Skip to main content

Full text of "The System of the Stars"

See other formats 


Google 



This is a digital copy of a book that was preserved for generations on hbrary shelves before it was carefully scanned by Google as part of a project 
to make the world's books discoverable online. 

It has survived long enough for the copyright to expire and the book to enter the public domain. A public domain book is one that was never subject 
to copyright or whose legal copyright term has expired. Whether a book is in the public domain may vary country to country. Public domain books 
are our gateways to the past, representing a wealth of history, culture and knowledge that's often difficult to discover. 

Marks, notations and other marginalia present in the original volume will appear in this file - a reminder of this book's long journey from the 
publisher to a library and finally to you. 

Usage guidelines 

Google is proud to partner with libraries to digitize public domain materials and make them widely accessible. Public domain books belong to the 
public and we are merely their custodians. Nevertheless, this work is expensive, so in order to keep providing this resource, we have taken steps to 
prevent abuse by commercial parties, including placing technical restrictions on automated querying. 

We also ask that you: 

+ Make non- commercial use of the files We designed Google Book Search for use by individuals, and we request that you use these files for 
personal, non-commercial puiposes. 

+ Refrain from automated querying Do not send automated queries of any sort to Google's system: If you are conducting research on machine 
translation, optical character recognition or other areas where access to a large amount of text is helpful, please contact us. We encourage the 
use of public domain materials for these purposes and may be able to help. 

+ Maintain attribution The Google "watermark" you see on each file is essential for informing people about this project and helping them find 
additional materials through Google Book Search. Please do not remove it. 

+ Keep it legal Whatever your use, remember that you are responsible for ensuring that what you are doing is legal. Do not assume that just 
because we believe a book is in the public domain for users in the United States, that the work is also in the public domain for users in other 
countries. Whether a book is still in copyright varies from country to country, and we can't offer guidance on whether any specific use of 
any specific book is allowed. Please do not assume that a book's appearance in Google Book Search means it can be used in any manner 
anywhere in the world. Copyright infringement liability can be quite severe. 

About Google Book Search 

Google's mission is to organize the world's information and to make it universally accessible and useful. Google Book Search helps readers 
discover the world's books while helping authors and publishers reach new audiences. You can search through the full text of this book on the web 



at http: //books .google . com/ 




I 



THE SYSTEM OF THE STARS 



rr — — *" ~^~T'l ^ 

11- ^ ' 



n 



/I 



— .- •--=- -- -:- '- 



u 



Plelone. 



AtlM. 



Sff 






+ 



• « 



• . 



i^it CWn ol xVft ^ 



THE SYSTEM 



OF 



IHE STARS 



BT 

AGNES M. CLERKE 

HOV. mxaBft or thx botal AaTB(utoiucu.i. aocinrr 

unam or ' histobt or utrmowmT dokdco tbv minnvxirrH cbvttkt ' Aim 

'noasjou or AtmopBTsica' 



SECOND EDITION 



LONDON 
ADAM & CHARLES BLACK 

SG-. 






'^ 






»5 



*Oj^ 






TO THK MSHOBT 07 

HT r ATHSB 

JOHN WILLIAM GLEBKE 

WHO DIED IN LONDON 
FlBBUABT 24, 1690 




FiPTMN years have elapsed since the original publication of 
the present work ; and ^teeii years count aa a long spell of 
time where sidereal research ia in queatioo* In preparing 
the Second Edition, accordingly. I have introduced extensive 
modifications. Considerable sectione of the book have been 
recaat, and all have been thoroughly revised. New chapters 
haye been inserted, old ones have been in large part suppressed. 
I^raetic meaBuree of reform have, in shorty been adopted, with 
reeolts that certainly Import progress and (it is hoped) con- 
stitate improvements. Most of the Illustrations are entirely 
new ; and I am under great obLigattonB for the use of valuable 
photographs and drawings, among others, to Sir David Gill, 
F-RS., to Professor Hale and the University Press of Chicago, 
[to the Rev. W. Sidgreavea, S.J., to Professors E. C. Pickering, 
Campbell, Barnard, and Frost, &nd to Br. Max Wolf of 





PREFACE TO THE FIRST EDITION 



SiDBREH. science has a great future before it. The proepects 
of its advance are inoalculable ; the poBaibilities of its develop- 
ment virtually mfinite. No other branch of knowledge 
attracts efforts for its promotion^ at once so wide-spreai3, so 
Tariedj and so enthusiastic j and in no other is anticipation 
BO continoallf outrun by the brilliant significance of the 
reaolts achieved. 

For the due appreciation, however, of these results, some 
preliminary knowledge is required, and 1^ pofiaessed by few. 
To bring it within the reach of ma&y is the object aimed at 

the publication of the present volume. Astronomy is 
tially a popular Bcicuce. The general public has an 
indefeasible right of access to its lofty halla^ which it is all 
the more important to keep cleared of unneceaaary technical 
impedimenta, that the natural teudency of all sciences is to 
become specialised as they advance. But literary treatment 
is the foe of speciaiiaationj and helps to secure, accordingly, 
the topics it is applied to, against being secluded from the 
interest and understanding of ordinarily educated men and 
women. Now, in the whole astonishing history of the human 
inteilectj there is no more astonishing chapter than that con- 
cerned with the sidereal researches of the last half century. 
Nor con the resources of thought be more eSectuaUy widened, 
or its principles be more surely ennobled through the vision 
of a Higher Wisdom, than by rendering it, so far as possible, 
iateUigible to all 



preiu 
■ Tob: 
^L^n t] 



X THE SYSTEM OF THE STAKS 

The following pages then embody an attempt to combine, 
in a geneml eurvej, some definite particulars of knowledge 
regarding our aidepeal surroundings. The plan pursued has 
been to instruct hy illustrative examples, to select typieal 
instancea from each claas of phenomena, dwelling upon them 
with Bufficient detail to awaken interest and aasist realisation, 
while avoiding the tediousness inseparable from exhaustive 
treatment. In developing the subject, it seemed best to 
proceed from the particular to the general ; to start with 
describing the physical constitution of individual bodies, and 
ascending by degrees through continually added complexities 
of mutual relationships, reaoh at last the crowning problem of 
the Construction of the Heavens. 

The writer gratefully acknowledges the assistance derived, 
in the preparation of the present work, from the kindness of 
Sir David Gill, H.M. Astronomer at the Cape, first and chiefly 
in affording her an opportunity of observing in southern skies ; 
secondly, in reading over several of its chapters in manuscript. 
Her thanks are also due to Professor E. S. Holden, to Messrs. 
Buniham, Keeler, and Barnard ; to Professor R C, Pickering, 
director of Harvard College Observatory ; to Sir William and 
lady Huggins, Sir Norman Lockyer, Drs, Vogel, Scbonfeld, 
and others for communications of great interest and value. 



LoKSOif, Sqitember 2g, 1A0O. 



^^^^^H^ ^ 


1 


r cm*r. 


PAQK ^1 


^^ 1. TuK Tas£ of Sideileal Astho^omt 


1 


^H S. Ths Methods or Sidzrigat. Rebeahch 


12 


^H 3. Tb£ StAftR AS Sgkb .... 


34 


^H 4, ThS CSEMtBTllT OF ReD StARS 


49 


^H & Gaseous Stars akv KebitIl^ 


58 


f ft. The Teufkhaturbs of tse Stabs 


71 


^K 7. Teufoiurv Stahs 


81 


^H 8. Stabs Va^abls in Lokq Periods and Irreoularlt 


97 


^H 9. Variablk Staha op Shobt E*eriod 


ue 


F 10, The Coloctm of the RTAitfl . . . . 


13& 


1 11. DotTBLE Stabs ..... 


160 


1 IS. SraiJiAR Obbits .... 


162 


^H 13i. Yariable Doube^ Stars .... 


176 


14. Sfbcteoscofio Binabies 


165 


lb. Multiple Stara ..... 


194 


16. The Etolotiojt of Mititiple Stars 


SOS 


17. The Pliiadbb ..... 


Sifi 


^^18. Star Cldstkkb ..... 


SS7 


^^■10. The FoRM.q or NxboLuS .... 


£42 


F 30, Tke Qseat Nebula .... 


267 


^^■21. Thk Nature and Chaxueh or Nebula . 


872 


^HiS. The Distances of the Stars 


SB2 


^^■£3. Tbanbi#atiok of the Solar Sybteu 


303 


^HS4. Tbb Pbdfbr Motioks op the Stars 


316 


^Bs5. T9B MruET Wat .... 


333 


^V £3. Status of trb Nrbui^ , . . . 


349 


^B SI. The Conbtructioh of the Heavens 


360 


^^^^^^H. 


1 




1. 

i "^ 

TIL 

1IV. 
VI. 
VIL 

H. 
X- 

XL 
XII. 

xux. 

XIV. 
H XV. 

XVL 




XX. 



Chart of th« Pleiadei 

Photograplied Stellar Spectra . 

Types of Stellar Spectra 

St«llar Spectra of Three Types 

Btell&r Spectm of the Second, Tbird, and Fourth 

Tjp^ in the Blue Kegioti . 
SputTograiu Qf Nova Aurtgi^ . 
Spectrum of Nova Aurigs 
Spectm of Temporary StarB , 
The Nehuloaitj round Nova Perael 
Spectrum of Alpha Phaeiiicia . 
IrreguLar Star CluEtera 
Fbotographe of Southern Sidereal Objecta 
PhDtogmpha of Spiral NehuliB 
Ad Irregular Kebula in Cygnue 
Photograph of the Looped, or " Spider " 

Nebula . . . , 

Photograph of M«siei 1 1 and adjacent Galactic 

CiDnd-formB 
photograph of the Key-Hole Nebula . 
Tba America Nebula 
Perforated Oalacltc Group iu SagittahuB 
The Cocoon. Nebula in Cyguus 



Franlupisee 


To/aae 


pa^t 36 




44 




46 




06 




00 




»l 




&4 




99 




136 




fisa 




238 




S60 


n 


S64 



ses 



sso 



S70 

33a 

30E 



ZIU 



^H^ LIST OF ILLUSTEATIONS 


1 


^^^P FAINTED EN XH£ TEXT 


H 


»ir 


Tua ^^H 


1, Bpectrom of Hydrogen in Wliite Stars (Huggina) 


3d I 


2. DistritmtioQ of 334 Pi^riods of VarUble Stan 


0d 


3. Two Consecutive Maxima of Mira Ceti 


100 


4. BiinimiUD of Miia Ceti, 1902-3 , . . . 


101 


^^m 5. Light-Curve of R Narmm . . . . . 


103 


^^B 6. Areas of FACola on the Stm, 1B78-99 


104 


^H 7. Light-Curve of XS Lupi (limea) 


10& 


^P 8. Light-Carve of Eta C«liiL», iaiO-1903 


loe 


i S. Oorve of Sun-spot Frequency, 1867-77 (EKiH) 


113 


^^-10. Distribution of 80 Variable Star Periods under 20''. 


117 


^^■11. Lighi-Ciirve of Bdta Cephei . . . . 


118 


^^■IS. Typical LtghtrCnrvee of Cluster VariAblea 


119 


^^ns. Light-Curve of No. 45 Omega Centauri 


120 


^^|l4. Light'CoTTe of 8 Ane , . . . 


ISO 


^^bs. LitjhvCarre of U Vulpecuhe 


12*2 


^^^ 16. Ligbl-Cur\ie of Beta Lyrae (Argelander) 


122 


^^17. Ligbt-Curv« of V Pttppie 


123 


^Kl8. TAgh|>CarTefi of Five Algol Stars 


1B4 


^V Id. Syttem of R^ Centauri (Roberta] 


125 


^^K to. Algol duricg an Eclipee 


1£9 


^^ftsi. MmiiQum of S Cancri .... 


131 


^^HS2. Apparent and Real Orbits of Alpha Centauri 


164 


23. Four Double Btara photographed at Patifi (llottche^, Phota 




graphii Aetrvnomiqw) . . , ■ 


166 


24. Orbite of the Component Stars of Gkunma Virginia . 


173 


26. Orbits of tbe Compotieut Stars of Alpha Ceutann 


174 


_ 28. Orbits of Siri ue and its compMiion . 


174 


^^■27. Stars of the Trapezium .... 


206 


^^H 


1 



CHAPTER I 



THE TASK OF SIDEEEAL ASTRONOMY 



When all the stars blaze out on a clear, raoonleBs night, it 
seams as if it would be imposBible to count them ; and yet it 
is seldoEQ that more than 2000 are visible together to the 
unaided eye. The nuiober, however, dependH very much 
upon climate and Bharpneas of sight. Argelauder enumerated 
at Bonn, where rather more than eight-teuths of the sphere 
come successively into view, 3256 stara." But of these no 
more than 2000 could be, at any one time, above the horizon, 
and BO many would not be perceptibly above it, owing to the 
quenching power of the air in its neighbourhood, Heis, with 
exceptionally keen sight, distinguished at Miinster 1445 
more stars than Argelander had seen at Bonn-, ^ Houzeaa 
recorded 5719 at Jamaica;* Gould 7756 within lOO*" of 
the south pole at Cordoba in South America.* The die- 
crepancies of these figures are due to the multitude of email 
stars always, it might be said, hovering on the verge of 
ibility. If, indeed, the atmosphere could l)e wholly with- 
iwn, fully 25,000 stars would, according to a trustworthy 

stimate, become apparent to moderately good eyes,^ 

Our system of designating the stars has come down to us 
Trom a hoar antiquity. It is a very emljarra&sing one. " The 

jostellations," Sir John Her&chel remarks," *' seem to have 

^ Dtanomttria AVma, 1649. 

* Hiiu, De MhffHiiitdim NuJtierofjue Stellarwn, p. 16, 1^52. 
» Uranuivtitrit OSrUrate, Anoalea de I'DUwrvatoire de BraselUB, t. i, 1678. 

* VratitowLctria Argentina^ 1879, 

* BtckhouAe* Jottr/iat Livtrpool Aair. SoeiA^, tqI. rii. ii. 33ti. 

* TrtiUUe on Jsironfrnj/, p* 168. ni4e. 

I 



THE SYSTEM OF THE STAES 



been almost purposely named aiid delineated to cause aa 
much confusion and inconvenience a» possible. Innumerable 
snakea twine through long and contorted iirEyis of the heavens 
where no memory can follow them ; bears, lions, and fishes, 
large and small, northern and southern, confuse all nomen- 
clature." And yet we could ill afford to dispense with the 
picturesque associatioue of a menagerie, largely stocked from 
the banks of the Euphrates. The signs of the Zodiac, which 
are undoubtedly of Chaldean origin, embody legendary cycles 
of thought already, some four thousand yeara ago, the worse 
for the wear and dilapidated by time. Homer and Hesiod 
were familiar with the Bear, Arcturue, and the Bog-star, with 
" the Hyades, and the Pleiades, aud the strength of Orion." 
The Little Bear was introduced from Phcenicia, when the 
Pole-star became the mariner's " cynosure." Finally, a number 
of individual stars have Arabic appellatiouB, dating from the 
epoch of Saracen supremacy over science. Thus " Vega " the 
current name of the brightest star in the Greek coasteUation 
of the Lyre, is the remnant of an Arabic phruse signifying 
the " Falling Eagle," while " Altair " stands for the " Flying 
Eagle ; " " Deneb " means the Tail of the Swan ; " Fomalhaut " 
the "Mouth of the Fish;" "Eigel" in Orion is the Leg, 
■' Betelgeux," the " Shoulder of the Giant " and so on. 

The constellations * now generally recognised are eighty- 
six in number, of which forty-eight are found in Ptolemy's 
" Almagest," From Ptolemy, too, is derived the method of 
classifying the stars by " magnitudes " This ia a most in- 
appropriate term, since none of the stars have any perceptible 
dimensions, Tht-y are literally what Shelley calls them, 
" atoms of intenaest light "- — ^globes shrunken by distance to 
the semblance of mere shining needle-points. Our own sun, 
removed to the place of the nearest fixed star, would be in 
the same condition ; contracted to xt^"^ ^^^ diameter would 
be utterly inappreciable with the largest telescope. It is 
true that the telescopic images of the stars appear to 
be of meaaiirable size ; but this is a purely optical effect, 
and the " spurious discs " shown by them actually grow 

' For an eaaj method of identifying the cti&f northern atars, aee Sir Robert 
Bill's Story of the HoavftiSt p, 37 'J ; »1bo the ' * Urauograpljy " itx Vouug't 
£iementt 0/ Astnmomi/, 1890. 



THE TASK OF SIDEREAL ASTRO>'OMV 



em^er instead of larger aa the power of the instrument 18 
increaeed, 

'• Magnitude " has, then, nothing to ili> with apparent ai^e, 
but refers entirely to apparent lustre, ^^bich depends upon 
distance and intensity of shining^ as well as upon actual 
{l]aien0xon£L The fainteet Btara have the highest numerical 
magniCudes; and it has been found that the gap between each 
mujoefiBive order^ as represented by the atars traditionnlly 
belonging to it, corresponds to a falling-off of light in the 
proportion of about 2^ to 1 . The arrangement by magnitudes 
js> of couTBe, entirely arbitrary ; natural gradations are not 
by a flight of steps, but along an inclined plane. Stars classed 
48 of the first magnitude (of which there are ten in each 
bezzkiephere) ^ differ accordingly very mudi among themBelvea. 
Sinus exceeds Eegulus no less than fourteen and a half times ; 
Vega is more than twice as brilliant as Aldebaran, Vega, 
CupelJa^ and Arcturus bold a co-ordinate primacy in the 
Borthem hemispheiv^but are outshone in the southern by Sirius, 
CanopU£, and a Centauri. Of second magnitude are the seven 
stars grouped to form " Charles's Wain," the Pole-star, and some 
of the ujOHt vivid gems ia Perseus, Cassiopeia, and .the Swan. 
Stars of the sixth magnitude are the faintest ordinarily visible 
to the naked eye; but those of the seventh can be seen under 
advantageous circumstances. The plan, introduced by Bayer 
in 1603, of naming the stare of the several constellations 
roughly in order of brightness by the letters of the Greek 
alphabet, established for each a kind of liglit-sefiuenoe, useful, 
though far from exact. The Buialler stars are usually dis- 
tisguished by the numbers attached to them in various 
GAbilogues. 

One of the most notorious circimi&tanceB about the stars 
is their "twinkling/' They undergo, especially when near the 
bomoQ, extremely rapid changes of lustre^ attended sometimes 
by the glinting of prismatic colours. Nor do all stand, in this 
reepect, on the same level White stars twinkle more than 
rud onea. Even early and untutored observers noticed how 

Tbe fieiy Sirius altera liue, 
And bickers uitc> red and emerald. 

*■ S«e Airpendis, T&ble I. 




4 THE SYSTEM OF THE STAES 

whence it was called by Aratus itokiXot-, the "many-coloured"; 
and chromatic unsteadiness was a marked peculiarity of the 
" new stars " of 1572 and 1604. 

It is easy to see that tbifi effect ia in Bome way due to the 
atmoaphere. Like Tefraction, it vanish^ at the zenith ; it 
varies in intensity with weather and climate. The first 
i&tional conjecture as to its cause was made in 1667 by 
Robert Hooke, who attributed it to iixegular refraction in the 
various air-strata. More exact Lnquiriea on the subject have, 
in recent times, led to some curious results. 

The impressions of light on the retina last, according to 
Plateau's c&ieful determinationj 0*34 — say one-third — of a 
second. This is the limit of their individual perceptibility. 
With more fi^uent recurrence, they become mei^d indis- 
tinguishably together. But the changes producing scintillation 
succeed each other much more rapidly than three times in a 
second Hence the need of some means of separating and 
(laalyeing them. 

These were provided by M, Montigny's " scintillometer," ^ 
in which the sensibility of different parts of the retina was 
skilfully turned to account for the registration of a swift 
aoocdasion of impr^sions. By the rotation of a glass-plate 
obliquely inserted in front of the eye-piece of a relracting 
tele*;ope, the image of a star viewed with it is made to 
describe an exact circle in the geld. The line of light traced 
out is, in perfectly steady air, continuous and of a uniform hue, 
but breaks up, under the influence of scintillation, into vividly 
tinted arcs, at times into prismatic " pearls." The addition of 
a pair of crossed wires facilitates the reckoning of the colour- 
fluctuations thus rendered separately visible; and they are 
found to occur, on an average, in white stars standing thirty 
degrees above the horizon, seventy-eight times in a second, in 
yellow and red stars similarly placed, sisty-eight and fifty-six 
times respectively.^ 

The explanation of these appearances is evidently to be 
sought in the refractive power, combined with the turbulence 



1 Described in £wU<iin de I'Aead. dt* Sciirux4, Brusefks, t. xvii. p. 261, 
2tid actha ; MoniAty Nttiiiet, to], xxxirii, p. 203 ; Ci<i ft Ttrrt (FievflE), t. i. 
p. 30&. 

^ BiiU. d* VAead. firuxellca, t Jtijcvii. p. 185. 2iid Khes. 



THE TASK OF SIDEREAL ASTRONOMY 



of our aedftl envelope. For a different path through its strata 
ia neoeasarily pursued by each of the differently refrangible 
beams united to form the itnage of a atar. The violet ent«ra 
them higher up, eiuce it is more bent in transit tlian the red ; 
and &o proportionately of the rest. Each then ia liable to 
^^mcouBter different vicissitudes on the way, betrayed to our 
^Bght by rapid flashes of colour. Each is affected by innumer^ 
^Blle small deviations and momentary capncea of refraction ; 
PHb that the bundle of rays picturing a star at a given instant 
jjt, as it were, a fortuitous and eminently unstable combination. 
It id dissolved, and a new one constituted, sixty or seventy 
titD€s in a second ; and the elements temporarily missing 
determine the reeultiug tint. The fundamental fact of the 
matter, in short, i8 that the light of every star near the 
horizon is drawn out into a tiny spectrum by the chromatic 
dispersion of the atmosphere ; and Respighi's study of the 
^actuations in these prismatic images ^ provided, accordingly, 
^^le ifirsct secure basis for a scientific theory of scintillation. 
^B That white twinkle more than red stats becomes intelligible 
^^■ben we consider that the sheaf of their beams being fuller, 
^^Kteroeptiona of them are more frequent But planets which 
are radiating discs, and not merely points, rarely show the 
effect, because the absence of rays from one part is compeasated 
by the arrival of rays from other parts of their surfaces. 
Similarly, the steady radiance of stars in large telescopes ia 
k^ie to the neutralisation of each casual stoppage by the 
P^^eat number of the beams collected together. Instead of a 
twinkling image, however, a blurred and distended one is 
formed under perturbed conditions, and observation gains 
nothing by the exchange. And since the degree to which this 
phenomenon is present varies very much with locality, regard 
should be had to its prevalence in choosing ^ sites for power- 
jil instruments. It diminishes, on the whole, with altitude. 
it the configuration of adjacent mountain-ranges is strongly 
uential, and Dr. Pemter found Sirius actually to scintillate 



^ Le* Monda, t lix. p. 6S8 {isa&) \ Lard Rayleigh, Phil M'ag. toL xxxvj. 
> 129, 1»03. 
' ExDCTf Attr, Xtteh, Tjlo. 37$! ; A. E. DouglaflB, Fopttlnr Aatronontt/, Jwae 
' ; I«oweU, Monthly Koti«s, Tol< Ixiii. p> 4D ; Eincr abd Villiger, Mtroph. 
Jowm. vqL xxL p. 36S. 



THE SYSTEM OF THE STABS 



more at the summit than at the foot of the Sounblick (10,C 
feet high).^ 

Scintiltatioti, like astronoooical refmction, augments as the 
thermometer falls and as the barometer rises. Thia is inevit- 
able, since the first requisite for its occurrence is differential 
refractiTe action on the various Ught-rays,* But it haa other 
leas obviously accountable meteorological relations. M. 
Montigny ascertained, by the experience of nearly forty years, 
that with the quantity of moisture in the air the twinkling 
of the stars increases so markedly as to serve for a useful 
pTOgnostic of rain. Cyclonic conditions promote it;^ and it 
is extremely sensitive to magnetic disturbances.* Useher was 
struck, in the eighteenth century, with the surprising vividness 
of scintillation during aurorse ; Montigny extended the coincid- 
ence to magnetic commotions perceptible only iuatrumen tally. 
Moreover, Weber remarked at Peckeloh in 1880, that stars 
aitoated near the magnetic meridian twinkled more than 
elsewhere in the sky;^ and although little attention has of 
late been paid to the possible dependence of the effect upon 
the points of the compass, yet the theoretic-al interest of 
scintillation would be much enhanced should it turn out to 
be one of the many terrestrial phenomena associated with 
vicisaitudes in the physical condition of the sun. 

The world of stars thrown open by the teleaoope may 
fairly be called boundless. Using a glass only two and a half 
inches across. Argelander registered 324,189 down to 9i 
magnitude, aU in the northern hemisphere, with the addition 
of a southern zone one degree wide. The work was extended 
to the southern tropic by Schonfeld, and completed to the 
southern pole under Sir David Gill's direction. At the Cape, 
the photographic method was employed, and the resulting 
enrolment, published 1896-1900, comprises 454,875 stars 
nearly to the tenth magnitude; while the Cordoba visual 
Dnrchmuaterung, executed by Thome and Tucker, 1835-1904, 
is a still more comprehensive register of southern objects. Yet 



* Obttrmtory, rol, lii, p. 1&4. 

* Montigny, Bull, de I'Acad, BrunBUeft, t. xlri. p. 813, 2t]d senea. 
' Roaentlial. Meicor. 2iitschr\/t, Bd. xi. p. 145. 

* CQmpltf R*nd%tMt t. icvi* p. 673- 

' lyoehtntchr^t/tlr Astrtmmut, 1360, p. 2&4. 



THE TASK OF SIDEREAL ASTKONOMY 



greftt works are merely preliminary to the International 
'chartJBg opCTationa in progress. Their accomplishment will 
yi«ld a " Catalogue of precision " including at least three 
million stars; aad somewhere about thirty millioas, taken 
from the submerged population of space^ will, by their record on 
the chart-plates, be admitted to the citizenship of astronomy. 
They can never thenceforward be excluded from the scope 
of research* Their light - changes, their movementB, their 
distribution, will present an inexhaustible, and probably a 
moot fruitM field of inquiry. 

Mr. Plommer showed in 1877^ that the lucid atare (those 
visible to the naked eye) in the Bonn Durchmusterung give 
as mach light as 7349, the telescopic stars as 23,337 sixth 
magnitude atara Those singly imperceptible thus really 
illuminate the sky just three lirues more than those in- 
dividually seen. Stamaing up, with the aid of the beet 
photometric data, the entire light of Argelander's 324,000 
atara, we get for its equivalent j^ full moonlight ; and we 
y roughly estimate the total light of all those similarly 
inumerated in both hemispheres, to the number of about 
900,000^ at -pj^ the lunar brightness. The amount of 
scattered effulgence dispensed by still fainter stars is exceed- 
ingly difficult to evaluate. Sir William Abney, using a 
photographic method, rated in 1896 the sum of starlight in 
both hemispheres at ^iv ^^ moonlight. Professor Newcomb, 
in 1901,^ from visual observations of diffused sky-radiance, 
condaded the light-power of all the stars to be just 723 timea 
that of CapeUa;* and 728 stars like Capella give ^ the 
light of the full moon.* But it is far from certain that the 
I T anlt of heaven would seem absolutely black if the stars were 
^^lotted out.^ Our tipper air is the seat of processes by which 
^^■iniinosity is at times strongly developed ; and we cannot be 
^^pore that they are ever entii'ely suspended. Hence, Frofe&aor 
[ Kewcomb's experiments ajford no assurance regarding the 

^^H 1 Monthly J^fotias, vol, xxxvii. p, 43(3. 

^V ^ Astrophytioil Jtmrnal, vol, xiv. p. 310; see iho OaTla J. Buras, ibid^ 
ToL iri. p. 169. 

' Newomab'B 600 stars a{ O'O fflftgnttude are «qatl to 723 of th« pbQtoniBtriQ 
bright&6» (0^1] 4f CApeUft. 

* VmtT, flictomarit dtr Qt^TM, p. 340. 

■ Cf. Bttrn», Jov.rn. BrU. Aatr, Au. toI. iv. p. 91. 



8 



THE 8TSTEM OF THE STABS 



iiudividi^B 



eqoiTaleat, to mf *****^f d Urn mitml BiuiLct of itaa 
oomuig dinctfy, or iadtnetly wHfata o«r fcea. Thaae last caa- 
QOt well &U ifaat of ntj ™ittwT«_ aad tbej m«« profaablj 
mun 1^ to OBie hazidred aillioBo; bat tbe aet* of iDqaiij can 
at prueeDi scmroefy be djswn ckser Yet tbe foot io Doto- 
worth^ that each clan of stsn sendi itt ^ppncnl^ more 
light tban tbe dam next sbofv iL Tbe I^it-Aggx«g«ta of 
BBOond magmtode ttaxs exeeeds that 'of ink, of third that of 
aeooud, and so oil Tbe ftinter tbe aUzs, in sbortj tbe gieatcr 
is their total luminooe power/ >iw»iiiift their 
uumbers more than «yMin t .f*' t;Miti^"fg tb^r diniiiiiabed i 
lufiti^ But this pfogreanao, it is eridetit^ caanoi go on 
iodefiuitelj, EJoce otbenrise an indtifinitelj inteoae radiaaca 
would fill tbe sky. Daricnen wonld be abolished through tbe 
fihiaing of invisible stara. It £ollo«a other Uiat tbe obeerred 
order of the etellar vorid baa aaagnabk limits — that the alar- 
depths, however proloand, are not abaolotelj tm&tbomable ; 
or that space, for whaterer reasoo, is not abeolately tiaaspareat 

The Laak of exploration, at anj rate, does not aeem to be 
altogether hopeless. It can neTer indeed be exhausted; bat 
it can fairly be grappled with by finite minds. It does aofc 
evade their efibrte with the paasve ecom of material infini- 
tude. Geauine. if partial, saooeases have otiwned them in the 
pastf and will, it may be hoped, continne to ciowu ihem in the 
future. 

We must not. however, in seeking enconittgement from 
the thought that it does not utterly defy oor powera^ under- 
rate the difficulty of the enterprise we have taken in hand. 
Tbe nature of our own sun oflers a vast and intricate problem, 
still very far from being solved ; but stellar epoce contains 
many millions of suns, variously couEtituted, varioualy circum- 
atanced, frequently snrpaesing our magmficent orb in size and 
splendour. Now each of these millions of suns challenges 
the closest pergonal attention ; no single one of them is 
exactly like any other, and their diffcrenoee and resemblances 
open endleaa vistas of instruction and interest. Their incon- 
ceivable remoteDese in no way derogates from their real 
dignity. An all but evaaescent speck of light in the field of 
the great Lick refractor may bo the life-giving centre of a 
* L*AttTonoini*t t. v, p. 400* 



THE TASK OF SIDEREAL ASTROXOMY 



of worldB, each aboanding as manrelloudj vitb proofs 
of creaiiwe wisdom and goodness aa tbe little planet in which 
our tempoml deetuues &re imprisoned. Such ligbtr-specks are 
Aea equally deeerring of stody with tbe moet effulgent orb« 
in the ekj, although it may never be practicable to bestow it 
apoD them. We can indeed hardly imagine the amount of 
teleBOppic improrement which would be needed in order to 
hnng them within the range of critical examination. For 
the preeent, accordingly, physical research must be confined 
to eome thonstuids of the brighter etare which may serve as 
Bpecuneiu of the rest. Nor need we lament the restriction. 
GenerAtioiiS of workers migbt expend their energies in 
^tbering facta from the field actually open to them, and 
yet leave a fiill harvest for their successors. In all ei- 
perimentAl inquiiies, it may with truth be said that the 
reaper, aa be gamers one crop of knowledge, sows another : 
Bo endless are the secrets of nature, ao untiring the inqniai- 
ivenese of man. 

The »tars in their combinations demand inquiry no less 
■n the stare in themselves. Stellar systems are to be met 
with in indescribable profusion and variety^ from mutually 
drcUng pairs, through groupa including thousands of phyei- 
caUy related objects, to tbe stupendous integrated collection 
which we call the Milky Way. But as yet investigation has 
barely aldrted the edge of this well-nigh infinite region. Before 
■^ can be penetrated by so much as a plausible conjecturej 
BMtati»tics are wanted of the distances and movements of 
r thousands, nay millions of stars. 

I Nor is the amassing of them any longer the Sisyphean 

I labour it seemed a short time ainca By the unhoped-for 
development of novel methods, the pace of inquiry has been 
qoickened all along the line. Particulars are accumulated 
faster than they can be assorted and arranged. Time has 
virtually expanded, as if for the purpose of gratifying curiosity 
^^phich becomes keener a3 its sublime objects loom more 
^Histtnctly above the horizon of thought Ten years now 
connt for a century of the old plodding advance. Express 
trains carry passengers on erraada of research, aa well as of 
btmnesa or pleasure. Problems ripen as if in a forcing- bousej 
and 80 numerously as almost to bewilder tbe attention. 




THE SYSTEM OF THE STAPS 

The whole sabjecc af ddereal natural historx ia 
and intricat« beyond wbAt il is eAsy to coarej to the 
approaching it for the fitBt time. TheT« ia ac»rcely a topic iti 
physical aHtronomy with which it is unconcected. The pro- 
gress of discovery has giadtially drawn ctoaer the geuenc 
relationships of the hearenly bodies The mm has oome to be 
recognised ns the grand exemplar of the Atars ; meteoritea 
show themseWes to be intimately associated with comets ; 
comets are perhaps the shreds of uaed-np nebnlfe ', while the 
stellar and nebular realms blend one with the other aa indift' 
tinguishably as the animal and vegetable kingdomB of organic 
nature. 

The strange cloud-like objects called "^ nebulie " may be 
considered as wholly of telescopic revelation. Only one of 
them — tlie famous object in the girdle of Andromeda — can 
be at all easily seen with the naked eye ; and even that 
escaped the notice of aU the Greek, and most of the 
mediseval astronomers. The " nebuloase " of the ancients 
were many of them small groups of stars accidentally set 
close together ; but among the seven enumerated by Ttoiemj 
were two real clusters like the Pleiades, only (presumably) 
much farther away, one in Perseus, and the other in Cancer. 
Indeed, to extremely ehort-aighted persons, the Pleiades them- 
selves put on a nebulous appearance, the individual ertars 
nmning together into one wide blot of light 

Halley was the first to form anything like an adequate 
conception of the importance of nebular observations. He 
was acquainted in 1716 with six -'luminous spots or patches, 
which discover themselves only by the telescope, and appear 
to the naked eye like small fixed stars ; but in reality are 
nothing else but the light coming from an extraordinary great 
space in the ether, through which a lucid medium is diffused, 
that shines with its own proper lustre." ^ Only two of Halley's 
half-dozen objects, however — those in Orion and Andromeda — 
were genuine nebnke ; the rest when viewed with better instru- 
ments than hia " six-foot tube," proved to be magnificent star 
clusters. 

This small beginning of knowledge was followed up by 
Lacaille in the southern, by Messier in the northern hemi- 
^ Phil. Trafit, vol, izLi. p. 3»0. 



THE TASK OF SIDEBEAL ASTRONOMY 



11 



ilibere. Then Hencbers great tdescopes opened the modem 
epoch in the sciesce of nebole. A» the result of his labours 
thef came to be reckoned by the tbooeaud instead of by the 
Kore. Portioofi of the Eky were foand to be crowded with 
them. Yet the vaat tnajoritj must always, owing to their 
^ctzeme faintness, remain imperceptible without powerful 
opCiical Aid, only sixty-four coming into view with the same 
tfllaoeope which showed Ar^lander 324,000 atara. Thua 
Aeoees to the nebular heavens can be gained only by making 
the yerj most of the little light they send iia> I^u-ge 
teleeoopes and prolonged photographic exposures are indeed 
pie-eminently uBeful in this department of celestial phyBic8» 
which, mainly through the application of the camera, has of 
late incalculably widened A pioneering survey with the 
Crosaley reflector led Profeseor Keeler to estimate at 120,000 
the number of nebulae which might be chemically recorded by 
it8 meftns ; and the specimen sheaves garnered by Dr. Max 
Wolf at Eomg;3tuhl in 1901 are no lees promising for a rich 
future harvest. But discoveries are of email account if 
Bucceeded by neglect. Assiduous and prolonged observation is 
indispensable for the detection of the cyclical or progressive 
changes doubtleas proceeding in these inchoate systems. 

This then is the task of sidereal astronomy — to investigate 
the nature^ origin, and relationships of 30,000,000 stars and 
of 120,000 nebalse — to inquire into their movements among 
themselves, and that of our sun among them — to assign to 
each ita place and rank in the universal order, and, gathering 
hinU) of what ■ has been and what will be from what is, 
distinguish hierarchies of cele.^^tial systems, aud thus at last 
rise to the higher eynth^is embracing the grand mechaniBm 
of the entire — the sublime idea of Omnipoteuce, to which the 
stars conform their courses, while " they shine forth with joy 
Him that made them." 



CHAPTER II 

TEDS HXTHODS OF SIDEREAL BXSEAItCH 

Sidereal science is, on its geometirical side, of modem dfiTekip- 
ment ; on its physical side, of modem origin. The places of 
the stars, as referred to certain lines and points on the sar&ce 
of an imaginary hollow sphere, are obtained now on eaaentiiiUy 
the same principles as by Hipparchus, only with incompazaUy 
greater refinement And refinement is everything where the 
stars are concerned. Significant changes among them can 
only be brought out by minute accuracy. To a rough dis- 
cemment their relative situations are immutable; and 
systematic inquiries into their movements hence became 
possible only when the grosser errors were banished &om 
observation. Bessel's discovery of Bradley's exactitode gave 
the signal for such inquiries. It seemed worth while to re- 
observe stars already so well determined that discrepancies 
might safely be interpreted to mean real change. 

Thus it is only within the last hundred years that the 
stars have been extensively catalogued for their own sakes 
and no longer in subordination to the interests of planetary or 
cometary astronomy. The scope of such labours now widens 
continually. For the objects of them are all but innomerable, 
and the inception of ambitious schemes is encouraged by 
modem facilities for executing them by combination. The 
project set on foot by the German Astronomical Society in 
1S65 of fixing the precise places of stars to the ninth 
m^nitude. found co-operators in all parts of the world ; and 
it« virtual completion in 1903 to the verge of the southern 
zones obseiTod by Qould and Gilliss* raised the number of 
stars not mer>L^ly rooordoil. but known in the strict astro- 
metrical sense, to not far from 400,000. 

19 



THE METHODI 



IDEREAl RESEARt 



A star is located in the LeaTens, just as a city or a 
oimtain ie located on the earth, by meaaurements along 
two imaginary circlea. Its " declination," or diBtance from 
the celestial equator, corraaponds to terreatrial latitude ; its 
'■ right ascension " to terreBtrial longitude. The astronomical 
prime meridian passes through the first point of Aries» that 
is, the sun's position at the vernal equinox ; intervab from it 

reckoned eastward &om 0"^ to 360"*, or in time from 0*^ to 
And since the zero-point retreats slowly westward by 
the effect of precession, it follows that the right ascensions 
of moat stars increase steadily year by year, apart from any 
moTementd " proper " to themselves. 

The diurnal revolution of the sphere furnishes the sole 
standard of time in sidereal astronomy. Sidereal noon at a 
given locality is the moment when the first point of Aries 
crosses the meridian of that spot; the right ascensions of 
the heavenly bodies indicating the order of their successive 
culminations. Thus, if the right ascension of a star be two 
hours and twelve minutes, it ^ill cross the meridian of any 
place on the earth two hours and twelve minutes after the 
first point of Aries has crossed it, coming up behind it, to that 
extent, in the grand diurnal procession. Differences in right 
aaoeuBion signify differences in times of culmination ; and 
tbeii measures in hours, minutes, and seconds need only multi- 
plication by fifteen (the number of times that 24 is contained 
in 360), to appear as measures of arc in degrees, minutes, 
and ^seconds. 

A transit-ciixle and a clock are the two essential inatru- 
tnenta for ascertaining the places of the stars. The instant, 
to the tenth of a second, at which a star stands in the 
meridian, is noted ; the vertical circle is read, showing its 
"aenith -distance" (giving at once its dechnatiou when the 
latitude of the observatory is known)> and the observational 
part of the work is done. The data thus obtained, after 
idergoing numerous corrections, aufi&ce to determine the 
ition of the star with reference to some other " funda- 
mental " star, the ahsobde place of which has been separately 
and laboriously ascertained. 

This business of star-location forma the substraturo of the 
older astronomy. But the precision given to it ia altogether 




THE SYSTEM OF THE STABS 



new, Bod alone has fitted it to be the meana of eliciting facts 
80 coy aa those that relate to stellar niovementa. For their 
diacloaure devices of accuracy are needed which our astro- 
nomical progenitors never caat a thought upon. Optical and 
mechanical skill has, in our days, reached a point of almost 
ideal perfection ; yet when the artist has done his utmost, the 
instrument is» in a sense, still in the rough. The astronomer 
then takes it in hand, and his part is often the more arduous 
aad anxious. The investigation of small surviving errors, the 
contrivance of methods for neutralising their effects, the carry- 
ing out of delicate operations of adjustment* the detection of 
microscopic detbrmatioD3, tremors of the soil, inequalities of 
expansion hy heat, fall to his share. Even his own rate 
of senBe-transmiseion has to be measured ^ and figures, under 
the title of " personal equation," as a correction in the final 
result. For, between the actual occurrence and the perception 
of a phenomenon, there is always a gap, more or less wide, 
according to individual idiosjoicrasy, and it is only after this 
gap — tiny though it be — has been crossed, that electricity can 
he called upon to play its prompt part as amanuensis to the 
observer. 

This detailed and painful struggle against eiTor has 
made sidereal aatronomy possible, by pr€cipitatinff from the 
mixed solution that held them the minute quantities it deals 
with. Just because the universe is almost infinitely large, 
these quantities are almost infinitely small They are small, 
not in themselves, but through the incomprehensible remote- 
ness of the bodies tliey affect. 

Sidereal astx-onomy is deeply concerned with the motions 
of the stars. These are of different kinds. " Proper " motions 
— so-called to distinguish them from " common " apparent 
displacements due to the slow shifting of the points of refer- 
ence on the sphere — advance uniformly along a great cirele ; 
orbital revolutions of one star round another are periodical in 
small ellipses; besides which annual oscillations, varying in 
extent with the distances from ourselves of the objects per- 
forming them, are barely measurable in a few of the nearest 
stars. The perception and characterisation of these orbital 
and " parallactic *' movements have become possible only 
through the attainment of exquisite observational accuracy. 



J 



THE METHODS OF SIDEKEAL RESKAECH 15 



rbe JDstrunients employed are the equatoieal with micrometor 
Attached^ and tbe heliometer. 

An equdtoreal la a telescope so mounted as to follow the 
diurnal Eevoltktion of the heavens. It i^ coimected with an 
ftXJB directed towards the pole, and revolving by clockwork 
oooe ID twenty-four hours. An object accordingly once brought 
into its field of view remains there immovably for any desired 
time, provided the tube be clamped in position^ and the clock 
3et going. The inconvenience of the earth's rotation in pro- 
ducing a continual ** march-past " of the heavenly bodies id 
thus neutraliged. 

To the eye-end of an eqnatoreal is uenally attached mi 
arrangement of spider lines constituting a " filar micrometer." 
Two sets of Buoh threads (which in subtlety and evennesa of 
texture far surpass any artificial product), crossing at right 
angles and some of them movable by fine screws, while the 
whole can be made to revolve together^ afford a most delicate 
means of ascertaining the distance and direction from each 
other of any two objects close enough for simulumeous 
obeervation. Mea-stixes of double stars are executed^ and some 
stellar parallaxes have been determined in this way. But for 
the latter purpose, the " beliometer " is the more appropriate 
instrument- 
Its designation is a misnomer, or rather represents the 
tradition of an original purpose to which it was never effec- 
tively applied. The true function of a hebometer is tbe 
critical measurement of two adjacent stars, or of a star and 
planet. Primarily, it is an equatoreal telescope ; its micro- 
gtrical powers are conferred by the division of the objetjt-glass 
Btu two halves sliding along their common diameter, and 
duplicating by their separation the combined image formed by 
them when together. The amount of movement igiven to the 
segments in bringing about alternate coincidences between 
opposite members of the pair of atars shown by each, suffices 
to determine with the utmost nicety the interval between them. 
That is to say, after endless precautions for accuracy have 
been taken, and endless care bestowed upon detecting and 
obviating occasions of infinitesimal error. 

The Badeliffe Observatory at Oxford possesses the largest 
heliometer in existence* Tbe diameter of its object-glass ia 



16 



THE SYSTEM OF THE STAES 



Beren and a half inches. A siiuilar idstrument, however, 
erected at the Eojal Obaervatory, Cape of Good Hope, nearly 
fortj years later, ia but slightly inferior in size, and is in other 
respects considerably its superior, Dr, Elkin, at Yale College, 
haa charge of the only hetiometer in the Kew World, while a 
good many are to he found in Germany and Huseia. The 
Kepsolds of Hamburg may he eaid to bold a monopoly in the 
mechanical part of their production ; and Merz of Munich 
stands almoet alone among opticians in bi3 readiness to take 
the responsibility of sawing a fine object-glass in two. Nor 
is the aptitude for the use of these instruxnenta by any 
ni^ns uniyersal among observers ; hence theii' comparative 
scarcity 

The science of the motions of the stars is only a part of 
modem sidereal astronomy. Within the last forty years a 
science of their nature has in defiance of forecast sprung up, and 
assumed surprising proportions. Sidereal physics has a great 
future in store fot it. Its e^cpausiveneas in all directions ie 
positively bewildering. The " What next ? " is hardly asked, 
when it is answered, and often in the least looked-for manner. 
In following its progress, the mind l>ecome5 so inured to 
novelties^ that antecedent improbability ceases to suggest 
dissent. Some details of what we have thus so far learnt 
will be contained in the ensuing chapters ; the means employed 
must be briefly indicated in this. 

They are of three principal kinds — spectroscopic, photo- 
metric, and photograpliic. The general theory of spectrum 
analysis has been explained elsewhere ; ^ here we need only 
repeat that it rests upon the constancy of the positions in the 
spectrum belonging to the taya of light given out by ignited 
vapours. These invariable lines serve as an index to the 
presence iu the sun or in a star, no less than in the laboratory, 
of the substance they are associated with. Whether they be 
bright or dark, the principle remains the same. They are 
bright when the vapour originating them is the chief source of 
illumination ■, dark, when a stronger light coming (rom behind 
ia absorbed by its interposition. Their appearance as " lines " 
is merely due to the transmission through a narrow slit of the 
light afterwards priamatically dispersed. 

^ 8«e tlie AQthor's Hittory t>/ AxlrOTiomy. 4th edit. p. 189. 



^k 



THE METHODS OF SIDEREAL RESEARCH 17 



Ifow a main difficulty in getting starlight to disclose ita 
secreta^, is that there is so little of it. It will not bear the 
Dficessarj amount of spreading -out, but evades analysis by 
&dinig into impei-ceptibility, like a runnel of water that widens 
only to disappear* Hence the absolute neoesaity in stellar 
spectroscopy for large telescopes. The collecting aeta have to 
be widely extended to gather in a commodity so scarce. Could 
we at all reaUse, indeed, the portentous expanse of the evet- 
broade&ing sphere filled by the stellar beams as they travel 

_towardB vm, we should be inclined to wonder, not at their 
itneaa, hat at their intensity. Bat the weakening effect of 

^diatattce is in some degree counteracted by powerful concentra- 
tion ; and this is one of the chief uses of the large telescopic 
apertureB bo much in vogue at the present time. 

Viewed with the lick refractor of 36 inches, any given 
Btar 18 32,400 times brighter than it appears to the naked 
eye, or 324 times brighter than when shown by a 2-inch 
telescope.* The large inBtrumentj that is to say, provide* 
324 times more material for experimenting upon, or raages 
further by 6^ stellar magnitudes than the small refractor. 
The interpretation of spectral hieroglyphics, by which we 

Jearn the chemical coostitution of a star, is a very delicate and 
iborious operatioa What is called a " comparison-specDrum " 
usually employed a*? an adjunct to it. Rays from some 
terreetrial source are reflected into one half of the sUt, through 
tht! other half of which the stellar rays are admitted. Both sets 
then traverse the same prisms, and form strictly comparable 
Spectra side by aide in the same field of view. Lines common 
to both can thus easily be identified ; and their genuine occiir- 
recce leaves no doubt that the element compared^hydrogeu, 
sodium, iron, magneaium, or any other^ — enters into the com- 
position of the star. But this process of matching can st^Mom 
or never be completely carried out. A dozen known lines 
may be attended by a hundred unknown onee, either too faint 
to be distinctly seen^ or in positions unfamiliar to terrestri»l 
light -chemistry. Nor is it safe to infer the absence of an 
ingredient from the absence of its representative rays. Many 
caufies contribute to render the display of lines in sbolkr 
ctra selective. 



18 



THE SYSTEM OF THE STARS 



Where direct comparisons can he dispensed with, & slit is 
not easential to stellar light -analysis. For a star, having no 
Benfiible dimensions, gives riee to none of the confused over- 
laying of images produced by gioasor light-sources imJess super- 
fiuoua rays be excluded by the iise of a fine linear aperture* 
Hence the poasibility of applying a " slitleaa spectroscope " to 
the stars. Their light is then simply passed through a prism, 
either before it enters, or as it leaves the telescope. The 
resulting variegated stripe, looked at through a cylindrical 
leas to give it some tangible breadth, shows the dark gaps or 
lines significant of the " type " of the atar, 

But prismatic analysis is not merely communicative as to 
the physical and chemical nature of the stars. It can tell 
something of their movements as well. And, what is especially 
fortunate, the informRtion that it gives is of a kind otherwise 
inacce^ible. ''End •on" motions, as every one knowa^ axe 
visually imperceptible ; the discovery that the spectroscope 
has the power to make them sensible ie of such far-reaching 
importance that Sir William Huggins, by bringing the method 
into effective operation, performed perhaps the greatest of his 
many services to science. Through the link thus established^ 
geometrical and physical astronomy have been placed in 
closer mutual relations than cotxld have been thought possible 
beforehand. 

The observiitions concerned are of great delicacy^ and can 
only he made with a powerful telescope* collecting light 
sufficient to bear a considerable amount of dispersion. Their 
object is to measure the minute displacements of known lines 
due to " radial " or end - on motion, and proportional in 
amount to its velocity. These displacements are towards the 
blue end of the spectmm when the star is approaching, towards 
the red when it is receding from the earth. The refrangibility 
of the luminous beams is changed, in the one case, by the 
crowding together of the ethereal vibrations, rendering them 
more numerous in a given time, in the other, by their being 
(as it were) drawn apart, and so rendered less numerous. 
The juxtaposition of a standard terrestrial spectrum, such as 
that of iron, gives the means of measuring deviations thus 
produced, and so of determining the rate of approach or reces- 
sion of the star examined. But the process is impeded to a 



THE METHODS OF SIDEKEAL RESEAECH 19 



degree hardly imaginable without personal experience by 
tioahles in the ocean of our air. The twinkling of the stara 
i& represented in their spectra by tremors and undulations 
oitan permitting only instantaneouB eatimates of line-pogitious. 
This inconvenience has been largely remedied by the use of 
^e camera. 

Stellar photometry has a twofold object It gives the 
means of investigating, first, the individual nature ; secondly, 
the collective relations of the stars. Stellar lustre is affected 
by endless gradations of change. It ia rarely, perhaps never, 
really constont. Periodical fluctuations are in many cases 
obvioua ; secular variations are suspected. The suspicion can 
be verified only by precise light-measurementa repeated at 
Jong interv'als. 

Their apphcation to the problems of stellar distribution 
becomes feasible through the dependence of brightness upon 
distance. The law of the decrease of light with the iwretue 
of the square of the distance is universally familiar. If all 
the stars were equal in themselves, their appai-ent differences 
would thus at once disclose their relative remoteness. We 
could locate them in spa^e just as accurately as we could 
determine their lustre. But in point of fact the stars are 
vastly diversified in siae and luminosity, and we can hence 
reason from distance to brightness only by wide averages. 
A statistical method alone is available, and its employment 
evolves the establishment of strict principles of light-measure^ 
lent. 

The firet requisite for this purpose was an unvarying and 
consistent scale, which was provided with the least possible 
diatorbance to existing habite of thought by regularising the 
antique mode of estimation by " ms^nitudee," Intervals 
loosely defined and unequal were made precise. A " light 
ratio *' was agreed upon. To this proportion of change from 
one magnitude to the next, the numerical value 2*512^ hwa 
been assigned. That is to say, an average first magnitude 
star sends ue 2-512 times as much light as an average star of 
tho second magnitude, which, in its turn, ia 2'512 times 
brighter than one of the third, and so on. From the first to 
ie third magnitude, the step is evidently measured by the 
' fiekotMl oM tbe uiuuber of which 0*4 ia the loguithm. 



20 



THE SYSTEM OF THE STAES 



square <if th« " liffht raiio " (2-5l2x 2-512"6-310); and in 
general the relative briiliaucy of any two stars may be found 
by raising 2'512 to a powyr represented by the numerit^'al 
diffei^nce of their magnitudes. One first magnitude star, for 

instance, is equivalent to one hundred of the sixth tank 
((2"512)^— 100) ; and to no less than a million atara of the 
sixteenth magnitude. 

All this IB a matter of pure definition, and definition is a 
useful leading-string to experiment. It is something to have 
a clear conception in the abstract of what a tenth, eleventh, 
twentieth magnitude star is, even though the conception be 
not altogether easy to realise. The problem of applying the 
aumerical standard set up waa practically solved ahnoat at 
the mme time by Professor Pritchard at Oxford, and by 
Professor Pickering at Cambridge in the United States. They 
first systematically and extensively employed instnunental 
means in stellar photometry, with the result of satisfactorily 
ascertaining the comparative lustre of till stars visible to the 
naked eye in these latitudes. 

Professor Pritchard adopted for his researches the ^' method 
of extinctions." The image of each star wag made to vanish 
by sliding between it and the eye a wedge of neutral-tinted 
glass, of which the thickness just needed to produce invisibility, 
waa found to give a very eiact measure of intensity. In this 
way tlie brightueasof 2784 stars from the pole to teu degrees 
south of the equator was determined and registered in the 
" Uranometria Nova OionieuBis." 

The Harvard " meridian-photometer " waa constructed on 
the principle of " equalisation/' The images of the pole-star 
(adopted as a standard of comparison) and of each star succes- 
sively experimented upon were reflected into a fixed telescope, 
and brought to an exact equality by means of a polarising 
apparatus. From the amount of rotation given, for this pur- 
pose^ to the double refracting prism, the actual difference of 
brightness was easily deduced. The method is of wider appU- 
cability than that by extinctions ; none the less, the '' wedge 
photometer," in the form given to it by Pritchard, has taken 
its place aa an indispensable adjunct to such inquiries. With 
either instrument the limit of clearly distinguishable diflerence 
is about one'tentb of a magnitude^ 



THE METHODS OF SIDEREAL RESEAl^CH 21 



The original Harvard photometry' included all atanf to 
the sixth maguitude as far as 30" of south latitude, to the 
number of 4260. But it was only the first in a series of 
similar and larger works. Its extension over the entire 
heavens was aceompliahed by the publication in 1895 of the 
San^ieTn Harvard Photometry of 792U stars ;^ and the 
Revised Harvard Photometry, embracing about 9000 stars in 
both hemifipheres, wae approaching completion in 1903>' A 
Photometric Durchmaatemng of stars to 7'5 magnitude within 
130"* of the North Pole was besides constructed with a larger 
iustmment,* while the measurement of thousands of stars in 
zones has established standards of exact comparison down to 
the ninth stellar magnitude. 

Potsdam is also the scene of extensive photometric opera- 
kions, which have now virtually reached their immediate term. 

Lbout 16,000 stars have there been very precisely observed 
by MM. Miiller and Kempf, with a polarising photometer 
employed on 26llner's plan of comparison with an " artificial 
Vtar." The results thus by various observers variously obtained 
1 in general satisfactorily accordant ; although the insecurity 
Itiending processes of correction makes some degree of diverg- 
ence inevitable. A Photometric Catalogue gives the bright- 
ness of stars in all parts of the aky " reduced to the zenith." 
But the reduction is not by a simple or Certain procedui'e. A 
law of light-absorption in the terrestrial atmosphere has first 
to be arrived at experimentally, and the esperimente are 
difficult and delicate. Even at the zenith a heavy duty has 
to be paid, estimated by the Potsdam obsei-vers at 16 or 17, by 
Frofeaeor Pickering at 20 per cent ; and the rate of increase 
towards the horizon differs at different allitudea, and probably 
in diSereut climates as well. Hence the adoption of a uniform 
plan of reduction for photometric observations seems to be 

^recladed ; yet its absence must involve more or leas serious 

ancies. They are, however, possible only when the 

ige of conditions is wide ; where they are fairly constant, 

treme accuracy is attainable. 

1 Harvard Annals, toI. xiv. pt. i. (IS&l). For i coupariBon with the 
)zfcrnl rrffuUs we ibid, voh xlii. |k ICi 

* Harvard ^ntuila, vol. rxiiv. 
« Jhul. vol L 

* Jbiii. ^-ol. x!v. 1901. 



22 



THE SYSTEM OF THE STABS 



In the Keviaed Harvarti Photometrjir two stars, Aldebaran 
and. a Crucia, are rated as almoflt exactly of standard first 
magnitude brightness, each being 2 '6 times more lumiiioua 
than the pole-star. They have in the northern hemisphere 
six superiors — Arcturus, Capella, Vega, Procyonj Altair, and 
Betelgeux, the standing of which has accordingly to be 
expressed by fractional numbers. Capella and Arctnras are 
' of magnitude 0'2, aignifying that each is eight-tenths of a 
magnitude brighter than Aldebaran, while the figure 0"06 
attached to a Centauri conveys the fact of its superiority to 
ft Crucis by just one magnitude. Carrying out the same 
system of notation, we get negative numbers for the designa- 
tion of still higher grades of lustre. Sirius, for instance, 
sends ug eleven times more light than Aldebaran ; it excels 
the standard by two magnitudes and six - tenths, a pre- 
eminence compactly expressed by calling its magnitude - 1'6. 
To find a star uutshining Sirius we must go to otir own 
fliun, to which a rank can be assigned on the same flcale. Its 
light, as measured by Alvan Clark in 1863, exceeds that of 
the dog-star 3600 million times. Bond made the dispropor- 
tion 5970, Steinheil 3B40 millions to one. From a mean of 
these insecure determinations, Professor Pickering fixed the 
sun's stellar magnitude at - 25'4 ; ^ but various lines of 
inquiry^ separately traced by Mr. Gore ^ and Sir David Gill» 
converge upon a much higher value lying between -26'5 
and - 26'8- It seems fairly certain^ accordingly, that the 
splendour of Siriua is some 10,000 million times fainter than 
the blaze of sunshine in which we live. 

The invention of the telescope itself does not mark an 
epoch more distinctly than the admission of the camera into 
the celestial armoury. All the conditions of sidereal research, 
in especial, have already been transformed by its co-operation. 
The versatility of its powers is extraordinary ; no task has 
yet found it unready or incapable. It is the very Ariel of the 
astponomical Prospero. 

This untiring serviceable ness was made possible by the 
sulistitution in 18*71 of gelatine for collodion as the vehicle 
for the Siilts of silver, the decomposition of which under the 

' Proeetdingi Ainerisan Aca(Lmy, toI. svi. p, 2. 
> Monthly Notiw^ vol Ixiii. p. 184. 



IHE METHODS OF STBEREAL RESEARCH 23 



infiueace of light forms the easeutial part of the photographic 
process. The new plates were, however^ first used for " astro- 
graphical" purposes by Sir William HugginB in 1876. Since 
they are five times more sensitive dty than wet, expoeurea 
with ihem can be indefiiiitely prolonged. They may, besides, 
be prepared any desirable time before, and developed any 
d^irable time after exposure^ thus accommodating themselvea 
in a really wonderful way to the needs of a&tronomers. 

The unique power of the pbotogniphie plate as an engine 
discovery is derived from its unlimited faculty for amassing 
liat impressions of light. By looking lontj enough it can see 
anything thtre is to be seen. Sir William Abney'a expet-i- 
ments convinced him that no rays are too feeble to overthrow 
the delicate molecular balance of silver bromide if only their 
separately evanescent effects get sufficiently piled up through 
repetition.^ By thia capability of taking time for its ally, the 
camaru leaves the eye far behind. Witli any given telescope 
much more can be photographed than can be seen, and the 
threshold has been crossed of a region of research, visually 
inaccesaible, but open to exploration by the far-reaching 
chemical method. 

The penetration of space has, nevertbelessj limits. A 
pixis -ultra is imposed, if not otherwise, by the restricted 
possibilities of continuous exposure to the skj'. Darkness 
does not last indefinitely, nor is it absolute while it pievalb. 
There is always enough light scattered abroad to " fog " 
sensitive plates left long under its influence j and when 
fogging begins, portrayal compulsorily terminates. Thus, the 
plan first adopted by Dr. Roberts of obtaining a single pict^ire 
by means of exposures renewed night after night can be 
availed of only with restrictions. 

The telescope forming the image which imprints itself upon 

prepared plate is always equatorially motmted, and has 
a motion given to it exactly concurrent with the revolution of 
^le sphere. Yet the utmost mechanical ingenuity cannot 
Bake the concurrence absolutely perfect. Minute inequalities 
survive and need intelligent correction. Even more sensible 
are disturbancea caused by the changes of atmospheric re- 
&ttCtiou with the ascent towards or decline from the meridiau 

* ObiertyitOTy, ffol. stu. p. Ift5. 



24 



THE SYSTEM OF THE STAES 



of the objects in course of delineation. For these reasons a 
photographic telescope has, aa a nde» a guiding telescope 
attached to its axis, through which an observer watches to 
counteract, almost to anticipate, nascent tendencies to die- 
jplacement- The strain upon the attention la severe ; ita 
endurance, upon nccaeions, for three, even four hours at a 
stretch, is no small proof of resolution, 

Expoaurea can^ however, be curtailed by shortening the focus 
of the photographic telescope, the image being thus rendered 
smaller, and — through the closer concentration of" the ^ame 
amount of light — more intense. For simply exploring the 
skies, sounding their depths, and dredging-up their contents, 
nothing can be better than the form of an ordinary portrait lens. 
With such a one^ only two inches in diameter, the picture of 
the comet of 18 82 was taken at the Royal Observatory^ Cape 
of Good Hope, the " thick-inlaid " background of which 
afforded the first palpable revelation of the star- charting 
powers of the camera ; and much of Professor Pickering's 
admiKible work in sidereal photography has been done with a 
Voigtlander's " doublet " [two achromatic lenses in com- 
bination) of eight inches aperture and about forty-five focus. 
Objects imperceptible through the Harvard fifteen-inch refractor 
can be photographed with this instnunent ; and it has proved 
extraordinarily efficient for the rapid charting of stars and 
their spectra. The " Draper Catalogue " was indeed com- 
piled wholly from materials collected by its means. In the 
*' Bruce telescope," completed by Alvan Clark under Professor 
Pickering** direction in 1S93, and mounted at Arequipa in 
1896, after transmission through the Straits of Magellan, the 
same plan of construction was carried out on a larger scale. 
The object-glass has a diameter of twenty-four inches, the focal 
length is eleven feet Stars down to the seventeenth magnitude 
are probably recorded on plates ex].H:>5ed to the strong con- 
centration of light thus effected. The portrait-fonn of lens 
has the additional merit of giving a large field of view. Each 
photograph taken with the Bruce telescope covers, with very 
slight distortion, five degrees square (25 square degrees) on 
a scale of one minute of arc to a millimetre. The whole 
heavens could be charted on about two thousand such 
plates. 



THE METHODS OF SIDEREAL EESEAUCH 25 



Where acciirate measurements are aimed at, however, the 
type of inaCrumeut represented bj the MM. Henry's photo- 
graphic telescope ia preferable. The object-glass in thk is of 
the ordinary achromatic kind, but corrected with reference to 
chemical, instead of to viBuiil jicti<jn. The rays selected to be 
brought to a foctis are those to which, not the human, but 
the "photographic retitia" is aeositiTe. The aperture is Id 
ioches, the focal length eleven feet ; a plate-holder is substituted 
for an eye-piece^ while a guiding- telescope of slightly inferior 
ilimensioaa ia enclosed within the eame rectangular tube. 
The field of view with the Paris photographic telencupe, within 
which definition may be considered as virtually perfect, is a 
le three degrees in diameter,' covering an area of not quite 
"five square degrees. Fully ten thousand of these plates 
(allowing for overlaps) will be needed to picture the sphere; 
and they ate being taken in duplicate for the purposes of the 
International Celestial Survey. Eighteen instnxments, modelled 
on that of the MM. Henry, are employed upon it ; and the 
high quality of the data they will provide is assured. Yet 
thought quails before the quantity of materials presently to 
be dealt with. Eventually, we may fairly hope, they will 
be brought within the unifying grasp of statistical research ; 

at only at a heavy cost of wearisome toil. 

Studies of the distribution of the stars," Professor Picker- 
ing remarked, " can now scarcely be undertfiken in any way 
except by photography." But photography, to be really in- 
structive on thia point, must be combined with photometry, 
The portrayal of millions of stars projected side by side oa a 
spherical surface tells us little or nothing of their relations to 

lie immensity of space. Thia can only be found out, for the 
St majority of them, by collecting atatiatic-a of the amount 
~f light they send us. Hence the importance of the photo- 
metry of small stars. Yet no visual means have hitherto 
proved competent to deal with it. Eye-estimates, however 

aided and succoured by instruments, break down when 
shed too far down the scale. The problem is evidently one 
of those reserved for successful treatment with the camera. 

What is called " photographic iiTadiatiou " affords one 
zaeans of attack upon it. This arises from the diffusion of 
' £ullttin Attrtmomiqut, t vi. p. 308. 



26 



THE SYSTEM OF THE STABS 



light within the eubatance of the gelatine film. The particles 
directly meeting the atellar rays reflect tbem irregularly all 
round to other particles, thus widening the area of oheruical 
decomposition, and creating circular imagea which, with the 
same exposure and on the same plate, are found to vaiy in 
size with the magnitude of the stars they represent.^ Thua, 
from a few stars of ascertained brightnesa, that of the neet 
imprinted with them may, dowa to a pretty low grade, readily 
be inferred. The faintest stars, however, give rise to dots so 
small that differential measures of them are scarcely practicable. 
The method is further ci:)mpromised by the uncertainty of the 
law connecting the size of chemical star-discs with brightness. 
It is indeed a mere empirical formula varying with the con- 
ditions of obaorvatioiL It ia not the same for rapid aa 
for slow plates ; it is not the same for twinkling as for 
steady images,^ There is no help but to treat each plate 
aa a document apart, and to assign its constants inde- 
pendently ; and this is to adopt an expedient, not to employ 
a system. 

Nor does the scale of photographic effectiveness agree at 
all closely with that of visual sensibility. Colour has in this 
respect a strongly disturbing influence. The quick vtbratioua 
at the blue end of the spectrum are those most active in 
releasing silver from chemical bonds. Blue stars are con' 
sequently far more, and red stars far less conspicuous self- 
printed than to the eye. Photographs of chromatic double 
stars thus show curious reversatsr a small blue companion 
often coming out superior to its yellow or reddish primary,* 
And tinted stai-s too ftiint for eolour-diaerimination with the 
telescope can sometimea be picked out on a negative, simply 
through anomalies of relative magnitude. Now differences 
of this sort occur not only in isolated cases but methodically. 
Their frequency, aa Professor Kapteyn discovered in measuring 
the Cape Dui-chmuaterung plates, depends upon celestial 
situation. It varies with galactic latitude. And the in- 
evitable inference was drawn that "the stars of the Milky 
Way are in general bluer than the stars in other regions of 

1 Asir. NacJi. No. 2SS4. 

• Gill, Inteoducti^n to Ca.p': Phatograj}hie DureJtirwstervng, p. 24- 

' Eipin.tJfewwitffJ/, Tol. vii, p, 247. 



THE METHODS OF SIDEREAL RESEAJICH 



^i 



the sky." ^ Thoflj photographic photometry has a very im- 
portaat beazing upoa studies of sideieal cotiHtructioti. Bat a 
settled baais of principle is needed to give its ne^oJta the full 
authority that should belong to them. At preseut. actioio 
magiutiid&s are derived very much at the discretion of individual 
obeervere. Perhaps the beat way of treating them would be 
to take them for what they are worth, abandouing attempts 
toveduce them to scales of visual magnitude. They are worth 
a great deaL^ In some respects, perhaps e^Tea more than 
magnitndea estimated by the eye. For their enumeration 
and classification may supply both geometrical and physical 
data regarding the stars ; indications^ that is to eay, as to 
iheir arrangement in space, and indications besides of their 
aaaortmenb by affinity of constitution into indefinitely vaat 
aggregations. 

The photometry of nebulfB haa, so far, obtained less than 
ite due meed of attention. Sir William Hu^ina ascertained, 
in 1866, the extreme intrinsic faintness of such objecta,* and 
there the matter rested until the universal agency of the 
camera was made available. Then, in 1884, Mr. "W". H, 
Pickering described a mode of constructing a scale of photo- 
graphic intensity by expoeing a number ot" email squares at 
one aide of a sensitive plate to a known light-source during 
different inten-'als of time.* These, developed with a nebula- 
pictore subsequently imprinted, afford so many terms of 
compansoa for the relative brightness of its parts. He drew 
in this way a set of " isophotal contours " in the Orion nebula, 
and the map representing them constitutes a record of present 
intereatj and of poesible future importance. The absolute 
brightness of the formation, in the central " Huygenian " 
region, was found to range between VO and 140 units, the 
adopted unit being one-millionth of the light given by a 
standard pentaue lamp. No further equally systematic 
attempts have been made to determine the luminosity of 
iiebulffi ; although its variabihty, assured in some cases, 

' Introduotiaa to Ca^i Fhotetfraphic DuTehmuattruntf, p. 22. 

* Oill, il^. p. xii. Tiote. 
' PKU. 3>tJM. vol clvL p, 302. CC J. E. Oew, Ohasrmiory, Mny 1£»05. 

* /Vm. Anutr. Aaad. ToL 3cx» p. 613; HarvoTd AnnaJs, va\. xxiii. p. 16. 
Sir W. Abney iDdepeudeitCly t^a ytat« l*ter inreDted tk simikt derice, iVWurv, 
Tol. i3. p. 472. 






THE SYSTEM OF THE STAES 



suspected in manj more^ strongly suggests the desirability of 
establishing a fixed plan of measurement. 

There is scaxcely one of tlie numerous taflka of nebular 
oatronomy that cannot be better performed photographically 
than visually. lu the simple perception of faintly illuminated 
surfaces, the tjuickly fatigued living retina is left far 
behind by the imperturbable gaze of a seneitised plate ; in 
their delineation, the subtlest human hand is at a similar 
disadvantage. Professor Holden gave his testimony to the 
effect that " every important result reached by his four years' 
study of the Orion nebula with the 26-inch Washington 
equatoreal/' and very many not comprised in it, were attained 
by Dr. Common*B subsequently taken photograph, which 
required an exposure of only forty minutes,^ Spectroscopie 
inquiries, both stellar *uid nebular, are enormously facilitated 
by the substitution of permanent autographic records for seta 
of quivering lineSj caught in their mean positions only by a 
keen glance at critical raomente, and constantly liable to 
effacement by atmospheric waves. It is true that the range 
of observation ia not the same in both cases. The plates 
in ordinary use ignore the lower end of the spectrum, but are 
affected by the higlier vibrations which, by their quickness 
and shortness, evade the eye. " Orthochromatie " plates, 
sensitive to yeDow and red rays, can, however, be produced 
by staining with eosin and other coabtar dyes; but their 
use is attended by some inconvenience. Uniformity of light- 
action can scarcely be secured with them ; they respond bo 
it with a certain caprice ; and are henoe apt to yield spectra 
of a somewhat patchy brightness. 

The wonderful comprehensiveness and adaptability of this 
method are strikingly apparent in the results obtained since 
1886 at Harvard College. By no other means could the 
spectroscopic stellar survey executed there and at Arequipa, 
its southern dependency, have been carried out on so great a 
scale. The results constitute a veritable spectroscopic Durch- 
musteruug, complete to about the ninth niagnitude ; and the 
work is now being extended to fainter stars. The manner in 
which it was conducted, although described by Frauuhofer and 
Secchi, was virtually novel. A prism large enough to cover 
' Owrlaiid Ifonthly^ NovembeT ISSfl. 



THE METHODS OF SIDEREAL EESEAKCH 29 



the entire object-gloas is placed in front of it. The stellar 

beams are thiia analysed before they we concentrated ; every 
stellar image is trunaforraed into a prismatic rii>aud ; and stara 
by the dozen or by the score print their eeparute sjjectra on a 
single plate with a single expoaure. SUt and cylindrical lens 
are alike rejected ; the diurual motion is employed to widen 
the sEpectral bands sufficiently to bring out their distinctive 
features. That ia to say, the stars are allowed to " trail " 
slightly across the direction in which their light is dispersed. 
The results are admirable ; innumerable lines are cleai-ly 
recorded ; but the highest degree of accuracy cannot, in the 
abeence of any system of reference-liues, be given to deter- 
minatioDB of their positions. Mrs. Flemiug's scrutitiy, mean- 
time, of the i*ecords thus profusely accumulated, has led to the 
discovery of hundreds of objects remarkable for the unusual 
quality of their light; and the spectra of about 1800 bright 
ebars, photographed on a larger scale with more powerful 
inBtrnmenta in the northern and Bouthern hemispheres, have 
been discussed and catalogued respectively by Mlbs Maury and 
Miss Cannon. 

For detailed identifications, the more laborious plan 
adopted by Sir William Huggins in 1879 is still pursued. 
The stars are taken one by one ; their rays are admitted 
through the postern-gate of a slit, and record their peculiarities 
aide by side with a comparison-spectrum providing starting- 
points for measurement The "Atlae of Stellar Spectra" 
published by Sir William and Lady Huggius in 1899, 
exemplifies the perfection with which details, otherwise 
iuaccassible, can thus be brought into view. Glass, which ia 
strongly abeorptive of the shorter wave-lengths, is excluded 
from their apparatus. The stellar raye admitted to it are 
concentrated by an 18-inch speculum, and dispersed by quartz 
prisma. Introduced by Sir William Huggine in 1868, the 
spectroscopic method of determining stellar motions in the line 
of sight was perfected, ten years later, by the change of venue 
effected on the initiative of Dr. Vogel, from the eye to the 
plate. Its superiority has, in this difficult branch, trium- 
phantly asserted itself. The precision photographically 
attainable in measuring spectral shifts is chiefly due to the 
virtual elimination of the effects of air-troubles. The lines 



THE SYSTEM OF THE STARS 



from which information &r to movement has to be gathered 
depict themaelvea in their normal places. Their waverings, 
so bafQiug visually, are chemically ineHective. 

Nor is it only here that the autographic mode of pro- 
cedure attains a refinement on a par with its power. The 
subtlest problem of atellar astronomy is that of aiminil 
parallax. It leaves room for no compromise in the matter of 
lu^cTU'acy, yet it has been solved with the aid of the camera. 
Professor Pritcbard's careful and persevering experiments 
established the validity of photographic determinations of 
parallax, and so furnished to the sidereal armoury a new 
weapon of pi-ecisiou and loug range. It happens, moreover, 
that the objects most inviting to the one mode of treatment 
are precisely those reached with difficulty by the other. 
StarB too faint for the eye to deal with satisfactorily come out 
on negatives in neatly measurable form ; while the brighter 
stars suitable for observation with tlie telescope and micro- 
meter give distended photographic images uupromiaiug for 
exactitude. 

There are intlicationa that reflecting telescopes will before 
long regain the position of preference which Fraimhofer's 
skill in grinding lenses forced them to abdicate. They have 
over their rivals the special advantage of being perfectly 
achromatic ; they collect at one focus all the ray^, visible and 
invisible, striking them. This the very best refractors fail 
to effect However skilful the combination of different kinds 
of glass, a largo amouat of ligbt is necessarily "thrown 
away" ^ Opticiana have to choose what sections of the 
spectrum they will turn to account, and neglect the rest 
Photographic re^jctors are for this reason useless in ordinary 
observation. The images they give are wholly built up out 
of blue light, while the light proper for seeing by wanders 
unserviceably astray. The plates exposed with them must 
accordingly be sensitised in correspondence with the mode of 
their correction. No tolerable results could be got with 
orthochromatic plates in the Henry telescope. 

These drawbacks are, nevertheleas, to a great extent out- 
weighed by tyauntervailing prerogatives, Refractors are moi^ 
manageable than retiectors. They are less sensitive to slight 
1 SiF H. Clrabb, Monihl^ AViw. yol xlvii. p. 30*. 



THE METHODS OF SIDEREAL EESEAECH 31 



less intolemnt of unequal pressures ; they Bcconuno- 
date tbemeelves better to mechanical exigencies, can be more 
rigidly mounted, hence made to follow more strictly the 
circling of the sphere, and so to ]seep & steadier hold of the 
Ejects in the 0el4 of view. Where measui-ee of precision are 
liefly aimed at, choice is thus naturally directed to them, 
and they have been stamped as the official inatrumenta of 
celestial photography by their adoption for the vast 
Btar-eharting operations decided upon at the Paris Con- 
gress of 1887. The splendid nebular pictures, on the other 
hand^ obtained with reflectors by Common and Roberts, 
and more recently in America by Keeler and Eitchey, prove 
their superlative fitness for tasks of delineation. More- 
over, their future extensive employment in spectrographic 
work eeema inevitable. With refractors, the range of 
good spectral definition is narrowed by their inequalities 
of focal concentration, while reflectors display in uniform 
distinctness the whole light-gamut from end to end. This 
point of superiority is very important, since a partial view 
of stellar spectra is, in many cases, not only unsatisfac- 
tory bat misleading. All telescopic varieties, in fact, find 
their place in the boundless fields of photographic research. 
Mirrors do not exclude lenses; instruments of short focus 
have A special function, while for other purposes those of 
long focus are preferable. The needs of sidereal investigation 
are manifold; they claim subventions from every quarter; 
they invoke the most diversified forms of assistance. And 
their demands have been met by a geueruus largesse of 
inventions and contrivances. 

The foundation of stellar astronomy is, as we have said, in 
infinitesimal accuracy. It could not otherwise exist, since the 
quantities concerned are so small as to disappear amid the 
errors of rough obaervatione. But for its progress, something 
more is required. A few scattered items of knowledge do not 
constitute a science. The word implies the suflusion of a 
subject with intellectual light derived from large inferences. 
Large inferences, however, must be based on a plentiful store 
of facta ; and the facta collected by sidereal study are, even yet, 
few compared with its innumerable objects. They are, indeed, 
being continually multiplied. The alliance with photography 



Jl THE SYSTEM OF THE STABS 

iifcs »:o=cr=»i tJi-r tiisi* .:: reaa>:ing without impairing il 
asccmj. Ft •:■ r-ibsr '" V Ar? co-idd the desired informatiB 
x 5:^rofc. -M .dj iz. « ': :ukdizice, but with the 
■^'dTcn-ir. F:c. :•: a ci=r:Aia exient, the work has to 
i-"of iij-i:za5 t^'t. r-zi- i* rsi'ii intuitions are necessary ] 
rcu."*:.::^ * '-^ irizz -:■: zi^iLrsiatical reasoning, beoiQ 
"wc.'K*: :!■:? «Ccrt> att li'tc-^r^c. the w^e^ed faculties at 
r^rL* :c :vc.r^ze tc toie them, so some degree of 
:3T«r^ is iTsi^^iSirZ-r r'.'-r sristdining the tmiversal int 
w^-* r>r* 1 ?z:iec: ::* TiiAlirr. but declines to foUow too] 
tATir ic jx:''.Ti~^ :;■ ;-•* bilrinr-plaoe for another. 

T-z*^ z*:c zLi-rrvZy wLa: i: c-^ji do. but the rate at which it] 
v%kr i: ::. his :c ':«r ;»:n5iirr*i*i in estimating the value dl 
yi?:c-vVt*rCT i^ m ai^i :c iSTr^zLomx. And there is little fear 
:c 'J^ ^^z:z.z*'jz^ Useznir thzv^c^h sluggishness of pace. It 
i»tfc«? -r i VST ■" S:;:r=i ;:2d Drang " of progress. The 
.•jivuTj: ^x-w«r* :■:' er-.:=i-=rii:c»=. dvsiied br Homer for cataloga- 
;z^ ;is ,r:vi :c* Gr«ek siije aie £ir outdone by it Its 
i:iS3r;:rie-i:TAl::T. riKWTer. came to hand just when the 
rs:*I::::3-ii:iv"^ :iir*::«rr c: the problem set by the heayeiia 
S.y*z :c ':« *T*srei in ill ::s f:rnudable reality. 

Tbtf $Tar'.:"^r^B*5 :z :hr r>i>:-:'^phic method is due, not 
il'.'Oii* :o :rj=' iT«i: =.;u:::tr o: vcoects it can register t(^ther. 
:;;: ;o ;>.-? ,l:<r%rr5;:- Ari division of labour it makes possible. 
K^vr.*.? -*cCA:~-vi *;y ;: hi^^ ih* enormous advantage of being 
•.vrrjiATs^r.^ Tb^y di :bc i:::in^ incidents of the heavens as 
:h<f v^'.*iv,vr*vi: lies :he iransien; accents of the human 
\ ,':^v. All ihr; sotriy hc'-irs of unclouded darkness can thus 
Ix- ,li-^c«v; :o AV-r^c n^icriils for subsequent investigation 
i:; ,l4xlxj:^.: .XT >i*d w-tither. Innumerable experts may be 
^xv.yls'w. :u ;>.:s wriy *: re::;o;c plaoes and with different ends 
'.*.'. \ vw A s;v«;li* :uy&::ve :::iT be communicative regarding 
*;>-lU? vhcCv ".•;:'; ry. d:s:r.ru:i^^n. ^xirftUax, proper motion; its 
,vr,;'Av.'.s.-r. w::h s-rher? ^rves ;o t-est variability, or even to 
r"\ ;S4' *"AV> .■:" s.'v.5e ::;ys;crl:uf sidereal cataclysm. Yet the 
».\vft*•««.^r, ,-: y:x;-rf# o: <vlest:al objects does not in itself 
xV-v.*; ;:;;•«' sr. uvrtvist ot kr.owle«,:ce. They contain latent 
*,;-»o^".;a:u".;. ;;:*: as :he siit^s ibeniselT«3 do, but the educing 
^\:\\\'«v Vv «hK'h i: is nude stf^sihif is as n etcom y in the one 
v'^k*^ *» u'» :>-^ v^:^.ef. Zeal in securing them !• ao 



THE METHODS OF SIDEREAL KESEARCH 33 



;ht avail without induatry ixi discusslDg them, whether 
inally or by deputy; for astrographical tasks are easily 
le of delegation. Star-prLnts are indeed usually dealt 
ith by specialUta in measurement, whose Bkill has been 
by many noteworthy diacoveriea. Those, too, 
the ftttuie has in i^^rre will doubtleaa fall macb more 
ty to the share of investigators armed with microscopea for 
bisection of star-dol^ on glasa plates thau of tekacopic 
lera of the heavens. 
Kot that the telescope is, or ever can be superseded. On 
'tiie oontraty, the enlargement of its capacity becomes more 
desirable with every fresh addition to the apparatus used in 
lOonjuDction with it. Modern sidereal astronomy may be said 
%o live an light. Large telescopic apertures are a sitw qud 
nan for its growth and activity, A considerable proportion 

Inf the objects it has to do with are, in fact, beyond the range 
4f AdaU instruments. It is. however, &a important to 
economise as to collect the far-travelled rays from the stars ; 
pUd m this direction little has been accomplished. Under 
l^lfae beet conditions, no more than 5 per cent of the light 
' Itriking the 40 -inch object-glass of the Yerkea refractor 
ictaiUly reaches the sensitive £lm after trausmiasion through 
t^ great Bruce epectiograph \ and although the waste in 
other forms of apparatus is less formidable in amount, they 
cannot yield results of equal precision. For obtaiiiing legible 
Rcords of the faintest spectra, none can compete with the 
slitless spectrograph employed in connection with the Crossley 
veflector at the Lick Observatory, How much of the incident 
Bgfat it tarns to account has not been ascertained ; but stars 
of the fifteenth magnitude have none to spare, and their rays 
ba Buccesafolly anadysed on Mount Hamilton. 



