A TE XT -B O O K

GEODETIC ASTRONOMY .

BY

J O H N F . H A Y F O RD , C E. .
,

A sso c z a te Me m b e r A m e ric a n So ciety

'

f s
o u E ng me ers
Expert Co mpute r and Geo desist U 5 . . Co a st a nd G eo deh c Survey .

F IR S T E D I TI O N .

F IRST T H O U SA N D .

N EW Y O RK

J O H N W IL E Y SO N S .

LO N D ON : C H A PM A N H A L L , L IM ITE D .

1 898 .
 Co pyrigh t , 1 89 8 ,

D
BY

JOh N F . H A YF O R .

7 9~ 77 é

ROBERT DRUMMO N D, PRINTER, NEW YO RK .

 PREFA C E .

( Ta be rea d by t fie s t u de n t as we ll as t h e t ea r/z e n )

purpose of this book is t furnish a text wh i h is

TH E o c

su fficiently short and easy t b mas t red by the studen t of o e e

civil ngine ring in a single c llege term but which shall give
e e o ,

him a sufficiently xact and xtensive knowledge of ge detic

e e o

astron my to s rve as a basis for practice in t hat line aft r

o e e

graduation Though the bo k has been prepared pri m arily

. o

f
or stud n t s the author has endeavor d to insert such sub
e ,
e ~

-matter tables and conveni nt formul as would make

j
ec t , ,
e a

it of value as a manual for t h engineer making as t ronomical e

observations I n order to make the book su fficien t ly short it

.

has b n necessary to mit all mathematical processes except

ee o

th s actually nec ssary for developing the working formul

e a

A d as t h o b j ect of the work is to t ach a certain limited

o e .

n e e

branch of astronomy ra t her than to teach mathematics the , ,

simpler and special means of d riving the working formul e a

have been chosen in ev ry case in which there was a chance

,
e

for ch ice instead of the more di ffi cult and general derivation

o ,

that would natu rally be ch s n by the math matician o e e .

Th i f the book is the fact that in the course of

e o cc a s o n
'

or

study prescri b ed for students of civil engin ering at Cornell e

Univ rsi t y but five hours per week for one term can be
e

d voted t the text -bo k w rk and lectures

e o astron my o o o n o .

Und r these conditions it is out f the q uestion to use o

standard work Even D oolittle s P ractical

e
’ '

Ch t
a u ve n e s .

iii
iv P RE F A CE .

A str nomy contains more mathematics than a student can be

o

exp ct d t master thor ughly in that period O f various

e e o o .

oth r text -books available none seem to fit the special condi

e

tions
I n the wording of the book it is taci t ly assumed that the
.

o b server is in the north rn hemisphere T make the word e . o

ing general w uld re q uir too man y circumlocutions

o e .

It is assum d that the stud nt has a knowl dg f least

e e e

s q uar s I f h wev r he has t such knowledge it will not

e o

e .
,
o e , no ,

d bar him from fol l wing nearly every part of the t xt exc pt
e o e e

7
10 3 d aling
—1 1 wi t h t
,
h tr eatm nt f transit tim b e e e o e o se r

va ti by leas t s q uares d
ons 54 57 giving the process , an 1 — 1
,

f combining the results for latitude with a z nith t lesc pe e e o

by t ha t m th d I f h reads car fully

o

e o . 83 8 5 stating e e 2 —2
,

the t chnical m aning f t h phrase pr babl rror the

e e o e o e e ,

statem nt f the unc rtain t y of a given observation in t erms

e o e

of t h pr bable rror or t h statement of the errors to be e x

e o e , e

p t
ec d from
e c rtain sou rces in such
e terms should convey t , o

him a d finite meaning e .

C nsiderabl spac has been devoted in the text t a dis :

o e e o

i
c u ss o n of the vari us sources of error in each kind of b
o o se r

vati treated Two separat considerations s em to t h

o n . e e e

auth r to j ustify t his O is that the special value of

o . ne

geodetic astro omy as a par t f the c urse of t i i g f an

n o o ra n n o

e ngin er dep nds largely upon t h fact that in s t udying it

e e e

h is brought face to face with the id a t hat ins t rumen t s are

e e

fallibl and that therefore their indications must be carefully

e,

scruti nized and interpr t d ; and that if t h best results are e e e

t obe s cured from them the s u rces of t h vari us minut e

e ,
o e o

err rs which c mbined c ns t itute t h err rs of observation

o o o e o

must be carefully studied The other consideration is tha t .

an observer s success in securing accura t e r sul t s with

’
e

moderate effor t depends to a considerable e x tent upon his

 PR E F A CE .

power to est mate rightly the l t i importance of the i

'

re a ve

various errors a ffecting his final resul t .

The accuracy of a man s thoughts as well as of his speech ’

, ,

when dealing with a given subject depends largely upon the

pr cis ion of his und rs t anding of t h sp cial vocabulary of
e e e e

tha t subject W ith t hat id a in V iew the finder list of

. e

d finitions given in 3 has b en pr pared The student

e 12 e e .

wh is not sure f t h exact meaning of a word may t u rn to o e

this list and find t h xact d finition quickly I n reading

o

so e e e .

definitions the context should also b r ad When a word is e e .

defined in the t ext it is prin t ed in italics .

Th effort has b n m ade t s lect h formul which have

e ee o e t e a

been found in prac t ic to lead t accura t e and rapid computa e o

ti ns They have been gathered at t h end of the v lume

o . e o

for convenient refer nce and adjacent to each formula will b e , e

found r f rences to t h corresponding portion f t h text

e e e o e

tha t f t hose who may use the b ook as a manual the list f
, so

or o

f rmul with th se references may serv as an index or finder

o a e e

for the text

I n t h five principal chapters the instrument has first been
.

described and the adj ustments given as well as dir ctions for
, , e

obser ing and an example f the record The derivation of

v ,
o .

t h formul
e t h computation etc follow I f the text -book
a , e , .
, .

work and t h practical w rk f t h observatory are carri d on

e o o e e

together during the same term one naturally wishes the

stud nts to b com familiar with the instrum n t s and th i r
e e e e e

manipulati n as soon as possibl In t hat cas it is

o m e . e re c o

m ended that the first portions only f c rtain chapters be o e

t ak n and t h
e lat r portions omitted temporarily The fol
e e .

lowing order may th n b used —

7 37 5 —6
3 8 3 9 e e : 1 2 , , 1 ,
— 1 ,

1 34 4 6 -1
77 8 7 3
, 5 to 1 middle of —1
73 —
7 6, 2 0 1 —2 0 ,
20 2 1 0, 2 2 ,

77 to t h end
8 5 — 064 8 — 6 — 8 8 — 2— 1
2 , 9 33 47 7 7 2 , , 1 1 , 1 2 2 ,
2 e .

D uring the preparation of this volume the text -books on

 PRE F A CE .

astronomy written b y Chauvenet D oolittl e and Loomis h ave

b een freely consulted as well as various reports of the Coast
,

and G eodetic S urvey of the Northern B ounda y S urvey of

, r ,

the U S Lake Survey and t h report of t h Mexic n

. .
, e e a

B oundary S urvey of 8 9 —9 3 A ppendix No 4 to

1 t h 2 . . 1 e

C oast S urvey Report for 8 8 which is wri t ten by A ssistant 1 0,

C A S chott and is used as a manual by the o fficers of th t a

survey has b e n extensively drawn upon as the b est p i

. .

, e ex os

tion of go d field methods known to the author Several

o

tables have b een taken from that source nota b ly t h table o f

.

, e

factors for the reduction of transit time observations given in

2
99 The. S uperintendent of the C oast S urvey has v ry e

kindly furnished certain data and photographs of i t ,

n s ru

ments .

J O H N F H A Y F O RD
p
. .

W A S H IN G IO N
'

, D . C .
, A ri l 23 , 1 89 8 .
 CO NTENTS .

CH AT
P ER

pp
I. IN TRO D U C TO RY O C O O O O O Q O O O O O O O O D O p O O o o o o o o o o o o o o o o

A a re n t

Th e
Pre c e s s io n d
l ll
an

P a n e t s Sa t e it e s S t a rs
, ,

D iu rn a l M o t io n

Ti m
Q
e .

I HT
u e s t io n s a n d
O
II C M PU TA TI ON R A SC E NS O N I
l
. OF G A ND

pl
Po s it io n o f Su n an d Panets
In te r o a t io n

Po s it io n o f Mo o n
Po s it io n o f S t a rs
A b e rra t io n . .

p
Mea n

C mp pp
Pro er M o t io n
t io n o fA
Q
o u ta a re n t

u e s t io n s a n d

III . TH E

D ire c t io n s fo r O b s e rv in g fo r
Re c o rd

D e riv a t io n o f

y
D isc u ss io n o f E rro rs
E c c e n t ric it . .

Ot h e r U se s o f t he

Q u e s t io n s a n d
V l il CON T E N TS .

CH AT
P ER

T I IT
A
P GE

p
IV . TH E A S RO NO M CA L TR A N S 96
D e s c ri t io n s o f Tra n s it s 96
A d j u st m en ts .
99

p
D ire c t io n s fo r O b se rv in g 1 03

Ch
C mp
Th e ro n o g ra h . 1 05

Re c o rd an d o u ta t io n . 1 07

Re d u c t io n to M e a n L in e 1 08

l
In c in a t io n C o rre c t io n

l
1 1 2

C o rre c t io n fo r D iu rn a A b e rra t io n 1 16

m C ll im i C
.

C mp
Az i u t h , o a t on ,
an d Ra t e o rre c t io n s . 1 1 7, 1 1 8, 1 1 9
Wi h
C mp
o u t a t io n t ou t Le a s t S qu a re s 1 20

Le a st S q u a re o u t a t io n 1 26

W e igh t s
l y
U n e qu a l 1 31
Au xi i a r O b se rv a t io n s 1 33
V a lu e o f t he 1 35
D is c u s s io n f E rro rs
ll
o 1 40

m pl
M isc e
Q
aneous . 1 44

Z N ITH T I N TI T I TU
u e s t io n s a n d E x a es 1 46

p p
V TH E TE L E S C O PE A N D TH E DE E RM N LA
l
E A O OF DE 1 50
Z
. .

dj m
D e s c ri t io n o f en it h Te e sc o e 1 51
A u st en ts 1 53
O b s rv i g
e n 1 56

F ml
D ire c t io n s fo r O b se rv i g n . 1 60

C mp
D e ri v a t io n o f o r u a 1 63
t a t io n 66
C mbi l
o u . 1

t io n o f Re su t s 68
m l
o n a 1

M ic ro e te r Va ue 1 74
D isc u s s io n f E rro rs 81
m i
o 1

m pl
Ot her Me thods f De te r in g L a t it u de 87
Q
o n 1

I UT H
u e s t io n s a n d E xa es . 1 94

p
VI A Z M

m 97
. 1

j m
D e sc ri t io n o f In s t ru en t 1 97

w
A d u st ts

m
en 2 00

D ire c t io n s fo r O b se rv i g it h D ire c t io n In s t ru t

m
n en . 2 01

Re c o rd — D ire c t io n In s t ru en t 2 05

Th e C irc l Re a d in g 2 06

lC
e

Le v o rre c t io n

m h ml
e 2 09

Az i ut Fo r u a 21 1

Cur v a t u re C o rre c t io n

l
21 3

mp
C o rre c t io n fo r D iu rn a A b e rra t io 6
m
n 21

p
Co u t a t io n — D ire c t io n In s t ru en t 21 8
Me t ho d o f Re e t it io n s 2 20
 C ON TE N TS .

CH AT
P ER

M ic ro m e t ric Me t h o d
D isc u s s io n f E rro rs
m
o

m pl
Ot h e r In st ru ts and Me t ho d s
Q
en .

I TU
u e s t io n s a n d E xa es .

l g ph pp
LO NG DE

Me t ho d— A
l g ph C mp
Te e ra ic a ra t u s a n d O b se rv a t io n s .

l g ph
Te ic M e t h o d — u t a t io n

l
e ra o

Te e ra ic M e t h o d — D is c u s s io n o f E rro rs a n d Pe rs o n a

y p
q u a t io nE
L o g it u d e b Tra n s o rt a t i o n o f C h ro n o m
p
n e t e rs

p Jp
O b se rv a t io n s u o n t h e M o o n
ll
. .

U
O b se rv a t io n s u o n u it e s S a t e it e s
’

I
r

VII I M SC E L L A N E O S
S u g g e s t io n s a b o u t O b s e rv i g
.

mp
n

S u gg e s t io n s a b o u t C o
.

u ti g

l
n

Pro b a b e E rro rs
Va ria t i o n o f L a t it u d e
.

m
E c o n o ic s o f O b se rv i g n

T TI IN I W O RK IN
IT FI NITI
NO A O N A ND PR C PA L G

L S OF DE O NS

F IG U RE S o o o o o o o o o o o o oooooooo oo o o o o o o 0 0 4 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
 G E O D E TIC A STRO N O M Y .

C H P TE R I
A .

I NTRO D UC TO RY .

1 . T H b ook is limit d to the treatment of astronomy as

IS e

applied t surveying or t what might be called geod tic

o , o e

astronomy O nly such matters are treated as

. p rtinent a re e

to this particular limited branch of the subj ect M r over . o e

the subject as thus limited is treated from the point f V i w

,

o e

of the engineer who wishes to obtain definite resul t s rather ,

than from t hat of a mathematician more interested in the

processes concerned than in th ir final outcom e e .

2 The b dies consid r d by t h

. o engine r in geodetic
e e e e

astronomy are the stars ; the S ; t h planets including the un e ,

Earth ; the Moon the Earth s satellit ; and to a very limited

,
’
e

ext nt some of t h satellites of t h other planets Th

e e e . e

engineer from his standpoint upon the surface of t h Ear t h e

sees thes di fferent bodi s moving about within the range of

e e

his vision — aided by a telescop if necessary Their apparent

‘

motions in the sky seen b y him are quite complicat d

, .

as

H is success in locating d orienting h imself upon the Earth

e .

an
2 G E OD E TIC A S TR ON O M Y . g 2 .

b y o b servations u ponthes heavenly bodies— for that is his e

particular purpose in observing them — depends first f all o

upon his having a clear and accurate conc ption of th ir e e

apparent motions and then upon his possessi n of and o

a b ility to use fficiently the instruments with which t h

, ,

obs rvations are mad M uch of the comple x ity in t h

, e

e e . e

apparent movem nts of these heavenly bodies is du e to the

e

fact that the obs rver sees them not from a fixed station in
e

space but from a standpoint upon one f the planets — the

,
o ,

Earth which is moving rapidly through space wi t h a motion

which is in its lf q uite c mplicated H e sees then in the
,

e o

t motion of each heavenly b ody

.

app a re n pon which he gazes u

no t only the actu l motion of that body b t also refl cted

a , u , e

back upon him to speak he sees the actual motion of t h

, so , e

seemingly s lid and imm vable earth upon which he stands

o o .

H is like a passenger upon a train at night who looks

e t

upon the many lights of a town H e se s the lights all

o u

. e

apparently in moti n I n one case the apparent moti n of theo . o

particular light may be entirely due to his own motion with the
train upon which he is riding t h light itself b eing at r st e

In another case the light may b e upon another moving train

, e .

and its apparent motion will then be due to the actual moti o n

of each of the trains I f the darkness is su fficient to conc al . e

the landscape he may be at a loss to determine what portions

,

of the apparent motions o f the lights are due to his own

change of position and what to the motions of the lights
themselves H e is then in the position of a man when he
.

first begins to study the apparent movements of the heavenly

b odies .

Let us first form concrete conceptions as to t h t l e a c ua

motion of each of the bodies u nder consideration including ,

the Earth itself We will then be in a position to understand

.

the pp t motions
a a ren .
g 4 . TH E E A R TH .

C onceive
3 . the S to be a very large self -luminous mass un

of matter F t h pr sent let it be supposed to be fixed in

. or e e

space A round this central S revolv ight plan ts

. un e e e ,

namely in rder of their distance from the S M rcury

,
o un , e ,

V nus Earth Mars j upit r Sa t urn Uranus and Nep t une

e , , , e , , , .

A ll th s plan t s m v e nearly but not exac t ly in t h same

e e o e , , e

plan passing through t h S

e The orbit path of any e un .
, o r ,

one of th m in its own rbital plane is very n arly a perf ct

e o e e

ellipse with f cus at the S and the v locity wi t h which

one o un ,
e

th eplanet m ves vari s at di fferent parts of its orbi t in such

o e

a way t hat the lin j oining t h planet and S describes e q ual e e un

areas in e q ual times This orbit and l w f velocity result . a o

from the fact that each planet is pursuing its path in obedi
enc to a single force gravity con t inually direc t d toward a
e , , e

fixed center the S ,

un .

4 The Ear t h may be taken as a repr sentative planet

. e .

It is the m st imp rtant f the planets fo p tp p o o or ou r re se n u r ose .

It mov s ab u t t h S in an elliptical orbit at a mean dis

e o e un

tanc fr m t h S i n r u nd numb rs of 9 8
e o e miles un , o e , 2 00 000 .
*

Though t h orbit is an ellipse its maj or and minor axes are

e ,

so n arly e q ual t hat if it were plotted to scale the unaide d

e

e
y e could not distinguish it from a circl The greatest dis e .

t ance of t h Earth from the S exc eds the l ast distance b y

e un e e

but little mor than The S is in that focus of t he un e

ellips to which the Earth is nearest during the wint r ( of t h

e e e

northern hemisphere ) The eccentricity of the ellipse and .

,

therefore the difference of t h t w axes is very sl wly e o , o

decreasing The plane of the Earth s orbit is not abs lutel y ’

o

fix d in direction in space I t changes with exceeding slow

.

e .

ness— slowly in fact that it is used as one of the

so , ,

astronomical reference planes M reover the p sition f t h . o , o o e

Se e Th e S o l ar Pa ra ll a xa n d it s Re l a te d C o n st a n t s , Ha r k n e ss , p . 1 40 .
4 C TR N M Y g 5G E OD E TI A S O O . .

elliptical orb it in the plane is slowly changing ; that is one ,

focus necessarily remains at t h S but the direction of the e un ,

major xis of the llipse gradually changes

a e .

Roughly speaking t h Earth mak s on e complete circuit , e e

of its orbit in the period f time which is rdinarily call d one o o e

year A t di fferent por t ions of the orbit it lin ar velocity

. s e

varies according to the law ( common to all the planets ) that

th line j oining it and the S describes e q ual areas in q ual
e un e

t imes Each portion f the nearly circular path b ing almost

. o e

perpendicular to the line j oining the Earth and S at that un

instant the linear velocity is nearly inversely pr porti nal to

, o o

the distance of the E arth from the S Evid ntly t h angular un . e e

velocity varies still mor largely than the lin ar since the e e ,

great st lin ar velocity com s at t h sam time as the least

e e e e e

distance from t h S and i e un ,

v ce v e rsa .

A t the same time that t h Ear t h as a whole is swinging e , ,

along in its orbit it is rotating uniformly about one of its w o n

diame t rs as an axis This rotation is so nearly uniform in

e
*

rate that it is assum d t b x actly u nif rm and is used to

.

e o e e o

furnish our standard of time Roughly sp aking the int rval . e , e

of time re q uired for one rotation of the Ear t h on its axis is

what is called one day The more ex act statement will be .

made later .

5 The axis of rotation of the Earth points at present

nearly to the star call d P olaris or North Star and makes an
.

e , ,

angle f about 665 with the plane f the Earth s orbit or

o
°
o
’

with t h p rpendicular t that plane The direction of

e e o .

this axis of rotati n is not fix d in space but changes j ust as

o e ,

m wh i k pl
wv
ly w p h w d lw y
Th e d ia e te r a bo u t ic h t he ro ta t o n ta es a ce is , ho e e r, n o t

ly
s t ric t fix e d it re s h ect to t he E a rt h — is not in t s t he
m m
, , o er o r s ,
a a

p p p h w v w ll
sa e d ia e t e r —b u t
,
v a rie s t h ro u g h a ra n ge o f
f a fe w eet o n on t h e su r

f ace o f t h e E a rt h Se e . 2 8 6 -7 . Fo r re s e n t ur o se s , o e e r, it i be
c o n si d e re d as fixe d .
g 5 . PR E C E S S ION AND N U TA T1 0N .
5

the axis of a rapidly spinning top is seen t wabbl abou t o e .

This change is q uite slow but xt nds thr ugh a large range ,
e e o

of m ti n It is c mp unded of two m t i ns call d p

o o . o o o o e re s ec

ti ly p
ve i and t t i B y virtu of t h mo t ion called
rec ess o n nu a on . e e

p i the axis f t h arth t ends t r main at an angl f

re c e ss o n o e e o e e o

ab u t 3 é wi t h t h p rpendicular t the plan f t h Ear t h s

o z ~
°
e e o e o e
’

orbi t ( usually kn wn as t h p l f M lip t i ) and to o e a ne o e ec c ,

r v lve c mpl tely around it d scribing a c n f t w napp s

e o o e ,
e o e o o e

wi t h an angl f about 4 7 (twic b e tween opposit

e o
°
e e

el men t s Th time r q uired to mak

e . such complete e e e one

rev l ution is at the pres nt rat ab u t 6

o y ars The
,
e e , o 2 000 e .
*

m ti n of the Earth s axis call d t t i is comp und d of

o o
’
e nu a on o e

several p ri dic moti ns the principal

e of which is such o , one o

as t caus t h axis to describe a con f which t h righ t sec

o e e e o e

ti n is an ellips and of which t h grea t st angl betwe n

o e ,
e e e e

opp si t l m nts is ab ut eight n seconds of arc and the

o e e e e o ee

leas t ab ut f t j o o ur een .
'

y l
pl p
Th e c h a n g e o f s e a so n s is c a u se db t he in c in a t io n o f t h e E a rt h ’
s a x is

ly w y m pp
t o t he b it
ane o f it s o r . At re s e n t the n o rt h e rn en d o f t he a x is is in
c lin e d d ire c t

w
f t he Su n t a bo ut De c 2 1 5t ; t h e Su n t he n
m
a a ro a . a e a rs

m p
t o b e f a rt h e r s o u t h t h a t a n yo t h her ti e d it is in t e r in t h e
mm
n a , an n o rt e rn

m ly w
a d su er i t he s t h e rn h e A t b o u t j u n e 2 o t h t h e r v rs e i h e re
mm
n n ou s . a e e

m p p w
is t u e n r e,
it is s u
a er i
,
r d i t r in t h e s n the rt he th n or e n an n e ou n

m p w ll J ly y
h is h e re O n a c c o u t f t h e r c ss io t h e i t e r o f t h e
e n o rt h e e e n n no rn

h e i s h re i o c c u r in j u e
.

pl
n d A u g u st b o u t 1 3 000 a s h c e
m ll
e n ,
u , a a e r en .

p l m m lp p ly
‘
1 A l l t h e v rio u s t as f t h e a e ts o d t h e ir s t e
io n it e s — theo n a n a

t h e v a ri b i it

pl pl y l
ecu i r aathe t ic a ro e rt ie s o f t h ir o rb it s
a o f t he e ,
a

e s e tc —

w
e s o f t h e o rb it s a d o f t h

m ply
rb it s in t h e b st i l

m
a n n cee o e an , . a r e a

p w
e c h a n ic s s h o t o b e d u e si to t h c t i n o f g ra v it a t io O r st a t i g
m m
n e a o n .
, n

lp
t he a t t e r f ro t h e c o v e rs e o in t f v ie g v e t h e se v io u s b o d ie s in
lm
n o , i n ar

w p w
t h e ir c t u aa o s it io n s d h a v i g t h e ir c t a a n ti s a t a g iv e n i s t a n t
n a u o on n ,

a ir o f t h e m

y ly p p p
a n d g iv e n t h e la w t h t g r v it a t io cts b t each it h a
l
a e a n a e e n n

ly p p p p
in t e sit in v e rse rt i t o t h e s q u re o f t h e ir d is t a n c e a rt a d
m
n r o o ona a a n

d ir c t o rt io n l t o t h e ro d u c t f t h e t wo a s se s t h e o sit io a n d
m m m
e ro a o , n

mp p p
o t io n o f e a c h o n e o f t h e t h e ir o rb it s e t c a t a n y o t h e r st a t e d t i e m a

m y
l m
, ,

p y ly
b e co u t e d f ro t he rin c i l e s o f c e e st ia l e c h a n ic s a l o n e Eve n t he .

re c e ss io n and n u t a t io n a re c a u se d b g ra v it a t io n a n d a re t h o ro u g h
6 G E OD E TI C A S TR O N OM Y . g 6 .

The Earth was taken as r presentative of the planets

6 . e .

Most or all of the various phenomena which have been indi

ca et d in the motion of the Earth are repeated in ach of the e

other planets Each has an orbital plane of its own which is .

slightly variable and does not in any case at present mak an e

angle of mor than about 7 with t h plan of the Earth s e

°
e e
’

orbit Each moves in that plane in an ellipse f which the

. o

eccentricity and positi n are slowly changing Each has a o .

rotation about its w axis —with which ho wev r the o n , , e , en

g i is
neer not concerned .

The M on is a satellite of the Earth revolving about it

o ,

u der t h action of gravity j ust as the Earth r volv s about

n e e e

th S e and in its orbit are again f und the same p culiarities

un ,
o e

as in the orbit of the Earth itsel f The orbit of the Moon is .

an llipse of variable eccentricity and position with the Earth

e ,

in one focus and lying in a variable plane making an angle of

,

ab ut 5 with the plan of the E arth s orbit The s veral

o
°
e
’
. e

variations mentioned are much great r in t h case of t h e e e

M on than of the Earth and t h motion moreov r is subj ect

o , e , e ,

to oth r perturbations Its mo t ion is theref r a mo st d iffi

e . o e

cult one to compute .

Each one of the other planets xcept M rcury and , e e

y p pl l lm h d v df m
p
d fo r b b v rin c i f c e e s t ia ic s e ri t he

l
a cc o u n te es o ec an e ro a o e

i-
ph ph d h vi g x
la w g v f it a t io n f h Th e s o la r re c e s s io n is d u e t o t h e a ct the

m
o ra . un t a t

mp
h
E a rt is b ut i f tte r in t he

l g
n o t a S e re , a s e ro , a n an e c e ss o a

g p ll y h f p
q t o ria io n s One t f t he t t ra c t io f t he M o o n d Su n
l x
e ua re . co onen o a n o an

w h h g g gl w h
a c t in h q u on dt is e u a t o ria e c e ss ten s c o n t in u a to s i t the o s it io n

pl p wh l d d w
o q
f t he e d ua t o r in o n e ire c t io n it out c an in it s a n e it t he e c lip

pl b l pl
t ic . Th e a c t io n h o f t he an e ts u on t he E a rt as a o e ten s to ra it

ll d p l y p pl y
f t he b h h g h
f it s it , t to c t he it a T is
l
o ut o ane o or o r ra er an e or ane .

p w h f ll d m p l y p
c h g an e is ca el t he an e ta r re c e ss io n . Th e un i-
so ar an d an e ta r

p d m wh h g l p
g h
re c e s s io n s t o et e r c o n s t it u t e at is te n ca si the re c e s si o n

md
o e .

wh h p d p
N u t a t io n is d f e rio ic t io n s ic du e e rio ic
fl m gh
a e u p o o a re t o re u ar

p d p
u c t u a t io n s f in the o rc e s ic ro uce re c e s s io n . Nu ta t i o n i t be
d b d
e sc ri e as the e rio ic a rt o f p re c e ss io n
8 G E O D E TI C A S TR O N O M Y . g 8 .

I f the stars had no motion r lativ to each other or to t h e e e

solar system as a whole the true direction of the line fr m , o

the Earth to any star would not vary fr m year to year

one o .

A s a matt r f observation however it is known that in

e o , ,

general the true direction f such a line d s chang although o oe e ,

the change is xceedingly slow in every cas This chang

e e . e

will be tr ated more in detail in a later chapter

The pp t motion of a y particular heavenly b ody as
e .

a a re n n

seen by an obs rver upon t h Earth is the compound result

e e

of t h mo t ion of t h Earth and f that body

e e o .

8 I n the case of a star the obj ct observed is for most

. ,
e

pu rpos s at what may b consider d an infinite distance

e e e

The line j ining the o b server and star preserv s th ref r a

.

o e , e o e ,

sensibly constant direction in spite o f the motion through

space of the observer u pon the Earth The apparent motion .

of the s t ar is caus d by the rotation of the Earth about its

e

axis and the change in the dir ction of that axis in space e .

The rotation of t h Earth causes the line of sight to a star to

e

seem to describe at a uniform rate a right circular con of e,

which the axis is the line joining the observer with a point in
the sky t an infinite dista nce in the axis of the Earth pro
a

du ce d I n other words the axis of the cone is a line fro m

o b server parallel to the axis of the Earth S uch a line
. ,

th e .

f or any point in the northern hemisph re pierces the sky i n

,

a point not far from the North Star P olaris The angl
,

, . e

be t ween any element of the cone and its axis is the angle
between the l ne j m g observer t star and the axis of the
i o in o ,

Earth This angle is called the p l d i t

. of the star o ar s a n ce ,

n or tl p l
t i t
o ar if measured from the north end of the
’
a s a n ce

Earth s ax s S long as these two lines are fixed in direction

’
i . o

in space the line of sight to the star continues to describe

,

the same right circular cone once for every turn which the
Earth makes on its axis F e xample the line of sight to . or ,
§ 8 . A PP A R E N T D I U R N A L M O TIO N OF TH E S TA RS .
9

P laris makes an angl of about f with the axis f t h Earth

o e I
)
o e ,

and describ s a corr sponding right circular cone O it may e r

be said t hat P laris seems to d scrib a circle in t h sky f

e .

o e e e o

which the radius subt nds an angle at the y of f With e e e I

)

a go d t lesc p P laris may be foll wed compl tely around

.

o e o e o o e

the circle ll f which would be above the horizon f any

,
a
'

o or

p int in t h Unit d States With t h naked y only that

o e e . e e e

p rtion of t h appar nt motion which occurs during t h

o e e e

h urs f darkn ss could b bserved F an obs rver t

o o e e o . or e a

Ithaca in latitude ,
th c ne for a star having a n rth e o .
o

polar distanc l ss than is ntir ly abov t h horiz n

e e e e e e o

G iv n one V i w of the star and an id a of t h position of t h

.

e e e e e

vanishing point of the Earth s axis i the sky an observ r ’

n , e

is able t trac out t h whole appar nt path f the star F

o e e e o . or

Ithaca a star f nor t h polar distance f 4 5 has its con

,
o o 2
°
e

tang nt to t h horizon ; and if greater than that value a par t

e e ,

of the cone m ust be bel w the horizon and t h star is o e n eces

I f the north polar distance

,

ily invisible on that portion

sa r .

is the cone b comes a plane Stars still farth r sou t h e . e

describ a right circular cone abou t the southern portion of

e

the E arth s axis pr duced t h angle of the cone b ing t h

’
o , e e e

sou t h polar distance f t h star o e

I n v ry cas t h diurnal rotatio of the Earth causes the

.

e e e e n

line f sight to a star to describe a right circular cone B t

o . u ,

as has already been stated the direction of the Earth s ’

axis is continually changing slowly and h nce the north p lar , e o

distance or angl betw en the Earth s axis and the line j oin e e
’

ing the observer and star is continually changing The cone .

of r volution therefore slowly changes from day to day

e .

If the obj ct observed is not a star at a practically infinite

e

distance but a planet the Mo n or t h S at a finite dis

, ,
o ,
e un ,

tance the lin j ining observer to obj ect describes a surface

e o

any sm ll portion of which may b e considered to b e a portion

,

a
10 G E OD E TIC A S TH o N o M Y . 10 .

of the sur f ace of a right circular cone B t the north polar u

distance of the o b j ect now continually changes not only

.

, on

acc unt of the change in the direction f the Earth s axis but
o o
’

still m re larg ly on account of the chang in the tru d i

o e e e re c

tion of the line j ining bserver and object — t w points o o , o

which are at a finite distance from each other and both in

motion This last cause also mak s t h t at which the
. e e ra e

su rface is describ d variabl e e .

9 The two principal refer nce planes of astronomy are e

the plane of the q uat r and the plan of the Earth s r b it

.

’
e o e o ,

or as it is generally called t h p l
, f lip t i The p l , e a ne o ec c . a ne

of t h q t
e e is a planeuapassing
orthrough the center of the
Earth and p rpendicular to it axis f rotation Neither of
e s o .

t hese t w planes from wha t has already been writ t en are

o , ,

fixed in space nor fixed relatively to ach other Th ir

, e . e

changes of position are how ver v ry slow e e

1 0 To avoid the necessity of using cum b ersom expres

.
, ,

. e

sions and circumlocutions it is c nv ni nt to make use f the , o e e o

l t i l pn
c e es aas an arbitrary c nception The l t i l p/
s e re o . c e es a s t e re

is a spher f infini t radius the eye of the bserv r being

e o e , o e

supp sed to be at its center A y cele stial obj ect is

o id . n co ns

er d t be project d along the line of sight to the surface of

e o e

this sphere and is r ferred to as occupying that position upon

e

the sphere Then for convenience one may speak of arcs

.
,

angles and triangles upon the celestial sph re instead of using

,
e

the complicated expressions necessary in speaking always of

the actual lines and planes which are under consideration .

The sphere is assumed to be of infinite radius so that lines

which are parallel and at a finite distance apart will int rsect e

the sphere in the same point or at least what is sensibly one ,

point since two points at a finite distance apart must appear

,

as one when seen from an infi ite distance S also parallel n . o

planes which are at a finite distance apart intersect the

 12 . D E FIN I TION S . 1 I

cel stial sphere in the same arc F example the axis of

e . or ,

the Earth and a line parall l to it thr ugh t h eye of the e o e

observer both intersect the c lestial sphere in t h same pair e e

of points called the p l f t i q t more bri fly the o es o te e ua o r, or e

p l
o es , — north and south r spectively A ls the plane of the e . o

e q uator and a plane parall l to it t hrough the eye of the

,
'

observer intersect the cel stial sph r in t h same great circle

,
e e e e

which is called t h q t f t n l t i l p/ or more f e e ua or o e ce es a s z e re , re

quently simply t / q t ze e ua or .

11 The q. t the l i
p t i h -ei lua and the l
o r,i ec c, o u r c rc es , z o r z on

are all gr at circles of the c l stial sph r f rmed by the

e e e e e o

intersection of various plan s wi t h t hat spher e e .

Th lip t i is t h inters cti n of the plan

e ec c f the ecliptice e o e o ,

or in oth r words the plane f the Earth s orbit with the

,
e , o
’

c les t ial spher

e Th S therefore is always s en pro
e . e un , , e

jec t d es me point
o n of the ecliptic
o .

A h u n- i l is the inters ction

o r c rc ef a plane passing e o

through the Earth s axis with t h cel stial sph r A ll hour ’

e e e e .

circles t h n great circles passing thr ugh t h

a re e p les o e o .

The l i is the intersecti n with the celestial sph re f

z or z on o e o

a plane passed through the eye of the observ r perp ndicular e e

to t h plumb -line or lin of action of gravity at the obs rv r

e ,
e , e e .

A ll horizontal lines at a giv n point the Earth s surface ’

e on

pi rce the celestial sphere in t h horizon of that point e

I n each of these cas s it is vident that the great circle on

e .

e e

the c lestial sphere would not be changed if t h intersecting

e e

plan were moved parallel to itself a finite distance — for

e ,

instan ce to pass through any oth r point in or upon the su r

,
e

fac of the Earth F example the horizon may be

e . id or , co n s

er d to b t h inters ction with t h c lestial sphere of a plane

e e e e e e

passing through the t f t l E t l and perpendicular to c en er o ze ar t

the observer s gravity line instead of that given above

’

, .

1 2 The angle between a line joining the center of t h

. e
12 G E O D E TI C A S TR ON O M Y .
g 12 .

Earth to a star ( or other c les t ial obj ct ) and the plan of the e e

I
e

e
q uator is called t h d li t i f that obj ecte ect is meas
na on o .

u re d upon t h celes t ial sph re by t hat portion

e f the h ur e o o

circl passing thr ugh the star which is be t w n the star and
e o ee

the q uator Th declination is consider d posi t iv when

e e e

m asured nor t h from t h q uator I t f llows from the

. e

e e e . o

definiti n f polar distance ( giv n in 8 ) that t h declina t i n

o o e e o

and p lar distance are compl ments of each oth r

o e e .

Th q uator
e ande the ecliptic intersec t ach oth r at an e e

angl of ab ut 3
e o Th ir t w p ints of intersection
2
°
e o o on

th e celestial spher are called the q i e That one at e u n o x es .

which t h S is found in t h spring is call d the

e un l e e v ern a

e q i
u n ox and tha
,
t at which it is found i n t h fall the t m l e a u u na

e q i
u n ox A b
. th t h qsuator and eocliptic m ve
e slowly
e in o

spac the q uin ctial points slowly shift in position upon the
e e o

celestial sph re e .

The mg/ u i of a star or th r c lestial bj ct is

r a sce n s o n , o e e o e ,

the angle m asur d along the equat r betwe n t h t w

,
e e o , e e o

hour -circles which pass through the star and t h verna l e q ui e

nox r sp ctiv ly I n ther words t h right asc ns on the

e e e . o ,
e e i IS

angl b tw en t w planes one passing through the Earth s

e e e o ,
’

axis and the star and the oth r through th E arth s axis and e e
’

t h v rnal e q uinox
e e I t is reckon d in degr s fr m t 3 6
. e ee o 0 o 0,

i n the directi n that would appear counter-clock wis i f

o e o ne

look d toward the e q uat r fr m t h north pol -fr m w s t

e o o e e , o e

to east Right ascensions still m r frequ ntly expressed

. a re o e e

in tim 4 h urs b i ng q uivalen t to 3 6 degrees

e , 2 o e e 0 .

The it n is the p int in which the action -line of gravity

z en o

produc d upward intersects the l t i l ph

e The oppo ce es a s e re .

site poin t on the celestial sphere is called the d i

‘

na r .

The intersection with the celestial sphere of a plane passed

through its center the zenith and the pole is called t h , , e

m erid i and the plane itself is called the m idi p l

an , er an a n e.
 I3 . D E F IN I TION S . 1 3

The intersection of the meridian plan with t h plane of the e e

horizon is called t h m idi li It c nnects t h north

e er an ne . o e

and south points of the horizon The int rs ction with t h . e e e

celestial spher of a plane through the zeni t h p rpendicular

e e

to the m ridian plan is called t h p im

e ti l
e The east e r e v er ca .

and west points of the horizon i n t h prim vertical a re e e .

1 3 The angle measur d along the equator between the

.
,
e ,

meridian and the hour-circl passing through a star ( or other e

cel stial obj ect ) is t h l

e g l of the stare I n th r words
z o u r -a n e . o e ,

the hou r -angle is the angle be t w n the meridian plane and a ee

plane passing through t h Earth s ax s and the star H ur e

’
i . o

angles reckoned like righ t ascensions either in degre s

a re , e ,

minut s and s conds of or in hou rs minut s and seconds

e a re e

I
e , , ,

o f tim n thise b . ok hour angles will be m asured f 8

- o e or 1 0
°

each way from the upper branch of t h m ridian and will e e

always be considered positive .

The student sh uld distinguish carefully between an hou r

o

angle and a right ascension Each is an angle b tween two . e

plan s I n each cas one of the t w p lanes is defined by the

e . e o

Earth s axis and the star and th refore chang s direction but
’

,
e e

slowly in space The second plane concerned in the case of

.

a right asc nsion is d fined by the Earth s axis and the v rnal
e e
’

e q uin x This plane changes its direc t i on very sl wly S

o . o . o

the right ascension of a star is an angl which is l wly chang e s o

ing t a rate f l ss than one min ute f p year for

,
— a o e o a re er

nearly all the stars The second plane conc rn d in the . e e

m asurement of an f
e g l is the plane of t h m ridian
l o u r -a n e e e

This acc mpanies t h E ar t h in its diurnal ro t ation H ence

.

o e .

t h hour -angle of a celestial bj ct vari s rapidly — 3 6 for °

e o e e , 0

each rotation f the Earth on its axis Th right ascension

o . e

and declination are spherical -ordinates locating a celestial co

obj c t with reference to the hour-circle through the vernal

e

e
q u inox and the e q uator The hour -angle and declination
.
 4 G E OD E TIC A S TR ON O M Y. 1 5 .

are two spherical ordina t es l cating a celestial obj ect with

co - o

referenc to the meridian and e q ua t or

e

1 4 I t is c nveni nt for som

.

purposes to refer t h posi

o e e

tion of a heavenly b ody by spherical -ordinat s t t h

. e

co e o e

planes f the meridian and horizon the two -ordinat s in

o ,
- co e

this case b eing the lt it d and im t n The lt it d f aa u e az u . a u e o

heavenly body is its angular distance abov t h horizon or e e ,

the angle between the line j ining the observer to the star o ,

and the horizontal plane A y great circle of the c l s t ial . n e e

sphere passi g through the z nith is called a t i l i l

n e v er ca c rc e .

T h altitude of a star is measured by tha t porti n of t h v rti

e o e e

cal circle passing t hrough the star which is included between

the star and the horiz n The im t h of a star or oth r o . az u , e

celestial body is t h angle b tween the plane f the meridian

,
e e o

and the vertical plane passing through the star Th sam . e e

definiti n appli s to a line j oining two terrestrial points Th

o e . e

azimu t h at station A on the Earth s surface of t h lin join ’

, e e

ing stations A and B is the angle betwe n the vertical plan, e e

at A passing through the lin A B and the meridian plan e e

of A The azimuth f a star is measured on the cel stial o e

sphere by that porti n of the h riz n included b etwe n the

.

o o o e

star s vert ical circle and t h meridian line I n g n ral the

’
e . e e

alti t ude and azimuth of a c lestial obj ect are bo t h changing e

rapidly because of t h Earth s r tation Th it / d i t

e
’
o . e z en t s a n ce

of a star is its angular distanc from the zenith m asur d e ,

- e e ,

of cou rse along a vertical circle Th zenith distanc and

,
. e e

altitud are complem ents f each other

e o .

1 5 The t
. m i l l t i t d of a station on the surface
a s ro n o ca a u e

of the Earth is the angle b tween the line of action f gravity e o

at that station and the plane of the e q uator It is m asur d . e e

on the ce lestial sphere along the meridian from t h q ua t or e e

to the zenith .

Th t m i l l git d
e of a station on t h surface of
a s ro n o ca on u e e
 1 6 G E OD E T1 C A S TR ON OM Y 16 .

system of spherical -ordinates frequently used by the co ,

—

astr n mer but s ldom by the ngineer I n this the ecliptic

o o e e .

and vernal equinox play the same part as do the e q uator and
vernal equinox in the ca e of declinations and right s a sc e n

sions .

1 6 In general when the engineer observes a heavenly

b ody he has one of four obj ects i n view nam ly to de t ermine
.
,

, e ,

his astronomical latitude the azimuth f a line joining his , o

station with some other terrestrial point the true local time ,

at the instant of observation or the longitud of his station , e .

The determination of longi t ude always involves a determina

tion of t h true local time together with additi nal operations
e o

which are in some cases quite complicated The instrument .

used in any case for the determination of time latitude or , ,

azimuth indicates the position of the horizon and , c o n se

qu ently of the z nith by m ans of attached spirit -l v ls or e ,

e e e ,

l l l
p pp l
f the v a rio u s v a u e s o f t h e a t it u d e o f t h is s i g e st a t io n a s t h u s o b t a i e d
ll m
o n n

is c a e d it s g d t i l t it d Th e c o rre s o n d in g st a t e e n t a ie s t o t h e
l m
eo e c a u e .

m w ly y w ll y
g it d It is e v id e t t h a t t h e g re t e r t h e n u b e r o f st a t io n s a n d t h e
l
o n u e n a

i t h e v e rt i c a a s g iv e n b t h e
.

id e
w ll p
sc a t t e re d t h e a re t h e

l ml
o e r e a re r n

w
g e o d e t ic a t it u d e a n d l o n g it u d e c o in c id e it h t h e o r a t o t h e e i so i d
m l
n .

p
Th e d iff e re n c e b e t e e n t h e a s t ro n o ic a
ll y ll
d g e o d e t ic l a t it u d e a t a g iv n
ml
an e

o i t is t h e re f o re u s u e d t he t ti i l t t d A si i a r
m I m
n c a a s a on e rro r n a i u e.

n t d e fin e s F o r f u rt h e r n f o r a t i o n o n
y pp
t ti i l git d
m
on

l k
st a t e e s a on er ro r n u e.

t h is s u b j e c t s e C a r 7 2 8 8 M e rri a n s G e o d e t ic S u r
y
s G o de s 2 8 - ’
e
’
e ,
.
,

e yi g p p 79—8 8 n y e xte d e d t re a t ise o g e o d e s

l l l
v n or .
,
a n n .
,

pl p w l
S t a t io e r o rs in n g it u d e o r d e fl e c t io n s o f t h e v e rt ic a a t rig h t a n g e s

m m m
n r o ,

w py yw l
t o t he e rid ia c h n ge the a n e o f the e rid ia n f ro the o s it io n it d
m m l
n a o u
,

w a n d so c h a n g e a ll a z i

w
t h r ise o c c t h s f ro the va u e s t h e o d u

m
o e u u

w
o the is h a v e H e c e t h e re a rise s t h e sa e d is t in c t io n b e t e e n t h e
l
r e . n

w
t mi l im t h o f in e a n d it s ge d t i i m t h a s is d ra n a b o v e
m l l l
a s ro n o ca az u a o e c a z u

y
b e t e n t h a st ro n o ic a a d g e o d e t ic a t it u d e s a d o n g it u d e s O n
l
e e n n .

p
t o f s t a t io e rro r t h e in e o f g ra v it a t a s t a t io n a d t h e a x is o f t h e
l m
a cc oun n n

t in g e e ra
y l y
Ea t h do in t e rs e c t He c e t o b e e xa c t t he e rid ia la n e
m
r n o n . n n
, ,

sa i d t o b e d e fi

y p w
st b d o t b t h e i e o f g ra v it a n d t h e E a rt h s a x is ’

l
u e n e ,
n n

p l p
i b b h i f g v -i t d t h i t i h ic h t h e a x is o f

I
f t t t t e e u r a e o n n n o a n
o r a
o n o y ,

r o t a t io r o d
n u c e d t r s e c t s t h e c e l e s t ia ns h e re
e .
 16 . P UR P OS E OF OE S E R VA TJ ON S . I7

by a basin o f mercury having a free horizontal surface The .

star or other c lestial obj ect is usually observed with a tele

,
e ,

sc pe The two p ints on the celestial sphere always

o . o

observ d are ther fore t h zenith and the object The right
e ,
e , e .

ascensi n and declination of t h object observed bec me

o e o

kn wn i nd p n d n t ly of the observations by the m ethods

o ,
e e
'

e ,

indicated in t h next chapter Th process most fr q uently

e . e e

us d is t acquire by instrumental observation and by the

e o ,

m ans indica t ed in the next chapter a knowledg f thre o f

e , e o e

th e elem nts arcs and angl s of some triangl on the celes

e ,
e , e

tial sph r of which e of t h unknown elements w

e one e , no

capabl of c mpu t ation is the q uantity s ught or is one

e o , o ,

from which t h r q uir d quan t ity can b readily derived

e e e e .

F xampl suppose that t h la t itude of the station of

or e e , e

observation is known and that the zenith d istance of a c rtain , e

star is accurat ly observ d Let t h tru local sidereal tim

e e . e e e

(se e 8 ) t t h1instant of bservation

a b re q uired
e I n t h o e . e

triangle the celestial sphere defi ned by the pole the zenith
on , ,

and t h instantaneous position o f the star the arc fr m t h

e , o e

z ni t h to t h pol is known being the complement of t h

e e e , e

la t itud f the s t ation The arc from the star to the p l

e o . o e

becom s known by the methods indicated in the next chapte

e r,

since it is the complem nt of h declination f the star at e t e o

th e instant of bs rvation The from the zeni t h to the

o e a re

star t h zenith dis t ance was direc t ly obs rved H enc in the
.

, e , e . e

spherical triangle pole -z nith -star all t hre arcs are known and e e

any of the angles may be computed The angl at t h pol . e e e

of that triangle is the hour-angle of the star at the instan t f o

observation This being computed by the me t hods f

. o

spherical trigonometry a mer addition to or subtraction e

from the right ascension of the star ( which b ecomes known

,

b y the methods of the following chapter) gives the true loca l

sidereal tim ( as will b shown later)e . e .
18 G E OD E TIC A S TR ON O M Y . 1 8 .

17account of the rapid appar nt motion of most

. On e

c lestial bj ects time enters as an important element into

e o ,

almost e ery astronomical problem with which the engin er

v e

has to deal Thr kinds of tim are in use in astr nomy

. ee e o

sidereal time apparent solar time and mean solar time

, , .

The passage of a star or other c l s t ial obj ct across the e e e

meridian is called it t it lm i t i s ra n s o r cu na on .

Th m ridian ( a gr at circle of the cel stial sphere ) is

e e e e

divided into t w half -circles by the poles I f the whole of

o .

the meridian be considered a star has two transits for each ,

complete rotation f the Earth on its axis one over that half
o :

of t h meridian stretching from pol t pole which includes

e e o

the zenith and the other over that half which passes through
,

th nadir The first of these is ca lled t h pp t it or

e . e u er ra n s

u pp lm i t i
er and t h s cond t h low r transit lower
cu na on ,
e e e e or

culmination Th word transit or culmination unm dified

. e o

usually means t h upper transit The xpressi n t h e e o e

passage of a star across the meridian r fers of course to the e

apparent motion f the star I t w uld b more accurate to

, ,

o . o e

say that t h meridian passes the star B t t refer dir ctly

e . u o e

to the apparent motion as if it w re r al sav s circumlocution e e e ,

is m ore clear in many cases and is not misleading if k ps , on e ee

in mind that this is mer ly a mod of speech e e .

1 8 A id . l d y is the interval b tween two successive

s ere a a e

t ransits of the vernal e q uinox across the sam meridian Its e .

hours are numbered fr m to 4 The sider al time is o 0 2 . e

o
h
ooat the instant when the vernal equinox transits
m
00
8

across the meridian Th id l t im at a g ven station and. e s e re a e i

instant is the right ascension of t h meridian or is the same e ,

as the hour angle of the v rnal equinox counted in the d i

- e , re c

tion of t h apparent motion of t h stars at that station and

e e ,

instant
Right ascensions b eing reckoned from west to east
.

.
,
 19 . A PPA R E N T S O L A R TIM E . 19

o pposite to the apparent motion of the stars it f ollows from ,

the above d finition that the si d ereal time at the instan t of

e

transit of a star is the same as the right ascension of that star

T he sidereal day is substantially the int rval of time
.

re q uir d for one rotati n of the earth on its axis and the
e o ,

uniformity of t he rota t ion of the earth is d pended upon to e

furnish the ultimate measure of tim B ecause o f t h moti n e . e o

o f t h v rnal quinox on t h c lestial sphere about 5

e e e
p e e , 0
”
er

year the sidereal day and h time of one r t ation f t h

, t e o o e

earth on its axis d i ffer by about one one -hundr d t h of a e

sec nd o

1 9 The interval b etween two successive t ransits of the

.

S un across the meridian is called an pp t l d y The a a re n so a r a .

app t
a re nl t im for any instant and station is the hour
so a r e

angle f the S at that instant from that m ridian

o un , B t , e . u

th e intervals between successive returns of the S to the un

same meridian are not exactly e q ual owing to the varying ,

motion of the Earth around the S and to the obliquity of un ,

the cliptic
e .

L t F ig erepresent a secti n of the universe on the plane

. 1 o

of the Earth s orbit as seen from some position in space on

’

the sid on which the nor t h p l is situa t ed The Earth is o e

seen moving around its or b it in a counter -clock wise direction

e .

while at the same time its rotation about its own axis app ars e

to be cou ter -clockwise The figure is t t scal but is

n . no o e ,

m r ly a d iagram i which certain dimensions are exaggera t d

e e n e

for t h sake of clearness S uppose that A is t h positi n f

e . e o o

the Earth a certain time about March

at when the S , 2 1 , un

is seen proj ected against the celestial sphere upon the vernal
equinox Let B be the position of the Earth one sidereal
.

day later Then A and B are parall l lines the vernal a a e

equino x b eing at an infinite distance ( on t h c l stial sphere)

.
,

e e e .

The Earth has made one complete rotation on its axis betwe n e
20 G E OD E TIC A S TR ON OM Y . 19 .

the two positions and t h vernal equinox has returned to the

, e

same meridian The Earth having moved a distance A B

.

along its orbit the S is now seen projected against the

,
un

celestial sphere at h instead of B fore the S will return a . e un

to the meridian o f positi n A again the Earth must rotate o

through the additional angle represented by t d d t re u ce o

t h pl
e f t h
ane E t h q
o t ( The figure represents
e ara ’
s e ua or .

section in the plan of the lip t i ) The apparent solar day e ec c .

will then b lo g r than the sidereal day by the time required

e n e

for the Ear t h to rotate thr ugh this angle — an aver ge o , on

'

a a

little less than fou r minutes .

L t the angle governing the excess of the apparent solar

e

over t h sider al day b examined further A the Earth

e e e . s

proceeds forward along its orbit the Su will apparently move n

backward on t h c l stial sph re along the ecliptic to p ints

e e e e o

h, d
c, t O ,
of the laws of gravitation governing the
e c . ne

moti n of the Earth in its orbit is that t h line j oining the

o e

Earth to the S swe ps over equal ar as in e q ual times

un e e .

The linear velocity then varies nearly inversely as the distance

to t h S and the angular velocity vari s still more than the
e un ,
e

linear The angular velocity is about 7% greater during t h

. e

winter ( of the northern hemispher ) than during the summ r e e .

The various arcs h h d t alo g t h ecliptic each a , c, c ,

e c .
,
n e ,

corresponding t one sider al day will vary in value through

o e ,

that rang B t the excess of the apparent solar over the

e . u

sid r al day depends upon t hese arcs proj ec t ed upon the

e e

equator along hou r -circles the rotation o f t h Earth being , e

uniform when m asured along the e q uator When such a e .

small arc as h near either equinox is projected upon the

a

e q uator it will be considera b ly red ced being at an angle

,
u ,

of 3 ? to the equator — the angle between the equator and

2

ecliptic at t h e q uino x es
,

e .

Oh t h other hand when a portion of the ecliptic about

e ,
 20 . TIM E . 2 I

m idway b etween the equinoxes is projected along its limiting

hour-circles upon the equator the projected l ngth will b e ,
e

greater than the original In short the di ff rence between .

, e

the sidereal and apparent solar day varies by a rather compli

t d law from ab ut 4
ca e 6 to 3 : being on an average
o
In
z
s m

3
m
( in sidereal tim ) e .

20 A pparent solar time is a natural and direct measure of

d uration inasmuch as it is indicated directly b y the hour
.

a ngle of the S t h most conspicuous of all the heavenly

un ,
e

b odies B t a clock or chronome t er cannot be regulated to

. u

keep this kind of time accurately sinc the di fferent days are , e

o f une q ual length To avoid the di fficulties thus arising from

.

the direc t use of the S as a m easure of time a fictitious un ,

m ea nS is used Th m S is supposed to move in the

un . e ea n un

e q uator with a uniform angular velocity and to keep as ear , n

the r al S as is consistent with perfect unif rmity of moti n

e un o o .

This m an S makes one complete circuit around the

e un

e q uator at a uniform rate while the Earth is making a com

p l t circuit
e e around its or b it at a variabl rate I t is s me , e . o

times as much as 6 minutes ahead of the real S and 1 un ,

som times behind it by that amoun t A m

e l d y is . ea n s o a r a

t h interval between successive transits of the mean S

e over un

the same meridian The m l t im for any instant and

. ea n s o a r e

station is the hour-angle f the m ean S at that instant from o un

that meridian F brevity mean solar time is often called

. or

simply m t im The mean solar day is about 3 56 longer

ea n e .
m “

than t h sidereal day —that being the amount by which the

e ,

apparent solar day exceeds the sid real day on an average e .

S tated more exactly 4 hours of mean solar time is the ,same

2

int rval as 4 3
e of sidereal time
2
h
0
m
.

The sidereal and mean solar time coincide for an instant

about March each year T he former gains 4 hours on
2 1 . 2

the latter in a year .

22 G E OD E TIC A S TR O N OM Y .

The q t i f t im is the correction to be applied to

e ua on o e

apparent time to reduce it to mean time I t is t h int rval

of tim b y which t h mean S preced s or follows or is fast
. e e

e e un e

o r slow of the real S at a given ins t ant I ts limiting valu s

,

, un . e

are ab ut 6
o and I t is given in t h A m rican
1
m
e e

E phemeris and Nautical A lmanac for every no n at Washing o

ton ( and G reenwich ) I t can be obtained for any intermedia t.e

instant with an error not great r than usually much less e 0
3
. 1 , ,

by a simple straight line int rpolation

- e

2 1 The i il d y according to the customs of society

. c v a , ,

commences and ends at midnight The h urs from midnigh t . o

to noon are counted from to and are marked M 0 12 A . .

The remaining hou rs from noon to midnight again num a re

bered from to and marked P M 0 12 . .

The t m i l d y commences at noon on the civil day

a s ro n o ca a

of t h same date Its hours are numbered from to 4

e . 0 2 ,

from noon of one day to noon of the next The astronomical .

time as w ll as the civil time may be either apparent solar or

e

mean solar The conveni nce of the astronomical day f the

. e or

astronomer aris s from the fact that he does not have t

e o

change the date on his record of o b servations in t h midst f e o

a night s work as he would be obliged to if he used civil

’

dates .

The zeros of sidereal apparent solar and mean solar tim , , e

are by definition the instants of transit across the meridian

, , , ,

of the vernal equinox the S and the mean S p , un , un , re s ec

t i ly
ve The time therefore ( of any of the three kinds ) will
.
, , ,

be the sam for two stations at a given instant only in case

e

those stations are on the same m ridian If the stations are e .

not on the same meridian the di fference of their times ( f o

any of the three kinds) is a di ff rence of two hou r-angles e

measured from the respective meridians to the same obj ct e ,

and is therefore the angle between the meridians or the di ffer ~

24 G E OD E TI C A S TR ON OM Y . 23 .

is merely ecessary i n each case t o apply as a correction

n

the di fference of l ngitude of the station and the standard o

meridian O nly the astronom r or engineer however is

. e , ,

oblig d to use this process The traveller has occ si n

e . a o

simply t change fr m o kind o f standard time to another o o ne

which di ffers fr m it by xactly one hour — interval which

o e , an

is easy to r m mber e e .

n rt M S l t S i d l Tim
To Co ve ea n o ar o e re a e.

23 To conv r t m an solar to sid real time or

. i e e e v ce

v ersa it is nec ssary t take account logically of two facts

,
e o ,

tha t t h zeros f the two kinds of day differ by a certain

e o

interval to be derived from the Ephem ris and that the t w

,
e , o

kinds of hours bear a fixed ra t io to ach other which is nearly e ,

but not q uite unity , .

The l cal mean solar time at St L uis M o 5

o . o , .
, 2
m

w st f Washington is 9
e o M J uly 9 8 9 ,
11
z 1
m
A . 2 , 1 2 .

What is the l cal sid r al time o e e

Lo c a lm l
so a r t i m
m m
ean e

Ti e o f ean n oon

Mea n l in t e r va l t o n e a re s t m 38 36 65
l l( §
so a r ean n oon 2 .

Re d u c t io n t o s i e re a d in t e r va se e 2 90 ) 0 00 26 . 06

l vl m
J ly
Si d re a in t e r to n e a re s t 71
l m m
e a ea n n oo n 2 39 02 .

S id e re a ti e o f ea n n oon , u

89 2 ,W h t in gto n 8 h m

l
29, 1 a as 31
C
l w)
o rre c t io n du e to o n g it u d e t o re

duce to St L o u is ( s e e b e o . O 00 08 . 64

J ly
.

Si d e re a l m ti e o f m ean n oon , u

2 9, 1 89 2 , a t St . L o u is 8 31 22 . 87

Re q u ire d si d e re a l m ti e at St . L o u is 5 52 20 . 1 6

first step is to obtain the mean solar interval betwe n

Th e e

the given time and the nearest mean noon and to reduce it
an equivalent sidereal interval by use o f the ta b les in § 9
'

to 2 0
g 3 2 . C ON VE R S ION OF TIM E . 25

( r printed
e from the back part of the Ephemeris ) The .

derivation of these tables from t h equation given at the end e

of each is fli i t ly obvious
su c en .

The next step is to derive from the A merican Ephem ris e

and Nautical A lmanac the sidereal time of that mean noon ,

or in other words the di fference of the z ro points of the t w

, , e o

kinds of time t noon of that day The Ephem ris in the

a . e ,

part headed Solar Ephemeris ( pp 3 7 7 3 8 4 in the volume .

—

for gives directly the sidereal time of every Washington

m an noon for the year What is re q uired is t h sider al
e . e e

time of St Louis mean noon The vernal e q uin x marking

. . o ,

th ezero f sid real t ime shifts 3

o e per mean solar day
,
m

wi t h respect to the mean S marking the zero of mean solar un ,

time Th sidereal time of mean noon for a given point then

. e

increases 3 per day St Louis being 5

”
. . 2
m

west of Washington its mean noon occurs at that interval of

,

m an solar time later than the mean noon of Washington

e .

Its sidereal t ime of mean noon is evidently t hat of Washing

ton increased b y the motion f the vernal equinox relative to o

the mean S in 5 un or 2
m

-
3 7 0 7)
8
56 -
555]
3

This proportional part is precisely that given b y the ta b le ,

92 for
0, the reduction of mean solar to sidereal time and
hence the correction is taken directly from that ta b le
,

H aving now the sidereal int rval to the nearest local mean
.

noon and the local sid r al time of that mean noon the
, e e ,

re q uired sidereal time is obtained by a simpl sub t raction ( or e

addition as the case may call for)

, .

Note that the longitude of the station is used only i n

r ducing the sidereal time of mean noon at Washington to the
e

local sidereal time of mean noon A error of 4 in the longi . n

3

tude produces an error of only in this reduction .

 GE OD E TIC A S TR O N OM Y
.

Ex ct i f m Sid l t M Tim
a m ple of t h e Redu on ro ere a o ea n e .

24 A t a certain instant in the v ning of May 89 e e 2 1 1 2

at H arvard O bservatory it was found by an observation upon

.
, ,

a star that the sider al tim was 3 4 What was e e 1

h
1
m

the mean time at that instant ? H arvard O bservat ry is 3 o 2

m

4 east
1
8
f Washington o .

G iv e n si d e re a l m ti h m

l m m
e 1 3 41
Si d e re a ti e o f ean n oon ,

Ma y 21 1 8 92 , a t W a s h in g t o n 3
h

C
,

o rre c t io n ,
du e to l g it d t
o n u e , o

re duce to Ha r v d O b se r
ar

va t o ry 2 90 ) 0 00 03 . 89

l m m
y
Si d e re a t i e f o ea n n oon , Ma y 21 , 1 89 2 , a t

H a rv rd O b s rv
a e a tor 3 59

l
d e re a in t e rv a a f t e r
Si l m
m m l
ean n oon

Re d u c t io n t o ea ti e n in t e r v a 291 )

Re q u ire d m ean t m
i e a t H a rv a rd O b s e rv a t o r y 9 40 44 . 10 P . M .

hm i Th e E p e er s .

25 The A merican E phemeris

. d Nautical A lmanac an

referred to in the above computation is an annual publi ation c

of the United States G overnmen t It ca n be obtained at any .

time by sending one dollar to the Nautical A lmanac O ffice ,

Washington D C It or its quival nt is a necessity to an

, . .
, e e ,

engin er making astronomical determinations as will be seen

e ,

by the many referenc s to it in t h following chapters A it e e . s

forms a part of the outfit of the as t ronomical observer and

computer the student should b come familiar with its general
,
e

arrangement should acquire a g n ral understanding of all

,
e e

part s of it and should obtain a t h d gh grasp of those par

,
or u

ti lcu arportions to which he finds especial refer nce in the e

text of this b ok To gain familiarity with the most o .

frequently used portions of the Ephemeris it is especially

desirable that the following pages o f the text at the b ack of
,
g 26 . Q UE S TION S A N D E XA M PL E S . 27

the E phemeris head d O the A rrang m nt and Use of the e n e e

A m rican E phem ris b r ad ; i the first fou r pages of e e e v z

the explanati n of P ar t I (pp 4 9 3 496 in t h volume f

e .
,

o — e

and t h first thr pag s f the xplanati n of P ar t I I

. or

e ee e o e o

( pp 5
.
5 3 f t h v
0 1—lume f o T h '
G overnm nts e o or e o e

of G rmany F rance and England and some others issu e

e , , , ,

similar publicati ns o .

QU E S T IO N S AND E XA M P L E S .

26 . p si t ion f the S project d upon the celes

1 . Th e o o un e

tial spher is always at s me point of the ecliptic Explain

e o .

why this statement is t true in regard to a planet no .

2 What is the r lati n between the latitude f a station

. e o o

and t h al t i t ude f t h p le t that stati n P

e o e o a o

3 G iv n
. the latitud of
e a station and the declination of e

a star how may the z ni t h distance of t h star at the ins t ant

, e e

of upper culmina t ion b determined e

4 I the case of a circumpolar star how may the zenith

1
n

distance at lower culmination b e det rmined the declination

.

e ,

and latitude being given

A i mp l t
c rc u is one comparatively near the pole say
o ar s ar ,

within ten d grees e

5 H ow would you determine the zenith distance at upper

.

culmination d also at lower for a circumpolar star of which

, an ,

the p lar distance is given ? Th latitude of the station is

o e

supp sed t be known

o o .

6 The hour-angle of the star V ega east of the meridian

.
, ,

at a certain ins t ant on the evening f J une 3 8 9 at the o 0, 1 2 ,

C orn ll O bservatory was The right ascension of 11 m s

e 4 2 l 1 1 .

V ega at that ins t ant was 8 3 3 What was the l cal h m 8

1 1
9 . o

sidereal time ? A lso what was the Washington sidereal time , ,

-Cornell b ing 6 east of Washington “n °

e 2 1
28 G E OD E TIC A S TR ON O M Y . 26 .

7 a certain
. instant
At the hou r angle and zenith distance
-
of a star are observed The declination of the star is known . .

In the spherical triangle star-zenith -pole what parts are known

and h w may the latitude of the station be computed
o

8 What was the hour -

. angle of the S on September 9 un 2 ,

189 at a station 4 9
2 , west of Washington when a h
1
m

fast o f lO l mean time indicated

,

clock which was ca

h m
2 4 1 PM ?
. .

The equation of time for apparent noon at Washington

on S eptem b er 9 was and for the 3 t h
2 o , 10
m

A 5 west of the m ridian ns . 2

h
0
m
e .

9 The mean
. time w 5 6 P M A ugust as
h
1
m
. .
, 1 0,

189 at a station
2 , 9 3 west of Washington What
2
11
w 1
m
1
8
. as

the sidereal time ? The sidereal time of mean noon at Wash

i gt
n on that day was 9 8
on
11
1
m

h m
A ns . 14 36
The sidereal time was 3 49
10 . the astronomical 2
h m

date A ugust 89 and the station

2 1 , 1
9 w est of
2 , 1
h
2
m
2 1
8

Washington What was the mean time and the civil date
.

The sidereal time of mean noon at Washington on the t 2 15

( civil date ) was and on the

10
h
d 01
m
22 , 10
h

A ns Mean time
.
45 A M C ivil date A ugust
1
11 m
. .
, 22 .

The apparent solar time at a station 46

II west 1
h m

o f Washington was at a certain instant on A pril 7 8 9

.

1 , 1 2,

10 33
h
Mm
What wasAthe.mean time. ? The equation
o f time at Washington apparent noon on that date was .

and on the 8 th was 1

h
A ns . 1o A . M .

