanation of any
specific problems.
b. Dates and general procedure of the project.
c. Acknowledgments of help received.
d. Geographic setting.
1. Location (a small index map will generally suffice).
2. Accessibility, if not obvious from maps.
3. Nature and distribution of principal geographic features.
4. Vegetation, climate, and land use (generally stated briefly).
II. Regional geologic setting.
a. Nature and distribution of principal rock systems, series, or
formations.
b. Major structures, in chronological order wherever possible.
c . . Summary of regional geologic history, if appropriate.
III. Rock units.
a. Introduction to the general nature, thickness, and grouping
of the stratigraphic sequence, including brief descriptions of
general stratigraphic problems, if any.
b. Systematic description of rock 'units, starting with oldest.
1. Nomenclature (explanation of name used for unit, with reference to past usage) .
2. General lithology, distribution, shape, and thickness of unit.
3. Detailed description of lithology, including lateral variations.
4. Definition of contacts, if not included in lithology.
5. Fossils, if any.
6. Age and origin of unit.
IV. Structures.
a. Brief introductory description of trends and interrelations of
principal structural features.
'
The introductory sections should" pertain directly to the main subject. Climate, vegetatio~nd land use can usually be described adequately in a few sentences, unless the region is especially remote or
the report is concerned chiefly with economics. Descriptions of geographic features can be kept concise and clear by including a pagesize map of the region. Special field or laboratory methods should be
described in an additional section of the introduction, and another
section may be needed to define important terms whose usage is not
widely standardized.
The section on regiorlal geology provides an important framework
for the detailed descriptions of the report. This is a section that will
require a thorough winnowing of the literature. When items unsupported by adequate' data are used, their source and probable reliability
must be noted. A separate section presenting a chronological review
of geologic work done in or near the area may be added in some cases,
but a critical evaluation of important contributions is generally preferred. A page-size map showing the main geologic features of the
region is always helpful, and it may be combined with the geographic
map used in the introduction. When the main body of the report is
completed, major structures and rock units should be reviewed to determine whether they correlate accurately with the descriptions of the
regional geology.
Large numbers of rock units that cover a considerable span of.
geologic time can be organized on the basis of systems or combinations
of systems. Accordingly, metamorphic rocks should be presented in
the order of their premetamorphic age. Intrusive igneous rocks are
..
best presented in chronologie order with all the other rocks, although
they may be described in a separate section. Rock units that are
formalized for the first time by the use of a geographic name must
be defined according to accepted rules (see Sections 11-4 and 12-2).
For all but the simplest areas, the section on structure must be organized with special care to describe both the individual features and
their interrelationships. This may require using a different order than
that shown in the outline; for example, a strict chronologie order may
be better in some cases. Data shown clearly on the geologic map
need not be described at length in the text. If the main map is too
cluttered to show the geographic distribution of the major structures
clearly, a page-size structure map should be used in context with the
structure section.
Most general reports conclude with a section on geologic history,
but this may be followed by sections on economic deposits, metamorphism, igneous activity, glaciation, systematic paleontology, or
other pertinent subjects.
188
l89
The one basic requirement for writing the report is that both data
and ideas be presented clearly. This means that sentences must be
constructed with care, and sentence order must be logical. Moreover,
words must be chosen so as to give the precise meaning intended. One
way to achieve logical sentence order is to expand an outline to sentence level. Subheadings can be used to orient the reader's thought to .
major topics, or each major paragraph can be introduced by a topic
sentence. There are many further ways to improve the clarity and
precision of a report. These are described fully in guides to general exposition and scientific writing (as Perrin, 1950).
The meaning of scientific terms must be clear to the readers for
whom the report is prepared. For general reports, the less technical
of two possible terms is preferred. If an important term has more
than one meaning, its usage should be defined in the introduction, in
context, or in a footnote. Roget's International ThesaurtlS is a valuable
aid in selecting nontechnical words, and the American Geological Institute's Glossary of Geology and Related Sciences provides brief definitions of many technical words. Suggestions to Authors of the Reports
of the United States Geological Survey contains many valuable instructions regarding both word usage and report writing.
terms.
Geographic names. Quadrangle maps of the U.S. Geological Survey
show almost all names correctly. Errors are likely to occur only on
old maps or on recent maps where it was not possible to check the
names of minor outlying features. Inquiries in the area may resolve
some of these errors; however, these inquiries must be thorough because local usage tends to vary with time and individual feeling.
Forest Service maps and county land plats may be helpful in areas not
covered by quadrangle maps.
Because most readers will not be acquainted with the geography of
the area described, all place names used in the text should appear on
at least one map in the report. Furthermore, descriptions of rock
units and structures should not depend on an endless chain of place
names, especially if the names may be difficult to locate on maps. Descriptions are followed more easily if locations are based on directions
<
190
l~l
The front matter consists of all materials appearing before the main
body of the report, generally a title page, a table of contents, a list
of illustrations, and an abstract.
192
Title page. In addition to the title of the report, the title page
should bear the name of the writer, the date or dates of the project
(or, in some cases, the report), and the name of the organization, if
any, for which the work was done. The title itself should be as short
as possible but should state clearly the nature or content of the report.
Titles of general reports should begin with Geology of . . . , followed
by the geographic name of the area described. It is unnecessary for
geographic names to be compounded to delimit an area completely,
for the area can be located readily by turning to the report.
Table of contents. The table of contents is prepared from the final
copy of the manuscript, when all section headings have been decided
upon. In many cases the lowest rank of headings may be omitted from
the table of contents if they are repetitive (for example, the subheadings under each rock unit).
List of illustrations. The list of illustrations gives page numbers or
other suitable reference to all text figures and plates, including those
folded separately at the back of the report. Wherever possible, titles
or captions of illustrations are shortened for the list. For example,
a figure caption that reads "Contact between Lucia Shale and The
Rocks Sandstone in upper Reliz Canyon, showing thin sandstone beds
and concretions in shale" could be listed as "Contact between Lucia
Shale and The Rocks Sandstone."
Abstract. The abstract is a very brief version of the report. It is
all some readers will ever see (or read) and should therefore be as
informative as possible. Abstracts are generally 150 to 600 words long,
although much shorter ones may be adequate for some papers and
1,000 to 4,000 words may be needed to abstract exceptionally long or
thorough reports. The length of the abstract also depends on the
nature of the paper and on the specifications set by the publisher.
Three suggestions for writing abstracts are:
193
of the report can then go quickly and easily from abstract to table
of contents to text.
3. Each major section of the report should be summarized in at
least a sentence, and if the report is long or thorough, each major section should be summarized in a paragraph. Favorite topics should not
be emphasized so much that other topics that may be of interest to
many readers are omitted. The main descriptive materials should receive the most thorough treatment. New lithologic units, unusual
associations of rocks or minerals, and new fossils or minerals should
at least be mentioned in the abstract, regardless of the main purposes
of the report. Where new or unusual methods have been used, they
should be described briefly.
_,;
I.
l94
195
1.
2.
3.
4.
Duplication processes. Since many illustrations must be limited because of the processes that will be used to duplicate them, these processes must be considered when the illustrations are planned. Some of
the factors to consider are: size of copy, costs, legibility of final copy,
accuracy of duplication, durability of copy (both to light and handling),
and nature and color of print surfaces (can they be colored and inked?).
The possibilities vary so with local facilities that it is necessary to
contact blueprinters or engravers and discuss items specifically. They
will be able to suggest processes that will give the best results, and
they may also be able to suggest how a map can be enlarged or reduced efficiently during the drafting stages. The following notes apply to the processes available most widely.
Ozalid (diazo-ammonia) process. Typically blue or black lines
on white base, but sepia tones are available locally; blue lines may
fade on some paper stocks; made from a transparent drawing; various
papers, including card stock, available locally; widths typically up
to 42 in. or 54 in. and any length; one of least expensive processes
for less than 50 prints.
BW or Bruning (diazo) process. About the same as for Ozalid
prints; white base may age yellowish or brownish in some paper stocks.
Vandyke, blueprints, brown-line prints. Various wet processes,
typically made in vacuum frame; require negative of original to give
positive copy; more expensive than those listed above, but highquality papers available; prints typically permanent; scale likely to be
somewhat different from original.
Photostat and other photographic prints. Various paper sizes and
surfaces, depending on facilities; some surfaces not suitable for coloring or ink work.
Lithographic, photo offset printing. Regular press printing, produces
excellent copy on various papers; size limitations vary greatly with
local press equipment; usually economic only where more than 30 to
50 copies needed.
Xerography, duplimat. Excellent black-line prints from any good
line copy; cheaper than photo offset printing, but image sizes limited
to approximately 9Y2 X 12Y2 in.
are desc;ibed briefly in this section. Ridgway ( 1920) has given many
additional suggestions.
Small-scale maps. Most reports require small-scale maps that show
area locations, geographic features, or generalized regional geologic
features. These maps are most useful if they are page size or smaller,
and are bound in the part of the report to which they refer. They
should be simplified by removing or subordinating nonessential geographic features. Place names should be shown clearly. It is easiest
to make these drawings for about 50 percent reduction.
Photographs. No illustration is as convincing as a good photograph,
but only the exceptional outcrop or hillside can b~ photographed effectively. Critical features generally lack contrast or are obscured by
patterns of shadows or Vf(getation. These difficulties can sometimes
be overcome by photographing the outcrop at a certain time of day
or by using flash attachments or colored filters. Some prints can
be clarified greatly by inking contacts and labels directly on them.
Photographs should not be altered in such a way as to modify geologic
features, but complicating shadows or other misleading effects may
be retouched. When 'only' a few copies of a report are needed, color
photographs cab be used to illustrate features that would not be clear
in black and white prints.
Drawings of outcrops and specimens. Drawings can eliminate disturbing and superfluous material and can clarify relations by showing
cut-aways and enlargements (as Fig. 15-11). Drawings from field
sketches or photographs are made most easily by tracing directly from
the original. Three-dimensional objects are generally shaded as though
light were coming from the upper left corner of the drawing. Ink
dots (stipples) are used to shade rough surfaces (as Fig. 13-4), while
parallel lines are used for smooth or striated surfaces (as Fig. 13-7).
Drawings can also be shaded with lamp black, carbon pencils, or watercolor wash, but they can be reproduced only by photographic processes.
Isometric diagrams. Isometric diagrams show structures in such a
way that the reader can measure distances parallel to the three
mutually perpendicular coordinate-axes on which the drawing is based
(Fig. 11-1). The scales on the three axes are the same, and the two
horizontal axes are typically placed at angles of 60 to the vertical axis
(lined papers of this type are available from suppliers of drafting materials). Other angular arrangements can be constructed. Detailed
instructions regarding both isometric ( axonometric) and oblique dimensioned projections are given by Hoelscher and Springer ( 1956,
chs. 16, 17).
196
197
I'
I
...
Manual of Field Geology
l98
200
200
200
100
100100
l99
Spherical projections. Stereographic and equal-area spherical projections can present large amounts of data relating to structural
orientations. The references cited in Section 15-11 describe some of
these possibilities fully. The final illustration must be labeled sufficiently so that the reader can orient it with the geologic map and
sections or with other relevant structural data.
Mine workings
Fig. 11-1.
it is based.
In addition to the m~terials noted below, a large pad or roll of inexpensive tracing paper will be useful for sketching layouts. If the
illustrations will be reduced, a redw;:ing glass can .. be u~ed to check
line weights and letter heights.
Layout. Illustrations should be composed and arranged before
detailed work is done on them. After their various parts are sketched
on scratch paper, they can be assembled on a drawing board, and the
overall composition can be made by placing a piece of tracing paper
over the assemblage and composing the illustration in pencil. The
distribution of labels and other lettering should be determined at this
stage.
'
Finished liae work. The finished drawing may be made directly on
a piece of top-quality tracing paper or linen placed over the layout
sheet, or it may be made on a piece of opaque drawing paper after
transferring the layout with carbon paper (or by blacking the back
of the layout sheet with a soft pencil). Engineering drafting methods
are used where illustrations require straight or smoothly curving lines,
as described in books on engineering drawing (as Hoelscher and
Springer, 1956, ch: 3). Most natural objects should be inked free-hand
over the pencil guide lines; these lines can be drawn neatly if made a
little at a time. Freely curving lines that must have a precisely even
thickness (as contours and contacts) can be drawn with a contour
pen or a pen from a lettering set. In order .to reproduce well, line
work must be black, not gray.
Shading. Stippling can be dQne evenly if the dots are made in small
spiraling groups (Fig. 11-2). Stipple board, a heavy white paper with
a bumpy surface, gives a roughly stippled effect when rubbed with a
black crayon (as Fig. 15-1 ) . Line-shading of flat surfaces should be
(:;
Hb'
2
()
J\!)M\~
5
,,
,.
!
200
done with parallel, even lines, while the lines on curving or irregular
surfaces should be made to curve back away from the observer. If a
fine, supple pen point is used (as Gillott's 290), the lines can be pressed
thick or thin to increase the effect of depth.
Delicately shaded drawings should be made on a hard white paper
with a satin-smooth surface, as Bristol board. Carbon lamp-black can
be rubbed on this paper to shade broad areas evenly, while a Wolff
carbon pencil can be used for more intricate effects. Drawings can
also be shaded with water washes and a brush, but this technique requires considerable practice. A machine eraser, razor, or china white
poster paint can be used to touch out errors or add highlights.
Printed patterns. Machine-printed patterns (as Zip-a-tone and
Craftint) are available in numerous patterns of lines and dots, as well
as in geological patterns. They should be used wherever a smooth,
flat effect is needed, as on geologic maps. Full directions for applying
them can be obtained from suppliers. When used on maps, the patterns must be chosen so as to contrast suitably with one another, but
must not be so bold or dark as to obscure other features.
Coloring illustrations. Colors can clarify complex illustrations
greatly, and can be used with little added expense if only a few copies
of a report are needed. Line-work, as for faults or other structures,
should be done with colored inks or pencils. Formational areas on
maps can be colored evenly by rubbing the long side of a colored pencil
point in small circles to build the color value up to that desired. Large
areas can be colored evenly by rubbing the pencil on a sheet of fine
sandpaper and applying the colored powder to the map with a bit
of folded cloth. Water colors cannot be used because they wrinkle
all but the heaviest papers, but printer's inks, artist's oil colors, and
some wax-base pencils can be applied lightly by dissolving them in
benzine or any other suitably volatile solvent. These dilute stains
should be applied with a brush or soft cloth and the excess fluid wiped
off the surface at once. The resulting appearance is typically excellent, but the colors cannot be erased and some mixtures cannot be
inked or penciled. It is therefore necessary to experiment with rna-
terials before using them on expensive copy.
Manufactured color sheets (as Zip-a-tone, Craftint) can be used
to give a flat effect where only one copy of an illustration is needed.
Many of these colors, however, are too bright or opaque for detailed
geologic maps.
Lettering. Lettering guides and machines provide a means to do
professional-appearing lettering, and full instructions come with these
201
..
"
202
t,
North
Scale
Legend
0 --D--
--
o-D--
o--D-Jl'
---
-r---
Topographic
symbols and data
Geology by - Dates - - Fig. 11-3.
they will appreciate a full record where it can be given. When exceptional amounts of structural data have been plotted, as with several foliations and fracture systems, it may be necessary to prepare an
accessory structure map; however, the possibility of showing these
data in colored inks on the main map should :S.rst be investigated.
Composing the legend and layout. When the :S.eld sheets are examined during the planning stage, a list should be made of all rock
units, structure symbols, and special culture symbols that will be
shown in the legend of the :S.nal map. These items should be penciled
out in order and to scale on scratch paper. The title, scale bars, north
arrow, and other accessory items should also be penciled to scale.
The :S.eld sheets should then be laid out so as to join, and the various
penciled items arranged with them until an optimum balance and use
of space are achieved. The :S.nal cross :sections should be selected
at this time so that they can be drawn on the same plate as the map
(Fig. 11-3). The map should be oriented so that north is toward
the top, unless it is costly or ungainly to do so.
Tracing paper can now be placed over the assembled parts and a
pencil sketch made of the layout. This sketch serves to guide the remaining steps of the drafting.
Preparing drafting sheets. If more than one color will be used
in printing the map, separate sheets must generally be prepared for
each color separation. They must match exactly at all stages of drawing and printing. The separates should be cut from the same roll or
stock of paper, linen, or acetate. They must be oriented so that their
203
Fig. 11-4. Use of line weight and patterns to show various kinds of features and
data in black ink.
204
be made for each color that will be used, and only the items that will
appear in that color are drafted on that particular sheet. For example,
a separate sheet might be made for drainage (to be printed blue),
another for contours (to be printed brown), and so forth.
When all features and patterns will be printed in one color, widths
and styles of lines must be selected so that the final copy will be clear.
The patterns applied to rock units will indicate which lines are contacts and which are contours (Fig. 11-4). The order of inking is important if erasures are to be kept to a minimum because many features
may cross or overlap one another. The following order may be used:
( 1) structure symbols without numbers, ( 2) faults, ( 3) contacts, ( 4)
cross-section lines, ( 5) numbers for structure symbols, ( 6) culture,
( 7) drainage, and ( 8) contours. Geographic names should be penciled in at an early stage lest they be forgotten. They can be shifted
as necessary so as not to cover important features, and they should
be inked after the contours. The grid should be inked next, but if
the map is crowded, it should be supplanted by short lines at the
margins of the map. Letter symbols for rock units should be inked
last. The order of inking implies that later items are broken to accommodate earlier ones, as shown in Fig. 11-4.
Most maps are completed by inking the title, legend, scale bars
(scale bars should be used whether or not a fractional scale is given),
north arrow, and other accessory items. Prepared patterns (Zip-atone, Craftint) should be applied after the map surface has been
cleaned with a soft eraser and brushed thoroughly. Great care must
be used not to thin the ink work by erasing, particularly where lines
will be covered by applied patterns.
Cross sections. The final cross sections should l;>e made after the
map is inked because they must correspond to it exactly. To make
an accurate section it is necessary to use a sharp pencil and to key
the section paper precisely to the end ticks of the section lines on the
map. Transparent cross-section paper is easier to use than opaque
paper. Completion of the geologic features under the profile line requires a careful analysis of the mapped features near the section line,
for many features will project into the line of section below the surface.
Some suggestions for these constructions are given in Section 3--8.
Lithologic symbols like those in Appendix 5 should be used when
sections will be printed in black only. Patterns for adjacent units
should contrast enough to indicate the positions of contacts. If colors
are used on the map, the cross sections may also be colored, but the
inked lithology should then be simplified. Unit symbols, fault arrows,
205
1000
Fig. 11-5.
Stratigraphic data are generally presented in detailed columnar sections that may represent either single measured sections or an average
sequence for a given area or region. The .basic procedure for constructing columnar sections .is give;. in Section 3--8. The drawing is
made at the smallest scale thai: will show the units ore> beds considered pertinent. Scales will typically range from 1 in. = 100 ft to
1 in. = 10 ft.
The formal units (formations, and so on) or the principal unnamed
divisions should be blocked out first after computing true thicknesses
from the survey notes or maps ( SeQtion 12-8), then the lithologic details are added by starting at the top and working down. If the section
was measured from the base up, this must be done with care in order
not to transpose data. In cases where the field measurements were
made in far greatefl detail than can be used on the column, the field
notes should first be transcribed into a sequence of somewhat generalized units that can be drawn to scale. Lithologic descriptions
should be added to most detailed sections, but lithologic symbols may
be adequate where several small-scale sections are used on one plate.
Serial sections and panel diagrams. If a number of stratigraphic
sections have been measured, they can be plotted together to illustrate
lateral variations in thickness and lithology. If the sections are located
along a simple linear feature, as the outcrop of a homocline, they can
be plotted in a geographic series, spaced in proportion to the distances
'
206
10 Miles
lfRe1pk.3l
207
.!:
References Cited
0
Fig. 11-6. Serial sections, with traverse lines shown on location map.
greatly compared to most full-scale illustrations.
t400 It
Wilton
Simplified
L Mann-3
200
0
Maills
Bornton Sh
Rigby Fm
Canon Crk.
Panel diagram based on drill holes.
most full-scale illustrations.
Fig. 11-7.
ill
Field Work with Sedimentary Rocks
12-1. Interpreting Sedimentary Rocks
209
2JO
11-2).
4. The definition must cite a type locality where the rocks are well
developed and well exposed.
5. The name must consist of a geographic name and a term for the
dominant lithology (as Dakota Sandstone); or, if the lithology cannot
be expressed in a single word, the term for the rank of the unit is
used (as Tobin Formation).
6. The geographic name should be that of a permanent feature at
or near the type locality and must not duplicate a name used for
another unit.
A detailed type section is often described for the type locality, but
this should be omitted if the unit has such lateral variations that a
single detailed section would be misleading. It is often more useful
to include a geologic map showing the continuity of the unit over a
large area.
Established units should not be changed in rank or redefined without
thorough field studies because each change in nomenclature makes
the published record more difficult to use. Sometimes, however, it
may be necessary to subdivide old, heterogeneous units into new for-'
mations or to redefine the contacts or critical features of formations
that were defined loosely. When these changes are considerable, it may
be best to abandon old names and use new ones.
Formal names should not be used for rocks that are poorly exposed
or are found as fragments of complete formations. The contacts of
these rocks may, for example, be faults or may be covered by surficial
deposits. Such units may be useful in mapping, but if named formally,
they will probably require subsequent redefinition. Their usefulness
is in no way impaired by informal names, as shale of Bradford Creek,
Table 12-1.
2JJ
Time Interval
Time-Stratigraphic Unit
.,,
of geologic time terms and the ranks of time-stratigraphic units. The
Stage, the smallest unit used widely for regional and interregional correlations, is established by a careful faunal study of a stratigraphically
continuous, well-exposed sequence o beds. The stage is named after
the locality where the initial study was made. Many stages used in
America are named after local rock-stratigraphic sequences, but European stages are being used more and more for American as well as for
world-wide correlations. Exceptionally thick and fossiliferous rocks
can oftep be subdivided locally by detailed stage classifications; in
the Tertiary of the Pacific Coast :fegion, for example, a sequence of
local stages has been based on microfossils ( Kleinpell, 1938; Mallory,
1959).
Faunizones are generally the smallest units used in a time-stratigraphic sense. They are considered by some to be purely biostratigraphic units because they do not have a usable time value. They are,
however, the basis on which most stages are delineated; thus they
comprise a very important part of time stratigraphy. The kind of
zone used most widely is based on the range of an assemblage of
fossils. The zone is named after a distinctive fossil of this assemblage,
but not necessarily one that is restricted to the zorie (as the Bulimina
corrugata Zone). Conditions of deposition limit the .use of a given
faunizone; zones based on pelagic foraminifera, for example, cannot
be used to classify nonmarine rocks.
The methods of establishing time-stratigraphiC units have been described briefly by Schenck and Muller (1941, p. 1424):
First, the stratigrapher carefully studies a contintfbus section of similar facies; collections of fossils are carefully allocated in this section.
Second, the species are identified, and their stratigraphic ranges in
this section are accurately determined. Third, these ranges are analyzed
so as to show a certain grouping of the strata; these are the arbitrarily
delimited, provisional, time-stratigraphic units. Each such unit contains
species or genera restricted to it; some fossils have their lowest strati-
,_
212
graphic occurrences in it; and some may occur above and below as
well as in the unit. Other species which occur first in the immediately
overlying beds range no higher than the superjacent unit. Fourth,
these time-stratigraphic units are tested by determining a similar distribution of fossils at some other more or less distant locality. Repeated testing proves the validity of the units for the entire geologic
province. The sequence of time-stratigraphic units established on such
a basis would serve as a yardstick, and its application to more distant
areas may be justified if supported by valid paleontologic evidence.
With regard to field procedures, the following points should be noted
especially:
1. The sequence of strata should be continuous. Unconformities
and faults must be detected by mapping around each provisional
section.
2. Fossils must be located accurately. The precision of measurements must be well within the limits of the faunal changes used.
3. The rocks must be of one facies. Lithologic studies must determine whether physical conditions were uniform; reworked materials should be noted with care so that fossils of two ages will not
be mixed.
Relations between lithologic and time-stratigraphic units. The
time interval represented by a vertical section through a formation
may vary considerably from place to place. Even where the interval
is constant over a large area, the contacts of the formation are not
likely to coincide with the boundaries of standard stages or series.
Lithologic and time-stratigraphic units cannot, therefore, be used interchangeably. Where fossils are so abundant that ages of rocks can
be determined at many outcrops, it may be possible to plot lines showing boundaries between time-stratigraphic units. These lines must,
however, be distinguished clearly from contacts between lithologic
units.
In some regions, extensive "formations" or "groups" may have been
defined on the basis of age, or "series" or "stages" may have been set
up and traced on the basis of unconformities or lithologic changes
that are not necessarily time horizons. It is then necessary to sort out
and understand the existing nomenclature before beginning a new.
field study.
12-3. Naming and Describing Sedimentary Rocks
213
meaning. Some names, as limestone, have only a compositional meaning; others, as sandstone, have only a textural meaning. Most sedimentary rocks, however, ~an be given names that are consistent and
meaningful within any one rock group. Several such systems are
presented briefly in this section. Not all the terms used are universally
standard, and therefore field notes should include actu.!l) quantities
that express textural and compositional characteristics. The basic
characteristics are: grain size, degree of sorting of detrital grains,
shapes of grains (especially degree of rounding by_ abrasion), fabric
of shaped grains, porosity, color, composition of the detrital fraction,
nature and amounts of authigenic materials (as cements), and degree
of recrystallization or replacement.
Grain size. The primary basis for I]Jlming detrital rocks is grain
size. The size scale proposed by Wentworth, used widely in America, is the basis for the classification shown in Table 12-2. The numbers in the table are the limiting sizes of the various size classes. The
pebble and sand sizes are divided into equal classes to facilitate determining the degree of sorting, as explained below. Pebbles can be
measured with a scale, while sand sizes are best determined by comparing a given rock to a sand-size card. These cards can be made by
Table 12-2.
Exact
Size
Limits
>256
64-256
32- 64
16- 32
8- 16
4- 8
2- 4
1- 2
lh- 1
mm
mm
mm
mm
mm
mm
mm
mm
mm
.! mm
~mm
116- ~ mm
.!56- l;iB mm
<lh5B mm
*-
Approximate
Inch
Equivalents
>10
2.5
-10
1.2
- 2.5
0.6
- 1.2
0.3
- 0.6
0.15
- 0.3
0.08
- 0.15
0.04
- 0.08
0.02
- 0.04
0.01
- 0.02
0.005 - 0.01
0.002 - 0.005
0.00015- 0.002'
<0.00015
N arne of Loose
Aggregate
in.
in.
in.
in.
in.
in.
in.
in.
in.
in.
in.
in.
in.
in.
Boulder gravel
Cobble gravel
Very coarse pebble gravel
Coarse pebble gra~el
Medium pebble gravel
Fine pebble gravel
Granule (or very fine pebble) gravel
Very coarse sand
Coarse san~
Medium sand
Fine sand
Very fine sand
Silt
Clay (clay-size materials)
214
Very
angular
Well
sorted
Moderately
sorted
Poorly
sorted
Very poorly
sorted
Fig. 12-1. Terms for degrees of sorting. The numbers indicate the number of sizeclasses included by the great bulk ( 80 percent) of the material. The drawings
represent sandstones as seen with a hand lens. Silt and clay-size materials are
shown diagrammatically by the fine stipple.
Subangular
Subrounded
Rounded
Well
rounded
Fig. 12-2. Terms for degree of rounding of grains as seen with a hand lens. After
Powers, M. C., 1953, Journal of Sedimentary Petrology, v. 23, p. 118. Courtesy
of the Society of Economic Paleontologists and Mineralogists.
llllrl~~
Very well
sorted
Angular
215
ti
I
j
...
216
larger pores can often be determined by hand lens study. In calcareous rocks, for example, some pores can be seen to be irregular channels formed by fracturing, leaching, or alterations (as dolomitization);
other pores may be partly filled openings in fossils. Porosity in
detrital rocks may depend on sorting and the degree to which grains
have been packed and cemented. Porosities are commonly increased
or decreased greatly by weathering. Fresh outcrop samples can be
used to estimate porosity at depth; however, the freshest rocks at a
given exposure may be misleading since they are generally those with
the lowest original porosity.
Color. Color can be used effectively in mapping and correlations
if its causes are understood. Rock color depends not only on the color
and grain size of the original sediments but also on their cementing
materials, the degree to which they have recrystallized, and, especially,
the degree to which they are weathered. The secondary origin of
colors can often be detected because the pigmenting minerals occur
along fractures, cut across bedding, or are related to the surface of
the ground. Even moderate weathering may change colors so much
that fresh colors can be seen only in deep cuts, tunnels, or well cuttings.
Color terms should be standardized by comparing rocks to color
charts, such as the rock-color charts that can be purchased from The
Geological Society of America, or the soil-color charts of the Munsell
Color Company (see Soils, Section 12-11). Colors for both dry and
wet samples should be recorded.
Naming mixtures of materials. Detrital rocks made up of grains
of various sizes derive their name from the size class that is most
abundant (except for certain conglomerates and breccias). The
names in Fig. 12-3 may be used for various mixtures of detrital and
nondetrital materials. Mixtures containing substances not shown in
the diagram can be named in a similar way, as ferruginous sandstone,
gypsiferous claystone, and so on.
Sandstones. Rocks consisting mainly of sand-sized grains are especially useful in field studies because their textures and compositions can
be determined readily. Furthermore, an average hand specimen is
ample for petrographic and permeability studies. Names for sandstones should be based both on degree of sorting and on composition,
as in the system proposed by Gilbert (Williams, Turner, and Gilbert,
1954). In this system, the name arenite is used for relatively wellsorted sandstones, and the name wacke for more poorly sorted sandstones (Fig. 12-3). These rocks may be classified further on the basis
iii
g-)..0
2J7
-'
.!.!
........
v
<))
liibOro.:!
t:(al
'"OP,..
ct
Ln
~itt
-~<::
Q)
:r-- --~:e
OEt:
.s"'
:g-5
::J
Ill
Q)
::J
Q)
Q)
<::
aa
Ln
.....
....,..........
~ ~
~.!!l.2
cn~u
Ill
Q)
C;
"' .
"'-.
llr.;
E-<Qf
u"'
Q)
::JC:~
en
g .s ~
- ;3
]E-<
>-
&.
C;
giDo
c:;
iii
.;
en::
---
---
::I:
...
O'o
Q)
----
"'
8~
c~
"'~
'
...."'~
s
t:i..Q
Q)
Q) 0
Q)
-' 0
g u;
s
Q)
.Elroo
' ~
0
uo
u-::no
..!S!QJO
:::E-o
e.o=
ct
t;;
~~~
$
'
~~
v; ~
~u
:ii
cn
.-:.