CHAPTER HI 



THE STAKS A& SUNS 



The Btara, speaking broadly, are buus. But what is a eun? 
We can only reply by taking function into consideration. A 
sun ia a great radiating machine, and the obvioufi chterion 
for admission to the order is fitiieaa for this office. Qualifi- 
cation to be a centre of light and heat is the dominant 
characteristic of each of its true members. Now the solar 
emiseive activity ia concentrated in a shining shell of clouds 
known as the "photosphere," which the entire energies of the 
organism (ao to speak) seem directed to maintain and renew. 
And with reason, since its efficiency as a radiator depends 
upon the perpetuation of the condensing process by which 
this brilliant surface is produced. 

The poaseasiou of a photosphere muat then be regarded &s 
an essential feature of the suns of space. But such a struc- 
ture can only he formed in an incandescent atmosphere, the 
action of which modifies^ more or less powerfully, the light 
traversing it. The spectroscope can, in fact, alone decide 
whether a given sidereal object be, in the proper sense, a sun. 
For it is not so much the quantity as the quality of its radia- 
tions that determines the point. They must be such as can 
be supposed to emanate from condensed and vividly glowing 
matter bathed in cooler, though still ignited vapours. That 
ia to say, they must, when dispersed by refraction or diffrac- 
tion, constitute a fundamentally unbroken prismatic band, 
marked iyicid^Titall^ by effects of absorption, A continuous 
ftinge of vivid light, crossed by dusky lines, is hence the dis- 
tinctive spectrum of a sun. 

The enormous light-power, and, 60 far, the solar nature of 



THE STABS AS SUNS 



35 



the stare, followed as a corollary from the Copernican theory^ 
since at the unimaginable distances implied by their apparent 
immobility while the earth performed its vast circuit, they 
should otherwise have been totally inviflible. But the analogy 
could be strictly tested only by spectrum analyaifi, and it 
proved virtually complete. Complete, that is, for the great 
majority of the etellar populace. There is a residuum in 
which it is impaired ; there are a few scattered inatancea in 
which it is actually overthrown. 
•■: This degradation of type shows itself in different ways. 
Absorption in some cases becomes so immoderate as well-nigh 
to smother the original light of the star ; the atmosphere in 
others outshines the photosphere^ giving rise to bright instead 
of dark lines in the spectrum ; while in certain objects a 
similar efi^ect appears to be produced rather by a paucity of 
photoepheriCj than by the intensity of atmospheric radiation. 
When the failure boa gone so far that the light of a seeming 
ite, analysed with the spectroscope, ia found to consist chiefly 
of isolated rays of various colours, then the object approximates 
more to a nebula than to a star. It certainly cannot lay claim 
to the designation of a sun. 

But as in the other kingdoms of nature, ao here ; there are 
no abrupt tranaitiona Continuity is everywhere maintained. 
The descent from a perfect sun to an imdoubted nebula 
ia effected without interruption. Hence, inevitably, some 
uncertainty in classification. Broad divisions are easily 
established, but hard and fast outlines to thoee divisions 
cannot be drawn. " Frontier-instances " abound, and compel 
recourse to somewhat arbitraty distinctions. We propose, 
in the present chapter, to consider only bodies of assured 
status, with radiative machinery in fuU working order — 
bodies, as to the essentially sun-like nature of which there 
c&n be no difference of opinion. 

The four spectral types discriminated by Father Secchi 
atill form the basis of arrangement. They may conveniently be 
designated as Sirian and solar, Antarian aud carbon stars* all 
showing continuous spectra crossed by dark lines of abaorp- 
tion, to which, in the two last varieties, dusky bands and 
flutings aj^ superadded. The Sirian order, however, as 
originally describedj included certain bright atars with seem- 



36 



THE SYSTEM OF THE STAES 



ingly blank spectra, prominent chiefly in the ooosteUation 
Orion ; and these have gained auch importance through the 

discovery of terrestrial helium as to compel their relegation to 
a class apart. Thus^ Secchi's ^hite stars have, as it were, 
epontaneously ranked themselves into two great battalions, 
and we shall consider firat those that came last to full recog- 
nition. 

Stars of the " Otloq " quality are purely white ; their light 
approximates to being io the native state ; that is to say^ it is 
Bcarcely modified by the arresting action of their atmospheres. 
Yet it 19 not wliolly characterless. Hydrc^en and helium, 
especiallyj have left their Btamp upon it. Fig. 1 of Plate 11. 
reproduoes, by the kind permission of Professor Pickering, a 
spectrograph of e Orioms, the middle star in the Belt, taken 
at Arequipa, November 10, 1896, with the 13-inch Boyden 
refractor. The dispersion was produced by means of a train 
of prisma in front of the object - glass ; no slit was used, 
consequeDtly uo comparison could be made available ; and the 
exposure allowed was of 106 minutes. The darkening at 
either end is due to the limited range in sensitiveness of the 
plate employed. Helium-lines are nearly as strong as those of 
hydrogen in the spectrum of e Orionis ; and their association by 
Sir Noiman Lockyer and others with the newly unearthed gas 
encouraged Dr. Vogel, late in 1895/ to search for further 
flignfl of their presence in a large collection of stellar spectro- 
grams, prepared at Potsdam by Dr. Wilsing, To his siirprise, 
he foimd Orion-stars to be fairly numerous all over Che sky . 
and the late Mr. McClean's researches proved them to be still 
more prevalent in the southern than in the northern hemi- 
sphere. 

They are, for several reasons, of particular interest. 
Sidereal genealogists assign to them a very early standing. 
They seem, indeed, to occupy a position intermediate between 
true nebulas and Siriau stars. This conclusion is enforced, 
first, by the visibility of nebulous appendages attached to 
many helium-stara ; next, by the low density which is one of 
their least doubtful characteristics. And it obviously implies 
that they have made less progress in contraction through 
cooling than globes of more substautiat build. But even 
^ SitxtngtinridUc, Berlin, October 24,, IStffi. 



THE STARS AS SUNS 



87 



among helium-BtarB themselves gradations of age are per- 
ceptible, Mias Mflury arranged theiu into six groups/ corre- 
flpondiug, it was plausibly supposed, to so many stages of 
sdy&nce in condeusatioa. One of the earli^t of her etars, 
$ Orionia, ia aigaiiicantly placed at the core of the great Fish- 
mouth nebula ; and la the species to which it belongs^ metallic 
abeorptioa is barely discernible, while some peculiar lines, 
identified with hydrogen by Pickering and Rydberg, come into 
view, together with a few delicate triplets ehowa by McClean 
to originate from oxygen. Helium and hydrogen are beaidea 
represented by their usnal acts of rays. Although some 
advance from the immature state of ^ is marked by the spectrum 
of c Ononis, it etiU bears the impre-ss of oxygen and "cosmic" 
hydrogen. Silicon-absorption is unmistakable in Bellatrix ; 
^ Crucis, one of the gems of the Southern Cross, is an 
exemplary oxygen-star; and, at the stage reached by Algol 
and Rigel, primitive symptoma have been superseded by the 
clear emergence of magnesium and calcium-absorption. 

To resume. Helium-stars have the&e special features. They 
pOBsess envelopes of helium capable of selective action upon 
li^t. Some among them^ and notably those with nebular 
affinities, show dark rays of oxygen and cosmic hydrogen. In 
aU, metallic absorption is feeble. The photospheric radiance 
o>f all spreads abroad into space sensibly unmodified by atmo- 
spheric stoppage. Finally, they tend to congregate in the 
Milky Way, particularly in the eouthem hemisphere. 

Sirian stars resemble their predecessors in their brilliant 
whiteness and comparative freedom from atmospheric encroach- 
loeDtB. There is little or no trace in them of the general veil- 
ing efiect by which our own sun is shorn of a large proportion of 
his more refrangible beams. The Sirian spectra, although not 
intact» are entire, and are hence especially strong in their 
ultra-violet sections. To this immunity from absorptive 
attacks there is one remarkable exception. The eign- 
manual of hydrogen is stamped upon them with extraordinary 
intensity, A number of metals are also present, but they 
show lines too faint and fine for easy recognition. Fraun* 
hofer*s " T)/' the ubiquitous double-line of sodium, is neverthe- 

obviotiB, as well as the " K " of calcium and the " blue " 
^ Harvard Awutls^ toI. xxviji, p, IS, 



38 



THE SYSTEM OF THE STARS 



line of magQeeium. With the best definition, too> mtilti- 
tudinoua fine striations, a large proportion of them due to iron, 
can be photographed. Thua, in the spectrum of Sirius (see 
Plate II. Fig. 2), Mr. S. A. Mitchell counted on his plates 75 
Hues in the restricted section compriaed between two consecu- 
tive hydrogen-lines {H^ and Hy).^ 

, Photography has indeed played an essential part in the 
investigation of Siriau spectra. The strength of their ultra- 
violet radiations lends to the method peculiar efficacy in 
dealing ^vith them ; and the advantage was turned to the 
fullest account by Sir William Huggtna in 1879.' The 
impressions secured by him with an apparatus from which 
glass was wholly excluded, afforded a discovery memorable, 
not in the history of astronomy alone^ but in that of mole- 
cular physics as welL 



Tia. 1.— 8|»rtnim of HjdnjeMi ^ Wliifca BUn (Hagglnm). 



A photograph of the spectmm of Vega obtained with one 
hour*8 expoaure contained twelve strong lines (shown in Fig. 1) 
forming a group in obvious rhythmical connection. They 
crowd together more and more as their wave-lengths shorten. 
A common ongin for the entire at once auggeata itself: and, 
on the ground that its two moat refrangible members were 
already known as hydrogea-lines. Sir William Huggiua did 
not hesitate to pronounce hydrogen reaponsible for all His 
inference has been amply justified. Seven of them proved to 
have been, a short time previously, photographed directly from 
glowing hydrogen, by Professor H. W. Vogel ; * and the com- 
plete set was similarly procured later by M. Coriiu»* but with 

^ AitTOpK Jaum. vol. r. p. 32. 

> Pkit, Trans, vol. clixi. p. 669. 

' MtT. NfKh, No. 2301. 

* JmrMtl dt Physique, E, v, p, 841 (1886). 



THE STAES AS SUNS 



S9 



extreme difficulty. Although the purified gas filling the 
capillary tube placed in front of the slit was excited by 
powerful electrical discbai^a, its highest radiations took no 
less than three hours and a half to get satisfactorily printed. 
Some idea may thug be gained of the intense incandescence 
leigDing in the remote stellar atmospbereB from which they 
wete fii^t derived. 

The law regulating the sequence of hydrogen -lines was 
aaBigned by Balmer in 1335.* It is a purely empirical 
formula that may be modi^ed at will ; yet since it has 
availedt not only to connect together numerically lines already 
obeerred, but also to locate additional ones, it must corre- 
ipond to something esacDtial in nature. The membem of the 
" Huggina-eeries " are really indefinitely numerous. They 
crowd up towards a limit in the ultra-violet, which, like the 
aaymptote of a curve, is continuously approached, yet never 
ftctnally attained. Subjoined are the wave-lengths, in ten- 
miUionths of a millimetre, of the first twenty, as given by 
Dr. Kayser.^ Their designation by the letters of the Greek 
alphabet is now all but univerBal 



Waw 


tODj^th. 




nsifgiLEtlan. 


Wavfl-tone^h. 


6583-07 (Frannhofer'e 


C) 


H A. 


3734-51 


4861-52 < 


I* 


F) 


H/£ 


3722-08 


4340-63 (near „ 


G) 


H V 


3712-11 


4IOl-90( 


n 


h) 


Hf 


3704-00 


3970-Sa ( 


n 


H) 


Ho 


3697-29 


3881) -30 






Hit 


3691-70 


3935-53 






Hp, 


3686-97 


379S-04 






Ho- 


36S2-9& 


3770-77 






Ht 


3679-4d 


3760-30 






H V 


3676-40 



^K Theoretical limit A 364613. 

^^ This series was believed to be solitarj' until Professor 
Pickering announced, in 189V* bis identification of a com- 
panion-set of line.^ in the peculiarly constituted southern star, 
^ Pnppifl. Much knowledge had by that time been gained 
regarding spectral aeries. The pattern set by hydrogen had 
been foxmd to be conformed to, with variations^ by the 

^ WiedemMHi'fl AnnaUn, Bd, xxv. p. 80. 

* Han/dbueh d<r Spietros'wpie, Bd. ii. p. &06. 

• AstTOph, Ji»tmal, toL t, p. $2, 



40 



THE SYSTEM OF THE STARS 



eimssions of a number of other subetanced ; and what might 
be called a typical arraagemeiiC of three series linked together 
by a slightly modified common formula^ had come to be recog- 
niaed as prevalent. This triple set consists of two " subordinate *' 
series^ conveiging towards the same Lmiting wave-number^^ and 
a " principal " Beries, composed of much more widely separated 
lines. Now Pickering's series is constructed accordantly with 
Huggins'g seriea ; it runs up to an identical " head " in the 
ultra-violet; and was hence at once» and beyond the possi- 
bility of mistake, associated with hydrogen. Yet it cannot 
be procured artificially. We are entirely ignorant of the 
conditions under which it is displayed. It is only certain 
that they prevail in early helium stars^ and have become 
abolished by the time the Sirian stage is reached. 

The fact is a memorable one that the true character and 
full extent of the hydrogen -spectrum became known through 
astro-physical inquiries. It shows with what curious un- 
expectedness the obligations of one science to another may 
be repaid, and exemplifies the advantages to be reaped by 
terrestrial chemistry from extending its experimental range 
to the heavenly bodies. The " white star " set of rays has 
fumiahed a clue to many a spectral labyrinth, and its under- 
lying principle of order has been proved, by extended 
laboratory -research, to be of vital importance in molecular 
physics. The modulated behaviour of different rhythmical 
series under varied treatment is profoundly significant, and 
indicates one of the most promising methods for probing the 
secrets of material structure and ethereal relationships. 

DisproportioE in etrength between hydrogen and metallic 
lines reaches a maximum in the spectrum of Sirius, a section 
of which, from a superb Harvard photograph, is shown in 
Plate II., Fig. 2. Helium makes no effect in it; and the 
change of type from e Orionig could be discerned — even if 
every other symptom of it were absent — from the substitution 
for the " blue " hehum-line (at X 44Y2) of a dark magnesium- 
ray slightly less refrangible. This trait of magnesium -absorption 
is relatively much stronger in Sirian stars than in the eun and 
his analogues; and a particular meaning is lent to the dia- 

* W»Te-numherB, or fi-aquenoiee of Tibrition, »» the reciprocala of waTB* 
]eugthE>, and are klwajA substitatod for them in nimierical calcDlationa. 



THE STABS AS STJNS 



41 



crepancy by the circumstance that the line in question iB (in 
Sir Norman Lockyer's phrase) " enhanced " in paaaing from the 
electric are to the spark Silicon is the only non-metallic 
element besides hydrogen identified La Siiius \ and here again, 
t^e identification i^ by means of lines intensified in the electric 
epork. 

An erpoeura of eighty-two minutes^ with the 13-inch Boydeu 
la&actor, was given at Areqtiipa to the negative from which 
Plate 11., Fig. 3, is copied The spectnim represented is that of 
the niagDi6cent Canopua {a CariniE), The diminished width of 
the hydrogen 'lines, and the augmented conspicnousness of the 
calcium K> signify a decline from the condition of Siriua 
towards that of our aun. Many distinctively solar lines have 
aaeerted themselvea \ yet no less than twenty members of the 
hydrogen -series may still be eeen or photographed, while only 
four can be clearly made out in the sun. There is reason 
to believe that the measure of their affluence givea a searching 

L -test of stellar constitution. 

IH Although the hydrogen spectrum is dominant throughout 
the first order of stars, it is not in all represented with equal 
empham The diifuee lines constituting it in Sirius and 
Vega show, in descending gradations of fineness, in Castor, 

IFomalhaat, and Altair. They appear, moreover, less solitary 
in proportion as they become less intense. Their all but 
eiclueive posfiession of the ultra-violet field is progressively 
encroached upon by the development of other spectral lines. 
The co-ordination of the two kinds of change may be expressed 
by the general statement that the prmninenci of Tays dut to 
ahtorption hy metal& in the speeira of whiU Btars varies in- 
veruly toiih Chat of ike hydrogen, series. When the hydrogen- 
niys become eflaoed from the invisible, and cease to dominate 
the Tiaible part of the spectrimi, the second, or solar type of 
stars is reached 

These are about one-sixth less numerous than the fiist 
kind. We may tafce aa examplea : Capellaj a Ursfe Majoris, 
a Casaiopeiae, a Arietis, c Carimje, a Serpeutis, Aldeharan, and 
Arcturua. The pole-atar, Procyon, a Leporis, and a Persei, 
stand nearly midway between the two groups. 

A golden tinge like that of sunlight betokens, in stai^ of 
tfafl second order, a spectrum more or less perfectly similar to 



THE SYSTEM OF THE STAHS 



that of the sun, delicately ruled from end to end through the 
absorptive effects of a great variety of metallic vapours ; non- 
metallic Bubstaacea are lepreseuted only by hydrogen, itasociated 
in the sun with ailicoa, oxygen (a trace), and carbon. Th« 
extent to which the Unea are crowded together may be judged 
of from the photographed spectrum of Capella depicted in Plate 
IL, Fig. 4. Dark hydrogen-rays are present, but with no pre- 
eminence, and No. 5 of the aeries (He) probably lurks concealed 
within the shadow of an obscure dififuse band which covers its 
place. This hand was named by Praunhofer H, and its com- 
panion, a little higher up, is designated K. The pair form the 
moat strongly^marked feature of the spectrum of calcium when 
raised to the highest pitch of incandescence ; the ordinary 
tiame of the substance shows them but dimly. 

They are peculiarly characteristic of the spectrum of solar 
prominences. Daylight observafcions at the edge of the aan 
disclose them as invariably brilliant in the chromosphere, and 
shining up to the very summit of each one of its flame'like 
extensions. Moreover, during the total eclipse of May 17, 
1882, the violet radiance of H and K Eooded the shadowed 
part of our atmosphere, and dimly UluuiinatLng the purple 
diBC of the moon, was scattered far out among the " aig;rettes " 
of the corona. 

Corresponding symptoms of the phenomenal importance of 
calcium-vapour in the aun'a economy had been already de- 
tected by Professor Young in the course of his daylight 
epectroecopic obaervatioas. But they were not fully appreciable 
until photography was made available for their continuous 
investigation. H and K lie at the verge of the visible 
spectrum ; the eye perceives tham with some difficulty ; 
while the chemical retina is extremely sensitive to their 
vibrations. Further, the black bands due to their absorption 
have, in the Frannhofer-spectrnm, dark shadings symmetri- 
cally attached to them, by which the bright rays of the 
same refrangibility used in prominence -photography are 
eflectively sheltered againflt atmospheric glara This con- 
venience was amply turned to account when Professor Hale 
and M. Dealandres independently adapted to the purposes of 
daylight solar photography Janssen's invention of a double 
slit for isolating individual qualities of light. In particular. 



d 



THE STARS AS SUNS 



43 



it led them both to the discovery that the sun's diac is over- 
spread with wreathed ghapea of glowing calcium, designated 
by Profeesor Hale " flocculi," and regarded aa expansions 
of vaporoufl niaaaea^ lying at the base of the chromosphere, 
uid fed hy mounting torreute &om the Bun'e interior.^ The 
intensity of their radiation appears anomalous, and may, he 
thinks, need to be explained otherwise than by mere elevation 
of temperature. 

Presumably, all solar etara poaaees calcium -appendages 
eqnally well-developed with those of the sun. Their epectra 
are impressed with the same powerful H and K abaorpfcion ; 
and it is hardly conceivable that a stamp so peculiar^ and 
to emphatically impressed, should have a different meaaiug 
in them from that deciphered in the one star accessible to 
interpretative research. The calcium H falls, neverthelese, 
very near indeed to the hydrogen He (see Plate II., !Fig, 3), 
the difference of their wave-lengths amounting to no more 
than one ten-miUionth of a millimetre, and it lb only on the 
best negatives that they can be separately distinguished. 
Usually^ either the hydrogen-line is so widened as to mask 
the calcium-line, or the calcium-line as to enwrap the 
hydrogen-line. Professor Young, however, saw them side by 
gide, bright, during his observations of prominences in 1879- 
1380 ; and they appeared dark in a photograph of the spectrum 
of a Cygni taken at Harvard College, November 26, 18&6. 
In this star, a spectrogram of which, taken under Professor 
Pickering's direction in 1893, ia reproduced in Plate IIL. Fig. 
1, the hydrogen rays have thinned down almost to their eolar 
condition, while other metallic lines are tine yet pronounced. 
Such a critical balance of conditions alone made possible the 
individnalisation of the two lineB on the earlier Harvard plate. 

Thus, while the double origin of H opens the way to 
mifiunderstandings, the state of the calcium-line at K is a 
most usefol index to the physieal condition of a star," Next 
to the mode of appearance of the hydrogen series, it is 
perhaps the feature most deserving of study in analysed star- 
light. The substance emitting both K and K is evidently 
of first'rate importance among the vapours surrounding the 

^ FuiAi<aiwtu of Vu Ytrkes OhKTvntoryi yo\. iil. parti, p. 16. 
=* A3 Sir William Huggitts pointed out in 137&. 



44 



THE SYSTEM OF THE STARS 



sun. And th^Lt it is, In fact^ the metal calcium in a highly 
rarefied state, was proved by the expeiimentB of Sir William 

aud Lady Huggine in 1897.' The hypothcBia of its dis- 
Bociation in the 8iiu thus remains unverified. It is true 
that the twin-lines H and K difler in their relationships from 
most of the other ccmponent-rays of the calcium spectrum. 
They stand aloof from any arrangement into seriea ; they 
exhibit special effects under pressure^ and — ^we may add with 
approximate certainty — in a magnetic field. Such personalities 
of behaviour in Bpectral lines are familiar in the laboratory 
as well as in the heavenly bodies They do not e^m to 
iudicate a compound nature (as ordinarily understood) in 
the emitting element; yet they undoubtedly suggest a lesa 
rigid uniformity of structure among its minute vibrating 
particles* than had, until reeentlyt been held compatible with 
the accepted principles of physical cbemietiy. 

Some alight diversity of detail is to be found among 
Bpectra of the second type. In the main, however, they are 
cloeely alike ; and the chemistry of the sun may be called 
normal for all his congeners. The model solar star waa 
long supposed to be Capella ; aud it is true that in the 
invisible, as well as the visible part of its prismatic light, all 
the characteristic solar groups exist iu about their solar 
strength. Dr. Scheiner identified with extreme precision 
255 lines photographed from Capella (between wave-lengths 
4124 and 4638)* with lines derived from the sun;' and it 
appeai-ed reasonable to infer an almost perfect constitutional 
simihirity between the two bodies. Capella has since, never- 
theless, provetl to be a double star with a compound spectrum ; 
and the secondary component gives light resembling that of 
Procyon or Canopus more than that of the sun. Thus, the 
ultra-violet hydrogen-lines distinguishable in the joint spectrum 
belong to this star, and not to the primary body, which is 
genuinely solar. In the sun itself^ the hydrogen-series has 
no more than four representatives. Professor Ames was able 
to ascertain, through the cloak of calcium -absorption, the absence 
of Hf, its fifth member ; * and some vague shadings, noticed 

^ Astroph. Joumaii vol. vi, p. 77. 
=* Rftyser, BandbueK der SptctroKopte, Bd, i, p. fi7fi. 
^ JwtT. Naeh,, No. 2»23, * F%il. Ma§. ToL xa. p. 54, cerioB t. 



THE STARS AS SUNS 



45 



hj KowUnd oear the poeitionfi of ^me of ita more refrangible 
associates, are of imcertuiii meaaing. Solar stars of pure 
type should, thea. display four, and only four, ilark r&ys of 
liydiogBn ; the presence of additional onea givea nee to a 
suspicion of duplicity. The chief member of the aouthern 
slar-pair, a, Centauri^ again oETers a perfect example of a sun- 
like spectrum ; and this stai^ being the sun's equal in mass, 
may be concluded to be similar to it also in temperature, 
dftnfiity, and light-power. The usea, for purposes of com- 
paiifloUf of an authentic model-eun have been expressly 
idierted to by Sir David GiLL 

In Aldebaran tbia standard is pretty widely departed 
from. The pale rose tint of its light is accounted for by the 
sUghtnesa of abeorptive e£fectB in the red end of ita spectrum, 
while numerous lines modify the yellow and green, and the 
violet rays are so feeble that, with an exposure yryily times that 
required for Siriu^, Sir William Hugginsj ita original spectro- 
grapher, obtained only an impression virtually conterminouB 
with the perceptive range of the eye. This feebleness of 
chemical action is perhaps due, not to intrinsic deficiency of blue 
light, but to its stoppage in the vaporous envelope of the star 

The spectrum of Arcturus (a Bootis) varies from the solar 
pattern in the flame direction, though not to the same extent, 
wi that of Aldebaran. A section of its blue part is copied in 
Plate rV,, Fig, 1, fi^m a photograph taken by Profesaora Frost 
and Adams in 1903 with the great Bruce spectrograph of 
the Yerkes Observatory. Spaik-lines of titanium fuinished 
a comparison-spectrum ; and the emphatic nature of the coin- 
cidences gives an idea of the strength of absorption by that 
metal in the star's atmosphere. In the sun also, titanium 
plays a remarkable part. The lines due to it, although 
mostly faint in the Frauuhofer spectrum, come out with 
augmented intensity in the spectra of agitated spots ; and 
the copious photographic record of its emissions, especially aa 
obtained by Mr. Everahed during the Indian eclipse of 1898, 
assign tt a place among the permanent constituents of the 
solar chromosphere and prominences.' In the spots and 
eruptive appendages of Areturus, we may be reaaonably sure 
that it is still more plentiful and active. 

i ICftimder, The Indian ^i^pse, p, 70. 



46 



THE SYSTEM OF THE STABS 



No line of demarcation can be drawn between helium 
and Sirian stairfc on Ibe ont^ band, or between Suian and 
6oLar stars ou the other. And the presumption is strong that 
the tiHDBition from each class to the next is ejected bj actual 
deyelopment ; that the same object passes successively throngb 
the indicated changea The chief reason for hesitation in 
adopting this view is that the various stellar oiders are not 
indifferently scattered over the heavens ; but we understand 
very little of what such distinctiouB really imply. There ia, 
besides, some direct evidence that heiium stars are those moat 
recently condensed from nebulje. Some of them appear to be 
as yet not wholly detached from their cloudy matrices ; while 
Sirian and solar stars, although they may be projected casually 
upon a nebulous background, are probably never, in point of 
actual fact, nebuloua. Kor ig the eupposition admissible — aa 
the case stands at present — that helium stars can progress 
to the solar condition otherwise than by the prescribed route. 
The intermediate Sirian type must apparently be conformed 
to before the third station on the long journey comes within 
reach. Some authorities, it is true» believe that it may be 
dispensed with, and that a Cygni is an example of stars 
traversing this evolutionary short-cut ; ' but the better 
opinion is adverse to such a possibility. 

We may now briefly reconsider the course and kind of 
modification undergone by the stars^ so far as we have yet 
taken them into account In the first place, there is a steady 
increase of general absorption. The light of Bellatrix (7 On- 
onis), for instauce, attains outer space substantially aa it 
left the star's photosphere. Its native bluish tinge subeista; 
the proportionate intensities of its variously refrangible 
sections remain unaltered. But our sun's proper atmosphere, 
or *' emoke-veil," arrests a large percentage of his violet radia- 
tions. The solar disc is, indeed, quite brownish near the 
edges, because near the edges the light cuts obliquely through 
the reddening strata, and suffers in consequence heavier 
encroachments. Could they be stripped off. the sun would 
blaze with at least once and a half times its present bright- 
ness, and would show a steely lustre very different from the 
golden radiance we are accustomed to. There is hence no 

' Scb«<iD«r, P^tdam PuHieationent Bd. tII. TI1. ii. p. 331. 




SteJlur Sjiectrn nf three Tyi"'*. 

Vx, Siwi'trmu uf ArcturuH wiih TiUmiim Com]Mui»on-Si*ctt-Hin (Frost). 
S]i«tTiiTii urn? Ct'ti (rifkeriwg). 




THE STABS AS SUNS 



47 



warrant for aeaerting that the emissive outpourings of solar 
ort» differ intrinaicaU)- from those of Sirian or of helium 
fitaiB. It is certaiB ooly that they are strongly inodilied by 
elective abaorption which has no appreciable effect upon 
spectra of the earlier types. 

The signs of speciSc abfiorption» as helium grow into solar 
stars, alter by insensible gradations, but according to a 
viaiblj conaistent method. From beginning to end qf the 
series there 19 a steady increase in the number and intensity 
of metallic lines, very notably in those of iron, whichj next to 
the giant-bands of caJcimn^ constitute the dominating feature of 
the solar spectrum, yet are imperceptible in the prismatic light 
of Bigel and BeUatrbc. Absorption by helium and oxygen ^ 
comes in at the start, but has a strictly limited time-range. 
It scarcely brooks competition, sharing the field only with 
hydrogen in its terrestrial and cosmic forms ; metals, in the 
earliest stars, showing only by a faint streak of ma^eaium. 
As they begin more numerously to assert their presence, 
helium lines become effaced, and those of hydrogen attain 
their mwdmiun development. But the Huggine-series is now 
alone repreaented, for the Pickering set is even shorter-lived 
than the helium rays, and it appears exclusively in what are 
supposed to be the earliest stars. This is an extraordinary 
circumatance, most battling to comprehension. There is no 
other known instance of such independent behaviour on the 
part of a spectral aeries. Nor can its meanings in the total 

I absence of terrestrial experience, be interpreted by so much 
j^ a plausible conjecture. If the Pickering lines could be 
feduced to show in vacuum-tubes some basis would he supplied 
pbr reasonings on the subject ; but attempts to elicit them 
have, so far, proved altogether fruitless. It is also important 
' to remember that, in the stellar succeasion, the character of 
I metallic absorption changes as well as its strength. The few 
traits of it which appear compatible with the display of 
I helium are — in Sir Norman Lockyer'e phrase^'* enhanced '^ 
lines ; lines which vivify notably when for the illuminative 
agency of the electric arc a disruptive spark -discharge ifl 

' FroHt ojid Adaina siupecl that ozygen-Btart may be trrayernDg a road 
pamUel to the maia tr&ck of evolution. FabL l/niversity of Chicago, vol. riiL 

^1(I7. 



CHAPTER IV 



THE CHEMISTRY OF EED 8TAKS 



Kbarlt aU Stars of a pronounced red colour show Bpectra 
crossed by dusky banda of absorptioii. Two varieties can be 
at once dificriiainated. They respectively form Secchi's third 
aad foarfch types. Kow epecti-a consisting of bands or flut- 
inga are given out in the laboratoiy by coinpoimd eubatances, 
tsach as oxides and chlorides, as well as by the chemical 
elemente glowing at a low pitch of excitement. They were 
formerly thought to originate from molecular, line-spectra from 
Atomic vibrations. But this opinion has been much shaken 
by recent experience/ Trowbridge and Eiehardg concluded 
in 1897* that, where electricity is the illuminating ^nt, 
an oecillatory discharge gives rise to iaolated rays, a " dead- 
bd&t'' discharge to flutings. And it is easily understood that 
Buccesffive shocks of transmitted energy should evoke more 
intimate thrillings than an even flow. Electrical stimulation 
of any kind is not, however, necessary for the production of 
banded spectra ; the cool oxygen of our own atmosphere 
stamps the red section of dispersed sunlight with three 
conspicuously dark flutings. Moreover, a profound difference 
in the relationships of the two kinds of spectra is signihed by 
the insensibility of fluted emissions to magnetic influence. 
They evince no trace of a " Zeeman effect " ; they remain 
unaltered and imperturbable in the field of the most power- 
ful magnet Pressure is similarly impotent to modify their 
wave-lengths or structure. Tliey are, however, displaced by 



1 B«8lukdroa, CvmpUt Jiendus, t. crxxrii. it. 1013, Deo. 14. 1903. 
* Anuriean Journal of Seimet, vot. iii. p. 117. 

4ft 4 




50 



THE SYSTEM OF THE STARS 



radial motiou in the same maiiDer, Aod to the same extent, as 
linear emiaaiouB. 

FititiDgs are made up of uumerous iodividual lin^ set 
more and raore cloeely together towards the " head," and at 
continually widening iutervaU towards the '* taiL" They in 
fact constitute series, on the model of tlioee of hydrogen and 
helium, but: condensed into a much smaller space. The 
numerical formuljs repreaentiug the principle of their con- 
BtruL-tion have, since 1885, been investigated with a fair 
promise of success, first by M, Deslandrea/ later by M. Thiele.' 
Enough has been done to show that the arrangement both of 
bands in the same spectrum, and of lines in the same band, 
is methodical, and may eventually be rendered intelligible ; 
but the underlying lawa are too intricate for immediate and 
full recognition. 

In the stars, fluting-absorption is 3uj>eradded to absorption 
by lines. Metallic groovinga, agreeing in position with those 
of the solar spectrum, but somewhat reinforced, aa a rule, in 
strength, aid the interception of photospheric radiationa 
dimmed further by a " smoky " envelope chiefly arrestiTe of 
the shorter wave-lengths. Thus, three distinct forms of 
jibaorption mark their effects in banded spectra, and seem to 
betray the action of deep and dense vaporous strata. Hence 
the redness of the stars. And, because their photographic 
rays are largely cut off, they come out on sensitive plates five 
or six times fainter than they appear to the eya Their 
spectra, too, ate recorded with much difficulty. In some 
nearly all the violet^ and all the ultra-violet, light is extiu- 
guiflhed ; while their blue beams are enfeebled by crossing a 
triple barrier of absorption. For purposes of effectual investi- 
gation, accordingly, orthochromatio plates have to be em- 
ployed ; and these are not free from inconveniences and 
drawbacks. 

From Autares {a Scorpii), a remarkably fine specimen of 
the third stellar type, Sir Norman Lockyer has conveniently 
designated its members " Antarian stare," The fiutings in. 
their spectra terminate abruptly towards the violet, but shade 

1 Compfes J^cridu$, tc. p. 1256; ibid. Oct 17> IWl ; Eaywr, Bandbueh 
dir Spectroscopie, Bd. ii. p. 175. 

^ Aitrc^h. Jounu^t toI. tl p. 65, vliL p. 1. 



TJtE CHEMISTEY OF BED STAHS 



51 



if pida&Hy towards the red, pnxluciag to the eje something 
^1 colonnaded effect. This, however^ is loat in photographs ; 
puti/ becaxise the flutiugs are discontinued in the higher 
^eccnil ranges, partly hecause the general impression is over- 
le by the wealth of self-recorded details. All spectra of 
th^ kind are constructed on the game fundamental design. 
princip^ bands, of which ten are counted, reappear with 
tial invariability in every Antarian star.' A homogeneous 
is thti9 suggested for them. It is unlikely thut many 
each acting independently of the others, are con- 
in the weaving of a pattern widely diffused, and 
pxaclically unchanging. Yet their chemical interpretation 
long remained a source of perplexity. Thii-d-type absorption 
^ no known analogue until Mr. Fowler, in 1904,^ thought 
^ comparing it with the Sutinga emitted by titanium oxide, 
tntdeied luminous by low-tension electricity. They were found 
to coincide very approximately with eight out of the ten star- 
baods; and particular resemblances in the arriiogeraent of 
kaea within the tlutinga emphaaiaed the general agreement 
Stellar spectrsi of the third type thus promise iinally to yield 
Dp their setsrot ; for it must be, at least provisionally, admitted 
ihdt titenium playe a leading part in giving them their 
channelled a8j)ect. Cyanogen flutinga, too, have been photo- 
graphed in these stars by Prafe-'ssor Hale and Mr. EUerman ; ^ 
1 they are of shorter wave-lengths than any of the bands 
ily measured They constitute a novel featui'e of 
.on interest, as the only symptom so far recognised of 
,ce of carbon, or its compounds, in third-type stars. 
Aldebaran, although a solar star, ia unmistakably of a 
reddish tinge. The blue end of its spectrum is heavily 
ptMCured ; and the tiutiugs chariictmstic of the type towards 
which it aeems to be advancing are just indicated as embryonic 
I^MMlings. The object might be called a Unking instance 
between Arcturua and Betelgeux (a Orionis), in which titanium 
latingB (if we may bo designate them) are fully developed, 
irhilo the main features of the Franuhofer spectrum remain 
and uneffaoed. The least possible departure is made 





52 



THE SYSTEM OF THE STARS 



from the aolar model that is recoadlable with the complete 
presentment of a different type. Bj'oomparing Plate II. » Fig. 
4j with Plate III., Fig. 2, our readers can estimate the close- 
nenB of the analogy between the linear elements in the spectn 
of CapeUa and Betelgeux. The four lo^'^st lines of hydrogen 
are ikirly stroag in Betelgeux ; but in specimens presumed to 
be more "advanced" they thin out, and perhaps disappeftr 
with the dosing in of denise banded absorption. A fine led 
Btar in the Southern Cross, y Crucis, ia of nearly the same 
standing as Betelgeux ; in a H^ercuUs the bands have acquired 
strength through the efflux of time, it Is supposed, and the 
progress of cooling. 

Eight or nine metals are easity reoognidable by their 
abaorption-liues in Antariau spectra. Calcium is the most 
prominent. The notable pair of difluse stri^ies in the violet 
due to it are fully the equals of their archetypes in the sun ; 
while the llame-line iu the blue, X 4227, shown as a black 
bar bisecting the spectrum in Plate IV., Fig. 2, is of greatly 
augmented intensity. Iron comes next to calcium in elective- 
neas for ligbt-stoppage; and with it are associated magnesium, 
sodium, chromium, titanium, vanadium, aluminium, strontiiun, 
and probably manganese. No lines distinctive of the electric 
spark are present; but those widened in sun-spots stand out 
significantly. 

Red stars, as we have said^ are encompassed with power- 
fully absorptive atmospheres ; and atmospheric density and 
extent are, for some recondite reason, accompanied by in- 
stability in shining. The radiative machinery tends to become 
clogged fitfully, or at definite intervals^ and irregular or 
periodical variability results. Nor do the spectra of Antarian 
stars remain imafTected by these vicissitudes. Their con 
tinuous radiance brighteuB and fades unequally in its differ- 
ently refrangible sections, though according to no traceable 
method; the bauds interrupting it alternately close in and 
thin off; above all, vivid rays of hydrogen and other eub- 
stances are kindled with the recurrence of the brilUant phases, 
and die out as each fresh outburst of energy becomes exhausted. 
This striking feature was first detected in a spectrogram of 
Mira Geti taken at Harvard College in 18S6; and scores of 
" long-period variables " have since, through Mrs. Fleming's 



THE CHEMISTKY OF RED STAKS 



53 



been recognised by their exbibition of the Bame 
peculiantj. Only two of them, however, have as yet been at 
all ftdeqoately studied. Mira has been made the subject of 
«OQie admirable mveatigations at the liek Observatory; unil 
^ Cygni has been dealt with to good purpose at Potadam 
dnring some of its recent ma^dma. 

The character of the flpectrum of Mira is well ahawn in 
Plate IV., Fig, 2, reproduced, by the kind permission of 
Profe^eor Pickering, from a Harvard photograph. The two 
Uae hydrogen-raya (Hy and H£) emerge resplendent ; yet the 
gxeen and violet linea (H^ and Hs) are alike imperceptible ] 
while four of the ultra-violet sequence shine brilliantly. 
Professor Campbell ascertained by direct observations with 
the great Lick refractor at the high majcinium of 1898, that 
the crim;&ou hydrogen-line makes no show in this star.^ 
Kevertheleas, in vacuum-tubes, in the solar chromospliere, and 
in bright-line helium stars, C is the most vivid lueuiber of 
the series. In nebuhe, on the other hand, it is usually 
invidble ; and thus the unexpected conclusion is forced upon 
us that hydrc^en glows in the great irregular nebulffi under 
similar conditions of excitement to those prevailing in variable 
Antarian stars. The suppression of the fifth line (He) is 
universal in anch objects, and was believed to be complete 
until the missing radiation faintly recorded itself on some of 
the spectrograms of Mira lately secured by Messrs. Wright 
and Stebbins at Liek. It seemed to glimmer through a dense 
Uyer of calcium, the nearly coincident absorption of which in 
g^eral masks it effectually. This explanation, however, needs 
to be further verified. It has much to recommend it ; yet it 
inTolvee an inversion of the order of stratification, as regards 
amitting and absorbing vapours, commonly observed in stars. 

The tripling of the hydrogen -Hnea in Mira, noted by 
ProfeaaoT Campbell in October 189S, is highly suggestive of 
niftgnetic action. But the effect, though eagerly looked for, 
has not since recurred, and appears to betoken exceptional 
agitation. On the sain© occasion, two iron-liuea, previously 
ngistered as dark, came out conspicuously bright; and they 
were again brightened at the maximum of 1902, together 
with a number of others, including the principal arc-line 
^ Attroph, Jmtm. toL is. p. SO. 



54 



THE SYSTEM OF THE STABS 



of silicon at X 3906. Mr Joel Stebbins^ followed the atar 
spectrograph ically down to a nmuiimm at ninth magnitude 
in January 1903. when a six hours' exposure barely sufficed 
to procure a legible, though not a measurable impreasioa from 
rays shorn of the adventitiovis glory lent to them by transitory 
gaaeous incandescence. The phenomena attending the brilliaiit 
phases of x Cygni, as observed by M. Eberhard,^ are essentially 
aimilar to those displayed by Mira. 

About 15 per cent of third-type stars are establiahed 
variahlea ; and few shine with the constancy of an average 
white star. Moreover, those subject to the widest vicissitudes 
have, on the whole, spectra the least like that of the sun. 
They seem to have descended furthest along an inclined plane 
of change, for there is no evidence of abrupt departures from 
what was presumably the more primitive type. The dis- 
tinctions between our sun and a Mira-variahle« though 
strongly pronounced, may have been brought about by in- 
sensible gradations ; and the inference invites, if it do not 
command our assent that no halt can be cried in the evolu- 
tiouary journey from a stable condition of sun-like luminosity 
to the stage of turbulent incandescence marked by the ruddy 
flare of recurrent maxima. 

The apeetra of fourth-type, or "carbon -stars/' resemble 
those of the Antarian kind in being marked by linear, as well 
as by banded absorption. Their aspect to the eye is illustrated 
in Plate IV., Fig. 3, copied from M. Dun^r'a drawings by 
permiaaiou of the University Press of Chicago. From the first, 
Father Secchi recognised the carbonaceous origin of the three 
deep shadings, distinguished in the Figure as B6, B9, and BIO. 
They face redward (to the right), and match respectively the 
yellow, green, and blue flutinga of the so-called "Swan" 
spectirum, derived, according to some authorities, from pure 
carbon — according to others, from carbon monoxide. But 
these stars are, without exoeptiou, faint ; the brightest is of 5*3 
magnitude ; and beyond the preseuce of some strong lines 
due to iron and sodium, little further could be learned about 
them until more powerful optical means became available. 
The completion of the Yetkes 40 -inch refractor at last 

* Aatropk. Joum. vol. Kviii. p. 341. 
* IhU. vol. itvUi. p, 20a. 



THE CHEMISTBY OF BED STARS 



55 



supplied tbe needed facilities; and in 1893 Professor Hale» 
aasisted by Mr. EUerman, undertook the spectrograph ic in- 
vestigation of fourth-type stars. Their overwhelming aelf- 
AbeorptioD in tha blue made the task seem unpromising ; but 
by the employment of isochromatic plates of different quality 
for the several spectral aectious, together with exposures up 
to twenty-four hours in duration, it was successfully executed. 
The definite results, published in 1903, are of the utmost 
Talae.^ They refer more particularly to eight stars, the 
tions and places of which are as follows :- — 



Nuns. 


B.A. (IMUX 


OaaQWO^ 


»«. 


74 8elu«IlBnip 

75 BnlueUeTap 
llA Schlellenip 
132SchjeUBnip(=UHrdi») . 
3IS BinDJDgham 

152 SclijeUenip 
Id Fiscitim .... 
280 Schjellerup 


8b]9Bt4B* 

a 29 40 
8 49 4fi 
10 32 36 
10 38 a 
12 24 2e 
23 41 17 
'2S ee 10 


+ 14' 47' 
+ 38 32 
+ 17 37 
-12 62 
+ 67 68 
+ 45 59 

+ 2 &d 
+SB 4S 


6 -a 

«-5 

«-6 (vw.) 

6-4 

5-8 (v*r.) 
7-6 



Of the above, six are known by their numbers in a 
CaifUogu* of lUd Stats, drawn up by the Danish astronomer 
Schjellemp in 1866 ; one ia taken from a similar work by 
John Birmingham of Tuom, dated ten years later; and the 
eighth and brightest was enrolled by Flamsteed. On the 
Yerkea plates of their spectra no leas than 307 dark lines 
were measured, and in part traced to their origins. They 
indicated the undoubted action of ten substances, namely, 
hydrogen, calcium, titanium, vanadium, iron, aodiumj mog- 
lieaiuin, chromium, nickel, and manganese, and suggested that 
of others. The record of cyanogen-flutings in the bine (see 
Plate V,) constituted in itself an important gain to know- 
lodge; and still more the discovery that bright lines are 
profusely scattered among the dark elements of such spectra. 
This bad been to some extent perceived by Father Secchi ; 
but his observations were only rescued from the discredit into 
which they had fallen by tbe sure testimony of the camera, 
Lt present some two hundred rays of emission have been 

^ DecenniAl Publieaiioru of the Univernty qf Chicago, vol. riii.^ fVom which 
PlfeU V. ifl Tcproduoed^ by the courtes^y of the ITuiventty Press AuthDri'CioH. 



56 



THE SYSTEM OF THE STARS 



determined ia carbon- stars, althotigh one only h&& been 
eecurely identified. This is the bydrogen-F, which at timea 
(for itB condition is subject to change) shines conspieuously in 
the peculiar star 280 Schjellerup (see Plate V., uppermoat 
apectrum), and more diinly in a few other members of the 
class; while, in 19 Pisciura, the blue and indigo hydrogen- 
lines are dark, and no trace of F, bright or dark, is to be 
found. Of the remaining crowd of vivified rays, some may 
be due to oxygen and nitrogen, one coincides with the funda- 
mental member of the Pickering-eeriea of hydrogen, and three 
or four besides fall very near emiesion-rays in gaseons Btars of 
the " Wolf-Rayet " variety. On the other hand, the absorption- 
spectra of carbon-etara are extensively oompo^ of lines widened 
in Hun-apots ; the metallic spark-lineB characteristic of early stars 
have vanished from them, and been replaced by arc and flame- 
lines> held to be an unfailing symptom of advanced stellar age ; 
and it is associated with variability in light even more pro- 
nounced than that shown by Antarian stars. Professor Hale's 
general conclusion is that third- and fourth -type stars "should 
be elaeaad together as co-ordinate brunches leading back to 
stars like the sun," from which they probably develop through 
loss of heat by radiation.^ That is to say, an outworn solar 
atar may descend to extinction by either of theee alternative 
downward paths. But why one is chosen rather than the 
other remains obscure. 

Two circumstances are, however, adverse to the view that 
suna like our own are liable to degrade into carbon stars. 
The 6rst is the lack of intermediate specimens. Profeesor 
Hale urges the plea often put forward by embarrassed palae- 
ontologists, that the record is incomplete ; and it may well 
prove valid. He has already picked iip some stray connecting 
links, and it would be premature indeed to despair of more 
being found. But there is another very serious difficulty. 
It relates to the distribution of the stars. Those of the 
fourth type are strongly condensed towards the Milky Way.'' 
Their relations to the fundamental plane of the sidereal 
system are quite different from those of solar stars. More- 

' Decennial Pi^lusaticns, vol. via jx. 136. 

' DuHtr, ilpteiret de la Troitiem^ CioiU, p. 126 ; Porkturet, Aatn^h. Joum, 
tqL ix. p. 230 : Espm, ibid. toI. x. p. 169. 



THE CHEM^STBY OF KED STAES 



57 



they &P6 evidently extremely remote. They eeem to 
occupy a region of space apart from that tenanted by our sun 
and his congeners. The incongruity is then obvious of sup- 
posing that they are united with them in a single evolutionary 
verier. 

About 250 carbon etara have, so far, been registered. 
Many were detecteil visually by Mr. Espin ; but Mrs. 
Flemings photographic gleanings have contributed largely 
lo swell the total Seven out of the 250 can just he seen 
with the naked eye ; the rest are telescopic objects. Oue 
of the finest specimens is Secchi's " La Superba " (152 
Schjellerup)^ situAted in Canes Venatici. It was so entitled 
for the extraordinary vivacity of its priflmatic rays, separated 
into dazzling " zones," red, yellow, and green, by broad spaces 
of profound obscurity. This Btar ahinea with approximate 
constancy; 19 Piscium at times surpasses it in brightness, 
but is subject to temporary deprivations of more than one- 
half its light. These happen irregularly ; many periodical vari- 
ables, however, belong to the same spectral type, among them 
XJ and V Cygui, S Cephei, V Hydrje, and K Leporis, noticed 
by Hind for its intense crimson colour. 

The comparative dimness of all such objects, though 
doubtless in part due to their great distance from the earth, 
is readily explicable by the opacity of their atmospheres. 
How much of their intrinsic light escapes absoiption can 
only be conjectured ; if we put it at a tenth, our estimate 
is likely to exceed rather than fall short of the trutli. 
Only loopholes, so to speak, are left for its exit; and, as 
usual, the violet and blue rays suffer most severely, being 
indeed almost completely smothered. The special depth of 
colour in carbon stars is thus accounted for. They are not 
merely suffused with red, like Betelgeux and Antarea ; they 
actually simulate the glow of carbuncles or rubies in the field 
of the teleacope. Vast and magnificent orbs must be found 
among themL Under circumstances less disadvantageous^ 
some might well take rank in the second magnitude of Siriau 
»nd solar stars ; and this apart &om any allowance for their 
apparently exceptional remoteness. The final darkening and 
death of fourth-type stars may then be reckoned as a con- 
tingency too far off for definite realisation. 



CHAPTER V 



GASEOUS STARS AND NEBULA 



Gaseous stars form a restricted aud peculiar class. The 
objects belonging tu it are characterised by the display in 
their spectra of isolated bright and dark lines on a mom or 
lesa perfectly continuous background. They present \m then 
with a triple combination — a direct gaseous spectrum, r 
reversed gaseous spectrum, and a spectrum due to glowing 
solid or liquid matter, all simultaneously made manifest by 
the unrolling of a single scroll, yet each originating under 
very different conditions. The true discrimination of those 
conditions tasks all the resources of physical eidereal 
astronomy. 

The state of bright emission is in 8om.& stars normal, in 
others it only supervenes as part of a great general increase 
of light. It ia not limited to any one period of sidereal 
existence. Stars barely condenaed from nebulae sometimea 
show bright lines, and the symptom is (as we have seen) apt 
to recur at the advanced stage reached by red stars with 
banded spectra. Only Sirian and solar stars are exempt 
from the tendency ; helium stara are peculiarly liable to it. 
It would thus seem that tlie stars capable of yielding mixed 
spectra of absorption and emission are those either unfinished 
or verging towards decay, while, during the iutermediate 
epochs of vigorous maturity, absorption is alone effective. 
The term " gaseous," however, should be reserved for inchoate 
orbs. Those giving bright lines are probably distended 
far beyond the solar proportion of size to mass ; and their 
spectra bear the distinctive marks of helium, frequently of 
cosmic hydrogen, and not rarely of oxygeu and aitrog:en. 



fid 



d 



GASEOUS STABS AND NEBULA 



59 



They have the farther diBCinction of showing the oidinary 

lirdfogen -series — the Huggins-series — in duplicate, a bright 
Une being coupled with, or Buperposed upon, a dark one of 
identical origin ; and the rule first adverted to by Professor 
Campbell ^ is invariable that, under these circumBtances, 
ibaorption becomes atraager, and emission feebler with 
dimiQutiou of wave-length. The brightest lineSj^ in other 
worda, are the least refrangible. Alcyone in the Pleiades is 
aa example. In this Btar one solitary ray is vivid, and it ia 
the fundamental (Ha), which glows cnjuBon beeide a dark 
CDmpatiioo. Plelone in the same group, on the other hand, 
and many stars besides, exhibit the direct radiance of several 
members of the hydrogen-aeries ; but that of C always pre- 
dominates. Similarly in the various helium-series, the stress 
of brightness is inevitably laid upon the lowest of their 
constituent lines. But to make this apparent, the fieries 
most be sorted out ; for the rule applies to each individually. 
An analogous order of progression was subsequently found by 
Professor Hale to regulate the reversal of lines in metallic 
spectra,^ As the vapours rendered incandescent by the 
electric spark taken under water became more and more 
self- absorptive, he perceived that the lines progressively 
darkened descended the scale of refrangibility with approxi- 
mate steadiness. That is to say, the rays most persistently 
bright were those of greatest wave-lengths. Here, undoubtedly, 
we hold a clue to the physical state of emissive stars, but 
one, unfortunately, by no means easy to follow up. 

The first specimen of a gaseous star was made known by 
Father Secchi's discovery, August 19, 1866, of the green line 
(F) of hydrogen conspicuously bright in 7 Caasiopeias," the 
middle star of five of the second and third magnitudes 
grouped into the shape of a W 011 the opposite aide of the 
pole from the Great Bear. Soon afterwards the same peculi- 
arity revealed it&elf in ff Lyrae, the emission being discerned 
in it as well as in 7 Cassiopeiae of three rays of hydrogen 
and one of helium. But these do not always present the 
same appearance. As early aa 1872 Vogel was struck with 

^ A*troph. Jowmalt toI. ii. p. ISl, 
» Thid. vol rv. p. 527. 
' Sugli Spettri I^ntmatici dclU UtetU Fisse^ H«m. i. p. ](], Mem. ii. p. 32. 



60 



THE SYSTEM OF THE STARS 



tbeir apparently imaecouii table capricea of visibility ; ' and 
M. von Gothttid watcheJ vainly during two years before he 
caught sight iu 7 Cassiopeiie of the crimson twinkling of 
C in paTdcnlariy unfuvonrabU weatkert August 13* 1883.* 
His subsequent obaervatione, and those of M. von Konkoly* 
fully established the occurrence, in both stars, of remarkable 
spectral fluctuations, which have eince been studied, with 
attentive curiosity, in widely separated parts of the world. 

The result has been to establish their reality, but strongly 
to disGriminate between the modes of their occurrence in the 
two objects. They are periodi*ail in /3 Lyrse, and their 
period is that of the star's light-changa The flickering of 
bright lines in its spectrum, accordingly, depends in some way 
upon the rise and decline of its visual luminosity.* But the 
white radiance of y Cassiopeipe, although including variable 
elements, is^ in its sum-total, perfectly steady. Since lh6 
advent of spectrography, fluctuations in the hydrc^en-linea 
have ceased to be apparent ; but then C, ivhich IJea below the 
range of the sensitive plate, has scarcely of late been looked 
for. And C was precisely the line most noted, in the days of 
viaual observation, for capricious shining. This property is 
unmistakable in the yellow helium-ray (B^); the sodium- 
flbaoi-ption near it is ;ilso apt to become effaced ; and the 
green and blue magnesium lines appear, with unaccountable 
alternations, bright and obscure^ Kor is it at present known 
whether these sundry changes proceed in concert, or each on 
its own account. 

The speetrum of fi Gentauri, which closely resembles that 
of 7 CaaaiopeiBB, is represented, from a photop'aph taken at 
Arequipa, February 8, 1897, in Plate III,, Fig. 3- The 
brilliant line to the right is H)3 (¥) ; but the dusky edgeSj 
faintly perceived to fringe it on the original negative, are 
swallowed up, in the print, by the general gloom due to the 
idiosyncraay of the plate. The luBtre of H7 is, however, 
strongly relieved by corresponding absorption, and HS dis- 
plays merely a thread of light dividing a profoundly dark 
band. Two clean cut, black lines to the right of H7 

' ^Qtfikamp Btobachiunoen, Heft iL p. 29. ^ Attr. Nach. Noa, 2&S1, 268fi. 
' Obseroalor'^, vol, vi. p. S^2 ; O Gyalla Beobachfungtny toI, viii. p. &. 
* See Chapter XXl. of FrobUma in Aflropkyria Ity the pJesBtit wTitflf. 



GASKOUS STABS AND NEBULA 



61 



ta from helium ; and there is room for farther research 
ui the ideutificiition of many additional rays, hotli bright 
aad dark, in this beautiful apectrum.* ProfesBor Campbell, 
in 1S94, observed in it the crlmaou shining of C. No 
Bospicion of variability attaches to the star, 

A etar-disc is rendered through distance immeasurably 
Bmall. It has no longer diversified parta The light from 
aU is blended indiscriminately into the radiance of a single 
point. Hence, Dr. Scheiner argued, the mere increase of 
atmospheric extent, apart from any increase of atmospheric 
enuBsive intenaity, would give overweening effectiveness to the 
bright-line ingredient/ an insignificant proportion of which 
ifl present in the luminous aheaf sent out by ordinary stars. 
Experience has, however, virtually negatived this view. The 
phenomena of spectral variability are entirely irreconcilable 
with it. And, in one form or another^ they prove to bo 
widely diffused. But the absurdity is patent of invoking 
fluctuations in atmospheric compass to account for them. 
de gaaeous surroundings of stars may conceivably bo subject 
to changes of luminosity ; but the hypothesis of their ex- 
pansion and contraction with the rapidity, and to the vast 
extent required, is plainly inadmissible. Moreover, the 
diverdty of conditions under which bright Unea are perceived 
to originate, their complex reversals, their peculiar individu- 
alities, intimate the working of profound physical agencies, 
and repudiate any single formula of explanation incapable of 

t adaptation to the lavish variety of nature, 
I Dr, Schuster proposed, in 1903, a theory in some respects 
BPeferable to Scheiner's. Both speculators alike sought to 
Irade the direct consequence of Kirchhoff' s Law, according to 
which, " a layer of gas in front of a radiating surface can only 
give bright lines if its temperature be higher than that of the 
radiating surface " ; ^ a state of things not easily to be supposed 
esdiBtent in stellar or solar atmospheres, A new expedient 
was accordingly resorted to by Dr. Schuster. He showed 
that bright lines might he produced by the dispersion of light 
in layers of incandesoant vapours overlaying a hotter radi- 

1 Cf. Loolcyar and Bkiendali, Month. Notices, April 1£K)&, App, No. 2. 

^ Scbeiner, Di« Spiclralana-iyse d4r GtsHrnt, p. 270. 

• OhterDtttery, vol. ixvL p. 379* 



e2 



THE SYSTEM OF THE STARS 



ating surface, His roaBoniiig ia no doubt correct, aod we 
have here a " true cause," although scarcely one that is 
sufficient for the part aasigued to it. Its inadequacy becomes 
patent when we consider that ita action ia at a minimum 
just where the eflects attributed to it are most conspicuoas. 
The scattering power of pure gaaea must be very small 
compared with that of the vapours, say* of carbon, ailicon, or 
titaniuoL Yet the bright lines in stellar spectra are chiefly 
those of the tenuous and transtueent gases, hydro^u and 
helium. The rationale is thus unconvincing. 

Another, designed expreasly to meet the case of temporary 
stars, has been advanced l^ M, Ebert.' It is based upon the 
principle of "anomalous refraction," by which light subtracted, 
through a special action of the refracting medium, from one 
narrow section of the spectrum, goes to reinforce the adjacent 
Strip, a show of paired bright and dark lines appearing as the 
reetilt. But here, again, the actual phenomena seem recalci- 
trant. Hydrogen and helium are among the least refractive of 
substaQces. They are eminently unlikely to act on transmitted 
radiance in the forcible manner demanded by M. Ebert's theory. 

The stellar cLlss, of which 7 Caaaiopeiae is typical, haa 
now upwards of fifty known representatives, most of them 
having been detected by Mrs. Fleming on the Harvard 
epeotiographic plates. Among them are the noted ex- 
v&riahles, at present shining with an uniform, unpretentious 
lustre^ 'rj Garinaa and P Cygni. The spectrum of the former 
object was found by Sir David Gill to bear a strong re- 
semblance to that of Nova Aurigoe ; ' and P Cygni is 
similarly characterised. The pemmnence in both of spectral 
relations transient in new stars has an import which can 
hardly be over-emphasised. 

The so-called Wolf-Rayet species of gaseous stars, although 
cognate with the helium variety, stands well apart from 
them. Hydrogen and helium rays are» in their spectra, of 
Bubardiuate importance. They are mainly distinguished by 
certain vivid bands in the yellow and blue, some of them 
due to cosmic hydrogen, others of untraced origin. 



> Aar. Jfach. No. 3917. 
^ Monthly I^otw$, vol. Ixi. p. AL6. 
Atmali, vol, xxviii. part ii. pv llb^ 



See alfto A. J> CumoD, Harvard 



GASEOUS STAICS AND NEBUL.E 



63 



The finest epecunea of the class is invisible in them 
Utitxbdea. Jirat detected by Kespigbi at Madras^ December 
24, 1871/ the pecaliarities in the light of y VelonuD were 
studied with fiome caxe and much delight at the " extmordiuary 
beauty " of the spectacle they present, by Dr. Copeland at 
Puno in the Andes, April 24, 1883.' "An intensely bright 
line in the blue" he remarked, " and the gorgeous group of 
tbT«« bright lines in the yellow and orange, render the apec- 
tram " of this star " incomparably the most brilliant and 
Btiiking in the whole heayens." There ia no eign that it is 
in any degree variable. Its appearance to the present writer, 
at the Cape, in October 18S8, tallied precisely with Dr. 
CopeUnd'e description, only that the additional feature of a 
deep band of absorption below the cobalt line seemed unmis- 
takftble.' A vivid continuous spectrum extends into the 
violet as far aa the eye has power to follow it, and accounts 
for the brilliant whiteness of the star. The diffiiseneae of 
radiation in this splendid object is an obstacle to its effective 
spectrographic treatment. A fair idea of its character can, 
however^be gathered from Plate II L, Fig. 4j which reproduces, 
by the courtesy of Professor Pickering, a photograph taken 
at Ajrequipa with 99 minutes' exposure, April 28, 1897. 
It does not extend to the lower tract illuminated by the 
golden and citron rajs. The spectrum begins on the right 
with the double or triple a2ure bands distinctive of this 
Bfcellar type. The less refrangible (X 4688) occupies a 
remarkable position. It is that theoretically assigned by 
Bydberg in 1897 to the one member accessible to view of the 
" principal aeries " of hydrogen. By analogy it should 'be 
there, and it ia This azure band seems to be an unfailing 
adjimct to the Pickering series, which it imitates aa well by 
refusing to be evoked in the laboratory. Professor Campbell 
was surprised, in 1893, to find C briUiant in the spectrum 
of 7 Velorum, the other hydrogen lines being dark.* They 
are nevertheless perceived photographically as projected on 
bright bands, an arrangement the inverse of that prevalent in 



' Oe-mpbit RmdvSf t Ixxiv, p. 610. 
* Vopemitnu^ vol, iii. p, 206. 
' Obtervatory^ vol. n, p, 430, 
' Pablieatiena Jstr. Fete. Soddy, rol. v. p. 106. 




THE SYSTEM OF THE STABS 

gaseouB stars of the helium kind^ Abeorptiott by o; 
WBA detected by Mr. McCleau amid the emissive spleudoius 
the southern luminary; metallic action is probibly a1 
throughout the class which it exemplifies. 

ItB original members, three small stars in Oygiius, havi 
in the course of thirty-eight years, obtained about one hundiw 
aaaociate-s, lurgi^ly through the extensive photographic researchfl 
of Mrs. Fleming. All theee are situated in the thivii 
of the Milky "Way, or of the Magellanie Clouda ; and ol 
except 7 Velorum, are insignificant when casually regarda 
None are known to varj' iu light; in none» so far, has any epectn 
clmnge be-en recorded. Indi^'idual diversities, however, abouitf 
Lines dark in one specimen may be bright in another ; band 
are mote or less hary ; the blue effluences differ from atar U 
stftT in relative strength ; hydrogen and helium lines are nO' 
piximinent, then again of quite subordinate importance. O; 
8uch star was discovered by Campbell to possess a 
hydrogen-envelope, indicated as existent only by the eenaibh 
length of the green ray originating from its glow. Perhapi 
some of its fellows have similar, if lesa conspicuous appurten- 
ancea ; though none of them are, in the proper sense, nebuloufl^ 
aiuue they give sharp images, teleacopically and photo 
graphically. 

Yet^ in the Wolf-Rayet stars, we have imdoubtedly reacheJ 
the borderland between the two great sidereal kingdonug 
where definitions cease to be valid, distinctions become inJ 
secure, and nondescript characters perplex classificationJ 
Between gaseous stars and stellar nebnlre there is but m 
narrow gap. Individuals of both species present the teleacopia 
Appearance of small star? ; and they discloBe, when a prisnt 
is applied, analogous peculiarities — analogous, not identicaL 
The light of a gaseous star so examined is ordinarily concen* 
trated in two points, or where a cylindrical lens is employed 
in two short lines, yellow and blue respectively; that of il 
ftteUar nebula gathers into one green knot. Its rays are, in a 
MOUse, incapable of analysis; they are so nearly monochromatio 
Ihul tliey can be refracted without being dispersed. Objects 
^'1' this nature can be picked out at a glance from ordinary 
*l'iil"n by Professor Pickering's method of "sweeping" with 4 

k^ A. J. C&Dnoit, Harvard Annait, voL xxviiL p. 247. 
^ i 



GASEOUS STARS AND NEBULA 



65 



tall direct-vision spectroecope used as an eje-piece.^ AI>ove 
flcote of them were in this m&nner found previously bo 
t90. and many more anbeeqnently by photDgraphic means; 
d it is remarkable that the exclusive prel'erence for the 
ilky Way of gaseous stars is shared by stellar nebulae. 

The green ray of the latter is the characteristic token 
gaseous nebulae. From the great Orion *' portent '* to the 
inteet "planetary," all without exception ahow it; and in 
any it appears at first sight ao pretlominant as virtually to 
and alone. But its origin remains an enigma. In position 
is almost coincident with an important line of nitrogen. A 
ifling divergence, however, shows them to be certainly dia- 
neL Nor can it be identideil, as Sir Iforman Lockyer 
loaght it might, with the sharp edge of a fluting emitted by 
agneslnm burning at about the temperature of the Bunseu 
ime.* The supposed agreement was made the corner-stone 
the meteoritic hypothesis of nebular constitution;* but it 
iled to obtain ratiticatlon by precise inquiry. 

Careful measurements by Sir William and Lady Huggins,* 
d by the late Professor Keeler, showed the fundamental 
steoric and nebular lines to stand very slightly apart in the 
Bctrum. But epectioecopic agreements must be absolute if 
ey are to be reckoned aignificanL Moreover, the nebular 
le is sharp ; and even the thinnest remnant of a luting 
ould have an attachment of unilateral haze. 

KebuJar radiance cannot then, it seems, he imitated in the 
jQiatory. PoBsibly it signalises a modification of matter 
sing only under extra-terrestrial tjonditions, " Nebulium/" 
the hypothetical element is called, has been vainly sought 
tongst the mixed leavings infioitesimally separated from 
Damon air by intense refrigeration, under favour of their 
equal boiling'pointa ; no mineral, however scarce, has Iwen 
eovered to occlude it ; no volcanic vent exhales it. Yet it 
is whole tracts of galactic space. Must we then admit, 
ier all, that there is a chemistry transcending text-booka ? 
jjgmy be so ; but to abandon the grand ideal of an universal 



*PioJcerifig, Obsermtoryj toIs. ir, p. 81 i t, p. 294 ; Axtr. JVacA. No. 2517 ; 

pelmnd, Mort-tkly NoIums, toL xlv. p. 91. 

■ Prot. R. Haciely, toI. xltU. pp. 124, 327. 

* Xkid. mL xliv. p. 2. * Ibid. toL ilri, p. 18. 



ee 



THE SYSTEM OF THE STARS 



physical acieuce because of a few outetanding incongruitiee 
would be to listen balf-heartedly to counsels of impatience. 
Stellar nebula, so far aa can be judged^ are exceasivelj remote, 
Toluminoua, faintly glowing bodies. The condensed or nuclear 
portions probably included in them give but slight eigns 
of incandesceuee ; spectroscopically (whatever they may be 
physically) they are mere spheres of glimmering gas. Now 
planetary nebuhe (so called by Sir Wjlliam Herschel becanse 
they exhibit a planet-like (Hsc) scarcely differ intrinedoally 
&om the stellar sort, but they either aFSk or owing to their 
greater vicluity, appear larger and brighter, and hence offer 
better facilities for the aualyaiB of their light 

It is found to consist mainly of three rays composing the 
fundamental spectrum of all gaseous nebulEe. Of these, the 
lowest, of wave-length 500 7> is invariably the strongeist* and 
accordingly survives alone in such a dearth of light as that 
created by the combined distance and faintness of the moncn 
chromatic discs or points detected by Professor Pickertng'e 
method. 

The nebular trio of lines, since they lie adjacent to each 
other in the middle or green part of the spectrum, inevitably 
give a resultant green or bluish colour to the objects they 
characterise. The intermediate ray at X 4959, as well aa 
the chief ray, is plauBlbly conueeted with uebulium since it 
preserves in each individuiU nebula the same relative intensity. 
The third or moat refrangible line, on the contrnry, is inde- 
pendently faint or bright. It is no other than the familiar 
F (K0); and the undoubted and unfailing presence of 
hydrogen betokened by ifc in theae fomiationa constitutes one 
of the very few links as yet recognised between nebular and 
terrestrial chemistry. From the former metallic elements are 
excluded ; they are, in uebulic, either non-existent or non- 
apparent. And we recall that stars supposed on other groimda 
to be in a primitive condition are just those in which metal- 
loids tend spectroscopically to suppress metals- 

Modcrn appliances, spectroscopic and spectrographic* have 
enriched the nebular spectrum with some three dozen extra 
lines, mostly in the ultra-violet. Some of those derived from 
plttnotary nebulffi agree with lines bright in Wolf-Eayet stars, 
nfitably with the fundamental member of the Pidtaring series, 




GASEOUS STABS AND NEBULA 



67 



utd with the Bjdberg blue baod^ Occasionally, a glint of 
red hjdrogen is seen ; eight of its more refrangible com- 
panions have been measured ; and helium is represented b^ 
ft domewhat axbitraiy selection irom its various seriea. 

The spectrum of the great Fiah-inouth nebula in Orion 
was first adequately studied by Sir William and Lady Hugging 
from 1832 ouwarda. Ita hydrogen -ingredient is in the peculiar 
condition denoted by itB emission of the higher members of 
the Huggins series, C being imperceptible. The yellow ray 
of hftliiirn was detected by Dr. Copeland, December 28, 1836 ;* 
and he remaxked " some iudicatioiis of resolvability into lines 
or bands " in the dim prismatic backgroiuid upon which the 
central group of rays in this spectrum are relieved The 
Q^iiion further expreaeetl by Sir William and Lady Huggina 
that the faint continuoi^ spectrum visible in most gaseous 
B6bula^ might, were more light available, " be found to consist, 
in great part at least, of closely adjacent bright lines'*^ hag 
been strongly confirmed by recent Lick spectrograms taken 
with the Crossley reflector, exhibiting tell-tale breaks and 
g^nnptoms of overlap in the hazy radiance previously assumed 
to be of '■ white " or unbroken quality.* On a plate exposed 
at Tulse Hill in 1882, an intense ultra-viotet ray of wave- 
length 3727 waa recorded,* and it has proved to be almost as 
marked a feature of the nebular spectrum as the "chief" line 
itself ; QOr can it be an emission of nebulium^ since its 
intensity varies disparately from that of the green lines.* It 
is fiingttlariy prominent in the Ring nebula in Lynu 

A gaseous nature was suggested for the four bright stars 
grouped into a " traiiezium " at the core of tlie Orion nebula 
by an abnormal photograph taken by Sir William and Lady 
Huggins, February 5, 1888J Several years later,® they per- 
ceived the hydrogen-lines to be not only arranged in bright and 

* Oopel&iid, Copemicttt, vdI. i. p. 2. The original "four-liue" pl&netariea 
AH Dnmbered in Dreyer'a JStw General CatAlogae, 7S^2, 7024, aod 7027. The 
two ]vt b*loag to a smgnlar group of allied ijuoous objetta in Cygno*. 

* Monthly N^otices, voL x^riii. p. 360. 

* Fnc -B. Hoeittif, vol. ilvi» p. 60. 

* Lick Sulictin, Sq. US. 

* JttpvH BrU. ABSoeialum, 1S62. 

* Attroph. Jaumat, vol. xxi. p, SflS (Hnftiuaiin). 
' Proc. Soy. Soeifty, toI. xlri. p, 4Q, 

* AHroph. Jottmat, vol. vi, p. 322. 



eS THE SYSTEM OF THE STARS 

dzLrk pairs, but to undergo shiftinga defined hy Frofuaflors 
Frost and Adams* as periodic in the chief member of the 
group, and dependent upon rapid orbital circulation. Theae 
etars are of the early helium variety ; they show absorption by 
oxygen * and cosmic hydrogen ; and it is an open question 
whether the bright lines crossing their spectra may not be 
deriveil from the folds of nebulous matter enwrapping them. 

Local variations in the spectrum of the Orion nebula are 
most likely to subsist. It can hardly be supposed that so vast 
and 80 agitated a formation can be throughout of uniform 
composition, Profeasor Campbell, accordingly, found in 1893 
that the relative intensities of the three lines, " which con- 
stitute nearly the whole of the viaible spectnmi, vary within 
wide limits as the slit of the spectroscope L^ moved over the 
different parta of the nebula." * That is to say, the two first 
lines, while preserving to each other a constant ratio, fluctuate 
largely by comparison with the third or hydrogen-line. In 
some sections of the nebulous Btructure, in fact, the hydrogen 
appears at least five times stronger than the nebulium-ray, 
although near the trapezium it lb four times fainter. 
Hydrogen- light, in conjunction with the unidentified ultra- 
violet emanatioUj, thus tends to become predominant with the 
decUne of illuminative power, whence the conclusion may 
reasonably be di-awu that the nebula verges more and more, 
with the increase of tenuity upward and outward, towards an 
almost purely hydrogenous composition. 

And this serves to remind us that the spectra of the great 
nebulaj, like those of " zoned '* stars, must be considered 08 
integrating the results of emissions taking their rise under 
notably diverse circumatances. Inniunerable strata of nebu- 
lous matter are piled one upon the other in the same line of 
sight. The eye is impotent to discriminate between them ; 
even the spectroscope can do so only indirectly. For, at the 
centre of the nebula, the lines coming from all its depths are 
seen or photographed together ; their different places of origin 



* Aatroph. Journal^ voL zix^ p. 15'3< 

a McCiean, Phif. Trans, vol. Cici. p. 128. 

"^ PuhL Attr. Pae. Sode(^, toL t. p. 300 ; Asir. Nack. No, 3471 ; tf. 
RuDgfl on the phyaiologioal variatJons of brightn«a3 dne to differeace of colour, 
J$traph. Jount. vol. viii. p. 33. 



GASEOUS STAKS AXD NEBULiE 



69 



^. 



are fuinoticed. Light, on the other haDd» t&ken from near 
the edges of the same object, emanates eicluaively from ita 
higher regions, and its characteristic peeuliariti&a may safely 
be localised. The possibiiily then seems at band of dtnding 
in this way the Orion nebula and others of tbo some claiia 
ioto various spectroscopic levels, distinguished by minor 
radiative differences. The helium- lines, for example, may 
prove separable from the hydrogen- lines ; and it would be 
eqtecially interesting to ascertain -ffhetber the three hydrogen 
fleriee change their relative intensities with remoteness from 

trapezium. The differing conditions along the same line 
sight must also hampt^r attempts to determine the internal 
mavemente of the nebula by the displacements or distortions 
of its spectral rays. For there must be a wide disc<»rdance in 
these effects at the heights and hollows projected together by 
perspective. Opposite motions might be progi*eaging in the 
lundry regions conjointly inspected • and the upshot would 
be an ambiguous blurring of lines. Such a possibility 
most be taken into account in estimating the value of some 
proiaifling experiments of this nature made by Dra. Vogel and 
Sberhaid at Potsdam in 1902.^ They hinted at tumultuous 
flows, or eddies, more probably than at a general potation. 
But until they have been repeated and confirmed, it would 
be vain to speculate upon what they seem to communicate. 

Only the species designated as stellar, planetary, annular, 
and irregular nebula give unmistakable signs of gaseity. 
Some five or six score have^ up to the present, been recorded 
to do so ; while several hundreds, and presumably the majority 
of thoee unobserved, shine with continuous light. Yet the 
disttuctiou is perhaps less profound than it seems. The gap 
is at any rate partially bridged. One connecting link 
Beeias to be supplied by the great " looped nebula ** in the 
southern constellation of Dorado, observed by Mr. C, E. 
Burton^ in 1874 to yield a strongly continuous spectrum 
croesed by the unfailing green nebular ray at 5007. Ita 
gaseous nature is thus shown to be modified by the presence 
of an unusually large proportion of dense material, and wliere 
this predominates, as in the Andromeda nebula, the spectrum, 

1 SiizuiufshtrriehU, B«TUa, Unralt 13, 1902. 
> Monihly Kotiees, vol. xxitI. p. 3S. See (7th Harvard JUporC, p. 7, 1392. 




70 



THE STSTKM OF THE STAES 



tboBgh nanhmOj ' ttntiamfrnt" is still nx&rkedLy dUferent 
from the coadnnooB wpettrrsm of a star. 

The light at Uiis "queen of the ne&qle,** prismaticall^ 
diflpened for the fint tizBO bj Sir WiHuun Hoggins in 1364, 
fltmcJE bim ae fu- &omiiufi»B; H aeemed wi^Ul^ thioughoat, 
whether hj the effects of aheorptuxi or of imgnl&r emisdon, 
it was impoesble to decide.^ Beee&t ofaaerratioDs at Tulse 
Hill point to the exiatesioe in Uua i|iectniiii of Twrions bright 
linee or wpatcea intennized with fcimilB of abeorptioiL- Dt. 
ScheJner perfonned in 1899 the diffieolt feat of spectrograph- 
iog the object, which he tolerred, &om the faint tndicalnons 
on hia pl&te, to be a claster of solar stars.' The truth of 
tile matter remaiD« still hidden ; but there are dg&s th&t it 
can before long be dicited. In the iiiTestigatioa of the diis- 
tinctive pecnljarities of continoous nebular light, a field 
lies open which can hardlj &il to be worked with profit to 
the rapidly advancing science of coemical phjsica 

* i'iliL Trmu. toL dir. }>. 441. 

* AUMtfSuUar Sf*eirt, p. 13S. 

> AMr. Ifatk. No. 3549. 



CHAPTER VI 



THB TKMPKRATHREa OF THE STARS 



T we mean by the temperature of a star is the degree 'of 
heat corresponding, on admitted principles, to the radiations it 
sends abroad into space. That ia to aay, the point at which 
4ti ide^il thermometer would stand if placed within the 
photoepibere, assumed for the purpose of simplification to be 
«flecttvely a single emitting surface. Now, the temperature 
of the sun, understood in this sense, has been fairly weU 
aaoertained. It is held by the latest authoriticB not greatly 
to exceed 6000* C» while that of the electric arc, determined 
by similar metboda, comes up to 4000".^ Thug, the gap 
dividing the heat of the solar furnace from the tu plus ultra 
of the laboratory has been greatly narrowed. It no longer 
seems hopeless to establish some degree of continnity between 
the state of things on either side of it. Nevertheless, there 
are qualifications to be taken into account. Conclusions 
legarding the sun's temperature depend essentially upon two 
conditions: first, that the photosphere is what is technically 
known as a "black body"; next, that "Stefan's law" of 
tadiation holds good over a range some thousands of degrees 
beyond the limit of experimental verification. Now, a " black 
body " is one that absorbs completely, and, through an inevit- 
able correlation^ emits perfectly. It radiates in the strict and 
due proportion of the heat communicated to it. But there is 
no such substance in nature ; even lamp-black meets very 
imperfectly the stipulated requirements. They have of late, 
^however^ been imitated by an experimental artifice;^ and the 

* D«7 ftttd OntrAnd, AatTOph, Jtmrnal, roL xix. p. 40. 
* Wien and Lmumer, fFitdtmanti's Anruittn, Bd. Ivi. p. 453, 1895, 

71 




72 



THE SYSTEM OF THE STAES 



facilities for investigation thus acquired prove of high valnti 
Yet the laws of radiation securely est&bliehed with their 
are, in all probability, conformed to only approxiniately 
the Bun. For the photosphere is certainly not, strictly 
ing, a " black body." 

Stefan's law defines the rate at which increase of radiative 
intensity corresponds with growing temperature. It is as the 
fovirth power ; a body rendered, for instance, twice as hot will j 
radiate sixteen times more powerfully than before. Nor ia I 
this a purely empirical rule. Boltzmann broitght thermo- 
dynatnical considerations to its support in 1884, and Planck 
deduced it in 1900 from the electro- magnetic theory of ligbt^ 
Stilli it luay break down before the soUr temperature is 

r^suhed. Its validity, experimeuUlly ascertained up to near i 
1200"» cannot be confidently relied upon at indefiniteljH 
exalted tenn^rstures. Such laws of nature have usually a 
limited fiold of iiction, and we do not know where to fix iu 
boundarii^ This element of uncertainty affects many other- 
wine well 'grounded conclualona. Nevertheless, high tempera- 
tUTtf n^Mitoh has made substantial progress; and much has 
en done in removing the difficulties which long stood in the 

[way of establishing a trustworthy relation between the he 

rStOilved from the aun and the sun^a proper heat. 

But ft8 regards the stars the case is widely different 
DirtOt mod« of procedure are here excluded. We cannot* 
determine stellar heat-constants, and thence infer stellar 
ttiwuMiriitures, This is rendered impossible by the funda- 
mental ciruiuustanee that the heat reaching the earth from 
the stars is all but insensible. The rays of the " wan. cold 
moon" have a thermal power more than 150,000 times 
t than that of the rays of Arcturus,* and Arcturus 
■« judging by the crude standard of direct heating eOects, the 
hottest star of onr acquaintance. No wonder, then, tba|^| 
quantititis bo minute failed to be elicited with the thermopile. 
Sir William Huggins made persiflteut experiments on the 
subject nearly forty years ago,"* and Dr. Stone* a little later. 

* Mandenhall uxd Saauden, Aalropk. Jounu toL xiii. p, 30. 

* Hicboli, Jttr&ph. Jgufn. toI. liii, p. 103, 

* JProc R(ryal Society, Toi. ivii. p. 309. 

* Ibid. ToI. iTiii. p. 159. 



I'ne 
be«lM 



THE TEMPEBATURES OF THE STARS 



tried his tadimeter in 1873, Boys his radiometer in 
888;^ both vainlj. At last, in 1896. Pi-ofea&or Miuchin 
'ntooeeded in converting stellar energj^ into electromotiTe 
Ukw ; ' jet his results, though genuine, were not wholly 
legitimate, owijig to the strong Belective preferencea of the 
onailive cells.* Thoee, on the other hand, ohtained by Pro- 
GEOor Xichols :it the Yerkea obaervatory in 1900 were of 
unequivocal signihcancd. The beams to be examined, after 
being concentrated by a two-foot mirror, fell upon the vanes 
of a radiometer eo delicately impressionable as to detect 
tempera tore-di^erences not exceeding one ten-inillionth of a 
JE^ree oetitigrade. Eesponsive detJections were, accordingly, 
undoubtedly due to radiated stellar energy ; and 
,ey varied from star to star, those caused by Artiturua being 
than twice as large as those which the white rays of 
A had power to produce.^ 

The outcome is in many ways hopeful. Not tliat the 
prospect can he discerned of arriving, by this straight' 
route, at the absolute temperatures of the Btfira ; but 
that it opens a way by which, with much further care and 
BOme knowledge of their relative temperatures may 
led. And even in this direction progress is seriously 
pered by diversities of absorption in stellar atmospheres. 
'e know by eveiyday experience that a glowing object 
fiwm red to white as it grows hotter. So likewise 
with the stars. The larger the proportion of light to obscure 
beat in their radiations, the higher their temperatures must 
be; But we do not see the stars as tliey are in themselves. 
Their faces are veiled by absorptive envelopes, and tar more 
[j veiled in some cases than in others. Moreover, the 
of absorption in the bodies strongly atfected by it is to 
modify very materially tlie relation of luminosity to thermal 
inleaaitj in the emissions they send out. The upper sections 
of the sjjecbram are those chiefly encroached upon ; the stars 
are rendered fulvous, and reduced to a lower grade of photu- 
tnetric magnitude than should be assigned to them if their 



* Pjroc, lioyal Societv, vol. ilvii, p. 480. 