The hour-angle of the S as observed at a certai

12 . un n

instant at a station 4
, 34 east of Washington on the 2
“
1
m 3
,

for noon of May

e 89 was found to be 2 1 ,
1 2 ,
2
11

What was the sidereal time The equation of time for

apparent noon at Washington was 3 — on May 111
2 0,
 26 . Q UE S TION S A N D E XA M P L E S . 2
9

and 3 on the
m
2 1 5t . The sidereal time o f mean
noon at Washington on the 2 1 5t was 3 59 11 m
1
"
A ns o6
m
. 1

3
1 uppose
. Syou are carrying a watch which is fast of 20
8

standard ( 7 st h meridian ) time and that you wish to start a ,

sidereal clock on correct time to within at Cornell ( 6 I

f5
2
m
1
3

east of Washington or 5 5 56 west of G reenwich ) S p

h
0
m 8
. u

pose the dat to be S ept 3 8 9 and the sidereal time of

e . 0, 1 2 ,

m an noon for Washington on that dat to be

e 39 e 12
h m

What tim should the sidereal clock indicate wh n your watch

e e

reads 7 h
?00
m
00
8
A 9 34 ns 1
h In

x
.

4
1 E . plain why the sidereal tim of mean noon as e

given in the last column of the Solar Ephemeris (pp .

— 8
3 77 3 4 in the volume for in the A m rican Ephemeris e

and Nau t ical A lmanac is not the sa m e as t h apparent right e

“

ascension at mean noon as given in the second column .

5 Why
1 . not the e q uation of time for appar nt
15 e

noon as given in the eighth column t h same as the di ffer ,

e

ence of the two columns mentioned in the preceding example ?

6 Wha t is the relation bet ween the right ascension of
1 .

the S at mean noon the e q uation of time at mean noon

un , ,

and the sidereal time of m an noon e

7
1 L ook
. up the sidereal time of mean noon for to day
-
in the Ephemeris Then knowing the time of day and you r
.
,

latitude hold two sheets of paper parallel respectively to the

,

p lane of the equator and the plane of the ecliptic .

3 0 G E OD E TIC A S TR ON OM Y .
gY
o
t

C H P TE R
A II .

C O M PU TA TI O N O F RI G H T A S C
C
E N SIO N A N D
DE L I N A TI O N .

27 I n the astronomical practice of the engineer the right

ascension and declination of the obj ect o b served are usually
.

known quantities determined from sources external to his own

o b servations The object of this chapter is to show how the
.

right ascension and declination for the instant of observation

are obtained from the available sources of information
The various heavenly b odies which the engineer is called
.

upon to bserve have all been observed fre q uently at the

o

various fix d o b servatories with large instruments and at

e

many diff r nt times xtending over a long period of years

e e e .

F rom these observations the positi ns t hat is right o , , a sc e n

sions and declinations at various stated times are det rmined

, e ,

and the m tions are carefully comput d This makes it pos

o e .

sible to compute t h position of each of these bodies at any

e

stat d future time with an accuracy depending on the pre

-
e

i i
c s on of the o b serva t ions and the remoteness of the future
time The results of such computations f positions made
. o

in advance and also the data for such compu t ations are
given in the ephemerides issued b y various governments the
, ,

A merican Ephe m eris B erliner Jahrbuch C onnaissance d

Temps ( P aris) B ritish Nautical A lmanac etc V arious other
, , u

, , .

occasional publica t ions als give the data for such computa
o

tions The engineer uses these computations o f position

made in advance and the p b lished data for such computa
.

,
u
3 2 G E OD E TIC A S TR ON OM Y .
3 0 .

of the S is a known function of time as the independent

un

variable I t is given in the Ephemeris for certain stated

.

times When it is required for any other time instead of

.
,

computing it directly from the known function it is much

more convenient and rapid to deduce it b y interpolation from
,

th stated numerical values

e .

In t erpolation always leads to approximate results which

may be m de more exact as the process f interp lation is
a o o

made mor complicat d and laborious The

e
f i t e . e rro r o n er

p l t
o a i is t h di
on f
f erence b tween a n
e interpolated value and e

th value which w uld be found if one resor t ed to direct

e o

computation from the known function O f the m ultitude f . o

m thods of interpolation with widely varying d grees of con

e , e

i
ve n en ce, rapidity and accuracy three methods will be f un d
, , o

su fficient for the grou nd covered by this book Th s thre . e e e

may be described bri fly as i t p l t i l g h d i t e n er o a on a on a c or , n er

p l t
o a i l g t
on g t and
a on i t p l at i lan
g p h l
en , n er o a on a on a a ra o a .

h dI n t erpola ti
on a lon g a C or .

3 0 In i t p l t i. l g h d the rate of ch nge of the

n er o a on a on a c or a

function between the two stat d values f the variable which

,
e o
\

are adjacent to the value for which the interpolati n is to be o

made is assu med to be constant and equal to the total change

,

of the function between those points divided by the interval

between the stated values of the variable I f the actual .

values of the function were repres nted graphically all inter e ,

p l
o at d valuesewould li along chords of the function curve e ,

connecting points on the curve corresponding to stated values

of the variable F xample the right ascension of J upiter
. or e ,

at 35 5
12 mean
“
time at Washingt
m
. n on O t 8 9 was o , c . 1 ,
1 2 ,

1
h
2 1 and at
m
3 on O t was 12
“
1
m
. 1 c . 2 , 1
11
z o
m

( Ephemeris p R e q uired,
its right
. ascension at 5 1
“

1 4
m
Washington
. 2 mean time on O t The interval , c . 1
 3 1 . IN TE R P OL A TION .
33

b tween stated values of the variable is 3 55 6

e 2
“ m
.

The change in the value of the function is The

rate of change is then + 3 93 per 2
h
.

hour The interval ov r which the interpolation is carri d

. e e

from t h nearest given value is 5 4

e 35 5 1
11
1
m
. 2 12
h m
. 2
11

3 8
m
7 . The chang during that interval is e

The re q uired right ascension is 1

11
z 1
m

3
'
. The
2o result would have
1 been
h
2 1
m

identical with this had the interpolation been made from the
other adjacent value nam ly that at 3 on O t d e 12
h
1
m
1 c 2

I n algebraic form this interpolation may be expressed thus

.
, , . .

the first form being used when the interpolation is made

forward from t h valu F and the second when it is made
e e , ,

backward from F ,
.

F is the re q uired interpolat d value corresponding t t h e

value V of the independent varia b le V and V are t h

, o e

, .
, , e

adjacent stated valu s of the argument to which correspon d e

the given valu s F and F f the functione 1 ,

o .

I t
n e rpola ti on a lon g a Ta n ge n t .

31 . is
In t e rp o la t io n g neral mor a lo ng a t a nge n t , In e , e

accura t e than in t rpolati n al ng a chord but can only be

e o o ,

used convenien t ly when the rates of change or first di fferen ,

tial coe fficients f the function giv n at the stated values

o , a re e

of the variable in addition to the values of the function

,

itse lf I n this interpolation the rate of change f the inter

.
, or

va l from the t stated value of the variable to the valu

n ea re s e
34 C A TR ON OM Y G E OD E TI g3 S . 1 .

for which t h interpolation is to b e made is assumed to be

e ,

constant and equal to the given rate of change at the stated

value of the varia b le .

The interpolated points represented graphically would lie

on a tangent at the nearest stated value of the variable to
the curve representing the function F e x ample let it be
, ,

. or ,

re q uired to find the dec lination of the S on S ept 5 8 9 un .

, 1 2 ,

at 9 3 A M Washington mean time O page 3 8 of the

“
0
m
. .
, . n 2

Ephemeris for that year t h ear st time for which the , e n e

declination is giv n is Washington mean noon of that day

e , .

t hat instant the declination is Its rate ° ’

F or 6 8 2

o f cha nge for that instant is stated to be per hour .

The inter al over which the interpolation is to extend is

v

b k w d f rom noon
ac ar Then by interpolation along the tan .

gent to the curve ( representing declinations) at noon of Sept .

5 t h there
, is obtained as the declination at 9 3 M
“ m
0 A .

6
°
8 2
’
6 3 In this method °
0
'

of interpolation the shorter the tangent the smaller the error

of interpolation and therefore care should b taken to int r e e

b
,

p l t
o a e from the t stated value of the varia
n ea res le The .

fo mula for this interpolation is

r

F, and
are the required interpolated valu e and the corre
V,
sp di g
ongiven argumne nt V and F are the earest tabular , , ,
n

value of the argume t and the corresponding value of the n

function and i the given first di ff erential coe fficient

, s
1

corresponding to V ,
.
8 33 IN TE RP OL A TION .
5

In t erpola t ion a lon g a P a ra b o la .

32I n i t p l t i l g p b l it is assumed that the

. n er o a on a on a a ra o a

s eco nd di fferential coe fficient of the function is constant

between adjacent stated values of the independent variable ,

or in other words that the rate of change of slope of the

, ,

function cu rve is constant betwe n those points This e .

a ssumption places the interpolated points along a parab la o ,

wi t h axis vertical passing through two points of the function

,

curve — the uniform rate of chang of slope being a property

,
e

of such a parabola There are t w case arising under this . o s

m thod depending upon whether the first di fferential is or

e , ,

is no t given for t h s t ated values of the variable e

an e x ample of the first case take the prob lem

.
,

33 F . or

pr posed in the p eceding section in which it is required to

o r ,

find the declination of the S at 9 3 M Washington un

“
0
m
A .

mean time S ept 5 8 9 The data given in the Ephemeris

, .
, 1 2 .

for 8 9 for Washington mean noon Sept 4 are declination

1 2 .

6 5 and the first di fferential coe fficient

°
0
’

and for Sept 5 declination 6 8 and first .

°
2
’

di fferential co ffi cient I t is proposed to place

e

the int rpolated value on a parabola (with axis ver t ical) coin
e

ciding with the curve of declinations at the two given points ,

and also having a common ta gent at each of these points n .

To make the interpolation the principle will be used that a

chord of such a para b ola is parallel to the tangent at a point
,

of which the abscissa is the mean of the absciss f the two a o

e nds f the chord The slope of the chord ( of t h para b ola)

o . e

corresp nding to the interval 9 3 to

o on S ept 5 is 11
0
m
.
,

then t h same as the slope of the tangent at the middle of

e

that interval 45 The slope of the tangent changes

,
h
10
m
.

by in 4 hours or 2 ,

- 8 per hour The slope at h m

0 . 010 45 . 10
3 6 G E OD E TIC A S TR ON OM Y .
34 .

The interpolated alue v at

9
h
3 0 is 6 8m °
2 6 3
’ °
0
’

This method of interpolation though most easily remem

b
,

e r d perhaps in the geometrical form may be put in con

e , , ,

ven en i t algebraic form as follows

V1 — V,

( )
1

( )
dF dF dF
F I = F I + [ V1 V I] —
j
7
2 ,7
l
7
0 V 271 Va — V I

— V9 )

( )
1
dF dF 7(

( )
V
F1 =F V1 — V2 ] +
1

dV Va — V I
l
a /
I
2 2

according to whether the interpolation is made forward from

V or backward from V The notation is the same as in the
preceding paragraphs The two results are ide ntical b ut t h
.
, ,

arithmetical work will b e shorter if the interpolation is made

.
,

f rom whichever of the giv n points happens to b e the nearer e

3 4 The second case of interpolation along a para b ol

.

. a

occurs when the first di fferential coe fficients are not given
The assumptions involved are j ust as before A an e x ample
.

take the problem proposed a f w paragraphs b ack of findin g

.
,

e ,

the right ascension of Jupiter at 5 4 Washington mean , 1

h
1
m
. 2 ,

time on O t ,
89 The Ephemeris gives the righ t
c . 1, 1 2 .

ascension
at 39 9 1on h
S ept 3 ; z 1
m
12
h m
. . 0

at 3 5 5 on O t
1
h
; 2 1
m
12
b m
. c . 1

at 3 1on O t
h
20
m
12
11
1
m
. 1 c . 2 .

I t is proposed to interpolate t h required point on a p e a

rabola with axis vertical passing through t hese three giv n

, , e

If t h e se c o n d d e riv a t iv e is c o n st a n t t h e n ,

is re al l y 55; C l l 2
a V1 V, AV . Th e n (3) pu t in t h e l l
c a c u u s n o t a t io n

m
’
’ 9
dF d F A V dF d F
be co es F1
2
F;
17
AV +al V i- In wh i c h a re l
va u e s

p p
.
i
2

c o rre s on d in g to t he o in t F 1 V
 34 . IN TE RP OL A TION .
37

p ints
o A gain using
. t h principle that in such
, a par b ola a e a ,

hord and the ta gent at a point of which the abscissa is the n

m an b etween the absciss of the two ends of the chord are

c ,

e a

f
,

p aralle l the slope o ,

the parabola at any point may be com
p t
u d eThe slope
. of the parabola at the middle of t h first '

interval at 3 7 7 6 3 on O t is
'

n m h
,
o . o . c . 1 ,

per hou r A t the middle of the second .

interval at 3 3 3 2 56 on O t it is
11 m 1
,
0 . 0 c . 2 ,

per hour The interval over which the .

int rp lati n is made from the near st given value is

e o o , e , 12
11

35 5
m
to.
5 4 on 1 O t or
11
3 8
1 7
m
. The
2 c . 1 , 2
h m
.

slope f the chord f this interval is that of the tang nt

o or e

at its middle 3 54 8 ,This slope is assum

1
“ m
.
,

ing the rate of change of t h slope constant e ,

h
63 o
5
6 6 h
.

per
2 4 5 3 o .

hour The right ascension at 5 4 is

. 1
h
1
m
. 2 1
h
z 1
m

This sample interpolati n is

1
h
z 1
m
o

made in the present form simply for the pu rpose of ill t t u s ra

ing the principles involved The numerical work of inter .

p l t
o a i should onordinarily be done as indicated in formula ( 4)
of t h following section
e

P utting this method in the algebraic language it takes the

.

following form Let F F and F be three successive given: , , , , ,

values of the function corresponding to the values V V and ,, , .

V of the independent variable ; and let V be the stated value

, ,

of the variable nearest to which lies the value for which the
i nterpolation is to be made Let F b e the required value of .
,

the function corresponding to V Then , .

V1 +V 2 V2 +Vl
F a—F 2 F 2 —F 1
%
2
A =a + Va — Va
_
Va — V I Vs-
l-Va V2 + V I [ m — mh
2 2
38 G E OD E TI C A S TR ON OM Y .
35 .

or in simplified form
, ,

— FI fP — 1 7 F2 _ E VI _ V
‘

l
s 2 ]
V — V
VZ — V. V. — V, V. — V. [ I J (4)

I f as is usually the case t h successive di fference e

b etween V V and V are all the same and e q ual to D thi

, , s

,, , , , , s

may be further simplified to the form

D
lV I c l ; ( )
4 a

in which d is the second difference or (F F ) (F

, , , , ,

The second term in the s q uare bracket will usually be com

p t i
a ra ly small
ve and ther fore easy to compute,
e .

F a more complete discussion of interpolation givin g

or ,

other more complex and accurate formul see Ch t a auve n e

’

S pherical and P ractical A stronomy

, s

l pp 79 9 vo I — 1

D oolittle s P ractical A stronomy pp 6 9 9 8 ; and Loomis

, . . .

’
—

P ractical A stronomy pp
.
,

, . 2 02 —2 1 2 .

A ccu ra cy t t i f P it i f S
of I d Pl
n e rpola t on o os on o un an an e s .

35 A interpolation
. along
n a tangent — th first di fferen , e

tial coe fficients or hourly changes being given — fr m t h , o e

values given for noon of each day in the Ephemeris ( pp .

— 8 of the volume for will giv the right ascension f

3 77 3 4 e o

the S at any t im wit h an rror of interpolation not exceed

un e / e

ing and the declinati n with an e rr r of interpolation t o o no

e xceeding F nearly all cases the rror of interpola

or e

tion will be much less than these xtreme limits A pp i e . ro x

mately the extreme error of interpolation along a tangent i

, s

one -eighth of the second di fference at that point — meaning ,

*

pl w ll v d l y g v
wh pl pi m w yb w
Th e in t e r o a t io n a on l g a ta nge n t i e i en t i e the g re a t e s t
e rro r en t he in t e r o a te d o nt is id a b let een t he t a u a te d va lue s ,
40 G E OD E TIC A S TR ON O M Y .
37 .

transit at Washington ( pp 3 3 3 of of every tenth

. 02 — 1

transit for ab ut stars ; and the right

o 2 00 i ly for a sc e n s o n
‘

on

every tenth transit visi b le at Washington of about more 2 00 .

O ther national Ephemerides contain similar lists which often ,

comprise about the same stars This list is made up of stars .

whose positions are well det rmined by many observations at e

various bservatories Th y are als chosen with especial

o . e o

reference to the needs of the engineer d navigat r as a n

o

regards brightness and distribution on the celestial sphere .

A idea of t h care with which their positions have been

n e

d termin d may be gained from the m ere statem nt of t h

e e e e

fact that in compu t ing many of these declinations fifty cata

l g
o of record d observations at many di fferent o b serva
ues e ,

tories made at various tim s during a total interval of a

,
e

centu ry and a q uarter were consulted and the various , ,

observati ns up n any one star combined in each case in a

o o

single least -s q uare computation .

*

The p sitions of the close i m p l at any time may

o c rc u o a rs

be btained with all needful accuracy by interpolation along

o

a chord from the values given in the Ephemeris F the . or

other stars given in the Ephemeris ( t -day intervals) an a \

10

interpola t ion along a parabola will usually be necessary .

When other stars must be observed than these Ephemeris

stars of which the places given at fre q uent in t rvals a a re e ,

complica t ed pr cedure is nec ssary to obtain the p sition of

o e o

the star at the time of the observation This proc ss forms

'
. e

t h subj c t of the remainder of this chapter

e e .

The positi n or place of a star is usually given in one of

o

y y m
h e rn B o u n d a r f ro t h e La e o f t h e W o o ds k
ky pp mp
Se e Su r ve o f t h e N o rt
M o u n t a in s ( W a s h in g t o n
p mp p yy w pp
t o t h e Ro c -6 1 le t e
0
4 9 , 5 fo r a c o .
,

y p
re o rt o n t h e c o u t a t io n o f s t a r l a c e s fo r t h a t su rv e b L e is B o s s (
l
.

p m mp
— 2 g iv t l g t d ) M f t h e st a r l a c e s g iv e n in t h e
m
4 2 1 4 4 e c a a o u e s c o n s u e a n . o

E h e e ris a re f ro t h is c o u t a t io n .
g3 8 . A B E RR A TION .
41

thr e ways which should be carefully distinguished Either

e , .

its pp t place t place or m place is given The

a a ren , ru e , ea n .

right ascension and declination as defined in indicate , 12 ,

th e t place of a star or other celestial obj ect B t the

ru e . u

apparent direction of a star even aside fr m the refraction of , o

th e lin of sig ht by t h terrestrial a t mosphere is a ffected by

e e ,

a b rration Th pp t place of a s t ar is its true place

e . e a a re n

m difi d by the aberration f light A observer sees a star

o e o . n

in a position which di ffers fr m what is tech ically called its o n

t place by the eff ct f refraction only It should be *

a pp a re n e o .

carefully noted that the Word apparent is t her used ”

no e

in the ordinary s nse but in t h special technical sense which

e , e

it must be understood to have hereafter throughout this book .

Ab r t i e ra on .

38 A b ti . is an apparent displacement of a star

e rra on

resulting froin the fact that the velocity of light is t infinite no

as compar d with the velocity of motion through spac of the

e e

o bserver stationed at a point on the Earth s surfac ’

I f one is standing in a rain which is falling in vertical

, e .

lines the umbrella must be held dir ctly verhead I f h w

, e o . , o

ever one is riding rapidly through such a rain -storm the

, ,

umbrella must be inclined forward I n t h first case a dr p . e o

of rain entering t t h centre of one end f a straight open a e o

tube held wi t h its axis vertical would pass along the axis of
t h t ub to the other end without touching the tub
e e I n th e . e

second case however if it is desired that drops which enter

, ,

the tube at the upper e d shall continu down the tube n e

without touching the sides it will b necessary to inclin the ,

e e

t b forward from the vertical t a c rtain angl which is

u e o e e

dependent on the relative velocity of the horiz ntal motion o

of the tube and the vertical motion of the rain S wh n a . o e

Fo r a l
deta i e d c o n si d e ra t io n o f re f ra c t io n se e 6 7-6 9 .
42 G E OD E TIC A S TR ON OM Y .
39 .

telescope is to receive along its axis the light undulations from

a star it must b e inclined forward in the directi n of t h
, o e

actual motion of t h telescope in space so as to make a slight

e

angle — the aberration — with the actual line j oining telescop e

to star This small angle the aberration amounting at

.
, ,

most to about is vid ntly d pendent upon the relative e e e

velocity of light and of the telescope and t h angle between , e

those two velocities (The student may easily draw . a

diagram for himself showing the geometrical relations con

cerned ) The motion of the telesc pe is compounded of that
. o

due t the diurnal rotation of the Earth on its axis and the
o

annual revoluti n of the Earth abo t the S o These give u un .

ris to t h d i l b t i and
e e l b
u rn a ti a
p e rra on a n n ua a e rra on , re s ec

ti ly
ve The diurnal ab rration evidently affects right asc n
. e o

sions has no effect upon declina t ions Th . e

annual abe rration in general a ffects both F an exampl . or e

of the w in whic h diurnal aberration is taken into account

av

in c mputations see 5 9 6 The effect of annual aberration

o , .

is included in the apparent place computation r at d lat r in c e e e

this chapter .

The velocity of light is according to the best d t m i , e er

nations abou t 8 6 3 miles p mean solar second It

\

*
,
1 00 er .

requires about eight minutes for light to travel from the S un

to the Earth A obs rv r then d es not see a c lestial

. n e e , ,
o e

bj ect in its tru position at the instant whe n the light enter
e s

the eye b ut in the position which it occupi d when that lig ht

o

,
e

l ft the obj ct — appr ciable interval earlier for all celestial

e e an e ,

obj cts This phenomenon is called p l t y b t i

e . a ne a r a e rra on .

With this form of ab rration the engineer is not concern d e e .

3 9 Th m. pl of a star is its position referred to the

e ea n a ce

mean equator and mean ecliptic as distinguished fr m its , o

ll x i R l d C W H kn
pp
S T ee S l P he o ar a ra a an d ts e ate o n st a n t s , m . ar e ss,

W sh i g t o n
a n , 1 891 , . 1 42 a n d 2 9-3 2 .
 39 . C MEAN
43 . PL A ES .

position as referred to the actual or true equator and ecliptic .

The e q ator and ecliptic as they would be if una ffected by

u

periodic variations in other w rds by nutation are called the

,
o ,

m ea n q t e and
uam li
orp t i ea n ec c .

The m an place of a star t hen at a given instant di ffers

e , , ,

from the true place by the effect of nutation at that instant ,

and from the apparent place by the e ffects of both nutation

and aberra t i n o .

T avoid inconveniences arising in the cou rse of computa

o

tions of star places if any other form of year is employed in

,

reckoning time the astron mer uses what is called the

,
o

Besse li fi t it
an i y The
c beginning
ous of the fi t it
ea r i y . c ous ea r

is t h instant at which t h c les t ial longitude of the m an

e e e e

Sun is or in other words when the mean S is 8

, , un 2 0
°

from the v rnal equinox measur d along the ecliptic The

e e .
*

b ginning of the fictitious year di ffers from the beginning f

e o

the rdinary year by a fraction of a day which varies for

o ,

di fferent years .

The places given in the Eph meris ref rred to in 3 7 e , e ,

for every day or every ten days apparen t places and are , a re ,

so mark d When t h ngin r is obliged to have recourse

e . e e ee

to stars which are t so given in the Ephemeris he consults

no ,

one or more of the various available star catalogues or star

li t }
s s Th se catal ogues and lists give the mean places of the
L
e

stars at the beginning of some stated fictitious year together ,

with other data relative to each star The problem which .

then confronts the engineer is to d rive from that given e ,

m a place the apparen t place at the time at which his

e n ,

observation was made This is done in two steps F irstly

’

. .
,

the mean place of the star is reduced from the epoch of t h e

catal gue to the beginning f the ficti t ious y ar t some

o o e a

Se e d e fi n it io n o f c e l e s t ia ll
g it u d e § 1 5
l l
on ,
.

1 Fo r re f e re n c e s t o a fe w o f su c h c a t a o g u e s a n d ist s se e g1 4 1 .
44 C TR ON OM Y G E OD E TI
4 AS . 1 .

part of which its apparent place is desired Secondly the .

,

apparent place of the star at the time of observation is

deduced from the mean place at the beginning f the fi t i o c

tious year .

t f M
Re du c ion Pl f m Y t Y o ean a c es ro ea r o e a r.

40 To serve as a concrete example let it be supposed

tha t the star 7 H ercul s was observed at its tra sit across the
.
,

1 e n

m ridian at S t Louis Mo on J uly 6 8 9 ; and the

e .
, .
, 1 , 1 2

authority depended up n for i ts position is B oss s Catalogue o

’

of 5 Stars for
00 This star is No 3 in that cata . 12

l gu and its mean place as there given for the beginning of

o e

the ficti t i us year 8 7 5 is

o 1

7 4 mean1 right
h
ascension
1 m
;
7 47 4 7 mean declination
2
° ’
2 1 .

( Throughout this book and 6 will be used to indicate a

the apparent right ascension and declination r spectively t , e , a

the t ime of the obs rvation u nder consideration The same

e .

letters with the subscript thus 6 will be used to indi m , , a m , m ,

cate the mean plac With a year as a subscript as above

e .
,

they will be understood t indicate the mean place at the o

beginning of that fictitious y ar ) e .

41 The reduction from the mean place at

. to that
at involves simply the change in the mean e q uator
and mean ecliptic during that time The determination of .

the laws of change of these two fundamental reference circles ,

and the method of computing the effect of those changes upon

right ascensions and declinations belong rather to the prov ,

ince of the astronomer than to that o f the engineer It .

y B y m Wo o d s
ky M pp
Su r ve o f t h e N o rt h e rn o u n d a r f ro t h e La k e o f t he t o the
-
Ro c oun t a in s (W a s h in g t o n 59 6 1 5
2
, . .
 41 . C 45 ME A N PL A ES .

su ffices for the engineer to accept the results o f the investi

g t i
a of the astronomer
ons in the following f orm :

u . sin a ”, tan 6m + ,
u ;

— n . co s u m
— p
’ °

d ot
c o s arm — u . sin a ” .

dt

in which m 46 6 3 8 49 ( t .8 ) (
0 t b
2 eing ex 0 . 000 2 1 00

pressed i y ars) and n e , 8 6 3 (t n 0 . 0000

The numerical values for m and as here given are those n

most ext nsively used and are the result of xhaustive

e e

inv stiga t ions by the astronomers P eters and Struve

,

e and . u
,

Ju are p p m t i per year in right ascension and declina

’
ro er o on s

tion respectively for an account of which see 44 4 5 F

, , . or

th pres nt thes proper motions may be considered simply

e e e

as changes at a uniform rat in each of the two -ordinates e co ,

without any refer nce to their meaning or method of deriva

e

d da
dt
an d
dt
are
a
t
'"
f c h ange per year
’”
ra es 0 .

The formul given above are neither complete nor exact

a ,

many terms of the exact formul having been dropped and a ,

those which are retained having been s omewhat modifi d e .

B t th y furnish t h
u e complete basis for a reduction with
e ,

su fficient accuracy for the purposes of the engineer from the ,

mean place given in a catalogue to the mean place at the

beginning of any other fictitious year within thirty or perhaps
fifty years F the formul in complete form adapted to
. or a

the use of the astronomer to bridge over long intervals of

time — sometimes more than a century — see S urvey of
, ,

the Northern B oundary from the Lake of the Woods to the

46 G E OD E TI C A S TR ON OM Y.
54 2 .

Rocky M ou ntains pp 4 6 4 and for a detailed discus 1 — 20 :

si n of them see D oolittle s P ractical A stronomy pp 56

.
,
’
o , . 0

57 8 and 8 8
5 3 5 9 — .

42 The engineer is however relieved o f the necessity for

p rforming the numerical operations indicated b y formul
.
, ,

e a

and F star catalogu s and lists give in addi or e

’
d d
tion to and the values of a m and the term t a
’
m
2

is tabulated in 9 of this b ook for the arguments

,
*
2 2 ,

a m and z o f which it is evidently a function

d ”,
.

S tudents will find slight di fferences b etween di fferent

authorities in regard to the nomenclature of this part of the
k w
Th is
pl
b l e is so fa r a s t h e a u t h o r n o s a n e w o e It wa s c o m
m
ta

m ml
, , n .

w p
f f u a ( 7) a b v e e d a t e 1 900 o s ix ce s o f deci a ls
d f r t
h t

l l
p u t e r o or o ,
o , a

mp w
a d a f t e r a rd re d u c e d t o fi v e
n It is h o e d t h a t a ll t h e t a b u a r v a u e s a re .
,

in s o fa r a s t h e c u t a t io n is c o n c e rn e d
o it h in o f a u n it in t h e fi f t h

pl h w v pp x m
,

f m l I lf
2
d 6m
ace Th e or u a u se d fo r 15, o e e r, a ro
O

i a te
O

In
0

t se in

p p m h y d
.
2
07

m d
h ld p d
h a v in g it t e t he ffe c t f t io n T in ic a t e s t h a t t h is
m m ll l
o e o ro er o . eor

p p p y
o is s io n s ou ro u ce a n e rro r s o s a as to be ne g ig ib l e fo r our

re se n t ur o se To t e st t hat c o n c u s io n l as we ll a x
s t h e a c c u ra c o f co m

p y y
.
,

bl
9
d 6m
uta t io n o f the ta e fo r fi ft s t a rs (e ve r te n t h) o f B o ss
’
L ist ,

B y p mp
,
afl
v d f m l
w y w mp
N o rt h e rn d a r Re o rt wa s d e ri e ro t h e t a b e a n d c o a re d
l v l m
o un ,

it h t h a t g iv e n b
w l
u te d f o
’
B o ss in t h e is t B o ss the

ml
a ues e re c o . r

e xa c t fo r

y y
Th e g re a t e s t d iffe re n c e f o u n d wa s o ooo o 3 Th is ou d
"
u a . . .

ly y y
c a u se a n e rro r o f o n l o 1 in a re d u c t io n e x t n d in g o v e r 3 0
"
e a rs a n d

l l
o . e ,

w w
50 e a rs It is b e ie v e d t h e re f o re t h a t t h e t a b e is a b u n d a n t l
”
04 i

lm m l
on o . n .
, ,

w y
it h in t h e i it s o v e r h ic h it s a rg u e n t s e x t e n d It s h o u d
m lm
a c c u ra t e .

not h o e v e r b e a ssu e d t o h o l d g o o d b e o n d t h o se i it s Th e t a b l e
m p iv ly wh
.
, ,

ga t i ve
da
doe s ve r the co in h — is
no t co a ra t e ra re c a se s ic
Bt
'
ne ( fo r
pl m l ( )m
l mp y m mp
the ) Fo r t h e se t he fo r d 7 u st be Th e
md
s t a rs n e a r o e . c a se s u a u se .

b w w ld g v
b is ute d fo r t h e Th e uta t io fo r
v l gf m h
ta e co ear 1 900 . sa e co n a e an y

bl by d m l pl
d t d i df fe rin t f t he
m
a te e een 1 700 a n 2 1 00 ou e a ues i ro o se o

n ot t ha n u n it in t h e l a stgv eci i in t he

bl
ta e o re one a ace en

ta e .
48 G E OD E TI C A S TR ON OM Y .
§ 44 .

If dt had not been g ven the star l st as frequently t

“ i in i , i

is not it could have been obtained b y ent ring the t abl

, 9 e e 2 2

d
with the argumen t s 7 4 6 and dl
um 3 45 1
h
1
m
.
a m
2
8
‘ ‘

The value as then found from the table would have been
and the final value for would have b een
identical with that given above .

r M ti P rope o on .

44 When the
.
-ordinates of a star as o b served directly
co ,

at widely separated times are reduced to the same epoch it

, ,

is usually found that asid from dis repancies arising from

, e c

accidental errors of obs rvation there are syst matic di ffer

e , e

e n cesin the various values indicating a steady movement of

the star in some one directi n with t h lapse of tim o e e .

O bs rvations on another star indicate usually that it has also

e

such a motion peculiar to itself which is wi t h ut any apparent ,

o

r lation to t h m tion f t h first star S each star is in

e e o o e . o ,

general found t have an u n xplain d moti n peculiar to

,
o e e o

itself call d its p p m t i

,
e This prop er motion is always
ro er o on .

exce dingly small and is assum d to t a k e place along an arc

e ,
e

of a great circle of the celestial sph re and at a uniform rate e

in each case P robably neither f thes assumptions are

,

. o e

strictly true ; but the accumulated proper motion for several

c nturi s even woul d be so small that observations of the
e e

highest degr e of accu racy now obtainable would not be

e

su flfi i t to prove the path of star to b e curved or its motion

c en

to b e other than unif rm

,

When the mean position f a star for a given date is to b e

.

derived from the results o f many observations at various

times in the past by the process indicated briefly in 3 7 an ,

unknown annual proper motion in declination and another in ,

 45 . P R OP E R M O TION .
49

right ascension are introduced into the least square adj ust ,

m nt The annual pr per motions in declination and in

e . o

right ascension as thus derived are then used in deducing the

place of the s t ar at any futu re da t e in the manner indicated
i n formul d 4
a 4 F a full discus an 1 ,
2 . or

si n of t h tr at m n t of pr p r m ti n fr m the astronomer s
o e e e o e o o , o
’

poin t f view with the r finements n cessary when r ductions

o ,
e e e

are t be mad c v ring very long periods of tim see e o e e

D oolittle s P ractical A str nomy pp 57 8 58 3 and Chauv

o ,

’
o — e
, .
,

net s A str nomy l pp 6 6 3

’
o , vo . I . . 2 0— 2 .

A c ncret idea of t h magnitud of h proper m tion

o e e e t e o

usually found may b gained from the fact t hat in the B oss e

C atalogu of 5 S tars for the p ch

e th r are only
00 e o e e

8 stars out of t h 5 for which the annual proper m tion i e 00 n

I n 3 6 7 cases it is less than

o

declina t ion exc ds ee 10

P roper mo t i ns in right asc nsi n are f t h sam ord r of

.

o e o o e e e

magnitud — k ping in mind of c u rs tha t of right

e ,
ee , o e, 1
°

ascensi n r presents for a star near ei t her pole a much

o e , ,

smaller displac men t up n t h c les t ial sph re than for a e o e e e 1

8

star near the q uator e .

45 Tha t t h -call d pr per motion is not really due t

. e so e o o

an errone us de t rmina t i n of the precessi n is put in i

o e o o , ev ~

dence by the fact t ha t t h various proper motions for diff rent e e

stars do t sh w t h systematic r lation which they mus t

no o e e

necessarily have if d to a shifting f the ref rence circles ue o e

P recessi n does not chang the l t i positions of t h stars

.

o e re a ve e

P r p r mo t i ns do T what are t h pr p r moti s d

.

o e o o e o

I f t hey due to a moti n of the solar system as

. e on ue

Is t . a re o a

whole through space t h stars which are ahead in the d i e

tion of motion must seem to b e separating in all dir c t ions

, re c

fr m the point toward which we are m ving must seem t be

o o , o

g ing backward at the sides and apparently closing together

o ,

behind — j ust as points of the landscape seem to a trave ller

u s,
50 G E OD E TI C A S TR ON OM Y .
§ 46 .

to mov d I f the proper motions are due to actual motions

e . 2 .

of t h stars th mselves acting as entirely independent

e e

b odies t h proper motions should seem to be without any

,

relation to each other 3 d I f on the other hand th y are

,

. .
, , e

d u e t actual mo t i ns of the s t ars which

o not however o , a re , ,

independ nt one would exp ct to fi d laws connecting the

e , e n

proper motions — laws however which w uld di ffer fro m , , , o

those called for by the last supposition ab v A cl se study o e . o

of the proper moti ns se ms to indicate that th re is some o e e

truth in each of t h three supp sitions e o .

C ompu t ations based upon hundreds of observed proper

motions made b y different astron mers at various times have
,
o ,

all agreed in a gen ral way in indicating th at there is a slo w

,
e ,

motion of the solar system as a whole through space t ward o

a point in the neighborho d of 7 6 F o a 1

h
,
z or

d e t ails in r gard to t h computations see Ch

e t A e a u ve n e
’

pp 7 3 7 8 I n regard to the third sup

, s s

t
ro n o my l , vo . I
. . 0 — 0 .

position it may b n t d that in a f w rar cases of double

,
e o e e e

stars t w stars apparently v ry near to each th r the

,
o e o e ,

obs rved proper motions indicat that t h two revolve about

e e e

som common cen t r

e link d together by gravitation e— a re e .

B t th ugh some laws connecting the vari us proper m tions

u o o o

hav been thus discover d t h salient fact to k ep in mind is

e e ,
e e

that the s cond supposition is very large ly tru -that the

e e ,

discovered la ws only account f an extremely small fraction or

of t h actually bs rved proper motions

e o e .

ct i fr m M t App t Pl
Re du on o ea n o a re n a ce .

46 T reduce from the mean plac at the beginning of

. o e

the year to the apparent place at a given date it is necessary ,

to reduce the mean place up to date and then apply to that ,

result the e ffect of nutation and aberration at that date .

§ 4 7 . C M EAN 5 To A PPA RE N T PL A E . 1

This computation if mad e directly from the known laws of

n utation and aberration is very la b orious
,

‘ *
.
,

B t such a direct computation is not necessary

u This is .

a gain one o f the cases in which it is advisabl for the engineer e

simply to acc pt the results of the astronomer s investigations

e
’

in the convenient form in which they are given in the

Ephemeris without going through all the details of the
,

derivation of those results .

4 7 S u ffice it to say that this reduction has been put in

.

the following convenient form :

at : d o +f + ru —
l l
Tgg
sin ( 6 + ) tan ar
e
6,
sin ( 17 + se c 6, ( in time) :
a . )
h cos (1 7 + 1 a o ) sin 6, i cos 6, ( in arc ); ( 1 1
)

in which and O the required apparent right ascension

a
‘

a re

and d clination at som stat d time

e and 6 are t h mean e e ; a , , e

right ascension and declination at the beginning of that

fictitious year ; is the elapsed portion of the ficti t ious year
1

expressed in uni t s of one year ; 7 and are the annual 1 ,

u
’

proper motions in right ascension and declinati n ; and f G o , ,

H g h and i
, , q uantities
, called i d p d t t - mb
a re n e en en s ar nu e rs ,

which are functions of the time only and are given in the ,

Ephemeris for very Washingt n mean midnight during the

e o

year ( pp 8 5 9 f the v lume for. 2 —2 expressed in

2 o o r

units of a year is als given in the Eph m ris on t h same o e e e

pag s Th values of these constants may be derived for the

e . e

e xact instant at which th y are r q uired with flfi i t e e , su c en

accu racy by interp lations along chords

, o .

ml mp m
my p
Fo r an e x h ib it o f t he fo r u a fo r t h e co u t a t io n if a de t h us , se e

D o o it t l le ’
s Pra c t ic a l A s t ro n o , . 61 0 .
52 G E OD E TIC A S TR ON OM Y . g 48 .

The compu t ation of the appar nt place of H ercule

48 . e u
,
s

at its transit at S t Louis Mo J uly 6 8 9 as proposed .

, .
, 1 ,
1 2 ,

in § 4 and partially carri d out i n § 4 3 may now be con

0 e ,

ti d as follows
nue F rom ,43 .

1 7 42
h m
2 65
°
31
’
(t o n e a re st m in u t e ) .

°
60
’
27 47

St L o u is is w e st o f W a s h in g t o n
l m m
.

St Lo u is si d e re a t i f t ra n s it ( sa )
l m ( m
. e o e a s cc

W a s h in g t o n
J ly )
s id e re a ti t o n e a re s t in u t e )
l m vl m md
e

y pl b w d
Si d e re a t i gh f i i t ( t end f t he ci i d a y, 6
m l im
e o ean n a o u 1

p m p J ly
b in t e r gv t io n si t f i
m )
o a e t een e re a es o ean n oo n a s en

in E h ( e ris , 3 81 fo r 6 d 7 t o n e a re s t in u t e
l v l b f W h g m d gh
e u 1 a n 1 1
.
, .
9 42
H d
en ce the si in t e r in t o n i ni t fo r t h e

d m
e re a a e o re as

v l w h y p p pl
st a te ti e is 1 07

h T is in t e r a is , it fi c ie n t a c c u ra c
su f fo r t h e ur o se o f in t e r o a t io n

- mb
h m

:
1 0
o f the st a r n u e rs , da y

p h m p g v d l y f ll w g v l
.

24

Th e E e e ris , 2 89, i es ire c t t he o o in a ue s

J ly W h g m m d gh
.

u 1 5, as in to n ea n i n i t

7 f G H lo g g log h lo g z
'

° ’ ° ’
26
'
I oo 3 15 1 57 43

J ly
.

u 1 6, W a s h in g t o n m ean md i n i ght
° ' ° ’
56
°
1 . oo 3 1 5 41 1 49

The signs attach d to log g log h log i in the Ephem ri e , , e s

are the signs of g h and i and not signs applying to th ir , , , e

logarithms as might naturally be supposed from the way in

,

which they are printed .

F the t t d t im
or day before Washington means a e
'

e,

midnight of J uly 6 t h following the order indicated b y l ,

formula
h m
i 7 42

f
(u no t b e in g g iv e n ) 131:

I
lo g TK

lo g g
§ 48 . ME AN TO A PPA R E N T PL A C E .
53
°

(G ) lo g s in (G )
°
0 3 5 1 a re 22 1 d o

lo g t a n 6 0

lo g Tg g sin l
(G (t o ) ta n 60

1 3
1
g s in (0 (1 0 ) ta n 60

lo g 3 1;

lo g h

H 1 56
°

( 17 + a , ) 62
°
lo g s in (H +a . ) 9-
9 475
lo g se c 60

lo g fgh s in ( H d o ) se c 60
i h s in
Ts ( H +a o ) se c 60

t 1 6 h
10
m
I4
°
I St Lo u is S i d e r a Ti e u 1 6
e l m J ly 1 892 z 1
h
7 42
m

mp ll w
a , a . . . , ,

Th e co uta t io n fo r 6 ,
f o o i g t he o de r f (1 1 )
n r o is

[u
’
is g iv e n 0 .
76
lo g g

lo g c o s (6 a o )

lo g g cos (0 m)

g co s (G (1 0 )
lo g 11

lo g c o s ( 17 + (to )
lo g s in 60

lo g h co s + ( t ) s in
(H o 60 0 63 84
h c o s ( 17 + 6 ) s in 0 50 4 -
35

lo g i

lo g c o s 6 0 1:

lo g t cos 60

i co s 60 3 04

m J ly
.

6 a t 16
“
1o m
St . L o u is S i d e re a l Ti e , u 1 6 ,
1 89 2 27
°
47
’
02 .
46

The above example sh ws how far out the computation o

needs to b e carried Where many star places are to b e com

b
.

t d the
e computation is materially shortened y sing
printed b lank forms so arranged as to facilitate the work
p u , u

.
54 G E OD E TIC A S TR ON OM Y 49 .

Especially convenient forms of that nature are in use i the n

Coast and G eodetic S urvey .

49 I n computing a n umber of stars on a single night

.

which will usually be the case in dealing with latitudes

observed with a zenith telescope — considerable time will be
s ved at an exceedingly small sacrifice of ccuracy by the
a a

f ollowing procedure F irst interpolate the values of the

.

independent star-numbers for every whole hour from Wash

i gt
n mean midnight for the period over whic h the b
on o e rv a

tion ext nds Then for ach star use the interpolated value
e . e

of each star-number for the nearest hour as interpolated ,

instead of making a special interpolation for each .

F an account f t h method of computatio of the

or o e n

independent star-numbers and t h method of computing sta , e r

places by the use of the B esselian star-numbers see D oolittle s ’

P ractical A stronomy pp 6 9 6 7 ; Ch
,

A stronomy
’
t , . 0 — 1 a uven e s ,

vo l I pp 6 4 5 6 5 ; and the Ephemeris pp

. . .
— 1 8 8 4 ( of t h , . 2 0— 2 e

volume for The B esselian star-numbers are not ordi

marily so convenie nt for the engineer as the independent star

num b ers .

If one has a great number of star places to compute

nder certain conditions the work may be a b ridged somewhat
,

b y using di fferential and graphic methods F the details

u ,

or

of a di fferential method which red uces the la b or of computa

.

tio about one -half in case the place of each s t ar is to be

n

com pu t ed on t h more nights see Coast and G eodeti

re e o r '

, c

Survey R eport 8 8 8 pp 4 6 5 4 7 1 A somewhat simila — 0 r

method to be used when the places are to b e computed for a

, , . .

f w stars on many nights will be found in the C oast and

P art I I pp 73 7 5 F
e

G eodetic S urvey Report for 8 9 1 2 , , .

— . or

a graphic method o f reducing from the mean to the apparen t

place i n declination see Coast and G eodetic S urvey Report
, ,

1 8 9 5 pp 3 7 —3 8
. . 1 0 .
56 G E OD E TIC A S TR ON OM Y .
50 .

What was the latitude of the station T he apparent

declination of J upiter at its meridian transit at Washington is
given in the Ephemeris (p 4 4) as follows J uly 5th 7 . 0 : 1
°

56
’
J uly 6 th 7 57 and J uly 7th 7 59 1
° ’
1
° '

3 2 . o . A ns . 2 3 .
5 .

What was the right ascension and declination of the

5 .

M oon at 8 3 9 P M l cal mean time t Cornell A pril 8 h

0
m
0
”
o a

I n the
. .
,

C ornell is 5 5 56 west of G re nwich “ m

89
8
1 2 e .

Ephemeris ( p 6 ) t h position of the Moon is giv n for . 2 e e

M G r enwich mean time A pril 9 t h

8 h m
2 A . .
33
e , a 1 1

3 4 ; di fference for minute in ’

right 02 . 2 I

asc nsi n e differ nce for minute in d li

o e 1 ec

nation 3 9 The di ff rences for minute inI righ t . 1 2 . e 1

ascensi n and declination respectively at

o M are I
“
A . .

and
B y interp lation along a ta g nt
h m
A ns .
33 o n e a 1 1
° , ”
6 7 39 I7 .
7 .

By t in erp lation alo g a parabola

o n a 1 1
h
33
m

° , ”
6 7 39 I7 . 6 .

What was the appar nt right ascension of the star 7L

6 e

A q uarii at transit at Mount H amilton C l S ept

.

89 , a . 1 , 1 2

F orupp r transit at Washington ( Ephemeris p 3 6 ) on

e , . 2

A ugust 7 t h 4 6 and dt
—
l
2 per ten a 22
h “

da
d ays lso for S ept 6 t h
. A and d t . a z 22
h

p ten days The longitude

er of Mount H amilton is .

2
“
58 west ofm
Washington 22
8
.

B y interpolation along a chord

h m
A ns .
47 a 22

B y interpolation along a tangent from S ept 6 t h

h
. a 22
m
47
By interpolation along a para b ola a z : 22
h
47
m

7 . F star B A C 57 6 m
or 6 0 g, s 1
h
50
111
46
°
 50 . Q UE S TION S A N D E XA M PL E S .
57

44 3
’
1 . 22 . I ts annual varia t ion in right ascension for that
da t e 1 and in declination ( including proper motion )
,

What was its mean declination for

° ’
A ns .
46 42
8 star 7 G eminorum
. Fo r 7 t he 7 a , 8 75
6 h
0
m

s per year annual pre

a

?
" ° ’
3 7 8 ,
22 2 2 . 1 ,

cession in declination 64 5 annual proper motion in 0 . 1 ,

d eclination What is its mean declination for

° ’
A ns . 22 3 2

9 the star
. B A C Fo r
744 9 39 0, “ 1 892
2 1
h
1
m 8
,

4
°
01 and annual
’
proper
1 motion in declination
0 68 . F0 th star B A C 74 8
. or 5 4 e 2 0 48 2
9
2 1 h
2
m
1
8
,

66
°
6 20and annual proper motion
’
1 . 00 , What
w the apparent declination of each of these stars at transit
as

on A ugus t 9 t h and 6 th at S B ernardino Ranch A rizona 1 an , ,

2
11
9 west
0 of Washington
m
The Ephemeris ( pp 8 9 9 ) . 2 , 2 0

g ives the following data for Washington mean midnight

A ug 9 . . Aug . 1 0. A ug . 1 6 . Aug . 1 7 .

° ’ ° ’ ° ’ ° ’
G 319 33 319 29 3 20 38 3 20 54
° °
H
° ' ’ ’ ° ’
1 34 38 1 33 30 1 27 38 1 26 39
lo g g
lo g h
lo g i v

The sidereal time of mean midnight at Washing t on on

Au g 9 t
. h was 7 and on A g 6 th 44 2 1
h
1
m
,
u . 1 , 2 1
h m
.

B A C 744 0 , A u g 9 t h , 6
°
A ns
’
. .
4 01 03 .
43 .

A ug 1 6 6 . th , 4
°
01
’
02 .
75 .

B A C 74 8 2 , A u g 9 t h , 6
°
66
’
. 20 I8 . 63 .

th A ug 1 6 , 6
66 6 .
°
20
’
2 1 . 2 .

10 . J ustify the half s q uare in the last term of formula

That is show that (9) is e x act if d i —
‘
’
d o m
42 .
, s 0 .
58 G E OD E TIC A S TR ON OM Y
.
50 .

da m
1 1 how from formula
. S g4 that dt
,cannot
1 , b e

n egative for any star u nless its declination is quite larg e

( near What other condition must also be fulfilled

12 D raw and explain the diagram ca lled for in t h
. e

parenthesis in 3 8 showing the geometrical relation betwee

, n

the aberration velocity of light and velocity and direction

, ,

of motion f the bservero o .

1 3 Look
. in the Ephemeris and see whether as giv n 1 e

with the ind pendent star -numbers for Jan of t h current

e . 0 e

year is z ro If not why not ? A t what time during this

e

year was it e x actly zero

.
,
 TH E SE X TA N T .
59

C H PTA ER III .

TH E S E XT A N T .

51 sextant is an instrument for measuring angl s ;

Th e
especially useful at sea and on exploratory su rveys b ecause of
. e

its lightness and portability and because it requires fixed , no

supp rt It also commends itself in c rtain other cases

o . e

b caus results of a su ffi cient degree of accuracy can be

e e

obtain d with it more conveni ntly than with larger i t

e e n s ru

ments with fixed supports which might oth rwise be employed e

I t is principally used i n astronomy for the determination of

.

local time and of latitude at sea and on explorations by land , .

It is also used x t nsively in hydrographic surveying for the

e e

measurement of horizontal angles serving to locate s und o

ings With it an angl may be measured from the d ck of a

. e e

r lling vessel wh re an engineer s transit or a theodolite

o e
’

would b u navailabl e e .

D sc ipt i f th S t et r on o e ex an .

52 A view of a sextant is shown in F ig 3

. The main . .

frame A B C carries a graduated arc D E of which the center , ,

is at F ; it carries a bearing at F which receives the axis ( p er

pen d i l to the
cu ar plane of t h fram ) about which t h arm e e e

G F swings ; a ring secured to the fram at H into which the e

telescope I is screwed and held in a fixed position relatively

to the frame ; a plan m rror at I called the h i -gl e i , or z o n a ss ,

which is fix d to the frame in a plane perpendicular to it ; and

e

c rtain c lored glasses at K and L which may b e used to

e o ,
60 C E OD E TIC A S TR ON OM Y 553 .

abso b some of t h light when obs rving the S ( the

r e e un or

M on with a star) so that the imag s s en in t h t lesc pe

o , e e e e o

may not be too bright for comfort N is the handl by which . e

the s xtant is held in the observer s right hand The arc

e
’
.

D E is graduated to fi -minute spaces but has the gradua ve ,

tion marked upon it if h p w T N m i t Th as ea c s a ce e re E n u es . e

arm G F carri s a vernier at G which reads against the arc

e ,

D E to five sec nds ( real ) I t is marked how ver as if it

o .
, e ,

read t t o d A y r ading
en sec onthe arc D E made by
s . n e on

m eans f t h vernier indicates th r fore twice the angle

o e , e e ,

b tween that position of the arm G F and the position c rre

e o

s
p d i g
on to the
n zero r ading M is a sma ll glass used in e .

r ading the v rnier

e e .

Th arm G F also carri s at F a plane mirror called the

e e ,

i d -gl
n ex which is perpendicular to the plane of t h frame
a ss , e

for any posi t ion f the arm The horizon -glass ] has only
'

o .

that half of its surface which is nearest the s xtan t frame e

silvered Th other half is m rely a plane clear glass or is

e e

cut away entirely The telescope I may b e adj ust d t

.
,

. e o

such a distance from the s xtan t frame that the edge of the e

silvering of mirror ] is in the axis of the telescope produced \

Th observer thus sees at the same time both the images

e

reflected from the silvered surface and wha t ever may be in

the line of sight of the telescop beyond the mirror e .

t t Th e P i cipl
r n e of t h e Se x a n .

53 The principle underlying sextant o b servations is indi

S uppose the sextant to b e in perfect
.

t d by F igs 4 and 5
ca e . .

adjustm nt Let OP F ig 4 be a ray of light from a dis

e .
, .
, ,

tant obj ect which passes through the unsilvered p rtion of

,
o

mirror j without change of directi n into t h tel scope 1

,
o , e e ,

parallel t its axis Let QR be a ray of light parallel to OP

o .
, ,

which strikes the index -glass F The positions of F relative to .

 53 . C P 6 TH E SE X TA N T PR IN I LE . 1

the arm F G and of ] r lative to the t l scop are such tha t

,
e e e e ,

if the reading f the arc taken from verni r G is zero the rayo e

Q R w ill be reflected from F l g the lin F] and reflected a on e ,

again from the silver d p rtion ff l g a line parallel t e o o a on o

OP into the t el p The imag of the object from which

e SC O e . e

Q R came will th refor be seen in the telescope in coincid nc

e e e e

wi t h t h im g of t h object from which OP came T

e '

a e e . o

secur the above result mirrors F d f m t for this zero

e an us ,

positi n of t h arm be parall l and t h perp ndicu lars to t h

o e ,
e ,
e e e

mirr rs F S and j T must bisect the angl s QF/ and Ff P

o , , e

N t that t h arc r ading is the sam as t h angle (z r in

o e e e e e e o

this case) b tw n the t w rays of light QR and OP which

e ee o

eventually ent r t h t lescope as parall l rays e e e e

In F ig 5 l t OP be as befor ; b t l t QR b e a ray of
.

’ ’
e e u e

light at an angl fl wi t h t h ray OF ( wi t h h ray QR f

.

e e or t e o

t h preceding paragraph ) and s t riking the index -

e glass Evi .

t ly QR will not b refl cted t f from F unless F i first

’ ’
den e e o s

r tated t hr ugh the angle 5by moving the arm F G to the

o o

positi n F G In this position of the mirror F the p p

o
’
. er en

F S will bisec t the angl QF] The angle G F G will ’ ’

di l
’
cu a r e

fl
.

be 2 but on account of the peculiar graduation of the arc as

’

indicated in 5 the reading of the arc will be fl The ray 2 .

F] will evidently be reflect d into t h t lesc pe as bef re e e e o , o ,

along a line parallel to OP and t h obj ect from which QR , e

' ’

came will b seen in the tel scope apparently i n c incid nce

e e o e

wi t h t h objec t from which OP came No t that here as

e . e ,

b fore t h r ading of t h arc is the angle b tween the t w

e ,
e e e e o

rays of light QR and OF ’ ’

.

S for any cas

0 if the sextant is in perfect adj us tm nt e , e ,

the reading of the arc is the angle between two rays f light o

one coming to the inde x -glass and the other through the
,

un
62 G E OD E TIC A S TR O N O M Y .
5 55 .

silvered portio of the horizon -glass which finally reach the

n ,

t l sc pe as parallel rays and produce coinciden t images

e e o .

When the images of t w celestial obj ects or any two objects o

at a great dis ance from t h o b server are made to appar ntly

,

t e , e

coincide in the telescop the reading is the angl at the e a re e

eye between the two bjects measu red i n the plane defined o ,

by t h two obj ects and the eye This plane may happen to
e .

b in a h rizont al a v rtical or an obli q ue position

e o I f the
,
e , .

t w obj ects observed are at a comparatively short distance it

o

may be necessary t take accou nt of the fact that the angle o

indicated by t h arc is t h ang le betw en the two obj ects

e e e

f m t h p i t U i which QR produc d intersects OP and

ro e o n n
’ ’
e

at t h y The di fference b etween these t w

,

n ot the angl e e e e . o

angles is called the t t p ll sex an a ra ax .

A d j st m ts f t h S t t u en o e ex an .

54 T o m a he t he in dex -
g la ss p e rp e n d ic u la r t o t he p la n e of
P lace the
.

t he v rnier near the middl of the arc

ta n t
*—
e e

H old the instrument with the arc away from you and look
sex . .

o b liquely into the index glass in such a way as to

,

- a por se e

tion of the arc both direc t ly and by reflection at the same

time I f the direct and r flected p rtions appear to form one e o

continuous arc the adj ustment is perfect I f not the inclina

.

.
,

tion of the glass to t h plane of t h sextant must be changed e e

b y whatever means have been provided on that p t i l ar cu a i

r

instrument This adj ustment once car fully made will not
l

. e

r q uire fre q uent attenti n ; f this reason some makers do

e o or

not provide a convenient m ans of making it e .

55 To m he t h e ho riz
a on -
g la ss p e rp e n d ic u la r t o t he p la n e of
H aving first adj usted the ind
.

th e se x ta n t —
. e x -glass point the
,

m pl
wh wh h ly
By pla n e
f
o t he Sex ta n t is ea n t the an e o f the g ra d u a t e d a rc ,

to ic h t he x b
is out ic t he a rm ro t a t e s i s n e c e s sa ri d ic
l
a a e
p rpe n
u ar .
64 C E OD E TIC A S TR ON OM Y .
57 .

This adj ustm nt may also be made as follows Place the e :

sextant face upward on a table or oth r firm horizontal sup e

port Sight across i the plane of the graduation or

. n , a

parall l plane and mark a point at that height and in lin

e , e

with t h telescope upon a wall distant fifte n fe t more

e , e e or .

Measure u pward from this point a distance e q ual to the

measur d distance at the sextant from the sight plane j us t
e

used t the axis of t h telescope and mark this second point

o e , .

The ring carrying the telescope must w be moved if no , n eces

sary in such a w y as to change t h inclination of t h t l

,
a e e e e ‘

sc pe until this sec nd p in t is xactly i n the centr of t h

o o o e e e

fi eld f the t l sc p
o e e o e .

I t should b noted that when this adj ustment has b en e e

accurately made a contact made upon one f the side lines o

will not necessarily be perfect when shifted to the middle of

the field Th reading f the arc is light ly less for a contact
. e o s

made at the middle of t h field than for one mad on it h e e e er

side when all adj ustments are perfect .

*

This adj ustment will usually remain sensibly perfect for a

long peri d o .

57 T m h t h i d o m — The reading of the

a e e n ex e rro r ge r

arc when the direct and r flected images of the sam point 1 ” e e

are made to coincide is called the i d of t h sextant n ex e rro r e .

The negative of t h index rror is the i d ti which e e n ex co rre c on ,

evide tly m ust be applied to every reading To make the

n .

ind x err r zero the horizon -glass may be rotated about a lin
e o e

perpendicular to t h plane of the sextant S crews for pro e .

d i g this rota t ion are

uc n ft n though not always provided o e , .
,

l m y m
p ll l m p pl
Fo r de ta i e d st a t e ent o f t he t he f t he e rro rs a ris in g f ro
l
a or o n on

my pp
a ra e is o f te e sc o e to the ane o f t he se xta n t , se e Cha u v e n e t
’
s

p
A s t ro n o , vo l
. 11, . 1 1 2 —1 1 4 .

p ll x
J
Pro v id e d f ha t t he t o in t is d is t t t ha t t he xt
l
( , o c o u rs e , so an se a n t

a ra a 53 ) m ay be n eg ecte d .
 58 . C T ON F OR R N D IRE65 I S OB S E VI O .

by the sextant -maker S ince t his adj ustment cannot rdi o

ily b e depended upon t

.

n ar remain perfect ven for a day it o e ,

is advisabl to determine the ind x error at the time of each

e e

series f observations and apply the derived correction

o , ,

instead of trying to keep the value of the err r down t z ro o o e

by adj us t m nt Wh n this procedu re is adopted it is only

e . e

nec ssary to make the adj ustment at rare intervals when the
e

ind x rror has become inc nveniently large The method

e e o .

of determining the index error will be found in 6 2 .

Dire c ti s
s rv i g t h S s A ltit d w it h S t t t
on for Ob e n e un
’
u e a ex an o

D t r m in the L l Tim e e e oc a e .

58 The altitude of t h S
. is a kn wn function of t h e un o e

latitude of the station of observation of the declination of the

and of t h local appar nt solar tim H nce if the lati
,

Sun ,
e e e . e

tude and d clination are known and the al titude is measured

e , ,

at a giv n instant t h l cal apparent solar time may be com

e , e o

rom this the mean tim may deriv d

J

t d F l b so a r e e e

I n determining the altitud of the S at a stati n on land

pu e . .

e un o

the ar t ificial horizon mus t be used The t ifi i l h i i . a r c a or z o n s

a shallow rectangular basin filled with mercury molasses , , or

oil prot cted from t h wind by a roof consisting of two piece

,
e e s

of plat glass h ld in a sui t able mounting Thes glass plates

e e . e

each have fac s which are as nearly as possibl plane and

e e

parallel so that rays of light may pass through them withou t

,

u ne q ual change of direction .

L t M N F ig 6 repres nt the surfac of the mercury i

e , .
, e e n

t h artificial horizon
e M N is n c ssarily a horiz ntal surface. e e o ,

that is a surface which is perpendicular at every point to t h

, e

acti n lin f gravity at that point A ray of light SB from

o e o .

th S e will b r flected along a line B A in the same vertical

un e e

plane with SB and such that the angle N B A is e q ual to the

,

angle M B S A obs rver at A will see the reflected imag

. n e e
66 G E OD E TIC A S TR ON OM Y . g 58 .

along the line A S H e may also see the S directly along un

the lin A S The distance A B being v ry small as compared

e
’
. e

with the distance to t h S S A and SB are sensibly parall l e un

’

and the angle S A S which is to b e measured with the

, e ,
’ ”
,

sextant is evidently the doubl altitude of the S

, e un .

B efore commencing the obser ations t h adjustments v e

should b e e x amined and corrected if necessary the telescope

,

should b e f lly focused t giv well -d fi d images of the

,

c a re u o e e ne

Sun and such col red glasses interposed in the path of theo

light that the two images of the S will b e of about the

,

un

same brightness and dim enough so that continued gazing at

,

them will not fatigue t h eye e .

To begin bservations place the eye in such a position

o ,

tha t an image of the S can b seen reflect d from the arti un e e

fi i l horizon
c a Without moving the eye bring the telescope
.
,

of the sextant up to it d point upon t his imag B eing , an e .

car ful to hold the plane of the sextant v rtica l swing the
e e

vernier slowly back and forth along the arc I f this is done
,

with suffi cient car a second image of the S formed by e , un ,

light coming to the telescope by way of the index -glass will ,

be seen in the telesc pe when the vernier is near the reading o

of the arc p d i g t the double

c o rre s altitude of t h
on S n o e un .

F convenienc let the two images of the S be called

or e un

the horizon imag and the index -glass image respectiv ly e e

I f t h o b server has not had su fficient experience to handle

, .

th esextant with facility it will b e well for him at this point ,

to familiariz himself with the followi g motions and their

e n

e ffects Rotate the s xtant about the telescope as an axis

. e

the horizon image will appear to remain fixed while the index
glass imag will appear to move sidewise horizontally Move
e .

the vernier slowly long the arc keeping the sextant frame a ,

and telescop fixed the index -glass imag will appear to

e : e

m ove vertically while the horizon image apparently remains

,
 59 . C T ON R R V N D IR E 6 7 I S FO OE S E I O .

fix d Rotate the sextant about a line in its plane p p

e . er en

di lcu ar to the t lesc pe at the eye end h images will e o : t e

appear to m ve sidewise without change of r lativ position

o e e .

Rotate the sextant abou t a line perpendicular t i ts plane at o

the eye and the images will appear t move v rtically with
: o e

out change of relativ posi t ion Move the y horizontally e . e e

in t h vertical plane passing thr ugh the artificial horiz n

e o o :

and t h horiz n image will appear to be cut off by a straight

e o

lin on the upper dg or lower as the y is mov d for ward

e e e, , e e e

or backward S imilarly i f t h y i moved sidewise t h

.
,
e e e .
s , e

horiz n imag will be seen par t ially cut off by the sid of the
o e e

artificial horizon on one side or the other as t h case may be , e .

Th ffects of thes vari us movem n t s of t h

e e sextant have e o e e

b n comm nted u pon because the ease and rapidity with

ee e

which one can use t h sextant dep nds larg ly upon having e e e

accurate conceptions of these e ffects as well as upon manual ,

skill To secure steadin ss of the sextant it is well in addi

. e , ,

tion to holding it by t h handle in the r ght hand to rest the e i ,

lower edge of the arc u pon the fingers and thumb of the left
hand Care must be taken h wev r t to touch the
.
,
o e , no

graduations at any tim nor to touch any part f the vernier e : o

ar m ,of the clamp and screws attached to it at the instant

or ,

wh n an observation is mad
e e .

59 The obs rv r having secured control of the images

. e e ,

let them be placed so as to be near together above the , one

oth r and approaching each other let us say F the

e , , . or

images will in general be moving re latively to each other

, , ,

since the altitud f t h S is continually changing Clamp

e o e un

the vernier in this position P ick up the b eat of the b ro

.

. o

no m t Th n watch the approaching images keeping

e er
*
e

th ir adjacent portions a b out in the middle of the field of the

.
,

Se e 5 60 .
68 C TR ON OM Y g6 G E OD E TI AS . 0 .

telescope and car fully k eping one image vertically above

,
e e

the other Note the exact time by the chronometer when

.

the two images are first tang n t to each ther O bs rve and e o . e

rec rd the corresponding reading of the vernier Unclam p

o .

the vernier and place the imag s in about the same rela t iv e e

position as before and repeat t h pr c ss until six reading ,

e o e s

say
( ) of time and the corr sponding angle hav been made e e .

Then rep at the wh le process with the di fference that w

e o , n o

the images slightly overlapped at first and allowed t

a re o

separate the instant wh n the tangency f the images takes e o

place b eing noted T comp lete the observation of the S u

,

’
. o n s

altitud it w remains to determine the index error (

e no se e

g
When a tangency is obser ed with images approaching v ,

t h noted time is too late unless the observer has accurately

e

kept the images in the sam ver t ical plane The reverse is e .

true of an observation made upon separating images To .

guard agains t an error in locating the vertical plane it is well

to continually rotate the sextant very slightly b ack and forth
,

around the t lescope as an axis so as to be certain to secur

e
~

t h first or last tangency as the cas may be

e , , , e .

60 T pick up the beat of the chronometer first look at

. o ,

s me second -mark t w s conds or m re ahead of the seconds

o o e o

hand F i t h nam of that second in mind as t h seconds

. x e e e

hand approaches it Name it xactly with the tick at which . e

the seconds hand reaches it keeping t h rhythm of the , e

chronometer beat Count it either aloud in a whisper or .

, ,

mentally I n counting it will be found easier to keep t h

. e

rhythm if the names of the numerals are elid d in such a way e

as to leave but a single staccato syllable in each The half

second beat should b e marked by the word half thus ”

one half ; two half ; three

, t w ty half ; t wenty -
, en ,
on e,

half ; twenty -t w and so on With practic an obs rve

o, . e ,
e r
 61 . D IR E CT ION S F OR OR S E R VIN O . 69

c an carry the co unt of the beat for an indefinite period with

o u t looking at the chronometer fac provided he can hear e

the tick I f he becomes expert he will ev n be a b le to carry

,

.
, e

the count for a half -minute or more during which he has not
ev n heard the tick When an observa t ion is made of the
e .

tim of tangency of two images or of any other visual event

e , ,

t h eye observes the event and the chronometer is read by

e

at the same instant I t is conducive to accuracy for the

,

c ar .

o bs rver to ac q uire the habit of deciding d finitely at once

e e , ,

without h sitation upon the second and fraction as soon as

e

he has s en the event H e who h si t ates is inaccurate

,

e . e .

The observati n of time may be made by the bserver at o o

the sextant calling tip at the ins t ant of tangency to an ,

”
,

assistant who reads the fac of t h chronome t r by eye e e e .