.:as~o
~ r:-
.sc.,c:
u
-gro
....,
'iii
C;
-o
v fj)Ol
rn
Pt
~ c:5 :-
-8asv... 8
>~"'t:='
"'c
,:_;~"'~a
C;
...
Q)~
-g
~
~
ct)_
Ill
=s :+=o Q)
o- c
.,
E o
"' Q) 0
en
E-o
:.::s
'-0-
~~4i'
~B
U '0- "Vl'
~=-g
~ 2.0 co
iil;,ltt.
"'(/)
0
.Ci
~0-{g
ro-oc
Oto/J
@ .s
Q)
. . . .s<::
~4Q
,;~
u-o
.!!!
o.-<
roo
Ill
~5~
~ >-
c:
.,=Q
~0
'"0'"0
._
--
"'
-o
O.s
>.:::g
"iil<i
--
._
Uo
c::M
'"0,
/ttiii,J
Ln
' .....
00
~.......
co ..
oc:3
.sc~o
----
.!!!>
2::9
,_.
/Jtt"
.<:::
: -g"
-~">;.;]
t:
Q)
u;
'"0""
;:;10
-a>
() .....
- ::J <0
C"-'=
::J
.....
<::
-0
Q)
"'
c::
Vl
i:o:t
v o; 8
;3 .., v(l.)
"'""'~
-~
8 .~
. . ~:t
..E s .
"'..s0 ~
Q)
' 8
3:"'"'
>.
"'
d
~
/Jtte- "'
"'~'vtea
'"t:l o"
...
0 -~
v ()
g :S r~
c/o itt.
iii,J-1&
IX: til-
Q'
-o
<::
8l
l-
()
4-<
"'
Q ..<::~
-~
"'
i;).
bh E
218
Quartz, quartzite,
and chert grains
Quartz arenite or
';.,--- (rarely) wacke
90% quartz-
oz
)~.;~lj~~thic I
/.?
~E&b'f;t?ke
il~~i~;iiW
. ,::;.::
Feldspar
grains
Lithic
arenite
arenite
or wacke
219
Fig. 12-4.
Gravel
Well-sorted
conglomerate
or breccia
or
breccia
Sandy
conglomerate
or breccia
Muddy
conglomerate
or breccia
Sand
Fig. 12-5.
9:1
Pebbly
mudstone
1:1
Ratio of sand to silt plus clay-size materials
220
may be far less abundant. Both the matrix and the gravel-size fragments should be described.
Most pebbles can be identified as rocks, and therefore conglomerates
are particularly valuable in determining sources of sediments. The
large fragments should be tabulated by systematic counts, as by marking off an area on an outcrop and identifying all the pebbles and
cobbles that fall within its perimeter. Fragments that are physically
weak (as claystones) or chemically unstable (as limestones) can
provide evidence as to the nature of the transporting agent or the
amount of transport. Surface effects such as polish, scratches, sandblast fluting, chemical corrosion, or abraded relics of old weatheringrinds can be used to interpret conditions of transport and deposition.
Mixtures of well-rounded and angular fragments, or presence of
broken rounded pebbles, may indicate reworking of older conglomerates. Conglomerates and breccias can form in many ways, and
accounts such as those by Pettijohn ( 1957) or Dunbar and Rodgers
( 1957) should be read before studying these rocks.
Limestones and related rocks. Although they are commonly recrystallized or replaced to varying degrees, limestones should be named
as much as possible on the basis of primary textures. This can be
done by examining many samples with a hand lens, both on freshly
broken and on slightly etched, weathered surfaces. The textural constituents of the rock should first be classified into the following groups:
1. Abraded biogenic grains, limestone fragments, and accretional
grains (ooids, pisoliths).
2. Unabraded fossils and fecal pellets.
3. Fine-grained carbonates (originally calcitic or aragonitic mud). ~
4. Noncalcareous detrital grains.
5. Cements (especially phanerocrystalline calcite).
Materials that have been worked and deposited by currents will fall
dominantly in groups 1, 3, and 4, and rocks consisting of these materials
can be named on the basis of their size sorting (Table 12-3). In addition to their textural characters, these rocks may show laminations or
cross-bedding. They are commonly interbedded or intermixed with
size-sorted, noncalcareous rocks, and if the limestones have recrystallized greatly, this relation may provide the only means of determining
their origin. Limestones with gravel-size clasts can be called limestone conglomerates or calcirudites, but if the large clasts are fossils,
such names as shelly calcarenite and spirifer calcarenite are generally
more useful.
Calcarenite
22J
Finegrained
l.isestone
9:1
1:1
1:9
Ratio of sand size to finer-grained detrital carbonates
-x
Table 12-4.
Limestone
Dolomitic limestone
Calcitic dolomite
1:1
9:1
1:9
Ratio between the minerals calcite and dolomite
Dolomite
j,
I
222
;,',~;',.~~-
Set of
cross-strata
Set of
cross-laminated
sandstone beds
Fig. 12-6.
223
Set of graded
silt-claystone
beds (varves)
Laminated shale
with %-in.
sandstone beds
i,~('\');;
. . . :2~+
';"';''(..,,,
. ......;.:
Fig. 12-7.
rocks.
"'1.
224
Fig. 12-8. Cross-bedded rocks. (A) Tabular sets with diagonal patterns. (B)
Wedge sets, showing considerable erosion between each set. (C) Tabular to
lenticular sets with tangential patterns; typically, these are laminated marine beds.
( D) Symmetrical trough sets with distinctly linear axes; typically, these are largescale fluvial features. The arrows indicate current directions.
225
of laminae or other subordinate bedding features. In, all cases, however, the precision of the estimates or measurements should be gauged
suitably to the regularity of the beds and the purpose of the study,
for measuring beds can become involved and time consuming.
Internal features. A bed may be internally homogeneouS"( structureless or massive); it may be graded, or it may be laminated in a variety
of ways (Fig. 12-6). Internal features can be used to interpret conditions of deposition as well as stratigraphic tops of beds. The dips
of cross-strata or cross-laminae give a valuable measure of the current
direction. This direction should be plotted on the map as an arrow
rather than a strike and dip symbol. Wherever possible, the average
direction of current flow and the degree of variance from this average
are determined by taking readings on several beds at each outcrop.
If the readings vary no more than about 10 from the average, a single
arrow may be plotted. If the variance is greater, the individual readings may be plotted as a radiating set, or may be recorded for more
thorough analysis in the office. Plotted dip arrows should represent
cross-bedding of one type, especially where rocks are not of marine
origin. Trough-type sets, for example, show a great range of dip
directions; where exposures are adequate the trough itself is the most
useful linear element. Lowell ( 1955) has described a field study that
utilized various types of cross-bedding.
Where fine sand or silt accumulates on a current-rippled surface,
it may build up a strikingly laminated bed by accumulating on the
downstream slopes of the ripples (Fig. 12-9A). This type of bedding
gives an excellent measure of current flow and depositional conditions,
but unlike true cross-bedding, the layers of superimposed crosslaminations dip upstream.
Folding and other internal flow structures in beds can sometimes be
used to ascertain the direction of slope of the original deposits (Fig.
12-9B ) . The nontectonic origin of these structures can generally be
proven because the beds are intercalated in sequence with undeformed
beds. Commonly, the upper surfaces of deformed beds have been
eroded moderately before the overlying beds were deposited. U ndeformed animal borings can also be used to determine the early deformation of beds (Fig. 12-9C). Slope and current directions need
not necessarily be parallel, and therefore a distinctive arrow should
be used to plot each on the map.
Abundant filled animal borings ("worm tubes") are characteristic of
some sequences. The slumping of fine laminations into large borings
may give an indication of tops of beds (Fig. 12-9D).
226
227
Current
~-~~
~/;}'' ?2""''2:..
Fig. 12-10. Arrangements of large fragments in cross sections of beds. (A) Bedding-plane orientation of mudstone fragments in Upper Cretaceous sandstone of
Santa Ana Mts., Calif. Note reverse size-grading of fragments. From a photograph by C. A. Hopson. (B) Imbricate structure. Planar cobbles tend to dip
upstream and linear cobbles tend to plunge upstream. Generalized from dissected
terrace gravels of Pacheco Pass area, Calif. (C) Stable position of large clam
shells relative to current. ( D) Orientation of clam shells in a graded Paleocene
sandstone bed in the Arroyo Seco, Monterey Co., Calif.
I,
I
228
229
...,
c~;;g:::: . ~
Fig. 12-12. Linear structures formed by currents. ( A) Flute casts with prominently
beaked ends on bottom of Eocene sandstone beds west of Stanford, Calif. Courtesy of E. A. Schmidt. (B) Simple flute casts on bottom of Upper Triassic sandstone bed from Santa Rosa Range, Nev. (C) Current groove casts, one set superimposed on another; Upper Triassic near Izee, Oregon. From a photograph by
W. R. Dickinson. (D) Subdued current lineations on top of sandstone bed. Generalized from various sandstone flags. See also Stokes ( 1953). All blocks are
1 ft across.
preserved in three dimensions because the underlying shale or mudstone is almost always eroded more readily than the overlying sandstone. The structures occur commonly, however, as reversed threedimensional casts on the base of the overlying sandstone beds. The
terms used most often relate almost entirely to these reversed forms.
Flute casts are elongated in the direction of the current and have a
roughly triangular, spoonlike, or beaklike shape, with the more prominent or pointed end directed upstream (Fig. 12-12A and B). Groove
casts are simpler elongate features that give only the orientation of
the current (Fig. 12-12C). Crowell (1955), Kuenen (1957), and
Wood and Smith ( 1959) have described these features in detail, particularly emphasizing their relation to, turbidity currents.
Another type of current lineation that forms in laminated or thinbedded sandstones consists of subdued lines or slabby, linear plates
and low ridges on bedding cleavage surfaces (Fig. 12-12D). This
230
A
231
Fig. 12-13. Idealized sections of cut-and-fill and loading structures. (A) Simple
cut-and-fill in laminated shale. Note concentrations of coarser materials. (B)
Cut-and-fill with moderate loading, as shown by bending of laminae. (C) Loading structures with little or no cut-and-fill.
linear structure is probably formed by accumulation and streamlining of grains into trains and windrow-like ridges by currents. Although
it may not be a cut-and-fill structure, it is included here for sake of
comparison with flute and groove casts.
Load casts. Where water rich muds are overlain by a layer of
coarser detritus, there is a tendency for the coarser materials to pocket
or load downward into the mud where the surface between them is
somewhat depressed. The original depressions can be formed by
current cutting, by rippling, by movements of animals, or by unequal
compaction, and therefore the downward-loaded bodies of sand may
have any shape from distinctly linear to equidimensional or irregular
pockets. They are called load casts because they are typically preserved on the base of sandstone beds. Crowell ( 1955, p. 1360) suggested that linear load casts may result directly from cut-and-fill where
a turbidity current is heavily charged with detritus. If laminations or
large sand grains can be found, relations such as those shown in Fig.
12-13 can indicate the relative degree of cutting and loading. In any
case, load casts can be used to determine stratigraphic sequence.
Structures formed by creep, drag, and sliding. Sediments can be
folded or disrupted by creeping or sliding downslope, and materials
deposited from dense turbidity currents may be contorted as the
currents drag against them. These movements commonly overturn
projections of mud between sand pockets (Fig. 12-14A). With
greater movement, the pockets may be complexly folded or rolled
(Fig. 12-14C). Internal laminations are an important key in determining the sense of these movements (Figs. 12-9B and C). Gravity slidingmay fold, disrupt, and interject sediments on such a large scale that
the effects can be seen only at large outcrops (Fig. 12-14B).
Fig. 12-14. Sections showing creep and flowage features, with sense of movement
toward left. (A) Asymmetric sand pockets at base of graded, laminated, and
rippled sandstone bed. Upper Cretaceous of Santa Lucia Range, Calif. (B)
Slump overfold of mudstone bed; Grindstone Crk., Glenn Co., Calif. After Crowell
( 1957, p. 1000). (C) Flowage folds, rolls, and pull-apart structures in sandstone
beds in argillaceous materials; there is a moderate lineation at right angles to the
page. Large core of Upper Pennsylvanian rocks from Runnels Co., Texas; courtesy
of M. K. Blaustein.
232
. 233
sometimes be detected by comparing structures of the younger sequence with those of the older (refolded) sequence (Sections 15-7
and 15-11).
Unconformities without appreciable angular discordan~ can be detected by the following means:
'
Fig. 12-15. (A) Distinctive small animal markings on bottom of fine-grained sandstone flag; Upper Triassic of Santa Rosa Range, Nev. ( B ) Tubular prominences
on bottom of Paleocene sandstone bed from Santa Lucia Range, Calif. (C) Mudcracks in silty claystone, filled by sandy claystone; from Colorado Plateau region.
(D) Cast of dinosaur footprint on bottom of red sandstone bed from Portland, Conn.
cussion marks of hail and rain, and impressions of ice or salt crystals
(Fig. 12-15C). As with other surface markings, these forms are
most commonly preserved as casts on the base of silty layers that overlie the layers of mud on which the marks were made originally.
12-6. Unconformities
234
c~<.-.:~::::;:.~i..
:-,
~~.
. . :.o.l~f
Fig. 12-16. Fossils as indicators of tops of beds. (A) Bryozoa, barnacles, and
other encrusting animals on exposed tops of shells and cobbles. (B) Rudists or
rudist-like pelecypods in growth position. (C) Shells of Schizothaerus (Tertiary)
or Pholadomya ( Mesozoic) may be found standing in living positions (left),
whereas inequivalved .Pelecypods will lie on their more convex valve if buried
alive. (D) Starfish and echinoids such as sea urchins tend to lie with their flat
ventral (oral) side down. (E) Root-like structures of crinoids branch downward.
(F) Solitary corals attach at base and branch outward and upward; colonial corals
and calcareous algae tend to grow outward and upward to give a convex form
facing upward (but not on steep or overhanging surfaces of reefs). (G) Animal
holes drilled in rock substratum tend to open upward. (H) Partial fillings in
cavities of buried shells indicate original dip of deposits. Condensed and modified from Superposition of strata by S. W. Muller ( 1958, Stanford University),
reproduced in Geotimes, 1959, v. 3, nos. 5 and 7.
235
I.
236
~~
~-:z
c -
g.._-.. _~------~--.~
Ideal geometric relations for measuring strata (top) compared to a
region where low dip, low relief, and variable thicknesses make measuring difficult.
Fig. 12-17.
basis of the degree of detail required, the physical nature of the terrain
and outcrops, and the time, funds, equipment, and personnel available.
Measurements and descriptions of thick unfossiliferous formations, for
example, need not be nearly as precise as those used to correlate fossiliferous units with subsurface data. The precision of the measurements is determined partly by the surveying methods and partly by
the geometric relations between the ground surface and the dip of the
rocks. Where the stratigraphic sequence is crossed at high angles
by the ground surface, precise measurements can be made far more
quickly and easily than when the two are nearly parallel (Fig. 12-17).
The amount of detail to be recorded and the spacing of fossil and rock
samples must be considered carefully when the project is planned;
in general, it is wiser to collect too much than too little.
Measuring sections from maps or photographs. Sections may be
measured and described by plotting a sequence of contacts and points
on a topographic map or aerial photograph. Because the scales of most
maps and photographs are too small for detailed work, this method is
used mainly to reconnoiter more complete studies, to measure gross
cartographic units, and to check total thicknesses accumulated by
more detailed measurements.
Compass-pace surveys. This method is described in detail in Chapter 3. It is generally adequate only where the ground surface is
level or slopes gently, as along a road, a smooth ridge, or by a waterway.
Its great advantage is that it can be executed by one man. Details
can be accumulated more quickly and accurately if a tape or graduated
pole is used to measure beds in cuts.
Measurements with the Jacob's staff. A Jacob's staff is a pole about
5 ft long graduated to feet and tenths of feet. It is used to measure
237
thicknesses of beds directly at the outcrop, and with it one man can
readily accumulate a detailed stratigraphic section where units are
and where they meet the ground surface at high angles.
well exposed
The measurements are started by placing the staff on the lowest exposure and orienting it at right angles to bedding (Fig. 12-18A). The
observer sights parallel to the dip of the beds across an even mark on
the staff, and notes the point on the outcrop that falls in his, line of
sight. The stratigraphic thickness to this point is equal to the- distance
subtended on the staff. The observer then moves uphil_~ and places
the staff either on the point sighted or on any other point on, the same
bedding surface, and he continues to measure the beds by repeating
the steps just described. If the staff can be oriented by estimatio~
so that it is within 8 or 10 of normal to the beds, the error in a 5-foot
measure will be 0.1 ft or less. For greater precision:"'it is necessary to
orient the staff and line of sight by using a sighting bar and clinometer
(Fig. 12-18B), or by holding or attaching a Brunton compass to the
staff. Dips of beds must be checked frequently, a procedure that is
facilitated by having one man do the measuring while a second checks
attitudes and describes the lithology.
Exposures are rarely continuous enough to be measured in a straight
line directly upslope, and even where they are, offsets must be made
to fossil localities. The Jacob's staff method enables these offsets to
be made easily. It is only necessary to walk along the trace of the
last bedding surface sighted and then to resume measurements at the
most useful outcrops (Fig. 12-18C). Because of the zigzag nature of
the course, important specimen localities must be described in detail.
238
Fig. 12-19. Plane table control for measurements across gently dipping beds, allowing for many offsets without reduction of accuracy.
239
_.,;
staked traverse course must be set up and where slopes are too steep
and rugged for an efficient plane table traverse. A small (mountain)
transit is ideal for this work, and the methods used are basically the
same as those described in Section 7-13. The stratigraphic details
between instrument stations may be surveyed by stadia methods or
by reading vertical angles and taping distances directly on the slope, as
described below.
Brunton and tape methods. Brunton-tape surveys are generally
precise enough for measuring stratigraphic sections where beds dip
more than 15 and intersect the ground surface at a fairly high angle.
Measurements are made most easily in a downhill direction. The
procedure that follows refers to a two-man party; however, one man
can make the measurements if the tape is anchored at the 0 end.
1. One man holds the 0 end of the tape at the starting point while
the second man carries the other end of the tape to the first break in
slope.
2. The slope angle is read and checked by both men.
3. The tape is secured on the slope (as by tying it to a stake), and,
starting at its 0 end, the rocks are examined and described; their positions are recorded according to measurements (slope distances) read
directly from the tape (Fig. 12-20).
4. Strikes and dips are measured as necessary, and the angle between the strike and the section line is measur.ed and recorded.
5. The 0 end of the tape is moved downhill to the first break in
slope, and steps 2 to 4 are repeated.
Where outcrops are rugged, or where beds meet the surface at close
to 90, the tape can be used as a Jacob's staff to measure true thicknesses directly. These thicknesses must be recorded as true thick-
'
240
24J
tl.
nesses. In any case, terms such as overlying and above must not be used unless they clearly refer to stratigraphic rather than to topographic position.
Corrections for slope angles and oblique sections. Unless true
thicknesses are measured directly, as with the Jacob's staff, they must
be calculated from the slope distances. If the measured slope distance is oblique to the dip of the beds, it must first be corrected by
the following formula (based on the notation shown in Fig. 12-21).
AC
= AB
cos a
= AB
Stn (r'~-y)
8C = AB cos (J'+Ij-90"}
or Be =Aa sin [iro- (x+yJ}
8
= AB
sin {K-'f)
BC
= AB
.sin {'f-X)
A
Thickness
Fig. 12-22. Forrnulas used foc the various possible combinations of direction and
amount of ground slope and dip of beds.
242
at this stage. If it is known, for example, that only the smallest calcareous foraminifera will be useful in a certain part of the section, the
finest claystones are collected in preference to silty shales that might
show large arenaceous foraminifera. Ideally, the sample localities at
this stage are scattered widely, both stratigraphically and parallel to
the strike of the beds. If there are thick, fine-grained units in which
no fossils have been seen, several samples of the most promising ( generally, the finest grained) materials should still be collected.
The faunal determinations from the first group of samples may solve
many stratigniphic problems, but they usually also demonstrate a need for additional sampling. More samples may be needed along certain
contacts or faults, or where relatively thin parts of a section appear to
contain many faunal zones or several stages. Unusually well-exposed
sequences of fossiliferous rocks might be sampled in great detail, both
stratigraphically and laterally, to test the validity of tentative faunal
zones or local stages (Section 12-2).
The stratigraphic measurements described in Section 12-8 are generally used as a basis for systematic collecting. The following suggestions apply to the samples themselves:
1. If calcareous fossils will be used, the samples should be fresh
(although moderately weathered rocks are sometimes useful because
they are friable).
2. Samples must be large enough; unless fossils are obviously abundant, 10 lb of material may be needed.
3. Locations must be plotted precisely on a map or described completely in the notes.
4. The stratigraphic position must be known. It will be determined
automatically if the section is being measured; otherwise the stratigraphic distance to the nearest contact or key bed should be measured
directly.
5. The source of the sample should be marked at the outcrop so
that more material can be collected later.
6. Sample bags must be silt-tight and marked on the outside so
they will not have to be reopened until they are in the laboratory.
12-10. Logging Wells and Drill Holes
'
'
,,
I
~
Ji
~
Ij
:\
I
~
24~
and reopening samples tend to increase the possibilities of contamination and mis-identification. The geologist should oversee these procedures if possible, and he should also make sure that as much material is saved as possible. Before sampling a drilling well, he should
_determine the exact drilling methods, the typical rates of drilling in
various units, and what is the expected performance of sampling
equipment available. These aspects of well logging have,.been described in LeRoy's Subsurface Geologic Methods ( 1950), as well as in
subsequent journals.
Logging cores. Large amounts of core can generall)\ be-logged by
the following steps:
1. Make a graphic log, either with lithologic symbols or colors, that
blocks out the main rock units and contacts and also indicates occurrences of fossils, key beds, and shows of petroleum or zones of high
porosity.
2. Fill in the uncored gaps in this record by reference to the driller's
log and other accessory logs.
3. Compare the graphic log with the expected stratigraphic or
structural sequence.
4. Re-examine the cores as necessary to make more detailed descriptions of rock units, to inspect critical intervals, contacts, and
structurally anomalous relations, and to collect samples.
'
245
proof bags and labeled as to well name and number, core number,
footage or interval, rock name, formation (or other unit name), and
probable stratigraphic position, as determined from the core.
The data in lithologic descriptions of cores should be recorded in
order of importance. The amount of detail needed will vary with
the project; somewhat more should be collected than appears necessary. Colors of cored rocks are much more diagnostic than those of
surface outcrops, and these colors can be recorded precisely by using
detailed rock color charts ( p. 216). Horizons of secondary coloration should be noted because they may indicate faults or unconformi-
ties. Porosities should be described completely.
Logs from cuttings. Lithologic logs of most wells are made by examining the cuttings brought up in the circulating drilling muds or
fluids. These materials are evolved continuously while drilling is in
progress, and samples are generally taken at certain time intervals or
at every 5 or 10 ft of depth drilled. Cuttings must be washed thoroughly; this is done by repeatedly decanting rinse washes until they are
free of clay-size materials. The samples are then dried, placed in
dust-proof bags, and labeled with the well description, driller's name,
and footage or time interval. If the footage has been corrected by
taking into account the circulation time from bit to surface, this should
also be noted on the label.
The size and appearance of cuttings vary greatly with drilling methods and local lithology. Cuttings from the bit are generally mixed to
some degree with fragments that have been caved or abraded from
higher parts of the hole. Picking contacts or key beds is not difficult
if rock units are thick and distinctive. The upper contact of a formation is drawn in the log at the first appearance of a cutting or of disaggregated grains from that formation. Where rocks grade into one
another, or where they are intercalated through considerable thicknesses, the local section must be known in detail.
Moderately friable clastic rocks are usually disaggregated by the
bit; thus a binocular microscope should be used to examine cuttings.
Samples stained with petroleum or asphaltic residues should first. be
cleaned with a volatile solvent. Accessories needed for most examinations include dilute HCl, chemicals required for staining tests, flat
evaporating dishes for examining materials, dissecting needles and
forceps for separating fragments and testing hardness, and a grid
ocular for measuring grains and estimating their amounts. Graphic
logs and brief descriptions are customarily entered on strips of paper
3 in. wide and 40 or 60 in. long. More detailed notes and comments
may be recorded in an accessory notebook that is organized in the
same order as the plotted log.
244
~
Manual of Field Geology
( 1951), and this book should be consulted for terms and information
relating to soils and soil profiles.
The total soil profile can be a complete gradation from surface soil
to unaltered rock or sediment, or it can consist of various layers that
may be defined sharply or may grade to one another. Field descriptions should give the nature, thickness, and regularity of each of these
layers and gradations. Where soil profiles are being cowpared or
correlated from one place to another, it is useful to classify the principal layers, customarily called the A, B, and C horizons of the profile.
The upper or A horizon i'S characterized by one or .more layers in which
organic materials have been concentrated and from which soluble
materials and colloids have been leached; it is typically grey to brown,
friable, or very rich in plant materials. The B horizon consists of a
layer or layers in which materials ( especially colloids) transported from
the A horizon are concentrated. Additions of clay and hydrated iron
or al~minum oxides make it more compact and more red, yellow, or
brown than the A horizon; however, additions of organic materials
from above make some B horizons gray or gray-brown. The C horizon
lies under the soil proper and consists of disintegrated rock or loose
sediments that are more or less weathered and are clearly related to
the overlying soil. In many areas, the upper part of the C horizon is
characterized by concentrations of calcium carbonate and other
soluble salts, which may be so abundant as to form a hard caliche layer.
The relative ages of soils and the land surfaces on which they lie
can sometimes be judged on the basis of thickness, degree of leaching
of carbonates, degree of disintegration of certain rocks, or by the degree
of argillization and the compactness of the B horizon. These criteria
must be modified and weighted, however, where downslope creep and
sheet wash have thickened soils on slopes and in vaJleys and have
thinned them on ridges, or where variations in underlying rocks affect
soil characteristics. In buried ancient soils, only the B horizon or some
part of it is generally preserved. Soils at different outcrops must then
be identified or correlated on the basis of texture, clay content, color,
and compactness (indicated in part by a tendency to develop columnar
or platy fractures in outcrop). Color is such an important measure of
oxidation and concentration in the B horizon that precise color terms
should be used. Standardized soil color charts are available from
the Munsell Color Company, 2441 North Calvert St., Baltimore, Md.
These. charts give a wide range of color names and symbols that can
be used for rocks as well as for soils.
246
!*
I
t
247
r
iI
-
248
References Cited
American Commission on Stratigraphic Nomenclature, 1961, Code of stratigraphic
nomenclature: Bulletin of the American Association of Petroleum Geologists,
v. 45, pp. 645-665.
Badgley, P. C., 1959, Structural methods for the exploration geologist: New York,
Harper and Brothers, 280 pp.
Bradley, W. C., 1957, Origin of marine-terrace deposits in the Santa Cruz area,
California: Bulletin of the Geological Society of America, v. 68, pp. 421-444.
---, 1958, Submarine abrasion and wave-cut platforms: Bulletin of the Geological Society of America, v. 69, pp. 967-974.
Crowell, J. C., 1955, Directional-current structures from the Prealpine Flysch:
Switzerland: Bulletin of the Geological Society of America, v. 66, pp. 13511384.
---, 1957, Origin of pebbly mudstones: Bulletin of the Geological Society of
America, v. 68, pp. 993-1010.
Dunbar, C. 0., and John Rodgers, 1957, Principles of stratigraphy: New York,
John Wiley and Sons, 356 pp.
Frye, J. C., and A. R. Leonard, 1954, Some problems of alluvial terrace mapping:
American Journal of Science, v. 252, pp. 242-251.
Hewett, D. F., 1920, Measurements of folded beds: Economic Geology, v. 15,
pp. 367-385.
Howard, A. D., 1959, Numerical systems of terrace nomenclature: a critique:
Journal of Geology, v. 67, pp. 239-243.
Ingram, R. L., 1954, Terminology for the thickness of stratification and parting
units in sedimentary rocks: Bulletin of the Geological Society of America, v. 65,
pp. 937-938.
Kleinpell, R. M., 1938, Miocene stratigraphy of California: Tulsa, American Association of Petroleum Geologists, 450 pp.
Krumbein, W. C., and L. L. Sloss, 1951, Stratigraphy and sedimentation: San
.
Francisco, W. H. Freeman and Co., 497 pp.
Kuenen, P. H., 1957, Sole markings of graded graywacke beds: Journal of Geology,
v. 65, pp. 231-258.
Leopold, L. B., and J. P. Miller, 1954, A postglacial chronology for some alluvial
valleys in Wyoming: U.S. Geological Survey, Water-Supply Paper 1261, 90 pp.
LeRoy, L. W. (editor), 1950, Subsurface geologic methods, a symposium, 2nd ed.:
Golden, Colorado School of Mines, Dept. of Publications, 1156 pp.
Lowell, }. D., 1955, Applications of cross-stratification studies to problems of ura-.
nium exploration, Chuska Mountains, Arizona: Economic Geology, v. 50, pp.
177-185.
McKee, E. D., and G. W. Weir, 1953, Terminology for stratification and crossstratification in sedimentary rocks: Bulletin of the Geological Society of America,
v. 64, pp. 381-390.
Mallory, V. S., 1959, Lower Tertiary biostratigraphy of the California coast ranges:
Tulsa, American Association of Petroleum Geologists, 416 pp.
Mertie, J. B., 1922, Graphic and mechanical computation of thickness of strata
and distance to a stratum: U.S. Geological Survey Professional Paper 129-C,
pp. 39-52.
i
t*
l
I;
249.
Pettijohn, F. J., 1957, Sedimentary rocks, 2nd ed.: New York, Harper and Brothers,
718 pp.
Phillips, F. _c., 1954, The use of stereographic projection in structural geology:
London, Edward Arnold Ltd., 86 pp.
Prentice, J. E., 1960, Flow structures in sedimentary rocks: Journal of Geology,
v. 68, pp. 217-225.
Schenck, H. G., and S. W. Muller, 1941, Stratigraphic terminology: Bulletin of
the Geologic{ll Society of America, v. 52, pp. 1419-1426.
.
Shrock, R. R., 1948, Sequence in layered rocks: New York, McGraw-Hril Book Co.,
507 pp.
Soil Survey Staff, 1951, Soil survey manual: U.S. Dept. of Agriculture I:landbook
.,.
No. 18, 503 pp.
Stanley, G. M., 1936, Lower Algonquin beaches of Penetanguisherie P~insula:
Bulletin of the Geological Society of America, v. 47, pp. 193~1959.