■ /bid, vol, iTiii, r- 1*2. 

* HichoU, Astropk. Joum. vol. xiiL p. IfSi 

• Ibid. p. 136. 





n 



THE SYSTEM OF THE STABS 



inUiDSU^ biiUiAncj were fiillj displayed. Kov^ Arcturos is a 
aoUr vbftT of a reddisb tii^ ; its blue rays paj heavy toll in 
tnTEtstDg its pbotospbenc ^ivelope. Not m those of Vega, 
which escape virtimllj scot-fi«e to the open. It follows that 
companaoQB between them are made on unfair term& If they 
were equalised, Arctums would seem much more brilliant than 
it does now, and the proportion of its heat to ita light would 
be redreesed. And it ia the proportion only, we repeat, which 
is significant as regards temperature Professor Nichols's 
resolt implies, on the taoe of it, that Arctums stands at a 
tower heat-level than V^a, becanae^ althoi^h its radiationja 
contain absolutely more heat, they correspond to a leas 
intense degree of incandeacenceL Bat no such deduction is 
admissible in view of the disparity of conditions. Nor is it 
possible to define the extent of the correction required. So 
far, theuj the relatire temperature of the two atars remains an 
open question. 

Professor Nichols himself i^e^rds his attack upon the 
problem of stellar heat as httlc more th^i a reconnaissance 
by which it has been learned that the position is not im- 
pregnable. With more powerful means at command, he hopes 
to achieve its capture. By pkcing his radiometer at the 
focus of a five-foot mirror, he believes it should be possible to 
arrange white stars to the second, and red stars to the third 
magnitude, " in the order of the thermal intensity of their 
radiations.'' That fainter red than white stars must, owing 
to the quality of their self -absorption, be thus measurable has 
been sufficiently illustrated by the case of Arctunia, Just for 
this reason, however, no instructive comparisons can be insti- 
tuted between stars with notably different atmospheric sur- 
roundings. The use of an identical platform of investigation 
for disparate objects can only lead to illusion. But by ranging 
white and red stars in separate series, and bringing aolaa: stars 
into relation with the sun, much may be accomplished. There 
is no surer criterion of temperature than the distribution of 
energy in the spectrmn. As it rises, the culminating point in 
the representative curve shifts steadily upward to shorter 
wave-lengths. Hence, by locating the maxima of emission 
from glowing bodies, their temperatures can be ascertained; 
and this Professor Nichols hopes to accomplish for the stars. 



THE TEMPERATURES OF THE STARS 



76 




It will be a matter of excessive delicacy; but, apart fioni 

instniinental difficxdties, the method seems promising. Not 

.t the esseDtial incompatibilities of stat« between the 

Brent spectral classes can be abolished by ita means; but 

ties of white stars might be examined together, while 

stars could advantageously be Bssimilated to the Bun. 

xm, Capella gives a spectrum almost indlBtinguishable from 

the Fraitnbofer spectrum eave through the slight discrepancies 

occaaioned by the differing light-quality of its spectroscopic 

companion. It is, however, an enormously larger and more 

ImzuDous globe than om: sun ; and it would be of crucial import' 

auoe to determine whether its vaster acale corresponds to a 

higher degree of heat. Again, Sir Norman Lockyer finds the 

oltra-violet spectrum of Kigel to be more extensive and intense 

than that of Siriua, and to be in turn surpassed, allowance 

being made for the lower magnitude of the atar, by that of x 

Orionis.* If these diversities can be securely associated with 

gradations of temperature, much would be made clear as 

regards the course of stellar development. And Professor 

Nichols's method of spectral energy-measurement pFomiBes, 

if efifectively realised, to afiford just the needed verification. 

Speculations concerning relative star-heat have heretofore 
been based chiefly upon the study of specific linear absoiption. 
But the observed indications refer immediately to the state of 
things prevailing in the reversing layers of the stars, and only 
indirectly to the condition of their photospheres. Moreover, 
their iDterpretation is, unfortunately, still a subject of debate. 
Indeed, it is more actively in debate than ever before ; for, ao 
the importance of reading their meaning aright has come to 
be more fully recogniyed, the pitfalls laid for their would-be 
decipherers have also become mote evident and formidable. 
^^ It has loug been known that the rays emitted by glowing 
^^aseoufl substances alter with alterations in tlie mode of kind- 
ling them to luminosity. Thus, metals rendered incandescent 
in the electric arc give out conspicuously bright tines that 
show quite dimly when the same metals are burned in a flame ; 
and similarly, lines subordinate in the arc develop iuto pro- 
minence through the excitement of the disruptive discharge 
in the electric spark- These progressive changes were, until 

» iV«. Boy*^ Soci«ij/t Feb. IS, 1904. 



76 



THE SYSTEM OF THE STAES 



lately, ascribed by general, though not univeiBal consent, to 
increasing temperature, the spark being, it "waa supposed, 
hotter than the arc, and the arc hotter than the flame. 
Adopting this view, Sir Norman Lockyer gave special 
attention to " enhanced " lines, and pointed out the 
remarkable ci^rcumstanoe of their selective display in th« 
spectra of certain daseea of stars.' It ia undeniable that the 
brightest Unea in the spark-apectra of the chemical elements 
are also highly characteriatic of white, and more especially of 
helium stars; that arc-lines to a certain extent supersede 
them in solar stars ; while in red atars with banded spectra^ 
even fame-lines take a position of importance. And the 
aaeumptioii that from these modifications trustworthy informa- 
tion could be derived regarding the comparative temperatunw 
of the stars in which they were observed, lay ready at hand, 
yet proved misleading. So early as 1888, Professors Liveing 
and Dewar were led by their experiments on the spectrum of 
magnesium to throw doubt on the received opinion that the 
electric spark ia, in the literal sense* hotter than the arc' 
Sir William and Lady Huggins demonstrated in 1897' that 
the colcimu-lines, H and K, owe their preponderance iu the 
stin not to extreme heat, but mainly to the inconceivable 
rarity of the emitting vapour, and they illustrated in 1903* 
by photographs of the magnesium-spectrum, the sensitiveness 
of radiation to changes in the mode of electrical excitation. 
The spark-discharge is oscillatory ; the released energy rushee 
with amazing velocity Lo and fro across the gap. But in the arc 
it assumes the character of a continuous flow ; means of trans- 
port are available ; it does not need to force a passage. That 
the ultimate particles of matter acted upon must respond very 
difiTerently in each case is then eaaily understood. A final 
overthrow was given to the older opinion when MM. Eherhard 
and Hartmann of Potsdam pronounced,* after laborious investi- 
gations, the spark-spectrum to originate, not from thermal 
radiation, but from ** electrical luminescence." 

The word is of subversive import. Its introduction by 

' Ptec Seyal Society^ toIs. Ixi. p. 441 ; Ixt. p. ii2. 
^ Ibid. vol. iUt. p. 341. 
■ Ibid. vol. Ixi. p. 433 ; Alias of JiUilar SpiCtra, p. 91. 
• * Aatroph. Joum. ?ol. xtu. p. 14&. 

= Sit^ngtbtriehtf, Berlin, Feb, 26, 190S. 



THE TEMPERATURES OF THE STARS 



77 



Wiedamauu in 1894 ^ to signiiy light without heat, or light 
in excess of bemper-atuie, liceaBed the adoption into orthodox 
phjuical science of certain new ideas which had long been 
gradually creeping to the front. They allow a wide latitude 
m the explanation of radixitive and spectral phenomena. 
Ltmiiaesoence may be evoked by chemical or electrical action ', 
it may be superadded to ordinary light, or appear by itself; 
either a continuoua or a discoutinuoua spectrum may be 
derived from it ; nor is it necessarily correlated with absorp- 
tion. Hence strict reasoning ia precluded where there is 
maon to suspect its ]>resence ; for the laws of its production 
w yet evade research. We find, accordingly, that the spectral 
peculiarities detected in the various classes of atars^ and long 
held to supply sure indlcationa of their thermal rank, fivow in the 
laboratory highly complex relations with vapour-density, with 
electrical '' damping," with chemical proceeaes, with luminescent 
iction. We are then thrown back in our search for tests of 
stellar heat-power upon the varied intensity of blue radiation 
from stellar photospheres, allowing, as beat we can, for the 
differences in absorption by which it is partiully masked. 
One pair of stars — Capella and Vega^thns compared by Sir 
William and Lady Huggina* yielded an mdooked-for result. 
The solar orb seemed intrinsically the hluet\ and was inferred 

-to he the hotter of the two. 

■ The temperatures of the atare are intimately related to 
their life-history. From the time when they first assume the 
phdtospheric vesture until, through decrepitude, they cease to 
ahine, a constant waste of energy must be going forward in 
them pari passu with contraction. Yet, up to a certain 
point, recuperation, ao far as eensible heat is concerned, takes 
the lead of dissipation. Cooling bodies, by a seeming paradox, 
rise in temperature until they cease to be wholly gasuous. The 
kw of thermal ascent, enounced by Lane of Washington in 1870, 
applies^ however, only to average temperature, while such indica- 
tions as can be gathered by us refer to surface temperature. And 
this depends not so much upon the amount of heat stored in 
ch globe as upon the facility with which it can be conveyed 
Kpward to the scene of action. Hence viscosity, gravity, chemical 

* Annalen dtr P/'i/nk, Bd. liv. p. fl04, 
■ Atlaa f)/ Stellar SjiesCrat p. 85. 



78 



THE SYSTEM OF THE STARS 



oompositioai, enter into the account as items of uaknown value. 
Nor is there any posibility of fixing & priori tho point ut 
which the extenul beat-power of a atar begins to f&ll ofl 
Too manj conditions are involved The main object of actual 
iuquihea is to establish the thennal reUtious of the varieties 
of stelUr spectra. Those of earlieet type are non-metallic ; 
they are markied hy the absorption or emission of hydrogen, 
heliom. 09^gen» and nitrogen. Sir Norman Lockyer alleges 
abnormally high temperature as the cause of this unusual 
chamcter. Metala, in his opinion, do not exi^t in 8uch stars 
because intense heat keeps their primal elements aannder. 
They are dissociated, or perhaps have not yet be^n to be 
associated Nebuh^ are nevertheless equally uou-metallic with 
incipient stars, and it is difficult, if not impossible, to ascribe 
an enormou^y high temperature to volumes of matter in the 
last degree of attenuation. However this may be, the fact 
can scarcely be gainsaid that metals show more and more 
distinctly in stars as they adranoe towards maturity, and that 
the ^t signs of their presence are the lines ''enhanced" 
in the laboratory when the spark -discharge replaces the electric 
are. Bat, a^ time and condensation go on, the arc-lines gain 
upon their nv'als, and in spectra of the solar type leave them 
utterly in the lurch. Finally^ in banded spectra strong flame- 
lines emerge ; while throughout the progression, as if to com- 
pUcate matters still further, the great violet ray^ of calcium, 
although eminently characteristic of the oscillatory spark, 
reach their highest development in reddening stars. The 
entire series of changes, nevertheless, is beyond question 
orderly and consequential ; and the obscurity of their origin 
will be in part cleared up should it be found possible to 
connect them with an equally consequential series of tempera- 
ture-changes. Yet it must be owned that the trend of pi-esent 
inquiries is rather towards establishing for them relations 
with varying modes of electrical illumination than directly 
with gradations of thermal intensity in stellar photospheres or 
their reversing envelopes. 

The hope of estimating temperature by locating maxima 
of intensity in discontinuous radiations is probably fallacious. 
Sir George Stokes, already in 1876, held it possible that the 
intenser blue radiation with increase of heat obvious in con- 



d 



THE TEMPERATUEES OF THE STAKS 



79 



tinuons spectra might extend to spectra composed of separate 
lajB — that the intensest line might shorten in wave-length ae 
Uiermal energy was gained by the emitting vapour,^ And 
M. l-Angenbiich obtained, in 1903, some experimental indica- 
taODft of the actual subsistence of some such relation.' He 
eonsideTa, indeed, that Professor CampbeU'fl observation of 
tbe greater strength in the Orion nebula than in vacuum- 
tubes of the mot* refrangible hydrogen-linea affbrda proof of 
ft difference of temperature which may eventually be measured 
in thernaometric degrees. Professor Xeeler, too* had inferred 
ixom the singular vividness of the green and blue hydrogen 
emanations in all gaseous nebulae " eithex a high temperature 
of the gasea emitting the light, or a state of strong electrical 
excitement/' ' 

The conditions of electrical excitement in heavenly bodiea 
caoiiot as yet be de&ned. That they are present in some casea 
moie fully than in others is. however, virtually certain ; and 
many anomaloufi appearances will doubtless ^nd an appropriate 
explanation when they come to be interpreted as electncal mani- 
feetations. Meantime, we can only reason about what cornea 
more or lees within the range of tangible acquaintance ; and 
experience is wholly contradictory of the notion that nebulffi are 
flZoeBBively hot bodies. On the adopted principle they should 
be at least 3000*^ C. hotter than the solar chromosphere, 
in which red hydrogen-light ia predominant, while nebular 
hydrogen shines exeluaivcly, it might he said, with green, 
blue, and ultra-violet emissions. Further, in many variable 
otans the leading bright hydrogen line is the dark blue 
HS» C and F (Ha and Hy5) being equally invieible. This 
relation, if interpreted on the basis of Wieu's law of spectral 
energy, would lead to the inference of fabulous temperatures as 
prevailing in such stars. It is, however, moat improbable that 
Wien's law is really applicable to bright-line spectra. " The 
radiation from gases," we are told on high authority, " is usually 
wholly or partially luminescent." * This means that its calorific 
ipendence is slight and secondary. Hence nothing can be 



^ Proe, Boyal SccUtj/t toL xsiv. p. 353, 

' Annalem der Fhyaik, No. 4, 1903. 

* Puhl. Litk QbitTWiUoTy^ vol. iii. p. 226. 

* Day and OrBCrand^ Attrvph, Joum, toI, lii. p. 31. 



80 



THE SYSTEM OF THE STABS 



authentically learned from its quality regarding the temperatiue 
of ita source. 

The science of stellar thermotics is, in fact^ still in a 
tentative stage. The most assured datum at ita command m 
the temperature of the sun, which at least supplies a term of 
comparison for the teraperaturee of the stare. It is fairly 
certain that the efEective lieat of the solar photoephere, measured 
on the thermometric scale, ia 6000" to 6500^ C. And there 
ia some probability that the sun has never been, and will 
never be, much hotter than it is now. But thia does not 
imply that none of ita compeers exceed tt in thermal power. 
Those of greater ma&s niuat develop proportionately more heat, 
60 that it is quite likely that giant suns of the solar type, 
aucb as Capella and Arcturus, radiate far more intensely per 
unit of area than our sun, and have photoepberea hotter m 
the due ratio of the " fourth power " increase^ 

Stars with banded spectra are generally admitted to have 
made a further advance in cooling. Yet, despite the heavy 
losses in total heat incident to protracted radiation^ their 
superficial temperatures may have euffei-ed only a slight 
decline. The development of flutinga in their q>ectra is not 
decisive on this point. Beceut investigations tend to show 
that it is an index^ not so much to the degree of excitement 
produced by heat, as to the kind of agitation set up by 
electricity in the originating vapour,* Nor have we any 
abfw^Iute ssBurance that heat -expenditure proceeds uninter- 
ruptedly. Current hypotheses on the subject possibly need 
revision. Eegenerative agencies may. under given circum- 
stances, be called into pltiy. The many suggestions of 
" radiology " (as the new science of radio-activity might be 
designated) cannot be inconsiderately set aside. They are^ 
however, too vaguely conveyed to be profitably discussed. 
The cosmical effects of these novel phenomena elude just yet 
our mental grasp. They may prove to be of stupendous 
Importance; to the future is reserved the task of unfolding 
their character and scope. 

* Trowbridge, Awer. Jowrn. of ScituvXt Tol. UL p. 117, 1697 ; Hale, Snct/, 
BriL Tfi). xxxii. p. 779, aiC. Spectroscopy ; Fringfthaim, Jiappi/rit dv Gon^f* 
Inttmationat di FAytiqv^^ t, ii. p. 108* 



CHAPTER VII 



TEMPORARY STARS 



Thk facta connected with the light-changes of stars arc in the 
highest degree strange and flUTprising ; and wonder does not 
lofBBn as faiailiaritj with them grows. They are of everyday 
occurrence ; they can be predict&d beforehand, in many cases 
with nearly as close accuracy as an eclipse of the 8un or 
m(X)ni and they affect in manifold ways a great number of 
objects. Stellar variability is of every kind and degi-ee. 
With the regulai-ity of clockwork some stars lose and legnin 
a fixed proportion of their light ; others show fitful aeeeB- 
iiona of luminosity succeeded by equally fitful relapses into 
obfloority; many waver, in appearance lawlessly, about a. 
datuna-level of lustre itself perhaps slowly rising or sinking. 
The rule of change of a great number is that of an evident, 
though strongly disturbed periodicity ; a. few seem to 
spend all their powei-a of shining in one amazing outburatt 
after which they return to their pristine invisibility or 
msignificancc. 

The amount is aa much diversified as the manner of 
fluctnation. Changes of brightness so minute as almost to 
defy detection are linked on by a succession of graduated 
examples to conflagrations in which emissive intensity ia 
multiplied a thoueaud times or more in a few honra The 
range of variation is in some stars sensibly uniform; they 
subaide during each crisis of change to the same precise point 
j.of dinmese, and recover without aubtraction or excess, just so 
lucb light as they had before. In others it is widely 
egular. The limits of fluctuation in one period furnish no 

31 6 




s: 



THE SYSTEM OF THE STABS 



precedent to be conformed to in the next Nothing is pre- 
deteTmined ; the intensitj of each phase seems to depend 
upon a complex set of conditions unlikely to recur twice in 
the Bame precise combination. 

The first effort to regularise the phenomena of variable 
stars was made by Professor K C Pickering in 1880,' His 
Bve clnsses, though often enough (as might be expected) con- 
fused at the borders, are still sufficiently distinct to form a 
useful framework for the facta. They are as follows : Class I. 
includes temporary or "new" stars; Class II., stare like 
" Ikfira " Ceti, strikingly variable in periods of several months; 
Glass IIL, stars subject to irregular fluctuations ; Class IV., 
variables with periods of a few days exemplified by B Gephei 
and /9 Lyrae ; Class V,, " Algol-variables," or stars like Algol 
in Perseus imdergoing brief obscurations at fixed intervaLk 
We will take each in turn, beginning with the first. 

A temporary star may be defined as a variable attaining 
one single, vivid maximum. A swift rise to such a height as 
to constitute a virtually " new " object, followed by a slower 
yet prompt decline, characterise these outbursts, close upon 
thirty of which have been more or less credibly recorded 
within historical times. The genuineness of those stated to 
bavg occurred in the follo^ving years ^ is, in only a very few 
cases, open to cavil. 

134 EC. in Scorpio; the star of Hipparchus, 

123 A.D. in Ophiuchua. 

Dec. 10, 173, between a and ^ Centauri. Conspicuous; 
scintillated strougly ; visible eight months. 

386 (April to July) between X and <ft Sagittarii. 

389, near a AquiUe, said by Cuspinianus to have eqmdled 
Venus ; vanished after three weeks. 

March 393, in the Tail of the Scorpion* 

827 (?) in Scorpio. Observed during four months at 
Babylon. There is some uncertainty about the date, none 
about the fact. 

May to August 1006, in Scorpio,^ Described by Epi- 
damnus, th© monk of St GaU^ as " oculos verberans." 



^ Proceedings Atasr. Acad. toL xvL p. 17. 

* S** Humboldt's CoMmat. toI. iiL j». 20& (Ott^'e tranfllation). 

' Sclibafeld, Astr. J^ttch. No. 30»4. 



TEMPORAEY STAES 



83 



all photographic detections. 



July 1203^ in the Tail of the Scorpion; said to have 
T«aeixibled Saturn. 

1230. Id OpMuchua 

15 72, Tyeho's star in Caaaiopeia 

1604^ Kepler*s star in Ophiuchus. 

1670, in ViUpecuIa, 

1848, in Ophiuchus. 

I860* in Scorpia 

1866, in Corona Borealis, 

1876, in Cygnus. 

1885, in Andromeda. 

1887, in Perseus; discoTered by photography. 

1891, in Auriga. 

1893, in J^orma ; 

1895, in Carina ; 

1895, in Centaurus; 

1898, in Sagittarius; I 

1899, in Aquila; J 
1901j in Perseus, 

1903, in Gemini; found on the Oxford chart -plates. 
Making a total of twenty-aeven^ besides four or five question- 
able instances mentioned in Chinese anuala. 

The moat noteworthy feature of thia liat ia the curiously 
partial distribution of the objects enumerated in it, All but 
one of them lie in the thoroughfare of the Milky Way, and 
nisie are clustered together in the section of it marked by 
the stars of the Scorpion and the Serpent-tamer. In time 
also the grouping of the apparitions is strikingly unequal. 
The oocurreoce of three within the seven yeaiB 386 to 
393 A.D. was succeeded by a blank of four and a half 
centuries. Kepler's came pretty dose upon Tycho's star ; 
none were recorded between 1670 and 1848; then, within 
little more than half a centviry, seven " Novae " attracted 
visual attention, and six made their marks upon sensitive 
plates. 

The brightest sidereal object known to us by authentic 
description was the " stranger-star " in Cassiopeia, observed 
by Tycho Brabe.^ He first saw it November 11, 1572, but it 

* Wolf, QiKhieMe der Aitrontymi*^ p. 414 j Kaisert De Sttrrenhemtl, Part L 
. £fiS ; Lyun, &>te.rvat.urifj vol, xvi. p. 268 ; Drejer, ihid.^ voU ixiv. p. 106. 



34 



THE SYSTEM OF THE STAKS 



bad ah-eiady been noticed by Wolfgang Schuler at Wittenbei^, 
November 6, and by Lindauer at Wmterthiir, Kovember 7, 
while Maurolycus entered upon its systematic study at Meaaina, 
November 8. Observed by Tycho to be the rival of Venus, it 
fihowed to keen eyee at midday, and at night through clouds 
thick enough to obscure every other star. After about three 
weeks, however, it began to fade, and in March 1574 die- 
appeartsd linally. Ita colour was at first dazzlingly white^ then 
for a while ruddy, and from May 1573 onward, pale with a 
livid cast. Rapid aciutillatiou distinguished it throughoaL* 
There is no reason to suppose the outburst other than solitary. 
The sppearaneea in the years 945 and 1264 connected with 
it by a Bohemian astrologer named Cyprian Leowitz * were 
almost certainly apocryphah' 

The " new " star (designated " B CassiopeiBB ") can atill be 
perceived smouldering in the sjiot where it once blazed. 
Tycho'a measurementa, reduced and discussed by Argelander. 
located it within one minute of arc of a reddish^ eleventh 
magnitude star, the character of which, as di8d.oa&d by the 
observations of Hind tind Plummer in 1870-4, and of Sofarik, 
188S-90,* fully warrants the inference of its identity with 
the famous " temporary " Not only is it variable to the 
extent of nearly a magnitude, but it frequently seems hazy 
and ill-defined, as if through some abnormality in the quality 
of its light. 

The star of 1604 ran a parallel course to that of 1572. 
Discovered by Maestlin, October 9, and observed by Galileo at 
Padua, October 10, it quickly overtopped Jupiter, but by the 
end of March 1605 had sunk to the third magnitude, and a 
year later vanished. Kepler describes it as " sparkling like a 
diamond with prismatic tints/' ^ but says nothing of progresBive 
changes of colour, " Nova Serpentarii," has left behind no 
clearly identifiable representative. 

The next " new star " wag discovered near fi Gygni on June 

^ Tjroho, De ifovd Stelld attT» 1672, p. 303. 
5 Tyoho, Judieixim de Novd Stettd. 

* Lynn, Observai&ri/, vol. vU pp, 126, l5l ; Sadierj BngiiBh Mechanic, toL 
iim p. 402 ; Tycho Bruhe, Frogymnaimataf p, 331. 

* Monthly Jfoticev^ voL xixiv. p. 168 ; i^n*& Caiatogug cf kncien VariabtM, 
p. 164; A3tr. Nach. Ko. 2960. 

* Kcplflri Opcfa^ t. U. p. 620. 



TEMPORAEY STARS 



d5 



20, 1670, by Anthelmue, a Carthusian monk at Dijon. It 

was then of the third magnitude, but its decline^ unlike that 

oE oth^B of its class, was iuterrupfced by two reappearaaceB 

Kparated by intervals of invisibility. Between March and 

May 1671, it rose from the fourth to the third raiik^ then 

died out, only ftickering up to the fliith magnitude in March 

1672.^ Almost exactly in its assigned position, Mr. Hind 

poked up, April 24. 1852, a star between the tenth and 

eleventh magnitude, which, when reobeerved in 1861, had 

loet more than half its lights and gave the bluiTed image 

eharacteristic of many auperaiinnated Novsj^.'^ The triple 

mft-rimnTTi of Anthelm's Btar assimilates it to Janaon'a 

wiable P Cygni,^ which has itself often been classed as a 

Nova, 

An object unequivocally aueh was detected by Mr. Hind in 
Ophiuchus^ April 28, 1848, when it was of 6'7 magnitude, 
and intensely reddish yellow.* Four days later it had 
mounted above the fifth magnitude, from which eminence it 
filowly descended, making no lasting halt until, in 1874-5, it 
bad got down to the thirteenth magnitude,* 

With the spectroscopic study of temporary stare, a fresh 
chapter in our knowledge of them opened. Through the 
magic of the prism, more was ascertained as to their essential 
nature in five minutes than could have been learned in as 
many centiiriea with the telescope alone. On May 12, 1866j 
Mr. John Birmingham, of Millbrook, near Tuam in Ireland, 
was amazed to perceive an unfamiliar star of the second 
magnitude shining in the constellation of the Northern Crown. 
On May 16, the application of Sir William Huggina'a spectro- 
gtiope showed the object to be wrapped in a mantle of blazing 
bydrogeo. Five bright lines (three of them due to hydrogen) 
stood out from a range of continuous light broken up into 
jEones by flatings of strong absorption.^ The incandescence of 
the star was hence largely atmospheric, and for the real, &om 
the rapid rate at which it fell away, could have been only 

^ J, Coaciiu^ £l4mmtMd'A9iTano7nUj p. &&, 

* MiMiKiji Notii*tf toL txL p. 231 ; A'atittif yoL xxxii p. 555. The staf ia 
Ko. 1814 in tb« Gri»«iwifib Cttologue for 1S72. 

> See anU, ^ 82. * Aatn Nock. Noi. 636, d3S, 672. 

» itmihiij Koti&a, toI. xii, p, 232. 

' I^daecdmgw Royal Society, rol. xv, p. 140. 



96 



THE SYSTEM OF THE STAKS 



" akin-deep." Although the light dwreased hj a. d&ily half 

magnitude, and its colour changed from white to orange, no 

alteration took place in the character of the spectrmu. Tha 
bright rays, however, faded fiomewhat less quickly than their 
continuous background. 

The vieibiUty of the object to the naked eye lasted only 
eight days, and al^eady^ in the beginning of June, it had sunk 
to the ninth magnitude. Its alow subsequent decline waa 
interrupted by fluctuations, thought by Schmidt to be 
periodical in about ninety-fauT days.^ When observed by 
Vogel, March 28, 1873, and again by Barnard in 1902, it 
wa3 of about 9*5 magnitude^ and gave an ordinary stellar 
spectrum.^ Virtually, it had resumed the conditions of ite 
existence when Schoufeld entered it aa of 9"5 magnitude in 
the Bonn Durchniuaterung. Ita leap upward to the aecond 
magnitude, involving a th&ttsand -fold gain of light, waa 
accomplished with extraordinary suddenness. Two hours 
and a half previously to Birmingham's discovery, Schmidt 
surveyed at Athena the constellation in which the blaze waa 
about to occur, and noticed nothing unuBual- He was certain 
that the star could not then have been aa bright as the fifth 
magnitude. 

Although its character aa a Nova seems undoubted, the 
name of " T Coronae " waa bestowed upon it in conformity 
with Argelander'a system of nomenclature, by which the 
variables in each constellation are designated, in the order of 
their discovery, by the Konoan capital letters from R onward. 
Only stars otherwi&e anonymoua, however, are included in the 
distijictive series thus created, so that many variables ure still 
entitled in the ordinary way by Greek letters. 

The stellar apparition that ensued after ten years was, in 
Bome of its featurea, the most remarkable of all. Dr. Schmidt 
noticed at Athens^ November 24, 13V6j a atar of the third 
magnitude near p Cygni, in a spot till then untenanted by 
any known stellar inmate. The weather having been cloudy 
during the previous four days, there waa no possibility of 
tracing the steps of its ascent, but it ran down very rapidly, 
and ceased to be visible to the naked eye on December 15* 

* MonaUUri^i, BerliD, ia73, p. 304 ; Monihly Ifaikes, vol. IxiL p. 41B. 



J 



TEMTORABY STARS 



B*r 



Its changes of colour pursued an inverse order to tfaoBe of 
ita predecessor From golden yellow it turned white, and 
eTeotually bluish. 

The earliest spectroscopic examination of " Nova Cygai " 
waa miide by M. Corau at Paris, December 2 and 4.^ Just 
tha same range of bright linea waa measured by him which 
would start into view in the solar spectrum upon a considerable 
augmentation of incandescence in the sun's gaseoua sur- 
roundings. Besides thiee, if not four, hydrogen lines, there 
were the yellow helium ray (wave length 6875), with severul 
of ite ffubsequently identified companions, the magnesium 
^^^up &, and probably the sodium D. The star, at the height 
^Bf ita outburst, scarcely seemed to diverge from the type of 
^^3 LyRe and y Cassiopeije. The C of hydrogen was vivid ; 
the continuous spectrum strong. But as the light diminished 
penmrkable changes supervened.* Bed hydrogen insensibly 
yielded its supremacy to greeu ; only a faded remnant of the 
general prismatic radiance survived in the yellow and blue ; 
helimn ceased to glow ; and the lazuHte band of y Velonim 
(nitrogen ?), identified by Copeland, January 2,* gained 

K expected prominence. 
Meanwhile, the chief nebula-line, usurping the place, as it 
Ip^ of an adjacent green line of helium, had been steadily 
■|iing to the front; and when observations, suspended in 
March owing to the encroachments of daylight, were resumed 
by Dr. Copeland at Dnnecht, September 2, 1877, it stood 
alone.* All the surviving liglit of tlie object — by that time 
Bonk to 10*5 magnitude — was concentrated in that solitary 
green ray, and a minute planetary nebula was^ in appearance, 
substituted for a star. But this too proved to be a phase 
scarcely less transient than the rest Three years later, when 
the Nova hatl dropped a couple of magnitudes lower still, 
^^indications were obtained at Harvard College of its affording 
^hm ordinary stellar spectrum.^ They were fully confirmed 
^^from the evidence of a spectrogram taken by Mr. Palmer, 
with the Crossley reflector^ August 12, 1901, whit;h showed 

^ Oomptit Rttvdus, t IxxxiiL p. 1172. 

* Loekyer* Proc Rayal Society, vol. xUii. p. 13B, 

■ Ct>p§Tr%icva, vol, ii. pp. X02, 112. * Ibid. p. 108. 

» Annual lifpoTly ie79-B0, p. 7. 



88 



THE SYSTEM OF THE STABS 



the faint rays of Kova Cygni to be uiimigtaka.bly contmuoi 
Profeesor Barnard observed them ahortly afterwards with 
Yerkes 40-iiich reftitctor to be vigually iU-defiued and bluisfai 
the star being estimated at about 15 5 m^oitude.^ He could 
detect no change in ita position rektive to the surrounding 
stars mapped twenty -four years earlier by Copeland and 
Lohse. 

The Novfe of 1866 and 1876 appear to have been of 
essentially the same character, notwithstanding some variety 
in the phenomena attending their decay. But they were 
not more closely assimilated by the analogous peculiaritiM 
of their light than the pair we are now about to describe by 
the singular circumstances of their situation. On May 18, 
1860j a nebula in - Scorpio, numbered 80 on Mesaier's list 
(6093 in Dreyer's New General Catalogue) was observed by 
Dr. Auwers at Berlin.^ It pre^nted its usual appearance 
of a somewhat hazy ball of light, brightening gradually inward, 
and resolvable with difficulty into separate stellar poiute* 
together constituting a closely -packed, and most likely ex- 
oessively remote globular cluster. Three nights later he 
looked again, and saw that these minnows had a triton in 
their midst. A seventh magnitude star shone close to the 
centre of the stellar group. The existence of the new-comer 
lasted visibly just three weeks. Before May 25, a decline set 
in ; it hiid made conaiderable progress when, on May 23, Mr, 
Pogson (uninformed of Auwers's discovery) was startled by the 
apparent substitution of a star for the nebula/ the dim 
luminosity of which seemed actually obliterated by the keen 
stellar radiance emanating from within it. It recovered, how- 
ever, very speedily from this merely optical effacement. On 
Jtme 10 its normal aspect was almost restored, and has never 
since been disturbed. 

After the lapse of a quarter of a century the significance 
of the event was accentuated by its repetition. This time the 
great nebula in the girdle of Andromeda was the scene of the 
outbreak. The unlooked-for addition to it of a " star-like 



^ Lick Svlletin^ No. 34 ; jrfatrepA, Jpum. vol. ndii. pp. 232, 233. 

'J Monthly A'cNtic«j, vol. Ixu* p. 10£i ; cf. J. O. Lohse, ibid. vol. xlriL 
p. 494, 

* Jttr. JVfflcA. No. 3287. * Mmthltf Noticta, va\. ni, p. 32- 



M 



DQcleiis " was aonounced by Dt, Hartwig at Dorpat, August 

31, 1885; bat it turned out that the change had already 

\wen perceived by Mr. Isauc W. Ward of Belfast, August 19, 

and two oights earlier at Eouen Uy M. Ludovic Gully, who 

«et it down aa au effect of bad definition.^ Concordant 

obeen'atioos by Tempel at Florence* Max Wolf at Heidelberg^ 

V)d Engelmaon at Leipzig showed decisively that the strange 

object made no show down to 10 p.m. on August 16 ;"^ and a 

photograph taken by the late Dr. Cotiinion in August 1884 

gave positive assurance that a year earlier its place bad no 

stellar occupant as bright aa the fifteenth magnitude.' What 

were virtually the first taya of the Nova reached the earth 

August IV, 1385. 

Between that date and Auguat 31 it mounted from the 

ninth to the seventh magnitude ; then without delay 

' Bitered upon nearly a& swift a downward course, cheeked 

ur only one decided pause. Even the largest telescopes 

^^Ped to keep it in view after March 1886. The full yellow 

oolour, by which the star at first contrasted effectively with 

the silvery background it was projected upon, faded with its 

light. No haze or glow blurred its image^ which remained 

sharply stellar with a power of 1100 on the great Princeton 

refractor, when the adjacent nucleus of the nebula melted 

into a confused luminous blot.^ Attempts, incomplete from 

the nature of the case, made by Dr. Franz at Konigsberg, 

and by Professor Hall at Washington, to determine the parallax 

of Nova AndromedjE, gave only negative results.^ So far as 

they were significant at ail, they indicated its immeasurable 

remoteness from the earth, nor should it be overlooked that 

Sir Robert Ball's similar experiment upon Nova Cygni had 

ifitimated a similar conclusion.^ 

The spectrum of Nova Andromeda was of a dubious 
character. It bore witness to a completely different order of 
incandescence from that of the " blaze-stars " in the Northern 
Crown and the Swan. The bright rays which it perhaps 
included were incouepicuoiis. None were definitely determined, 



1 CUltt Titrtc. Oct 1, 1885. » AmIt, Nadt. Nas, 2682, 2fl83, 2691. 

* Nature, vqI, xjntiL p. B22. 

* VoojQg, SuUreai Mcutngw, vol. Iv, p. 282. 

* AMr. Niiuk, 2816. • Dunsink 0&wn«W<rt«, Part V. p. 24. 





THE SYSTEM OF THE STAItS 

though the presence of several io the green and yellow ' -v 
atroriglj suggested to Sir WilUam Huggins on September 
and Dr. Copeland succeeded, on September 30, in getCis 
rough measures of three vaguely disceraible accessiouB ( 
brightneBs.* The light, however, waa mainly continuous ;, aw 
a general resemblance in quality of radiance was one o 
many arguments proving a physical relationship between tht 
atar and the nebula. This was, indeed, superHaously evident 
That one stellar coufiagmtiou should by chance be projects 
almost accurately upou the core of a nebula in reality difl 
connected from it, is just conceivable ; that ttco aucb higbl; 
improbable events should occur within twenty-five yeara c 
each other distances possibility. A third was barely rescue 
by photographic agency from irrevocable oblivion.* Diacoveie 
by Mrs. Fleming when its course was already nearly ra 
enough waa nevertheless learned about Nova Centauri to plai 
beyond doubt its analogy with its two predecessore. Tl 
nebula, in an outlying part of which it was lodged, ia catalogui 
as N.G.C. 5253 ; the spectrum of the star, fortunately reeordi 
by a casual exposure in July 1S95, showed the same irregular 
continuous character with that of Nova Andromedae* "W 
may then feel assured that the Kovse of I860 and 1885, 
well as that of 1895, were situated within the substano^ 
the several nebnlre which they temporarily illuminated. ■ 
This collocation obviously falls into line with the galact 
aflnnitiea of other temporary stars. The Milky Way is a plai 
of condensation for all small stars, but more especially, and 
a marked degree, for stars as well as nebuliE of a gaeeoi 
nature; Temporary stars are closely cognate with these, n 
merely through the brief gaseous incandescence bringing the 
to our notice, but through the symptoms of nebuloaity whii 
Burvive it. Opportunities for their study have lately bei 
nnmetoua and varied. Dr« Anderson of Edinburgh la di 
tiuguished as the visual discoverer of the two moat remarkat 
Novie that have appeared for some centuries. He anjaotmci 
the addition of a guest to the stellar family of the Chariots 
February 1, 1892 ; but the Harvard photographs were fom 

* SepijTi Brit. AtaacieUion, 1885, p. 93f>. 
" AfotUhly Noticts, vol. xlvii. p. 64- 
' F^ty^fiTBt ffarmrd Rtpart, p. 7. 



TKMPORAEY STARS 



91 



have Biiently noted the event on Deceuibet 10, 1891, and 
^ stars omximum brightness was fixed from their evidence at 
ii magnitude on December 20. Speetrographic methods, 
iy\iUed for the first time to anch an outburat. disclosed some 
profoimdly significant peculiarities, since ascertained to be 
^oeraUy characteristic of " temporary '* star-light. They con- 
sist mainly in the great width of the spectral lines, in the 
duplication of the bright by a corresponding dark series, and 
k tbeir large relative displacements. A fine spectrogram of 
yova Aurigffi, taken at Harvard College during au early stage 
of its development, is reproduced in Plate VI. Three briUiant 
hjdrogen-lines, beginniDg with F on the right, are visible in 
it with their obscure, more refrangible companions; the H 
and K of calcium are aimilarly conspicuoua to the left ; while 
the hydrogen-aeries ia continued in a less pronounced manner 
bejrond them in the ultra-violet. It is instructive to compuie 
with it (oee Plate VII.) a map of the same spectrum, drawn by 
Father Sidgreaves from two photographs taken at Stonyhurst, 
February 3, 1892. It ib in two sections. The upper contains 

Cthe lines apparently present, the lower those that were 
ent and indubitable. The shorter wave-lengths, it will be 
noticed are here on the right hand ; F is in the middle, and D, 
undivided and distended, is placed near the extreme left. The 
hypothesis of a double origin for this extraordinary spectruin 
irresistibly suggested itself. Two stars, one shining with vivid 
gaseous emissions, the other showing heavy absorptiou-Unes^ 
were supposed to be in the act of rushiug past each other 
with enormous opposite velocities, a grazing collision, or tidal 
influence being invoked to account for the conflagration by 
which they were rendered suddenly conspicuous. Accumulated 
incongruitiea, however^ have thrown discredit upon the " two- 
Btar " theory, and it has now few adlierents. That an encounter 
BO wonderfully circumatanced shoidd occur just once was con- 
ceivable ; but the repetition of virtually identical occurrences 
year after year transcended the powers of reasonable assent 

Nova Aurigve followed the example of a nebular trans- 
formation set by Nova Cygni, the change being in its 
case accompanied by a very considerable recovery of lustre. 
^he ator, then of sixteenth magnitude, was lost sight of in 

ril 1892; in August, its light, multiplied some three 



THE SYSTEM OF Tl 

hnjidred times, waa mostly collected into tbe greeo ray of 
nebulium (X 5007). Nor did it begin to ebb away again for 
many months, aud thea only at the slow rate of about half a 
magnitude yearly. Mr, C. I), Perrine observed it in August 
1903 to be of the fourteenth magnitude ; and actually 
succeeded in obtaining a legible epectrographic impreasioc of ^_ 
its faint rays. They proved to have nearly lost their gaseooa ^M 
character ; the continuous streak yielded by them included 
only bare traces of bright liuea.^ The rule, in fact, seems 
general for temporary stars of reversion to a stellar type after 
a more or less prolonged nebulous interlude. 

Four new stars, discovered photographically by Mra 
flemiiig,— Nova Normie in 1893, Nova Carina in 1895»Nova 
Sagitt^rii in 1898.andNova Aquilseiu 1900,— followed closely 
in the track of Nova Auriga^ They deviated very little from 
the pattern of its spectrum* and they underwent similar 
Tnetamoiphoses, only vanishing more quickly and more com- 
pletely. They were then essentially phenomena of the same 
order, though probably on a reduced scale^ since the duration 
of a Nova's phases may serve as a rough measure of its real 
mf^itude. Clearly outbursts of the kind ate incidents that 
occur with method, and call for an explanation capable of 
being uniformly applied. 

The difficulty of finding one comprehenaive enough for 
the purpose was not lessened by the strange disclosures con- 
nected with the apparition of Nova Vex-sei. Early in the 
morning of February 22, 1901, Dr. Anderson saw with 
aatouishment that Algol had a twin-companion, which, thirty 
hours later, was super-eminent in the northern sky. It was 
not until Februaiy 24 that its light was distinctively that of a 
Nova ; but the seeming delay may have been due to its 
prompt discovery. None of its predecessors had been caught 
on the rise and at once spectroscopically examined. If they 
hiid, it would probably have been found that the gaseous 
blaze invariably needs some time to develop. It took place 
in Nova Persei precisely in the same way as in Nova Aurigse 

^ Lick Bulletin, No. 48 ; at Palmv, ibid. Kow 3fi -, Buiurd, Mb^Uhty 
IfoHces^ vol. liii. p. 418. 

' That of Koira Sfligitt*rii sbowod no conspicuoua d»rk linos wlien photo- 
graphed April 1&, 1893 ; but this was a comparatively late rscord ; it is funply 
|NOsatble that the usual cbianKscuro effects liad been apparent a month previaudy. 




TEMPORABY STABS 



n 



and its l^s noted imitators. That is to saj, the bright bands 

weie shadowed on their blue sides by beavy bars of fibsorption ; 

And these were shifted from their normal places by amouDta 

oorrespoiidiag, oa Doppler's principle, to the well-uigh ld- 

credible approaching velocity of one thousand miles a second 1 

The decline of the gbar was interrupted and irresolute, 

X)tiru)g March, indeed, the intemiptiona amounted to vivid 

spasms of recovery periodical in three days. Yet the loss of 

light progressed despite of them. By September 1-901 the 

once brilliant Nova had ceased to be visible to the naked eye.* 

The singular nature of the corresponding spectrum t^an be 

gathered from an inspection of Plate VHI., Fig, 1, in which, by 

the kind permission of Father Sidgreavea, two of the Stonyhuret 

photographs are reproduced. The differences between them, 

it shonld be noted, are mainly of instrumental origin ; so that 

the sum of what each records gives a fair picture of the 

emission^raya of Nova Persei after it had dropped below the 

aith magnitude. It had descended to the tenth in October 

1902; and its colour throughout that year was dull white 

or bluish,' whereas it had at fiist been strongly red. A 

Bptxitrogram taken by Mr. Perrine, July 30* 1903, when it 

was of about the twelfth magnitude, indicated that the nebular 

rays for some tiuae predominant had lost strength concomitantly 

with the usual restoration of continuoua Hght;^ and we may 

be sure that the equalisation will progress until every spectral 

trace of the strange cataclysm of February 1901 has becomef 

efifaced 

One of its accompaniments or conaequeDces was absolutely 
without precedent. A circumferential nebidit was partially 
photographed by Dr, Max Wolf, August 23, 1901 ;* and it 
came out fts & complete series of spires issuing, it might be 
said, from the Nova as their origin, in a Yerkes photograph 
of September 20. Later impressianSj secured at the two great 
American obaervatorieB, supplied evidence, the startling purport 
rf which was independently perceived by Mr. Perrine * and 

> Jost, JUr, Nach, No. 3811, 

■ Bftrsurd. Uor^tkl^ Noiisa, to]. IxiL p. 41S ; Rimbaut 1.1111 Williwna, ibidf 
Jtuu 1902. 

> LukButletin, No. 43. 
* Aftr. ^afJi. Nob. a73fl. 3762, . . fi3. 
' Lick Bulhliix, No. 30. 



94 



THE SYSTEM OF THE STABS 



Mr. Bitchey, that the nebula was expanding at a portentous 
rate, its awathed folds spreading outward, month by month, 
like the ripploa on a water-aurfmie disturbed by the tall of a 
Htone. Moreover, thia had been going on steadily from the 
initial date of the star's visibility. A Croaaley -reflector plaie 
was found to have been impressed in ten minutes, March 29, 
1901, with narrow nebulous coils, the obvious progenitors of 
thoBe fully disclosed six months later. For they were narrower 
in the computed proportion of the time elapsed, and would, at 
the Bame rate of e^cpanaioii, have started from their focuB in 
the star a few days before it? observed outbreak. In Plate IX. 
the Lick photographa of March and November are shown in 
juxtaposition. The augmented spread of the nebulosity during 
the interval can be gauged by using the small star to the 
north-weat of the Nova as a point for comparison. In the 
earlier picture it lies far out in the clear sky, like a rock near 
high-water mark at low tide ; in the second, it is heavily 
involved in folds of coamie cloud. 

Theories of the expanding nebula round Nova Persei had 
to meet arduous conditions The star has no appreciable 
parallax ; aad the indefinite remotenesa thus implitd for it 
gave almoet an assurance that the speed of the observed 
movements iu its neighbourhood could not have been infeiior 
to that of light transmission. Hence the idea occurred Bimul- 
tftneously to Professors Kapteyn and Seeliger, and to Mr. 
W. E. Wilson, that the motion concerned was ethereal, not 
material^ Wliat we saw resulted, they thought, from the 
progressive illxmiinatioa by the sudden stellar blaze of a pie- 
existent dark formation. The rays photographed were those 
of the Nova reflected from nebulous particles multitudinoualy 
strewn round its place. The supposition^ indeed, of a dark 
star lurking m the midst of a series of spherical shells of 
obscure filmy material "^ ia not one that invites ready aaaent ; 
nor did the observed movements within the nebula fit in quite 
satisfactorily with the illumination-hypotlieais. It had to 
encounter, besides, the difficulty that the light proceeding 

' Jsir. Na^h, NoH. 3756, 37&9 ; Nature, vol Uv, pp. IflS* SOS. 

^ The etf&ot, viewed ftom the earth, wculd bftre b«en that of a ji&rmholDidikl 
surfftco projected on a piano. S«q the demonstration hy O. Lu^ftics, Atiropk. 
Jowrn. vol. xix. p. 131. 



J 



TEMPOBAKY STAES 



9G 



firom the nebiUa was not polarised, aa it s)iould have been 
reflected ; and Professor Newcomb considered that the 
Lce of the nebulous spires from the Nova must have been 
Bi> great as to precludo the po&eibiUty of their shining per- 
eeptiblj by means of its enfeebled light.' Thus the weight 
of opinion ultimately favoured Professor Very's hypothesis* 
that the nebulous &tructm'e connected with Nova Peraei 
resulted from its actual emission of minute particles under 
the stress of electrical repulsion, or the subtle agency of light- 
p rean mre. 

Kova Gerainoruin was discovered by Professor Turner as 
an intruded star of the seventh magnitude ou a plate taken 
at Oxford, March 16, 1903. The Harvard College plioto- 
graphic archives were then consulted, and a record of the 
star's tna:cimum on March 6, at 5-0 magnitude, was extracted 
from them; while a negative of March 1, showing stars down 
nearly to the twelfth magnitude, preserved no trace of it. 
Mr. Newall found its spectrum^ on March 26, to be ablaze 
ftdth hydrogen and helium ; and a photograph of so much of it 
as could be brought into focus at one exposure with the Yerkea 
Bgfractor, obtained by Professor Frost, March 28, is repro- 
duced in Plate VIIL, Fig. 2, Its predominant feature la the 
strong effulgence of the Wolf-Rayet lazulite band ; the green 
hydrogen ray to the right is of only half its width ; the faint- 
nB9S of H7 (blue hydrogen) is partly due to its unfavourable 
situation ; but Ho was perceived visually to be of great 
intensity, and lent to the star its characteristic crimson tint.' 
Colour and brightness faded together, and very rapidly, the 
transience of the conflagration suggesting that it affected a 
comparatively small mass. The decadent Nova was observed 
by Professor Barnard, September 1, 1903, as a hazy star of 
11 '5 magnitude,^ yielding, it had been ascertained at Lick, 
August 17, a purely aebular Bpeetrum ;* and it will doubtless 
sink ere long to virtual extinction. 

The manifold experience of recent years has taught m 

' Aitr. Journ. No, 5^0, 

* A$tr. AS'ac/t. No. 3771 ; Amer. Journal qf Seifnct, vol. xri. p. 49. 
' Ytrkt^ BaUetiii, No. 19; Astrx^k. Joum. vdL xvlL p^ 376i 

• Turnerf Montfdy Nfflicet^ vol. IxiiL p. 5fi6. 

" H. D. Curtis, Lick £%U€tin, No. 48 ; PublwUions Aatr, Faeific Societ^f 
OCL 1903, 



96 THK SYSTEM OF THE STARS 

that a representative " temporary " specbrum passes through 
five well-marked stages. To begin with, it seeme^thongh for 
a very brief intervul — that of an ordinary helium-star. Next, , 
it displays a series of brilliant rays^ set off by dark eatellites, 
always of shorter wave-lengths. Later, blue bands emerge, 
and the light partakes markedly of the Wclf-Eayet peculiarities. 
Fourthly, it becomes ooncentrated Into the green rays of 
nebulium. Finally, it reverts to whiteness, and disperses into 
a dim, featureless prismatic band. A strictly methodical 
course of change is then traversed by tbe^ bodies. They 
undergo transformations of a prescribed kind in a settled 
order. Nor are any of thoir phaaea neceaaarily and essentially 
unstable, since each is exhibited permanently by the members 
of other sidereal families. The conditions to which Novee are 
temporarily subjected cannot, then, be adequately explained 
without reference to the fact that they are durable elsewhere. 
They cannot depend for their production upon the fleeting 
effects of catastrophes. This consideration seems to dispose of 
collision and explosion theoriea of stellar outbursts. For the 
displaced and coupled lines accompanying their vivid phases, 
which such theories are specially designed to account for, do 
not exclusively distinguish " new " stars. They occur as well in 
the spectra of permanent though peculiar denizens of the sky. 
The most familiar examples are P Cygni and rj Carina. Both 
these stars now shine steadily, though both underwent striking 
vicissitudes in. the pa^t ; and both show spectra perfectly similar 
to those of Nov(e near their maxima. Their remarkable quality 
of light accordingly corresponds to a state of things capable of 
persisting year after year, decade after decade. The causes 
which produce it must act uniformly and for an indefinite lengtli 
of time. They are not brought into action casually through 
some -momentary combination. Now thie inferenc-e obvioualy 
applies likewise to Novee. The spectral phenomena shown by 
them are, it is true, of a transient nature. Yet they are the 
same observed to be constant in other stars. 'No rationale, 
accordingly, which expounds them on the exclusive basis of a 
passing catastrophe can be true. 





The Nelmlnaity round Narii IVrw^i, 
1. lu Mink '2. Ill NtvvL'Ujl.er, UHJI. 






EEN two and three thousand stars are certainly or veiy 
probably variable, and known objects of the kind multiply 
with the more syBtematic use of photographic methods. In 
Chandler's Third Catalogue (1896)/ 393 were enumerated; 
aeven years later, 1309 figured in the Harvard Provisional 
Catalogue^ with its Firat Supplement,^ . In 1904, no les& 
than 407 new variables were brought to light by llifl» 

vitt'a comparisoQB of Harvard plates taken at diGTerent 
and detections go on apace^ both at Harvard * and 
at Heidelberg.* So far, only a small proportion (417) of 
the stars recognised as variable have had periods assigned 
to them ; with most acquaintance has been made too re- 
cently for the puipose ; yet a large residue seem entirely 
Uwless in light change. Periodical stars are divided into 
those with " long " and those with "short" periods. Nor is 
the distinction an arbitrary one. The 6tars seem to separate of 
themselves into two principal groups, undergoing fluctuationa 
in cycles of respectively less than thirty, and of 120 to 460 
days. The paucity of stars with periods of intermediate 
leogtha is shown graphically in Fig. 2, where the height of the 
curve represents the number of stars subject to changes pro- 
portionate in duration to the horizontal distance from left to 
right. 

Variations requiring several months for their completion 

» Aiir, Joumai, No. 378- 

• Harvard A-miah, vol. xJviii. p. fll. 

' Jfarvard Circuiar, No, 77. 



' lind. No* 9S. 




» M, Woir, Aftr, Jfach. No. 400C. 
97 7 



J 



98 



THE SYSTEM OF THE STABS 



differ both in degree and kind from those run through in ft 
few days. They are of much greater amplitude, ranging over 
five to eight instead of, at the most, two magnitudes ; they 
are accomplished with less punctuality; and they are fre- 
quently attended by symptoms of gaseous ignition almort 
wholly foreign to quicker vicissitudes. Most important of all, 
they affect bodies of peculiar constitution. Nearly all long- 
period variables are red stars with banded spectra ; those of 
short period are white or yellowish in colour, and display 
Sirian or solar spectra. Quality of light is thus the pro- 



50 


— 


■~ 


— 


-~ 


_— 


~~ 


— 


— 


— 


— 


— 


""■ 


ys_ 


- 


- 


^ 


n 


^ 


1 




^ 




















4B 




































J 


\ 
























1* 














r 


























/ 


\ 
































































f 




L 






















40 






































j' 




\ 




















































A 


^. 




/ 






^ 
















































> 


^ 


|l 


/ 
























































> 


r 






\ 








\ 




















a* 


























/ 








































C 331 
























i 
































































t 








































t; ojd 
























f 
























\ 
























—I 
















> 


























\ 








































/ 


























\ 
















1 *" 


- 


„ 


- 


- 


- 










. 


' 




















































/ 






























\ 
















= u 


■ 










J 


r 






























1 
















- 
















f 


































i 








■ 
















y 








































\ 


























/ 








































1 


























f 










































s 












^ 


h 








/ 












































_ 


S 


-^ 


^ 








\ 








/ 










































^ 




s 


_ 


^ 


f 






















































s 






-1 






" 
























































between 


60 


!6 

a 





5 
1^ 





1, 







fe 



1* 








to 


2 
Z 




r 


b 


3 


QO 
30 


3. 
3 


3D 
>0 


3i 
3 


10 


4. 




4 
4 


JO 

id 


4f 





4 
B3 


to 

LO 


MO 
540 



Flu. 2.— Distribution of 8S4 Periodit of Variable Staw. 

dominant factor in determining the law of stellar light-changes, 
and is itself dependent, as we have seen, upon atmospheric 
properties. The rule is, in fact, almost unfailing that short 
periods are attended by slight, long periods by strong 
absorption, and that the conditions producing redness in stars 
not only favour variability in general, but almost absolutely 
prescribe its type. 

Periods of between one and two hundred days may be 
called " long " ; but as Fig. 2 shows, they are not of plentiful 
occurrence. Such fluctuations as now engage our attention 
usually demand more than 200 days for their accomplish- 



STAKS VARIABLE IN LOKG PEKIODS 



99 



It, and are seldom prolonged beyond 450. Dr. Chandler 
riders 320 days as the average duration of change for long* 
iod variables;^ the prevalencej however^ among them of 
iod& of about one jeai is remarkable, and cannot be 
ited for by mere accidents of observation. Tlie first 
best known apeeimen of the olaaa. the members of which, 
January 1. 1904, numbered 397,^ anticipates by about a 
' aooth the rule of annual recurrence. 

When Bayer, in 1603, affijted in his cbarta the Greek 
letter o to a small star in the neck of the Whale, he had no 
sospicion of its identity with a supposed " Nova " which had 
di^ppeared seven years pre\'ioiiBly, after blazing up to the 
second magnitude. Its diBCOverer, on August 13, 1596, was 
David FabriciuB, of Osteel in East Friealand ; but though he 
»w the object again, February 15, 1609, he left it to John 
Phocylides Holwarda, pi-ofeasor of philoaopby at Franeker in 
Holland, to ascertain its true character in 1639 ; aud the 
repetition of the phasea in cycles of 333 dayjs was eBtablished 
in 1667 by Boulliau.^ The name '* Mira," bestowed by 
Hevelius upon the changing star in Cetus, commemorates the 
iaiazemeut excited by the detection of stellar periodicity. 

The phenomena it presents would 3eem incredible were 
they less well established. Once in eleven months the star 
mounts up in about 125 days from below the niuth to near 
the third, or even to the second magnitude ; then after a pause 
«f two or three weeks, drops again to it3 former low level in 
once and a half times, on an average, the duration of its 
rise. The brightest maximum on record was observed by 
Sir "William Hersehel, November 6, 1779, when Mira waa 
little inferior to Aldebaran ; * the faintest minimum, that of 
1783, is said to have carried it below the tenth magnitude- 
An eitent of eight magnitudes may then be aesigued to 
the oBcillations of this strange object, which accordingly 
emit&t at certain timeg, fully jif teen hundred times as much 
light as at others. That each maximum is a genuine con- 
flagration has been proved by Bpectroscopic ob&ervation ; the 
conflagrations recur yearly, with approximate regularity, and 

^ Aatr. Jcttm. Ko. Ifi3. 

■ ffarvard CirmlaTj No. 7*. * Monilwa ad Attronomot, p. 7. 

* Phil. Trant. t*]. In. p. 338. 




5:]G0K^ 



100 



THE SYSTEM OF THE STAES 



after three centuries of notified activitj, give no signa of 
exhaustion ! 

The height of the maxima, however, varies greatly. Th0 
consecutive ones of 1887 and 1888 (represented £rom Colonel 
Markwick's observations in Fig. 3) showed a nearly fourfold 
difference of intensity ; but Heis's remark that high and low 
maxima tend to alternate has not in the long run proved con- 
sonant with facts. There is no rule by which the brilliancy 
of impending phases can be predicted. That of November 
1868, in which the star just failed to reach the fifth magnitude, 
was, it is true, preceded by a high maximum, but several 
average or low maxima followed it. All that can be said is 
that exceptionally bright apparitions are isolated ; they do not 
come in sets, but one by one, at considerable intervals, — at 

Mae 
35 

4-0 

4-B 

S-0 

flJ S 10 15 

ias7 

FiQ, 8.— Tvo consecutive Hudmft of Mlis Get!. 

intervals, if M. Guthnick's conclusion proves correct, of 59^ 
years. The high maxima of 1779, 1839, and 1898 suggested 
this period, prescribed — it is thought possible — by extra tidal 
disturbances due to the periastron passages of a distant satellite 
revolving in a very eccentric orbit.^ The next brilliant phase 
need not be looked for, on this showing, until 1957. 

At minimum, Mira rarely descends far below the ninth 
magnitude. Fig. 4 portrays, from the observations of Mr. 
Stebbins,^ the gradations by which it lost and regained light 
in 1902-3. Although the star appears never to become 
actually inert, its changes during six weeks amount to no 
more than flutterings about the lowest level of brightness. 
The progressive reddening of its rays as they grow dim is 
held by M. Osthoff to be explicable as a physiological effect * 

» Astr. Nach. No, 3746. 

* Attroph. Joum, voL xviii. p. 346. 

s Aatr. Nach. No. 3940. 

















n 
























— 1 
































^ 


^ 


















/ 


'^ 


^ 


L> 


■^ 


"■^ 


\ 


k 
















s 


K 












f 












- 
















^^ 
























< 


















■^ 


























^ 


-.^ 




















^ 


■ 
13S' 


)^l 


t 1 


1 


s ^ 


2 


S 3 


3 


5 4 


4 


5 a 


18 


aa 


> \ 


1 


b 2 


2 


S 3 


3 


6 W 45 5tt 



H STAilS VARIABLE IN LONG PERIODS 101 '^M 

^Ipctf^ variatioQg are, nevertbekas, evident,^ and must cou- ^^M 

tribute to the result. ^^M 

The periodicity of Mira obeya a highly complex law. ^^B 

Beviations to the extent of a fortnight &0La the '* mean period '* ^^M 

of 331 days are commoD, and the maxiiaum of Septemher 29, ^^^ 

1840, wafi a full month late.^ Its perturbations may, indeed, ^^^ 

be themaelveB periodical, but if so, their law has not yet been ^^H 

saccessfuUy formulated. Aj^lander detected the influence of ^^M 

ft wave of disturbance with an amplitude of tweuty-five days, ^^M 

and embrar,ing eighty-eight periods ; ' Schwab's obeervationa ^^M 

indicated subordinate ripples of change In six and a half days ;* ^^H 

and there are half-efifaced traces of several oscillationa besides.^ ^^H 

Tet none are quite eecui^ly established Guthnick, howeverj ^^H 


4-0 


N 






































1 


^ 


s 


































ho 

•-0 






\ 


r^ 


s 




























/ 










S 


^^ 
























/ 
















'^ 


k 


















/ 






















^ 


^ 








^_, 


•^ 




























'■^ 


"^ 














JijweSO JuLvaO AUCL29 &(i>r.2a Oct.2S 1Nov^7 De«.27 Jk>j^6 F(ft.S5 HalS' 
Fio. 4.— Mlulmuai Of Xli* Oat), IWS4, 

haa entire confidence in the reality of an inequality coverin 
200 cycles (180 years), which is donbtleaa the equivalent t 
one suspected by Argelnnder with a period of 160 years. Tli 
tc»t pf verified prediction will eventually decide whether he i 
right The shape of the light-curve, too, varies notably. It 
peaks are aomelimeB much blunter than at others ; and th 
Btar» which usually retains its full lustre during a fortnigh 
has been known to remain twice that time stationary. Sti 
more aingulaTly, the otherwise invariable rule of an increos 
more rapid than the ensuing decrease waa reversed in 1841 
Sixty-two days were occupied in aacending from the eixth t 

^^fe 1 Astroph. Jourrt. toL xviiu p. 300. 
^^^^^^^^K ^ ArgeUtider, Astr, Nadv. No. 419. 
^^^^^^^^H ' BofVMr Bcohacht'angtn, Bd. vii. p. 332. 
^^^^^^^ « Attr, NaeA. N<i. 27S1. 
^^^^^V " Afgelancter ia Humboldt's Cosmot, vol. iii- p. 234. 



102 



THE SYSTEM OF THE STABS 



the third magnitude, forty-mne only in smking back to 
datum level The anomaly, due to what might he called 
unprepared retardation of the maximmu^ recalls the ahn 
course of the sunspot cycle which culminated at the 
of 1883. 

The spectrum of Mira is a splendid example of 
third type. Eleven bauds of profound shadow, sharp to' 
the violet, gently gradated towards the red, throw out ini 
strong relief the iutervening brilliant zones ; while dark lia 
of metallic abgorption, and vivid hydrogen -rays, vary the effei 
and add to the intricacy of the characters to be deciphft 
The more refrangible members of the hydrogen series are thoai 
chiefly brightened in this star ; no trace of C betrays itself U 
the moat attentive scrutiny ; F has never been seen, and i 
only occnaioually photographed,^ Its violet aBsociate, He, il 
also exceedingly dim, either intrineieally, or because of ttw 
obacuring effect of a coincident cakium-band The brilliant' 
phase of Miia in 1393 was attended by a curious triplicatioQ 
of the blue hydrogen -lines,* as if through powerful magnetic 
action. But the phenomenon has not recurred; hence the 
polarising esperimente, by which its nature could be estab- 
liahed, have yet to be tried. The dark lines in this 8i>ectrum 
are affected by motion-shifts corresponding to a recession of 
the star from the earth at the rate of 66 kilometres (40 miles) 
a second ; ^ the bright lines are much leea displaced. Yet 
there is no evidence that the absorption- and emission-raysj 
although they act thus independently, belong to distinct bodies. 
Mira is^ to all appearance, a slngly-coustituted star. 

The same may be said of a variable in the neck of the 
Swan^ which Bayer, ignorant of its changing character, set 
down in his maps aa of the fifth magnitude. It still retaim 
the name he gave it of " j^ Cygni" Missed by GJottfried 
Kirch in July 1686/ it reappeared October 19, and sub' 
aequently disclosed to his vigilant watch fluctuations even 
wider than those of the " wonderful " star in Cetus. It deacendf 
below the thirteenth and rises nearly to the fourth magnitude 

> StebbioB, Aatroph. Joum. vol, xviii, p. 3Q4. 

3 CunpbeU, ibid. tqL ix. p. S\. 

* StebbiuBr ibid. rol. iviiL p. 352. 

*■ Miaceltanta S*rolinensiA, U L p. 20S. 



I STAHS VAKIABTK m LONG PERIODS 

^kaetimea indeed stopping short when barelj visible 
^bad eyft, bub more commoulv temainini? lucid fot a 


103 ^^B 

to the ^H 
couplo ^1 


^B liiontb& iN'or is its course much better regulated as regards ^H 
^pna. "EzTor&" up to forty days often attach to its phaaes, ^H 
Hbd the attempt to correct them by the introduction of cyclical ^H 
Herms haa proved only partially Buc(^e3eflll.^ The period, ^H 
Hitimated at 402 days by Kirch, now averages 406. Gibers ^H 






1 






















































r 


V 














^ 


\ 
























f 




\ 


\^ 








/^ 


X' 




\ 






























N 






^ 








1 














H B-2 
















\ 


/ 






























/ 










\ 
























■ 








/ 










\ 


J 












\ 




















/ 
























\ 












f »« 
































1 








































\ 














J 




























\ 
















/ 




























\ 
















/ 




























^ 


L 












































^ 








108 




f 






























1 


I 








/ 
































Y 










/ 










































/ 










































11-4 

1 

noti 

and 
cipa 
aace 
it, o 

1 


















































































j 








































/ 








































^- 


r 


J'' 3S M 79 uiO 12S tao rr& soo 225 350 ats uo 32s 390 376 400 435 450 479 soo 

Fia. 5.— Llgbt-cniTS of R Normw. 

sed that it had b^n ete-adily lengthening down to 1818 
it 18 lengthenixig still. The compensatory process ant 
ted by him has not set in. As usual in such casea, th 
nt to maximum is much more rapid than the descent fror 
Octtpyittg at present about 171 days.* 

Tewwr B*6b, Bd. vii. p. 33{( ; Ma'n.^eiifut Jahreaberieht, Bd. xL {h ItO, 
_ ' Sobnmacber'i Jmhrbwh, iSil, p. S3, 
k * Chandler, Third Catalogue. 18&e. 


^1 

u ^^^1 




THE SYSTEM OF THE STARS 

The colonnaded spectrmn of x ^ff°' ^ appropriate to 
•oarlet colour. Moreover, OQ May 19, 188 9» when tb© star 
was near a maximtmi, Mr. EapiE perceived evidence in it of 
diiect radiation by hydjwgen which seems to he quite similarly 
conditioned to the blaie in Mira. The firet and fifth members 
of the series (Ha and He) are missing ; the second (H)3) i» 
perhaps leae evasively present. There is besides a discrepancy 
in the positions of the bright and dark syatemB of linee 
analogous to that perceived in the spectrum of Mira/ and 
obviously connected with the peculiar processes of light-change 
characteristic of such stars. 

The light-curve of E Norma is depicted in Fig, 5 from 
Mr. Lones's observations in 1898*99 at the Eoyal Obaerva- 
tory, Cape of Good Hope.* It is of a very unusual character. 




Fia- A,— Amu of ITtciiIe od tba Snn, 1B7S-[)f». 

Each cycle includes a pronounced double maximum, resulting, 
we can scarcely doubt, from the full and unrestricted develop- 
ment of a tendency, half- suppressed in most variable stars, 
to arrest the ebb of light at a certain interval after the 
culminating crisis. The vicissitudes of solar agitation, too, 
undergo a similar phase of hesitancy, as may be gathered from 
Fig. 6, which portrays, from the Greenwich observations, the 
virtually double facular maxima of 1882 and 1884, and of 
1892 and 1S95. The variability of li Norm© was first 
noted by Grould in 1871. An average period of 481 days 
is conformed to ; hut ext^ifiive divagations from it are 
patent. 

They are, however, insignificant compared with those of 

^ Ebnhvd, Atr, JVocA. Ko. 3766 ; Aatroph. Jtmm. ToL xriii. [h IftS. 

Annalt ofiht Cape ObaervaUiry, voh Is. p. 107 B. 



STAES VARIABLE IN LONG PERIODS 



105 



M«E 


- 








^ 


\ 


10-5 




\ 


{\ 


\ 


/^ 


\ 




\ 




y 




\ 




J 





















OJ bQ WO ■}bO ZOO 260 
Fifl, 7.— Lfghl-curr* of U Lupl <IonM), 



U Lapi. Indeed, thiB atar harely pretends to regularity. 
Sometimes, it is true, the eoidden failures of light distinguish- 
ing it succeed each other at intervaJa of about eighty-eeven 
days ; but an interruption i& sure to supervene, baffliog all 
attempts at anticipation, and the oBcUlatioas are then resumed 
aa if by a freah impulse. In Fig. 7 two auoceaaive minima 
are represented — one complete, the 
other aeemingly abortive- Mr. 
lo&ee diacovered this singular 
object in 189 8>^ through its unex- 
plained absence from the Cape 
DurchmuBteriLng plates, and 
vatched the capricea of its in- 
stability durlDg three years. 

By fer the longest period 
attributed, on reasonably 
good evidence, to a variable star was assigned to e Aurigee 
bj Dr. Lttdendorff of Potsdam in 1903.* As a white star 
with a spectrum resembling that of a Cygni, it should not be 
predisposed to instability ; and the amplitude of its changes 
is, in fact, of no more than aeven-tentha of a magnitude, 
implying a reduction of brightness by one-half. Moreover, 
they take place very gradually, and are rarely observed. For 
the most part, the star shines with approximate constancy at 
'i 3'3 magnitude. Its phases of dimness were believed to occur 
quite irregularly, until Dr. Lndendorff showed, by an exhaustive 
discussion, that they are subject to a periodicity of slightly 
more than twenty-seven years. Decline and recovery occupy 
thirteen or fourteen months ; an intervening stationary time 
of least light lasta just ten; so that the entire oscillation 
needs close upon two years for its accomplishment. Three 
were recorded in the nineteenth century, a fourth in 1901-2. 
^^b renmins to be seen whether futui*e minima will honour the 
^Bl^^ifis of calculation as they fall due. Illusory effects of 
^periodicity have often been perceived in stars fundamentally 
irregular, and e Aurigte may still evade the law of order 
which it has temporarily obeyed. Its character as a spectro- 
scopic binary, established by Drs. Vogel and Eberhard in 

^ Aslr. Journal, No. 143. 
' ji9tr. Nach. Noa. 3918-20. 



106 



THE SYSTEM OF THE STARS 



1902,^ cannot well be supposed extraneous to its luminoaa 
T&riability ; but the mode of cotmection remains enigmatical, 
atEice the oocurrence of edipBes ia preeludeil by the eonditionB 
of tnovemeat that appear lo prevail in the system.^ 

We now come to that unique star, r^ Caiinse, Its actual 
appearance is insigolficaDt. Invigible to the naked eye, it is 
teldBoopically distinguished only by it« leddiab colour and 
slightly superior brightness from the crowd of Bmall etats 
embroidering one of the finest of the southerti nebulffi, named 
the " Key-hole Nebula " from the aperture of that ah^pe with 
which it is centrally perforated. Close to one edge of the 
aperture, in the densest part of the nebula, tj Carin^e is placed 
Nor can we suppose its poeitioa fortuitous. There is every 
reaaon to believe the star to be really plunged in nebuloufl 
substance ; and the peculiarity of its enviromuent combines, 
we must suppose^ with peculiarities of constitution to produce 
the exceptional character of its changes. 

The first observation of 17 Carinae was made by Hatley at 
St Helena in 1677. when it was of the fourth magnitude; 
had it been as bright in the second century A<D., Ptolemy 
would presumably have recorded it, since, as Mr. Innee points 
out.* the chief stars in Crux and the Centaur, which culminate 
about the same altitude, are included in the Almagest. 

Soon after the youug English astronomer n^ade his 
hurried survey of the southern sky, the variable had a notable 
accession of lustre. Pfere Noel, a Jesuit missionary in China, 
rated it as of the second stellar rank, 1685-89,* and so it 
appeared to LaoaiUe in 1751; yet the discrepancy with 
Halley's appraisement remained unnoticed. The higher 
estimate was confirmed by those of Fallows, Brisbane, and 
Johnson in 1822, 1826, and 1832 respectively, au intervening 
decline having been noted only by the traveller Burchell, who, 
familiar with the star as of the fourth magnitude in 1811-15, 
was surprised one night in 1827, at San Paolo in Brazil, to 

it temporarily raised to a level with the finest brilliants 



' SitzungsbtrichU, Berlia, November 27, 1902. 
- A. M. Gierke, Oiatrvaioryt toI. titu, p. 118. 

' Annah Cape Obftrvatpry, Tol. ix. p. 75 B, Henderson lb referred ta ei : 
auttaur af tbe remark. 

* Wiimeeke, Attr. Nach. Ko. 1^4 ; Klein, SiHm, Bd. vi, p. 2&B. 



STAKS VAEUBLE IN LONG PERIODS 107 



of the aky. Another, and a atUl more vigorous outburst, 
was witnessed by Sir John Herschel, December 16, 1837. 
■Without previous note of warning, the star all at once 
iieftrly tripled ite Ught, and before the end of the year fully 
tu&tched a Centauri. Since then it has been kept under 
strict Hurveillance as a notorious charaoter^ and not without 
reason, After a partial decline and seveKil preliminary 
•* Sutterings," it reached a final maximum m April IB-IS, 
when Sirius alone among the fixed stars slightly outshone it. 
This high position was moreover fairly wt:il maintained for 
a^ or ten years, Gilliaa, at Santiago in 1850, found it very 
'Httle inferior to Canopua in liglit, and in colour more deeply 
tinged with red than Mars.^ Still of the fir&t magnitude in 
1856.* it fell to the second in 1858, to the third in 1859, 
and ceased to be visible to the nalied eye early in IS 68.* 

For sixteen further years the slow ebb of Ught continued, 
and the magnitude of the once e£^gent orb* carefully 
determined by Mr. Fiulay at the Gape, was in March 1886 
only 7 '6.* This proved to be the beginning of a stationary 
minimum of indefinite duration. The star hag suspended its 
fluctuations, and it is impossible to say when it may resume 
them. Quite piobably ita history is one that " does not repeat 
itself," Our coutinuouB knowledge of it is embodied in the 
accompanying diagram (Fig. S). A single vast oscillation is 
indicated, occupying about a century for its completion, and 
diversified by secondary fluctuations of a very conspicuous 
character (innumerable minor ones ai-e ignored in the figure). 
The data at present available^ however, afford no grounds for 
concluding this oscillafcion to occur regularly. Attempts to 
aaeign a period to the variations of this object have signally 
failed. Wolf's of forty-six,^ and Loomie'e of seventy years* 
were both palpably too short ; larger time allowances encoun- 
te'rfid obvious difficulties ; and y CariuiL^ was, by general con- 
sent, abandoned to ita own lawless cotii-sea. Protracted periods 
of light-change have, in several other easee^ been suggested, 

^ Abbott, Monthly yoiicM, voL xxi. p. 230. 

* Moe»t4, Attr. yaek. No. 1054. 

' Tebbutt^ MonOUy Nolwt, vol, mi. p. 210. 

* MoiUKty NoliCAt, Tol, ilvi. p. 340. 

' Ibid. Tol. iiUL p. 20&. 

'■ Ihid. ToL siii. p. 2e8. 



THE SYSTEM OF THE STABS 




bul Iwve in lume been verified^ onkn I>r. Lodendorff^s 
tveoty-seTen jean fisr c Ami^ prove an exception. 
stvB thej were aactibed to, wben Ukft Um£ caotc for a 
tioo of their preemned cjclical changes, sbowed a total 
of oonibnuity with what wae eipect^i of tbem. As eram| 
may be mentioned 63 Cygni^ to which Mr. Eepin attribatedj 
period of five, and R Cepbei, tbooght by Mr, Pogson to ol 
one of seventj-three years. 

Thus the gro&t southern variable cannot be depended 



I 



Mas- 


n 
























i 






^ 




























J 


^. 


s 












































f 






s 


























•04 
















/ 








\ 






















1 










. 






M 


f 








\ 






















1-4 








h 




J 










1 




















_1 


30 








J\ 


./ 


' 












I 




























, 


y 




















\ 


















































\ 
































1 








































/ 




















'i 






















*^ 




' 








— 1 














i 






















XM 
























\ 


































































































\ 


















































^ 


v^ 


















































V 




















































— 


— 


— 


- 


















































1 < 


k « 


i 


li 


' i 


li 


f i 


M 




!: ^ 




n 


M 


\i 


1 ( 

F - 


il 


' 4 


'2 


I 


li 


i 


!i 


j 





upon to revive its past splendoors. There is nothing inevit- 
able about the kind of fluctuation it exemplifles. One might 
indeed any of it, in Jtick Cade'8 phra&e, " then is it in order 
when it is tnost out of order/' IrregiUar light-change seems 
to develop all its resources in the vicissitudes of ij Carinte. 
They have included quick, yet sustained ascents in brightnesa, 
and also evanescent kindlings. Stationary epochs huve been 
followed by epoclis of inetabiUty ; at some times the star has 
shown a tendency to establish itself at halting-places, at others 
to slip along an inclined plane of change. In all this it 
differs materially from temporary stars, which leap up, as if 



STAES VARIABLE IN LONG PERIODS 



109 



hj a single impulse, to their Bolitaiy maximum, after which 

ihsj loae m a few months the whole of the light they had 

•equired. The nature of its light, on the other hand» 

jasimilate& it to the temporary clasa. Sir David Gill's study 

photographs takea with the McCleaa apparatus in 1899 

QELStrated a close agreemeDt l>etweea the speictra of the 

aoaldering variable and of Nova Aurigse when near 

lunL* Hydrogen-lines, broad and bright, with dark 

apanions on their more refrangible sides, were the leading 

iture of both; and a feature no leas enigmatical in the one 

than in the other. Yet the fact, as already remarked, 

of profottud importance that the conditiousj transiently 

praeent in Novte have been rendered permanent both in 

ff Caiinse and in P Cygni, the qiiaai-Nova of 1600^ now an 

imexceptionally steady fifth-magnitude luminary. 

Among the miscellaneous objects comprised in Pickering's 

class of variables are some slightly changeable third- 

brilliants. Sir William Herschel added in 1V95 a 

HeixuHs to the list of eeven fluctuating stars then known* 

A period of two months assigned by him to its oscillation 

between 3'1 and 3'9 magnitudes has proved illusory. During 

some years the swing appears almost to ceaae, then is 

hurriedly resumed, but with no settled order. The analogous 

rariationg of Betelgeux and ^ Pegasi are equally unmethodical. 

A. conspicuous brightening of the former star attracted much 

attention in the autumn of 1902, but speedily subsided. 

^_ The extraordinary character of a star long known as 

^V Variabilis CorouEe/' now called " R Corouse," was discovered 

^Hy Figott in 1795 ; a near neighbour of the "blaze star'* of 

^■1366, its changes are of the nature of extinctions rather than 

^^f outbursts. Ordinarily of 5 '5 magnitude, it occasionally 

drops out of sight with small telescopes, and after lingering 

below the tenth, or even the twelfth uuigtiitude for many 

months, slowly regaiiifl its lost light. But its phases, at times 

suspended, as during the seven years, 1817-13 24,* are 



^tiid 



1. ' Monthly ybtiutt toL IxL p. 456, App. iv.; cf. Earvavd AnnalSf vol. 
nriii. p. 175 (Miu A. J. Cunnoii), 
' Phil. Trans. toI. Ixxivi. p. i52. 
* ArgelAucler, Soniur Biob. Bd. »'ii. p, 374 ; Olbera, Berlinf-r Jahrhixh, 




110 THE SYSTEM OF THE STABS 

&l others iU-Kkuked. Thxa at tfa« v&mimiuii obaerred bj 
Sawyer October 13. 1885, the etar waa ettU of 7 4 magni- 
tade.^ It abows ft peculiar spectrum oeeditig closer in- 
vestigation, 

B Cepbej is a star which, ^ce the beginning of thft 
present century, has loBt JJ of its ladiaace, aad at preiieiit 
in no way tends towards rec<»verr. In the time of HeveliuB 
it was of the fifth magnitude, and Grodmbridge's observati<m 
of it in 1S07 showed it to be then stUl unchanged. Bj 
1840} however, it bad sunk to the tenth, and has never aiitoa 
ii»ji aJ:*ove the eighth magnitude.* Its identity with the 
" 24 Cepbei " of Hevelius, itacked out by Pogaon in 1856/ is 
tmiversoUy admitted. Of late years it has shown no eign 
of variability. Its light, considered by Schonfeld to be 
tinged with red, appeared bluish to Farley in 1838* 

Genuinely red stare are ordinarily subject to shiftings 
of photometric staading. Among twenty - two such, kept 
in view by M. Safarik at Prague, from 1883 to 1888, 
for the express purpose of testing their constancy, only 
nine remained without noticeable change, two were found 
periodically, six irregularly variable, and five either vanish^ 
or lost great part of their light. Earlier observations of 
several of these objects certified the progress of their decline 
during twenty to twenty-five yeara.^ An example of a sudden 
acqtiiaition of lustre is a£forded by a small red star in the 
same field of view with y Cygni. Between December 1886 
and June 188G, Mr, Espin perceived it to have risen iu rank 
by a whole magnitude,^ that is. to be giving out two and n 
half times as much light as six months previously. And so 
far as is known the gain has been kept. 

The track of recent astronomical progress is strewn with 
the JiUpidated remnants of hypotheses invented to explain 
the strange phenomena of stellar variability. Nevertheless, 
much has been learnt as to their relationships and essential 
natiire. The gaseous incandescence^ for instance, visible in 

1 Jttr, Joum. No. 151, 

« Bcb<JiLfeld, Mawtheimgr Jahrabmcht, Bd. xL p. 118 ; Oore'a Oataloyue^ 
08B4J, p. 200. 

' Monthlif NoUeett toL xtiL p. 39, 

• Aitr. Jfae\, No. W74. 

' JourTtal Liv. Atir, iSoc toL v. p. 2. 



4 



STAES VABIABLE IK LONG PERIODS 111 

Ibe spectra of periodic&l »lara near their maxima brings them 
■into 8nch close physical i-elationship with temporary stoi'S as 
: absolutely to prohibit the speculative separation of the two 
kinds of change they respectively exhibit A theory stendfl 
eelf-condemned which deals with them on diHereut principles. 
Moreover, the association of variability with processes of 
liuniooiis change in stellar atmospheres has been rendered 
obrioiiB; and this at once diapoaes of darkening expediente, 
M by Blag-fonnation, vaporous obBCuratioQs, and axial rotation, 
bringing bright acd dark sides alternately into view. Equally 
inadmissible is the rationale of stellar fluctuations by inter- 
mittent chemical associations and dissociations at the atmo- 
spberic outskirts of cooling bodies.' For the increase of light 
&t maximum demonstrably ensues upon a real access of 
incandescence, and is not a mere appearance due to the 
dissipation of absorbing vapours. The fires really die down 
and leap up at regular intervals ; they are not merely screened 
oCr and disclosed. 