This is an asier process but is also a much less accurate

’

e ,

process than the one described above The n rve tim s (or . e e

intervals of time required f the nerves concerned to perform or

their functions ) and errors of j udgment of t w men inst ad

, , o e

of ent r into t h result Moreov r t h assis t ant at the

one ,
e e . e , e

chr nom t r is observing an event which comes upon him

o e e

sudd nly instead of one of which h s es t h gradual appr ach

e e e e

I f however an ordinary watch is used instead of a h m

o .

, , c ro n o

e ter it is nec ssary to let an assistant read the time both on

, e ,

account of the faintness of t h tick and b cause it is d ifli lt e , e cu

to carry by ear a beat of five ticks p second er .

61 The imag of the S

. seen in the surface of the mer
e un

cury is reversed by the reflection in such a way that the

apparent upp r edge or lim b of the image is r ally the
e , ,
*
e

i mage of t h lower limb of the S The imag of the S

e un . e .
un

received by way of the ind x -glass and horizon -glass is e

rev sed at each of the two reflec t ing surfac s and is finally
er e ,

w wh m
y y
Th e or d lim b is h e re u se d in the tec h n ic a l s e n se h it e a n s
in ic
v i si b l e d isk — o f t h e
,

t he e d ge o f t he Su n , Mo o n , o r o ther hea ven l bo d .

70 C TR ON OM Y g6
G E O D E TI AS . 2 .

seen as if no reversal had taken place I f then one makes

the lower lim b of the index glass image tangent to the upper
.
, ,

limb of the mercury image t here are r ally at t h point of , e e

tangency two c incident images of the m p i t of the S

o sa e o n un ,

namely the lower limb I f an inverting telesc pe is us d

, . o e

b t h imag s are rev rsed again in addition t

o e the reversals
e o

stated above The record of observations must be mad to

. e

show which limb of the S is used in each case Th bject un . e o

of making each complete set of observations include pointings

upon each of the two limbs is the limination of certain errors e ,

which will be commented upon later

62 To determine the index error point the telescope at
.
,

the S with the vernier set near zer and with the sextant
un ,
o ,

in such a position that a line in the plane of the sextant per

pen di l to the
cu ar telescope is horizontal Make the direct .

and refl cted images of the S tangent to each oth r with

e un e ,

t h zero of the vernier on the positive part of t h graduated

e e

arc and read the vernier Make the two images tangent to
, .

each other in the reverse position with the zero of t h verni , e er

on the negative portion of the graduated arc and read t h , e

vernier again Repeat the process two or three t im s f

. e or

gre t er accuracy A y reading on the positive portion of t h

a . n e

arc evidently gives a measure of the S u n s apparent diameter ’

.

S also does any r ading on the negative arc

o B t these t w
e . u o

measures are a ffected qually and in pposite directions b y

e O

the inde x error H ence both the inde x error and the Su n s
.
’

diameter become known .

The Sun s h i t l diameter is measured in order tha t

’
or z o n a

the results may not be a ffected by the refraction in the ver

tical plane I n making readings on the negative arc car e

must b e taken to mentally reverse the numbering of t h

.

graduations on the vernier .

Se e 6 7—
69 .
72 G E OD E TI C A S TR ON OM Y .
65
.

l
A t it u de A
Z e n it h d ist a n c e C
L a t it u d e lo g
l
z cos

D e c in a t io n 6 lo g co s

— 6
(4 ) )

10
1 2 ( 6b lo g s in m (o 6 )l= 9 8 7 56 596
.

i ll: ( cb lo g s in 1} [C (g
o

it
t h o u r-a n g l e
TA t t i l m
m
z a pp a re n s o a r e

E E q u a t io n f ti
l m
o e

m by
TM = M e a n so a r t i e

To Ti e c h ro n o m

e te r

Me a n o f s ix g iv e n
re a d in g s 1 1 04

A Tc z C h ro n o m e t e r co r

re c t io n —2 1 1

t f R d d C m p t ti
E x pla n a ion o ecor an o u a on

6 5 The a b ove o b servations were mad

.

at uniform inter e

on the sextant arc b y setting the vernier

.

vals of 20
’

each observation to that exact reading instead Of taking the ,

readings on the arc after each random pointing A rough .

method of detecting any sing le wild observation was furnished

by t h fact that t h time interval s between successive read
e e

ings must be nearly the same throughout .

The mean of the arc re dings is assumed to correspond to a

th mean of the observed chronometer times This would

e .

b strictly true if the rate of change of the S un s altitude as

’
e ,

affected by refraction were constant during the period

covered b y each half set of observations The rate of change
,

varies so little d uring this short interval that the error intro
d d is n gligible
uce e .
g66 . PA RA L L A X OF S UN .
73

The method of computing the index error has alread y

b en indicated ( 5
e

The eccentricity of this sextant had not been determined ,

but was known to be small F the method of determining . or

eccentricity see 76
H alf the corrected r ading of the sextant is the pp i
.
,

e a rox

m t altitude i th S to which must be appli d the

o e un e co r

r ctions f t h Sun s s mi -diam t er for paralla x and for

a e ,
’
e or e e e , ,

refraction as indica t ed in t h f llowing s ctions

,
e o e .

The pointings were made upon the S un s upper limb ’

B t t h p sition of t h S as giv n in the Ephemeris is f e un e or

the S un s cent r H ence t h angl subt nded at the observer

u e o
’
e . e e e

by t h S un s semi -diamet r must be subtract d to reduce the

e
’
e e

a ltitude to the value which would have been obtained had

t h observati ns be n made upon the center
e o This angular e .

s emi -diameter f the S is given in the Eph meris ( pp o un e .

3 77 3 48 of
— the volum for 8 9 ) for every day at Washingt n e 1 2

appar nt noon I t can be obtained for any other time with

o

e .
.

all needful accuracy by an in t erpolation along a chord .

P a ra lla x .

Moreover since the right ascension and d clination of

66 .
, e

the S define its position

un the celestial sphere as seen on

from the Earth s t it is necessary to reduce the observ d’

ce n e r, e

altitud to wha t it would have been had the observer placed

e

himself at t h center f the Earth d had used the samee o an

horizon as before
I n F ig 7 let S represent the position of the Sun s center
.

’
.
, ,

and let the circl B F G r present a section of the Earth madee e

by a plane passing throug h the observer at B and the centers , ,

of the S and Earth t S and C r spec t ively L t B D

un a , e . e

r present the plane of the observer s horizon L t CE be

e
’
. e

parallel to B D Then D B S is t h altitude of the S . seen e un as

74 G E OD E TIC A S TR ON O M Y . 66 .

by an observer at B E CS is the ltitude that would be . a

observed by him if h were at t h Earth s center and us d a e e

’
e

horizon plane parall l to the one he used at B Th di ffer e . e

e nce betwe n thes t w angles which is evidently equal t

e e o , o

the angle B S C is t h r ducti n required , e e o .

I n gen ral the p ll of an bject is its apparent dis

e , a ra ax o

placement due to a change in t h position of the obser er e v .

The p ll of any c l stial object is the di fference of di

a ra ax e e re c

tion of two straight lines drawn to it from two di ffer nt points

of view I t is t hen the angle t t h bj t b tw n t h two
e

.
, ,
a e o ec e ee e

straight lin s drawn to it from the two points from which it

e

is suppos d to be viewed The word p ll u nmodified

e . a ra ax , ,

will b us d in this book to indicate the di ff rence of directi n

e e e o

of a c lestial bj ec t as seen from the center of the Earth and

e o

from a station on t h surface The h i t l p ll is the e . or z o n a a ra ax

parallax for an obj ct which is i t h h i f t h b e n e or z on o e o se rv e r .

The q t i l h i t l p ll is the parallax of a c lestial

e ua or a or z on a a ra ax e

obj ct seen in the horizon by an obs rver at a station t h

e e on e
’
E a rt h s e qu a tor

I n F ig 7 if S represents a position of the S in the

.
,
’
un

horizon of the bs rver at B the angle B S C is the horizontal

o e ,
’

parallax f t h S I t is the angle subtended at the S by

e un un

the radius B C of the Earth I f p is the horizontal pa rallax

o .

.
,

of the S in seconds of arc is the radius of the Earth and

un , r ,

d is the distance between the centers of the Earth and S

,
un ,

then
about 9
’
7
p ,,
d sin 1

The e x act value of the equatorial horizontal parallax of

the S is given in the Ephemeris at intervals of ten days
un

( p 7 8 of the
2 volume for The di f
f erent radii of the
Earth are so nearly e q ual that the Sun s horizontal paralla x
.

’
§ 67 . RE PR A C TION .
75

for a station anywhere on the surface will not di ffer from that
for an e q uatorial stati n by more than —
q
o uanti t y a

which may b e disregard d for our present purpose e .

R turning to the figure l t B H be drawn perpendicular e

to CS I f p is the parallax of the S at any positi n S

e ,

. un o , ,

above the h rizon th n keeping in mind that p and p are

o , e , ,,

very small angles ,

p ptz H z B C,

p :
p ,, c o s A .
( 1 3)

The table in 9 3 a b ridged from C onnaissances

2 des,

Temps serves to give the parallax of the S for any date

,
un

and altitude The distance of the S is so nearly the same un

f the same date in di fferent years that the ta b le may be used

.

or

for any year for several centu ries t come o .

r ct i Ref a on .

6 7 The path of a ray of light from any celestial o b j ect to

an o b server upon the Earth s surface is to t h best of our
.

, e

knowledge a straight line until the ray ent rs the Earth s

, e
’

atmosphere F rom that point onward the ray encounters at

.

each successive element of its path a stratum of air which is

mor d nse than the stratum left behind since t h d nsity of
e e , e e

the air con t inually increases from the top downward with the
increasing pressu r due to w ight of the superincumbent
e e

strata The ray is therefore continually being refracted or

.
, , ,

b nt out f the straight line and this portion of its path is

e , o ,

a curve .

The two general laws of refraction are That when a ray :

passes from a rarer to a denser medium it is refracted toward

the normal t the separating surface b y an amount which is a
o
7 6 G E OD E TIC A S TR ON O M Y 67 .

function of the angle b etween the ray and the normal and of
t h densities o f the two media ; and that a plane containing
,

the ormal and the original ray also contains the refracted
n

ray The refraction is reversed in passing from a denser to

.

a rarer medium .

Let F ig 8 represent a portion of a section of the Earth

.

and its atmosphere made by a plane passing through the

c nter of the Earth and the straight portion of a ray of ligh t
e

from the celestial obj ct 0 to the poi t of incidence of the

e n , a ,

ray with the Earth s atmosphere A t the ray is r fracted

’
. a e

out of the straight line 0 to a new dir ction b nearer to a e a ,

t h normal C and still in t h plane O C

e a , A t the point b at e a .

which the ray passes to a denser stratum the ray is again

b nt toward the normal b C to the new direction b The
e , , c .

ray is thus refracted at the succ ssive points b d etc e a , , c, , .

,

remaining always in the plane O C u ntil it finally reaches the a

observer at A In reali t y the path is a continuous curve

.
,

since the increase of density is continuous A observer t . n a

A sees t h obj ect in the direction A O along the tangent t

'
e a

A to the path of the ray The angle between the rig nal . o i

direction of the ray 0 and its final direction O A is called

,
a , ,
’

th t mi l f t i or for conv ni nce simply the

a s ron o ca re ra c on e e

I t sh uld be noted that the refraction as de

e , , ,

ref tira c on . o

scribed above a ffects altitudes directly always making the ,

observed altitude too great but has no ffect on azim uth ,

e .

In the above treatment it is assu m ed that the various

strata of air are horizontal at very point F a statem nt e . or e

of the extent to which the azimuth is a ffected by refraction

becaus of the er or f t h above assumption e
e r 9
o e , s e 2 1 .

Even if the l w of variation of density of t h air with the

a e

height were a simple one the computation f t h refraction ,

o e

would be a complicated process B t the laws governing t h . u e

variation of density are themse lves complicated and not ,

g 68 . R E PR A C TION .
77

t horoughly known H ence the theory of refractio is l ng . n o

and di fficult The treatment of refraction in this book will

t h refore b e limited to an xplan tion of the ta b les given in
.

e e a

94 9 7 which
2 contain
—2 the r sults
,
of the astronomer s e
’

investigations i convenien t form f t h engin er n or e e .

68 94 gives. the m f 2 t i or refraction u nder ea n re ra c on ,

*

the m d it iea n at t h sta t i n f bservation stat d at

co n o n s, e o o o , e

the head of the tabl viz pressure 76 mm in )

'

e , .
,
0 . .

and t mperature e C ( 5 F ) The mean refraction

10 is
°
.
2 0
°
”

a functi n of t h al t itude since t h refraction of a ray of light

o e , e

in passing from one medium into another is a function of the

angl b tw en the ray and t h normal t the dividing surface
e e e e o .

95 gives the2 factor C by which the mean refraction , B,

must be multipli d if t h r ading of the baromet r is not e e e e

exactly 76 mm Th argumen t f this table is the barom

0 . e o

eter reading u ncorrected for temperature but corrected if

n cessary for its ind x err r when its tempera t ur is
,

°
e C e o e 10 .

If a mercurial barom ter is used with a brass reading -scal e e

it is n cessary to apply a corr cti n to the reading to take

e e o

account of the di fferenc of expansion f the brass scale and e o

t h mercury
e This correction is usually applied directly as a
.

correc t i n to the barometer reading B t for convenience in

o . u ,

dealing wi t h refrac t ions it has been xpressed as a correcti n , e o

to the mean r fraction and is given in § 9 7 in terms f e , 2 o

th r ading f t h th rmometer which is at t ached to the

e e o e e ,

y m b g bl m f l
my p w h h g v
*
Th is b wa s bta i e d b in in the ta f ra c t io n s

l l l
ta e o n co e o ea n re

p
gv i in D o o
’
Pra c t ic a A s t ro n o 62 8 , it i in t h e
v l l l
en it t e s t a t e n

C
.
,

bl w
f
issa n c e s d e s Te m ps fo r 897, 6 58 Th e f Pro D o o it t
'

m b h bl
on n a 1 . . a ues o . e s

p d gv l k
d d hfi rs t t o t h e sa a s is a s t f t h e F re n c d
m m
ta e e re re uce e o se o ta e , an

b d p l bl w h h w
h d the in isc ri f in a t e f c o rre s in Pro Do o
l l bl
t en ea n o on a ues ta en . .

b d p d d by
it t d ’
ta is i to b e B e sse
’
ic in t u rn

l g l g
e s e sa a se u on s ta e s, e re

h y l b d by
b v c e rt a in s e rie s o fo t io n s a s re B e ss e u s in

h bl
a se u on on se r a uce ,

h h d d p d p p di h y
the f t eor o f re ra c t io n e a o ra te h im . Th e F re n c ta e s, o n the

pl
o t er h b v
an , e en u on o t er o se r a t io n s a n d u on a ffe re n t t eor

th a t of La a ce .
7 8 G E OD E TIC A S TR ON OM Y . 69 .

barom eter I n case t h barometer used is not of the kind e

ref rred to a b ove — if for example it is a mercurial b m

.

e , , , a ro e

ter with an ivory or st l scale or if it is an aneroid — the ee , ,

proper corrections ( including the temperatur correction ) must e

be applied to its readings to reduc to millimeters of mercury e

at C and then t h table of

10
°
.
9
, 5 must b e used but that e 2 ,

of 9 7 ign red 2 o .

9 6 shows the factor

2 C by which the mean refraction ,

m ust be multipli d to take into account the temperature of e

the open air at the station of observation .

To sum up the refraction R as computed from these , ,

tables is ,

in which R is the mean refraction as given in 94 and C

M 2 , B,

C Dand C are the factors giv n in

, A 9 5 97 e 2 —2 .

Th density of the air along the line f sight and there

e o ,

fore the refraction is dependent upon the pressure and ,

temperature at all points along that line B t observations .

*
u

of t emperature and pressure can be made at the station of

observation only The refraction is ex pressed as a b ove in .
, ,

terms f the pressure and temperature at the station and the

o ,

t mperature and pressure are assumed to vary with the height

e

according to certain laws which depend to a considerable

extent pon theory only u .

6 9 I t is in order here to in q uire what errors may be

.

expected in t h refractions as thus computed The values e

for R as d rived from a long series of o b s rvations extending

.

M e e ,

over several years at one observatory when compared with ,

th corresponding values derived from a similar series at

e

l d p d v y m ll x p h mid i y
m ll x h w v h m p i d i il y m d
iIt s a so e en en t to a er s a e te n t u on the u t o f t he

k h h m id i y i
ai r, — t o so s a a n e ten t , o e e r, t at n o a tte t s or nar a e to

ta e t e u t n to a ccoun t .
80 C E OD E TIC A S TR ON OM Y 70 .

evanescent clouds the large margin of uncertainty in the ,

refraction stated above seems not only reasonable but i , n

evitable I f however more direct evidenc is needed to

.
, , e

convince one it will usually be found in making any l ng

, o

series of astronomical observations in our climat F e . o r, on

some nigh t s in attempting to make an accurate pointing upon

,

a star with a t lescope it may b obs rved to be apparen t ly

e e e

o cillating irr gularly through a range of three or fou r seconds

s e

say
( ) about it m an position on account of momentary
s e

chang s in refractione .

t f F m l Deriv a ion o or u a.

70 T h t hree corr ctions for

e S un s semi -diameter paral e
’

la x and refraction b ing applied to the approximat altitude

.
,

,
e e

A th result is t h measured altitude A of the S un s

e e , ,
’

c nter The complement of this altitude or the zenith dis

e .
,

ta ce C f the S un s center is a side of the spherical triangl

n , ,
o
’

, e

( g 9)
F i S .
- z nith pole
- upon
un the celestial sphere
e The , . a re

z nith to pole of that triangl is the complement of the la t i

e e

tude which is supposed to b known The declination f

,
e . o

the S at the instant of observati n may be obtained fr m

un o o

the Ephemeris by interpolation along a tangent as indica t d e

in 35 T h necessary
. interpolation xtends over the inter e e

va l from t h n ar st Washingt n m an noon

e e To b t ain t his
e o e . o

interval may assum an rror for the chron m ter ( to be

one e e o e

checked later) F this purp se it is only necessary t . or o o

know the error with n one minute F xample in the i or e

c mputation i n hand it was known from previous o b serva

.
,

tions that the error of t h chronom ter on Washington mean e e

time was about The Washington mean time of observa o

n“
.

tion was then 4 and the interpolation in t erval

1 1
h
56 0
m
,
m

Th complem nt of the declination is t h third side S

e e t e , un o

pole of the sph rical triangle S -zenith -pole The thre

,
e un . e
 70. D E R I VA TION OF F ORM U L A . 81

arcs of this spherical triangle b i g k w the angle at the e n no n ,

pol which is the hou r-angle of the S may be comput d

e un e

b y spherical trigonometry
, ,

Let A B and C be the angles of any spherical triangle

,, ,, , ,

and b and t h sides opposite r spectively F rom

a ,, ,,
c, e ,
e .

sph rical trigo n om try

e e

sin ( b ) sin ( ) s s c,
sin 5A sin b sin ,
,

Ct
9

in which

In the triangle S -zenith -pole l t A be the angle at the

un e ,

pole namely the h ur-angle of t h S t Let b b the

, , o e un , .
,
e

side zenith to pole ( 9 and b the sid S to 2 0

°
c, e e un

pole 9 m0ust °
b C the zenith dis t ance of the
a, e ,

Sun Making the substitutions indicat d s q uaring b th

. e , o

members of t h e q uation and simplifying there is obtain d

e , , e

sin i fC (v sin HQ ( ct
Si n i t
cos gt ) cos 6
by the use of which the hour-angle may b computed as i e n

d i t d in
ca e 64 .

The hour -angle subtracted from the S being eas t 12

18
, un

of the meridian is the local apparent solar tim T The

, e , A .

e q uation of time E ( may b obtain d fr m t h

, se e e e o e

Ephem ris with fli i t accuracy by an in t rpolation al n g

e su c en e o

a ch rd from the near st Washing t n pp t no n T E

o e o a a re n o .
A

is t h local mean solar time T A T t h correcti n t the

e , M .
C, e o o

chronometer to giv local mean tim is vid ntly t h di ff r

e e , e e e e

ence between T the reading of the chron m t r and T

C, o e e , M .

It should b kept definitely in mind that A T is strictly t h

e C e
82 G E OD E TI C A S TR ON O M Y .
72 .

correction to the reading of the chronometer only at the one

instant when t h reading of the chronometer is T A t any
e C.

other instant t h chronometer will have a di fferent correction

e ,

depending upon its rat e .

I f the chronom t r used in the sextant observations keeps

e e

mean tim and it is proposed to obtain from it the error of a

e

sidereal chronom ter t h two chron meters should b e com

,

e , e o

pared by the method indicated i 5 n 2 0 .

c si f E Dis us on o rrors .

7 1 The various errors which a ffect t h final result in any

. e

astronomical bservation may b grouped i n three classes

o e :

Is t t
, ex l
ern a or errors arising from conditions outside
e rrors ,

th e ins t rument and observer d i t m t l or ; 2 , n s ru en a e rro rs ,

errors due to the instrument arising from lack of perfect ,

adj ustment from imperfec t construction from instability of

, ,

the relative posi t ions of di ff erent parts etc ; 3 d b , .

, o
’

serv e r s

e rrors , or errors due directly to the inaccuracies of the

observer arising from his unavoidable errors in j udgm ent as
,

to what he sees and hears and from the fact that his n rves , e

and brain do not act instantaneously B y the phrase . e rro rs

of b
o se rv at i is meant the
onerrors arising from all these sources
combined .

E t l E x e rn a rrors .

72 F ollowing the order indicated above let us first con

.
,

sider t h errors m g from conditions outside the instrument

e a ri s

and observer .

The accuracy of a determ ination of time from observations

upon the S depends largely upon t h part of the day at
un e

which the o b servations are made Near apparent noon t h . e

altitude of the S is changing quite slowly A few hours

un .

later or earlier the change of altitude is comparatively rapid .

 7 2 . E R R OR S .
83

E id ntly the e ffect of a given error in the measured altitude

v e ,

pon the computed time will be less t h greater is the rat of e e

chang of t h altitude I t may b shown from the differen

u

e e . e

tial f rmul applicable t the spherical triangl used i n t h

o a o e e

preceding computation that this rate of change is

g rea t st when the
e S is in the prime vertical when it is un , or

n arest the prim vertical in case it does not reach it while

e e

above t h h i This condition by itself would fix the

e
‘

or z on .

m s t favorable time for observations at sunrise

o sunset or

d uring the months when the S is south of the e q uat r and un o ,

fr m thre to six hours from t h meridian d uring the

o e e

remainder of the year for nearly all points in t h United , e

S tates .

A nother condition how ver must also be c nsider d in ,

e , o e

determining the most favorable tim for observing A indi e . s

ca et d in 6 9 the uncer t ain t y in the computed refraction

incr ases rapidly as the altitud diminishes I n so far th n
,

e e .
,
e ,

as t h refraction is concerned t h nearer to apparent noon the

e ,
e

obs rvati ns are made the be t ter Taking b th conditions

e o . o

into account the m st favorable time for observing is from o

tw oto four hours from the meridian Within these limits .

and for stations in t h United Stat s (excluding A laska) the e e

rat of change of alti t ude is from 5 to 4 seconds of a per

e 1 rc

second f tim and the error in the derived chronome t er cor

o e ,

rection arising from t h uncertainty of the computed f e re ra c

t ion adopting the

, stima t e of that uncertainty as giv n in e e

6 9 may be from
, under t h most favorabl conditions e e

( latitud 4 midsumm e r
2
) t for
°
an bservation at e o o

two hours fr m t h m ridian in latitude 4 9 in midwinter or

o e e
°
,

ev n if this last observation is mad near sunrise or sunset

e 2 8
e .

my pp l
my p
Se e C h a u v e n e t
’
A s t ro n o v o l. 1 . D o o it t le ’
F ra c
l
s , . 21 3, 21 4; or s

t ic a A s t ro n o , . 223 .
84 G E OD E TI C A S TR ON OM Y .
73 .

The position of the S is so well determined that ther un e

is sensible error i n the result from that cause

no .

t t E I n s ru m e n a l rrors .

73 I f the telescope is
. t perfectly focused upon t h no e

S or if the colored glasses introduced make the images f

the S very dim or leave them t bright to b e gazed at
un , o
‘

un ,
oo

with comfort there is a tend ncy to

,
the images eith e se e er

larg r or smaller than they real ly are and so to misj udge t h

e , e

position of tang ncy This is not eliminat d by the d t m

e e e er

nation f inde x error as described in § 6 for the ff c t

. i

o ,
2 , e e

would be to incr ase ( or d creas ) both plus and minus r ad

e e e e

ings by the same amo nt and so leave the computed index u ,

error u nchang d B t it is eliminated by taking half of t h

e . u e

obs rvations on the upper limb of the S and half on the

e un

lower limb as sh wn in t h t of observa t i ns given in § 6 3

,
o e se o .

Th inclination of the ind x -glass to t h perpendicular t

e e e o

the sextant plan and the inclination of t h axis f t h tele

e, e o e

scope to t hat plane combine to produc an error which vari,

e es

as the t ang nt of one -quarter of the m easur d angle The

e e .
*

error of adj ustment of the index -glass and of the t lescop e e

may each be made less than 5 by t h methods given i ’

e

I
n

54 5 6
,
f each
. is the maximum error introduced int o

a measured angl of is and for oth r angles in the

e 120
°
e

ratio indicated a b ove This rr r is therefore small provided . e o

the adj ustments are carefully made and frequently verified

b ut it is sensibly a c nstant a ff cting t h mean of a set o f
,

o e e

o b servations made at n arly the same r ading of the arc If e e .

the telescope is parallel to the plane of the sextant but in , ,

observing the contacts are made with the images out of the
,

center of the fi ld the sight line is inclined to the plane of the

e ,

Cha u v e n e t
’
s A st ro n o my , vol . 11 . p . 1 1 6 .
 73 . E R R ORS . 85

s extant and the ffect on the measured angle is the same as

,
e

i f the telescope were so inclined It is important therefore

that v ry bservation should b e made nearly in the middle
.
, ,

e e o

of t h fi ld of the telesc pe o

I f the horizon -glass is not perpendicular to the plane of

e e .

the s xtant t h error intr duced is great r the smaller the

e , e o e

angle bserved and will ordinarily b appreciabl only in the e e

d e t rminati n of the inde x error I n d termining the index

o ,

e o . e

e rr r by observing the S un s sem -diam t r the rr r in any

o
’
i e e , e o

reading from this cause will b l ss than ven if the ”

'

o ne e e 1 e

horizon -glass is inclined as much as 3 to it normal position 0

”
s

( which is about the maximum error of t his adj us t ment made

as indica t ed in This rror is liminat d from the e e e

d rived index correction for b th p si t iv and negative r ad

e , o o e e

ings are num rically t small by the same am unt oo

I f t h c nt r ab ut which t h index -arm swings does not

e o .

e e e o e

c incid with the c nt r ab u t which t h graduated arc is

o e e e o e

d scribed an rr r d t t his cc ntricity will b introduc d

e ,
e o ue o e e e e

into every reading Th magnitud of this rror wil l evi . e e e

d en t ly d pend upon t h esize f t h angl m asur d as w ll as e o e e e e e

up n oth r conditi ns S
o e 7 6 f t h m thod f deter
o . ee or e e o

mining and correctin g for ecc ntricity

, , e .

Th rrors t reat d in the last three paragraphs are func

e e e

tions f the angl measur d but are constant for a given

o e e ,

reading of the sex t ant so long as the condition of the i t n s ru

m nt r mains unchanged Their e ff ct may therefore be

e e . e

eliminated almos t wholly from t h final result in d t rmining e e e

tim by measured altitudes of the S b y making observa

e un ,

tions both in the f d ft t b t th m

ore n oo n an e rn o on ou

Th compu t ed altitude will b e t

a a a e sa e

a lt it d u e . e gr at or too oo e ,

small b the same amount in both cases if t h two altitudes

, y , e

a re equal and one computed time will be as much too late as

,

the other is too early This procedure will l lim i t the . a so e na e

86 C E OD E TIC A S TR ON OM Y .
74 .

error in the computed time arising from an error in the

assumed latitude .

Th errors arising from changes in t h l ti position of '

e e re a ve

di fferent parts of the sextant due to stress s or to changes f e o

temperature are probably small in comparison with the othe r

errors consider d under the next heading S also

e the . o a re

errors of graduation of the s xtant arc e .

The rror which may arise from either or both glasses of

e

the horizon roof being prismatic instead of plane may be

liminated by reversing the roof when half the o b servations
,

have been taken .

To avoid errors arising from the prismatic form of the

shades some instrument mak rs provide a contrivance by
,
- e

which the colored shades may be rota t ed 8 from their 1 0

°

original position ; but it is better to use c lored shade o

b etween t h eyepiece and the eye instead of the c lored

s

shades in front of the inde x and horizon glasses A shade

o

in

this position may b e of a prismatic form without vitiating t h

.

o b served results .

Ob serv er ’
s E rrors .

74 The errors which are class d as instrumental depend

. e

to a consid erable e xtent upon the care and j udgment with

which the sextant is manipulated B t aside f rom t h . u e

manipulation which is an important as well as a di ffi cult

,

portion of the observer s duty the final result also dep nds ’

, e

u po his estimat s of the positions of contact of t h t w

n e e o

images and f the chronometer times of those c ntacts

o o

H is estimate of the position of contact is subj ect to bot h

.

an id t l and a
a cc en t
a t error The accidental erro co n s a n .
*
r

wh m p
p
A c o n st a n t e r ro r is one ic h ha s the sa e ef
fe c t u on a ll t h e o b se rv a

y
t io n s o f t h e s e rie s , o r o rt io n of a s e rie s , u n d e r c n s id e ra t io n A id t l
m
o . cc en a

e rro r s a re no t c o n st a n t f ro o b se rv a t io n to o b se rv a t io n ; t h e a re a s a p t t
o
88 G E OD E TIC A S TR ON OM Y.
76 .

E rror of t h e Com pu t ed Tim e .

The mean r sul t fr m a set of observations such as

75. e o

that given in 6 3 is subject then to an accidental error of , ,

about l on an av rage arising almost entirely fro m

: : e ,

t h observ r s accidental err rs It is lso subject to an error

’
e e o . a

which is constant for the set arising from un liminated i t , e n s ru

m ntal rrors and the error of t h computed refraction ( neglect

e e e

ing for the time being the personal equation of t h observer) e .

A fair estimate of t his constant error under average field

This mak s the proba b le rr r
con

d it i seems to be d
ons : e e o

aside fr m personal e q uation I t is evident from the above

o .

estimate that incr asing the numb r of observations in a set

e e ,

or number f sets taken u nder the same circumstanc s

o e ,

diminish s the final error but lit t le ( only

e term under t h o ne e

radical abov being reduced) The constant instrumental

e .

error may how ver be almost entirely eliminated by making

, e ,

observati ns at about the same altitude in both forenoon and

o

afternoon as indicated in 7 3 There is no feasible way of .

eliminating the personal e q uation error i the field n

Th a b ove estimat
.

e f the rrors from various sources

e o e

is b elieved to be a fair one for average conditions A .

special investigation for a particular observer and set of con

d it i may show errors either somewhat smaller or much
ons

larger than those indicated .

ct i f E t i ity Corre on or cc e n r c .

76 Unless the center about which the inde x arm swings

-
coincides e x actly with the center of the graduation every
.

sextant reading will be in error by the e ffect of this ecce n

t i it y ( as noted in
r c which e ffect is di fferent for readings
taken on di fferent parts of the arc To eliminate the e ffec t .
 76 . E CCE N TR ICI T Y . 89

of eccentricity upon the s x tant readings one may proceed as e

follows
Fi t The values of angles as measured wi t h t h sex t ant
rs . e

m y b
a compared with their true values det rmin d in some
e e e

other way .

F example t h angl s be t we n certain terr strial bjec t s

or ,
e e e e o

may b m asured with the sextan t and then wi t h a g od

theod lite I n making this comparison it must be k pt in
e e o

o . e

mind that a theodoli t e as ordinarily us d measures horizontal e

and ver t ical angles while the sextan t measures directly the
,

angle between the two bjects in the plane (horizontal o ,

obli q ue or v rtic al ) passing through t h two obj ects and the

,
e e

sex t ant A lso in this cas t h sextant parallax 53 must

.
, e , e , ,

be tak n into consideration unless the objects

e very a re

dis t ant .

Th angular distance between two kn wn stars may be

e o

observed and compared with its valu as compu t ed from the e

kn wn right ascensions and declinations of the s t ars c rr cted

o , o e

for t h ff ct of refraction at the tim of observation This

e e e e .

computation will be found unfortunately t be rather , , o

laborious .

O the altitude of a known star (or of the S ) may be

r, un

m easur d at a known time at a stati n of which t h latitude

e o e

is known The true altitude of the star may b compu t ed

. e ,

and becomes comparabl after correction for refraction with e, ,

that m asured with the sextant

e .

S d F
e co n each s xtant observation which is compar d
. or e e

with a known angle an observation e q uation f the f rm o o

( I7
)

is formed in which , x , y , and are unknowns t be deter

v o

m ined and D , 0 A , am is the di fference between the true

90 G E OD E TI C A S TR ON OM Y .
577
value of the angle 0 and the measured value 6 , m . is t he

reading of the sextant corrected for index error .

0m
and K sin
4

The m st probable valu s of y and

Th ird . o e x , , v a re

d termined fr m thes observation e q uations by the m t h d

e o e e o

of least s q uar s e .

F th Th se values f y and may now be b t i

our . e o x , , v su s

t t d i
u q
e ua t i n ( 7)
n and
e a tabl f correcti ns computed
o by 1 e o o

substituting successively f (9 the corr or e

sp d i g
on comput d
n values of D being vidently t h e A e e c o rre c

ti ns f eccentricity which m ust be applied to measur d

o or e

°
e
an gles .

many and such accurate obs rva t i ns are requir d for

So e o e

a sa t isfact ry determina t ion of the eccentricity f a sextant

o o

that it will usually b found more conveni nt to liminate the e e e

e ffect of eccentricity upon time o b servations by obs rving e

both i the forenoon and aft rno n with the S at about the
n e o un

same altitude as indicated in 7 3 B t in s xtant obs rva

,
. u e e

tions for latitude a special de t erminati n of t h eccentricity o e

is nec ssary if the highest attainable degree of accuracy is

e

d sired
e .

Oth Us s f t h S t t er e o e ex an .

7 7 I n d etermining time with a sextant by the pr c din g

. e e

method the latitude is supposed t b known If conversely o e .

, ,

t h time of obser ation of t h altitud

e of the S ( a star) v e e un o r

m h d d mi i g pp l i d f
Th
wh h h g v w
is i f te r t h e c o rre c t to be a

m
e t o o e n n o n s e o r c ece n

f m l w ll
t ric it y, ic is i de
d f o r a n d it h o u t t h e d e riv a t io n
in
l l l
e re en co n n se ,

m y pp w
b e f o u d t e a t e d in f u l in D o o it t l e s Pra c t ic a
’
o f t he i
m m
or u a n r

C e rt a in re fi n e e t s t h e re g iv e n h ic h a d d u c h
,

mp y mp
A s t ro o 1 9 6-2 06

lb
n , . . n ,

t o t he a o r o f co u t a t io a n d l it t l e t o t h e a c c u ra c o f t he c o ute d

m
n

re su l t h a v e h e re b e e n o
,
it t e d .
9 2 G E OD E TIC A S TR ON O M Y . 80 .

a re of a small circle on the Earth s surface of which the pole ’

is at a point in the line joining the object and the Earth s ’

center and of which the polar distance is equal to the

,

ob erved zenith distance of t h obj ct This small circle or

s e e .
,

such a portion of it as is necessary may be plotted on a ,

sphere or chart A second such observation on an obj ct in

. e

som other azim uth s rves to locate the observer on another

e e

small circle int rsecting the first i n two poin t s Th se two

e . e

poin t s f inters cti n o usually so far apart that t here is no

e o a re

di fficul t y in discriminating betwe n them and the observer s e ,

’

p sition becomes defini t ly known This proc ss of de t er

o e . e

mining p sition at sea is known as S umner s m thod F

a o
’
e . or

a m r c mplet state m ent of this method see Ch

o e o e t a uven e
’
s

A stronomy l pp 4 4 4 8
, vo . I
. . 2 — 2 .

8 0 If for t h purp s of d termining local time an b

.
,
e o e e , o se r

vati n is taken u pon a star east of the meridian and the

o ,

observation is repeated west of the meridian at the same

reading of t h sextant t h computation of tim may b made
e ,
e e e

indep ndently of any knowledge of the index error

e e cce n

t i it y or other er ors of the se x tant and ind pendently of

,

r c ,
r ,
e

any computation of t h refraction upon the assumption that e , .

thes quantities retain the same values at the second observa

e

ti n which they h d t t h first and that therefore the two

o a a e ,

observations made at t h m ( unkn wn ) altitude D

a re e sa e o . ur

ing such an interval of a f w hours th declination of any e , e

star is for the p resent purpose sensibly constant U pon these .

assumptions it may be shown that the m an of the two e

o b serv d chronometer tim s is the chronometer time corre

e e

ond i g to the ntransit of the star across the meridian ( Let

I f the obj ect o b served is a planet
s p .

t h student prove this )

e .
,

th eMoon or the S the same method of computation mayun

b e used b ut it will b e necessary to apply a correction for the

, , ,

,
g 81 . C
M IS E L L A N E O U S .
93

change of d eclination during the interval b etween the t w o

observations .
*

The advantages of this method are the ease and simplicity

of the computati n Its disadvantag is the liability of o . e

losing the second observation on account of clouds or other

hindrances I f observati ns are taken within one hou r ( say)
. o

of t h sam hou r-angle east and west of the meridian p

e e re s ec

ti vely are c mputed as indicat d in

,
6 4 and the mean
o e ,

taken the elimina t ion of errors is almost as c mplete d

,
o , a va n

tage may be taken f t mporary br aks in t h clouds and the o e e e ,

o bserver is t put to t h inconv ni nce of being ready at

no e e e

s m particular m om nt The computation will consum a

o e e . e

littl m re t im
e o e .

8 1 Th Covarrubias method of observing developed by

. e ,

the M xican astronom r of that nam serv s to eliminate the

e e e , e

instrum n t al errors and acc mplish s t ha t purpose without

e ,
o e

the necessi t y of t h l ng wait betw en obs rvations which is e o e e

re q uired in the m thod stat d above Two stars are select d e e . e

which are s veral hours apart in right ascension and have

e ,

declinations not very di ff r n t A t a c rtain time each night e e e

which is first es t imat d roughly by t h b server these t w

.
,

e e o , o

s t ars will f an instant be t the same altitude

or east and a , one

th o t h r w st of t h m ridian A f w minutes b fore t his

e e e e e . e e

tim he observ s one of t h stars noting the chronom ter

e e e , e

time and t h s xtant r ading H then turns to the sec nd

e e e . e o

star which h finds approaching the same altitud

,
e d e , an

observ s the chron m ter time at which the sextant reading

e o e ,

and th r f re the al t itude is the same for this second star as

e e o ,

that bef r obs rved upon the first star F rom t his obs rva
o e e . e

tion of the two chronometer times at which t h two stars reach e

mp l l l
m y pp my pp
*
Fo r t h e co u t a t io n o f t h is c o rre c t io n ,
se e D o o it t e s
’
Pra c t ic a

m y pp
-2 0 1
; C h a u v e n e t s A st ro n o v o l. I.
’
A 8
m l
st ro n o , . 2 30, 23 1 , . 1 9 ; o r

Lo o is
’
Pra c t ic a A st ro n o , 1 2 6-1 30 . .
94 G E OD E TI C A S TR ON OM Y. 82 .

the same altitude the error of t h chronometer may be com e

p ted independently of any knowledge of the exact absolute

a

value of that altitude This method will sometimes be found

desirable esp cially in case the only available se x tant is an
.

,
e

inferior one or has been damaged to such an extent that its

indications are unrelia b le It is not dev l p d in detail here
,

. e o e

for lack of space F a complete stat ment of the method

. or e

see Nu vos M t d A t mi para determinas la hora

e o os s ro n o cos

F D Covarru b ias
e ,

el azimut la latitude y la longitude ,

. .
,

Mexico 8 67 ,
1 .

Q U E S T IO N S A N D E XA M PL E S .

82 . 1 . P rove
using figures if necessary that the test as
, ,

given in 54 for determ ining whe t her the index -glass is per
p d
en i l to the plane
cu a r of the sextant is valid .

Explain why two images of the same object cannot be

2 .

made to coincide in the sextant telescope if the ind x -glass is e

in perfect adj ustment but the horizon -glass is inclined to the

plane of the sextant ( Explain also h w it is possi
se e o

b le that such coincidence may be secured if b t h the index o

and horizon glasses inclined Why is it advisable to make a re .

certain of the ind x -glass adj ustment before adj usting the
e

horizon -gla s s

3 S uppose that the ind x correction of a certain sex t ant

e

is found to b e
.

Through what angle and in what

direction must the horizon -glass b e rotate d to make the cor
rec t ion zero
4 S how
. that the errors due t the inclination to the plane o

of the sextant of the sight line and of the index glass are -

not eliminated b y the process of eliminating the inde x error

indicated in 6 2 .

5 T h . radius of the graduated

e circle of a certain sextant
is 4 in What linear movement of the vernier corresponds
.
96 GE OD E TI C A S TR ON OM Y .
g 83 .

C H P TE R
A IV .

TH E A STRO N O M IC A L TRA N S I T .

83 . astronom cal transit is designed primarily to be

TH E i

used for the determination of time wi t h its tel scope in the e

plane of the meridian Its essential parts are a telescope an

.
,

axis of revolution fixed at right angles to the telesc pe a o ,

suitable support for said axis such that it shall b stable in , e

azimuth and inclinati n and a striding level with which the

o ,

inclination f the axis may be determined

o .

F ig shows an astronomica l transit which is now and

. 10

has been f several years past in use in the U S Coast and

or . .

G eodetic S urvey for time determinatio s of the highest order n

of accuracy The focal l ngth ( distance from the lines of the

. e

eyepiece diaphrag m to the optical center of the obj ective ) of

th t lescop A B is 94 cm ( 3 7
e e e The clear aperture of
.

the obj ct -glass is

e cm ( 3 and the magnifying power
.

with the diagonal eyepiece A ordinarily used is 4 10

diameters I n the focus of the eyepiece is a thin glass

, ,

diaphragm u pon which are ruled lines which serve the same
purpose as the spider lines or cross wires more commonly
placed in t h at position in a telescope Th system of lines . e

consists of two horizontal lin s near the middle of the field e ,

and thirteen vertical lines The milled head sh own at C con

.

tr ls by means of a rack and pinion the distance of the

o , ,

diaph ragm from the obj ect -glass and serves therefore to focus
the o bject -glass — in other words to b ring the image
,

, o r, ,
g 83 . CR P T N TR N TDES 97 I IO OF A SI .

formed b y t h obj ct -glass into coincidence with the dia

e e

p hrag m The diaphragm

. and the corresponding image ,

form d by the obj ect -glass are much larger than the field of
e ,

view of t h eyepiece To enable the observer t see various

e . o

parts of the diaphragm and the corr sponding portions of , e

the image t h whole eyepi ce proper is mounted upon

,
e e

a horizon t al slide controll d by the milled h ad shown at D e e .

The lines of the diaphragm seen black against a light a re

field The illumination of t h fi ld at night is obtained from

. e e

one of the lamps sh wn at E The light from t h lamp o . e

passes in through t h perforat d end of the h rizontal axis e e o ,

and is reflect d down to the eyepi c by a small mirror in the

e e e

in t rior of t h telesc p fastened to a spindle of which t h

e e o e , e

mill d head G is t h outer d The perforated disk shown

e e en .

a t F carries plain ground and color d glasses to be used by

, , e ,

the bserver to t mp r t h illumina t i n

o e e e o .

The horizontal axis F F is cm ( } in ) long ending . 20 .

,

in pivots of b ll m tal H H is the striding level in positi n

e e .
, o ,

resting upon t h piv ts of the horizontal axis

e o .

Th ir n sub -bas a p rtion of which shows at l i

e o e, o , s

cem nted firmly to t h pier The transit base is carri d by

e e . e

three fo t -screws r sting upon this sub -base The devic

o e . e

shown at j to give t h instrumen t a slow mo t ion i

s e rv e s e n

azim uth .

The lever K actuates a cam to raise the cross -pi ce L e ,

and with it the colu mns M M The horizontal axis is the . n

rais d su fficien t ly upon the forks at the upper end of M d

e an

M to clear t h V The cross -piec L is then free to turn

e s e

until arrested by the fi x ed stops and thus to revers

.

the horizontal a x is FF in t h V
, e

e s

T he setting ircles N N are cm (4 in ) in diameter

.

c 10 . .
,

are grad uated to spaces and are read to single minute

20
’
, s

by verniers T hey are set to read zenith distances

. .
9 8 TIC A S TR ON OM Y . 84 .

shows anoth r typ of transit in use in the

F ig . II e e

U S C G S Its peculiarities are the folding frame the

. . . . .
,

graduated scale at Q t facilitate putting the t lesc pe in t h o e o e

meridian and the fact that the ey piece is furnish d with a

,

, e e

movable line carried by a micrometer screw while one of the ,

setting circles carries a it i level so that the instru ment se n s ve

m y be used both as a zenith t l sc pe ( see C hapter V ) and

a e e o

a transit The screw T m . the upper base 55 in azimuth o ves

wi t h respect to the low r bas RR e e .

o y f t h T it Th e Th e r o e ra n s .

8 4 I f a transit were in perf ct adj ust m

. ent the li f e ne o

co llim t i of the telescope as defined by the mean lin of

a on e

the r ticle would be at right angles t the transv rse axis upon
e o e

which it revolves and that transverse axis would b in the , e

prime vertical and horiz ntal Under these circumstances

, o .

t h line of collimation would al ways lie in t h meridian plane

e e

Th e lin e f c o llim a t io n o f a t e e sc o l p is t ha t l
in e f si ght wh ic h ll
l ll m
o e o to a

w y
b s e rv a t io n s a re re f e rr d I g in e e r s t ra n sit t h e in e o f c o i t io n ’

l m
o e e . n an n a

h ic h a ll b se rv a t io n s a re a d e a d is d e fi e d b a
w l p
is t h e in o f s ig h t o
ll
m l l
e n o , n n .

v e rt ic a i e in t h e idd e o f t h e fi e d o f v ie o f t h e t e e sc e In a
m l m ll
n o . n

y
ic a t ra n s it t h e o b s e rv a t io s a re a d e o n t h e s e v e ra i e s o f s ig h t
ll l
a st r n o o n n

y w
d e fi d b t h e s e v e a i e s o f t h e e t ic e Th e se v a rio u s o b se rv a t io n s a re
m l ll m
ne r n r .

w y m y m l
a ll re f e re d t o a n i a gi a r l i e o f s ig h t o r in e o f c o i a t io n h ic h is
l
r n n , ,

p
d e fi n e d h e v e r b t h e e a o f a ll t h e in e s n d o t b t h e idd l e i e
m l m l
, o , n a n n .

ly m p l
Th e in e is f cou s n ea r to t he id d le in e t h e s a c in g o f t h e
l m
ean ,
o r e , ,

pl p l p
l i e s i t h e re t ic e b e i g a d a s e a r u n if o r a s o s sib e
m l l
n n n e n .

p p
I a g in e a a ss e d t h ro u g h a i o f t h e re t ic e o f a t e e sc o e

l m
ane y n e n

a n d t h ro u g h t h e c e n t e r o f t h e o b j e c t -

ly y j l ly y p w
g a ss I a g in e t h e l a n e ro d u c e d .

b e o n d t h e b e c t -g a ss E v id e n t e v e r o in t o f h ic h t h e
l p pp w
in d e fi it e
m l
n o .

pl p
i a g e is s e e in t h e t e e s c o e in a a re n t c o i c id e n c e it h t h is in e o f t h e
l m l
n n

p y b e s a id
re t ic e u s t lie in t h is an e i s ace Th e li o f t h e r t ic e m
n . n e e a

p ly
t o d e fi n e t h is pl n in S a c e o r a p i t o f t h e re t ic l e l in e m a y b e s a id t o
l l
a e ,
o n

pk
d e fi e a li e in s a c e F o r c o n v e n ie n c e a in e o f t h e re t ic e is o rd in ri
n n .
a

S o e n o f a s d e fi n in g a lin e o f s ig h t ra t h e r t h a n a pl n e o f s ig h t it b e in g

ly y
a

d e rs t o o d t h a t o n e po in t o n l o f t h e re t i c l e l i n e is re f e rre d t o
,

y m p
t a c it u n

o rd in a ril the idd l e o in t .

IOO G E OD E TIC A S TR ON OM Y . 86 .

turned up to the sky The focus for most distinct vision of

.

the reticle lines is what is required Now direct the telescop . e

to some distant obj ect ( at least a mile away and pref rably
m uch farther) and focus the o b ject -glass by changing its dis
, e

tance from the reticle so that when the eye is shifted abou t
,

in front of the eyepiece there is no apparent chang of rela e

tive position ( or parallax ) of the lines of the reticle and of the

image of the obj ect I f the eyepiece itself has been properly
focused this position of the obj ect -glass will also b e t h
.

position of most distinct vision The focus o f the o b ject -glas

,

. s

will need to be inspected again at night d corrected if an

necessary using a star as the o b j ect None but the brightes t

,

stars will b e seen at all unless the focus is nearly right

.
,

d fi d distan t o b j ect using the apparent

, .

B isect some well - e ne ,

upper part of the middle vertical line of the reticle R tat

the tel scope slightly a b out its horizontal axis until the obj ct
. o e

e e

is seen upon the apparent low r part of this same line I f t h e . e

bisection is still perfect no adj ustment is needed I f h w

ever the b isection is no longer perfect the reticle m ust b
, .
, o

rotated ab out the axis of figure of the telescope until the line
, ,

is in such a position that this test fails t discover any error o .

8 6 Now bisect the distant obj ect with the middle line of
the reticle Reverse t h telescope axis i its Y I f t h
.

e n s e

b isection still remains perfect the line of sight d efined by the

. .

m id dl line of the reticle is at right angles t

e the horizontal o

axis and t h m line may be assumed to b e su ffici ntly near

e ea n e

to that position I f necessary however make the adj ust

.
, ,

ment by m ving the reticle sidewise so as to mak the

o , e e rro r

f llim t i small B y on f llim t i is meant the e rro r o co a on

angle b tween the line of sight defined b y the mean line o f

o co a .

the reticle and a plane perpendicular to the horizontal ax is of

the telescope
Level the horizontal a x is o f the telescope A dj ust the
.

.
 87 . A D J U S TM E N TS . 101

l evel so that when it is reversed its reading will change b ut

little .

Test the finder circles t see that they have no index o

error P oint upon an o b ject and read one of the finder

.

circles Reverse the telescope point upon the object again

.
, ,

an d read the same finder circle as before The mean of these .

t w readings is evidently the zenith distance of t h obj ect ( if

o e

the circle is graduated to read zenith distances) and their half

'

difference is the index error of the circle This index error .

m y be made zero by raising or lowering one end or the other

a

o f the level attach d to the vernier of the finder circle

e This .

same process will evidently serve if the circle reads elevations

inst ad of zenith distanc s I f the circle is to b e made to read
e

d eclinations direc t ly the same process is still applica b le

e .

, .

F or if the circle be made to read zenith distances with an

index error e q ual to the latitude of the station its readings ,

will be decli nations for one position f the telesc pe (though o o

n o t f the other after reversal ) The other circl may be

or , . e

m ade to read declinations for the other position of the tele

scope
finder list of stars showing for each star to b e
.

A ,

observed its name magnitude setting of finder circle and

, , ,

the right ascension or t h chronometer time of transit to the e

nearest minute will b e fou nd to be a convenience in the night

,

work I n making out a finder list the refraction may be

.

n glected not being su ffi cient to throw a star out of the field

e ,

of the telescope The zenith distance of a star is t hen

. 6 ,

south zenith distances being reckoned as positive .

t j tm t z
Th e A im u h A d us en .

8 7 In the evening before the regula r o b servations are

.
,

comm nc d it will be necessary to put the telescope more

e e

accurately in t h meridian H aving estimated the error of

, .

e .
1 02 G E OD E TI C A S TR ON O M Y . 87 .

the chronomet r i n any availabl way within say five minut s

e e , e ,

and having carefully l velled up t h axis set the tel scope f e e , e or

some bright star which is ab ut to transit within ( say ) of o 10

°

th zenith O bs rv the chr n meter tim f transit f the

e . e e o o e o o

star This star at transit being nearly in t h z ni t h its time

. e e ,

of transit will be but little aff cted by the azimuth error f e o

the instrument Th collimation err r and l vel error have. e o e

been m d small by adj ustm nt Ther fore the di ffer nce

a e e . e e

betwe n t h right ascension of t h star and its chronomet r

e e e e

tim of transit will b a close approxima ion to the error f

e e t o

the chronome t er Now set the tel sc p for som slow . e o e e

moving star which will transit well to the nor t hward of t h e

zenith (let us say of declination gr ater than 6 and a n rth, e 0

°
o

z nith distance of more than

e Compute its chronom t r e e

time f transit using the approximat chronometer rror j ust

o , e e

obtained A that time approaches bisect the star with the

. s

middle line of the reticle and keep it bisected following the , ,

m ti n of t h star in azimuth by t h
o o eof whatever means e u se

have b e n furnish d on that particular transit for that purpose

e e

Keep the b isection perfect till the chronometer indicates that

.

the star is the meridian The telescope is w pp i

on . no a ro x

mat ly in t h meridian
e e .

The adj ustment may be te ted by repeating the process s ,

i
. eby obtaining a closer approximation to t h chronometer
.
,
e

error by obs rving another star near the zenith and the
e , n

comparing the c mput d chronometer time of transit of a o e

slow moving north rn star with t h obs rved chronomete e e e r

time of its transit I f t h star transits apparently too la t e

the object -glass is too far west (for a star a b ove the pol )
.
, , e ,

e ,

and i The slow -motion azimuth -screw may then b

v c e v ersa . e

used t reduce the azimuth error This process of reducin g

the azimuth error will b e much more rapid and certain if
o .

instead of simply guessi g at the amount of movement whic h n

 04 G E OD E TIC A S TR O N OM Y .
g 88 .

tenths f seconds it is necessary to divide the half -second

1

interval given directly by the chronometer into fifths by

some mental pr cess To secure accuracy and ease in mako .

ing this stimate it is advisable to transf rm it into a proc

e o

ess of estimating relative distances The apparent motion .

of the star image is nearly uniform ( not q ui t e so on account

of disturb ance by irregular refracti n ) Let us suppos that o . e

at the instant when the chronometer tick which indicates the

time to b 5 is h ard the star
e is se
1 n8
at A
.
( g
F i L t e e . 1 e

the observer re t ain a mental picture of this r lative position e

of the star and the line Wh n the chronom t er tick for . e e 2 8

. 0

is heard h s es the star at B I f he has re t ained ( for

,
e e .

only) the m ntal picture r ferred to above he h before his

e e as

mind s y exac t ly what is indicated in t h figur H e esti

,

'

e e e e .

mates the ratio f the d i t from A to the line and from

o s a n c es

A to B and conclud s that the ratio is n are to

,
than to g e e r r
,

or } and calls t h time of transit of t h star across line I

A
,
e e ,

Though this mental process may seem awkward at first it ,

will ultimately be found to be both easier and more accurate

than the direct process for all cases in which the star has an ,

apparent motion which is su fficiently rapid to make the dis

tance A B appreciable in a half -second A exp rienced . n e

observer using this process is able to estimate t h t ime of

, , e

transit of a star s imag across each line of the r ticle with a

’
e e

pro b able error of about i 0

5
. 1 .

It is well here to bear in min d the sugg s t ion given e

i g 6
n that h wh o hesitates is inaccurate The successful
0, e .

o bserver decides promptly but without hurry upon the , ,

second and tenth at which t h transit occu rred e .

A t convenient intervals betwe n stars the striding level e

should b e read in each of its positions upon the horiz ntal o

axis A t a b out the middle of the observations which are to

.

consti t ute a set the tel escope should be reversed so that the ,

e fl t of the error of collimation ( and inequality of pivots)

ec
 89 . CH R ON OG R PH TH E 5 A . 10

upon the apparent time of transit m y be reversed in sign a .

Each half -set should contain one slow -moving star (of large
declination) t furnish a good determination of the azimuth
o

error of the instrument .

The telegraphic longitude parties of the C ast and o

G eodetic S urv ey make the best time determinations that are

made with p rtabl instruments at present in this coun t ry
o e .

In their prac t ice ten stars are observed in each set five before ,

and five after reversal of the telescop F rom t w to four e . o

readings of the level are taken in ach of its positions in ach e e

half -set Care is taken to have the telescope at different

.

inclinations during t h di ffer nt readings of the level inclined

e e ,

s metimes to the south and sometimes to the north so that

o ,

the level may rest in turn up n various parts of the pivots o .

This is done to eliminate in part at least the e ffect of , ,

irr gularity in the figure of the pivots upon the determination

e

of the inclinat ion of the axis .

Th e Ch ron ogra ph .

The preceding directions for observing were given on

89 .

the supposition that the eye and ear m thod f observing the e o

t imes of transit is to be used I f instead the time b .

, , o se r

vation proper is made with a chronograph the method is ,

changed in that one particular only .

A common form of the chronograph is shown in F ig 3 . 1 .

The train of gear-wheels partially visible through t h back e

glass of the case at F is driven by a falling weight and drives ,

th e speed governor at A E CCD D the scre w 1 and the , ,

cylinder H A the speed of rotation of the governor

. s

increases the weights CC m ve farther from the axis until a

, o

small proj ction on one of them strikes the hook at E and

e

carries it along This hook carries with it in its rotation the

.

small weight A The result of the impact and of the added

iction at the b ase of A is to cause the speed of the governor
.

fr
to decrease until the hook E is released The speed then .

increases until the hook is engaged decr ases again un t il it i , e s

r leased and so on The total range of variation in the speed

e , .

is howev r surprisingly small — small that in int rpreting

,
e , , so e

the record of the chron graph the speed is assumed to be o

u niform during the intervals between clock br aks By e .

moving the adj usting nuts D D upon t heir scr ws t h critical e , e

speed at which the hook is engaged may be adj ust d The e .

carriage M is moved parall l to the axis of t h cylinder H by e

‘

the screw 1 The p G carri d by an arm proj ecting fr m en e

the carriage M t nds to trace a heli x at a u niform rate upon

.
, o

,
e

the paper or chronograph sheet stretched upon the

, ,
”

cylinder The magn ts XX are in an electric circuit ( through

. e

the wires L ) with a break circuit clock or chronomet r

- , e .

Whenever the circuit is broken the armature N i released and , s

the back portion of the m carrying the p is drawn back ar en ,

by a spring to contact with the stop at f The pen then

, .

makes a small o ffset from the helix I t returns to the helix .

as soon as t h current is renewed A a result the equal i

e . s n

t l of t im between the instants at which the chronometer

e rv a s e

breaks the electric circuit are indicated by qual linear i e n

t l b etween o ffsets on the line drawn by the pen the

e rv a s

sp d being kept constant b y the governor The h m

,

ee . c ro n o e

ter is usually arranged to break the circuit every second or

every alternate sec nd and to indicate the beginning of each

o ,

minute by omitting break The hours and minutes may on e .

be identified by r cording at some point upon the sheet t h

e e

corresponding reading of the face of the chronometer .

The electric circuit passing through the magnets XX t h , e

chronometer and battery also passes through a break -ci rcuit

key in the hand of the o b server To record the exact time
.

of occurrence of any phenomenon he presses the key at that

instant b reaks the circuit and produces an additional o ffs t
, , e

in the helix of which the p it i indicates accurately t he

, os on
1 08 C TR ON OM Y G E OD E TI 9 AS . 2 .

Tim f T it A M Li
e o ra n s cross ea n n e.

92 If the transit of the star across every line of the reticl

. e

i observed the time of transit across the mean line or line o f

s

collimation is evidently obt ained b y taking the m an of the

, ,

s veral o b served times I n o b taini g the sum of the several

, e

e . n

times for this purpose an error of a whole second in any one

observed time which might otherwise remain unnoticed will
be detected b y the use of the auxiliary sums shown in the
, ,

li t tle column j ust after the o b s rved tim s namely the sum e e , ,

of the first and last times of the second and last but one
third and last but t w etc These auxiliary sums should b e
, ,

o , .

nearly the sam and n arly equal to double the time on the
e e

middle line The unexpressed minute f r each is the same o

as that for the middle line The sum of these au x iliary sums
.

and of the m id d l time is the total sum required in comput

e

ing the mean .

I t will fre q uently happen especially on p rtially cloudy , a

or hazy nights that the transits of a star across several lines

of the reticle will be successfully o b served and yet the
,

o b server may fail utterly to secure the transits across the

,

remaining lines I t then becomes necessary to reduc the

. e

m ean of the observed times of transit across certain of the

lines to the mean of ll of the lines a

b e the obs rved times of transit across

.

Let t t t
,, ,, ,,
e

the successive lin s and let t be their mean or the time of

e , m ,

transit across the mean lin e .

Let 21 , 23 »be the q t i l i t l of the succes

z ,, e ua or a n erva s

sive lines fr m the mean line or the intervals of time which

o

elapse for an e q uatorial star (star of zero declination ) b etween

,

transits across the separate lines and the transit across the
mean line .

Then for an equatorial star i t t i t t l , m , , 2

i, t' tm ,
§ 9 2 . RE D U C TION TO M E A N L IN E . 09

To determine the relation for any other star b etween i

the equatorial interval for any line and t ( in which t , ,, ,,

is the time of transit over that lin ) deal with the spherical e ,

triangle defi n ed by the pole the star at the instant when it is ,

on any line and the star at the instant when it is on the mean
, ,

line I n F ig 4 let P A and B represent these points

. . 1 , , ,

respectivel y The sides PB and PA are the polar distance of

.

the star ( 9 2 The angle at P expressed in sec nds

0
°
o

of arc is 5(t t ) ( t t ) being expressed in seconds of

1 ,, m , n m

time Th side A B expressed in seconds f arc may be

. e , o ,

taken e q ual to i being expressed in seconds of time n .

Using t h law that the sines of the sides of y spherical t i

e an r

angle are pr por t ional to the sines of the opposite angles

o
,

there is obtain d e

sin 5i sin (9 1 6) sin 5(t t ) sin A ( 9)

,,
0
°
1 ,, m . 1

The arc A B corresponding to the equatorial interval for a

line will seldom e x ceed 5 in any transit and is usually much 1
’
,

less F such an is sceles spherical triangle as this A B

or o

b eing short angles A and B are n cessarily nearly e q ual to

.
,

,
e

A ssuming as an approximation that A ( 9) may 1

be written sin 5i sin s(t cos 6 A gain assuming

1 ,,
1 ,, .
,

that the small angles and 5(t t ) are proportional to 1

their sines and dividing both members b y 5 we obtain

,, m

, 1 ,

( tn co s 6 .

I n the derivation of ( ) besides the approximations men 20

ti d there is another in the assumption that A B corr

one , e

sp ond to the perpendicular

s distance between t h two lines of e

the reticle concern d whereas none but images of e q uatorial

e ,

stars will pursue a path across the reticle which is p p d i er en c

ular to all the lines Every other star image follows an .

1 10 G E O D E TI C A S TR ON OM Y .
93 .

apparent path which is curved (the apparent radius of curva

ture being less the greater the declination ) and therefore not
, ,

perpendicular to more than one line of the reticle 5 A B 0

corresponds except for equatorial star to an o b lique dis

.

an

tance between lines of the reticle the obli q uity b eing exceed
, ,

i gly small
n I n spite of all these approximations it has been
.

found by comparing ( ) with the formula expressing the 20

exact r lation betw en i and (t — t ) that for an e x treme

e e

value of i 6 the computed value of (t t ) will b e in

n n m

°
n
0 n m

error by less than for any star of declination less than

and that for a star of declination 8 5 the error is only °

Let us deal now with such a case as that of star 7

93 1

H C V in the preceding computation in which the s t ar

.

an en

image was o b se ved to transit across the first ten of the eleven
. . .
,

lines of the reticl and the transit across the eleventh line was
e

missed S uppose that the equatorial intervals z

have previous ly b een determined b y special o b servations as

.
,,
z,, z,,

indicated in 4 1 1 .

F rom ( ) we may write 20

— t , — i1 sec 6 ;
— t , — i, sec 6 ;
t, ia sec 6 ;
— t
w
— z sec 6 ;
whence
+ t +
i sec 6

m his p m my
pp m y pp
Fo r t h e e xa ct t re a t ent o ft ro ble , se e Ch a u v e n e t
’
s A s t ro n o ,

vol . 11 . . 1 46 -1 49 ; or l le
D o o it t
’
s A s t ro n o , .2 9 1 -2 9 3 .
1 12 G E OD E TIC A S TR ON OM Y .
94

F or t h g neral case (
e e 22
) and ( 24 ) may b e written , re spe c

t iv e ly,

(su m o q u a t o ria l in t e rv a l s
fe

tm m ea n o fo b se rv e d t im es
o f o b se rv e d l i e s) se c 6 )

n u m (
b e r o f o b se rv e d i e s
n

l n

( su m
m dl l q
f e u a t o ria in t e r a
o v l s

t m ean o fo b se rv e d ti m es
n u
of

mb
isse in e s) ( se c 6 )
er o f o b se rv e d l in e s

The first or second of these t w formul is to b e used o ae ,

respectively according to whether less than half or more than

,

half of the lines were observed .

ti C ti I n clin a on orrec on .

94 I n the following treatment it will b e assumed that the

.

two pivots upon which the telescope turns are circular

cylinders having nearly b ut not exactly equal radii and that
, , ,

the two inverted Y forming a part of the striding level d

s , an

the two Y in which the telescope pivots rest all have equa l
s ,

angles
I f F g 5 represent a cross section through one of the
.

i . 1 -

pivots perpendicular to its axis it is assumed that the curve ,

G E B D is a perfect circ le of which t h center is at C and that

, e ,

the angles G F E and D A B are qual to each ther and to the e o

corresponding angles at the other pivot With these mp . a ssu

tions the distances F C and A C from the vertex of the level ,

Y and from the vert ex of the supporting Y to the center of

the pivot are equal
,
.

I n F ig 6 let CC be the line j oining the centers of the

1
’

two pivots the axis a b out which the telescope rotates of

.

which the inclination is required Let F and F b e the

, ,
’
. v er

tices of the level Y and A and A the vertices of the sup

s,
’

porting Y S uppose the radius of the pivot o f which t h

s . e

centre is C is greater than that o f the pivot at C Then

’
.
 94 C T N 3
IN C L IN A TION C
O RRE IO . 1 1

the distance A C ) is greater than the distance

’ ’

F C( = A C ) D raw C G para ll l t F F and CH parall l to

’
. e o e

Th inclination f the line FF ( e q ual t the inclina

’ ’
AA . e o o

tion f the line CG ) is dir ctly measured by the level readings

o e .

G CH
Th angl betw en FF and C C is the angle
’ ’
e e e
2

This is the correction to be applied to the inclination as given

directly by t h level readings to obtain the inclination of the
axis I f the telescope axis wer rev rs d in t h Y the line
e

. e e e e s

A A j ining t h v rtic s of t h suppor t ing Y would remain

’
o e e e e s

unchang d and the lin FF w uld assume a w position

e ,
e
’
o ne

F F
”
such t h at t h angle betw n F F and F F is f
”"
e ee
’ ” ”’
ou r

t im es Let fi and B designate t h W E e

inclination as given by the l vel r adings for lamp west and

,
e e ,

lamp ast r sp ctively ( using the position of the lamp which

e ,
e e

ill mu t the interior of t h tel sc p as a convenient

rn a es e e o e

m ans of designating the positi n of t h telescope) and O

e o e , w

and b the c rr sp nding inclinations of t h axis CC Let

e o e o e
’

the angl G CC be call d p or p i t i q lity L t all

e
’
e vo ne ua . e

inclinations b consider d positive when the west end i

e e s

higher than the ast Then e .

bw : 1
6 W +p
be flE _
p u

flE — B
I W

4 .