Stokes, W. L., 1953, Primary sedimentary trend indicators as applied to ore finding in the Carrizo Mountains, Arizona and New Mexico: U.S. Atomic Energy
Commission, RME-3043 (pt. 1 ) , 48 pp.
Williams, Howel, F. J. Turner, and C. M. Gilbert, 1954, Petrography: San Francisco, W. H. Freeman and Co., 406 pp.
Wood, Alan, and A. J. Smith, 1959, The sedimentation and sedimentary history
of the Aberystwyth Grits (Upper Llandoverian): Quarterly Journal of the Geological Society of London, v. 114, pp. 163-195.
!'
ill
251.
1/"
I
I
t
13-1. Volcanic Sequences and Unconformities
Fig. 13-2. (A) Basalt flow resting against steep edge of somewhat older rhyolite
flow. (B) Anticlinal section through flank of eroded volcanic cone. (C) Onlap
of lavas against steep mountain front. (D) Cliff-section on the Island of Kahoolawe, Hawaii, showing how caldera-filling lava flows are almost parallel to older
lavas of the main volcanic shield. Their surface of contact (heavy line) is the
eroded fault that was once the caldera wall. The irregular vertical lines are dikes.
After H. T. Stearns ( 1940, p. 133).
Fig. 13-1. Misleading stratigraphic relations between contemporary flows that come
to lie at different levels.
250
Fig. 13-3.
252
it is uncertain that either will project evenly for a long distance. Volcanic .piles accumulate rapidly; hence associated sedimentary units
tend to be lensoid or to interfinger with the volcanic rocks through con
siderable vertical thicknesses (Fig. 13-3).
Unconformities. Each flow and pyroclastic deposit in a "continuous"
accumulation can be considered as bounded by unconformities whose
time values may be anything from a few days to tens of thousands of
years. The tops of some flows and ash layers are scarcely altered or
modified before being covered by another deposit, while many others
are weathered and eroded appreciably and some severely. Although
these minor unconformities are useful in interpreting the periodicity of
ancient volcanoes, they can make it difficult to detect and trace major
unconformities unless large areas are mapped. Major unconformities
can be recognized by: ( 1) persistent angular discordance, ( 2) truncation of volcanic necks or other volcanic intrusions, ( 3) layers of sedimentary rocks that contain materials foreign to the underlying volcanic
sequence, ( 4) zones of weathering that persist across all types of
underlying rocks, and ( 5) abrupt change in the general magma-type
of the rocks, as from calc-alkaline to alkaline. The most obvious feature
at any one outcrop is the zone of weathered materials, which is typically
colored in reds, yellows, or greens by ferruginous clays or iron-oxide
pigments. Channels filled with sand or gravel may occur here and
there, as may nne-grained sediments containing vegetative materials
or fossils. Because subaerial surfaces can develop as a series of
pediments rather than as simple peneplains or wave-planed surfaces,
some pass into irregular upland surfaces. The surfaces may also disappear laterally into lake or marine deposits.
13-2. Cartographic Units of Volcanic Rocks
Tabular sequences of plateau basalts and ignimbrites can be divided into cartographic units comparable to simple sedimentary formations. Procedures and rules for setting up and describing these
units are given in Sections 4-6 and 12-2. Many flows, pyroclastic
deposits, or composite units, however, may be irregular or stratigraphically unpredictable (Section 13-1); these units cannot be defined by measuring and describing isolated stratigraphic sections.
Mapping must be extensive enough to relate each unit accurately to
adjacent units, especially to named sedimentary formations with which
the volcanic rocks interfinger.
The basic unit of a volcanic terrain is the individual flow or pyroclastic bed. The more of these units that can be mapped, the more
I
II
1
253.
accurate will be the interpretation of the volcanic sequence. Distinctive, relatively extensive flows or tuffs can constitute separate members
within formations, as, the Wenas flow of the Ellensburg Formation
(Waters, 1955). A formal geographic name is usually omitted for a
member or bed whose lithology is unique within a formation, as
dacite flow of Umptak Formation. Informal names should be used in
cases where the area mapped is small, the units discontinuous, or relac .
tions to formalized units undetermined.
'
Formations consisting of a number of flows or pyroclastic beds are
selected on the basis of their physical individuality ana the close
genetic relation between their constituent rocks. The ideal vo~canic
formation is formed by closely succeeding eruptions of one kind of
magma; examples are sequences of basalt flows or dacite tuff beds.
Two units of similar lithology which are in contact can often be divided on the basis of an unconformity (but see Unconformities, Section 13-1). Formations or Groups that include a variety of rocks are
generally delineated at major unconformities. Some can be recognized in a general way on the basis of magma-parentage (magmatic
consanguinity); however, considerable petrographic study may be
needed to resolve such units and cases are known where nonconsanguineous rocks are associated closely.
Intrusive bodies in volcanic areas should be mapped individually.
If they cannot be plotted to scale, appropriate symbols can be used,
such as colored lines for swarms of parallel dikes. Most intrusive
units should be named in an informal way, as dacite neck, but if theii
genetic relation to effusive units is clear, they may be considered as
members of formations, as dikes of Koolau Volcanic Series. The term
complex is sometimes used for composite units of intruded, collapsed,
and faulted rocks that commonly form in the central parts of large
volcanoes, as caldera complex of Wailuku Volcanic Series (Stearns and
Macdonald, 1942).
13-3. Naming Volcanic Rocks
254
255
256
257
The term ignimbrite may be applied to all rocks laid down by glowing avalanches or pyroclastic flows (Section 13-6). The adjective
welded (as welded dacite tuff) applies to pyroclastic materials that
have been heated sufficiently to cohere. The most thoroughly welded
tuffs are massive and look like vitrophyric lavas; others are light
colored but with streaky or lenticulAr bodies of darker glass (Fig.
13-5), and still others are so devitried or altered as to be red or
light colored, stony, and without visible glass. Although most welded
rocks are ignimbrites, intrusions and lavas can also weld ash or lapilli
deposits .
Essential or juvenile.
magma.
B$.,,.
~1\.,.~-:
'
-v-.f~
'::~. -::'..
:..
'
258
259
Fig. 13-7. Tumuli (left), pressure ridge with squeeze-out 'of lava (center), and
hornitos (right) on a pahoehoe flow. Mter various photographs from Wentworth
and Macdonald ( 1953).
'""'
Idealized section through flow of fluid basalt, showing partly filled
lava tube, wrinkled and vesicular pahoehoe top, and pipe amygdales near base.
Although the flow wrinkles can often be used to determine the stratigraphic top
of a folded sequence, wrinkles can also form at the base, as shown in the drawing.
Fig. 13-6.
~=:
~
.~~--~~-~-~~-~-~~ ~~- :.~~--:..:-~~:1 .. ~=--. .
-~~~)
Fig. 13-8. Small pahoehoe tongues ("toes") accumulating to form a body whose
transverse cross section (right) has a pillowed appearance. The right view is
after a photograph from Stearns ( 1940, facing p. 88).
260
Levees of rubble\
Brecciated but
solid wall
4!6~
~~f~][''wl~"'
l~tt~~t~-~l
~p- )} ~J~...iJtfu
,.--~.J / '"' "'"
~\,tsG.N4+
Fig. 13-9. Restricted lava stream in a large flow. (A) Map showing generalized_
pattern of ridges and lava movements, with major stream cutting through older
lobes of the same flow; based on an aerial photograph (Krauskopf, 1948, facing
p. 1278). (B) Enlarged section across major stream channel, showing morainelike levees of scoria and brecciated but solid walls; after Krauskopf ( 1948, p. 1279).
are generally aligned in flow patterns, but these may be obscure. The
more silicic rocks may show a vague banding similar to How structures
in viscous lavas (Section 13-5). Stratified or partially filled amygdales
can be used to determine the original attitudes of some flows (Fig.
13-10A).
Columnar joints form in the nonscoriaceous cores of many flows,
and they are commonly divided into two tiers (Fig. 13-llA). Some
flows, however, have only one tier whereas others have three, so that
interflow sediments or soils should be used where possible for counting
How sequences. In determining the dip of jointed flows at isolated
outcrops, it cannot be assumed that the columns lie normal to the con-
..,,Altered
basalt
~
Fig. 13-1 0. ( A) Hand specimen showing amygdales and geodes with stratified
fillings of chalcedony and quartz. (B) Spiracles of laminated siltstone (left) and
baked mud in basalt flow near Fallon, Nev.
'
base
Fig. 13-11. (A) Section thr6ugh a two-tier columnar lava flow. (B) Cross section
through a columnar fld\v in which creep of the liquid interior (toward the rlght)
resulted in inclined growth of the central tier of columns; after Waters ( 1960,
p. 352).
I
:>
tact of the flow, for widely divergent fans of columns are present in
some flows. Waters ( 1960) described a particularly useful relation
where columns in the center of flows are tilted in the direction of
How (Fig. 13-llB).
Pipe amygdales are uncommon in these more viscous flows, but
large spiracles form where air, mud, ~r steam from wet ground rises
into flows when they are still fluid (Fig. 13-10B). The upper ends of
the spiracles are typically overturned in the direction of How (Waters,
1960). Although spiracles can be used to determine stratigraphic sequence, they must be distinguished from clastic dikes formed by sifting
of sediments into cracks in the tops of flows, and from dikes of partially fused sedimentary rocks that may intrude the borders of larg~
sills or dikes (Section 13-7). Large-scale steam explosions at the
base of a flow may form pipes that lead upward to surficial cinder
cones, structures that might be mistaken for the eruptive vents of the
flows themselves .
Subaqueous flows. If lavas are erupted under water or flow into
water, pillows form wh~re spurts of lava burst upward or outward and
expand until chilled (Fig. 13-12). The glassy skins arid zones of
vesicles shown in. the figure are characteristic, though some have
hollow cores like the pahoehoe tongues described above (Fig. 13-8).
Pillow lavas, however, are generally associated with aqueous sedimentary rocks or with beds of fragmented basaltic gl~s1> or palagonite.
Palagonite tuffs are dark greenish-gray to black, have a subvitreous to
waxy luster, and commonly carry calcite amygdales or cements. They
may weather brown or may crystallize to iron-rich clays or chlorites,
which give them a gray-green color and more stony appearance.
262
skin
Fig. 13-12. Pillow lava, showing generalized section through deposit (left) and
enlarged segment of single pillow. The pillows' rounded tops and cusped bottoms can be used to determine stratigrapWc sequence in deformed rocks.
are shears which cut across earlier flow structures. The platy cleavage
is more closely spaced in the lower parts of a flow or along the walls of
a lava channel or conduit, where it may dip steeply (Fig. 13-13D).
The overall pattern of flow layers and flow lineations in a ~scous
tongue is shown in Fig. 13-14.
,
In summary, five types of directional structures may be mapped in
viscous flows: ( 1) plap.ar, and perhaps linear, orientation of phenocrysts, ( 2) flow layers (generally parallel to phenocrysts), ( 3) shear
surfaces (generally but not always parallel to earlier flow layers), ( 4)
fold-axis lineations Jorm'ed at right angles to flow, by drag of layers,
and ( 5) lineations on surfaces of cleavage plates, formed parallel to
direction of m.ovement.
Domes. Viscous lava commonly accumulates directly over a vent to
form a steep-sided dome. The lava in most domes is so nearly solid
it rises as steeply inclined sheets and prisms, which protrude at the
"'
13-5. Structures of Silicic Lavas
'!263
~ J...!L ~l,
~~~~~~.=,~~~
pumiceous glass
:;:-::-"'?':'_"':.::~--
Banded stony
rhyolite
.
Field Work with Volcanic Rocks
264
'..<:
E\
:::<;.. ~
r'f r6o
L
....
~ -.:::
-....,...-.-
~----Sd-!!t-- ...........io
j!
,., 6000
~10.)()
;.""
1000 It
..._.....
70
...-
---.....rc
~;:......
--
,~.,. ~
V\1'
...
'
;I
'tiT'""~'" 4l~ ~
h.
""i
-4:.
IO
...,.
....
3o 6s
1tl
SO,<o -+-.6!{
....L
~~
1
""lo
"
-+-~..sa...:.:!:_
...._
~->
00
~-wZ
"C..
.............
,,.to-
lOi
fi
265
Map and sections of the Watchman dacite How, Crater Lake National
Park, Oregon. After Williams ( 1942, p. 45). Courtesy of the Carnegie Institution
of W ashin.gton.
Fig. 13-14.
upper surface of the dome as plates and spines. As these masses disintegrate, avalanches of angular fragments cascade down the sides of
the growing dome (Fig. 13-15A). Inward-dipping flow layers in the
solid interior of the dome are hidden by the apron of fragmental debris except at the summit, where partly broken prisms and plates can
be seen. The steeply inclined surfaces of the prisms may be grooved
in a way comparable to the shear surfaces described above.
2000 It
Fig. 13-15. (A) Viscous dome with protruding spines and talus apron. Mainly
after examples described by Williams (1932). (B) Dome of Divide (Raker)
Peak, Lassen National Park, protruded through Hank of andesite cone. After
Williams ( 1932, p. 145). (C) Eroded dome forming a small island off Ischia,
Italy; it is cut by a fault so that its internal structures are shown clearly. After
Rittmann (1930, p. 43).
Although magmas of all compositions may form pyroclastic deposits, the more viscous (silicic and alkaline) ones are far more frequently fragmented by explosions or frothed by rapid expansion of
gases. The characteristics of pyroclastic deposits depend not only on
the eruption itself but also on geographic conditions such as rainfall,
drainage, and relief. Pyroclastic units may consist of simple, tabular
beds or they may have linear or irregular shapes. Their textures and
internal structures may be of great variety. The origin of a given
deposit is determined by tracing out its shape, examining its composition and small-scale features, and recording lateral variations of these
features. Ash-fall beds or distinctive ignimbrite sheets that can be
traced laterally are especially valuable because they provide ideal
stratigraphic datum planes (see, for example, Mackin, 1960). Useful
characteristics of a deposit include the size, sorting, shape, and composition of fragments and the degree of their compaction,. welding, or
alteration. Brief descriptions of the principal kinds of deposits follow.
1. Cones of scoria, pumice, or ash. Formed by mild to moderate
explosions from a central conduit; vague to distinct beds of various
grain sizes dip putwar9- from conduit; abundant bombs or shaped
lapilli and coarse ash indicate nearness of vent; accidental blocks may
be common; relation with interbedded lavas likely to be complicated
by mixing and channeling; agglutination and gas alteration very close
to vent.
2. Widespread blankets of ash and lapilli. Formed by major ( Vulcanian) explosions; materials driven high in atmosphere so that they are
cooled during fall (except very close to vent); differentiation produced
266
267
....
deposits of basic glass and palagonite commonly associated with pillow
lavas (Section 13-4).
Various kinds of deposits may be interstratified around one volcano,
and some deposits may grade laterally into other types. The upper
or distal parts of an ignimbrite deposit, for example, may be incorporated into a volcanic mudflow, or an ash blanket may. be overridden
by, and incorporated into, a pyroclastic flow. Where the textures and
compositions of successive ash-blanket deposits are very similar, the
sequences can be subdivided at weathered zones, thin vegetative layers,
or layers and lenses of stream deposits or wind-worked materials.
Contacts between closely similar ignimbrite sheets can often be drawn
at the base of each zone of welded or compacted pumice fragments.
13-7. Volcanic Feeders and Related Intrusions
268
Dikes and sills. In mapping dikes and sills, the major difficulty is
to distinguish them from associated lavas. Near the surface, steep
basaltic feeders may be as fine-grained and vesicular as the lavas
they intrude; however, they can be distinguished by their steeply inclined flow structures and steep, chilled contacts. At greater depth,
where the dikes are less vesicular, most of them have simple sets of
columnar joints lying at right angles to their contacts (Fig. 13-17A).
The central parts of more deeply seated basic dikes consist of coarsergrained rock (diabase or dolerite), even though the margins are
aphanitic. The central parts of thick basalt flows, however, may also
be coarse-grained.
Dikes of silicic rocks are typically fine-grained, even at considerable
depths, and are commonly less tabular than dikes formed from more
fluid magmas. They generally have flow structures like those of silicic
lavas; these structures lie parallel to their steep contacts and may be
flow-folded or brecciated along them. At moderate depths, most
silicic intrusions have margins of dark glass or very fine-grained spherulitic rock that grade inward to stony, more coarsely spherulitic, or very
fine-grained phaneritic rocks. They may also show steeply inclined
autobrecciated bands. Pyroclastic rocks may be formed where pyroclastic explosions tap deeply into viscous feeder-dikes.
Sills may be particularly difficult to recognize because their flow
structures and contacts lie parallel to those of the lavas and tuffs that
they intrude. They can be identified most readily where they cut
across flows, send apophyses into overlying beds, or carry inclusions of
the overlying rocks. They are also typically less vesicular and have
269
I
!.
simpler primary joint patterns than lavas with which they are associated, and they lack the highly vesicular- or eroded tops that char:
acterize many flows. Thick or more deeply seated sills are generally
rather coarse-grained and alter the overlying rocks perceptibly. Many
basic sills and dikes weld silicic vitric tuffs for distances of a few inches
or feet from their contacts. Some melt siltstones or sandstones which
may then be mixed or injected into the intrusions.
Volcanic necks. The necks or conduits of central-type volcanoes are
roughly equidimensional in plap view, although they may be connected
with complexes of radia'ting, intersecting, or concentric dikes and sills.
Some are only pipelike enlargements or channels along dikes. In
deeply eroded areas it can be difficult to determine if an intrusive pipe
was ever a volcanic conduit. However, exposures near the original
level of eruption show a systematic outward dip of lavas and pyroclastic
beds erupted from the vent. High-level exposures of necks are also
likely to contain bodies of agglomerate or agglutinated breccias that
have slid down from the erath walls. Some may contain fragments
of water-laid sediments that were deposited in lakes or ponds in the
overlying crater or caldera and have subsided into the neck. The
border zones of silicic necks are commonly perlitic or pumiceous near
the surface, especially where they intruded water-rich wall rocks. Intense alterations of rocks by escaping volcanic gases are also indicative
of exposures near the original vent level (see Domes, Section 13-5).
Finally, the pattern of columnar joints in otherwise massive necks
may indicate the relative depth of exposure (Fig. 13-17B).
Pyroclastic textures alone do not mean that the rocks formed near
the level of the original vent, for explosions can reach to considerable
depths. Most clastic necks consist of nonvitric breccias or tuff-breccias
containing various amounts of broken country rocks. The fragments
of country rock may be angular, abraded, or well rounded 'and in some
pipes they constitute almost all of the breccia. In general, the deeper
the source of the fragments, the smaller they are,. and the more they
are rounded (see Williams, 1936, p. 132).
Intrusive sequence in composite or multiple nttcks is determined by
the usual relations of cross-cutting, diking, chilled margins, and inclusions. Where basaltic and silicic magmas are erupted simultaneously into a neck or dike, the basic magma can chill against the silicic
one and at the same time be intruded by it (see Bailey and McCallien,
1956). Anomalous fine-grained margins can also be produced by
contact-alteration of older rocks by younger magmas.
..
270
Wentworth, C. K., and Rowel Williams, 1932, The classification and terminology
of the pyroclastic rocks: National Research Council, Bulletin 89, pp. 19-53.
Williams, Rowel, 1932, The history and character of volcanic domes: University
of California Publications in Geological Sciences, v. 21, pp. 51-146.
---, 1936, Pliocene volcanoes of the Navajo-Hopi count:r.y: Bulletin of the
Geological Society of America, v. 47, pp. 111-171.
---, 1942, The geology of Crater Lake National ,Park, Oregon: Carnegie Institution of Washington, Publication 540, 162 pp.
Williams, Rowel, and Helmut Meyer-Abich, 1955, Volcanism in the southern part
of El Salvador: University of California Publications in Geological Sciences, v. 32,
pp. 1-64.
-;:,
271
References Cited
Anderson, C. A., 1933, The Tuscan formation of northern California: University
of California Publications in Geological Sciences, v. 23, pp. 215-276.
Bailey, E. B., and W. J. McCallien, 1956, Composite minor intrusions, and the
Slieve Gullion complex, Ireland: Liverpool and Manchester Geological Journal,
v. 1, pp. 466-501.
Fisher, R. V., 1960, Criteria for recognition of laharic breccias, southern Cascade
Mountains, Washington: Bulletin of the Geological Society of America, v. 71,
pp. 127-132.
Fuller, R. E., 1931, The aqueous chilling of basaltic lava on the Columbia River
Plateau: American Journal of Science, v. 21, pp. 281-300.
Gilbert, C. M., 1938, Welded tuff in eastern California: Bulletin of the Geological
Society of America, v. 49, pp. 1829-1862.
Krauskopf, K. B., 1948, Lava movement at Paricutin Volcano, Mexico: Bulletin
of the Geological Society of America, v. 59, pp. 1267-1283.
Mackin, J. H., 1960, Structural significance of Tertiary volcanic rocks in southwestern Utah: American Journal of Science, v. 258, pp. 81-131.
Rittmann, A., 1930, Geologie der Insel Ischia: Zeitschrift fur Vulkanologie, Erglinzungsband 6, 265 pp.
Shelton, J. S., 1955, Glendora volcanic rocks, Los Angeles Basin, California: Bulletin of the Geological Society of America, v. 66, pp. 45-90.
Smith, R. L., 1960, Ash flows: Bulletin of the Geological Society of America, v. 71,
pp. 795-842.
Stearns, H. T., 1940, Geology and ground-water resources of the islands of Lanai
and Kahoolawe, Hawaii: Hawaii, Div. of Hydrography, Bulletin 6, 177 pp.
Steams, H. T., and G. A. Macdonald, 1942, Geology and ground-water resources
of the island of Maui, Hawaii: Hawaii, Div. of Hydrography, Bulletin 7, 344 pp.
---, 1946, Geology and ground-water resources of the island of Hawaii: Hawaii,
Div. of Hydrography, Bulletin 9, 363 pp.
Waters, A. C., 1955, Geomorphology of south-central Washington, illustrated by
the Yakima East quadrangle: Bulletin of the Geological Society of America, v. 66,
pp. 663-684.
---, 1960, Determining direction of flow in basalts: American Journal of Science,
Bradley Volume, v. 258-A, pp. 350-366.
Wentworth, C. K., and G. A. Macdonald, 1953, Structures and forms of basaltic
rocks in Hawaii, U.S. Geological Survey, Bulletin 994, 98 pp.
~
Field Work with Igneous
and Igneous-Appearing Plutonic Rocks
273
..
A
'l,
I
:
274
Jacks Valley.
Units based on alterations. Units based on deuteric or secondary
alterations can be utilized to trace movements of fluids in plutons and
may be essential to interpreting ore deposits. These units are fundamentally alteration zones, superimposed on the original compositional
or textural units. If geographic names are needed to designate alteration zones, they should be used informally, as ko.olinized Jo1UlS Granite,
uralite zone of Hartford Gabbro, and chloritized breccias of Haley fault.
14-3. Naming Plutonic Rocks
275
!
:\
276
~
e
~
E
-~
"'E
/~~:
~
.E
------~
ii
~~
~ \
1
e
i---~1 ~i
-~~
&;~
~"-J. e.=
i
~
~~~
1
I
i5:
,.~
-c..-
-E~~-E~~
~:5~~-=~
~~~i~~
_g -a. !::
"[ ~ .8
~]
Q.>
<,.)
bll
.s
Q)
~~~
~~g
"';:l
g
Glt
~i ~a.~~
<V
~0
....
:E ~ .8
~
,.I~:
Q.>
solid state. Oriented fabrics are also important, and they will be
considered in Section 14-5.
C alar index (percent of mafic grains, by volume) affords a simple
means of classifying some rocks. The charts of Appendix 3 can be
used to estimate color index, although variations of overall GOlor with
grain size must be considered also. In addition to color index, the
ratios between distinctive mafic minerals can often be used to identify
rocks whose essential minerals are difficult to estimate in hand specimen. Many adamellites, for example, carry only biotite as a m~fic
accessory, whereas most" granodiorites carry both biotite and hornblende.
Specific gravities. Rocks that look alike in hand specimen can sometimes be named and mapped according to their specific gravities.
Specific gravity balances are too cumbersome to carry in the field,
but they can be used in c'G.mp to determine large numbers of specimens. Readings within about 0.03 of absolute specific gravity can be
made quite quickly if rocks are fresh and not too porous. This degree
of precision is sufficient to distinguish many pairs of similar rocks.
Another method, suggested by R. G. Coleman (personal communication), is to carry wide-mouthed bottles of graduated heavy liquids in
the field and to drop small pieces of rock into them to see if they sink
or float. Three or four 1Yz X 3 in. amber bottles should be adequate
if the approximate specific gravities of the rocks are known in advance.
Mineral chips of suitable specific gravity should be placed in each
bottle so that changes in the specific gravities of the liquids can be detected. Because the liquids are volatile and reactive, the bottles
should be carried in a strong case, not in a knapsack.
~
~
~ 6:::
~
-g
"';:l
-~
"0
.
~~~ ~~
~~A~~
=-=E-
1ii'i5.. ___
~e'i
-'
~
0'
.52
iiic;
~~
z
n:l
::1
1;;
~ ~
~~ ~
~= c:
(..)
CI'J
>J: u
~:, ~
-E
~~
"'
.g:Ei::~
E ~ @'E
~:;; 8.~
~i~S
~i
~:
g.~~
~~
~~
~~
~
277
I
1
278
279
~\
'1
r
I
Fig. 14-2. Veined (left), permeated (middle), and veined and permeated gradations between homogenous igneous and metamorphic rocks.
Intrusive (dilative) dike (left) and replacement dike. The beds and
black vein in the left block have been separated at right angles to the dike walls,
as can be proven by matching their lines of intersection across the dike. In the
case of the replacement dike, all structures project through the dike as though
it were not there, showing that there has been no dilation.
Fig. 14-3.
lit
;; : \ . ' \
~.. <<:: ..
Fig. 1.4-4. Intrusive relations. (A) Horizontal granitic dike with flow structures
that are discordant to the fabric of wall rocks, a rounded inclusion of schist that is
foreign to the walls, and geometrically matching walls. Near Monadhliath pluton,
southwest of Aviemore, Scotland. (B) Wedge-shaped apophyses that compress
bedding adjacent to their walls; at contact of granodiorite stock, Santa Rosa Range,
Nev.
280
the fact that the younger rock cuts across veins or inclusions in the
older.
Gradational internal contacts are typically obscure and therefore
difficult to map. Some gradations within plutons are so broad that
the term contact loses its structural significance. In feldspathic rocks,
the gradations can often be traced on the basis of color index, or on
the basis of the ratios of critical mafic minerals, as biotite and hornblende in granitic rocks. Although fresh, dark-colored rocks are dif- _
ficult to map by color index, weathered surfaces often show minerals
clearly. Feldspars are generally whitened by weathering, while olivine
is altered to limonitic pits and hypersthene is bleached to golden or
bronze-colored cleavable grains. Gradational contacts between mafic
rocks can also be keyed on the first appearance of distinctive accessory
minerals, as biotite, olivine, or quartz.
After the criteria for drawing gradational contacts have been established by studying several complete gradations, rock chips from these
gradations can be carried in the field as a standard for plotting other
outcrops. The color-spot method is especially useful for this type of
mapping (Section 4-10).
Zones of mixed (hybrid) magmas. Some gradational zones suggest
that one magma intruded into or against another. If both magmas
were largely liquid and had similar physical properties, the only
textural evidence of mixing may be corroded grains that have survived
interaction in the hybrid magma. If one magma was nearly solid, the
hybrid zone can be expected to show partially disintegrated and reacted inclusions of one rock in the other. Where the more crystalline
magma was pliable, the inclusions may be spindle shaped or streaked
into lines and stratiform mixtures of the two rocks. A striking structural relation is that where one rock occurs both as dikes and as inclusions in another. This condition may arise because the two magmas
had different temperatures of crystallization. Gabbroic magma, for
example, may locally be chilled against granitic magma.
Although the rocks in the hybrid zone may have compositions suitable to simple mixing of the two adjacent magmas, volatile materials
can cause strikingly heterogeneous textures. Where basic and granitic
magmas have mixed, for example, the basic rocks may have large
poikilitic grains of potassium feldspar or biotite, and the granitic
rocks may have clots of mafic grains or orthoclase phenocrysts with
rims of plagioclase.
Deformed contacts. Convincing evidence of intrusion or metasomatism may be difficult to find where contacts have been deformed
t
Field Work with Igneous and Igneous-Appearing Plutonic Rocks
281
I
Granitic rock with a foliation that passes into country rocks without
deflection. Contact zone of Flamanville granite at Anse de Sciotot, Normandy,
France (see also the figures of Martin, 1953).
Fig. 14-5.
Fig. 14-6. Contact of the Ardclach granite on Hill of Aitnoch, Nairnshire, Scotland.
Note that the planar fabric of the granite crosses the contact at various angles but
is typically parallel to layering in the adjoining metamorphic rocks. There is a
strong lineation in the schists but not in the granite.
282
283
Fig. 14-7. Planar fabrics in plutonic (here, granitic) rocks. Quartz is stippled,
feldspar unpatterned, and mafic grains black. (A) Simple flow fabric in hypidiomorphic granular rock. (B) Fabric showing effects of late flow (note change
in shapes of quartz grains compared to A). (C) Strongly protoclastic fabric.
(D) Granitic gneiss, showing lenses of biotite grains and simple granoblastic texture which have resulted from recrystallization during and following deformation.
tj~
if
:;
284
285
Fig. 14-8. Metasomatic dikes of light-colored granitic rocks in more mafic tonalite,
Santa Lucia Range, Calif. Gneiss-like structure of tonalite is in vertical plane
perpendicular to page, and passes through dikes. Note ( 1) less modified septa
that cross the dikes, ( 2) continuation of dark inclusions from tonalite into dikes,
and ( 3) control of the dikes by a set of cross fractures in the tonalite, locally .
forming sharp contacts.
when the magma was largely solid (type 2, above). Inherited fabrics
in metasomatic rocks can also pass through contacts as well as inclusions (Fig. 14-8). Further relations between fabrics and related
structures are described in the next three sections.
Methods. Oriented fabrics that are obscure in hand specimens can
often be seen in clean outcrops measuring several feet across. Rocks
tend to split along the aligned cleavage surfaces of feldspars and
micas, and both the shape of outcrops and the grain of the topography
may reflect the fabric of the rocks.
The number of readings and symbols needed per unit area will depend on the consistency of the fabric and the purpose of the project.
Several readings should be taken at each large outcrop or group of
outcrops before plotting a symbol. If all the readings are within 5
or 10 of their average, the symbol may be plotted with a straight
strike line. If attitudes are more variable, the outcrops should be
examined to trace out the continuity of the fabric. Moderate rolls or
irregularities will be found in many cases, and a sinuous strike line
can be used for their symbol. When different ages or kinds of fabrics
are present, distinctive symbols should be used for each, and the
relations between the fabrics should be sketched and described in
::
'
le
286
~D~~li~~!