Attempts have several times been made to explain the 
periotlicity of stars through the influence of satellites revolving 
round them in highly ew^ntric orbits. Klinkerfues suggested 
great atmospheric tides, raised at successive peribeliou-paasages,^ 
as a means of bringing about periodic interceptions of light. 
Plaasmann's ' view of tidal effects is wider, and perhaps 
embraces a partial truth. For, just aa in the earth the 
unequal attractions of aun and moon on its centre and 
surface sometimes provoke, though they could not produce, 
earthquakes, so the tide -raising power of bodies making 
very close approaches to stars in a critical state of heat- 
equilibrium may serve as the occasions of lujuinous outbursts 
of a temporary or recurrent nature. 

Sir Norman Lockyer's meteoritic hypothesis included the 
pregnant idea that variables are to be regarded as " incipient 
double stars." * But it had a different application to that 
imagined for it by its author. There is no sound reason for 

1 Br«Hter, Sssai d'un< TVoWe du SqUU ti /ka £tmUt Variahkt, Delft, 1880. 
* OoUingiteAe ^achriehUn, 1&S5, p, S ; aee also Dr. WilaiDg's conunents Id 

> Did Veriitiderlichtn Bemty Ebla, 18S8. 
' Ftoc Soyal ^eUty^ toL xliv. p. SO. 



112 THE SYSTEM OF THE STARS 

believmg stare like Mira to be oomposed of twin meteor- 
flvanns blaitiig tbrougb pemstral ooUlsione. That tbeir 
lommoeity maj be affected hy the tide-raismg power of 
UBseeu companioDfl, it would indeed be rash categorically to 
deny ; yet thero is no evidence to support the opiniou, 
while the periodicity of the objecta in question is of so diS' 
turbed a kind as to raise almost insuperable obstacles agalEBt 
connecting it with moTements necessarily punctual. More* 
over, periodical cannot be sbaiply divided off from irtegd&t 
variablea Every degree of perturbation, up to the total 
Bubversion of laws of change, lb met with among them. Man)r 
stars seem at timsa diapoaed to conform to a period which 
they later ignore* In others, method is indicated, though too 
vaguely to be defined, while the majority oscillate, with wide 
allowance of amplitude, about a period itself often subject to 
periodical or secular change. It is evident that the immediate 
and unmodified interaction of revolving masses cannot explain 
breaches of regularity widening out to its total destruction. 

The time has scarcely yet come to formulate a general 
theory of gteltar variability, but we may, at any rate, try to 
render our ideas on the subject coherent, admitting provision- 
ally those that are consistent with known facta, rejecting 
summarily those that contradict them. It will then become 
possible to realise with some distinctness the conditions under 
which alone any such theory could be regarded as adequate. 

As long ago as 1852^ M. Rudolf Wolf adverted to the 
analogous character of the curves representing sunspot 
frequency and stellar light-change.^ They are not only of 
the same general form, but they are marked by precisely the 
same kind of irregularities. Both are steeper in ascent ttian 
in descent ; both rise into peaks of unequal heights at unequal 
distances apart. Mira, ^ Cygni, K Hydrae, and the rest, have, 
like the sun, retarded and accelerated, high and low or abortive 
maxima. The representation in Fig, 9, from the Greenwich 
observations, of the changes in sunspot frequency during the 
decennial period 1867-1877, is the very counterpart of the 
Ught-curve of a variable star. Especially charaeteristie is the 
break in the descending branch reflecting a partial recovery 



1 ifiUhnlitngen Naiur/orKh. OueilaeJui/i, Bam, lSb% {>, 201. 






STARS VARIABLE IN LONG PERIODS 113 ^H 

i& the downward course towardB miaimum, to which variable ^^^k 

Etara of HifTerent classes are proue, and which maj even^ as ^^^k 

ttk R Nonn^, assume the importance of a Becoud co-ordinate ^^| 

mudmum. Moreover, the How of change in sun and stars ^^| 

alike is broken and disturbed by the superposition upon the ^^| 

normal period of subordinate and superior cycles ranging from ^^H 

i few days perhaps to centuries, ^^H 

The presumption tlien of their similai' ori^u is very ^^H 

sLrung: nor are we wholly without evidence of a physical ^^H 

nature to the same effect The development of bright lines ^^H 

' in the spectra of variable stars near their maxima is paralleled ^^| 


1^ .» 








jTX 


























f 
















=— 






f 


V 


















} 


V 


















A 


f 


\ 


















{ 




y 


f^\ 












W feo 






1 






\ 








■ 1^ 






/ 








V 




' 




1^ o 






^ 








\ 








■■ »!^ 














\ 










■ ^40 




} 








\ 








^1 £ ^" 




/ 












L 






■ ^^° 




/ 












\, 








f 


_ ... . 








\ 








/ 
















^ 


=— 


*-• 


















^^^H 1807 (558 1B6Q TQ70 1071 1fi72 1873 1574 1375 1678 1877 ^^H 

^^^^B Year* ^^H 

^^^^^P Fto. 0,— Gnrv* of Soiupot Fnniucncy, 1347-77 <BlllA> ^^^| 

in the sun by the iucrease of emissive intensity in the oorona ^^| 

as eunspots increase^ as well as by the attendant development ^^| 
of calcium Hocculi shining by direct radiation. Atmospheric ^^| 
incandescence is thus in both cases heightened, although in ^^| 
irnrneusely different degrees: and continuation is afforded to ^^| 
what was already certified by the congruous shapes of the ^^^k 
two curves, namely^ that the maximum of spots in the sun ^^^k 
corresponds with the maximum of light in stars, and iiic^ ^^^k 
verad. It is the more necessary to bear this in mind, because ^^U 
obscurations by spots have sometimes been alleged as a cause ^^| 
of stellar variability. That just the opposite is the truth has ^^U 
been further certified by a beautiful mathematical investiga- ^^^W 
tioQ set on foot by Professor Turner in 1904.' Submitting y 1 

^^k ' ifvnihly N^tieatf ToL bdr. p. S13. ^^^| 



THE SYSTEM OF THE STARS 



to the methods of harmonic analysis eleil&r Ught^urres and 
the curve rej'resentiag solar cycJiDul activity, he obtained 
cc9ults of emphatic import aa to the identity (it might be 
said) of the constructive principle at their base, provided that 

maxima of spote be regarded as the coiTelatives of maxima of 
brightnesa. But tba agreement vaniahed when the trial was 
made of inverting the reUtionahip., i 

The conclusion that Bolar and at^llar disturbances are alike 
in kind clears the ground for further investigation, for, 
beeidea obliging u8 to reject cauBed for the latter which 
are demonstrably unconcerned with the former, it renders 
snnapot studies tlirectly available for Bolving the problem of 
stellar variability. Now what do we really know about the 
production of sunspots ? Not much more than that they arijs^ 
incidentally to the great cii-culatory process by which photo- 
flpheric radiations are maintained. They arise, beyond question, 
when it ia most active and tumultuous. Nor is there anything 
to show that the variation in their niuubers depends upon au 
external causa It seems, on the contrary, to result &om 
peculiarities inherent to the solar constitution — from the in- 
tncate movements proceeding withijj the vast globe, and 
accommodating themselves somehow to those due to the 
swirl of its rotation — from fluctuating relations of heat and 
pressure — from the alternate accumulation and discharge 
of explosive forces, perhapa of a molecular nature. Similarly, 
the secret of stellar light-^ioiBsitudes is held by the stars 
themselves, although superinduced modifications may also, in 
some cases, be recognised. But that their periodicity is 
essentially self-regulated becomes manifest through the con- 
sideration that it is materially influenced by colour. Not 
only a very large proportion of red stars are variable^ but 
nearly all variables of long period are red. The length of 
the period^ too, ia very distinctly connected with the intensity 
of the colomr. This was first noticed in 1873 by Dr. Schmidt 
of Athens;^ it was reaffirmed by Mr. S. C. Chandler, who 
concluded, after an elaborate study of all the facts, that " the 
redness of variable stars is, in general, a function of the lengths 
of their periods of light variation. The redder the tint the 
longer the period." '^ And Mr. Yendell has quite lately 

* A^iir, Nach. No. 1SS7. » At(r. Journ. Nos. ISO, 1»3. 



STARS VARIABLE IN LONG PERIODS 115 

leached a similar conclusioa^ Many individual exceptions to 
the rule might be cited. But it prevails in a large sense ; and 
its prevalence enforces the obvious truth that the explanation 
of redness in stars lies very close to the explanation of their 
vambilitj in long periods. 

1 A^r. Joum. No. S64. 



CHAPTER iX 



TARUBIR STLRS QF SHORT PSKIOD 



We have seen, in the lost chapter, that stara vajyiiig their 

light in perjotis of leas thau thirtj' days stand apart in 
several important respects from those undergoing slower 
changea. The distinction \a aocentuated bj the tendency 
apparent in each claae to group its members as far as possible 
&om the frontder-line of separation. Thu^j most long periods 
exceed two hundred days, while a large majority of short 
periods fall below eight. The total number of stars so far 
found to be variable within a calendar month is eighty 
(besides a multitude of faint objects crowded tt^ther in 
clusters); of these seventy complete an oscillation in less 
than ten days, while sixteen bare periods measured by hours. 
Fig. 10 gives a graphical conspectua of these facte. 

Variables of short period are, as we have said, nearly all 
white or yellow stars, showing spectra of the Sirian or solar 
type. They fluctuate much less extensively and much more 
precisely than Mira-variable-s. In many rapid stara the light 
ebbs and Hows like clockwork as to titue^ and as to measure, 
viitb deviations scarcely of the tenth of a magnitude ftom a 
settled standard. These remarkable changes progress gradually 
and continuously in Pickering's fourth class of variables; in 
hiB fifth class they only interrupt, although at perfectly regular 
intervals, the usually steadfast shimng of certain stars. Of 
these two kinds the former is conspicuously exemplified in 
Lyrse — a star of which we have already made the acquaintance 
in connection with its gaseous spectrum — the latter in Algol. 

Further distinctions, however, have to be mada Class IV, 
really comprises three separate families, which may conveni- 
ng 



■ 


^^ 


■ 


n 


IRfW 


1 


n 


W 


1 


p 


i 


P 


i 


1 


!t 


TESTi 


i 


■ 


■ 


ff 


n 


■ ently be designated as Cepheid variaMea. Cluster variables^ and 
I Geminid -variables. Their several characteristica we ehall now 
1 briefly indi(?ate. 

H One peculiarity, full of meaning in itself and in its 
m implications, is common to thero all. They are, probably 

■ without exception, close binary eysteme revolving in the 

■ period of light-cbange. Tbia was long ago suspected of atara 
1 undergoing brief phases of obscuration ; and the fact, as regards 

^^L Algol, was definitely ascertained by Dr, Vogel in 1888. Aboat 
^^M thirty analogous objects ars already known, and there ia slight 


^1 












































1 








































































































































































- 


— 


— 












































i 


I 




































^1 








/ 


\ 










































/ 


^ 




j 


i 
































1 




V 


1 








































N 


f 




\ 










































\ 












































V 


/ 


I 






































V 


\ 












































) 




J 


V 








/ 


V 






























s 


/ 


> 


/ 








N 


/ 


S 




















































Under 1'^ 2 3 4 S b 7 B d 10 11 t2 13 {4 15 16 17 IB 19 2fl ^^^H 
Fta. la—DtKtiibaUoQ of 60 Vuiible-Bur Periods under 30^. ^^H 

ri^ of ertor in describing theiu genericaUy as ** eclipsing atata." 
"Far more surprieiug was the discovery that stars fluctuating 
in a manner inconfiistent with the eclipse hypothesis shared, 
nevertheless, their compound nature. Take^ as an example, 
& Cephei, the light-curve of which is depicted in Fig. 11. It 
is by no means Bymmetrical. The ascending is much steeper 
th&n the descending branch, and the latter is besides markedly 
f inflected. Now the great majority of sporadically occurring 
short-period variables belong to the type thus illustrated. In 
many, it is true, the pause in the decline from maximum is feebly 
accentuated or imperceptible ; but most gain brightness about 



118 



THE SYSTEM OF THE STARS 



twice aa quickly as they part with it, and all accomplish their 
changea by uuinterrupted gradations. They are continually 
on the moTe ; tliey have no definite halting-places, either at 
maximum or at niiuimum. This mode of yariation is clearly 

irreconcilable with eclipse-conditions, and the anticipations of 
reason have been confirmed by experience. Yet the objects in 
question are, in point of fact, binary stars, and their revolutiosa 
Btrictly control their light-changes. This is known from the 
synchroniam of the two kinds of observed effect— of the 
spectroscopic alterations due to orbital movement, and of the 
photometric periodicity. The variable radial velocity of 
S Cephei was detected by Belopobky in 1894;^ and that of 
7} Aquilio, a star similarly Tariable, in the following year.* 
In both systems the companion -body is obscure ; in both the 
occurrence of eclipses is precluded by the circumstances of 



Mat 






/ 


'\ 


V 
















f 




\ 














j 


' 






^ 


■^ 






At 




/ 












"^^ 




4. 

+■7 


[y 
















^■^ 


• 


H 








j ^ 


i 




4 


b 



Fio. 11.— Liglit-OlUTo of DoltoCeptini. 

movement. Both atar8 give spectra of the solar type. The 
further recognition, aa spectroacopically double, of the Cepheid 
variables, X, W, and Y Sagittarii,^ of T Vulpecula? * and 
S Sagittse, adds weight to the accumulating evidence that 
the peculiarity may be generalised. Thei-e need be no hesita- 
tion in affirming that the pattern of variation set by B Cephei 
ia prescribed by the circling, in an identical period, of a 
usually non-luminous companion. 

" Cluster variables " are met with by the score in certain 
globular clusters, and scarcely at all in the open sky. Hence 
their current title. Their discovery by Professor Bailey in 

* Aatr. Nach. No. 3257 ; Jatroph. JourK. vol. t p. 160 ; Bull, de L'Ac 
ds St PiUrthouTQ, Not. U94, No. 3, p. 288. 

* Atiroph. Journ, rols. vi. p. 3&3 (B^Jopolaky) ■ ix. p. 59 (Wright). 
■ Sli|)het^ B-uileUn of iht LovMii ObservaU/ry, No. 11 j B. H. Ciuttss, Lick 

Bull No. 62. t 

* Frn«t, Attroph. Jonm. vol. xx. p. 206. 



iCffl^^ 




VARIABLE STABS OF SHORT PERIOD 119 

1895 remarkably illoatraled tlie perfection to which the, 
photographic method has been brought. To individualise th*i 
minute^ throagiiig components of compressed clustera would, 
outil lately, have been regarded as «. notable feat ; to follow 
their variationa of lustre through brief cycles of about twelve 
hours, and to determine their special character, might well 
hare seemed impossible. The camera alone is competent 
to undertake work at once so delicate and so comprehensive. 
With two or three of the most powerful telescopies in the 
world, these tiny Ughb-apecks can, indeed, be observed to 
good purpose (^ Professor Barnard haa shown) ; but only 



Mac- 

13-0 


V. 


















\ 


<, 








{"] 


\ 






14*0 




\ 


s^ 




J 


N 


b 




53^ 




^^ 




/ 












' 


! • 


r ' 




1 


1 


' 
























13*0 




















^ 














/^ 


\ 


HO 






'^ 


^ 


- 




/ 






















\ 


W - 


r -i 


'. • 


J 


*' • 


b ■ 


i 


f . 


i -^ 



Flo. 13.— TyTdcoJ Lighl-Can-H ol Clii«t«r VarUbles. 

one by one^ and they demand wholesale treatment. Some 
etar-globes contain shoals of variablea ; over 500 have so 
far been regiatered, besides upwards of a thousand in the 
Magellanic Clouds,' The variations of stars in clusters, mote- 
over, are by no means vague or indeterminate. They are 
executed with punctuality and precision in perioda^ very 
generally, of leaa than one day. Three light-curves, two 
typical and one individual, are shown in FigS- 12 and 13. 
They are ci>pied from Professor Bailey's dravrings illustrative 
of his elaborate discussion of the conditions of variability in 
the great eouthem cluster to Centauri.' And they are, in all 

' Pickering, Harvard Circ*ilar, Noa. 82. 9&, 100. 
* Uarvani Annals, toI. xxsviii. 1902* 



120 



THE SYSTEM OF THE STARS 



13-0 



13-6 



14-0 



0^ ,t -2 ^a '4 -5 -fi -7 
Fid. 19.— LletiMTorre oJlTo. 4i Onega CeDtcari. 



auch groups, strangely imiform. A swift riae to maiimum, 
and & proloaged halt at minimum, are their lading tmita 

Cluster variables are dis' 
"^'^^ ^ tmguished from Cepheid 

vnrjablea. partly by the 
auddeimeaa of their up- 
springing, but most eesen- 
tially by the relatively 
protracted duration of their 
dim phases. About half of 
each period is gpeut at a dead level of low light ; then all at 
once tripled brightness supervenes, to ebb away again by slower 
gradations. 

Why stars thus singularly afifected should so strongly tend 
to herd together, none can at present attempt to divine ; but 
the exceptions to the rule of aggregation claim particular 
attention^ if only for their rarity. One of them is S Arse, 
noticed as variable by Mr. Innes in 1898.^ The curve by 
which he delineated its course of change is copied on a reduced 



Mac. 

9-0 

9-2 










j 






/> 














I 






r\ 




\ 














1 




' 


\ 






\ 










/ 














V 


■-^ 






^ 


/ 


































^ 


i 


> • 


I 


1 


( 


i ^ 




I 


^ 


D 1 



Fio. U.— Ught-OurT«i>f S Ana. 

scale in Fig. 14. Scarcely distinguishable from one of Bailey's 
typical tracings for the variables in q> Centauri, it shows a 
stationary minimum lasting four houra, then a more than two- 
fold increase of lustre within an hour and half, followed by a 
leisurely decline in approximately five and a half, by which 
the cycle of nearly eleven hours is completed, Y Lyrse is 
almost the alter ego of S Ats&. Mr. Stanley "Williams deter- 
mined its period to be 12^ 4^^, the riae from 11-3 to 12"3 



' Capt AniuiU, vol. ix. p. I2Gs; 
Monikly Notka, vol, Ixi. p, 108- 



AstT, Jitu/m. Nob. 401-92 (Eoberti) ; 




VARIABLE STABS OF SHORT PERIOD 121 

m&goitade being accompLiahed in 1^ 30*^; and he baa eince 
asBOciated with it a star in Cygnus (designated " UY Cygni **) 
following nearly the same preecription of change.^ Madame 
Oer&aM's cliister-vaiiable, also situated in Gygnus, i& beUeved 
to have the extraordinarily short period of 3*^ 12"^ and an 
oaeillat ion -amplitude uf approximately one magnitude. Dia- 
covered photographically in 1904, it demands special and 
adroit study. 

The third aub-claas of short-period variables, called 
* QezDinids" from their exemplar ^ Geminorum, alter in 
brightness by continuous and aymmetrical gradations. Their 
maxima are placed about midway between tlteir minima. All 
(we need not hesitate to eay) are binary systems^ but some at 
least are eicempt from eclipses. The line, however, separating 
thidDi from occulting variables is very feebly traced. Whether 
or no eclipses occur has to be decided by a distinct investiga- 
tion for each individual star, and the process of decision 
advances slowly. Only in one ease a conclusive reply has 
been obtained, and it is in the negative. B^lopoleky and 
Campbell independently, in 1898,^ recognified f Geminorum 
as a spectroscopic pair revolving in the period of light- 
change ( 1 0"* 4^), One component ia invisible, yet the 
observed minima are not due to its intervention, siiice their 
epochs are not those of conjunction. Eclipaea can naturally 
only take place when the two bodies concerned are in the 
same line of sight ; and the spectroecope intimates their being 
in the same line of sight by the reduction to zero of their 
radial velocity. They must, in other words, at the time of 
ocjcultation, be moving across the line of sight. The non- 
fulfilment of this condition by f Geminorum excludes it 
peremptorily from the number of eclipsing variables. 

The light-carve of U Vulpecuhe (fiee Fig. 15)* so closely 
imitates that of ^ Geminorum that we need have no doubt of 
the atarB being similarly circumstanced, S Antliae, formerly 
taken for an Algol variable, belongs to the same category. At 
its maxima the How of change is so slack as to suggest au 



^ Monthly N^oliixg, vols, liiii. p. 304 ; liv. p. 686. 
* AHr. yacA. No. 8565 ; Aitroph. Joum. vols, ir, p. S6 ; xLii. p. 90. 
' MnJierind Eflropf, A9tr. Nadi. No. S483 ; LmKutj ibid. Wo, 3670; 
Itarnirii Cireviar, No, 41. 



122 



THE SYSTEM OF THE STAES 



actual Btatidstill, vrhile the minima are oompaiatiyelj Bharp^ 
jMid the entire cycle u aocamplished in 7** 47* It hag 
yieldeii, so far, no spectroscopic signs of duplicity. 



«« 


s. 






1 


; t 












,^ 


S. 












\\ 






















/ 




!> 












Is 


















/ 








\ 










— 


s 














/ 












N 


















/ 






















' 






s 








/ 


































S 




y 






















'^ 


































i^ 








p 




3 


4 




S 




t 




1 




1 











Fio. li-— Ll^t-Gumt 0* tJ Vu!i»«rai». 



Some Greminid stars are subject to what may be called a 
double periodicity. They dip, that is to say, to a secondary 
miDimum placed half-way between two equal maxitna. Thia 
mode of variation is brilliantly illustrate in )3 Lyrse, the 
saddle-back curve of which is shown in Fi^. 1 6. Its fluctua- 



Mkc V\ 







15 




tions, detected by Goodricke in 1784, were completely tracked 
out by Argelander in 1844.^ Their cause ia nevertheless slill 
involved in perplexity. The star is known to be binary, but 
the complex changes in its bright-line spectrum profoundly 
embarrass measurements of its velocity. The evidence at 
hand does not preclude the hypothesis of disparate obscura- 
tions at the primary and secondary minima respectively ; it 
does not, however, enforce it, and a solution to the enigma 
presented by this star will perhaps be most easily obtained by 
the indirect means of studying it at second-hand in analogous 
objects of less complicated relationships. 



D% SUllA ff Lyrtt DitquUitio. 



VAfilABLE STABS OF SHORT PERIOD 123 

One Bucb is found in V Puppis, the light-curve of which, 
as drawu by Mr. A- W. Roberts, is shown in Fig. 17- Its 
similarity with that of ^ Lyrie doee not need to he pointed 
out. If the oecurreDCti of a double eclipse can be proved in 
one esse it may be preHuined in the other. A Bpectroscopic 
pronouncement on the point is awaited with much interest, 
and should not he difficidt to procure. Already, in 1895, 
V Puppis was ascertained by Professor Pickering to be a 
spectroaoopic binary, composed of two unequally bright stara.^ 
Nevertheless we are still ignorant aa to whether their move- 
ntentB satisfy the requiremeata of the occultation- theory. 
Meantime the photometric data collected by Mr. Koberts have 
been shown by him to agree remarkably well with the light- 
variation resTilting from the mutual eclipsea of two bodies 



0E3 t 9 r5 iS S S 2A~^ S3" 

Flo. 1?.— Llght-Ciirv«of VPuppts. 



i'i 4i 



circulating in contact.- Actually in contact they should be, 
perliape even confluent ; but this involves no mechanical 
imposeibility* It would, however, involve the consequence that 
th@ mean density of the double globe of V Puppis could mot 
exceed ^-^ that of the sun. Other analogues of ^ Lyne are 
U Fegasi, E Sagittae^ with a period of 70, and V Vulpecuhe, 
accomplishing its changes in 75 days.^ That the spectroscope 
will eventually supply evidence of their binary chara^iter is 
scarcely doubtful, but it does not follow that all or any of 
them are eehpsiug binaries. 

"We would now invite oiu* readers* attention to the five 

hght-curvea grouped together in Fig. 18. They are copied 

torn an instructive paper presented by Mr. A. W. Eobeits of 



* Harvard Cireulart No. 14, 

■ Attropk, Jaum, vol, JEiii, p, iSl. 

> Arlr. J^aek. Ko. 3929 (SUuIef Williama}. 



VARIABLE STARS OF SHORT PERIOD 125 



Lovedttle to the South Afrioim Aaeoctation for the Advaace- 
ment of Science at its first meeting lu 1903. It embodies his 
principal cooclusions regardiog stellar cclipseg, and vividly 
illostrateft both the variety of conditions under which they 
tike pl&c« and the difficulty of pronouncing, In certain casea^ 
for or against their genuine occurrence. Thua, the light-^urvo 
of R' Centauri (No. 5, the lowest In the dhigram) is that of a 
Geminid star. The stationary maximum character is tic of 
AJgol variables is absent. Mr Roberts, nevevthelessj consider* 
the star to be composed of two egg-shaped bodies rotating 
oa a common axis in 14^ S2"^, and sending us more or less 
light according as we see them broadside or end-on.* The 
forms corresponding to the observed variations in brightneas 




are depicted in Fig. 19, the dotted lines indicating Darwin's 
figure of ecjuilibrium for a liquid globe on the brink of fission 
through accelerated rotation. The close agreement between 
;,be forms arrived at from photometric and mechanical con- 
Jderations reepectively U of strongly peraufisive import. 

Curve No. 4 in our figure is of an intermediate type. 
X Cajime verges towards the condition of unceasiog change 
viaible in p Lyrae and S Antliae ; yet brief intervals of stable 
shining appear to interrupt the proceaeea of decline and 
recovery^ which Mr. Roberts expounds as the phases of a 
protracted double eclipse. Two stars, somewhat uoequaUy 
brilliant, are beUeved to be concerned in them, the minima 
being slightly unequal The orbital period of 26 hours thiis 
includes two light-periods. 

' jiftr. Jourti. Noa. 37B, 384; Aslroph. Jmrn. vol. x. p< 312; Uonlhly 
Noticti, roU btiii. p. 627. 



126 



THE SYSTEM OF THE STARS 



The variability of K* Velorota, suspected by Kapte 
eetabliahed by Innes in 190 1^ and further defined by Rol 
It has ft pericul of 1** 20^" ; and the light-curve (Fi^ 
JSo. 3) intimates the occun^uce, within that apan, of 
abortive, as well as of a sharply pronounced eclipse, 
bright and a dusky st^ir are thus perc^iyed to be combined 
thia system. Each timo that they come into coujiiDCti 
there is a noticeable diminution of their joint light, but the efll 
Ib conspicuous only when the lu£fcrous body pas&es behind 
companion. 

The curve of R Ane (No, 2) includes no seoondl 
drop. This shows tlio occulting globe to be sensibly obscn 
Its concealment mukos no difference in the sum -total 
light. I 

S Velorum is one of many variablofi detected during t 
construction of the Cape Photographic Durehmusterung. 1 
Ci Eay Woods first noted its fluctuations in 1894,^ and M 
Roberta ably investigated the oatui'e of the fiystcm in whi( 
they arise.' A large semi-obscure star has, it appears, a ixA 
paratively small but much brighter comiianion revolving i 
5^ 22^^ The obscuration represented by the profoul 
gap in the curve (Fig. 18^ !No. 1) is due to the tram 
of the dim primary over its lucent satellite, which 
completely hides duj'ing a minimum lasting 6^ 35". F 
so long, in short, one star is, to our vision, substituted I 
another. Thia ingenious theory^ however, is hampered by tl 
anomalous consequence that the hiatrons globe must, on i 
showing, be twenty times denser than ita obscure attendai 
Ite final acceptance depends upon the verdict of the specti 



We have insensibly passed on from Geminid stars to € 
consideration of Algol variables. These, which foim Pickering 
Class V,, are of very peculiar interest from the amount of pi 
cise knowledge which they place at our disposal. Tht 
characterSj accordingly, invite close scrutiny and minute coi 
parison with theory. Variability is in them by short aceesa 
and consists always in a temporary loss of light. Tbi 
undergo, in fact, what are now known to be real eclipses 

1 ^str. Jou-m. No. 508. ^ Monthly Nctix^t^ toL Iv. p. 2n. 

' A^r. /oum. No, 327 ; A^to^k, Jowik, tdU, it. pr 270 i x, p« 314, 



A 



intervalB, wMle shiniug, for the most part, as steadily 

■ binary stars. Their detection is for this reason so difficult 

iintil the era of pliotogruphic discovery began, acxjuaint- 

with them was casual and scanty. "Novr, however, about 

ty figure in our catalogues, the designations of which are 

BQ in Table III. of our Appendix ; and Che list excludes auch 

ious instances as S Antliae, in which the occurrenoe of 

lipees is still auh jvdice. 

The eponym of the claee is curiously exact in its changes, 
''iich have been long and accurately observed. Their extra- 
linary character was determined, and an explanation of them 
interpositions of a dark satellite suggested by Goodricke in 
r83, since when some 15,600 minima have occurred in a 
Etnuer perfectly consistent with the hypotheaia. It became 
en of great interest to test its absolute truth, and the first 
Qeane of doing so were afforded by Professor Pickeringa strict 
L'^ntiuiry into the conditiona of the gupposed recurring eclipse.^ 
t*They proved to be all but perfectly complied with. Outside 
«f the twelve hours during which the " Demon-star " parts 
ritb and regains two-thirds of its light, it displays the re- 
red uniform lustre. The oac-iUation ia the same^ or very 
rly the aame^ in duration and extent now that it was fifty 
ago, and that it probably will be fifty years hence. The 
cision of itn performance seemed to correspond far better 
with the reaulta of geometrical rule and measure than with 
those of the complex interaction of physical causes ; and the 
spectroscope testified in the same sense by showing the sur- 
iving light at minimum to be of unchanged quiility. It \b 
amed, as if in large meaaut'e cut off, but betrays no symptom 
of intrinsic modification. These singular correspondences have 
not proved dec^iptive. The poatulated eclipses actually take 
place. 

The manner in which their genuineness has been estab- 
lished illustrates the singular versatility of modern metboda 
of research. No problem in which distant lighb-sourcea 
are concemefl seems hopelessly beyond their grasp. The 
received explanation of Algol's chfinges evidently involved 
the mutual revolution, in a period identical with theirs^ of the 
eclipsed and eclipsing bodies. And since their orbits, to admit 

' Proc Am*r, Acttd. ti>1. viii. (ISSl), p, 17 ; OhtcrveUory, vol. It, p. 116, 



128 



THE SrSTEM OF THE STAES 



of a transit of the satellite over the prmiaiy, should Me ai 
edgowise to our sight, practically the whole of their velociV 
should, in the course of each revolution, be direct*^ alti3n]»i 
straight away &om and stmigbt towards the earth. 
accordingly, spectniecopic meaeuies, recommended by Profi 
Pickering/ were clearly applicable ; and their phofcoj 
execution by Prafeasor Vogel in 1888-89 * eventuated in one 
of the most remarkable verifications of theoiy on record. 

Before each minimum Algol wa* found to be movicg 
away from the eun (independently of a continuous translation 
towards him of 2^ miles a aecoud) at the rate of 26^ Euglisb 
miles per aecx)nd ; after each minimum, to be approochiog 
with an equal speed ; while at intermediate timee the Im* 
printed lines, by resimiing their normal positions in the 
spectrum, proved the star to be then moving perpendicularly 
to the viBuul ray< Multiplying this velocity (of 26 J miles) hj 
the number of seconds in Algol's period (247,735) we get ao 
orbital ciivimifepence corresponding to a diameter of (ia round 
numbers) two million milea. Moreover, since the proportionate 
dimensiuns of the bright and dark bodies are aliowu by the 
amount of obscuration of one by the other to be vety nearly 
as 100 to 83, their relative miisees would also be known, if 
we conld be sm-e that they are of the same mean density 
The assimiptiou as regards a mass sbining with great brilliancy 
and one almost totally dark is certainly a hazardous one> bot 
it receives some warrant from tlie example of the sun and 
Jupiter. By its aid Profeesur Vogel mrived at the following 
ppoviaional data for the system of Algol -. — 



DianieUr of AJgol 

„ satellite 
Difitance from centre to centre 
Orbital velocity of Algol . 

„ ,^ sateUite 

Uaffi of Algol 

,. aateUite . 



1,061,000 English iiuie& 

834,300 „ 

3,230,000 „ 

2&'Z milei -pes sea 

05-4 

* mUt DlUIL 

t 



In the accompanying diagram G marks the centre of 
gravity round which both stars revolve with velocities in- 
Tersely proportional to their maases. Thus, Algol travels in 

^ Proe. Ayfter. Aoad, toI. viiL p. 34. 
* Attr. N*i6k. No^ 2H7. 



VARIABLE STAES OF SHORT PERIOD 129 



\Algcl 



•C 



AH orbit of only half the compass of that of its compamon. 

because possessed of twice its attractive force. It is easy to 

aee, too, that the duration of the eclipse compared with the 

length of the period gives the relation hetween the diameter of 

the occulted body and the diameter of the orbit of the occulting 

bcNiy, 60 that the absolute dimensions of one becoming known, 

thoee of the other follow. 

The density allte of Algol and of its satellite must be less 
than a quarter that of the sun, or 0"38 that of watet. They 
are both then presumably gaseous. Some slight dLssynimetry 
in the phase-curve, formerly per- 
ceived or imagined, and set down 
by Dr. Wilsing ' to the account of 
ellipticity in the path pursued, 
has not of late been verified.^ 
Although the interval between 
AlgoVe succeaaive eclipses shortened 
by eight seconds between 1790 and 
1S80, when the process bociime 
reversed; it does not follow that 
the star's orbital period is subject 
to alteration. Dr. Chandlez holds 

^^e inequality to be merely apparent 

^K-to represent the time occupied by light in crossing a wide 
ellipse described by the occulting pair round a supposed dark 
primary.* The visibility of the phases would, if that were so, 
be alternately accelerated and delayed according as the body 
undergoing them was on the hither or the farther side of its 
great orbit Nor is it impossible that Chandler's theory may 
ultimately be directly verified by the micrometrical measure- 
meut of undulations in the proper movement of Algol * 
corresponding to its suggested spacious citcuitings in a period 
of about 118 years. M. Tisserand, on the other hand, rejecting 
the idea of a triple system, explained the deviations of the 
eclipses from their normal epochs by a progression of the 

^^De of apsides due to the presumed spheroidal shape of the 

^^ ' Astt, Na£h. No. aoao. 

! " G. MiUler. Pfid. No, 3732 ; H. 0. VogeU F. J, S. Astr. Qei, Bd. xiiri, 

^ 140. 

> AHr. Joum. Noi. 26&-5S, 509. * Boaa. Hid. No. 813. 



•^□Ark SatellTto 
Fid, 20 — Algol during an E;cUpM. 



130 



THE SYSTEM OF THE STAHS 



contiguous globes.^ This would necessarily bring about a 
cjclical fluctuation in the eclipse -period. Yet here, too> 
contouation iii etill UcJcing ; the implied eccentricity of tb« 
star's orbit not being spectroaoopieally apparent. 

It needa no argument to ptove tliat the edipee-theory of 
the variable in the head of Medusa mnst apply to all other 
memberg of the same sharply characteriBed class. Many d 
them, however, present anomalies which are the more de- 
serving of careful etndy that they may one day throw an 
important light on the circumstancea imder which combina- 
tions of the indicat*ad kind exist. 

The light-change of S Canuri was discovered by Mr. Hiad 
in I84S. and its peciUiar nature ascertained by Argelander in 
1852,' The atar remains steady during thirteen-fourteentha 
of its period, then declines, in eight hours and a half, to \ss» 
than one quarter of its usual brightness, which it regains in 
the course of thirteen hours more» The process of recovery, 
besides being abnormally Blow, is interrupted soon after it baa 
begun by a miurked pause,' represented graphically from 
Schonfeld^s observations in Fig. 2 1. The compass of this 
atar'a change appears to be by no means invariable. On April 
14, 1832, Schmidt observed at Athens a minimimi nearly two 
magnitudes :fednter than any he had seen before. During one 
hour the star remained sunken nearly to the twelfth magnitude.' 
The period of S Cancri is subject to a perturbation with a 
range of about forty minutes^ and embracing rather more than 
three hundred minima.^ 

Inequalities of this kind, which in Algol sum up to a few 
seconds in a century, and grow to many minutes in S Cancri, 
BIO in \ Tauri counted by hours.* Their method and cause 
have stiH to be unravelled The companion of X Tauri ia 
not, like that of Algol, entirely obscure. M. E^lopolaky 
spectrographically resolved the variable, in 1897/ into an 
uneqUiiUy bright pair, revolving in 3*^ 23^ the period of light- 

^ Compttfi Stndua, t cii. p. 125. 

* jistr. XiuJi. Nob. /&6, eCH, SOft. 

• rurtetJahrtsehTi/i Aatr. Qes. Bd. ix, p. 230. 

* AKr. Nach. No. 24B1. 

AfgeUnder. Banner Peob. Bd. viii. p. 397 : Schonfeld^ Smtw, Bd, t. p. 

' SobbitfeM, Jahrfsbtriehi, HaEiubelm, Bd, il, p. 7@. 

» Ajtr. iVaeS. No. S47<. 



VARIABLE STAKS OF SHOKT PERIOD 131 



ige, A secondary minimum, detected by M. Plassmatm in 

i90.^ corresponds to the octultation of B^lopolsky'a faint 

iponent, as the chief minimum tloea to that of its prim&ry. 

iie disparate couplad eclipses of R* Velorum (Fig- 18, No. 3) 

thns repeated by X Tauri 

The variations of U Cepheij first recognised by M. Ceraabi 
M06COW, Jmie 23, 1880, are unusually rapid and extensive. 
In four and a half hours the star is reduced to about one- 
ninth its ordinary lustre, losing light, at one stage of its 
ddcline, at the astonishing rate of more than one maguitude 
an hour ! The obecurity laets an hour and a half, but not 



MftC. &-2 




























s. 
















^-^ 


■" 








V 














> 


/ 






a-fl 




1 


\ 














/ 












^ 


t 








/ 










S'2 








\ 


























^ 


1 






-j 












B-4 

Be 










V 


— = — - 




7 




















\ 


, > 


^ 


/ 
























\^ 

















O* 2 4 S a "lb 12 14 I© f8 20 22 

Via. 21.— UtDimnm of S OsocrL 



with entire umformity.^ The lowest point is touched at 
first ' and a pause in the ascent, like that inflecting the light- 
curve of S Cancri (see Fig. 21), is indicated. Mr. Yendell, 
on the other liand, pronouncea the light-curve to be sym- 
.etrioal, and regards the minima aa annular eclipses of two 
inra' duration.* Some complicated irregularities in the period 
of U Cephei have been ascertained by Dr. Chandler, and M. 
Plasamann observed the minimum of April 27, 1902, to occur 
2* 27" later than the calculated time.* 

The period of Y Cygni, added to the list of Algol variables 

* Plftsstoami, DU vtrandtrlKhin Sterne, p. 42 ; Journ. BrU. Avtr, A»9, toL 
L pp. 137, 255. 5 Bohlin, Astr. Naeh. No. 3762. 

* CliMdler, AUr. Joum, ITo. 199. p, 63, * A$tr. Jovm. No. 551, 
9 Aatr. JVooi. No. 37M. 



132 



THE SYSTEM OF THE STABS 



by Dr. Chandler, December 9, 1886, averages about a dnj 
and a half,^ but iSuctuutes to an extent unparalleled is this 
kind of star. The retardation of its phases between 1887 
and 1888 amoiHiEod to seven hours, totally disconcerting pre- 
diction, and the jieriod soon afterwards set about shorteniu^' 
as rapidly aa it had lengthened before.* The actual change 
does not exceed half a second at each of the returns, but these 
are so numerous that the accumulating errors sum up in a 
short time to a startling aggregate. They have been success- 
fully rationalised by M. Dun^r ' on the principle applied hj 
Tisaerand to the irregularities of Algol The phases of Y 
Cygui, however, are duplicated, the conjoined stars being twins 
in size and splendour. Hence they take it in turn to be 
eclipsed, and revolve iu twice the mean interval between their 
conjunctions. Their orbit has an eccentricity of 0^145 i its 
major axis completes a gyration in forty-one years; and its 
varying position with regard to the hue of eight determines 
the amount by which any individual eclipse falls behind or 
anticipates its due epoch. The system of Z Herculia, although 
composed of diasiniiJar stara, exhibits analogous symptoms of 
perturbation to those presented by Y Cygni* 

Stellar eclipses, it need hardly be said, are purely relative 
phenomena. Their occurrence depends upon the situation of 
the observer. But the chance of their being visible from a 
distant point augments very greatly yrith the closeness of the 
revolving stars ; for which reason, and also because the phases 
recur more frequently when the orbit is narrow, Algol 
variables with periods exceeding four or five days are of 
extreme rarity. Until 1902> S Cancri was the only such 
instance known. In that year, however, Mrs. Fleming dis- 
covered in VZ Cygni a star with an eclipse-interval of 31*3 
days,^ or more than thrice as long as that of S CancrL Each 
of its minima lasts two days, and the loss of light amounts 
to two magnitudes. Secondary obscurations, symmetrically 
dividing the period, have lately been detected by Hartwig,** 

' 'ChftudtEr, Asir, Journ. Nos, 163, 185. '^ Ibid. No. 204. 

^ Aitr. Naeh, No. 3467 i Astroph, Jovm. rol. xi. p. 175 ; Kncwkdgej toU, 
icy. p. 87 ; rTi, p. ItJe, 

* Dmn^r, Attr. Joum. Noa. 374, 3S4, 422 ; AstToph, Joum. toL i. p. 2S6 ; 
Yendel], Aair. Journ. Noa. 328, JS6. 

» B^CLTvard Circular, No, 66. ■ Aatr. Jfach. No. 3944. 



VAEIABLE STAilS OF SHORT PERIOD 133 

and occasion some perplexity hj their apparent recalcitrance 
to the hypothesis of a double eclipse. Their observer^ in fact, 
diacajrck that hypothesis, und suggests Insteud the view that 
VZ Cygni is a " pear-shaped " body or " apioid," rotating in 
what was taken for the period of its systemic circulation. 

Eclipsing Rtara are remarkably tenuoua bodies. The com- 
parison of the duration of their transits with the periods of 
their revolution supplies means for deducing a limiting value 
for their densities/ and it was fixed by Mr, H. N- Ruasell in 
1899/ from an average of seventeen stars, at one-fourth that 
of the sun. Mr. A, W, Roberta, working independently on 
the aame lines, found the mean consistency of four southern 
Algol-paira to be about one-eighth the solar."^ The range 
of variety in this respect among such stars is evidently very 
con^derable ; but all appear to be less compact than our buq. 
The circumstance is also noteworthy that eclipsing stars agree 
in showing a Siriau or a helium spectrum. Nu exception to 
the rule at least has yet come to our notice. The solar type, 
on the other hand, predominates among ordinary short-period 
variables. Variable stars of all classes are probably at 
enormous distances — even on the celestial scale — from the 
^arth. There is no sign that any of them are included among 

le stars in our comparative vicinity. One of the best means 
forming a rough judgment on this point is by the amount 
of apparent motion ; and variables remain in general nearly 
fixed in the sky, Mira, perhaps the most mobile, shifts its 
po&ition indeed to the not wholly iuconaiderebla (jxteiit of 
twenty-five seconds of arc in a century^ but measures for 
parallax would be much embarrassed by its changes of magai- 

.de, and have not yet been attempted. Dr. Chase'a deter- 
ation of a distance for Algol ' about ten times that of 

firiua is the only piece of direct information yet obtained as 
to the remoteness of variable stars. 

Their distribution over the sphere presents some noticeable 
pectdiarities. Contrary to what might have been expected, 
short-period variables^ although on tlie whole much brighter 
objects than those of long period, tend much more decidedly 
to concentration in the Milky Way, while Che preference for 

' Maiven H*1l, Ob$ervfif<rnj, vol, ii. p. 22^. 
* Attroph. Jtmm. tdL x. p. 316. =* Jhid. p. 30S. * AaCt. Jmim. No. 31B. 






^^d( 
^Bhin. 




134 



THE SYSTEM OF THE STABS 



its plane belongs cbieSy to tadnt Btarfi amoDg those not van- 
M&. In Algol 6tar9 it is lees strong than in periodical stars 
of FicXering^B fourth class. These be, for the most part, 
alo&g a gre&t circle nearly, but not quite coincident with the 
medi&l tine of the gal^n'.^ It is remarkable that their cod- 
densation-levei (as i& showa by its being projected into a ffreat 
ciicle) pasaee thit^ugh the sun. Within the zone itself, there 
is an evident disposition towards clnaterLag. Where the 
Milky W*aj divides in Cyguufl, the variables follow ita southern 
branch, and they are thickly sown over the whole sky-r^on 
horn Lyra to Sagittarius.^ Indications abound that the con- 
ditions of variability, and even of particular kinds of van- 
ability, are localised in apace. Thus, in Sagittarios no lees 
than four stars fluctuate in periods of six to seven days, and 
many others are subject to slower or undetermined rioiflsitudes. 
Two adjacent stars in the Southern Triangle vary in unusually 
short periods. A small region near 17 Carini]& includes six 
stars changeable in brightness.* Dr. Max Wolf has lately 
puhliahed * a list of thirty - six variables grouped round 
S AquiUe. The Orion nebula and the Magellanic Clouds are 
veritable nests of fluctuating light-pointa. The new star 
which appeared in Scorpio in 1860 marked the centre of a 
group of nine or ten objects, all widely variable irregularly or 
in long periods. Five stars of a similar nature, including two 
virtually extinct Novxe, are collected in a small section of 
Ophiuchua, and in general the sites of temporary stellar 
apparitions are more or lesa closely dotted round with variables. 
There is reason to suppose that the circumstances favouring 
instability of light do not exist anywhere in the neighbour- 
hood of the sun. 

' Thfl northern pole of this oimle, according to Pickering, is sitnatod ia 
E.A. lSh.,D«:. + 20'. Tliat of the Milky Tilay '» »a ^■■^- I2h 40m., Doc.+M". 
* Chandler, Asir. Jottni. No. 103 ; riasiamami, Die ter&ndcrtii^ieu SUme, p. &5. 
' Goto, ^Tnowltdge, vol. xiv. (i. 193. * Aitr. Naeh. No. i005. 




stars differ obviously in colour. Three or four among 
'the brightest strike the eye by their ^xlent glow^ others are 
tinged with yellow, and the white light of several baa a 

IMoiah gleam like that of polished steel. Keddinh tints are, 
however, ia the few cases iu which they affect lucid stars, 
^e most noticeable, and were the only oues remarked by the 
ftncients. 

Ptolemy designates as '* fiery red " (uTroKtppot) the following 
Bix atars: Aldebaran, Arcturua, Betelgeux, Antares, Pollux, 
and — mirah'Ue dictu — Sirioa ! all the rest being iudia- 
criminately classed as " yellow " {^dv&oi'). Now Pollux at 
present^ though by no means red, is at least yelLowiah, but 
Sirius is undemably white with a cast of blue. A marked 
change in its colour since the Alexandrian epoch thua appears 
all but certain, the more so that Seneca makes express mention of 
the dog-8tar as being " redder thau Mara." ' Horace baa " rubra 
Canioula " as typical of the heat of summer,^ and Cioero, iu 
his translation of Aratus, speaks of its " ruddy light." Sig- 
nificant, too, in the same sense was the custom described by 
the gramraarian Festus of sacrificing red dogs at the feast of 
Bth« Floralia for the parpose of placating in the interests of 
the approaching harve:at, the formidable and inimical dog-star. 
The wliole auhjeet has been learnedly disciissed by Dr. See,^ 
who may be said to hare fairly established his contention 
that the present white lustre of Sirius does not date back 
more than a thousand or 1200 years. In the tenth century 

1 QwBxl. Nat. I. L 3 Sat. ii. 6, 39. 

' AaiTt and Axfrophytiw, voL ii» p. 269, etc. 

1S6 




136 



THE SYSTEM OF THE STARS 



the star was as^ui'edly no longer red. Al S^, 3 
observer, pointedly omits it from the list of those excepi 
coloured, while adding to it " Cot Hydrse " and Algol ^^ — Algol. 
now a silvery orb, if there be one in the sky ; yet it ie worth 
recording that Schmidt noticed the Perseus variable asyeUowiah 
red in 1841, although he never in later years saw it otherwise 
than white.* 

The aame observer was amazed, March 21, 1852, to per- 
ceive Arcturus without a trace of the Btroug colour familiar 
to him in it during eleven previoua years. In comparisoii 
with its paleness, Capella seemed bright yellowj Mara and 
Betelgeux glowed almost like fire.^ It was some years before 
the star resinned its original hue, and the reality of the 
change, admitted by Argelander, was certified by the obeerv*- 
tiona of Kaiser at Leyden.'' 

The periodical variations in colour of a Utsjb Ma} 
the "Pointer" nejtt the pole, announced by Klein in 186' 
were long tliabelieved in. Nevertheless, a aeries of obsej 
tions with ZoUuer's polarising cotorimeter, executed in 181 
by M. Kovesligethy, of the O Gyalla observatory in Hun] 
gave evidence of alternating fluctuations between red 
yellow in a period of 54^ days;*^ and essentially confirmai 
results have recently been obtained by MM. Lau 
Wirtz,^ 

The colours of the stars visible to the naked eye are f? 
and pale compared with those disclosed by the telescope. 
The real gems of the sky are found low down in the scale 
of brightness. To some extent this is only what might 
be expected- Intense tints result from strong Belective 
abaorption in the atmospheres of the stars they distiuguieh, 
and strong absorption impUea large loss of light. Stars shine 
witli the rays that have sxirvived transmission tlu^ough the 
glowing vapours in their neighbourhoodt and the more 
nearly those rays are limited to one particular part of 

* Schjellwup, Deaeription dei £tifiUi FitseSt p. 25. 

" Astr. Nach. Ko. 10&9, 

3 Ibid. No. &99, 

' De Sterreniiemel verklaart, Ft. L p. 597. 

" Afir. Nach. Nob. 1663, 2131. 

" Siriu*, Bd. lir. p. 253. 

^ Jahrlmeh der AslrmtanUfSj lid. liv. p. 135. 



THE COLOURS OF THE STABS 



137 



rspectrum, the pmer and clearer the resulting tint will be. 
A true prismatic hue could accordingly be produced only 
through a vast reduction of brightness; but true prismatic 
hues do not exist among the stars,^ the colours of which 
aie always more or less copiously diluted vdth white light. 

The science of stat'Coloiu's has hitherto made little 
pipgiest^. Attempts to set up a standard chromatic scale 
h&ve not been successful,^ and instrumental devit-ea for ensur- 
ing just and equable judgments may sometimes induoe larger 
erroTfi than they avert.^ Simple visual estimations} on the 
other hand, must be treated with great reserve, since " personal 
equation " in this matter often assumes enoi-mous proportions. 
llie extreme of colour-bUndnese is reached by comparatively 
few, but endless minor individuaUties of perception vitiate the 
greater part of an accumulated mass of evidence which might 
otherwise justify inferences of real change. From the com- 
plex bundle of raya forming the image of a star, each retina 
picks out and accentuates those to which it is most highly 
sensitive, precluding the possibility of ^reement aa to delicate 
tints between many different observers. With both the 
Herschela, for instance, the equilibrium of colour was shifted 
towards the red end of the spectrum. Struve's assistant, 
Knorre, saw all stars indiscriminately white ; Admiral Smyth, 
on the contrary, discjiminated between shades of colour alto- 
gether inappreciable to most of those who hav6 profited by 
his " Cycle of Celestial Ohjecta." 

Even of the same observer the impressions do not always 
agree. Fatigue and advancing years modiliy the colour- 
sense ; and M. Safarik stated that stars invariably appeared 
redder to his left than to his right eye.'* Atraaspheric condi- 
tions* toOf are powerfully operative. Misty air blots out faint 
tints and alters strong ones, azure visibly turning green 
through its influence. Height above the horizon ia another 
circumstance to be taken into account before any useful com- 
pari^na can he made, while instrumental causes tend further 



^ StriavB, Mensural JUicrotnelrictt, p, Ixiivi. 

' See the system proiKised by FratLta, Monthly Notieu, vol. ilvii. p. 269. 
' S«e the reaolti) giTon by Kijvoalii^othy, Ucber Hns neue Mtthodt der 
FarheTifxetimmung cUr SlenKt H*J1g, 1^37. 

* VisrUijahrasekri/t A^r. Qes. Jahrg. xjv. p, 378. 



138 



THE SYSTEM OF THE STABS 



to perplex their npabot l^rge apertures help af theaiselves 
to bring out colour/ espedallj in bihaU stora, bat the ooloor- 
oorrection or groat re£ractors is always imperfect, ukd tlw 

outHtanding blae fringe uaually conapicuoua in them roust bj 
cunLntat give a retldiah coat to the image,' Baflectors produce 
a aimilar effact through absorption of many of the higher rajs 
by the silvered glass or apeculuni -metal fonning their mirtoa. 
And since with high magnificat ion all hues merge, or ten<L to 
merge, into yellow, only medium powers should he naed in 
colour obserratioDA, 

In this department then, di^repant statements by no 
means necessarily imply actual variation. The former abound; 
instances of the latter are met with, but can only be admitted 
with extreme caution. 

The study of star-colours divides naturally into two 
bmnchea — one concerned with isolated, the other with com- 
pound obJBcte. Inq^uiriea in the first case are simplified by 
the curious and unexplained fact that elngle stars are never 
markedly tingetl except with red or yellow, Vega makes the 
nearest appruitch in the northern hemisphere to an independ- 
ently blue star ; 7 Toucanse, a Eridani, and e Pavonia are 
the " pale siipphires " of the southern sky. But they are very 
pale indeed — so pale as to produce no definite impression of 
colour upon ordinary eyes. Nor is the " emerald " tinge of 
^ Librre much more decided. We have accordingly to deal 
juat at present only with " red stars." 

The earliest lint of thirty-three of them was drawn up by 
I^lande in 1805." '* Ces dtoiles " von Zach remarked in re- 
publishing it in 1822, "anncnceut toujoure quelque choae de 
particulier ; or toute particularity merits d'etre obeerv^e." * 
We have to a great extent got rid of the notion which pre- 
sented itself to John Michell in 1767^^ that what they 
"announce" is the impending extinction of their own fires; 
but their pecuUarities have l>ecome, on that very account, all 
the more worthy of attention. Eed stars are commonly 
variable both in light and colour ; the display of colonnaded 

1 StrnTe^ Jtf^Hfl. Mieri>mr. p. Ixxxiri, Mr. Frnoks held thfi oppEisite opiitioD. 
» Webb, Studeni, vt>L v, p. 487. 

• Connoisaanee des Ttmt pour Van 1808. 

* Corratp, Astr, t. vii. p. 295. * PhiL Trans, vol. Ivii. p. 38A. 



THE COLOURS OF THE STARS 



139 



zoned spectra belongA exclusively to them \ and they are 

jiiently characterised by atmospheric incand^cence as well 

by atmospheric instability. Few of them can be watched 

and attentively without being caught in some singular 

of change. 
Their systeToatic study began with the publication in 
1866 of SGhjellerup's Catalogue of 280 red stars ;^ ten 
years later, Mr. Birmiugham of Tuam completed a eimilar 
work compiisittg 658 entries,' and Mr. Chambers laid before 
the Koyal Astronomical Society, April 6, IBS 7, a catalogue 
founded on his personal observations during seventeen years." 
Of V H nominally red stars in both hemispheres he had 
examined 5 &9| being virtually all those visible in England} 
with the result of finding the colour of most exaggerated. 
" Orange" was to his eye the tint prevailing among them ; true 
" reda " were scarce ; of stars meriting to be qualified ae 
" carmine " or '* ruby " he had not met above a dozen. 

More recent works of the kind have their value enhauced 
by spectroscopic indications. Such were added in 1888 to 
Sir. Espin'a edition of Birmingham's Hed Stars, augmented to 
the number of 1472, and to Kriiger'a Catalog der Fdrhigen 
Stfrn€ (Kiel, 1893), whicli includes 2153 more or less deeply- 
tinted objects. 

None present saturated colours. The perfect crimson of a 
solar prominence cannot be matched among the stars. Their 
hues result from atmosphenc action, and stellar atmospheres 
are only partially effective in lifting the prismatic rays^ Eed 
fitars are none the less striking telescopic objects. Their light, 
even in the less distinguished speciinens, hafi a lurid glow 
which at once marks them out from ordinary stars, and those 
deeper tints shine with an ardour recalling the wrathful 
itensity of a stormy eunaet. The contrast between a red and 
a whtte star in the same field of view is sometimes most 
viTid and beautifuL Thus, in the southern constellatiou 
Grus, 71^ and tt* show like little burnished discs of capper and 
silver respectivelyj seen under strong illumination. 



■^ ^ Astr. Nach. No. 1591 ; reprintad with nuraftroua udditiona in VurUljahrs- 
tehrift Attr. Qta. Jalirg. ix, p, 253. 

* Memffiri It. Irish Acad, vol, xxvi. p. 249. 

* Mpntkljf Notices, toL xlrii. p. S48. 



140 



THE SYSTEM OF THE STABS 



Among odnFpicuoua stara, Antaree, in tba heart of the 
Scorpion, is the ruddiest, Betelgeux comes next ; while Alde- 
l^j&u and Arctunzs have figured isimemorially in th« short 
list of visible fiery objects, to which Al Sftfi (as alieadj mat^ 
tioned) added a Hydrse. aud Father 'Soij-X y Cruds. But their 
colouis are mere pale shades compared with those inatni- 
meutallj brought into notice. '* Hiod'a criniBOa star" oihss- 
wise knowTi aa R Leporis, appeared to ita diaeoverer in 
1845 like " a drop of blood on a black field.'' As with most 
other variables, however, increaee of light brings with it a 
paling of colour. Near maximum, intense redneea givee place, 
partly through a well -known phygiological effect/ to a ooppefy 
hue. Its spectrum is of the fourth type, with particularly strong 
absorption of the blue rays, a very small proportion of which 
pfiuetrate ita dense veil of carbonaceous vapours. 

A similar objet-t, now known as V Hydne, is No. 16 of 
Lalande's, Ko. 136 of Schjellerup's Red Stars, and was re- 
cord^ by Dr. Oopelaud at Dunsiok, March 32, 1876, ftS 
" browu red," and of 7 '2 magnitude.* But three years later, 
Dr. I>reyer found it risen to the gixth magnitude, and of a 
" most magnLficent copper red," while Birmingham observed it 
in 1874 as of the eighth, Duult in 18S4» as faded to S'S 
magnitude. Ita fluctuations of light are comprised in a 
nominal period of 575 days. 

Close to one of the gems of the Southern Cross, an eighth- 
magnitude star wag described by Sir John Herecbel to be of " the 
folleet and deepest maroon red^ the most intense blood-red of 
any star I have seen. It is like a drop of blood when con- 
trasted with the whiteness of ^ Cnicis." * Among other 
wuthem stars remarkable for colour are E Sculptoris, no 
less "intensely scarlet" now than when Gould saw it nb Cor- 
doba, R Doradi^s, glowing like a live coal out of the darkneaa 
of space, and Lj Puppis, all of them noted variables. 

In the northern hemisphere, V Cjgni bears the palm for 
deptli of tint, especially as its light diminishes ; and not far 
inferior to it are K CassiopeifB, R Leonis, R Crateris^ and 
Mira, with U Cygni and U Caseiopcite, both splendidly 8et oif 

1 Oflthoff, Aair, Kf7ck. No. 3910. 

* Duntink Olacrvationa, vaL iv. p. 55. 

* Capt Oiuervati<m$j p. 4*6. 



THE COLOURS OF THE STARS 



141 



by the vicinity of blue attendants. Crimgon, indeed^ vergeB, 
in these and other periodical starSf more and more towards 
orange in their brightening phases, yet they remain pretty 
constantly " red." A few casea of complete if temporary 
change of colour have, however, been recorded. Thus, a 
seventh-magnitude atnr in the Lynx (90 Schjellerup) noted 
by Stritve as " rubra," by Secchi as " bella giaUa/' seemed to 
Birmingham, January 13, 18*74, blue or bluish white, a con- 
firmatoi-y and nearly cantemporaneous observation being made 
&t Greenwich.* A star of 8^ magnitude (148 Sclijellerup) 
called "scarlet" by Lord Eosae in 1861, "dark red" by 
d'Arreat, December 8, 1866, showed " uo colour" to Birming- 
ham, May 8, 1874, Dun^r found it, nevertheless, of a deep 
orange red in 1884, and it is characterised by a fine colon- 
naded spectrum. Again, Schjellerup was struck with the 
redness of a star in Aquila^ in 1863, which, after an interval 
of ten years, struck Birmingham as actually blue ; and 
similarly a bluiBh-white object occupied the place. November 
14, 1850, of a atar in Taurus (Schjellerup, 64 i) marked 
" very red " by Hind, September 3, 1848,* which Dreyer per- 
ceived as once more red in January 18T9» but Espin as white 
with a continuous spectrum, January 10, 18BS. One further 
instance may be mentioned. A fifth-magnitude atar in Argo, 
known as r Velorum,^ was natorioualy red during Gould's stay 
at Cordoba. But. !t seemed leaden white to the present 
writer in the antumn of 1888, and was observed by Mr. 
Tebbutt in March 1891 as barely tinged with red* Chough the 
tinge became more decided on substituting an 8 -inch for a 
4i-iDch equatoreal. The spectrum of r Velorum closely 
resembles that of Aldebaran. 

The changes of colour visible in tempomry stars have 
generally been in an opposite direction to those of ordinary 
variables. Their sanguine tints faded, instead of deepening 
with the decline of their light. Thus, Tyeho's atar, though 
it passed through an intermediate stage of redness, was of a 
leaden white when it disappeared. T Coronse ran nearly the 

^ Mevwira S. Iriah Aead, voL situ p. 2B9. 

* No, 6S03 or the CctpenAo^t Caialoijui ; No, 214 of Se^tlUrup'f Eed 
Start. 

' Jf<»UA/y Xotieta, tqL xi. p. 46, 

« The plftce of the Bt»r for 1900 u E.A* 10* 18»'l, D-ll* 9'. 




same course. Nova Ophiuchi (1848) and Nova Aiadros 
were ruddy lit tirst, colourless later. Nova Cygni from orange^ 
turned bluish. The colour changes of Nova Persei were not 
thtj least curious part of its history. Purely white at its 
outburst, February 22, 1901, it rapidly fluBhed to a deep red. 
whic'li lightened to orange during the spasms of intermittent 
brightening observed in March and ApriL The steady decline 
of the star was nevertheless accompanied by a progressive loss 
of colour, until on February 5» 1902, it showed to Frofesaar 
Barnard greenish -white, like Neptune.^ Its nebular spectrum 
was by that time fully developed ; but its antecedent redness 
was unaccounted for by any marked absorption iii the blue or 
green. Nova Geminorum similarly glowed vividly at first, 
but blanched with the waning of its sudden access of ligbt 
In its ease, however^ the predominance of red hydrogen in ita 
emissions explained the initial ruddiness of its hue. Professor 
Barnartl * was able, by suitably drawing out the eye-piece of 
the YiTkea 40-iuch refractor, to form on the C-line a crimaoD 
image of the star as purely tinted as a solar prominence; 
This is the only ascertained instance of the production of 
stellar colour by direct radiation instead of by complementary 
absorption. 

Red stars are very unequally scattered. Certain wide 
tracts of the sky are nearly destitute of them; in some, they 
oeour profusely. Tlie Milky Way between Aqnila, Lj-ra, and 
Cyguna was called by Birmingham the " Red Beglon " ; * yet 
other galactic consteUatious^ such as Perseus and Oa^opeia, 
might be said, on a preliminaiy survey, to consist of white 
atars.^ 

Evidently, however, real partialities of colour-distribution 
must be to a great extent masked by the projection, to the 
eye, of objects at indefinite distances from each other upon 
the same portion of the sphere* Hence extensive loeal 
eolloctions of similar stars may be so confused with overlying 
tuid underlying aggregations as to be completely unrecognis- 
fcble. Smaller groupings are more readily detected. It is by 

* Uimthly Notiies, toL liiL p. 418. 

■ A$trt!ph. Joum. ¥ol. irii. pp, 302, 376. 

■ Mcmoira E. Jrixh Acad. va\. xxvL p. 26B, 

* JVuika, Monthly Noticta, vol. xlvl p. 343 ; wtt ■icoOatholT, WfckcMehrift 
/Ur Attr. Bd. xli. (1876) p. 328. 




no accMetit that, in the immediate neighbourhood of one red 
star» others are so apt to be met with; and the "brick red" 
and " ruby " pairs included in Herschel's Cape list, may with 
oonfidenoe be assumed to be severally in some sort of physical 
coDoectioD. Red atarft, it was remarked by the same authority, 
«re conspicuous in many clusters both by brightness and 
situatiou ; and Father Seccbi wae etmck with the critical 
positions of euch objects as regards spiral or radiated stellar 
anangements in the Milky "Way.^ 

The principle of colour by association, barely indicated in 
clusters* is in double stars carried out to the higb^t perfec- 
tion. Kature is inexhaustible in her display among them of 
harmonies, contrasts, and delicate gradations of hue. They 
not only vividly sparkle in green and gold, azure and crimsoa, 
but shine in the sober radiance of fawn and olive, lilaCj deep 
purple, and ashen grey. Chalcedony, aquamarine, chrysolite, 
agate, and onyx have counterparts in the heavens as well as 
^mbies and emeralds, sards, sapphirea^ and topazes. 
B Mariotte of Dijon, a physicist, but no astronomer, was the 
vast to speak of blue stars. "Les ^toiles qui paraissent 
blauefi/' he wrote in 1681, "out une lumi^re faible, maia 
pure et sans melange d'exhalaisons." * But he gave no 
examples, and it is not easy to divine what class of objects he 
alluded bo. The chromatic observation of double stars was 
really begun by Father Christian Mayer at Mannheim in 
1776; although the interest of his preliminary efforts wag 
absorbed in the splendour of Herschers similar but vastly 
more extenaive and assured results. He not only discovered a 
great number of exquisitely tinted couples, but by liia success 
emphasised the importance of systematic attention to colour in 
double stars. 

His example was followed by F. G* W. Struve, who in 
1837 clasaified from this point of view 596 of the brightest 
known stellar pairs. The upshot was to prove agreement in 
colour the tulo, contrast the exception.^ Just half, or 295 of 
the objects examined were uniformly white ; 118 had both 
components yellow or reddish with slight differeucea of 

' AUi dei Nvovi Xinc«i', t. Til. p. 72- 

* Oetivret, t. i. p. 287. 
' £Ml«a Doi^h*, pp. 33-34 ; Aftnsura Mierouu p. Ixxxii. 




144 



THE SYSTEM OF THE STAKS 



intensity ; sixty - three were tinctured with blue. The 
iustances of getiuioe contrast nuinbered 120, and in aii of 
these the Bmall star was called blue. The rule is, moreover, 
without exception^ that no prLmary member of a disaimiltirly 
tinted pair is blue. 

The reality of chromatic contrasts in double stars was 
established by the persistence of colour in satellites during the 
obliteration of their primaries by an interposed wire or bar; 
and besides^ as Struve remarked, optically produced huee 
should be invariably complementary, which is far from being 
the case in stars. A curious proof of this independence was 
afforded by a double occultation of Autares and its coijnpSLaion 
observed by Dawes in 1856, The small star, emerging first 
from behind the moon, seemed as perfectly green viewed thns 
alone as when half lost In the glare of the great red star ii 
is attached to.' The same phenomenon was reobsorved in 
1SY8. 

The connection between inequality of brightneaa and un- 
litenesB of tint in coupled stars did not escape Struve's notice.' 
He found a mean difference of less than half a magnitude 
between the exactly similar members of 375 pairs, of over 
one magnitude for 101 etara showing varied shades of the 
same colour, and of nearly two magnitudes in 120 cases 
of contrasted tints. Professor Holden, taking account of 
physical pairs only, reached, in 1880j an analogous rseulb' 
Where there was identity of colour the average difference of 
lustre proved to be only half a magnitude, where there was 
diversity the luminous inequality mounted to two and a half 
magnitudes. One hundred and twenty-two of the stars con- 
sidered belonged to the first class, forty to the second. Now 
markedly unequal are generally wide pairs^* so that disparity 
of hue is seen to prevail in systems formed by a large star and 
a comparatively small and remote companion ; while genuine 
twin suns^ of not very different radiative power, and of 
similar radiative quality, circulate as a rule rapidly and closely 

^ Monthiy Notices, toI. X¥i. p. 113 ; Ntesten, del et Terre, t. ii. p, 96 ; 
Webb, C«/, ObjedSt p. 389. 

' jfmt, Jfiei-uwi. p, IsTtiiL 

' Aintr. jQvm. o/Seiena, rgj, lix p. 467. 

* Doberck, Astr^ J^ack. No. 2276. The Bc&rclty of dBiiilU cloae companioua 
to bright sUri tuay be partly due to the dilEctilty of dtacoranug them. 



J 



THE COLOURS OF THE STARS 



145 



yd their common centte of gravity. Why this is so we 
lot tell ; the bare fact is before ua. 

Some beauMfullj coloured stars are, nevertheless, ascer- 
led to bti in mutual revolution. The yellow and rose-red 
mpobents of 7} CaaaiopeiEe Sni&h their circuit in about two 
idred yeiira; those of e Bootis, chrome yellow and sea-water 
blue, in probably upwarda of twelve hundred ; f Bootis and 
y Cepbei, orange and purple, o Cephei and t Cygni, golden 
and azmie pairs, are all in swifter or slower orbital movement. 
A good many richly-tinted stars, on the other hand, appear 
atationary, doubtless because their diietancea apart are so con- 
sidexable ^ to make thair revolutions inordinately slow. 
TbtiB the emerald-green companion of a Herculis has preaerved, 
during a century, an invariable position with regard to the 
ruddy star it depends upon, and Antares forms with its sea- 
green satellite a somewhat similar and equally rigid combiua- 
tion- The fixed pair j3 Cygni (Albireo) shining with "yellow 
topaz " and " aquaccelestia blue " light, presents perhaps the 
most lovely effect of colour in the heavens, nearly matched^ 
however, by the variable S Cephei and its ccerulean companion. 
Among numerous other examples of contrasted or harmonising 
tints in double stars may be mentioned y Androraedse, orange 
and green ; 7 Delphinij yellow and pale emerald ; ij Persei, 
golden and azure ; 24 Comae Berenices, orange and lilac; 12 
Canum Venaticorum, pale yellow and fawn ; v Serpentis, sea- 
green and lilac; a pair in Cassiopeia (S 163), copper colour 
and blue; 17 Virginis, light rose and duaky red; o Draconia^ 
or&nge and emerald. A few red and green pairs seem 
abnormal through the near approach to equality in the 
roagoitudea of the components. One such was observed by 
Herschel in Pisces; and Burnham noted in 1900 the reality 
of the contrasted tint it presents.^ Another wag discovered 
by Burnham himself in Pisces, But the fiery primaries are, 
in both cases, likely to prove variable. 

Bright white stars have not unfrequently small blue ones 
in their vicinity. A distant companion of Eegulus seems as 
if steeped in indigo ; Higel has an azure attendant ; \ Gemi- 
noium one of an amethyatine shade; 84 Ceti and 62 Eridani 
made up each of a white and a Ulac star; while the 
' Mniaurt$ of DoithU Stan, p. 10. 



146 



THE SYSTEM OF THE STAKS 



sapphire t Bootis La grouped closely with one, more loosely 
with two olht-r subordinate blue objeots,* 

Two questions at oncij suggest themselvea about the 
colour* of double stare. To the first, Are they real ? a 
decisively affirmative answer can be given ; but the second, 
Are they permanent ? cannot be dispoeed of with 8«ch 
promptitude. The subtlety of hues reBulting from a highly 
complex set of retinal iuipresaions renders thera peculiarly 
liable to subjective variation. As evidence of ohJecH've vaiia^ 
tion, then, mndom notes of colour ace of little or no uee. 
Only the estimates of skilled observers, trained to the needfiil 
precautions, furnished with suitable instnuuenta, above all, 
owning normal eyesight, are worth weighing and comparing. 
Under this rule of exclusiveness, the testimony requiring the 
admission of real change eliriuks surprisingly in compass, but 
does not whoUy disappear. Colour-vaidablee are to be found 
among compound, no lees than among single stars. 

Owing partly to instrumental, partly to personal causes, 
the elder Struve perceived as purely white many stars seen 
by Herechel with a tinge of red or yellow. Disagreemeata in 
the opposite direction merit, then, particular attention, and 
they are especially mai-ked in two cases,^ The components of 
the splendid couple y I^ouis were described by Herschel in 
1784 aa both white, the smaller inclining slightly to pale r«d* 
But Struve saw them in 1837 golden yellow and "reddish" 
green ; Admiral Smyth " bright orange and greenish yellow"; 
and strongly, though unequally yellow, they still I'emain, 
Here then we have a presumption of genuine change, which, 
in the companion instance of y Delphini, is raised almost to 
certainty. Thcae last stars, noted by Herschel in 1779 as 
both perfectly white, showed golden yellow and bluish green 
to Struve's scrutiny. The progress of alteration may perhaps 
be marked by the younger Herschel'a and South*a record of 
them as white and yellowish in 1824.* Their dissimilar tints 
of orange and green now strike the eye at the first glance with 
the smallest telescope.'' 

' Fl*mm&rion, Catalogue des ^toila Dvubles, p. 76. 
■ MlensuTie Mierom, p. IjEwvii. 
t* Phil TruTis. vol. Ixxv. p. 48. • Ibid, vol, ciiv. p. SftS. 

* NoblC| Hourt with a ThrU'ituh T«li$cope, p. Ill; Fi&x)It8, Jvum^ Brit, 
Aatr. Au, voL v, p. 4&7. 



THE COLOURS OJ THE STARS 



147 



Another pair famoas fqr colour-fluctaationB is 95 Herculis^ 
composed of tw<j equal Btars of 5^ magnitude^ ptatiM (to 
appearance) immovabljr within 6" of each other. Familiar 
with them as nvidly lioted objects, Professor Vl&zzi Smyth 
waa astonished^ on pointing his telescope towards them from 
the Peak of Teneriffe, July 29, 1856, to perceive them l>oth 
white.' In the following year, nevertheless, they shone as 
before in " apple green and cherry red " and were bo observed 
by Admiral Smyth, Dawes, and others. Captain Higgens * 
actually watched these colours fade and revive in 1362-63, 
in the course of about a year; but no trace of them has 
been seen of late; the stars of 95 Herculis are now of an 
identical palish yellow.' Their spectra are not identical. 
Dr. Vogel, in 1899, classed one as solar, the other as Sirian 
in type. The history of these stars goes back to 1730, when 
Herschel observed them as bluish white and white ; J. Her- 
schel and South called them " bluish white and reddish " in 
1824 ; Struve, 1828-32, greenish yellow and reddish yellow, in 
precise agreement with Pickering's appraisement in 1878.* 
Thus the " magnificent tints of orange and green " which 
Secchi admired in 1855, and Piazzi Smyth missed in 1856, 
were of a transitory chai^eter. 

In the well-known binaiy, 70 Ophiuehi, there has been an 
equally undoubted change. Except an " inclination to ted " in 
the smaller, the elder Herschel perceived no colour in either 
of these stars ; his son and Sir James South called them white 
and " livid " ; yet they were recorded by Struve as of an espe- 
cially intense yellow and purple, by Admiral Smyth as " pale 
topaz and violet." They are now both yellow» very much as 
they were seen by Secchi in 1S55, and by Franks in 1876 ; 
the companion was, nevertheless, marked " purplieh " at 
Harvard College in 1878i "rose-coloured" by Flammarion in 

<!79. 
i The three stars of f Cancri are usually yellow, but Dem- 
waki noticed them as all white, 1854-56, the remoter 
component turning yellowish or olive in 1864-65.^ This form 
of concordant change through various shades of primrose or 

' 8myth^ Sidtreal Chramotia, pp. 35, 76. * Ihid. p. flO, 

]* Koble, Op. cit. p. 10&. * Harvard Annaht vol. xi. p. 150. 

• AftK Nadi. Noa. lUO, 1674. 



148 



THE SYSTEM OF THE STAES 



cowslip is not very rate amoDg revolving stars, while ll 
devdopmeiit of colour la othex pairs teads towards 
production of ooDtrast. It ofteii huppena, too, that 
component only varies iu hue, in which case the change ed- 
^ectB the Batellite star. The attendant, for example, of 
H«rcxdis has appeared by turns ashen, " grape-red," blue, and 
bluiah-green ; that of B Cygai was observed by Struve iig grey 
in 1826-1833, but conspicuously red in 1836, blue by Dawes 
in 1839-1841^ alternately red, blue, and violet by Secchi is 
1856-57, grey once more by Denibowaki in 1862-63, red bj 
Engehnann in 1865, aince when it has commonly seemed 
light blue.* Again, the multiple star <r Oriouis includes two 
if not three colour-variables ; the distant companion of 7 
Leporis changed from pale green in 1S32 to garnet in 1861 
and 1874. and the satellite of p Serpentia from lilac in 1332 
to "native copper" in 1851.^ 

Eye - estimates of colour do not reach below the 
surface ; they are mere indiqatiousj which the spectroscope 
and the spectrograph can alone help us to interpret. But the 
task is delicate for the eye. and until of late, was impossible 
for the camera of discriminating the varieties in quality of 
closely adjacent light-sources. Miss Maury managed^ never- 
theless, to pick out on the Draper Memorial plates eighteen 
" composite " spectra, in which the characters were so mixed 
as to suggest a twofold origin ; * and her acumen has been 
vindicated by the spectroscopic resolution of several of these 
dubious objects into swiftly-revolving, unlike couples. Exact 
determinations of the kind, however, were rendered possible 
only through Sir William Huggins'a invention of a " reOecting 
slit/' by means of which the spectra of stars no more than 2" 
apart can be separately photograplied.* Successful impressions 
were thus obtained from y Leonis, Cor Caroli (12 Canum 
Venaticorum), and Cygni. Fnam previous observations of 
the last- mentioned pair with a visual spectroscope, he had 
found their complementary colours explicable (wholly or 
in part) by complementary absorption ; * but this was not 

' Eng«lm»]m, Asir. Xaeh. No. 1676. 

» aroyth, Sid, Okromatia, p. 2fl ; Webb, Cet. Ohjeeit, p. 3&0. 

^ Rarvatd AjiTiaU, vol. xxriii, p. 92. 

* CifrnpUf Jievdv-t, Oct. 11, 1SB7 ; Aatropkys, Jov>rn. voL vi p. 824. 

■ fhH. Tr&nt. vol. cUt. p, 431. 



THE COLOURS OF THE STARS 149 

tible in the more re&angible sections of their dispersed 
photographed by himself and Lady Huggins in 1897.^ 
ue star yielded a Sirian, the yellow star a solar spectrum, 
of perfectly normal quality. Their exceptional tints 
• to be reserved for visual explanation. 
it the improvement of methods has brought within view 
alisation of the chief desiderata in stellar chromatic& 
are, first, the definite correlation of the integral effect to 
re with the analytical data furnished by the prism; 
ly, the recognition of some fixed mode of correspond- 
etween spectral and colour-variations. The foundations 
len have been laid of a true science of stellar chromatics. 

^ Atlas o/SUllar Spectra, pp. 168, 163. 




I 



A DOUBL£ Star is oue that divides into tvo with the help of a 
more or leas powerful telescope. The effect ia a strange, and 
might have appeared beforehand a most unlikely one. Yet it 
is of quite ordinary occurrenca Double stars aiie no fre$k 
of nature, but part of her settled plan ; or rather^ they enter 
systematically into the design of the Mind which is in uid 
above nature. 

The fii'St recognised specimen of the class was ^ Uraae 
Majoria, the middle " horse " of the Plough, called by the 
Arabs " Mizar/' which Riccioli found at Bologaa, in 1650^ to 
consist of a 2^ and a 4 magnitude star within foiirtees 
aeconda of arc of each other. Both are white, and they 
make a radiant dJaplay even in a very small telescope. The 
accident of a bright comet observed by Robert Hooke passiiig. 
on February 8, 1665, close to y Arietta ('* Mesarthini ") led to 
his diecovery of its duplex nature. The components, each of 
the fourth magnitude, and eight seconds apart, are perfectly 
alike both in li^ht and colour. Meanwhile Huygens had, in 
1656, seen $ Ononis to be triple — it disclosed itself as 
quadruple in 16S4; a Crucis, in the southern hemisphere, 
was divided by some Jesuit missionaries sent by Louis XIV. 
to Siam in 1685, and a Centauri by Eichaud at Pondicberry 
in 1689 ; making in all five double stars detected during the 
seventeenth ceutur)*. Four more — 7 Virginia, Castor, 61 
Cygni, and ^ Gygni^ — were taken note of by Bradley before 
1765 J and in 1776, Father Christian Mayer began at 
Mannheim a deliberate search for stellar couples. His thirt] 

i&a 




DOUBLE STABS 



151 



three di^coveriea in two years might be de^Hhed as the 

prelimiDary washings from the rich lode atmck a few months 
later by Sir William HerscheU 

The plentifulnees of double &tars was in itself an iire- 
fiistible argument for tbeir reality. That any two unconnected 
bright stars should be projected closely side by side upon the 
sphere was improbable; that such a eoutingency could be 
repeated hundreds of times was what no sane man ought 
to have been capable of beUeving. But human credulity is 
nowhere more conepicuous than in what it is prepared to 
attribute to chance ; and it needed such clear evidence of 
mutually circling movements as Herschel was able to produce 
in 1803 to establish the conviction of the "physiccU existenoe 
of double stars- 

The fact is one at which we can never cease to wonder. It 
brings US face to face with a state of things entirely unfamiliar, 
and of which the purpose lies beyond the scope of our limited 
underatandinga. So accustomed are we to the " sole dominion " 
of our own great star, that the presence of two suns in one 
sphere might well at first sight appear incredible. Yet there 
are many things " undreamt of in our philosophy " which are 
uevertheless true. Every drop of stagaant water is a world of 
uninterpreted mysteries ; what we choose to call the " order of 
nature " is violated at every instant, inexplicably, by our own 
volition ; and if that order be attended by anomalies upon the 
earthy how much less shall we veoture to prescribe its course 
in the heavens ? 

The term " double star " is obviously quite indefinite^ apart 
from some agreement as to its meaning ; and it was in fact 
used by early observers in a far wider sense than it is now 
usually considered to bear. Many of the small and remote 
attendants upon brighter stars recorded by the Herschels 
could scarcely be presumed to have any real connection with 
them ; 3 2" was fixed by Struve as the maximum interval be- 
tween the components of a genuine double star^ or 16", tuiless 
both were britj;hter than the ninth magnitude ] the younger 
Struve'a " Pulkowa Catalogue " included no at-ars beyond the 
narrower limits and Mr. Biirnham rejects oil pairs below the 
eighth magnitude above 5" apart. This progressive restriction 

' See the writer's History of Astronomy, p. 18, 4th ed. 



152 



THE SYSTEM OF THE STAES 



has almost necessarily accompanied the improvement 
telescopes. With the powerfii! and perfect refractors now in 
use, really close pairs accumulate faater than thej can con- 
tinue to be obfserved ; and the t'ollectton of the innimieraUe 
loosely yoked and ill-assorted couples they fuitber reveal would 
be inane waate of time. 

Already above twelve thoueand double stars, in the Her- 
Bchelian sense, have been registered^ of ivhich some six 
thousand correspond, by the closeness of their combination, 
to strict ideas of what a double star should be; about 1400 
are separated by 2" or less, and between 600 and 700 are 
visibly revolving. These laat interesting cases multiply jusfe 
now with especial rapidity. They are most apt to occur, as 
might be expected, among those etars at the smallest apparent 
distances &om each other, and requiring accordingly the 
highest optical powers for their detection. Our acquaintance 
with most of these ia so recent that their movements are only 
coming to be recognised as one pair aftet another is re- 
measured after a few years' interval 

The singular profusion with which stars are planted sideh^ 
side with a bare hairhreadtk of sky between, became manifest 
through Mr Burnham'a discoveries made at Chicago, 1871 to 
1879. while he still pursued the profession of a stenograpber. 
His thousand new pairs included 743 at an average distance 
of l"'58.^ This means that the total interval from centre to 
centre of these objects was just equal to the width of a human 
hair held thirty-si^ feet from the eye. About oue-tentb of 
that distance is the minimum at which, even with the great 
Lick telescope, stars can be divided, but by no means the 
minimum at which they can separately exist The spectro- 
scope has demonstrated what it was logical to infer, that 
numberless stars which must always, either through their 
distance from ourselves or the closeness of their companions, 
remain optically single, are nevertheless compound ; hence 
of any given star, as of a chemical " element/' we can say, 
not that it ia indivisible, but only that it has never been 
divided. 