Let w and be the readings of t h west and east endse e ,

respectively of t h bubbl f the striding lev l f a given

, e e o e or

ml ly l l l
pp y w
*
Th e s e fo r x g e o f t h e e v e Y s is t h e
in c a se t h e a n

m gl
u ae a re e act o n

wh l w ly y
the a n f t h su o rt in g Y s F o r o rd in a r c s e s h o e v e r in
m ll
sa e as e o e . a , ,

ly x p lm
ic hp is d t he Ys ha ve g e s h ic h d o o t d iffe r g re a t t h e
ll m
, s a an an n ,

my pp
fi c ie n t
a re s u f e act Fo r t h e f
. tr t e t f t h is ro b e
u fo r t h e g e n e ra l
ea n o

c a se se e Cha uv e n et
’
s A st ro n o l —1 58
, , vo . 11 . . 1 53 .
1 14 G E OD E TI C A S TR ON O M Y .
95 .

position of the telescope axis L t w and b e the corre . e

’
e
’

sp d
on i g west
n and east readings after the level is reversed ,

the telescop axis remaining as it was L t be the value

e . e
’
a

of a division of t h l vel in seconds of arc Then f B

e e . or ,

the apparent inclination f the telescope axis xpressed i n

o , e

seconds of time we may write if the level divisions are

num b ered in b oth directions from the middle
, ,

or in more convenient form for numerical work ,

’
fl = (m a i
g ,

in which is a constant for the level

a
.

I f the level divisions are numbered continuously from one

end of the level to the other ( 9) takes the form , 2

’
a

o :
f( w + a
>
—
-

65
;

in which the primed letters ref r to that position of level ine

which the division marked zero is at the western end .

9 5 It still remains to derive the relation between 5 the

.
,

inclination of the rotation axis of the tel scope and the cor e ,

rection to the observed time of transit of a star to reduce it

to what it would b if the rotation axis w re truly horizontal
e e

I f the error of collimation w re z r and the r tation axis

.

e e o o

of the telescope horiz ntal and in the prime vertical the line
o ,

of collimation would describe the meridian upon the celestial

sphere when the telesc pe was rotated upon its axis If w
o . no

the other errors of azimuth and collimation be assumed to

, ,

remain zero but the rotation axis is moved slightly out of the
1 16 G E OD E TIC A S TR ON OM Y 6
.
9 .

The approximations in the derivation o f ( 3 ) are of the 2

same order as those made in deriving I f 4 b e allowed ’

as a maximum value for O the error of the formula will be ,

much less than for any star of declination less than

Under ordinary circumsta c s O will seldom be as great as n e

In deriving 8 from the level readings it is sometime

,
s

assumed that the inclination is variable and that each set of ,

level readings giv s the inclination at that particular time

e .

Under g od conditions however the variation of the inclina

o

tion during any half set is pro b ably less than t h error of any
, ,

one determination of the inclination I t is advisable there .

,

fore to assume the inclinati n constant during a half set The

,
o .

method of computing this mean inclination from the level

readings is su ffi ciently shown in the example in 9 A d is 1 .

tinction is made between level readings wi t h bj ctive north o e

and rea dings with objective south on account of the possibility

that the level r adings may b e a ffected by irregularities in t h
e e

shape of the pivots .

C t i f Di l A b t i
orrec on or u rn a e rra on .

96 The effect of the annual aberrati n due to the motion

. o ,

o f the Earth in its orbit is taken into account in com

puting the apparent star place B t t h effect f the diurnal . u e o

aberration due to the rotation of the Earth on its axis must

, ,

be dealt with in the present computation In round number . s

the velocity of light is 8 6 miles p s cond and the lin ar

1 000 er e , e

velocity of a point on the Ear t h s e q uator due to the diurnal ’

rotation is mile per sec nd The line r velocity of o . a

the Earth in latitude is then cos mile

\

an yp i to n on q)
per second The apparent displacement of a star on the
.

meridian is
k tan ’ "
 97 . A Z IM U TH AND C OL L IM A TION CORR C TE ION S . 1 17

t he motion of the observer b eing at righ t angles to the line o f

sight The hour -angle k corresponding to the displacement
. ,

I is f und by applying the meth d of t h latt r part of § 9

’
e ,
o o e e 2

to the spherical tria gl defined by the true position of the n e

star its displaced position and the pole I t is thus found

, ,
.

that
O sec 6 tan sec 6 ”
I
’
? .

Keeping
in mind that the angles O and ’
O are ve y small
r ,

this may be written

1 86 tan
0 00
3 cos
1
ob
” cos
6 ( p sec . 19 se c

cos sec 6 .
( 3 5)
Fo r convenience O is tabulated in terms of qb and 6 in
3 0 1 .

the aberration causes the star to appear too far east

As ,

the observed time of transit is too late and R is n gative when e

applied as a correction to the o b served times ( e x cept for su b

,

polars) .

A im t h C ti z u orre c on .

97 I f the transit is otherwise in perfect adjustment b ut

.

has a small error in azimuth the line of collimation will ,

describe a vertical circle i a great circle passing through , . e .

,

the zenith at an angle with the meridian ( measured at the

,

zenith ) which we will call a

I n F ig 8 let be the zenith B b e the position of a star

.

. 1 , 2

wh n it is on the meridian and B its posi t ion when observed

e ,

crossing the line of collimation of a transit which has an

azimuth error B y applying the process of the latter part
a .

of 9 t this spherical triangle it may be hown that

2 o s

BB sin C
,
2 a .
1 1 8 C TR ON OM Y 9 8 G E OD E TI AS . .

A pplying the same process again to the spherical triangle

defined by B B and the pole it may be shown that t h
,
’
, , e

corr sponding hour -angle is B B

e 6
’
se c .

L t the angl be express d in seconds of time and be

e e a e ,

call d positive when the object -glass is too far ast with t h
e e e

telescope pointi g southward Th n the re q uired correcti n

n . e o

to the observed times e q ual to the time elapsed between ,

position B and B of the star may be written

’
,

A zimuth correction sin C sec 6 A a 1: a ,

in which A is written f sin C sec 6 and is tabu lated in terms or ,

of C and 6 in 99 2

from the time o b servations will

.

Th me t hods of d riving
e e a

be treated later 1 00- 1

t Collim a ion Corre ct ion .

9 8 I f the instrument is otherwise in perfect adj ustmen t

.

but has a small error of collimation th mean line e

d scri b es a small circle parallel to the meridian at an angula

e , r

distance the error of co llimation from it when the telescope

c,

is rotated a b out its horizontal axis By the same line of

, ,

reasoning that was used in 9 in d aling with t h intervals 2 e

of t h vari us lines from the mean line it may b e shown that

e

e o ,

if be expr ssed i n t ime t hen the

c e ,

C ollimati n correction sec 6 C o c c,

j m hd d d v g m
m y pp x m
O b e c t io n y be
m de to t he et in e ri in t he fo rm u lm

v lv d h m
a a o s u se

d pp x m
f b 2 — 8 e c a u se o f t h e an a ro i a t io n s in in t F o r,
o 9 9 o e e .

in dd it io n h h v b
to t hed st a t e i t io t t m in t he
f m l l m h
a a ro a ns a a e een a e

d p d
d v e ri f h t io n s , t he a ct t t t he fo r in c in a t io n , i d l
lm gl
a a or u a: az ut an co
,

ly d p d
i a t io n ba re d n o h
t in d e en en t has een ne ecte , an d eac t re a t e a s if

h bj p p p
e n t ire h in t o f t he o t

v ld
e en en er .

w h hm hm lf ml ( g h w h
Is f h
e c t io n i ? Our is t o is the
ll g
su c o a re s e n t ur o se u rn

eng in e e r it su c at e a t ic a or u a to et er it an in t e i ent
 o C TR N M Y g G E OD E TI AS O O . 101 .

chronometer will b e sensibly e x act but the computed prob ,

able errors will be slightly too large .

If the rate of t h chronometer is very large it may even e ,

be ecessary to apply a rate correction to the reduction to the

n

mean line in case of an incomplete transit as derived in 9

,
~

, 2 .

Com pu t tia on of t
A z im u h , Collim a ion , t an d Ch ron om et er Correc
t t t h U f L t Sq
t ion s, W i h ou e se o ea s u a re s .

1 00 H aving corrected each observed time of transit f

. or

inclination and aberration the azimuth error and the colli , a

m ation error as well as the required chronometer correction

c,

may b e derived from the observations b y writing an o b serva

,

tion equation of the following form for each star o b served ,

A TC + aA + cC —
( a
(3 )
8

forming the corresponding normal equations and solving for ,

th required quantities A T the chronometer correction

e , , , a ,

and I n ( 3 8 ) is the apparent right ascension of the star

c . a

( reduced to mean time if a mean -time chronometer is used ) ,

and T is the observed chronometer tim e of transit f the

,
’
o

star t f i
c o rrec e a l b t i i l
’
i t i d
or t f a
’
u rn a a e rra on ,
nc na on
,
an ra e o

c lz ro n o m e t e r

T his least square process is rather laborious and a shorter

.

method is desirable for obtaining appro x imate results S uch

,

a short method without least s q uares will now be treated *

I t is a method of successive appro x imations to the required

.

results
1 01 T he e x act form of the computa t ion is shown below
.

a numerical e x ample dealing with the observations shown

.

in

in 9 1 .

Th i m h d w h i h i l fi ld
l gi d p v yf m y y
*
s b i
et o i c ha s een n con t n ua u se n t he e on t he
C
,

y w m y
on tu e d ti
a rt ie s o w f t he o a st a n d G eo e c S ur e or an e a rs , as

v
d e ise d in t h e
’
7o
’
s b Mr Ed . in S it h . t hen an a id on t ha t Su r ve .
§ IO L COM P U T A TION W I TH O U T LE A S T S Q UA R E S . 121

| assa a
| mmmmm
fl 3 3
] mm
” ga
l mm

c
o
> w
.
$ m
i 2

cE
c
2
.

2 m m
o
m n m m
fi t
g
. L “ a a .

o o o c c c
.
o o fi o Eo o o
mD m m fi m mm m D
5 a a C d a od e m
122 G E OD E TI C A S TR ON OM Y . 1 04 .

The first fi columns of the main portion of t h

1 02 ve

computation are compiled from 9 and from the ta b le of

. e

1 ,

§ 992 ( factors A B C ) The remaining columns , are filled , .

t after the computation of and shown in the lower part a c,

f the ta b ular form is completed

ou

I t should be noted that the five stars o f each group

o , .

1 03 .
,

observed in position of the instrument hav be n

one , e e so

selected that one is a slowly -moving north rn star at a con e

sid b l distanc
e ra from the zenith ; whil the other four
e e e a re

all comparativ ly near the zenith some ransiting to t h

e , t e

northward of it and some to the s uthward and so placed o ,

that their mean azimuth factor (A ) is nearly zero These .

four stars of each group are for convenience called t im t e s a rs ,

since the determination of the time falls mainly upon t hem ,

while the slowly -moving star serves to determin the azimut h e

error of the instrument and is called the im t n t az u s ar

1 04 I n the computation
.

. to d riv and t h four tim e e c a , e e

stars in each position of the i t m t combin d and treated n s ru en a re e

as one star by taking the means of their ( and of

their factors C and A respectively the m ans b eing writt en
, a

b low the separate stars in the computati n form together

, ,

e o ,

with the azimuth stars} O the assumption that the means n

of the time stars in the two positions of the instrument are

e q ually affected by the azimuth error the first approxi mation ,

to is found by dividing t h di fference between the t w mean

c e o

values of T by the di fference between the two mean

a ,
’

C Or
’
s .

T T
,
( a c ) W c )E
G OV — CE

mpl m mp wh
pl k w hl l
*
Th is xa f t he t h d f u t in g d it out le
m h
e e o e o o co a an c a st

m pp v y p
s q u a re s an d uc o f the e x a n a t io n o f it , is t a en it it t e m o dific a
C
,

y
t io n f ro A d ix No f t he o a st d Geo d e t ic Su r Re o rt fo r
9
m
en . o an e

1 8 96 b , A sst . G . R Pu t n a
. .
1 24 C TR ON OM Y G E OD E TI s AS . 1o .

in the last column to b e due to error in the first approximate

value f ; substituting from that column in formula
o c and
thus obtaining a correction to the first approximate Thus
i the present case the second mem b er o f ( 3 9) b ecomes
c .

_
(4 3 )
8
O ' 01 9 ’

T he second approximation to the tru e value of is then c

P roceeding as b efore
improved values for W and are, a aE

found by the use of formula Thus in the present case

there are obtained as second approximations
3 98 4-
6 0)
+ -
0
8
0 559 ,

o oo.
+ O
8
°
SO 6 ‘

T his process of making successive approximations to the

values of and may b e continued until the values (
c a T a
,
’

C A ) show a su fficiently good g

a m a re e e

t be
c

with a well -chosen time set the final value for 1 no

changed b y as much as
,

by any number of pp im a ro x a

tions made after the a b ove agreement has been b rought within
the limit
Wh n satisfactory values for
e w and h ave b een c, a , aE

obtained the corrections C and A are applied separately to

,
c a

each star as shown in the sixth seventh and eighth columns

, , ,

of the upper part of the computation form and the values of ,

the chronometer correction (A T derived separately from ach e

star The residuals furnish a check on the computation

. .
 1 06 COM P U T T ON W TH O U T
. T QU R A 5
I I LE A S S A ES . 12

A y large error in observation or computation will be indi

n

t d by the r siduals and may often b e located by a careful

ca e e ,

study of them The mean value of A T is the re q uired .

.

chronometer correction at the epoch of the mean of the

observed chro omet r times n e .

1 06 A study of the above process of successive pp

. i a ro x

matio to the values of

n w and shows that the rapidi t y c, a , aE

with which the true values are approached depends upon

three conditions The mean A for the time stars for each .

position of the instrument sho ld be as nearly zero as possi

ble I n each position of the instrument the A for the azim u t h
u

star should di ff r as much as possible from the A for the mean

e

time star while corresponding C should di ffer as lit t l as

,
’
s e

possible Th last two condition s are di fficult to satisfy

. e

simultaneously but the fact that both mus t be considered

,

leads t avoid obs rving sub -p lars The conditions here

one o e o

stated sho wwhy t h stars for the abov time set w re chosen
.

'

e e e

as indicated i § 3 It is not advisable to spend time in

n 10 .

observing more than azimuth s t ar in each half t one se .

It sh uld be noted that the choice of stars indicat d above

o e

also insures the maximum degree f accuracy in the d t m i o e er

nation of A T for a giv n expenditu r f time regardless of

,
e e o ,

the method of computation .

The two things which especially commend this pp i a ro x

mate method f computing time to those observers who have

o

used it much in the field are the p id ity with which the ra

computation may be made ( specially whe C ll multipli e n re

’
e s

cation -tables are used ) and t h y which results from , e a cc u ra c

the fact that the derived values f and depend upon all o a c

the observations and not upon obs rvations upon a few stars e

only as is f requently the case with other appro x imate

,

met ods h .
1 26 GE OD E TI C A S TR ON OM Y . 1 07 .

Com pu t ti
a on of t h e A z im u t h , Collim a t ion , a n d Ch ron om e er t
t Corre c ion s b y Lea st Squ are s .

1 07 . We start with the observation equations indicated *

in F lamp west each of these equations are of the

or

477 1 + A w a w + CC ( a T!)

and for lamp east of the form ,

A E + A E4 E + Cc " —
( a “
TO 2 0 ( )
4 6

The su b scripts added to A discriminate b etween factors

applying to stars observed with lamp east and those observed
wi t h lamp west This is d ne for the purpose of avoiding
. o

confusion in the normal equations The parenth sis ( T) e

’

is an appro x imate value for the clock correction after taking

. a c

accoun t of inclination rate and aberration It is the observed

b
, , .

q uantity The coe f

.ficients A and C may e obtained from
the tabl in g 99 e 2 .

Tr ating the observation equations ll together as a single

e a

group as many equations as stars the four derived normal

, ,

e quations are of the form

+2 A W 4 W +Z A E aE + 2 Cc - z Ty) =
'

2 A T. (a o
—
E A W A TC +2 A Wa W
9
+ A WC
2 c Z A (a
W
Z A A Tc
'

=
+2 A E a E + Z — 2 A =o
E
A E
Cc
E ( a .

Z CA TC +2 —- 2 CA
+2 9 — E C( a — Tc =0
CA W a W f C
’
4 c
E E
.

The solution of these e q uations gives the re q uired quan

tities A T W and ” a aE c

To obtain the probable error of a si gle o b servation su b

.
, ,

O b se rv a t io n e q u a t io n s a re l
a so ca ll e d co n d i t io n a l e q u a t io n s b y m
so e

aut h o rs .
128 G E OD E TIC A S TR ON OM Y .
g 1 08 .

the use o f least squares in 1 06 . T he o b servation equation s

are

1 . 1 3c o . 1 1 o

1 . r1 c — o . o6 = o

1 0 6c
. o . 1 9 o

1 .
46 c +o . 23 z o

1 . 64c +o . 29 = o

4 . 1 8c + 0

’
Z v

rom the absolute term in each equation

F has been
dropped as is fre q uently the case in l ast s q uare computa
,
e

tions for the pu rp se of shortening the numerical work The

,
o .

true value of ( T ) is then in each case that written

’
a c , ,

above
The four normal e q uations formed from the a b ove b
,

o se r

vation e q uations in the usual way are

+ 1 0 00 A. T,
0 .
7 3a W 2 .
5 E
1 a 2 08c
.
+ o ;

o .
73 A T ,
1 .
33 W
a 2 . 2 4c o .
7o = o

2 .
51 A T, 6 77a E
.
+ lo . 8 4c
2 08A
. T,
2 2 4a W
.
+ 1 0 8 4a E
.
3 6 64c
. z: o .

The so lution of th se equations for the unknowns gives e

+ u p;3 4 and A T c o
s
. o , ,

which combined with t h which was dropped e

t
o ease the numerical work gives A T .

If these values are now substituted in the observation

equations 8 the residuals ( ) there shown are o b tained
, 10 , v .
g 1 1 0. COM P U T T N
LE A S T S 9 Q UA RE A IO .
12

F rom these the probable error of a single observation see ,

formula is
e j ;

The modifi ed n rmal quations being solved for Q as indi

o e

t d in
ca e 7 its value is
10found to be 58 0 1 1

H ence the probable error of the result ( A T) is see

. .
,

:
c ,

formula
e
, j; Vo . 1 1 58 : l:

1 09 . If
instead of computing a separat value for the
,
e

azimuth error for each of the positions of the telescope

, a ,

axis before and after reversal the azimuth error is assum d

, , e

to be the same throughout the whole set the princi pl s , e

involved in the compu t a t i n are the same as before ; t h dis o e

tinction between W and is dropped ; th re are b t three

a aE e u

unknowns and three n rmal equations instead of four ; and o

the work of solving the normal q uations is corr spondin gly e e

shortened The loss of accuracy in the computed resul t

.

depends upon the magnitude of the actual change in t h e

azimuth error at rev rsal If no more than six stars e . a re

observed in a set it may be advisable to use this process

, so

as to reduce the umb r of unknowns n e .

1 1 0 Exp ri nce shows that the process outlined in

. e e
5 10

10 6 gives such an accu rate value f that t h value b or c e su se

q uen t ly derived from a l ast s q uar adj ustment is found to

e b e e

subs t antially iden t ical with it Wh n such a preliminary com . e

p ut t
a i h onb en
asm d t hel t q dj tma t i shortened
e, e eas s u a re a us en s

considerably with little loss f accuracy by accepting this

, o ,

preliminary value f applying the collimation correcti ns

o c, o

( as well as the inclination rate and aberration c rrections )

b efor the l ast square adjustment and treating the clock
, , o

e e ,
1 3 0 C TR N M Y G E OD E TI AS O O .

c rrection and the two az m uth errors as the only unknowns i

I t is well in this case to tr at ach half set separately The

o .

e e .

discrepancy between the two values for the clock corr ction e

thus deriv d when r duced f cl ck rat to t h same poch

e , e or o e e e ,

indicates t h amount of rr r in the assumed value f

e e o or c .

To illustrate this m thod we may use t h same set of e e

obs rvations as i n 9
e L t it be assum d that the 1 , 101 . e e

preliminary computati n sh wn in has be n made d o o 101 e ,

an

l t t h value
e e for given there b accepted as a basis c e

f this computation
or The observation quations now b come . e e

A T, 0 02a
.

W + o
'
. 1 1 0

A T, +o
l
o .
3a . 1 4 0

A T, 0 .
3 6a W
o . 28 o F o r t h e fi rs t ha f o f se t .

A T, 0 2 2a
.
W o . 07 z 0

A T, 1 03 a
.

W +o . 60 o

A T, + 0 2 .
5 a
E
-o . 10 o

A T, +0 —

l
3 5a o 22 0
E
. .

A T, o z .
a o . 1 8 z o Fo r t h e se c o n d ha f o f se t .

A T, - o .
3 8a E + o . z 4 z o

A T, - 2 .
53 a E 0

The normal e q ua t i ns o half of the set are

5 . 0 0A T ,
0 .
7 3 W
a + z 0;

0 .
73 A T , + 1 .
3 3a W o

and for the second half o f the set ,

5 . 00 A T ,
2 .
SIa E + o ;

2 .
51 A T, 6 77a E
. o .

The soluti n gives for the first half -set A T

o ,

and Q and for t h second half -set e

and Q aE

The probable error of a single observation derived from

1 3 2 G E OD E TIC A S TR ON OM Y g 1 12 .

little error to instruments of the same nature which are con

sid b ly larger or small r
e ra e .

112 T introduce the une q ual r lative weights

. o in t e , 20 , o

the l ast -squar adj ustment it is necessary to multiply each

e e ,

observation equation by V and to make the usual b iv , su se

q uent modifications in t h least square compu t ation Th s e . e e

modifications are indica t ed in the following xampl -t h e e , e

same problem as that trea t ed in 5 8 without the of 10 , u se

une q ual weights I f an incomplete obs rvation is made upon

. e

a sl w star that both the tables of 3 and of 3 3 mus t

o , so 02 0

be used first multiply the two r lative weights t geth r

,
e to o e ,

and then take the s q uare root of that pr duct as the multi o

plier for the observation equations .

The s q uare roots of the weights given to the ten stars in ,

o rd r of bs rvation are respectively

e o e ,

and The obs rva t ion q uati ns shown e e o

in 8 10 multiplied by these factors r sp c t ively Th

a re e e . e

normal q uations r sulting from the weighted bservatio

e e o n

equations so obtained are

6 8o A T
.
, 0 01a W
. 0 04 a E
. 0 .
53 c o ;

0 .
57 c 0
8
.
34 = o ;

o . 8 7a E 0 .
9 3 c o ;

T ,
0 .
57 a W + 0 .
9 3 E
a 1 3 7g. o o .

The solution of these e q uations g ves A T i ,

and a,
’

The probable error of an observation f w ight uni t y is o e

:t
‘

The pr bable error of o

A T e e VQ j : 0 04 1
8
.
/
1 o 1 47.
j;
, ,
 1 14 . A UXIL IA R Y OE S E R VA TION S .
I3 3

Unl ss an extreme d gr e of accuracy is r q uir d the

1 13 . e e e e e ,

assumption that all obs rvations are of equal weight is ffi e su

i t ly exac t
c en The introduction of un q ual w ights adds so
. e e

lit t l to the accuracy of t h computation that con mic con

e e e o

siderati ns will of t n indicate that t h least s q uare adjustment

o e e

should be ma d on the basis f e q ual weights e o .

*

A x ili y Ob s ti s u ar erv a on .

1 1 4 A side fr m the bservations and computations which

. o o

have be n t reat d in d tail certain o t hers are n cessary f

e e e , e or

th d t rminati n of t h instrumental constants which have

e e e o e

b en assum d in the preceding tr atm nt to be known

e e e e .

The e q uat rial intervals of the lines of t h r ticle may b

o e e e

d termin d from any series of compl te transits i

e e observa e , . e .
,

ti ns in which the transit of each star was obs rved acr ss

o e o

e very lin A sp cial s ries of observations is not r q uired

e e e e ,

for t h complete transi t s f the particular s ries f time

e o e o

observations u nd r treatmen t may b utiliz d for this purpose e e e

i addition to using th m t
n d termine t h cl ck c rrection e o e e o o .

F or e v ry compl te transi t every t rm in equa t i n ( ) (see

e e e o 20

namely i ( t t ) cos 6 is known exc pt i E v ry e e

mp y
, ,, n m ,

Th e t a t io n b se rv a t i s t a
fa b t he a u t h r
s e rie s o fti m k
lk l l k l m
co u o e o o n en o

t h e s h re o f C h i

l yl lw
t I e t A a s a ( in a t it u d e 59 in 1 8 94 wa s a d e
°

l
o n o a n , ,

in t h e fie d b e st s q u e s g iv i g a ll s t rs e q u e ig h t re g rd ss o f

l m m l l
a ar ,
n a a ,
a e

p
t h e ir d e c in a t i n s a d o f t h e

y
b f is s d i e s I t he fi
l m
o u n co m
n e r o e n n na

d e a t t h e C s t a n d G e o d e t ic S u rv e O ffic e in
.

w lw w
u ta ti n su b se q
o n t y ue a oa

ll y p k p
W a h i gt o n u e q u a
s n ig h t sn e re a s s ig d In t h e se i s t h e re
e re 46 ne . r e e

w w
s e t s e a c h c o n s is t i g g r i g f o b se rv a t io s u o n 1 0 s t a rs
m
,
S ea n , en e a n ,
o n .

mp wy m m
Th e a v e ra g d iff e e c e e it h u t r g rd t o sig n b t e e n t h e c h r n o e t e r
n , o e a r ,
e o

p l
c o rre c t io u t e d in t h e t wo s f ro t h e sa e se t o f o b se rv t io s

l k
s as c n o a a n

mp
wa s Th is is a b o u t e q u a l t o t h r b b e e rr r o f t h e c o c

m m mm
co r c e o a o re

w m y
ti n c te d fr B t it u s t b e re b e re d t h a t t h e c o d it i s
m l
o o a se t
u o e . u n on

w l wm
e re e x t r O a c c o u n t o f t h e h ig h a t it u d e f t h e s t a t io n
e e . n n f t he o a o

- o v i g s t a rs ( e v e n t h s e b se rv e d in t h z e n it h ) Th e
m yl m pl
s t a rs r s o e e n o o e . re

w ly
w s so
a u c h i t e rf e re n c e b cn ds th t c e t e o b s e rv a t io s n a ll t h e
ou a o n o

l l ly l w
s t a rs e e s e c u re d o n o n
r 10 ig h t s o u t f t h e 46 a n d o b se rv a t io n s o n a n o ,

s i g e in e o n
n o f t h e re t ic e e re n o t in f re q u e n t .
1 34 G E OD E TIC A S TR O N OM Y . 1 1 5
.

complete transit observ d furnishes th n a determination of e , e ,

the e q uatorial inte val of every line The transit of a slow r .

moving star gives a more accurate determination of the

e q uatorial intervals than the transit of a tar of small declina s

tion for the rrors in obs rving t do t incr ase so rapidly

,
e e ,,
no e ,

with increase f declination as cos 6 decr as s F this

o , e e . or

reason s me observers pref r t make a sp cial series of

o e o e

o bs rvati ns for e q uatorial int rvals using s t ars of large d li

e o e ec

na t ion only I n compu t i g and using the quatorial intervals

. n e

it m ust be b rn in mind that wh n the tel scope axis i

o e e e s

reversed in its Y t h ord r in which t h star transits acr ss

s e e e o

t h lines is revers d and also the alg braic sign of the

e e
q ua, e e

torai l interval f e ch line o a .

1 1 5 Th porti n of a set of obs rvations given below wil l

. e o e

serv t sh w h w the pivot in quality p (see

e o o o i e , ,
s

de t ermined by a series f r adings f the striding level upon o e o ,

t h telesc pe axis placed alt rnately in each of its two possi

e o e

b l p siti ns with clamp west and clamp eas t ( t h

e o o clam p e

instead f the lamp being here used to indicate t h positio

o e n

of the axis) .

O B S E RV A T O N S F O R I INE Q U A L ITY O F PIV O TS O F

A W
T RA N S IT N o .
4 .

S T TI ON Se a t o n , W a sh m gt o n
'

.
—G . D .
, o bse rv e r — J u n e . 1 9, 1 867 .

CL A W MP ES T . CL A ATMP E S ..

b
O j ec t -
gl a ss S . O b je c t -
gl a ss N .
h _ Ow
T im e
“ 2 7” H2 ?”
Le v e l . c el .
4

W . en d . E . en d . W . en d . E . en d .

—o .
3oo

05 A . M .
1 3 6 G E OD E TIC A S TR ON O M Y
. 1 19 .

ment of precision This time is greater the smaller is the .

value of a division ( of a given l ngth ) expressed in arc but e ,

for levels of the same division valu is less the more perfect is e,

the inner upper surface I f the level tube is so held in its .

m tallic m unting that th re is any possibility that it may be

e o e

put u der stress by a change of temperatur it is advisable to

n e ,

d termin the value of a division wit / t / t q i it m t i g

e t ze n s

at two or m re widely different temperatures I t may be

e ou n n

o .

well als to d t rmine whether changing the length of t h

o e e e

bubble by changing t h amount of li q uid in the chamber at

,
e

the end f the level tube changes the apparent valu e of one
o ,

division
1 1 7 I f an observer is forced to determin
.

. the value of a e

level division in the field remote from a lev l -trier aft r some , e , e

accid nt let us say which leads to replacing an old well -known

e , ,

( b t broken
u ) lev l by another of which the value iseunknown ,

his g it y will lead him to devise a method of utilizing

1n en u

whatever apparatus is at his disposal The thre methods . e

giv n b low will be found suggestive

e e

1 1 8 I f a tel scope having an eyepiece micrometer similar

.

.
.
e

to that of a zenith telescope measuring altitudes or

zenith distances is available t h unknown angular valu of a
, , e e

division f the level may be found by comparis n with the

o o

known ngular value of a division of t h micrometer P lace

a e .

the lev l in an extemporized mounting fix d t the telescop

e e o

P oint with the micrometer upon som dis t ant w ll -d fi d

e .

e e e ne

fixed obj ect and read the microm ter and l vel Chang the e e . e

m icrome t er reading by an int gral number of d point e 1 v 1 5 1 0 n s,

to t h same object again by a mov ment of the t lescop as

e e e e

a whole d note the w reading of the level Every

, an ne .

repetition of t his routine gives a determination of the value

o f a level division .

1 1 9 If in his instrumental outfit he has another well

.
 1 20 . TH E L E VE L VA L UE .
7

d t rmined l vel of sufficient sensibility the o b server may use

e e e

it as a standard with which to compare the unknown l vel e .

P t the unknown level in an extemporized mounting fasten d

that of t h known level A dj us t so that both b ubbles are
u e

t o e .

near the midd l at onc C ompare corresponding movements e e .

of the t w bubbles for small changes of inclination c m mon

o o

to both levels .

1 20 Th following m th d
. e giv s fully great precision e o e as

as ei t her of the other two outlin d above and is especially e ,

v aluable because the requir d m ans apt to b at hand in e e a re e

the fi ld ev n when t h apparatus re q uired for the other t w

e e e o

m e t h ds is wanting
o .

t hi s method t h nly ins t rument re q uired is a t h d

.

F or e o eo o

l it or an engin er s transit or any oth r ins t rument having ’

e ,
e , e

b th h riz n t al and vertical circl s ( not n cessarily wi t h a fi

o o o e e ne

g radua t i n ) and a goo d vertical axis M unt the l vel t h o . o e on e

plat of the t m t parallel to the plane of the t el scope

e 1 n s ru en e

a n d adj us t it as i f it w re a p late l vel Make t h v rtical e e . e e

axis t ruly v rt ical in t h usual way M asure t h z nith dis

e e . e e e

t anc f s me w ll -d fi d stationary bj ct t aking r adings

e o o e e ne o e , e

w i t h ( vertical ) circle right and circl l ft t liminat index e e o e e

e rr r N w incline t h v rtical axis directly toward from

o . o e e or

t h objec
e t from to , by of t h foot -scr ws The
1
°
u se e e .

d irection f this inclination may be assured by use of t h

o e

p lat l v
e l which e is t right
e angles to the plane of the t le a e

scope Measure the apparent zenith distance of the bject

. o

a gain The apparent change in the zeni t h distance is evi

.

d t ly the inclination f t h axis to t h ver t ical which we o e

I f now the instrument is r volv d complet ly

en e ,

w ill call y .
, , e e e

l y f Cm k ll
y be d fu l b Pro
pp y
D e sc ri in . G . C . o st o c in t h e B u e t in o f the U n i

g ll y B
ve rs it f W is c o n s in S c ie n c e S e rie s v o l. No 6 8—74 d i db
l
o , , 1 , .
3, . . an sa

l f m wh
h im b e du e o ri in a Th o se d e s irin g fu t h e r d e ta i s
m
to to ra u n . r a re

re fe r e dr to t ha t a rt ic e , ro ic h t h is st a t e en t is c o d e n s e d
n .
1 3 8 G E O D E TI C A S TR ON O M Y . 12 1 .

around its vertical axis two positi ns will b e found at which , o

th ebubble of t h level is in the middle of the tub F

e e . or

p sitions near th se two the bubbl is within such limits that

o e e

i t may be read I t is from r adings of the bubbl in such

. e e

positions in connecti n with r adings of the horizontal circl

,
o e e

and t h e ab ve ou t lin d det rmination of y that t h value of

o e e , e

a divisi n f t h l v l is deri ed
o o e e e v .

1 2 1 L t F ig
.
9 repr
e s nt a portion of
. the cel stial
1 e e e

sph re adjac nt t the zenith Z and let V and S be t h

e e o , , e

poin t s in which t h axis f t h theod lite and the lin drawn

e o e o

from the c nter of curvatur of the level tu b e through t h

, e

e e e

middle of the bubble respectively intersect the sph r , , e e .

”

S ince the bubble always stands at t h highest part of the e

t ub its position 5 and the corr sponding valu of q are

e , , ,
e . e

f und by l tting fall a perpendicular from the z nith upon the

o e e

arc VS and in the right a n gled spherical triangle thus formed

,
-

we h ve the relation
a

tan q tan y B (5 ) co s . 0

S inc t h lev l t ub tu rns with the th odolite when t h

e e e e e e

latter is rev lv d in azimuth while the positions of the po nts

o e , 1

V and Z r main unchanged it appears that t h a gle 6 must

e e n

If
,

vary directly wi t h the r adings of the azimuth circl e e .

w repr sent by A t h reading f t h circle when the arc V5

e e ,
e o e

is made t c incid with VZ we shall have corresponding to

o o e ,

any oth r reading A / e

tan q tan y cos (A (5 ) ,

1

The valu of A may b obtain d by taking the mean of

e ,
e e

any two readings of t h circle for which the bubble stands at e

the same part of t h t b

‘

e e

I f A and A denote slightly di fferent readings of the

u .

“ ’

azimuth circle O and O the corresponding readings of the

,
’ ”
1 40 G E OD E TIC A S TR O N O M Y
. 123 .

Discu ssion of E rrors .

ollowing the same general plan as in discussing the

1 22 F
errors of s xtant observations the e x ternal errors i t
.

e , , n s ru

mental errors and observer s errors will be discussed sepa

,
’

ra t ly and then their combined effect will be consider d

e , e .

The two principal t l are the error in the ex e rn a e rro rs

a ssumed right ascension of the star and the lateral refraction ,

of t h light from the star

e .

I f only such stars as are given in the va ious national r

ephemerid s are observ d for time the probable errors in the

e e ,

right ascensions will usually be on an av rage j or e :

j : and no appr ciable constant err rs need be pp e o a re

hended from this source .

F rom considerations which ne d not be sta t ed in detail e

here ,
is led to the conclusion that t h e ffect of lateral
one e

r efraction upon transit tim observations must be q uite mall e S

in comparison with the other errors ; but it is di fficult to

estimate b cause it is always masked by other err rs foll w
,
e o o

ing about t h same l w of distribution F furth r

e a id . or e co n s

c ration of this matter 9 se e 2 1 .

1 23 A mong t h instrumental errors may b m ntioned

. e e e

those arising fr m chang in azimuth collimation and inclina

o e , ,

tion from non -ver t icality of the lines of the reticle from poor
, ,

focusing and poor c ntering of t h ey piece from irregularity

e e e ,

of pivots and from variations in the clock rate

,
.

The errors of azimuth and collimation being de t ermined

from the observations themselves are q uite thoroughly can
celled out from t h final lt p i they remain constant
e re s u , ro v a e a
’ '

d uring the p riod over which the observations xtend and

e e ,

pr vided also that the stars observed are so distributed in

o

declination as to furnish a good determination of these con

t
s an t Their h g however during that interval arising
s . c an es, , ,
 1 23 . E RR OR S . 141

from changes of temperature shocks to the instrument

other causes produce errors in the final r sult I t is in this
, , o r

,
e .

connecti n that t h stability f t h pier is f especial imp r

o e o e o o

tanc Such changes will vid n t ly b smaller t h m ore

e . e e e e

rapidly t h bs rvations are mad and the more carefully the e

instrument is handl d I n general they are probably small

e o e

e .

but not inappreciabl e .

To considerable extent t h sam r marks also app ly to

a
'

e e e

th inclina t i n rror Th / g in inclination during each e c za n es

half -set evidently p roduc errors direc t ly H ence again the

e o e .

e .

d sirability of rapid manipula t ion B t the mean value of

e . u

the inclination is d t rmined from r adings f the striding e e e o

l v l not fr m t h tim observations and t h lev l may give

e e ,
o e e , e e

an erron ous determinatio of t h m an inclination D i ff r

e n e e . e

ent obs rvers se m to d i ff r radically as t t h probabl mag

e e e o e e

n it d u f rrors from this source but t h b st obs rv rs are

e o e e e

pron to use the striding l vel with great care H w ver

, e e

e e . o e

small t his error may b und r the best conditions and most e e

ski lful manipulation there can b d ubt that car less , e no o e

h andling and slow reading of the striding l vel or a li t t le *

e ,

heedl ssness ab ut bringing a warm reading lamp t near to

e o oo

it may easily make this rror

,
f t h larg st a ff cting t h e one o e e e e

result A err r of . inch in t h d t rmina t ion of t h

n o e e e e

di ff rence of l vation of t h two pivots f such an instrum nt

e e e e o e

as that described in 8 3 produc s an rror of more in e e or

the d duc d t ime f transi t of a z nith star

e e o e .

If t h lin s of the reticle e not car fully adj usted so as

e a re e

to define v r t ical plan s stars will be observed too e

early too late if b s rved above or below the midd le of the

e

or o e

reticl S uch err rs may b mad v ry small by careful

e . o e e e

m mp g lvl
p p
It is h e re d t h a t be f e t t t in d the it h a s be
l l m
a s su e or a e to re a e e en

in o s it io n g o u g h fo r t h e b u bb to co t o re s t in t h e o s it io n f
l m
on en e e o

e q u i ib riu .
1 42 G E OD E TIC A S TR ON O M Y . 1 24 .

adj ustment and by always bserving within the narrow limits o

giv n by the two horizon t al lines of t h reticl e

P oor focusing of either the object glass or the eyepiec

e e .

- e

leads to increased accid ntal errors because of poor d finiti n e e o

of the star imag B t poor focusing of t h obj ct -glass is

e . u e e

esp cially obj ctionable because it puts t h r ticle and t h

e e , e e e

star image in di ff rent plan s and so produc s parallax The

e e , e .

parallax rror may larg ly b av ided by c nt ring the eyepiece

e e e o e e

each time over t h lin of the r eticl upon which the star is
e e e

next to be obs rved This r peated c nt ring sh uld never be

e . e e e o

omitted even though t h bserver may be confident that the e o

focusing is perf ct I t also serves in a measure to avoid

e

err rs which might otherwise be produced b y the im p f

.

o er ec

ti ns of the eyepi ce e

I f the ine q uality of the two pivots has been carefully

o .

determined as indicated in 5 the errors arising from 1 1 ,

d fects in their shapes may ordinarily be depended upon to

e

b e negligible .

C hang s in the rate of the timepiece during a set

e f o

observations evidently produce errors in the d duced clock e

corr ction at t h m an epoch of the set Under ordinary

e e

circumstances such errors must be exceedingly small I f

e .

. ,

h wev r an observer is forc d to use a very poor timepiece

o e ,
e ,

or if cl uds interf r so as to extend t h interval re q uired f

o e e e or

a t of observations ov r sev ral hours this error may

se e e ,

b come appreciable It is less the more rapidly the b

e . o se r

va ti are made
ons .

The errors introduced by irre gularity in the action of a

chronograph f the form described in 8 9 too small to be
o a re

c nsidered esp cially if its speed is assumed to b constant

o ,
e e

simply du ring the int rval between successive clock breaks e ,

and the chronograph sheet is read accordingly

are b y far the most serious in
.

1 24 The b
’
. O se rv e r s e rro rs
1 44 G E OD E TI C A S TR ON OM Y . 1 27 .

1 25 . In
addition s t ill to these errors there is anot h er
, ,

which is c nstant for all the observations of a set Every

o

exp rienc d observer though doing his b est to record t h

.

e e , e

time of transit accurately in reali t y forms a fixed habit of

observing t late or too early b y a constant int rv l This
,

oo , , e a .

interval b tw en the tim when t h s t ar image actually tran

e e e e

sits across line of the reticle and the recorded tim of

a e

transit is called the O l t p l q ti of the observer

a so u e erso n a e ua on

The di fference b etween the absolute personal quations of two

.

observers is called their l t i p l q ti The rela re a ve e rs o n a e ua on .

tiv pers nal e q uation of t w experi nced bservers has been

e o o e o

kn wn to be as great as
o and values greater than 2 5 0 2

are comm n F a more detailed discussion of pers nal

o . or o

e q uation see 43 44 2 2

sum up it may b e stated that the id t l

.
, ,

1 26 T . o , a cc en a

errors in the determination of a clock correction from b o se r

va ti with a portable astronomical transit upon ten star

ons s

may be r duc d within the limits indicated by the probabl

e e e

e rr r i
o t j b t that the r sult is subject to
o ; u e a

large t t error the observer s absolut p rsonal e q ua t ion

con s a n ,
’
e e ,

which may be t times as great as this probable error

en .

Miscella n e ou s .

I n the field i t is of t n necessary to use other i t

1 27 . e n s ru

ments as transits for the determination of time A theodolite .

when so used is apt to give results of a higher degr e of e

accu racy than woul d be expected fro m an instrument of its

size as compared wit h the astr nomical transits whose per o

f m
or has j
an ce ust b en discussed - unless indeed one has it
e , , ,

firmly fixed in mind that the principal errors in a transit time

determination are those due directly to the observer O t h . n e

o t h r hand a zenith telescop o f the common form in which

e , e

the telescope is eccentric with respect to the vertical axis has

 1 30 . C
M IS E L L A N E O US . 1 45

b en found to give ra t her disappointing results -perhaps

e ,

because f t h asymmetry of the instrument and of the fact

o e

that there can be no rev rsal of the h i t l i in its

'

e or z on a ax

b earings but nly f t h instrument as a whole

s

,
o o e .

1 28 The math matical theory for the d termination of

. e e

time by t h use of the transit in any posi t ion t of the

e o u

meridian has be n thor ughly dev loped T ha t practic has e o e . e

b en adv oca t d B t the additional di ffi culty f making the

e e . u O

computation ov r that for a transit nearly in the meridian and

,
e ,

other incidental inconveniences much more than o ffset the ,

fact that t h djustment for putting the transit in t he a e

meridian is unnec ssary The transit is g n rally used in t h e . e e e

m ridian f tim t leas t in this country

e or e , a .

1 29 The use of t h
. transit for time in the vertical plane e

passing thr ugh Polaris at the time of observation has als

o o

b n advoca t ed and has been used to a considerable extent i n

ee

Europ The obvious advantage which this mod f

. e e o

observing possess s li s in the shorter period of time durin g e e

which the obs rv r dep nds upon the stabili t y f his i t

e e e o n s ru

m ntal constants F meridian obs rvations this period i

e . or e s

rarely much l ss than half an hour whil by t h meth d sug

e , e e o

gested — i which t h whole tim t consists of a pointin g

n e e se

up n P olaris imm diat ly followed by an observati n of t h

o e e o e

transit f a z ni t h or south rn star acr ss that vertical plan

o e e o e

it need n v r exce d five minut s This m thod i

e e e e .
"
e s

O p n to a less x t en t to t h sam obj ctions that f t h

e ,
e , e e e as o e

preceding paragraph This in connection with t h fact tha t .

, e

it is rarely us d in t his country makes its ex t nd d discussio

e , e e n.

inadvisable h re e

1 3 0 I f t h transit is turn d at right angles to the plan

.

. e e e

of the m ridian in th r words is put in t h prime vertical

e , o e , e ,

Se e
8 -
Bu ll e t in o f the Un i v e rsit y of W isc o n sin , Sc ie n c e S e rie s , v o l. 1

No .
3 .
PP 1 93
~
1 46 G E OD E TIC A S TR ON OM Y .
g 1 33 .

an bs rvation of the tim of transit of a star across the mean

o e e

lin of its reticle furnish s a g d determination of the lati

e e oo

tude f the station if t h clock c rr ction is known O if

o e o e . r,

both transits east and west of t h zenith are bserv d t h

,
e , O e , e

latitude may be computed with ut a knowl dg of t h c l ck o e e e o

correction F ormerly this me t h d was oft n us d for the

. o
*
e e

de t erminati n of latitude N w it is almost entir ly super

o . o e

se d d by the use of the z ni t h telescope for latitude

e e .

1 3 1 Th S or a planet may s metimes b bserved for

. e un o e o

time I n t h case of t h S the transit f both the pr ced

. e e un o e

ing and t h following limb may be observ d and t h m an

e e , e e

taken as t h time of transit of the center B oth limbs of a

e .

planet may possibly be observed if a ch ronograph is used .

O th rwise the pr c ding nd f ll wing limbs may be obs erved

e e e a o o

alternately on successive lines of the reticle t aking car that , e

the number of observations on ach limb is t h same and t h e e , e

mean of all t aken as t h t ransit of the center across the m an e e

line
1 3 2 I t is not advisable to observe t h Moon for time for
.

. e ,

its place is not well determined U sually but one limb can be .

observed the other being either bscure or invisible ; and the

, O

O bs rvation of t h limb on a sid line of the reticle is aff ct d

e e e e e

by the rapid change in the Moon s right asc nsion and by a ’

e

parallax due to its c mparative n arness to the Earth o e .

Q U E S T IO N S A N D E XA M P L E S .

obse rver who is trying to get his transit into

133 . 1 . An
’

the meridian to b gin observations for t ime finds that an e

observation upon 7 D rac nis ( 6 6 indicat s that his 1 o 1

°
e

chronometer is fast of l cal sid real time while an b o e o

vation upon 6 H erculis ( 6

, se r

indicat s that his 2 1

°
e

l m l l my
pp my pp
Fo r t h e de ta i o f t h is et hod se e D o o it t
’
e s Pra c t ic a l A st ro n o ,

- v o l. 2 3 8-2 7 1
’
.
34 8 3 77, o r C ha u ve n e t s A s t ro n o , . .
1 48 G E OD E TI C A S TR ON OM Y .
8 1 33

Po sit io n C o rre c t e d T ra n sit Rig h t A sc e n s io n

St a r . of A c ro ss Re d u c e d t o
La m p . Me a n L m e .
* M e a n T1 m e .

Virg in is
’
W
° h
56
m
8 8 h
48
m

Cm
e 1 1 3 1 2 2 2 66

43 o a 28 24 W 9 06 1 8 93 . 8 58
20 Ca n . Ve n 41 07 W 9 1 2 09 .
90 9 04

C U rsa M a j 55 28 W 9 1 9 00 . 84 1 1

Gr . 2 00 1 .
72 56 W 9 22 4 7 60 . 1 4
1 7 HCa n . .
37 43 9 29 21

77 U rsa M a j .
49 50 E 9 42 3 9 58 .
34 42 -
52
77 B o o t is 1 8 55 E 9 48 54 94 .
9 40
1 1 B o o t is 27 53 E 9 55 3 7 63 .
9 47 40 8 3
or D ra c 64 52 E 10 oo 45 68 .
9 52 48 27

A ns . By t he m thod of e 1 04 , A T, 7
m
56 -
73 ,
8
6

a W and aE

By t he m t h d of e 7 o 10 , A T, 56 8
-
74 . :t o . 02 ,

a, c

The following t stars wer observ d for t ime at

6 . en e e

Washington D C ( q5 3 8 on j une 8 9 6 with a , . .

°
22 ,
1 ,

sid r al chr nom ter G iv n t h following data compute

e e o e . e e ,

t h chronomet r c rrection on local sidereal time

e e o

Po s1 t io n Co rre c t e d T ra n sit R1 h t A sc e n sio n

of A c ro ss e d uce dt o
La m p . Me a n L 1n e .
*
Me a n Tm i e .

h h
E 1 s 10
m
1 s 10
m
O4a 1 4 .

E 1 5 1 3 1 5 1 3
0 0 0 0 0 0 0 0 0 0 0 E 1 5 20 37 . 07 1 5 20

E 1 5 22 41 . 21 1 5 22

E 1 5 27 1 5 . 1 1 1 5 27

W 1 5 35 30 . 1 8 1 5 35
W 1 5 44 05 . 86 1 5 44 06 -
54
W 1 5 47 52 -
19 1 5 47 51 -
32
W 1 5 53 19 42 0 1 5 53 1 9 -
92
W 1 5 59 59 -
56 1 5 59 59 -
92

A the m t hod
n s. By e o f 1 04 , A T, c

OEand a W

B y the method f § o 1 0 7, A T, j: c

aEand a w
*
Tra n s it d fo r d iu rn a l b e rra t io n pv i ot in e qu a ly it d
l
c o rre c t e a , ,
an

in c in a t io n .
 1 33 . Q UE S TION S A N D E XA M P L E S . 49

7 uppose
. t hat
S a striding l v l carries a continuou
e e s

graduati n of one hundred divisions each one -twentieth of an

o

inch l ng and that each division represents

o , second of arc one .

B y about h w much does the o which is t h longitudinal

a re e

section of the upp r inner surface of the l v l tube d part

e e e e

fr m t h cho rd of tha t arc joining the end gradua t ions

o e

A inch ns .
1 50 G E OD E TIC A S TR ON OM Y . 1 34 .

C H P TE RA V .

TH E Z E N ITH TE LE S O PE A N D TH E D E TE RM I NA TIO N O F C
LA TI T U D E .

Th e P i cipl
r n e of t h e Z en it h Tele scope .

The zenith distance of a star when n the meridian

134 . o

is the difference between the latitude of the station of b

vati n and the declination f the star H ence a measurem nt
o se r

o o . e

of the meridional z nith distance of a kno wn star furnish s a e e

determination of the latitude In the zenith telescope or

H orr bow-Talcott m eth d f determining the latitude there
.
,

e ,
o o

is substituted for this measur m nt of the absolut zenith e e e

distance of a star t h measurement of the small lif of e c e re n c e

zenith distances of t w stars culminating at about the sam e o

*

tim on opposite sides of t h z nith The e ffect of this sub

e e e .

s t it t i
u is the attainm nt of a much higher d gre of pre
on e e e

i i
c s on arising from t h increas d accuracy of a di fferential
,
e e

measurement in general over t h corresponding absolute

, , e

measuremen t ; fr m the elimination of the use of a graduated

o

circle f in the measurement ; and from t h fact that the com e

p ut d result
e is a ffected not by the error in estimating the ,

absolute value of t h astronomical refraction but simply by e ,

the rror in estimating the very small di fference of re fraction

e

of t w stars at nearly the same altitude

o .

wh m
p m ply
A is i d to lm in a t e t t he in s t a n t it t he e ri d ia
l l
st a r sa cu a en c ro s s e s n .

mp
1 Th e z en it h t e e sc o e c a rrie s a g r d u t e d c irc e
a a . b u t it is u se d Si

fin de r o r se t t in g c irc e l and it s re a d in g s d o n o t e n t e r t he uted

l
as a , co

re su t .
1 52 C TR N M Y 35 G E OD E TI AS O O . 1

purpose but two w re placed upon this instrument so that

,
e

increased accuracy might b secured by reading both B y e .

means of the clamp at G and the tangent screw at H operat ,

ing upon the sector I the t lescope may be brought to any , e

desir d inclinati n o

The o b ject -glass has a clear aperture

e .

cm ( in ) . z : .

in diameter and its focal l ng t h is ,

cm e .

Th ey piece has a magnifying power of

e e diam ters The 1 00 e .

focal plane of the objec t -glass lies in the rectangular brass b ox

shown at f The m m t screw of which the graduated

. 1 c ro e er ,

head is shown t K controls a rectangular brass frame liding a , S

in parall l guides within this b

e The movabl line with ox . e

which t h star bisections are m ade is stretch d across the

e e

s liding frame While in use the object -glass is so focused as

.

to mak the focal pl ne coincide with the plane in which this

e a

lin m ves
e o .

To facilitate counting the whole turns of the micrometer

s crew a small brass strip is placed i n one side of the field of
v iew of t h ey piece early in the plane of t h microm ter
e e n e e

line Th edge of the strip is filed int notches

. e in o .

apart The pitch f the screw being

. l the micrometer o n .
,

line app ars to m ve one notch along this comb for ach m
e o e co

p l t
e turn
e of the screw The whole turns are thus read from .

the comb and the fractions , read from the h ad of t h a re e e

screw which is graduated into one hundr d equal divisions

,
e .

I n F ig drawn in a vertical plane thr ugh the center

. 2 1 ,
o

o f the t lesc p let 0 be the optical c nter of the object

e o e ,
e
*

g lass Let. S be the position of a star The s t ar image is .

formed at the focus T which is necessarily in the line S O ,

produc d If the star is to appear bisected the micrometer

e .
,

line must be placed at T If another star later occupies the .

l p wh
yp w p m
Th e pt ica
o l ce n t er o f a e n se is t h a t o in t t h ro u g h ic h a ll in c id e n t
ra s a ss it ho ut e r anen t c ha n g e o f d ire c t io n .
 1 3 6 . A D J U S TM E N TS . 1 53

p si t ion S it image will be formed at T in S O produced

o
’
, s
’
,
’
,

and to mak a bisection the micrometer screw must be turned

e

until t h micromet r line is at T Th record d numb r of

e e
’
. e e e

tu rns of t h micrometer screw re q uired t move t h line from

e o e

T to T gives a measurement f the linear distance TT

’
o .

F the small angles conc rn d this linear distance is pr por e e o

the angle TO T the q ual of S OS H enc the

or
’
l t
’
ti ona o , e . e

observ d mov m nt of t h micrometer screw gives a measure

e e e e

m nt f the di ff rence f zenith distanc s S OS of the two

’
e o e o e , ,

s tars I n this particular instrument t h pi t ch of the scr w

.
, e e

being ab ut in and the focal length O T about

o i . n .
,

t urn f t h screw m asur s an angl f abou t sin “

one o e e e e o
-
’

45 9
or about
Th m re c mmon form of zenith telescope di ffers fr m
e o o o

the h re shown in having the telescop moun t ed cen

one e e ce

t i lly
r ca one side of the vertical axis instead of in fr nt f
o n o o

it as in this case ; in having a clamp which acts dir c t ly upon

, e

the horizontal axis in t h plac of the clamp at G acti ng on e e

t h sect r I ; and in having only one latitude level inst ad of

e o e

t wo .

j tm t Ad us en s .

1 3 6 The v rtical axis must be made truly vertical

. In e .

adj usting and using the instrum nt it will be found e con ven

i t to hav two f t h thr e foot -scr ws in an east and wes t

en e o e e e

dir ction The v rtical axis may be made approximately

e . e

vertical by use of the plate l v l if there is on t h i t e e , o ne e n s ru

ment and t h final adj ustm nt mad by using the lati t ude
, e e e

l m wh i y
y m ly ll
Th e va f t be d in e d it s u ffi c e n t a c c u ra c

l m m
ue o o ne t u rn c a n no e te r

p pl p by wh h
b y su c in e a r h Th e g iv h i
l v l
e a s u re en ts . a re en e re e re to u s t ra t e

w ll
the rin c i e in vo ve d Th e in d ire c t ic t he is di
m
. ro c e s s a ue or

n a rily dete r in e d i be f o u d d e sc rib e d

n in 1 5
-
8 16 4 .
1 54 G E O D E TIC A S TR ON O M Y . 138 .

level The process i n each cas is precis ly the sam as that

. e e e

of using the unadj usted plate l vels of an engine r s transit e e

’

to adj ust its ver t ical axis .

The horizontal axis mus t b perpendicular t t h ver t ical e o e

axis This may be t s t d aft r t h v rtical axis has been

. e e , e e e

adj ust d by reading t h s t riding level in both f its positions

e , e o .

I f the h rizontal axis is inclin d it must be made h rizontal

o e , o

by using t h scr ws which chang the angle betw e t h hori

e e e e n e

z on t l and ver t ical axes

a .

1 3 7 Th lin f c llimation must be perp ndicular t the

e e o o

h rizontal axis I f the instrum nt is of the form shown i

. e o

o . e n

F ig this dj us t ment may b mad as f an astron mical

2 0, a e e or

transi t 8 6 ) by r v rsing t h h rizon t al axis in the Y I f

. o

e e e o s .

t h instrument is of t h f rm in which the tel sc pe is

e e o e o eccen

tric with r sp ct t t h vertical axis t h meth d of making

e e o e e o

I t may b made as
,

t h test must be mod ified accordingly

e . e

for an ngineer s transit b t using t w fore and t w back

e
’

, u o o

points t h distanc apart of each pair of poin t s being mad

,
e e e

d uble the distanc from the v rtical axis to the axis of the
o e e

t l scope O a single pair of p ints at that dis t anc apart

e e . r, o e

may be us d and the horizo tal circle trusted to d termin

e n e e

wh n the instrum nt has been turned 8 in azimuth If

e e 1 0
°
.

one consid rs the allowable limit f rror in this adj ustm n t

e o e e

(
se e it b comes evident that a telegraph pole small
e or

tre if fi i t ly d i t t from the instrument may b

e ,
su c en s an , e

assum d t be of a d iameter qual to t h re q uir d distance

e o e e e

b tween the t w points O a single point at a kn wn dis

e o . r, o

tance may be used and a computed allowance mad on t h

horizontal circle for the paralla x of the point when the tele
e e

scop is changed from one of it positions to the other

e s .

1 3 8 D uring daylight the object -

. glass should b carefully e

focused on the most distant well -d fi d object available t e ne o

insure that stars may b e seen at night A neglect to do thi

,

. s
1 56 G E OD E TIC A S TR ON OM Y. g 1 40 .

of t h field by turning t h instrument a b out its v rtical axis e

I f t h b isection is t still perf ct half the correction should

e e .

e no e ,

b emad with the micrometer and half with the slow -moti n
e o

screws which ro t ate the whole eyepi ce and reticle about the e

axis of figure of the telescope The adj ustment should be .

car fully tested at night after setting the stops by taking a

e ,

s ri s of pointings upon a l w -moving star as it crosses the

e e S o

fi ld with t h t lescop in the meridian I f the adj ustm nt

e e e e . e

is per fect the m an reading of the micr meter before the star
e o

reach s the middle of the fi ld should agree with its mean

e e

reading after passing the middl xcept for the accid n t al e , e e

errors of pointing It is esp cially important to mak this. e e

adj ustment car fully for the tend ncy f any inclination is to
'

e , e o

introduce a t t error into the computed values of the

co n s a n

latitude .

t Th e Ob serv in g Lis .

1 40 B efor comm ncing the obs rvations at a station an

. e e e ,

obs rving list should be pr pared showing for each star to e

b observed it catalogue num b er or its name its magni t ud

e , ,

e , s , e,

m an right ascension and declination ( at the beginni g of

e
*
n

the year) z nith distance whether it culmina t es north or

,
e ,

south of the z nith ; and for each pair the setting of the ver
e

tical circle (the mean of the two zenith distanc s ) the di ffer e ,

ence of the zenith distances with it algebraic ign as given by s S

formula and finally the microme t er comb setting for each

star F the purposes of the bserving list the right
. or o a sc e n

si ns to within one second of time and the declinations and

o ,

derived quantities within one minute of are su fficiently a re ,

accurate I f the micrometer comb r ading is one minute per

. e

n otch d the middle n tch is called the comb setting for

, an o 2 0,

Fo r t h e d e fi n it io n o f t h e m ea n pl a ce o fa s t a r se e 3 7, 3 9 .
g 141 . OE S E R VIN O L IS T . IS7

one star is t h half di ference of z nith distanc s for one

20 e - f e e

s t ar and , t hat half -di ff rence f t h ther star of a pair

20 e or e o .

The re q uisites for a pair of stars for this list are that t h ir e

right ascensions shall t di ffer by more t han to avoid n o

t oo gr at rrors from instability in the rela t iv positi ns f

e e e o o

differ nt par t s of t h ins t rument ; nor by less than

e e that I
“
,

int rval being re q uir d to t ak t h readings upon the firs t s t ar

e e e e

and prepare for the s cond star f a pair ; that their di fferenc e o e

of zeni t h distanc s shall not exceed half the leng t h of the

e

micrometer comb for t h usual t ype of instrum nt ; t hat

,
20
’
e e

each star shall be brigh t n ugh to be s n distinctly — t e o ee no

fainter than the seven t h magnitude for t h instrum nts h re e e e

described ; and t hat no zenith distance shall xceed t e o

guard against too great an u cer t ain t y in t h refracti n The n e o .

selecti n of a s ries f such pairs from t h stars of a catal gue

o e o e o

r quir s much t ime and pati nce

e e e .

Th t tal range f the lis t in right ascension is gov rned

e o o e

by t h hours f darkness t h propos d dates of observa

e o on e e

tion and by the convenience of t h bs rver Th third f

,
e o e . e o

the above conditions may perhaps be us d m r conv ni n t ly e o e e e

in this form t h sum f the t w declinations must t di ff r

: e o o no e

from twic the latitude by more than

e T pr par the o e e

list the latitude of the stati n should be known within a o

minu t e I t may possibly be s cured from a map ; if t then

. e no ,

from a sex t ant observation of the S or from an obs rvati n un , e o

o f th m ridional zenith distance of a star with the find r

e e e

circl of t h zenith telescope

e e .

1 41 The stars selected should be such that th ir com

. e

p ut de m an places may
e be made to d pend in each individual e

case upon observations at several different observa t ries o .

The declinati n of a star as deriv d from bs rva t ions at a

o e o e

single observat ry will not in g neral be su fficiently accurate

o e

for the purpose in hand .

 58 G E OD E TIC A S TR ON O M Y . 141 .

I f the bserver knows that in making the computation an

o

ample collection of catalogu s of original observations at e

*

e ch f various observatori s will be available he may select

a o e ,

all t h suitable pairs he can find in any t i list of stars

e ex en s ve ,

say t h B ritish A ssociation Catalogue or any of t h G r n

e
“

, e ee

wich Ca t alogues and trust to finding af t erward in the vari us

, o

catalogues a su ffi cient numb r of obs rvations upon each star e e

at various obs rvatories t giv an accurate determination of

e o e

its place This is the usual procedure in the Coast and

.

G eodetic S u rvey A computer in the o ffice at Washington

.

calculates the declinations of the stars w hich have be n e

o bs rv d for latitude by bringing together in a least square

e e

adj ustment all the observations upon each star that he finds
in his large collection of star catalogues
I f the necessary coll ction of catalogues of original b
.

e o se r

va ti is not known to be available the observing list had

ons
'

best b made up from star lists in which the d eclinations given

e

a re the result of the compilation and c mputation of original o

observations at various observatories as utlined above o .

A mong such available lists of m an places are those in the e

various national eph merides ; P reston s Sandwich Island Lis t e

’

in the Coast and G odetic S urvey Report f 8 8 8 A pp ndix e or 1 , e

No 4 pp 5 5 3 ; the list giv n in A ppendix No 7 pp

. 1 , . 1 1— 2 e .
, .

83 —1 2
9 C G S R
, port for .8 7 6 ; and B
. oss list in.the e 1
’

re

port f the S urvey of t h North rn B oundary from the Lake

o e e

of the Woods to the Rocky Mountains pp 59 6 5 These , . 2— 1 .

lists given in about t h order of the accuracy of their star

a re e

places when reduced to the present time the mor recentl y

computed places b eing more accurate if other conditions are
, e

about the same I n the report of the M exican B oundary

,

S urv y of 8 9 9 3 which is about to b e published there will

.

e 1 2— , ,

wh m y m pl l
y
A n in d ic a t i on o f a t is ea n t b an a e co ll e c t io n o fc a t a o g ues
m a ybe g i ed b
a n re a di g n 37 .
1 60 G E OD E TIC A S TR ON O M Y . 1 43 .

n early zero so as to make t h computed latitude for the

, e

station nearly fre from any effect of error in the mean value
e

o f the m icrometer screw .

t Direc ion s for Ob serv in g .

1 43 The instrument being adjusted set t h vertical circl

. , e e

to read the mean zenith distance or circle set t ing as ,

”

marked in the observing list of the first pair D irect the , .

tel scope to that side f the zenith on which the first star of O

P ut the bubble of the latitude level

e

t h pair will culminate

e .

n early in the middl of the tube by using the tange t screw e n

which changes the inclina t ion of the telescope P lace the .

microm ter thr ad at that part of the comb at which t h star

e e e

is xpect d as sh wn by the observing list Watch the

e e ,
o .

chronomet r t keep posted as to when the star s h ould

e o

appear When t h star en t rs the field place the micrometer

. e e

thr ad approximately upon it and center the ey piece over

e , e

the thread A soon as the star com s within the limits indi
. s e

ca et d by the v rtical lines of the reticle bisect it carefully

e .

A the star moves along watch the bisection and correct it if

any error can b e detected B ecause of momentary changes
s

in the r fraction the star will usually b e s en to move along

e ,
e

t h thr ad with an irregular moti n

e e w partly above it w o , no , no

partly below The mean position of the star is to b covered

. e

by the line A att mpt is being made to secure a result

. n e

which is to be in err r by m uch less than the apparent width o

of the thread henc e too much care cannot be bestowed upon

,

the bisection It is possible but not advisabl to make

.
, e ,

several bise ctions of t h star while it is passing across the e

field A soon as the star reaches the middle vertical line of

. s

the reticle read ff promptly from the comb the whole tu rns
o

of the micrometer read the level and then the fraction f a , , o

micrometer turn in divisions from the micrometer head S t

, , . e
 1 4 5. CT N R RV NG D IR E 6IO S FO OE S E I . 1 1

pr mptly for the next star even though it is not expected

soon I n setting f the second star of a p i ll that is
o ,
’

. or a r a

necessary is to reverse the instrument in azimuth and set the

micrometer thread to a new position .

1 44 The instrument must be manipulat d as carefully as

. e

possibl Es pecial care should be taken in handling the

e .
'

micr meter screw as any longitudinal force applied to it pro

o ,

duc s a fi
e of the telescope whic h tends to enter the
e x u re

result directly as an error The last motion of the micr meter . o

head in making a bisection should always be in the same

direction ( preferably that in which the screw acts positively
against its opp sing spring) to insure that any lost mo t ion is
o ,

always taken up in direction The bubble should b read

one . e

promptly so as to give it as little tim as p ssibl t change

,
.
e o e o

its position after t h bisection The desired reading i that

e . s

at which it st od at the instant of bisection A void car fully

o . e

any heating of the level by putting the reading lamp warm ,

br ath or face any nearer to it than necessary D uring t h

e , . e

observation of a pair the tang nt scr w of t h setting circl e e e e

must not b e touched for the angle between t h l vel d e

telescope must b kept constant I f it is necessary to relevel

, e an

e .
,

to keep the bubble wi t hin reading limits use the tangen t , :

screw which changes the inclination of the t lescope Even e .

t his may introduce an error due to a change in the fl of , e x u re

the telescope and should be avoided if possible

, .