0(*~JffW~
(,~)\~]:~~~~;#~~;
%;.w;~;;J;{:
' ~'"'If~!:: ,;. " ,,:t;' .f.Jf:;,
.<~?~,~~;;g\:..!.;;~~
,.,,,,,,,};J,I,I.'
~j{~~~~;Yfi;!;)/
I'
"'~
:;~~~~;;Jxlft{~I ;:~~~~~~~!t}:.::Y:~~
;., .,.-.,.,.
:~:tM;~:1~J'S;'
_,. .,.,.::--'-:.'
__
~~;((~If~ .
0
- - ' .--.,c'
llllt
Fig. 14-9, Relations between inclusions and rock fabrics (see text for explanation).
287
tures are analogous to sedimentary beds, and, like beds, they may be
thick or thin, obvious or obscure, and tabular for long distances or
lenticular. They may also be folded,. broken, or deformed into various
shapes. The German term Schlieren is sometimes used for these layers,
but it is more appropriately applied to streaky, wisplike, or nonstratiform concentrations of dark or light minerals. Alternately, schlieren
may be called streaks, wisps, or other descriptive names. Most layers
and schlieren can be measured and plotted as planar structures~ some
have linear shapes or associated linear structures that c;m also be
measured and plotted.""
Layers and schlieren can form in different ways, and it .is possible
for several types to occur in a given pluton. It is therefore necessary
to classify th9fi in a useful descriptive way that will serve, if possible,
to distinguish them genetically. The foliation symbol used for each
type should be designated by a small letter or other suitable mark.
Layers and schlieren are here organized, for conveni~nce, into
five genetic clasres. Although the list is tentative and some of-the .
structures described are problematical, all known possibilities should
be considered in the field; otherwise large-scale relations might be
grossly misinterpreted.
a. Layers formed by accumulation of crystals on a floor.
288
.?:.;B;::
;0
~:: ,: .
. -
:~&. :k~,;rfY
(f+~\ \
t~:
,
~~'.
-~:
<'
.
'
". ot
<~:>". ~
a
North
-'-'-'---
Fig. 14-10. Layers- and schlieren in plutonic rocks. All views are essentially vertical sections except B and D, which are horizontal. (A) Systematically graded
and eroded(?) layers in tonalite; Tioga region, Sierra Nevada, Calif. (B) Combfabric in layered pyroxene diorite at contact with earlier tonalite (left); Lake
Tahoe, Calif. Courtesy of A. A. Loomis. (C) Flow layers and schlieren in
granodiorite; Mt. Conness, Sierra Nevada, Calif. (D) Dike showing mafic minerals and large feldspar tablets concentrated into layers and schlieren oblique
to walls. Arrow indicates sense of late flow. Near Mt. Conness, Sierra Nevada,
Calif. (E) Thin mafic layers in dike, showing fabrics that lie oblique to contacts
of dike but parallel to fabric of the surrounding Aar granite. South of Goschenen,
Switzerland. (F) Hornblende gabbro, showing two sets of feldspathic "layers."
Near Enterprise, Sierra Nevada, Calif.
289
290
inclusions can also be used to determine the stage when certain layerstructures formed (Section 14-6).
14-8. Fractures and Related Structures in Plutons
7~,/L
291
Trace of planar
fabric in
granodiorite
''~fi! ...z>~'$~~
Fia, 14-12. Dark tabular dikes in granodiorite, passing laterally into inclusion
swarms. Note intrusive relation of granodiorite to dike at lower left. French
Lake, Sierra Nevada, Calif.
I ,
'
292
fig. 14-13. Quartzofeldspathic dike ( unpatterned) cut along medial and marginal
fractures by tourmaline veins (black). Near English Mt., Sierra Nevada, Calif.
l'
Field Work with Igneous and Igneous-Appearing Plutonic Rocks
293
''il
~:rv.0g~y,~~~r~;J~ifJ~
-~.. _.Jil.-:;.,
1 ft
fig. 14-14. (A) Sheet of protoclastic granodiorite cutting planar fabric; South Fork
of Feather River, Calif. (B) Feather joints and bleached rock related to shear
in granodiorite; west of Donner Summit, Calif. (C) Large outcrop surface that
dips steeply toward the observer, showing set of nine late shears in granodiorite.
The two dikes have a component of dip toward one another, thereby recording a
systematic sense of offset on the shears. Fabric and schlieren dip steeply into
the outcrop, as shown by the map symbols on the right. Near Enterprise, Sierra
Nevada, Calif.
294
where altered and unaltered rocks can be related to the mapped features. Alteration zones should be mapped as an overprint on primary
units and structures, as by placing transparent overlays on field sheets
and plotting color spots or patterns for a given alteration. These maps
can show that the alterations are connected with certain rocks or structures in a pluton, or that they are related to structures and unit,s in
the surrounding terrain. Alteration minerals should be matched with
fracture sets or systems so that age relations between crossing fractures
can be used to date alterations. Some hydrothermal alterations are
closely connected with contact metamorphism, and the two should
therefore be examined concurrently (see Section 15-13). Examples
of detailed studies relating various alterations to fracturing and igneous
processes have been given by Gilluly ( 1946) and Mackin ( 1947, 1954).
A large number of deuteric and secondary minerals can be used as
a basis for mapping. Some occur in a variety of rocks; others are restricted to certain types. Sericitization of plagioclase and chloritization
of biotite take place in most feldspathic rocks, especially in granitic
rocks. Silicification typically forms fine-grained tough rocks with
blurred to obscure relict textures, while argillization (production of
clays from silicate minerals) yields softer rocks that generally show
clear textural relics. Thoroughly albitized feldspathic rocks -may be
nearly white, light green (colored by epidote and chlorite), or light
red (colored by hematite or hydrated ferric oxides). Other minerals
that may characterize various alterations of feldspathic plutonic rocks
are zoisite, carbonates, barite, fluorite, zeolites, alunite, pyrite, and
various ore minerals, especially chalcopyrite, galena, and sphalerite.
Hot fluids in particular can alter granitic rocks to coarse-grained
quartz, tourmaline, and muscovite, locally with scheelite, cassiterite,
magnetite, pyrrhotite, wolframite, or topaz. Common alterations in
basic and ultrabasic rocks are olivine and enstatite to serpentine
minerals, pyroxenes to uralite (aggregates of acicular actinolite) and
chlorite, and calcic plagioclase to whitish or greenish aggregates of
fine-grained albite, epidote, zoisite, sericite, and calcite. Other important secondary minerals in basic rocks are dolomite, ankerite,
magnesite, siderite, magnetite, pyrite, prehnite, talc, and various basemetal sulfides. The nepheline in feldspathoidal rocks is commonly
altered to cancrinite or to a variety of zeolites or clays.
References Cited
Balk, Robert, 1932, Geology of the Newcomb quadrangle: New York State Museum
Bulletin 290, 106 pp.
295
I
I
quence.
2. Map the premetamorphic rock units and structures.
3. Classify and map the metamorphic rocks and structures.
4. Determine the amounts and kinds of deformation.
5. Map zones of metamorphic minerals or textures.
The relative importance of these studies and the order in which they
are undertaken will depend on the individual case. Typically, several
can be done simultaneously, and each kind of study will contribute
information to the others. Determining the original lithology and
distribution of rock units is generally very important because metamorphic changes can be guided or limited by premetamorphic features.
Primary textures and structures, for example, can control metamorphic
deformation considerably, and primary minerals (or minerals formed
during an early stage of metamorphism) often limit those that can
form subsequently.
The depth and therefore the pressure at which a given metamorphism took place can sometimes be approximated by broad stratigraphic and structural studies. Zone mapping may indicate directly
whether metamorphism was caused by igneous intrusion, by deep
burial, or by deformation. Metamorphic temperatures and pressures
can sometimes be estimated by comparing mineral assemblages with
laboratory data on the synthesis of minerals. In order to do this reliably, the total metamorphic history of a suite of rocks must first be
296
297
I
~
;
I
~.
298
Schistose. Platy or elongate grains oriented in a parallel to subparallel way so that the rock cleaves readily; may be classified further
as foliated or lepidoblastic where the preferred orientation and cleavage
are planar, and lineated or nematoblastic where they are linear.
299
~~ '
.I
300
I
!
':\
i'~- ~
301
e. Cataclastic rocks
)-
i.
j
303
302
The successful interpretation of metamorphism depends coll'!liderably on determining the original lithology of a given suite of rocks.
Not only must the original rock types be determined, but also their
textures, compositions, and small-scale structures. An argillaceous
sandstone, for example, is likely to react differently to metamorphic
processes than a well-sorted quartzose sandstone, yet each may bear
relict sand grains after metamorphism. Similarly, thermal metamorphism of phyllite may produce a decidedly different product than
thermal metamorphism of a clay shale. These differences are caused
in part by compositions (including volatile substances), in part by
mechanical response to deformation, and in part by the effects of
permeability and grain size on the rates of mineral reactions.
Some mineral assemblages are directly indicative of original rock
types. Almost all quartzites, quartz-rich schists, marbles, and calcsilicate rocks are of sedimentary origin. They should be examined
for relics of coarse textures (as in metaconglomerates and biohermal
limestones), sand-size layers, or thinly laminated beds that suggest
fine-grained sediments. Micaceous rocks with grains of kyanite,
andalusite, or sillimanite form from claystones rich in illite or kaolinite,
while schists with staurolite, chloritoid, or abundant pink garnets
form from more iron-rich ( chloritic or montmorillonitic) claystones or
siltstones. Ultrabasic rocks yield such unique rocks as talc schists
and anthophyllitic schists, but few other meta-igneous rocks can be
recognized on the basis of composition alone. Amphibolites, for example, can form from basalt, diabase, gabbro, basic tuffs, tuffaceous
sediments, mafic graywackes, or dolomitic mudstones. At least as
great a variety of rocks can yield metamorphic rocks consisting of
micas, feldspars, and quartz. Relict textures and structures must
therefore be sought carefully in these rocks. The brief suggestions
that follow can be supplemented by the descriptions and figure~ in
the three preceding chapters.
Metasedimentary rocks are most easily recognized by their relict
bedding, especially where some layers or lenses are calcite bearing or
quartz rich. Amphibolites of sedimentary origin are commonly intercalated with gray-green calcareous layers rich in diopside or scapolite,
or brownish layers containing much biotite. Associated micaceous
~-
304
6 in.
305
ceous (upper) parts (Fig. 15-2C). The larger of the various sole markings can be found in many hornfelses and low-grade schists. The
tops of beds can also be determined from the geometric relations between folded beds and secondary cleavages (Section 15-8).
15-5. Studying Metamorphic Deformation
If fqrmational units cannot be mapped as in step 1, it may be possible to map the rocks in zones characterized by certain textures and
structures (Section 15-13). The remaining studies could then be applied to each of these zones.
15-6. Foliations and Lineations
Fig. 15-2.
Metamorphic foliations and lineations can form in a number of different ways, some of which are not well understood. Although they
should be interpreted as fully as possible in the field, mapping must
306
The maximum dimension limiting the terms closely spaced is generally about ~ in. in fine-grained schistose rocks and about 2 in. in
coarse-grained schists and massive rocks.
Where it is possible to discern two or more .foliations of one kind
but of different ages, the sequence can be numbered, as slip cleavage 1
and 2. Notations can also be simplified by assembling foliations into
groups related by age or geometry. This is commonly done by using
the letter S with a subscript number; for example, compositional
layering and a parallel schistosity could be labeled sl, slip cleavage
and a parallel schistosity 5 2 , and so forth.
Lineations. A lineation is a preferred (subparallel) orientation of
one or more kinds of linear features. A great variety of lineations may
be studied and mapped. Some give direct clues as to the amount,
kind, and direction of deformation; others may be problematical until
the map is complete and their patterns can be compared to all other
deformational features. If the scale of the map is small, lineations
can be plotted and compared on spherical projections. Where lineations can be related to distorted primary structures, their genetic implications will be particularly clear.
'
II
&
9,
307
Not only can folds be plotted as linear elements, but their general
shapes or geometric styles can be used to assess the distribution and
intensity of deformation. Marked variations in fold style over an
area should therefore be described or mapped. At the same time,
the relations of folds to foliations and lineations should be noted, as
described in Section 15-8. These data will be particularly useful in
determi:q:ing the distribution of two or more periods of folding.
The following suggestions regarding terminology are illustrated by
Fig. 15-3. An important aspect of fold style is the degree of divergence
of the limbs and their symmetry relative to the axial plane (A, B,
and C in the figure). Terms relating to the degree of curvature at the
hinge line, to the degree of similarity between the folds in successive
layers, and to the attenuation and slipping out of limbs are shown by
308
A~
. . ..
..
.....
Open symmetric
~
Asymmetric
BINI\fA
...
Tight
AMI.
M.
Isoclinal
D~~;j
t(/~v
309
Nor
"'
'" -0
Fig. 15-4.
Open chevron
~
Asymmetric chevron
Tight chevron,
disharmonic
mrmL
Parallel
Similar
~
Axial or
cylindroidal
:0:0
Curving axes with
simple axial planes
Axes
straight
but axial
planes
divergent
Fig. 15-3.
Similar, isoclinal
Limbs attenuated
and slipped out
l)[)fJ
Curving axes with
curving axial planes
p~.
~;~,
old limbs and axes (Fig. 15-4A and B), or the outright evidence shown
by folded isoclinal folds (Fig. 15-4C). Hinges of the earlier folds
are likely to be destroyed in most places, but crudely quadrate or
rhombbid patterns may still be seen on smooth surfaces.
1S-8. Foliations and lineations Related to Folds
310
The general size of the folds and the strength of the rock materials
should be noted as these various features ar~ studied. A large body
of incompetent beds, for example, is likely to fold less regularly than
a few beds within the body. The relative amounts of flexure, solid
flow, and slip should be assessed at each group of outcrops. Time
relations between deformation and recrystallization are also important,
as illustrated by the examples described in the next paragraph.
Relations between older and younger foliations and lineations can
be seen most easily in medium to coarse-grained schists, gneisses, and
granulites. Figure 15-5A shows a case where schistosity is exactly
parallel to the layering (probably relict bedding) in parallel-folded
rocks. It is almost certain that this schistosity is older than the folds.
In case B, the rocks were also schistose before they were folded, but
, , , <Ifill,--
,;({:;!,1'1\\~\\' ':
Micas
.'~.li?:it(::~:.l
Iff',
.t,~,v,-.-.:. :.,n
..~i--~ '~,....
.-.:.-.-__:_ :~~~
,,
1,
\~ ... -:.
'''''i"i, ,
l:
'
,~;,!,~ !,,
~'''11'''\t
:. '
~\{(1',!:~:~,;5~
_.I \,,.
l,jl'
',oil
..
:.-;.:~~,f!J.;
'tJfr~-:::
~/: ,,:.:-:-?e1
: ...... ~7
\",
i::-//;),~/ ; ,:,..<
Ji}1):!.:..,,f:\1i~ : ~\~?i;:~v
,'\, .. i ~I
n;,
'"
1~1,1!'1
'
'1'~-::
r
\II'
sin.
L -
,,.i.\~i{,';:l, .~':,,'. :,
/Micas
.I
''''ii" I 't'tf'.
t~~-;.~1:~., ~liftiff..
::.,-~
~~~J~i~,~~i~,\:::~
'~'
6 in.
I
~
&'
if~
:*
II
3ll
'ik
I
c
t
a new foliation had begun to form along some limbs and axial planes.
At the outcrop, the rocks cleave almost as readily in this new direction
as along the older schistosity. In case C, the rocks recrystallized considerably during folding, for the micas at fold hinges have been pressed
or sheared so as to grow parallel to the axial planes. The mica-rich
layer in case D has also recrystallized during folding, but here the
new orientation is similar to that of secondary cleavages in many
finer-grained rocks (compare Fig. 15-7). Finally, the case shown in
Fig. 15-6 strongly suggests slip folding (translation of thin sheets of
rock parallel to the new schistosity). The critical relations are: ( 1)
the schistosity is approximately parallel to the axial planes of the folds,
regardless of the compositions of the layers, ( 2) the folding is similar
(harmonic) from one layer to the next, and ( 3) the distance subtended by any one cleavage surface on a given layer is approximately
constant, regardless of where it occms in the fold (this results in an
apparent attenuation of layers on fold limbs) .
Cleayages in fine-grained rocks. Individual grains cannot be seen
in slate_s and phyllites, but relations between cleavages and bedding
can beused to interpret deformation. Slaty cleavage commonly lies
parallel to. the axial planes of argillaceous beds that were folded tightly
as they were metamorphosed. Figure 15-7A shows this relation diagramll}atically; the rocks cleave on an infinite number of surfaces
parallel to the lined pattern of the drawing. This cleavage can form
. when compression normal to the axial planes flattens the rocks as
they recrystallize. It can also form in consequence of shearing on
312
313
ll
..._!_!!___.
,
Fig. 15-7. Secondary cleavages in fine-grained rocks.
(A) Simple axial plane
cleavage in slate. (B), (C) and (D) Secondary cleavages in folded phyllites and
quartzites of Santa Rosa Range, Humboldt Co., Nev.
closely spaced surfaces that parallel the axial planes. The origin of
the cleavage can sometimes be determined if deformed relict fossils,
ooids, or sand grains can be observed. Oriented samples should therefore be collected for petrographic studies (Section 15-12).
Where competent and incompetent beds are intercalated, secondary
cleavages are commonly oblique to axial planes, except at fold hinges.
Figures 15-7B, C, and D show several cleavage patterns in folded
quartzites and phyllites. The cleavages illustrated are either fracture
cleavages or slip cleavages with displacements of 1 mm or less. Regardless of how they form (this is .much argued}, oblique cleavages
such as these are valuable for determining tops of beds at isolated
outcrops and for estimating the position of an outcrop relative to the
overall geometric form of a fold. The angle formed where cleavage
surfaces pass from phyllite layers into quartzite layers is especially
useful. Outcrops at the crests or troughs of folds show cleavages
that pass from one bed to another without deflection, whereas outcrops on limbs of folds show a distinct angle in the cleavage at the
contacts between beds. The tops of the beds can be determined by
the sigmoidal curve of cleavages in micaceous or feldspathic quartzite
beds (Fig. 15-7B) or by the unilateral curve in graded beds (Fig.
15-7C). A useful empirical rule for determining tops of obliquely
Fig. 15-8. Folded secondary ckavages. (A) Gently dipping quartzite bed showing vertical fracture cleavage deformed into steeply plunging minor folds; Santa
Rosa Range, Nev. (B) Older cleavage (thin arrows) folded and cut locally by
a younger cleavage (heavy arrow). The patterned bed has been offset into a
sawtooth shape along the earlier cleavage. From Shenandoah River Valley, south
of Harpers Ferry. After Nickelsen (1956, p. 262).
cleaved beds is that the acute angle between the cleavage and a
bedding surface points in the direction the adjoining bed slipped
toward the hinge of an adjacent fold (Fig. 15-7D). These determinations of sequence should be tested by reference to relict sedimentary
-structures because an entire sequence may have been overturned before it was folded and metamorphosed.
. Examples of field studies in which cleavages and related lineations
were used to interpret larger structures are given by Wilson ( 1951)
and Nickelsen ( 1956).
Folded. cleavages. Folded cleavages provide evidence of a second
. folding of beds (Fig. 15-SA). If a second set of cleavages formed with
the younger folds, however, the beds themselves may be difficult to
recognize (Fig. 15-SB). Statistical methods may be helpful in sorting
out the structures formed during each deformation (Section 15-11).
.. Boudin's and rods. Thin beds, veins, or dikes enclosed in incompetent materials are f~equently broken, rotated, and squeezed into
linear bodies during metamorphism and folding. The commonest of
these are boudins (literally, sausages); two types of these linear struc.. tures are shown in Figs. 15-9A and 15-16D. Rods or rodding structure
are chiefly quartz or quartzofeldspathic bodies that have been folded,
dismembered, or rolled out at right angles to their lengths (Fig.
15-9B). Judged by their streamlined profiles, these bodies are typically deformed as they are emplaced; many appear to be segregated
metamorphically from the surrounding rock (Gilbert, 1953).
Boudins and rods should plunge parallel to the fold with which
1~
314
.315
6 in.
Fig. 15-9. (A) Cleaved siliceous beds in limestone, partially disrupted into linear
rhombs; Palmetto Mts., Nev. (B) Rods and related linear relics of silicate veins
in marble; Santa Lucia Range, Calif.
Foliations and lineations iThtmassive rocks are less striking than those
in bedded rocks, but they can indicate local stress-strain relations more
reliably because of the homogeneity of the rocks. Large intrusive
bodies are ideal units, and thick lava flows, tuff-breccias, quartzites,
metaconglomerates, or marble beds can also be useful.
0.5 in.
6 in.
Fig. 15-11. Deformation features of massive rocks. (A) Deformed metaconglomerate near Mariposa, Calif. (B) Slipped-out quartz vein and inclusions in metamorphosed dike; near Bidwell Bar, Calif. (C) Metabasalt cut into lineate forms
by sheets of phyllonite; heavy arrow indicates plunge. Near Oroville, Calif.
(D) Semischistose meta-andesite cut into rhombic prisms by intersecting cleavage
surfaces; near Bidwell Bar, Calif.
than the metamorphism. Segregations of quartz, epidote, and carbonates are especially common on fractures formed during low-grade
metamorphism. Fractures in higher grade rocks are likely to bear
quartz, alkali feldspars, and various metamorphic minerals. Sequences
of veins can be determined ill some areas, while folding or slicing out
of veins may indicate ~he relative ages of two deformations.
316
Fig. 15-12.
317
I~
!
\I,..
318
N
s
tions from one subarea to the next; perhaps it was c~sed because the
structures in one or more of the subareas formed at different times.
4. Re-examine outcrops to determine the relative ages of structures.
5. If there were two ages of folding, it may be possible to unfold
( undeform) structures to determine the patterns of earlier folds (see
Phillips, 1954, p. 28; Badgley, 1959, p. 195).
Depending on the area concerned, many useful variants of these
methods can be used. Weiss and Mcintyre ( 1957) and Ramsay
( 1958) have described studies that combined mapping and structural
analyses.
15-12. Oriented Samples for Microscopic Studies
319
l -
I
I
!
B
Fig. 15-14.
readily recognizable, and sensitive to variations in 'metamorphic conditions. Zones can be mapped by plotting color spo.ts or letter symbols for outcrops with suitable characteristics. Because many metamorphic changes are broadly gradational, the first appearance or disappearance of a key mineral, texture, or structure must often be used
to map zone contacts (Section 4-9). Reversals and other complications should be described in the notes. New key minerals, for example,
may be concentrated locally on shear zones or near veins, indicating
localized effects of granulation and fluids. Relics of low-grade rocks
in high-grade zones can indicate that the high-grade rock passed
through the states of lower grade zones.
Because of variations in rock properties and metamorphic histories,
methods that succeed in one area may fail in another. Each area must
be considered individually; the suggestions that follow present only
a few possibilities.
Mineral zones in regional terrains. The most widely used mineral
zones in regional metamorphic terrains are based on the first appearance of certain index minerals in schistose metashales. Slates containing clays and clastic chlorite and micas are first recrystallized to phyllites or schists with new metamorphic chlorite and muscovite; then,
the following minerals appear with increasing degrees of metamorphism: red-brown biotite, almandine garnet, staurolite (if the rocks
are rich in iron), kyanite, and sillimanite. This well-known sequence
is particularly useful because the minerals are commonly distinctive in
hand specimen. Several samples should be examined at each outcrop
because only a few layers may have suitable chemical compositions or
be coarse-grained enough to show index minerals. Age relations between minerals must also be noted, for chlorite and muscovite can
form by retrograde reactions in high-grade rocks.
Basic rocks, as metabasalts, can be zoned by the following criteria:
( 1) disappearance of chlorite (to form hornblende), ( 2) disappearance of epidote (to form plagioclase and hornblende), ( 3) appearance of red garnet, ( 4) change of green hornblend~ to brown hornblende, and ( 5) appearance of pyroxenes, especially hypersthene.
Retrograde metamorphism must again be considered; for example,
epidote and chlorite in amphibolites may be retrograde.
Mineral zones can be mapped in impure calcareous or dolomitic
rocks, although moderate variations of chemical composition may result in misleading variants. The following changes may be useful:
( 1 ) reactions of chlorite, muscovite, and carbonate~ to biotite and
epidote (or zoisite), ( 2) appearance of hornblende, ( 3) disappearance
320
~J
321
t
I
I
322
323
Fig. 15-15. Structural varieti~s of migmatites, with inserts showing patterns that
may be plotted in color on field sheets. The left-hand and middle examples are
from the Feather River region, Calif., and the other is from Nairnshire, Scotland.
walls, bending and breaking of septa in veins, moved inclusions, foreign i~clusions, and planar fabrics that are parallel to vein walls but
not the foliation of the adjoining rocks (Figs. 14-3, 14-4, and 14-9).
Metasomatism may be indicated by proof of lack of dilation of walls, incluslons ( skialiths) in place, relics of former textures and structures,
gji(dational contacts, and mane segregations at or near the contacts of
veins
, (Figs. 14-3, 14-8, 14-9E, and 14-9F). Metasomatism is indicated strongly where vein materials are closely related to the compositions of the rocks in which they occur. Veins in metashales, for example, might contain abundant potassium feldspar, mica, and garnet,
whereas those in adjoining amphibolite layers consist mainly of plagioclase, hornblende, and pyroxene.
Observations should also be made on deformation structures associ,ated with migmatites. The contrast between the angular appearance of some veins and the complexly folded, plastic appearance of
others indicate important differences in the histories of rock masses
(Figs. 15-15 and 15-16A). The irregular rotation ("swimming") of
schist relics may suggest deformation of soft, perhaps partially molten,
granitic materials (Fig. 15-16B). In contrast, deformation of solid
veins tends to produce slipping-out, boudinage, and rodding (Fig.
15-16C and D). Extreme post-vein deformation may yield porphyroclastic schists or phyllonites. Sequences of structural types can indicate that deformation went on concurrently with vein formation.
Maps of deformation structures may show that the migmatites moved
as gross plutonic bodies, perhaps intruding less mobile parts of a metamorphic terrain.
The original definitions and descriptions of migmatites given by
Sederholm ( 1907, 1926) are particularly instructive. Read ( 1931)
to
324
325
I
and Crowder ( 1959) have described terrains where migmatites were
treated as cartographic units.
References Cited
Badgley, P. C., 1959, Structural methods for the exploration geologist, New York,
Harper and Brothers, 280 pp.
Burnham, C. W., 1959, Contact metamorphism of magnesian limestones at Crestmore, California: Bulletin of the Geological Society of America, v. 70, pp. 879920.
Cloos, Ernst, 1947, OOlite deformation in the South Mountain fold, Maryland:
Bulletin of the Geological Society of America, v. 58, pp. 843-918.
Crowder, D. F., 1959, Granitization, migmatization, and fusion in the northern
Entiat Mountains, Washington: Bulletin of the Geological Society of America,
v. 70, pp. 827-878.
Gair, J. E., 1950, Some effects of deformation in the central Appalachians: Bulletin
of the Geological Society of America, v. 61, pp. 857-876.
:1
Ij
i
t
I
I
'r
..
Appendixes
"'
.,
1,
328
~~
Aerial photographs
Ephemeris, solar
Erasers
Field glasses
Field notebooks
Flagging for signals
Hand level (or Brunton compass)
Hatchet
Keel (crayons), red and blue
Logarithms, 7 -place
Lumber for signals and stakes
Nails, assorted
Paint for signals
Paint brushes
The abbreviations suggested here are only !!- few of many possibilities. Most words of six letters or less are omitted because little is
gained in abbreviating them. Meanings can be clarified in some
cases by capitalizing the first letter of nouns. A more complete list
has been given by J. G. Mitchell and J. C. Maher ( 1957, Suggested
abbreviations for lithologic descriptions: Bulletin of the American
Association of Petroleum Geologists, v. 41, pp. 2103-2107).
abundant
acicular
aggregaJe
amorphous
amount o\
angle
angw~
andesite
anhedral
an\'drite
apptoximate
arenaceous
argillaceous
arkosic
asphaltic
average
beddf<l
bedding
bentonite
biotite
bituminous
bou~der
brachiopod
calcareous
carbonaceous
cavernous
cemented
chalcedony
claystone
cobble
conglomerate
contact
cross-bedded
cross-bedding
cross-laminated
abnt
acic
Aggr
amor
Amt
L
ang
And
anhed
Anhy
approx
aren
arg
ark
as ph
Ave
bdd
Bdg
Bent
Bio
bit
Bldr
Brach
calc
carb
cav
cmt
Chal
Clst
Cbl
Cgl
Ctc
xbdd
Xbdg
xlam
crystal
crystalline
diameter
different
disseminated
dolomite
elevation
equivalent
evaporitic
exposure
feldspathic
foliated
foraminifera
formation
fracture
fragmental
glauconitic
granite
granodiorite
granular
graptolite
graywacke
greenstone
gypsiferous
hematitic
horizontal
hornblende
hornfels
igneous
inclusion
interbedded
intrusion
irregular
joint
laminated
329
XI
xln
Diam
diff
dissem
Dol
Elev
eqniv
evap
Exp
feld
fol
Foram
Fm
Frac
frag
glanc
Gr
Grd
gran
Grap
Gwke
Gmst
gyp
hem
horiz
Hbl
Hfls
ign
Incl
intbdd
Intr
ireg
Jnt
lam
330
limestone
maximum
member
metamorphic
mudstone
muscovite
nepheline
nodular
oolitic
orthoclase
outcrop
pebble
peridotite
permeability
phenocryst
phosphatic
plagioclase
point
porphyritic
probable
pyritized
quartz
quartzite
Ls
max
Mbr
met
Mdst
Muse
Neph
nod
oi:il
Orth
Otcp
Pbl
Perid
Perm
Phen
phos
Plag
Pt
porph
prob
py
Qz
Qzt
regular
rhyolite
rocks
rounded
sandstone
saturated
secondary
sedimentary
shale
serpentine
siliceous
siltstone
soluble
station
structure
tabular
temperature
unconformity
variegated
vegetation
vertebrate
volcanic
xenolith
reg
Rhy
Rx
rod
Ss
sat
sec
sed
Sh
Sp
sil
Sltst
sol
Sta
Struc
tab
T
Unconf
vrgt
Veg
Vrtb
vole
Xen
.-
,.