Such stellar pairs as are known to be in orbital move- 
ment are called " binary stare," to aigniiy that they form real 
1 JftmotV* it. Atti". Soe. vol. iItIL p^ ai7. 



J 



DOUBLE STABS 



153 



al aystpems. The finest specimen qf this kind in the 
northern heavens is Castor, or a Geniinomm, composed of 
a second and a third magnitude star 5"' 18 ap^rt. They are 
both white with a greeniah tinge, and can be divided with a 
very modeiate telescope, so that the sight of this brilliant 
and suggestive object is not reserved for the inner circle of 
astronomera. Now it happens that Bradley observed the 
relative eitttation of these stars in 1719, and the comparison 
of hifi record with measures of tlie present day shows that 
they have shifted in the interim to the extent of ISI"", or 
ooaaiderably more than a third of a revolution. To complete 
an entire one they would need at the same rate about 500 
years. But they are likely to advance upon it. The moat 
trustworthy orbit yet computed fixes their period at 347 
years ;^ and although their movements have, in the past, 
falsified many prediotionSj each sTiccessive investi^tor is in a 
better position, because commanding a wider range of data 
than hi* predeceBsors. The limits of uncertainty as regards 
time of circulation shrink of themselves with every decade 
that goes by. 

The brightest ia ^o the widest pair of revolving stara in 
the sky, and a third distinction — that of being nearer to ua 
than any other known sidereal object — accounts for the first 
two. In a Centauri are combined two stars so brilliant that 
the lesser, though emitting only one-third as much light aa 
ite neighbour, ia still somewhat above the second rank. It 
ia of a deeper yellow than the primary star^ and must have 
gained considerably in lustre during the last century unless 
FeuiU^, Lacaille, Brisbane, and Bunlop all erred egregioualy 
in calling it of fourth magnitude.^ Since they were observed 
by the Franciscan monk, Louis FeuUlde, at Lima in 1709, 
these stars have executed nearly two and a half revolutions. 
Tliey traverse in eigbty-one years' an orbit about as much 
elongated ae that in which Faye's comet travels round the 
sun, and diverge, accordingly^ at '* apastron " to more than 
thric-e their "periaatron" distance. They are now 22" apart, 
and are separating fast, having in 187^ awept through their 

■ Dobenk, ^j^r. Nac^t. No. 3070. 

^ Seft flutamario^n'jj Ga-idtogiu, p. 81. 

' T. J, J, See, ExxliUi&ti o/ihe StcUar Si^«(«™, p, 14S. 



154 



THE SYSTEM OF THE STAES 



point of nearest approach. Th^ " mean radius/* or half tbe 
Duyor axis of tbe c-ooiputed ellii^e, if seen square from tbe 
tvtfa, would siibt^ud an angle of IV^'7, corresponding at the 
star's distance of twenty-five biUiou miles to an actual spui 
of (in roaud numbers) 2100 million milea; so that them 
lustrous globes are sometimes almost aa close together as 
Saturn is to the aun^ then, after two-^score years, at IJ timw 
the distance of Neptune; Their mosa ia just twice, their light 
about Ij times that of the sun. 

The spectacle is beyond doubt amazing of two such 
bodies united tbua organically into a single stately system. 
That it includes many other members xxmj be taken for 
granted, although we may never succeed in observing them, 
and are unable, even in imagination, to bestow or arrange 
them satisfactorily. Evidently, no planetary scheme or schemes 
at all resembling our own can depend upon the stars of 
a Centaim, A Mercury or a Vulcan, at the most, might 
Gad shelter in the close vicinity of one from tbe dis- 
turbing power of the other, its poBaible inhabitants enjoying 
the combined or alternating radiance of a greater and a lesser 
8un- Comets entering these predncta must be perplexed to 
decide betweeu the two potentates claiming their allegiance, 
and perhaps on occasions pay their court to each in tiuna, 
throwing out tails, as they do so, of a highly anomalous 
character. It has, however, been suggested that tbe clienta of 
double stars circulate about both simultaneously, in orbits wide 
enough to keep them beyond the reach of dangerous pertur- 
bations from either. This is, of course, conceivable, if for 
many reasons unlikely ; but the aurmiee can neither be verified 
nor disproved. 

The stars of 61 Cygni. like those of a Oentauri, share a 
rapid onward movement through space They resemble them 
too in spectrum and colour, and are counted among our nearest 
stellar neighboura. Yet they are inconspicuous, one falling 
short of the fifth,' the other of the sixth magnitude. 

Although they have been under continuous ecrutiny since 
1753, when Bradley noted the differences in their times of 
transit, it is only within the last few years that the curvature 
of their path haa become perceptible. Besides the forward 
movement possessed in common by the two, the smaller alao 



J 



DOUBLE STARS 



155 



I 
I 



lifts its place senfiibly aa regards the larger star. But for b 
lentury and upward the shiftiug appeared to take place along 
straight line. If this had really been the case, the fact 
3uld have aboUflhed the presumption of their binary char- 
Facter, and compelled the belief, which was adopted by Captain 
[Jacob in 1858/ and maintained so lately as 1891 by Mr. 
Bombam,^ that the etars would eventually part company 
and cease to have even an apparent connectioiu This, we 
can now see clearly, was a false alarm. They are really 
inaeparable ; although the circumstances of their revolution 
must long remain unknown. Moreover, Dr. Wilaing'a an- 
notincement in 139^^ that the motion of one or both com- 
ponents was disturbed by the attraction of inviaible attendants, 
Icicks coulirmation. 

The systems of 70 Ophinchi and jj Cassiopeiee have much 
in common. The stara forming them show elmilar spectra 
and (apart from incidental variations) similar colours. They 
progress through space at about the same rate/ and both are at 
neairly the same distance of twenty light years from the earth. 
Both, too, have proved somewhat recalcitrant to computation. 
The orbit of 70 Ophiuchi, more especially, though one of the 
earliest experimented upou^ can still only be regarded as 
approximately determined. The stars have hitherto so 
persistently refused to keep to their predicted places that 
Madler, Jacob, and Sir John Herschel suspected disturbance 
by an invisible member of their system calculated by Dr. See 
in 1895* to revolve in a period of thirty-six years. The 
bright companion, on this view, de8cril>e3 in eighty-eight years 
an eccentric orbit with a major axis slightly less than that of 
Neptune, while simultaneously tracing out, in thlrty-six years, 
•' another ellipae, which in size considerably surpasses that of 
the planet Mara." ^ Yet its vagaries of movement are not 
even thus completely explicable. The mechanism of 70 
Ophiuchi has still obscure springs. 

The path of 13 Cassiopeia, traversed in 196 years (accord- 

* Etit'jthurffh Hev} Phil, Joum, vol. vii. p, 107, 

3 Aalr. Nach. No. 8047. 

* Sii3U7tffsberichte^ Berlin, October 26, 1393, 

* Sftdler, En^luh JfteJutnic, vdIb. xM. p, 410 ; xUv. p. 322. 

* Anfr. Joum. No. 383, 

' T. J. J. a««, Stellar ^ysteioi, ft, 220. 



: 



166 



THE SYSTEM OF THE STARS 



ing to Dr. See'a elemeats), is of an ampler Bweep. Its mean 
radius is fifty-throe times that of the terrestrial, oearlj twice 
that of Nepfcuue'a orbit. The stars, nevertheless, at their 
neareat approaches come within less than once and a half 
times the distance of Uranus from the sun, and since thej 
together contain more than four times the Bolar quantiiy 
of matter, perturbations of no slight intensitj would at 
aueh times affect their perhaps visionary trains of attend&at 
bodies. 

Laplace's conjecture that space might hold as many dark 
as bright masses has received some countenance from Ihd 
phenomena of double stars. For among them are reckoned 
effects of the attraction of unseen upon the movements of 
seen bodies, wliile in two cases the detection of an imperfectly 
luminous object has, like the discovery of Neptune^ ensued 
upon the theoretical indication of its place. 

From the nature of the proper notion of Sinus, Bessel 
inferred in 1844 that it did not travel alone. The line traced 
out by it must, were it solitary, have been straight, whereas 
it undulated markedly and regularly once in about half a 
century. Revolution in that period round an obscure com- 
panion was indicated j the elements of the hypothetical Sirian 
system were computed by Peters and Anwers, and the precise 
position of the Sirian Katellite was assigned by Safford la 
September 1861. On January 31 following, it was found 
juflt in the right spot by Aivan G. Clark, of Cambridgeport, 
MaeeachuBetta. 

The companion of Sirius is a dull yellow star of tenth 
magnitude, almost lost in the glittering radiance of its great 
neighbour. Their apparent distance having diminished from 
10" in 1862 to 4" in 1890. it was then barely distinguishable 
to Mr. Burnhnm with the 36-inch Lick refractor.^ For six 
years it remained wholly invisible ; unseen it passed periastron 
in 1894; but in November 1896 re-emerged in its predicted 
place, and has since steadily followed the track laid down for 
it. The elements assigned to the pair by Dr. O. Ix)hse in 
1904* when it had performed all but 30** of an entire circuit, 
may accordingly be accepted as authoritative. 

The system brought to our notice is a very remarkable 

' Aatr, iV^wft. TTo. 2884.. » Ibid, No. 3965, 



^ 



DOUBLE STARS 



157 



■ one. Its chief member i& a body ertremely bright in pro- 
portion to its mass ; its secondary member b a body abnormally 
niassive proportionately to its light. Siriua shines like four 
thouaand, it gravitates like two of its companions. There 
must hence bfl an enormous disparity of temperature between 
them, with a probably corresponding difference of mean density. 
Yet they are presumabl)'' of contemporaneoua origin. 

At the distance of Sirius (about fifty billion miles) the 
sun would appear aa a star of the second magnitude. A 
cioUection of tweuty-one auna would barely supply the 
emissions of that brilliant orb, the attractive energy of which 
is, nevertheless, little more than twice that goveniing the 
solar system. The revolutions of its satellite are completed 
in 50^ years at a mean distance twenty times that of tha 
earth from the sun^ with e^^cursions, at apastron, two hundred 
milliojia of miles further than Neptune's. Now to con* 
trol motion so swift in so spacious a path, 3|- times the 
solar quantum of matter must be present; of which one- 
third belongs to the satellite itself, constituting it a body 
rather more ponderous than the sun, though giving no more 
than xsW ^^ ^*^ light. Thus the contrast between the 
components of this binary star could scarcely among visible 
objects be more pronounced. And its significance ia accentu- 
ated by its essential repetition in Procyon, the lesser dog- 
star. 

Since its motion was known to be disturbed in precisel}' 
the same way aa the motion of Sirius, no doubt was enter- 
tained of its belonging to a binary combination. But the 
second member of that combination long evaded search, and 
was at last identified by Professor Sehaeberle at Lick, November 
13, 1896, in the modest guise of a thirteenth-magnitude star 
4"' 2 from its primary. So far. its movements tally well with 
the period of forty years, hypothetically attributed to them by 
Anwers in 1361 ; but indicate probably a markedly eccentric 
orbit instead of the nearly circular one assigned by him.' 
Schaebeile's satellite again exemplifies the strangely disparate 
nature of soma stellar couples. It exerts tliree-fifthe the 
gravitational power of the sun, while emitting no more than 
g^ of its light. Procyon ia further off from the earth than 
1 Newcombj The Stan, p. 162. 



158 THE SYSTEM OF THE STAES 

Sinus, thoQgb cot in the proportioD of its inferiont/ in magni- 
tade. Its actual lummomtj evaroelj amoonu to otte-fiftli tfast 
ci the gretiter dog-sUr. 

The detection of porti^Uj obscure stellar schemes goes ob 
apiOQL The disturbances of motiou telescopicaUy app^mt 
•Oggeited analogous diaturbaDces of motiou recogniadile 
onlj witb the spectroscope ; and the hint has proved extt«- 
ordinaril^ fniitfuL Stars like Procyoa are coiinected bj 
innumerable gradatioos with stars like AlgoL This we shall 
lAam in some detail further on. 

There are other criteria besides that of visible revolution 
in au orbit by which physical can be distinguished &om 
optical double Btara. Since 1812, when Beeeel pointed out 
tho concluaiveuess of the argument for real cotmection implied 
in the advance together of the st&xs of 61 Cygni,* "common 
proper motion " baa been ujiiverBally admitted aa a proof of 
geiuiiue usaotiifition. Thua the lustroua pair y Arietis has 
continued relatively fixed aince Bradley measured it in 1755; 
yet its members are fellow -iravellera through apace, and 
aiwuredly keep mutually circling aa they go, although so 
slowly that a century and half count almost for nothing in 
the majeBtic cycle of their revolutions. Again, the brightest 
atar in the Soutbetn Cross is made up of two stars of re- 
■pectively 1'6 ami 2*0 magnitude 5" apart, the situatione of 
which have not penxsptibly changed since Duntop determined 
thcni in 1826. This amounts to saying that their small 
proper movement is identical And even independently of 
thi» jH>9itive teat, the probabilities are enormously against the 
accidental dose juxtaposition of two stars so brilliant and so 
naarly «quai as those of a Crucis. 

The circumstance teetifiea strongly to the prevalence q^| 
phynioil conuectioD between stellar pairs, that the averag^^ 
difffirence of brightaess between them growa steadily with 
thuir distance apart^ approximately equal being usually con- 
tiguous objects. Every degree of inequality is indeed found 
iu undoubted systems ; still the chances of optical association 
must obviously increase vastly, even at the same distances, witl^l 
increase of optical disparity. ^^ 

The background of the sky is so thickly strewn with small 
* Mimat, Cttrrf^iondau, Bd. xzvL p. 160. 




DOUBLE STARS 



159 



8tarB, that we cannot be surprised if some of them happen to 
occupy critical situations. Bather, there is ground for 
astonishment at finding that certain remote satellites of 
bright stars seem indissolubly united to them. Reguhis, for 
example, carries with it, as it pursues its way across the 
Bphere, a star between the eighth and ninth magnitude^ 
discovered by Winlock at Washington to be itself closely 
double. The interval between the pair and its governing 
body amounts to nearly three minutes of arc. Castor, too, 
has att-ached to it a tenth-magnitude star at 74'' ; one of 
fleventh magnitude 90" to the south-west of a Crucis evidently 
belongs to its ooit^go ; ^ and Aldebaran forms with a minute 
object at 31" what seems to be a permanent eombination 
resembling in its effect to the eye that of Mara with his outer 
aatellite.'- 

Where two close stars seem fixed, relatively and absolutely, 
the ca^ for their physical union must dei>end upon circum- 
stantial evidence alone. But this is sometimes of cogent 
import Conti-ast of colour, for instance, may afford grounds 
for a strong persuasion of real relationship. Certain tints, as 
blue, greeuj and violet, only occur among mutually associated 
stars. We cannot, then, suppose the association upon which 
they depend for their production to be merely apparent. 
The topaz and azure components of ^ Cygni have no 
appreciable motion of any kind, and they are separated by a 
gap of 34", exceeding the limit of distance of real double stare 
as defined by Struve. Yet it is impossible to doubt that their 
brilliant hues are truly expressive of the systemic union from 
which they in some unknown way result. The same may be 
said of S Cephei and its blight blue attendant, and of the 
much closer and nearly equal stars 95 Hereulis, the inference 
being here strengthened by the concerted changes of colour 
recorded of them. We might be surej too, of the dependent 
status of the emerald satellites of the red stars a HeicuUs 
iind Antares, even if it were not independently proved by 
a community with their primaries in very slow progressive 
movement. 

Nevertheleaa, some highly coloured pairs have been eon- 

' Inues, RtfeTtnoe Cntaiogue, p. Ill a, 
' Bumham, A»tr. ycteh. No& 21S9, 'iS7£ ; Generai Cai'ilogue, p. 49. 




Ym 

ttl^ mpplm tMlfecr wiMUt udioiiBa of 
Wbn ecamoB to both aeabomof a pMc;^ leam bo n»m 
Cpr doolit OB the anlQeetu We abill racttr to this bopac in « 
ktcr cfaApCer. 

8Un v^th MoertAtned pRifir fiMtioao cfaAnctenB tf 
tlieiMBlv«* tlM lubire of their eompnooniiiipL For eiyier 
Uujf lc«ep oa together or tbej ihoir npu of incipieBt wepan- 
tioD, iknrlj bat sorely marking the diwtinfitinn between & 
looting ttttion And mexe tempofuy oootiguity. In the latter 
otto the moraaeat of one of the stozs referred to the other 
neoanarily proceeds along a straight line, so that rectilineal 
diJvplaoecuent is an infallible mark of an optical coupIeL One 
caiiouHXy cloMe (S. 1516; occora in the oonstdlation Df»ca 
Two ob&n, of 7 and 7'5 magnitude, pasaed in 1356 so near to 
one another by the hazard of their paths nearlj inteisecting, 
M to preiwnt Ibc e(f*ytst of two points of light, one inch apart, 
at u furlong'ft dlHtaace from the eye. Their angolor aepara- 
tioQ, Uven only 'i'^'6, ia now 13'', and it will continne to grow 
indefinitely. Their absolute disconnection haa heen confirmed 
by direct meaiiurenienta showing them to diifer extremely in 
raiuoteneM from the earth. The larger of the two, by one 

^ Oov^$ Rmdu$, i. UuviL pp. S36, 872. 
' Manok, ICnoieledge, toL xil p. 17<}. 



DOUBLE STAES 



161 



of ehose aingularitiefl which Hbound in the heavenft, fomifi a 
genuine pair with a etar very much fainter than ita spurious 
companion. 

From what has been said, it is clear that a good deal of 
patience is needed for the inyestigatlon of double gtars> Tbe 
facts about them must often be allowed to ripen for a long 
time before they can be turned to account. Sooner or later, 
however, their fruit cannot f^il to appear. There is. perhaps, 
no other branch of science in which induatry is 30 sure to bo 
rewarded with definite results. The £rst step i6 to separate 
perspective from physical couplea ■, this can only be done by 
the persistent repetition of exact measures. The next is to 
detect nascent circulatory movements in true binary pairs, 
or to keep watch on them us they progreaa. Their careful 
comparison with theory may at any time bring surprising 
novelties to light. For each steUar system is in effect a 
world by itself, original in its design, varied in its relation- 
ships, teeming with details of high signi&caiice. But at 
present only an imperfectly traced outline of the construction 
gf some three-score among hundreds of them is before us ; 
their multitude distracting attention. Yet it would be better 
to make intimate acquaintance with one than to know a score 
superficially. All the rasources of modem inventiveness 
should be enlisted in these inquiries. Not only the revolu- 
tions, distances, and masses of double stars, their movements 
etcross and in tbe line of sight, should he determined with 
ever-iucrcasjng precision, but their colours and magnitudes, 
and above all, their separate spectra, both visual and photo- 
graphic, should be recorded. By such means as these, real 
knowledge will be augmented far more than by the most 
brilliant auoceaa in the telescopic detection of new pairs. 
This has its own interest and value, but the recesses of 
sidereal structure must be otherwise explored. 



11 




CHAPTER XII 



BTKLLAB OKBITa 



The strong presomption that the law of graTitation wotild 
prove trujj univerBal has been fully borne out hj mvest^a- 
liouB of stellar orbits. Binary stara circulate, it can be 
unhesitatingly asserted, under the inf uenca of the identical 

force by which the bud sways the movements of the planeta, 
the earth the movements of the moon. This, it is true, does 
not admit of mathematical demonstration, but the over- 
whelming improbability of any other supposition enforces an 
almost equal degree of certitude,^ The revolutions of the 
stars are hence calculable, because conducted on faniiUar 
principles; their velocities have the same relation to masB, 
their perturbationB may lead to similar inferences as in the 
Bolar system. 

Observations, however, must precede calculations; and 
they are rendered arduous in double stars by the extreme 
minuteness of the intervals to be measured Many revolving 
pairs never separate to the apparent extent of a single second 
of arc ; yet this fraction of a second may represent, in abridg- 
ment, a span of some thousands of millions of miles. In- 
finitesimal errors, magoified in this proportion, become 
of colossal importance, and often impenetrably disguise the 
real aspect of the facts. 

For determining the relative situations of adjacent stars, 
two kinds of measurement are evidently needed. The fiiut 
gives their distance apart^ the second the direction of the line 
joining them as regards some fiied line of reference. That 
selected is the " hour-circle/' or great circle passing through 

* Tl^aerihnd, SuiUtin Jstronamiqtte, t. iv. p, 13. 
142 



J 



STELLAK ORBITS 



163 



the pole and the larger star, and the angle made with it by 
the line of junction of the pair ia called their " poaition^agle." 
It is couuted from 0*" to 360°, in a direction opposite to that 
of the movement of watch-hands, and a star is hence aaid to 
be ia direct resolution if its circuit \s from north to south 
through east, but in retrograde revolutiou^ if it ia oppositely 
pursued. 

i^ow the succeesive places, from year to year, of the moving 
Btar, obtained in this way with absolute accuracy, would 
fall into a perfect ellipse, the foresliortened delineation, as it 
were, of the real ellipae traversed in space. For the star's 
path ia seen by us projected against the sky, or rather upon the 
plane touching the sphere at that point, while the actual orbit 
may lie in any one of an infinite number of planes. The two 
curves, nevertheless, have relations by which one can be 
deduced from the other with geometrical certainty. Both are 
eliipaea, and in both the " radius vector," or line drawn from 
the satellite to the primary, describes equal areas in equal 
times. But the position of the chief star at the focus of the 
real ellipse is not maintained in its perspective repi-esentation, 
in which the " projected focus " is often quite unsymmetrically 
placed. Fig. 22 shows in juxtaposition the apparent and 
actual ellipses traversed by a Ceutauri, as delineated and com- 
puted by Dr. See. The distorting effects of perspective can 
be estimated by noticing that, white the true epoch of 
peiiastron was in 1875, the stars nevertheless continued to 
draw together optically until 1877- Once then the seemin/f 
orbit of a binary star is thoroughly ascertained, the problem 
of determining its actual orbit is as good as solved, the 
transition from one to the other being effected by a mathe- 
matical operation of no considerable difficulty. Even when 
the seeming orbit is a straight line the process remains 
feasible, and in fact one of the most reliable stellar theoiiea 
relates to a couple, the movements of which are conducted in 
a plane passing, it may be said, accurately through the earth.' 
The stars of 42 Comai Berenices appear simply to oscillate to 
and £ro in a period of somewhat less than twenty-six years, 
never diverging to a greater extent than about half a second, 

^ O. Struve, Monthly Notices, toL xxit. p. 367 ; AUi dtlV Accad. Font. t. 
xLt. p. 259. 




Wia. %— Appftnnt and real Orblta at Alpha CenUnri^ frbs r«duc«l dii^nin oT Uig Api»n«,t 
aUlpae La Sawcomti'a Stars hu bean anilnd of.) 

travelling in paths eeen nearly edgewise are 7 Coronae 
Borealis with a period of ninety-five, 44 Bootis with one of 
261 years, and a bioary in Ophiuehua (S 2173) revolving in 
forty-six years. 

Nothing would at iirgt sight seem Baster than to lay down. 



STELLAR ORBITS 



165 



from a sufficient store of data, the apparent track of a star. 
Yet the task is often a moet embarrassing one. Owing to the 
excessive imnuteness of the quantities concerned, the best 
observ'ations can give only loose approximatioue to the real 
facts. The margin of uncertainty, always veiy wide, at times 
exoeeds any reasonable Umit, and computers are hence obliged, 
as a rule, to reject some part of the materials before them as 
mifileading and incompatible with the rest. But tlie exerdae 
of discretion leads to diversity of results, and totally different 
orbits can thua be deriTed from the same set of obeervationa 
by varying their treatment so as to distribute differently 
their inherent errors. Where only a moderate arc of the 
orbit has been described, the problem of ideally completing 
it admits, from the iudeterminateness of its conditions, 
of no rigid solution. One might, in Mr. Burnham*a opinion, 
as well guess the period as go through the formality of 
calculating it.^ When the cftmpanion of Sirius» for instance, 
had been eighteen years under scrutiny, it was Btill impossible 
to decide whether it would return to its starting-point within 
the halfKientury allotted to it while still an unseen agency of 
disturbance, or depart on a remote excursion from its primary, 
demanding some hundreds of years for its accomplishment.^ 

In no department of astronomy is the mischief of " per- 
sonal equation " so sensible aa in the measurement of double 
stara. Nearly all available data are prejudicially affected by 
it, and those emanating from different individual Gources are 
thus often rendered exceedingly inharmonious. Much labour 
and ingenuity have been spent in determining its direction 
and amount for various observers, with a view to freeing their 
resulte from its effects ; and after all, it remains a question 
whether the observations so elaborately corrected are not more 
misleading than in their *' raw " state. 

All these complications might be at once swept away if 
it were possible effectively to substitute the camera for the 
micrometer. The photographic method leaves no room for 
systematic, very little for accidental e^Tors, G. P. Bond, of 
Cambridge (U,S.), showed in 1857, long before the intro- 
duction of the modem " dry plates," its wonderful capabilitias 

1 Popular Aitronofn^t vol. i. p. 248. 
' Plummen Monthly Notices, vol. iliL p. 83. 



166 



THE SYSTEM OF THE STARS 



for the accurate regisiration of the varying relative situatitms 
of double stara ; * and those capabilities were more fully 
realised in 1886 by the skill of the MM. Hemy. Somespeci- 
meua of the Paris photographs of double stars may be inspected 
iQ Fig. 23. The repeated impressious shown of each p^ wen 
obtained by allowing free play to the diumal motioa daring 
certain definite intervals between succeaaiTe short exposures. 
The line of displaceioent of the stars traces out conseqaentlj 
part of a circle of declination, aad their angles of position are 
directly measurable from plated embodying the data for their 
own orientation. The eiactitode of determinations from them 
proved very remarkable ; for ^ XJissd Majoria the *' mean error " 
of eingle measured of distance amounts to only 0^*077* of 



TV..i;.rus 



fia. 33.— Four Daiibl* BUm photojcnpfanc! &t Puii. (Mouehs^ 

angle to 0*56.* And this ia no illusory preciBion, but the 
statement of an unalterable fact. Unfortunately, however. 
the calculation of stellar orbits from photographic meaatuces 
must long continue impracticable. 

Their application is at present restricted to Bucb pairs as 
are neither very unequal nor very cloee. The ditfuaive bright- 
ness of Sirius, for instance, leaves no pogsibility of getting a 
separate print of its compinion, nor could even the much lesser 
disparity between the stars of B Cygni be made compatible 
with the distinct self-portraiture of both. Again, the minimum 
interval at which even perfectly equal couples have hitherto 
been successfully photographed exceeds two seconds, cloaer 
objectB running together on the plate. But the closer objects 
are just those most likely to be in rapid circulation ; and 
measures of any others are only in a remote sense useful. In 

I Aatr. yaeh. No. 1129. 
* Monchu, La Photoyraphie jt9tr(»uiniq'U4, p. i4. 




STELLAK ORBITS 



167 



>partioii, tbeOj ae Hfcar-couplea lead themselves to photo- 
"graphic delineation, they are unprofitable to compntera ; while 
those that repay computational labour for the most part 
evade Tegistration with the camera.' l^ere are no doubt 
exceptions, such as Coator and a Centauri \ and a jearly photo- 
graphic rec^^rd of each pair of the kind would eventually supply a 
stock of facta impaii'ed in value by no perplexing inconsistencies. 
The apparent orbits of revolving stars will be virtually 
inflcribed, one may hope, in future collections of negatives. 
The process of drawing a smooth eurve will then no longer 
admit any wide latitude of discretion ; and the representative 
ellipse, instead of threading its way amid a straggling crowd 
of outlying observational will paaa, not indeed actually through 
(which would imply the annihilation of error), but very close 
to all the given places. 

Above eighty stellar orbits have eo far been computed by 
Doberck, Gore, Glaaenapp, Bnrnham, See, and others. But 
for the moat part tentatively, nor always with success. Pre- 
dictions are often at fault; orbits are assigned only to be 
discarded ; the moving stars show themselves regardless of the 
trammels of theory. Thus, the persistent approach, since 
188T, of the components of Castor makes it certain that the 
millennial cycle ascribed to them by Thiele is much too long^ 
and suggests^ even apart from calculation, that Sir William 
Herschera estimated period of 342 years may be nearer the 
mark- The movementa, on the other hand, of the fine 
southern pair, 6 Eridani, demand a continually lengthening 
time-allowance; and Mr. Bnmham was inclined, in 1893, to 
regard them as rectilinear, that is, of a definitively separatist 
character.' The stars, none the less* seem at present to be 
conforming pretty closely to Mr Gore's orbit, which should be 
completed in 302 years.* 

Premature attempts to determine stellar revolutions are 
strongly deprecated by Mr. Bnmham, Some computed orbits 
are so purely provisional that, ae he points out, many others 
of a totally different form might be substituted for them with* 
out doing any violence to the obaervationa. Among the 

1 W, J, Hussey, PulL Astr. Soc. Pacijic, No, 74, p, 102, 

* Aitr, and Aatrophyaia, vol. xii. p. 698. 

^ lunea, Re/ertnct Cat, p. 13 A. 



I6d 



THE SYSTEM OF THE STABS 



earliest of known couples is ^ Aquarii^ divided 1^ Fatba 
GbxiBlian Mayer in 1777 into eqnal compoueats of the fourth 
snagnltuda They are in alow circulation, having described in 
the course of a centuiy and a quarter, an arc of leos than 60\ 
From the scanty data thus supplied, Dr. Boberck baa derived 
an orbit traversed in 1578 years, the Lotigeel; period yet found 
for a binary systenL But it i% or may be iUasory, Mt 
Buroham Bcaroely goes too far in saying that the zuaBfitiTeA of 
at leafit 500 years must be available before a aatisfactoiy 
theory of this Btar's revolutions can be formulated.^ 

An opposite in&tance of uncertainty in calculation it 
afforded by £ Equulei, which now ranks as the quickest of 
visual binaries. Discovered by Otto Struve in 1852, it was 
watched through many complete ciTcoits^ each of which was, 
until 1900, held to occupy 11^ years. In that year, however, 
Professor Hussey expreeaed a suspicion that the admitted 
period was just twice too long ; * and it was amply justifidd 
by the Btar's subsequent behavionr.* Confirmatory spectro- 
Bcopic observations have also been made ; * and the pair 
promises to supply a most useful link between stars measuiabte 
with the micrometer and stai^ only resolvable on pmmatic 
negatives. It was long &go perceived * that the combination 
of both methodB, if rendered practicable, would lead to a far 
more complete acquaintance with the systemic conditions of 
stars than the employment of either of them separately. The 
capabLUtiies of one, in fact, fill the lacunse left by those of the 
other. Stellar orbits as computed from visual data are of 
undetermined size. Their distances from the earth have to be 
further ascertained before the scale of their construction 
becomes known. But it can also be derived, under favourable 
circumstances, from spectroscopic measurements of velocity In 
the line of sight. For these give^ in miles per second, the 
rate of circulation of such star-pairs as prove amentible to them ; 
and the rate of travel and the period virtually combine to state 
the dimensions of an orbit lying in a known plane. The mass 

' Fop, Aiir, voL it. p, 475. * Pubi. Aatr. Soq. PoctjSc, No. 76. 

* F«bL Liik Obtervatortf^ vol. v. p. 20Cp* 

* lAck SvlUlin. No*. 4, 52. 

' Fox Tftlbot, Heport Brit. Aa. \871, pt. ii. p. 34 ; Niven, UoTUkty ybtiwi^ 
Tol, XXI jv. p, 339 ; Pftlica. Attr. Nttth. No. 2941 ; Hattibaat. Proc. M. /rwA 
Aead. vol. iv. «ct. ii. p. «63 ; T. J. J. See, AUr. Nath. No. 3314. 



STELLAK ORBITS 



169 



of the atara pursuing it and their parallax can thence eosily 
be deduced- And Profeaeor Huasey accordiugly derived for 
5 Equulei a combined attractive power neatly twice that of 
the eun^ and a parallax corresponding to a light-journey 
of 46 yeara. These resultg are the first of aubetantial value 
obtained by the application of the spectroscope to a telescopic 
double star. 

The pair that comes next to B Equulei in rapidity is 
ic Pegasi, divided by Mr. Bumham in 1880 into 4*3 and 5"0 
magnitude componentSf not much above one-tenth of a eecond 
apart. They liave einc* then performed more than two 
revolutions (period 11^ yeara) in an orbit viewed by ug under 
an angle of only 9°. and coubeq^uently foreshortened into an 
exceedingly narrow oval, the oscillations of the stars in which 
be followed only with the most perfect appliances. 
A Btandard collection of forty stellar orbits, with new 

"elementa calculated by himself, was published by Dr, See in 
1896/ Twenty-eight among them have periods falling short 
of 100, fourteen of 50 years; and these comprise some of the 
beat-determined^ though most recently discovered pairs — 
^ Sagittarii, travelling in 18 years, 9 ArgCls in 22, Sb Pegasi 
in 24, ^ Delphini in 27 years. The longest stellar period 
likely to be authenticated is that of 370 years, assigned by 
)r. See to o- CorouK Borealis, discerned by Herschel as double, 
August 7, 1780, Yet, thirty years ago, Doberck found it to 

'need 84G years to finiBh a circuit; and he was an expert in 
researches of the kind. In general, as Mr. Burnhani has 
insistently urged, revolutions require to be finished, or almost 
finished, before they can be said to be ascertained. Among 
the beat stellar theories extant is that of f UrsaB Majoria, one 
of two fourth 'magnitude stars marking the hindmost paw of 
the Great Bear. Divided by Herschel in 1780. this couple 
was made by Savary, in 1828,^ the subject of the firBt experi- 
ment in the extenaion of Newtonian principles to the sidereal 
universe. It Hucceeded ; for the stars were found to describe, 
as nearly as could be expected, the orbit csilculated for them 
on the supposition that their mutual gravitation was the 
Inence binding them together into a moving system. The 

' S^ltUviA of the SUllar SyftiiM, p, 2*3. 
* Conn, des Ttmps, 1830, pp. 56, 163. 



n 



170 THE SYSTEM OF THE STABS 

valiUUy throughout the universe of Newton's law has nerer 
since beeQ opeu to serious question* 

The path of f UrstB h&& of Late been seYeraJ times re- 
invegiigated, and with results so concord&nt ^ to give & 
atrong assurance of their approximate accuracy. It is & 
considerably elongated ellipse, the e-ccentriciiy being expreaaed 
by the fraction |» which ia just twice that of the orbit of 
Mercury, htdf that of the orbit of Encke'a comet. The period 
of traversing it is 60 years ; its semi'iuajor axis vould 
subtend, if seen without foreshortening, an angle of 2^". and 
it lies in a plane inclined 56'* to the plane touching thf 
sphere at that poiut.^ 

We are ignorant of the mass of this system because we 
are ignorant of its distance from the earth ; but whatever its 
distance and whatever its mass^ there seems no doubt that 
the stars conjoined in it are intensely luminous. If of the 
same meau density, they must be, st^uare mile for square mile 
of aurface, of about two and a half times the solar brightnesa 

Be terminations of the distances of binary stars are of speoia! 
interest from their leading to a knowledge of their masses. 
The connection is easily explained. Angular measurements, 
which are the only ones possible to be got of objects out of 
tangible reach, are convertible into definite linear values when 
the radiua of the sphere they refer to becomes knowo^in 
other words, when the interval of space between the eye and 
the objects measured is ascertained. So that the dimensions, 
in seconds of arc, of the orbits of stars at determined distances 
give at once their dimensions in millions of miles, whence, 
with the help of their periods of revolution, their masses 
easily follow. For by the law of gravity the attractive power 
of any system is proportional to the cube of the mean 
distance of the bodies composing it, divided by the square of 
their period. Employing^ then, as a unit of space in this 
little calculation the distance of the earth from the sun, and 
the year as our unit of time, we get the mass of each pair of 
revolving stara in terms of the sun's raaaa. It comes out. of 
course, larger in the ratio of the cube of the distance for the 

1 Date oD tlifiSfl seToral faaadB, together with others defining the titiulaon of 
petioatrou and of the liine of interaectioQ of tho orbiUl »nd refecum pluutt 
coDfitituta whftt an eaU«d the "alemeiita" of ft«tAr*i moTementB. 



STELLAH ORBITS 



171 



Bame period, and smallei* m the ratio of the square of the 
period for the same distance. Swiftly-moving and spacious 
syatems cotibaiti accordingly great quantities of matter, 
aluggish ones comparatively little. Many radiant couples 
maintain, decade after decade, an all but absolute hxity. 
Alpha Crucis, 95 and a Herculia, 7 Arietis are eiamples. The 
nascent diBplacemente of ^ TJrsai Majoris suggest the possible 
accomptishment of a circuit in 1 0,00 yeaa-s ; those of 7 Delphini 
might conaiafc with a period of four millenniums. This strange 
inactivity intimates for the stars displaying it either an exceed- 
ingly small mass^ or an inconceivable remoteness from the 
earth. 

A list of binaries moving in known orbits for which 
par^a^es have been ascertained, will be found in Table IV. 
of our Appendix, It is unfortunately brief; yet it embodies 
some important information. One of these pairs. 35 Pegaai^ 
was divided by Burnham in 187S, and completed in 1904 
its first observed round. The components, although no 
brighter than sLxth and eleventh magnitudes, exert a gravitating 
power (from Eumham'a elements)' eleven times that of our 
fiun, Tliey wear the aspect, as Mr. Gore remarks, rather of a 
aun and planet than of two suns. The primary eentuples the 
light emitted by its satellite, and there is just the disparity 
between them that would be presented by the sun and Jupiter 
if of the same intrinsic brilliancy. These would, on the con- 
dition supposed, and, at the distance (attributed to 85 Pegasi) 
of sixty light-years, show as a pair of 7^ and 12j magnitude, 
never above 0""23 asunder. The utmost powers of thegr&tt 
Lick refractor would scarcely be adequate for their separation ; 
but the real stars being respectively four times brighter than 
the iUustrative sun and planet, can be kept under watch and 
ward. 

The masses of eight visual couples are fairly well known, 
and their total value is just that of thirty-one suns. On an 
average, then, any one of those systems contains nearly four 
times as much attraetive ener^ as the solar system, each 
individual star being equal in this respect to a pair of 
Buns like ours. Were the extension of this result legi- 

' Otneral Gaialcgue, p. 370 ; ComBtock, Aatroph. /(mm, roL xril. p. tf23. 
The pftralUx of 65 PegAai U taa staaM to be well-usured. 



t?2 



the; System of the stars 



timate, Che-iiiataDcee of all stars revolving in asccft&ised 
orbits might be inferred from their assiuaed massiTeneaB (am 
the relation between distance and mass is convertible^ ibI 
upon this principle Madler derived what he eddied tbi 
" hypothetical parallaxes " of binaries/ rodconing, howeTH, 
the mass of each pair to be only that of a single sun. TIib 
estimate is uow seen to be much too small, and the distaixa 
founded upon it to fall proportionately short of the truth.' 
But, indeed, no general conclusions of the kind are fit fioi 
application to individual c^aea. The range of variety is » 
great that only simulated knowledge can be obtained in ihi 
way. Yet collective inferences are not therefore worth- 
leas. Thus, from averaging the masses of only eight binaria, 
we have gathered plausible gi-ounds for believing our sun to 
occupy a low rank as a centre of attraction. It may be, 
nevertheless, that the swifter binarieSj which can at preeec; 
alone figure on such a list, give too high an average mass. 

Calculationa based upon the orbital elements of revolving 
stars t«ll nothing of their reiative masses. They apply onlj 
to the common stock of matter in each system, leaving its 
distribution to be otherwise tested. This cannot be done 
except through the due apportionment of movement between 
the members of the system — an arduous task, "just begun to 
be grappled with. 

There ia no such thing in nature as a stationary body 
round which other bodies circulate. Answering tnotiou there 
must always be, though on a scale reduced in the exact pro- 
portion that the mass ia increased. The earth, for instance, 
describes, under the influeuce of the moon, an ellipse precisely 
similar to that described by the moon under the influence of 
the earth, but eiglity-one times smaller. And the sun corre- 
sponds in the same way to the revolutions of every one of the 
planets, notwithstanding that the centre of his movement as 
regards each of them, with the single exception of Jupiter, 
lies far beneath his own surface. Binary stars, however, are 



an 

J" 

m 

itt 

\i 



^ Thb nimseB of r^TolvJug stan vary, eaUns paribtu, u the cubes of tbflii' 
diataDces from the eartb. Of s7at«Qi& ideuticAl ib period &a<) ftpparent tnorc- 
mtmU on<i twice u remote u anotber would be eight times as mw»iye, out thrct 
tliUM u remote, twcDtj-Mveti timei ma maasiTe, and so on. 



irobably often almost equally maasive, and therefore almoat 
uaUy mobile bodiea. The fixity of one member of each 
-jkair 13 purely conventional — an iudifipenaable lictiou without 
"which meaeurementa would be impracticable. Those actually 
made give the sum of the movenLenta of both stars, and an 
orbit computed from them represents the Bum of their dis- 
tixict orbits. Identical in shape and position with the tnie 
eUipeeSj it differs from them only in size, its linear dimensions 
in any direction being equal to both theira taken together. 

The genuine centre of movement of two mutually ciitling 
stars is their common centre of gravity, which lies on ^ 
straight line drawn from one to the other> at a distance from 
each inversely proportional to its masa. The Btrictly similar 
orbits traversed by each are then spacioua in the aame inverted 
ratio. The larger fitar performs the smaller circuit, and vice 



1626 




I92S 



fia. U— OrblU of Uie OompOnClit 8Um vf daasa^ Vlr^iali. 



vend. In the case of their equality, their orbita must inter- 
sect if elliptical, but coincide if circular, when the stars will 
pursue each other along the same tracks while occupying in it, 
at each successive moment, diametrically opposite poaitions } 
nor could either, during an eternity of undisturbed revolution, 
gain a hair's breadth upon the other. But circular move- 
ment is not even approximated to by telescopic binaries, 
which usually follow highly eccentric paths. 

"We may take 7 Vii'giuis as an example of a pair moving 
in equal ellipaes, the relations of which are shown in Fig. 24. 
They have, it will be seen, a common focus, the seat of the 
centre of gravity, from which the stars (being of equal mass) 
must always be equally distant. Neither can approach to or 
recede from this point of origin of the force acting upon them 
without the other simultaneously doing the aame ; the two 
muat be in periastron and retire towards apastron together, 
losing and subsequently regaining velocity by the same grada- 




m 



THE SYSTEM OF THE STAES 



tioQS, The Btara of a Centauri also travel in orbits se^Mf 
eqiitil, but much leas elongated than those of 7 VirgiiUB {see 
Fig. 2B> 

The movements of unequal stars are Rimilaily oonducted 
That is to eay, the proportion of their respective distances Trom 
the common focus is invariable. They are accordingly always 
found in cyarreaponding parts of their orbits, and at opposite 
ends of a right line paasLpg through the focua. The manner 
of thtiir revolutions can be realised by a glance at Fig. 26, 
repreaenting the orbits of Sirius and its companion, the smaJI 
ellipse belonging of course to the bright star. 

Obviously, from what has been said, knowledge of the 
relative masses of binary stars would ensue upon knowledge 
of the relative dimensions of their separate orbits. This, 
indeed, is not within easy reach. Befiaed measurements, in a 

( 1S7J \iB75 j ( 'Ms/ W J 



FlO. S6.— O^bft* prtiin CompaDcnt Stan 
of Alptu Centiurl. 



Pio. adw-'Orblta of SiHn« mai tt> 
compuiion. 



prolonged eeries, of the individual components must first be 
secured* either with the transit-instrument or by micrometrical 
refei*ence to some adjacent star.^ From the proper motions 
of binary syetems information of the desired kind may, in 
some cases, be elicited ; for the track pursued by each star is 
necessarily, if their orbit be seen under a fairly wide angle, 
a Binuous one^ like that of the moon round the sun» while 
the centre of gravity of the two advances uniformly in a 
straight line. Now this neutral point was, by Mr. Stone in 
1876,'' and again after twenty years by Mr. Roberta,* fixed 
about midway between the components of « Centauri. They 
are, accordingly, almost equally massive; yet the ratio of 
their luminosity is as four to one. Sirius, again, deviates 
from rectilinear motion to nearly half the extent that its 
companion does; it is, then, twice as maaaive, but 4300 

* Pickering, iVmr. Anur. Acad, vol viii. p. « {1881J. 
3 Monthly Natitts, voL xxxtL p. 258. ^ Aair. Nach. No. 33S0. 




STELLAR OKBITS 



175 



rtimee more lumiaoua. And evidence of the same nature 
shows ProcyoD, while outshining its satellite 4700 times, 
to outweigh it no more than seven times. Analogous 
instances are gradually multiplying. Profeseor Comstock 
"has aftcertained that the faint companion of 85 Pegasi con- 
itaina more matter, in the proportion of five to three, than its 
'coxnparatively brilliant primaiy;^ and similarly, a fourfold 
predominance in mass is assigned by M. Adalbert Prey to the 
fiix times lees brilliant member of the interesting pair 70 
' Ophiuchi.* There remalD si^ couples, the relative masses of 
which have be.en determined by Mr. Lewia from the Greenwich 
meridian cbaervations. They are: 7 VirgimSj f Heroulis, 
Tf Casstopeiffi, 0^ Eridani, f Urwe Majoris, and f Scorpii. 
Only for the first a normal regult (if we may call it so) was 
obtained. The stars of 7 Virginis seem perfectly matched 
in every respect — in average lustre, in speetrumj and m 
atti^etive power. In all the other pairs, varied degrees of 
disparity were met with, the component inferior as regards 
light being uniformly superior in the government of motion. 
In the system of ^ Herculis, which revolves in thirty-five 
years, the mass seems evenly divided, while no more than 
one-aixteenth of the total light is given by one of its members. 
Mr. Lewis has, moreover, detected anomalies in the move- 
ments of this pair suggesting the perturbative influence of 
fttL invisible body. The disproportion between mass and 
luminoaityi which appears to be ratber the rule than an 
eicception among binary stars, has a most important bearing 
upon their physical history; but we shall not learn its full 
meaning until distinctions of spectrum are correlated with 
discrepancies in relative brightness. 

Professor Pickering * and Mr. Monek * of Dublin separately 

lived the existence of a rehition between the movements 

magnitudes of bmaries, rendering it possible to determine 

' their comparative luminous power proportionately to mass 

quite independently of their distances. It is necessary, 

K^»"«Tever, to assume either that the components of each pair 
AitTcqfh. Joum, vol. xvii. p. 223. 
fine, ^m. Jcad. vol. vlli p. 14 (IfiSl). 
(Aitrvatory^ toI, z. p. 96 ; ICncwUdge, tq], jai. p. lil ; Jtntm. Sri*. Attr. 
An, rol viii. p. 179. 



^ AatT, Nach. No. 394S. 



1T6 THE SYSTEM OF THE STARS 

are of similai- quality in this respect, or elae that one d 
them is of negUgeable mafia ; nor doea the formula distioguidi 
between extent of surface and intently of ablaing. TU 
bodies cotididered may owe their differences in "maa- 
brightness " either to variety in mean density, consequently in 
the extent of shiumg siirfdce, or to variety in actual briUiancj-, 
area per area. The etfectfi observed by us would be the supt 

The results obtained confirm the prevalence of astonishing 
v&rietieA in the emissive powers of stars. Taking ( Vztse u 
the standard pair, Mr. Gore calculated in 1891 ' the relafcire 
brightneas of about ^ty binaries moving in determined 
orbits. The tabulated figures are curious, even setting aside 
such objects as Castor and 7 Leonis, whieh came out phe- 
nomenally brilliant beoauee of the exaggerated length of the 
periods assigned to them by earlier computers. The duski^t 
stars in the list are those composing 0^ Eridani, which revoN 
in ISO years, and considerably exceed the sun in gravitatii]^ 
power, although claiming respectively only ninth and tenth 
magnitude rank. Admitting their density to be the BamB 
as that of f Ursae, they seem to be intrinsically 660 time* 
leas luminous. On the other hand, X Ophiuchi and y Coron* 
Borealis are, the first, twenty-nine, the second, twenty-oofl 
times more brilliant ; if HercuKa and 7 Virginis, notwith- 
standing the difference of their spectral types, are about on a 
par in this respect, each couple sliining with five times the 
standard lustre ; while TO Ophiuchi and if CasBiopei^ main- 
tain the similarity of their general character by their agree- 
ment in poflseasing a aurface-brightneea only one-third that 
of f Ursaj. The only rule governing these diveraitiea is that, 
on the whole, Sirian stars are much more lustrous propor- 
tionately to their masses than solar. Corresponding data 
regarding stare with banded spectra would be especially 
valuable ; but are unattainable, since objects of the kind are 
usually inimobiJef or inordinately alow in circulation. 

The mo8t striking general peculiarity of binary orbits 
their high eccentricity. Nearly all of them are greatly more 
oval than the planetary paths round the aun, and a large 
proportion approach to cometary shape. In the "mean 
stellar ellipse," the focus of revolution is situated just half- 

^ iV«, Soy, IrUh Acad, yqL i, Nft t, 8rd x^riBS. 




STELLAE OKBITS 



X7T 



between tlic centime and one of its extremitiee. The stars 
are, acoordingljr, aeparuted ut apaatron by four times their peri- 
aatron tlistance. " The average eccentricity among the double 
fttara/' Di\ See writes/ " is more than twelve times that found 
in the planetary system, and this extraordinnry reault is 
manifestly the expression of a fundamental law of nature," 
Indeed, an improved star-orbit is commonly more elongated 
th&u the one it replacea Pi'ofesfior Kuasey, for instance, in 
Tecalciilitting the elements of B Equulei, found it necessaiy 
to triple the previously assigned ecceutricity. 

The circulation of binary stars is indifferently retrograde 
or direct. No tendency is perceptible towards concordance 
regarding the direction in which their tracks are pursued 
Nor do they affect a common plane. They lie at all posdble 
angles towards the Milky Way, and seem wholly exempt from 
ita influence.* Ths existence of a fundamental plane of 
movement would be of high significance as regards the history 
and relations of the sidereal world ; yet the whole drift of 
modem research suggests, rather than the close and rigid 
union between its parts which it would indicate, a loose con- 
nection destined to be extensively modified by tijue, 

^ EvoWtitnt GfUtA stellar Syxtems, y. 261. 

' Ihid. |i, 24h, An aaalogoiia conclusion ha.d been aomevliat eatlier ruched 
by Miw Alice Ereritt. 



IS 




The light-changes of double stars are commonly of a fitfal 
aad icdeciaive kind. They may a£fect one or both members 
of stationary pairs ; but visibly revolving stars, aa a riile, 
conspire to vary, if they vary at all. The alternating fluc- 
tuations of y Vii'giuis, discoverable only by close attentiou 
to the swaying balance of lustre between the components, are 
in this respect typicftL Eaeb may be described as nonnally 
of the third magnitude, and eaeh in turn deeliues by about 
half a magnitude and recovers within a few days, yet so that 
the general preponderance during a cycle of several years re- 
mains to the same star. The eiistenee of this double perio- 
dicity was recognised in 1851 by M. Otto Struve, who» how- 
ever, despaired of inveetigating it "with suceess in a latitude 
where the stars subject to it never rise more than 30" above 
the horizon.^ Xor has anything more definite been sjnoe 
learned on the subject. Instability still persists. It is made 
evident by frequent inversions of the position-angle, accord- 
ing as oae or the other component, taken to be ^perior in 
brightneaa, is chosen as the origin of measurement;^ but no 
trace of regularity ia apparent in these inversions. 

The pair circulates in the moat eccentric of ascertained 
stellar orbits (see Fig. 24). The elHpse traversed by 7 Virginis 
in 194 years is, in fact, proportionately somewhat narrower 
than the path round the sun of Encke*s comet> bo that the 
stars -mil in 1934 be separated by at least thirty timeB the 
sky-interval between them in 1336, when they merged into 

' OhMTvations de Pautkovsa, t, ii. p. \2Z. 

' OorSj Knoioiedgi, vol. xxil. p. 201. 

17s 



VAltlABLE DOUBLE STABS 



179 



single telescopic object A spectrum of late Sirian pattern 
combined -vnth a perceptible tinge of yellow in tlieir light 

Kelative variability ia in 44 Bootis still more marked than 
, 7 Virginia, But here a fundamental disparity between the 
iponenta ia seldom and temporarily aboliahed. Noted by 
[ereehel aa considerably unequal in 1781, they appeared to 
perfectly matched in 1787- And it is worth noting that 
iey had in the interim passed pcriastron* Struve observed, 
fune 16, 1819, a difference between them of two magnitudes, 
^hich had fiunk to half a magnitude in 1833. Argelander 
Bund them precisely equal June 6, 1830; Dawes perceived ^ 
Ipril 27, 1841, a slight advantage on the Ride of the usually 
smaller star; while the superiority of its companiou was recorded 
by M, Dun^r at Lund as ranging^ during the years 1869 to 
1875, from 0'4 to 1'3 magnitudea^ Since their changes, 
often simultaneous, are not always in the same direction, their 
combined variability has never been conspicuous, the brighten- 
ing of one tending, on the whole, to neutruliee the fading of 
the other. The stars of 44 Botjtia, according to Doberck's 
elements, traverse a highly eccentric orbit in a period of *261 
years. Their tints, varying from yellow and sky - blue to 
white and dull grey, cannot be without infliueuc-e upon their 
photographic magnitudes, determined at raria in 1886 to be 
5"3 and 6. Their joint light, though of the same spectroscopic 
quality, has then only one-twelfth the intensity of that of 7 
VirginiB. 

The component stars of ^ Bobtia when photometrically 
measured at Harvard College in 1883 were of 4'4 and 4"8 
magnitudes, but the order of their brightness has been at 
least three times reversed during a century of observation,' 
Their period of revolution must be of prodigious length. 
From 1796 to 1841 they appeared fixed; then a very slow 
wheeling movement became perceptible, accompanied by a 
diminution of distance, and it now taxes the powers of the 
best telescopes to divide them.^ Their apectnim is of the 
Sirian type. 

' Lwtd Obtervatie^^9, 1876, p. 74. 
Satvard AnnaUj vol. xir, p. 4fi8 ; Obxrvaiunu de Povlkffuxij t- Ix. p. 
; Duater Mituret Mi6rovLariquei, p. €8. 

" Croas]ey, Handbook cf Bmtbie SiaT$, p, 299 ; Tftmut, Jottrn. Liverpool 
A»tr. Society, vol. v. p. 77. 



THE SYSTEM OF THE STABS 



An aoalogoaa object is a Pisciom, made ap of a fourth «i>d 
ft fifib auigoitude stAr at 3* distaooe, and revolriog in a penoi 
ttolikdy to be mucU lea» than two thoonad jeafR. Tht 
larger certainly rahea in light, and perhaps also in ooIooe, 
the RomUer oertaitily in oolour, and perbape alao in bght* 

An obeenradon made by Mr. Tebbatt in New South WaW 
August 22, 1887, gave a unique proof of the relative Tuit- 
bility of a close double star in Virgo (OS 256). At its ooculto- 
tion hj the moon on that night, the chief pftrt of the Light 
went out with the disappearance of the reputed leaser «ttf, 
the Lomponent whicb had of late passed for its |Htmaif 
ling still for a few momeuts sepdrately but dizuty 
\T«bIe.* Similar but less marked reversab had already been 
noticed by 0, Strave and Beiubowski in this elowly circidat' 
ing pair.' 

Dr. Anderson haa alleged convincing proofs that Kridani, 
now of the third magnitude, was assigned first rank by PtolMn5 
and Al Sdfi/ The star is a fine pair just perceptibly revolv- 
ing;* but there is nothing to tell whicb component hw 
Buffered most from the ebb of light since the tenth century, 
or whether it has affected both equally. The kind of varia- 
bility mofib distinctive of double stars might be described aa 
a tilting of the luminous balance, now in one direction, now 
in the opposite, according to no settled law. It is exemplified, 
^•ocordmg to M. flanunarion in 7 Artetis; it is, or has been, 
ftlao present in 6 Serpeutia, 38 Geminorum, w Bootis, e Arietis, 
and many other couples, most, if not all of which give spectra 
of the Sirian type. Their agreement in the possession of this 
particular quality of light is the more remarkable Irom its 
being tije badge in solitary stars of e:[ceptional stability. 
Every *' white star/' bo far known to be variable, has proved 
also to be compound; and those of the Algol class are so far 
from making an exception to this rule, that they rank with 
the swiftest of poasihly existing binades. 

Amoug the very few helium stars which have had 
periods of light-change assigned to them are S Orionis and 



' EtMrvarU Annali, vols. xi. [i, 112 ; xtv. j^ 433 ; Flammama, Catali 

p. n. 

^ Obtervatory^ vol, x, p. 381. 
* KnowUdQtt vq\, rri. p. 134, 



^ Obt. dA Pwlkcwa, t. ix. p. 
" tmiea, lU/trtnce Cat. p, 20 




VAKIABLE DOUBLE STABS 



181 



MonoccTotis ; and in each case on dubioue gi-ounde. The 
first is widely double ; the second is the leading member of a 
straggling cluster, and was thought by Winnecke in 1867 
to change from 4'9 to 5*4 magnitude in 3^ 10'' 38°. It 
has two close attendants^ both probably fixed. The system 
has no appreciable proper motion. Both S Orionia and 
S Moaocerotia are now known to be spectroscopic binaries. 
The first was detected ae such by M- Deslandres, the second by 
Professors Frost and Adams. 

y Virginis combines the display of a firet-type spectnmi 
with liability to considerable, though inteniiittent, variations 
in light.^ Struck by its peculiar brilliancy, .Tune 6j 1866, 
Schmidt investigated its history from the tenth century down- 
ward, and concluded it to fluctuate irregularly from the fifth 
to the eighth magnitude. The anomaly of such changes in 
a Sirian star was brought more into harmony with other 
example© by Biirnham's division of it at Chicago in 1879, 
into two nearly equal components less than half a second 
(0*''47) apart.^ His subsequent observations have given no 
satisfactory evidence of alteration, eittier in brightness or 
position, during twenty years.* This singularly interesting 
system should not be neglected by the poBseasora of great 
telescopes. 

The variations of U Puppis in fourteen days, from 6 to 
6*8 magnitude^ detected by Mr, Espin in 1883,* derive added 
interest &om the strong probability that they integrate the 
changes of two close components. The star is the chief 
meanber of one of Struve's wide fi^ced couples {% 1097) ; after 
being "elongated" by Dembowski in 1865, it was fuUy 
resolved by Burnham in 1875 into an unequal pair at an 
interval of 0"^80. No symptoms of orbital movement have 
aa yet been derived from it, and those of luminous in- 
stability seem suppressed. The case recalls the effaced 
periodicity of S Monocerotis, and is, in a measure, typical. 
lie spectrum is of the solar type. 

A corresponding long-period star is ij Geminorum, per^ 



' AitT. Xtah. No. 1597 ; ffrtrvard Annals, vol xiv. p. 156 ; Qore, JTjuw- 
ledgt, vol, xiii. p. 304. 

» Obtervatory^ vol. iii. ji. 192. ' Oiri. Cat. p, 128, 

* MonChiy Notices, vd. xUiL p. 433. 



182 




THE SYSTEM OF THE STABS 



ceived by Mr. Buruham during a visit to Mount Hamilton in 
1881, to form "a splendid unequal pair, likely to prove u) 
interesting sjstem." ^ Its revolutions, shown by remeasu 
menta in 1897 to be in alow progress,- will deserve the mc 
attention that they are the first attributed to an Antaria^ 
8tar. Aa a variable^ ij Greminonim may be described as od 
abortive Bpecimen of the Mira class. Its phaaes, run through 
in a period of 229 days, are always ill-marked» and at tima 
almost whoUy suspended. The share of the companion in 
bringing about these arreats of change has yet to be de- 
termined. Indeed, the more potent influence may belong Ui 
a SGOoud, closer attendant apparent only through motion-shlfte 
in the spectrum of the variable, detected by Professor Campbell 
in 1903. Should it prove to revolve in 229 days, the fact will 
be of great importance to the theory of long*period variability 
The ahort'period variable, & Cephei, governs a somewhat 
aimilarly constructed Bystem ; but its sea-blue satellite is 
remote, and appears stationary. The primary of the chro- 
matic pair, ^ Cygni, too, is undeniably variable, though its 
variations are of an elusive kind. It shifts quite capriciou^y 
ou the light-scale from 3*3 to 3'9 magnitude;* and recovers 
by the same imperceptible gradations that it lost brightneae. 
The fluctuations long imputed to the companion of S Cygni * 
are probably of atmoapherio creation.^ The object, at all 
times difficult and delicate, ia readUy obliterated by air- 
troubles ; the more readily, perhaps, because of its variability 
in colour. Struve found it of an ashen shade from 1326 to 
1833; in 1836 of a bright red.'' It has since generally 
appeared blue ; bub Dun^r saw it once olive, at other times 
red ; and intervals of greyneea are on record. No orbit yet 
computed for this pair inspires much confidence; Mr. Gore's, 

kwith a period of 377 years, is doubtless the best derival 
from inaurtieient materials. 
The presumption of sympathy in light-change betwe 
intimately united stars, although recommended by appearanc 



Mtmlhiy Notices, vol. ilvii. p. 204 ; Astr. Nacli, No. 2&30. 

* Gen. Cat. p. 76. 

* Klflin, Astr. Nach. No, 1663. 

* HeTeefael and South, PkiL Tratu. voU. ciiv. p. 331) ; oiri. p. 376. 

^ Duair, Miituret Jilicriftnitriquet, p. 118. 

• ^^OoloreamitiaegWigiUB," Jfnw. JUicrom. {k 297. 



VARIABLE DOUBLE STAES 



I; not yet been strictly tested; but the converse proposLtion, 
iftt agreement in ILght-cUange implies physical conneclioD, ia 
of all but eelf-evideut trutb. Two variables in Cygnus/ for 
example, situated 24" apart, may safely be assumed to con- 
stitute a system, their " ruddy and ccerulean " tints being a 
confirmatory circumBtance. A still more etrikiug combination 
presented by U Caasiopeiae and a blue companion at 59'', 
rith which its atroug red glow at times contrasts splendidly. 
le principal star fluctuates irregularly from the sixth to 
low the ninth magnitude; the attendant from the eighth 
''to the tenth. The probability of their being united by a 
special tie is overwhelming. Accordant variabiUty of a 
conspicuous kind is an argument for its existence to the full 
as convincing as the posaeasion of a common proper motion, 

The crimson tint of U Gygni, discovered by Mr. Knott in 
1871 to vary from above the eighth to below the eleventh 
magnitude in a period of 466 days, was described by Webb as 
" one of the loveliest lq the sky." It is set off by the blue 
rays of a companion at 6'', which seems to fluctuate in 
colour, but little, if at all, in light. Their azure is, however, 
no mere optical efifeet of contrast, since (though capable of 
fading independently) * it survives without alteration the 
telescopic extinction of the adjacent red luminary. V Cygni 
is the only star belonging to the fourth spectral order open to 
a suspicion of being in systemic connection with a neighbour. 
A good many variables have satellites as to which no such 
j8picion arises. Thus, the ninth-magnitude star within 46" 
of the beautiful " carmine " tinted object S Orionis is undis- 
tinguiflhed either hy colour or change, and they hence very 
likely form only a perspective couple. The same inference 
applies to three small stars contiguous respectively %o 
R Cassiopeiae, R Crateris, and Mira Ceti, detachment through 
the proper motion of the variable beings in the last case, 
visibly in progress. 

The light-changes of connected stars indicate duplicity as 
le of the causes tending to produce fluctuations of a certain 
ill - pronounced type. For we can safely assert, from the 

k ^ h l470 = LBlaDtio 3B428. 

" * Birroingham, Trofts. E. Iritfi Acad, rol. ixiri, [i. 3O0 ; GommiU, English 
Maehanie^ ral. xlvi. |>, 340. 




TKE SY3TKM OF THE STA£S 



184 



f chuacter of their ^ectn, that todst of tbe objects 
tbem would shine stndiljr if single. This relation is 

the more significaDt through tbe poeaibility first obscurelj 
Lroi^ht into view bj Sir Norman Lockyer's meteoiitie Ui«0>7, 
that rariabibtj of evei^ kind depends for its prodnctios 
upon external action bj doflely drcoUting and to us invisitile 
bodiM. A te«t too may be fumisbed by flnctuatibg couples 
to tbe opinion that luminous insUbility belongs to a late 
stage of Btellar existence. The contemporaneous origin and 
similar constitution of members of binary systems axe indi- 
cated to our minds as highly probable. If this be so. 
development should, according to the received opinion, pco- 
oeed, other things being equal, quickest in the amalleBt 
m&aaes, mote slowly in the larger.^ Hence, if it were troa 
that variability accompanied decline^ compauion-stars should, 
one might suppo^, be far more unstable in light than their 
primoxies. But this anticipation is far from being realised. 
Variable primaries are more frequently met with than vari> 
able satellites, tt is scarcely, indeed, too much to say that 
in every undoubted case of variability in one member only of 
a pair, the member it distinguishes is the principal star. 
Nor, even if we were sure that the pace of evolution in 
Btar-syatema in prescribed by masa, would it be safe to 
conclude anything as to its distribution from observed diflfer- 
enoee of brilliancy. Mass and luminosity are not comparable. 
Predominance in light, as we have learned from repeated 
experience, is no valid argument for predominance in quan- 
tity of matter. This circum9tani« adds both difficulty and 
interest to the far-reaching question of the origin, history. 
and mutual relations of conjoined stars. 

' Lwk^er, Prvc. Jt Socid^, voL jUf. p. 90. 



CHAPTER XIV 



SPBCTROSCOPIC BINARIES 



The epectroHCopfc method of determining velocity has already 
beeu frequently referred to in these pages. We may now 
briefly explain the principle upon which it reats. Tins is 
easy to apprehend, but extremely difficult to apply. Cliriatian 
Doppler of Prague pointed out iu 1842 that the motion of a. 
} iimi nous source towards or from the eye, or of the eye 
towards or from a huninoua ft^urce. must alter the refrangi- 
bility of the incident light. It must dimiuieh it if the 
relative movement be of withdrawal ; it tnuet augment it if 
it be of advance. For refraa|pbility depends upon wave- 
lengthy and wave-length is the reciprocal of wave-frequency. 
In other words, the more vibrations are pressed into a given 
apace, the shorter they necessarily are, and the more re- 
frangible the ray they conspire to form. But clearly, the 
light-waven emitted by a body travelling towards iia arrive in 
quicker auccesaiou than if it were at rest ; they are shortened 
in the proportion of the rate of travel of the Inminoua object 
to the velocity of the propagation of light; and being 
shortened, they are rendered, ipso facto, more refrangible. 
An opposite effect is produced by receding movement. The 
Ught-wavea from a retreating body are lengthened ; there are 
fewer of them in an inch or a mile ; and they are accordingly 
rendered leas refrangible. 

AJl this was fairly obvioua ; bnt it seemed questionable 
whether any practical outcome conld be derived from the 
recognition of its abstract truth. The doubt was removed 
when Hippolyte Fizeau showed, in 1848, the feasibility of 
using spectral lines as staudards of reference for measuring 

1^5 




186 



THE SYSTEM OF THE STAES 



the changes of refraugibility due to ead-on motion. Tba 
rays, bright or dark, included in the spectrum of the travBl- 
ling object Daturally slxifb with the whole of its light-gamut. 
Bj juxtapofdtioti, accordingly, with correepondiug rays other* 
wise derived, and, of coui-se, in their normal places, the 
amount of their ahiftings becomes known, and heoce the 
direction and rate of the movement prodaciag them. Ev^ 
then, twenty years bad to elapse before Sir William 
HugginB made his pioneering experiments on the radial 
velocities of the stars; and twenty years further before the 
method was perfected by the adaptation to it, through 
Dr. Vogel's initiative, of the camera. 

The effects of motion on stellar spectra are illustrated in 
Plate X.J which reproduces, by Sir David Gill's kind per' 
mission, one of the admirable Bpectrograraa takett with the 
McClean telescope by h is asaistan t , Mr. Lun t. The raja 
analysed were those of a Phcenicis, a atar of ^ar type, which 
retreats from the buh at the rate of 82 kilometres per 
second, or nearly thrice as fast as the earth circulates in its 
orbit. Let ua consider how this is learned from the photo- 
graph. Five strips of dispersed blue light are included in it 
A section of the solar spectrum is shown in two of them ; 
the same part of the spectrum of the star adjusted to precise 
correspondence with the solar prototype, appears in the 
fourth strip from tlie top. The dark lines are in both 
almost identical i and that many of them claim an origin 
from iron-vapour, can be Been by a glance at the iron-spark 
comparison-spectrum, in which the rays, obscure in the sun 
and star, ahiue by direct emission. Now the iron*spectrum is 
given twice : in the second, and in the fifth, or lowest strips ; 
and there ia a conspicuous difference between the present- 
ments of it. One is strongly displaced relatively to the 
other. This displacement measures the motion-shift in the 
spectrum of the atar. The iron-lines below are fiducial ; they 
were photographed with the star, and served to determine its 
receding speed. The liues above were photographed with the 
sun, and being designed purely for purposes of identification. 
are ai^anged to coincidence with the star-lines. The glaring 
nature of the discrepancy between the adjusted and the 
non^adjust^d iron-spectrum enables us to realise the poaai* 




SPECTROSCOPIC BINARIES 



187 



bility of appreciating, by aimilof means, alterations of wave- 
length due to yelocities of no more than one or two kilometres 
a second. 

Solitary atarSj lite our own, travel uniformly. Their rate 
of transport does not vary sensibly from month to month, 
from year to year, perhaps from century to centoiy. But 
with binary atara the case is diflerent. Besides the constant 
velocity belonging to the system as a whole, they circulate 
round their common centre of gravity ; and the direction of 
their orbital movement as regards the earth is continually 
changing. Take, for instance, a pair revolving in the visual 
plane. At some given moment their total speed of revolution 
will lie across the line of sight. It will then be spectro- 
Bcopically ineftective ; the rays characteristic of its light will 
be in their standard places ; no shifts will be perceptible. 
But when a quarter of a circuit from the points of con- 
junction ha.s been described, and the stttrs are at the opposite 
nodes of their orbit, the direction of movement, having 
turned through a right angle, will be straight end-on. Hence 
the spectral lines of the components will show contrary dis- 
placements corresponding to the entire velocity of their 
mutual ctrcUug. Two good spectrograms, accordingly, taken 
at the critical instants of conjunction and elongation, suffice, 
at least theoretically, both for the elimination of translatory 
speed, and for the determination of the actual rate of 
revolution ; and this, when the period is known, gives the 
size of the orbit in miles or kilometres. Evidently, spectro- 
scopic meaaui-ea at such times of visual pairs with securely 
computed elements would make us acquainted with their 
distance from the earth and joint gravitating power — 
even if their paths were inclined, at a considerable, though 
a known angle. The success of such investigations, it is 
true, long seemed remote ; but it has now been achieved. 
Prafesaor Hussey, in 1903, determined for B Equulei a 
" spectroscopic parallas " which inspires no small confidence.* 
And still more recently, Dr. Palmer of the Lick Obaervatory 
has computed, from the radial velocities of a Oentauri, a 
value for its parallax in all bat exact agreement with that 
.derived from the most skilful heliometric observations.^ Few 



^ Lick Bulletin, No. 32, Aee ante, p, 1S8. 



3 lUd, No, «0. 



THE SYSTEM OF THE STAES 






telescopic bitmries, liowever, are circmnat&noed favoiirably 
enough to invite the application of the method. The con- 
ditions propitious to it a.re mpict motioti in a plane not much 
inclined to the sight-Une, and considGTD.bie brilliancj in both 
the moving objects. Pairs tlxat are easily divided mrely 
circulate i|uickly enough for the discrimination of differences 
in their radial velocities; and the most rapidly circulaLing 
pairs are those most apt to defy visual separation. The 
epGCtroscope, in fact, finds its opportunities just where the 
telescope encounters batHing difficidtiee; and this reciprocal 
relation, while it hiifi lielped to give almost indefinite et- 
teneion to our acquaintanee with stellar systems, restricts 
opportiuiities for combining the methods, bo as to connect 
the angular measures furnished by one witb the linear orbital 
dimeneioiiB obtained by the use of the other. 

8(ai« revolving in or vety near the visual plane must 
undergo, at any rate, partial eclipses. They are comparatively 
fow; the great majority move in. patba sensibly, thou, 
variously inclined to it. But hs the tilt increases, tbe 
portion of the velocity available for spertroscopic measniC' 
ment, because directed ratlially, obviously diminishes. Wholly 
inacce^ible to it, in fact, are such couples — i^nd they may 14|I 
very numerous — as circulate approximately at riglit anglee to 
the li Lte of sight For those with orbits intermediately 
situated, the apparent or measured radial velocity divided by 
the cosine of the angle of deviation from horizontality gives 
the actual radial VL>locity. If, for instance, the inclination 
amoimted to 60°, only half the speed of revolution would tell 
in Hue - displacements. An average value for it of about 
30" is regarded as probable;^ and this would give 100 
to 115 as the proportion of the observed to the true rates of 
circulatory motion. But estimates of the kind serve only to 
mislead if applied to inttividual systems. The safer oourae is 
to admit that, in the absence of occultabion-phases, we cau 
secure only minimum values for the velocities, distances 
apart, and masses of binaries spectroscopically detected and 
observed. ■■ 

They are extraordinarily numerous. Professor Campbell' 
showed, in 1902,^ the probability that, among the entire 
* Barr, Astroph. Jexim. toI. %i. p. 248. ^ Litk Bultetinf No, 




SPECTKOSCOPIC BINARIES 



189 



kidtitucle of stars, one in six or seven is m constituted j and 
Professor Frost added wonder to wonder by the Btatement 
that helium-stars in particular yield ^ large a harvest of 
binariea that the chances are nearly even whether a given 
object of the kind will r^ist oc euixender to the reBolviog 
powers qf the spectroscope.^ One of the moat curious facts 
about close star-pairs is the prevalence of extreme inequalities 
of lustre between their coiuponentB. In very many, one is 
brilliantly luiniuaus, the other so far obacure that no spectro- 
graphic impression can be derived from it. Some Batellites, 
again^ shine dimly ; traces more or leas pronounced of their 
action on the sensitive plate are perceptible ; while finaily, 
twin-stars are occasionally met with scarcely distinguishable 
by quantity or quality of light To this category belong ^ 
VtHiQ Majoris and ff Aurigso, the former detected by Professor 
Pickering in 1889, the latter shortly afterwards by Miss 
Maury. In both, the spectral lines are periodically photo- 
graphed aa doublets, when the components of each pair, at 
their " elongations " are travelling full speed ahead and aback 
respectively, towards and away from the earth. The separa- 
tion of the lines thus corresponds (allowance having been 
made for the constant velocity of the system aa a whole) to 
twice the rate of circulation, diminished, as we have ex- 
plained, in proportion to the inclination of the path traversedr 
Twice in the coutsb of each revolution, stars of the kind we 
are now consideiing exhibit the phenomenon of doubled lines ; 
at the intermediate epochs of conjunction, they give a single 
spectmrn. By timing these alternations, a period of twenty 
days has been deduced for ^ Ursfe^ one of four days for )3 
AurigEe. Miss Maury's binary, accordingly ^ changes its 
spectral aspect from day to day ; and the indicated relative 
velocity of 150 miles a second shows it to possess at least five 
times the massdvene^s of our sun.^ 

A magnificent system of this type, but in comparatively 
slow circulation, is formed by Capella.* The light of a purely 
aolar star, and of a companion somewhat less bright and some- 



* Frost and Adftms, AHr^h. Jonm. vol. irriii. p. 383. 
> Vogel, Silzungebfrichfc, BerLiD, Minih 10, 190*. 

* Campbell, Lick Sulietin, No. 6 ; Attroph. Jovm. vol. z. p. UT ; Newa11« 
JiMlhijf Notiat, V6t. Ix. p. il6. 



190 THK SYSTEM OF THE STAKS 

what leas maaaive, of the Prtx:yoa variety, are combined m 
this object; and they mutually revolve in 104 days at th« 
apparent or tuinimum rate of about seveoty luilea a Beconl 
The parallax of Capella, though email, la assured ; it gives t 
distauce of forty light-years, wheuoe we can infer that the joint 
lustre of the components exceeds more than a hundredfold th^ 
splendour of the aun. 

Gouplea coiiaisting of two bright membora are much 
of detection than those of which one is obscure. Duplicated 
rays in a Bteltar spectrum attract attention even on a crowded 
plate exposed with an objective prism, iu a manner to secure 
wholesale records ; but the oscillations of single lines reqaire 
the aid of a slit and a comparison-spectrum to make them 
evident. They evade notice unless the stars are separately 
studied. Yet, despite observational difficulties, about 75 per 
cent of the apectroaoopic binaries (to the number of nearly 
130) known at the end of 1904 have sensibly obscure com- 
pauious. The first noti-eclipsiug star recogniaed as double by 
the mere swinging to and fro of solitary lines was Spiai 
(a Virginis). Dr. Vogel thus determined in 1890 its revolu- 
tion in four days round an attendant later shown to be very 
feebly luminous. Both members of the stately visual pair, 
Castor, on the other hand^ circulate swiftly round bodies giving 
no sensible light. The quaternary system thus formed is of 
unique interest. Dr. Curtis^' the discoverer of one of the 
dark attendant stars (the other was found by Belopolsky in 
1896), has determined for it — provisionally assuming tha 
correctness of Doberck'a elements — a spectroscopic parallax of 
O''*05, corresponding to a light-journey of 65 years, and 
implying a total mass for the four constituent globes nearly 
thirteen times that of the sun. But it seems to be very 
unequally apportioned. One of the revolving pairs exerts thrice 
the attractive power of its companion, while emittiug only half 
as much light. The gravitational primary is the visual satellite, 
and vice versd. Only through the association of telescopic with 
spectroscopic measures, practicable by a rare exception for 
Castor, can inferences regarding the mass of bright and dark 
pairs become legitimate. Nothing definite can, under ordinaiy 
circumstances, be affirmed oa the subject. For one com- 
' Liek Bulletin, "So. 70. 





SPECTROSCOPIC BINAEIES 



m 



ponent beiog iuvisible, the size of their relative orbit remains 
absolutely undeterrained ; and the size of the relative orbit is 
a datum esaeubial for computing the mass of the system, or 
even a minimum value for the mass, which, in our ignorance 
as to the be of the orbit, is the utmogt we could hope to 
arrive at. The relative orbit, aa already explained, is the 
Biim of the separate orbits; or, putting it otherwise^ the path 
described by one &tar round a companion taken to be at rest. 
Trom its span, in conjunction with the period, the noaas 
of the pair la dcdncible. There is no means of learning it 
without knowledge of these two elements. The masses of 
bright and dark pairs can then^ aa a rule, only be guessed 
at with the unwarranted help of arbitrary hypotheses. That 
of Spica> for instance, has been estimated at 2*G times the 
solar mass ; but the value has only the authority of a plausible 
conjecture. 

From the measured movements of stars with shining com- 
panions, as already explained, the least passible amount of 
matter present in each system may be learned. The orbit spectro- 
scopically determined being the foreshortened representation 
of the orbit actually traversed, is smaller in an unknown 
degree ; and the mass of the moving bodies varies aa the cube 
of the orbital radius. The error of mass- determinations hence 
grows in a tripliciLte ratio to the eiTor of the linear dimensions 
directly obtainable. Minimum values, nevertheless, are better 
than none at all ; they fti a limit of great importance ; and 
for one species of crypto-double stars they may almost count 
as absolute values. These are eclipsing pairs in which both 
eomponente are bright. Y Cygni is a favourable example. 
The opposite movements in its system may be speetroscopically 
elicited ; the size of the relative orbit will thereby be given ; and 
since its plane is iixed by the occurrence of a twofold eclipse, 
^he joint mass of the components can be ascertained without 
ubious asaumptioaa. Several Algol -variables appear to be 
Similarly constituted; but their investigation awaita the 
leisure of overtasked spectrographers. 

Few compound objects are more inviting to research than 
stars variable in short periods, though not through eclipses. 
Six have already been detected && spectroscopic binaries; and 
the resolution of all may be regarded as de[)ending merely 



m 




THE SYSTEM OF THE STAIiS 



upon time and facility. The extreme interest of the qucsti 
connected with them is exempli^ed in ^ Oominonim^ 

radial velocity of which was found by Profefisor Campbell 
b© subject to a double periodicity, due, he surmised, to 
influonco. of a second dark satellite ; ^ yet the stable const] 
tiou of n triple system on the indicated conditions seems 
a hopelesa task. Again, * Pavonis varies from 3*8 Uj 
5'2 magnitude iti a mean jjeriod of nine days, affected, 
according to Mr. Roberts's observations,^ by an inequality 
amounting to ten hours, comprised in a cycle of eight yeara 
The duplicity of the star was ascertained by Dr. Palmer in 
1904;^ and before long an answer may be furnished to the 
critical incjuiry whether the disturbance of the light-period 
is reflected, or perhaps originated, by a disturbance of the 
gravitational period. 

Nor IB it short-period variables alone which yield to the 
motioU'test. The division of stars that fluctuate irregnlarlj", 
or in long periods, is of not infrequent occurrence. Such ia 
1} Geminoi'um, one of Burnlxam'a unequal pairs, raised to a 
higher plane of complexity by the addition of CampbeL'a 
spectroscopic attendant.* 

A period of about forty days is partially conformed to by 
the ebb and flow of luminosity in u Herculia, a helium-stAr 
visible, even at its dimmest stages^ to the naked eye. Detected 
aa binary by Mr. Adams* at the Yerkea Observatory, it revolve* 
round a seemingly dark companion at a high speed, but in a 
period which has still to be ascertained. The investigation 
of the system constituted by e Aurigse is alflo incipient 
Dr. Vogel's resolution of this object in 1902 into a bright 
unequal pair, with a relative velocity in tlie line of sight of 
nearly twenty-five miles a second, will be iu the minds of our 
readers.^ Now « Aurigae ha.s, during the greater part of a 
century, been remarked for slight phases of light-change held 
to be quite capricious in their occurrence. Should Dr. 
Ludendorff's regolarisation of them be verified by future 
observations, their twenty-seven year period will rank as by 



^ Astroph. Jounk, toL xiii, p. M. 

^ Hd does Bot yilBtio implicit confideuco in them, Astr. Joum, No. 
» Lick Bu-lirtin, No. flO. * Ibid. No. 20. 

" Jttrapft. Joum. toL ztiL p. 261. * 3e« anU^ p. 10£. 




SPECTEOSCOPIC BINARIES 193 

/he longest that can be ascribed to &ny form of stellar 
-change. And it will be of great importance to correlate 
it, if possible, the star's period of orbital revolution. 
3 is, perhaps, no object in the heavens of mote pregnant 
est than this variable spectroscopic binary. 



13 



CHAPTKK XV 



ITDUXPltE ETTABS 



Tuz further r^olvability of a great nuBuy double stars iB 
one of the most cuiioua resulta of modem improvenieDte in 
the optical means of obBerring them. With every additiao 
to the defiaiog power of teleeoopoB, the visible complezily 
of stellar eji'atema has increased so rapidlj as to inspire a 
Buspicion that simple binary combinations may be an excep- 
tion rather than the rule. The frequency with which those 
token to be such have yielded to disintegrating scrutiiiy 
Buggeets at any rate some innate tendency, indicating that 
the duplicity of etara is no accident of nebular condensa- 
tion, but belongs essentially to the primitive design of 
their organisation. Although we can never become fully 
acquainted with all the detailed arrangemeuta of stellar 
syBbemfi, we are then led to suppose them far more elaborate 
and varied than appears at first eight E^h, we cannot 
doabt> IB adapted by exquisite contrivances to its special end, 
reflecting, in its untold harmonies of adjufibjuent, the Supreme 
Wisdom from which they emanate. 