1 45 F fi t -class observing it is desirable to have a

. or rs

rec rd r H may count seconds from the face of the

o e . e

chronom ter for a minu t e before culmination in such a way

e

as to indicate when the star is to culminate according to the

right ascension given on the observing list taking t h known , e

chronometer correction into account S uch counting aloud

serves a d uble purpose I t is a warning to be ready and
.

o .

indicates where to look for the star if it is faint and di ffi cul t

1 62 G E OD E TI C A S TR ON OM Y . 1 46 .

to find It also gives for each star a rough check upon the
position of the azimuth stops and warns the o b server when
.

they need readj ustment It is only a rough check b ecause

t h o b serving list gives mean right ascensions ( for the begin
.
,

ning of the year) instead of apparen t right ascensions for the

date B t in view of g 6 7 it is su fficiently accurate The
. u 1 .

bserver can easily make allowance for the fact that all stars
will appear to b e fast or slow according to the o b serving list
o

by about the same interval to 5 ( h di fference between 0

° “
t e

the mean and the apparent place) If a star cannot b e

,

observed upon the middle line on account of temporary ,

interference by clouds or tardiness in preparing for the b o se r

vation observe it any where within the safe limits of the field
,

as indicated by the vertical lines of the reticle and record the

chronometer time of obs rvation e .

E XA M PLE OF C
RE O RD .

1 46 S t a t io n — N o 8 n e a r Sa n B e rn a r in o Ra n c
p
A riz o n a d h
m
.

Z
. .
, ,

In s t ru e n t — W u rd e m a n n e n it Te le sc o e N o 2 0 h . .

O b se r v e r—
J . F . H .

D a le — A u g u st 9 , 1 892 .

M i c ro m e t e r . Le v
e l .

No o f
.

Pa i r. B
S t a r No
.A C . .
N .

S
or
.

T u rn s . D 1 v 13 1 o n s . N . S . Re m a r k s
.

19

56
9
o
Sk y e r
C ro n oh
p mf l y
ect

e te r 2 1
c le a r.
'
f a st .

52 9 1 7
1 7 9 52

52 5 1 7
1 6 4 51
1 64 G E OD E TIC A S TR ON OM Y g 1 49
(

. .

continuously from one end to the other with the z ro neares t e

the eyepiece the inclination of the ve tical a x is i

,

, r s

iW —-
( i
'

l S
) ”

If the zero is nearest the o b j ect glass the algebraic ign mus t - S

be chang d from that given above

e .

The inclination of the vertical axis makes the south zenith

distance too mall b y the amount indicated by ( 55) or ( 56)
S

and the north too large by the same amount H ence the .

'

c rrection to %(
o ) is { ( z ) ( z
'

or the
a
corre n n
’
s
4
s p d i
on g xpression
n e

1 49 The refracti n makes each apparent zenith distance

o

too small I f R and R represent the refraction for the south

.

’
.

and north s t ar respectively the correction to (

, ) i , z z
’
s

( R and to } ( is } ( R i z 1

Let m be the correction to the apparent zenith distance

of a south star observed slightly ff t h meridian to reduce it O e

to what it w when on the meridian and m the correspond

as ,
’

ing r duction for a north star observed ff the meridian Th

e o . e

c rrection to £ ) will then 3 be é -

( m z
’
m and m are ’

of course zero in the normal case when the o b servation is

o

made in the meridian .

F ormula ( 54) may w be written for an instrument with

no ,

the level graduated both ways from the middle ,

i(6 + (M >
'
n

+i < R ~
1 0+
 F ORM UL AE 1 65
1 51 . .

This is the working formula for the computation b ut the

alues of the last two t rms may be conveniently ta b ulated
,

v e .

1 50 The di ff rence R R being very small the variation

’
e

f the state of the atmosphere at the time of o b servation from

. ,

mean state (see r fraction ta b les

o

9 4 9 7) may be
it s e 2 —2
,

neglected exc ept for stations at high altitudes It has been

,
.

shown by the investigations of the laws of refraction which

,

have b en r ferred to in 6 7 6 9 that this di fferential f

e e — , re ra c

tion for the mean state of the atmosphere is with su fficient

, , ,

accuracy for the present purpose ,

R R 7 sin ( sec
’ ’ 2
z :
57 . 2 z
) 2 .
(5 )
8

By computation from this formula the value of the term

M R R ) of formula ( )
5 7 has
’
b een ta b ulated in 3 4 for , 0 ,

t h argumen t s fi
e as directly observed with the m i m
e c ro

eter and the z nith distanc e e

I f t h s t ation is so far above sea-level that the m an

.
,

e e

barometric pr ssure is less than say of 76 mm

e , 0 .

th e mean pressure at sea lev l it is necessary to take this fact

- e ,

into account by diminishing the values of the di ffer n t ial e

refracti n given in 3 4 in the ratio of the mean pressures

o 0 .

That is if the mean pr ssure is % less than at sea -level

, ,
e 10

diminish the values of 3 4 by if % less sub t ract 0 20

and so on I p t of the table shows that this allowance

ns ec lo n

need only b e made roughly since the tabular values are smal l
.

, .

1 51 The value of } and of its equal

.

Z 1
3 is tabulated in , ,

3 5 0 The table gives directly the correction to the latitude

.

for any case of a star observed ff the meridian but within o ,

one minute of it If both stars of a pair are observed ff the. o

meridian two such corrections must be applied one for each ,

star F the di fficult derivation of the formula from which

. or
1 66 G E OD E TI C A S TR ON OM Y . 1 52

this table is computed see Ch t A stronomy auve n e

’
s v o l. II

pp 3 46 3 4 7 ; or D oolittle s P ractical A stronomy pp

, .

.
, , .
5 5,
0

50 6 .

E XA M PLE OF C O M PU T A TIO N .

1 52 St a t io n — N O 8 n e a r Sa n B e rn a r in o Ra n c , A riz o n a
p
d h
m l
.

Z
. , .

In s t ru e n t — W u rd e m a n n e n it Te e sc o e N o 2 0 h . .

Ob se rv e r— J F H . . .

D a t e — A u g u st 9 1 8 92 , .

L e t -
[ f b a n d pa ge f
o C om pu t a t io n .
]
M 1c ro m e t e r
1 t u rn 1 00 dv 1 .
Le v e l .

1 d iv .

D e c l i n a t io n .

Re a d ln g D i . ff . Z . D .
fl

’ ’
t . a . t . a .

7 52 8 S 22
88
.

7 544 N . 1 6 5
7 566 N 24
-1 1
.

7 586 S . 1 3 1 7 9

7 63 1 N 30
—2 1
.

7 66 2 S .
9

[ g
R i Ot -
O a n a pa e
g
’
o f C om
pu t a t io n ] .

Su m a n d M e a n C o rre c t io n s .

O f D e c l i n a t io n s .
L a t it u d e . Re m a r k s.

M 1 c ro m e t e r . Le v e l . Re f ra c t io n . M e rid 1 a n .

3
’
01 . 05 + 0 . 04 31
°
1 9
’

5 42 . 26 + 0 . 04 0 . 08 58 .
71

—1 0 57 . 01 + 0 . 64 -0 . 1 8 58 .
77
1 68 GE O DE TIC A S TRON O M Y . 1 54
.

The corrections for refraction were o b tained b y su b tract

ing % from the values of 3 4 the b arometric pressure
20 0 ,

being only about four-fift h as great at S B ernardino as at s an

sea level
-

The latitude is obtained from each pair b y adding the

.

various corrections algebraically to the mean declination as ,

indicated in formula
With su fficient accuracy for some purposes the d i im i 1n sc r

nate mean f all the individual values may be taken as the

O

final value of the latitude I f the best or most proba b le .

, ,

value is desired the procedu re outlined below must be fol

lowed .

C m b i ti
o f I di id l R
na lt b y L
on o t Sqn v ua e su s ea s u a re s .

1 54 Let us first deal with the simplest case Namely

let it b e supposed that p separate pairs hav been observed
. .
,

on each of nights at a s t ation each pair being observed on

n
’

every night F this case it wi ll b e found that the indis

,

. or

criminate mean is after all the most probable value of the

. ,

latitude but t h principl s develop d will be found useful i

, e e e n

dealing with other more di fficult cases in which this is not true
The differences A obtained by su b tracting the mean r sult
.

for any one pair from the result on each separate night for
that pair are evidently independent of errors of declination .

We may compute from these differences or residuals t h

probable error of a single o b servation
, , e

This error of e

o b servation includes the observer errors instrumental errors

.

5 , ,

an d all external errors except the errors of the assumed

d eclinations Then by least s q uares
.

( 59)
k mm
ly
*
Th is q u a re b ra c et j is h e re d t o in d ic a t e a t io n it fre
k l
s u se su ,
as

quent is in t e x t -
boo s on e a st s q u a re s .
 54
1 C
. OM N T ON R U T BI 69 A I OF ES L S . 1

in which [A A ] stands for the sum of the squares of all the

residuals A obtained from all the pairs .

The probable error of the mean result from any one e

pair may also b e compu t ed fr m the observations by the

,
o

formula

in which is the residual obtained by subtracting the mean

v

result for the station from the mean result for each pair .

Th re are p such r siduals [ ] stands for the sum of the

e e . vv

s q uares of these r siduals e .

L t be t h probable error of the mean of the two

e e, e

d eclinations evid ntly includes the declination errors of

. e
,
e

the t w stars of a po F rom the ordinary law of transmis

an .

sion of accidental errors

Whence

may thus be o b tained from the observations for lati

e, ,

tude by substituting the values and computed by ( 59)

, e e
,
and (6 ) in 0

being the same for all pairs it is evident from (6 ) that

n
'
, 1

the means from t h various pairs have e q ual weight The

e .

m st probabl valu for t h latitude is then the indiscriminate

o e e e

mean of the results from the separat pairs or what is e , ,

numerically the sam in this case t h indiscriminate mean of

e , e

all t h individual results for latitude

e .
1 70 C TR N M Y G E OD E TI AS O O . 1 55 .

The probable error o f the final result for latitude is

The simple case j ust treated seldom occurs in p

1 55 . ra c

tice O bservations upon c rtain pairs are missed on some of

. e

the nights by accident or by cloud interferenc ; work may e

be entirely stopp d by clouds after half the observations of an

e

evening have been made ; or on the later eve ings of a series n

the observer may purposely with a vie w to more ffectual , e

elimination of declination rrors includ in his observing lis t

e , e

certain pairs which have not b for b en observ d at that

station in the place of pairs which have already b een observed
e e e e

onc e moreor .

In the usual cas then a total of p pairs are observed

e , , ,

pair No being observed times

1 on nights) pair n n

and t h total num b er of observations

.
, l ,

No . times
2 n, , e

isn n n n,

B y the same reasoning as before we have b y the ordina y

, ,

, r

least s q uare formula the probable error of a sing le observa

,

tion

T o o b tain
pro b able error o f the mean result from
t he e
,
any one pair with rigid exactness it is necessary to take into
account the fact that di fferent pairs must now b e given
d ifferent weights since some observed more times than
,
a re

others To do this would make the computation consider

.

ably longer th n is otherw ise necessary F ortunately i

a .
, n ves

t ig t i
a of the numerical values concerned shows that the
on

results are abundantly accurate i f formula (6 ) is here used 0

1 7 2 C A TR N OM Y
G E OD E TI S O .

and may w no be computed from the known values of e,

and e .

The m ost probable value 45 for the latitude of the station ,

i the weighted mean of the mean results from the various

s

pairs or ,

wA b . w
[ e]
20, “

l
“
we w
I e " "
w
[ ]

in which 95 is the mean result from the first pair

, , 5,
(7 from the
second pair and so on , .

A lso the probable error f this result is

, o

( 0 4 55) l
(e Ol ] av

in w h ich [w ) stands for the sum of the products of the

v

weight for each pair into the square of the residual obtained
by subtracting 95 from the mean result for that pair and [w]
, ,

is the sum of t h weights e .

In case two north stars are observed in connection with

the same south star or i and the computati n is , v ce v e rsa ,
\
o

made as if t w ind pendent pairs had been observed the

o e ,

weight f each of these pairs as given by (69) should be mul

o

t ip li d by e to tak acc un t of t h fact that they are bute o e

partially independ nt Sim ilarly if three north stars have

e .

been observed in connection with the same south star the

weigh t s from (6 9) for each of the three resulting pairs should
be multiplied by
If however a giv n north star is obser ed in connection
, ,
e v

with a certain south star on a certain ight or nigh ts and n ,

C v y R p p
P p f h C p k v yTi g l i ) p
d d i
o a st an 8 Geo
55 P f
et c i l Su r e e o rt , 1 8 0, . 2 ; or ro e ss o n a

a e rs o f E gi t Ne ( or s o n n e e rs , o . 24 La e Su r e r an u at on , .

62 5 .
 1 57 . COM N T N O R U T BI 73 A IO F ES L S
’

. 1

on a cert in t l night or nights is bserved in conn ction

a o t er o e

with some t / south star the case is di ff rent and the m

o z er , e , co

putati n is su fficiently accurate though not exact if each f

o , , o

such pairs is given the full w ight resulting from e

I f v ry few pairs are obs rv d more than once at a sta

e e e

tion the determination of from the latitude observations e,

obviously fails and it must be estimated in som other w y

,
e a

from t h t ca t al gu s f xampl
/

e s ar o e , or e e .

1 57 A an exampl of the application

. s f formul ( 4)
6 to
e o a

t h process e f c mbining t h various values for the lati

o o e

t ud e f s t ati n No 8
o th Mexican B oundary Surv y may
o . on e e

b given
e A t t his station
. bserva t ions were made on 7 5 1 00 o

pairs 5 f the pairs b ing observed twice each and t h

,
2 o e , e

oth r 5 once each The obs rvations ext nded ov r f ur

e 0 . e e e o

nights .

The um of the s q uares of the fif t y residuals A obtained

s , ,

by subtracting the mean f each pair which was observed or

twice fr m ach of the two values from that pair was

, o e ,

s q uare s conds The probable error of a single observation

e .

was then from ,

: l: O . 2 1 3 .

The indiscriminate m an of t h 7 5 results one from ach e e , e

pair was found to be 3 9

,
B y subtracting t his 1
°
1
’

value from each of the 7 5 s parat values s q uaring and e e , ,

adding it was found hat [ ]

, s q uare seconds t vv .

F rom (6 5) it followed that

Vo . o 71 4 j: o . 2 67 .
I74 G E OD E TI C A S TR ON O M Y
. 1 58 .

The term + ;
7 3
of formula ( 7)
6 is here
50 and

( 5)
6 2 -
0 0 3 79 .

Whence from 6
( )8

or e, l
: : o . 1 83 .

ubstituting t h values of and in

S e the weights e,
2
e
2

for the pairs observed twice was found to be and for

th se up n which
o o observation only was made one ,

S ince it is the l ti weights only which a ffect the final

re a ve

re S lt these wei ghts w re for convenience written

u and e

respectiv ly e .

The resulting weighted mean as indicated in ( 7 ) was 0

found to be 3 9 1
°
1
’

F rom the residuals corresponding to this value it was

found that [w ] s q uare seconds H ence
v
”
.

( 0
Vo -
. 000 94 9 : l: 0 O3 I °

t m i t i f Mi m t
De er na d L l V l
on o cro e er a n ev e a ue s .

1 58 The most a d vantageous method of determining the

.

screw value is to bserve the time re q uired for a close circum

O

polar star near elongation to pass over the angular interval

measured by the screw N ar elongation the apparent motion . e

of the star is very n arly vertical and uniform e .

That one of the four close i m p l given in the c rc u o a rs

Ephemeris namely 6 and l Urs Minoris and 5 Ceph i

, ,
a
, , a ,
1 e ,
1 76 G E OD E TI C A S TR ON OM Y
.
5 59I

is considered desirable I f desired an observation may be .

,

made at every half-turn or even every quarter -turn by allow ,

ing an assistant to read the level .

1 59 I n F ig let P b the pole S the position of the

. . 22 , e ,

star wh n observed t ransiting across the micrometer line and

e ,

S its position at elongation

E
The small circle SS is a por .
E

tion of the apparent path of the star Let SK be a portion

of the vertical circle through S limited b y the great arc PS
.

E .

Let the length ( in seconds of arc) of SK measuring the ,

change in zenith distance of t h star in passing from the posi e

tion S t elongation be called o Let the sidereal interval 2

of time in seconds from position S to elongation b e called

.
,

I n t h spherical triangle SKP the angle at K is and tha t

, , 1 .

at P is in s conds of arc 5 Therefore

,
e ,
1 1 .

cos 6 sin ( 73 )

The various values of corresponding to o b served value 2 s

of might be directly computed from

1 B t the compu u

t ti
a b comes much asier and shorter though based upon
on e e ,

a more di fficul t concept on if one proceeds as follows i ,

(7 3 ) may be writ t en
Z :

F rom the known e xpansion of the sine in terms of the a re

there is obtained
sin ( 1 5 1 ) ( 1 5)
1 sin I
” —
1 sin 1

By su b stitution from ( 7 5) in ( 74)

z : 1 5 cos 6 i r sin T5 05
3 sin
 1 60 . M ICR OM E TE R VA L UE . I77

inspection of ( 76) it is evident that 5 cos 6 is the rate

By 1

of change f ( in seconds of arc per second of time ) t l g

‘

a e on a

I f the rate of change of were always constant at this

o e

ti
on .
2

value for any position of t h star w uld be ( 5 cos 6 )

, 2 e o 1 r .

Compare this with and it becomes evident that if the

obs rv d value f is corr cted by applying to it the small
e e o r e

correction 405 sin 3 05 sin the resulting ” “ 3

1 1
T 3
value is that t which the value of would be the same as the
a 3

actual value if inst ad of the actual star ther w re substi t u t ed

e e e

one whose motion is t i l t t h t t t 5 6 v er ca a e con s a n ra e 1 c os .

These c rr ctions o sin e § sin —

f a re

tabulated in 3 6 for the argument B y using t h

0 signs 1 . e

th re indicat d th y may be applied directly to t h observed

e e e e

chr nome t er times The corrected times corresp nd t uni

o . o o

f rm moti n in a vertical great circle in t h place of the

o o , e

ac t ual mo t ion in a small circle .

1 60 S uppose t h reading of the level has not remain d

. e e

constant during t h observations The change of reading e .

indicat s that the inclination of the telescope as a whole has

e

chang d L t e b the r adings of the north and south

. e n , , s, e e

ends of t h bubble respectively at some selected time d uri g

e , , n

the ob ervations L t it be proposed to reduce t h observed

s . e e

tim s t what they would have been if the level readings h d

e o

throughout the observations I f and are the

a

b en
e n, , s, . n s

north and s uth readings respectively at a given observation

o , , ,

the correction to the inclination of the telescope t r duce it o e

to the st te of the level is evidently in seconds of arc

a n, , s, , ,

for a grad uation numbered both ways from the middle ,

1W
1 S
) (fl . a nd ,

in which is the value of one division in seconds of

a
’

noting the smallness of t h ta b ular values of 3

a re .

F rom e 06 ,
1 78 G E OD E TI C A S TR O N OM Y. 16 1 .

itis evident that with su fficient accuracy for the present pur
p se the rate of change f is 5 cos 6 ( in seconds o f o z 1

b
o a re

p second
er of t ime ) at any instant during the o servati ns
H ence the correction in seconds to the observed chronometer
o .

, ,

tim is e

— — n o — So —s —
i % l( fl S
) ( )i :l
: {( n
) ( n,

The upper sign is to be used for western elongation and the

lo wer for eastern The corresponding formula for a l vel with
. e

a graduati n numbered continuously from one end to the

o

ot h er with the zero nearest the eyepiece is

, ,

i jQ l )
Z — -S ( 72, + » 50

1 61 . H aving
applied the corrections of 3 6 and of 0 ,

formula ( 77) or to the observed chronometer times the

results are the times which woul d have b een obtained if the
,

star had moved uniformly in a vertical circle and the telescope

had remained fixed in position We may w subtract the . no

first corrected time from the middle one of the series the ,

sec nd from the next following the middle and so on thus

o b taining a series of values of the t ime interval corresponding
o , ,

to a known number of turns of the micrometer The mean .

of t hese values may be taken and from it the time interval t

in seconds corresponding t one turn may b e derived The
, ,

,
o .

val e of one turn expressed in seconds of arc is ( 5 cos 6 )t

u , , 1 .

To this must still be applied corrections for di fferential

refraction and rate of chronometer The refraction correction .

to the derived value f one turn is the change of re f raction

o ,

or di fferential refraction corresponding to one turn at the

,
1 80 G E OD E TIC A S TR ON OM Y .
g
1 63 .

r
o e e
f e e e
r . r
e r r
g n u .

C o rre c t e d
r r
C h ro n o m e t e r
t e
e m . o t o o
T im T im
t
a
md n
i
s .
t
C a C
a
e. c v
r
t N es .

o r e l
S l

r e u r u e 6 e
v o v
c
RT r
.

1
o C e t 0 e
M C L
5
L
l l I l

M e a n fo r 1 0 t u rn s ( ro m f a ll t he o b se rva t io n s) 3 1
m
:t

M e a n t im e fo r 1 t u rn

O n e t u rn
C o rre c t io n fo r re f ra c t io n
ra t e

F in a lv a ue l 62 . 01 6 : l: o
”
. 01 1

I f the values of both the level and the micromete

1 63 . r

are u nknown one may observe for micrometer value as out

,

lined above and also derive the value of the level in terms o f
,

the micrometer as indicated in 8 W may first compute 1 1 . e

the micrometer val ue omitting the level corrections ; then ,

derive the value of the level division from this approximate

value of micrometer The level corrections being now intro .

d d into the micrometer computation will be found to

u ce

modify it so slightly that a second approxi m ation for the level

value will not ordinarily be re q uired .

Re f ra c t io n a t t h is s t a t io n wa s on l y i? f t
o ha t a t se a -e l ve l .
 1 65 . E RR OR S . 1 81

I f no special obs rva t i ns for micrometer value have

1 64 . e o

b een made or if such observations have pr ved defective the

,
o ,

micrometer value may be derived directly from the latitude

observations Let 915 b e the mean latitud as d duced with
.
1 . e , e

an approximate micr meter value from all pairs for which the o ,

micr meter difference (taken S N ) was positiv ¢ the mean

o e , N

latitude fr m pairs with minus micrometer di fferences D t h

o , P e

mean of t h p si t ive micrometer di ffer nces and D t h m an

e o e , ,v e e

of the negative di ffer nces Then the corr ction t the e e

appr x imate valu of one turn is

. o

o e

Fo r methods of determ n ng the level value alone i i , se e

1 1 6— 1 2 1 .

s Di sc ss i
u on of E rror .

1 65 Th t . l affec t ing a zenith telescope

e ex e rn a e rro rs

observation are th s due to def c t ive d clina t ions and those

o e e e

d to abnormal r fraction
ue e .

The d clinations us d in the computati n have probable

e e o

errors which su ff ciently large to furnish much often more

a re i ,

than one -half of the err r in the final computed result This
,
o .

arises from the fact that a good z nith telescope gives res lts e u

b t little inferior in accuracy to t hos obt ined with the large

u e a

instruments of the fix d observat ries which are used in e o

d termining t h d clinations
e e e .

Th following three exampl s will serve t indicate the

e e o

m m l p w
m p l l
f o r u l a is n o t e x a c t f r t h e e a st s q u re o in t o f v ie t h a t is
Th is o a

ro b a b e v a u e o f t h e re q u ire d c rre c t i n
, , ,

ly
it d o e s n o t g iv e t h e st B u t it
m m l
o o o .

mp mp
g iv e s so n e a r t h e sa e n u e ric l re su l t s a s t h e e x a c t e st s q ua re t re a t
m
a a

en t a n d l e a d s t o s o s h o rt a n d s i l e a c o u t a t i o n t h a t it s u se is a d vis
l
, ,

ab e .
1 82 C E OD E TIC A S TR ON OM Y. 1 66 .

improvement in t h available declinations during the last f w

e e

years and the magnitude o f the declination errors t be

expected The pro b able error of the mean of two declina
, o

.
‘

tions was found to be 1

e, , f the list of st rs *

furnished to the U S L ake S urvey by P r f T H Safford

, 3 ; or a

. . o . . .

in 8 7 Similarly for t h list f stars furnished fr m the

1 2 .
,
e o o

C oast and G odetic S urvey O ffice for use in d termining t h

e

variation of latitude at the H awaiian I slands i n 8 9 9 3

e e

1 1- 2

i 8 ; for the list furnished to the M exican B oun

0 1

dary S urvey b y P rof Sa fford in 8 9 9 3 = i

e, r
:
z z .

”
8 . 1 2— , e, 0 . 1 .

S uch a high degree f precision as that of t h last t w exam o e

ples is only attainable b y an p-to -date computation from

o

many catalogues of many observatories B y the tim such . e

lists are available in print their accuracy has ordinarily

diminished considerably with the lapse of time .

Th errors in the computed di fferential refractions are

e

pr bably very small and it is not lik ly that they increase

o , e

much with an increase of t h mean z nith distance of a pair e e ,

up to t h limi t I f there were a sensible tendency as

e , ,

has been claimed f ll stars to be seen too far north or , or a ,

south on some nights —because of t h existence f a

, , e o

barometric gradient for example — it should b det cted by a , e e

comparison of the mean results for di fferent nights at the sam e

station Many such comparisons mad by the writer indicate

. e

that in zenith telescope latitudes there is no error peculiar to

the night The variatio in the m an results from ight t
. . n e n o

night was found in all the cases examined to be about what

should be exp cted from the known probable errors of b
e o se r

vation and declination .

1 66 The O are those made in bisecting the ’

o se rv er s e rro rs

star and in r ading the l v l and microm ter H er als may

.

,
e e e e . e o

lPp C p pp -
d i S v y R p p
P f i ro e ss o n a gi N a e rs o f the or s o f En n e e rs . o 2 4, 62 2 63 8

1C
. . .

d G
o a st a n P eo et c ur e e o rt , 1 892 . a rt 2 , . 1 58 .
1 84 G E OD E TIC A S TR ON OM Y . 16 7 .

position of the star The whole instrument b eing elastic the .

,

force required for even such a displacement is small A . n

experienced observer has found that in a series of his latitude

observations during which the level was read both before and
af t r the star b isections t h former readings continually
,

e , e

di ffered from the latter from to always in one ,

direction .
*

1 67 A mong the i t
. m t l may be mentioned n s ru en a e rro rs

those due t to an inclination of the micrometer line to the

: Is ,

horizontal ; d to an err neous level value ; 3 d to inclina

2 , o ,

tion of the horizontal axis ; 4 t h to erroneous placing of the ,

azim uth stops ; st h to rror f collimati n ; 6 t h to i g , e o o , rre u

l it y f micrometer screw 7 t h to an erroneous mean value

ar o ; ,

of t h micrometer scr w ; 8 th t the t b ilit y of the relative

e e , o 1n s a

posi ions of di fferent parts of the instrum nt

t e

Th fi t -m ntion d source of error must b e carefully

.

e rs e e

guard d against as indicated in 3 9 as it tends to introduce

e , 1 ,

a t t error
c on s a n The observ r even if he attempts to make
. e ,

the bisection in the middle of the field (horizontally) is apt to ,

make it on one side or the other according to a fixed habit .

If the line is inclined his micrometer readings are too great on

,

all n rth stars and too small on all sou t h stars or i

o , v ce v e rsa .

The err r from using an erroneous level value is smaller

o

the smaller are the level corrections and t h more nearly the e

plus and minus correc t ions in a s ri s b alance each other To e e .

insure that it h ll b negligible it is necessary to r level every

s a , e e

time the correction becomes more than t w seconds t most o , a .

The errors from the third fourth and fift h sources may
asily b e kept negligible A inclination f one minute of
, ,

e . n o a rc

in the horizontal axis or an error of that amount in either ,

collimation or azimuth produces only about error in the , o

”
. 01

C o a st a n d G eo d e t ic Su r ve y p Re o rt , 1 8 9 2 , Pa rt 2, p .
58 .
 1 68 . E R R OR S . 1 85

latitudes A ll three of these adj ustments may easily be kept

.

far within this limit .

Most micrometer screws are so regular that the lim i un e

mated error in the mean result for a station from the sixth

cause is usually very small B t it should not be taken for . u

granted tha t a given scr w is regular L arg irregularities e . e

may be detected by inspection of the computation of the

microm ter valu Errors with a period of one turn may be
’

e e .

d tected by making the observati ns for micrometer value at

e o

ev ry quarter-turn and then deriving the value of each

e ,

quarter to 5 divisions 5 to 5 divisi ns t of the head

, 0 2 ,
2 0 o , e c .
,

separat ly Th four mean values thus deriv d should agree

e . e e

wi t hin t h limits indicated by their pr bable errors

e o .

1 68 To guard against error fr m t h seventh source the

. o e

pairs must be so select d as to make the plus and minus e

micrometer differ nces at a station balance as nearly as possi e

bl e F. example at t h fifteen astronomical stations

*
or , e

occupied on the Mexican B oundary S urv y of 8 9 9 3 t h e 1 2— e

m an micr meter di fference taken wi t h regard to ign n ver

e o , S , e

exceeded turn at any station and was less than turn 0 10

at nine of the stations I f the plus and minus microm ter

, .

. e

di ff rences balance exactly at a sta t ion an erroneous m i m

e , c ro

eter value does t a ff ct t h compu t d la t itude but m re ly no e e e , e

increases the compu t d probable rrors e e .

m ym mk m
xp h w h
It tt e r to de te r in a t io n f mi
v l m
se e s an eas a a e an a c c u ra te o

j xp dly l g xpl m pl
c ro m e t e r a ue . But e e rie n c e s o s t a t su ch de te r in a t io n s a re

w
b ect to e cte d d in e Fo r xa in
l d m
su une ar e an un e a e rro rs . e e ,

f ll y d m d w lv m w
t he Ha iia n Is b v se rie s o f t io n s , t io n e d a bo ve t h e m ic ro m

l l
a an o se r a en ,

ly p
e te r va wa s in e ti Th e h
lvl
ue c a re u ete r t e e e s. re s u ts S o a

j
ge f I
f t he Th is ds ra n g e f
l
ra n o n ea r
“ n; o to ta a ue . c o rre s on to a o

-
F l wh v
bo u t s ix t h f in c h i t he f f t he b ect- g In t h e Sa n
m m
a one o an n o cus o o a ss .

p ly m p
ra n c is c o s e rie s , a n d in ge n e r t he v a lu e h a s b e e n
ic ro
m l g d
a e re er e te r

y p
re ea te d e a s u re d , t he sa e ar e isc re a n c ie s h a v e b e e n e n c o u n

ph d g t h e su gg e s t io n s o f t h e a b o v e
t e re d . He n ce t he n e e o f c a rr in out ra a

gra .
1 86 G E OD E TI C A S TR ON OM Y. 1 69 .

The errors from the eighth source may b e small on an

aver ge but they undoubt dly produce at times some of the
a , e

largest residuals They may be guarded against by protect .

ing the instrument from sudd n temperature changes and e ,

from shocks and careless handling and by avoiding l ng waits , o

be t ween the t w stars of a pair The closer the agreement o .

in t emperature b tween the instrument room and the outer e

air t h more secure is the instrument against sudden and

e

u n qual chang s of t mperature

e e e .

The computed probable error of a singl observa t ion e , e,

including all errors exc p t thos of declinati n was found t e e o

be as follows in thre recent fi t -class latitude series I n t h

, o

e rs : e

observations for variation f latitude at S F rancisco in o an

"<

1 89 9 1—77 obs rvati2 ns ( in two eries

12
) gave j e o S O
”
1

in a similar series at the H awaiian I slands 1

,
9 e z
: .

and j e z ;
°

for t h m purpose in 8 9 9 4 3 4 observations gave

‘

e sa e 1 1— 2 , 2

e : l : : from 3 6 obs rvations at fift en stations on 1 2 e e

t h M xican B oundary in 8 9
e e 93 i 9 to l 1 2— , e z z 1 : :

1 6 9 W hen an bserver b gins planning a s ries of b

. O e e o se r

va ti to det rmine the latitude of a giv n point t w q ues

ons e e o

tions at once arise H w many observations shall b e mad o

H w many separate pairs shall be observed P I ncreasing the

. e

number of obs rvations increases the cost of both field work e

and c mputation A increase in the total n umber of

o . n

separate pairs adds proportionally to the w rk of computing o

th mean places but oth r wise has little ffect on the

e , e e

total cost The economics of the problem demand that the

.

ratio of observations to pairs shall be such as to giv the e

maximum accuracy for a given expenditure Tw xtremes . o e

C y p p
y p pp
o a st a n d G eo d e t ic Su r ve Re o rt , 1 8 93 , Pa rt 2 , .
494 .

J
[ C o a st an d G e o d e t ic S u rv e Re o rt 1 8 9 2 , Pa rt 2, 54 , 1 58
C iv il C
.
, .

y p
t Tra n sa c t io n s o f t h e A ss o c ia t io n of En g in e e rs of o rn e ll Un i
v e rs it ,
1 8 94 , .
58 .
 88 G E OD E TIC A S TR ON OM Y . 1 71 .

to the plane in which the t lescope rotates u pon its horizontal e

axis F the co venienc however of those who may

. or n e , ,

d sire to det rmine the latitud with a s xtant on explora

e e e e

ti ns or at sea and of those who may be forced by circum

o ,

stances to determine the latitude by a measurement of the

al t itude of the S or a star with a the dolite or an lt
un , , o a

azimuth the following formul are here coll cted

, a e

To co m p ute t he la t it u d e f ro m a n o Ose rv e a
’
a lt it u de o f a st a r ,

or t he S u n , in a ny p osit io n , t h e t im e Oe in g hn o wn .

1 71 . The requisite f rmul are o a

tan D tan 6 sec t ,

cos D) sin A sin D cosec 6 ;

m which 6 is the declination and t the hour -a gle of the star n

( or S ) at t
unh instant of observation ; D is
e an auxiliary
angle intr duced merely to implify the c mputation A is
o S o
°

the altitude resulting from the measurement aft r pplym g all e a

instrumental corr ctions the correction for refrac t ion and

e , , ,

if the S is observ d the corrections for parallax and semi

un e ,

diamet r ( see 6 5
e D is to be tak n less than and
, e

or according to the lg b r g of the tangent a e ra c s1 n .

F ormula ( 8 ) is ambiguous in that

2 D determined from ,

the cosin may be either positive or negative B t the

e , . u

latitude of the station is always known befor hand with ffi e su

ci nt accuracy to decide between these two values Th se

e

formul are e x act no approximations having b een made in

. e

a ,

deriving them .
*

litt l m y pp
my pp
F o r t is h d e riv a t io n se e Do o e s
’
Pra c t ic a l A st ro n o , . 2 3 6, 2 3 7 ;

Ch a u v e n e t A st ro n o v o l. 1
’
or s , . . 2 29. 23 0 .
g 1 72 . L A TI T UD E F R OM Z E N I TH D IS TA N C ES . 1 89

To com pute t he la t it u de fr om z e n it h d ist a n c es f

o a st a r , or

t he Su n , oOse rv e d n ea r t he m e rid a ni , t he t im e Oe ing hn o wn .

The rate of change of zeni t h distance (or of altitude)

1 72 .

f a given star is smaller t h nearer the star is to the meridian e

H ence the ffect of a small error in the time which is

o .

e ,

assumed to b e known is less the near r the bservation is ,

e o

made to the meridian and is z r for an observation made ,

e o

precisel y O the meridian O nly a singl pointing can be

'

n . e

made when the star is the meridian whereas it is desirable on ,

to take several pointings so as to decrease the e ffect of rrors e

of observation H enc the desirabili t y of a rapid method for

. e

computing the latitud from circummeridian observatio s e n .

Let to be t h re q uired latitude of the station ; 6 t h

e e

declination of the star at obs rvation ; C C C succ s e “ ” ,"

e

sive observed valu s of the zenith distance of the star corre

e

sp d i g
on to the hour
n -angles t t t and C the , , , , ,

meridional zenith distance of the star Then .

B =A ’
cot C C: 3 cot
’
Co) : ( 3)
8

i l l 7 ( 4)
8

A , B, and C are evid ntly constant for a series of obs r

e o

v a t io n s mad n ar a giv n meridional passage of the s t ar

e e e
-
.

n n

values of and corresponding respectively to t

m , n , O , , t, .

t ,, Then for a star upon which observations u a re

mad near upper culmination

e

( I Hr
' -
fQa
: l:
u

in which the upper signs are to be used if the star crosses t h e

1 90 C TR N M Y g7 G E OD E TI AS O O . 1 2 .

meridian south of the zenith and t h lower signs if it cross s ,

e e

north of the zenith .

Similarly for a star upon which observations are mad

,
it e

n ar l wer culmination
e o

. g
u
. m
a
err i . n
“ ° o
u
O= I 80 — 6 °
A B

t b ing w r ckon d from lower culmination

e no e e .

Th se f rmul are not exact They are derived by an

e o a .

e xpansi n into a converging series of which nly the first

o , o

thre terms are retain d The rrors of the formul are

e e . e a

gr ater the gr ater the hour-angle and the s m aller t h zenith

e e , e

distance of the star Their accuracy is su fficient for the .

reduction of s xtant observations if the hour-angle is limit d

e e

to 3 and is also not allow d to xceed (in minut s of t ime )

0
m
,
e e e

the zenith distance of the star ( in d grees) T be certain of e . o

su ffici nt accuracy for the r ducti n of obs rvations mad

e e o e e

with a theodolite or altazimuth t h hou r-angle must be limited e

to 3 and to one -half ( in minut s of tim ) the zenith dis

0
m
,
‘

e e

tance of the star ( in degrees)

I t will b noted that go and are re q uired at the start for
.

e s,

the reduction since they occur in the s cond member of ( 8 5) e

and O nly approximate values are re q uir d howev r e , e ,

as the terms involving A B and C are always small S uch , , .

approximat values may be deriv d from previous computa

e e

tions ; from a preliminary approximate reduction of a single

observation ; if t h observations extend on both sides f
o r, e

meridian t h s mallest o b served zenith distance may be

o

th e ,
e

assumed for this purpose to have been made upon the

meridian .

The values of m and will b found tabulated in terms , n , 0 e

of t in 3 7 for values of t up to 3
0 ,
0
m

l l m y pp
my I pp
Fo r t h e ir d e iv r a t io n s e e D o o it t e s
’
Pra c t ic a l A st ro n o , . 2 3 8-2 44

A st ro n o vol 2 3 8 -2 40
'
or Cha u v e n e t s , . . . .
1 92 G E O D E TIC A S TR ON O M Y. I74 .

the four t h term is abou t 3 3 The preceding statem nts 0 . . e

are based upon as the value of P and the terms are

1
°
20
’
,

correspondingly reduced when P is below that limit .

1 74 F ormul and (8 7) furnish t h m

a of ‘

. e ean s

computing t h latitude from a i gl bs rva t ion of altitude

e s n e O e .

H w m ust one proceed if a series f observations of t h al t i

o o e

t ude have been made If the chosen f ormula is applied to

each observation separately ach computation will b exact ; , e e

the computations will be nearly alike as to furnish so

i t rough checks by differences ; b ut if there are many

con

ven en

observations the computati n of t h series b ecomes tediously o e

long I f one takes the mean of all the measured alti t ud s

. e ,

assum s t hat it corresponds to the mean of the hour-angl s

e e ,

and applies ither formula to this mean al t itude and mean

e

hour -angl h result is approximate F the mean of t h

e , t e . o r, e

altitudes does not correspond to the mean of the hour-angles ,

in general b cause of the curvature of t h apparent pa t h of

,
e e

the star and the corresponding variation in the rate of change

o f the al t itude The error involved in the assumpti n hat
. o t

the t w means correspond evidently decreases as t h interval

o e

covered by the seri s of observations decreases I f then

,

e .
, ,

one b reaks up t h series into su fficientl yshort groups the

e ,

means for each group may b treated by a singl application e e

of the formula without s nsible error F sextant bs rva e or o

ti ns it su ffices when observing upon P olaris to break up the

. e

series into groups not more than fifteen minutes long F or

more accu rate o b serva tions made upon P olaris with a t h d

.

, eo o

li t e or altazim uth one cannot be certain of su fficient accuracy

u nless each group is limited to ab out five minutes In using
,

( )
8 1 and ( )
8 with the S
2 or with other stars than
un circum ,

polars it will usually be necessary to make the groups v ry e

much shorter than indicated a b ove

,

.
 1 75 . L A TI T UD E F R OM OB SE R VE D A L TI T UD E .
93

The process of making a dir ct measurement of alti

1 75 . e

tude with an instrument having a vertical circle ( theodolite or

altazimuth ) is so simpl that a detailed statement of it is e

hardly necessary h re Su ffice it to say that the ff ct f t h e . e e o e

ind x error of the ver t ical circle must be liminated fr m

e e o

e ach obs rva t i on or group f observations before applying t h

e o e

formula for the la t i t ud c mputa t ion ; that t h chr nom t r e o e o e e

t ime of each poin t ing upon the star must be n t d wi t h a o e

chronom t r of which t h error is kn wn (t fix t h hour e o o e

angle) ; and tha t t h inclina t i n f t h v rtical axis must b e

e e

e o o e e

determ in d during the bservations by r adin gs f a level

e o e o

having its tube parallel to t h plane of the telescope The e .

e ffec t f ind x rror may b liminat d either by making

o e e e e e

sp cial bserva t ions to determine t h index error by making

e o e ,

half f t h bservations f each group with t h t lescope in

o e o o e e

a r vers d posi t ion or by a c mbination of both these p

e e , o ro

cess s The accuracy wi t h which the chronome t er rror must

e . e

be known may b inf rr d fr m the observed rat f change e e o e o e

of t h al t i t ude The correc t ion t t h m asured altitude due

e . o e e

to the inclination of the vertical axis may be computed by

form ula ( 55) or 4 8 using the a lgebraic signs th r 1 , e e

giv n if t h star is north f the zenith and rev rsing t hem if

e e o , e

s outh f t h zenith o e .

The method f observi g the altitude of the S or o n un a

star with a sex t an t has already been giv n in Chapter I II

,

, , e ,

in connection with t h d t rmination of time e e e

I f for a given purpos it is only re q uir d to determin t h

.

e e e e

latitude within 3 the table given in 3 8 may be utilized

0
”
0

as there indicated .
1 94 G E OD E TIC A S TR ON OM Y .
g 1 76 .

Q U E S T IO N S AND E XA M PL E S .

1 76 tate in d tail how you would make the com

. 1 . S e

pu t d allowance
e for parallax upon the horizontal circle if a
single point at a kn wn distance is used in making the colli o

mation adj ustment f a zenith telescope mounted on one side o

of it v rtical axis (S e ee

P rove formul ( 55) and

s .

2 . for the level corrections a

to be applied to bservations with a zenith t lescop o e e .

3 A t t.h sam sta t i n and with the same

e instrument as e o

given in 5 the following latitude observations were lso

1 2 , a

mad Compute t h results

e . e .

Date , 1 892 .
B .

NO o f
A
.
. C .

N . or S . D e c l in a t io n s .

St a r .

Au g u st 9 73 8 0
74 1 7

Au g u st 9 7440

Au g u st 1 6 7440
748 2

D iminish the differential refractions of by % ( see 3 04 20

g A p 3 9 andns . ( 1
°
1
’

4 What . is the correcti o n to the computed latitude d ue

to observing the star ff t h m ridian with a zenith telescope O e e , ,

in ach of the following cases The declinations of the f ur

e o

stars were respectiv ly 3 , 5 4 6 8 e , 2

° ° °
1
°

and they were observ d too late by the following intervals e ,

respectively 8 ,
12
3
,
2 1
3
, 1
3
,

A +ns 5 and
. o . 01 , 0 . 0 , o . 1 1 .

5 A t A.stronomical S ta t ion No 5 on the M xican '

.
, e

B undary 99 observations wi t h a zenith telescope were taken

o ,

o n 5 pairs 0each pair excepting one

, being observed twice .

It was found that [A A ] square seconds and [ ] , vv

1 96 C TR N M Y 5 76 G E OD E TI AS O O 1 .

separate o b ser ations as read from the face of a chronometer

v

known to be fast of local sidereal time were p ,

re s ec

t i ly
ve 3 ,
2 1
h
372
m
and z 1
h
z 1
h m
2 1
h

39 5
m
C ompute
0 the
8
latitude
1 of the station C arry y u r o

c mputation to tenths o f seconds of arc and treat ach b

. . .

o , e o se r

vati n separately
o .

A Mean value of o 3 8 4 ns .
°
1
’
1 1

The f llowing observations of the altitude of the S

1 1 . o un

were m de wi t h a theodolite on F ebruary 5 8 9 The

a 1 ,
1 2 .

times are given as read from the chronometer which was ,

known to be fast of l cal mean solar time and the alti

1 1
8
o ,

tud s are given after all corrections have been applied The
e .

following data in regard to the S were derived from the un

Ephemeris
C o rre c t e d A lt 1 t u d e s
II
.

° , " ° ’

m
42 05 12 4o 3 9

p
6 Eq u a t io n o f ti

l
05 1 e

Th e se v a dt t he

m m
05 27 u e s c o rre s on o

05 ean o f th o b se rv e d t i

l
29 e es .

05 3S W h a t wa s t h e t it u d e o f a t he

st a t io n ?

° ’
A ns 35 1 3 46
our observations of the altitude of P olaris with
.

12 . F an

altazimuth gave for its mean altitude corr cted for r fraction , e e ,

4 0 9
°
1The sidereal
’
times f the bservations were o O ,

respectively g 6 3 ,
22 5 and
h
1
m
1
8
, 22
“
z o
m
1
3
,
22
h
22
m
1
8
, 22
h

24
m
A t the
10 time of
5
observation the right ascension of
P olaris was
.

and its declination 8 8 44

1
h
z o
m ° ’

What was the latitude of the station ?

A n s. 3 9
° ’
26 06 . 8 .
 1 78 TH E IN S TR UM E N T . 1 97

C H P TE R
A VI .

AZ I M U TH .

1 77 The instrumental process of determining the azimuth

.

o f a terrestrial line astronomically consists of a measurement

of the angl between two vertical planes one defined by the
e —

azimuth mark and the vertical line through the instrum nt e ,

and the other by the observed star and t h vertical at the e

instrum nt Since the angle between these two planes is

e .

continually changing the exact time at which ach pointing

, e

is mad upon the star mus t be no t ed upon a chronometer of

e

which t h error is known F rom this record d time the

e . e

hour-angle of the star and its azimuth as seen fr m t h station o e

may be c mput d Th compu t ed azimuth of t h star com

e e e

b i d b y addi t ion or subtraction ( as t h cas

o .

ne may b ) with e e e

the measur d horizontal angle at t h station between the star

e e

and t h azimu t h mark gives the azimuth of the mark from

e

the station .

D ipt i f I t m t e scr on o n s ru en .

1 7 8 A y one
. f th
n many the d lites used for the
o e o o

measurement f the horizontal angles of a t riangulation may

o

be us d for azimuth d terminations — provided the telescope

e e ,

can be inclined enough to point to the star — each instrument ,

giving a degree of accuracy dependent upon its size power , ,

and workmanship The larger instruments often called

.

altazimuths designed primarily for astronomical work do not

, ,

di ffer in principle from the smaller theodolites The i t . n s ru

 1 98 G E OD E TIC A S TR ON OM Y
. 1 78 .

ment shown in F ig 4 is a theodolite which b elongs . 2 -

2 0 in
*
.

to the U S Coast and G eodetic S urvey

. . .

The Troughton and Simms altazimuth now in use in the

C ollege of C ivil Engineering of C ornell Univ rsity may be e

taken as a type of the larger altazimuth Its horizontal s .

circle is 3 6 cm 4 in ) in diameter and is graduated t 5 1

’

spaces I t may b q uickly unclamp d from the fixed center

. .
, o

. e e

of the instrument and its zero shifted to a new position , .

The index microscope carries in its field of view ( as the comb

is carried in the micrometer f the zenith telescop ; see 3 5) o e 1

a small pointer which is se n proj ected against a portion of the e

horizontal circle and gives t h degrees and the nearest p

, e re

ceding five minut s of the reading The remaining minut s e . e

and the s conds of the reading are o b tained from thr e

e e

r ading micr sc pes Each of these microscopes is furnished

e o o .

with an eyepiece micrometer m l to that of t h zenith Si 1 ar e

t l scope described in 3 5 A line of the circle graduation

e e 1

the bject -glass and the micrometer line b ear the sam e
.
,

o ,

r lation to each other here as the star the object -glass and
e , ,

the micromet r line bear to each other in the zenith telescope

e

( see F ig and
. the corresponding
2 1 t ext
, Each ,

microscope is so adj usted that five turns of the microm te e r

screw correspond as n arly as may be to one space on t h e e

circle Each turn represents therefore approximately

. one

minute and each of the sixty e q ual divisions of the micrometer

,

head one second The reading of the micrometer i . n c re a ses

as the micrometer line appar ntly moves in the direction of e

d i g graduations The reading given directly b y the

com b and head is the distance measured in the back ward
ec rea s n .

dir ction along the circle graduation from the zero of the
e ,

micrometer ( middle of t h comb ) as seen projected upon the e

circle to the nearest gradu tion representing t m ll a a oo s a

l w w y
,

Th a t is , a t he o do it e it h a h o riz o n t a l c irc e l t ent in c he s in d ia m

e t e r.
2 00 C E OD E TIC A S TR ON OM Y
. 1 80 .

separate index microscope and the vertical circle a of en , re t

omitted A reflecting prism is sometimes placed at t h

. e

intersection of the tel scope with the horizontal axis to turn

e

the light rays at right angl s nd t h eyepiece is th n placed e , a e e

at the end of the horizontal axis The proportions f the . o

various parts and th ir absolute size may be varied greatly

e .

The graduated circle may b furnished with a clamp and tan e

gent screw similar i principl and purpose to that of the

,
n e

lower motion f an engineer s transit Th se various m d ifi

o
’
. e o

ca t ions produce great changes i the outward appearance of n

the t m t may introduce certain obvious limitations t

1 n s ru en ,
o

th ir use but the principles involved are changed only in

e ,

minor details .

t Adj u st m e n s .

1 8 0 The vertical axis must be made truly vertical b y the

.

same process that is used in levelling up an engineer s transi t ’

Whatever levels are t b used should for c nv nience b o e o e e

adj us t ed so that eac h will re erse with but little chang in the v e

position of the bubble .

The adj ustm nts of the focus f the telescope of the colli
e o ,

mation and for bringing the middl li e of the reticle into a

,
e n

vertical plane should be made precis ly as for the astronomi

, e

cal transit (see 8 5 ,

The r ading microscopes must be kept in adj ustment

e .

O rdinarily the only adjustment that will be fou nd necessary

is to fit t h m croscope to the eye by drawing out or pushing
e i

i n the eyepiece until the m st distinct vision of the m i m o c ro

eter lines and of the grad uation is obtained Sometimes it .

may be found that the micrometer lines are apparently t n o

parallel t o the graduation upon which the pointing is t be o

made This may be r medied by rotating the micromet r e e

bo x about the axis of figure of the microscope I f to do this

.

.
 1 81 . CT N R D IRE
RV N IO S FO OE S E I O .
201

it is necessary to loosen the microscope in its suppor t ing

clamp great caution is necessary t insure that t h distance
,
o e

of t h obj ective from t h circle graduation is not changed e

I f one turn of the micrometer is f und to di ffer very much

e .

from its nominal value in terms of the circle gradua t ion ( one ,

tu rn one m m t for the instrument described i § u e n

it may b restored to its nominal valu by changing t h dis

e e e

tanc f the objective fr m t h circl graduation A

e o
'

o e e . n

inspection f F ig and t h corresponding t xt

o . 2 13 5 will
,
e e , 1 ,

show t hat f a given microscope t h nearer the obj ective is t

or e o

the graduation the small r is t h valu of one turn and i e e e , v ce

v e rsa A change in this distanc also nec ssitates a chang in

. e e e

t h distanc
e fr m the objective t t h microme t er lines— these
e o o e

lines and the graduati n b ing n c ssarily at c nj ugate foci of o e e e o

th e o bj ctive This adj ustment of the micr me t er valu is a

e . o e

di ffi cult one to make and should t b at t empted unl ss so no e e

it is certainly necessary O nc w ll made it usually . e e , re

mains su fficiently good for a long period .

cti f Ob i g Dire on s or se rv n .

1 81 The azimuth mark and t h instrument fix the t w

. e o

ends f a line of which the azimu t h is t b determined

o o e .

Th azimuth mark must be placed so far away from the

e

instrum nt that no change from the sidereal focus of the tele

e

sc p will be re q uired t give a w ll -d fi d image of it O

o e o e e ne . ne

mile will usually be found su fficient A conv nient mark is . e

a bull s -eye lantern shining through a small hole in a b

’
ox

which s rves to protect it from the wind A rdinary

e . n o

tubular lantern m unted in that way has been found t i f

o sa s ac

tory on distanc s from to three miles The size f the

e one . o

h le must be suited to the distance and telescope Too large

o .

a hole gives a blur of light instead of a well -d fi d point e ne

T oo small a hol makes t h light appear too faint A

.

e e .
2 02 O E OD E TIC A S TR ON OM Y . 1 82 .

diameter of one inch has been preferred by the writer at dis

tances of one to three miles for observations with a t lescope e

having an objectiv 4 5 mm i diameter and an eyepi ce e . n e

magnifying 3 diam ters M st observers prefer a much

0 e . o

smaller h l I f the distance to the mark exce ds five miles

o e . e ,

a strong r light wi t h a parabolic r fl c t or behind it or a lens

e ,
e e

in front f it may b r q uir d B lack and whit stripes on

o ,
e e e . e

the front of the box or a pole accu rately in lin may serve , e,

as a targe t for the measurements in daylight of the horizontal

angl s n cessary to conn ct with a triangula t i n A ground
e e e o .

mark (stake bolt or s t one monument ) must always b pro

, , e

v id d to h ld the point in cas

e o the b and light are acci e ox

dentally disturbed I f t h lin f sight passes near the . e e o

ground t h light usually appears m r u nsteady than if t h

,
e o e e

line is high abov the gr und A l ng cut through wo d e o . o o s

along t h lin of sight e th pr s nce f objects very n ar

e ,
or e e e o e

t th
o lin on ei t her side t nds t mak the light app ar
e e , e o e e

unsteady and introduce a liability t a small constant error

, s o

in the observ d azimuth Th direction to t h azimu t h m ark

e . e e

is immaterial xc pt when a micrometer is t b us d as

, e e o e e

indica t ed in 5 20 .

1 8 2 O ccasionally it is di fficult to find a satisfact ry loca

. o

tion for an azimuth mark F xample t h astro omical . or e , e n

station may be of the stations of a triangulation loca t ed

one

on t h flat top of a mountain in such a posi t i n that non but

e o e

very near or very distant points are visible from it In such .

a case one may resort to a collimator for azimuth mark an .

The collimator is an auxiliary t l escope rigidly mounted so e

as to fac the instrument and adj usted t sidereal focus

e o

The instrument tel scope b eing als at sid real focus and
, .

e o e

pointed upon t h collimator the two object -glasses th n being

e , e

t ward each other the lines of the reticle of the collima t or

o ,

may be seen as if they were at an infinite distance for rays of ,

2 04 O E OD E TIC A S TR ON O M Y
. g 1 85 .

pointin g pon the star and note the chronometer t ime of

u
*

th ebisection ; read the horizontal circle ; point again upon

the star noting the chronometer time ; read the circle read
, ;

the striding level and rev rse it ; point upon the star and note e

the chron meter time ; read the circle ; read and r m ve the
o e o

striding lev l ; point upon the mark and read t h circle twice
e e ,

each This completes a half - t Reverse the t l scope in

. se . e e

altitude and the instrument in azimuth and repeat t h same

, , e

routine for a similar half-set The whole t will t hus be . se

made up of six poin t i gs upon the star and eight pointings n

upon the mark with the corresp nding circle readings and
,
o ,

four readings of each end of t h bubbl of the striding level e e .

I t will be found a convenience wh n mak ing the computation e

if the altitude of t h star is r ad to t h near st minut from

e e e e e

the vertical circle once for each half -set The special id . co n s

e rati that le d to t h recomm ndation of t h ab ve routine

ons a e e e o

a re : that the eff ct of twisting of the instrument in azimuth

e

Should be eliminated from the result for each half -set as far
as possible ; that the level bubble should have time to settle
witho ut d laying t h observer for that purpose ; and that the
e e

observations of a set should b e compl t ed as q uickly as possi e

ble to avoid the e ffects of instability of instrument More

pointings in a set would serve to decrease the e ff ect of errors
.

of observation bu t tend to increase the errors due to i

, n

stability
To secure accurate results the pointings upon the star , ,

mark and graduations must be carefully made ; all heating

of the instrument a b ove the t emperatu re of the outside air
,

and especially all unequal heating of its parts must be avoided

as far as possi b l ; the manipulation must be made with as
,

little applied force as possible especially at t h i t t when , e . n s an s

ll w b l m mp ly l m
w k m pl y ll
Th e a o a e e rro r in ti e b e in g co a ra ti ev arge in az i ut h
t he o b v m y si Tip
”
t b i se c t io n d le t a n a s sist a n t

m
or , se r er a ca a an

re a d t he c h ro n o e t e r.
 1 87 . C RD 5 E XA M PL E OF R E O . 20

b isections are being made O ther conditions b eing unchanged .

,

the more rapidly the observa t i ns are made the grea t er the o

accuracy because the errors due to instability are smaller

, .

1 8 6 B efore commencing the n xt

. t of bservations the e se o ,

graduated circle should be shifted to another position that so

each micr sc ope will come over a different part of the gradua
o

tion To insure as compl te an elimination of peri dic errors

. e o

of graduation as possible it is advisable t shift the circle , o

betw n t h successive s ts of a series so tha t the vari us

ee e e o

p si t ions of a giv n microscop when pointing up n the mark

o e e , o

in a giv n position of the telesc pe shall divid the interval

e o , e

betwe n successiv microscop s (

e or usually) into e e 120
°

as many q ual parts as th r es ts in the series e e a re e .

1 87 . X MP O R C R E A LE F E O D .

W e st b v
S t a t io n —
s41
D a t e — S e t 1 3 1 8 77 p
°
29
’
o3
B a se
”
.
. O se r e r— A F Y
m
In s t ru e n t — T
.

S A ta z i u t
d
.

gl v l
N o 72 l m
.

h . . .

S t a r— 6 U rs a M in o ri s
(1 1 8h 1 1
m
.

.
1 d iv
Ch
Ch
ro n o
ro n o
.

mm
o f s t ri in

e t e r— N e
e e

u s 1 43 1

e t e r c o rre c t io n
g
( Si e re a ) d l .

° ' "
6 86 3 6 41 o .

Q)
E
"

“
H o riz o n ta l C irc l e .

6
“ 5t
3531
10
5
9
g W . E
, ,

In d e x . M ic A . . M ic . B . M ic C . .

M a rk

Ma r k

Th e m e a n z e n i t h d i st a n c e o f t h e st a r d u rm g t h e o b se r a t io n s ro m t wo a p pro x 1 m a t e re a d v f
w
,

m gs o f t h e v e rt c
1 a l c 1rc le, w a s o nd t
u o b e 40 °
f
Th e re a d m g o f t h e h o l e t u rn s fo r t h e o t h e r m i c ro m e t e rs i s n o t re pe a t e d i n t h e re c o rd
b u t in m a k m g t h e re a d m gs i t is c a ll e d o u t t o t h e re c o rd e r
m ic ro m e t e r A , t h e re c o r 1 5 a s a b o e d v
,

If 1 t a ll s a u n 1 t b e l o
If h e n ds
f
t h a t fo r A
.

w
.

fi it
he
t he
I dn
sa m e
ic a t e s
as fo r
11 b y
,

b o ve
.
,

w rrt m g a m i n u s S ig n o v e r t h e re c o rd e d re a d i n g o f t h e h e a d ; a n d 1 f 1 t i s a u n it a , he ca ll s
a t t e n t i o n t o it a n d re c o rd s it a s 60 t h e g1 v e n re a i n g + d .
2 06 G E OD E TIC A S TR ON OM Y . 1 88 .

Th e Circle Rea din g .

1 88 . I n the record above

two readings are given for each ,

positi n of each reading microscope When commencing t

o . o

read microscope A f the fir t time for xample in the above or s , e ,

se t of observations the fi ld of V iew looked as sh wn in , e o

F ig The r ading w evidently 5’ plus the angle

. e as 2

represented by the interval from t h zero of the micr sc pe e o o

( the position in which the micrometer lines are shown ) to the

2 5 grad
’
uation This plus q uantity is read directly from the
micromete when the 5 graduation is b isect d ( called the
.

’
r 2 e

f rward reading) namely 3

o p id d ther is no error
, ,
’
ro v e e

of p inting and p id d each turn of the micrometer p

o , ro v e re re

sents xactly B t neither of these conditions are realized

e u

i practic and a more r liable r sult may be secur d if a

n e ,
e e e

r ading is also made on the 3 graduation (called the back

e 0
’

ward reading) With perfect pointing and perfect adj ustment

.

of t h micromet r the screw would n cessarily be turned

e e , e

e xactly five revolutions backward to pass from a pointing on

th 5
e line to 2a pointing on t h 3 line and the reading of e 0 ,

the head of the micrometer would be t h m in both cases e sa e

I t is actually less Neglect for a moment all considera

.

tion of possi b le errors in pointing and reading These two

.

readings would then indicate that one t urn of the micrometer

H
/
5
re p t f re se n 59 8
s 5 6 ence the measured interval . .

of 3 from the zero of the micrometer to the 5 line

’
2
’

represents (3 2 g ) This procedure d es

1 ’
( 59 8 56 ) 3
——
9 . 1 . 2 . o

ld w w l y pp
p y
Th e fi e f vi is h h o n a s it a c t u a l a e a rs t o t h e o b se v
m
o e e re s r e r.

d t h e re f o re t h e g ra d u a t i o n re a ll
pp w
Th e ic ro sc o v e rt a
in in c re a se s in
m
es ,
n

the o s it e d ire c t io f ro t h a t h e re s h o

I
o n n

d iv i sio n
.

J
{ 0. 7 t u rn . t u rn .
)
2 08 G E OD E TIC A S TR ON OM Y
. g 1 90.

the h ad d considered to repr s n t s conds The nominal

e , an e e e .

reading of the micromet r b ing the corr ction C e e e , , ,

to this nominal reading or correction for run to be applied ,

to the forward readi g is n ,

£43
2
B
F )

The valu s of C will be found tabulated in 3 9 for the

e , 0

argum nts F the nominal forward r ading and B F ex

e
’
,
e , ,

pr ss d in divisions
e e nominally in seconds This table ,
or .

applies of course only to a reading microscope f the above o

type in which five turns are nominally q ual to space e one

of the circle graduation and each division of the h ad is , e

n minally
o The t able may be used to corr ct each forward e

reading of a series or the mean value of C taken out from

, ,

the tabl for each reading may be applied to t h m an f the

e e e o

forward readings A similar table may be co struct d on the

. n e

same principle f any other micrometer or .

1 90 I n developing the preceding method of computing

.

the true reading of the circle the accidental errors of pointing ,

upon the graduation have been t i i y ignored it being en re ,

tacitly assumed that t hey are small as compared with the

error of runs L t us now make the converse supposition
. e

that the error of runs is small as compared with the error of

pointing O this supposition the forward and back readings
n

o f the h ad diff r simply because of errors of pointin g

.

e e

H ence they are e q ually good determinations of the seconds

.

of the reading and t heir mean is to be taken O this sup

, . n

position t h true reading so far as the seconds are co cerned

e , n ,

15
 1 92 . L E VE L CORR C T E ION .
2 09

The use of ( 8 9) instead of

1 91 . or the corresponding
tables leads to quite a saving of time Two other considera
tions also point to such use as advisa b le F irstly the tru
.
,

.
, e

result sought is in reality O t w the results given by these e ee n

two methods since errors of run and err rs of pointing both

,
o

exist and in general neither are insensible as compared with

,

l
F
t h oth r
e S econdly
e : in (8 8) is as apt to be greater than ,

3 as it is to be less than 3 I f
; is 3 the use of ( 8 8 ) give .
5 , s

numerically the same result as H ence the results from

the of ( 8 8 ) are as apt to b e greater as to be less than thos
u se e

from and the greater the number of observations treated

the near r the results from the t w formul agree H enc
e o a . e ,

in gen ral there is not a su fficient gain in accuracy over the

e ,

procedure indicat d in ( 8 9) to j ustify the time required to

e

correct for errors of run .

*

t Th e Lev el Correc ion .

1 92 A y inclinati n of t h
. horizontal axis ffects the
n o e a

circle reading corresponding to the pointing upon the star ,

and nec ssitates a corr ction which is to be determined from

e e

th readings of the striding l vel e

I n F ig 6 let N E S W repre ent the horizon L t b

e .

2 s e

the star and Z t h z nith I f the instrum nt is in perf c t

. . s e

. e e . e e

m m m l m m
p l p p
So ti the va g iv e ic ro e t e r is d e
fo r t h e f
v d f m
e es ean ue ru n o a n

p
ri e c ia se ri s o f o b se rv a t io n s fo r t h a t u r o s e : t h e ru n is

md m l
ro S a e e

pp l d mp y p
to b e c d a c o rre c t io n b s e d u o n t h is v a u e is
m
st a t
:
a ss u e a on n n ea a n

w k pp y g
ie to the s u lt s c o ted b Th is ro c e d u re s h o rt e s
m l
a ea n re u n

mp d mp
th e o ro fa l in t h e c o rre c t io f r ru f t t h e e a n v a u e o f the
n o n ,
a er

w
b
ru n h a s Th e v a lid 1 t y f t h e a ssu t io n t h a t t h e ru is a
m
ee n co u te . o n

pp p l
c o n st a t is so d o u b t f u l h o e v e r t h a t it se e s t h a t if t h e c o rre c t io f r
l
n , , n o

y
ru n is t o b e a lie d t ll it s h u d b e b a se d u n a v a u e fo r t h e ru de
m
a a o o n

t h e v e r re a d i n g s t h a t a re t o b e c o rre c t e d
,

riv e d f ro .
2 IO G E OD E TIC A S TR ON OM Y .
g 1 9 2 .

adj ustment when the telescope is pointed upon the star the
plane in which the telesc pe is free to swing ab out its hori
,

z on t l axis is defined by the arc Z P in which P is t h pole

a s , e

of the grea t circle passing through Z and A the point in , ,

which the horizontal axis produced pi rces t h celestial e

sphere If now t h horizontal axis b e given an inclination

e

.
, , e

é t h west end being placed too high A will move to a point

, e ,

A
’
at a distance 5 along A The zenith of t h instrument
,
3 . e

will virtually be shifted to Z ( such that the arc A Z Z ’ ’

and Z Z b ) that being the nearest point t the true zenith

’
, o

to which the telescope be pointed The t lescope is w can . e no

free t swing in the arc Z P B t this arc does not pass

o
’
. u

through To bisect t h star it is nec ssary to turn the

5 . e e

instrument about its vertical axis which w passes through , n o

Z
’
until the telescope swings in the arc Z
, A will then
’
s .
’

be in such a position as A The change in t h circle read ”

. e

ing due to the inclinati n of the axis is vidently m asur d

,
o , e e e

by the angle PZ The circl if graduat d clockwis now

’
s . e ,
e e,

reads too small by that amou nt which will be called C

I n this triangle t h
, L .

Consider the spherical triangle Z P

’
s . e

a gle at Z is the required C the angle at P i b the side

n
’
L, s ,

P is the altitude of the star A and the sid Z is the zenith ’

s , e s

distance of the star as measured with the displaced i t n s ru

ment B t the displacement being small Z may be for

. u , ,
’
s

the present purpose considered e q ual to Z the z nith dis s, e

t ance C of t h star
, ,
F r m the proportionality of the sines
e . o

of angles and opposite sides in the triangle Z P we may ’

s,

sin
write sin A C sin 6
Replacing sin C and sin b by C and
L
'
L L
m C

6 those angles b eing small and solving f or C

s

,
there is , L.

obtained
CL z b b tan A
2 12 G E OD E TIC A S TR ON OM Y
. 1 93 .

triangle defined b y the star the zenith and the pole Certain, , .

sides and angles of this sph rical triangle have the alues e v

indicated in F ig 9 in terms of the angles the azimuth of

.
, , 3

the star reckoned from the north A its altitude and z ,

and 6 F rom the principle that the sines o f the angles of a

, ,

spherical triangle are proportional to the sines of the pposite O

sides we may write

,

sin 3 sin t
cos 6 cos A '

A lso from the principle that in any spherical tr angle the

,
i

cosine of any side is e q ual to the product of the cosines of the

other two sides plus the product of their sines into the cosine
o f the opposite angle we may write the two formul
, a ,

sin 6 = sin sin A + co s (b co s A cos 3 ;

sin A sin sin a cos (15 cos 6 cos t .