~
1%
2%
Appendix 3
APPENDIX 3 (Continuecl)
7%
15%
20%
25%
10%
30%
Prepared by R. D. Terry and G. V. Chilingar for Journal of Sedimentary Petrology (v. 25, pp. 229-234, 1955); reprinted as Data
Sheet 6 of Geotimes, available from the American Geological Institute,
332
3%
'
5%
333
tI
40%
50%
4--------------
5 '"""""'"'""'""""""'"'"""'""""""
6 ..
3------
___go
27
---:7-+
28
{-_r. }--
29
30
31
30
-.L.
-E-+
9--?---
...,.
33
11 - - - - - - - - -
34
12~
13---=;----
14~
35
36
__.__
38
-+-
~~_.:::;;V_./'""~~
63
39
40
41
18-?-+-?-
42
19
20-
20
Jv
60
~0
TO?
_,_.
....6..._~
,_......
+ + +
30
I--"~
_L
="===
--'"'--
--- -
+.+
'
..:.....a..-
o
....__
'
"""""'
,. \:9
37
TO
ts#g~--
,.
32
10 ......... .
8------
335
26-
90
1
Appendix 4
...
25
=!!=
?-t
21~~
44
22~
45~
46~
23 -
24
+T
t
)
t1""T"'Tw
l
1
47
-o
48
<1~0
49
)--
50
51
~hbJ...._
52,....
..........
........
54
(}-
>f---
53
Appendix 4
336
55
69
~~
@Ji;J/JIY
~~
71
1:1
72
>l>>H>
~
~
75
"'
!21
k8J
~
-Q
l'l-
57
>-
>+-
58
>-<
59
'X'
60
61
62
63
..
64
65
..
--
-9-
-6-
-0-
cp
66~
6 7 = ;::::::::> 0
68
:=:
:::::r=
i.77~
l!l
56
337
Note: map symbols used for topographic, hydrographic, and cultural features
are explained and illustrated in color on the folded sheet "Topographic Maps,"
free on application to the Director, U.S. Geological Survey, Washington 25, D.C.
I__,
,
oo
10
20
so
40
'
--
Sine eosin Sine Co sin Sine Cosin Sine Co sin Sine Cosin
.06976 .99756 60
0 .00000 One. .01745 .99985 .03490 .99939 .05234 .99863
.99861 .07005 .99754 59
1 .00029 One. .01774 .99984 .03519 .99938 .05263
.99860 .07034 .99752 58
2 00058 One . .01803 .99984 .03548 .99937 .05292 .99858
.07063 .99750 57
3 00087 One . .01832 .99983 .03577 .99936 .05321
.99857 .07092 .99748 56
4 .00116 One. .01862 .99983 .03606 .99935 .05350
.07121 .99746 55
.99855
.05379
.99934
.03635
.99982
.01891
.00145
One.
5
.07150 .99744 54
6 .00175 One. .01920 .99982 .03664 .99933 .05408 .99854
.07179 .99742 53
.99852
.05437
.99932
.03693
.99981
.01949
.00204
One.
7
.99851
.07208 .99740 52
.05466
8 00.233 One .01978 .99980 .03723 .99931
.99849
.07237 .99738 51
.05495
.99930
.03752
.99980
.02007
.00262
One.
9
.07266 .99736 50
.99847
10 .00291 One. .02036 .99979 .03781 .99929 .05524
.07295 .99734 49
11 .00320 .99999 .02065 .99979 .03810 .99927 .05553 .99846
.07324 .99731 48
12 .00349 .99999 .02094 .99978 .03839 .99926 .05582 .99844
.07353 .99729 47
13 .00378 .99999 .02123 .99917 .03868 .99925 .05611 .99842
.07382 .99727 46
.99841
.05640
.99924
.03897
14 .00407 .99999 .02152 .99911
.07411 .99725 45
15 .00436 .99999 .02181 .99976 .03926 .99923 .05669 .99839
.07440 .99723 44
.99838
.05698
.99922
.03955
16 .00465 .99999 .02211 .99976
.07469 .99721 43
17 .00495 .99999 .02240 .99975 .03984 .99921 .05727 .99836
.07498 .99719 42
.99834
.05756
.99919
.04013
18 .00524 .99999 .02269 .99974
.99716 41
19 .00553 .99998 .02298 .99974 .04042 .99918 .05785 .99833 .07527
.07556 .99714 40
.99831
.05814
.99917
.99973
.04071
20 .00582 .99998 .02327
.07585 .99712 39
21 .006ll .99998 .02356 .99972 .04100 .99916 .05844 .99829
.05873 .99827 .07614 .99710 38
22 .00640 .99998 .02385 .99972 .04129 .99915 .05902
.07643 .99708 37
.99826
23 .00669 .99998 .02414 .99911 .04159 .99913
.07672 .99705 36
24 .00698 .99998 .02443 .99970 .04188 .99912 .05931 .99824
.99703 35
.99822
25 .00727 .99997 .02472 .99969 .04217 .99911 .05960 .99821 .07701
.07730 .99701 34
.05989
.99910
.04246
.99969
.02501
.99991
.00756
26
.99699 33
.99819
.07759
27 .00785 .99997 .02530 .99968 .04275 .99909 .06018
.99696 32
28 .00814 .99991 .02560 .99967 .04304 .99901 .06047 .99817 .07788
.99694 31
.07817
.99815
29 .00844 .99996 .02589 .99%6 .04333 .99906 .06076
.9%92
30
.07846
.99813
.06105
.99905
.04362
.99%6
.99996
.02618
.00873
30
.07875 .99689 29
31 .00902 .99996 .02647 .99965 .04391 .99904 .06134 .99812 .07904
.99687
28
32 .00931 .99996 .02676 .99964 .04420 .99902 .06163 .99810
.07933 .99685 27
33 .00%0 .99995 .02705 .99963 .04449 .99901 .06192 .99808
.99683
26
.99806 .07962
34 .00989 .99995 .02734 .99963 .04478 .99900 .06221 .99804
.07991 .99680 25
35 .01018 .99995 .02763 .99%2 .04507 .99898 .06250 .99803 .08020
24
36 .01047 .99995 .02792 .99961 .04536 .99897 .06279 .99801 .08049 .99678
.99676 23
.06308
.99896
.04565
.99960
.02821
37 .01076 .99994
.99673
22
.08078
38 .01105 .99994 .02850 .99959 .04594 .99894 .06337 .99799
.99197 .08107 .99671 21
39 .01134 .99994 .02879 .99959 .04623 .99893 .06366 .99795
.99668
20
.08136
40 .01164 .99993 .02908 .99958 .04653 .99892 .06395
.08165 .99666 19
41 .Oll93 .99993 .02938 .99957 .04682 .99890 .06424 .99793 .08194
.99664 18
.99792
42 .01222 .99993 .02967 .99956 .047.11 .99889 .06453 .99790
.08223 .99661 17
43 .01251 .99992 .02996 .99955 .04740 .99888 .06462 .99788
.99659
16
.08252
.06511
44 .01280 .99992 .03025 99954 .04769 .99886
.08281 .99657 15
45 .01309 .99991 .03054 .99953 .04798 .99885 .06540 .991~
.99654
14
.08310
.99784
.06569
46 .01338 .99991 .03083 .99952 .04827 .99883
.99782 .08339 .99652 13
47 .01367 .99991 .03ll2 .99952 .04856 .99882 .06598 .99780
.08368 .99649 12
48 .01396 .99990 .03141 .99951 .04885 .99881 .06627
.99647 11
49 .01425 .99990 .03!'70 .99950 .04914 .99879 .06656 .99778 .08397
.08426 .99644 10
.99776
.06685
60 .01454 .99989 .03199 .99949 .04943 .99878
.99774 .08455 .99642 9
61 .01483 .99989 .03228 .99948 .04972 .99876 .06714
.99639 8
52 .01513 .99989 .03257 .99947 .05001 .99875 .06743 .99772 .08484
.99770 .08513 .99637 7
53 .01542 99988 .03286 .99946 .05030 .99873 .06773 .99768
.99635 6
.08542
.06802
.99872
.05059
.99945
.03316
54 .01571 .99988
,08571 .99632 5
55 .01600 .99987 .03345 .99944 .05088 .99870 .06831 .99766
4
.99630
.08600
.99764
.06860
.99869
.05117
.99943
.03374
56 .01629 .99987
.08629 .99627 3
57 .01658 .99986 .03403 .99942 .05146 .99867 .06889 .99762 :o8658
2
.99625
.99760
.06918
.99866
.05175
.99941
.03432
58 .01687 .99986
.99622 1
59 .01716 .99985 .03461 .99940 .05205 .99864 .06947 .99758 .08687
0
.99619
.08716
.99756
.06976
.99863
.05234
.99939
.03490
.99985
60 .01745
eosin Sine eosin Sine Cosin Sine eosin Sine Cosin Sine
'
89
ss
s7
340
so
srs
'
341
Appendix 7
0'
'I
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
.. . I
57
58
59
60
--
. go
90
Sine Cosin
.12187 .99255
.12216.99251
.12245 .99248
.12274.99244
.12302 .99240
.12331 .99237
.12360.99233
.12389 .99230
.12418 .99226
.12447.99222
.12476 .99219
.12504 .99215
.12533 .99211
.12562 .99208
.12591 .99204
.12620 .99200
.12649 .99197
.12678 .99193
.12706 .99189
.12735 .99186
.12764 .99182
.12793 .99178
.12822 .99175
.12851 .99171
.12880 .99167
.12908 .99163
.12937 .99160
.12966 .99156
.12995 .99152
.13024.. 99148
.13053 .99144
.13081 .99141
.13ll0 .99137
.13139 .99133
.13168.99129
.13197 .99125
.13226 .99122
.13254 .99118
.13283.99114
.13312 .99110
.13341 .99106
.13370 .99102
.13399.99098
.13427 .99094
.13456 .99091
.13~85 .99087
.13514.99083
.13543 .99079
.13572 .99075
.13600 .99071
.13629 .99067
.13658.99063
.13687 .99059
.13716.99055
.13744.99051
.13773.99047
.13802.99043
.13831 .99039
.13860.99035
.13889 .99031
.13917 .99027
Sine Cosin
.13917 .99027
.13946.99023
.13975 .99019
.14004 .99015
.14033 .990ll
.14061 .99006
.14090.99002
.14ll9 .98998
.14148 .98994
.14177.98990
.14205 .98986
.14234 .98982
.14263 .98978
.14292 .98973
.14320 .98969
.14349 .98965
.14378 .98961
.14407 .98957
.14436 .98953
.14464 .98948
.14493 .98944
.14522 .98940
.14551 .98936
.14580 .98931
.14608 .98927
.14637 .98923
.14666 .98919
.14695 .98914
.14723 .98910
.14752 .98906
.14781 .98902
.14810.98897
.14838 .98893
.14867 .98889
.14896.98884
.14925 .98880
.14954 .98876
.14982 .98871
.150ll .98867
.15040.98863
.15069.98858
.15097 .98854
.15126.98849
.15155 .98845
.15184 .98841
.15212 .98836
.15241 .98832
.15270 .98827
.15299 .98823
.15327 .98818
.15356.98814
.15385.98809
.15414 .98805
.15442.98800
.15471 .98796
.15500.98791
.15529.98787
.15557 .98782
.15586.98778
.15615 .98773
.15643 .98769
Sine Cosin
.15643.98769
.15672.98764
.15701 .98760
.15730.98755
.15758 .98751
.15787 .98746
.15816.98741
.15845.98737
.15873 .98732
.15902 .98728
.15931 .98723
.15959 .98718
.15988.98714
.16017 .98709
.16046 .98704
.16074 .98700
.16103 .98695
.16132 .98690
.16160 .98686
.16189 .98681
.16218 .98676
.16246.98671
.16275.98667
.16304 .98662
.16333 .98657
.16361 .98652
.16390 .98648
.16419 .98643
.16447 .98638
.16476.98633
.16505 .98629
.16533 .98624
.16562 .98619
.16591 .98614
.16620.98609
.16648 .98604
.16677 .98600
.16706 .98595
.16734 .98590
.16763 .98585
.16792 .98580
.16820 .98575
.16849.98570
.16878 .98565
.16906 .98561
.16935 .98556
.16964 .98551
.16992 .98546
.17021 .98541
.17050 .98536
.17078.98531
.17107.98526
.17136 .98521
.17164.98516
.17193.98511
.17222.98506
.17250.98501
.17279.98496
.17308.98491
.17336 .98486
.17365 .98481
eosin I Sine
Cosin I Sine
Cosin I Sine
50
7o
Sine eosin
.08716.99619
.08745.99617
.08774 .99614
.08803 .99612
.08831 .99609
.08860 .99607
.08889.99604
.08918.99602
.08947 .99599
.08976.99596
.09005 .99594
.09034 .99591
.09063 99588
.09092 .99586
.09121 .99583
.09150 .99580
.09179.99578
.09208.99575
.09237 .99572
.09266 .99570
.09295 .99567
.09324.99564
.09353 .99562
.09382 .99559
.09411 .99556
.09440 .99553
.09469 .99551
.09498 .99548
,09527 .99545
.09556.99542
.09585 .99540
.09614.99537
.09642 .99534
.09671 .99531
.09700.99528
.09729 .99526
.09758 .99523
.09787 .99520
.09816 .99517
.09845.99514
.09874 .995ll
.09903 .99508
.09932.99506
.09961 .99503
.09990 .99500
.10019 .99497
.10048.99494
.10077 .99491
.10106 .99488
.10135 .99485
.10164 .99482
.10192.99479
.10221 .99476
.10250.99473
.10279.99470
.10308.99467
.10337.99464
.10366.99461
.10395.99458
.10424 .99455
.10453 .99452
Sine Cosin
.10453 .99452
.10482.99449
.105ll .99446
.10540.99443
.10569 .99440
.10597 .99437
.10626.99434
.10655.99431
.10684 .99428
:10713.99424
.10742 .99421
.10771 .99418
.10800 .99415
.10829 .99412
.1085 199409
.10887 .99406
.10916.99402
.10945.99399
.10973 .993%
.ll002 .99393
.11031 .99390
.11060.99386
.ll089 .99383
.11118 .99380
.11147 .99377
.lll76 .99374
.ll205 .99370
.ll234 .99367
.ll263 .99364
.11291 .99360
.ll320 .99357
.ll349 .99354
.ll378 .99351
.ll407 .99347
.ll436 .99344
.ll465 .99341
.ll494 .99337
.ll523 .99334
.ll552 .99331
.11580 .99327
.ll609 .99324
.11638 .99320
.11667 .99317
.11696 .99314
.11725 .99310
.11754 .99307
.11783 .99303
.ll812 .99300
.ll840 .99297
.11869 .-99293
.ll898 .992~
.11927.99286
.11956 .99283
.11985.99279
.12014.99276
.12043 .99272
.12071 .99269
.12100.99265
.12129.99262
.12158 .99258
.12187 .99255
eosin I Sine
Cosin I Sine
ss
82
81
so
l
l
6i}
59
58
57
56
55
54
53
52
51
50
49.
48
47
46
45
44
43
42
41
40
39
38
37
36
35
34
33
32
31
30
29
2827
26
25
24
23
22
21
20
19
18
17
16
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
0
Appendix 7
342
I
to
11
12
'
-~
eosin
Sine
79
Sine eosin
.20791 .97815
.20820 .97809
.20848 .97803
.20877 .97797
.20905 .9779I
.20933 .97784
.20962 .97778
.20990 .97772
.210I9 .97766
.21047 .97760
.21076 .97754
.21104 .97748
.21132 .97742
.21161 .97735
.21189 .97729
.21218 .97723
.21246 .97717
.21275 .97711
.21303 .97705
.21331 .97698
.2I360 .97692
.21388 .97686
.21417 .97680
.21445 .97673
.21474 .97667
.21502 .97661
.21530 .97655
.21559 .97648
.21587 .97642
.21616 .97636
.21644 .97630
.21672 .97623
.21701 .97617
.21729 .97611
.21758 .97604
.21786 .97598
.21814 .97592
.21843 .97585
.21871 .97579
.21899 .97573
.21928 .97566
.21956 .97560
.21985 .97553
.22013 .97547
.22041 .97541
.22070 .97534
.22098 .97528
.22126 .97521
.22155 .97515
.22183 .97508
.222I2 .97502
.22240 .97496
.22268 .97489
.22297 .97483
.22325 .97476
.22353 .97470
.22382 .97463
.22410 .97457
.22438 .97450
.22467 .97444
.22495 .97437
Cosin I Sine Cos in Sine
78
77
13
Sine Cosin
.22495 .97437
.22523 .97430
.22552 .97424
.22580 .97417
.22608 .97411
.22637 .97404
.22665 .97398
.22693 .97391
.22722 .97384
.22750 .97378
.22778 .97371
.22807 .97365
.22835 .97358
.22863 .97351
.22892 .97345
.22920 .97338
.22948 .9733I
.22977 .97325
.23005 .97318
.23033 .97311
.23062 .97304
.23090 .97298
.23118 .9729I
.23146 .97284
.23175 .97278
.23203 .97271
.23231 .97264
.23260 .97257
.23288 .97251
.23316 .97244
.23345 .97237
.23373 .97230
.23401 .97223
.23429 .97217
.23458 .97210
.23486 .97203
.23514 .97196
.23542 .97189
.23571 .97182
.23599 .97176
.23627 .97169
.23656 .97162
.23684 .97155
.23712 .97148
.23740 .97141
.23769 .97134
.23797 .97127
.23825 .97120
.23853 .97113
.23882 .97106
.23910 .97100
.23938 .97093
.23966 .97086
.23995 .97079
.24023 .97072
.24051 .97065
.24079 .97058
.24108 .97051
.24136 .97044
.24164 .97037
.24192 .97030
Co sin Sine
76
14
Sine
.24192
.24220
.24249
.24277
.24305
.24333
.24362
.24390
.24418
.24446
.24474
.24503
.24531
.24559
.24587
.24615
.24644
.24672
.24700
.24728
.24756
.24784
.24813
.24841
.24869
.24897
.24925
.24954
.24982
.25010
.25038
.25066
.25094
.25122
.25151
.25179
.25207
.25235
.25263
.25291
.25320
.:/5348
.25376
.25404
.25432
.25460
.25488
.25516
.25545
.25573
.25601
.25629
.25657
.25685
.25713
.25741
.25769
.25798
.25826
.25854
.25882
Co sin
eosin
~97030
.97023
.97015
.97008
.97001
.96994
.96987
.96980
.96973
.96966
.96959
.96952
.96945
.96937
.96930
.96923
.96916
.96909
.96902
.96894
.96887
.96880
.96873
.96866
.96858
.96851
.96844
.96837
.96829
.96822
.96815
.96807
.96800
.96793
.96786
.96778
.96771
.96764
.96756
.96749
.96742
.96734
.96727
.96719
.96712
.96705
.96697
.96690
.96682
.96675
.96667
.96660
.96653
.96645
.96638
.96630
.96623
.96615
.96608
.96600
.96593
Sine
75
u~o
59
58
57
56
55
54
53
52
51
50
49
48
47
46
45
44
43
42
41
40
39
38
37
36
35
34
33
32
31
30
29
28
27
26
25
24
23
22
21
20
19
18
17
16
15
I4
13
12
11
10
9
8
7
6
5
4
3
2
I
0
I
16
'
'
6iJ
1
17
343
tso
19
'
Sine Cos1n Sine Cosin Sine eosin Sine Cosln Sine Cosin
-0- .25882 .96593 ::27564 .96126 .29237 :95630 .30902
.95106 .32557 .94552 60
I .25910 .96585 .27592 .96118 .29265 .95622 .30929 .95097 .32584 .94542 59
2 .25938 .96578 .27620 .961~0 .29293 .956i3 .30957 .95088 .32612 .94533 58
3 .25966 .96570 .27648 .96102 .29321 .95605 .30985
.32639 .94523 57
4 .25994 .96562 .27676 .95094 .29348 .95596 .31012 :~~8iZ .32667 .94514 56
5 .26022 .96555 .27704 .96086 .29376 .95588 .31040 .95061 .32694 .94504 55
6 .26050 .96547 .27731 .96078 .29404 .95579 .31068 .95052 .32722 .94495 54
7 .26079 .96540 .27759 .96070 .29432 .95571 .31095 .95043 .32749 .94485 53
8 .26107 .96532 .27787 .96062 .29460 .95562 .31123 .95033 .32777 .94476 52
9 .26135 .96524 .27815 .96054 .29487 .95554 .31151 .95024 .32804 .94466 51
10 .26163 .96517 .27843 .96046 .29515 .95545 .3ll78 .95015 .32832 .94457 50
11 .26191 .96509 .27871 .96037 .29543 .95536 .31206 .95006 .32859 .94447 49
12 .26219 .96502 .27899 .96029 .29571 .95528 .31233 .94997 .32887 .94438 48
13 .26247 .96494 .27927 .96021 .29599 .95519 .31261 .94988 .32914 .94428 47
14 .26275 .96486 .2795&1'96013 .29626 .95511 .31289 .94979 .32942 .94418 46
.26303 .96479 .27983 .96005 .29654 .95502 .31316 .94970 .32969 .94409 45
.26331 .96471 .28011 .95997 .29682 .95493 .31344 .94961 .32997 .94399 44
17 .26359 .96463 .28039 .95989 .29710 .95485 .31372 .94952 .33024 .94390 43
18 .26387 .96456 .28067 .95981 .29737 .95476 .31399 .94943 .33051 .94380 42
19 .26415 .96448 .28095 .95972 .29765 .95467 .31427 .94933 .33079 .94370 41
20 .26443 .96440 .28123 .95964 .29793 .95459 .31454 .94924 .33106 .94361 40
21 .26471 .96433 .28150 .95956 .29821 .95450 .31482 .94915 .33134 .94351 39
22 .26500 .96425 .28178 >95948 .29849 .95441 .31510 .94906 .33161 .94342 38
23 .26528 .96417 .28206 .95940 .29876 .95433 .31537 .94897 .33189 .94332 37
24 .26556 .96410 .28234 .95931 .29904 .95424 .31565 .94888 .33216 .94322 36
25 .26584 .96402 .28262 .95923 .29932 .95415 .31593 .94878 .33244 .94313 35
26 .26612 .96394 .28290 .95915 .29960 .95407 .31620 .94869 .33271 .94303 34
27 .26640 .96386 .28318 .95907 .29987 .95398 .31648 .94860 .33298 .94293 33
28 .26668 .96379 .28346 .95898 .30015 .95389 .31675 .94851 .33326 .94284 32
29 .26696 .96371 .28374 .95890 .30043 .95380 .31703 .94842 .33353 .94274 31
30 .26724 .96363 .28402 .95882 .30071 .95372 .31130 .94832 .33381 .94264 30
31 .26752 .96355 .28429 .95874 .30098 .95363 .31758 .94823 .33408 .94254 29
32 .26780 .96347 .28457 .95865 .30126 .95354 .31786 .94814 .33436 .94245 28
33 .26808 .96340 .28485 .95857 .30154 .95345 .31813 .94805 .33463 .94235 27
34- ~26836 .96332 .28513 .95849 .30182 .95337 .31841 .94795 .3..>490 .94225 26
35 .26864 .96324 .28541 .95841 .30209 .95328 .31868 .94786 .33518 .94215 25
36 .26892 .96316 .28569 .95832 .30237 .95319 .31896 .94777 .33545 .94206 24
37 .26920 .96308 .28597 .95824 .30265 .95310 .31923 .94768 .33573 .94196 23
38 .26948 .96301 .28625 .95816 .30292 .95301 .31951 .94758 .33600 .94186 22
39 .26976 .96293 .28652 .95807 .30320 .95293 .31979 .94749 .33627 .94176 21
40 .27004 .96285 .28680 .95799 .30348 .95284 .32006 .94740 .33655 .94167 20
41 .27032 .96277 .28708 .95791 .30376 .95275 .32034 .94730 .33682 .94157 19
42 .27060 .96269 .28736 .95782 .30403 .95266 .32061 .94721 .33710 .94147 18
43 .27088 .96261 .28764 .95774 .30431 .95257 .32089 .94712 .33737 .94137 17
44 .27116 .96253 .28792 .95766 .30459 .95248 .32116 .94702 .33764 .94127 16
45 .27144 .96246 .28820 .95757 .30486 .95240 .32144 .94693 .33792 .94118 15
46 .27172 .96238 .28847 .95749 .30514 .95231 .32171 .94684 .33819 .94108 I4
47 .27200 .96230 .28875 .95740 .30542 .95222 .32199 .94674 .33846 .94098 13
48 .27228 .96222 .28903 .95732 .30570 .95213 .32227 .94665 .33874 .94088 12
49 .27256 .96214 .28931 .95724 .30597 .95204 .32254 .94656 .33901 .94078 11
50 .27284 .96206 .28959 .95715 .30625 .95195 .32282 .94646 .33929 .94068 10
51 .27312 .96198 .28987 .95707 .30653 .95186 .32309 .94637 .33956 .94058 9
52 .27340 .96190 .29015 .95698 .30680 .95177 .32337 .94627 .33983 .94049 8
53 .27368 .96182 .29042 .95690 .30708 .95168 .32364 .94618 .34011 .94039 7
54 .27396 .96174 .29070 .95681 .30736 .95159 .32392 .94609 .34038 .94029 6
55 .27424 .96166 .29098 .95673 .30763 .95150 .32419 .94599 .34065 .94019 5
56 .27452 .96158 .29126 .95664 .30791 .95142 .32447 .94590 .34093 .94009 4
57 .27480 .96150 .29154 .95656 .30819 .95133 .32474 .94580 .34120 .93999 3
58 .27508 .96142 .29182 .95647 .30846 .95124 .32502 .94571 .34147 .93989 2
59 .27536 .96134 .29209 .95639 .30874 .95115 .32529 .94561 .34175 .93979
1
60 .27564 .96126 .29237 .95630 .30902 .95106 .32557 .94552 .34202 .93969 0
eosin Sine eosin Sine eosin Sine Cosin Sine eosin Sine
.u
'
74
73
72
71
70- - - ' -
-----
344
20
Sine Cosin
-0- .34202 .93969
1 .34229 .93959
2 .34257 .93949
3 .34284 .93939
4 .34311 .93929
5 .34339 .93919
6 .34366 .93909
7 .34393 .93899
8 .34421 .93889
9 .34448 .93879
10 .34475 .93869
ll .34503 .93859
12 .34530 .93849
13 .34557 .93839
14 .34584 .93829
15 .34612 .93819
16 .34639 .93809
17 .34666 .93799
18 .34694 .93789
19 .34721 .93779
20 .34748 .93769
21 .34775 .93759
22 .34803 .93748
23 .34830 .93738
24 .34857 .93728
25 .34884 .93718
26 .34912 .93708
27 .34939 .93698
28 .34966 .93688
29 .34993 .93677
30 .35021 .93667
31 .35048 .93657
32 .35075 .93647
33 .35102 .93637
34 .35130 .93626
35 .35157 .93616
36 .35184 .93606
37 .35211 .93596
38 .35239 .93585
39 .35266 .93575
40 .35293 .93565
41 .35320 .93555
42 .35347 .93544
43 .35375 .93534
44 .35402 .93524
45 .35429 .93514
46 .35456 .93503
47 .35484 .93493
48 .35511 .93483
49 .35538 .93472
50 .35565 .93462
51 .35592 .93452
52 .35619 .93441
53 .35647 .93431
54 .35674 .93420
55 .35701 .93410
56 .35728 .93400
57 .35755 .93389
58 .35782 .93379
59 .35810 .93368
60 .3582>7 .93358
Cos in Sine
I
69
21
22
23
24
'
68
67
. 66
65
'
345
Appendix 7
I
25
Sine Cos in
-0- .42262 .90631
1 .42288 .90618
2 .42315 .90606
3 .42341 .90594
4 .42367 .90582
5 .42394 .90569
6 .42420 .90557
7 .42446 .90545
8 .42473 .90532
9 .42499 .90520
10 .42525 .90507
11 .42552 .90495
12 .42578 .90483
13 .42604 .90470
14 .42631 .90458
15 .42657 .90446
16 .42~ .90433
17 .42 09 .90421
18 .42736 .90408
19 .42762 .90396
20 12788 .90383
21
.90371
841 .90358
22
2l .42867 .90346
24 .42894 .90334
25 .42920 .90321
26 .42946 .90309
27 .42972 .90296
28 .42999 .90284
29 .43025 .90271
30 .43051 .90259
31 .43077 .90246
32 .43104 .90233
33 .43130 .90221
34 .43156 .90208
25 .43182 .90196
36 .43209 .90183
37 .43235 .90171
38 .43261 .90158
39 .43287 .90146
40. .43313 .90133
41 .43340 .90120
42 .43366 .90108
-43 .43392 .90095
44 .43418 .90082
45 .43445 .90070
46 .43471 .90057
47 .43497 .90045
. 48 .43523 .90032
49 .43549 .90019
50 .43575 .90007
51 .43602 .89994
52 .43628 .89981
53 .43654 .89968
54 .43680 .89956
' 65 .43706 .89943
56 .43733 .89930
57 .43759 .89918
.43785 .89905
~g .43811
.89892
60 .43837 .89879
Cosin Sine
~r15
'
64
26
27
28
Sine Cosin
.43837 .89879
.43863 .89867
.43889 .89854
.43916 .89841
.43942 .89828
.43968 .89816
.43994 .89803
.44020 .89790
.44046 .89777
.44072 .89764
.44098 .89752
.44124 .89739
.44151 .89726
.44177
.44203
.44229 .89687
.44255 .89674
.44281 .89662
.44307 .89649
.44333 .89636
.44359 .89623
.44385 .89610
.44411 .89597
.44437 .89584
.44464 .89571
.44490 .89558
.44516 .89545
.44542 .89532
.44568 .89519
.44594 .89506
.44620 .89493
.44646 .89480
.44672 .89467
.44698 .89454
.44724 .89441
.44750 .89428
.44776 .89415
.44802 .89402
.44828 .89389
.44854 .89376
.44880 .89363
.44906 .89350
.44932 .89337
.44958 .89324
.44984 .89311
.45010 .89298
.45036 .89285
.45062 .89272
.45088 .89259
.45114 .89245
.45140 .89232
Sine Cosin
.45399 .89101
.45425 .89087
.45451 .89074
.45477 .89061
.45503 .89048
.45529 .89035
.4S554 .89021
.45580 .89008
.45606 .88995
.45632 .88981
.45658 .88968
.45684 .88955
.45710 .88942
.45736 .88928
.45762 .88915
.45787 .88902
.45813 .88888
.45839 .88875
.45865 .88862
.45891 .88848
.45917 .88835
.45942 .88822
.45968 .88808
.45994 .88795
.46020 .88782
.46046 .88768
.46072 .88755
.46097 .88741
.46123 ,88728
.46149 .88715
.46175 .88701
.46201 .88688
,46226 .88674
.46252 .88661
.46278 .88647
.46304 .88634
,46330 .88620
,46355 .88607
.46381 .8b593
.46407 .88580
.46433 .88566
.46458 .88553
.46484 .88539
.46510 .88526
.46536 .88512
.46561 .88499
,46587 .88485
.46613 .88472
.46639 .88458
.46664 .88445
.46690 .88431
.45166
.45192
.45218
.45243
.45269
.45295
.45321
.45347
.45373
.45399
Cosin
.46716 .88417
.46742 .88404
.46767 .88390
.46793 .88377
.46819 .88363
.46844 .88349
.46870 .88336
.46896 .88322
.46921 .88308
.46947 .88295
Cosln Sine
Sine Co sin
.46947 .88295
.46973 .88281
.46999 .88267
.47024 .88254
.47050 .88240
.47076 .88226
.47101 .88213
.47127 .88199
.47153 .88185
.47178 .88172
.47204 .88158
.47229 .88144
.47255 .88130
.47281 .88117
.47306 .88103
.47332 .88089
.47358 .88075
.47383 .88062
.47409 .88048
.47434 .88034
.47460 .88020
.47486 .88006
.47511 .87993
.47537 .87979
.47562 .87965
.47588 .87951
.47614 .87937
.47639 .87923
.47665 .87909
.47690 .87896
.47716 .87882
.47741 .87868
.47767 .87854
.47793 .87840
.47818 .87826
.47844 .87812
.47869 .87798
.47895 .87784
.47920 .87770
.47946 .87756
.47971 .87743
.47997 .87729
.48022 .87715
.48048 .87701
.48073 .87687
.48099 .87673
.48124 .87659
.48150 .87645
.48175 .87631
.48201 .87617
.48226 .87603
.48252 .87589
.48277 .87575
.48303 .87561
.48328 .87546
.48354 .87532
.48379 .87518
.48405 .87504
.48430 .87490
.48456 .87476
.48481 .87462
Cosin Sine
62
61
:~m~
.89219
.89206
.89193
.89180
.89167
.89153
.89140
.89127
.89114
.89101
Sine
63
29
'
Sine Cosin
.48481 .87462 60
.48506 .87448 59
.48532 .87434 58
.48557 .87420 57
.48583 .87406 56
.48608 .87391 55
.48634 .87377 54
.48659 .87363 53
.48684 .87349 52
.48710 .87335 51
.48735 .87321 50
.48761 .87306 49
.48786 .87292 48
.48811 .87278 47
.48837 .87264 46
.48862 .87250 45
.48888 .87235 44
.48913 .87221 43
.48938 .87207 42
.48964 .87193 41
.48989 .87178 40
.49014 .87164 39
.49040 .87150 38
.49065 .87136 37
.49090 .87121 36
.49116 .87107 35
.49141 .87093 34
.49166 .87079 33
.49192 .87064 32
.49217 .87050 31
.49242 .87036 30
.49268 .87021 29
.49293 .87007 28
.49318 .86993 27
.49344 .86978 26
.49369 .86964 25
.49394 .86949 24
.49419 .86935 23
.49445 .86921 22
.49470 .86906 21
.49495 .86892 20
.49521 .86878 19
.49546 .86863 18
.49571 .86849 17
.49596 .86834 16
.49622 .86820 15
.49647 .86805 14
.49672 .86791 13
.49697 .86777 12
.49723 .86762 11
.49748 .86748 10
.49773 .86733 9
.49798 .86719 8
.49824 .86704 7
.49849 .86690 6
.49874 .86675 5
4
.49899 .86661
.49924 .86646 3
.49950 .86632 2
1
.49975 .86617
.50000 .86603 0
Cosln Sine
60
346
30"
Sine Cosin
31
32
33
34
.51504 .85717
.51529 .85702
.51554 .85687
.51579 .85672
.51604 .85657
.51628 .85642
.51653 .85627
.51678 .85612
.51703.85597
.51728 .85582
.51753 .85567
.51778 .85551
.51803 .85536
.51828 .85521
.51852 .85506
.51877 .85491
.51902 .85476
.51927 .85461
.51952 .85446
.51977 .86431
.52002 .85416
.62026 .85401
.52051 .85385
.52076 .85370
.52101 .85355
.52126 .85340
.52151 .85325
.52175 .85310
.52200 .85294
.52225 .85279
.52250 .85264
.52275 .85249
.52299 .85234
.52324 .85218
.52349 .85203
.52374 .85188
.52399.85173
.52423 .85157
.52448 .85142
.52473 .85127
.52498 .85112
.52522 .85096
.52547 .85081
.52572 .85066
.52597 .85051
.52621 .85035
.52646 .85020
.52671 .85005
.52696 .84989
.52720 .84974
.52745 .84959
.52770 .84943
.52794 .84928
.52819 .84913
.52844 .84897
.52869 .84882
.52893 .84866
.52918 .84851
.52943 .84836
.52967 .84820
.52992 .84805
.52992 .84805
.53017 .84789
.53041 .84774
.53066 .84759
.53091 .84743
.53ll5 .84728
.53140 .84712
.53164 .84697
.53189.84681
.53214 .84666
.53238 .84650
.53263 .84635
.53288 .84619
.53312 .84604
.53337 .84588
.53361 .84573
.53386 .84557
.53411 .84542
.53435 .84526
.53460 .84511
.53484 .84495
.53509 .84480
.53534 .84464
.53558 .84448
.53583 .84433
.53607 .84417
.53632 .84402
.53656 .84386
.53681 .84370
.53705.84355
.53730 .84339
.53754 .84324
.53779 .84308
.53804 .84292
.53828 .84277
.53853 .84261
.53877.84245
.53902 .84230
.53926 .84214
.53951 .84198
.53975 .84182
.54000 .84167
.54024 .84151
.54049 .84135
.54073 .84120
.54097 .84104
.54122 .84088
.54146 .84072
.54171 .84057
.54195 .84041
.54220 .84025
.54244 .84009
.54269 .83994
.54293 .83978
.54317 .83962
.54342 .83946
.54366 .83930
.54391 .83915
.54415 .83899
.54440 .83883
.54464 .83867
.54464 .83867
.54488 .83851
.54513 .83835
.54537 .83819
.54561 .85804
.54586 .83788
.54610 .83772
.54635 .83756
.54659 .83740
.54683 .83724
.54708 .83708
.54732 .83692
.54756 .83676
.54781 .83660
.54805 .83645
.54829 .83629
.54854 .83613
.54878 .83597
.54902 .83581
.54927 .83565
.54951 .83549
.54975 .83533
.54999 .83517
.55024 .83501
.55048 .83485
.55072 .83469
.55097 .83453
.55121 .83437
.55145 .83421
.55169 .83405
.55194 .83389
.55218 .83373
.55242 .83356
.65266 .83340
.55291 .83324
.55315 .83308
.55339 .83292
.55363 .83276
.55388 .83260
.55412 .83244
.55436 .83228
.55460 .83212
.55484 .83195
.55509 .83179
.55533 .83163
.55557 .83147
.55581 .83131
.55605 .83115
.55630 .83098
.55654 .83082
.55678 .83066
.55702 .83050
.55726 .83034
.55750 .83017
.55775 .83001
.55799 .82985
.55823 .82969
.55847 .82953
.55871 .82936
.55895 .82920
.55919 .82904
Cosln Sine
Cosin Sine
Cosin Sine
Cosin Sine
Cosin Sine
5UV
58
57
56
51So
---------------------
I
l
I
i
.~
!