The continuance of the proceas of optical dissociation, begun 
by the splitting up of an apparently aimple star, sometimes 
shows the primary, sometimea the eatellite, not unfrequently 
both primary and satellite, to be very closely double. Ternary 
ayatems are accordingly of two kinds. In one, the smaller 
ebar conaiats of two in mutual circulation and concurrent 
revolution round a single governing body ; in the other, an 
intimately conjoined pair guides the movements of an un- 
attended attendant. The planetary type of construction is 
uncommon or unknown. No star has been ascertained to 



MXTLTTPLE STAIIS 



195 



possess two or more companions circulating co-ordinately. 
Groups poeeibly indicating such a disposition of parts exist, 
but perspective may have a share in producing them. The 
variable S Monooerotia, with its two client-stars, is an example. 
Another is mentioned by Dr, See. Four extremely faint com- 
panions, at distances between V &nd 39" from a si^th- 
magnitude star in the Poop of Argo (Cord. Q. C. 10534), 
have been detected by him and his predecessors. The group 
is thus commented on by him : — " Quintuple ! Probably a 
complicated physical system ; the only stellar system I know 
of OGnstructed on a plan in any way analogous to that of the 
planets/' ' There is, however, at present no proof that the 
artaagemeut may not reaultj at least in part, from the chances 
of perspective juxtaposition. 

Among tlie most interesting triple stars of the double- 
satellite description is the brilliantly coloured 7 Andromedffi- 
The original components, of third and fifth magnitudes 10" 
apart, remain in statu quo since they were seen by Father 
Mayer in 1777, but their secular journey together over an 
arc of 10" establishes the genuiuenpsa of their relationship. 
In 1842 the sea-green companion wag found to be itself double. 
With the 15 -inch Pulkowa refractor. Otto Struve caught 
sight of a thin black line (representing probably a gap of 
Home thousands of millions of miles), dividing it into two stars^ 
the fainter of about sixth magnitude, which within fifty-five 
years completed a revolution in an orbit not much less eccentric 
than that of y Virginia.^ 

A pair in some respects similar, but much fainter, is 
attached at about the same appareut interval to the lustrous 
white star Eigeh An excessively difficult object at the time 
of its detection, it later became impossible. Its elongation, 
suspected both by Burnham and Herbert Sadler in 1871,^ was 
verified at Chicago and Mount Hamilton in 1378-79. Then, 
for nearly twenty years, no sign of duplicity could be elicited 
from it.* Change^ under the circumstances, seemed much 
more probable than error ; and its rapid progress was indicated 
by Professor Aitken'a reobservatlon of the sapphire attendant 

* ^str. Jt»*rn. No, 431, 

* W. J. Huasey, iVW. Lick Ohurvaiory, vol. v. p. 4&. 

* A^r, Begisier, vol. itlu. p^ 16, * BumLom, Otn. Oat. p. 69, 



OH 



THE SYSTEM OF THE STABS 



upon RigeU* s» pslpoUy doaUe in 1898. Xevenh«ka, 
^fUr two yeuB, (he enttve satellite wu ftg«m undiaoeniibk' 
The riddte of ita exifllanioe maj be finally aoJved by the tid d 
a «pBctro|^niphic appflffstus of gteat light-collecting power. 

A ternaiy groap, ooim^iaatidiiiig to these two in plan hm 
gioatly enlarged in ai^ular scale^ oooiinsto of a 4*5 magnitiMh 
fftar^ deeignabed bj FUmsteed 40, by Bayer o* Eridani, wiifc ■ 
faint and iar-away double satellite already referred to,* all 
thtiee diaoovered by Herachel in 1783. The aflBooiation of the 
pair with the large star at bo gr^t an interval as 92* would b» 
improbable^ were it not certified by itieir poeseesioQ in oommtn 
of an exceptioaally swift proper motiotv Aa advance during 
the laat century over a spaoe nearly equal to a quarter of the 
raoonV diameter has modified their relations only by a trifling 
approach to their primary of the depeodeat &tars, due perhaps 
to slow circulation round it in an orbit presented edgewise 
to our flight 

They have, in the meantime, almoet fini&hed a circuit of 
one another, and will have completely finiahed it within about 
160 years from the date of their detection.' And since their 
distance from the earth has be«n measured, the real size of 
their orbit and their joint mass are also known. We find then 
that the average interval between them is thirty-eight tim« 
that separatitig the earth from the sua. &o that (their path 
being only moderately eccentric) they never approach aa near 
to each other aa Keptune does to our oeotral orb, which they 
together surpass 1"6 timee in gravitative power. But in their 
place, from which light reaches U6 in twenty years, the stm 
would shine as a fourth-magnitude star, while they combine 
into one of only ninth magnitude. Their feeble luminosity 
thua once more forces itself upon our attention, and compels 
US to reflect upon the possibility of whole aystems existing in un- 
impaired mechanical perfection, but wrapped in perennial dart* 
neea. For what purpose eiisting, who can tell ? The flight 
of our thoughti9 is short and the ultimate aims of the Maker 
are remote. Attempts to compass them are foredoomed to 
failure. 



' Pttp. Attr. December 189&, p. 585. 
' Liek BulUtin, N«. II, " See ante, p. 

* Bttmliiim, Monthly AbCieu, vol, Uii. pi. 478- 




iMULTIVLE STARS 



197 



Double primaries occur as freely as double satellites \ and 
'iheir comjnon centre of gravity presumably conatitutea the 
focua of attraction for their remote attendants. Castor 
Bplendidly illustrabes tliia plan of conatruction, carried out on a 
Taat scale in its system, which includes five members; the 
locid components, a far-off captive star borae in their train, 
I'And the obscure adjacent masses disclosed only by their 
ctroscopic eflects. Another specimen is e Equulei, one 
Herschel's pairs, the larger member of which was again 
ivided by Struve in 1835. The feat had become possible 
3ugh the progress of orbital motion, the continuance of 
ioh has since rendered it easy. Signs of circulation in 
7'5 magnitude star at 11"^ are scarcely, if at all, 
eptible. Yet it is an undoubted satellite of the close 
iple. 
The movements of the third star (of T"! magnitude) in 
lie ternary combination f Scorpii, seem to progress in an 
pposite direction from that of the close double star which 
controls them at an apparent distance of 7" ; but their nature 
and method are etill imperfectly developed. The primary in 
this system consists of two fifth-magnitude stars, formerly just 
eeparable with a gootl 4-inch telescope, but now only 0'*'7 
ajort. The orbit assigned to them by Dr. See ^ approximates^ 
in an unusual degree, to a circle, and ia traver&ed in 104 years, 
Their spectrum is of the Sirian type. The eighth-magnitude 
companion of e Hydrte has described, since its discovery by 
Struve in 1825, an are of over 40", at a distance of 3". Its 
blue tint ia charmingly set off by the warm yellow of the chief 
star, divided by Schiaparelli in 1888 into a difficult pair re- 
volving (by Professor Aitken'a elements) in leas than sixteen 
yeara PoMibly, the process of resolution has not reached its 
term; for Dr, Curtis measured, on 1900, fluctuations in the 
radial velocity of the leading component. They more probably, 
however, depend upon its revolution in the visual orbit,* Un- 
doubted spectroscopic binaries nevertheless frequently occupy 
analogous positions. Besides the systems of £* Ursae Majoris, 
and If Geminorum, those of f Ura^, ^ Scorpii, and tc Pegasi 
are thus composed. As a wide pair, k Pegasi was noted by 

' FlMUDarion, Catalogw, p. 139. » SltlUxr Syjfem*, p. 178. 

' LiciBulUtin, Noa. 4, U. 



Ida 



THE SYSTEM OF THE STARS 



Herschel in 1786, The queatlott is still open whether hb 
eleventh-magnitude companion has any physical cotmection 
with the nuclear group. This consists of a fairly equal piur. 
recognised by Borahom in 1880/ and found to revolve in 
11 '4 years; together with an unseen companion to one of 
them, detected by Campbell in 1900, and completing ite 
circaits in about six days,* 

The relations of tlie stars compounded in ^ Scorpii are no 
less remarkable. Two, of respectively third and sixth magni- 
tndes^ were first observed by Herechel in 1779; the third, 
discovered by Burnham a century later, makes with the 
primary an extraordinarily difficult, unequal pair at 0''"96.' 
A fourth has been Bpectrographically measured by Mr. Adama' 
It is a lustrous object with a helium spectrum, and revolves 
at high speed in a period estimated at 6** 2l\^ No visible 
relative movements have yet been perceived in the ^tem of 
y3 Scoipiij which nevertheless asserts its organic unity by the 
harmony of its advance througli apace. 

One of the moat curiously interesting of all the stellar 
systemfl known to us is ternary from an optical, quaternary 
from a physical point of view. It is composed of one 
obscure and three bright members, all in comparatively rapid 
mutual circulatiou. The division of f Cancxi, by Tobias 
Mayer in 1756, into a fiftli and a sixth magnitude star about 
5-J-" asunder was tlie preliminary to Herschel's further ajujlysia. 
" If I do not see extremely ill this morning," he wrote ou 
November 21, 1781, "the krg© star consists of two,"* Thifi 
was the earliest example of the decomposition of a double into 
a triple star. The next distinct view of these close objecte 
(called for convenience A and B, the remoter star C) was ob- 
tained by Sir James South at Passy in 1825, but Struve's 9-inch 
Fraunhofer showed them easily, and they have never since been 
lost eight of. Eeohaervation at once rendered patent their 
swift movement of revolution. Before the close of 1840 they 
had, by resuming the positions in which they were originally 

* Gemral CaUiiosuet p. 231. 

' AairopJi. Jmi-m. toI, lii. p. 267- 

* Monthly Noiiees, vol. xL p. 100 ; Geiierat Catalogve, p. 14S. 

* Aatroph, Jvum. vol. xiriii. p. 69. 

" Lowgll ObKrvaiiirti Bulletin, No. 1. 

* Croasley, Kandbaok, p. 247. 



MTTITTPLE STARS 



199 



observed, authoiitratively declared their period to be not far 
from sixty years. And their orbit lies iu a plane bo nearly 
square to the line of sight, that foreshorten ing takes little 
effect upon it, and occultations are hence imposaible. Although 
the maximuia inteiTal between the stare stsarcely e^cceeds one 
second, and the nLinimum interval ia no more than 0''^'2, they 
never doM up beyond the dividing powers of firgt-clas^ 
instruments. 

But the orbital movements of the couple A B make only 
port of a complex scheme of displacements. " This atar /' Sir 
John Herschel remarked in 1826, "presents the hitherto 
unique eonabination of three individuals, forming, if not a 
Bystem connected by the agency of attractive forces, at least 
one in which all the parts are in a state of relative motion." ^ 
He added that, if really ternary, its perturbations must 
present ^' one of the most intricate problems in physical astron- 
omy " ; and Professor Newciomb holds it probable that the 
laws of motion in such combinations must, in general, be 
" too eompUcat^d to admit of profitable mBtberoatical investi- 
gation." " 

The star C apparently retrogrades round A B at an 
average rate of half a degree a year, indicating (if maintained 
with approximate uniformity) revolution in a period of 600 
or 700 years. But this average rate is subject to very 
remarkable irregularities. The path traced out in the sky, 
far from being a smooth curve, is looped into a series of 
epicycles, in traversing which the star alternately quickens and 
slackens, or even altogether desieta from its advance, while 
increasing or diminishing, by proportionate amounta, its 
distance from the centre of motion. This anomalous 
behaviour, detected by M. Flammarion in 1873,* was both 
detected and interpreted by Otto Struve in 1874.* The 
vagaries of the third component of ^ Cancri proved, from his 
investigation, to be very far from unmethodical. The accelera- 
tions which they included were shown to be perfectly com- 
penaated by retardations, and to be accompanied unfailingly 
by expansions outward of the parts of the track where they 

^ PhiL Trana. toL exvi, p. 32G. 

» The Stars, p. 181. = Catalopif, p. 49. 

* Cojfiptu Stndus, t. Ixzij. p. U63, 



200 



THE SYSTEM OF THE STAHS 



OGCurred, while contructiona iuw&rd attended slackened m<xv^ 
ments. An explanation too was hazarded^ the Bafastantiftl 
truth of whicli was attested by M. Seeliger's eluborftte Kaearcliee.^ 
It seems then that the atar G is merely a satellite to « 
dark body round which it describes, in 17^ ye&is, a Utile 
ellipBe with a mean radius of one-fifth of a eecond. Together 
this singular pair circuits, or, more probably, is cinsuited bjr 
A B, the iuviaible disturbing body being the moBt tnas&ive <d 
the system- If this be the case, it is also^ of oouree^ the moet 
nearly alaticiiAry, and should be regarded as the centre roond 
which the lucent trio revolve — an arrangement hinting to as 
that the collocation in the same orb, familiar to us in cIm 
solar domain, of the functions of rule and light-givingj mijt 
on Qccaaions, be dispensed with. An anti-Copemican system, 
at any rate, appears to be to some extent exemplified bf 
^ Oancri Here a ooolj dark globe^ clothed possibly with the 
vegetation appropriate to those strange climes, and plentifully 
stocked, it may be, with living things, ia waited on, for the 
supply of their needs, by three vagrant suns, the motions of 
which it controls, while maintaining the dignity of its own 
comparative rest, or rather of its lesser degree of movement 
For the preponderance of this unseen body cannot approach 
that of a sun over its planets ; hence its central position ia by 
no means imdisturbed. We must not forget, meantime, that 
its existence is to some extent hypothetical. Mr. Bumham 
thought it an evanescent creation of accumulated micro- 
metrical errors;^ and Professor Frost in 1904 failed to elicit 
any spectroscopic evidence of the eighteen - year period 
ascribed to the third star.* The close pair (according to 
Pickering's calculation) possesses nine times the solar emissive 
power relatively to maaa ; and all the three visible components 
show spectra of Sirian quality. Their real differences of 
magnitude, too, Heem to be slight, although at times 
exaggerated by relative variability. The entire group is 

^ SiUungalrricht'T^ Wiem Bd. Ixxiiii. Abth^ 2, p. 1016; £fcnk9eKriJtm, 
Munich, Bd. ZTii. Abth. I, 1SS9; Barzcr, Attr. Naeh. No. 2764; ObstrviHory, 
vol. sii. p, 116, 

* MoiUhly NoticM, April 1691, November 1B92 ; Aatr, arid AelTV^hyna^ toL 
lil. p. a72 ; SeeligGr, ibid. voL xiii. p. 802 ; A$tr, Naeh. No. 31fl& ; ^zungsb. 
Bt^ytf. Akad. Bd. xxiT, H«a. ill. 1S94. 

* AMtroph. Jvurn. vol xix. p. aSG. 




MULTIPLE STABS 



201 



irted through Bpace at the rate of IS"" a century^ but 
distance from the earth ia unknowru 

A quadruple system of remarkable type ia formed 
f Ursffi Majoria with three variously related bodiee. 
iden its spectroBCopic and telescopic attendants, the one aa 
;e aa the other is sedate in its revolutions, it clauns the 
escort, on it& indefinite journey onward, of the fifth-magnitude 
star Alcor» the two making the combination popularly 
designated the *' Horee and Bider." Since the interval 
between them is of 11' 30", they can easily be dietinguiBhed 
with the naked eye ; neverthelesa Alcor, totally overlooked by 
the GreekSj was regarded sis a test-object for keen eyesight by 
the Arabs. Its gradual brightening is thus strongly sug- 
gested.^ The probability that Mizar and Aloor mutually 
revolve ia strong, but not overwhelming ; their connection 
nUffht be otherwise explained. If they do, their annus magnus 
must be of enormous, to our ideas of interminable length. 

Heal quaternary stars are often self-discriminating ; their 
arrangement into two adjacent couples asserts physical con- 
nection more strongly than any possible distribution of three 
stars can do. And in effect, several perspective groups of a 
single star with a genuine pair, such as S Equulei, 35 Pegasi, 
7 Tauri, and ^ Delphini, are visibly in course of being dis- 
solved by proper motion, while no " double-double " combina- 
tion has yet given signs of breaking up. 

A representative specimen of the latter class offers itself 
in c Lyrse, a atar of the fourth magnitude, a little to the 
north-east of Vega. Exceptionally keen eyes show it as 
double, and one of the brilliaDt surprises provided by the 
heavens for Sir WiUiam Herschel was that of finding each 
component further diviaibk. The discovery, though beautiful 
and interesting, was easy ; all the four stars can be seen with 
a good 3-inch telescope. The "preceding" pair, or that 
which crosses the meridian first, is distinguished a^ e^ the 
" following " pair as f^ Lyr^ ; and Flamsteed attached the 
numbers 4 and 5 to them respectively. The former consists 
of a fifth and a sixth magnitude star 3" asunder; the con- 
stituent stars of €^ are nearly equal (5*3 and 5*5 magnitudes), 
and are set a little closer together (at ^"■45). Their revolu- 

* FlunnmrioGL, Caialogat, f, 75. 



THE SYSTEM OF THE STAItS 




tiona, too, appear to be performed about twice as ijuicklf tf 
those of the neighbouring couple. From the shifticig of th( 
relative aituatioua since 1779 by more thaa half a rigi 
aiiglo, their period maj he estimated at about 800 
while that of c^ Lyrse is likely to exceed on© thousand, 
practicabilitj of computing either orbit is still remote. 

The small common proper motion (9" a century) of 
t^ght coupler affords poaitive evidence of their unioa Into 
va^b system. At their unmf^osured, perhaps immeasanUs 
dxatance, the gap between them of 3J' may well stand ia 
a ch^m costing light itaelf Home months to bridge ; yet the 
atreas of their mutual gravity reaches across it, compelluiii; 
their circulation in orbits bo spacious that a sirtgle round d 
them must occupy an era of no insignificant duration^ even in 
the life of a star. The four star* of e Lyrje give a spectrum of 
the firat type, combined, in the leading couple, with a decided 
cast of yellow. But thig is often the case with double stars. 

" A miniature of e Lyrse " ^ is offered to our regards in 
tf SeorpiL This is perhaps the most beautiful quadrapls 
group in the heavens, from the narrow limits within which 
the brilliant objects composing it are crowded. As a widi 
doable it was noticed by C. Mayer in 1776 ; after seven^ 
years the smaller star was divided by Mitchel at Cincinnati, 
and the larger one of fourth magnitude yielded similarly, iii 
1874, to the insistence of Burnham. Both pairs share with 
several neighbouring stars a slow drift through space.^ They 
are 41" opart, and have as yet developed no systemic 
movements.' 

The sixth -magnitude star 86 Virginia may be said to 
consist of a double primary with a double satellite at 27". 
Full acquaintance with the group was made through Burnham'a 
amdyais of one of Struve's " rejected " pairs. Its internal 
relations will need time to develop/ A quaternary combina- 
tion of p^uliar interest was detected by Mr. Innes in 1897.* 
It consists of two close pairs, k Toucaui and Lacaille 353, 
separated by the wide interval of 5' 20*^; yet, notwithstanding 



^ FUmmftrioH, Caealoffve, p, 96. 

* Iqdsa, Re/erenc£ Catahgue, ■p. 157 \. 

* Bamhfttn, ffer*. Cat. p. 149, * Ibid. p. 128. 

* MimihJy Notica, toL LriL p. 456 ; Re/. Cat. p. 10 a. 




MULTIPLE STARS 



203 



th« antecedent improbability of their connection, it is emphatic- 
ally asserted by the unanimity of their rapid rate of travel 
across the sphere. 

Eighteen "double-double" star-groups — one (% 2435) 
with a span of no more than 15" — wete enumerated by Bum- 
bum in 1882/ and three have, since been added by Hough 
and Innea.* They perhaps exist more numerously than we 
have as yet any idea o£ 

A " double-treble " star, so-called by Herschel, has been 
the subject of numerous succeaaive discoveries. With the 
slightest optical assistance tr Orionis, a star of 3*7 iniignitude, 
just beneath the middle star in the belt of Orion, separates 
into two wide and unequal components, each of which was, 
October 7, 1779, perceived by Herschel to be triple.' As 
usual in such cases the process of resolution was continued, 
and the assemblage was described by Barlow as " double- 
quadruple, with two very tine starB between the sets/** 
These last, however, are not unlikely to be mere optical 
associates. To this intricate group Burnham added a further 
element of complexity. At Lick, in the autumn of 1888, he 
found its chief member to be formed of a fourth and a sixth 
magnitude star, a quarter of a second apart, and yielding, after 
ten years, signs of mutual circulation.^ The disclosure, like 
some others, raised a question as to the point where stellar 
subdivision can really be said to cease, That it is not where 
visual limitations interfere with our recognition of it was 
emphatically reasserted in 1904 by Professors Frost and 
Adams's spectroscopic discovery that the primary of Burnham's 
pair is intimately, though invisibly coupled with an obscure, 
tuafisive body. 

The essentia! character of tr Orionis is that of being made up 
of two diatinct, yet evidently connected knots of stars^ and the 
same knot (S 762) *" contains all the four brightest componeata 
These ditfer, and perhaps vary in colour, and their influence 
may be assumed to predominate in this remarkable system. 

* Obaenxitofy, vol. iv. p. 176. 

' Astr, Nach. No, 377a ; AF^tUhly Noiicts, vol. Itu, p, *W. 

■ PAH. frniM. Tol. liiii. p, 124. 

• Smjtti, Ci/cte o/Otl Ob;eci9,vd. 1881^ p. IBS. 

* AHr. Naeh. No. 2875 j Oeti. Gat. p. 68. 

* Strar6> Mens. Mier&ai. pp. 149^ 24fi. 



THE SYSTEM OF THE STABS 



The multiple star 45 Ijeporis ia organised on apbnla 
definite tbao thftt governing tb« stin>cture of tr Ononis. I 
visible to the naked eye> it ooosista of four principal and 
puboTvlinate members, ooccessirelj diaeovetBd t^ Sir 
Herachel aud Burnham.^ One of the st^iBy of e^tb 
ttide, stands ou t through its ruddy colour firom its 
coinpaiiioii&' The entire group of nine objects covob 
extent of 125". 

The nebular relations of double and multiple 9tais 
noticed with surprifie by Sir John Herachel at the outset of 
career.' Although admitting without hesitation their ph 
character, he was without the means of establishing it 
made available^ aud coold enpport his conviction only b^ 
utter improbability of such collocations as he pointed out 
fortmtou£i Thus, a close, minute stellar couple is planted 
the exact csentre of a faint round nebula in Leo (New Gen. 
Cat, 3230); and the same kind of coincidence recurs twice 
in the southern constellation Dorado (K.G.C. 1732, 1961 
Two pairs in Sagittarius, each set in the midat of a nel 
(N.G.C. 6589, 6590) may from their contiguity be 
pected to constitute one system; and two ninth- tnagnitude 
Btara at 15", marking very nearly the foci of an elliptical 
nebula in the same region (N.G.Q 6595), are certainly not 
accidentally projected upon it. " One of the most curioui 
objects in the heavens " (according to Sir John Kerschel),* is ft 
trio of stars arranged in a minute equilateral triangle, relieved 
upon a shield of milky light (K.G.C. 1931) j and its 
siogularity was enhanced by the duplication, under Biimh&m's 
gaze in 1891, of one member of the combination.^ By the 
same obaorver, again^ an eighth -magnitude star riglib at the 
heart of a round nebula in Monoceros (N.G.C. 2182) was 
divided, also in 1391, into a delicate pair,'^ the remeasurement 
of which after twelve years, by Professor Aitken, gave evidence 
of slow circulation. Of circulatioDj we remember with surprisa, 
conducted in a nebulous medium, and therefore presumabl 



t 



' Burnbam, Mtmnri H. Attr. Sec. toI. xUt. p. 238 ; Attr. Nach. No, 204S ; 
i^MTvatory, voL iv- p, 177 ; Otn, Cat. p. 68, 
^ O. Kfiottr OhatTWitiir^f^ T&L iv. pp. 1S4, 212. 
* Mijrufirt B. Afir. Soe, vol. vL p. 7&. * /Net toL uL pi &4, 

' ffm, OtU. p. 66. • Ibid. p. 7&. 




MULTIPLE STARS 



205 



ipeded. But if eo, the syatena coiJd not be a pennanent 

and a temporary star should result from its collapse. 

difficulty is intensified by the consideration of many 

slUl more noteworthy instances of the asaociatiDQ of 

ipoedte stars with nebulae. The whole framework of the 

\t nebulous structure in the sword of Orion seems to rest 

the stellar group designated 0, or rather 0^ Oriouia ; for 

is a second & not far off. itself a wide double star ; and 

two together form, to the eye, one diffuse oKject, singly 

logued by Ptolemyj Tycho Bralie, and Hevelius. But it ia 

itb 0^ exclusively that we are at present concerned. 

On the very slightest teleacopic persuasion, it allows itself 

be eeea as quadruple. The four atara into which it divides 

severaUy of fifth, sixth, seventh, and eighth magnitudes, 

greatest interval between any two of them not exceeding 

None of them is in visible subordination to any other ; 

By stand, it might be said, on an equal footing, at the four 

lers of a rudely quadrilateral figure, or " trapezium.*' They 

itain their places, too, botli absolute and relative, with 

rigidity. After two and a half centuries of observa- 

1, no shifting of them can be detected. They are hence 

bIj to be at a prodigious distance from the earth. 

The rule that such groups seem more crowded as they are 
tt«r seen, has not b«en infringed here. A fifth star of the 
renth magnitude was added to the company by Struvcj 
|ovember 11, 1826, and a sixth, still fainter, by Sir John 
erschel, February 13, 1830. Both of these, though closely 
Ijaoent, each to one of the larger stars, share their apparent 
immobility.^ Variability in light has often been ascribed, and 
as often denied to them. Burnham'a experience is against it; 
yet the curious fact that Itobert Hooke saw the fifth star in 
16G4 with a non-achromatic three and a half iuch tele^ope,' 
is strongly indicative of temporary brightening; and M, 
^omas SolfL was convinced that the sixth star was shining 
itU unusual lustre when he observed it, after an interval of 
months, on Kovembcr 10. 1901.^ Individual and 



' Burubam's meiLBurea seem decUive oa tUis [K>iiit. Sea Iftrmoirs R. Astr 
Soc, voU. xlir. p[x 203, 2S7 ; xIviL p. ^14 ; ifonXhiy H'otiut, vol. xlix. p. 2U7 
I*ubt. Lick Obttrwttory, vol. ». p. 46- 

' Jiieroyrapfita, p. ^2. ■ AstT. AbtfA. No. a7SL 



206 



THE SYSTEM OF THE STARS 



atmoapherk conditioua are, however, largely conoemed in such 
perauasiotis ; when they conspire favourably, it is weU asoor- 
tained that all six of the trapedum-etars con be made out 
with achiomatics or reflectois three to four inches id apeitme.^ 
Hooke^B observation can scarcely theu be said to demonstnte 
change. 

Further members of this group have, at varioiis times, 
been half-seen, balf-Burmisod j but their existence, alwavs 
problematical, has been disproved through the application of 
the Lick thirty-eix inch^ for the three new stars perceived 
feom Mount Hamilton by Alvan G-. Clark and Barnard could 
certainly not have been detected vath any less powerful 
instrument. Two of them lie within the trapezium ; th« 
third, a double star of extraordinaiy minuteness and difficultr, 



Bunard 






• F 



X 

AlnoCUrk 
a . 



* B 

Pio. ST.— fitan of Ibe TnpfKiaiii. 

just outside it* Their positions are shown on the accom- 
panyiug diagram (Fig, 27) where G and H represent two of 
the recently discovered stara, the third, as an imperfectly 
determined object, being provisionally marked with a cross. 

The fact that the leading star in the trapezium has 
proved to be a spectroscopic binary is not the lees astonishing 
becauae of its congruity with much that had previously been 
learned. Sir William and Lady Huggina recorded in 1897' 
the ^XAt eigiLB of its dupUcity ; they were assured and defined 
by the Yerkes measures in 1904.* Ita nature is moreover 
shared, and attested through a far larger variation of radial 

^ Webb^ GaI, Olgtcts. p. Sfl7. 

* Burnham. MoTtthty yoiices, vol. ilix. p, 352 ; ji3fr. Nadu No, 2930. 

' Atiroph. Joum. vol. vi p. 322. 

* Frost uid Adima, Udd, toI. xtx. p. 153. 



MTJLTTPLE STAES 



207 



ed, by ^ OriomB^ which crof^ses the meridian six seconde 
itter than the septuple object bo the westv Half a degree to 
south lies t Ononis, a triple star immersed in an ouLlyitig 
ikred of the Sword-handle formation. Here again visual has 
an followed up by spectroscopic resolution. The diffuse 
rk lines in the prismatic ligbt of t Ononis oscillate to an 
ttent indicating to and fro velocities of 60 kilometres (37^ 
liles) a second. And by siraikr tokens^ the nebulous stars S 
[onocerotis, 5V Cygni, and <r Scorpii are known to be swiftly 
Bulating couples. The investigation of Hystems so singularly 
cumstanced will doubtless serve to elucidate the relations of 
je glimmeriog fields of apace to maasea of matter traversing 
lem. Should they appe-ar to slip through unopposed, our 
ieas ae to the essential properties of material substance will 
ive to be considerably modified. 

In one other great nebula besides the Orion structure a 
'multiple star seems dominant. The nuclear group in the 
thM nebula (N.G.C. 6514) consists of a close quartette 
covering an angular extent of only 19', with two extremely 
faint additional stars discovered by Professors I^ngley and 
Holden. Complete apparent fixity characterises the arrange- 
ment 

The frequent association of compound stars with nebulje 
is no mere isolated fact. For they pass by insensible degrees 
into star-chistera, the nebulous affinities of which have been, 
ja many case^, established with the aid of photography. The 
conjecture is even plausible that the formation of a multiple 
stajr in a great nebula represents the initial stage of the 
development from it of a crowded cluatre, minor nebulae giving 
rifle to leaser groups ; and if objects of the kind have not yet, 
BO to Speak, been turned out of the workshop, it is no wonder 
that fragments of their raw material still cling round them. 
Compositeness of structure may thus measure primitiveness of 
condition^ illustrating, though to us dimly, the sequence of 
Divinely decreed chatigos by which cosmiejil order is gradually 
more and more fully disengaging itself from the " loud mis- 
of chaos. 




&TAB8 joined tc^tber ia sjntems peremptoriljr aaaerC m conuaon 
origia. Tbeir companioiiBbip U not due to ch&nceL Bodies 
moving independently in space may circuit one anoth^ !□ 
a hj^rboUe orbit ; but the event ahould be noiqnjei They 
cun never meet again. Permanent capture is pnurtioUly out 
of the questioiL The mterveatioa of a third body would be 
Inquired to bring it about, and the tbiid body should be 
critically situated, and of enormous toslsr to produce th€ 
neeowaTy amount of retardation. The poflaibility of a ctmaaH 
aplffoftcb having led to an indissoluble unioti is indeed too 
remote to be worth counting. Multiple etais, we naay rest 
asBured, were such potentially from the first. 

Yet the manner of their origin remained long ao im- 
penetrable myBterj% Clearly, it belonged to the regular otder 
of fiidoreal arrangements ; and quite aa clearly^ it difiEered 
toto cotlo from the aeries of operationa by which the planetary 
Hystem had come into existence. Nor could any reasoa for 
the divergenoc be BBsigDed until Dr. See published, in 1893] 
hia researchea on the part played by tidal friction in moulding 
the relations of double stars. Baaed on Professor Darwin's 
memorable inquiries into the antique history of the earth and 
moon, they followed a line essentially original. No attempt 
had previously been made to trace the consequences in Blellar 
systems of a mode of action known to have been powerfully 
effective within the narrow procincts of the lunar sphere. 
Yet, since it gains efdcacy as the ratio between the masses 
submitted to it approximates to equality, it should reach a 
maximuiD of influence in modifying the relations of co- 

209 



THE EYOLTTTION OF MULTIPLE STABS 20t 



ite bodies such as double stars. Tidal friction is, in 
See*e words, " a neccBsary adjunct oi" gravitation wherever 
oystems of Ouid bodies exist Id a state of relative motioD ; 
it ie a physical agency aa universal as gravitation itself, 
operating more or less powerfully in all the syBtema of the 
universe." ^ 

The key to the enigma of double-star development was 
given by the high eccentricities of their orbita. Two bodies 
levolving very close together are not only pushed asunder by 
tidal reaction, but are forced to retreat along trocka that 
become elongated aa they widen. The process is terminable 
after an indefinite lapse of time ; and is even theoretically 
reveTBible, though to an almost evanescent degree. What we 
jttflt now have to do with, however, are the dii'ect work- 
imge of a cause, aatiefactorily shown to be adequate to the 
effects assigned to it. If this be so, telescopic star-pairs set 
forth on their careers as spectroscopic binaries ; while spectro- 
scopic binaries must, through the mfluence of tidal friction in 
widening their paths, be steadily growing into visual couples. 
One class ia complementary to the other, and Dr, See's 
hypothesis obtains fresh confirmation from each additional 
discovery of a star with variable radial motioru 

Moreover, these excessively close systema are strongly 
marked by signs of progressive change. Some appear to be 
still inchoate. The eventual stars are intimated by certain 
phenomena of their light-changes to be perhaps as yet 
undivided. Connected, it ia thought^ by a surviving ligament, 
they revolve as a dumb-bell might, pivoted on its neck, and 
aptly iUustrate the " apioidal " forma dealt with in the 
formuhe of Poincar'i find Darwin. From the rupture of a 
spinning " dumb-bell/' then, to stately binaries in secular 
revolution, a virtually unbroken series of instances can be 
traced ; and that the advance is really a development, is a 
fairly irresistible conclusion. A set of specimens presenting 
gradual modifications of a given type proclaims of itself a 
transforming agency ; and double stars exhibit, not only all 
the linked instances that could be expected, but the requisite 
agency for producing them ready at hand in the grinding and 
modelling power of ttdal friction. 

* ^Jtn afui A$lTOpfij/ties, rgl. xiL p. 290. 

14 



210 



THE SYSTEM OF THE STARS 



I 



The theory of double-star evolution through its exoep- 
bional otficuuy in embryonic stellar sy^temB is thus, at &a 
eight, strongly reooumiendud. It must be true to some 
extent Bub it has limitations and quali^cations which 
should not l)e overlooked. The dynamical and pbyaol 
histories of double stars are iiitimateiy counect^ Theit 
relative masaes and densities, their epectr&l congruities or 
divergencies, their colour and brigbtne^, are 80 many itemi 
of evidence as to the course of their companionship, and ihi 
destiny awaiting them. Comparative mt^uiriea are indeed 
especially difficult for objects in many cases barely eepoiable 
in the aky, yet they have been set on foot ; and the few ficti 
BO far collGcted are valuable, both in themselves, and sa a& 
earnest of a fuller harvest. Meantime, they seem to warmnt 
one or two provisional generalisations. The first is, that 
contrasted pairs of yellow and blue tints are formed each of a 
solar and Sirian star, the lesser component showing the morf 
primitive spectrum. Since, however, their revolutions are (^ 
inordinate elowneeSj their mass- relations cannot be apportioned 
It is only permissible to say that if satellite-statB resemble in 
constitution solitary stars of similar light - quality, theii 
brilliancy must exceed the proportion of their mass. Their 
gravitative disparity should, in this case, largely outmeasure 
their visible inequality. 

An opposite rule seems to apply where the secondary star 
is of a roseate or purplish hue. It is founded, indeed, upon 
only two ascertained instances ; yet they do not appear to be 
eioeptionaL The companion of 70 Ophincbi, while of less 
than one-fourth the brightness of its primary, quadruples its 
mass. Admitting it to be of the same density, it gives eleven 
times less light per square mile of photospheric aurfaca And 
for the analogous pair, i} Cassiopeiae, the ratio is ten to one. 
The lesser luminary ehinea with one-tenth the areal brilliancy 
of the primary, equal densities being again assumed. Un- 
fortunately, the rays of these heavy satellites have not been 
separately analysed ; so that, in the few case^ where spectial 
distinctions are on record, nothing is knon-n aa to relative 
maira; and where relative masses have been determined, 
differences in light -quality are undiscriminated. An ex- 
ception is furnished by a Centauri, the components of which 




THE EVOLUTION OF MULTIPLE STAES 211 



ftre thoroughly individualised. But they ehow no hetero- 
geneity of colour or apectrum. One is a deeper yellow than 
the other, simply because its ab&orption is stronger than, 
though fiimiJai: to that of its comrade. Binary systeme 
undoubtedly offer the most promising field for investigating 
the stagea of stellar development. Their relative antiquity, 
to begin with, can be roughly estimated ; for there is good 
leasou to suppose the closest and swiftest pairs to be in 
general the most primitive. Further, the compouents of 
each are necessarily of the same age. Whatever spectral 
differences they present cannot then be set down to the 
exclusive account of tima They are either aboriginal, or 
have supervened as the residt of innate diversities. Thus, 
bodies unequal in mass are unlikely to develop at the same 
rate. The long'accepted opinion was that it should be slow 
in proportion to the quantity of matter contained in each 
globe. Sir William and Lady Huggins, however^ suggested 
in 189Y doubts on this point^ They indicated the probability 
that a high gravitational constant might hasten the transition 
from a Sirian to a solar spectrum ; and the phenomena of 
double etara seem expreaaly adapted to serve as a criterion 
I whether this is so or not. A deciaive condition could be drawn 
I from a comparison of the masses of such a chramatlc pair as 
^^3<50tis ; but the possibility of instituting it ie in the dim future. 
^H Something, on the other hand^ may be learned from the great 
spectroscopic binary Capella concerning the influence of massive- 
nesa upon spectral history. If the plane of its revolutions passes 
through the earth, the attractive power residing in the system 
is more than twice that of the sun ; and this minimum esti- 
mate should be increased eight-fold to correspond with an 
orbital inclination of 60" to the line of sight. Now the star 
is enormously brilliant. It has a measured par»Iliix, and at 
its diatance, our sun would appear 102 times fainter. Assuming 
the components of the binary to be equal globes^ of solar 
density aud intrinsic lustre, each should posseas the mass of 
363 suns. Their mutual attraction, in other words, should 
exceed 726 times the gravitational pull exerted on the planets. 
Keverthfeless, the highest value that can plausibly be assigned 
their joint mass ia twenty times that of the sun. The 

^ Astropk^ Jovm. toL tI. p. 320. 



212 




THE SYSTEM OF THE STAKS 



tMBModous diflcrepancy thus mftde apparent obliges u i» 
BOppooe that Capella, while iipectroeoopioaUj alinost « t«plin 
of the Bvn, is an orb far leas oondenaed and more lanmioaa. 
It then eridentljr became inveeted at a much earber st«ge 
of oooling, with a reversing lajrer Bimilarly oompoaed hi tfait 
producing the Fraonhofer lines. Further experience most 
decide whether giant sons of thia type are inyariablr mon 
C«]iuoQJs than their light would le&d us to expect, or wbeUk9 
Capella is in this respect peculiar. Its example seircs «t 
leaflt to give an idea of the modes of evolutioDAiy iuqoiij 
rendered feasible by the study of bdnaiy syBtexna. 

Inequality between mass and light is pushed to the iM 
extreme in the obecore, though etrangly attractive attendaztha 
of many lustrously white stars. A large numbex of sucb 
ineongriious couples have been discovered spectaxtgraphicaUr, 
and the enigma they present; finds no ready solution. Clearly, 
diApaiate loss of energy by radiation will not reasonably ei- 
plain the contrast of their present state. Rather, some 
profound diversity of constitution mnst be supposed to have 
brought it about The brilliancy of stellar photosphere 
essentially depends upon the activity of interior circulator? 
process^. These are perhaps, in abnormally dark globes, pre- 
maturely retarded or arrested. But the bow ? and why ? evade 
divinatory efforts* 

Ftom dwelling on the origin of binary stars, our thoughts 
insensibly range beyoud them to larger combiDations. Be- 
volving couples very often form only part of an extended 
system- Is it conceivable that such varied aggregations were 
fashioned throughout by the same kind of influence ? Waa 
tidal friction the factotum in all their developmental changes I 
To affirm it would be periloua There is no warrant for 
ascribing an iron consistency to creative methods. In the 
planetary system, at any rate, they seem to have been cod- 
siderably varied. Professor Darwin has virtually demonstrated 
that lunar-terrestrial relations lay apart. No other member 
of the solar family could have originated by fission, as the 
moon presimiably did, or raised a tidal wave of overruling 
magnitude ou the etill plastic surface of its primary. Hence, 
uniformity in the processes of cosmogony need not be taken 
for granted "We may^ for example, reasonably admit that the 



THE EVOLUTION OF MULTIPLE STAKS 213 



close pair in 2^ Vt^x Majorifi divided at clo6€ qimrters Bimilarly 
to the earth and moon, without extending the inference to the 
telescopic compaDion of the same star ; still leaa to Alcor» its 
distant feUow- passenger through space. Each case should be 
ooDsidered on its merits, without prepossession in favour of 
hannoniaing the results. Indeed^ the long series of operations 
traced out hj Dr. See is scarcely capable of being duplicated. 
Throagh over-rapid apiumug, a fluid globe may split asunder 
ooce ; but a repetition of the event is virtually precluded by 
the very consequences of the first disruption. For the new- 
born satellite exerts, by the drag of the tidal wave it raises 
on its primary, a powerful retardative influence on ita rotation ; 
the speed of which inmost unlikely to reach a second time the 
pitch needed for instability ; and the satellite itself is uuder 
the same prohibition. Multiple stars, then, cannot^ so far aa 
we are able to judge, have been formed by succesflive sub- 
diviaiouB of one parent mass. In stellar systems, as in the 
solar system, many degrees of relationahip are diatinguiahable. 
Groups in loose mutual connection may have originated con- 
temporaneously in different sections of the same nebula. 
Binaries of co-ordinate rank are doubtless frequently included 
in a single complex mechanism. This species of remote kin- 
Bhip is forcibly suggested by the movements of f Scorpii, one 
of the few triple stars, all the members of which are in visible 
revolution. Tet not in the same sense. The close pair 
circulates directly, the third, more distant companion (as 
already stated), in a retrograde direction. The wide 
divergence thus probably indicated of the two orbital planes 
betokens unmistakably a remarkable dissimilarity in the 
eonditions under which the near and the remote satellites 
respectively took their origin. A more intimate acquaintance 
with such systems will perhaps show the case of f Scorpii to 
be typical j it is assuredly most significant. Again, the 
six stars forming the Orion trapezium are, beyond cavil, a 
physical group, in immediate genetic connection ; yet it would 
be extravagant to suppose one the parent of the others. Only 
the dark body circulating contiguously to the brightest of the 
aextett can be regarded as ita proper ofispring ; the rest claim 
an independent footing. Ws can conceive them as having 
condensed from distinct knots in one vast nebulous structure 



214 



THE SYSTEM OF THE STA£S 



imfmmed with & dowly wheeling iiiimiMnit, We euiaol 
flotto dy e thflB as thiowmaff one After Mindier hy oae evifi^ 
rotatiBg ^^eh&. The sne Biglit be sid of ibe qnednqfc 
fljitem ff Ljia» and of meaj mat. 

The BOfe •tfeeDtirely coaoie prooesies m etodied, tte 
Bkon Tmrioee tiwj e^ypeftr. 7hii§» oometo vma be cleeziy ra 
to bare ongtnftted dilfeiciitlj from planete^ aad sonte pUnHs 
diflezaiUf from others; while our o>wn aaleUte Btrndt Mk 
e line for itsell The large ooafWDanceB of the volar sjnlai 
snlwist amid divenctj of detail, attesting etroog indiTidiialitifli 
of history and statua. A^d vocb diTersitj' has its foUec* SDvpi 
among the etare. It la much if we ean eatch glimpees of I 
partial truth in meditating on their evolutioQal order; the 
profundities of its meaning, and the intiicauaeB dt ila roois i> 
tbe pa£t, and lamificaticins into the futme baffle our scniti- 
niaing efforts. 



CHAPTER XVU 



THE PLEIADES 



Tbom multiple fitats the transition is ewsj to star -clusters. 
These eeem to embody completely the idea contained in germ 
in the former class of objects. They are collections, often on 
the grandest scale, of sunlike bodies small and large, united 
in origin and history, acted upon by identical forces, tending 
towards closely related ends. The manner and meaBure of 
their aggregation^ however, vary widely, and with them the 
cogency of the evidence as to their organic oneness. There 
are innumerable cases in which it absolutely excludes doubt ; 
there are some in which it is rather persuasive than con- 
vincing. It i& not then always easy to distinguish between 
a casual " sprinkle *' of etaifi and a genuine cluster, Kor can 
the movement-test, by which so many physical have been 
discriminated from optical double - stars, be here applied. 
Internal difiplacements of a circulatory character have not 
yet become apparent in any cluster, and there Ib only one 
with an ascertained common pioper motion, 

Thia ia the inunemdrial group of the Pleiades, famoue in 
legend, and instructive, above all others, to exact inquirers — 
the meeting* plaee in the skies of mythology and science. 
The vivid and picturesque aspect of these stars riveted, from 
the earliest ages, the attention of mankind ; a peculiar 
sacredness attached to them, and their concern with human 
destinies was believed to be intimate and direct Out of 
the dim reveries about them of untutored races, issued their 
association with the seven beneficent sky-spirits of the Vedaa 
and the Zendavesta/ and the location among them of the 

' BuDaen, Die PUiade% und dtr TkUrkreiM **, 4S4. 



216 



THE SYSTEM OF THE STABS 



centre of the universe and the abode of the Ddty, o( 
the tradition ia etill preserved b^ the Berbers imd 
With November, the " Plfidad- month," many pmnitive 
began their jear ; ' on the day of the midaigbt-cuL 
of the Pleiadea, 17th Xovember, no petition was 
in rain to the ancient kinga of Persia ; ' and the samt.' ^t^: 
gave the signal at Bosirifi for the commencement of the im\ 
of Isis, and regulated leae immediately the celebration m\ 
nected with the fifty-tvro year cycle of the Mexicans. Samp] 
AuBtralian tribes to this day dauf^ Id boaour of the " Seval 
Stars," because " they are very good to the black felloin.* [ 
The Abiponea of Paraguay regard them with pride as tlicir | 
aneeators,* Elsewhere^ the origin of fire and the knowledge of 1 
rico-culture are tiaoed to them. They are the " hoeing-stan" 
of South Africa,* take the place of a farming-calendar to tha 
Solomon Islanders, and their last visible rising after sunael 
ia, or has been celebrated with rejoicings all over the southen 
hemisphere, as betokening the " wakiug-up time" to agricultural 
activity. 

To the Greeks of Hesiod'a age their " heliacal rising " (the 
first visible before sunrise) announced, each May, the opening 
of the fieasou for navigation ; and their name thus came lo 
be interpreted (from plein, to sail), the " sailing-staTS." But 
this etymology was doubtless — like the derivation of "elf" 
and " goblin " from Chalf and GkibelUne — an afterthought ; 
and it may be confidently maintained that the word '" Pleiades," 
bearing like its Arabic and Hebrew equivalents, the essential 
signification of a '*cluBter/' came from the Greek pleiones, 
many, or pleids, full,* It was represented in Latin by 
"Vergiliai" (from ver, gpring), a designation possibly com- 
memorative of the ancient coincidence of the atats with the 
vernal equinox. They were, moreover, ehoseu about the same 
epoch — say 2700 B.C. — by the Hindus to mark the first lunar 

' Hs-Hburtosi J^efure, roL xxt, pp. 100, 317 ; V»ii SAndibk, L'A^nnomit, 

t iv. p. 3S7. 

* HAliburtoOj Ftativai of the Dtad, p. 46. 
' Ihid. p. 13. 

* Lubbock, Ori^n 0/ CivHisation^ p. 316, 4tli ed, 

' .1. HamnioDd Tooke, in «D ioteresting p&per read in Jutuaiy 1SB9 before 
tbs S. African PliUosophical Society. 

* iffliwr*, Yol. III?, p. 608. 



THE pleiahes 



217 



aion, called " Krittika," general of the celestial armies ; ^ 
long occupied the same post in Ohaldea under the title 
inrayya," the crowd.* 

The similarity of the traditions respecting the swarm of 
tial " fireflies " 

Quie septem did Bex tamea eese soIent, 

1^8 suTpriaing as their universality. That they " were seven 
now are six," ia asserted by almost all the nations of the 
If from Japan to Nigritia, and variants of the classical 
of the ** lost Pleiad *' are still repeated by sahle legend* 
Dugers in Vi<itom and Western Australia, by "head-hunters" 
L Borneo, by fetish-worshippers amid the mangrove-swamps of 
Gold Coast An impression thus widely diffused must 
Ither have spread from a common source or originated in an 
?ioUB fact, and it is at least possible that the veiled face of 
seventh Atlantid nmy t^ify a real loss of light in a pre- 
atorically conspicuous star. Some members of the collection 
at preaent, there is little doubt, slightly or slowly variable," 
id progressive tendencies of the kind are in more than one 
suggested to he present. Thus Alcyone, the chief of the 
[liUeotioD, now of the third magnitude, and just twice aa bright 
the brightest of its companions, was either not oue of the 
JUT Pleiades observed by Pfcolemy, or was then much fainter 
it has been from Tycho Brahe'a time to our own. So at 
least Francis Baily concluded from a careful examination of 
the records/ and he knew better than most men how large 
an allowance has to be made for ancient inaccuracy. Al 
Sflfij boo, the competent reviser of Ptolemy's observations, 
expressly states that the Alexandrian quartette appeared to 
him, in the tenth century, the most lustrous among the 
Pleiades,* Yet none of them can be identi6ed with the 
pTQBent imid^i. A literal explanation of the old legend may 

' R. H. AUen, Star I^ames arid IfitiT Meaniiigi, p. 39S. 
■ Wab«r, iTuiiKhe Shtdien, Bd. x. ii. 21£. 

* C. Wolf, Amtaleg dt I'Obaermtoirre fie Farit, t. xU. ii. p. 28 ; Lindemnnn. 
M^moira dt fJcad., 9t. P^tw^bourg, t. XKcii, vii. S«r. No. S, p. 29 ; Vogel, 
Pottdam Jte^ort., ISPO. 

* Jitia&irs H, Aatr. Soe* vol. ilii. p. 0. 

* Sofajellerap, Ik$eriptiofi dta ttaiUt, p. 132; Flaminarioii, Lta SU/iUs, 
3M. 



21S 



THB SYSTBlf OF THE STABS 



heooe be feasible, and ProfwBor Pickering's voggestod iderit- 
ficfttion of PleioDe with tfae waanag A tlanrtd htm mnA ti 
reoommend it* Tba dmpbj bf this star of a guet 
Bpectram resembling that of P Cfgni ooimifeiiaiioes the tvw 
that, like P Cygni^ it formerlf abatie with teznpoiaij a 
intermittent briUiancj. It is now of 5'4 magnitude, or jati 
twice aa bright as it wa^ hj Azgeknder's estiutate, fifty 
/ears aga 

The fire stars ordinanlj Tttible bftrirtm Alcji^ie (flee 
Plate L Frontispiece) are Sectn and Atlas^ each fluctiiating 
slightlj abore and below 3*8 magnitude; Maia« now of the 
fourth, or one magiutnde fainter than Akyone ; Merope and 
Taygeta, the inferioT? of Maia b^ respectively a quarter and 
a half magnitude. Celaeuo, the seTeath or concealed Btar, 
gLveB only about one^third the light of Tajgeta, 

Yet it ean he eeen with many otheTB> under faTOur* 
able circumfitances. Maestlio, the tutor of Kepler, perceiTcd 
fourteen, and toapped eleven Pleiadea previousl)^ to the inven^ 
ticm of the telescope ; Carringtoii and Denning ootmted 
fofirteen,' Misa Airy marked the places of twelve with the 
naked eye.' The faintest of these fell but little short of the 
sixth, and there are tweatj-tbree Pleiades down to the seventh 
m^nitude,^ each of which (with perhaps one or two exceptioiis] 
might be separately visible in a transparent sky or &om an 
elevated station. But their crowded condition makes thia im- 
possible, and givee rise rather to the efifect described by 
Ka^wini in the thirteenth century, of " six bright stara with 
a number of dusky ones between.'' ^ 

With the use and increase of telescopic powers, the popu- 
louflneaa of the cluster has been amazingly increased. An 
object-glass scarcely exceeding two inches diameter showed 
Robert Hooke in 1664 seventy-eight Pleiades,* and Miehell*8 
conjecture, in 17&7, that there might be more than a thousand 
of them,' has been superabundantly verified by the results of 
modem labours. Over an area about Alcyone me&surii^ 




Webb. Ctl (Hgteif, p. SM. 



* Attr. JVacA. No. 2934. 
" Monlhly Noti^tt ToL xxiii. p. 17B. 

* Harvard Annala, rol, xir^ pt, iu p. S98 ; cf. HuUer Hud Kemrf, Aatr, 
Naeh. Xo. 9Afl7. " Ideler. Sltmnamoi, p. 147. 

■ Miero(fraphia, p, 241, t PhU. Tran*. ToL IviL p, 9&S 




THE PLEIADES 



219 



15' X 90' M, Wolf catalogued, at the Patis Obeervatory in 
176, 625 Btara to the fourteenth magnitude; on the MM. 
^s aensitive ptatea, in 1885^ 1421 made their appearance 
amaller space, and the number was brought up to 23 2d 
"by exposures of four hours in November and December 1887. 
The faintest objects thus registered were probably of about the 
sixteenth photometric magoitude. 

How many of them really belong to the group, and how 
ay are referred to it by perspective, can be determined with 
!ie help of time and patience. As regards some of the better 
lown stars, the process of discrimination haa already begun. 

Beseera meaaurementa of the places, relative to Alcyone, 
of 52 Pleiades,^ executed with the Konigaberg heliometer 
during the twelve years from 1829 to 1841, furnished a 
starting-point for inveetigatioDS of their internal movements. 
The upshot of the first effective comparisons was to exhibit 
these as null. From a coUodion-print of the cluster taken by 
Eutherfurd of New York in 1865, Dr. Gould redetermined 
nearly nil Bessel'a stars with such accuracy as to make it 
certain that no appreciable interstitial shiftinga had occurred 
in the course of a quarter of a century ; * and his conclumon 
was, through additional photographic comparisons of ten of 
the same stars, extended by Professor Jacoby to the year 
1900,^ 

Now this seeming rigidity implied a great deal. For the 
point of origin of the measures in question is not immovably 
fixed in the sky. The chief Atlantid has a secular proper 
motion (according to Newcomb) of 6", the possession of which 
in common by the whole stellar band virtually demonstrated 
their effectual union. Where one among many objects is 
ascertained to be moving, relative fixity can only mean that 
all drift together^ and so the unique phenomenon was brought 
to light of the transport in block across the sphere of some 
scores or hundreds of congregated suns. Even if the whole of 
this apparent displacement should prove to be as it were 
rejected from the sokr advance, its significance of physical 
kinship among the objects affected by it would be nowise 
impaired. For an identical parallactic shift would equally 

' AUt. J^ach. Ko. 430. ' Observatory, vol. il p» Ifl. 

■ AttfopK Joum. ToL xitt. p. 66. 



220 




THE SYSTEM OF THE STABS 



' Bsffioe to locate tbem in Ibe sune region of sp«oe note 
imtnediate inGueooe of their confitrainicg mutaal grsTitj. 

The BRlAbliahment of a genenl uQaaimitj of moveraeU 
among the Pleiades waa the first Bt«p towards inveetigatii^ 
their retationa; the next waa to seek evidence of ii j i il iiiM ri r 
change. This haa still to be found ; its highlj reooiidifie 
nature has been rendered tinmistakable hj the laboois of 
Wolf/ Pritchard* and ElJdn.' Bisplaoemanta within the 
duster, though necessarilj in progresa, are bar^y naMxaL 
Sometbiug, however, haa been done towards its aimljBi% M 



the result of Dr. EUdn's work at Yale College in 1884-85. 

Leaving nothing to be deeired in the way of skill and care, 
it was the more Btnctly comparable with Beasel's from having 
been executed, like his^ with a heliometer, one of about seren 
inches aperture, completed in 1832 by the Messrs. Bepsoild 
of Hamburg. Sixty-nine stars, down to 92 m^nitude, wan 
included in the survey, only one of Bessel's being omitted, 
while seventeen were added from the Bonn Dorchmusterung 
The close agreement, on the whole, between the places detei^ 
mined, after an interval of forty- five years, at Konigaberg 
and Yale, lent importance to some minute discrepancies 
the most considerable of which intimated the probability that 
eix of the objects on Bessel's list were only apparent members 
of the cluster/ They should probably be regarded as peendtK 
Pleiadee, intruders into a company from which they will 
eventually be expelled through the cumulative effects of in- 
compatible movementsL Exempt from the influence of the 
current bearing Alcyone and its true associates slowly towards 
the south -south -east, they remain almost absolutely stationary, 
and are accordingly in course of being left behind. Recent 
counts by Pickering* and Stratonofif^ make it &irly certain 
that the majority of the small stars within the area of the 
Pleiades will be left behind with them. The ground for this 

1 AnnaUa de VOhKrvaioint t, xit, u, ; Comptes J^outut, t IxxxL p, 6, 
' MtjTtthty Natiai, toL aOiv. p. 857- 

rran*. Yrds CoUege Obtfrvaforif, Tol. L pt. L IS87. Pt viil contwns 
reialta ofn SBcoud triaiigul&tiou by M. F. Smith id lfiOO-2. 

Dr. Klkin expresaed this tIqw with iwuBicierable reMm u ngardi roar of 
the sis. stATB, Cf. hi» reviBMl oocduaLotifl, Traiu. Yah ColL Obtfrvatory, Tol L 
pt. vii. p. 3G6, ]»34, 

' Harvard Circular, No. 17. ' Attr. Nath. Nff. 34il. 




THE PLEIADES 



221 



;ce is that they ore leas densely strewn than the multi- 

.0U8 stare in the adjacent sky. The true glomerahiU 

is then formed of brighter objects, the number of which, 

"tlioiigh still unknown, can be ascertained after a moderate 

lapse of time by renewed photographic comparisona 

The proper motion of Alcyone reverses, with approximate 
accuracy, the direction of the aun'a progress through space. 
It may hence be regarded as parallactic, that is, transferred 
by perspective from our own. If this be so, the distance 
between Alcyone and the earth can be calculated, given the 
direction and velocity of the sun's translation. Now we know 
that the sun is travelling towards a point in the conBtellation 
Lyra at the rate of about twelve miles a second. On this 
showing, the Pleiades are so remote that their light takes 190 
years to reach us (parallax = "'01 7). Nor is the estimate 
likely to be materially diminished, 

Our own sun, thus far away, would shrink to a star 

of 8' 6 magnitude. There can be little doubt, in fact, that it 

is surpaased in brilliancy by fifty to sixty of the Pleiadeg. 

And it must be, in some cases, very greatly surpassed ; by 

Alcyone 170, byElectra 83, by Maia about 70 times. Sinus 

itself takes a subordinate rank when compared with the five 

I most brilliant members of a group, the real magnificence of 

l^^hich we can thus in some degree apprehend. 

^H The scale of its construction is no less imposing. No 

^judgment can of course be formed as to the interval of space 

' B&parating any two of the stars belonging to it. All of them 

are seen projected indiscriminately upon the sajne plane, 

without regard to the directions in which they lie one from 

the other* The line joining Maiaj for instance, with Alcyone, 

may be foreshortened to any extent, or not at all. No 

criterion is at hand which we can apply. Of the dimensiona, 

however, of the cluster as a whole, some notion can be 

gathered- For its shape — irrespectively of some outlying 

streams of small stars — ^may be taken to be rudely globular ; 

and since a circle described from Alcyone as a centre with a 

radius of 48', includes all the princix>al stars, sixty of Klkin's 

sixty-nine, fifty-two of Bessel's fifty-three falling within it, 

the apparent diameter of the denser part of the aggregation 

cannot differ much from 96'. But the proportion of the 




THE SYSTEH OP THE STABS 

radioB to the disUnoe of a ^LcA» is known from eleme&ti 
trigoDOOBfttrf , and hoe ooaes oofc (in roand nombere) m i 
to wTOufcx-ooe; ao tfant tfaa bodin ntnated doae to iia snrfHt 
•M serentj-one times neuer to Uieir c^stzal lamioaiy U« 
tlifiir oen^sl IimiiiiuT' la to as. If thej revolve round it, ii 
IB at &XI interval exceeding fifteen billion miles, costing lig^ 
not far from three joaia (o cnaaa; and the period of their 
ciTciilation m&j well be reckoned by millions of yearn Upc« 
these dependent otba, Alcyone shines with sixty timee the 
loBtre of S'liiua in terresCrial ^es ] yet the presence of 130,000 
Alcyonea would only juat compensate for the withdrawal d 
even such a diminished sun aa brightens the firmament of 
Neptune. From staia moi« centrally placed, the chief of the 
closter doubtleas appears a Teritable aun^ although it may not 
be to all the primary light-giver. An assemblage like the 
Pleiades distributed round our sun would extend eompaetlf 
three -quarters of the way to a Centauri, its feelers and 
appendages indefinitely farther. Hence there would be ample 
room in it for secondary ayatems and particular associations of 
luminoua bodies. And, in point of fact, the actual dusttf 
contains several of Bumham's close double stare, one certainly,* 
all presumably in mutual revolution, to say nothing of the 
doubtful companion of Atlas which, distinctly visible only 
onisd to Struve in 1827, gave nevertheless some sign of itB 
preeence during an occultation by the moon, January 6j 
1876.* 

The discovery of Maia as a spectroscopic binary suggests 
an indefinite range of hidden complexities in the mechanism 
of the cluster. It ensued in the course of a research on the 
radial velocities of the six leading Pleiades, carried out by 
Mr. Walter Adams with the Bruce spectrograph of the Yerkes 
Observatory in 1903-4.* The diffuseneaa of the spectral lines 
in these stars impaired the precision of most of the determina- 
tions; but Maia offered more facilities than its companions; 
and the variation of its speed to the extent of about eighteen 
miles a second seems indisputable. The period of its re- 
volutions baa atill to be assigned. The other five stars all 
proved to be receding from the sun at rates corresponding 

1 AUt. JtfoM. No. 8047. * Ibi<L No. 2074, 

■ AsiropK. Jqw%. toI lix. p. 336. 




THE PLEIADES 



233 



■ly well witb the 8un*B celerity of withdrawal from them. 

^That ifl to say, no evidence of individual motion (unless 

3K)B8ibly in the case of Taygeta) was elicited from them. 

Spectroecopic measures are then no moi^ immediately hopeful 

"iban micrometrical measures ^ for obtaining a clue to the 

djruamicHl condition of this marvellouis star-group, 

A iSpectrum of helium type characterises itB genuine 
members. A simultaneous spcctrographic impreseion obtained 
by Professor Pickering from close upon forty of tlieBe etars, 
January 26, 1886, demonstrated the nearly identical quality of 
their light, and fumifihed " strong confirmation of their common 
origin." ^ Only in two cases, a stronger " K line " recorded 
itaelf than auch light ordinarily includes, and the divergence 
was, in one of the two, both accentuated and explained by 
diversity of motion. The star in question {a Pleiad um) haa 
been already signalised as an incipient fugitive from the group 
to which it never truly appertained. 

The stars of the Pleiades, while shining with so poignant 
a lustre as to make the sky-ground they are relieved upoa 
show to the eye as blacker than elsewhere, are in reality 
wrapped and entangled in an immense cosmical cloud. Some 
indications to this effect caught by optical means have been 
autographically amplified to so surprising an extent that the 
discovery of the nebulous condition of the Pleiades ranks among 
the most important achievements of celeatial photography. 
The " Merope nebula " was compared by M. Tempel to a 
,ia of breath upon a mirror. Discovered by him at Venice, 
October 19, 1859, it envelops and stretches back in comeCary 
shape from the star to which it is attached, covering a space 
oi about ^5' by ^O'.* But this large size only makes its 
perception more difficult, by impairing the effect of contrast 
with the surrounding sky. High magnifying powers (which 

e imply narrow fields of view)^ render it on this account com- 
pletely inviaible, and a haze so slight as to permit the observa- 
tion of stars of thirteenth or fourteenth magnitude sufhces 
I I Ellcin, Trans. YaU CoUtgt Obweraiiory, toL L p. 101. 
* ittmoin Amer. Atad. voL xi. [\ 21D ; Soirvard AnmaU^ vol. zzti. pt ii. 
p. 3d2, whore the apcctrk pf Diiuty^one membein of tha group vo BpeoilieaUy 
Noarded 

« A^r. Jfaeh. No. 13W ; MonlMv Noiiu$, toI. zL p. 622. 



, the 

^^i 




THE SYSTEM OF THE STARS 

to oblit«tate it. This evasiveness auggeeted varubtHtj; If. 
Tempers contrary opinion has been fullr justifietL 

The idea was entertained both by Goldschmidt^ aai 
"Wolft thftt the filmy veil flung round Merope was but a fag- 
ment of a larger whole ; and as time went on, gUmpees ven 
snatched of imsty ahreds and patches in cotiuectioa with otbff 
members of the gmup. Alcyone appealed to Searle at Harrari 
College, November 21, 1875, surrounded by whitish hght;* 
the effusion about Merope, eyidently to Schiaparelli in lS7h? 
and to Maxwell Hall in 1880, reached Electra and eTm 
Celaono ; * while a leniarkable view afforded to the late Dt 
Common by his three-foot retlector, February 3, 1880,* rf 
three feebly luminous blotches between Merope and Alcyooft, 
prompted his comment that " there ia a great deal yet to be 
settled as to the extent and number of the nebulf& in thii 
duBter.'* 

Its import, however, became apparent only when 
photography was brought to bear upon the subject. ITm 
first nebula discovered by the new method was a small BpiraJ 
appendage to the star Mala, which printed itself on plates 
exposed by the MM. Henry, each during three honra, io 
December 1885." Only the accumulating faculty of the 
" chemical retina " co»ild have revealed the presence of aa 
object BO excessively faint in a telescopic sense ; but what fe 
known to exist is, by that alone, rendered more than half 
visible, and the Maia nebula was accordingly discerned. 
February 5, 1886, with the FuDcowa thirty-inch refractor, 
then newly erected, and later with smaller iuBtruinents.^ 

Besides the Maia vortex, the Paris photographs depicted 
a series of nebulous bars on either side of Merope, and a 
curious streak extending like a finger-post from Electra towards 
Alcyone. But all these were mere samples of what lay 
behind. Impressions of the Pleiiides secured by Dr. Roberts 
with his twenty-inch retlector in October and December, 1886, 
showed the whole western side of the group to be involved in 

^ Lf> Mtmdtt, t. lit. p. B29. ^ Harvard Annah, vol. xiii. p. 71. 

" Attf, Nadt, No. 2045, 

* Mtmthiy ^'oficet, toI, xU. p. 515. * Tbid. vol. iL p. 37«. 

* Similarly recorded a month earlier at Harrard Oolkfef it vm takon for a 
Saw in the D£gaciv«. 

T Attr. Jfaeh, Noa, 2719. 3726. 2730. 




THE PLEIADES 



22S 



Doe vast nebulous formation.^ " Streamers und fleecy iBasaes " 
of coBiuical fog seem, in these aatomsbrng pictures, almost to 
fill thi3 gpaeeB between the stars, as clouds choke a mountain 
valley. The chief pointB of its couceutratiou are Alcyone, 
Idierope^ and Maia ; but it includes as well Celseuo and Tay- 
geta> atid is traceable southward from Asterope over an arc of 
1' 10'. These photographs, in fine, as Mr- Wesley wrote, 
•• not only prove beyond a doubt the existence of the much- 
disputed Merope nebula, but they also combine and harmoniee 
itt a very satisfactory manner the apparently irreconcilable 
drawings." ^ 

The matter was not allowed to rest here. Early iu 
1888 the MM- Henry succeeded in giving to several plates 
eacposures of four hours, with results identical in each case, and 
very eurioua Their nature can be estimated from our frontis- 
|Kiece, which reproduces the final chart of the Pleiades prepared 
by the MM. Henry. The greater part of the constetlation 
is shown in it as veiled in nebulous matter of mot^t unequal 
dsnsity. In some places it lies in heavy folds and wreaths, 
in others it barely qualifies tho darkness of the sky. The 
details of its distribution come out with remarkable clearness, 
and are evidently to a large extent preecrihed by the relative 
Bituatiou of the stars. Their lines of junction are frequently 
marked by nebulous rays, establishing perhaps between them 
relations of an unknown nature ; and masses of nebula, in 
niuxkerous instances^ seem as if pulled out of shape and 
drawn into festoons by the attractions of neighbouring stars. 
But the strangest exempUlication of this filamentous tendency 
ifl in a fine, thread-liko process, 3" or 4" wide, but 35' to 40' 
long, issuing in an easterly direction from the edge of the 
uebula about Maia, and stringing together " like beads on a 
rosary " * seven stars met in its advance. Two similar rectilinear 
nebuhe run parallel to the first, and a fourtli was photographed 
by M. Stratonoff iu 189G.'' 

Wliethcr these luminous highways are due to material 
condensations, or merely indicate tracks of electrical excite- 

' JfoivfA/y Jfotiert, voL ikii. p. 24, 

* Jount. Liv. Astr. Soc^ voL v. p. 160. 

* Motiohex, Comptes Jiewiug, t. cvL p. 013 : H. C. Wilton, AHr* and Attirs- 
iricf, Tol. xiil [J. 192. 

* Attr. HmK. No. SSS«. 

X$ 



22fl THE SYSTEM OF THE STABS 

meDt, Uiej &re equall^r coimiiumcAtiye upaa one point Tbi 
oozmection by their meaas of stars into zows virteoUy dt 
monfltiabes tbeir real aligcuneDt, and thtiB oooakknib^ 
strengthens the presampdon that the linear mxnagaimt 
preT&lect in dusters is no optical iltaoon^ but depeoda Vfm 
intrinsic conditions, the outcome of univerBal laws.. 

The wonder of this aggregatioQ of stan and ixebol^ h 
bGen enhanced hy Professor Bami&rd'e diacoTeries. In l89U.* 
he detected visually a bright round object, like a comet with- 
out appondagea, in Buch close ptoximitj to Merope aa to foca 
with it probubly a nond^ript binary combiaattcm;, T^ 
three years later,^ two exposures with the WiUArd lena broo^ 
into view a «et of curving streaks iBsoing from the cbisltf 
as a whole, and enfolding it exteriorly in far-reaching diiD 
nebulosity. Ita genetic history can hence be judged to b« 
Btill at an aarly atage ; yet the unity marking it is already 
singularly diversified. Many orders of stars are there gatjiei«d 
together iato what might be called a micdatuTe sidereal sjrstem, 
the largest of such " surpassing gtory " aa to dim by comparisoo 
the splendour of Siriua and Vega. The " act of order " in tins 
** peopled kingdom " is not easy to divine ; we eao only stt 
that the mutual relations of its denizens must be highly 
intricate. Within the wide framework of the association 
room is found for subordinate groupings of various obaradfcen 
and degrees of closeness, from stars far apart, but drifting 
in company, to pairs aa unmistakably united by contiguity as 
two nuts within the same shell. Thus, the polity governing 
the entire system of the Pleiades would seem to be of the 
federative kind. Nor can we be yet sure that its boDd^ 
while evidently so loose as to give lutshackled play to looal 
liberties, are nevertheless sufficiently strong to restrain tlie 
slow worktugs of disruptive tendencies. 

1 A9lT, JVocft. No. SOI 8. * /M, No. S353. 



CHAPTER XVin 



STAK CLUSTERS 



CSbuT five hundred clufiters are at present tolentbly well 
mown to astronomers, and a large number besides, their 
ibaracter rendered ambiguous by distance, are probably in- 
Inded among botb " resolvable " and " unresolved " nebulee. 
Inch aggregations may be broadly divided into " irregular " 
iiid "globular" clufiterB. Although, as might have been 
ixpected, the line of demarcation between the two claaaea 
a by no means eharply drawn, each has its own marked 
»eculiaritie& 

Irregular clusters are ftamed on no very obvious plan ; 
bey are not centrally condensed, they are of all shapes, and 
heir leading stars rarely occupy critical positions. The stare 
n them are collected together, to a superficial glance, much 
iter the fashion of a flock of birds. Alcyoue, it is true, 
leems of primary dignity among the Pleiades, and the Pleiades 
n&y be regarded as typical of irregular clustei's; yet the 
lominance, even here^ of a central star may be more apparent 
han real 

The arrangement of stain in clusters h, nevertheless, far 
rom being unmethodical, even though the method discernible 
Q it be not of the sort that might have been anticipated. 
'.% seems inconsistent with movements in closed curves, 
ind suggests rather the description of hyperbolic orbits, 
ifet its true nature must obviously be greatly obscured to 
►ur perception by the annulmentj through perspective, of 
■he third dimension of space, whereby independent group- 
ngB, projected indiBcriminately side by aide, are rendered 
jarely if at all recognisable. That they should, under these 



%2B 



THE SYSTEM OF THE STABS 



circamstanoeA, stand out to Anj extent is mora sorprais; 
than that thej should aometimee be inextrjcabljr e&UngIri 
with sprinkled etais belonging to the fore- or haefcgranl 
The nebidous ImlHng together of a septupie aet in the FUld* 
aaaureB oa, nevertheless, that Btar^^alineatioiis are o«>i illcwT 

Nearly oil observers have been ixupreaaed wntt tU 
streaming and reticulatied character of many stnllar m»a. 
blagea. Thus where the feet of the Twins <U] 
Way, an object is enconnlered so "nrnrvelix.^.'-.j ^ww».-_ 
with a Urge telescope^ that "^ no one coold see it for Lb« fij<c 
time/' Mr. Lassell declared, "without an excIamatioB.* A 
field 19' in diameter " is perfectly fuU of briUijuit atan sa- 
nanally equal in magnitade and distribution over the v^U 
area. Nothing but a sight of the object itself can cor 
idea of its exquisite beauty/'^ Admiral Sm}'th dcscj;.-^ ■ 
aa " a goi^ecma field of etaia &om the ninth to the sixteenlti 
magnitudes, but vrith the centre of the maas leas xich Uud 
the reet. From the small stars being inclined to fonn cunes 
of three or four, and often with a large one at the rcvot of th' 
curve, it somewhat tieminds one of the burning of a ekj- 
rocket." * A photograph of this duster ' by Professor Bamar>], 
teproduoed in Plate XL, Fig. 1, leaves very little doubt of its 
intimate galactic affinities. The sinuous linens of etArs i 
compose it/ although more dosGly entangled than the 
catenary arrangements on the leas crowded parts of the 
can scarcely be organically distinct from them. 

Yet the radiated aspect of stellar throngs lends them a 
quaai-individuality. Tbe singular looped conformation \*i£ib)o 
in the " gold-dust " duster in Auriga (M 3 7) attracted the atten- 
tion both of d'Arr«st and Lord Eoese ; * about one bundretJ 
connected stars in Ophiuchus (N.G.C, 6494) "run in lines anJ 
arches";* a superb assemblage in Cassiopeia (N.G.C 77S9; 
was deecribed by Lord Roaae as formed of jagged branches with 

' Monthly Noiicta, voL xir. p. 7A> 

* Cycle, p. IftS (Cbunbera's ed.). 

' U 3fi = N.0.C. 2163. NcibuUe ud closters tli»aghoat thk toIcbic 
■n dutingnished bf Ucuier's welMfQOwn oambera, irbeD uooog Uie 10^ 
epumflratod bj him^ atherwiae by 7>rejeT'£ in the Kne Ofnerai Caimtsgm'. 

* 3*cchi, jftti deir Aeead. Font. t. rii. p. 72. 

* Tm«. li. Irish Acad, to], ii. p, 51. 

* PML Tranf, vol cinii. p. 460. 




^ 

^ 



STAR CLUSTEBS 



irk holee between, and by Dr. Roberts, from hia photographs, 
^s exhibiting curved and wreathed patterns in stars.' The 
constituents of a large group near the Poop of Argo (N.G.C. 
2667) struck the elder HerBchel by their arrangement "chiefly in 
rows/' illustrative, to his mind, of the mechanical complexities 
of such eystemg. Each row, he observed, while posseraing its 
own centre of attraction, will at the same time attract all the 
others; nay, "there must be somewhere in all the rows to- 
gether the seat of a preponderating cluBtering power which 
will act upon all the stars in the neighbourhood."* Specula- 
tions, indeed, upon the dynamical relatione of " stars in rows " 
are still prematui'e, nor are they likely, for some time to come, 
to be accounted aa " of the order of the day/' But the con- 
tinual recurrence in the heavens of this mode of stellar aggre- 
gation cannot fail, to suggest the development of plans of 
systemic dissolution and recomposition on a grand scale, and 
involving the play of, by us, unimaginable forces. 

The more attentively clusters are studied, the more intri- 
cate their construction appeara That which challenged 
HerscheFs notice is not singular in lEtimating a Icjigue of 
several co-ordinate groups. There is rarely evidence in the 
conformation of irregular clusters of their being governed 
from a single focus of attraction,; there are irequent indica- 
tions of the eimultam^Dus ascendancy of several. A cluster in 
Sagittarius (K.G.C, 6451) is distinctly bifid It was re- 
marked by Sir John Herschel at Feldhausen as " divided by a 
broad, vacant, straight band";° and his figure shows the 
separation as absolutely complete, the sections, 

Like cliffs which had been rent asunder, 

facing each other with a chasm between. 

A beautiful cluster in Sobieski's shield (M 11), first noted 
by Kirch in 1681, seems to be extensively dislocated. Sir 
John Herechel succeeded, by the use of high powers, in break- 
ing it up " into five or six distinct groups with rifts or cracks 
between them." * Father Seochi perceived in it a three-lobed 

^ Cfiestial Fhoio'jTapht^ vol. i. p. 129, 

2 Phil. Trans. toI. civ. p. 289. 

' Cape Obiervaiiont, p. llO. 

* F\ii, Trans, vol. oniii. p. 4fl2. 




230 THE SYSTEM OF THE STABS 

central vacuity;* and M* Fenet, who in 1895 m&pped from 
Dr. Roborts's phatograpbe 395 of ita oompocenta, ' remufaii 
their most likely progresaive sepftxation into aeven or ag^ 
difitinct aUotmeuta. 

Plate XYI. shows M 11 in iba Milky Way kkc&tiaii u 
photographed by Frofe-saor Barnard with the WillArd km 
In hia opmion it ia *' hardly queetionabU " that the cloUtt 
(here ueceaB&rily contracted into a simple cltimp) really octo at 
nucleus to the vast atar-clouds * attached to it like volondoois 
wings. But bow strangely related auch a nucleus must be to 
such a, formation I 

This " glorious object " (as Sir John Herachel called it) ow 
just be made out with the naked eye on a perfet:tly cle«r 
night. HaUey mentioned it in 1716 as " of itself but a small 
obscure spot, but with a atar that ahiaes through it whii^ 
makes it the more luminous." * Some years lat6r, Deiliam 
found it to be " not a nebulose, but a clui^ter of stars, somewhat 
like that which ia in the MUky Way." ' A catalo^e of tw& 
hundred of the components, prepared iu 1870 by M. Helmert, 
of the Hamburg Observatory," provides material for the futur* 
investigation of relative changes. 

The presence in a cluster in Monooeroa (N'.CC. 2269) of 
** a double seat of preponderating attractions," was observed by 
Sir William Herschel ; ' and a throng of some two htmdied 
stars in Cancer (M 67), discernible with an opera-glaas, falb 
no less obviously into two divisiona.^ In a collection seen at 
PaiBonatown to be riddled with absolutely dark '* lanes and 
openings"* {N.G,C. 2548), the principle of local self-govern- 
ment has evidently been already carried a long way. A 
"reticulated mass of small stars" iu Cj-gnus (K.G.C. 6S1&) 
was there described as " a most gorgeous cluster, /u// of Kolu"; 
and the drawing published by Lord Bosse depicts a winding 

1 AUi deW Jccad. Pwt, U Tli. p. 75. 

* £ul£. Soei&i Astr, de France, 1895, \\ 6fi. 

' Aitrsph. /imm. vol. i. p. 11, 

* Phil. Tr&M. vol. ixix. p. 3Q2. 

■ Ihid, vol, jutviii, p. 72. 

■ P^lieatimen der Hamhurgw Stemware, No. I, 1S74. 

' Phil. Tranit. vol. civ. p. 208. 

• Smjtb, Cffctf, p. 241 ; L. Fenet, L' Attronomie, t. vi. p. 145, 

* Trant. R. DuHin. Soe. vol. ii. p. 6fl. 





*.lltll. 



Fk»tngi«|>l!i ipr ili'SRiw 1 1 \iw\ ailjiiTUt (ialftctic Cli^ini funiis, \ltai'jirti-cl. ) 



I 



STAK CLUSTEKS 



2S1 



bbon of aters inclosiDg three blank circular spaces of sym- 

letrically varying diameters. 

Ainong the '* curioaitiea " of the heavens are to be reckoned 

laaters within clusterBL Thus, a large loose collection in 
emini {N,G.C. 2331) involves a neat group of "six or 

5veQ stars close together, and well isolated ixom the rest." ' 
parallel instance is met with in N.G.C. 2194, situated where 

le Milky Way passes between Gemini and Oriou ^ and within 
|ha bright cluster M 67 in Cancer^ Dr. Roberta was struck 

rith a knot of five stars of tenth to twelfth magnitudes, 

■their photo -images touching. If they should be proved/* 

le added, " to be physically connected, the revelation would 
be astounding."'^ 

Star- groupings of singularly definite forma frequently 
occur, A triangular swarm (N.G.G, 7836) presents itself 
in Cetus; a rectangular area in Vulpecnla {N,G.C. 6802) 
is densely strewn with fine star-dust. Clusters shaped like 
half-open fans are tolerably numerous. One in Gemini, if 
removed to a sufficient distance, would appear^ according to Sir 
John Herschel, "as a fan-shaped nebula with a bright point 
like a star at the vertex." Another specimen of an " acut- 
angular" cluster 2' in length (N.G.C. 7510), is bounded by 
" two principal linea of stars drawing to one." ^ 

In Cygnua is an oval annulus, 4' across (N.G.C, 7128), 
of stars centrally surrounding a ruddy one of the ninth 
magnitude^ A similar elliptical group, with a double substi- 
tuted for the red star, is centrally placed in one of the two 
great adjacent clusters in Peraeus (K.G.C. 869).* This 
superb object, like the scarcely inferior assemblage (N.G.C. 
884) it immediately precedes, was regarded by Herschel as 
merely a protuberance of the Milky Way, and his intuition 
was probably correct. The two togetlier form a telescopic 
pageant such as, in the wildest flight of imagination, 
Hipparchus could little have dreamed would one day be ■un- 
rolled before the eyes of men out of the "cloudy spot" in the 
Bword-handle of Perseus which he (it is said) was the first to 



1 Trans. J?. liuhtiii Sag, p. 66. 

^ CeUstial Phoiographs^ voL i. p. fl9. 

* Phil, Trans, yo], csKiii. pp. 476, 603. 

* J, HcrBcliel, Phil. Tram. yqI, cxiiii. p. 573. 



» I »— ♦ 






^- X - 

-r - ; r — - 






• T \r- T 






A 



.k.V . A.V ... 



.■..'. :r:-i V. ::;:■ • ::.: .::•::.:- :y Mr. H. C 
Hi ].^72.' T:.-- :! -:.■■•. ^vi- :■ '^ liiu *'''.: ■ 
it 'Iw ill i'mI iJiov-;. .-.:.".-. '.•.'.■.lii :■►■ rif ox'.r 
iKj'i'-r Ui*- < ij'-uiij'*. ii:- ■■-. th-y 1,'uly liii--.- u 
'■I..iii;'f, -ill' •: H'-l -•■!.■ 1'- liiL'.i-'Urcl:;L-:its ^\ 
h.T-,iy t'j i/" iiiihul<-]y i"li.il']t.'. 

' .Sr, rl,, OW-, ji. f;.'. - As*r. . 