By su b stituting sin A from (9 5) in the first term of the second

mem b er of and solving the resulting equation f or cos 2 ,

there is obtained
( 1 — sin ’
sin 6sin cos (15 cos 6 cos t
cos 3
,

cos 90 cos A

By su b stituting cos for sin ’

I
2
q) ,and dividing
numerator and denominator b y cos 95 , (9 )
6 reduces to
cos sin 6 sin 15 cos 6 cos t
(
cos 3
cos A
 1 95 . CORRE C TION 3 C UR VA T URE . 21

I f ( 9 3 ) is w divided by and both denominators of the

no

resulting equation are divided by cos 6 there results ,

sin t
tan z
cos 45 tan 6 sin cos t ’

from which may be computed from the known values of Q5

3

I n ( 9 8) is the azimuth of the star east or west

,

6 and t
, . z ,

respectiv ly of north as the hour-angle is reckoned from

e , , ,

upper culmination to the eastward or westward respectively , .

t C ct i Th e Cu rv a u re orre on .

1 94 To apply this form ula to the star for each pointing

.

m ad up n the star during a set of observations w uld be too

e o o

lab ri us a process If for the t of the formula is taken the

o o .

m an of the hour-angles of the set the comput d azimuth is

e , e

that corresponding t t i m h -
g l but is not the
o le ea n ou r a n e,

re q uired m f t h m t l p
ea n d i g t ot / p te az z
’

u zs c o rres on n o ze se a ra e

fl -
gl
ou r a n since the rat ofes ,change of t h azimuth is con e e

ti nually varying The di fference between the t w q uantities

. o

indicated by the italics is small though not usually negligible

for the interval of time cover d by a set of observations I t
, ,

e .

is proposed to derive a su ffi ciently precise expression for this

di ff rence and to apply it as a correction to the result
e

o b tained by using a mean value of t in formula

,

1 9 5 Let t t t . t be the observed hour-angles ,, n ,

and th z respective corresponding azi e

L t t be the mean of the o b served hour -

, ,

m th u s . e o angles ,

and let be the azimuth correspond 3 ,

in g t o t , . Let d t, t1 t At, t, to , At” : t, to .

Then
A tn z o .
 C TR ON M Y
G E OD E TI AS O .

I f the t hi rd d following derivatives of with respect to

an 3

t b e neglected we may writ , e

dz d
’
e
3 0
E
A !"
( it
? % tn
A

“
d
’

in which 2; and d ; are the first and second derivatives corre

a

sp d
oni gnto and t zThe,
mean value of is o
. z

+ + Z c — z 0
’
dt 7:

The terms of the form 22 ”d sappear b y V rtue of the rela dz

4 i i

tion e x pr essed b y
approximate value of d ; may b e derived from
?
d
1 96 A
. n

equation ( 9 8 ) as follows Since only close circumpolar stars :

are b eing considered is always small and as an pp i. 2 , a ro x

ma t ion we m y wri t e ( in arc measure) for tan

a A lso
2 .2 .
,

since for t h stars under consideration 6 is nearly

e cos gb
tan 6 is much larger than sin cos t and the q uantity ,

cos tan 6 sin 95 cos t 2

interval of a few minutes over which the o b servations o f a t se

extend d will b e called C Equation (9 8) becomes afte

, an . r

these su b stitutions
d
C sin t whence t and
z
C3 : co s
it
, ,
2 16 G E OD E TI C A S TR ON OM Y . 1 99 .

Correct ion for Diu rn a l Ab erra t ion .

1 98 . T o the results as computed b y the a b ove f ormul a

there must still be applied a small correction for the e ffect of

diurnal aberration B ecause of the rapid motion of the
.

observer du e to the rotation of the Earth on its axis the star ,

is seen slightly displaced from its real position and the ,

apparent azimuth of the star is correspondingly a ffected .

S uppose t hat at the instant of observation the station from

which the o b serva t ion is made is moving in the direction A 8 1
of F ig 7 A B necessarily passes through the east point on
. 2 .
,

the celestial sphere of the observer s horizon at that instant

,
’
.

Let SA b the true direction of the ray of light from the star
e .

The figure is drawn then in the plane defined by the star , ,

the observer and the east point of the observer s horizon

,
’

I n conse q uence of the aberration the star will be seen in the

direction A S L t CA and A D be drawn proportional to the
’
. e

distance V t d by a ray of light in one second and the

ra v e rs e ,

distance traversed by the station of bservati n in one

r) o o

s cond respectively Complete the parallelogram CA B G

e

C all the angle SA B fl

. .
,

and the angle CA G SA S dfl ,

2
’
.

fl is the apparent displacement of the star measur d in the

’
a e

plane of the figure F rom the tr angle CA G sin 5 V.

sin fl
i
a
’
7;
,
1

a
’
fl sin 1? ‘

S ubstituting for cos and for V 1 86 000 ( see

there is obtained
dfl o .
3 19 cos sin 13 .
( 1 04
)

It remains to determine the e ffect o f this displace

1 99 .

ment upon the star s azimuth In F ig 8 let N E S W be ’

. . 2
 CORR C TION F OR O TURN L B RR TION
1 99 . E 7 A A E A . 21

the horizon of the observer Z his zenith and the star s

P rolong the great arc Z to F I n the spherical triangle FE

, , .

S . r

the a gle at F is
n the side F is the altitude of the s t ar A S ,

the side FF is 9 ( is the azimuth of

0 the
°
star )zand the z ,

side E is the 3 of the preceding paragraph Call the angle

S , .

sEF Then from Napier s rules for a right spherical tri

' ’

angle we may write

cos A sin F E cos s

sin F E sin fl cos ; s 2

cos A sin fl 27 .

the value of sin F E from ( 6) in s 10

cos A cos sin fl cos ( 8 ) 3 10

We require the effect upon of a small change in fl 3 .

This may be found by di fferentiating ( 7) and ( 8 ) with 10 10

r spect t A and [3 ( but not with respect to since that

e o , z , 71 ,

i not changed by the change in

s Thus we find
cos A cos d sin A sin s sin fl d fl ; ( 9)
e e 10

cos A sin d sin A cos s cos [9

e g dfl ( ) co s v . 1 10

Multiply ( 9 ) by cos and ( ) by sin and subtract the

10 3 1 10 3 ,

s ec nd of t h resulting e q uations from the first and ther is

o e , e

ob t ained
cos A d cos sin fl dfi sin cos [3 cos dfl ( )
z 3 2 v . 1 1 1

S olve for d multiply both numerator and denominator of

by b t i
z

t h resulting fraction by sin fl and eliminate cos

,

e 7} su s

tuting its value f rom

,

and t h result is e

CO S 2
dz
sin [1 cos A
2 18 G E OD E TI C A sTR ON OM Y . 2 00 .

S u b stituting the value of dfl from we have finally , ,

cos 15 cos ( 3
3 9
cos A O ' 1

F ora close circumpolar cos will always be nearly u nity A

,
co s

and so will be cos e x cept f stations v ry near the pole 3 or

H ence for such stars it is nearly exact to use

, e .

o .
3 2 .
( 1 14
)
The greatest variation from this value for the four circum
polars mentioned i 8 3 and for stati ns bel w latitude
n 1 o o

H nc for most purposes it su ffices to use

,

is e e

that is t disp nse with the computation of the fac t or

,
o e

cos 15 cos of
(
That will
z
be don in th xamples of e e e
A
co s

this b k oo .

Th sign in ( 3 ) and ( 4) is for the correction when t h

e 1 1 1 1 e

azimuth of the star is expressed as an angle west of north .

I f the azimuth of the star is expressed as an angle east of

north the sign must be changed to
,

t ti Ex a m ple of Co m pu a on .

200 The computation of the

. t f observations given in se o

8 7 is as follows
1 No correc t ion w applied for run of . as

micrometers the true reading of the circle being derived by

,

( 9)
8
F o r F i rst H a lf-
se t

l
.

Me a n c irc e re a d in g on st a r

Le ve l c o rre c t io n

C d e m di g o n st a r

m k
o rre c t e a n re a n

Me a n re a d in g on ar

Ma r kw e st o f st a r 1 5 52 oo .
50
2 20 G E OD E TI C A S TR ON OM Y . 2 02 .

P rogra m oft f R p t it i
Ob serv in g for t h e Me h od o e e on s.

201 To measure a horizontal angle by repeti t i ns one

. o

must use an inst ument havi g a clamp and tangent screw to

r n

control the motion of the lower or graduated circle in addi ,

tion to a similar clamp and tangent screw to control the rela

tion betwe n the upper circle carrying the verniers and t h
e e

lower graduated circle A t the beginning of the measurem nt . e

th ecircle is read B y a suitable manipulation of the t w

. o

motions upper and lower the angle to be measu red is multi

, ,

plied mechanically three fi or more times The circl , ve, . e

b ing read again gives the measured value of the m ultiple

e

an gle from which t h re q uired angl is r adily derived e e e

This pr c ss serves t gr atly decr ase the err rs arising f rom

, .

o e o e e o

err ne us readings f t h verniers and errors f graduation

o o o e , o .

B t any lost motion or fals

u motion in clamps and tang nt
, e , e

scr ws affects t h m asur d angle directly To eliminate

e ,
e e e .

this error as far possible one may measu re both the r quired
as e

angle and its explement al ways rev lving the instrument in ,

*
o

a clockwise dir cti n with t / motion loose and making ll

e o ez
'

z er , a

pointings with either tangent screw so that t h last motion of e

that screw is in t h direction in which t h opposing spring is

e e

being compr ssed With this procedure unless the action of

e

clamps and tangent scr ws is varia b l the d riv d va lues

.
,

th e e e, e e

of both the angle and its explement will be too large or too
small by t h same amount The mean o f the measured
e

angle and 3 6 minus the measured explement will b e the

.

°
0

c rrect value of the angle una ffected b y the constant errors

o

of t h clamps and tangent screws

e .

202 The f ollowing is a convenient program f or the

measurement of an azimuth b y repetitions A fter all adj ust
.

ments have b een made and the instrument car fully levelled
.

e ,

3 60
°
m in u s a g iv e n an gle is ca ll e d t he ex ple m en t o f tha t an g le .
 2 02 . M E TH OD OE RE RE TI TION S .
22 1

clamp t h upper circle to the lower in any arbitrary position ;

e

point approx mately upon the star ; place the striding level
i

in position and read it ; reverse it read it again and r move , , e

it ; point accurately upon t h star noting the chronom ter e , e

time and using the lower clamp and ta gent only ; read the
,
n

horizontal circle unclamp the upper motion and point upon

;

the mark using the upper clamp and tangent screw ; unclamp
the lower motion and point upon the s t ar using the low r , e

clamp and tangent screw and taking care to note the h c ro

no m t etime of bisection ; loosen the upper motion and point

er

again upon the mark using the upp r clamp and tangent , e

screws ; take anoth r pointing upon t h star with the lower

e e

motion noti g the time ; p int again upon the mark using
,
n o ,

the upper motion ; read the horizontal circle This compl tes . e

th observations of a half-set if three rep titions are t be e o

made I n passing from the star to the mark and i

e

.
, v ce v e rsa ,

the instrument should always be rotated in a clockwise d i re c

tion and t h precaution stated in the preceding paragraph as

, e

to the of the tangent scr ws must be kept in mind

u se e .

B efore commencing the second h lf t th lower motion a e

se e

should be unclamped the telescope reversed in alti t ude and

, ,

the ins t rument rev rsed 8 in azimuth The program will

e 1 0
°
.

be as for the first half -set except that now the first pointing ,

is to be upon t h mark ; ll pointings on the mark are to be

e a

made with the l w clamp and tangent screw and upon the
o er ,

star wit h the pp clamp and tangent screw and the strid
u er ;

ing level is to be read j ust a f ter the l t p inting upon the as o

star The direction of motion of the instrument must always

b e clockwise as b e fore and the tangent screws must be used
.

as before .
222 G E OD E TIC A S TR ON OM Y g 2 04 .

203 E XA M PLE O F RE C O RD ; M E TH O D O F RE PE TITIO N S

ll
. .

S t a t io n — D o Po in t Te x a s O b se rv — A F Y
m l
ar , . er . . .

t—Gam bey Th e o d o it e
pl l
° ’
26 6 In s t ru
I l l
29 02 . . en .

y
Da te — A ri 8 48 d iv f s t ri d in g e v e
5,
m
1 o

Ch
. .

S t a r— Po a ris . e t e r— H a rd No ( S i d e re a l )
m
ro n o 50
Ch
. .

11 e t e r c o rre c t io n
m
a 1 04 ro n o
° ’
6 88 29 57 . 82 .

C irc l e Re a d i n g s .

C h ro n o m e t e r
O bj e c t Tm
V V B
.

i es .

e rn ie r A . e rn i e r .

b 1“
9 1°
10
’
30 2 71
°
10
’
40

91 1 3 40 2 71 1 2 50

No reading of the altitude was taken The altitude

204 . .

may be derived with su fficient accuracy for use in computing

the level corrections from the table in 3 The lev l cor 10

rection must here b e applied to t h angle betwe n the star and

. e

e e

mark not direct ly t t h circle reading F ormul ( 9 ) and

, 0 v
e . a 1

(9 ) will
2 not g t h sign of the level orrection that
1v e mus t e c :

b derived from the consideration that the star appears to be

e

farther west than it really is if t h e west end of the horizontal

axis is too high and i , v ce v e rsa .

The angle between the star and mark computed from the ,

first half- t is se ,

’ °

)}
° ° ’

(
’
12 8 1 4 50 9 1 10 35 a 12 2 1 2 5 . o,

and from the second half -set is

° ’ ° ’

9 1 1 3 45 12 20

ly h m mp l l
y lw y g v mp
E i v de n t t is ho d f u t in g the d va f t he g
m m l
et o co se c o n ue o an e

p m
n e c e ssa ril i e s t h e sa e ric a l t fi rst u t in g t he
gf m
a a s e nu re su as co

e x le en t and t hen su b t ra c t i n ro
 24 O E OD E TIC A S TR ON OM Y 2 08 .

night The mark may be p laced either to t h northward or

. e

to the southward of the station .

207 The adj ustments of the vertical axis of the levels

.
, ,

of focus (see end of g and for b ringing the movable

microm ter line into a vertical plane must b e made as indi
e ,

ca et d in 8 1 0 .

208 The following is a go d program for the observa o

tions P lace the micrometer line at such a reading that it i

.

nearly in the line of collimation of the telescope I f thi

. s

. s

reading is not a lready known it may b determin d by takin g , e e

the m an of two readings upon the mark with the microm ter
e e ,

the instrument b eing rigidly clamped in azimuth and the ,

horizontal axis of the telescope rev rsed in its Y between t h e s e

t w readings
o C lamp the low r circle and point upon t h
. e , e

mark by use of the upp r clamp and tangent -screw Th n e . e ,

with the instrument clamped rigidly in azimuth tak five , e

p intings with t h micrometer upon t h mark dir ct the

o e e ; e

telescope to the star ; place t h striding l vel in position ; e e

tak three pointings upon the star with the microme t r noting
e e ,

t h chronometer tim
e f each ; read and reverse t h striding
e o e

l vel ; take two mor pointings upon the star noting the
e e ,

times ; read the striding level This completes a half -set . .

Reverse the horizontal axis of the telescop in its Y ; point e s

approximately t t h star ; place striding level in position ; take

o e

t hree pointings upon the star noting the chronometer times ; ,

read and re erse t h striding level ; take two more pointings

v e

upon the star noting the times ; read the striding level ; and
,

finally make five pointings upon the mark

S uch a set of observations may b e made very quickly ; the
.
,

e ffect of a uniform t wisting of the instrument in azimuth is

eliminated from the result ; and the b ub b le of the striding
level has plenty of time to settle without delaying the
observer for that purpose .
 2 10 . C M TH 5 M ICR OM E TR I E OD .
22

209 With the instrument used f

. the following observa or

tions increased readings of the micromet r c rrespond to a

,
e o

movement of t h line of sight toward the east when the e

vertical circle is t t h east and toward the west if the o e , ver

tical circle is t t h w st o e e .

21 0 X MP O R C R
. N C O M PU T T O N
E A LE F E O D A D A I .

St a t io n N O O b se rv e r— J F H
m l
. 10 . . . .

e n t— Fa u t h Th e o d o it e N o
° ’
O In s t ru
9 35
lvl 72 5
31 1 . o . . .

Da te— Octo be r 1 3 892 dv

One i is io n f s t ri in d g
l Ch m
, 1 . o e e

e t e r— N e g
S t a r— Po a ris , No 71 6 ( Sid e
l l)
n ear ro n o us . 1

g a t io
m m m
e a s t e rn e on n re a

Ch
. .

* “
O n e t u rn o f ic ro eter 1 23 .
73 . ro n o e t e r c o rr . 2 1 1
m

Le v e l Re a d in gs .

Ch ro n o m e t e r
T im e

w
d . t . t .

W
9
11 06
‘n
3 8 '
. o 3
m
58 °
6 31 05 1 8 3 79 L o n git u d e
6 h 1 2m
07 32 . o 3 04 3 88
. .
3 5 1 2 e st
08 05 5 2 31 . 1 .
400 .
3 15 o f a sh
1 8 o 09 31 O 1 23 . 6 .
42 4 .
31 1 in gt o n .

09 48 . 0 0 48 . 6 .
43 0 .
316

Me a n s .

9 0 9 1 2 01 8 1 2 5 . 2

1 2 24 1
1 0 9 .
7 . 1 00 . 2 75
1 2 48 .
3 2 1 1 .
7 9 46 . 090 . 2 79
1 3 36 .
3 2 59 7 1 7 61 . 08 6 . 2 81
1 3 58 . 1 3 2 1 .
5 08 0 . 2 79

Me a n 9 1 0 36 . 6 Me a n s .

f Po l a ris
h ‘r
n ’

m
l
a o 1 z o o7 4
b se rv a t i o n
.

at t he t i e o fo
f Po
° ,
3
.

o a ris 88 44

Al t it u d f s t a t t h
e o ch f t he r a e m p ean e o o

b s rv t i s p r s c t i 3 08 ° '

l m l 31
o e a on ,
e e on , 1 3
A t it d f t t th idd e f t h e fi rst
l
u e o s a r a e o

h f -s t p s e c t i n 3 0 8
l m l
a e , er o , 31 I2

A t it d f st a t th id d e f t he
l
u e o r a e o

d h f - t p r c t i 3 08
ll m
s o
ec n a se e se on 31 1 4
C i ti r ds
, ,

o a on ea ,

Th e h e a d o f t h is ic ro e t e r wa s g r d u a t e d m m qua l p
m
a t o 1 00 e a rt s .

the e a n in g f a n in c re a se d re a d in g s e e
o 2 09 .
2 26 GE OD E TIC A S TR ON OM Y . 2 1 1 .

M a rk e a st o f co l lm i a t io n ,

63
C l ll m
0 . 01 02 . 02

irc e E st a r E . o f co i a t io n ,

ll m
o 1 2 52

Ci l
.

rc e W st a r E . o f co i a t io n

i
Me a n st a r E o f co ll mi a t io n

kw
.
,

Mar f st a r
v l
e st o

Le e c o rre c t io n

M k w
,

ar e st o f s t a r, c o rre c t e d 1 9 . 82

Me a n h ro m ti m f b se rv a t io n
m
c n o e te r e o O

Ch ro n o e t e r c o rre c t io n

Me a n si d e re a l m ti e o fo b se rv a t io n
(1

H o u r-a n g l e t ) e a st o f u pp e r cu lm in a
O ’
t io n 95 14 6 20

lo g c o s t lo g s in 1
< ) lo
g s in t

lo g t a n 6 lo g c o s t 8 96 1 08 n lo g
.

8 6 7 70 1 n
. lo g t a n 20
° ’
1 28 1 6 .
92

%A t x
Q
I 2 S irl
ta n Z o
N
e

71 5 1 11 I

C o rre c t io n fo r d iu r a l a b e rra t i o n
n

kw
éA t h
(
1 2 s in ’ , St a r e a st o f n o rt 2
e tc
Ma r f sta r
.

s in
"
e st o

m b
fl 1

lo g 2: f ro a o ve
10 8 t a n z O 84 09 6 9
Ma r k e a st o f n o rt h

0 33

H ere again the sign of the level correction as applied

.

21 1 .

to the angle betw n t h s t ar and mark must be deriv d ee e e

directly from the fact that the star appears to be farther west
c o s in e
° '

( n a t u ra ) l
l
31 1 2 .

( n a t u ra )
° ’
c o s in e 31 1 4 .
228 O E OD E TIC A S TR ON OM Y
. 2 1 5 .

hour-angle of the star and take out the corresponding altitude

for use in deriving the lev l correction then the other two e ,

angles may be derived by interpolating over the interval to

the middle of each half -set with the rate of change of altitude
taken from 3 8 The altitude ne d only be known within
0 . e
'

one minute ordinarily

any other star than P olaris the table of 3 8 t
.
,

F or 0

b eing available
, no

must either read the re q uired altitude

,
one

from the vertical circle of t h instrument if it has one or else

e , ,

resort to the computation upon which the t able of 3 8 i 0 s

founded .

Th e use of the factor sec A is not necessary with t h e

pointi gs upon the mark both b cause the line of collima t i n

n e

was purposely placed nearly upon the mark and b ecause

, o

se cA is very n arl y unity for the small altitude of the mark

e .

2 1 4 I nspection will show that there is nothing in this

.

method of bserving or computing which limits its use to the

o

time near elongation The micr meter may be used in thi

. o s

way and the azimuth computed as abov with the star at any e

hour -a gle even at culmi ation B t if the star is not nea

n ,
n . u r

elongation its motion in azimuth is more rapid it remains

, ,

near the vertical plane f the mark a shorter time and larger
o ,

angles must be measured with the micrometer or else the

eries of observations made shorter Errors in the time also
,

s .

have less ffect t h earer the star is to longation

e e n e

I f the azimu t h mark is placed to the t l w d of the

.

soa z ar

station the program of observing and the computation are not

,

materially modified .

t Vl Microm e er a ue .

21 5 T o determine the value of one turn of the m i

. m c ro

eter the observer may use a process similar to that used in

determining the value of the zenith telescope micrometer .
 2 16 . C M TH OD M ICR OM E TR I 9 E . 22

That is one may observe the times of transit of a close cir

,

cu mp l star near culmination across the micrometer line set

o ar

at successive positions one tu rn apart ( or one -half a turn) the

instrument b eing rigidly clamped in azimuth The corre
,

. e

tions for curvatu re may be made by use of the same table ,

3 6
0 as f , the z nith telescope
or but w using e in the place ,
no

of the hour-angle of the star reckoned from the nearest

r

culminatio and making the corrections to the observed

n,

times p sitive befor culmination and n gative after The

o e e .

striding level may be read during the observati ns and a o

corresponding correction applied .

The correction in s conds of t ime to be applied to each e

observ d time to reduce it to what it would have been wi t h

e

the axis level is

f
‘
- sin A
4 1 5
sec 6
w
d . sm A
{( w w) (
’
e f ,

for a level with a graduation numbered b oth ways from the

middle The o b server must depend upon his instrument t
. o

remain fixed in azimuth —unless fortunat ly he has an , , e ,

azimuth mark so nearly in the meridian that he can occa

i
s ona lly take a pointing upon it withou t unclamping the ,

horizontal circl during the progress f t h observations and

e , o e ,

so d termin the twist of t h instrument

e e e .

2 1 6 A nother convenient way of d termining the m i

. m e c ro

eter value without doing any w rk at night is to measure

, o ,

a small horizontal angle at the instrument bet ween two

terrestrial bjects both with the horizontal circle and the
O

microm ter I f the two objects pointed upon are much above
,

e .

or below the instrument t h measu red angle between them ,

e
2 30 G E OD E TIC A S TR ON OM Y .
g 2 19 .

may be reduced to the horizon for comparison with the circ le

measurement by use of the factor sec A as indicated in , 2 12 .

A the micromet r value is depended upon to remai con

s e n

stan t for the station t h focus must b left undisturbed if e e

possible after the micromet r value has been determined e

‘

Di s cu s si on of E rrors .

21 7The t l are those due to errors in the

ex e rn a e rro rs

right asc nsion and declinati n of the star o b served to lateral

.

e o ,

refraction of the rays of light from t h star or mark to the e

instrum t and to error in the assumed latitude o f the station

en ,

o f observation .

2 1 8 Errors of declination enter the computed azimuth

.

wit h full value when the star is observed at elongation and ,

errors of right ascension ente r with a maximum e ffect wh n e

it is observ d at culmination A t i nterm diate posi t ions both

e . e

errors ent r the computed result with partial values The

e .

errors arising from this source are usually small as compared

with the rr rs of o b servation but are nearly constant if all
e o ,

the observations at a station are taken with the star at about

the same position in its diurnal path y eastern e longa , sa t n ear

tion They may b eliminated to a considerable extent by

. e

observing t h same s t ar at various positions of its diurnal

e

path or by observing upon two more di fferent stars

,
or .

2 1 9 When the compu t ed r sults f a l ng seri s of accu

. e o o e

rate azimuth observations t a station are inspected it is a

usually found that they tend to group themselves by nights .

That is the results for any one ight gree b tter with each
, n a e

other than do the results on di fferent nights They thus .

app ar t i di t that some source of error exists which is

‘

e o n ca e

constant during each night s observations but changes from ’

night to night F example from 44 sets of micrometric

. or , 1

observations of azimuth made on 3 6 di fferent nights at 5 , , 1

2 32 G E OD E TI C A S TR ON OM Y
. 223 .

with his own instrument I n designi g instruments the n

attempt is often mad to so fix the relative power of the e

telescope and means of reading the horizontal circle that the

errors arising from telescope pointings and circl readings e

shall be of the same order of magnit ude

The e ffect f errors in tim may be estimated b y noting
.

o e

the rate at which t h azimuth of the star was changing at the

e

time it was being bserved The table of 3 will serve o 10

this purpose for P olaris Such errors are usually small b ut

.

not insensible e x cept near elongation

.
,

222 O f the relative magnitude of the i t

. m t l n s ru en a e rro rs

arising from imperfect adj ustment and imperfect construction

little of g neral application can b e said because of the great
e ,

variety of instruments used With the more powerful i t . n s ru

ments however it may be stat d that the errors due to

, ,
e

instability of the instrument become relatively great and

must be guard d against b y careful manipulation and rapid
,

o bserving .

The rrors due to the striding level becom more serious

e e

the farther north is the sta t ion ( formula I f the se e

level is not a go d one it may be advisable to take more level

o ,

readings than have been suggested in the preceding programs

of observation Wi t h a very poor l vel or at a station in a e

high latitude it may b e well to avoid placing any dependence

.
,

upon t h level by taking half of the observations upon the

e

star s image reflected from the free surface of mercury ( an

’

a rtificial h orizon ) Th e ffect of inclination of t h axis upon

. e e

the circl reading will be the negative for the reflected star of
e

what it is for the s tar seen directly Considerable car will . e

b enecessary to protect the mercury from wind and from

trem rs transmitted to it through its support
o .

223 The micrometric method treated in

.
5 4 gives 2 0 —2 1

a higher degree of accuracy than the other methods described ,

 22 5 . C M IS E L L A N E O U S
33 . 2

if a good micrometer is availa b le It avoids several sources

of error and the observations may b e made so rapidly that
.

the conditions are q uite favorable for the elimination of errors

due to instability The error in the final result for a station
. , ,

due to an error in the value of the micrometer screw may be

m ade as small a desired by so placing the azimuth mark and
.

so timing the observations that the sum of the angles meas

u red eastward from the mark t the star shall be nearly e q ual o

to the sum of the angles measured westward from the mark

to the star .

Oth I st m t d M th d er n ru en s a n e o s .

224 A astronomical t ransi t furnish d with an eyepiece

. n e

micrometer is esp cially well adapted to give results of a high

e

degre of accurac in determining azimuths by the micrometric

e y

me t hod
225 I f the transit has no micrometer a secondary azimuth
.

.
,

may b det rmined incidentally to time observations with

e

littl e x tra x penditure of time P ut an azimuth mark as

e

e e .

nearly as possible in the meridian of the transit A t t h . e

b ginning of each half-set of the time observa t ions point upon

e

t h mark with the middle line

e f the r ticle If the mark is o e .

nearly in t h horizon of the instrum n t t h collimation and

e e , e

azimuth errors of the transit as derived from ach half-set e ,

reduced to arc and combined by addition and subtracti n o

with ach other and with the equatorial interval of the middle
e

line give the azimuth of the mark The azimuth of a certain

, .

mark was so determined from the t ime observations re q uired

for a d termination of the longi ude of a station at A nch r ge
e t

P oint Chilkat Inlet A laska in 8 94 The computed

o a

azimuth of the mark from 3 8 nights o f o b servation varied

.
, , ,

through a range of 8 The probable error of a single

12
”
.

det rminati n was l

e o : : 2 . 1 .
2 34 G E OD E TIC A S TR ON O M Y
.
g 2 28 .

226 The transit may also be made to furnish a good

.

determination of azimuth by observations in the vertical of

P olaris b y the method already referred to in 9 12

227 I f the allowabl error f a given azimuth d termina

.

. e o e

tion exce ds a convenient method is to observ up n e

P olaris t any hour -angl and use t h table given in 3 to

e o

a e e 10

compute its azimuth at the time of each obs rvation The e .

tabulated valu s w re computed by formula Th only

correction t b e applied to the value as taken from the table
e e e

is t h at du e to t h di fference between t h appar nt declination

e e e

of P laris at the tim of observation and the value 8 8 46

o e
‘ ° ’

with which the tabl was computed T his may be comput d e . e

by use f the columns given at the right -hand side of the

o ,

tabl headed C rrecti n for increas in declination of o o 1

’
e

P olaris by assuming that the correction is proportional to

e,
”
,

the increase and must b chang d in sign if t h declination

,
e e e

is l ss than 8 8
e The tabl may also b us d as a
°
e e e co n

i t rough ch ck on compu t ations made by f rmula

ve n e n e o

If a star which is not a clos circumpolar or the S is e , un ,

O bserved for azimut h at a known hour -angle its azimu t h may ,

be computed by formula ( 9 8) for each o b servation or the

o b servatio ns may be tr ated in groups covering short nt rvals e i e

of time B t formula ( 3 ) will not apply sinc certain

. u 10 , e

approximations were m ade in its derivation which are o l y n

allowable when 6 is nearly 9 0

°

2 28 F rough determinations of azimuth in daylight

. or ,

say within when the tim is only approximately known

[

e ,

the S may be observed with a small theodolite or with an

un

engineer s transit as follows P oint upon the mark and read

,

, :

the horizontal circle ; point upon the S making the h ri un , o

z on t l line of t h transit tangent to the upper limb and t h

a e e

C v yR p 5 A pp
ll w
S d d Geo d ix N f h
e t ic Su r 89
l
ee o a st a n e e o rt , 1 , en o . 1 0, o rt e

o rig in a o ft h is a n d t he fo o in g ta ble .
2 36 G E OD E TIC A S TR ON O M Y . 2 29 .

229 .
Ob se r v a t io n s f
o Su n for A z im u t fi.

N ia n t ilik C mb l d
d B rit is h A e ric a
So u n m Se t p 8, 8 96 , P M
m l m
, u er an . , . 1 1 . .

e n t — Th e o d o it e M a g n e t o e t e r No
°
In st ru db 64
m m
19

Ch
. .

e t e r c o rre c t io n o n G re e n wic h M e a Ti m
ro n o n e 2

H o ri z o n ta l C irc l e . V e rt ic a l Circ l e .

O bj e c t

B
.

. Me a n . A . Mea n .

A z im uth m ar k .

Su n
’
s fi rst an d u pp e r l im b . .

Su n
’
s se c o n d and l o we r l im b D

Me a n s
Su n
’
s se c o n d a n d lo w er l im b

Su n
’
s fi rst a n d u pp e r l im b . R

Mea n s .
 2 30 . M IS C E L L A N E O US . 2 37

Com pu t ti
a on .

F irst Ha l f . Se c o n d H lfa .

Ch
Ch
ro n o mm ete r ti m e
w m
ro n o

G re e n ic
’
Su n s A
wh m
e t e r c o rr

ppf m p h m l
M e a n Ti
.

D e c in a t io n , 6 ,
a re n t
on

e
G re e n ic h M e a n Ti

in t e r pl o a
e

te d
b v d l
ro E e e ris

O se r e
C
C
d
A t it u e
o rre c t io n fo r a ra a p f ll x .

C
o rre c t io n fo r re ra c t io n

o rre c t e

P ( = 90
°
d
A t it u e A
6)
l d ,

d
L a t it u e , d) .

i( ¢ A P ) s

L o g s in ( s
L o g s in ( S A)
L o g n u e ra t o r
Lo g c o s s
m
L o g c o s (S
d
P)
L o g e n o in a t o r
Lo g t a n 5
’
2 1
m
L o g t a n {52

H o riz
m l l
o n ta

d
z

c irc e re a
d
ds
Tru e
Az i
Ma r
mk wh m k
ut
e ri

e st
a
ia n
r

o f n o rt h
re a

re a d s
s

The derived value of the azimuth is more exact the

23 0 .

farth r t h S is from the m r dian and becomes u nr liable

e e un e i ,
e

wh n t h S is very near the meridian These are the same

e e un

conditions that limit the use of t h s lar transit I f h w

.

e o .
, o

ever t h rror of the tim piece has been d t rmined earli r

‘

, e e e e e e

in the day by formula or is d t ermin d later in the day e e ,

mp ly wh
mp pl
It is i o rt a n to n o t ic e ha t t he wa y in ich t h is o b se rv e d
m m m
t t on

ph m
h ro n o eter ti t e rs t h is c o t io n is a s a e a n s o f in t e r o a t i g t h e

l m
c e en uta n

dec in a t io n f ro the E e e ris .

2 38 C TR ON M Y G E OD E TI
3 AS O 2 1 .

or b oth an o b servation at noon will give a good determina

tion of azimuth b y computing the hour -angle from the
,

observed times and then computing the azimuth f rom the

f ormula
cot 4
2 tan %t

which may be derived readily from ( 1 1 8) and

Q U E S T IO N S A N D E XA M PL E S .

23 1 1 I n the azimuth computation of

a correction 2 00

for diurnal a b erration Why is

. .

is applied not a correction .

also applied for the aberration due to the motion of the

Earth in its orbit
2D i fferentiate equation ( 9 8) with respect to g
. o and show ,

in a general way the relative errors introduced by a given

err r in go into the computed azimuth from o b servations
,

o ,

taken u p on stars of various d clinations observed at various e

hour-angles and at stations in various latitudes Check

your c nclusions b y inspecti on of the table i 3
.

o A lso n 10 .

do the sam with respect to errors of time

e .

3 To . what indet rminate form does formula

e
(9 )
8 reduce
for a star in the zenith ? Is that case in nature i d t m i n e er

n ate P

4 What
. di fficulties would you expect to encounter in
determining the azimuth accurately at a station of which the
la t itude is nearly aside from those arising from the
climate I n considering this q uestion remem b er t hat in most
o f the methods for determining the azimuth the determination
of the error of a chronometer on local time is of the one

n cessary auxiliary bservations

e o .
2 40 G E OD E TI C A S TR ON OM Y . 23 1 .

C p l k W a s h in g t o n C
S t a t io n —
pm
it o E a s t Pa r D
l
a , , , . .

Su n riv e rt ic a Au g u st 1 5 A M 8 56 O b se rv e r— C A S
m l l m
n ea r e 1

C h ro n o e t e r
. . . . . .
, ,

In st ru e n t — s-
in M g . a n e t ic Th e d o it e o . Si d e re a .

V e rt ic a l C irc l e .

Ch ro n o m e t e r
T1 m e .

Su n
’
s u pp e r a n d fi rst l im b . T l
e e s co pe D .

’
h m ° ’ ° ' ° ° '

5 z o 28 5 oo 2 2 08 25 00 58 2 9 00 58 2 9 3 0
22 2 8 3 7 45 2 08 38 1 5 58 1 4 4 5 58 1 4 3 0
25 29 1 3 30 2 09 I4 00 57 3 6 00 57 3 5 4 s

Su n
’
s l o w e r a n d se c o n d l im b . T l
e e sc o pe R .

° ' ° ° '
° ° ' '

5 2 09 01 30 29 00 30 57 48 00 57 47 3 0
28

30
2 09
2 09
1 2

2 7
45
oo
29
29
12
26
Is

30
57 3 4 3 0
57 1 9 1 5
57 3 4 1 5
57 1 8 3 0 78
°
Fh a r .

w w
pl m ph m
D 38
° '
8
" h
08
m
e st o f G re e n ic h
m m
C 2
53 1 5 .

6 a t e d f ro e ris )
° ’
(a t ea n o f the ti es ) 1 3 55 ( In t e r o E e .

10 . I f both available which should be used in formula

a re ,

(9 )
8 -t h e geodetic the astronomical latitude o r

1 1 . P rove formul and a

 2 33 . L ON G I T UD E .
241

C H P TE R
A V II .

L O N G ITU D E .

To determine t h longitude of a station on the

23 2 . e

Earth s surface r ferred to the meridian of G r enwich is to

’

,
e e ,

determine the angle betw n t h two m ridian plan s passing ee e e e

through the station and G reenwich resp ctive ly (S e . ee

This angle b tween the t w m ridian planes is t h same as

e o e e

the diff rence f t h local times of the t w stations con

o e
*
o

H
e ,

id i g 4 to r pr sen t ( ence to deter

h
s er n 2 S e e ee

mine the longitude f a sta t ion is to d termine the di fference o e

b t w en t h l cal tim of that station and the local tim of

o e

I n gen ral the longitude of an unknown s t ation

e e e e

G reenwich . e

is not referred to G re nwich directly b t to s me station f e , u o o

which the longitude is already known The astronomical .

det rmination of t h longitude of a station consists then i

e e , , n

a d t rmination of the local time at each of t w s t a t i ns t h

e e o o , e

l ngitud of one which is known and of the other is to b

o e e

d t rmined and t h comparis n of thes t w times Th i

e e ,
e o e o . e r

diff rence is t h diff r nc f l ngitude xpr ss d in t ime

e e e e e o o e e e .

This may be re duced to arc by t h relations 4 e 2

h

1 = 5
h
1 and °
, 5
1
m
1
°
1
”

23 3 The principal methods of d termining di fferences of

. e

m l m l
pp ly m k mpl
Th e ti m y be ei /fi er si de Th e l e q u in o x
v l
es a re a or ean so a r . v e rn a

pp ly m k mpl
u t io n bo u t t he h in 2 4 s id e re a l

l
a a re n t a es one co ete re o a e a rt

h o u rs , a n d the m ea n S u n a a re n t a es one co ete re vo u t io n in 24

m ea n so la r b oa rs .
2 42 G E OD E TI C A S TR ON OM Y . 2 35
.

longitude are b y the of the telegraph by transportation

u se ,

of chronometers by observations of the Moon s place and

,
’

by observations of eclipses of j upiter s satellites The ’

.

me t h ds of making the necessary determinations of the local

tim in each of these methods need not b e c nsider d here
o

as they have already been exploited in Chapters I II and IV

e e ,

We need here consider only the methods by which some

signal is transmitted between the stations to serve for the
comparison of the times .

Th e Ob serv in g P rogra m an d A ppara t u s of t h e Telegra phic

th d Me o .

23 4 The telegraphic method has been used very exten

.

i ly in this c untry by t h C oast and G eodetic S urvey and

s ve o e ,

during the fifty years of its use has been gradually modified .

Th emethod and apparatus at presen t used will be here

described .

The nightly program at each station is to observe two sets

of ten stars each for time with a transit of the type shown in
F ig . Each half-set consists in general of four stars
10 .

having a mean azimuth factor A (see 99) nearly e q ual to 2

0, and one slow star ( f larg declination ) observed above

o e

t h pole
e Two such half -sets with a r versal of the t lescope
.
, e e

in the Y between them give a strong determination of the

s ,

time The same sets of stars are by pr vious agreement

. e

observed at each station B etw n the two time sets or

rather at a b out the middle of the night s o bservations certain
. ee ,

arbitrary signals are exchanged by telegraph b etween the two

,

stations which serve to compare the two chronometers and

, ,

therefore to compare the t w local times which have been o

determined from the star observations .

’

23 5 F ig 3 shows the arrangement of the electrical

. . 0

apparatus at each station d uring the intervals when no arbi

2 44 G E O D E TI C A S TR ON OM Y . 2 37 .

chronograph record Each signal is transmitted b y the .

signal relay at station A to that local circuit and its time of ,

receipt recorded automatically by the chronograph A t the .

sam instant xc pt for the tim r quired for the electrical

e , e e e e

wave to b transmitted over the main line betwe n stations

e e ,

the signal r lay at station B transmits the signal to the local

e

circuit and chronograph there I f these signals coincide with .

the cl ck br aks on the chronograph at B at any time the

o e ,

observ r at B breaks into the main circui t with his tel g

e e

ra p h key
er s and
’
produces a rattle at A , sou nder which ’
s

informs him that he must change his signals a fraction of a

second t n ther part f the intervals given him by his
o a o o

chronograph beat Thirty signals are sent from station A at .

intervals of about two seconds The observer at A th n . e

clos s his key and t h observer at B proceeds to s nd thirty

e e e

similar signals from B to A The Western Union line is

th n releas d t h apparatus at each stati n is again arranged
e e ,
e o

as shown in F ig 3 and each observer proceeds to finish his. 0,

time observati ns o .

F a fi t -class determination this program is carried out

or rs

for five nights at each s t atio n the o b servers then change :

places ( t eliminat the ff ct of personal e q uation ) the

o e e e ,

i nstrumental e q uipm n t of each station being left undisturbed ; e

and the same program is again followed for five nights .

E x m pl f C m p t ti a e o o u a on .

23 7 A determ nation of the di ff rence f longitude of i e o

Mass and of I thaca N Y was mad May 6

.

Cambridg e , .
, , . .
, e 1

J une 3 8 96 The following is a portion of the field com

'

,
1 .

putation
y l l k p m
j wh l wh
Th irt si g t t wo -se c o d i t rv a s h ch in
m l l
na s a n n e ee eac ro n o e te r u se

p g
ti in g si gn fo r e v o u t io n ( 1
00
) f t he t o o t he d ic h
k m
a s u st on e r o ee

w l
b re a s t h e c irc u it in h ro n o e t e r a n d t h u s a n y e rro rs i n
the c t he S a c in

lm m l l
, on

t ha t he e a re e i in a t e d f t h e fi n a re s u t
ro .
 2 38 . C OM P U TA TION OF TE L E G R A PH I C L ON G I T U D E S . 24 5

b
A r it ra r y Si g na l s, Ma y 2 7, 1 896 .

F ro m It h a c a t o C a m b ri d g e . F ro m C a m b ri dg e t o It h a c a .

Ca m b ri dg e R e c o rd . It h a c a Re c o rd . Ca m b ri dg e Re c o rd . It h a c a Re c o r d .

27 -
39
29 -
39

h m “
9 3 8 505 53 2 9 39
m

07 57 1 0 7. 07

9 3 0 53 4 2 5 9 32
I 3 3 78 3 1

4 22 59 3 70 4 22 59 -
3 70
13 55 1 8 13 56

21
m
D iffe re n c e 21
m

The h ading shows which way the signals were sent over
e

the main line and the four columns giv the times of the
,
e

signals as r ad directly from the chron graph sheets at the

e o

s t ations indicat d There were 3 or 3 signals in ach series

e . 1 2 e ,

of which nly a portion are h re printed The means

o e . a re

given for the whole s r s each case A mean -time clock ie in

was used in the chronograph circui t at I thaca and a sidereal

e .

chronometer at Cambridg e

23 8 The first time set of the evening at Cam b ridge gave

.

for t h chronomet r c rr ction on local sid real tim at the

e e o e ,
e e ,

m an epoch of the set when the chronometer read 3 3

e ,
1
11
0
m
. 2 ,

The second set gave the chronometer correction

t the epoch when the chronometer read
“
a 4 1

31
m
3 . B y taking
. the means of the epochs and corrections ,
2 46 G E OD E TIC A S TR ON OM Y. 2 39 .

on the assumption that the chronometer rate was constant

d uring this int rval it was found that the correction w
e , as

at the chronomet r r ading 4 7 A lso fro m e e 1

h
oo
m
. .

the di fferences of epochs and correcti ns it was found that the o

rate of the chronometer during this in t erval was p er

minute A pplying this rate for the interval ( 4 7 3 4

. 1
h
1
m
. 1
h

to the value of t h correcti n ther is e o , e

obtained for the chronometer correction at the mean epoch

of the signals sent from I thaca to Cambridge
S imilarly from the time bservations at I thaca it was found
,
o

that when the clock read 9 its correction was 7 h m

( on local mean solar time ) The computation shows .

h w t h m an poch of the signals was derived in Ca m bridg

e e

and in I thaca sid real time

o e e

sid real time ( 4 6

e 1
h
1
m
e

( 3 55
1
h
The
‘n
difference of these two , 2 1
m

is the di fference of longitude of the stations a ffected by t h e

transmission time of the electric w

,

ave and by the relative ,

personal e q uation of the two observers .

23 9 The lon gitude difference as computed from the othe

. r

set of signals shown in the computation is evidently a ffected

in the rev rse way by the transmission time H nce t h
e . e e

mea of t h t w derived values namely

n e o 6 , , 1 1 2 1
1m

is the longitude di fference una ffected

2 1
m

by transmission time provided such t ime remained constant ,

during the two minu t es of the xchange A lso the transmis e .

,

sion time i itself i 3 ( s, 54 2 1

1m
z i
m
0
°
. o .

mp w ld mpl m
mp
Th is co u t a t io n ou be si ifi e d in a n o b v io u s a n n e r if s i d e re a l
ti ie c e s h a d bee n d at bo t h st a t io n s

l m m m
e u se .

+ “
4 22
m
is t h e si d e re a ti e o f ean oon a t It haca Ma y 2 7,

8 96
m l m m l
1 .

p
1 Th e va f t h e t ra n s iss io n t i in e ig h t s v e r t h is in e
l m l
ea n ue o e on n n o

p l y y
wa s v a u e s v a rie d f ro
d t he t o o o 84
te Th e t e e ’

m m k
an se a ra

g ra h in e f ro I t h a c a t o C a b rid g e b wa y o f S ra c u se N e w Y o r a d
. .

B ml p p k
, , ,
n

o st o n wa s 59 2 i e s l o n g a n d a s se d t h ro u g h o n e re e a t e r ( a t N e w Y o r )
, , .
 G E OD E TI C A S TR ON O M Y .

Discu ssion of E rrors .

rom the final comput ed result there has thus b een

24 1 . F
eliminated the average relative personal equation during the
series of obs rvations and the average value of the transmis
e ,

sio time during the sh rt interval covered by the exchange

n o

of signals on ach vening The rrors f the adopted right

e e . e o

ascensions are also eliminated from t h result becaus the e , e

sa m stars have be n observed at O t l s t ati ns

e e o t o .
*

242 The final computed result is subject to the follo wing

.

errors t that arising from the accidental rrors of obs rva

: 1s ,
e e

tions of stars t each station which must b q uit small

2 00 a , e e

af t r the limination d ue to 4 repetitions ; d that arising

e e 00 2 ,

fr m t h variati n of t h relative personal e q uation of the two

o e o e

observers from nigh t t night f whic h the magnitude may o , o

be estimated fr m t h following paragraphs ; 3 d tha t due to

o e ,

lat ral refraction to which r f rence will be made in 4 5

e ,
e e 2

4 t h that
, d to varia t ions in the rates of the chronometers
ue

during the period covered by the observations which m ust ,

usually b q uite small as the chronom ters are not disturbed

e , e

in any way during the observations and are prot cted as far e

as possible against chang s of t mperature ; 5t h that arising e e ,

from t h varia t ion of the transmissi n time b tween t h two

e o , e e

halv s f the exchange f signals on each night which is

e o o , ,

probably insensible as this interval is usually only a minut ;

, e

6 t h the di fference of transmission t ime through the two

,

signal relays s ince this di fference always ent rs with t h sam

,

, e e e

sign as may be seen by an inspection of F ig 3 This last

, . 1 .

e rr r is made very small by using specially designed r lays

o e

which act very q uickly by adj usting the t w relays to b as , o e

yl
p p
W h e re the d iff e re c e o f l o g it u d e is v e r a rg e t h e o b s e rv e rs m a y b e
l m
n n ,

f o rc e d d iff re t s t a r is t s t o a v id d e e n d i g u o t h e ir c h ro n o e t e r
l l
to u se e n o n n

ra t e s fo r t o o on g a n in t e v r a .
g 2 43 . PE RS ON A L E OUA TION . 2 49

n arly alike as possible by controlling the strength of the

e ,

current passing through the relay so that it shall always be

n arly the same and by exchanging relays when the o b servers
change places or b y a combina t ion of these methods
e ,

*
, .

P l Eq ti erson a ua on .

243 The extent to which the r lativ personal e q uation

. e e

may b ex pected to vary may be estimated from the foll w

e o

ing statem nt of the experience of two observers who hav e e

mad the major portion of the primary longitude d termina

e e

tions of the Coast and G eodetic S urvey during the period

indicated Th plus sign indicates that Mr Sinclair observes e

later than M r P utnam

. .

. .

P RS N
E Q T ON
O TW N C H
AL S NC R N R E UA I BE EE . . I LA I A D G . .

P TN M SS T N TS C
U N G RV R SU T N
A , A IS A . A D . SU EY , E L I G

R M O R C O N N CT
F W TH TH
O T G R PH C N E ED I E E LE A I LO GI
L
TU D E W O RK OF TH E SU RV E I M
By d ire c t co mp a riso n a t W a s h in g t o n , D . C .
, 1 8 90 , Se p t .

mp
1 7 ,
1 8 , 1 9 S P

F m
By d ire c t a ris o n a t St L o u is , M o 8 90 , N o v 2 78

l m
co . .
, 1 .
4, 1 5 o .

w k m pl ll y m y
in t e rc ha ge b s rv e rs d u ri g
fo g it u d d e t e r i a t io ft e r '

l
ro n o e n on e n n s, a on e

p l y
li a lf wa s c o et d ge n e r f 4 r 5 da s
’
o f the or e a ro o re s u ts

C
,

Ma y N J d A b N Y 1 891

J
a e , . a .
,
n n a ,
. .
, ,

lb y
May d S P
J
an une . .
i

J ly
D e t ro it , M ic h .
,
an d A an , N Y . .
, 1 89 1 ,
une

ly
an d u o 1 40 5:
Ch g
.

pl
ic a Ill d D e t ro it , M ic h 89 1 ju 006

ll
o an 1 o 1 72 : t 0

Chi
. . . .
, , ,

M in n e a o is , M in n .
, an d ca go I , .
, 1 89 1 ,

pl
—
A ug l 61 i
mh
. o . 1 o . 01 0

p
O a a , Ne b .
, an d M in n e a o is , M in n .
, 1 89 1 ,

—
A ug . an d Se t .
l o . 1 76 : l: o . 01 1

L o s A n ge le s , C a l .
, an d Sa n D ie go , Ca l .
, 1 892 ,
Fe b . an d Ma r 0 . 1 60 : l: 0 . 00 6

C y p p
p
Se e o a st a n d Geo d e t ic Sur ve Re o rt , 1 8 8 0, . 2 41 .

y
:

l Fo r h e se d a t a t h e a u t h o r
t is in de bte d t o t he Su e rin t e n den t o f t he

C o a st a n d G e o d e t i c S u rv e .
2 50 G E OD E TIC A S TR ON O M Y
. 2 43 .

S a n D ie go , C a l d Yu m A riz 8 92 , M a rc h
l m
.
, an a ,
.
, 1

pl
Lo s A n ge e s, Ca l an d Yu a , A riz .
, 1 89 2 ,

pl
Ma r d A ri

m l 40 : l:
. a n o . 1 o . 002

Yu A riz d No g A riz 89 2 , A ri

l
a , .
, an a e s, .
, 1 o . 1 50 i:
No ga
pl
e s, A riz .
, an d E l Pa so , Te x .
, 1 8 92 ,

ll w
A ri d Ma y 1 2 6 :t

l k
an o . o . 004

J ly
He ena ,
Mo n t an d Ye o St o n e La e ,

W yo 892 , j d
l k
.
, 1 un e an u o . 1 09 : t

E l Pa s o , Te x a n d L it t e Ro c , A rk .
, 1 8 93 ,
Fe b . an d M a rc h

The f ll wing values depend on unrevised field com

o o

putati n o

K y W
e F Ch l C
e st , la .
, an d a r e st o n , S . .
, 1 8 96 ,
Fe b . an d M a rc h

pl
A t la n t a , G a d Ke y W e s t F la 8 96 , M a rc h
l k l
an , .
, 1 .

L it t Ro c A rk d At a n ta Ga 8 96 , A ri

Ch l
e , .
, a n , .
, 1 .

W a s h in g t o n
pl
a r e st o n , S . C a n d ,
D . C .
, 1 8 96 ,
A ri d Ma y
C m b idg
an

W a s h in g D C d M a ss 8 96 ,
J
ton , . .
, an a r e , .
, 1

y
Ma y a n d
l
une

C
Wa sh i n g t o n
y
D Na v a b se rv a t r a n d
O o

C
.
,

W a s h in g t o n D C o a st a n d G e o d e t ic S u rv e

J
. . ,
,

O fic ef , 1 896 , une an d +o . 1 1 7

Note that t h period cover d by this record is nearly six e e

years and that t h locali t i s show that t h obse vers were

,
e e e r

surely submi tt d to a great variety of climatic conditions e

Y t if the first t w determinati ns made when Mr P utnam

.

e o o , .

w as comparatively w to the work be omitted the total ne , ,

range of the results is only It must be remembered 0 8

. 1 10 .
,

h wever that each of these results exc pt the first two

o , , e ,

d pend upon from eight to ten nights of obs rvation four or

e e ,

five nights each before and af t er the interchange of observers

It is quite probable that the actual variation o f the relative
.

personal e q uation from night to night is somewhat greater

than that sh wn ab v o o e .
2 52 G E OD E TI C A S TR ON OM Y . 2 47 .

t Person a l E qu a ion .

246 If in making a longitude determination circum

. ,

stances prevent the interchange of observers the e ffect of the

”

relative personal e q uation upon the computed longitude may

still be eliminated in part at least by a special determination
, ,

of the equation by joint observati ns at a common station o .

Th two obser ers may place their instruments side by side

e v

in the same observatory observ the same stars and record , e ,

their observati ns upon the same chronograph The differ

o .

enc of the t w chron meter corrections computed by them

e o o ,

c rrected for the minut longitude difference corresponding

o e

to the measured distance betwe n their instruments is then e ,

th ir relative personal equation O they may observe wi t h

e . r,

the same transit as follows O the first star A observes t h : n e

transits over the lines of the first half of the reticle and th n , e

q uickly giv s place to B who

e o b serves t h transits across the , e

remaining lin s O t h second star B observ s on the first

e . n e e

half of the reticle and A f llows A fter obs rving a s ries

, o . e e

of stars thus h l d i g alternately each observer com

,
eac
’

ea n ,

putes for ach star from t h kn wn e q uatorial intervals f the

e , e o o

lin s and fr m his own observa t i ns the time of transi t of

e o o ,

the star across t h m an line of t h whole r ticle The

e e e e .

difference of the two deduc d tim s of transit across the mean e e

line is the r lative personal equation I f each has l d t h

e . e e

same number of times in observing the mean r sult is inde , e

p ndent i any error in the assum d e q uatorial intervals of

e o e

t h lines
e No readings of the s t riding l vel need be taken
. e ,

and the result is less aff ct d by the instability of the i t e e n s ru

ment than in the ther m thod o e .

247 I n certain cases in which it is not feasible to use a

.

t legraph line for a longitude det rmination the same prin

e e ,

c ipl may be used with the substitution of a flash of light

es
g 2 48 . L ON G I T U D E E Y CH R ON O M E TE R S . 2 53

b tw en stations in the plac f t h electric wave F

e e e o e . or

example one might de t ermine t h longitudes of the so e

A leutian I slands of A laska the successive islands being in

,

general intervisible .

h m t Lon g t t
i u de b y C
E q ipm ron o e ers— u en .

248 I f a t l graph line is not available between the t w

. e e o

t ti
s a th
o n s,next m thod in ord r of accuracy asid from
"
e e e , e

the flash method allud d to abov is that f transporting e e , o

chronom ters back and forth b tween them The transported

e e .

chr n m t rs then p rform t h same duty as t h t legraph

o o e e e e e e ,

nam ly that of comparing the local t imes of t h t w stati ns

e , e o o .

Th chr n m tric m thod may p rhaps be best xplain d

e o o e e e e e

by giving a c ncr t example Th longitud of a stati n at

o e e

A nch rag P in t C hilkat I nl t A laska was d t rmin d in

. e e o

o e o , e , , e e e

18 9 4 by transpor t a t ion f chr nome t ers b twe n that station

, o o e e

and S itka A laska f which the l ngi t ude was known A t o o

A nch r ge P int obs rva t ions w r t ak n on ev ry possible

, , .

o a o e e e e e

nigh t fr m May 5th to A ugust t h namely in 53 nights

o 1 12 , , ,

by t h y and e emethod with a transit f t h typ shown

e ear , o e e

in F ig using as a hack for t h observati ns chron meter

. 1 1 , e o o

B ond 3 8 ( sider al ) A t t h stati

0 n th r were alsoef ur . e o e e o

other chronomet rs t w sid real and t w m an Th se f ur e , o e o e . e o

were n v r removed during the season fr m the padd d

e e o e

doubl -wall d b in which they were kept for pr t ction

e e ox o e

agains t sudd n chang s of temp rature and in which the hack

e e e ,

chronom t r was als kept wh n t in use The i t

e e o e no . n s ru

m ntal q uipm nt t S i t ka was similar A sidereal h

e e e a . c ro

n om t was us d as an bserving hack and t w other

e er e o , o

chr nome t ers one sider al and one mean were us d in

o , e , e

additi n Nine chr n m ters eight k eping mean time and

o o o e

sider al tim w r carri d back and forth b etween the

.
, e

one e e, e e

stations on the st am r H assl r

e

e e e .
 54 G E OD E TIC A S TR ON OM Y.
g 2 49 .

t d b y Ch m t — Ob Lon gi
ti u e ron o e ers serv a o n s.

249 A side from the time obs rvations the procedure w

. e as

as foll ws J ust before beginning the time observations at

o :

A nch rag P oint and again as so

o as they were finished on
e on ,

each night t h hack chronomet r No 3 8 ( ide eal ) was

,
e e . 0 S r

compared with the t w m an time chron meters by the o e o

method f coincid nce of b ats to be described later

o e e ,

Thes t w wer then each compared with each of t h two

e o e e

remaining (sider al ) chronometers at the station These com e .

parisons tog th r with the transit observations served to

,
e e ,

d termine the error f each chr nometer on local time at the

e o o

epoch of the transit observa t ions Whenever the s t amer .

*
e

firs t arrived t the station and again when it was about toa ,

l av the hack chro ometer No 3 8 was compared with the

e e ,
n . 0

o ther station chronom ers as indicated above was carried on et ,

b ard the steamer and compared with the nine steamer

o

chronome t ers and then immediately returned to t h s t ation

, e

and again compared with the four stationary station h

A an xtra precaution both the station o b server
c ro

no m t e e rs . s e

and the observer in charge of the steamer chron meter made o

e ach of th s comparisons I n the comparisons on the e e .

st am r the hack ( 3 8 ) was c mpared by coincidence of beats

e e ,
0 o

wi t h each f the eight mean time chron meters and the

o o ,

remaining ( sid r al ) chronom ter was then compared wi t h e e e

some of t h ight The comparisons on shor b f re and e e . e e o

after the trip to the steam r served to det rmine the error of e e

the hack ( 3 8 ) at the poch f the s t eamer comparisons 0 e o .

The s t eamer comparisons determined the errors f each f the o o

m w l mp y wh
w md m ly d
Th e st a t io n c hr on o e t e rs e re a so in t e rc o a re d on da s en n o

mp w mp
b se rv a t io n s Bu t t h is wa s to a sc e rt a in h e ir pe r
l
o e re a e . e re on e t

fo rm a n c e , an d t h e se co a riso n s e re n ot u se d in co u t in g t he o n gi
tu de .
2 56 C TR ON OM Y g 5 G E OD E TI AS . 2 2 .

certain e x tent av ids the waits— f a minute and a half on o o

an average— which would otherwise be necessary to secure

an o b servation on a pair of chronometers selected arb itrarily .

t t i f L git d b y Ch m t
Com pu a on o a on u e ron o e ers .

251 The following example ( taken from anoth r set of

o b servations) will show h w the chronome er comparisons are
. e

o t

c mput d A certain D ent mean time chronometer was

o e .

compared with a certain Negus sidereal chronomet r on O t e c .

1 4 8,9 at a
1 sta t ion2 , west f Washington It was 2
°
12
1m
o .

found that M D nt Negus

h m h m
5
1 14 z o A . . e 1z .

The correction to the D ent on local mean time was known

to b e and the correction of the N gus t
2
h
1 1
m
e o

local sider al time was re q uired e .

Ti m by
e De n t c h ro no m ete r 23
h
z o m

C o rre c t io n to De n t 2 1 1 53 4 1
.

Lo c a lm l
so a r t im 08

l l
ean e 21 29 59.

Re d u c t io n d
t o s i e re a in t e r va 2 90 ) 03 28 38.

l v l f m p m II
p
Si d e re a in t e r di g
m m
a ro re c e n ean n oo n 21 57 9 7
.

ti e o f re c e d in g ea n n oon (O c t . 1 3 31

l l m
m y
Lo c a d e re a t i e
si

m
IO 43 I2 . 02

Ti e b N e g u s c h ro n o e te r 1 2 54 41 . 00

C N g h m
o rre c t io n to e us c ro n o e te r 2 1 1 28 .
98

The computation is modified in an obvious manner i f it is

the error of the id real chronometer that is known S e .

252 This pr c ss of comparing chronometers is so accu

. o e

rat that it was f und that the t w values of the error of

e ,
o o

either of the station sidereal chron met rs as derived from o e

comparisons des ri b ed above in two di ff rent ways from

,

th e c e

the hack chronometer seldom di ffered by more than ,

This corresponds to an error of in noting t h time of 1 1

8
e
§ 25
4 CH R N OM T R L ON G I T U D E
57 E Y O E E S . 2

c incidence of beats on the supposition that all the error was

o

made in one f the four comparisons concerned I f t w

,

o . o

chronometers of the same kind both sider al or both mean , e

time are c mpared directly it re q uir s v ry car ful observing

,
o e e e

t osecu re th ir diff r nce within of t h truth

e e e 0 5
. 10 e .

253 The c omparisons of the oth r f ur station chr no

. e o o

meters with the hack chronometer imm diat ly b fore and e e e

aft r t ransit obs rvati ns gave the err rs f ach of th se four

e e o o o e o

chronometers T comput t h errors f the steamer h . o e e o c ro

no m t at t h tim f the comparis ns mad on t h steamer

e e rs e e o o e e ,

it is first n cessary t secur as g od a det rmina t ion as p ssi

e o e o e o

ble of the err r of t h hack chr n m ter at t h poch of t hose

o e o o e e e

comparis ns O valu f tha t rr r was ob t ained in an

o . ne e or e o

o bvi us mann r by assuming t hat t h hack chr n me t r ran

o e e o o e

a t a unif rm ra t b t w n t h last prec ding and t h n xt

o e e ee e e e e

f llowing transit obs rva t i ns F u r o t h r d t rminati ns of

o e o . o e e e o

th err r of t h hack at t hat epoch wer btained by making

e o e e o ,

t ha t same assump t i n for ach f the o t h r four statio o e o e n

chr n m t ers and d riving the rr r of t h hack fr m t h

o o e ,
e e o e o e

c mparisons mad with that chr nom t r at t h station b f r

o e o e e e e o e

and af t r t h s t am r comparisons Th weighted m an f

e e e e . e e o

th s efi valu s of t h rr r f t h hack was us d F t h

e ve e e e o o e e . or e

m t h d f d riving the r la t iv w ights which w r assign d

e o o e e e e e e e

to th s five r sul t s see

e e 6 A t A nchorag P in t 3 e 2 0 . e o 1

comparis ns w r made with the st am r chr n m t rs I

o e e e e o o e e . n

six cas s t f the thirte n t h rang f t h five d rive d

e ou o e e e o e e

valu s of t h error f t h hack was less t han o e

2 54 H aving now t h
e e

. rrors of t h st am r chron me t ers e e e e e o

on t h local tim of each s t a t ion t t h t ime of arrival t and

e e a e a

departure from ach sta t i n the diff rence of longitud was e o , e e

c mput d in the mann r indicated by the f llowing illustra

o e e o

t ion S uppose chronometer No

.
3 t have been found to . 2 1 o

have the following rrors on a certain round trip e :

2 58 G E OD E TI C A S TR ON OM Y . g 2 56 .

A h o ra g e Po in t at de p Ma y 1 5 h '“
A M
°
fa st fA P ti m
k l
a rt u re

k
nc , , , 9 oo . .
40 o . . e.

p
S it a rri va Ma y 1 6 A M S it u

k
a on 9 00 1 1

k
, , . . a

S it a d e a rt u re
,
a t Ma y 22 , 9 oo A M . .
4 S it a

A n c h o ra g e Po in t Ma y 2 3 9 oo A M 32 A P

rom the t w A nchorage P oint observa t ions it app ars

.
, . . .

F o e

t hat the chronometer has lost 8 in t h eigh t days it was gone

°
e

from t her F rom t h t w S itka obs rvati ns it appears that

e e o e o

H
.

7 w°
re los tewhile at S itka ence t h chronomet r lost . e e 1 °

only while trave lling bo t h ways between the sta t ions or its ,

trav lling rat was per d ay losing A pply ng this rate

e i

to the errors as det rmined at A nchorage P oint we find that

e , .

e ,

th rr rs of the chr nomet r on A k g P i t time at t h

e e o o e nc o ra e o n e

e pochs of the Sitka steamer comparisons were fast and

fast and that the difference of longitude re q uired is
A P west of S
,

4 We °
. o . . .

have thus d rived t h longitude di fference on the supposition

e e

that the steam r chronometers have a travelling rat which is

e e

constant during the round trip and without any assumptions ,

as t the rat s while in port The assumptions as to the

e

station chronom ters hav b een simp ly that each pres rves a
o .

e e e

constant rate betwe n successive transit observations e .

2 55 The longitude was thus comput d fr m each round e o

trip starting from A nch rage P oint and the mean taken I f
.

o , .

the chronometers had continually accelerat d ( or retarded) e

rat s this mean was subj ect to an error arising from that fact
e , .

To eliminate such a possible error and to serve as a check ,

upon the computa t ion a second compu t ation was made from ,

each round trip starting from Sitka and the mean taken , .

The error from acceleration ( or retardation ) of rates was

necessarily f opposi t e sign in this mean The mean of th se
o . e

two results is then subject only to accidental errors in so far ,

as the chronome t ers are concerned .

256 The f llowing table shows the separate results

. o

obtained and the manner of combining them

2 60 C TR N M Y G E OD E TI g 59 AS O O . 2 .

the first part of the above computation and the first half -trip
omitted in the s cond part I f there had been imply s ven
,

round trip start ing from A nchorag P int t h procedure

. S e

s e o e

would have been to deal regularly with all trips in the first
half of the c mputation ; and in t h last half in additi n t
o e o o

the six regular round trips starting from Sitka the last half
trip ( S to A P ) and the first ha lf -trip (A P to S ) would
,

. . . . . .

have b n used t g ther as a seventh round trip from Si t ka

ee o e .

258 Let N be t h
. numb r of days during which thee e

chronom ters were depended upon to carry the tim during

e e

each r und trip reckoned as follows A d d tog t her the two

o , : e

int rvals between comparisons f the s t eamer chronometers

e o

with the shore chr nom ter at the beginning and at the end
o e

of each half trip and increas t his by adding the int rval
-
, e e

from each comparison f t h bserving chron m ter and o e o o e

steamer chronomet rs t the t transit bservations made

e o n ea res o

at that stati n The w ight assigned to each trip in the

o . e

above computation is pr p rtional to /N o o I .