'
II
I
I
Appendix 7
I
31So
36
37
Sine Cosln
Sine Cosin
.58779 .80902
.58802 .80885
.58826 .80867
.58849 .80850
.58873 .80833
.58896 .80816
.58920 .80799
.58943 .80782
.58967 .80766
.58990 .80748
.59014 .80730
.59037 .80713
.59061 .80696
.59084 .80679
J60662
.80644
.59154 .80627
.59178 .80610
.59201 '.80593
.59225 .80576
.59248 .80558
.59272 .80541
.5'1295 .80524
.59318 .80507
.59342 .80489
.59365 .80472
.59389 .80455
.59412 .80438
.59436 .80420
.59459 .80403
.59482 .80386
.59506 .80368
.59529 .80351
.59552 .80334
.59576 .80316
.59599 .80~99
.59622 .80282
.59646 .80264
.59669 .80247
.59693 .80230
.59716 .80212
.59739 .80195
.59763 .80178
.59786 .80160
.59809 .80143
.59832 .80125
.59856 .80108
.59879 .80091
.59902 .80073
.59926 .80056
.59949 .80038
.59972 .80021
.59995 .80003
.60019 .79986
.60042 .79968
.60065 .79951
.60089 .79934
.60112 .79916
.60135 .79899
.60158 .79881
.60182 .79864
.60182 .79864
.60205 .79846
.60228 . 79829
.60251 .79811
.6027 4 79793
.60298 . 79776
.60321 .79758
.60344 .79741
.60367 .79723
.60390 .79706
.60414 .79688
.60437 .79671
.60460 .79653
.60483 .79635
.60506 .79618
.60529 .79600
.60553 .79583
.60576 .79565
.60599 .79547
.60622 .79530
.60645 .79512
.60668 .79494
.60691 .79477
.60714 .79459
.60738 .79441
.60761 .79424
.60784 .79406
.60807 .79388
.6'0830 .79371
.60853 .79353
.60876 .79335
.60899 .79318
.60922 .79300
.60945 .79282
.60968 .79264
.60991 .79247
.61015 .79229
.61038 .79211
.61061 .79193
.61084 .79176
.61107 .79158
.61130 .79140
.6ll53 .79122
.61176 .79105
.61199 .79087
.61222 .79069
.61245 .79051
.61268 .79033
.61291 .79016
.61314 .78998
.61337 .78980
.61360 .78962
.61383 .78944
.61406 .78926
.61429 .78908
.61451 .78891
.61474 .78873
.61497 .78855
.61520 .78837
.61543 .78819
.61566 .78801
.62932 . 77715
.61589 .78783 .62955 .77696
.61612 .78765 .62977 .77678
.61635 .78747 .63000 .77660
.61658 .78729 .63022 .77641
.61681 .78711 .63045 .77623
.61704 .78694 .63068 .77605
.61726 .78676 .63090 .77586
.61749 .78658 .63113 .77568
.61772 .78640 .63135 .77550
.61795 .78622 .63158 .77531
.61818 .78604 .63180 .77513
.61841 .78586 .63203 .77494
.61864 .78568 .63225 .77476
.61887 .78550 .63248 .77458
.61909 .78532 .63271 .77439
.61932 .78514 .63293 .77421
.61955 .78496 .63316 .77402
.61978 .78478 .63338 .77384
.62001 .78460 .63361 .77366
.62024 .78442 .63383 .77347
.62046 .78424 .63406 .77329
.62069 .78405 .63428 .77310
.62092 .78387 .63451 .77292
.62115 .78369 .63473 .77273
.62138 .78351 .63496 .77255
.62160 .78333 .63518 .77236
.62183 .78315 .6.>540 .77218
.62206 .78297 .63563 .77199
.62229 .78279 .63585 .77181
.62251 .78261 .63608 .77162
.62274 .78243 .63630 .77144
.62297 .78225 .63653 .77125
.62320 .78206 .63675 .77107
.62342 .78188 .63698 .77088
.62365 .78170 .63720 .77070
.62388 .78152 .63742 .77051
.624ll .78134 .63765 .77033
.62433 .78II6 .63787 .77014
.62456 .78098 .63810 .76996
.62479 .78079 .63832 .76977
.62502 .78061 .63854 .76959
.62524 .78043 .63877 .76940
.62547 .78025 .63899 .76921
.62570 .78007 .63922 .76903
.62592 .77988 .63944 .76884
.62615 .77970 .63966 .76866
.62638 .77952 .63989 .76847
.62660 .77934 .640ll .76828
.62683 .77916 .64033 .76810
.62706 .77897 .64056 .76791
.62728 .77879 .64078 .76772
.62751 .77861 .64100 .76754
.62774 .77843 .64123 .76735
.62796 .77824 .64145 .76717
.62819 .77806 .64167 .76698
.62842 .77788 .64190 .76679
.62864 .77769 .64212 .76661
.62887 .77751 .64234 .76642
.62909 .77733 .64256 .76623
.62932 .77715 .64279 .76604
Cosin Sine
Cosin Sine
Co sin Sine
Co sin Sine
Co sin Sine
54
53
52
51
50
Sine Cosin
~~
347
:~mr
38
39
:61566 .78801
;-1
~~
56
55
54
53
52
51
50
49
48
47
46
45
44
43
42
41
40
39
38
37
36
35
34
33
32
31
30
29
28
27
26
25
24
23
22
21
20
19
18
17
16
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
0
I
i
I
!
40
41
-120
43
44
Sine Co sin Sine Cosin Sine Co sin Sine Co sin Sine Cosin
-0- .64279 .76604 .65606 .75471 .66913 .74314 .68200 .73135 .69466 .71934 60
1 .64301 .76586 .65628 .75452 .66935 .74295 .68221 .73ll6 .69487 .71914 59
2 .64323 .76567 .65650 .75433 .66956 .74276 .68242 .73096 .69508 .71894 58
3 .64346 .76548 .65672 .75414 .66978 .74256 .68264 .73076 .69529 .71873 57
4 .64368 .76530 .65694 .75395 .66999 .74237 .68285 .73056 .69549 .71853 56
5 .64390 .765ll .65716 .75375 .67021 .74217 .68306 .73036 .69570 .71833 55
6 .64412 .76492 .65738 .75356 .67043 .74198 .68327 .73016 .69591 .71813 54
7 .64435 .76473 .65759 .75337 .67064 .74178 .68349 .72996 .69612 .71792 "53
8 .64457 .76455 .65781 .75318 .67086 .74159 .68370 .72976 .69633 .71772 52
9 .64479 .76436 .65803 .75299 .67107 .74139 .68391 .72957 .69654 .71752 51
10 .64501 .76417 .65825 .75280 .67129 .74120 68412 .72937 .69675 .71732 60
ll .64524 .76398 .65847 .75261 .67151 .74100 .68434 .72917 .69696 .717ll 49
12 .64546 .76380 .65869 .75241 .67172 .74080 .68455 .72897 .69717 .71691 48
13 .64568 .76361 .65891 .75222 .67194 .74061 .68476 .72877 .69737 .71671 47
14 .64590 .76342 .65913 .75203 .67215 .74041 .68497 .72857 .69758 .71650 46
15 .64612 .76323 .65935 .75184 .67237 .74022 .68518 .72837 .69779 .71630 45
16 .64635 .76304 .65956 .75165 .67258 .74002 .68539 .72817 .69800 .71610 44
17 .64657 .76286 .65978 .75146 .67280 .73983 .68561 .72797 .69821 .71590 43
18 .64679 .76267 .66000 .75126 .67301 .73963 .68582 .72777 .69842 .71569 42
19 .64701 .76248 .66022 .75107 .67323 .73944 .68603 .72757 .69862 .71549 41
20 .64723 .76229 .66044 .75088 .67344 .73924 .68624 .72737 .69883 .71529 40
21 .64746 .76210 .66066 .75069 .67366 .73904 .68645 .72717 .69904 .71608 39
22 .64768 .76192 .66088 .75050 .67387 .73885 .68666 .72697 .69925 .71488 38
23 .64790 .76173 .66109 .76030 .67409 .73865 .68688 .72677 .69946 .71468 37
24 .64812 .76154 .66131 .750ll .67430 .73846 .68709 .72657 .69966 .71447 36
25 .64834 .76135 .66153 .74992 .67452 .73826 .68730 .72637 .69987 .71427 35
26 .64856 .76ll6 .66175 .74973 .67473 .73806 .68751 .72617 .70008 .71407 S4
27 .64878 .76097 66197 .74953 .67495 .73787 .68772 .72597 .70029 .71386 33
28 .64901 .76078 .66218 .74934 .67616 .73767 .68793 .72577 .70049 .71366 32
29 .64923 .76059 .66240 .74915 .67538 .73747 .68814 .72557 .70070 .71345 31
30 .64945 .76041 .66262 .74896 .67559 .73728 .68836 .72537 .70091 .71325 30
31 .64967 .76022 .66284 .74876 .67580 .73708 .68857 .72517 .70ll2 .71305 29
32 .64989 .76003 .66306 .74857 .67602 .73688 .68878 .72497 .70132 .71284 28
33 .650ll .75984 .66327 .74838 .67623 .73669 .68899 .72477 .70153 .71264 27
34 .65033 .75965 .66349 .74818 .67645 .73649 .68920 .72457 .70174 .71243 26
35 .65055 .75946 .66371 .74799 .67666 .73629 .68941 .72437 .70195 .71223 25
36 .65077 .75927 .66393 .74780 .67688 .73610
.72417 .70215 .71203 24
37 .65100 .75908 .66414 .74760 .67709 .73590 :~~~ .72397 .70236 .7ll82
38 .65122 .75889 .66436 .74741 .67730 .73570 .69004 .72377 .70257 .7ll62 ~I
39 .65144 .75870 .66458 .74722 .67752 .73551 .69025 .72367 .70277 .71141 21
40 .65166 .75851 .66480 .74703 .67773 .73531 .69046 .72337 .70298 .7ll21 20
41 .65188 .75832 .66501 .74683 .67795 .73511 .69067 .72317 .70319 .7ll00 19
42 .65210 .75813 .66523 .74664 .67816 .73491 .69088 .72297 .70339 .71080 18
43 .65232 .75794 .66545 .74644 .67837 .73472 .69109 .72277 .70360 .71059 17
44 .652.>4 .75775 .66566 .74625 .67859 .73452 .69130 .72267 .70381 .71039 16
45 .65276 .75756 .66588 .74606 .67880 .73432 .69151 .72236 .70401 .71019 15
46 .65298 .75738 .66610 .74586 .67901 .73413 .69172 .72216 .70422 .70998 14
47 .65320 .75719 .66632 .74567 .67923 .73393 .69-193 .72196 .70443 .70978 13
48 .65342 .7S700 .66653 .74548 .67944 .73373 .69214 .72176 .70463 .70957 12
49 .65364 .75680 .66675 .74528 .67965 .73353 .69235 .72156 .70484 .70937 ll
50 .65386 .75661 .66697 .74509 .67987 .73333 .69256 .72136 .70505 .70916 10
51 .65408 .75642 .66718 .74489 .68008 .73314 .69277 .72ll6 .70525 .70896 9
52 .65430 .75623 .66740 .74470 .68029 .73294 .69298 .72095 .70546 .70875 8
53 .65452 .75604 .66762 .74451 .68051 .73274 .69319 .72075 .70567 .70855 7
54 .65474 .75585 .66783 .74431 .68072 .73254 .69340 .72055 .70587 .70834 6
5
55 .65496 .75566 .66805 .74412 .68093 .73234 .69361 .72035 .70608 .70813
4
56 .65518 .75547 .66827 .74392 .68ll5 .73215 .69382 .72015 .70628 .70793
57 .65540 .75528 .66848 .74373 .68136 .73195 .69403 .71995 .70649 .70772 3
58 .65562 .75509 .66870 .74353 .68157 .73175 .69424 .71974 .70670 .70752 2
1
59 .65584 .75490 .66891 .74334 .68179 .73155 .69445 .71954 .70690 .70731
0
60 .656061.75471 .66913 .74314 .68200 .73135 .69466 .71934 .707ll .70711
I
349
Appendix 7
348
Cosin Sine
Cosin Sine
Cosin Sine
Cosin Sine
Cosin Sine
49
48
47
46
45
Tang I Cotang
-0- .00000 !Infinite.
1
2 :88&~~ 1 tm:~~
3 .00087 ll45.92
4 .00ll6 859.436
5 .00145 687.549
6 .00175 572.957
7 .00204 491.106
8 .00233. 429.718
9 .00262 381.971
10 .00291 343.774
11 .00320 312.521
12 .00349 286.478
13 .00378 264.441
14 .00407 245.552
15 .00436 229.182
16 .00465 214.858
17 .00495 202.219
18 .00524 190.9M
19 .00553 180.932
29 .00582 171.885
21 .00611 163.700
22 .00640 156.259
23 .00669 149.465
24 .00698 143.237
25 .00727 137.507
26 .00756 132.219
27 .00785 127.321
28 .00815 122.774
29 .00844 ll8.540
30 .00873 ll4.589
31 .00902 110.892
32 .00931 107.426
33 .00960 104.171
34 .00989 101.107
35 .01018 98.2179
36 .01047 95.4895
37' .01076 92.9085
38 .Oll05 90.4633
39 .Oll35 88.1436
40 .Oll64 85.9398
41 .Oll93 83.8435
42 .01222 81.8470
43 .01251 79.9434
44 .01280 78.1263
45 .01309 76.3900
46 .01338 74.7292
47 .01361 73.1390
48 .01396 71.6151
49 .01425 70.1533
50 .01455 68.7501
51 .01484 67.4019
52 .01513 66.1055
53 .01542 64.8580
54 .01571 63.6567
55 .01600 62.4992
56 .01629 61.3829
57 .01658 60.3058
58 .01687 59.2659
59 .01716 58.2612
60 .01746 57.2900
Cotang Tang
'
89
t
Tang
.01746
.01775
.01804
.01833
.01862
.01891
.01920
.01949
.01978
.02007
.02036
.02066
.02095
.02124
g'M53
.02 82
.022ll
.02240
.02269
.02298
.02328
.02357
.02386
.02415
.02444
.02473
.02502
.02531
.02560
.02589
.02619
.02648
.02677
.02706
.02735
.02764
.02793
.02822
.02851
.02881
.02910
.02939
.02968
.02997
.03026
.03055
.03084
.03114
.03143
.03172
.03201
.03230
.03259
.03288
.03317
.03346
.03376
.03405
.03434
.03463
.03492
Cotang
Cotang
57.2900
56.3506
55.4415
54.56!3
53.7086
52.8821
52.0807
51.3032
50.5485
49.8157
49.!039
48.4121
47.7395
47.0853
46.4489
45.8294
45.2261
44.6386
44.0661
43.5081
42.9641
42.4335
41.9158
41.4106
40.9174
40.4358
39.9655
39.5059
39.0568
38.6177
38.1885
37.7686
37.3579
36.9560
36.5627
36.1776
35.8006
35.4313
35.0695
34.7151
34.3678
34.0273
33.6935
33.3662
33.0452
32.7303
32.4213
32.ll81
31.8205
31.5284
3L2416
30.9599
30.6833
30.4ll6
30.1446
29.8823
29.6245
29.3711
29.1220
28.8771
28.6363
Tang
88
20
Tang Cotang
.03492 28.6363
.03521 28.3994
.03550 28.1664'
.03579 27.9372
.03609 27.7117
.03638 27.4899
.03667 27.2715
.03696 27.0566
.03725 26.8450
.03754 26.6367
.03783 26.4316
.03812 26.2296
.03842 26.0307
.03871 25.8348
.03900 25.6418
.03929 25.4517
.03958 25.2644
.03987 25.0798
.04016 24.8978
.04046 24.7185
.04075 24.5418
.04104 24.3675
.04133 24.1957
.04162 24.0263
.04!91 23.8593
.04220 23.6945
.04250 23.5321
.04279 23.3718
.04308 23.2137
.04337 23.0577
.04366 22.9038
.04395 22.7519
.04424 22.6020
.04454 22.4541
.04483 22.3081
.04512 22.1640
.04541 22.0217
.04570 21.8813
.04599 21.7426
.04628 21.6056
.04658 21.4704
.04687 21.3369
.04716 21.2049
.04745 21.0747
.04774 20.9460
.04803 20.8188
.04833 20.6932
.04862 20.5691
.04891 20.4465
.04920 20.3253
.04949 20.2056
.04978 20.0872
.05007 19.9702
.05037 19.8546
.05066 19.7403
.05095 19.6273
.05124 19.5156
.05153 19.4051
.05182 19.2959
.05212 19.1879
.05241 19.0811
----Cotang Tang
87
30
Tang
.0524T
.05270
.05299
.05328
.05357
.05387
.05416
.05445
.05474
.05503
.05533
.05562
.05591
.05620
.05649
.05678
.05708
.05737
.05766
.05795
.05824
.05854
.05883
.05912
.05941
.05970
.05999
.06029
.06058
.06087
.06ll6
.06145
.06175
.06204
.06233
.06262
.06291
.06321
.06350
.06379
.06408
.06437
.06467
.06496
.06525
.06554
.06584
.06613
.06642
.06671
.06700
.06730
.06759
.06788
.06817
.06847
06876
.06905
.06934
.06963
.06993
Cotang
Cotang
19.08ll
18.9755
18.87ll
18.7678
18.6656
18.5645
18.4645
18.3655
18.2677
18.1708
18.0750
17.9802
17.8863
17.7934
17.7015
17.6106
17.5206
17.4314
17.3432
17.2558
17.1693
17.0837
16.9990
16.9150
16.8319
16.7496
16.6681
16.6874
16.5075
16.4283
16.3499
16.2722
16.1952
16.ll90
16.0435
15.9687
16.8945
16.82ll
15.7483
15.6762
15.6048
15.5340
15.4638
15.3943
15.3254
15.2571
15.1893
15.1222
15.0557
14.9898
14.9244
14.8596
14.7954
14.7317
14.6685
14.6059
14.5438
14.4823
14.4212
14.3607
14.3007
Tang
seo
60
59
58
57
56
55
54
53
52
51
50
49
48
47
46
45
44
43
42
41
40
39
38
37
36
36
34
33
32
31
30
29
28,
27
26
26
24
23
22
21
20
19
18
17
16
15
14
13
12
ll
10
9
8
7
6
5
4
3
2
1
0
I
'
350
40
Tan11 Cotang
- 0 -.06993 .14.3007
I .07022 14.24ll
2 .07051 14.1821
3 .070~0 14.1235
4 .07110 I 14.0655
5 .07139 14.0079
6 .07168 13.9507
7 .07197 13.8940
8 .07227 13.8378
9 .07256 13.7821
10 .07285 13.7267
ll .07314 13.6719
12 .07344 13.6174
13 .07373 13.5634
14 .07402 13.5098
15 .07431 13.4566
16 .07461 13.4039
17 .07490 13.3515
18 .07519 13.2996
19 .07548 13.2480
20 .07578 13.1969
21 .07607 13.1461
22 .07636 13.0958
23 .07665 13.0458
24 .07695 12.9962
25 .07724 12.9469
26 .07753 12.8981
27 .07782 12.8496
28 .07812 12.8014
29 .07841 12.7536
30 .07870 12.7062
31 .07899 12.6591
32 .07929 12.6124
33 .07958 12.5660
34 .07987 12.5199
35 .08017 12.4742
36 .08046 12.4288
37 .08075 12.3838
38 .08104 12.3390
39 .08134 12.2946
40 .08163 12.2505
41 .08192 12.2067
42 .08221 12.1632
43 .08251 12.1201
44 .08280 12.0772
45 .08309 12.0346
46 .08339 11.9923
47 .08368 11.9504
48 .08397 11.9087
49 .08427 11.8673
50 .08456 11.8262
51 .08485 11.7853
52 .08514 11.7448
53 .08544 11.7045
54 .08573 11.6645
55 .08602 11.6248
56 .08632 11.58&3
57 .08661 11.5461
58 .08690 11.5072
59 .08720 11.4685
60 .08749 11.4301
-----Cotang Tang
-I
85
60
50
Tang
I Cotang
~~TI:430f
.08778 11.3919
.08807
.08837
.08866
.08895
.08925
.08954
.08983
.09013
.09042
.09071
.09101
.09130
.09159
.09189
.09218
.09247
.09277
.09306
.09335
.09365
.09394
.09423
.09453
.09482
.09511
.09541
.09570
.09600
.09629
.09658
.09688
.09717
.09746
.09776
.09805
.09834
.09864
.09893
.09923
.09952
.09981
.10011
.10040
.10069
.10099
.10128
.10158
.10187
.10216
.10246
.10275
.10305
.103.34
.10363
.10393
.10422
.10452
.10481
.10fi10
Cotang
11.3540
11.3163
11.2789
11.2417
11.2048
ll.1681
ll.l316
11.0954
11.0594
11.0237
10.9882
10.9529
10.9178
10.8829
10.8483
10.8139
10.7797
10.7457
10.7119
10.6783
10.6450
10.6118
10.5789
10.5462
10.5136
10.4813
10.4491
10.4172
10.3854
10.3538
10.3224
10.2913
10.2602
10.2294
10.1988
10.1683
10.1381
10.1080
10.0780
10.0483
10.0187
9.98931
9.96007
9.93101
9.902ll
9.87338
9.84482
9.81641
9.78817
9.76009
9.73217
9.70441
9.67680
9.64935
9.62205
9.59490
9.56791
9.54106
9.51436
Tang jcotang
-----84
'10
Tang Cotang
.10510 9.6T436
.10540 9.48781
;~ 9.46141
9.43515
.10628 ~:40904
.10657 9.38307
.10687 9.35724
.10716 9.33155
.10746 9.30599
.10775 9.28058
.10805 9.25530
.10834 9.23016
.10863 9.20516
.10893 9.18028
.10922 9.15554
.10952 9.13093
.10981 9.10646
.llOll 9.08211
.ll040 9.05789
.11070 9.03379
.11099 9.00983
.11128 8.98598
.11158 8.96227
.11187 8.93867
.11217 8.91520
.11246 8.89185
.11276 8.86862
.ll305 8.84551
.11335 8.82252
.11364 8.79964
.11394 8.776!)9
.11423 8.75425
.11452 8.73172
.11482 8.70931
.ll5ll 8.68701
.11541 8.66482
.11570 8.64275
.11600 8.62078
.11629 8.59893
.11659 8.57718
.11688 8.55555
.11718 8.53402
.11747 8.51259
.11777 8:49128
.11806 8.47007
.ll836 8.44896
.11865 8.42795
.11895 8.40705
.11924 8.38625
.11954 8.36555
.11983 8.34496
.12013 8.32446
.12042 8.30406
.12072 8.28376
.12101 8.26355
.12131 8.24345
I .12160 8.22~44
1.12190 8.20352
1.12219 8.1R370
1.12249 8.1(,398
'.12278 8.14435
'
Tana Cotan11
.12278 ""8.fi435
.12308 8.12481
.12338 8.10536
.12367 8.08600
.12397 8.06674
.12426 8.04756
.12456 8.02848
.12485 8.00948
.12515 7.99058
.12544 7.97176
.12574 7.95302
.12603 7.93438
.12633 7.91582
.12662 7.89734
.12692 7.87895
.12722 7.86064
.12751 7.84242
.12781 7.82428
.12810 7.80622
.12840 7.78825
.12869 7.77035
.12899 7.75254
.12929 7.73480
.12958 7.71715
.12988 7.69957
.13017 7.68208
.13047 7.66466
.13076 7.64732
.13106 7.63005
.13136 7.61287
.1316517.59575
.13195 7.57872
.13224 7.56176
.13254 7.54487
.13284 I 7.52806
.13313 7.5ll32
.13343 7.49465
.13372 7.47806
.13402 7.46154
.13432 7.44509
.13461 7.42871
.13491 7.41240
.13521 7.39616
.13550 7.37999
.13580 7.36389
.13609 7.34786
.13639 7.33190
.13669 7.31600
.13698 7.30018
.13.728 7.28442
.13758 7.26873
.13787 7.25310
.13817 7.23754
.1:1846 7.22204
.13876 7.20661
.13906 7.19125
.13935 7.17594
.13965 7.16071
.13995 7.14553
.14024 7.13042
.14054 7.11537
T.fng Cotang Ta.'.'!_
83
351
Appendix 7
'
60
59
58
57
56
55
54
53
52
51
5o
49
48
47
46
45
44
43
42
41
40
39
38
37
36
35
34
33
32
31
30
29
28
27
26
25
24
23
22
21
20
19
18
17
16
15
14
13
12
11
10
9
8
7
6
5
4
3
2
I
0
------ -82
go
'
Tan(l
( ) .14054
1 .14084
2 .14113
3 .14143
4 .14173
5 .14202
6 .14232
7 .14262
8 .14291
~ .14321
10 .14351
11 .14381
12 .14410
.14440
.14470
.14499
16 .14529
17 .14559
18 .14588
19 .1~18
20 .1 648
21 ~14678
2,;! .14707
23 .14737
24 .14767
25 ;14796
26 .14826
27 .14856
28 .14886
29 .. 14915
30 .14945
31 .14975
32 .15005
33 .15034
34 .15064
35 .15094
36 .15124
37 .15153
38 .15183
39 .15213
40 .15243
41 .15272
42 .15302
43 .15332
44 .15362
45 .15391
46 .15421
47 .15451
48 .15481
49 .155ll
50 .15540
51 .15570
52 .15600
53 .15630
54 .15660
55 .15689
56 .15719
57 .15749
58 .15779
59 .15809
oo
Tan11 Co tang
.15838 6.31375
.15868 6.30189
.15898 6.29007
.15928 6.27829
.15958 6.26655
.15988 6.25486
.16017 6.24321
.16047 6.23160
.16077 6.22003
.16107 6.20851
.16137 6.19703
.16167 6.i8559
.16196 6.17419
.16226 Ji.16283
.16256 6.15151
.16286 6.14023
.16316 6.12899
.16346 6.11779
.16376 6.10664
.16405 6.09552
.16435 6.08444
.16465 6.07340
.16495 6.06240
.16525 6.05143
.16555 6.04051
.16585 6.02962
.16615 6.01878
.16645 6.00797
.16674 5.99720
.16704 5.98646
.16734 5.97576
.16764 5.96510
.16794 5.95448
.16824 5.94390
.16854 5.93335
.16884 5.92283
.16914 5.91236
.16944 5.90191
.16974 5.89151
.17004 5.88114
.17033 5.87080
.17063 5.86051
.17093 5.85024
.17123 5.84001
.17153 5.82982 .