' /Ar St,.n\liiv<' i\ X I'crsei, p. S 

* Moathhj Xnii-ys. v-l. 1. p. :jl; 

^ '.V/^'c (itistrriitiotis, }>. 17. " Month? u Sot 



STAB CLUSTERS 



233 



la the constellation Cancer may be seen, any fine night in 
iter, a blot of dim light placed midway between two fourth- 
agnitude stam The stars were called by the ancients the 
Aselli, the interposed cloudlet representing to their 
a " Manger," FroEsepe. Since its disappearance was 
£oned a sure presage of itiin/ a good deal of popular 
attention was paid to it, and its stellar constitution was one 
of the earliest telescopic discoveries ; but only preliminary 
steps have been taken towards its exact investigation. Of its 
components^ thirty are measurable on Rutberfurd'a photo- 
gmpbs, and 363 were mapped over an area of three square 
degrees by C, Wolf some sixteen years later, eighty-two among 
them being carefully determined as points of reference for their 
fellows,^ Asaph Hall's catalogue for 1870 of 151 of these 
stars has already been turned to account by Sch ur of 
Gottingen for testing the relative fixity of 45 among their 
number,^ previously (in 1858) measured by Winnecke with 
the Bonn heliometer ; * but no assured results as regards 
either their concerted, or their individual movements, have 
yet been elicited. Most of the stars in Praesepe yield spectra 
of the solar type."^ 

The particles of a drop of water are not in more obvious 
mtitual dependence than the constituent stars of globular 
clusters ; *' the most magnificent objects/' in the elder 
Herschel's opinion, " that can be seen in the heavens " Were 
there only one such collection, the probiibility of its separate 
organisation might be reckoned " infinitely infinite," and one 
hundred and eleven of them were enumerated by Sir John 
Herscbel in 1864. It does not, however, foUow that the 
systems thus constituted are of a permanent or stable 
character ; their configuration, in fact, points to an opposite 
conclusion. There may, of course, be an indefinite number of 
arrangements by which the dynamical equilibrium of a " ball 
of stars " could be secured ; there is only one which the 
present resources of analysia enable us distinctly to conceive. 

> AtBtOB, IHoaeineia, tt. 160-180, 265 ; Tbeophrastus, Ds S^Kis PluviaruTTi, 
ed. EeiutttQi, p. 419. 

= OoTtiptcs RfndKS, t. lOT. p. 333. 

' Attr. Milth. der 06ltingenji^n Stemtcarte, 1S95. 

• Jbui, Tb. ii. ; Naturt, toL lii, p. &1B, 

' Fiokerlng, Harvard AT^-naltf vol. uyi. pt. iL p. 26i< 



S34 THE SYSTEM OF THE STABS 

Tbi» was adverted to, mAaj years aluce, by Sir John H( 
Kqual revolviog maascs, uniformly distributed throughoot i' 
ephezieat space, would be acted upon by a force Taijiug 
diruUy as the distanoe from, the centre. The reason of Om 

is easily seen ; for the further out a component of such a 
system is located, the more matter there will be ioside, and tk 
> leflB outside ite orbit. The strength of the oentral pull thm 
j ffeaohes a maximum at the surface of the sphere, the velccity 
by wliich it ia baUncefl growing in the same propomoa 
Ellipses described under these conditions would all, acoord- 
i ugly, have an identical period; whatever their eccentricilit*, 
iu whatever planes they lay, iu whatever direction they were 
traversed, each would remain invariable j and the harmony of 
a system in which no perturbations could possibly arise, would 
remain unbroken for ever, provided only that the size of the 
circulating bodioa, and the range of their immediate and inteti^ 
attractions, were insignificant compared with the spatiui intei- 
vals separating them. 

But this state of nice adjustment la a mere theoretical 
possibility. There is no likelihood that it has anywhere an 
aotual existence; and the stipulationa, upon compliance with 
which its realiaation strictly dependa, are certainly disregarded 
in all the stellar groups with which we have any close 
acquaintance. The components of theae are neither equal 
nor equably distributed. Central compression^ ov^r and above 
the merely apparent effect of the gradually increasing depth 
of the star-strata presented to the eye, is markedly efifecUve 
iu globular clusters. Professor Pickerings from careful photo- 
graphic counts of the three typical specimens « Ceutauri, 47 
Toucan i, and M 13 (in Hercules) deduced the rule that the 
number of stars per square minute of are increases in arith- 
luetical progression with approach to the middle point.* Real 
crowding thus intensifies the " blaze," where the stars run 
together, even with powerful telescopes, into an indiscriminate 
silvery eflFulgence. 

Sir John Herachel acknowledged his embarrassment in even 
trying to imagine the *' conditions of conservation of such a 
system aa that of ta Centauri or 47 Toucani without admitting 

' OuUin^s of Astronomy, 9tli ed, p, €36, 
* Harvard AnnaU, vol. xxri. pt. il p, 21&. 





SXAB GLUSTEE3 



235 



pulaive forces on the one hand, or an interposed medium 
the other, to keep the etara asunder." ^ Tims compacted 
kto a whole, they might, he thought, instead of revolving 
iividnally, be supposed to rotate in their corporate capacity 
a single body. But the efltablishment in such aggrega- 
Dns of a " statical equilibrium " by means of an " interposed 
liam," ia assuredly chimerical. The hypothesia of their 
station in one piece is countenanced by no circumstance 
inected with them. It is decisively negatived by their 
Tularitiea of figure. The shai-p contours of bodies whirl- 
on an axis are nowhere to be found among these objects, 
aeir streaming edges betmy a totally different mode of 
misation. 

Globular clusttira commonly present a radiated appearance 
their esterior parts. They seem to throw abroad feelers 
ito fipace. The great cluster in Hercules ia not singular 
in the display of " hairy-looking, curvilinear " branches. That 
in. Canes Venatici (M 3) has "i^ys running out on every side" 
from a central mass, in which " several small dark holes were 
disclosed by Lord Rosse's powerful reflectors;' showing pretty 
plainly that the spiral tendency, visible in the outer regions, 
penetrates, in reality to the very heart of the system. From 
a well-known eloster in Aquarius (M 2), " streams of stars 
branch out, taking the direction of tangents." * That in 
OphiuchuB (M 12) ia provided with long straggling tentacles, 
gf a "slightly spiral arrungement," according to the same 
aathority. And a remarkable assemblage in Coma Berenices 
(M 53) was described by Herschel and Baily as "a fine com- 
pressed cluster with curved appendages like the short claws 
of a crab running out from the main body." * The peculiarity 
in question is the more significant that it is shared by many 
undoubted nebula?, 

We find it difficult to conceive the existence of " streama 
of stars" that are notjlowin^; and accordingly the persistent 
radial alignment of the components of clusters inevitably 
gests the advance of change, whether in the direction of 



1 Cape 0b*crmtiQn9, p. 139. 

" Trans. It. Dub. Smi. vol. ii. p. J32. 

' Ibid. p. 162. 

*■ FhiL Tram. rol. cixiii. p. 4GB. 



236 



THK 3TSTEM OF THK STABS 



€opoeate»tkio or td diflbvcn. Either the tide ai mtntiual i 
Httuig iavmrd^and the * f Iwtf wring pows** (to ue & bvonisi 
phnMB of Sir William BacadifllV) is ttSXl «^M*wfcg itadt to 
collect atus from smrooiidii^ ii{»oe; or elae a ffintfifiyl 
ini|witoft predomxaato^ bj niicb (lUl-grovn orts are diii« 
from tfae Dfuaesj of ■obs in wbieb Uiey irei« leared, to Mek 
tbesr 80pante fijitma^ and enter on an independent qmbk 
6hBwbcire> Bat the qneoCaon «• to whether aepazatiit <r 
a ggrr gtti-fMifirt tendeociai pietail in globular closteis in, for the 
p ree c ot^ beyond the range of profitable di^cnasioEL All tbil 
ean be eaid i« that, after Ihe lapee of aome oenturiea^ photo- 
graphic maararenients 10^7 help towardfi deciding it. 

An object viaoallj teaembling a blurred star below th< 
fourth tnagnitttde, woa named hj Bajer u C«ntanii It neter 
rUeb in theeo latitudes^ but Herschel^s great reflector rerreaied 
it to Mm at the Cape aa a ** noble globular cluster^ beyood all 
compahsoQ Che rtcheat and larg^t object of the kind in the 
heaTen&" ' The stara coDtained in it are, strictly epeaJdng, 
innumerable About 6400 have been enumerated from the 
best photographic platea;^ but a residuum of mottled bax 
indicates the reckoning to be far from exhaustive. Those 
iudiTidoalised are nearly all brighter than the fonrteeatk, 
fainter than 125 magnitude; and no less than 125 amotig 
them have been found by Professor Bailey to flash and fade io 
periods ranging from 475* to 6** 11". The diameter of this 
stellar swarm is put at 40', but eome 1600 of its appatent 
coufititueute are held to beloug to the general population of 
the aky. 

The lovelineaB of the cluBter 47 Toucaai near the Loooor 
Magellanic Cloud was, to Herschers view, set off by a divemty 
of colour betweeu an interior mass of ro&e'tiuted stars and 
marginal strata of purely white ones.' But the effect was 
doubtless eubjecLive ; it met with no later recognition ; and to 
the present writer, in 1888, the fiheeny radiance of thifi 
©xquiflite object appeared of uniform quality from centre to 
circumference. A photograph of it, secured November I7j 



Oapt ObMervations, p, 21, 
) Aiir. avd Aair^h^ticMf tqL xii p- SH ; Sarvitrd Annalt, vol. nxviiL 

' Cape ObaervationM, p. 18. 




STAB CLUSTERS 



23r 



)02, aC the Royal Observatory, Cape of Good Hope, is, 
jugh the kiudnesa of Sir David Gillj reproduced in Plate 
[L, Fig. 2, Probably no other cluster exhibits an equal 
ee of conipresaioD. Withia a sphere of 11' radius are 
luded mgh upon 10,000 stars, of which 5019 have been 
ly counted.* The blanknesa of the auirounding sky 
idera 47 Toucani all the more obvious to unaided aight; it 
indeed, for several nigbta after his arrival in Peru^ 
staken by Humboldt for a cornet-^ Only eight variables 
ive been detected in it — a scanty gleaning compared with 
rich harvests gathered in & Centauri, and in the starry 
sberes M 5 and M 3, situated in Serpens and Canes Venatiei 
ectively. The last-named collection shelters at least 132 
ckering lights, being one in seven of its individualised 
ttmponente. Those of several other elusterSj nevertheless, shine 
ith remarkable stability ; and these eontraata seem unrelated 
I diverBitiea of Btructure or quality In the groupa manifesting 
lem. 

The gradationa of lustre are, in many of these aggrega- 
tions, distributed on a clear!y traceable plan. As a general, 
if not an invariable rule, the smaller stars are gathered 
together in the middle, while the bright ones interpenetrate 
them in rows and branches. ThuSj of a magnificent cluster 
in Sagittarius (M 22), known since 16G5, the central portion 
accumulates the light of multitudes of excessively minute, 
and is freely sprinkled over with larger stars. Sir John 
Herachel remarked of a cluster in the southern constellation 
of the Altar (N.G.C. 6752) : " The stars are of two magnitudes ; 
the larger run out in lines like crooked radii, the smaller are 
maBsed together in and around the middle."" A similar 
arrangement was noted by Webb * in the Canes Venatiei and 
Coma Berenices clusters (M 3, M 63), aa well as in the 
imposing collection in Serpens above referred to, the more 
condensed part of which (compared by Sir John Herachel to a 
snowball) seems as if " projected on a loose, irregular ground 
of stars." s 

* Bailey, Edrvard Annaht rol. xixriii. p. 349. 

' Cosmos (Otti'strana.), vol. iii. p. 193. 

* C&pe ObserAktiiona, [}. lie. 

* Student, vol. L p. IflO. 

' PMl. TranM. vol. csuciii. p. 3£0, 



i»9 



TOR STSTDH OF THS STABS 



ZrreguUhtitt of distiibatmt in 



A Gm 



highi J ffiigtitttKTi 



At TummmowvL, in 1850.^ tins 




" du-k Udm,* Mftwting fti a point cooaiAetmhkj remofvd fm 
tii«i ceiatret wen peiueired to mterropt the briHiaai^ of tbs 
globe of 8Un in Heieaki (M 13> They wen aftenrai 
noogniaed bj Bnffhun and Wdib, and reoorded tbaoaadva 
with emphaAA i& a pbotogr^h taken by Dr. Soberts in 18 ST. 
Globular da«t«n in Ophttichixii (M 12^ ia PegaooB (M U),' 
and ia Caoee Venatici (M 3), appear to be sinailari/ toBneUeiL 
PreooDceiv«d ideas as to the awcfaanism of oelfiotial spbem 
are ntteiix ooofoiuidod bf pheDomena not eaaOr recoDdUhk 
with the proaecuUcHi erif any otderly schente of cimdatefj 
movement. The aeeniix^ rifts, bowe^'cr, are not afaaolutilf 
vacant. A slndy of the cluster in Hcrcnlefl (M 13} Vt 
Mr. H, K* Palmer,' &om plates ezpoeed with the Croesky 
reflector, brought oat the aingoJar fact that its cozopodtttt 
fall generally into two orders, the distiibuUAii of which ■ 
radicaUy dlBTeiieiit. The faint stats, or those below 
miigoitude, are scattered with fair uniformiCy throng] 
spherical epoce ; thoee brighter obey a streaming 
and the gaps between their ramifications show ae dark 
Tha Bame exphmatioti is doubtless valid in all siioilar 

BiffereDcee of distance are alooe adequate to account 
the variety of texture observable in globular clusteraL 
in Aqoahiis^ for instance, likened by Sir John Heivcbel to 
heap of golden sand/' might very well be the domewhat 
grained Hereulea group withdrawn as fer again into 
At a still further stage of remoteness, the appear&noe 
presumably be reached of a stellar throng in the Dolpl 
(N.G.C. 6934), which, with low powers, might pass for a 
planetary nebula, but under stronger optical ootDpuIaion 
assumes the granulated aspect of a true cluster. And many 
more, their genuine nature rendered impenetrable by exceeeive 
distance, are possibly reduced to the featureless semblance of 
" irresolvable ** nebulse. 

But there are real, as well as apparent diversities in these 
objects. Although smaller and mote compact clusteia must, 

* Pfiil. Trans, vol, cli. p. 732. 

» Webb, Oil. of^t€tM, p. ara. 

* AalToph. /ourm. tah i, p. 240, 






riiul't^taplis of tSuiillierii riideieal Ulijctits. 

fi«. 1. l>iiui*4£nM'l' -^1 Me-isJtT ^. Liki'M with tlu^ Bruce Titl*iin.Dpe liy H. I. 
g. "^ PliifiogMiili nf 47 ToTicjinw, taken nt llit^ Mayai Ohw-rTatory, Cspe of 







STAB CLUSTEKS 



239 



on the whole, be more remote than large, looaely-formed onea, 
yet " this argument," Sir William Herachel remarked, " does 
not extend so far aa to exclude a real difference which there 

tmay be in different clusters, not only in the eize, but *lao in 
the number and arrangement of the stare." There may be 
Ifc globular clusters with coinponenta of the actual magnitude of 
ll Sirius ; others, optically indistingulBhable from tbem^ may be 
aggregated out of eelf-luminoufi bodies no larger than Mara^ 
or even than Ceree or Pallas. Our total inability to locate 
them in Bpace leaves us without the means of judging. Nor 
are we likely to be better provided in this respect for an 
indefinite time to come. AU that can be done is to make a 
supposition, and trace the coneequences. Let our example be 
the great cluster in Hercules. 

"This is but a little patch," Halley wrote in 1716, but it 
shows iteelf to the naked eye when the sky is serene, and the 
moon absent.^ Messier termed it "n^buleuse sans ^toiles";* 
yet a " twinkling " indicative of ita stellar character may 
l^e caught with a tele^ope four inches in aperture ; and a 
powerful instrument resolves it to the core. Within the 
precincts of Halley's " little patch," Sir William estimated 
fourteen thousand stars to be " cribb'd^ cabined, and confined!" 
The apparent diameter of this object, including moat of 
the " scattered stars in streaky maasea and lines," ^ wliich form 
a sort of "glory" round it, is 18'; that of its truly spherical 
portion may be put at 14'. Now a globe subtending an angle 
of 14' must have a real diameter y^^ of its distance from the 
eye, which if we assume to be such as would correspond to a 
parallax of -^^j of a second, we find that the cluster, outliers 
apart, measurea 1^600^000 millions of miles across. Light, in 
other words, occupies about ninety-seven days in traversing 
it, while it needs siity-five years to journey thence hither. 
Its components may be regarded, on an average^ as of 13'5 
magnitude; and Mr. Palmer reckoaed at 5482 the number 
distinctly printed on the Crossley plates. 

If, then, 6500 atare be supposed uniformly distributed 
through a sphere 1,600,000 million miled in diameter, an 

^ FMl. Tra-ru, vol. iiii. p. 392. 

* Conn, dea Timps, 1784, p. 233. 

* Fhil, TroHM. toL oxxtu. p. 4fi8. 



240 



THE SYSTEM OF THE STARS 



interval, roughlf^ of 61,000 millJOD» or more tbaa 
five time» the dbtance of Keptune from the son^ 
each froia its ne&reBt uclgbboar.^ Upon a spectator ia 
intermediate eituatton* six stars (beddeB crowds of gt^i 
inferiority) would shine with about sixteen times the 1 
that Sirius displays to uis. Yet since 625 milljoa st&n 
thia brilliancy would be needed to supply the light we 
from the sun, the general illumination of the cluster 
greatly exceed the qualified darkness of a star-lit aight. 

At its aurmised distance, oar 8un would appear as a 
of 6'3 magnitude ; it would shine, that is to say, about 7' 
times ss brightly as an average one of the grouped objectt 
Each of these, otxordingly, emits ^-^ of the solar light; uA 
if of the same luminosity relative to mass as the sun, it 
exercises just j^^ of the solar attractive power. The tniSB 
of the entire system of 5500 such bodies is aooordiog^ 
lees than one - third that of our sun. This may be 
r^arded aa a minimum estimate. The probabilities are in 
faTOur of the cluster being vuatly more remote than we haw 
here assumed it to be ; hence composed of larger or brighwi 
and presumahly more mussive individual bodies than reeoltt 
from our calculation. 

No insight has yet been obtained into the mode of fo; 
tion of globular clusters. Their antecedent state re: 
wholly obscui-e. True nebidosity appears to be absent fr^sm 
them. All those examined by Professor Barnard with the 
great Yerkes refractor proved throughout unmistakably stelLir. 
The nebular relations of leas condensed groups are, on the 
other hand, frequently very close ; and their photographic 
study has amply justified the conjecture that the two dasflflB , 
of object form an UDbroken series — that clusters exist i^fl 
every stage of development from nebulae, and that the advan^ 
cing condensation of many nebulae will eventuaEy transform 
them into veritable clusters. Suggestions to this effect derived 
from analogies of forra^^ are corroborated by numerous observa- 
tions of the actual coexistence with grouped stars of nebulous 
masses. The Pleiades is the palmary but not a solitary 

^ Seo Mr. J. E. Gore's analAgoiu calculatioiui in Jown, Liv. Attr. Soc* 
VoL V. p. 169 ; Studies in Astrvnomy^ p, SO (1(^04). 
* Lockyar, Ptw:. R, Soeiety, toL xliv, p. 29. 





STAR CLUSTEK>S 



241 



stance of a hybrid system. A bright cluster of the same 
:ral character in Sagittarius (N.G.O. 6530) is obvioualy 

mnected with a great nehida (M 8), in tlie meshes of which 
it ^eems as if entangled. Yet tht^ two formations are not 
Strictly c<;>ncentrie. The cluster followSj while it overlaps 
the nebulii. Nor do the coamic llowules adhere 80 closely to 
individual stare as in the Pleiadtra. Plat^j XII., Fig. 1, shows a 
photograph of the combiuatioii taken ut Arequipa with the Bruce 
telescope, June 11. 1896 ;' and a group of nebulous stars iu the 
ittitiit-'diate neighbourhood was detected by Professor Barnard 
on it plate exposed during four and a quarter hours, June 

K. 1892. 
A suggestive photograph of a "cloudy vortex"' in Mono- 
,j^^'os (N.G.C, 2237-39) was taken by Dr. Roberts, Marah 5. 
1899. Within a series of annular undulutious (so to c^ll 
them) of nebulous matter covering a aky-space one degree 
across, a straggling cluster is centrally situated. The associated 
effect strongly recalls that of Nova Persei with its web of 
glimmering whorls, and intimates possibly an analogous 
genetic tie. 

The gpectm of clusters while, in the main, continuous, 
are probably not devoid of individual peculiarities. Sir 
William Huggins was struck, in 1866, with the absence of 
red rays from the analysed light of the great cluster in 
Hercules, and perceived in it irregularities due either to bright 
or dusky bands." They were construed in the latter sense by 
Vogel in 1371,^ since when the inquiry has been unaccountably 
neglected. With the powerful Lnstrmnenta of modern con- 
struction, it might, nevertheless, be profitably extended by 
spectrograph ic means to many globular assemblages; and the 
leeults to be expected from it are a nne qud non for advancing 
much further our imperfect acquaintance with them. 

' Harvard Uircular, No. 15. 

^ Mrii. Roberta, Journ. Brit. jtMtr. Art. vol. xii. p. Ill ; aae ftlsc Barnud, 
Astr. ^ach. No. 2916, Jstr. and Aatrophj/ticf, vol. xili. pp. 173, 642. 
* rhii. Trans, vol, clvi. p. 339. 
,* Aatr. Nach. No. 1B64. 



I6 




with 
fMfcoa ty gwiHiUKin^ hmbI Khe —C^crte of 
ia tiw adovia flpivo^ HeMs wnv to be ibend in tk 
«i tiagib CiMl brasfaai^ hwimTW ; ihej 
eaMBltf7> •u^iuau, oamaam^ Taaeiii 
Aieldiy mihf—il with rtai^ or liaded like tbe cpa i( 
irhwia, diiplaTfld itwiiiianlw^ m weB m nebttloin diar^ ag^ 
ih— ill, triipglw^ panlMogniM^ twin and teiple rjjihf 
One ndnlft^ Uioaght to nfWw the fMse ai an owl, sm 
Mined Moovdingli'; aaother waggeatad a oah; * tiuzd a 
■wan ; « fotxitb (tbe great Onon fomation) became known ■ 
the Flab Month nebula, trom it* toppcned UteieiB lo tbe 
g^ing jaws of a marine moorter Fancy ranged nl Urge 
UuDogb thia wide radn, attemptang to fiamiliariae itaetf 
with the strange objects ooDtained in it by* finding for thoa 
terreatrial aimtlitudeaL 

Within the la«t few yean, however^ — indeed, it maj Ik 
naid, Hioce the completion of the Roase reflector in 1845, — 
iMbiUar ioquiiied have entered upon a new phase. A " glim- 
Otfrmg of reaaon " haa begun to hover over what kmg appealed 
a aoeoe of hopeless bewilderment;. With improved tel^oc^ 
mains above aU, with the aid of photographj— ^m^iurf hae 
blOOBM iuere&»iiigly manifeat among all n1«iffp^f of nebulie. 
^tmcture, not of a finiflhed kind, bat indicating with great 



212 



^ 



THE FORMS OF NEBULA 



243 



probobility the advance of formative processes on an enormous 

scale, both as regards space and time. Masses that seemed 
all but asDorphous when imperfectly seen, ehow to a keener 
scrutiny nodes aad nuclei of condeDsafcion ; curbing lines of 
light, telling of the presence of movement and force, furrow 
theoi ; they are perceived to be rifted as if by a colossal 
thunderbol t, or riddled as if by a portentous cannonade. 
Simple milky eflusions prove to be far less common than had 
been supposed, and exceeaive comple^ty of constitution is 
already a rec<»gnisable characteristic of most nebula>. 

It is one which adds greatly to the interest of their study. 
For a5 the curious details of thi?ir organisation are laid bare 
by the intricate inequalities of their light, the prospect grows 
hopeful of gaining some insight into the nature of the systems 
formed by, or in preparation from them. Optical discoveries, 
while gradually acquiring phyai&il significance, are helping 
to lay the foundation of a " nebular theory '* emanating from 
augmented knowledge, and the discreetly adventurous thoughts 
which it may be supposed to countenance. 

Meanwhile, some mode of nebidar classification has to be 
adopted for the guidance of our ideas ; and since their rapid 
modification through fresh detections allows no arrangement 
to be at present more than provisional, it will be best to 
depart aa little as possible from that already in use. We 
may, then, for descriptive purposes, divide nebulse into the 
following eight classes, which, nevertheless, frequently over- 
lap so widely as to be barely distinguishable : — ( 1 ) Nebulous 
stars; (2) Planetary nebuliE ; (3) Annular nebulee; (4) Comet- 
ai-y nebulffi; (5) Spiral nebuhe; (6) Double nebula; (7) 
Elliptical nebulae; (8) Irregular nebulaa. 

In the course of one of his " reviews of the heavens," Sir 
William Herschel discovered a star in Taurus " perfectly in 
the centre " of a " faintly luminous atmosphere" about 3' in 
diameter.^ The consideration of thig object (N.G.C. 1514) 
and of some others like it, led him in 1791 to the memorable 
concluaion ttiat there exista in space " a shining fluid of a 
nature totally unknown to us." Nothing, indeed^ could be 
chiartir than that " the nebulosity about the star was not of 
a starry nature/' and there is just as little doubt that it is 
' rhif. Tram. toI. Ixni. pp. 71, S2. 




Roberti't pbtai is 1889;' end Pt u J U M y i r BartiArd's phol»- 
gmphic explontuaw hȴe jielded munemos exampln of a 
nmiUr nature aoiuetuMe tifmbtH into multiple gro^ 
Anmig ^ir John HeracbeTa aoathem dtnoTeries wms * dtfi. 

dlATply'^tLned doitt:^ star stntoonded by a bright laminoos 
** fttmoiiphere " 2*10 extent * (N.G.C. 536") : one of Bumltam's 
close pMis lA in atow revolutioa, as already mentioned, at the 
b«arl of the nebula N.G.C 2182 ; while nebulous spoctroooo|nc 
binArieii, despite their upjtarentlj lucongmoua cimuastancM, 
are b^iiiiUDg to take luuk auiong the ordinary products itf 
the Mdereal world. 

A " nebiilouB star " proper foroi8 the centre of an ill- 
defined aureola ; but nebulous adjuncts to stars exist in em; 
variety' of branches and cberelureav wiaps and whorlo, In^ 
dtAd the sequence is so eonttnuou« between bright stars with 
filmy appendages, and pronounoed nebutie involving minute 
stars, that it is often difficult to say vrbether the stelltf 
or the nebular character predominates. Thus, "planetary" 
nebulfir haTe^ witli mre exceptions, stellar nuclei; they can be 
discriminated, however, from nebulous stars, finat, by their 
protlnrninanlly gaaeooH i^ctra, next> by the definite terminatioo 
of their disn:n. 

It ifl then no wonder if, among Herschel's nebulous stars, 
one wiiH found not strictly entitled to bear that name, Thia 
nooduKript objtrl C^*G-^'^ 2392) is situated in Gemini, and 

' Tram. Ji. Dub. Soe. v&i. ii, p. 40. 

• Sidertat MesMetigrr, vol, it, ji, ^8. 

' Uvnthly yoticet, vol ilix. |». 363. 

* Qt)nt ObKTvations, pp. 23, 10/,, 




THE FORMS OF NEBULA 



245 



that it abruck him as somBthing ucusual luaj be inferred from 
his designating it "one of the most remarkable pUenonema I 

Ke ever seen." * With the Parsoiistown reflector it presented 
most " astonishing " appearance. Herachera ** equally 
used nebulosity" was replaced by Eieverul bright and dark 
rings, various in breadth, and perhaps spiral in their arrange- 
ment.^ The diameter of the combination is about 45"» and 
donbta about ito nature were set at reat by the dictum of the 
spectroBCope. D'Arrest" found its light to be concentrated in 
the green ray of uebidiuni (wave-length 5007) — the central 
star or nucleus a^^erting its superior t-xjudensution by the dis- 
play of a faint continuous radtauL^e. It ift then essentially a 
nebula, and generally paBses for one of the annular, or 
perforated kind. A truly nebuloua star of a reddish colour, 
make8 with it (not^ we may Burmiae, fortuitoualyj, an open 
pair at 105''.* 

A nebula in Afjuila fX.G.C, 6781) presents analogies 
to HiTscliclft burred star in Gemini. -T. Herschel con- 
sidered it to be of planetary nature ; but the RoBse reflector 
showed a sudden diminution of brightness towards the 
middle,^ and lUmard unhesitatingly pronounces for its 
annularity.^ 

Planetary nebuLe were first distinctively adverted to by 
Sir William Hersehel Their classification caused him a good 
deal of perplexity. " We can hardly suppose them " he re- 
marki'd at starting, '- to be nebulae ; their light is so uniform 
ae well aa vivid, the diameters so small and weU-defin<idj as 
to make it almost improbable they should belong to that 
Bpecies of bodies." After he had weighed and found wanting 
the hypotheses of their being actual planets belonging to 
distant suns, or distended starsj or comets near aphelion, he, 
nevertheless, (it last decided — rightly, as usual — in favour of 
their nebulous nature/ 



^P Hit. Trana. toL btutL p. 81. 

'^ Trana. B. Dub. Soc. rol. ii. p. &0 ; ue also H. C. Key, Monthty Nctitts, 
vol. xxviii. p. ]&1. 

' Atir. Nach. l^o. \^%h ; Abhandlunfjeii, LuLpsi^, Bd. ili. p. 521. 

■* lAieellj, Menwira R. Aeir, Soe. vol. IXXTJ. pp. 42, Q\, 

'■ Tram. Ji. Dub. Soc. vol. ii. p. 50; Fkil. Trtmt. 1801, p. 752. 

* Monthly Noiit<s, voL Ix. p. 250. 
I? Phil. Trtrnt. vol. Ixxt. p. MS. 



THB SYSTEM OF THE STABS 

Tifij of them vere kmemn wben PidDehBg tifgin 

lag" in 1881, for "ttmn with reouxkablft np&gUm*;^ tad, 

vrithiin a few yeai% npwanU of tweotj more were identiiiei 

throng the qtuiUtf of their light aloDe, bj him and Dr. 

CopeUzMi Tbe«e are. ])ow«Ter, devoid of the oooagitBooA dis 

which waa the ori^nal had^ of their das ; thef are eithtt 

very smalls or very remote pUnetariea ; and are oftca & 

LLDf^ubihcd as " aIoUat nehal^e." 

A true " plaaeUiry " aapect haa not, indeed, in an^ oMt 

8unriTe<l the acrqtiiiy of modem obserrere. What had weemA 

equably iUummated discs arc broken up hj the poweriiil tefe- 

■copea now iu use, into brighter und darker portiocw, & 

tributed in evident rebitiou to some unknown conflict of forott 

Some of these discs include atrongly-tuarked nuclei ; others i 

iprinkliDg of minute stars ; oondeasation towards a spheixtl 

mir&oe givefl to muny the uspei-t of a ring-shaped encloeon; 

few (if any) ure clean at the t'dges. 

Not li few are seemiii|^lj multiplex. Two or three super- 

polled djsoa are traceable in them, hinting at a morphotogicdl 

history Bimilur to that of disturbed comets. 

^fl^^^ Thus a small oval planetary (N,(_i,C. 6572) in 

^^^Hl Ophiuchus wa£ resolved by Vogel with the great 

^^^H I Vir-nrtti Ti^fractot in 1883,^ into three strata of 

L v|f . ii^'i'iilosity ilifiposed as in Fig. 28, representi^. 

y no doul.it, flUL-tesaive apherical imd elliji^pidfll 

- - ' ujivtilupi'B uT lUminishiiif^ luminous power, 

""mWJ 01^; An abJLHit CN.G.C. G210) with an "intense' 

N.b,.iMV«ff.i). i^j^^, ^.^j^^j^ j.^^-j,^ ^p^ ^^ ^j^g distance aU round;' 

ttiid huzy *4t tho odges,^ was i>eTCeived by Vogel as triple^ A 
faint oval husk (ao to speak) seemed to enclose a vivid keruel. 
aud that agtiin to include a stellar nucleus. This nebula is 
aituated iu the constellation Hei'cules ; and one of a simi 
oharocter in EridanuB (N.G.C. 1535), waa described at Pi 
aonstown as preseutiny a grauular. Nile-water blue disc 18 
(MT088, including a atelkr nucleua, and encircled with a faiui 
atmosphere** Mr. Lixssell noted the combined eflect 

* Obtervatory, voI». iv. p. 81» r. |.. SB-I ; MotUhlif iVotiett, ToL xIt. p. 91. 

* Potsdam PukHeatimm, No. H, p. M. 

' 2Va)t3. li. Dub. See, vol. ii, p. J60, 

* /Wrf. p. 41. 






THE FORMS OF NEBULA 



247 



Jextraordiuary and beautiful" ^ Such " nebulous nebulae " (to 
row Dr. Swift's phrii^)'^ are among the most enigmatical 
' celestial objects. 

or special interest ttmoDg planetary nebulse is one lying 
lite close to the pole af the eeliptiic, near the star tn DracouL* 
r.G.C. 6543). Its longer diameter (for it is slightly 
Uptical) measures about 30"; it is of a blue colour, and 
iowe a white star of eleventh magnitude giving a perfectly 
itinuons spectrum exactly in the middle of a diac from 
rhich a purely gaseous one ia derived. Sir William Huggina'a 
fit experiment in the analysis of nebular light wae in fact 
tied upoti the planetary in Draco, which has in varioua way& 
en used aa a test object- Attempts to determine its parallax 
ere vainly made by d'Arreat, Briinnow, and Bredichim* For 
aper motion, too, it was tested by d' Arrest in 1S72, with a 
lilarly negative reatUt. During the eighty -two years* 
iterval since a careful observation by Lalande in 1790^ the 
ebula had remained to all appearance completely stationary'. 
lut this fixedness was really to some extent communicative 
regards its •minintum distance from the earth. D'Aireat 
iowed that unless this exceed a light-journey of forty-seven 
yeatB, the nebula must have become sensibly displaced in the 
course of eighty-two years by the simple perspective effect of 
the sun's advance at the rate of five miles a second.* Now the 
solar velocity assuredly does not fall short of twelve miles a 
second, while the term of the nebula's seeming immobility 
has become protracted from 82 to 115 years. The estimate 
of its distance needs, on both these gi-ounde, to be augmented ; 
and we find, accordingly^ that its light can only reach oui- 
eyes after an interval of nearly 160 yeara,^ if then ', for it may 
spend a much longer time on the road. The real size of the 
^^lobe it emanatea from must be vast iu proportion to such 

^H ^ Memtriri R. A^r. Soc. VoL xnvi p. 40. 
^B ' Aitr, Nash. No. 3474. 

^^B ^ Brdniiow abtamod fbr H iv. 37 a notnlTu.! parallax^ bat Brediohin, from 

^^p&arly douMe thi.- number of obserTatiom, derived n negatW4 citie implying thi^ 

nel>u]a to l>e more rffinotei iliao the tt^ntb-nrngnitudi? star with wbii::b it was 

ciJiQpu-ed. AttT, Nach. No, 'i9l€ (OudetuADs) ; Kn'j. Mec, rd. xliii p, 604 

(H. Sadler;. 

• AaiT. Koch. No. 1885. 

' CerrwapoaJtDg to a pAtalUi of O""*)^. 




t '^ // I ^ **•** - *"•■ 

^^^^_^ II y y shape iBdic«t«d w«s t» 

^^^^^^ ^fe^^^^^ phatoEBBoii taken at Jtoad* 

■ ^^ iB 18»9.* A -bnga, H' 

pUiKUrr'* in Sigitta (S.G.C 6905X with a ck«elf U$adr 

abt Atf at each mdt,* was eMUj wa ul y wl i&fco a ■yiit ky 




M. Antoniadi, with the great Parifl dderosUtt.' His drawrag, 
rcproducetJ in Fig. 30* shows the object under a form sjiuilar 
to that prcHeuted by the " Great Whirlpool *' nebula uDtO 



* Hi/fitKly JifolWt, vol. xlviii. p. sa$ : PuhL A. 8. il vol i p. 36. 

• Dii»JiitidrB«, B^il AmIt. Feb, 1900, 

' KMler, Putfl, Liek ObMrvatory^ vol. ilL p. 213. 

* JCneiciedget vol- iiiii. p. SSO- 



THE FORMS OF NEBULiE 



249 



compelling power of the Rosse reflector was brought to 
*T 00 it. 

Tlieae are nut iaolattid exiiuipliiB. A consklerable number 
' planetary nebulce, as we have Jilready partly seen, Tnaniiest 
:)th Hunulir Miid spinil t<?iulBi)cieB, In some a marginal 
brightemng gives, with sufficient light-concentration, the eftect 
of a ring ; in others, curvilinear effecta of chiaroacuro betray 
incipient spirality of conforniRtion. Since they partake of the 
nature of all tbe three species, their classification is to a great 
extent arbitrary. Five of Herachel's planetaries assumed, in 
fact, at Pargonstown a ring-shape.' One of these (N,G.C\ 
^438). remarkably situated within the eluater M 46 in Argo, 
observed not alone to be pierced with a nearly central 
n^vity, but to contain two, perhaps three atars, towards one 
of which the exterior nebulous ring wound spirally inward.* 
A " bole " too diaclosed itself in a planetary in Andromeda 
(N.G.C. 7662) observed by Lassell ae biannalar, "a ring 
within a ring," To Father Secchi, it had, with liigh powers^ 
the effect of a " magnificent horseshoe of scintillating points," ■ 
the glitter of which was alao evident to Vogel.^ It is. never- 
theless, like all the memlH^rs of its clasa^ of a purely gaftcous 
constitution. The "biannular" planetary is either hazy or 
" fringed *' at the edgea, of a bluish colour, and measures 
32" by 28",^ An etfuaion from its south-eastern extremity 
was pliotc^'uphed by Deslandrea.* 

The blue, or greenish tinge distinguishing, to some extent, 
all gaaeoufi nebulee, is especially conspicuous in a planetarj^ 
(N.G.Q. 3918), discovered by Sir John Heischei in the 
Centaur, and described as very like Uranus, only half as large 
again, its "colour a l>eautiful riuh blue, between Prussian and 
verditer green."' And a " sky-ldue likenesa of Saturn" 
replaced, in Mr. Lnasell'a reflector, a round, faintly luet^nt 
object (N.G.C, 7009) observed by Herschel in 1782 near 
the star u Aquarii. With higher powers, the diec became a 
26" by 16", hazy within and without; and the whole 

* FMl Trails, vol. cil. p, 507, - TbitL p, 5ia. 

■ AtCT. Naeh. No. 1018 ; i*s ^oilts, t ii. p. 14. 

* Faitditm Fi^litatitmen, No. U, p. 37. 
* LaaHQLt, JfemmMjS. Atir. JSm. voL xixtL ]i. 51. 
" BtUl. Aalr. FbU, 1«00, 
' Ca^ ObstTvalioM, p. 100, 



260 



THE SYSTEM OF THE STARS 




<^ 



Fto. Bl,— Anriii! ..- ■■!■','. 



I Aqusrlii^ 



interior assumed, under Vogel'a examtcatioa, a corioos team* 
shaped atructiirc.' Profo«s8or Holden remarked an uneipeCUd 
point of likeness between ihm nebula and the one in Aiidi«- 
med& (K.G.C. 7662)^ in the possession by both of a 
interior oval ring, singularly vrarped and twisted out of tbe 
central plane ; the pei^ulmrtty associating them alaa with 
th« helical planetary in Draco. The atue, or handle-Hh 
appendages producing in the object near p Aquarii its 
resemblance to Saturn with half-opened rings, are a unique 
feature among nebula;. First represented bj I^rd Koa£«, thej 
were resolved hy I^rofeafior Holden * into distinct luminotf 
niaeaes, just tmcrsihly connected witli the main body(8ee Fig.31). 

The analogy with sateIUte-«ten 
ia tenkptiiig, but may l>e am- 
leading. The frequent attend-^ 
ance of small stars upon nebulc 
both planetfiry aad annular long 
ago attracted the attention of 
Sir John HeracheL' One such 
group (N.G.C. S618) struck him as exactly like a planet 
and pair of moons, and stan; in slow ti-ausit across a uebuk 
of which they are the dependents, may often appear projCNAed 
upon it. But not the slightest evidence of movement in theef 
ancillary stare has yet been detected. 

The typical annular nebula (M 57) was detected by 
Darquier of Toulouse in 177^, between and 7 Lyrae. It 
confliats of an oval bright ring, 80^' by 60", the interior uf 
which is filled with a dim nebulous haxe like " gauze stretched 
over a hoop/'* Hardiugj already in 1797, perceived inv^- 
larities in the illumtiiation of the ring ; ^ and vivid patches 
emphasise the extremities of the minor axis. Minima of 
light, on the contrary, terminate the major axia ; the nebula, 
as photographed at Her<?ny, September 1, 1886, taking some- 
what the shape of a pair of parentheses set a. little apart, 
^ but with spiral links between."^ With the Roase 

' Potsdam Publicalionen, No. 14, p. 37. 

' 3f(nUhiy ^atieet, vol. xlviii. p. ^91. ' Phit. TVafu. tdI. cxxiiL p. 500. 

* Sir J. Her»t^lkQj'8 Outlines of J ftr, {>, 644, Sth I'd. 

"[alcii'H, Monthly Nolieta, vol. irxvi, p. 64, 

• VonGotliwd, Aatt. Naeh. Nft. 274». 




THE FORMS OF KEBULiE 



251 



reflector, filaments of u^bulositj were seen streamiug outward 
from the edges ; ^ and the realitj of these singular appendages 
was, after fifty-three years, photographically attested with the 
Croasley reflector.^ 

A central star, discerned in thie nebula by von Hahu at 
emplin, towards the close of the eighteenth century, was 
tnissed by him in 1800,* and has often since evaded the 
scrutiny of better provided observera. It appeared on the 
Her^ny and Liverpool photographs,* but notwithstanding its 
intense actinic quality, left no trace on plates exposed at PaiiB. 

T*r. Max Wolf's recent photographs lend aome countenance 
to the explanation of these anomalieH by genuine variability.'' 
This insignificant light-speck appears to be the very punetum 
stdieTis of the aurrouuding nebulone orguniam. Professor 
Schaeberle's photographs, taken in 1903 with a reflector of 
extraordinarily short focus, display it as the origin of a pair 
of oppositely issuing branches which, coiling in a clockwise 
direction, produce by their amalgamation the effect of a conj- 
plex ring,* MopeovePj the ring is now seen to be the nucleus 
of a larger formation, 15' in diameter, which embraces in its 
dim folds a email nebula, also a " two-branched, left-handed 
spiral" discovered by Professor Barnard in 189S. These 
surpriaiug results, although needing full verification, are re- 
commended to acceptance by the unity they impart to facts 
previously disjointed and fragmentary. 

A nebula in Cygnus (N.G.C 6894) might be called a 
i-eduued copy of that in Lyra, It measures 47" by 41", the 
interior vacuity^ which is partially tilled with faint light, 20"- 
A conspicuous star is included within it.' An object of the 
some kind in Scorpio (N.G.C. 6337) was described by Sir- 
John Herschel as " a beautiful delicate ring of a faint ghost- 
like appearance, about 40" in diameter."** Two stars, or 
nebulouB uode^, are placed in it exactly opposite to each 

' PAt'it. Trans. vo\. cirxiv. p, 322. 

* Kaoler, Ailroph. JonTti. vol. x, |j. 196- 

* Asir. Jahrbue/t, 1802, p. 10. 

* Vou Oothftrd, AtCr. A'adt. No. 27fi4 ; 3pitAl«r, ibid. No. 2&00. 

B A*lr. iVacft. No. ZaU. 

'^ A%(r. Jottrn, NiK<i. &39, 5^7. 

' Lord Roue, TraTu. R. Irish Auttl; vol, ii. p, 156. 

* Caps ObUFvaiitfns, p. 114. 




oUier, aad the whole aspect of the nebiila suggests 

remoteness.^ 

Four ring-nebalm, two in the northern, two in the 9ont 
hemisphere, were known to the Herachels ; and^ as we 
Been, many 80-caU«d " plimeturiee " show iLunular, as aaM 
uebuls ahow spitftl proclivities, Bitigs, in some ins 
visibly curve inward towards a nucleus, giving rise to U* 
variety which we have designated " eometary " nebuUe. Thiit 
a ninth - magnitude star with nebulosity attached (N.O.C 
1999) appeared with the Kosse reflector like "a comet ooiM 
into a ring " * and was photographed, precisely tindar the ame 
aspect, by Dr. Common in 1883.' The triple star, 4. Onmk, 
leas than a degree distant, is enveloped in a nebula of au- 
logoufl form, which ih also less definitely shown bj tiie 
appeudnge nf the star Maia in the Pleiadea. Sir John 
Herscbel's *' falcated " nebula ar^ of the same kind. Os# 
such in Argo, 10' in extent (N.G.C, 3199) displayed to bio 
H " semi'luiiar shape/' dilfu»e ouUide, but with a sharp izmar 
edge.* Another (N.G.C. 346) occurs far to the south in 
Hydrua "A complete telescopic comet, a praiect miniatuR 
of HalleyV''' was encountered in Eridanue (N.G.C. I325'i. 
and star-like coudenaatious, with brush, or fan-like appor- 
tenances, are not iuifre([uently entered on bis lists. A pair 
of amnll nebiihe pliotographed by Wolf and Barnard near Ibe 
bright Btar y Oassiopei^e * are singularly perfect examples of 
rectilinear cometaries. They imitate precisely half -opened 
fans ; the emanation from the star at the vertex has had do 
curvature impressed upon it ; it spreads from a point withooi 
enfolding it; the rotational twist seema absent. 

The discovery of spiral nebulse was beyond question the 
moat important result of tlie construction of the great Parsons- 
town reflector. Its significance is continually enhanced as ths 
wide prevalence of convoluted forms among this whole class 
of sidereal objects is rendered more fully apparent by the 
increasing advance of esplomtion. 

The Whirlpool nebiihi in Canes Venatici (M 51) presents, 



^ li&BBell, MsrrunT» Ji. AHr. Sot, toI. xxivi p. 47. 
TVonj. R, Duh. S&c, vol. ii. p. 50. * ObMervaiorit, vol. xii p^ 84. 

* Cape Obtervations, gjp. '20, 94. ° Ibid, p. &l, 

" Aitr. Jfaek. Nm. 3214, 3317 ; Aar. and Astroph, toI. xiii. p. 1S2. 




THE FORMS OF NEBULA 



263 



M^ a gi"cat telescope, a truly amiLzing appearance. Two 
slei separately oiitnlogued by t^ir John Herschel ai*e then 
to ha conntjcted by an exterior faint sweep Irotii an 
^tter system of ivreathiug nebulous banda. These, too, show 
cloeities and angularities/ in obv^ious mutual relations, as 
the knots, instead of aimply forming upon the spires, bad 
entiined, or at least deflected their course, A still clearer 
awledge of their arrangement was gained tiirough a photo- 
iph taken by Dr. Koberta with four hours' exposure. April 
1889.- The nebula displayed itself, no longer simply 
led like a watt'; h -spring, hut aa composed of a pair of curving 
issuing fi'Dm oppoaite extremities of an oval central body. 
Pne of these loses itaelf in a vague effusion, as a comet's tail 
iee out into darkness ; the other attains the secondary nucleus, 
id there terminates. The spiral character of this great vortex 
perhaps rendered exceptionally conHjiicuouB by its being 
ore favom-ably placed than moat others for our inspection. 
Te seem to get nearly a bird's-eye view of it, and are thus 
labled to take in the design of its construction at a glance, 
spectrum is continuous. 

An object in Virgo 3' acroas (M 99) is a right-handed 

Its branches turn the opposite way from those of 

51. Their tendency to form ncMie.s and angles was strik* 

ziy ahown in a piiotograph obtained in two hours, by M. 

Dn Gothard, April 12, 1888.* A dift'uae nebulous masa in 

Priauyuluiu (M 33) Just discernible to the naked eye, appeared 

rith the Boese rejector in the guise of a " large spiral, full of 

iota."* Since it measures 62' by 35', ita real si^e must 

prodigious, and its structural dt^tails are of correspond- 

ag intricacy.'* Fundamentally, however, the nebula, like 

other known members of itM class, doubtless consists of a 

"Bucleua and two winding branGln?s, the complex ramiHcations 

of which are of what may be culled incidental origin. A 

photograph of it, taken by Dr. Max Wolf, Septemljer 26, 1902, 

is, with his kind permission, exhibited in Plate XIII., Fig, 2. 

Curved furrows of light and shade, conuentrically c^spoaed, 



' VogffI, Puhlicationen, No. 14, p, 32. 

|[gg*1, AitT. Naeh. No. 2864. * fhit. Trans, vol. i-li. |i. 711. 

* KobtrlB, CifMstvU PhotoyraphJ^ vol. ii. ji, 8&. 



THE FORMS OF NEBULA 



25& 



Gt85,^ makes probably the closest pair known; and a 

k^ifl reproduction, with greatly widened gpatial intervalu, 

^tiar syatems like that of y Aadi'omedBe, occurs in a triple 

^Xxla in Virgo, conaiatiiig of a bright round nebula attended, 

distance of 5', by an extremely faint one which is itself 

t>\e* (N.G.C, 5813-14). ^Viiother compound object of a 

ig character was noticed by Sir John HerscheP in Canes 

atici (N.G.C. 4631), where an enormously long ray of 

kolosity has a round, dimly luminous companion, a tenth- 

^^itude star placed between i*er\Hng perhaps as a centre of 

taction for both. In other cases, the syateiuic association 

individual stars and nebuke seems tolerably obvious. Merope 

the Pleiades, as we have seen, claims the attendance of a 

ay satellite ; and Dr. Swift observed at Echo Mountain an 

irently quadruple star to be really composed of two pairs, 

apart, each condsting of a star and nebula. 

A very close double nebula in Gemini (N.G.C 2371-72) 

also an intervening star symmetrically located in the line 

their centres.* Cirrus -like atreaks of nebulosity 

tially encircle the two objects. Duplicity is, in other 

9, still less clearly defined. Thus, a pair of nebulie near 

Leonis (N.G.C. 3226-27) are together enclosed in a faint 

linous envelope, the effect recalling that of the celebrated 

'Dumb-bell*' nebula in Vulpecula (M 27).'^ which ia only 

eived to be eseentially single wlien the " neck " uniting 

vo conspicuous hazy masses is brought into view with a 

ttwerfnl telescope. Sir John Hei-scbel first observed the 

Uptical outline of the entire to be rounded out by faint 

iminosity, and thus saw it in its true aspect as a large, 

iversified oval disc^ measuring about o' by S'. It might 

adeed be called a magnified planetary nebula not devoid 

annular inciinations. The eventuality that by the 

reaa of the central contraction and marginal spreading 

indicated by its present hour-glass shape, the chief part 

its maaa may become di£[\ised into a ring^ is strongly 



' Sid, Mfta. vol. iv, p. 39. 

* O'Arrwftt, Astr. Nuch. No. 1360. • Phit. Trans, rol. oxxul p. 43) . 

* IamwU, Memoirt Ji. Attr, Soc vol. xxUL p. 02 ; Lotd Uo«a0, Phil. 2'ran*. 
dI. 0x1. p. &n. 

* D'Arre&t, AbkoTtdlti^ngtnt Leipzig, 1857, p. 335. 



256 THE SYSTEM OF THE STARS 

auggeeted by the au^ilog}' of the bright Sipot« at either end of 
the minor axia of the ring-nebula in Lyra, A planetary in 
the southern hemisphere (N.G.C 1365) appears, in Etict, tu 
have already reached a more advanced stage on the eame roaii. 
and Beveral '* miniatures " of thu " Dumb-bell " art* included 
among that clas8 (if objects. Oue especially in Cygnu*^ 
(N.G.C. 6905). depicted by Vog/dH with the Viemis 27-iDch, 
eaaily givefi an impression of actual duplicity,^ and showed mI 
Pareonstown as a " beautiful little spiral." It has a centni 
star and four " satellitea." Its visual spectrum, like that of 
the Dumb-bell nebula, ia approximately monochromatic The 
leading nebular my at 5007 concentrates nearly the whole of 
its light. 

A photograph of the Dumb-bell nebula, taken by Dr. 
RoberUf with an exposure of three hours, October 3. 1888, 
mtiinates pretty clearly the advance towards conapleteuees of 
the oval bright border of the disc,^ as well aa its superpoeition 
upou a fainter, more elliptical one, visible as a kind of effiuam 
at the extremities of its longest diameter. Vogel's drawing* 
likewise »u|^6ts, though after a different fashion, the presence 
of two ellipses, one partially concealed behind the other ; and 
there hence Beems reason to think that this singidtir fonuatioD 
partakes in more ways than one of the compound character 
evident in many planetary nebuhe. Nor is it pt-rhaps isolated 
in space. On I'rofeaaor Sc^haeberle'e plates it appeared as the 
bright central part of u right-handed vortex ; coamical relations 
Ijeing thus ascribed to it no leas extensive than those claimed, 
on eimilar evidence, for the spiral Ring in Lyra. 

' PrtbliaitutncH, FoUidflhi, "So. 14, p- 36. 

^ K&lioui-diii, Cvmptet Jitndw, t- imwi. p. ^SO, oommentod ou the aiulog; 
belweaQ Ibc Dumb'beU apd the Lyre bebulK reodared muitfeat by the Moadoa 
j'hotoifriiphH. 

' Publieatianm, No. 11, p. 8S. 



CHAPTER XX 



THE GR£AT NEBDL£ 



elliptical and irregular classes of nebulae are illustrated 
euch splendid examples, that an entire volume rather than 
' fiiugle chapter might well be devoted to their consideration. 
One member especially of each towers above the rest, like 
Ajax among the Argive host, and the two are so different 
that it is not eaay to award the palm of superiority to either. 
Needless to say that we allude to the objects in Andromeda 
and Orion, the types respectively of the elliptical and irregular 
plans of nebular tonatruction. 

The former (M 31) ifi the only genuine nebula which 
could easily be detected, without previous knowledge of its 
existence, with the unaided eye, and it ia the only one, acicord- 
ingly, which was discovered in pre- telescopic times. AX Sflfi 
was familiar with the " little cloud " near the most northern 
of the three stars in the girdle of Andromeda ; ^ and its place 
wae marked on a star- map brought £rom Holland to Paris 
ty De Thou, and believed to date from the tenth century,* 
Simon Mariua, who was the first to turn a telescope upon it, 
December 15, IG 12, called it "stellam quandam admirandte 
figurte," and compared its dull and pallid mys to those of a 
candle shining by night through a semi-transparent piece of 
horn. Yet this strange phenomenon was only rescued from 
neglect by Boulliaud, whoae attention was directed to it by 
the passage of the comet of 1664 across that part of the eky. 
So surprising did the disregard of it by Hipparchus, Tycho, 




^ Scl^jellBrap, Ikterij^ion de» ^oitgs, p. 120. 
* Lfl Oantil, M^nvirti tU VAcad, 1769, p. 469. 

367 



t1 



258 



THE SYSTEM OF THE STASS 



ftDd Bayer then appear to hita that hd coDclitded it to nry 
in light, a hypotbeau which derives no support from reoetii 
obserTAtioiifl, 

With powerful light-conceDtratiou this " most magniiceDi 
object'* {ia Sir John Herachel'a phrase) asaamee Twt pi»- 
portiona They were extexided by G. F. Bond, nsiDg thi 
15-ixicb refractor of Harvard College, to cover an area d 
4" X 3(^ and be probably did not reach their absolute limits. 
Two adjacent nebule^ one (M 52) described by Le GeDtil to 

1749, the other (N.G.C 
205) by Caroline Hencbd 
in 1783, "undoubtedly fifl 
within their oompasB.' ' 
The light of this nebolt 
ifl " of the most perfectly 
milky, absolutely irreaolf- 
able kind" * It does nat 
collect into " floccules>" and 
producer none of the sdntil- 
latiDg effect giving to maor 
guaeoua nebulsB a dt^usive 
appearance of reeolvabilit^. 
From the circumference 
towards the centre, hov- 
ever, it gradually brightens^ then abruptly condenses to » 
small nucleus, of indistinct outline under high magnifyifig 
powers, and possibly (like the nuclei of many comets) granu- 
lated, but asauredly not stellar. 

Thia progressive brightening inward ahows, nevertheles, 
interruptiouB. On September 14. 1847, Bond discerned two 
long dark rifta running nearly parallel to one another, and to 
the axis of the nebula.' Their detection was a consequence of 
the widened area of luminosity perceptible with hia inatrumentr 
the inner rift having been taken, until then, for its boundary in 
that direction. The outlines of Bond's drawing are given in 
the accompanying diagram by Mr. Wasley (Fig. 32), in which 
the " rifta " are marked A and £. G represents Le Gentil's 

* Bond, UcTruiira Amsr. Acad. vol. iii. p. 83. 
' J. HenKhel, Mtmoirs £u A tit. Soc. vaL il. p. 4M, 
' Mtuwirs AmtTv Aead, vol. iii p. 80, 







L 




THE GREAT NEBULA 



259 




\ 



¥?■ 



..D Miss Herechel's attendant nebula, E an exoeptionally lucent 
^ region crowded {it has Binoe been found) with hoata of minute 
Altera.' 

^B Tbeae enigmatical appearances at last assumed an intelli- 
gible form ill a photograph 
taken by Dr. EobertSj October 
1» 1888.* The view given by 
this magnificent picture of 
the Andromeda nebula aa a 
BymmetncaJ, though still 
inchoate Btnicture^ ploughed 
up by tremendouBj yet not 
undisciplined forces, working 
harmoniously towards the 
fulfilment of some majestic 
design of the Master Builder 
of the universe, is of a 
stature to modify profoundly 
CUT notions as to how such 
designs obtain their definitive 
embodiment. An impression 
obtained three months later 
with four houre' exposure 
was carefully studied by Mr. 

Wesley ; and hia tracing of the lin^ of conformation brought 
out on the sensitive plate is copied in Fig. 33, 

Bond's " canals " are now seen prolonged and curved into 
two vast rings (AA and BB), which prove, on attentive con- 
sideration, to be dusky intervala separating the succeseive 
epirea of a single great stream of nebulous matter, winding 
outward from near the primary to reach the secondary nucleus 
(M 32). The similarity of the relations between the two 
nuclei here, and in the " whirlpool " nebula in Canea Venatvci, 
can scarcely escape notice. Thousands of stars are scattered 
over and around the Andromeda nebula; the situation of 
which in a prolongation of the Milky Way perhaps sufficiently 
explains their profusion. A beautiful photograph of it taken by 
Dr. Max WoK, August 18, 1901, with the 16-inch Bruce lens, 




Flo. 33. — Splnl Stmcian of thfl Qre^t Andro- 
meda NebuU {Knotelw^, yoL slU p. 16), 



^ lUnyard, 



wUdgty vol. xii, p. 7fl. 



to], xlix. pp. €C, 130* 



2d0 THE SYSTEM OF THE STABS 

is ahovm in PUte XIII., Fig. 1 ; and serenU striJoAg pkiur*^ 
of tlie same taAfveUouiS object have been obtained br lb 
Ritchey with a two-foot redector of bis own oonsUtacUoii 
Near the ceotre on the oti^noiU aegHtir^, he tolla vb} " sh&H'k 
detmed narrow rifts and d&rk holes " ard ditftinguiduble, tft 
truces of which evade direct telooccypic observrntioa. Thai 
peculiar eSeota of chiaroBcuro are probablj indicatiTe tri 
tumultuary local movements, but attempts to ttxplaia tbem m 
detail must loug reuiain futile. 

Opinioias are divided as to the cooBtitution of the Andxt)- 
meda nebula. Sir Norman Lockyer inferred it to be meteomc. 
comparing its state to that of " a comet vrithin a tnoDth ol 
perihelion." ^ Dr. Scheiner, on the other hand, r^ards it 
as virtually a cluster of sun-like stara* A ^)ec£rographic 
impression, wrung from it by meaoa of an exposoz^ of 
Y^ hours with a short-focus reflector in January 1899, flaemed 
to him a faint replica of the solar apectrum, marlnd br 
the familiar dusky ruliuga, but by no bright lineei Y«t 
Sir William and Lady Huggine hnve more tb&n ODce, 
observing the same spectrum visually, caught tmequivocil 
traces of the admixture in it of emissive with absocptire 
elements.* 

The real shape of the formation must be that of a disc. 
oval or inund^ but with a spheroidal mass both at the origin 
and extremity of the nebulous spires. Their convolutions, if 
actually circular, must lie in a plane inclined about 25* to the 
line of sight ;^ but we can only estimate their extent in space 
by making a precarious assumption as to their remotenesR 
Taking the distance of the nebula, for instance, to be of 
sixty-five light years, that already attributed, for illustrative 
purposes, to the cluster in Hercules, we find its radius to 
measure 162,000 times the radius of the earth's orbit; so that 
the frontier of this glimmering realm, as determined by Bond, 
is much more than half as remote from its centre as the 
nearest fixed star (a Centauri) is from ourselves I In travelling 
from end to end of it, light spends nearly six years ; and if it 

» Attroph. Jmtmatf foL iit. [i, 228, 

* Obaervaiory, vol. lii. p. fla. ' Aatr. Naeh, No. 3M9. 

* AUas o/StettaT SptdTa, p. 135. 

° Scheiner, Photographu d€r Ghaivnw, p. S32. 



THE GKEAT NEBUUE 



261 



obliquely towards uSj then onr view of the further margin 
ly be of an earlier date than our view of the hither margin 
a coaple of years or more. Extensive changes within the 
ehula might then manifest themselves to us successively, 
Ithough they had really occurred simultaneously ; while, con- 

ely, the coincidence in time of widespread variations would 
^ue a position of the nebula nearly square to the line of 
sight. That it is Btill in the pla&tic 8t&ge there can be little 
doubt; and the outbreak of the "new star*' of 1885 haa 
shown that the action of the powers engaged in ruoulding it to 
ite predestined shape may occasionally be attended by a cata- 
strophic liberation of energy. 

Two adjacent nebulae in Leo, both enrolled by Messier, 
have finally quitted the ambiguous position long occupied by 
them. They are elliptical spirals, analogues of the colossal 
structure in Andromeda. Photographed by von Gothard in 
1888, the one (M 65) showed a bright centre with four 
appendages resembling the sails of a windmill ; the other 
(M 66), a complex arrangement of envelopes partially sur- 
rounding a nucleus, somewhat like the paraboloidal veils flung 
round the head of a comet near perihelioiu These, howeverj 
were incomplete views. On Dr. Koberts'a plates/ Loth objects 
took shape as ovoid formations composed of closely winding 
luminous coils, "thick inlaid," in the case of M 66, the chief 
of the pair, with nebulotis condensations. 

Sir John Herschel frequently noticed an approach to 
&nnularity in the members of this class of nebulce, and added 
the remarkable comment that " as the condensation increases 
towards the middle the eUiptieity of the strata diminiahea" ' 
This, if verified, would imply their ovalness to be not a purely 
visual effect ; and the inference that the appearance of elonga- 
tion corresponds to its reality is supported by such critical 
circumstances as the situation of a pair of stars either at the 
fod of a nebulous ellipse (N,G.C, 6595), or at the extremities 
of its major axis (N.G.C 6648), 

The longitudinal clefts often visible in ray-shaped nebulas 
corroborate this opinion. For otherwise, why should they rim 
lengthwise rather than in any other direction ? If the bodies 

' Cdeatial Phoio^aphSt vol. ii, p. 76. 
* Cape Obtervalwnt, p, 22, 



262 THE SYSTEM OF THE STABS 

t\mj cbarocieriae were, in &ct, cticakr diaca, none o^ tbii? 
pbjsical featoroB ooold hAve anj relation to the appeiaacE 
they happen to a»ume bj projection. Such « nktaoa m 
eixtremely conapicuoua in an elliptical nebaU, ia tlks 
Centaur (K.G.C. 5123), divided along ite entiro lengUt br 
" a perfectly de&nit^ atruigfat cat, 40" broad.** A dafioBe 
nebulous streak runs between, and parallel to the habia. 
which are shaipl/ bounded on the aides facing each other, bot 
hazj on thoee averted. " The internal edges of thin tst 
problematic objeet," Sir John Herschel remarked, ** hftw - 
gleaming light like the moonlight touching (hd ootliBe la & 
tinutspaiencj." ^ It is by no meaoa a solitary exampla A 
long, narrow nebula in Leo (N.G.C. 3623) ia " split faalo two 
parallel iaj&" * A black chasm, into which the nucleiB pto- 
trades, eepatates a lucid from a faint atreak in Coma BerenieH* 
(N.G.C. 4565); and a nebula in Andromeda (N.G.CL 891). 
with " a chink in the middle and two stars," sappoeed bj Sir 
John Herachel to be " a thin, flat ring, of enormous dimeiUMU; 
seen very obliquely," and photographed under that aspect 
by Dr. Boberta in 1891/ may really belong, as waa indicated 
by the Farsonstown observations^ to the numerous cabegoi^ 
of cloven rayt. That their distinctive peculiarity dopenda 
upon some general conatnictive principle cannot readily be 
doubted, but we ehould vainly attempt to apecuIuLe upon ita 
nature. 

No descriptive formula is wide enough to include all the 
capricious forms of "irr^ular" nebulae. In regarding tboe 
singular structures, we seem to see surges and epray-flakea of 
a nebuloujs ocean bewitched into sudden immobility ; or a rack 
of tempest-driven clouds banging in the &ky, momentarily 
awaiting the transforming violence of a fresh onset. Some- 
times, continents of pale light are separated by narrow straits 
of comparative darkness ; elsewhere, obscure spaces are hemmed 
in by luminous inlets and channels. The "great looped 
nebula" (30 Doradds)^ one of the inmates of the greater 
Magellanic Cloud, resembles a atrip of cellular tissue. It 
serves not to conceal, but to ornament in a wide, openwork 




' Cape OfiiKraoiion*, pp. 20, 105. 
Trant, R. DuK Soe, vol. ii. p. 9C. * Ihid, p. 118^ 

'* Celestial PhotoffrapA^ voL I p. 41. 



1 



1 -■ 



p 



THE GREAT KEBUL^ 



263 



ftttem the ekj behind it, and was described by Sir John 

Herschel as ^' an aBsemblage of loops," the " complicated 
windings" of which constitute it "one of the moet extra- 
ordinary objeeta which the heavens present.'* ^ It gives, 
nevertheless^ a dense photographic image, Plate XV. reproduces 
an admirable picture obtained in two hours at the Royal 
Observatory, Cape of Good Hope, in which claws and tentacles 
of faintly lucent stuff are seen to be thrown abroad in all 
directions from the main trunk, itself extensively riddled with 
dark spaces.* From the Arequipa plates Professor Pickering 
judged this object to be the core of a spiral," probably embracing 
the whole contents of the Nubecula Major, and even to the eye 
it appears as in some sort their kernel. The " looped nebula " 
may then very well be the hub of ita own particular universe. 
Its spectrum is gaaeoui*, but with an unusually strong inter- 
mixture of continuous light. 

The efficiency of the camera in disclosing the intricacies 
of nebular structure is vividly exemplified in Plate XIV., which 
exhibits, eelf-portiayed, one of the many celestial marvels 
gathered together in the constellation Cygniis, The object 
(N.G.C. 6992), though very imperfectly characterised to 
visual observation, is marked in our picture (the original of 
which was taken by Mr. W. E. WilBon of Daramona, October 
7, 1899) by a surprising wealth of detaiL Its texture is 
perceived to be throughout delicately filamentous/ the fila- 
ments being, as it were, drawn out in the same general 
direction ; and the bright patches composing the nebula 
have besides, as Mr. Wilson noted, a parallel trend which 
coincides with that of the separate nebulous threads, A con- 
cordant play of forces is thus evidently brought to bear upon 
the entire formation, which, measuring from end to end 80' 
of arc,* must be of a vastness transcending imagination. 
There is, beeides, some reason to believe it organically related 
to an undulated nebular mass in its immediate neighbourhood 
(N.G.C. 6990).* 

1 Capg Oburvationtf p. 12. 

' Oore, World* of Space, p* 208, 

* ffaroard Amvsh, vol nri, p. 208, 

' Sw fetao Ritcheyi Astroph^ Joum. tuI. zir, p, 22B. 

" Roberta, C*I. Photifffrupha, vol, ii. p, 14&, 

* H. C. WilBon, Fop. AmIt. Oct. 1002. 



M4 



THB SYSTEM OF THE STABS 



But all other irreguUr nebubs oiikk into iiwigtufieutt 
compared with that ahown by an opaxs-glsaB as ft ffliv«ij 
patch round one of the atora in Orion's swonL Thii extzft- 
ordioarj object (M 42) has been under eSecdTe obesntm 
for 250 jeATi, and during the last cdghty haa baen aao»> 
graphed^ mapped, me^ured, figured, and photographed with i 
diligence worthy of its pre-eminence. Hence, fulare chai^ 
In it, ahould they take place, most be slight indeed to 
- detection. 

The multiple star Orionia might be called the foondAtii 
etone of the edifice. All the lines of ite architeotore an 
down with reference to it, iind the intimate physical 
tioa of the stare with the gaaeous stulf surroundiug them, 
been spectrographically demon at rated by Sir William 
Lady Huggins^ and amply ratified by Professors Frost 
Adams. 

Surrounding the trapezium is the brightest part d 
nehuUi called, from its ^t delineator, the " Hu; 
region," Over this epace, shaped like a right-angled 
the light 13 collected into " flocculent ma&aes" which, with 
best aoeing, prove to be throughout of a " hairy " 
The effect was compared by Sir John Herschel to 
" breakiog-up of a mackerel-fiky, when the clouds of whiolLl 
oonsista begin to assume a cirrous appearance," * and su, 
to Mr, Laseell* " larg« masses of cotton -wool packed 
behind another, the edges pulled out so as to be very film^ 
The " pea-green colour " of the whole object struck 
forcibly, but is apparent to moat obserTers as little more 
a greeuiah tinge. 

Emanations (or what seem such) from the Huygeniau 
core stretch away in wide curves to form the outlying porti 
of the uebula* One great effluence runs out into a " proboi 
attached to the upptT jaw of the nondescript creature liuiu< 
in unearthly radiance, on the eky. Another, represeni 
ing the lower jaw, bounds on the northern side the 
of the distended fauces (the " Sinus M^nus "), and ia sub- 

L sensibly connected — as if by a ahoal leading to an island — 



I 




Memoirs R Aitr, Soc, vol, ii. p. 4&1 ; 0, Stone, fu6(. Ltaiuier ^Corrnkk 
Obrarvatori/t vgl, L ^t. 7, p. 274. 
^ Ibid. voL xxiii. p. 64. 





iVii Irivt^iilui" X^buJii ill I'v^iuih. I'lmto^rujilicil by W. K. W'Jhon, KJl S, 




THE GREAT NEBULA 



266 



ith tha ''nebula minima" (M 43), a roonded mass vhieh 
^appears as if just drawing together into a star.'* ^ Behind, 

id between the two, misty efiTusions Bpread far afield; 
olved by O. P. Bond into an intricate fabric of convoluted 
ftnd branching filaments, the brighter region from which they 
spring diapkying a similar, but more compact mode of aggre- 
gation. "^ It IB now impoaeible/' he wrote after a particularly 
fine view of the nebula, on February 26, 1861, "to see it in 
any other aspect than as a maze of radiating, spiral-like 
wreatlia of nebulosity, or filamentous tentacles, the centre of 
the vortex being about the trapezium/' ^ And to Mr. Safford^ 
using the then recently constructed 18j-inch Clark equatoreal, 
it appeared as " an assemblage of curved wispa of luminous 
matter, which, branching outward from a common origin in 
the bright massea in the vicinity of the trapezium, sweep 
towaj^ a southerly direction, on either side of an axis passing 
through the apex of the Regio Huygeuiaaa." ' 

The appLcation of photography to this amazing object 
has not only supplied records of its actual condition in- 
definitely more authentic than any producible by the human 
hand, but has served to combine the peculiaritiea of its con- 
formation into a strikitiigly suggestive whole. The outcome 
can best be appreciated by reference to Fig. 34, exhibiting an 
" index diagram " prepared by Messrs. Eanyard and "Wesley 
from several negatives procui'ed by Dr. Roberta with varied 
exposures. Two points are made perfectly clear by it. First, 
that the whole fabric of the nebula is concave towards an 
axis " passing through the trapezium in a north-easterly and 
south-westerly direction."* Kext, that the effluences from the 
trapezium liave a predominant tendency to assume ramified, 
or tree-like forma. Thus the seemingly eruptive jet marked 
with the letter b mimics the shape of a stone-pine, and bears 
a less equivocal resemblance to the " stemmed " type of solar 
prominencee, aa well as to certain arboreal structures visible 
in some photographs of the solar corona. The latter analogy 
ia rendered still more apparent by comparing the vast nebulous 
out-growths, a and A, with the group of coronal rays repre- 

^ J. Hdsehe], Metrtoirt B, Attt. See. vol. li. p. 49.'>. 

' B^kTvard Annalt, vol. v, p, 168. * Ih%d. p. 159. 

* Raujard, Knawttdyt^ vol. lii, p, 147. 




THE SYSTEM OK THE STABS 



asnted in Fig. 35. The aanie kind of atmctoreB seeBa, ii M 
otsee. At ODce to spring npwafd ftad to carre i&wd. as if 
nnder the iuduence of a two-fold actioo — oatwaid fen a 
centre, and inward towards an axi& Tbia oi^ganic aboilint; 
— Gmt detected by Ur. Banyard — btt t wat u i the OrioD aabok 
and the lominoua appendages of the sun, is borne oat fef tk» 
fesemblance to apon glan of their minute texture. 

The limits of the greet " sword-handle ** nebula an obb* 




Flo. St. — Index tHagnm to fltmetum photogmpbed la the Ortcm HcbalE 
(gwioLHtffe, voL xlL p. 140). 

tinually being pushed further back, and there ia no reason to 
believe the process nearly terminated. On Mr, Kobert^fl 
plates/ firet the " nebula minima " waa joined on to the main 
body, then, with lengthened exposures, a cloudy mass to the 
north (N.G.C. 19 77) was reduced to its true position of an 
offset, by which the forma of the parent-body are pretty 
closely counterfeited.'' But even the combined object is far 

' MinUMtf ^^otica^ toL xliz, p. 290, 
» Ratiywd, tec, c«. p. 1*8. 




THE GREAT NEBULAE 



26r 



om representing the nebulous contents of this part of the sky. 
Br an area of 150 square degrees, iri which it IB nearly 
antral, twelve new nebulie were photographed at Harvard 
ollege in the spring of 1888, and ludicatioog were obtained 
it Sir William Herschel's surmise ' of the union into 
le immense stratum many degrees in length of the " great " 
gbula with othera lying north and south of it, might 
be veriSed in the immediate future.^ It was soon after- 
wards more than verified. On January 14, 1890, Profeeeor 
W. H. Pickering exposed a plate during 6^ 22"" at the 



■k 



k 



Fro. S5.— Qnrap of OqphuJ: &tniet(iT» (W. 11. WHler^ XiwvMgt, vtfL ili p. HT\ 

aummit of Mount Wilson, California, with a 2l-inch portrait- 
lens. A singular diaclosure resulted,' A great " snake " of 
diffuse nebulosity (aa Professor Wolf has since called it), 
issuing from the trapezium, was perceived to enclose the whole 
central part of the constellation with dim folds 15 degrees in 
diameter. Cosmical vortines on this gigantesque scale are 
beginning to appear leea rare phenomena than the strangeness of 
their aspect and implications might have led us to anticipate. 
They are discoverable only by means of short-focus photo- 
graphy. 

In the Orion nebula we perhaps see an undeveloped 

1 Phil. Tratu, vol. Ixxii. p. 2111. 
' Harvard Annais, Toi. inii. p. 117. ' Und, voL rxnL p. 36. 



268 



THE SYSTEM OF THE STAHS 



diiBter on tb» model of tbe rieiade&. Nearly & tbooduul ttan 
were catalogued b^ Bond Id the portion of it (3*36 aqiine 
d^rees in area) examined by Liio. and las list provw to be &j 
from exhaiutiva.^ Many of them ahoulJ, uideed, in Pcotoor 
Hotdeik'a opinion,' be accounted rather aa nebular oondeoi^ 
tiona than as true stars ; but that they will eventoaily gra* 
to he auch^ aa they slowly absorb the nebulona niaterial now 
enahrouding them, is a justifiable coojectureL Tbe aame pn>- 
oesa appears already far advanced in the PIeiades» with the 
reault that the stellar now altogether predominate orer the 
nebuhir element in the compound Bystem. It may not hai^ 
been always so. The balance inclined perhaps^ in tbe remote 
past, as decisively the other way aa it now does in the Orion 
nebula. Tbe variability of nearly a hundred of its in^njlved 
stars ia an additional fealiii^ of resemblance to compieased 
duatera. Noted in a few cases by Dr. fioberte in 1890/ it 
was rendered astonishingly prominent in 1904, by Mis 
Leavitt'a comparieona of records stored in the Harvard photo- 
graphic archivea.^ The discovery gains added aigmficaaoe 
from the fact that many, or moat of tbe twinkling crowd 
thus brought into evidence belong to the cluster- 1}*^ of 
variables.* 

It iB not easy, Sir John Herschel tdla na, '^ for laognage 
to convey a full impreasion of the beauty and sublimity of 
the spectacle offered by the Argo nebula^ when viewed in a 
sweep, ushered in as it is by so glorious and innumerahle a 
procession of stars, to which it forms a sort of climax.'** 
"Situated in one of thoee rich and brilliant ma&sea, a sue- 
oeaaion of which, curiously contrasted with dark adjaoant 
ipacea, constitute the Milky Way between Centaur and Argo' 
its branches with their included racuitiea cover more than a 
square degree, and are strewn by above twelve hundred stars 
These, unlike the components of the trapeEimn-cluBter, show 
no general tendency to vary in light The peculiarity from 
which the Argo formation derives its current title of the 

* Pftrkhurst Astrop\. Joum. toL ir. p, 130. 

* fFiuhinyltni Observatioiu, 1S7S, App. 1, p, 2S1, 

3 MiffUhly ^'^atuxs, yd]. 1. p. 818. 

* Attr. Koch. No- 3fl50 ; Bartard Oirculara^ Nob. 78» 7», B«. 

' J. S. HAgen, Jstroph, Jtmrn. vol. x\x. p. 3H4. 

* Cape ObKfvaiiontt p. 88, 






^iy^^-^ijj- 







'_•■*" 






I 








Phntogrnjih of tlie Key-Hole Nebula. 
Tiiki.>u Willi tlie Knicb 'Jl4-iuch Effractor at Antqttipft. 



THE GREAT NEBTILiE 



tej-hole " nebula, ia a large lemniscate-ehaped opening in 

central and brightest part, the blackneau of which ia 

Etlified only by the veiling of one corner by a strip of thin 

Bbulous haze, Fonr stars are placed precisely at the edgea, 
none perceptible to the eye within the vacuity; and the 
famous variable rj Carinae. lies close to its eaatern border. Such 
was the brilliancy of the etar In 18S8 that it ahnoBt obliterated 
the " key-hole/' now, and previously to the outburst, the 
individualising feature of what would otherwise seem a chaotic 
sea of luminous biUowa. Duplication emphasises the meaniDg 
with which it i& fraught. A second oval aperttue, completely 
dark, hut for the faint sparkle of four minute stars, occurs in the 
eouthera^ sparser part of the nebula. It might be compared 
to the echo of a catch-phrase. Plate XVII. reproduces a photo- 
graph of this nebula, taken by Professor Bailey at Arequipa, 
June 1, 1896, with an exposure of four hours. The light 
and shade being reversed, the most intensely actinic parta of 
the portrayed objeet are printed black, while its vacuities are 
left white. Thus, the "key-hole" appears as a crooked, 
flhadowless aperture in the moat strongly himiuous section of 
the inchoate structure, which is rendered in a manner bifid 
by a vast dusky channel running from south-eaat to north- 
west between two nebulous continents. The second " key- 
bole/' which ia larger than the first, and has some similarity 
to a kidney-bean, lies far to the south, in an outlying effusion. 
A very peciJiiir gaseous nebula in Sobieski's Shield (M 17) 
has revealed itself piecemeal. Messier, first of all, noticed, in 
1764, a apindle-shaped, starlesa ray, about 6' in lengths Sir 
William Herschel added an arch springing from its western 
extremity, and the combined object became known as the 
" Horseshoe" or " Omega" nebula, its form reaembhug that of 
the Greek capital letter n, with the left-hand base-line turned 
up obliquely,^ Again, it suggested a " Swan " to observers 
whose instruments were inadequate to show the complete 
arch, and Flammarion compared it to a smoke-drift, fantastic- 
ally wreathed by the wind. In the cleai' aii* of the Cape, Sir 
John Herschel detected a second very faint " horse-shoe/* 
attached to the opposite end of Messier's streak from the first,* 

1 J. Hom-hel, Phil. Tranf. vol. cJCiiiL p. Ml, 

'* Cape ObMrvaiiftiti, p. 10. 




Omega nebula^ obtained by Br. Roberta Id 1893. eahiUi 
Meesier's my ha the axis of &n oval maas covering an ajM IS* 
hy 12';* but the intimate secrBt of its orgaaiaatioa bn Mffl 
to be penetrated It ia iudeed situated too Car to the nott 
to be usefully inveetigated in these latitudes. 

The " America " nebula is apparent oiilj to th« obttBial 
retina. Some inditiAlioufl of ita existetice were caofht bj iIm 
cider Her^bel ; but its virtual diaoovery wae mAde pbsle* 
grapliically by Dr. Max Wolt On a plate exposed DmmIm 
12, 1890, the vast object portrayed in Plat« XYITL etoagdiot 
the Sret time into full view, and received ^ name deecriliEI9 
of its striking resemblance in &hape to the contitient of Kott 
America. Among the features attracting attetition oo ov 
plate are a amall, intently black hole piercing the neck of 
light that corresponds to the Isthmus of Panama, and 
nebulous atar, like an island in the Atlantic Ooean off 
coast of Kew Jersey. This is 57 Cygni, lately found to I 
spectroacopic binary in ewift circulation round an ot 
companion.' It yields a helium spectrum in con^spondeiM* 
with its nebulous relation ships. 

Ko logical distinction can be established between irreffo 
nebuhe and those indefinite tracts of milky radiance 
by the elder Herschel " diCfufled nebulosities." The total 
of fifty-two separate tracts perceived to be thua phosphorescent 
was estimated by him at 152 square degrees, and he added 
the judicious remark that *' the abundance of nebulous matter 
diBuasd through such an expansion of the heavens most 
exceed all imagination."* The few visual observations subse- 
quently made of these regions* intimate irregularities in their 
illumination doubtless dependent upon the slow advance of 
cosmic processes of large scope. Their investigation, bow- 
ever, ia properly the task of the camera. Already, some of 
Herschera " affected " regions have been proved, by the photo- 




' Sidgrtat JfeJKujj*?, t*l. iv. p. 38. ■ C*i. Photagrapkt, voi i p. lOL 

' Frost and AddiCDJ, Astrcph. Joum. toL xrii. p. 381. 

* Phil Tritns, vol cL p, 277- 

• Dreya^^ K J. S. Aatr. Oea. Jahrg. iiii p. 63 ; Littrowj SUrngntppm vmd 

^ehelmaam, p. 29 ; T«inp«l. Attr, ^a^. No. 2C11. 



THE GBEAT NEBULA 271 

iphs of Barnard/ Boberts/ and Wolf,* to indade strongly 
p-eloped nebulous formations ; the condition of the rest has 
LL to be ascertained. Prolonged exposures with appropriate 
truments will be needed to determine it satisfJEkctOTilj. 

' JMroph. Joum. vol. xvii. p. 77. 
• Monthly Natic$$, toL ixiiL pu 28. * Srid, ju 80S. 




BpKCULAnotrs aa to an identity of uuture between nebnbl 
u&id oometd are no novelty ; they pr^entod themsdvM, u 
they could hardly fail to do, to the mind of Sir William 
Herschel,^ and Sir Norman Lockyer sought to prorida Uieta 
with an experimeDtal b&sb bj his researchce oa meteqrites in 
1888-89.'' The results were scarcely encouraging. Indeed, UiC 
two classes of bodies have epectroecopically little in commoc 
Ko gaseous nebula gives a trace of the carbou-bandfi chancte^ 
ietic of comets; while the bright lines derived trom nebalft 
are absent from moat comets certainly, from all probably. 
Yet a physical analogy snbsiats, and is evidently charged with 
meaning. 

Both comets and nebulae consist of enormous volumes of 
gaeeoufl material, controlled by nuclear condensationSt whether 
of the same or of a different nature in the two genera we need 
not now stop to inquira. Both, tht^re is the strongest reason 
to believe, shine through the effects of electrical excitement 
In both, thei-e ate manifest signs of the working of repulsive, 
as well aa of attractive agencies. 