259 What relative weights shall b assigned to t h resu lts

. e e

from t h differ nt chronometers ? Some evidently run at

e e

a more nearly constant rate than others L t l l l l . e ,, , , ,, ,,

be the s parate values of t h longitude as giv n by any

e e e one

chr nometer and their mean and l t be the number of

o , , e n

such values or t h number f trips Then by least s q uares

, e o .

the probable error of any one valu is e

lm ) _
l ( ” a
1m )
?

(n
Z 1m ) ?
71 I

By the rule that the weight of a result is inversely propor

ti l to the s q uare of its pr babl error the relativ w ights
ona o e , e e

to be assigned to the chronometers are proportional to

fl — I

(?
2 a
1m ) 1m )
.

Z 1m ) J
§ 26 1 . CH R N OML ON G I T UD E
T R 6 E Y O E E S . 2 1

The factor is dropped for simplicity since we are deal

ing with l t i weights onl In the above computation
re a ve y .

the sum [( l 1 ) (l , ( l wasm

determin
’
d , , e

from each half of the computation and t h mean us d in the , e e

d enominator of The remainder of the computation

n eeds xplanation
n o e .

260 The r lative w ights assigned to the station b ro

. e e e

no m t e indicat d in
e rs
'

as 53 may be determined by an e 2

anal gous process Let b the error of a chronom t r at

o . 0 e e e

the poch of the transit tim obs rvations as det rmined from
e e e e

those bservati ns L t I be its error at that same instant

o o . e

int rpolated betw n its rr rs as d termined at the last pre

e ee e o e

ceding and first f ll wing t ransit time obs rvati ns on the o o e o

assump t ion that it rate during that int rval is constant s e .

Th n 1 e is a measure of the b havior of the chronometer

0 e .

It is t h amount by which t h chronometer has gone wrong

e e

on the suppositi n that the transit observations may be con

o

s id d exact
e re Th chr n me t r apparen t ly indicat s that
. e o o e e

t h station at t h middl
e bs rva t i n was at a distanc I
e e o e o e 0

in longitude from its p si t ion at t h prec ding and f ll wing o e e o o

o bs rvations F a group of chronom ters wh se rr rs are

e . or e o e o

all d term ned a numb r of times in succession by t h m

e i e e sa e

t ransit obs rvations t h rela t ive weights are vidently pro

e ,
e e

portional to the quanti t ies

The above example serves to illustrate the principles

26 1 .

involved in the c mputa t ion of a longitude by chron meterso o .

The accuracy of t h d rived longitude is greater t h greater e e , e

the number f chronomet rs used the greater the number of

o e ,

trips the smaller the average value of N

,
and f o

course depends intimately upon the quality of the h m c ro n o

2 62 G E OD E TIC A S TR ON OM Y
. g 2 63 .

and the care with which they

e t e rs protected from jars a re

and from sud den changes of tempera t ur Unl ss the round e . e

trips are quite short the errors of the transit time observati ns o

will be small as compared with t h other errors of the proc ss e e .

I f considered necessary the r la t iv personal e q uation f t h e e o e

observers may be eliminated from the result by the sam e

methods that are us d in connection wi t h t legraphic e e

det rminations of longitude

262 I f the trips are very long it may possibly be d i
e .

able to determin by a sp cial seri s of o b servations the

.
, a v s

e , e e ,

t mpera t ure c fficient f ach chronometer and also a

e oe o e

c e fficient expressing its acceleration ( or retardation) of rate

o
,

and to apply corresponding compu t ed c rr ctions to t h o e e

travelling rates The chronometers are compensat d for

.
*
e

t mperature as far as possi b le by t h maker of course but

e e , ,

such compensation cannot be perf ct Th thick ning of the e . e e

oil in the b arings tends to increase the friction with lapse o f

e

time and by diminishing the arc of vibrati n of the b alance

,
o

wheel to increase t h rate of running A ttempts to use rate e .

corrections depending upon the computed coe fficients f a o

chronome t er have usually be n rather unsatisfactory and e ,

should not be made except in extreme cases .

t d D t m i d b y Ob s i g t h M
Lon gi u e e er ne e rv n e oon .

2 63 I f none of the pr c ding methods are availab le one is

. e e ,

forced to use those meth ds which d p nd upon t h motion , o e e e

f the Moon or perhaps to observe upon J upiter s sat lli t s

’
o , e e .

The place of the Moon has be n observed many times at e

th fix d observatories F rom these observations its orbit

e e .

and the vari us perturbations to which it is subj ect have b en

o e

comput d I the A merican Ephemeris and similar publica

e . n ¢

—
Fo r de ta i s l o ft h is p ro c e ss se e l l
D o o it t e s
’
Pra c t ic a l A s t ro n o m y pp , .

3 83 3 88 .
2 64 G E OD E TI C A S TR ON OM Y. 2 68 .

o bj ects with a sextant or other suitable instrument H is

, .

measurem nt d d t z b E t / e t g
rei t
u ceh lunar o e ar t s
’
cen er ves e

distan ce ; from which wi t h the use of the E phemeris the

G re nwich time of the o bservation becomes known ; and also
e

his lon gitude if he noted the local time of t h Observation e .

2 66 A star is said to be occulted during the t ime it is out

.

of sight b hind the M n The beginning and d of the

e oo . en

occultation that is t h instants of disappearance and

, , e re

app aranc f the star call d its im m i and m i

e e o , e e rs o n e e rs o n , a re

phenom na capabl of b ing obs rved with c nsid rable

e e e e o e

accuracy The Eph meris gives the n c ssary el ments for

. e e e e

computing the W k i gt tim s f occultati n f various as n on e o o o

stars as s en fro m any p int up n the surface of the Earth

e o o .

The l cal t im f t h occultati n being observed either of the

o e o e o ,

imm rsi n em rsi n and the Washington time being m

e o or e o , co

p t
u de the ,l ngitude b comes known
o A bs rvati n of thee . n o e o

local tim s f t h ph nom na of an clips f the S or

e o e e e e e o un

Mo n fu rnish s a similar d termina t i n f longi t ude

o e e o o .

26 7 Th computa t i ns requir d in the las t t w m thods

. e o e o e

a re q uit lon g ande c mplicated and the the ri s inv lvedo ,

o e o

re q uir much study f their mastery (Th m th d f cul

e or . e e o o

mi ti giv s rise als to rath r di ffi cult compu t a t ions

though not di fficult as hos j ust menti n d H ow ver
na o ns e o e ,

so t e o e . e ,

the tim and lab r exp nded w uld be fully r warded if

e o e o e

accurate resul t s w re obtain d B t any of th se m h ds e e . u e e t o

giv ris t r sults which are crude in comparison with those

e e o e

given by t h tel graphic m thod or

e by transp er t a t ion of e , o

chr nom t ers The meth d of ccultations requ ires the

o e . o o

gr atest amount f computing but also giv s the greatest

e o , e

accuracy of the methods named

,
.

2 68 Thr e c nditions stand in the way f the attainm nt

. e o o e

of accuracy by any meth d involving the Moon F irstly t h o .

, e

Mo n re q uires about 73 days to make one complete circuit

o 2
§ 2 70 . L ON G I T UD E E Y TIIE M O ON . 2 65

in its or b it a b out the Earth The apparent motion of t h . e

Moon among the stars is then about one -twenty -s venth as e

fast as the apparent motion of the stars relative to an

o bserver s meridian which furnishes his measure of tim
’

, e .

A y rror in det rmining the position of the Mo n is th n

n e e o e

multipli d by at least twenty -seven when it is conv rted into

e

tim in t h progress of the computat ion I f then the tim of

e

e e

transi t of t h M on for exampl could be o b s r ed as

. e

e o , e , e v

accura t ly as that of a star one would expect the errors in a

e ,

longitude computed from Moon culmina t ions t be tw t y o en

seven times as gr at as t h rrors of t h local tim deriv d e e e e e e

from the same numb r f star obs rvati ns e o e o .

2 6 9 S ec ndly t h m tion of t h
. M on is so di fficult t
o , e o e o o

compute t hat its p siti ns t vari us tim s as given in o o a o e

th E phem ris and als

e of cours t h data there given in
e ,
o e e

r gard t lunar distances and ccultations

e o in error by o , a re

amoun t s which becom whole sec nds wh n multipli d by the e o e e

fac t r tw n t y -s v n This s urc of rr r is oft n avoided

o e e e . o e e o e

in the meth d of Moon s t ransits ( culminati ns) by using in

o
’
o

the c mputation f ach nigh t the M n s right ascension as

o or e oo
’

c rr ct d t G r nwich som t her stati n f known longi or e o o o

tud b y dir ct bs rvati n on t hat same nigh t

o e e a ee ,

e , e o e o .

Thirdly t h limb dge f t h visible disk of the

,
e , or e o e

M n is nec ssarily the bject r ally observ d and this is a

oo , e o e e ,

ragged dg ra t h r than a perf ct arc for pu rp s s f

e e
”
e e ,
o e o

accurat m asurem nt e e e .

27 0 Th d t rmination of the p ints at which the

. e e e o

boundary b tw n A laska and B ritish A m rica ( 4 t m ri e I 1s e

dian ) cross s t h Yukon and P rcupin rivers was one of the

e ee

e e o e

comparativ ly f w ins t anc s in la t e y ars in which it was

e e e e

necessary to resort t obs rvations up n t h Moon to deter o e o e

mine important longitud To d t rmin the longitude

an e . e e e

by transportation of chronometers w uld have been exceed o

2 66 O E OD E TIC A S TR ON O M Y
. 271 .

in gly di ffi cult and costly for there is more tha a thousand ,
n

mil s of sl w i
e ig t i b tween the mouth of the
o r ver , n av a on e

Yuk n Riv r and either station A t a stati n near the point

o e . o

wh r t h Yuk n cross s t h boundary M on culmina t i ns

e e e o e e , o o

w re bs rv d
e o3 nigh t s F our
e f the results
e w ron 2 . o e e

r j ect d as w rth l ss Th other 9 gav r sults ranging

e e o e . e 1 e e

from 9 t with a w ight d m an f

h
22
m
o e e e o

F ourte n of t h se comput d r sults d pend upon t h M ns ’

e e e e e e oo

plac as c rr ct d by corresponding observati ns at G reenwich

e o e e o

or S F rancisco A t t h same station t w observ d

an l . e o e o cc u

t tia gave for t h s conds f the longitude

ons and e e o

and a solar clips gav A t a stati n n ar t h point e e

wher the P orcupin R iv r cr ss s th b undary 3 bserv d

e o e e

e e e o e e o , 1 o e

M on culmina t i ns computed by the

o f corr sp ndingo u se o e o

o bservations t h same nigh t s at S F rancisc Washing on e an o ,

G reenwich gav l ngitud s varying from 9 h m

t on , or 3 , e o e 2

t o wi t h a weight d m an of O obs rv d e e ne e e

occul t ation b t h imm rsion and mersi n gave for the

, o e e o ,

s conds of t h longitude
e Thes xampl s will serve
e e e e
*

t oindicate roughly t h possibili t i s of the lunar m th ds of e e e o

d t rmining l ngitude Experi nced observers took the

e e o e

obs rvati ns at both stations I t sh uld b not d how v r

. .

e o . o e e ,
e e ,

that in such high la t itudes ( the stations were near the A rc t ic

C ircle ) the trig n m tric condi t ions are u nfavorabl t
, o o e e o

accu rate time det rminations and t h climatic conditi n e , e o s

w r such as to make observing di fficult

e e .

2 7 1 Th se wishing to s t udy t hese lunar methods of

. o

determining the longi t ude are r ferred for details to D e OO

littl s P ractical A stronomy ; to Ch

:

A stronomy
’ ’

e t auven e s ,

vo l and in the A merican Eph meris ( aside from t h tabl s )

. 1 . e e e

m m pl C
y p pp
Fo r a o re co e te a cco u n t o f t h e s e o b se rv a t io n s se e o a st an d
G eo d e t ic S u rv e Re o rt ,
1 8 9 5, .
33 -336 1 .
2 68 G E OD E TI C A S TR ON OM Y . 2 73 .

C H P TE R
A V III .

I C
M S E LLA N E O U S .

t b t Ob i g Su gge s ion s a ou se rv n .

273 A mong t h . characteristics of a good o b server that

e ,

is f an obs rv r wh will s cure the maximum accuracy

,
o e e o e

with a given expenditure f time and mon y in making such o e ,

astronomical d terminations as are tr ated in this book may

e e ,

be mentioned the following :

H is with ut bias as to the results t be obtained his

e o o ,

prime motive being always to come as n ar as possible to the e

truth H e has that kind of self-control which makes it p ssi

. o

ble for him to pr ven t the knowledge that t h result he is

e e

securing is too small ( or too larg ) to check with other e

determinations from having the slightest effect upon his

obs rvations F xample h may know that his o b serva
,

e . or e , e

tions in making a tel graphic d t ermination of the longitude

,
e e

of a station are placing that stati n

,
farther west than it o

has b en fixed by a primary triangulati n and yet have no

e o ,

tendency t observe stars earli r or later than usual O

o e . r,

when in reading a microme t er upon an azimuth mark several

tim s in q uick succession he s cures thre or four readings
e e e

which agree almost exactly and th n one which di ffers from , e

th m by t w seconds (say ) thus maki g a bad looking br ak

e o , n e

in his record he wil l not suppress or spring this reading

, ,

th ugh it may serve to make him more careful with following

o

readings .
g 2 76 . OR S E R V IN G . 2 69

2 74 . H e is well aware
minuteness of the allowable o f t he
errors A student when warned that he must t apply any
. , no

longitudinal f rce to the head of the microm ter of a zeni t h

o e

tel scop will p rhaps xp rim nt for himself by purpos ly

e e ,
e e e e , e

applying a li t tl pr ssure whil making a bisection and not e e e ,

b ing able t o
e any appr ciabl m ti n will become i
se e e e o o , n

credulous as to t h cessi t y of t h warning A good e ne e .

obs rv r ; on t h oth r hand kn ws t hat h can secur b

e e e e , o e e o se r

va ti such that the combina t i n f errors from ll s urc s

ons o o a o e

produce an error in each result which is as apt to b l ss than e e

as gr ater than that valu ( j e although e e ;

0 3 is a fraction f t h appar nt width of t h line wi t h

o which e e e

he makes t h bis ction I n other w rds h kn ws that h

.

e e . o , e o e

can make p intings und r g od conditi ns f which t h errors

o e o o , o e

a re s o smal l as to b invisible in t h t el scop H kn ws e e e e . e o

that he can mak p intings wi t h a pr babl error of l e o o e : :

say with a tel scope with which it w uld be h pel ss to t y

,
e o o e r

t o se ea d - sixth of an inch i diam t r plac d

ro one mil n e e e one e

away (4 inch subt nds at mile ) e one

27 5 H e is consci us that t h m s t d licat manipulation

.

o e o e

is r q uir d H e knows that his instrument is built of elas t ic

. e

e e .

ma t rial and tha t unl ss h is xc dingly car ful to apply

e . e e e ee e

o nly such forces as n cessary he may r adily pr duce a re e e o

deformations in his ins t rum nt which though stric t ly in e ,

accordance with the modulus of elastici t y of the material

comp sing it are y t as larg as the larg s t all wable errors
o ,
e e e o

o f obs rvation O e may s m t im s secure striking ocular

. ne o e e

evidence of this by watching a bisec t i n in a r ading m icr o , e o

sc pe a horiz ntal circle ( in t h t l sc pe) while a p r

o on o or e e e o , oo

o bserv r mak s his p intings wi t h an th r of t h r ading

e e o o e e e

microscopes on t h ins t rumen t e .

27 6 A good bs rv r d es not consider his instrum nt t

. o e e o e o

be of fixed dimensions or shape even wh n no ext rnal , e e

 70 G E OD E TI C A S TR ON OM Y . 2 77 .

forces are applied to it H e knows that it is constantly .

undergoing changes of shape due to chang s of temperature ; e

that these changes even under the best conditions that he can
secur may produce errors f the same order of magnitude as
e o

t h observer s errors ; and under adverse conditions may pro

’
e

duce errors which are larger t han all the others concerned in
the m asurem nt e e .

With resp ct to movements u nder stress and under

e

thermal chang s the support of the instrument ( trip d

e , o ,

bl ck or pi r) should b consid red as a part of the i t

o ,
e e e n s ru

ment .

S gg st i ns b t C m p ti g u e o a ou o u n .

2 7 7 A lm s t t h firs t q uestion that arises on commencing

. o e

a giv n kind of computati n for the first time is To h w

e o o

many decimal places must each part of the computation be

carri d ? I f too f w figures are used the errors from the
”
e

cast away decimal places become larger than is allowable I f

e

too many plac s are used the computation b comes slow r

e e e

than is necessary — the w rk r q uired for interpolati ns i n

, o e o

whatever tables are used being especially liable to incr ase e

rapidly with an increase of d cimal plac s A good g n ral e e . e e

guide in this matt r is to carry each part of every computati n

e o

to as many d cimal plac s as correspond to two doubtful fig

e e

ures in t h final result That is when the compu t a t i n is

e .
, o

finished and the probable rror is computed there should in e

general b e two significant figur s in the probable error O

, ,

e . r.

in other words the probable err r should b between and o e 10

units in the last place I t may be allowable to drop one

,

1 00 .

mor figure than above indicated if to do so decreases the

e

w rk of computati n very much as in computing sextan t

o o ,

bservations for tim ( foot -note to Example 8 at the d e se e

of Chapter III ) I t is important after deciding upon t h

o , en

.
, e
2 72 C TR ON OM Y 5 3 8
G E OD E TI AS . 2 .

l be obtained by subtracting a mean from each of the

ua s

separate values the sums of the positive and of the n gative

, e

residuals so obtained must not differ by more than units 72

—
2

corr sponding t the last place f t h m an wh re is the

e o o e e , e n

numb r f the separate valu s

e o e

280 I n conv rting angl s in t o time or

.

. i e it is e , v ce v e rsa ,

abou t as rapid t use t h r la t ions 3 6 5 o e e 0

°
1
°

1
°
4
m
5 , as it is to use t h tables given for
1
’
1
”
e

tha t purpose on pag 56 f V g Logarithmic Tables and e 0 o , e a s

’

els wh re Th tabl s may b used as a check

e e . e e e .

28 1 When sev ral computations of the same kind are to

. e

b emade it usually saves time t carry along corresponding

,
o

porti ns t ge t h r F
o xample i n c mputing apparent
o e . or e , o

places all t h star numb rs m y be tak n out at one time

e e a e ,

later all t h valu s of l g ( G l e and on e o co s - so .

Th use of a fixed f rm for a computati n saves time and

e o o

mistakes Th f rm sh uld r present a logical ord r f work

. e o o e e o ,

and shou ld involv as little rep titi n of figures as p ssiblee e o o .

A ll scribbling mul t iplying dividing int rpola t ing t

, , ,
e , e c .
,

sh uld b done on separate shee t s of paper from the regular

o e

computa t i on .

s P rob a b le E rror .

28 2 The reader wh d
. s t und rstand the principles o oe no e

of least squares can ot h p to u nderstand t h l gic f the n O e e o o

formul giv n in Chapt rs IV and V f certain l ast -s q uar

a e e or e e

computations B t after a car ful perusal f . 83 85 a u e o 2 —2

statement of t h uncertainty in a certain value in t rms of

e e

the -called probable error should not be unint lligible to

so e

him
In the expression pro b able error the word
.

283 .

probable is not used in its ordinary sens but in a special e ,

 23 3 PR OB A B L E E RR OR . 2 73

technical sense To assert that t h probab le error of a cer

. e

tain sta t d value is i is to asser t the chances are equal for

e z z e,

and against the truth of the proposition that the stated value
does not di ffer from the truth by more than Thus t e .
, o

ass rt that the zimuth of a c rtain lin w st of nor t h as

e a e e e

d riv d fr m a certain series of observati ns is

e e o l 5 o : : o
”
.
,

is t ass rt tha t it is as likely that the t rue value f that

o e o

azimuth is bet ween and W f N as that it is . o .

s me value outside of t h se limi t s To assert that the prob

o e .

abl error f a single obs rvati n in a s ri s j

e o is to e o e e ;

assert tha t it is an ven chanc that any particular bs rvation e e o e

is wi t hin in either dir ction of t h t ruth O what is e e . r,

th sam t hing it is to assert that if a l ng s ri s f such

e e , o e e o

obs rva t ions were mad the chanc s that one -half f the
e e , e a re o

o bservati ns would giv r sults wi t hin

o f t h tru t h and e e o e ,

one -half w uld give r sul t s di ffering from t h

o truth by m re e e o

than
Mor accurat ly perhaps t h probable error should be
e e , , e

regard d as r f rring to accid ntal err rs nly withou t

e e e e
°
°
o o ,

r fer nce t possibl constant errors Thus the above state

e e o e .

m n t s should be modifi d to read as foll ows To assert tha t

e e :

t h azimuth of a certain lin

e west f north as derived from e o , a

c r t ain s ri s f obs rva t i ns is

e e e t o is to assert tha t
e o , :

if an infini t number f such obs rvations were taken unde

e o e , r

t h same averag conditions th ir mean would be as lik ly to

e e , e e

lie betw n and ee W of N as to fall outsid t h s . . e o e

limi t s A d t assert that the pr bable error of a singl

. n o o e

observa t ion l is t ass rt that it is an ev n chance

: : o e e

that that particular obs rvation is within 5 in eith r d i e 0 . e re c

t ion of t h mean which w uld result from an infinite number

e o

o f such obs rva t i ns mad und r t h same average condi

e o , e e e

Fo r t h e d ist i n ct io n be t w een a cc i de n t a l an d c o n st a n t e rro rs se e fo o t ~

n o te to 74 .
2 74 G E OD E TIC A S TR ON OM Y . 2 86 .

tions The second form of the statement is non -committal

.

as to possible constant errors a ffecting all the series alike

which would not be eliminated by increasing the num b er f
,

observations S uch a constant error would be introduced

.

int an observ d azimuth by placing the azimuth light

o e ,

unkn wingly a li t tle to one side of the monument which it is

o ,

suppos d t indicate e o .

28 4 There se ms t be some confusion between these two e o

conceptions of the probable error I t is a common mistake

.

among those wh use least s q uares to derive a probable err r o o

by methods which correspond t the econd form of stat o s e

ment above and then to assume that the first form f state
H ence one is always on the safe side t
, o

m en t is true . o

assume the second form of stat ment t give the true mean e o

ing of the probable error and to form an stimate of t h , e e

possibility of a constant err r from ther sou rces of informa o o

tion If there is no p ssibility of a constant error in t h

. o e

observations the two forms f statem nt ,

identical o e a re .

28 5 The r lation be t we n the probable error of a single

. e e

obs rvation and the total range b tween the largest and e

smallest values given by such observations is as follows I f

e ,

, :

th eprobable err r of a single observation is j one is to o : e,

expect that if a large number of such observations were

made only about one per c n t will fall outsid a total ra ge
,
e e n

of 75 tim s O if the probable error of a singl observa

e e . r, e

tion is l only about: : obser ation in hundred one v one

w uld b xpected t fall outsid a total range of

o e e o e

t t d V i ti Th e La i u e ar a on .

286 Until a f w years ago it was supposed that the lati

. e

tude of a given station was invariable D uring the last few

years a vigorous investigation of that assumption has b een
.

made b oth by means of w series of observations of the

, ne
2 76 G E OD E TIC A S TR ON OM Y . 2 88 .

made with zenith telesc pes for the special purpose

v a t io n s o

of de t ermining the latitude variation may be found in Coast

and G eodetic Surv y R ports f 8 9 part pp — 59
e e or 1 2 , 2 , . 1 1 ,

and for 8 9 3 p rt pp 44 5 8 The principal inv sti

1 a 2 0— 0

b
, , . . e

g at i of
onsthe variation by means of latitude o servations t

specially planned for the purpose have b n made by P rof

no

ee .

S C C handler Indeed his investigations first proved satis

a fac t H is r sults will be
. .
.
,

f t ily that such variations

ac or a re . e

found publish d in various n umbers of the A stronomical

e

J ournal for several years past A general statement o f the .

Mechanical interpretation of the variations of latitudes ”

by P rof R S Woodward will be found in the A stronomical

,

.
. .
,

Journal No 3 4 5 May 895

.
, 2 1 ,
1 .

S t ti
a on E rrors and t h e E con o m ics of Ob serv in g .

28 8 .auth r cannot close this book with ut calling

Th e o o

attention bri fly to phase of geodetic astronomy to which

e one

lit t le attention has apparently been paid but which is of gr at , e

importance in planning the as t ronomical work in connecti n o

with a g odetic survey namely the relation b tween the

e , , e

ec nomics of observing and station errors

o .

B roadly stated the purpose of the astronomical observa

‘

tions made in connection with a geodetic survey is to det r e

min the relation between the actual figure of t h Earth as

e e

defined by t h lin s of gravity and the assum d mean figur

e e e e

upon which t h ge odetic c mputa t i ns bas d This is

e o o a re e .
*

the purpose whether the astronomical observations b used

, e

simply as a check upon the geodetic operations wh ther , or e

they be used as a means of determining the mean figure of

the Earth I n determining the relation between the actual
.

figure and the assumed mean figure three classes of errors a re

Se e f o o t -n o t e to g1 5 .
 2 89 . C O OR R V N E C77 ON O M I S F SE I G
'
. 2

encountered t h errors of the geodetic observations ; the

: e

err rs of the astron mical obs rvations ; and the errors due to
o o e

th fact that only a few scatter d stations can be occupied

e e on

the large area to be covered and that the station rrors as , e

d rived f these f w points must be assumed to represent

e or e

th facts for the whol

e e a rea .

Negl ct the first class f err rs as b ing in t h province

e o o , e e

o f geod sy ra t h r than astronomy

e Th duty of the engineer
e . e

when planning t h astronomical work of a survey is to so fix e

th number and character of t h obs rvations t ach sta t ion

e e e a e ,

and t h numb r and posi t ion of the s t a t ions as t make t h

e e , o e

c mbined errors f the s cond and t hird classes a minimum

o o e

for a given xp ndi t ur B y incr asing t h numb r f b

e e e . e e e o o se r

va ti at a station the rrors of t h s c nd class may be

o ns e e e o

diminished the relation betwe n the numb r of observations

, e e

and the error of t h r sult being t hat said rr r is inversely e e e o

prop rtional t the s q uare f t h numb r of obs rvations in

o o o e e e

t h most favorab le case ( of no tend ncy to cons t ant

e rrors in e

the series of observations ) I f ther

e

any constant errors . e a re

a ff c t ing the s ries then the increas in accuracy with increase

e e , e

in t h umber f observations is slower than that stated

e n o

ab ve Th third class of errors may be reduced by i

o . e n c re a s

ing t h number of stations and distributing them as uniformly

e ,

as possible so as to diminish the area to which the result

,

from each station is assumed to apply .

28 9 To illustrate suppose the latitude observations for

.
,

a g odetic survey f a S t a t e are being plann d L t us sup

e o e . e

pose that t h engin er knows that wi t h the available z nith

e e e

t el sc p s and s t ar plac s an obs rv r can secur

e o e a la t itud e e e e e

with a probabl error of ab ut i from observations

e o 10 on

a single vening and that he can reduce this to l

e , by : :

observing on four evenings L t us assume t hat he estima es . e t

that it will cost the same on an average to observe on four , ,

2 78 C TR N M Y g 89 G E OD E TI AS O O . 2 .

nights at station as to observe t three stations on three

one , a

di fferent nights S h uld h plan to observe at t di fferent .

*
o e en

stations dis t ributed uniformly over the State on four nigh t s

at ach sta t i n or at thirty stations uniform ly distribut d f
e o , e or

one night only at each O bviously the answer d p nds e e

mainly on t h magnitude of the station errors to be expected

e

I f he estimates the station error by c nsulting the r sul t s

.

o b tained on the U S Lake Survey he will expect an av rage

e

. .
, e

station rror of nearly e with a maximum exc eding 1 e 10

”
.
L

I f he consults the p t gt of t h Survey of the Northern re or e

B oundary fr m the Lake of the Woods to the Rocky Moun

o

tains h finds that the average station error there was

e

with a maximum of and that in one case six successive

stations a total distance ofon miles along the line showed 1 00

a n arly uniform change of ab ut 4 p mile in di o 1 er one

tion If he consults the pu b lished results of still other

e re c

surveys his estimate of the stati n errors to be exp cted will o e

not be mat rially altered D oes it not seem evident that

e .

under such circumstances the thirty stations should be o ccu

pied one night each Yet the usual practice of g od tic

ou e e

surveys in this country corresponds ra t her to the plan of

observing at sta t ions on 4 nights each even though the
10 ,

observation rror in t h result from a single night is upon an

e e

average only -twentieth say of the station error one , , .

With longitudes and azimu t hs it will be found that the

ratio of t h rrors of the astronomical observations to the
e e

stati n errors is somewhat larger but not enough larger to

o ,

materially m dify the above economic problem o .

p k
y
b e in g e x e c t e d t h a t t h e o b se rv a t io n s a re t o b e t a e n a t t ria g u l a
It
m m l
n

t i n st a t i s b t h e sa e o b se rv e rs w h o e a s u re t h e h o riz o n t a l a n g e s o f

l
o on

p p
t h e t ria n g u a t io n

1 S e e Pro f ss io n a l Pa e rs o f t h e C o r s o f E n g in e e rs N o 2 4 ( La k e S u r
.

p p
'
e .

pp pg p
Re ) 81 4
l
vey o rt , . .

1: P ate o o s it e a e 2 67 o f t hat re o rt .
2 80 G E OD E TI C A S TR ON O M Y
. 2 90 .

C O NVE R SIO N OF M E A N SO LA R TIM E I N TO S I D E RE A L .

C o rre c t io n to b e a d de d to a m e a n so l a r in t e rv a l t o o b t a in t h e c o rre s p o n d in g s id e re a l in t e rva l .

1
m

fl 1 9 . 01 6
u 1
9 . 1 86
fl
fl 1 9 -
5 9 0
d
i 1 9 673
N x9 83 7
-
‘
20 . 00 2

20 . 1 66
20 .
3 30
2° '
49 5
20 . 659
20 . 82 3
20 .
98 7
2 1 . 1 52

2 1 .
316
2 1 .
48 0
2 1 . 644
2 1 . 809
2 ‘ ~
9 73
2 2 . 1 37

22 .
302
22 .
466
22 . 63 0
22 .
7 94
22 .
9 59
2 3 . 1 23

2 3 .
451

N
N

-
O
I
U

N
l
fi

fl
fi

fl
fi

fi
fi
fl

fl
h

h
H
 2 90 . C ON VE R S ION TA B L E S . 281

C O N V E RSIO N O F
’

M E A N S O L A R T IM E IN TO S ID E RE A L .

C o rre c t io n to be a dd d t
e o a m e a n so l a r in t e rv a l t o o b t a in t h e c o rre s p o n d in g sid e re a l in t e rva l .
2 82 G E OD E TI C A S TR O N OM Y
.
g 2 90 .

C O N V E RS IO N OF M E A N SO L A R T IM E IN TO S ID E RE A L .

C o rre c t io n to b e a dded to a m e a n so l a r in t e rv a l t o o b t a in t h e c o rre s p o n d in g sid e re a l in t e rva l .

24 m l
e a n so a r h o u rs 2 4
h c
3
m
d
555 o f si e re a l t im e .
2 84 OE OD E TIC A S TR ON OM Y . 29 1 .

C O N V E RS IO N OF S ID E RE A L TIM E IN TO M E A N SO LA R .

Co rre c t io n to be su b t ra c t e d f ro m a Si d e re a l in t e r v a l to o bt a in t h e c o rre s p o n d in g m e a n t i m e

in t e r v a l .
g 29 1 . CON V E R S ION TA B L E S . 2 85

C O NV E RS IO N O F S ID E RE A L TIM E IN TO M E A N S O LA R .

Co rre c t io n to be su b t ra c t e d f d e re a l
ro m a s i in t e r v a l to o b t a in the c o rre s p o n d i n g m e a n t im e

in v
t er a l .
 G E O D E TIC A S TR O N O M Y. 29 1 .

C O N V E RS IO N O F S ID E RE A L TIM E IN TO M EA N SO LA R .

C o rre c t io n t o b e su b t ra c t e d f d
ro m a s i e re a l in t e rv a l to o b t a in p
t h e c o rre s o n d in g m ean t im e
in t e r v a l .
88 G E OD E TI C A S TR ON OM Y . 2 92 .

CH A NG E PE R Y E A R IN T H E A N N U A L PRE C ES S IO N IN
D E L IN A TIO N C
fif
.

In un it s o f t he th d e c im a l p l a c e ( u n it 0
da

( 77 )
m
Th e d
s i e ar gu m e n t
1
is t he a n n u a l v a ria t io n in ght
ri a sc e n sio n .

R i g h t A sc e n sio n ( a m ) .

h oo m h 2 0m h m h 00m h 2 0m h 0 m h oom h 2 011) D 0m

3 3 3 40 4 4 4 4 5 5 S 4

3 2 2
1 40 1 43 I4 S
1 54 I S7 1 60
1 67 1 71 I74
1 81 1 86 1 88 1 90
1 95 2 00 2 03
2 08 21 4 21 7
222 228 23 1
2 36 2 42 2 46
2 50 2 56 2 60
2 63 2 70 2 74
2 77 2 89
291 3 03
3 04 317
318 332
3 32 3 46
3 46 3 61
3 59 3 75
3 73 3 89
387 404
400 41 8
41 4 43 3
\ 42 8 44 7
44 I 46 1
4 55 4 75
4 69 490
48 2 504
49 6 51 8
51 0 53 3
52 4 54 7
53 7 56 2
551 576 58 2
6 88 70 7 72 0
82 5 847 8 63
962 989 1 00 7 1 01 8
83 1 8 99 96 1 1 01 5 1 06 1 1 09 9 1 1 29 1 1 50 1 1 63

h 6h 6h 6h 40 111 h h h
oo m 511 5h oo m oo m 2 0m
1 2 0m 1 1 2 0m 1 1 1 1 m
5 1 1 7 7 7 40

Rig h t A sc e n sio n ( a m )

Ch a n g e a ll S i gn s w he n u si n g t h is l o we r a rg u m e n t .
2 92 . TA B L E S . 2 89

C H A NG E PE R Y E A R IN T H E A N N U A L PRE E SS I O N IN C
DE C L IN A T O N I
fi
.

In u n it s o f ti : f th d e c im a l pl a c e ( u n i t
}
0
a.

Th e sid e ar gu m e n t “

2 ? is t h e a n n u a l v a ria t io n in ght
ri as c e n sm n .

R i g h t A sc e n s i o n ( a m ) .

611 4010 7
11 00 1 0 7
11 2 011 1 11
7 40
01 sb oo m 8b z ow 8 11 4001

+2 +0 +4
1 42 1 39 1 34 1 28
1 56 1 53 1 48 1 42
1 71 1 67 1 61 1 55
1 85 1 81 1 75 1 68
2 00 1 95 1 89 1 81
21 4 2 09 2 02 1 95
228 223 21 6 2 08
2 43 237 230 221
2 57 2 51 2 44 234
2 71 2 65 2 57 2 47
2 86 2 80 2 71 261

3 00 2 94 2 85 2 74

315 3 08 2 98 2 87

3 29 322 312 3 00
3 43 336 3 26 3 14
3 58 3 50 3 40 327
3 72 3 64 3 53 3 40
3 86 3 78 3 67 3 53
40 1 3 92 3 81 3 66
41 5 4 06 3 94 3 80
42 9 42 0 40 8 3 93
444 43 4 42 2 406
4 58 4 49 43 5 41 9
4 72 4 63 449 43 2
48 7 4 77 463 446
50 1 49 1 4 77 4 59
51 5 505 490 4 72
53 0 51 9 504 48 5
544 53 3 51 8 499
559 54 7 53 1 51 2
573 56 1 54 5 52 5
71 7 702 68 2 6 57
8 60 8 43 81 9 790
1 00 4 984 9 56 92 2
1 1 48 1 1 25 1 09 3 1 0 54

1 8h oo m 1 8h 2 010 1 8 h 40 20 1 9h oo m rgb 2 0m rgb

40
10 2 oh 0011 1 2 0h 2 0m 2oh 40111

Ri g h t A s c e n sio n ( a m )

w w
.

Ch a n g e a ll S i gn s hen u sin g t h is lo g
e r ar u m e n t.
90 G E OD E TI C A S TR ON OM Y .
5 2 92 .

C H A NG E PE R Y E A R IN TH E A N N U A L PRE C ESS I ON IN
D E LIN A TIO N C
fif
.

In u n it s o f t h e th d e c im a l pl a c e ( u n i t
do .

Th e d g
si e a r u m e n t is t h e a n n ua l va ria t io n in ri gh t a sc e n sio n .

Rig h t A sc e n sio n ( a m ) .

h oo m h 2 0m h m 1 0h oo m h oo m h 2 0m h m
9 9 9 40 1 1 1 1 1 1 40

3 91 2 93 1 94

21h oo m h 2 0m 2 1 h 0 m 2 2h 001 0 2 2h 2 0m 2 2h 40111 2 3 11 0011 1 2 3 h 2 0m h 40m

2 1 4 2 3

Rig h t A sc e n sio n ( a m )

w
.

C h a n ge a ll si g n s wh e n u sin g t h is ol er ar g u m en t .
2 92 G E OD E TI C A S TR ON OM Y
. 2 94 .

294 . F
M E A N RE RA CT IO N ( RM ) B A RO M E TE R 760 M I L L I M E TE RS
I NC H E S
F
.

T em pe ra t u re 10
°
C .
50
'
.
( Se e
5 2 95 TA B L E S . 2 93

295 . C O RRE C TI O N To M E A N RE F RA C T IO N

M E A N RE F RA C A S G IV E N IN 5 94, D E PE N D IN G
2 U PO N

TI O N ( KM ) — o n d
. C ’
.
T H E RE A D IN G O F T H E B A RO M E TE R
R
.

( Se e

t.
0
1 1 5 1
c)
1 :
c
0 u

3 m 3 0
5 11 5 s
o
5 15 3
°

93 a? CB 83 a ? CB 8 3 a ? CB
0 0 0: 0 0 o : 0 0 0:
a c tag as 55 a”: 35
2
m “
m m m m m
“

0 0 . m 7
o 01 . 0 0 . 01 7

o cn . o 0 . m 7
 O E OD E TIC A S TR ON O M Y. g 2 97 .

C O RRECTI O N TO F C
M E A N RE RA TI ON A S G I VE N IN § 294
C H ED
.

D E PE N D IN G U PO N TH E RE A D IN G O F TH E D E TA
TH E RM O M E TE R .

R ( Se e

CD
8

0 -
89 5
43 -
9
44 .
4
1 1 3 45 . 0

1 1
4 45 . 6
1 15 46 . 1

1 1 5
1 1 47 2
7 .

1 1 8 47 . 8

1 1 9 48 .
3
1 20 48 .
9
1 21 49 .
4

50 . 0

50 . 6
1
51

51 .
7
52 . 2

52 . 8

53 .
3
53 .
9
- -
T
54 4 0 8 67

29 7 C
O RRE C
. ON I
F TI N
To
M E A N RE RA C O

OF 2 94 , D E PE ND

R
IN G U PO N A TTA C H E D
T H E RM O M E
( RM XCB XCD XCA )
T ER .

Se e 3 68 .

F
T em
a h r.
p . T
Ce n t
em p
.
.
2 96 G E OD E TI C A S TR ON OM Y . g 2 99 .

29 9 . F A C TO RS F O R TH E RE D U C TI O N O F TRA N S T TIM E I
O B S E RVA TIO N S .

Th e

w
si

fo r su b -
g n o f A is e x c e p t

AT
p o l a rs a n d fo r C it i s
,

po l a rs it h l a m p E ST ( Se e
.
fo r s t a rs

95, 9 7 ,
b e t we e n t h e z e n it h a n d
e x c e p t fo r su b -p o l a rs W i t h l a m p W
t he
ES
po l e
T ,
,

an
fo r B it is
d e x ce pt fo r
e x ce

su b
pt

h is t o p argu m e n t is t h e
‘
st a r s d e c l i n a t io n 3; 8 .
2 99 .
TA B L E S . 2 97

F A CTO RS F O R TH E RE D U
O B S E RVA TIO N S
CT IO N
.
O F T RA N S IT TIM E

Thi s g
t o p ar u m e n t is t h e st a r s
’
d e c l in a t io n :t 8 .

Th e b o t t o m l in e on t h is p a ge is l
t he c o li m a t io n f a ct o r C se c
2 98 O E OD E TIC A S TR ON OM Y . 2 99 .

FA CT
O RS F O R TH E RE D U CT IO N O F T RA N S IT TIM E
O B S E RV A TIO N S .

Th i
s top gu m e n t
ar is t h e
’
st a r s d ec l in a t io n : l: 8 .
3 00 O E OD E TIC A S TR ON OM Y . 2 99 .

F A CT
O RS FO R T H E RE D U CT IO N O F TRA N SIT T M E I
O B S E RV A T IO N S .

Thi
s gu m e n t
to p a r is t h e ’
st a r s d e c l in a t i o n : l: 8 .
g 2 99 . TA B L E S .
30 1

F A C TO RS FO R TH E RE D U CT IO N O F TRA N S T T I IME
O B S E RVA TIO N S .

Thi s top argu m e n t is fo r t h e st a r s

‘
d e c l in a t io n : l: 8 .

Th e b o t t o m l in e o n t h is p a ge is t h e co l l im a t i o n f ac t o r C se c
3 0 2 G E OD E TI C A S TR ON OM Y .
g 2 99 .

FA C TO R S F O R TH E RE D U CT IO N O F TRA N SIT TIM E

O B S E RV A T IO N S .

T hi s gu m e n t
to p a r is t h e st a r s
’
d e c l in a t io n z iz 8 .

°
61 1 °
62 °
62 1 ° °
63 4 64 °
644 °
65 °
651 66
°
661 ° 67 °
3 4
0 G E OD E TIC A S TR ON OM Y 29
.
9 .

F A C TO R
S F O R TH E R E D U CT IO N O F TRA N S T TIM E I
O B SE RV A TIO N S .

Thi
s top ar gu m e n t is t h e ’
st a r s d e c l i n a t io n : l: 8 .
2 99 . TA B L E S .
3 05

F A C TO RS F O R TH E RE D U
O B S E RVA TIO N S
C TI ON O F TRA N S T T I I ME
.

Thi s g
t o p a r u m e n t is t h e st a r s
’
d e c l i n a t io n :t 8

0 2 27
1
44
00 2
1
30 43
1 98 2 02 2 07 2 1 2 0
1 0 2 34 42
06 6
2 01 2 2 1 1 2 1
38
0
1 0 2 41
2 04 2 09 2 1 4 2 1 9 00 2
1
41 40

Th e bo tt o m line o n t h is p age is t he c o l l i m a t io n f a ctor C se c

3 06 G E OD E TIC A S TR ON O M Y.
g 2 99 .

FA CT O RS FO R THE RE D U CT IO N O F TRA N S T T I I ME
O B S E RV A T IO N S .

Thi
s top ar u g m en t is t h e
’
st a r s d e c l in a t io n : 1: 8 .

° ° ° ° ° ° ° ° °
1 73 1 73 1 73 1 7 41
° °
71 5 72 72 1 72 72 3 73 74
8 G E OD E TIC A S TR ON OM Y .
5 2 99 .

FA CT O RS F O R TH E RE D U CT IO N O F TRA N S IT TIM E
O B S E RVA TIO N S .

Th is to p gu m e n t
ar is t h e
’
st a r s d e c l in a t io n :t 8 .
2 99 . TA B L E S .
3 09

F A C TO RS FO R TH E RE D U CT IO N O F TRA N S T T I I ME
O B SE RVA T O N S I .

Thi s g
t o p a r u m e n t is t h e s t a r
’
s d e c l in a t io n 3; 8 .

Th e b o t t o m l in e o n t h is pa g e is t h e co l l i m a t io n f ac to r C se c
3 10 G E OD E TI C A S TR ON OM Y .
g 2 99
.

F A C TO R S F O R TH E RE D U CT IO N O F TRA N SIT TIM E

O B S E RV AJ TO N S

T hi
s top a r gu m e n t is t h e ’
st a r s d e c l in a t i o n : l: 8 .

° ° ° ° ° ° ° °
78 782 7 81
°
771 7 71 781 79 1
°
79 79 1 7 91 80 °

— o
3 12 G E OD E TIC A S TR ON O M Y .
3 00 .

3 00 . F A C TO RS FO R TH E RE D U CT IO N O F TRA N S T TIM E I
O B SE RV A T IO N S A T C O RN E LL U N IV E RS ITY .

w Th
h o se
is
l
ta bl e is c o m
d
a t it u e
ute
oes no t
d
p d fo r l a t it u d e 42 °

d iff e r f ro m t h a t v a l u e
It m a
b y m o re t h a n
wv
y b e u se d , h o e e r, fo r a n y
F o r a s t a t io n i nll st a t i o n
a t it u de
42
°
l
.

o r in a t i t u e 42
°
d t he m
O

a x im u m e rro r i n t he t a bl
e i s t wo u n it s in t h e l a st p ace .
 3 03 . TA B L E S .
3 I3

3 01 . C O RRE C TI O N TO T RA N SIT O B S E RVA TIO N S FO R

Th e c o rre c t io n is ne g a t ive w D IU RN A L A B E RRA T IO N

hen a ppl ie d to o b se rve d t im e s , e x c e
.

p t fo r su b -p o l a rs
.
( Se e

D e c l in a t i o n 8 .

3 02 RE L A T IV E W E IG H TS F O R TRA N S IT O B S E RV A T IO N S
C
.

D E PE N D IN G O N T H E S TA R S D E LIN A TIO N
’
.

Vw 6 Vw '
6 Vw

0 -
33 0 -
57

b
In t h e
ut o n e g
ppal ic a t io n
s i n ifi c a n t fi gu r
o f the
e .
m l pl
u ti ie r it ge n e ra ll y su f
fice s to e m pl y o

3 03 . RE L A TIV E W E I G H TS F O R IN C O M PL E T E T RA N S ITS .

F o r e ye a n d e ar o bse rva t io n s .

5 Lin e s in
Re t i c l e .
3 14 G E OD E TIC A S TR ON OM Y
.
3 04
.

C O RR C T E IO N TO L A TIT U D E F O R D IF F E R E N TIA L
RE F RA CT IO N .

Th e gn
si o f t he c o rre c t io n is t h e sa m e a s t h a t o f t h e m ic ro m e t e r d iffe re n c e .

( Se e 8 1 50 )

Z en it h D is t a n c e .
i6 G E OD E TI C A S TR ON O M Y.

m ( Se e 5
3 18 G E O D E TI C A S TR ON OM Y
.
3 07

771
 2
3 0 G E O D E TIC A S TR O N OM Y .
3 08 .

( 500 1 75

H o ur Th e Co rre c t io n to b e a ppl i e d
t o t h e L a t it u e o f t h e s t a t io n t o o t a ind b
an gl e be t he a pp a re n t a t it u el d o f Po a ris l
C o m u t e fo r t h e e c in a t io n
. p d d l
fo re o r 88 °
46 a n d t h e m e a n re ra c t io n
’
f .

a ft er
u ppe r
Cu l m in a L a t it u d e
°
L a t it u
°
d e L a t it u d e
°
L a t it u d e
°
L a t it u d e
°
t io n .
30 35 40 45 50

o
h oo
m
+ 1
°
+ 1
°
+1 °
1
°
1 4
’
1 4
’
1 4
’

o 1 5 + 1 1 5 1 5 . 2 +1 1 1 1 4 . 6 + 1 1 4 .
4 4 —1

+ 1 1 4 9 + 1 1 1 4 3 +1 + 1 1 4 8

+
0 30 . . 1 3 .

0 45 + 1 1 4 2 . 1 1 3 .
9 +1 1 3 .
7 1 1 3 .
5 + 1 1 3 .
3 + 1 1 3 . 2 +1 1 3 o

+ +
.

1 1 1 3 0 . 1 + 1 1 2 .
5 1 1 2 .
3 1 1 2 . 2 + 1 1 2 . 0 + 1 1 1 .
9

1 1 5 + 1 1 1 . 6 + 1 1 1 .
3 +1 1 1 . 1 +1 1 0 .
9 + 1 10 8 + 1 1 0 6 1 0 4
+
.

1 30 + 1 09 .
9 +1 1 09 .
4 + 1
+ 1 09 . 0 +1
1 45 + 1 + 1 + 1 07

2 + 1 05 05 04 04 4
2 1 5 + 1 03 . 0 +1 02 02 02 01 8

2 30 + 1 00 0 59 59 58 9 + 58 -
7
+0 + + +0
1+ 56 56 56 56 55 8
57 o 0 7 0 5 0 2 0 0
2 45 . . . .
55 .
5
3 0 53 4
--
l 0 52 52 52 3 +
3 1 5 0 0 0 49 + 5 o + 0 49 48 8 + 48 5
3 30 + 0 46 46 -
0 45 -+0 7 45 -
5+ 0 45 2 ‘
l0
’

45 0 + 44 7

3 45 + 0 42 4 + 0 42 1 +0 41 8 0 + 41 + 5 0 41 3 + 0 41 0 40 7
+
. . . .

+ +0 +
.

4 0 38 .
3 0 38 0 37 6
-
0
37 .
4 0 37 1 +0 36 8 36 5
+
.

-+
.

4 1 5 0 0 33 33 3 1 0 33 c c 32 5 32 -
1

4 30 +0 2 9 . 6 o 29 . 2 + 0 28 9 0. + 2 8 5+
. 0 2 8 3 + 0 2 8 o 27 6
+ 6 + +
.

4 45 0 2 5 . o 0 24 . 0 24 3 0. 2 4 .
+ 0 0 2 3 7 +0 23 4 2
3 . 0

5 +0 20 4 +0 20 . 0 +0 1 9 .
7 +0 9 1 4 + 0 1 9 1 + 0 1 8 . 8 + 0 1 8 4
+ +
. .

5 1 5 0 1 5 . 6 + 0 1 5 .
3 0 1 4 .
9 +0 1 4 . 6 + 0 1 4 .
3 +0 1 4 . 0 + 0 1 3 6
5 30 +0 1 0 8 +0 10 4 +0 10 . 1 +0 09 9 + 0 09 6 +0 09 2 +0 08 8
+
. . .

+ +
.

5 45 0 06 . 0 0 05 6 0 05 .
3 + 0 05 . 0 + 0 04 7 +0 04 4 +0 04 0

+ + +
.

6 0 0 00 . 8 o — 0 00 1
— 0 — 0

6 1 5
6 30
6 45
7
7 1 5
53 09 TA B L E S .
3 2 1

3 09 C O RRE C TIO N F O R E RRO R O F RU N O F A M IC RO M E TE R

w
.

Fw
Th e bu l a r v a l u e is t h e c o rre c t i o n t o t h e f d d in g
B
ta or ar re a .

Th e sig n of t he c o rre c t io n is t h e sa m e a s t ha t o f t he d iffe re n c e a c k wa rd — o r d

ar .

U se t h is ta bl e W it h su c h a m ic ro m e t e r a s IS d e sc rib e d in 1 89 a n d n o o t h e r
, .

B .
—F .

”6 ”
o
.2 o 1 1 2 1
0 . . .
4 3 . o
22 G E OD E TIC A S TR ON OM Y .
3 10 .

A Z IM U T H O F PO LA R IS C O M PU
‘
TE D FO R
D E C L I A IO 8 8° N T N

La t . La t . La t . La t . La t . La t . La t .

0
11 1
111 ° /
0
° 1 "
0
°
05
! ” 0
o }
5 0 05 05 43 47 05 51
o 30 0 1 1 1 8 0 1 1 25 0 1 1 33 0 1 1 41
0 45 0 1 6 53 0 1 7 04 0 1 7 1 5 0 17 2 7
1 o 22 23 0 22
38 0 22 53 0 23
1 1 5 0 2 7 48 0 2 8 06 0 2 8 25 o 28

1 30 0 33 05 0 33 26 0 33 49 0 34 1 3
1 45 0 38 1 3 0 38 38 0 39 04 0 39 32
2 43 1 2 0 43 4° 0 44 09 44 40
2 1 5 0 47 58 0 48 29 0 49 02 o 49 36
2 30 o 52 32 0 53 06 0 53 42 0 54 1 9

2 45 0 56 52 o 57 29 o 58 07 o 58 48
3 1 58 1 01 37 1 02 1 8 1 03 01

3 1 5 1 04 47 1 05 28 1 06 1 2 1 06 58
3 30 1 08 1 9 1 00 02 1 09 48 1 10 36
3 45 1 1 1 1 1 2 1 8 1 1 3 06 1 1 56
33 3

1 1 4 28 1 1 5 15 1 1 6 05 1 1 6 57
4 1 5 1 1 7 04 1 1 7 52 1 1 8 44 1 1 9 37
1 2 1 02
4 30 1 1 9 1 9 1 20 09 1 2 1 57
1 22
4 45 1 2 1 1 4 1 22 05 59 1 23 55
5 1 22 48 1 23 40 I 24 35 1 25 32

5 1 5 1 2 4 1 24 53 1 25 48 1 2 6 46
0 26
5 30 1 24 51 1 25 44 1 40 1 2 7 38
5 45 1 25 20 1 26 1 3 1 27 09 1 28 07

6 1 2 5 2 7 1 26 1 9 1 27 15 1 2 8 1 4

6 26 1 26 59 1 27 57
1 5 1 2 5 1 2 1 04

6 30 1 2 4 34 1 25 2 7 1 2 6 21 1 2 7 1 9
6 45 1 2
3 36 1 24 2 7 1 25 2 1 1 2 6 1 8
7 1 22 1 6 1 23 06 1 23 59 1 24 55
7 1 5 1 20 35 1 2 1 2 5 1 22 1 6 1 23 1 0

7 30 1 1 8 34 1 1 9 22 1 20 1 2 1 2 1 05

7 45 1 1 6 1 3 1 1 6 59 1 1 7 48 1 1 8 39
8 oo 1 1 3 33 1 1 4 1 7 1 1 5 04 1 1 5
8 1 5 1 1 0 34 1 1 1 1 6 1 1 2 01 1 1 2

8 30 1 07 1 7 1 07 57 1 08 40 1 09 2 5
8 45 1 03 43 1 04 22 1 05 02 1 05 44

1 01 07 1 01
9 0 59 54 1 30 47
9 1 5 0 55 49 0 56 23 0 56 58 0 57 34
9 30 0 51 3 1 0 52 01 0 52 34 0 53
9 45 0 46 59 0 47 2 7 0 47 57 0 48
10 0 42 1 6 0 42 42 0 43 08 0 43 36

x0 1 5 0 37 2 3 0 37 45 0 38 08 0 38 33
10 30 o 32 20 0 32 39 o 32 59 0 33
10 45 o 27 09 0 27 25 0 2
7 42 0 2 8
1 1 o 21 51 0 22 04 0 22 18 0 22 32
1 1 1 5 0 1 6 2 8 o 1 6 38 o 1 6 48 o 1 6 59

1 1 30 0 1 1 01 0 1 1 08 o 1 1 1 4 0 1 1 22

1 1 45 0 05 31 0 05 34 0 05 38 o 05 42
E l o n ga t io n :
1 25 2 7 1 26 20 1 27 1 6 1 28 1 4
11 m s. 11 m s. 11 m s 11 m s

H o u r-
. . . . . . . . . .

a n gl e d 57 09 5 57 02 5 56 55 5 56 48
3 24 G E OD E TI C A S TR OZ VO M Y .
§ 3 10 .

A Z IM U TH O F PO L A RIS C O M PU TE D FOR
D E C L I A IO 88 °
N T N

La t . La t . La t . La t . La t . La t .

’ ” ' ” ’ ” 1; , H I
06 06
0 O
06
° 0
6 06
°
0
2 0 3 2 0 39 45 52 o 07

1 2 1 03 0 I3 1 5 I3 29 0 I3 43 0 x3
50 0 3
1 9 1 1 o 1 9 30 0 1 9 48 20 08 0 20 29 0 20
0 2 6 1 6 26 0 2 10
2 5 2 7 o 25 51 43 7 0 2 7
31 36 o 32 05 0 32 36 33 09 33 44 0 34

37 36 0 38 1 1 38 48 39 27 0 4° 09 0 40
43 2 6 0 44 0 7 44 50 45 35 0 46 22 0 47 1 2

49 04 0 49 50 so 39 51 29 0 52 2 3 0 53 I9

54
59
29
4°
0

”
55 2 0
35 ON 56
o r
1 4

34
57
02
10

36
0

1
58
03
10

41
0
4
1
59
04
1 2

49

04 34 H 05 34 m 07 44 r 08 54 n 10 O8

—
H 1 6 4 1 2 35 1 1 50 u 15
09 1 2 1 0 3
-
1

H 38 06 1 8
1
1 1 25 n 19
1 3 30 I4 7
H 8 n 2 1 1 6 1 22 n 2
1 7 2 9 1 41 39 u
4
08 2 5 04 1 26 N 8
a
21 A 22 23 r
32 2

24 25 H 2 5 43 28 3 1 1
3 0 01 -
t
v
31
2 20 N 2 8 40 3I 33 f 33 07 H 34
7

N
2
52 H 31 I4 34 1 2 r 35 48 H 37
9
M 33 2 4 36 2 5 1 38 04
Q
t
- 39
N 35 1 0 38 1 4 1 39 54 H 41

H 36 30 r 39 37 1 41 1 8 i
t
- 43
H 37 2 6 1 40 34 1 42 1 6 N 44

N
H 37 57 1 41 05 1 42 47 H 44
H 38 02 1 41 10 1 42 51 H 44
M 37 43 1 40 49 1 42 30 N 44

36 58 1 40 0 3 41 42 43
‘
! a 1

M 35 48 n
u 1 3 8 50 1 40 28 42
H 34 1 5 -
t
t I 3 7 I3 I 38 49 40
H
N
32
29
I7

56
u

n
r

1
35 n
3 2 46
r

1
36
34
45
1 7 H 38
35

N 27 I3 -
c
1 2 9 56 1 31 2 5 m 32
-
y

H 24 07 n 1 2 6 45 1 28 1 0 4
29

20 41 1 23 2 5
n 1 2 1 24 33 n
‘
F
‘
F 1 6 55 a 1 1 9 1 8 1 20 35 u 2 1
‘
P 1 2 49 n 1 1 5 05 1 1 6 1 8 n I7

N 08 2 6 m 1 10 33 1 1 1 41 -
4

-
1 2

-
H 03 45 m 04 43 r 05 06 47 07
1

43 t 1

58 59 42 1 38 02
4
O 49 0 1 01 37
O S3 39 0 54 27 0 55 1 8 0 56 1 1 0 57
O 48 IS 0 48 58 49 44 0 5° 32 0 51

0 42 39 0 43 I8 o 43 58 o 44 4° 0 45
0 36 0 37 2 6 0 38 01 o 38 38 0 39
31 26 26
0 30 0 0 31 55 0 32 0 32
0 2 4 0 25 1 8 0 25 42 0 2 6 06 0 2 6
0 1 8 0 1
9 04 0 1 9 22 0 1 9 40 0 20

I! 30 1 2 2 3 0 12 0 1 2 45 0 1 2 57 0 1 3 09
l l 45 06 1 2 O 06 0 06 23 0 06 29 0 06 36
E l o n ga t io n :
A z im u t h .
36 36 1 38 03 I 39 35 1 41 1 1 1 42 53
fi t 5 11 m . S. 9
. . 11 . m s b . m . s

H -
our an gl e 5 55 52 5 55 43 5 55 34 5 55 2 4 5 55 1 4
§ 3 10 . TA B LE S .
3 5
2

A Z IM U T H O F PO L A RIS C O
E C L IN A TIO N 8 8 ° D M PU TE D FOR

La t . La t . La t . La t . La t . La t . La t .

m ” 1 ” ° ; ”
0
h 1 5 0
0
O7
,
60 0
0
07 08 0 07 1 6 0
0
0
0
34
II
0
0

0 30 0 I3 58 O I4 1 3 O I4

52 O 2 1 IS O 2 1
O 45 0 20

I 0 27 40 0 2 8 1 1 O 2 8
I IS 0 34 2 1 0 34 59 O 35

38
-
u 30 0 40 52 O 41 0 42
45 1 2 O 48 05 0
c
0 47
n
w IS
0

0
53
59
I

1 2
9
-
O

c
54 I9

1 8
o
t
30 4
1 04 49 H OK

u 45 H 10 2 6
w ‘
F 15 32
w 5 I N I9 I7

w3
w 45
0 M
N
2

2
4
8 N 40
41

F - H 31 I7

b- 1 5 34 29
‘
F

F -3 0 H 37 r IS

- 45
A N 39 a
n
L H 41 m 29

M
1
IS 43
0 SS
1
0 30 H 44 54
1

0 45
C
O
- 44 2 6
0
0 IS
M
H
44
44 H 31
08

Q 30 H 43
Q 45 H 42 01

Q H 40 1 8
Q
Q 30
I
S
H
H 38
35 H
Q
D
45 32
X
C
"
? 29 I4
O
1 5 d 25 29

N
O
O
O 30 N 21 2 3
O
O 45 ” I7

1 2

IS 07

30 02

45 57
51

IS 0 45
30 0 39
45 0 32
0 2 6
IS 0 20

II 30 0 1 3
06
E l II
o n ga t io n :
A z nm u t h
4S 0

1 44
11 m
H o u r-
ange l .
5
.

55
.
3 2 6 G E OD E TI C A S TR ON O M Y .
§ 3 10
.

M
AZ I
EC L D
U TH O F PO L A R IS C O
IN T IO N 88
A °
M PU TE D FOR

Lat La t . La t . La t . L a t 54°
. . La t . La t .

, 0 ! 11
08
0
07 0 0 05

46

1 1 30 o 1 4

E l on
g
1 1
t
a io n
45 0 07

r
1 55
A m
Ho u -
ang e l .
5
.

54
.
3 2 8 G E OD E TIC A S TR ON OM Y .
g 3 1 1 .

311 . NOTATION AND PRINC IPAL W O RKING

FO RMU L/E .

Th e f o l l o w i g g l t i i ed t h r g h t t he
n e n e ra n o at o n s us ou ou

b ook Th e s p e i l t ti i l ed i e h w rk i g f
c a no a on n vo v n ac o n or

d b el w e h gr p f f r l
.

m u l a w ill be f o un o ac ou o o m u ae .

G E N E RA L N O TA TIO N .

an d 5 E t he pp r e t ri gh e i
a a n de l ti t a sc ns on and c in a
r e p e ti e ly t he ti e t he e r i
0! on ,

s c v at m o f o bs va t o n

de r i de r i
,

un co n s at o n

t he e ri g h t e i de l i ti
.

a
m
and E m an a sc ns on an d c na on , re

spe c t iv e ly

an d ( w it h yae r ript
a ) t
as he m e a ri g h t
s u b sc E an

a sc e i ns on de l i ti r e p e ti e ly t he
an d c na on s c v at

e gi i g t he fi ti i y e r i d i ed
, ,

b nn n o f c t o us a n cat

t he v l e t he e g i i g t he
.

“ an d 6 a u s o fa an d 6m at b nn n of

fi ti i y e r d ri g wh i h t he e rv ti
o m

c t o us a u n c obs a on

un de r i de r ti w m deco n s a on as a

pr p e r i p e r y e r ri gh t e
.

d [t in n sio n an d
’
4
1 an o m o t o n s, a a sc

de l i ti r e p e tiv e ly
,

c na on s c

l tit de
.
,

A a u

e it h d i t e
.

C n s an c

tr m i l l tit de t he t t e rv
z .

as ono ca a u o f s a io n of obs a

ti on

h r- g l e m e r ed e t w r d w e t w rd f r m
.

t an a su as a or s a o

t he pp e r r h t he e ri d i t he e
ou ,

b an c o f m an as cas

i de r ed p itiv e
u

m ay lw y
b e, b ut a a s cons os an d

e v e r e x eed i g (
,

°
n c n 1 80 or
 31 1 . N O TA TION AND W ORKIN G F OR M U L / E .
3 9
2

t he i az ut h
r e r ed t he e t w r d
m o f a st a m a su to
we w r d f r r h t he e y b
, as a

or st a om no t as m
i de r ed p itiv e d e v e r
cas a e,

b utlw y a a co ns

ex eed i g
s os an n

c n

va u l e i f
, di i i f l l
n a rc , o one v s on o a eve .

W O RKIN G F O R M U L /E ,

W ITH TH E IR SPE C IA L N O TA T IO N , AND W ITH RE FE RE N C E S

TO C O RRE SPO N D IN G PO RTI N S O F O TH E TE XT A N D TO
TH E TA B L E S .

To c o n v e rt m ea n so la r to s ide re a l t im e

t b le
.

Se e 2 3 an d t he a o f 2 90 .

To con ve rt s id e rea l t o m ea n t im e

e
.

Se e g 2 4 an d t he t a b l s of 2 0
9 ,
To in t e rp o la t e a lo ng a c h ord .

t he fi r t f r m b e i g ed w he t he i t e rp l ti i de
s o n us n n o a on s m a

f rw rd f r m h
o a l e F d t he e
t e va u d whe it i de an co n n m
b k w r d f r m t he v l e F F i t he r e q ir ed i t e rp l t ed
o ,,
s s a

ac a o a u s u n o a

v l e f t he f ti rr e p d i g t he l e V f h
,
.
,

a u o unc on co on n to va u o t

i de p e de t ri b l e V d V t he dj e t t ed
s , e

n n n va a an a re a ac n st a

l e f h i de p e de t v ri b l e wh i h rr e p d t he
.
, ,

va u s t e n n n a a to c co s on

gi e l e E
o

v n va u d F f h f
s i an S 3 ,
o t e u n ct o n . ee 0 .

num b r assi g n d t o ac h formula corr s p on d s to t h at us d in

b o d y of t x t
Th e e e e e e t he

the e .
 G E OD E TIC A S TR ON O M Y .

To in t erp ola t e a lo ng a t a nge n t .

f
o
V
’
a ,

F , is t he r e q ir ed i t e rp l t ed l e f t he f ti
u n rre o a va u o un c on co

spo n di g n t he l e V f h i de p e de t v ri b l e V d
to va u , o t e n n n a a an

r e p e iv e ly t he gi e l e f h i d p d
.
,

F, ct n ea res t v n va u o t e n e en

rr e p d i g l e f t he f ti
a re , s ,

d h
'

en t var ai bl e, an t e co s on n va u o un c on .

g i e fi r t d i ffe r e ti l e ffi i e t rr e p d i g
F
’
a
i h s t e v n s n a co c n co s on n

to V ,
. Se e 3 1 .

To in t e rp o la t e a lo ng a p a ra bola
t he fi r t d i ff e r e i e fi ie t ie
.

If s n t a l co f c n s a re g v n ,

w — VI)

(3
’
aF
a
l
F1 = F x + I 1
V — Vx
] 7 + W
1

%( VI — Va )

a cc o rd i g
he t he r t he i t e rp l ti m de f rw rd fr m
n to w n o a on is a o a o

V w rd fr
or V
b a ck at he re q ir ed e rp l ed l e
om F , is u in t o at va u

t he f ti rre p d i g t he l e V i de
.
, ,

o f unc on co s on n to va u o f t he n

p e de t ri l e V V t he dj e t ed l e
,

n n va ab and a re a ac n st a t va u s o f
t he i de p e de t v ri l e wh i h rr e p d t he g i e
.
, ,

n n n a ab to c co s on v n

vl e t he f i t he g i e v l e F
’
a
a u s F, an d F, o f un ct o n , and v n a u s

t he fi r t d e r e t l
dF
of S s i ff n ia c o e ffic xe n t . ee 33 .

t he fi r t d i e r e ti l e fi i e
If s ff g iv e n a co f c n ts a re n o t n ,

F. F.— F , —
VI V.
— V
[ VI a] , (4)
33 2 O E OD E TIC A S TR ON OM Y .
3 1 1 .

er l v l e da l6
et e
n
cl
‘

c d
t he nu m ic a a u s o f a n d so m im s fo r
dt d, ’ ’
a t

e h tr t ed i
”

t
:
6“ d
i
a a

2
m m
ac s a . is ab u a l n 2 9 2 .
( N B . . an d

i l de
nc u b o t /t t h e e ff e t f pr e e i
c o c ss o n and o f pr p e r o m o t i on .
)
Se e 4 0—43
H a v ing giv e n a ,
a nd 6 t he m ea n pl a ce f
o a st a r a t t ne

begin n ing f
o tb c fi c t it io u s yea r d u r ing co n c /
z i t /z e o bse rv a t io n s

un ile r c o n s ide ra t io n were m a de , t o com p ute t /z e a pp a re n t pl a ce ,

t t ne in st a n t bse rv a t io n
’
a ana 6, a of o .

at : a fo —-
If I — 1 74 + Tgg l
s in
( 0 + ) t an 6,

i e)
a c

s in ( +
H se c 6, ( in t m ; ( 10
)
-- a
l o )
+ 4 co s ( H + ) a o
sin 60 + i co s 6, ( in a rc
) ; ( II
)

in wh ic h f G H giti e ll ed i d p d
It an d i a re q uan t s ca n en

e r h i h f ti f t he ti m e ly d
, ,, , , e

r
e n t st a -
n um b s w c a re un c ons o on an a re

g i e i h E p he m e ri f e e ry W h i g
v n n t e me i d i gh s or v as n to n an m n t

The ir l e f t he i t f b er i de ri ed
.

va u s
y b
or b n s an t o o s vat o n m a e v

i e rp l ti l g h r d b e wee t he E p he e ri l e
y
m
‘

nt o a ons a on c o s t n s va u

i t he e l p ed p r i fi i i y e r expr e ed i
s .