.17183 5.81966
.17213 5.80953
.17243 5.79944
.17273 5.78938
.17303 5.77936
.17333 5.76937
.17363 5.75941
.17393 5.74949
.17423 5.73960
.17453 5.72974
.17483 5.71992
.17513 5.71013
.17543 5.70037
.17573 5.69064
.17603 5.68094
~.15838 ~ .17633 5.67128
-----Cotang Tang
Cotang Tang
1:
Cotan(l
7.11537
7.10038
7.08546
7.07059
7.05579
7.04105
7.02637
7.01174
6.99718
6.98268
6.96823
6.95385
6.93952
6.92525
6.91104
6.89688
6.88278
6.86874
6.85475
6.84082
6.82694
6.81312
6.79936
6.78564
6.77199
6.75838
6.74483
6.73133
6.71789
6.70450
6.69116
6.67787
6.66463
6.65144
6.63831
6.62523
6.61219
6.59921
6.58627
6.57339
6.56055
6.54.777
6.53503
6.52234
6.50970
6.49710
6.48456
6.47206
6.45961
6.44720
6.43484
6.42253
6.41026
6.39804
6.38587
6.37374
6.36165
6.34961
6.33761
6.32566
81
80
10
Tan11
.17633
.17663
.17693
.17723
.17753
.17783
.17813
.17843
.17873
.17903
.17933
.17963
.17993
.18023
.18053
.18083
.18113
.18143
.18173
.18203
.18233
_.18263
.18293
.18323
.18353
.18384
.18414
.18444
.18474
.18504
.18534
.18564
.18594
.18624
.18654
.18684
.18714
.18745
.18775
.18805
.18835
.18865
.18895
.18925
. 18955
.18986
.19016
.19046
.19076
.19106
.19136
.19166
.19197
.19227
.19257
.19287
.19317
.19347
.19378
.19408
.19438
Cotang
11
'
Cotang
Tan(l Cotan11
5.67128 .19438 5.14455 60
5.66165 .19468 5.13658 59
5.65205 .19498 5.12862 58
5.64248 .19529 5.12069 57
5.63295 .19559 5.11279 56
5.62344 .19589 5.10490 55
5.61397 .19619 5.09704 54
5.60452 .19649 5.08921 53
5.59511 .19680 5.08139 52
5.58573 .19710 5.07360 51
6.57638 .19740 5.06584 50
5.56706 .19770 5.05809 49
5.55777 .19801 5.05037 48
5.54851 .19831 5.04267 47
5.53927 .19861 5.03499 46
5.53007 .19891 5.02734 45
5.52090 .19921 5.01971 44
5.51176 .19952 5.01210 43
5.50264 .19982 5.00451 42
5.49356 .20012 4.99695 41
6.48451 .20042 4.98940 40
5.47548 .20073 4.98188 39
5.46648 .20103 4.97438 38
5.45751 .20133 4.96690 37
5.44857 .20164 4.95945 36
5.43966 .20194 4.95201 35
5.43077 .20224 4.94460 34
5.42192 .20254 4.93721
5.41309 .20285 4.92984 33
32 I
5.40429 .20315 4.92249
~~ I
.20345
4.91516
5.39552
5.38677 .20376 4.90785 29
5.37805 .20406 4.90056 28
5.36936 .20436 4.89330 27
5.36070 .20466 4.88605 26
5.35206 .20497 4.87882 25 '
5.34345 .20527 4.87162 24'
5.33487 .20557 4.86444 23
5.32631 .20588 4.85727 22
5.31778 .20618 4.85013 21
5.30928 .20648 4.84300 20
5.30080 .20679 4.83590 19
5.29235 .20709 4.82882 18
5.28393 .20739 4.82175 17
5.27553 .20770 4.81471 16
5.26715 .20800 4.80769 15
5.25880 .20830 4.80068 14
5.25048 .20861 4.79370 13
5.24218 .20891 4.78673 12
5.23391 .20921 4.77978 11
5.22566 .20952 4.77286 10
9
5.21744 .20982 4.76595
8
5.20925 .21013 4.75906
7
5.20107 .21043 4.75219
6
5.19293 .21073 4.74534
5
5.18480 .21104 4.73851
4
5.17671 .21134 4.73170
3
5.16863 .21164 4.72490
2
5.16058 .21195 4.71813
1
5.15256 .21225 4.7ll37
5.14455 .21256 4.70463
- - - - - - - 0Cotang Tang
Tang
1
-----'190
'18
352
12
Tang
( ) .21256
1 .21286
2 .21316
3 .21347
4 .21377
5 .21408
6 .21438
7 .21469
8 .21499
9 .21529
10 .21560
11 .21590
12 .21621
13 .21651
14 .21682
15 .21712
16 .21743
17 .21773
18 .21804
19 .21834
20 .21864
21 .21895
22 .21925
23 .21956
24 .21986
25 .22017
26 .22047
27 .22078
28 .22108
29 .22139
30 .22169
31 .22200
32 .22231
33 .22261
34 .22292
35 .22322
36 .22353
37 .22383
38 .22414
39 .22444
40 .22475
41 .22505
42 .22536
43 .22567
44 .22597
45 .22628
46 .22658
47 .22689
48 .22719
49 .22750
5o .2278I
51 .22811
52 .22842
53 .22872
54 .22903
55 .22934
56 .22964
57 .22995
58 .23026
59 .2:>056
60 .23087
Co tang
I
13
Cotang
4.70463
4.69791
4.69121
4.68452
4.67786
4.67121
4.66458
4.65797
4.65138
4.64480
4.63825
4.63171
4.62518
4.61868
4.61219
4.60572
4.59927
4.69283
4.58641
4.58001
4.57363
4.66726
4.56091
4.55458
4.54826
4.54196
4.53568
4.52941
4.52316
4.51693
4.51071
4.50451
4.49832
4.49215
4.48600
4.47986
4.47374
4.46764
4.46155
4.45548
4.44942
4.44338
4.43735
4.43134
4.42534
4.41936
4.4I340
4.40745
4.40152
4.39560
4.38969
4.38381
4.37793
4.37207
4.36623
4.36040
4.35459
4.34879
4.34300
4.33723
4.33148
Tang
77
Tang
.23087
.23117
.23148
.23179
.23209
.23240
.23271
.23301
.23332
.23363
.23393
.23424
.23455
.23485
.23516
.23547
.23578
.23608
.23639
.23670
.23700
.23731
.23762
.23793
.23823
.23854
.23885
.23916
.23946
.23977
.24008
.24039
.24069
.24IOO
.2413I
.24162
.24193
.24223
.24254
.24285
.243I6
.24347
.24377
.24408
.24439
.24470
.24501
.24532
.24562
.24593
.24624
.24655
.24686
.247I7
.24747
.24778
.24809
.24840
.24871
.24902
.24933
Cotang
Cotang
4.33148
4.32573
4.32001
4.31430
4.30860
4.30291
4.29724
4.29159
4.28595
4.28032
4.27471
4.26911
4.26352
4.25795
4.25239
4.24685
4.24132
4.23580
4.23030
4.22481
4.21933
4.21387
4.20842
4.20298
4.19756
4.19215
4.18675
4.18137
4.17600
4.17064
4.16530
4.15997
4.15465
4.14934
4.14405
4.I3877
4.13350
4.12825
4.I2301
4.11778
4.11256
4.10736
4.I0216
4.09699
4.09182
4.08666
4.08I52
4.07639
4.07127
4.06616
4.06107
4.05599
4.05092
4.04586
4.0408I
4.03578
4.03076
4.02574
4.02074
4.01576
4.01078
Tang
76
14
Tang Cotang
.24933 4.01078
.24964 4.00582
.24995 4.00086
.25026 3.99592
.25056 3.99099
.25087 3.98607
.25118 3.98117
.25149 3.97627
.25180 3.97139
.25211 3.96651
.25242 3.96166
.25273 3.95680
.25304 3.95196
.25335 3.94713
.25366 3.94232
.25397 3.93751
.25428 3.93271
.25459 3.92793
.25490 3.92316
.25521 3.91839
.25552 3.91364
.25683 3.90890
.25614 3.90417
.25645 3.89945
.25676 3.89474
.25707 3.89004
.25738 3.88536
.25769 3.88068
.25800 3.87601
.25831 '3.87136
.25862 3.8667J
.25893 .3.86208
.26924 3.85745
.25955 3.85284
.25986 3.84824
.26017 3.84364
.26048 3.83906
.26079 3.83449
.26110 3.82992
.26141 3.82537
.26172 3.82083
.26203 3.81630
.26235 3.81177
.26266 3.80726
.26297 3.80276
.26328 3.79827
.26359 3.79378
.26390 3.78931
.26421 3.78485
.26452 3.78040
.26483 3.77595
.26515 3.77152
.26546 3.76709
.26577 3.76268
26608 3.75828
.26639 3.75388
.26670 3.74950
.26101 3.74512
.26733 3.74075
.26764 3.73640
.26795 3.73205
Cotan11 Tan11
7~0
Appendix 7
1~0
Tant
.26795
.26826
.26857
.26888
.26920
.26951
.26982
.27013
.27044
.27076
.27107
.27138
.27169
.27201
.27232
.27263
.27294
.27326
.27367
.27388
.27419
.27451
.27482
.27513
.27545
.27576
.27607
.27638
.27670
.27701
.27732
.27764
.27795
.27826
.27858
.27889
.27921
.27952
.27983
.28015
.28046
.28077
.28109
.28140
.281'/2
.28203
.28234
.28266
.28297
.28329
.28360
.28391
.28423
.28454
.28486
.28517
.28549
.28580
.28612
.28643
.28675
Cotan11
Cotang
3.73205
3.72771
3.72338
3.71907
3.71476
3.71046
3.70616
3.70188
3.69761
3.69335
3.68909
3.68485
3.68061
3.67638
3.67217
3.66796
3.66376
3.66957
3.65638
3.65121
3.64706
3.64289
3.63874
3.63461
3.63048
3.62636
3.62224
3.61814
3.61405
3.60996
3.60588
3.60181
3.59775
3.59370
3.58966
3.58562
3.58160
3.57768
3.57357
3.56957
3.56557
3.56159
3.55761
3.55364
3.54968
3.54573
3.54179
3.53785
3.53393
3.53001
3.52609
3.62219
3.51829
3.51441
3.51053
3.50666
3.50279
3.49894
3.49509
3.49I25
3.48741
Tan11
74
"""6(f
59
58
57
56
55
54
53
52
61
50
49
48
47
46
45
44
43
42
41
40
39
38
37
,
~
-0I
2
3
4
5
6
7
8
9
10
11
I2
13
14
I5
I6
I7
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
46
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
36
35
34
33
32
31
30
29
28
27
26
25
24
23
22
21
20
19
18
17
16
15
14
13
12
11
10
9
8
7
6
6
4
3
2
I
0
I
-
17
16
Tang
.28675
.28706
.28738
.28769
.28800
.28832
.28864
.28895
.28927
.28958
.28990
.29021
.29053
.29084
.29116
.29147
.29179
.29210
.29242
.29274
.29305
.;19337
.29368
.29400
.29432
.29463
.29495
.29526
.29558
.29590
.29621
.29653
.29685
.29716
.29748
.29780
.29811
.29843
.29875
.29906
.29938
.29970
.30001
.30033
.30065
.30097
.30128
.30160
.30192
.30224
.30255
.30187
.30319
.30351
.303ll2
.30414
.30446
.30478
.30509
.30541
.30573
Cotang
Cotang
3.4874f
3.48359
3.4?977
3.47596
3.472I6
3.46837
3.46458
3.46080
3.45703
3.45327
3.44951
3.44576
3.44202
3.43829
3.43456
3.43084
3.42713
3.42343
3.41973
3.41604
3.41236
3.40869
3.40502
3.40136
3.3977I
3.39406
3.39042
3.38679
3.38317
3.37955
3.37594
3.37234
3.36875
3.36516
3.36158
3.35800
3.35443
3.35087
3.34732
3.34377
3.34023
3.33670
3.33317
3.32965
3.32614
3.32264
3.31914
3.31565
3.31216
3.30868
3.30521
3.30174
3.29829
3.29483
3.29139
3.28795
3.28452
3.28109
3.27767
3.27426
3.27085
Tang
73
Tang
-:30573
.30605
.30637
.30669
.30700
.30732
.30764
.30796
.30828
.30860
.30891
.30923
.30955
.30987
.310I9
.31051
.31083
.31115
.31147
.31178
.31210
.31242
.31274
.31306
.3I338
.31370
.31402
.31434
.31466
.31498
.31530
.31562
.31594
.31626
.31658
.31690
.31722
.31754
.31786
.31818
.3I850
.31882
.31914
.31946
.31978
.32010
.32042
.32074
.32106
.32139
.32171
.32203
.32235
.32267
.32299
.32331
.32363
.32396
.32428
.32460
.32492
Cotaq
18"
353
19
72
7!0
Cotang
2.90421
2.90147
2.89873
2.89600
2.89327
2.89055
2.88783
2.88511
2.88240
2.87970
2.87700
2.87430
2.87161
2.86892
2.86624
2.86356
2.86089
2.85822
2.85555
2.85289
2.85023
2.84758
2.84494
2.84229
2.83965
2.83702
2.83439
2.83176
2.82914
2.82653
2.8239I
2.82130
2.81870
2.81610
2.81350
2.81091
2.80833
2.80574
2.80316
2.80059
2.79802
2.79545
2.79289
2.79033
2.78778
2.78523
2.78269
2.78014
2.77761
2.77507
2.77254
2.77002
2.76750
2.76498
2.76247
2.75996
2.75746
2.75496
2.75246
2.74997
2.74748
Tang
70"
60
59
58
57
56
55
54
53
52
51
50
49
48
47
46
45
44
43
42
41
40
39
38
37
36
35
34
33
32
31
30
29
28
27
26
25
24
23
22
21
20
19
18
17
I6
15
14
13
I2
11
10
9
8
7
6
5
4
3
2
1
0
I
!I
354
Appendix 7
----o
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
I
20
Tang
.36397
.36430
.36463
.36496
.36529
.36562
.36595
.36628
.36661
.36694
.36727
.36760
.36793
.36826
.36859
.36892
.36925
.36958
.36991
.37024
.37057
.37090
.37123
.37157
.37190
.37223
.37256
.37289
.37322
.37355
.37388
.37422
.37455
.37488
.37521
.37554
.37588
.37621
.37654
.37687
.37720
.37754
.37787
.37820
.37853
.37887
.37920
.37953
.37986
.38020
.38053
.38086
.38120
.38153
.38186
.38220
.38253
.38286
.38320
.38353
.38386
Cotang
21
69
68
22
Tang
.40403
.40436
.40470
.40504
.40538
.40572
.40606
.40640
.40674
.40707
.40741
.40775
.40809
.40843
.40877
.40911
.40945
.40979
.41013
.41047
.41081
.41115
.41149
.41183
.412-17
.41251
.41285
.41319
.4~353
.41387
.41421
.41455
.41490
.41524
.41558
.41592
.41626
.41660
.41694
.41728
.41763
.41797
:41831
.41865
.41899
.41933
.41968
.42002
.42036
.42070
.42105
.42139
.42173
.42207
.42242
.42276
.42310
.42345
.42379
.42413
Cotang
Co tang
2.47509
2.47302
2.47095
2.46888
2.46682
2.46476
2.46270
2.45065
2.45860
2.45655
2.45451
2.45246
2.45043
2.44839
2.44636
2.44433
2.44230
2.44027
2.43825
2.43623
2.43422
2.43220
2.43019
2.42819
2.42618
2.42418
2.42218
2.42019
2.41819
2.41620
2.41421
2.41223
2.41025
2.40827
2.40629
2.40432
2.40235
2.40038
2.39841
2.39645
2.39449
2.39253
2.39058
2.38863
2.38668
2.38473
2.38279
2.38084
2.37891
2,37697
2.37504
2.37311
2.37ll8
2.36925
2.36733
2.36541
2.36349
2.36158
2.35967
2.35776
2.35585
Tang
67
23
~
.42447
.42482
.42516
.42551
.42585
.42619
.42654
.42688
.42722
.42757
.42791
.42826
.42860
.42894
.42929
.42963
.42998
.43032
.43067
.43101
.43136
.43170
.43205
.43239
.43274
.43308
.43343
.43378
.43412
.43447
.43481
.43516
.43550
.43585
.43620
.43654
.43689
.43724
.43758
.43793
.43828
.43862
.43897
.43932
.43966
.44001
.44036
.44071
.44105
.44140
.44175
.44210
.44244
.44279
.44314
.44349
.44384
.44418
.44453
.44488
~
Cotang
Cotang
2.35585
2.35395
2.35205
2.35015
2.34825
2.34636
2.34447
2.34258
2.34069
2.33881
2.33693
2.33505
2.33317
2.33130
2.32943
2.32756
2.32570
2.32383
2.32197
2.32012
2.31826
2.31641
2.31456
2.31271
2.31086
2.30902
2.30718
2.30534
2.30351
2.30167
2.29984
2.29801
2.29619
2.29437
2.29254
2.29073
2.28891
2.28710
2.28528
2.28348
2.28167
2.27987
2.27806
2.27626
2.27447
2.27267
2.27088
2.26909
2.26730
2.26552
2.26374
2.26196
2.26018
2.25840
2.25663
2.25486
2.25309
2.25132
2.24956
2.24780
2.24604
Tang
66
----o1
60
59
58
57
56
55
54
53
52
51
50
49
48
47
46
45
44
43
42
41
40
39
38
37
36
35
34
33
32
31
30
29
28
27
26
25
24
23
22
21
20
19
18
17
16
15
14
13
12
ll
10
9
8
7
6
5
4
3
2
1
0
"'
'
..
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
45
47
48
49
50
51
52
53
54
55
56
57
58
59
60
I
24
Tang Cotang
.44523 2.24604
.44558 2.24428
.44593 2.24252
.44627 2.24077
.44662 2.23902
.44697 2.23727
.44732 2 .23553
.44767 2.23378
.44802 2.23204
.44837 2.23030
.44872 2.22857
.44907 2.22683
.44942 2.22510
.44977 2.22337
.45012 2.22164
.45047 2.21992
.45082 2.21819
.45117 2.21647
.45152 2.21475
.45187 2.21304
.45222 2.21132
.45257 2.20961
.45292 2.20790
.45327 2.20619
.45362 2.20449
.45397 2.20278
.45432 2.20108
.45467 2.19938
.45502 2.19769
.45538 2.19599
.45573 2.19430
.45608 2.19261
.45643 2.19092
.45678 2.18923
.45713 2.18755
.45748. 2.18587
.45784 2.18419
.45819 2.18251
.45854 2.18084
.4581l'9 2.17916
.45924 2.17749
.45960 2.17582
.45995 2.17416
.46030 2.17249
.46065 2.17083
.46101 2.16917
.46136 2.16751
.46171 2.16585
.45206 2.16420
.45242 2.16255
.45277 2.16090
.46312 2.15925
.46348 2.15760
.45383 2.15596
.46418 2.15432
.46454 2.15268
.45489 2.15104
.46525 2.14940
.46560 2.14777
.46595 2.14614
.46631 2.14451
-----Cotang
Tang
65
25
Tang
Cotang
A663f 2.14451
.46666 2.14288
.46702 2.14125
.46737 2.13963
.46772 2.13801
.46808 2.13639
.46843 2.13477
.46879 2.13316
.46914 2.13154
.46950 2.12993
.45985 2.12832
.47021 2.12671
.47056 2.12511
.47092 2.12350
.'47128 2.12190
.47163 2.12030
.47199 2.11871
.47234 2.11711
.47270 2.11552
.47305 2.11392
.47341 2.11233
.47377 2.11075
.47412 2.10916
.47448 2.10758
.47483 2.10600
.47519 2.10442
.47555 2.10284
.47590 2.10126
.47626 2.09969
.47662 2.09811
.47698 2.09654
.47733 2.09498
.47769 2.09341
.47805 2.09184
.47840 2.09028
.47876 2.08872
.47912 2.08716
.47948 2.08560
.47984 2.08405
.48019 2.08250
.48055 2.08094
.48091 2.07939
.48127 2.07785
.48163 2.07630
.48198 2.07476
.48234 2.07321
.48270 2.07167
.48306 2.07014
.48342 2.06860
.48378 2.06706
.48414 2.06553
.48450 2.06400
.48486 2.06247
.48521 2.06094
.48557 2.05942
.48593 2.05790
.48629 2.05637
.48665 2.05485
.48701 2.05333
.48737 2.05182
.48773 2.05030
----Cotang Tang
64
355
26
Tang
.48773
.48809
.48845
.48881
.48917
.48953
.48989
.49026
.49062
.49098
.49134
.49170
.49206
.49242
.49278
.49315
.49351
.49387
.49423
.49459
.49495
.49532
.49568
.49604
.49640
.49677
.49713
.49749
.49786
.49822
.49858
.49894
.49931
.49967
.50004
.50040
.60076
.50113
.50149
.50185
.50222
.50258
.50295
.50331
.50368
.50404
.50441
.50477
.50514
.50550
.50587
.50623
.50660
.50696
.50733
.50769
.50806
.50843
.50879
.50916
I .50953
Cotang
Cotang
2.05030
2.04879
2.04728
2.04577
2.04426
2.04276
2.04125
2.03975
2.03825
2.03675
2.03526
2.03376
2.03227
2.03078
2.02929
2.02780
2.02631
2.02483
2.02335
2.02187
2.02039
2.01891
2.01743
2.01596
2.01449
2.01302
2.01155
2.01008
2.00862
2.00715
2.00569
2.00423
2.00277
2.00131
1.99986
1.99841
1.99695
1.99550
1.99406
1.99261
1.99116
1.98972
1.98828
1.98684
1.98540
1.98396
1.98253
1.98110
1.97966
1.97823
1.97681
1.97538
1.97395
1.97253
1.97111
1.96969
1.96827
1.96685
1.96544
1.96402
1.96261
Tang
63
27
Tang Cotang
1.50953 1.96261
.50989 1.96120
.51026 1.95979
.51063 1.95838
.51099 1.95698
.51136 1.95557
.51173 1.95417
.51209 1.95277
.51246 1.95137
.51283 1.94997
.51319 1.94858
.51356 1.94718
.51393 1.94579
.51430 1.94440
.51467 1.94301
.51503 1.94162
.51540 1.94023
.51577 1.93885
.51614 1.93746
.51651 1.93608
.51688 1.93470
.51724 1.93332
.51761 1.93195
.51798 1.93057
.51835 1.92920
.51872 1.92782
.51909 1.92645
.51946 1.92508
.51983 1.92371
.52020 1.92235
.52057 1.92098
.52094 1.91962
.52131 1.91826
.52168 1.91690
.52205 1.91554
.52242 1.91418
.52279 1.91282
.52316 1.91147
.52353 1.91012
.52390 1.90876
.52427 1.90741
.52454 1.90607
.52501 1.90472
.52538 1.90337
.52575 1.90203
.52613 1.90069
.52650 1.89935
.52687 1.89801
.52724 1.89667
.52761 1.89533
.52798 1.89400
.52836 1.89266
.52873 1.89133
.52910 1.89000
.52947 1.88867
.52985 1.88734
.53022 1.88602
.53059 1.88469
.53096 1.88337
.53134 1.88205
.53171 1.88073
----Cotang Tang
62
.,.
60
59
58
57
56
55
54
53
52
51
50
4<)
48
47
46
45
44
43
42
41
40
39
38
37
36
35
34
33
32
31
30
29
28
27
26
25
24
23
22
21
20
19
18
17
16
16
14
13
12
11
10
9
8
7
6
5
4
3
2
1
0
I
356
~.
'
-0I
2
3
4
5
6
7
8
9
10
ll
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
'
28
29
Co tang
-:53i7l 1.88073
.53208 1.87941
.53246 1.87809
.53283 1.87677
.53320 1.87546
.53358 1.87415
.53395 1.87283
.53432 1.87152
.53470 1.87021
.53507 1.86891
.53545 1.86760
.53582 1.86630
.53620 1.86499
.53657 1.86369
.53694 1.86239
.53732 1.86109
.53769 1.85979
.53807 1.85850
.53844 1.85720
.53882 1.85591
.53920 1.85462
.53957 1.85333
.53995 1.85204
.54032 1.85075
.54070 1.84946
.54107 1.84818
.54145 1.84689
.54183 1.84561
.54220 1.84433
.54258 1.84305
.54296 1.84177
.54333 1.84049
.54371 1.83922
.54409 1.83794
.54446 1.83667
.54484 1.83540
.54522 1.83413
.54560 1.83286
.54597 1.83159
.54635 1.83033
.54673 1.82906
.547ll 1.82780
.54748 1.82654
.54786 1.82528
.54824 1.82402
.54862 1.82276
.54900 1.82150
.54938 1.82025
.54975 1.81899
.55013 1.81774
.55051 1.81649
.55089 1.81524
.55127 1.81399
.55165 1.81274
.55203 l.8ll50
.55241 1.81025
.55279 1.80901
.55317 1.80777
.55355 1.80653
.55393 1.80529
.55431 1.80405
----Cotang Tang
Co tang
1.80405
.55469 1.80281
.55507 1.80158
.55545 1.80034
.55583 l.799ll
.55621 1.79788
.55659 1.79665
.55697 1.79542
.55736 1.79419
.55774 1.79296
.55812 1.79174
.55850 1.79051
.55888 1.78929
.55926 1.78807
.55964 1.78685
.56003 1.78563
.56041 1.78441
.56079 1.78319
.56ll7 1.78198
.56156 1.78077
.56194 1.77955
.56232 1.77834
.56270 1.77713
.56309 1.77592
.56347 1.77471
.56385 1.77351
.56424 1.77230
.56462 1.77110
.56501 1.76990
.56539 1.76869
.56577 1.76749
.56616 1.76629
.56654 1.76510
.56693 1.76390
.56731 1.76271
.56769 1.76151
.56808 1.76032
.56846 1.75913
.56885 1.75794
.56923 1.75675
.56962 1.75556
.57000 1.75437
.57039 1.75319
.57078 1.75200
.57ll6 1.75082
.57155 1.74964
.57193 1.74846
.57232 1.74728
.57271 1.74610
.57309 1.74492
.57348 1.74375
.57386 1.74257
.57425 1.74140
.57464 1.74022
.57503 1.73905
.57541 1.73788
.57580 1.73671
.57619 1.73555
.57657 1.73438
.57696 1.73321
.57735 1.73205
----Cotan11 Tang
Tang
61
Tang
r-:ss.m-
60
30
Tang
Co tang
.57813
.57851
.57890
.57929
.57968
.58007
.58046
.58085
.58124
.58162
.58201
.58240
.58279
.58318
.58357
.58396
.58435
.58474
.58513
.58552
.58591
.58631
.586&0
.587 9
.58748
.58787
.58826
.58865
.58905
.58944
.58983
.59022
.59061
.59101
.59140
.59179
.59218
.59258
.59297
.59336
.59376
.59415
.59454
.59494
.59533
.59573
.59.12
.59651
.59691
.59730
.59770
.59809
.59849
.59888
.59928
.59967
.60007
.60046
.60086
r=-==Cotang
1.72973
1.72857
1.72741
1.72625
1.72509
1.72393
1.72278
1.72163
1.72047
1.71932
1.71817
1.71702
1.71588
1.71473
1.71358
1.71244
I.7ll29
1.71015
1.70901
1.70787
1.70673
1.70560
1.7041!6
1.70332
1.70219
J..7Ql06
1.69992
1.69879
1.69766
1.6965.)
1.69541
1.69428
1.69316
1.69203
1.69091
1.6'8979
1.68866
1.68754
1.68643
1.68531
1.68419
1.68308
1.68196
1.68085
1.67974
1.67863
1.67752
1.67641
1.67530
1.67419
1.67309
1.67198
1.67088
1.66978
1.66867
1.66757
1.66647
1.66538
1.66428
Tang
r:sms
T.ill05
.57774 1.73089
59
31
Tang Co tang
.60086 1:66428 60
.60126 1.66318 59
.60165 1.66209 58
.60205 1.66099 57
.60245 1.65990 56
.60284 1.65881 55
.60324 1.65772 54
.60364 1.65663 53
.60403 1.65554 52
.60443 1.65445 51
.60483 1.65337 50
.60522 1.65228 49
.60562 1.65120 48
.60602 l.650ll 47
.60642 1.64903 46
.60681 1.64795 45
.60721 1.64687 44
.60761 1.64579 43
.60801 1.64471 42
.60841 1.64363 41
.60881 1.64256 40
.60921 1.64148 39
.60960 1.64041 38
.61000 1.63934 37
.61040 1.63826 36
.61080 1.63719 35
.6ll20 1.63612 34
.6ll60 1.63505 33
.61200 1.63398 32
.61240 1.63292 31
.61280 1.63185 30
.61320 1.63079 29
.61360 1.62972 28
.61400 1.62866 27
.61440 1.62760 26
.61480 1.62654 25
.61520 1.62548 24
.61561 1.62442 23
.61601 1.62336 22 I
.61641 1.62230 21 I
.61681 1.62125 20
.61721 1.62019 19
.61761 1.61914 18
.61801 1.61808 17
.61842 1.61703 16
.61882 1.61598 15
.61922 1.61493 14
.61962 1.61388 l3
.62003 1.61283 12
.62043 l.6ll79 II
.62083 1.61074 10
9
.62124 1.60970
8
.62164 1.60865
7
.62204 1.60761
6
.62245 1.60657
5
.62285 1.60553
4
.62325 1.60449
3
.62366 1.60345
2
.62406 1.60241
1
.62446 1.60137
.62487 1.60033
0
----Cotang Tang
58
~~
.~
'k
"
'
t
I,
"'
~
o1
'2
3
4
5
6
7
8
9
I
1o
'11
~ 12
I 13
' 14
15
16
17
18
19
1
2P
...
357.
Appendix 7
.);&
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
.40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
65
56
57
58
59
60
'
32
'
33
57
56
34
35
Tang Cotang
1.70021 1.42815 60
.70064 1.42726 59
.70107 1.42638 58
.70151 1.42550 57
.70194 1.42462 56
.70238 1.42374 56
.70281 1.42286 54
.70325 1.42198 53
.70368 1.42110 52
.70412 1.4202.2 51
.70455 1.41934 50
.70499 1.41847 49
.70542 1.41759 48
.70586 1.41672 47
.70629 1.41584 46
.70673 1.41497 46
.70717 1.41409 44
.70760 1.41322 43
.70804 1.41235 42
.70848 1.41148 41
.70891 1.41061 40
.70935 1.40974 39
.70979 1.40887 38
.71023 1.40800 37
.71066 1.40714 36
.71110 1.40627 36
.71154 1.40540 34
.71198 1.40454 33
.71242 1.40367 32
1.4~92 .71285 1.40281 31
1.4 01 .71329 1.40196 30
1.45410 .71373 1.40109 29
1.46320 .71417 1.40022 28
1.45229 .71461 1.39936 27
1.45139 .71505 1.39850 26
1.45049 .71549 1.39764 25
1.44958 .71593 1.39679 24
1.44868 .71637 1.39593 23
1.44778 .71681 1.39507 22
1.44688 .71725 1.39421 21
1.44598 .71769 1.39336 20
1.44508 .71813 1.39250 19
1.44418 .71857 1.39166 18
1.44329 .71901 1.39079 17
1.44239 .71946 1.38994 16
1.44149 .71990 1.38909 15
1.44060 .72034 1.38824 14
1.43970 .72078 1.38738 13
1.43881 .72122 1.38663 12
1.43792 .72167 1.38568 11
1.43703 .72211 1.38484 10
1.43614 .72255 1.38399 9
1.43525 .72299 1.38314
8
1.43436 .72344 1.38229
7
1.43347 .72388 1.38145
6
1.43268 .72432 1.38060
6
4
1.43169 .72477 1.37976
1.43080 .72521 1.37891
3
2
1.42992 .72566 1.37807
1.42903 .72610 1.37722
1
~ 1.42816 .72654 1.37638 0
Cotaq Tana Cotan11 Tana
Tang
.67451
.67493
.67536
.67578
.67620
.67663
.67705
.67748
.67790
.67832
.67875
.67917
.67960
.68002
.68045
.68088
.68130
.68173
.68215
.68258
.68301
.68343
.68386
.68429
.68471
.68514
.68567
.68600
.68642
.68686
.68728
.68771
.68814
.68857
.68900
.68942
.68986
.69028
.69071
.69114
.69157
.69200
.69243
.69286
.69329
.69372
.69416
.69459
.69502
.69545
.69688
.69631
.69675
.69718
.69761
.69804
.69847
.69891
.69934
.69977
Cotang
1.48256
1.48163
1.48070
1.47977
1.47885
1.47792
1.47699
1.47607
1.47514
1.47422
1.47330
1.47238
1.47146
1.47053
1.46962
1.46870
1.46778
1.46686
1.46595
1.46503
1.46411
1.46320
1.46229
1.46137
1.46046
1.46955
1.45864
1.45773
1.45682
55
54
358
I
-01
2
3
4
5
6
7
8
9
10
ll
12
13
14
15
16
11
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
I
36
Tang
.72654
.72699
.72743
.72788
.72832
.72877
.72921
.72966
.73010
.73055
.73100
.73144
.73189
.73234
.73278
.73323
.73368
.73413
.73457
.73502
.73547
.73592
.73637
.73681
.73726
.73771
.73816
.73861
.73906
.73951
.73996
.74041
.74086
.74131
.74176
.74221
.74267
.74312
.74357
.74402
.74447
.74492
.74538
.74583
.74628
.74674
.74719
.74764
.74810
.74855
.74900
.74946
.74991
.75037
.75082
.75128
.75173
.75219
.75264
.75310
.75355
Co tang
Co tang
1.37638
1.37554
1.37470
1.37386
1.37302
1.37218
1.37134
1.37050
1.36967
1.36883
1.36800
1.36716
1.36633
1.36549
1.36466
1.36383
1.36300
1.36217
1.36134
1.36051
1.35968
1.35885
1.35802
1.35719
1.35637
1.35554
1.35472
1.35389
1.35307
1.35224
1.35142
1.35060
1.34978
1.34896
1.34814
1.34732
1.34650
1.34568
1.34487
1.34405
1.34323
1.34242
1.34160
1.34079
1.33998
1.33916
1.33835
1.33754
1.33673
1.33592
l.335ll
1.33430
1.33349
1.33268
1.33187
1.33107
1.33026
1.32946
1.32865
1.32785
1.32704
Tang
53
37
Tang
~
.75401
.75447
.75492
.75538
.75584
.75629
.75675
.75721
.75767
.75812
.75858
.75904
.75950
.75996
.76042
.76088
.76134
.76180
.76226
.76272
.76318
.76364
.76410
.76456
.76502
.76548
.76594
.76640
.76686
.76733
.76779
.76825
.76871
.76918
.76964
.77010
.77057
.77103
.77149
.77196
.77242
.77289
.77335
.77382
.77428
.77475
.77521
.77568
.77615
.77661
.77708
.77754
.77801
.77848
.77895
.77941
.77988
.78035
.78082
.78129
Cotang
Cotang
1.32704
1.32624
1.32544
1.32464
1.32384
1.32304
1.32224
1.32144
1.32064
1.31984
1.31904
1.31825
1.31745
1.31666
1.31586
1.31507
1.31427
1.31348
1.31269
1.3ll90
l.3lll0
1.31031
1.30952
1.30873
1.30795
1.30716
1.30637
1.30558
1.30480
1.30401
1.30323
1.30244
1.30166
1.30087
1.30009
1.29931
1.29853
1.29775
1.29696
1.29618
1.29541
1.29463
1.29385
1.29307
1.29229
1.29152
1.29074
1.28997
1.28919
1.28842
1.28764
1.28687
1.28610
1.28533
1.28456
1.28379
1.28302
1.28225
1.28148
1.28071
1.27994
Tang
52
38
Tang
r:7sU9
.78175
.78222
.78269
.78316
.78363
.78410
.78457
.78504
.78551
.78598
.78645
.78692
.78739
.78786
.78834
.78881
.78928
.78975
.79022
.79070
.79ll7
.79164
.79212
.79259
.79306
.79354
.79401
.79449
.79496
.79544
.79591
.79639
.79686
.79734
.79781
.79829
.79877
.79924
.79972
.80020
.80067
.80ll5
.80163
.802ll
.80258
.80306
.8\).354
.80402
.80450
.80498
.80546
.80594
.80642
.80690
.80738
.80786
.80834
.80882
.80930
.80978
Cotang
Cotang
1.27994
1.27917
1.27841
1.27764
1.27688
l.276ll
1.27535
1.27458
1.27382
1.27306
1.27230
1.27153
1.27077
1.27001
1.26925
1.26849
1.26774
1.26698
1.26622
1.26546
1.26471
1.26395
1.26319
1.26244
1.26169
1.26093
1.26018
1.25943
1.25867
1.25792
1.25717
1.25642
1.25567
1.25492
1.25417
1.25343
1.25268
1.25193
1.25ll8
1.25044
1.24969
1.24895
1.24820
1.24746
1.24672
1.24597
1.24523
1.24449
1.24375
1.24301
1.24227
1.24153
1.24079
1.24005
1.23931
1.23858
1.23784
1.23710
1.23637
1.23563
1.23490
Tang
51
359
Appendix 7
Cotang
1.23490
1.23416
1.23343
1.23270
1.23196
1.23123
1.23050
1.22977
1.22904
1.22831
1.22758
1.22685
1.22612
1.22539
1.22467
1.22394
1.22321
1.22249
1.22176
1.22104
1.22031
1.21959
1.21886
1.21814
1.21742
1.21670
1.21598
1.21526
1.21454
1.21382
1.21310
1.21238
l.2ll66
1.21094
1.21023
1.20951
1.20879
1.20808
1.20736
1.20665
1.20593
1.20522
1.20451
1.20379
1.20308
1.20237
1.20166
1.20095
1.20024
l.l9953
l.l9882
1.19811
1.19740
1.19669
1.19599
l.l9528
1.19457
1.19387
1.19316
1.19246
1.19175
Tang
50
60
59
58
57
56
55
54
53
52
51
50
49
48
47
46
45
44
43
42
41
40
39
38
37
36
35
34
33
32
31
30
29
28
27
26
25
24
23
22
21
20
19
18
17
16
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
0
I
'
-01
2
3
4
5
6
7
8
9
10
ll
12
13
14
15
16
17
18
19
20
. 21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
. 60
I
400
Tang
.83910
.83960
.84009
.84059
.84108
.84158
.84208
.84258
.84307
.84357
.84407
.84457
.84507
.84556
.84606
.84656
.84706
.84756
.84806
.84856
.84906
.84956
.85006
.85057
.85107
.85157
.85207
.85257
.85308
.85358
.85408
.85458
.85509
.85559
.85609
.85660
.85710
.85761
.858ll
.85862
.85912
.85963
.. 86014
.86064
.86ll5
.86166
.86216
.86267
.86318
.86368
.86419
.86470
.86521
.86572
.86623
.86674
.86725
.86776
.86827
.86878
.86929
Cotang
Cotang
1.19175
1.19105
1.19035
1.18964
1.18894
l.l8824
1.18754
1.18684
1.18614
I.l8544
1.18474
1.18404
I.l8334
I.l8264
I.l8194
I.l8125
I.l8055
l.l7986
I.l7916
l.l7846
1.17777
1.17708
1.17638
1.17569
1.17500
1.17430
1.17361
1.17292
1.17223
l.l7154
1.17085
1.17016
1.16947
l.l6878
116809
1.16741
l.l6672
1.16603
1.16535
1.16466
1.16398
1.16329
l.l6261
1.16192
1.16124
1.16056
1.15987
l.l5919
l.l5851
I.l5783
1.15715
1.15647
1.15579
l.l55ll
1.15443
l.l5375
1.15308
1.15240
1.15172
1.15104
1.15037
Tang
-----49
41
42
48
47
43
t--Tang
.93252
.93306
.93360
.93415
.93469
.93524
.93578
.93633
.93688
.93742
.93797
.93852
.93906
.93961
.94016
.94071
.94125
.94180
.94235
.94290
.94345
.94400
.94455
.94510
.94565
.94620
.94676
.94731
.94786
.94841
.94896
.94952
.95007
.95062
.95ll8
.95173
.95229
.95284
.95340
.95395
.95451
.95506
.95562
.95618
.95673
.95729
.95785
.95841
.95897
.95952
.96008
.96064
.96120
.96176
.96232
.96288
.96344
.96400
.96457
.96513
.96569
Cotang
Cotang
1.07237 60
1.07174 59
1.07ll2 58
1.07049 57
1.06987 56
1.06925 55
1.06862 54
1.06800 53
1.06738 52
1.06676 51
1.06613 50
1.06551 49
1.06489 48
1.06427 47
1.06365 46
1.06303 45
1.06241 44
1.06179 43
l.06ll7 42
1.06056 41
1.05994 40
1.05932 39
1.05870 38
1.05809 37
1.05747 36
1.05685 35
1.05624 34
1.05562 33
1.05501 32
1.05439 31
1.05378 30
1.05317 29
1.05255 28
1.05194 27
1.05133 26
1.05072 25
J.05010 24
1.04949 23
1.04888 22
1.04827 21
1.04766 20
1.04705 19
1.04644 18
1.04583 17
1.04522 16
1.04461 15
1.04401 14
1.04340 13
1.04279 12
1.04218 ll
1.04158 10
9
1.04097
8
1.04036
7
1.03976
6
1.03915
5
1.03855
4
1.03794
3
1.03734
2
1.03674
1
1.03613
1.03553
0
Tang
46
l~
~
360
I
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
I
44
Tang
.96569
.96625
.96681
.%738
.%794
.96850
.%907
.%963
.97020
.97076
.97133
.97189
.97246
.97302
.97359
.97416
.97472
.97529
.97586
.97643
Co tang
1.03553
1.03493
1.03433
1.03372
1.03312
1.03252
1.03192
1.03132
1.03072
1.03012
1.02952
1.02892
1.02832
1.02772
1.02713
1.02653
1.02593
1.02533
1.02474
1.02414
~97700 : 1.02355
Cotan11__l Tang
45
60
59
58
57
56
55
54
53
52
51
50
49
48
47
46
45
44
43
42
41
I~
I
44
Tang Cotang
20 -:97700 1.02355 4o
21 .97756 1.02295 39
22 .97813 1.02236 38
23 .97870 1.02176 37
24 .97927 1.02117 36
25 .97984 1.02057 35
26 .98041 1.01998 34
27 .98098 1.01939 33
28 .98155 1.01879 32
29 .98213 1.01820 31
30 .98270 1.01761 30
31 .98327 1.01702 29
32 .98384 1.01642 28
33 ,98441 1.01583 27
34 .98499 1.01524 26
35 .98556 1.01465 25
36 .98613 1.01406 24
37 .98671 1.01347 23
38 .98728 1.01288 22
39 .98786 1.01229 21
40 .98843 1.01170 20
-----Cotang Tang
1
45
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
I
44
'
'
-I
"
Tang
Cotang
.98843 1.01170 20
.98901 1.01112 19
.98958 1.01053 18
.99016 1.00994 17
.99073 1.00935 16
.99131 1.00876 15
.99189 1.00818 14
.99247 1.00759 13
.99304 1.00701 12
.99362 1.00642 11
.99420 1.00683 10
.99478 1.00525
9
.99536 1.00467 8
.99594 1.00408 7
.99652 1.00360 6
.99710 1.00291
5
.99768 1.00233 4
.99826 1.00175 3
.99884 1.00116 2
.99942 1.00058
1
1.00000 1.00000 0
------Cotang
Tang
45
'
--
361
:~
,.
lnd~x
DIRECTIONS
50
<0':<:J.."'
~"
~"'
'r>"<J.."\~
40
..,
""
Cj,
;!
30 ....
-~
\,.)
.::"'
Q
20
10
Aa lava1 259
Abbreviations for field notes, 8, 329
Abstracts of reports, 192.
Adamellite, 276, 277
Adit, 9f mine, 175
Adjustments, of alidade, 93, 94
of barometer (altimeter)' readings, 56
of Brunton compass, 34
of control networks, 125
of radial line plots, 166
of traverses, 54
Aerial photograph compilations, 154169
from controlled photographs, 163
cross sections from, 157, 169
marking a:l\.d transferring points for,
161
materials for, 160
J?lotting control for, 162
radial line method for, 159
transferring details for, 167-169
from uncontrolled photographs, 165
Aerial photograph projects, preparations
for, 154
Aerial photographs, compilations from,
154-169
computing scale at any elevation,
83-84
cross sections from, 157, 169
determining average scale of, 81
determining exact scales of, 81
distortion of, by relief, 76
distortion of, by tilt, 75, 165, 166
geologic mapping on, 84, 155
inking, 85, 167
kinds, 73
locating points on, by bearings 'and
pacing, 83
150
Age (of geologic time), 21}
Age relations, 3, see also Intrusive sequence, Stratigraphic sequence
Agglomerate, 255
Agglutinate, 255
Ahrens, L. H., 181
Alidade, see Peep-sight alidade, Telescopic alidade
Alluvial deposits, mapping, 245
Alterations of plutonic igneous. rocks,
291, 292, 293
of volcanic rocks, 257, 261, 265, 269
Alteration zones as map units, 27 4, 293
Altimeter, see Barometer
American Commission on Stratigraphic '
Nomenclature, 210
American Geological Institute, lSi, 332
American Society of Photogrammetry,
74, 154
Amphibolite, 300, 302
Amygdales, 260
Anderson, C. A., 266
Andesite, 254
Anorthosite, 276
Aplite, 275
Apparent dip, used in sections, 46, 362
Areal geologic mapping, see Mapping
363
364
Index
Index
365
366
Index
Index
Drawing detailed maps and sections,
201
Drawi~ methods, 199
Drawings, duplication of, 196
in field notes, 10
in reports, 197, 198
Drift, in mine workings, 175
Drilling, in detailed studies, 172
diammad, 173
Dunbar, C. 0., and John Rodgers, 209,
220.
..,
Dunite, 276
Duplication processes for illustrations,
.196
Eardley, A. J.,' 74
Elevation, difference in, see Difference
in elevation
Elevation of instrument, 106, 108
Entrenching tool, folding, 173
Epoch (of geologic time), 2U
Equipment, field
for aerial photograph compilations,
160
for compass traverse, 38
for control surveys with transit, 328
for general geologic work, 5, 327
for mapping on photographs, 80, 327
for mine mapping, 174
for plane table mapping, 328
for taping base line, U9, 121
for trimming fossils, 16
Error of closure, 54
Errors permissible in plane table mapping, 137
Errors of plotting on plane table, 133
Estimating positions on a map, 53, 145
Excavating and cleaning outcrops, 172
Extrusive rocks, see Volcanic rocks
Fabric, of metamorphic rocks, 283, 298,
305-307, 309-313
of plutonic igneous rocks, 282-285,
286,287-289
of sedimentary rocks, 215, 226
of volcanic rocks, 257-260, 262-265,
266
Face (breast) in mine workings, 175
367
364
Index
Index
365
366
Index
Index
Drawing detailed maps and sections,
201
Drawi~ methods, 199
Drawings, duplication of, 196
in field notes, 10
in reports, 197, 198
Drift, in mine workings, 175
Drilling, in detailed studies, 172
diamo~ad, 173
Dunbar, C. 0., and John Rodgers, 209,
220.
..,
Dunite, 276
Duplication processes for illustrations,
196
Eardley, A. J.,' 74
Elevation, difference in, see Difference
in elevation
Elevation of instrument, 106, 108
Entrenching tool, folding, 173
Epoch (of geologic time), 211
Equipment, field
for aerial photograph compilations,
160
for compass traverse, 38
for control surveys with transit, 328
for general geologic work, 5, 327
for mapping on photographs, 80, 327
for mine mapping, 174
fo~ plane table mapping, 328
for taping base line, 119, 121
for trimming fossils, 16
Error of closure, 54
Errors permissible in plane table mapping, 137
Errors of plotting on plane table, 133
Estimating positions on a map, 53, 145
Excavating and cleaning outcrops, 172
Extrusive rocks, see Volcanic rocks
Fabric, of metamorphic rocks, 283, 298,
305-307, 309-313
of plutonic igneous rocks, 282-285,
286,287-289
of sedimentary rocks, 215, 226
of volcanic rocks, 257-260, 262-265,
266
Face (breast) in mine workings, 175
367
368
Index
Gneiss, 300
Gneissose structure, 282, 306
Goldsmith, Richard, 300
Gradational contacts, 64
mapping, 64, 66, 280
new symbol for, 64, 334
in plutons, 278, 280, 284
Graded beds, 222, 225, 227, 230, 231,
234
in metamorphic rocks, 304
Gradienter screw, procedure with, 109
Grain sizes, cards for determining, 213
of metamorphic rocks, 299
of plutonic igneous rocks, 275
of sedimentary rocks, 213
Granite (potassic granite), 276
Granitic rocks, 275, 276, 279, 281, 282,
284, 288, 294
Granodiorite, 276, 277
Granofels, 300
Granulite, 300
Graphic symbols for rocks, 338
Graphs and curves, 198
Gravel, classification of, 213
Graywacke, 218
Great circle, in three-point locations,
144
Greenly, Edward, and Rowel Williams,
63
Grids, plotting on illustrations, 203
polyconic, 132
rectangular, constructing, 131
in transferring map points, 138
Groove casts, 229
Groups, 209
of plutonic igneous rocks, 27 4
Grout, F. F., 178
Hammer, geologist'&, 7
Hand lens, 7
Hand level, difference in elevation with,
27
Hewett, D. F., 240
H. I., in measuring difference in elevation, 101, 103, 104
Hodgson, R. A., 56, 123
Hoel, P. G., 178
Index
369
370
Index
Index
t
I
I'
'
'!.
37J
1
.,
372
Index
Photogeology, 74
Photogrammetry, 74
Photographs, see also Aerial photographs
described in notes, 8, 10
as illustrations in reports, 197
for plotting data at cliffs and cuts,
172
Photo indexes, 75
Phyllite, 300, 312, 314
Phyllonite, 300, 301, 315, 323
Pillow lava, 261, 262
Pipe amygdales, 258
Pitcher, W. S., 286
Pitcher, W. S., and R. C. Sinha, 178
Planar fabrics, in metamorphic rocks,
282-283, 298, 305, 309-313
in plutonic igneous rocks, 282-285,
286, 287-289
in sedimentary rocks, 215, 226
in volcanic rocks, 257-260, 262-265,
266
Plane table, 96
in aerial photograph projects, 156,
159
detailed mapping with, 171
geologic mapping with, 135-153
measuring stratigraphic sections with,
238
precision of plotting on, 133
setting up and orienting, 99
surveying steep cuts with, 172
traversing with, 146
tripods for, 96
Plane table maps, inking and completing, 151-152
Plane table projects, 135, 170
scales for, 135, 137, 154
Plane table she~ts, 97-98
selecting layout of, 137
Plane table traverses, 146
Planning field studies, 4
detailed examinations, 170
of metamorphic rocks, 296, 305
plane table projects, 136
of plutonic igneous rocks, 272
sampling, 178-182
of sedimentary rocks, 208
Index
Planning field .studies, triangulation
. surveys, !14
Plant fossils, ~
Plats, township, 82
Plotting, control networks, 131, 138, 162
strike and dip, 42
traverse data, 40, 44
Plunge of ]4n~ar features, 33
Plutonic igneous rocks,
age relations in, 272, 273, 279, 280282, 288,289-290,291,~94
alterations of, 291, 292, 293
contacts of, 277
hybrid zones of, 280
inclusi~ns in, 285
layers and schlieren in, 286
naming, 274
planar and linear fabrics in, 282
studies of, 272
units of, 273
Pocket stereoscope, 77
Polaris at elongation, observing, 123
Pqlyconic grids, 132
Polymetamorphic rocks, 301
Porcellanite, 217
Porosity of rocks, 215
Porphyrie1f, 275
Powers, M. C., 215
Precision of plotting on plane table, 133
Prentice, J. E., 226
Primary structures in metamorphic
rocks, 296, 302-304, 315
Prints from illustrations, 196
Printed patterns, 200
Protoclastic textures, 281, 282, 283,
291, 293
Protractor, for interconversion of true
and apparent dip, 362
Pseudoscopic image, 80
in delineating streams, 167
Pseudotachylite, 301
Pyroclastic cones, 265
Pyroclastic rocks, classification of, 255
deposits of, 264
grains in, 255, 256-257
Pyroxenite, 276
373
374
Index
Index
375
Stadia measurements,
D. Nl., 178
computations for, 104, 108
Shear fractures in plutons, 291-292
with gradienter screw, 109
Shells, fabric of in sediments, 226
by half intercept, 107
Sheltollii.T. S., 2t17
by level sight, 107-108
Shrock, R. R., 235
notebook for, 105, 106-107
sills, 268
principles of, 102
Siltstone, 218
steps in, 106
Skarn, 300
Stadia rod, uses of, 112, 151, 238
Sketch m;ps, of control networks, 118
Stages, 211
for pl~ne table mapping, 137, 141,
Stanley, G. M., 246
148
State plane coordinate systems, 130
by reconnaissance, 69
.,
Statistical methods in sampling, 178,
Sketches in notes, 10.:11, 309, 318
179, 180, 181, 182
Skialiths, 285
Steams, H. T., 251, 259
Slate, 300, 312
Steams, H. T., and G. A. Macdonald,
Slaty Cleavage, ~06, 311
253, 259
Slip (shear) cleavage, 306
Stebinger screw, 109
Slip (shear) folds, 311, 315, 324
Stepping method with alidade, 110
Slope direction in sediments, determinStereographic projection, uses of, 227,
ing, 225, 230
285, 293, 317
Slump structures in sediments, 225, 230
Stereo pairs, 77
Smith, H. T. U., 74
Stereoscope, mirror, 78
Smith, R. L., 2(36
pocket, 77
Soils, color ch;rts for, 247
Stereoscopic image, 73, 77
in detecting unconformities, 233, 252
obtaining with stereoscope, 77
in geologic mapping, 61, 246
obtaining without stereoscope, 79
hqrizons of, 246
Stokes, W. L., 229
profiles of, 246
Stapes, 175, 177
Soil Survey Staff (U.S. Department of
Strata, see Beds
Agriculture), 246
Stratigraphic horizons on key beds, 59,
Solar ephemeris, 124
265
Sorting in sedimentary rocks, 214, 219
Stratigraphic sections, measuring, 235
Spads, use in mine mapping, 177
by Brunton-tape method, 239
Specific gravity, used in mapping rocks,
by compass-pace surveys, 36, 236
277
corrections for slope and oblique
Specimens, see Samples
measures, 240, 241
Spilite, 255
by Jacob's staff, 236
Spiracles in lavas, 261
from maps or photographs, 236
Spot samples, 179
by plane table methods, 238
Stadia constant, 109
t
requirements for measurements, 235
Stadia intercept, 103
by transit-tape traverses, 238
Stadia interval factor, 109
Stratigraphic
sequence, determining,
Stadia mapping, 147, 151, 171
in metamorphic rocks, 303, 312-313,
instrument stations for, 139
314
by moving instrument around rod,
in sedimentary rocks, 225, 226, 228,
151
229, 230, 231, 232, 233
by traversing, 146
Shaw~
376
Index
Index
377
Varves, 234
Traverse, error of closure of, 54
Vegetation, used in mapping, 62, 84
general scheme of, 36
Veins, in metamorphic rocks, 314, 315,
planning, 38, 54
316, 322, 324
Trav~ing, with alidade and plane
in plutonic igneous rocks, 290-293,
table, 146; 238
294
with barometer ( altimeter) , 56
sampling, 181-183
on a base map, 54
Vertebrate fossils, 15
with G:ompass, 36, 54, 236
Vertical aerial photographs, 73
in mines, 177, 178
Vertical angles, see Difference in elevawith transit and tape, 133, 238
tion
by turning angles, 41
Vertical control for surveys, 113, 123
Trend and plu~ge of linear. features, 33
Volcanic complexes, 253
Triangle of error, in intersecting, 52
Volcanic intrusions
in radial li1;_1e plots, 165, 166-167
identifying, 267-269
in three-point (plane table) locations,
mapping, 253, 267
144-1'46
naming, 253
Triangles, computations from, 125
superficial, 270
Triangulation networks, ll6
Volcanic mudflow deposits, 266
Triangulation surveys, for aerial photoVolcanic necks, 253, 269
graph projects, 156
Volcanic rocks, alterations of, 252, 257,
with alidade and plane table, 132
261, 265
computations for, 125
cartographic
units of, 25~
.with peep-sight alidade, 90
flows of, 257-265
plans for, 114
initial dip of, 250
for stratigraphic measurements, 238
intrusive
bodies of, 267
with transit, 122
pyroclastic deposits of, 265
Tripods, plane table, 96
stratigraphic relations of, 250-252
.True north by observation on Polaris,
unconformities in, 252
123
Volcanic vents, indications of, 265, 267,
Tuff-breccia, 255
269
Tuff, classification of, 256, 257
Turner, F. J., 321
Wacke,216,217
Type locality (area), for rock unit, 210
Walking contacts, 60
Type section of rock unit, 210
Wallis, W. A., and H. V. Roberts, 178
Waters, A. C., 253, 261, 262
Ultramylonite, 301
Waters, A. C., and Konrad Krauskopf,
Unconformities, in columnar sections,
274
47
Weiss, L. E., and D. B. Mcintyre, 318
evidence for, 232, l\')2
Welded tuff, 257, 266
stratigraphic significance of, 232
Well logs, 242
in -surficial deposits, 245
Wentworth, C. K., and G. A. Macvalue as contacts, 59, 70
donald, 2S9
Wentworth, C. K., and Rowel Williams,
in volcanic rocks, 252
Underground mapping, see Mine map255
Wentworth scale of grain sizes, 213
ping
Wet weather, preparations for, 7
Units of weight and measure, 361
Williams, Rowel, 264, 266, 269
Urtite, 276
.I
378
Index
....,
'