1 s f h a s o t on o t e ct t o us a ss n

un it f y e r I i g i e i t he E p he e ri wit h t he t r
s o one a t s v n n m s s a

mb er
.

nu S 46 49
s . ee — .

To com p ute t he c o rrec t io n to a t im e ie c ep on m ea n t im e f ro m

a o u ble wit /z
’
o bse rv a t io n s f
o tb e a lt it u de of tb c Su n a sex ta n t

c ia l b o riz Se e 6 2 — 70
’
ana
fi
a rt i on

e re d i g t he ext t rr e t ed i de x
. .

Th e m an a n of s an a rc co c fo r n

e rr r e e tri i y
o an dA t he ppr xi m e l i de
cc n c t is Z A is a o at a t tu

t he t he e t h d rr e ti g i de x e rr r
.
“ “

o f Su n Fo r m o o f co c n fo r n o se e

e e ri it y g 7 T l tit de
.

62 , an d fo r cc n t c se e 6 . he a u

e m d i m e t e r +p R
A A, j ; Su n
’
s s i- a .
g 3 1 1 . N O TA TION A ND W ORKIN G F OR M U L /E .
333

e i d i m e t e r t e fr
Th e S u n
’
s s m - a as ak m t he E p he e ri m is
dded i f t he S l we r l i w e r ed
, n o s,
’

to be a un s o mb obs d su b

tr ed i f pp e r l i w e r ed
as v , an

act t he u mb as o bs v

p r ll x p g i e le
.

T he a a a is n in t h e t a b f 2 93

r e fr i
, , v o .

T he R R act on is RM CB CD , an d

g i e i t he le re fr i
M , ,

C a re n n t ab s o f 2 94
—2
97 Th e act o n is
r e q ired l i de
A v .

u A
fo r t h e a t tu A not .

o
— A
n O .

srn
2
é;
s in 4 3K ( qb s in 3K ( 5
0 ( ”
I
cos q) co s 6

Se e g e er l
n a n o tat o n i 6 m ay be ob ta i ed f r m t he
p he e ri
i e rp l i l g ge
. n o

E m by n t o at on a on a tan n t

e e
s

pp r r i
.

i 12
h
l t T a a n t so a t m

e l r ti e T
A .

T E m an so a m

E i h eq i f i e wh i h i g i e i t he E p he m e ri
, M .

t uat o n o t m c s v n n

f W hi g pp r e ed
s e s

d tonb b i f h
a a n t noon an m a e o ta t
f b er i b y i e rp l i l g h rd
or as n y , n or e

in st a n t o o s vat on n t o at o n a on a c o

me r e d i g f i e pi e e
.

T an a n o t m c

r e q ir ed rre i i m e pi e e
.

T T AT
M
_ : u co ct o n to t c .

t ko o bse rv a t io n s t a ke n u s in g t /z e t u ra l b o ri
f a re a t sea na

t he m e re d i g
an n o f t he s ex tan t a rc c o rr e ed f i de
ct or n x

e rr r e e ri i y
z on , a

o an d cc n t c t is A “ ( no t

A e i d i e e r +p R
A “ j: Su n
’
s s m - am t D ip .

T fi r f r t e rm t he e d m e e r
he st in s co n mb a re t he sa m e
e f r e T d ip d w w r d i l i ti t he l i e f
ou s

b he or o n a nc na on o f n o

i gh t he pp r e t h ri he i gh t f t he e x
as o

t
.
,

t to o z on d u e t o t he o s t an

e t he r f e t he gi e e
s a a n

ab o v o f se a , is v n in t h e t a b l in 2 98

r e i de r t he m p t ti gi e e
su ac .

T he m a n of co u a on is as v n ab o v .
3 34 G E OD E TIC A S TR ON O M Y .
3 1 1.

TO C O M PU TE TH E C O RRE C TIO N TO A SID E REA L TIM E PIE C E

F RO M O B SE RVA TIO N S W ITH AN A STR O N O M IC A L TRA N SIT

PL A C E D IN TH E M E RID IA N 9 0—
99
i e me (b
. .

r it f t r r
If t h e t m s o f tll) ans o a s a a c o ss s o ut not a

f t he l i e f h r e i l e we r e
n b e r ed t he ti e
s o t ef t c o s v m t o

r i r t he f t he r e ti l e ed
o ,

p t
,

t an s t li
ac o ss y b m m ea n ne o c m a e co u

by h f r t l e o m u a

sum
of o of
b s qua
rv t orial in t rvals
d lin s )
(
m an of o b s rv d t im s num b r of o b s rv d l n s
e e

e e e ( se c 6)
e e e e 7
e e e i e

( sumf ofm ssquatorial int rvals

m an of o b s rv d
e e

) ( 5 °C 5
‘

0 i ed 1m es
t ”. e e e

F o r t he pr e oc ss o f fi din n g t he e q t ri l i t e r
ua o a n va s l se e

1 14 .

--B b -- b ;
tm I I
“ ‘

in w hi h T is
’
t he bs e r ed i e t m o f t rahsit a c o ssr t he m e an

i e rr e t ed f i l i i f t he h ri t l xi d f
c c
o v

l n co c or nc nat o n o o z on a a s an or

d i r l b e rr i at o n

ed
u na a .

B Q 6 i b l i 5 99 ta u at
‘

se c n 2

e he e de ed
co s s .

b fli p p b i g t pi i q l it y riv n vo t n ua as

i d i t ed i 94 5
h ,

n ca n , I1 .

i f t he l e v e l d i i i v s ons a re n um be red f r m t he m i dd l e t w rd
o o a

e h ac en d, w and w b
’
ei n g t he we t s r e d i g t he
en d le a n s o f b ub b
 33 6 G E OD E TIC A S TR ON OM Y .
3 1 1 .

Th e obs e r ti e q ti
va on ua o n s a re o f t he f r m o

A T —
I—A w a W
c
—
I Cc
( a = O,

Cc ( a T4)

fo r t he b e r ti o de wi h i ll m i i
s va ons m a t u n at o n w et s

r e p e ti e ly
r m l e q ti
s c v .

Th e no a ua o n s a re

—Z A
E A Tc —I m Te )
'
E A B AE 2 66 2
Wa W
E A W A T. Q
E A Wa W +2 41 W 6 0 2 A W m — Tc )
’

Z A A Tc +Z A ’
A C —
+ E C 2 A E (a
2 — Te')
E EaE
——E C QC — 2 C( a
+ 2 CA W a —-2 CA Tc ) =
’
E CA TC I
W f E aE
0 .

Th e so u l ti on o f t he e f r e q ti
s ou ua ons g iv e t he v l e
s a u s of

A T, , w an d c

e e rr r i gl e e r ti
a aE

pr
.
, ,

Th e obab l o o fa s n ob s va on is

6
no

in w h i h t he c t he r e i d l ’
v s a re e rv ti e q t s ua s o f t he o bs a on ua io n s,
n, t he m e r
is nu e r ti b o t he m e r
f obs va o n s, an d n ,, is n u b o f
u n kn ow ( ns e
an q ti
d o
) f n o rm a l ua on s

pr l e e rr r t he m p t ed T
.

T he o b ab o VQ
o f co u A is in
wh i h Q q tit y t i ed f ll w eq i ( )
, e, e ,

c is a uan ob a n as o o s: In uat o n 47
wri e Qt t he p l e
in T t he p l e
ac o ( fA I in ac o f2
p l e t he ot he r l t e t e r m t he l v e
, ,
u

and o in t h e ac o f a b so u s, a n d n so

fo rQ
w m d i fi ti t h i m et h d p ti g w h i h
.

Fo r t o o ca ons of s o o f co m u n c

ed i f i de red d i l e
,

m ay be us co n s a v sa b se e 1 09 , 1 10

t he f r m i f e q l w e i gh t ( e p e d
,

mp t i
.

Fo r o o f co u at o n un ua s d n
g 3 1 1 . N O TA TION AND W ORKIN G F OR M U L A I .
337

in g p h de l i ti f t he t r d t he
u on t e c na on o s a an nu mb er li e of

f t he r e ti l e b e r ed i g ed t he e p r t e
n s

p ) c o s v on a re a ss to
b er
o u n s a a

o s i v a t o n s, 3 se e 1 1 1 —1 1 .

To C O M PU TE TH E L A TITU D E F RO M O B SE RVA TI N S M A D E O
W ITH A Z E N ITH TE L E SC O E P 1 46

tit de f r i g l e p ir
.

Th e l a u o m a s n a o fs tr
a s is

40 + ( M 44 3
2+ u
f ) 44
’
4 n n 1

+ (
4 R

( 57)
In t he pri m ed l e tt e r rr e p d t he r he r r s co s on to no t n st a

f t he p ir M i h m i r a e t e r r e d i g expre ed i r
s t e c om a n tu
g l r l e f t he r h - d d
o : ss n ns ; r

i h t e an u a t r va u d o o ne u n ; n an s a re no t
t h - d r e d i g r e p e ti e ly f h l e e l f h r he r
s en an

a n s, s c v o t e v t t
t he t he r t r R i h r e fr i
so u en , or e no n

r d
,

’ ’
st a an d f n an s or so u n s a ; s t e act on

h r ed i t he m e ri d i f t r b e r ed ff t he
, ,

and m t e u ct o n to an o a s a o s v o

m e ri d i
Th l e e l rr e ti gi e ei f l l b
an .

co b c on as v n a ov s or a tu

wh i h rri e g r d i f wh i h h m b e ri g i r e e
e v e ve e

s a a uat o n o c t e nu n

e h w y fr m l e e l b e gr d
c ca nc as s

h idd l I f h
t e - m i i e t e v tu a uat o n

m b e r ed t he t he r h l e e l
ac a o . s

nu i ly f r co n t n uo us d om one en to o t e v

rr e i b e m e Z { (
7
a
(
’ '
S (
co ct o n ) co s g n s n ee

Th e r m } (R R ) i b l ed i 3 4 (S
e t 4
’
s ta u at n 0 . ee

Th t e r e i t b l ed i m 3 5 s(S a u at n 0 . ee
3 38 G E OD E TIC A S TR ON OM Y 3 1 1 .

To co m b in e t ne se p a ra t e v a lu es f
o b
g and t o com pu t e tb e
ro ba b le e rr o rs —1
p 1 54 5
e e rr r i gl e er i
.

Th e pr o b ab l o o fa s n o bs va t o n is

in h i h t he
w c t he d i fe r e e t i ed
A
’
tr ti g
s a re f nc s ob a n by su b

e r e l t e h p ir f r m t he r e l t e h e p r t e
ac n

t he m a n su fo r ac a o su on ac s a a

i h t r h t p ir [AA ] t he
n g f om t a t he q r e t hea ; is su m o f s ua f
t he t t l er e rv ti
s o

A
’
n is o
p a t lnum b o f obs a ons ; an d is t h e ta
e r p ir e r ed
s ; o

n um b of a s o bs v

pr l e e rr r t he m e r e l f r m
.

T he o b ab p ir o o f an su t o any o n e a

in w h i h t he t he r e i d l t i ed y tr ti g t he
c
’
v s a re s ua s o b a n b su b ac n

i d i ri m i t e m e
n sc nat he t i f r m e an re su lt fo r s a t on o t he m
r e l t f r e h p ir t he q r e
an

su om[ ] ac a ; an d vv is t h e su m of s ua f
t he
s o
’

pr l e e rr r t he m e w de l i ti
v s.

T he ob ab o o f an o f t he t o c na f
t he r
ons o

p ir
st a s o fa a is

in w hi h c
3 40 G E OD E TIC A S TR ON O M Y .
3 1 1 .

tE dded r ed f r m t he ri gh t e i
to o r su b t t he
act o a sc ns o n of

we e r e e r e l g ti r e p e ti e ly g i e
a

tr fo r a st n or a st n on a on s c

t he i de r e l ti e e l g i
s a ,
v , v s

a m o f on at o n

rr e ti r re gi e
s .

T he c on fo r c u va t u is v n in 3 06
l e el rr e ti gi e
co

f r m l ( 7)
.

T he v co c
(7 )on is v n by o u a: 7 an d 8 o f
1 60

ti e i t e r l e d rr e p d i g
.

If t is t h e m n va in s co n s co on n to
i r e t e r he t he v l e r e xpr e ed
s one

t r n o f the m c o m t n a u o fone tu n

e d
u , ss

in co n
( 5
s o ) ( S
f a rc is g 1 co s 6 t ee

h i l e t ti l l pp l i ed rre ti
s .

T o t s va u m us s be a t he co c ons fo r
c hr e er r e
o nom t r e fr ti ati d i t ed t he
and fo r ac o n as n ca at end o f
161 .

To Co mp u t e t he L a t it u de fro m an Obse rv e d A lt it u d e of a

S t a r, or tb e S u n , in a ny p osit io n , t b e t im e b e in g fin o wn .

Se e § 1 7 1 .

To Co mp u t e t /z e L a t it u de fro m Z e n it n D ist a n ces of a S t a r,

or t ne Sun , Obse rv ed n ea r t ne M e rid ia n , t ne t im e b e ing k n o wn .

Se e 1 72 .

To Co mp u t e t ne L a t it u de fro m Obse r v a t ion s of t b e A lt it u d e

of Po la ri s a t a ny H o u r-a ngle , t b e t im e be ing k n o wn .

Se e 1 73 .

To com put e az im u t h fr om o bse rva t ion s u pon a c irc u m p ola r

st a r wit /z a d irec t io n in st ru m e n t .

x m p l e f re rd 8 7
E a o co 1

E x mple f
.
,

a m p t ti o co u a on 2 00

L evel rre ti
, .

co c on

{( + )
CL w w
’
—tan A
,
 31 1 . N O TA T1 ON AND W ORKIN G F OR M UL / E .
341

fo r e el h i
a l v av n g it s di i i
v s o ns nu mb e red t h w y fr t he
bo om

m i dd e
a s

l .

CL w ——e
{( l ) ( w +fl } ;
’
tan A

fo r e e l m e red
a l v nu b i
c o n t n u o u sl d ire i
y in one ct on t he
pri ed t e r r e f e rri g re d i g e p ii
,

m le t s n t o t he a n t ak in t h e t

wh i h t he e ri g i r e e w rd e C
s n os on

in c n um b n nc as s to a t he a st

gi e t he e f r l rr e ti ir l e r e d i
.
L as

n by s o m u ae is t h e co c on t o t he c g
t he r p t he pp iti h t ir l e g r d i
v c a n

fo r st a u on su os on t a t he c c a uat o n

i re e
nc l
as s w i e d ir ein a c o ck s ct oni
sin t
t an 2
c o s ob t an 6 s1 n ob co s t

Se e § 1 93
r t r e rr e ti
.

Cu va u co c on

t an 3

in w hi h At At, , A t, a re d i f e re e e t wee e h
t he f nc s b n ac

h r gl e e t he h r- gl e l e
c

ou -a n a nd the m an of ou an s, a n d t he va u s o f

t he e r 2 s in
w f r m t he le
’

é ,
d t
kn o t ab in
t m s e tc .
, a re n 7 o 3 0 .

( S 1 94

rr e ti di r e rr ti
e e

Co c on fo r u na l ab a on ,

w it h fi i e t r y e rly
su f c e a cc u a c fo r n a a ll c a s s 0 3 2 Se e
t he rr e ti i f pp l i ed
o r, n .

1 98 , 1T ig he o f co c on is a to

m p t ed i m t h t he t r e xpr e ed gl e
n
99 . s

t he co u az u o f s a ss as an an e a st
342 G E OD E TIC A S TR ON OM Y . g 3 1 1 .

o f no rt h and i f t he an gl e is m e r ed w e t w rd fr
a su s a o m t he

rt h
,

no .

To com pute az im u t n fro m o bserva t io n s u p on a c irc u m p o la r

st a r wit /z a pre ea t ing in st ru m en t

x mple r e rd
.

E a o f co 3 20

p ti i m de i d i t ed e wi h
.
,

Th e com uta on s a as n ca ab ov t t he

e ep i
x c t on o f e rt i
c d i fi ti
a n m o ca ons i d i ed i
n cat n 2 04 .

To co m p ute a z im u t h fro m o bse rv a t io n s u p on a c irc u m p o la r

st a r wit a a n eye pi ece m ic ro m e t e r

ex m le e rd
.

Se e a p o f r co an d co m p t ti
u a on , 2 1 0—2 1 4 .
3 44 G E OD E TIC A S TR O N OM Y .
g3 12 .

on g itu d g o d tic Po l ar D istanc

unar D istanc P r c ssion
L e , e e e 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0

an cli p tic P rim V rtical

L e e e o o o o o o o o o o o o o o o o o o o o o

an quator P ro b a b l E rror
Me E e e o o o o o o o o o o o o o o o o o

an Plac P ro p r M otion
Me E . e 0 0 0 0 0 0 0 0 0 0 0 0 0

an fraction fraction
Me e e

an olar fraction m an
Me Re Re 0 0 0 0 0 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0

an olar im lativ P rsonal quation

Me S Da y Re , e

an i gh t sc nsion
Me S T e Re e e E

an im of a microm t r
Me Su n R A e 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0

ri d ian x tant Paralla x

Me T e Ru n ( e e )

ri d ian in id r a l D ay
Me Se 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0

ri d ian Pl an i d r al im
Me L e S e e 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0

a d ir tan d ar d im
Me e S e e T e 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0

utation tar num b rs n d p n d n t

N S T e

-
b s rv r s rrors tation rror
N S e , I e e e

p tical ntr ransit

O E
’
e e S E o o o o o o o o o o o o o o o o o o

Paralla x ru Plac
O Ce e T

P aralla x s x tant rtical ircl

T e e

P rsona l quation
, e Ve C e

Plan tar y b rra t ion nit h D i tanc

e E Z en hh o o o o o o o o o o o o o o o o o o o o o o o o o

Po l
e A e Z e s e 0 0 0 0 0 0
0 0 0 0 0 0 0 0 0 0

e
 NDEX
I .

facto -
PA G E S

b rration corr ction

A , B , C, rS o c 0 0 0 0 o o o o o o o o o o o o o o o o o o o o o o o
2 96—3 1 2

b rration d i rnal annual an d

A e e
, ,
313

for w i t h transi t o b s rvations

A e ,
u , 1
4 4
- 2

corr ction for in a imut h com p u t ations 6—1

c o rre c t l o n , e . 1 1 1 8

d j ustm nts of
e , z .
21 6—2 1 8

of
A e , z o o -2 0 1

of transit 6 2 -6 5

of nit h
lta imut h d scri p tion
—
z e 1 53 1 56
A z , e 1 9 7—2 00

ltitu d d n d 2 0 0-2 0 1
a d j u st m en ts o f

of Polaris
A e , e fi e 14

pp ar n t p lac s com p uta t ion of

3 20

p p ar n t solar d a y
A e e -
, 5 54
0

A e 1 9

stronomical d a y
A pp a re n t s o la r 1 —2 1
9

imut
A
h
22

imu t h a d j ustm nt of transit

Az —2
.
1 9 7 40
Az
imu t h astrono m ical g o d tic
1 0 1 —1 03
e .

imut h corr c t ion tim co m p utation

Az ,
v e rs u s e e 1 6

imut h curvatur corr ction in com p utation of

Az e —1 8
e , . 1 1 7 1

Az ,
e e . 21 -
3 21 5

d ir c t ions for o b s rvin g

I4

201 —

d iscussion of
e e 2 05 2 20, 2 2 1 , 2 2 3 -2 2 5

2 3 0—2 3 3

d iurnal a b rration
x a m p l of r cor d
e 21 6—2 1 8
e e e 2 05 2 2 2 —2 2 5

m t h o d of
2 1 1 -2 1 3

micro m tric
e 2 2 0—2 2 3

o b s rvations u p on
e 2 2 3 —2 2 8

e Su n 2 34—2 3 8

3 45
346 IN D E X .

am p l s
w it h an astronomical transit
q u e st io n s an d e x e

w it h an n g in r s
.

’
e ee

Barom tric corr ctione e ,

Ce l stia l s p h r d n d
e e e ,
e fi e 10

h rono m tric
1 0 5—1 0 7

ircl r a d in g 2 53 —2 6 2
C e

C e e 2 0 6—2 09

ollimation corr ction transit o b s rvations 22

lin of d n d
C , e ,
e . 1 1 8—1 1 9

om b ination of r sults nit h t l sco p

e , e fi e .
98

o m p u t ation of a p p ar nt 6 8—1 74
C e ,
z e e e e . 1

of
C e -
, 4 , 5 54
0 0

of c h ronom tric lon g itu d s 22 5

of m an p lac s 6—
e e .
2 5 2 62

of o b s rva t ions w i t h nit h t l sco p

e e .

of t l g ra p h ic lon g itu d s
e z e e e e .

of ti m from s x tant o b s rvations

e e e

of transit tim o b s rvations w it h out l ast squar s

e e e .

e e , e e ,

of transit tim o b s rvations by l ast 1 0 7, 1 20 —1 2 5

o m p tin g su gg stions a b out

e e ,
e 1 2 6 -1 33

onv rsion an to si d r al tim 2 70—2 72

C u ,
e .

si d r al tim to m an solar
C e , m e e e e . . 2 4—2 5, 2 79—2 8 2

ountin g s con d s 2 8 3 —2 8 6
e e e e . . 26,

ovarr b ias m t h o d of o b s rvin g w it h 6 8-6 9

C e .

ulmination d n d
C u e e —
, 93 9 4

urvatur corr ction

C , e fi e 1 8

microm t r com p utation

C -2 1
e e , 21 3 5
e e .
31 5

Day
a pp ar nt solar
,

m ean s o la r .

clination d n d
c iv il a n d

nitions list of
De , e fi e

i f r ntial r fraction corr ction

De fi ,

of
D fe e e e ,

D ip se a
 3 4 8 IN D E X .

L atitu d e ,

from altitu d s of e c irc u m p o la rs .

from circumm ri d ian nit h d istanc s

fro m a lt it u d e o f

from o b s rv d al t itu d s
e z e e .

from nit h t l sco p o b s rvations

e e e .

z e e e e e .

v l va l u by com p arison w it h anot h r l v l

w it h y p i c microm t r
Le e e , e e e .

w it h l v l t st r
e e e e e e .

w it h t h o d olit
e e -e e .

ist o b s rvin g for nit h t l sco p

e e

L
of
,
e , z e e e e .

on g itu d astrono m ical

by c h rono m t rs
L e ,

e e .

by oon o b s rvations
b y fl a sh e s

by o b s rva t ions u p on u p it r
M e

e
J e .

L unar
sin
2
’
%t
3 6— 1
1
3 9
s in I

an p lac d n d
1 08— 1 1 2

r d uction of from y ar to y ar
Me e , efi e .
4 2
43
-

an
- 8
e , e e 43 4

to
Me 2 9 2 —2 9 3

an
Me a n s o la r 2 4—2 5, 2 79 —2 8 2

ri d ian d n d
Me Su n 21

Me , efi e 1 2 -1 3

M icrom t r corr ction f r un

e e ,
e o t ,
315
3 21

icrom t r val curvatur corr ction

-1 8 1 2 2 8—2 3 0
1 74 ,

icrom t ric
M e e ue , e e . .
315
—2 2 8

icrosco p s
M e 2 23

M
oon
1 —
8 1
e , 9 99
M
o b s rv d
,
0 . 6
e e 2 6 2 —2 6 7

p o sit io n 39

n
sin I
—
"
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 o o c o . 0 0 . 0 0 . 0 0 0 0 c o c o -
0 0 0 0 0 0 0 0 0 0 0
 IN D E X.
3 49

ort h PA G E S

ota t ion
N

u t a t ion
N
N

sin
sin
‘
2

b s rvin g cono m ics

I

b s rvin g list for ni t h t l sco p

O e , e

b s rvin g su gg stions a b out

O e z e e e e

ccultations o b s rv d for l on g itu d

O e ,
e .

O , e e e .

P rsonal
Pivot in qualit y
e

P lan ts
e .

Polar d is t anc
e ,

of
e .

Po la ris , a lt it u d e

Positions of oon an d stars

of
Su n ,
M ,

n
Pre c e s s io n , c ha n ge

Prim v rtical
defi d e

Pro b a b l rrors
e e

Pro p r
e e

Qu s t ions
e . . 2 7—2 9 , — 8 —
55 5 , 9 4 9 5, 4
1 6—1 —1 6 2 8 —2 0
49 , 94 9 , 3
1 4

at corr ction t i m
a d in g
R e e , e

Re

fraction corr ction nit h t l sco p

lativ w i gh t s transit
Re e , z e e e e

m t h o d of in o b s rvin g
Re e e ,

i gh t asc nsion co m p u t ation of

Re pe t it 1 o n s ,
e , e

R e ,

Ru n of a ta b l of
m ic ro m e t e r, e

h ori on d i p of
cular variation in p r c ssion
Se a z ,

x tant a d j ust m nts of

Se e e .

com p utation of
Se , e .

corr ction for cc ntricit y e e e

3 50 IN D E X
.

Se x tant ,
d e sc ript io n

o b s rvation at qua l a l ti t u d s
e s e e

r cor d
i d r al d a y
e

i d r al tim d n d
S e e . .

conv rt d to m an solar
S e e e , e fi e

ol ar tim a p p ar nt
e e e .

S e ,
e

tan d ar d tim
tars a pp ar nt motion of
S e

d istanc
S ,
e

u gg stions a b out
S t a t io n e rro rs

a b out
S e ,

Su n
osition of
, m ean

g ra p h ic
m p ratur corr ction
Te l e

im a pp ar nt
Te e e e ,

T e , e

-6 9
quation of e
,

m an e

T ransit a d j ust m nts of

,
e

d scri p tion of
d ir ctions for o b s rvin g
e

e e w1 t h

in p rim
in v rtica l of
e

out of
the e

the

t h or y e
 FIG | .

C E L E STIA L S P H osition
E RE of o
— C b s rv r
h is na d ir v r t ical circlnors t h sout h p ol snor t h ast souhtorip nit h
on an d e ; Z
e h is z e N
; P P z
h an d w st
HME OW : M
’
: the an d e z Z

p oin t s r s p ctiv l y qua t or a s t ar s m all

, ,

2 11 2 0 e e ; 0, E , H W : e e

circl d nin g m driiurnal m o t ion of s t ar our circl s

, , , , ,

QW
h
’
e e e ; A LE e ; S ; S SD 2 :

d ian la t i t u d of s t ation ni t h d is t anc

,

e e fi t he the PA P
’
PS F
’
- e ;

of star h our an gnort p olar d istan of star d clina t on of star

.

O PZ A H
h the e Z A e Z S z e e
2 :

l of star a l titu d of star a imut h of star

; PS ce ; LS e i ;
Z PS - e MS e OM : z .

F IG la

nort an d sout h p ol s quator

.

C E L E STIA L S P
w s H
cli pt ic
arro in E REh
d ica t in g a pp ar n t m o t ion of
— P P’ e E V QA e

vs t arrnal quino x au t umnal quino x ri gh t asc nsion of

,

VD A B e ( t he the e t he Su n ) ,

V e e ; A e ; VI , e

the S .
 RO NOC RA PH

C HRO N O M ETE R
B S E RV ING KEY S ( C LO S E D)

TE LE G RA PHE RS KE Y
MA IN LIN E

LOCAL BATTERY
MA IN LINE
S OUNDER

FT
G 30
.

LOCA L B ATTE RY

RO NO G RA PH

O B S E RVING KEY S ( C LO S E D)

C HRO NO M

NA L RE LAY

MA IN LIN E
E LEG RA PHERS KEY

RE LAY

A LBAT
MA IN LINE
SOUNDER
FK1 3 I
 -
S HO RT TI
TlLE C ATA LO G U E
’ ' ‘

O F THE

PUB L ICA TION S

J OH N W IL E Y SO N S ,

N EW Y O RK .

LON D ON : CH A PMA N H A LL, LI M D ITE

BJE CTS
.

A RRA N G E D UN D E R SU

Ds r
.

l
s ar
e c rip ti v e c i c u a rs se n t o n a p p lic a t 1 o n

l l
r
Boo k m k e d wi h t t
a n a s e risk a re so d a t n et p 1 1c e s o n
°

y .

A ll b o o k s a re b o u n d in c o l th l
u n e ss o th e wise st a t e d

A G RIC U LTU RE

D G D YP C C D
.

D G
CA TTL E FE E IN — A IR RA TI E — ISE A SE S O F A N IM A L S
GAR E N IN E TC

M nual of d ing
, .

Ca t t le F e e
’
A rm sb y s

D ow ning s ruit an d it r s
a l 2m o, $ 1 75
'
F T
’
E m
P rinci pl s of o d rn D airy P r ctic W oll
ee 8v o ,
5 00

G ro t e n fe lt
’
Th e M ( )
mo
s e e a e . .

K m p s an d sca p ard ning

12 , 2 00

L G
l o yd s ci nc of g icultur
’
e e e 1 2111 0 , 2 50

L S A 8vo
ard ning for ad s D ow nin mo
’
e e r e ,
4 00

L o n do n G L ie

t l s r atis on D is as s of D o g
’
s e .
( g) 12 , 1 50

S T
’
th e 8vo
r is D is as s of
ee e e e e t he .
, 3 50
T the th e Ox 8 vo, 6 00

ocks an d
ea t e on e e

St o c k b i idge R
’
8vo 2 50

Wo l s n db ook for rm rs an d D airy m n mo

s ,

Ha Fa 12 1 50
’
l e e . ,

A RC HITE C TU RE

B D G Y
.

U IL V
IN O
— CA RPE N TR — STA IR S— E N TIL A TI N , E TC

B rg s B ui ld ingsan d of m rican ailro d s

.

R
P lanningan d
St 1 u c t u re s A 4t o ,
’
e e a . . .

B irk m i1 e Am e 1 ic a n Th e a t i e s— Co n st ru c t io n 8 v o
t l
’
s .
,

8vo

om p ou d iv d ird rs
A rc h it e c t u ra l h em and S ee ,

R
k l ton o struction in B uild in gs
C n ete G e 8vo,

8vo

P l n ing onstruction of igh c Buil d ings

S e e C n ,

a n an d C H Offi e .

8 vo ,
 ints to rc h it cts H A
s rc h it ctural ngi r ng
e 8y o
F 1 e it a g A
’

anitary ous ns p ction mo

e E n ee i ”
1
1
s 8vo ,
7
. . .

G e 1 h a rd S H I
’

ir s an d Panics mo
3 e e 16 ,

Th ea t 1 e F
m rican ous ar p nt r
e 12 ,

:
I

H a t fi e ld H
’
A
’

olly s ar p nt an d Join r
s e e C e e

H C
’

K i dd r s rc h it ct s Morocco
e er e .

’
A d d B u ild e 1 POc k e t b
s ton s B uil d ing
’
e e an ok “ fl ap,
i
M e i rill S 1 01 d D e c o Ia t io n 8v o
t ir B uil d i g W on an d b to
e an . n : .
,

M o n c k t o rrs S ’
— l
W ait s ngin ring rch it ctural
a n o o d, I1 , St ne . 4 ,

E
’
ee an d A e 8v o
h p
. .
,

S
W orc st r s mal l os p ita s s a b l is h m nt an d int nanc
ee ,

S H l — E t Ma
’

incl u d ing tkinson s ugg stions for osp ita rch i

e e e e e

S H
’
A l A
t ctur mo
e

W or ld s o l um b ian xp osition of
e e 12 ,

C E 1 8 93 zit o ,

Y Y A RM NA V Et c
M Y G G O D C PO H G
, , .

IL ITA R EN IN E E R IN — R NA N E— RT C AR E S— LA W E TC
q

B ourn s cr P ro p ll rs to
, .

w
'

S 4 I5 001 )
’

rd nanc an d unn ry
e e e e ,

B ru fi 8 vo , 6 00

u b marin in s an d
s O e G e

B u c k n ill s S M ' ‘
Io rpe d o e s 8 v o , I 4 002i
h s s cr
’
e e .

C ae w P1 0 1) e lle rs 8 vo , 3 00
’
S
oo s val rd nanc
e I

ke
'

Na 1 2 50 1
’

C O 8v o ,

unn ry for m f c rs mo mor cco

1 1. 1 ‘

e . .

N o n -c o !O fi
'

C IO n k h it e G 18 2 00

avis s r atis on Mi l itary

’
1
1
s e e . . .
,
o ,
1

'

D T 8V 0, 7 00
’
1
e e

h
.

1 S ee p , 7 50- f

8 170 2 50

B rack s Ca v a l ry O u p ost D uti r morocco

I 1

De ; ( Ca 1 8 111 0 , 2 00
D i t s So l d i r s F irst A id
’

morocco
t es i .
,

1 2111 0, 1 25
’ ’

D r d g s Mo d rn F r nc h A rtil l ry to h al f morocco
e z e .
,
1

4 1 5 00

of x h i b i ts Buil d ing
’
e e e e e , ,
.

Re c o 1 d Tra n spm t a t io n E
W or ld ol um b ian x p osition of to h al f morocco
th e ,

C E 1 8 93 1 0 00
sis anc d P of h i ps
4
’
8 . .
, ,

D u 1 an d Re p u lsio n S 5 00
’
8vo
Dy r s i gh t
8 t e an 1o ,
i l '
1 1

1 2111 0, 3 00

o s aval actics
L
’
e

H ff N T l 50

unt r s P o t h arg s h al morocco

81 m
’

3 I
.
I

H 8vo , f 1 3 00
’
e r C e .
,

2
B ay for d x t book o od ic tr nom y Te - f Ge As I 8 vo
M c h i an d arl w s P ractical tronom y
’
et l 1
’

s o . . r
1

H As 8 v 0, $3 00
’
i
W h it s h r tic l an d D scri p tiv stron my
' ’
e o . .

e
’
T eo e a e e A o .
1
. . . 1 2m o , 1 2 00

O Y B TA N .

D GAR E N IN G F O R L A D IE s, E TC .
’

l dw i s O rc h i d of ngl an d
’
N ew E 8 vo ,
ard ning for ad i s D ow nin mo
Ba n s .
1 50

Lo n do n G L ( g) 12 1 50
’

h om s ructur B otan y mo
s e e .
,

T e St al 18 2 25
’

ral Botan y ch n i d r
,

W es t e rm a ie r s G e n e
’

.
(S e e .
) 8vo , , 2 00

B RIDG E S , ROO FS , E t c .
1

CAN TIL E VE R— D RA W — H IC H W A Y — SU SPE N SIo N .

( See a lso E N G IN E E R IN G , pf

i ghw ay B ri d g s
B o lle r s H 8v 0 2 00
h am s iv r B ri d g pa p r
’
e . . .
,

Th e T R 4t o
B rr s tr s s in
e e e ,
e ,

S B rid ge S x 8 vo 3 50
’

M c hanics of ird r
u es e ,

Cre h o re the G 8 vo 5 00
’

D r dg s h am s B ri d g s p arts p art
s e . e ,

T
’

7 1 25

D u s tr ss s in ram d tructur s to
e e e e , pe r ,

B 0 1s S F S 4 1 0 00
ost r s W T B i d g to
e e e e ,

F o o de n re st I r 4 5 00
’

r n s rch s in W oo d
e e es . .
,

G A 2 50
’

B ri d g russ s
ee e e , e tc . a

T 8vo 2 50
of russ s
e e ,

Ro T 8vo , 1 25
ow s r atis rch s
e

H ’
T A 8 vo 4 100
Joh nson s od rn r m d tructur
e e e on e 1 .
,
.

M F S 4t o 1 0 00
M rriman J x t b ook o oo s an d B ri d g s
’ 1
1
e a e es . .
1. .
,

a cob y s
’
Te - f R f
P art
'

e r e .

I
M rrim n J x t b ook o ofs an d B ri d g s
. ,

& Te - Ro
’
a c ob y s f
P art ra p h ic tatics
'

e a s e .

II G S
M rrim n Jacoby s t boo o oofs B ri d g s
1
. .

Te x - k R d
Part D sign
f
’
e a an e .

III B i id ge 8vo 2 50

M r iman J x t b ook of ofs an d B ri dg s

e .
,

ac ob y s
’
Te - Ro
P rt ontinuous D raw ti l v r usp nsion
e r e .

IV C Can S an d

rch d B r d g s
a , , e e , e ,

A i 8 vo 2 50
Morison s m ph is Bri d g O bl g
1 e e . . ,

*
Th e M e
’
e . . on 4t o , 1 0 00 1

4
W add ll s r n i ghw y B ri d g s “

H
1
8v o
a Pock t Ofor d
I
’
e o a e . . . .

D e Po n t ib u s ( bo k B Ii ge E n gm e e rs)
W oo d s of B ri dg s oofs
e .

d R
W r g h t s D signing of D ra w p ans
Co n st 1 u c t 1 0n
’
e an 8 vo,
S
’
i e . 8 vo ,

Y C HE M ISTR .

Q V Q U A L ITA TI E —V O G C O G C U A N TITA TI E R ANI — IN R A NI E TC

‘

ab oratory alcu latio

.
,

A d ria n c e L 1 C
ll n s a b l s for ron alysis
1 21
’
s ns 1 25
1
. 0,

A T I An
ust n s ot s for h mic l tud nts
’
e e 8 vc , 3 00
A
’
N C S 1
B o l ton s tu d nt s ui d in Quantitati v naly sis
e e e a e 1 2m o, 50
S G A
’
8 vo ,
nalysis by l ctroly sis
’
e e e . . 1 50
C la sse n A E ( H e rric k d B o lt wo od ) 8 v 0

Qualitativ nalysis
’ '

s e . an .
, 3 00

( Sc h a efi en )
'

A
h mical actions rril l mo
C ra ft s s 1 2 11 1 0,
’
e . 1 50

D re c h se l s C Re ( Me
Quantitati v h mica na l y sis
)
’
e . . 12 , 1 25
F re se u iu s s
’
ll C lA (A 8 v0
ual it ti v Jo h nso
e e . en ) , 6 00

Q ( 8 17 0, 3 00
W ll rans
a e n )
s) T
rman d ition
( l 6t h
1 1
e . .

Ge E
F u rt s W at r an d Pu bl ic al th '

He
ill s u l nal ysis
’
e e e
u
G G as an d F
’

P hy siolo g ical h mistry

e A .

H a m m a rst e n 0 C
elm s of at h matical h mistry
8v
’
s e .
,

H Prin c iple s M C
Kol b s nor g anic h m istry mo
’
e e .

I C
add s Quantitativ h mical na ysis
12
’
e e ,

L C A l 1 2111 0

n d au r s p ctrum
’
e e .
,

( S A n lIyS lS Tin gle 8vo

’
La
Man d l s B io c h mical mo
e e .
,
.

L a b o 1 a t o ry 12

ason s W at r su pp l y
’
e e ,

M - 8vo

nalysis of P otab l W at r
’
e ,

A ( I h )
Mi ll r s h mical Phy sics
e e . n t e p ress .

C 8 vo ,

l m ntary x t b ook of mo
’
e e

M ix t e r s E Te - 12

Morgan s h ory of ol utions an d its sults mo

’
e e . .
,

Th e T Re
‘

S 12

W at r su p p l y h mical an d anit r y
’
e ,

- S 8 vo

ab oratory ui d to h m cal naly sis

N ic h o ls s (C )
’
e e a .
,

G C i A 8 vo ,

ualitativ nal sis

L
’
O B rin e
’
s e e

Pe rk in s s Q y A
P inn r s h mistry ust mo
’
e .

O i ga n ic C (A ) 12

P ool s P o w r of u ls
’
e e . en ,

8 vo

ick tts an d uss l l s ot s on I

e
’
Ca IO Iifi c e F e ,

C h e m ist 1 y ( N o n
b l ong morocco
R R N n ic
’

m tall ic
e e e n o1 ga

e ) . O 8y o ,
Ru dd im a n s ncom p tib il iti I
'

V o l u mtr c l y sis
a es 1 n
n
8
l $2 00
Sc h im pf s i Ana
’

s g an d book orocco a s
e . . 1 2m o , 2 50
1

Spe n c e i
’
Su M a n u fa c t u 1 e 1 H 1 2m fl p
n dbook r h mists B t ugar ou
’
a1 s . o, m , 2 00
Ha fo C of S H
morocco
e ee se .

1 2m o ,

tockb ri d g s ocks an d oi l s
, 3 00
S R S
’
8 vo ,
h m try o ron
e . 2 50
Tro iliu s s C
’
is f I 2 00
W ll s s norg nic Qu l itativ aly sis mo
e 8vo ,
1

I An
’

b or tor y ui d Qualitati v Ch mic l na ly sis

e a a e . 12 ,

La G A 1 50

h mical ctur mo
a e ln e e a , 8 vo ,
W ie c h m a n n
’
C Le 12 3 00

uga naly sis

8 e e . . ,
l
S 8 A 2 50

Wu ll i g s P h an d mo
1 v o,

Me d Ch e m ist i y
’
n In m ga n ic ar . . . . . 12 ,
2 00

D RA W ING
Y
.

E L E M E N TA R — G E o M E TR Ic A L — TO PoG R A PH ICA L

i ll s h d s an d h ad o w s P rsp ctiv
.

H S S d
’

D scri p tiv om try

a e an e e e 8 vo ,
M a c Co rd Ge 8 vc ,
’

Kin matics
s e e e

8 vo ,
M c h anical i
e

D win g 8vc ,
M h ns g h om p so v ol s
e a

D
’
In d u st i ia l w1 n (T 8vo ,
d op ographical D aw ing to
a a 1a n .
,

Re e T ( II A ) 4

i d ours in ch nic l D raw i g

’
3 r . . .
,

Re A C Me 8 vo

c h a ical D raw ing l m nt ry M ch D sig

’
3 e a a n .

Me n an d E e e a a i
n e e n .

8vo

mit h s op ogr p h ical D r w ing acmi l l

.

S T v (M 8

W rr n s D scri p tiv omtry vol s

)
’
a a . an . o,

Ge 2 8vo
D fting nstrum nts
’
a e e e e .
, ,

I 1 2m o ,

r han d D raw i ng
ra e

F ee- 1 2m

igh r in r P rsp cti v

o,

H L 8 vo,

in ar
e ea e e e

L Pe 1 spe c t 1 v e 1 2m o ,
Mach in onstruction vo l
e

C 2 8v c ,

P l P ro b l m
e s ,

1 2m o

P rim r y om try
an e e s .
,

Ge 1 2m o

P ro b l s h or ms
a e

d T 8 vo ,

P ro j ctio nD r w in g
em an e e

1 2m

h ad s an d h d ow s
e a o,
, i
S S 8v o

t r oto y to utti g
e a ,

S — S C 8vo,

W h l p l y s tt grav i g
e e m ne n . ° . .

e e
’
Le e1 En n 1 2m o
l ‘ lu
‘
.

6
 D r d g s P nn Rail ol io h l mor d Co n St ru c t ion
'

t ‘
F $ 20 00
to h al morocco
’
f
D rink r s Tunn ll ing
e e e . ro a , e . .
. a .
,

i
'

4 f 25 00

p l osiv s an d Dynamit
’
3 1
e e . . .
, ,

E issle r s E x e — N it Io glyc e rin e 8 4 00

’

owl r s dam Proc ss for P i rs

e . vo,
: .

Co fi e r- 8 vo
’
F
r ar d s nitary ous ns p ction mo
e e e . .

Ge h Sa H I 16 1 0
0
o d w in s ailroa d ngin r s i l d b ook mo p ock t orm
’
e e ,

G
’
R E
’
F - 12 -
bk f 2 50
r s l m nts of o d sy
ee e . , e .
,

Go
'

E Ge 8vo , 2 50

ow ar d s ransition rv i l d book mor cco

e e e e

T Cu F - 1 2m o , fl ap 1 50
ow s taining W ll s d ition
H ’
e e . . o ,

H Re (N e w E 1 2m o , 1 25
'

)
d son s x cav ation ab l s Vo l
e a . .

Hu E T II 8 vo , 1 00
’

utton s M ch anical ngin ring of Po w r Pl a ts

e . .

H 8 vo , 5 00
’

E
oh nson s t ri l s of onstruc t ion
e ee e n

J
’
Ma C 8 v o, 6 00

tad i d uction D iagram h t inch s

e a

S Re S 2215 x 28 0 50

h ory an d Practic of
a . . ee ,
e ,

T Su 1 v e yin g 8 vo 4 00

K nt s M c h anica l P ock t b ook mo orocco

e e . .
,

E 11 gin ee r s - 12 5 00

w ag D is p osal mo
’ ’

e e e . . .
,
m ,

K ie rst e d Se
K k w oo d s ad Pi p for rvic P i p
’
s e 12 ,

ir Le Se 8 vo 1 50
ahan s ivi l gin ring W oo d
’
e e e ,

M ’
C En ( 8 vo , 5 00

rrima an d b ook for urv yors mo m or

ee . .
)
Me n an d H S
'

B ro o k

rrim n s od tic urv y ing

12 2 00
’
s e . .
, .
,

Me ’
Ge S 8 vo , 2 00

t ining W al ls Maso ry D ams

a e e

Re d 8v o 2 00

os l y s c h a ical n in ring
a an n ,

M ’
Me E g 8 vo , 5 00

gl s M nua l for ailro d ngin rs mo morocco

e n ee . .

Na R
’
E 12

tton s i vi l ngin ring

e a a ee . .
, , 3 00

Pa
’
C E 8 vo , 7 50
oun d tions
ee .

F 8 vo , 5 00

ockw ll s oad s an d Pa v m nts in ranc mo

a .

R R
'
’
e e e F e 12 , 1 25

R u ffn e r s N o n -t ida l Riv e rs

’
8 vo ,

i l d ngin ring mo morocco p s

'

1 25

S ea rle s s F
’
3 00

ilro d p iral morocco aps

e E ee 12 ,
fla ,

Ra S 1 2m o
-
fl 1
i b rt an d ton utting an d Masonry
a , . 50

M o d e rn S
'

B iggin
’
S C 8 vo , 1 50

mit h s a bl ram w ay s to
e e s e . .

C T
’
S
W ir anufactur an d s s to
e 4 2 50
'

. .
,

M U 4
p l d ing s oa d s an d P a v m nts
e e e , 3 00
S R
’
1 2m o ,

y d rau ic m nt mo
a e e . 2 00

H l Ce 12

h urst n s t rials of onstr ction

e , 2 00
T Ma C 8 vo 5 00

i vi l ngin r s P ock t book mo ap s

’
o e u .
,

Tra u t win e C
’

E 12 m or fl 5 00

ross s ction h t
’
s ee e - . . .
, .
,

C -e S
x cav tions
ee , 25

E a an d 1 . 8 vo , 2 00
8
i

add s a
i Tra u t win e s L yi ng O ut u v s
C
Poc t b ook f B
’
morocco
ngin rs
1 e . . . 1 2m o ,

W
mo mo occo
e ll D e Po n tib u s ( A ke
’
i id
or ge E ee ) .

W t s ngin ring an d Juris p ru d nc

12 , r ,

ai E A rc h 1 t e c t u 1 a l
’
8 vo ,
h p
ee e e

i l d O p ration in ngin ring S ee ,

W arr n s ton utting

La w o fF e e E ee , etc 8v o .

n g in ring nstrum ntsasonry Dams mo mor ccoto

St e 1 e o t o m y— S
’
C
W bb
e e .

e 8 E ee I e : 12 ,
o ,

W
W ll ington s ocation of ai l a y s
e gm a n n s
’
Co n st ru c t io n of M 4 ,

W h l r s ivil n g in ring
e
’
L R w 8 vo ,

Win d mi ll as a P ri Mo v r
C E
’

ee e ee . . 8 vo ,
W o lff s
’
me e . 8vo ,

Y H D RA U L ICS
H W D VC P
.

W A TE R -
W D GE E L S— IN M IL L S— SE R I E IPE — RA IN A E, E TC .

G G p ( Se e a lso EN

B in s x p rim nts u p on ont actio of i ui d V in

IN E E RIN , .

E the C the L q
’
az e e r n e

B ov y s r tis on y d rau l ics

( Tra u t vvin e ) .

ap h ical olution of yd rau l ic Pro b l ms mo

T ea e H
’
e . 8 vo ,

rr l s r atis on W in d s y cl o s an d orn do s S
C o ffi n Gr H
’
s e . . 12 ,

u rt s W at r an d Pu b l ic al th mo T
Fe e
’
T e e t he , C ne , a e . . 8 vo ,

Kutt r s l o w of W t r
F He
’
e e e 12 ,

n s i ltration of P b l ic W at r u pply ( He rin gdz

’

G a n gu ille t e F a e .

rsch l s p rim nts

H az F u S
'

e e 8 vo ,

w a D isp osal mo
He e
’
1 15 E x e e 8v c ,

Kirk w oo d s a d Pi p or r v ic Pi p
K ie rst e d
’
s Se ge 12 ,

ason s Wat r u ppl y

Le f Se 8 vo ,
’
e e e

M rriman s r atis on yd raul ics

M S 8vo
’
e ,

W at r u pply h mica l an d anit ry

e
’
T e e H . 8 vo , S

m p rov m nt ti d al i rs
N ic h o ls s
’
e S (C e S a ) 8 vo , NS

W at r u p p l y of ity of York to
R uf
fn e r s
’
I e e fo r N o u - R ve 8 vc , HR
-

W is b ac h s yd au l ics Du
W N ew 4
1

e gm a n n s
’
e S the C . .
,
yOS .
.

W ilson s rrigation ngin ing

8 vo , S
I

H
’
e r .
( O

yd r u l ic an d P l ac r Mining mo E 8v o R
-S
I
’
eer , I

W olf s W in d mill as a P rim ov r

12 O S
H a e .
,
I

W oo d s h r y o u b in s
M
O
f
’
e e Z 8v o , O S
’
T eo f T r e : NS

M ANU F A CTU RE S
CH
.

A N IL IN E — BO IL E R S— E XP L O V SI E S — IR N — O SU G A R— WA T ES

W OO L L E N S E TC

lB la nmont
s a bs l s oolforl r ann d nalWorst
y sis
.
,

d anufactur
A e
’
T e Io A

B o l lan d s ncy clo p d ia of oun d ing rms

e u W ’
en e M e
’
E ae F Te
9
B ol lan d s T Ir n F oun d r
Su p p l m n t
’
he o e ”

B oot h s Cl ock an d Wa c h Mak r s Manual mo

l “ 1
e e I 0
1 8

B o v i r s Han d b ok on O il Painting mo
’
8
’
t e 12 ,
0
1

E i l s E xp losi v s N an d D y namit
’
u e o . 12 ,
0
1 8

Fo d s B Making f Boi l r Mak mo

’
ss e r
gy e ,
1 t1 o l c e rm e e 8vo , 5 1 8

M tcalf ost of Manufactur s

’
r o ile 1 or e e rS J . . . 18 , 4 8

M tcal f s t l nual for t l s rs mo

C
’
e e S e . 8 vo , 08

ni l in ol ors
e
’
S ee — A Ma S ee U e . L 12 , Q 8 0
1

isig s ui d to P i c Dy ing
Re im a n n s A ’
e C . J . 8vo ,
‘
0
( 8

p nc r s ugar an dbook mo mor

Re G 1 l1 i
’
f
‘
e e e e . . . . . . 8vo , 25 00

an db ook for h is s of B t ous s

S Ma n u fa c t m H
’
S 12 fl a p, 8
’
e e er s “
, .

mo mor
H C em t ee H e .

an db ook or h arcoal B urn rs mo

12 , . fl ap, O
O OO

th ss
‘

Sve d e liu s s H f C H8
’
e . . 12

h urston s anua l of t am Boi l s

Th e La e an d It s U e .
D
C 8

W al k s ctur s on xpl osiv s

T ’
M S 8 vo
I

e er .
,
J
C 8

W st s m rican oun d ry P ctic

Le E 8v o - 0O
P
’
e e e , I

1 2mo ,
’

oul d r s x t b ook mo
A
’
F N 0O
1

e e r
a e .

ugar nalysis
M Te -
‘

12
J 1

e ,
N 0O
W
W o db ur y s r P rot ction i ls
ie c h m a n n S A
’
8v o O
0O
1

s , l

’
Fi of M l 8vo, O 0O
1
o e e l

M A TE RIA LS O F E NG INE E RING

STRE N G TE — E L A STICITv — RE SISTA N C E , E TC .

( See a lso EN G IN E E R IN G p
Bak r s Masonry Construction
, .

B ar d sl an d K t s Str gt h of Wroug h t Iron

’
e .

B ov y s Str gt h of Mat rial s

’
e ee en en .

B urr s E lasticity an d R sistanc of Mat ria l

’
e en e .

By n s Hig hw ay Construction
’
e e e s .

a p nt s sting ac h in s an d th od s o sting Mat rials

’
1 e

h urc h c h anics of an d lui d s

C r e er Te M e Me f Te e .

D u B ois s tr ss s in ram d tructur s to

C
’
S Me E n gIn e e rIn g— So lrd s F ”

ransv rs trains S F S 4
’
e e e e ,

Joh nson s at rials of onstruction

H a t fi e ld T S 8 vo,
’
s e e . .

an a s pp i d ch anics
’
M e C 8v
p ,

ril l s to s or B uild ing an d D coratio

L z
’
A l e Me . . 8 vo ,

M rriman s trc h anics o at ria s

Me r ’
S ne f e n 8 vc ,

gt h of Mat rial s Me f M 8v o ,
’
e e l . .

Patton s r tis on oun d ations

S en e 1 21 11 0,

ockw ll s oa d s Pav m n s ranc mo

T F 8 vo ,
’
ea e .

p al d ing s oa d s Pav m nts mo

R R an d F 12
’
e e e t In e .
,

h urston Mat rials of ons ruction

S
’
R an d e e . 12 ,

T 3 e C t
10
 War r s Pl n P robl ms
P ro b l ms h or ms $1
’
’ en a e e 25

P o j ction D raw in g
d T m 2 50
1

e an e e 1

W oo d s ord in t om try
1 50
‘
r e

”
igonom try
Co - Ge 1 Ui 8
’
1
a e e . . . vo, 2 00

W l f s D sc i p tiv om try
Tr e 1 2 11 1 0 , 1 00
Ge Ro y al 8 v o
’
1 3 00'
‘
oo e r e e
1

M E C HA NICS — s
M A C IIINE RY ‘

TE XT-B OO KS A ND PRA CTICAi .

( See a lso EN G IN E E R IN G p
B al d w in s St H ating for
, .

B j amin s W ink l s an d R ci p s
’
ea m e

a p nt r s sting ach in s an d M t h od s sti g

’
en r e e e
“

C r e e Te M e e of Te n
1

h ord a l s tt s oMch anics mo 8vc .

h urc h s ch anics of g in ring

C
’

Le er t e 12 , 2 00

ot s an d xam p l s in M c h anics
C Me
’
En ee . .
1

8v c , 6 00

M c h anics of ir d r
N e E e e 8 vc , 2 00
'

om w ll s B lts P ul l ys 8 vo
C re h o re
’
s e the G e , 5 00

aring mo
Cr 1 2n io,
’

e e an d e . 1 50

om p ton s ir t ssons in tal

To o t lre d G e . . 12 1 50
m
,

D a s l m ntary c h nics
C
’
F s Le Me . 12 o, 1 50
m
D ing y ac h in ry Pat n Making mo
E Me
’
na e e a 12 o, 1 50

rans x h i b its B ui l d ing W l d p osition

e
’
3 M e te r . f 12 , 2 00
D red ge
to h al f morocco
l

T E Ex
’
s , or ,

D u B ois s c h a ics Kin matics

4 , , 1 0 00

V ol tatics
’
Me n . Vol I . e 8 vo , 3

V ol Kin tics
. II . S 8 vo ,
i

t g ral d s B oston Mach inist mo

III e f 8v o ,
8
l

mo
1

Fi 18 1
’
z e .
, .

mo
M
Fla t h e r s Dyn a m o m e t e rs ” . . 12 , 2

all s u b rication ROpe

mo
12 , 2

ol ly s li g mo
’

H Ca r L 12
’
.
,

Joh nson s h or tical M ch a ics Sa w Fi

l m ntar y r atis
H ’
n 18 ,

T An E T
’
e e e n . e e e e .

Ji D si g n Pat Kin matics

( In the p ress ) .

P t an d P ro p ortio of
o re s Ma éh in e e . r I .
. e 8 vo , 1 50
I
i n
' '

M ch in Parts
ar II , St re r gt h
.

an p p li d M ch nics
a e .

L A
’
z a 3 e e a

:
f

M rriman s ch anics of at ri l s
M a c Co rd ’
s i
i‘
i
i

i
'

M tc l f ost of anufactu s
e
’
Me M e a .

I 1 11

Mich i aly tical ch anics

e a e 8
’
C M re “

Mos ly M ch nical ngin ing Mah a

e 8
’
An Me
’ I'l l “ ! I‘ II
e 3 e a E ee r .
( n .

12
 om pr d
b inson s P rinci pl s of c h anism
Ric h a rd s s C
’
e sse A ir . 50

mit h s P r ss w orking of tal s

Ro Me
’
e 8v o, " 3 00

t h an d s
S - Me
’

e
1
. . . 8 vo ,

h urston s riction an d ost W o k

Th e L a e It s Use 8 vo , 6 00
T
nim l a ac h in mo
F L
’
r 8vo , 8 00

W arr n s Mac h in onstr ction vo ls

r
1 ‘
r Th e A a as M e 12 , 1 00

W is b ac h s yd r ulics an d Du
C
’
e e u
1

. 2 .
, 8 vo , 7 50

M ch anics of ngin ing V ol Part

H Hyd ra u lic Mo t o rS
’
e a ( 5 00

l in
E III I
‘
e ee r . .
, .

(K e
W is bac h s M c h anics n g in i g Vo l Pa t
Se e I . . . 8 vo , 5 00

l in
of E III , I
’
e e . ee r n . . r .
,

W is b ac h s t am ngi s D u B ois
Se c II . .
(K e .
) . 8v o ; 5 00

W oo d s nal y tical c h anics

S E ( )
’
e e ne . .
1
. 8 vo , 5 00

l m ntary c h anics mo
’
A Me 8vo , 3 00

u ppl m t an d K
E e e Me 12 , 1 25
S e en ey 1 25

M RY
’

E TA L LU G

O D
.

O V OY IR N — G L — SIL E R— A L L S, E TC ‘

l l n s a b l s for on na y sis
.

o dl and M rcury
A T Ir A l
’
e e

M tallurgy o i l v r G
’
E gle st o n s e .r

rl s tall urgy opp r an d ron

e f S e

Ke M
t l ul
’
e — C e I

D ssin g in uro p
S ee ,
F e . etc

M tcal f s l anual for t l s rs

K h a rd t O re E
’
r
rn s re e

r atm nt of ol d O r s
e
’
Ste e — A M S ee U e

h urston s ro an d t l
O D risc o ll s T G
’ ’
e e e

l l oy s
T I S
’
n ee
1

W i l son s y ani d P roc ss s mo

A
C 12
’
e e e . .
,

Y M INE RA LO G A ND M IN IN G
CC D
.

MIN E A D GI E N TS— V E N TIL A TIo N — O RE RE SSIN E TC

in rals of omm rcial Valu g morocco

.
,

B d s V ntilation of Min s mo
B a rrin ge r s
’
M e C e e . . o b lo n ,

B y d s sou c s of out h W t rnVirginia

12
’
ea r e e “ ,

o out h W st n Virginia Pock t b ook fo m

Re S 8 vo ,
’
o r e es e . .

B us h an d sD t in ral o y
Ma p f S e er e r ,

h st r s talogu of in ral s
r Pe n fi e ld
’
e e rm rn a t rv e M e g 8v o ,

C e e
’
Ca e M e 8vc ,

D ict io aryof am s of in r l s
D ana s m rican oca iti s of in rals
n t he N e M e a
’
A e L l e M e

13
 n s D sc i p ti v Min ral ogy h l f rocco $
in ra logy an d JD
Da 1 2 50 1
’
a e r e e ”
(E S ) . . . 8vo , a mo ,

in als an d to tu d y h m l (
1
“
1 M e Pe t ro gra ph yv ( ’
. . 1 . i
. 1 21 11 0 ; 2 00 [

x t b ook of in ralog
. M er
1

. Ho w S T e . E . . 1 2m o
‘
, 1 501 4 1

g xp losi v s poun d s k D r ll s
“
o Te ~
M e y .

ll
h alf mor cco
*
D rin k e r s Tu n n e llin E Co m a nd Roc
’
l
, e , , i .

g ston s atalo gu of Min ls an d y ony ms

1
H '
1 3 l 4t o
, o , 25 00
E le
s xpl osiv s itroglyc rin an d Dy m t C S 8vo 2 501
1
e e ra
1
n ” . .
,

ood y ar s oal in s of W st rn o st mo
E isslér
’
E e — N e e na i e . 8vo , “
4 00

ock forming in ral mith

G C :M the C
’
12 2 50
‘
e
1

e e e a . . .
,

Manual of ining
H u ssa k R M
'

(S J
’
s - e s . . . . 8vo , 2 00

D r s ing in uro p
Ih lse n g s
’
M 1
8 vo , 4 00'
K
r tm nt of old O r s
rrn h a rd t s O re E l 8 vo
’
e s e . . .
,
‘
2 1 50

os n b usch s icrosco pic l Phy siograp h y Min ra s an d l

O D risc o ll s T G ‘I 8v o
’

2 00
’

ea e e .
,

ocks d d ings
R e M a lo f e l

wy s cci d n s Min s
R (I 1 8vo 5 00

tock b ri d g s ocks an d oi l s
Sa A 8 vo 7 00
’
er e t rn e ,

W lk s ctur s on xpl osiv s

S R S 8vo
’

e 2 50

it h ology Le E 8vo 4 00
’
a e e e .

W f
W i l s n s Min V nti l tion
L
1

illia m
’
ss . 8 vo , 3 00

yd rau l ic an d P l c r ining
l 6m 1 25
’
e e a 1. o,
o 1 .

H a e M 1 2111 0 .

S TE A M A N D E LE CTRIC A L E NG INE S , B O ILE RS , E t c .

STA TIO N A R Y M — A RIN E —O TIVE — G A S LO CO M EN G IN E S, E TC .

( See a lso E N G IN E E RIN G p

B l d w in s St am ating for ui d i gs
.
,

B l n
C l rk s G E ngin
He
’
a e .

F or d s B oil r M aking for B oi l r Mak rs

’

e as e

w y In d icator P ractic
’
e e e

H oa d l y s W arm b last F urnac

H
’
em e n a s e

P ractic an d Th ory o f h In j ctor

e
’
- e

M C d Sl i d V l v
K
’
n e ass s e e t e e

aw s Marin ngin s olio h al f mor cco

’
ac or s e a e
.

MP aybordsyMoandd rni l l ocomotiv onstructio

*
M
’
e E e . Z ’
. F o ,

r s t am B oi l rs
L C
’
e e e e n . 4t o ,

P a bo dy a l s of aturat d t am
e M e
’
S e e 8v c ,

h rmo d y namics o t am
e
’
8 T b e S e S e . l l 8vo ,

V al v ars or t am gin T e f t he S e 8 vo ,
’
50
0
P ay w nty Y a s w it h n d ic tor yal
e Ge f th e S e En e 2 50

P u p in an d O st b rg s h rmody am cs
T t he I Ro 8 vo , 2 50
’
r S e e r a
‘

agan t m an d l ctrical ocomotiv s

T n
’
i 1 2m o , 1 25
‘

er e e .

Re
s h mo dy namics B ois
5 S ea E L 00
1

1 2m o ,
’
e e . 1 .

inclair s comotiv unning

Rb n t ge n T e r ( Du
’
. .

h urston s B i l r x p l osion l l Lo R 3 I
’ ’
S e . .

T i ’
o e E .
 1 1

M IS C E L LA N E US PrJ B c O ' ‘

ATIO Ns

lcott s ms ntim nt anguag ilt dg s

.

B ai l y s a of a
A Ge Se
’

, e , L e G e e ,

Bal la d s ol ution of P y rami d P ro b l m

Th e N e w T le
’
e 8vo ,

Barna d s M t o l ogical y st m of r at Py r m d
S
’
r t he e 8vo ,

D vis s l m nts of
Th e G
’
r e r S e the e a i 8v o .

s ol ogical ui d b ook of ocky Mountains

E La w
’
a e e 4 . 1 8v o ,

rr l s r tis on W in d s
Em m o n R
’
Ge G e- t he . . 8v o ,

in s s dd r ss s D l iv r d ssn mo
Fe T
’

e ea e the 8 vo ,

ott s all acy of P s nt h ory o oun d mo

Ha Ry A
'

A
’
e e e e e e b e fo re t h e A m . . . . . 12 .

orn ll niv rsi y Ob l ong to

M Th e F
’
the re e T e f S . . Sq 1 6 . ,

ick tts s istory nss la r Polyt c h nic nstitut

Pe rk in s s C ’
e U e t 4 ,

iticall y m p h asi d
R e H o f Re e e
’
e I e 8vo ,

mo
Ro t lre rh a m s Th e N e w Te st a m e rrt l C r ’
. ,
E z e .

m p h asi d N st translation 12 ,

ar
Th e E w Te A w
g
z e e s ne .

m p ortant Qu stion in tro ogy L e 8vo ,

W h it h ous s ak o is Pa p r
To tt e n
’
s An I e Me l . 8 vo ,

W il y s l ask Y llo w ston M

’
e e L e er . . . e ,

e
’
Y o se m it e A ,
I
a, an d e e I4t o .
,

H E B RE W A N D C HA L D E E TE XT=B OOKs
CH OO H O OG C
.

F 01 1 S LS AND T E L I AL SE M IN A R IE S

b r w Ch a d l s m nt
.

G e se n iu s s H e
’

O
m ll h alf morocco
d e an d l ee L e x rc o n to
,
Te ta e .

r n s l m entary b r w rammar mo S a 4 10 , ,

rammar of b r w an guag (
G E He G
’
ee e e 12 ,

b r w h r stomat h y
G ) the He e L e N e w E d it io n . 8 vo ,

b r w B i bl ( Massor tic ot s ngl ish )

He e C e .

He
ara b squ e
L e t t e ris s N in E
’
e e e e .

s ramma B i b l ical h al d aic an uag an d

8vo e ,

L
al mu d a b li d i oms mo
G r Of th e C L th e
’
u z z a to g e

T B I 12 ,

M E D IC A L
B u l l s Matrnal nag m nt in al t h an d D is s mo
.

P h y s olo g ic l h mistry
’
e Ma e e He ea e . . 12 ,

ott s om p osition D ig sti b il ity an d utritiv

Ha m m
Valu of oo d
i C 8 vo ,
’

a rst e n s a e .

M
rg N
mou t d ch art
’
C , e , e e F .

ncom p a i b iliti s in P r sc i p tions

La e n e ,

t l s r atis on D is as s of O x
Ru dd im a n I 8v o
’
s t e e r ,

r is on D is s s of Dog
S ee
’
T e e the e e the 8v o ,

W oo dh u ll Mi l itar y mo
T ea t e the ea e t he 8 vo ,

W orc st ma l os p i als in anc

12
’
3

ta l
,

inclu d ing tkinson s u g stions for ospit l c h i

e er 3
’
S l H t — Es b lis rm en t an d Ma te n e,

t ctur mo
A
’
S g e H a Ar
e e 12 ,