A telescopic comet is indistinguishable except by it9 
motion from the ordinary, centrally condensad, rotind nebola. 
which is itself indistinguishable from an exceedingly remote 
globular cluster. Superficial likenesses do not, it is true, 
count for much ; one object may counterfeit another without 
bearing any true relationship to it. What it ta really im- 
portant to note is the structural resemblance of nebahe to 
comets. The ports of a comet become differentiated exclusively 

> I^il, Trwtf. roL oi. p. 306. * Tht MUmnilic B^pathaiM, p. 2S8. 

373 



J 



PHE NATUBE AND GRANGES OF NEBULAE 273 



Icander solar InSuence. Hence their symmetrieal arrangement. 

^ms regards au axis paesing through the sun, is modified only 

tbj the orbital displacement of the body to which thej belong, 

l^nd from which they emanate. It is extremely curious to 

tifind that, notwithstanding the absence of these conditions, the 

features of certain nehulfe are impressed with a correaponJiDg, 

"though different kind of polarity. Definitely directed outflows 

(or what aeem to he such) are in them of frequent occurrence. 

Prom many of these objects, aa from many comets, only a 

' single stream of effueion is manifest ; and we then get " stars 

' with tails," which might well pass for miniatures of the 

' bearded travellers through oui^ constellations. Those effusions 

ftre, in some cases, boxmded by straight lines ; they are fan- 

' shaped ; in others, they are curved, often so strongly that the 

! " brush " is bent into a " coil" Such difierencea may be 

plausibly associated with the varied conditions of aiial 

rotation in the star-like nucleus. Wbere there is none, the 

isauing matter naturally proceeds straight outward ; its 

corvature depends upon its being left continually further 

I behind in the widening cirelea reached by it as it ascends 

I from an advancing point of efflux. 

It is easy to see that this process, if carried far enough, 
will result in the production of a helicoidal nebula. Fully 
developed spirals are, however, constructed oa a less simple 
principle. Their convolutions are duplex. Two opposing 
branches combine to form them. That these branches 
represent intakes of matter from space is hardly a tenable 
opinion. It has, in fact, been virtually abandoned owing to 
its inconsistency alike with the design and the details of the 
objects under consideration. The conviction is now dominant 
that they originated through some kind of repulsive action. 
Professor T. C Chamberlin regards them aa products of 
catastrophic disruption. Each is the wreck of a sun-like body, 
dispersed through the tidal influence of some similar orb to 
which it made a fatally close approach,^ They illustrate, 
accordingly, " the effects of explosive projection combined with 
conourreut rotation." M. Deslandrea adverts to the corre- 
spondence between the diametral wings of the solar corona 
id the opposed arms of a spiral nebula, the ramifications of 
1 Aitroph. Joum, voL xir. p. 31. 



274 



THE SYSTEM OF THE STABS 



which he hold^ to 1^ explicable on Br6dikhine*B ihtotj 
multiple comet&ry tails ; ^ wliile Mr. W. £. WUson w«ft stmdc 
vrith the " coinetary taila, curved like plumfcs away fram iKe 
central nuck'UB,'' attiichcd to the nodoaitiee beeCrewiog tq 
photograph of the Whirlpool in the Honting Boga' 

With very rare exceptions, xiel-iuli' are seen by na, nAt ' 
pUn, but in fxirapeetive. '' The only thing," Profeaeor 
wrote in 1889,^ "we really know about the form of a 
in general. Is that it is projected into a cert-Ain hIi 
problem is to €nd the true curves in space, knonin^ 
projected curves." His efforts to solve this arduous problem I 
actual trinls with helices of wire projected in all ima 
varieties of poaition, mtt with a certain mc-asure of buc 
The " type-cun'e " at which he fiually arrived, though 
perhap** really conformed to in nature, served, at any 
usefully to define ideaa regarding " the actual aituation of 
different branehea of a nebula in space of three dimensiooSi.^ 

Kebulfe of the spiral and " brush " kinds, are not the 
ones exhibiting cometary relation ships. A charact 
feature of nebulous trains and " wings," exempli Bed liotb' 
the Orion and Pleiades formations, is their va^e diffluioai 
one aide, but sharp terminatii>n on the olher^ just as in 
luminous vanes appended to the heads of cometa. The 
tinned ejection of matter agaiiist a counter-current of force' 
which it iH unceasingly driven backward, seems iudicatedj 
both cas«s. 

Planetary nebulae, too, imitate, in a fashion of their 
the heads of comets under energe tic solar action. Their 
nxultipla discs correspond most strikingly to th« multiple 
envelopes of comets, and intimate a similar origin through 
interior espansive or repulsive agencies. Only that in the 
absence of the dh-ective power of the sun, tlie waves of «id*- 
nation ^lead equally in uU directions, producing suoceflsive 
approximately globular, instead of parabolic surfaoea. Und 
tlie combined iuOuences of rotation and contractiou, 
shells might be expected eventually to subside into rinc 



* CompUf Jifiviffi*, Juno 23, 1902. 

* Aftr. ReArarcKtuit lktr<imr»nn., 
• PublUtUiana Attr. Soc. of the Pa 

* Roch«, itimniint tU t'Acad. dt4 StUtiut, W 



' p. 2«. 



(THE NATURE AOT5 CHANGES OF NEBULA 275 



but it would be rash to affirm that annular nebulaj, in poiut 
of fact, acquired in this way their present aspect 

No visible sigus of movement have, up to this, been 
ived from nebulae. In the ordinary sense they appear 
absolutely etationary. Accurate observationH of them are, 
indeed, comparatively recent; they go backj nevertheless, for 
enough to juHtil'y the statement that not one among about 
four hundred well-determined nebulae becomes progreaaively 
displaced by so much aa one second of are yearly. The 
nebula in the Pleiades ought perhaps to rank as exceptions to 
thia general immobility, since we need no direct proof to 
aesure us that they drift with the cluster of which they form 
an integral part- The drift, to be sure, is slow ; but it is 
aecnrely ascertained, and atforda grounds for the ijole estimate, 
that is not mere guess-work, of the distance of a nebulous 
syatem.^ The apparent indiffewnce of all other uebulie to 
the perspective effects of the eun'a swift advance through 
Space leaves little probability that any of tbem lie as near 
to us as the nearer stars ; and the probability is raised aknost 
to certainty by the apectroscopie discovery that many of thaiXL 
ar« animated by rapid individual movements. 

Thirteen planetary nebulffi measured by Profeasot Keeler 
in 1890,* proved to have an average radial velocity of 17 
miles a second. The swiftest among them is the helical 
structure in Draco, which ahowa an advancing speed of 
40 mile^ a second ; the " Saturn " planetiiry in Aquarius 
oXbo approaches the aun, though somewhat more slowly ; while 
a stellar nebula in Aquila (N.G.C. 6790) increases its distance 
by 30 miles a second, or by ten times the earth's orbital 
ladiufi in each year. As regards motion, then, no distinction 
can be drawn between stars and this kind of nebulae. Tlie 
■ irregular class has bL'cn investigated for spectral displacements 
only in its leading representative ; and the concordant results 
of several inquirers prove the Orion formation to be in course 
of withdrawal from the earth at the rate of about 11 miles a 
second.* But, since thia velocity must be largely, if not 
wholly, an effect of the sun's translation through space, the 

^ Se« ante, p. 106. 

* Liek PnUifaHim9, vol iii, pp, 217, 228. 

* Froit Aud Adama, Aitroph, Journ. vol. sii, p. 354, 



S74I 



THE SYSTEM OF THE STABS 



Xflbaift mftjr in bet be «lroaet st rest in the sldlar 



Indkabom of poanble marvow&tB wifchia it i 
Dr. Vogel in 1902 Irom spectrogisiDB of its 



1 ikntsd bf 

HMft ptftS,' 

and the doe will periiapa eretitqAll/ be followed op to ft definite 
mne. But no ciebiila has jet been pfoved, by line of s^X 
memmaim, to loUte oa an axis. Nor are double nebdc 
pooeptibtj circnlAting. Tho ajsUmn tbey ptesiunably ood- 
stKtnte rem&in ngidl/ fixed. The oontz&rf baa often been 
MBerted ; jet revelations, alleged on the sfaength of inoxsci 
obeervakiona, and brought to a rtaindiitili bj precise ooee, 
nuut plaintj be diamiesed ae tUoaoiy ; " unlees," aa Dr. Ih^-^ 
wsjM,' " we are to believe tbat nebulas in the good old d^r^ 
moved about an tbey liked^ but bare boon on their good 
behaviour dnoe 1861 and kept <)aiet'' The existence, 
frequently obaerTed, of a oebulous connection between gnmped 
objectiL intimates a atat« of things h&rdlj oompatihie iHlb 
mntnal drcnUttou, The relations of theee imper&otlj 
separated individuab ^re perhaps in ft state of tr&osition, likl 
those of multiple comi^tarj masses, at times enclosed togetbs 
like double nebulse^ in a dimlj Uimlnous aheath. 

The idea of the accompauioient of pUnetarj nebuUe bj 
satellites, was snggested to Sir John Herschel bj the Crequeut 
and clc^e proximity to such objects of minute stars/ and be 
recommended their careful micrometrical measurement as a 
criterion of poeeible future changes But d' Arrest found thA 
attendant stars just in the positions Herschel had assigned to 
them/ and only one case of suspected displacement has sinee 
attracted serious attention. Professor Barnard noticed a minute 
discrepancy between his measures in IS 99 of a twelflti- 
magnitude star closely following the ring nebula in Lyra 
and those made by Burnham eight years previously.* He 
inclined, nevertheless, to attribute the alteration (if real) to a 
shifting of the point of reference rather than to movement in 
the star* Now the point of reference having been the nebular 
Qucleus, the verifioation of his surmifie would involve the 
detection of proper motion in a class of objects heretofore 

1 Sitiatit'jaberKhU, Beirlui. Marcb 13, 1903. 

" Jfmifilif Kvtieet, vol. ilvii. p. 418. ' PAiL Tnmt. roL cxuU. p. BOO. 

* Leif&ig Abhandiungtn, IM. iii. p, 30S. 

" Mimthly Noiictt, vol. li. p. a<& ; cf. Lcrnvflmrorth, ibid, vol, bci. p. :25. 



THE NATUEE AKD CHAISTGES OF NEBULA 277 



imperturbftbly fixed. The lapse of a very few years will 
render practicable the decision of this interesting question. 

Variability in light is a quality of nebulie && sui'ely as of 
etarSj although the cases are rare in which it can be eetablislied 
with certainty. Nebulae are peculiarly seneitive to atmo- 
spheric influences. Their finer detailsj always hovering on the 
verge of viaibility^ are completely shrouded by the lightest 
mist. Hence, even to the same eye, and with the same instru- 
ment, the aspect of the same nebula often varies greatly from 
night to night, and since personality is nowhere stronger than 
in the perception of the delicate luminous gradations delineat- 
ing to our aighfc the forms of nebulfe, a difference of observers 
adds a further incalculable element of uncertaJBty. Rumours 
of change then easily arise, hut are with difficulty sub- 
stantiated. 

Theystart.neverthelesSjfrom a presumption not unwarranted 
by facts. The occLurence of luminous fluctuations in nebulae 
has been proved by the complete extinction of one^ and the 
fading to evanescence of a second in the same vicinity. On 
October 11, 1852, Mr. Hind discovered, near the group of the 
Hyades in Taurus, a small round nebula (N.G.C. 1555) with 
slight central con den so lion. It was then very faint, but 
brightened steadily until 1856, when d'Arrest ranked it as 
belonging to the first, although verging towards the second 
class of brilliancy.* His amazement then was extreme to 
find on October Z, 1861, iU place apparently vacant! Some 
glimmer of its light was indeed made out for a year or two 
longer with the Pulkowa 15-inch refractor; but that too 
waned, and the object remained invisible for above a 
quarter of a century. Then, in 1890, it was just barely 
descried by Professor Barnard with the SS-inch Lick telescope; 
and even that Buperb instrument failed to show it in September 
1895. In 1897 it was not to be eeeu on any terms; in 1898» 
it barely stained the field of the 40-inch Yerkes refractor ; 
but Professor Keeler managed in 1899 to secure impressions 
of its somewhat complex fonn on long - exposed Crosaley 
plates.^ The apparition^ so far as can be judged, was a 
strictly temporary one* Hind's notice probably did not 

^ Attr. NfKh. Noa. \m^, T689 ; Auwera, ihid. No. 1393. 
' ProhitTTis in ABirophyticSi p. 52^ (Gierke), 



f8 THE SYSTEM OF THE STABS 

lag far behind its first perceptibility with modente insbKH 
mentB, 

A curious feature of the occunenoe was the sjmpathetk, 
or at any rate simultaDeoos, decay in light of a bhiaII star- 
since known as " T Tauri "' — placed almost in contact with tk 
nebula. The atar, however, recovered in 1868 about the same 
time that a second new nebnla (N.G.C. 1654) came into riev. 
First discerned by O. Strave, it was observed by d'Arrest. wbo 
was fully convinced of its novelty, and bis opinion was borne 
out by its subsequent total disappearance,^ 

The ligbtr^jhanges of nebuhe do not offer the same diversity 
as those of stars. Only two kinds of variability — those 
producing respectively ephemeral appearances and capricious 
brightening^ and fadings — axe represented among them. No 
periodical nebulae have yet been shown to exist. Tlie influenceSv 
of whatever nature^ bringing about the rhythmical pulsations 
of stellar light» would seem to be absent from the nebular 
kingdom. A distinotioD, however, peculiar to themselves, can 
be established among Tariable nebulae. Their fluctuations 
m^j be either general or partiaL They may aGTect the whole 
of a moderately compact object, or certain sections of aa 
extensive formation. Examples of both kinds^ and of all 
degrees of authenticity, abound; but we will only mention a 
very few, in which the reality of change seems scaroely 
disputable. 

One such is afforded by en elliptical nebula in Leo (N.G.Cl 
3666), "very bright" when discovered by the elder Herscbd 
ia 1734, but noted by hie son as abnormally faint for the 
first class. Subsequently observed altemationa have made it 
all but certain that the discrepancy indicated genuine change.' 
A nebida in Cetus, too {N,G.C. 955), is evidently subject to 
similar vicissitudea SchdnfeM in 1&61, and Vogel in 1865, 
failed to see it, although it was, at sundry other epochs, easily 
visible to the former observer, aa well as to d' Arrest and 
Winnecke, and fully justified in 1887, I>r. Dreyer considered, 
Herschel's ascription of it to the second order of brightness.' 

' Dreyer, Mimairt Jt. A»ir, Soi. voL xlii. p. 211, 

* Winnecke, Aatr. Hath. No. 2283 ; Dreyer, XemnTt M. AHf. Sae. roL klix. 
p. 218. 

■ Winnecki, Monthly Noiic£t, roL xnTiiL p, IM ; D«yWr toe. eit. p. 21S. 




THE NATURE AND CHANGES OF NEBULAE 279 



iham found it without difficulty in 1891, and attributes 
"its occasional evasivenes8 to '' uafaYourable atmospheric con- 
ditions."' Yet they aeem hardly adequate to explain the 
marked anoinali&a of its record. 

Intrinsic change indubitably affects a amall nebula 
attached in fan-shape to the variable star R Coronae Australis. 
Suspected by Schmidt in 1865, it has been rendered patent 
by Mr, Innes's observations.* When further materials are 
available it will be of ^n^at interest to determine the relations 
in light-change of these diasimilar, though nasoeiated objenta, 

Again^ one of a group of nebulce in Virgo, observed by 
Schmidt at Athens in 1862 (N.G.C. 5655), could not be found 
by d'Ar«st in 1855, two minute stars appearing as its loctim 
tenentes? If, as seema probable, the identical object was 
inserted by Hersehel, from an observation of December 28, 
1785^ as No. 498 of hia second class, its re-emergence to view 
may at any time he looked for, Tlie collection to which it 
belongs^ were judged by d'Arrcat (no doubt tightly) to be the 
brightest '* knots " of a wide - spreading nebulous structure 
(M 49). The vai-iability of one of them approximates to 
the local changes of irregular nebulae, exemplified with most 
certainty in the virtual efFacement of the " swan *' eection of 
the formation in Argo. Prominent in Sir John Herschera 
drawing* it waa missed visually by Mr H. G. Russell in 1871, 
and is very feebly represented in the best modern photographs. 
A bare mottling of the lacuna to the south-east of the " key- 
lole " in Plate XVIL now marks the place where the nebulous 
Ffiwan " formerly paraded its radiant plumage. 

No change so striking haa, within two centuries, occurred 
in the Orion nebula. Its organiaation appears, on the whole, 
to be wonderfully stable ; for certain alleged light- fluctuations 
of a partial kind * have not been thoroughly substantia ted,^ 
Those believed to affect the Omega and Tritid nebulae, seem to 
have resulted in a modification of what we may call caast-lineef 

^ Lick Publicntions, vol. ii, p. 172. 

* Annals Cape Ohtervaii>r\i, vol. ii. p. 134 B. 
' D" Arrest, Attr. NacK No. 1620. 

* 0, StruiTB, V, J, 3. Aitr, Qe§. Jnhrg. lii. p, 35 ; M^fanget Math. t. ii. p. 
BO ; d'Arreat. Astr. NaeK. No. 1366 ; H<ilden, fVask. Obxrvaiiifnt, 1S7B, 
kpp- i- PP- 12If 23^. 

* O. 3tone, Publ. Lwandtr WCormick Olaerv. vol. i. Plate VII, p. 271. 



280 



THE SYSTEM OF THE STABS 



here advancing^ there encroached upon hj the oea of daikfiMi 
which eiuToundfl them. In the latter oliject, a smgalu 
apparent alteration in t^e relatiTe places of the multiple star 
and the nebulous n^assea involving it, i& perhaps dne to Ibifi 
instability of outlines. Sir John Herschel ia 1827 and 1833 
described the star as located " exactly in the central Tacuitjr " 
of tJie nebula^ and just at the point of convergence of tbe 
three rifts dividing it throughout.^ But a drawing made 
by him at tbe Cape in August 1335 exhibits the star, no 
longer aa central, but as adhering to the eaet^n tnaes of 
nebulosity, A similar state of things waa noted hy IAmod 
and Smith in 1839,^ and subaista so obviously at the pmenl 
time, aa to render a mistake about it inconceivable. Th« 
implied change, however, must have taken place abruptly, 
between 1833 and 1S35, aJid then cea«ed ; so that proper 
motion, either of the star or nebuU, had certainly nothing to 
do with it,' There seems no alternative but to admit that 
the frontier4iues between luminosity and obscurity were, at 
the epoch in question, very considerably "rectiiied." 

Speculation regarding the uature of white nebulae must 
be left in abeyance until their spectra have been satisfactorily 
investigated ; but there is a consensus of opinion that those 
aho wing a discontinuous radiance are lum inous through 
electrical excitement. The fact of their incoroj^rable tenuity 
was established by Mr. Ranyard in 1892, on the ground of 
their ineffectiveness in imparting motion to bodies in their 
vicinity; and highly rarefied gases in Rpaoe cannot be hoi 
Gaseous nebula ate, in fact, reasonably believed to be at a 
temperature not much above absolute zero. They are not, 
then, incandescent, but rather " luminescent " ; their light is 
independent of thermal conditions. The phospliorescencie 
produced in a Crookea's vacuum-tube exemplifies, according to 
M. Dealandres, their mode of illuminatioa.'* Cathode-rays 
emanating from central condensations set their materials aglow. 
M. Arrhenius, on the other hand, derives from exterior 



^ Mtmoirt &. Aatr, Soc vol. iii. p. 43 ; Phil, Trans, vol. cxxiiL p. 400 ; 
Hold^u, A7n«r, Jaurn. of Sf:ii:ncij vol. »iv, ji. 434 (1377). 
* Tram. Apier. Phil. Sac vol. vij. p. 175. 
' Dreyor, Manthly Notices, toI. xlvii. p. 419, 
*■ Comptet Hfiidui, Ma; 20, Judb 23, 1002. 



J 



THE NATURE AJSD CHANGES OF NEBULA 281 



acies the light -stimulus acting upon them. There are 
ay indications, he tells us/ that space is pervaded Ly 
itively charged particles expelled with enonnons velocity 
the stars. These being absorbed by nebulse, occasion 
ctxical discharges through the gaseous volumes composing 
lem, their frigid condition notwithstanding; and Professor 
Dewar's experiments assure us that excessive cold is no bar 
to light-emiseioa. Moreover, none that is truly continuous 
seems to be included in nebular radiations. Wliat passed 
for such baa been analysed by Mr. Palmer into ranges of 
faint, superposed bright lines,^ Associated stars, however, 
yield genuine etellai- spectra, and have perhaps taken an 
active part in the weaving of the textures they begem. The 
unmistakable analogy, at any rate, between solar-coronal and 
cometary forma on the one side, and nebular forms on the 
other, indicates for all a kindred origin in the play of 
opposing forces, generated by certain foci of condensation, one 
of which is our sun, while the others can be safely designated 
only as nuclear points. Where there is only one such nucleus, 
the enveloping gases assume a simple globular or oval shape ; 
where there are many, the result is exceedingly complex. 
Irregular nebul£e are thus most likely potential star-clusters ; 
they consist of a stellar framework, draped with nebulous 
folds, spirals, and festoons, disposed along lines of force laid 
down by the rival or concurrent energies of the compact 
masses which it is permissible to regard as inchoate suns. 

* Zthrbudi dtr Kosmischen Ph^aik, Bd. i. p. -la. 

' Lick BulUtin, No. 35. loeq^UiOlitiea of the kind were long ago ftunpected 
by Sir WilliAiEL and I^y HuggiiUL 




THS DI3TA>'CE3 OF THS STASS 

Ths most arduous among the problems of BieQu' artroDOiDf 

was, filnguUrlj enough, the first to be attacked. Tt 
attacked, indeed, before the possibiiiiy was eren remofeely di»- 
cemed that Htollor astroDomjr might come to be regarded as i 
Bubfitantive branch of science^ In the hope, not af penetntiag 
the mocratable oecjeta oi the remote sphere of the fixed staz^ 
but of solving doubts about the motion of the earth, Copemiea^ 
Tycho, and Galileo led the way in the long series of experi- 
ments on the appisrent displacements of tJie stars reBottiiig 
from oar own annual travelluig round the son. Tlie interest 
of the question whetlier such displaceinents existed or not, was 
for them of a wholly '* parochial" kind; it lay in the test 
they afforded as to the reality of the terrestrial reYolutioii& 
Should the stars be found to shift ever so little by the effect of 
perspective, then the heliocentric theory could no longer be 
gainsaid ; if. on the contrary, they ignored sublunary ciroling^ 
the " pill " (as Kepler termed it) to be swallowed by Coper- 
nicaos was Indeed a huge one. For the distances to which 
the &xed stars had^ in that case, to be relegated, seemed in 
those times monstrous and incredible; and monstrous and 
incredible they would appear still, were we not fortsed by 
iirecuaable evidence to believe in them. 

Throughout the history of these inquiries, at least in its 
earlier part, it may be taken almost as an axiom that the 
largeat ostensible parallaxes were obtained by the worst 
means. With each successive improvement in methods and 
instruments, as the limits of possible error shrank, the dis- 
placements apparently measured dwindled, and the stars 

282 



THE DISTANCES OF THE STARS 



283 



ae less accessible to attempted determinations. DuriDg 
Bme tliree centuries, the ill-success of an aatronomer in thie 
atter wa3 a measure of bis akiil and judgment. Eesulta 
bbtaiaed with euapicious facility by inexpert obBervere utterly 
raded the guarded scrutiny of such men as Tycho Brahe, 
S'tadley, and Pond. Flamsteed, indeed, just at the close of 
le seventeenth century, detected in the pole-star annual 
itioaa which were certainly Twt illusory. Yet here too 
leie was a caveat. Theory and fact did not correspond. 

Let ua conBider for a moment what must be the viaual 
Beets upon very distant objects of the comprehensive and 
nceasing rounds of the planet upon which we are borne aa 
%toTs. Unmistakably, to begin with, we see them in 
dififerent directions at different times of year. In Jiinuary and 
July, in March and September, and bo on, we are at opposite 
ends of baae-lines 186 millions of miles in length. The stars 
then must be continually thrown, now a little to one side, now 
to the other, of the true, or " mean " places which they would 
severally occupy if viewed from the immobile sun. In other 
words, each describes round its mean place in a period of a year, 
a small apparent orbit, which is nothing else than the orbit of 
the earth projected in miniature on the sky. For stars flituated 
in the ecHptic^ — -that is, in the plane of the earth's motion^ 
thifl orbit contracts into a right line, along which the star merely 
swings to and fro ; for stars near the pole of the ecliptic, the per- 
spective orbit is sensibly a circle ; while intermediate latitudes 
afford all degrees of foreshortening. Every star — unless those 
few lying close to the pole of the ecliptic^haa thus its epochs 
of maximum parallax, six months apart^ when it seems to stand 
alternately at opposite extremities of the major aits of the par- 
allactic ellipse, and it is then that measures of its apparent 
displacements can be most advantageously made. These 
opportune seasons occur when the earth's longitude falls short 
of, or exceeds by ninety degrees the longitude of the star. They 
are accordingly different for stars with different longitudes. 

The precise /orm of displacement due to the earth's revolu- 
tion round the sun is thus strictly calculable for each individual 
star ; the am4}'unt alone cannot be predicted, but must be obtained 
by observation ; and from tliia amount the distance of the star 
is deduced, Por each pamltactic orbit is a perfect model, both 



284 



THE SYSTEM OF THE STABS 



in slukpe and nxes, of the e&rth's orbit as it would be nen froal 
the iter, abridgiaeiit of compasB (down ta oontnotiOB into « 
vbtttal point) ooKreaponding to a more and more profonnd i»' 

mermon of the point of survey in the abywoB of apaeo. 

'Hve patalUx of a star is then the dlfferenea b e ftw em 
pofiitiooe aa seen horn either side, and from the centre of tte 
earth's orbit. U is, in short, the angle subtended, at tbe 
diAtaoce of tl^iat particnlar star, by the mean interral banran 
earth and sou. Kow we can tell in a moment how &r off i 
spectator most be to eee a line ninetj-three millions of 
in length diminished to the angular dimension of, let u» 
one second. He must be 206,265 x 93 millions of mila 
distant But no star has yet been foond so n^r to as as tiu 
Tliat is to say, the shift of no known star amounts to as mneh 
aa the width of a sixpence held up at Charing Croes to 
spectator at Stanhope Gate or at Millbank. 

We are now in a position to understand why it was 
Flatnsteed'a obeervatioDB of the ap]>arent diaplacetnenta 
Polaris could not, when critically examined, be set down to 
the account of parallar. The Btar seemed indeed to deecdbt* 
regularly each year, a little ellipse of exactly the right tlu^ 
and as to ite sii^, there waa no d priori reason why the pole- 
star should not hare a parallax of upwards of twenty seoonda 
But there waa one irreconcilable discrepancy. The displaoe- 
menta noted occorred at wrong times. Had they been of t 
jiamlkctic nature, the position of the star in its minttte 
fictitious orbit should have been invariably ninety d^reea in 
advance of what it actually was. They were not then due to 
parallax ; but obtained their proper explanation from Bradley's 
discovery of the aberration of light in 1729. 

During the ensuing century and a quarter, the only valid 
results obtained in thia direction consisted in demonstrations, 
renewed and enforced from time to time aa more conclusive 
evidence presented itself, that with the instrumental means 
then available, stellar parallax waa an inappreciable quantity. 
Bradley showed that it must fall short of half a second, and 
although his reasoning applied strictly to only a limited number 
of stars, it was rendered at once more general and more cogent 
by the investigations of Pond and Airy, of Struve and BeaseL 
It thus seemed that astronomers should content themiselves 



^ 



THE DISTANCES OF THE STARS 



285 



with the knowledge that the etara were exorbitantly remote, so 
remote that tight spent at least four or five years in travelling 
to us from the brightest of them, and might, for anything that 
appeared, need an indefinitely longer time for the journey. 
The labours and refinements of two centuries had issued in 
fixing a loyjeT limit for distance, an upper limit for parallax ; 
in isolating the sun from his compeeia by setting between 
bim and them an unmeaBured stretch of desert space ; in 
widening to a startling extent the boundaries of the visible 
universe. Kepler's "mighty bolus" had to be swallowed in 
its entirety. 

At laat, in 1827. Savary of Paria brought forward a 
method for fixing an upper limit to the distances, a lower 
to the parallaxes of binary stars moving in known orbits. 
The further off from ua such orbits are, the greater of course 
their real si^e* and the longer the time taken by light to 
cross them. Hence, the deviations of the moving stars from 
their true places due to inequalities in light -transmission 
must increase with their remoteness, and thus serve in theory 
to determine the distance from the earth of the pair. Or if 
no such deviations are apparent^ it should at least be possible 
to fix an amount which tliey could not exceed without becom- 
ing so. Savary, accordingly, professed to demonstrate in this 
way that f Ursa Majoris, the couple most favourably situated 
for the purposes of the inquiry, must have a parallax exceeding 
^^ of a second * — must, that is, be at a less distance than 
would be traversed by light in 1000 years. But the informa- 
tion, however credible in itself, was not fully authenticated. 
Villarceau showed, in 1878,^ that the method was inapplicable 
except to stars of known relative masses ; and it was by that 
time already obsolete. The light-equation in stellar orbits ia 
too small to be extricated from errors of computation and 

I observation, unless where the occurreaoe of eclipses sharply 

I defines positions and epochs. 

^K Besides, the end in view can now be compassed by 

^Bvpectroscopic means. Measures of radial velocity in an orbit 
of known inclination give at once its linear dimensions, and 
eomparison with micrometrical data, its distance from the 

* Co-nn. dts Temps, 1830, p. IflS. 
• Ibid. 187a, p. fig. 



THK SreTEM OF THE STAJtS 



> deemed nwiMwry, ih—hiI lanfe^d^ i-niimliiii , T1» < 
wctc of tbe dilfeiotial kind, and lfa» i»«™^«t»fc 

, of Ibe e«rT«B into whkh thej wete proiectod by Me. 

[fnnptod tlw ooorktioo Ihftt ben at lart wba a st^kr 
tte gtnviaaMM of wfakh wu beyood cxTil The 

I impartaooe of ita dftectioa, proDOonoed b jt Sir John 
Ibe grooteot tmoiph et«r ochkrTed faj pnctinl 

I eoa be wlimolnit firara Bendji deeUzmtMm that, otittl tl «i 

1 001110117 ooppoowd. be was OBaUo lo fona an opinicn 4s 1 
wbeiber fcbo pocoUasooof the neozoofcatezB ohonki be 
hj tenthi, or by thooootidUw of a secood 1 ' 

The difltaoce from the earth of 6 1 Cxgni has been i 
freqoeQtlf laTcatigated thaD that of anj other otar, and Km 
Infling diflciepaziciBa notwithstanding, may bo coiieadeied M 
OitiifactorUy oacertained Beeeel'a p&ralUx of aboat a Hasi 
of a oecood was aogmented to 0*'42 bj Aawers's 
in IS66 of the o&me dau, and to 0**47 hy Sir fiobert 
raeosurGO at Dunslak.' Dr. HertnAati Davia's diacnanm rf 
the Butherford platee in 1898 yielded a Talne of 0'*36;' 
Profeaeor Bamard'a obeervationa in 1900-2 with the Teriai 
re&actor proved closeljr coa^ta&totj} This nnobtnuaTc etar- 
pair may then be contidentl}^ located a( a distance of 
li^bt-years from the earth/ 

We aooordingl/ aee the coupled stars, not where ihej on, 
but whore thej wfrc nine years ago ; that ie (mnoe tbor 
proper motion is about b'*'l2 yearly) just forty-sijc aeoondfl of 
arc behind their true places. The effulgent pointe terreetria% 
determined are then mere sinmlacra of the real stars; they 
pursue, without ever overtakiiig them ; they would continue to 
shine and to travel for nine years after their originals had been 
blotted out of the visible creation. Our views of all moving 
obJButs are of course to some extent affected by this curious 
kind of Ugh t -aberration j but in the sidereal heavens it 
attaint* proportions that are not only large, but, for the most 
purl, incalculably large. Our survey of the background 

*■ MejnoirB R, Awir. Sx. yq\. xii, p. 42^ 

» A9tT, Nath. No. 386. 

■ Abhandl%n^tn Sim, Akad. Berlin, laGS, p. lU. 

* CtftuTnUa Uniwraity ContribittioTiSf No. 13, p. \3&. 

• ititport rcrkei Observatory, 187fl-I902, p. 18. 

* BftrreU, Jttrt^h. Journ. toL xriii. p. 398. 



THE DISTAJS-CES OF THE STARS 



289 



tbe sky ma.j lag ce&tmiea, even miUeuniums, behind om 
aimultaneoua survey of its fot«ground ; and the disturbed 
^fuchroaoua relatioDa between the varied luminous coutenta 
the sphere are, to our perception, incapable of readjustment. 

Transported to the place of 61 Cygni, our auu would 

ar about eighteen times brighter. It would represent 
a star not of the fifth, but of the second magnitude, such ae 
Polaris, or one of the Pointers. Nor ia it likely that the 
Swan binary is massive much beyond the proportion of its 
luminosity. The extreme slowness of its revolutions, on the 
contrary, intimates a comparatively slight power of mutual 
Attraction. 

Fraunhofer'a construction of the instrument with which 
BdBsel observed 61 Cygni, marked the turning-point from 
failure to success in parallactic inquiries. The heliometer is 
apecially adapted to facilitate them. It has two chief points 
of superiority over the ordinary equatoreal and micrometer. 
In the first place, much wider pairs of stars can be grasped 
with it, its compass being, by the mobility of the semi-lenaesj 
extended far beyond the limits of a single field of view. The 
selection of comparison-stars is thus greatly enlarged, and the 
qhanoes of a systemic connection with the central star fatal to 
the purpose of the designed operation, are reduced to a mini- 
mum. In the next place, the stars under observation can be 
visually equalised by placing a wire -gauze screen of any 
desirable opacity over the segment of the object-glass forming 
the image of the brighter one, whereby batSing personal errors 
are completely eliminated. These are the chief, but not the 
only features of the heliometer tending to promote critical 
precision ; and not the least among Sir David Gill's services 
to astronomy was his development of the powers and upplica- 
tioua of this imique instrmnent- 

Bessere success with 61 Cygni gave the impulse to 
numerous undertakings of the same kind. Their result 
depended mainly on the skill or luck of the observers in 
picking out from innumerable indefinitely remote stars the 
few near enough to be sensibly displaced through the efiFecta 
of the earth's motion. Two circumBtancea maitily determined 
their choice. 

That distance is a factor of stellar brightnesd is so obvious 

19 




be ipeJL D M cd m tnoBB. Admic;. . 
t nmfft id vvfiet^r u actual light-povBr, v 
HMJKorf mSSk mmum tkal the mart lostrcNM ottyects inQ : 
foood Bttoog tboM m doeest proximity to (be earth. Kiccp- 
Jimtm^abooBd; bat^ one vide even^ kbe tfaeonCioU 
iavctve ntio lail wiwe rtkttaffr «2>d the aqaerB not o( toul 
ligbt fonuafaeB at leeet a flaeble gude to aotoal £ict' 

Sewal l UMna i' iaw ilus^^Vcga, Arctorus, Ckpdk a 
Cygnti and Pofaav— wer^ <n thk gnNmd, fixed upon t-i 
iawitigrtinm I37 & A. F. Pecera in 1842-43.' Hk n^i 
obtaimed with the PuIkQwa vertical dicle, were absolote, bt: 
imapeetiTe of oompadfioiu with other stars ; bat the deduc- . 
paraUazca were «> small. &nd their " probable erron " m xAf 
tlydy Urge, tb&t it woa di£Gcult to place much cxnifidence ift 
IbeiD. Vet LhejT came «urprifiiiigly near the truth. 

A Becond ciiterioii of neameei vas found in the appear- 
anae of mpid tnobion. Thia varies m the same proportioii tf 
diitiiioe. but in the revenB aense. At twice the dist&Dce, u 
ideotical velocity produces only half the angular disphu:eineiit, 
at tbroa times the diBtonoe, one'third, and so on. Tbuti, 
apparent Bwiftnees^ oo less than apparent lu^re^ depeuds in 
part upon vicinity, and the largest proper motions must beJcQgi 
on the whole, to the nearest stars. 

Aud, on the whole, parallax -hunters taking mpiditj d 
advance for their guide have prospered the best. A 7*5 
umgiiitude stair in Ursa Major, flitting annually over 4} 
necoudfl of angular space, was found by Winue<:ite in 1858* 
to liave a parallax ( - O^^S) inferior only, among those as 
yot determined, to that of a Centauri. This insignificsut 
object, numbered 21,185 in Lalande's great catalogue, is 
separated from the earth by a light-journey of seven years, 
ttud to thtit extent our observations of it are retarded. So 
tlmt it ia in reality always 33" in advance of the place we 
ore Gomi>eUed lo o^ign to it For a body claiming the tank 
of a Buu, it is either very small or very obscure. Our own 
ruling orb ia 260 times more luminous. 

' Btruvtt Jftm^ ^icrom. p. clxii. 

* XtUttkH/t fv^r pvp. Mitth. Sd. iii, p. 104 ; Mimoirtt, St Fitanbourg, 
t> Til. ^ 1*0, I&53 i A$tr. Kadi. No. 1147. 
■ AitT. Kadi. No. 1147. 



t 



THE DISTANCES OF THE STAKS 



291 



An 8^5 magmtude star in the eame constellation (Lalande 
Jl,258), also dtstinguifihed for apparent velocity, discloaed to 
Luwers'a measuremente with the Kbnigsberg heliometer in 
1860-62 a parallax of 0"*26,' corTesponding to a ligbt-journey 
12^ years, and a permanent displacement on the sphere, 
le to its proper motion in that interval, of 55". The real 
lirilliaDcj of the star is only ^^ that of the Bun. A sbill 
Jer 8tar in Draco (Oeltzen 17»415) gave au even more 
aphatlc warrant to confidence in ewiftnees, rather than in 
fcre, ag a certificate of proximity. Kriiger, induced by ita 
^eEirly movement of 1""27 to subject it to experiment, obtained 
parallax of one quarter of a eecond j ^ while the fine binary 
fstem, 70 Ophiuehi, with an annual motion of 1""13, proved to 
removed from tha earth by twenty years of light travel 
lax 0"'16).^ And all these results seemed, from the 
illness of their '* probable errors," to be exceedingly tniat- 
rorthy. 

The probable error of any result, however, repreeenta only 
phat we may call the uncaiLsed inaccuracies of the observationa 
upon which it is founded. It sums up» according to the 
doctrine of 'probabilities^ the effect of their deviations, in 
either directioDj, from the mean. But it takes no heed of 
"systematic" errors due to causes working steadily in one 
sense, but, so to speak, underground. These are the real 
eourcea of mischief from which fallacious parallaxes have 
abiindantly sprung in times past, and which cannot, in the 
present and future, be too carefully guarded against Especially 
formidable are certain slight idiosyncrasies of perception, by 
which measures of distance become modified with the varying 
positions of the line of direction between the objects measured, 
lelative either to the vertical or to the line joining the 
observer's eyes. And since this subtle spring of error rises 
and falls hanuoniouflly in a period of a year (because dependent 
upon the uranographical situation of the stars under scrutiny), 
it would be capable not only of completely^vitiating observa- 
tions apparently accordant, but even of simulating parallactic 
changes that had no real existence. Instrumental erroiB, too, 



^ MoTttkty Noticts, vol. ixiii. p. 74. 

* Ada, SocielfU. £tieni. Frnnitte, t. vii. p. 

' Krager, Attr. A'ach. Koa- 1210-12. 





««& 

ii 

•Uj 111 Mill iBlo dbct bf Dc. Bruaiii>« tn • uniua of 
|itioMoritaikrpKaIUz«iX>pbtuibrt«ttD 1868 and 1874 
Th> Ma« pi> tiwM »t of tbe Uwr^igb fllimiiiAtkn of errata it 
M0» p«MB«i And ptnodicAl, Iim mm» hma gmw\i ully fblkwi 
Mom dfcctnaUjrltun b/ tzioifcoUisBHn»Uia&inoiis''Ka0V 
tbjmlf" of til* old Greek phOiMopiMo hu faaen tdban to 
liMrt by MtrouomMk Tbcir naiam and eUbot&te tnqoibBi 
reftfd not aMselj BJocoacopic ttiequiUtiaa of Bcale-diriaQM 
and aemr-rdiiea, dwngw in tefractim, oortectioiiB for aben*- 
llon kdU proper motion, tml the cuxming tricka of their own 
DcrVM^ ibe oaprioes of cerebiBiion, all the wying ooaditioas 
of p«ro»pUoa in ^to orgaiuim «.( their individual oommaad. 

N'on« of thene precautioDA were neglected in the imporUnt 
work executed bj Gill and Elkin at the Cape In 18^1-33.' 
Fully aljvo lo ita subtle reqiuremeata, tbey g*Te to th«ir 
deterinjaatioufl a procisiaij entitling them to standard tank. 
Sir Dnvid GiU'i diacuasion, especially, of the parallax of « 
Ontauri in a uiodel inquirj. It leaves, one maj eay, no stone 
unttirnoU bcnt-ath which a eource of illuaion might lie oon- 
OBftldd. Tlie reflulting parallax of 0"'75, accordingly, obtained 
by iaUt»periik'iit cotiiiwiri»ons with no leas than four pairs of 
■djaoent itara, im probably more nearly accurate than any 
value of tlifl uort yot rugistered. The fact is definitely assured 
that liKht, whiuh llioH from the suu hither in eight minutes, 



> iftutrtf f)intrvatiotu, 1848-53, Appendix ; Mtmoirt IL Aair, Soc 
ixvil, ]i. 44. 

< liuthiin dt I'Amd, SI, F4tAnboaig, t. xiii. Suppl 
' Utnuiri R. Atir* Soq, vol. Klviii. 




THE DISTANCES OF THE STAES 



293 



ads four years and four months on the journey from our 

Bt known neighbour among the stars. The corresponding 

einee is, io round numbers^ twenty-five billion miles. 

A joint attack on Sirius discloaed a parallfix of 0""38, 

iplying a light-journey of 8"6 years. These were the first 

Bure^ of the dog-star made imder perfectly Buitable con- 

itions, and their repetition by Sir David Gill in 1888-89 

fith the 7-inch Cape heliometer, finally established their 

curacy,^ 

Of the nine eouthem stars investigated by GUI and Elkin, 
76 — e Indi, Og ^^^ ^ Ei'idani, Lacaille 9342, and f Toncani 
-were chosen for their large proper motions, and all proved 
be meaaurably near the earth.^ Canopus and ^ Centauri, 
on the other hand^ included in the list l>ecause of their 
distinguished brilliancy, averred their extreme remoteness. 
Prom the former in particular, no symptom of displacement, 
progressive or periodical, coukl be elicited then or subsetj^uently. 
This is really, when we come to consider it^ an astonishing 
result. 

Second only to Sirins in the southern hemisphere, the 
star of the Nile far outshines every star north of the celestial 
equator. As chief of the great constellation Argo, it seems 
to command, while standing slightly aloof from, the daz^Ung 
array of all stellar ranks spanning the heavens from the 
Greater Dog to the Cross. And since its spectrum is marked 
by nearly the same kind and amount of absorption aa that of 
Procyon, we cannot safely conclude its mass to be abnormally 
small in proportion to its light- 
Both, regarded absolutely, must be enormous. The failure 
of persistent efforts to detect any parallactic shifting in 
Canopus obliges us to suppose it so far o& that its 
light needs ai least SOO years to reach us; how much 
longer, it is impossible to tell. At this minimum distance, 
our 3UU would shrink to a tenth-magnitude star; it would be 
one of the dense shoal of telescopic objects im|>ercept!ble to 
unaided sense, and scarcely yet individualised by the industry 
of astronomers. But 22,000 stars of tenth magnitude give 
only the light of one Canopus ; whence it follows that Canopus 

^ Annah o/tfu Oape Obtervatory, vol. Tiii. pt. it. p. 24. 
* See Appeudu, Tabk V. 




Sir Bnii 
feo it a f""---- oi O'^S^ aqnvBknl to 
of 109 yemtm. On asn, ae maJu i^, 
mUk m»mfitK ihu ^^ the loiiRar this wIuSb orK 

TbeUne had now cocne wbc« a 
vi'biob innwiriw of tbis Inid 

Hittetoi, ntaw^ww ^d faMa CBBtt to aefeq^ tlv 
rr'"*'"*g ■objecta for tEeir iinmiiamln vitboot anj 
to ihB otMitJiiMtHiti of resttlta. Tbe ootoome mn a miniiitiw 
' detached statemmts as to fltelkr ilkUiajiiH, iatereatiBgp flMft 
I7 itadf^ in a high degree, yet fnmpahiw of 
for the paxpoee of any genenl con dM on. So long age ■ 
1853, Dr. Peten had pointed ont that what was needod fr 
obtaining a funduaaital acquaintaoee vtth the etnictizt« 4f 
the sidereal world waa not 00 much the detecminackD <rf 
flxoeptional paraDaxee, ae the steady compLUtion of dsta lor 
■one well-grotuided inference relative to the ^i«t»««i* of 
defined star-cUeeee.^ Bat it was not antH thii^ yeas letar 
that it became poeeibk to act on tbe soggeetioo. 

Encouraged by the eoooera of tbe work just aooomplidied* 
Sir David Oiil propoeed. January II, IS 84, a acheooe of 
attack upon the problem of etar-difitanoee in its widest 
bearings. Two " great cosmical que^tionfi " presented thtifr- 
selveQ to him aa answerable hj the judicious distributioa of 
some yeare^ continuous labour* The first related to tbe 
average parallaxes of stata belonging to sucoesaive orders of 
brigbtneHB; the jsecond to the connection between poialkx 
and proper motion.^ 

A plan was accordingly concerted by which I>r» SUkxa 
undertook the measui'emenbt with the new Yale Collie beUo- 
loeter, of a coOBidetable number of representative northern 
atare, while Sir David Gill dealt with a corresponding 
Bouthem list at the Cape. Its final outcome was tbe anthem 

' U47MirsM de Si. PMerxbour^. t. vu. p, 149. 
Mimoin B, Att/r. Soe. ^L sdviii p. IBl. 




THE DISTANCES OF THE STARS 



295 



Btermination of thirty-two parallaseB — tea in the northern, 
Iwentj-two ia the southera hemisphere, including those of 
all stars of the first magnitude.^ And thus at last, a scale- 
unit for the stellar universe was provided. For, once we 
know the distance in billions of miles or light jears corre- 
sponding to the first magnitude — the distance, that is, at 
which a " mean star " would shine with about the luatre of 
Spii^a or Kegulus — the distances corresponding eeveraUy to 
the lower magnitudes follow as a matter of course. They are 
linJced together (unless we are deceived by systematic changes 
of brightness) by an invariable proportion. We have already 
explained what is meiint by the "light-ratio,"^ but it may 
here be repeated that a atar of any given magnitude is^ by 
definition, one 2'512 times brighter than a star of the magni- 
tude next beloWj and 2'512 times leas bright than a star 
of the magnitude next above it. But, since light varies 
inversely aa the square of the distance, any star removed to 
^2*512 = 1-585 times its actual distance, would show exactly 
one magnitude fainter than it did before. This number, then, 
1"585, the square-root of the light-ratio, may be desiguated 
the " distance-ratio." It represents the difference of dietanee 
equivalent to a difference in light of one stellar magnitude. 
The relative mean distances of the various classes of stars 
are then known ; to render them absolute, we only need to 
ascertain the real mean distance of any of thoee classes. 

It is true that, within each class, vast diaparitiea 
escist. Small stars, comparatively near the eartb^ take their 
stand on the same level of apparent brightness with indefinitely 
large, but indefinitely remote bodies. What is invariable for 
each magnitude is the proportion between real brilliancy and 

the sq^uare of the distance. Symbolically expressed, — is 

constant. That is to say, photometric unifonnity results 
from a certain balance being atruck between remoteness and 
light-power, by wluch the effect of equality ia produced. The 
law, however, connecting average dietanee with apparent 
lustre ia not invalidated even by the limitleea variety included 

> Anitals Goft ObttrvatOTij, vol. yiii. pt, ii. 1900 ; PublicAiiom Vale Ob' 
jervatonf, vol, i. pt. vi. 1903. 
* Sea antt, p. 19. 




does to 

tbti catao of brightnw to 

disUDc« ii probttblj not wiwitaral wiUuD the | 
oatmdt of it. 

Mtaawhik, the Cape aad Y«le resolts give ^ of a 
•teood of ftrc ■« tbe sfersge panOax of firatHxtf^nitade snxm 
in aU parts of the ■ky. Tbej eatabUi» at a distance of tfantx- 
iJuw light fcan. the first haltug-atage lor expkmtioaB of 
Mata^ IN'ce. Thuit inoonoedTat^jr reieotev taken all romwL 
are the brightaet of the itellar host Got sun, bo placed, 
would Hjnk nearlj to the fi/tli magmtude. Its fellow anna, 
then, far aiLrpaw ita glory* 

On the acale determined at Yale College, the mean dJBUnc^ 
of atan of the gecond magnitude h Sfty-two light jean 
(pandlfix 0'''063); stan of the third magiiitode are at. eighty- 
two light yCAFH (parallax 0"'04), and ao on ; the invaria>bb 
ratio of l'G85 rL'gulatiEig the incxease of distance and decneeae 
of magnitiido for eiich descent of one step, provided only that 
light flufffltH no ethereal abnorption. "When we get down to 
tho MixUHintli iiiiiguitiide. which is ahout the minimum visibiU 

' IVnwwimb, The Stan, p. 313, vhtte (u Ur. T, K. Hutta hu renurksd) « 
Tftloft of ] '414 for the diMU&«0-rAUd ia implied. 




THE DISTANCES OF THE STAHS 



297 



Sn the largest telescopes (the Yerkes re&actor attains to one 

"Snaguitude lower still), we find the theoretieal light-interval 

lengLheued to 33,000 years; but there is no certainty that 

m:Dj aucb far-travelled rays reach ub. The regular progression 

of distances may not extend so far. It must stop eomewliere» 

if the stellar syetem be — as we have reason to think it 

is — ^of finite dimensions; at what particular magnitude the 

break occurs, it would at present be futile to conjecture. 

All that can be said ia that, distance becoming at length 

eliininated as a factor of magnitude, the diffeieuces of the 

fainteat stars represent, chieSy or solely, real inequalities 

in shining. There may possibly, for instance^ be no " mean 

distance " corresponding to the sixteenth magnitude. The 

stars of that rank would not then, on the whole, be further off 

than those of the rank next above them, but would> on the 

whole, possess only ^^^ of their real light. This must be the 

case — so far as we can see — at some stage of the descent into 

the abysses around ua 

Besides Oanopus, four among the twenty-one brightest 
etaifl in the sky were found by Gill and Elkin inaccessible 
to parallactic research. These are Kigel, Spica, a Cygni, 
and ;3 CruciB. AH five must be Brobdingnagian orbs ; their 
magnificence defies the realising efforts of imagination. In 
singular contrast to them are certain swift, but dijn etars 
i measured by Sir David Gill, notably one in Pictor of 8"6 
magnitude, which cannot, since it has a comparatively large 
parallax, shine with more than ^j^ the lustre of our modest 
sun- Ten, in fact, of the tweuty-two stars determined at the 
Cape, proved to fall short, in sundry degrees, of the solar 
standard of light-power.^ But to what extent they depart 
from the wide average of the sider&al system, we have at 
present no means of judging. All that can be said is that 
the variety embraced by it has a prodigious scope.* 
^L Of the ten stars measured at Yale, Procyon, with a parallax 
^%f C-334, giving a light-journey of 9'8 years, was found to be 
the nearest to the earth. Altair, at a distance of 14 light- 
years (parallax 0"'233), came next; Aldebaran, at 30 light- 
years (parallax 0''"109), third. From Arcturus and Betelgeux 

' Cape AniuiU, vol. viii, pt. ii. p, HI, 
* OiU, Prec, S. J/riean. Phil. Society. Sept, 17, 1002, 



298 



THE SYSTEM OF THE STABS 




an aIiuo8t identical result was derived. Kach is plonged ia 
tt profundity of Bpnce represented hj 12fi years of li^t- 
travet ; and tbi^ implies tht? r«al luminosity of An^turus to be 
itpproximaU'ly 1000, that of Betelgeux to v&ty a, Httld on 
either side of SCO times the solar brightQeas. 

So clooely and so coELfiequeati&llj hare advanoeii in thii 
arduoLii^ bronch followed the growth of improvemetit in 
holiometere, that direct visual measurements for the purj^Me 
with auj other instrumeut might almost s(^m waste of labour 
Nevertheless, the traneit^circle has been imexpectedlj rendered 
thus available. Professor Kapteyn originated, in 1SS9, t^ 
determination of stellar paiallaioa by diSerences in right 
ascension, giving for fifteen stara results at once acknowledged 
ft6 authentic ; * and hie method woa applied to ninetj-aii start 
by Professor Flint at the Washburn Observatory (I7.S.A.) ia 
1803'OG.' It is safe and expeditious, and serves as a useful 
adjunct to work with the camera. Photographic poraUai 
researches were effectively set on foot by Frofesaor Pritchard 
His first experiment was with the classic 61 Cygui, of which 
330 separate impressions, obtained in 1886-87, furnished the 
materials for 30,000 measures^or " bisections" of star-images.* 
For the immediate end iu view, these extraordinary pains 
were largely superfluous; but they had the ulterior object, 
fully attained by their means, of establishing the credit of a 
□ovel and unfuniiliiiT method. The most delicate of all 
astronomical inquiries was thenceforward^ with the full a^enl 
of experienced judges, admitted to be within the competence 
of the celestial photographer. 

The advantages of determining parallaxes from setiaitive 
plates are manifold. Perhaps the chief of them is the nearly 
indefinitt) power of control they afford. Any of the imprinted 
stars, situated at all near the prolongation of the major axis 
of the parallactic ellipse (in other words, with a tolerably 
large " parallax-factor ") may bo used aa a point of reference. 
Comparisons can thus be multiplied almost at pleasure, and 
inferred displacements with regard to one star checked by 
recourse to another, duplicate plates being at hand for addi- 

* AHr^ Jfioeh. No. 203S ; Annaltn dtr ,*iUmieartf in Lcidtn, Bd. tU. 

" Publ. fVarhbttrn Obtervatoiy, vol. ri. 1902. 

* Monthly ytflka, v<A. xJviL i*, 87. 



THE DISTANCES OF THE STAES 



299 



[>nal safety. By the proper as8 of euch safeguards, delusive 
results can be all but certainly excluded. Moreover, relative 
paralittx becomes virtually absolute when coinparisODS are niade 
with a great number of stars, moat of which are presumably 
too remote to complicate the result by perspective movements 
of their own, 

"Within its peculiar province photography compriseB stars 
too faint to be conveniently dealt with by visual means. For 
the images of those much brighter, over-exposed through the 
necessity of giving the small etais in their neighbourhood 
time to imprint themselves, become diifused into blurred disca 
unfit for accurate bisection. Plates, on the other hand, taken 
by Dr. Schlestnger with the Yerkes refractor in 1903-4 proved 
well adapted for obtaining the parallaxes of eighth and ninth 
magnitude stars.' Trial waa made of only three ; but one of 
them was a double star in Cepheus. found to be ten times 
nearer the earth than Vega or Arcturus (parallax = 0"'278). 
Yet the pair give no more than ^^ the sun's light, though 
their rapid mutual revolution suggests the presence in the 
system of considerable attractive power. It will be curious to 
aecertain, when their orbit develops sufficiently to be com- 
puted, the relation in its members of mass to luminosity. 

The main object of present inquiry is to obtain a wider 
basis for general conclusions regarding the distances of the stars. 
For this purpose it is mote important to secure a considerable 
number of parallaices reasonably well determined than a few 
reduced by scrupulous care within the narrowest possible 
bounds of error, Eesearch in this sense is already well on its 
way. From statistics of proper motion, Frofe^or Kapteyn 
has derived trustworthy estimates of the mean distances of 
the stars according to their photometric rank." In doing 
so he found it advisable to distinguish between the 
members of different apecti-al classes, since the fact clearly 
emerged from his discussion that^ — as Mr. Monck had indicated 
several years previously" — Sirian and helium stars are fully 
twice aa remote^ on an average, as solar stars of the same 
magnitude. A colossal scheme of direct inquiry was besides 

^ Astraph. Jtmm^ roL xx. p. 123. 

* Attr, Naeh. No. ^487 ; Onmii^en Astr, Fubl No. 8, 

' Aatr^ and AMtrGphysica^ toL xt. p, 701, 



300 



THE SYSTEM OF THE STABS 



sketched by the Groningen astronomer in 1889' azid iti 
feMnbility eBtabliahexl by experimental measurementa ot pUlM 
UJcan hj Ftofesaor Banner at Helsin^ors, the results of which 
were made public in 1900,* But the proposed pbotograpluc 
stirvey of the heavens, by which the parallaxes of 800,000 
Stan might poeaibly become known, is stlU in abeyance^ Itt 
execution would be most costly in time, labour, and monej; 
and their lavish outlay ia discouraged by the bofSing irtalliMM 
of the quantities sought to be discloeed. 

Meanwhile we may attempt to summarise the outcome of 
preliiuiuary explorations. It is, in the first place, decidedlj 
unfavourable to the existence of any large parallaxes. Tbs 
possibility is, of oouree, by no means excluded that stars mftf 
be found much nearer to the solar system than a Centauri, 
but their diflcovery is growing every year less and less probable. 
Sir Bobert Ball^ examined some years ago about 450 objecti^ 
in a manner which, though summary, would have Buffieed to 
bring to view any parallax of a single second of arc. If one 
was forthcoming. His list comprised a number of red and 
variable stars, Nova Cygni, Webb's planetary nebula, and the 
Wolf-Rayet gaseous stars in Cygnus ; and it may be noted 
in passing that spectral peculiaritiea are almost invariatily 
associated with an uncommon degree of immobility id the 
sky. The ninety-six stars reviewed by Professor Flint on 
Kapteyn'a method were of a more promising character; and 
ninety'two were selected by Dr. Chase for rapid scrutiny 
with the Yftle heliometer, because their large proper motions 
supplied an argument of proximity ; yet, among them all, not 
one aeema to be within a five years' light - journey of the 
earth. Indeed, parallaxes even of one-tenth of a second^ 
signifying a Ught-iDterval of thirty-three years, are held by 
Proftissor Knpteyn to be extremely scarce.* On the whole, 
perhaps one hundred stars * have been shown to swing to and 
fro sensibly in response to the earth's orbital vibration ; and 
the number of fairly well determined parallaxes may be put 
at about seventy.*" 

1 Bull, da la Carte dit tSW, No. 3, p. 262. Ct PulMoh, A»tr. Nach. No. 
4013. 

* Oroningen Aatr. Pnhl, Ko. 1. " Dunsink OhaerrutiiimB, voL T. 

* Qnmingeti Publ. No. 1, p. S3. * Nowcomb, The Stan, p. 149. 

* Sm Appendix, TftbU V. 




THE DISTAilCES OF THE STARS 



301 



The cardinal truth emerging from theae inquiries is that 
the extreme isolation of the solar system. A skiff in the 
Iddfit of a vast unfurrowed ocean is not more utterly alone, 
out the same proportion would be borne by an oasia one 
tnile across to a desert twenty times as extensive as the 
Sahara, that our sun with hia entire planetary household 
bears to the encompaasing void of space. The enormity of 
its blank extent is strikingly illustrated by Father Secchi's 
remark that the period of a comet reaching at aphelion the 
naiddle point between our sun and the nearest fixed atar» 
would be of one hundred million years ;^ and, by recent 
ujeasures, the nearest fixed star has been pushed further back 
into space by one-quarter the distance assigned to it when he 
wrote. Yet the sun is no isolated body. To each individual 
of the unnumbered stars strewing the firmament, down to the 
fainteat speck of light juat shimmering in the field of the 
Lick refractor, it stauda in some kind of relationship. 
Together they master its destiny and control its movementa. 
Independent only so far as its domestic affairs are concerned, 
it is bound, aa a star to the other atara, by inQuencea reaching 
efficaciously across the unimagiaablo void whicli separatee it 
from them. The outcome of those influences in the trana- 
latory motion of the eolar system we shall consider in the 
next chapter. 

1 h«9 M»k9, t. iL p, 146. 




Tkk «tadj of the stars inevitablj leads ua to oooeader 
ttdvancsDg movement in the midst of tbem of tbe son Aad 
atteodftiit train of pLaneUi. There c&q be oo resaonable 
-^^tid the thought is an aBtaandmg one — that wv mn 
ga^ed on a voyage throagb space witboat starting-proitt or god 
that we can know of, which to&j prove not wholly nueventfiiL 
Ite progress may possibly bring aboat» as millenniiuns go by, 
chaog«e powerfiolly influential upon human destiniaB ; nay, «a 
incident in its course inay at any time, by the injscnUalik 
decree of Providence, terminate the terrestrial existeztoe of oor 
race, and consign the records of its ciTiliBation in dnst 
cinders to the arid boBom of a dead planet. A ciirioiia 
of helplesanesa, tempered, however, by a higher trust, is 
duced as we thus vividly realise how completely we are 
the mercy of unknown forces- — how Irresiatibly our liti 
" lodge in the vaet wildernees " of the universe is 
onward over an aoDual etietcb of some four huadred mUlioos 
of miles, under the mysterious sway of bodies reduced by their 
almost infinite diatancea to evanescent dimensions. 

But, aa things are constituted, the tranalation of the sun' 
household is a necessity, albeit one of startling import to o' 
I selves. The stellar system is maintaiued by the balanoe 
' forces, and motion is the correlative of force. Aa a star amon, 
stars, the sun can only maintain a separate existence by con- 
tributing its share to those harmonies of movement by which 
"the heavens show forth the glory of God." Destruction 
woidd be the eventual penalty of even a moment's immobility 
- — a penalty, indeed, which might not be exacted until after the 

dO'2 



Htd9 




TBANSLATION OF THE SOLAE SYSTEM 303 



Ipse of many miliioos of years. It may reasonably be 
Bumed that a Centauri exercises upon the sun the strongest 
Attraction of any individual star; but a collision would ensne 
very tardily upon abandoninent to its influence. The aua (if 
undisturbed by competing piiUs) would fall from a position of 
Test to-sra-rds its next neighbour, less than the third of an inch 
in the first month ; the second month would see despatched 
nearly a full inch of the journey of twenty-five billions of 
miles; and although the acceleration would of course grow 
more rapid as the distance diminished, upwards of fourteen 
million yeara should pass before the firea of aun aud star. 
probably become extinct during their gradual approach, could 
be rekindled by the catastrophe of their impact. 

There is then an A priori certainty that the sun moves ; 
and assurance on the point is rendered doubly sure by 
inferences from observed facts. For besides their annual 
paraLai due to the earth's motion round the sun, the stars 
have a "secular" or "systematic" parallax depending upon 
and attesting the reality of the sun's motion round an 
unknown centre. Let as see bow this -systematic parallax can 
be investigated. 

If the sun alone were in motion, and the stare at rest, the 
reaolts in perspective displacements would be simple aud iin- 
mistakahle. Each star would appear to travel backward along 
a great circle of the sphere, passing through the two points 
towards and from which the sun's course was directed. So 
that there would be the acmblance of a general retreat from 
the " apex " or solar point dit 7/im, coupled with a thronging-in 
from all sides towards the opposite point, or " anti-apex." For 
each particular star^ the amount of displacement should vary, 
inversely oa its distance from ourselves in space, directly as 
the sine of its angular distance from the apes. Hence, if the 
annual parallax of even one such aensibly shifting star were 
determined, not only the rate, in miles per second, of the 
solar progression would at once follow, but the parallax of 
^very other sensibly shifting star in the heavens could be 
ieduced by a simple calculation from the relative quantity of 
its apparent movement. 

But the stars are not at rest. They have movements of 
their own, greatly swifter, in many cases, than that of the sun. 



$04 THE SYSTEM OF THE STABS 

Fts^eotiTS fiflecU Are thos ta a great «xtooft wmakeL l€ 
thtff ffabttiL U U tnAthemstacttUj cearUin that era; rttt; 
whatever its own ootme or «poed« refiecto the son's moCiott a 
the itiiet tne&sure of its positioa with regard to it What in 
called the "proper tnoiions" of the stare are then madsupflf 
two parU. one real, the other apptuneat. They inehide i 
eomtuOQ element, the separatioo of which from the hetEtf- 
g«neoU5 admixtures dis^oiaing it, constitiOes the prohiem W 
be solved. 

With the instinctive appreciation of genius^ Horscbel vest 
■Izaight to the heart of the matter. Wliat had to be done, bt 
saw clearly, was to fiud out the direction which abould )» 
given to the sun'e couraet in order to make it account far« 
targe a proportion as po^ible of the sum-total of $telUr 
moTements. "Our aim must be," he wrote in 1805, "» 
reduce the proper motions of the stars to their lowat 
quantitiea." ^ And again: "The apex of the solar motioa 
ought to be BO fixed iis to be equally favourable to every star." 
But how is this to be done ? Very simply^ if we only cotiaider. 
as Herschel did, a few of the brightest stare. 

Take, for example, four stars with conspicuous movemente, 
two in the northern, two in the southern hemisphere ; namely. 
Vega, Capella, Sirius, and Fomaibaut. The great ciitles, of 
which each annually describes u minute arc, traced backward 
on the sphere, yeiy nearly intersect in a single point situated 
in the oonetellation Hercules,* Had we only the motions of 
those four stars to consider, we should accordingly infer with- 
out hesitation the " suu's wa^ " to He thitherward. Nor 
ahould we be very far wrong. The moat refined modeni 
determinations of tbe solar apex, founded upon the motions of 
several thousand atars, differ among thomselTes to an extent 
comparable with their mean deviation from the result of the 
extremely summary proceeding juat indicated. 

The graphical methodj however, is evidently applicable 
only to a very restricted stock of data. When a crowd of 
atars have to be taken into account, the points of intersection 
of their respective circles of motion become spread over too 
wide an area for a " mean apex " to be struck out fairly 

' Phtt. Trant. tdI. %vv. p. 2*8, 
* TbLs irat rotQ«rkud by Kltukerfuw, QminfriH^Ae ^achnefOM^ 1S73, ^ 3tK). 



TRANSLATTOI^ OF THE SOLAB SYSTEM 305 



gtweea them, even bj the exercise of a judgmeut as dieh 
criminating as tliat which in 1783 led Herschel to place the 
go^l of solar travel in the vicinity of \ Herciilis, The 
accumulated facts must then be dealt with by a method at 
once stricter and more comprehensive. A glance at the 
itura of the task in hand easily auggeats to a mathematician 
rhat that method should be. 

The proper motions of the stars give, as already hinted, 
le plainest evidence of individuality. The lines pursued by 
lem run in all possible directions. But a aub-stratum 
regularity underlies this seeming confusion. A mere 
spection of the signs plus and minus, eiguifying respectively 
ist and weatj and north and souths attached in catalogues to 
le CDinponeijt& iu right ascension and decUnatiou of stellar 
movement, auihces to show a general prevalence of law through 
le unequivoi;al tendency of the signa to vary concordantly in 
sing from any one to an adjacent region of the heavena.* 
Lt a c&up d'eeil^ Argelander fixed the point from which this 
ider-cuiTent of motion flowed, and so gave an improved apex 
ifor the course of the sun, confirmed in the main by Bubsequent 
irch.'' It ia then clear, in the first place, that no movement 
possibly assignable to the sun can explain all stellar displace- 
ments; a large residuum being real, and therefore by no ingenuity 
to be got rid of. While in the second place, the nearer the truth 
is approached as regards the direction and amount of the sun's 
motion, the smaller obviouflly this residuum will be. Iu other 
words, the most probable value of the solar motion will be that 
which renders the "sum of the squarea of the residuals" of 
stellar motion a minimum. 

But why the sum of the squ^^s, and not the simple arith- 
metical sum of the outstanding proper movements ? It needs 
only common senae, aided by the most elemeutar)' geometry, 
to get a sufficient insight iuto the reason. Any one can see, 
with the help of a pencil and a piece of paper, that, if a line 
be divided into two segmeute, and squares be constructed on 
the scgmeuts, the sum of those squares will be the least possible 
{when the line is equally divided, and will increase continuaUy 
|Vith the inequality of the segments. This simple fact gives 

' stone, Monthly Notices^ voL xxvii p. 239. 
^ Mhnoirtt -priawdis d VAmd. St. P^tcreLourg, t. m. p. SSfl. 




$oe 



THE SYSTEM OF THE STAS8 



Uie clae to tbe prioctpW of * lawt ■qiiBz«&*' lu oibjeci ii to 
«lidt toch A qoaaXi!^ m will nuike the ^^*»*-"^'iTg enw* «( 
olfrrttiiMi, or mbj other kiod of rasidtaAla, sa miall «s poaiUB 
off mnuf^ Not iner&lf stnftll takan in the ftggre^^U, Us 
fadnead impartially to a tmifbDU ievBl of imrigniBniin 
Ihidflr tbene dfcmnstaujces, as we have seen from tbe on* 
nderation of our divided line, tbe sum of their sqiiareB wiQ te 
a mmimmn ; and it oan be mathematicallr demoosttsted UMt 
the most probable result of goch investigations as are vm- 
oeptible of tbiB kind of trsatment, is arrived at when tbs 
condition of " least squares " is fulfilled. 

This mode of attack upon the problecn of the son's tiatis^ 
latton was first employed by Argekmder in 1837. Afisomiag 
|kra\'iBionaUy the correctness of Herschel*s apex, he |jtoo ce il id 
to compute for each of 390 stars with aaoertained piQpv 
motioos tbe linea along which those motions fihould [Hrooeed 
if due to ayatematic parallax alone. Their deviations bom 
the prescribed directionfi gave him '* angles of error," wbidi, 
placed in the category of casual errors of obserration, and 
treated by the method of least squares, indicated a corrected 
apeXi such that by its adoptiou, the sum of the aquares of 
the dI3eren£«e between what was calculated and what was 
observed — that is, between the purely jiarallactic drift of the 
stars and their actual displacements — was reduced to the least 
possibie amount. The solar movement wos^ in a word, so fixed 
as, in Herachel's phrase, " to be eqimlly favourable to every 
star/' a condition fulfilled by directing it towards a point in 
right ascension 260" 51', north declination 31* 17V Bat 
there is much r^son to believe that the position of TnATcininm 
neutralisation — so to call it — really lies some fifteen degtees 
further to the east. 

An important modificatinu of his method was introduced 
by Sir Gem^ti Airy in 1859-^ Abolishing the conception of a 
spherical surface of reference, he defined the linear move- 
ments in space of the aufl and staiB with regard to three 
directions at right angles to each other (" rectangular co- 
ordi Dates"). No assumption of any kind was then needed; 
the subject was treated with the utmost strictness and gener- 

' Mimoires pr^tnUt, 1 iii. p. 690. 
^ Meitutirs B. A^. Site toI xxTiii. p. 143, 



TRANSLATION OP THE SOLAR SYSTEM 307 



lity, and some possible causes of error were removed. Airy'a 
id many points of theoretical superiority over Argelander^s 
lethod. That, however, of introducing the oonsideration of 
le quantity of each star's movement was to a great extent 
juaterbalauced by the necessity which it involved of adopt- 
ig precarious suppositions as to the distances of the classes 
stars employed. The apex for the solar movement resulting 
3m the coDfiideration of 113 stars was situated in R.A, 
IGV 29\ Dec. -1-24° 44'; while Mr. Main's similar treatment 
1165 stars shifted it to R.A. 263' 44', Dec. -J- 25V 
This great subject waa again investigated by M. Ludwig 
Struve in 1887-' The incitement to undertake a task 
Bndered formidable by the very wealth of the materials at 
disposal, was afforded by Auwera's fresh reduction of 
iley's Greenwich observations. From a compariBon of 
le star-places authoritatively determined for 1755» with 
given in the St. Petersburg catalogue for 1855, a list 
2814 proper motions waa derived, of which 2509 were 
Available for M. Strnve's purpose. Among the stars for 
various reasons excluded, were the seveti swiftest travellers^ 
as unduly affecting the result through motions no doubt 
mainly ohginaL 

The outcome of this exhaustive diacussion waa to place 
the apex of the solar motion in "R.A, 273° 21', north declina- 
tion 27" 19', a rate being assigned to it such that the space 
traversed in a century, viewed square from the average dis- 
tance of a sixth-magnitude star^ would subtend an angle of 
4*-36. Admitting that stellar distance varies inversely as the 
square toot of stellar brightness, hence that stars of the first 
are, on an average, only one-tenth as remote as stars of the 
sixth magnitude, we can, with the help of Dr. Elkin'e mean 
parallax, for the former class, translate this angular into linear 
velocity. It comes out 14^ miles a second. 

Well-nigh the whole of the stars visible to the naked eye 
in the northern hemisphere concurred in M. Struve's deter- 
mination. It was conducted on Airy'a method, likewisa 
adopted in 1890 by Lewia Boaa in a discussion of 253 proper 
motions extracted from the '* Albany zone," the observation of 

1 Mcnoirs R. Asir. Soc tdI, Txxn. p. 27. 
■ Mi»ioirA» dt St. r^irgbourg, t. luv. No. 3, 1S87. 



808 THE SYSTEM OF THE STABS 

which had just then been completed for the AjrtzoooiBkJH 
Oeaallaohaft Catalogue. He obtained an apex in RA. 2^0'. 
J>^40V near thti quadruple star e Lyrse; aod recurring to 
the subject in 1901, finally concluded for a point five degicM 
further north.* A Dotable attempt^ too, was nude "by IL 
Oscar StumpQ in 1890' to show th^t the apexefl aepaiatelj 
deduced from various clasaea of stars shifted ^^temiitically on 
the spliere. This aeemed to involve the important difxjasare 
that the groupa of stars oom^ider^ had distinctive aggr^tM 
MorementSj and were hence dominated bj difj^rent gtaiita- 
tional influences ; but the effects brought to Tiew have ptobahlj 
aa much to do with the correction of catalogue-placee as with 
the laws of sidereal construction/ That theae are alao ooa- 
cemed, was proved by a reseitrch based on the proper niotioM 
of Groombridge'a circumpolar atars^ executed by Measra. Byaen 
and Thackeray at Greenwich in 1 905.* Solving their equation* 
separately for stars of the first and second spectral types^ they 
obtained from the separate eoUectioos markedly divergmt 
directions for the sun's route» and thus, aa the upshot of 
their experiment^ laid bare one of the hidden lints between 
the dynamical and the physical relations of the stellai 
world. 

A material advance was made towards disentai^Iing the 
intricacieB of tho solar movement by an innovation in the tnat- 
ment of proper motions. It might have been supposed that eveiy 
device for their manipulatian had been exhausted, and that tha 
decipherment of their perspective significance was complefce, at 
least in principle ; yet Professor Kapteyn contrived, in 1893, 
to give it a novel stamp of clearness and certainty. ^Resolving 
one by one the whole stock of star-movements at his disposal 
along, and at right angles to, the great circle passing through 
a solar apex aagumed as the most probable, he succeeded m 
isolating their parallactic element much more perfectly than 
had been done before. The fundamental nature of the 
problem was thus laid bare ; obscurities wero dissipated ; and 
there ensued a determination of first-rate authority, accor 
1 Astr. Journ. No. 213. s jj^^ ^^^ ^qi^ 

* Astr. Nach. Jfos. 29M-S00O. 

* Kaptejro, ihid. Kos. 3721-22, 3869-60 ; Pwu. A-m*terd^m Acad. ^^finaS} 
Jiiti. 27, 3W0. 

* Sfonlhli/ Jiotiea, voL Izv. p, 43a. 



■din^ 




TEANSLATION OF THE SOLAK SYSTEM 309 



which the sun's path is directed towards a point in R.A. 
174°. D + SO", just six degrees south of x Lyne. Ita flub- 
otial accuracy was vouched for by ProfeBsor Newcomb'a 
Bterly researches.^ Profiting by long experience in evading 
Wtfallg and estimating aources of error, he deduced from 2627 
lall proper motions a solar apex in E.A. 274°, B + 31° ; from 
SOO larger, one situated in E.A. 277', D + 31^ Which result 
ervea more confidence, cannot off-hand be decided ; but we 
ay hope — although this is by no means sure — that their 
ifference representa the surviving extent of uncertainty. 

The plan of inquiry just sketched, altliough it serrea 
wonderfully well, on the whole, for the ascertainment of the 
oute followed by the solar syetem in space, avails little 
Dr determining its velocity. For this purpose, the distances 
the stars employed aa indexes should be known ; and they 
AQ only be estimated for ranks and classes, more or less 
ariously. Kapteyn and Newconib htive, however, vastly 
iproved the method of evaluation, and they agree in fixing 
dn miles a second as the approximate rate of the sun's 
Burney, But a more direct way of arriving at it has, in 
ent times, been thrown open, 
We have elaewliere explained the principle of spectroscopic 
determinations of motion." Their peculiar value consists in 
their independence alike of distance and of visible displace- 
ment. Referring to movements visually imperceptible, they 
complete knowledge of stellar velocities by giving their other- 
wise unknown " radial components." Apart from this mar- 
vellous application of the spectroscope, the real directiona 
pursued by the stars as they ti^avel could never have been 
ascertained, since we can immediately discern only that part 
of their motion lying ticross the Une of sight, which, in 
individual cases, may be all or none. By the spectroscopic 
revelation, however, of motion in the line of sight, the miaaing 
element is supplied, precise and particidar knowledge may be 
had for the asking, and the etara voyage under astronomical 
crutiny, no longer as mere flitting bright specks on the surface 
an imaginary sphere, but as buds in space of three dimen- 
sions, each with its secret in petto, and its deatiny in reserve. 

' Astr. Joum. No. 457, 189S ; The Start, p. 91, 
' Siti. o/Attr, 4th ed. jip. 200, 38S ; ante, p. ISiB. 



310 THE SYSTEM OF THE STABS 

The ejects of reoesBioD and approacb on the ligbt emitted 
bj moviog objects being phygioal and real, the^ ranun 
unimpaired by distanoa Out at the verge of the stdonL 
Bystem, or cloae at hand withiD our own atmospbeie, thej ue 
the HLme for the same velocities, and can, with a sufficient 
Ught-fiupplj, be detected with equal facility. Hence their 
special applicability to the problem of the suu^a speed. Tb 
determine it with reij approximate accuracy, it needs oulj to 
compare the average radial celerity of a good number of stan 
lyiug in &t>Qt of the son's way with ibat of otheiQ he is tear- 
ing behind. Movements of upproaeh must* on the whole, 
predominate in the oue direction, movcinents of reoeasion io 
the opposite, half the mean difference representing the late of 
tranaport of our system relatively to the stais used for the 
ooroparisou. The apectroseopic method, nevertheless, did not 
become ideally effective for this purpose until the twentieth 
centuvy had begim to run ita couree. Experiments with the 
51 stars radially measured at Pot&dam were evidently tenU- 
tiva;^ they forecasted rather than a£brded results. At Ust, 
in 1901, Professor Campbell,^ having collected with the Mill* 
epectrograph data lees iuadequate to the end in view^ deduced 
from them ii movement of the solar system towards an apeat 
in R.A. 277' 30/ D + 20^ at the rate of 12^ miles a second 
The velocity may be depended upon — it is unlikely to bo 
erroneous by more than a mile per eecond ; but the direction 
ia subject to a somewhat wide uncertainty, eapecially aa 
regards declination. For the 280 staia taken into account 
being situated for tbe most part in the northern hemisphere, 
the goal determined by their means was probably displaced 
towards tbe equator. The deiiciency of southern stare wiU, 
however, bo supplied by tbe work of the " Mills Expedition " 
now in progi'eas at Santiago ; and a research, based on sym- 
metrically an-anged materials, will then be practicable. 

Thus, both the course and speed of the fiun and planets 
are not only included iii the category of things knmoahle, but 
there ia every prospect of their becoming known with more 
and more satisfiictoiy exactness in the immediate future. All 

1 Homana, J3lr. Nath. No. 3714 ; Schtinfeld, K J. S, Atir, 0«. Jthig, 
Eli. p. 6& ; Voxol, J3tr. JVtwA. No. 31&0. 
* Asf:r<fph. Jmij-n. toL liii, p* 80. 



TRANSLATION OF THE SOLAE SYSTEM 311 



\t is needed la a closer and a wider application of means 

iy in the bands of astronomers. Still our cnriosity will 

even Iheii be satisfied. The value of the two items of 

formation within our reach is indeed incalculable. They are 

Htu qud nOTi for the furtherance of intjuiriea into stellar 

nechanics ; are they to be a n« plus ultra as well ? 

The Bun, we are well assured, is not travelling along a 
light line. The universality of gi'uvitation makes recti- 
aear movement next to impossible, since no cosmical body 
fcn traverse apaee under the sole guidance of ita own primi- 
ve velocity. It is true that, supposing primitive velocities 
Itogether abolished (and we know of no necessity for their 
istence), any number of bodies might be united into a 
fatem eudowed only with pendulum-like motions. The eun 
ad stars might tbua, by an abstract possibility, be totally 
void of advancing or circulatory movements^ each swinging 
for ever to and fro through their common centre of gravity. 
But it ia practimlly certain that this plan is not realised in 
the sidereal system. 

The path of the sun is then a curve, but a curve moat 
likely of such vast proportions as to remain for ages indis- 
tinguishable from a right line. Strictly speaking, its direction 
is continually changing; the apex of to-day will not be the 
apex of to-morrow ; atili less will it be the apex of a milUou 
years hence. Yet in a million years, it may quite conceivably 
not have shifted from its present place in the sky by more 
that! the width of the full moon ; and our beat determinations 
still fall far short of the accuracy which would enable ua to 
detect a change of half a doxen times that amount. Directly, 
that 18 to say; indirectly, a much more insignificant alteration 
might disclose itself. We will endeavour to explain how. 

Pond, who in 1811 succeeded Maskelyne as Astronomer 
Royal, made the remark that the sun's motion must produce 
a kind of secular aberration of light, by which the stars are 
permanently displaced from their true positions.* The well- 
known consequence of aninitd aberration is to make them appear 
to describe little ellipses, the aemi-axes of which depend upon 
the ratio of the velocity of light to the velocity of the earth 

^ Li&gre, B'ulL d* VAcad. BnuolleB, t. till, p. 168, 1859 ; O. StniTe, 
Mifiwirtt, St. P^terftbourg, t. r, p. 106, 6* Svrie. 



313 



THE SYSTEM OF THE STARS 



ia its orbit. But the svLn*s orbital moTement being 
BO far as experience yet goes> in one direction, the abemtko 
due to it is in one direction too, and is heuoe constant, ud for 
the present beyond the reach of obeervation. It is, boweTK.ooo- 
stant only ao long as the movement producing it remains ssniti^ 
so. As the latter changes, it will change too, and may in thk 

[manner be brought within the domain of bunaan cogmaurae: 
For upon the acceleration, retardation, or deflection of llw 
sun'a movement systematic changes among the stars sbo^ 

I ensue, the nature of which would at once betray t<beir oii^Ow 
The total amount of this secular aberration may be roughly 
stated as oae second of arc for every mile per second of the 
sun's Telocity. Hence, stars 90^ &om the boIat apex an 
pushed forward towards it by about 12", the effect upon 
other stars diminishing with the sines of their distanoes im 
the sphere from the same poinU These abeiralionaJ can be 
distinguiahed from the parallactic displacements simiJarly 
occasioned by their indifference to remoteness in space, Sl»n 
far and nea.r, bright and faint, swift-moving and tardy, are 
equally affected by them. But while it is quite certain that 
visual disturbances of this kind are produced, their interest 
must for a long time remain purely theoretical Indeed, it 
may well be th^tt the modifications reodering them sensibls 
and inBtmctive, will proceed with such exorbitant slowness 
that not even astronomical patience will avail to nnmask them. 
We do not know the plane of the eun'e orbit- — only the 
direction of one line in it. And that line, pointing towards 
the constellation Lyra, makes an angle of about 60" with ths 
Sim's equator. Thus, the solar movements of rotation and 
translation would seem to be unrelated one to the other ; and 
the planetary revolutions to be similarly independent of inter- 
spatial travelling. Our whole system is driven obliquely 
upward by a power which, taking no apparent account of ita 
domestic economy, must have a Bource dist'onuected from 
originating impulse of the helicoidal gyrations illustrated 
Fig. 36, from a diagram by Professor Yoimg. 

A remarkable feature of recent improvements in the deter- 
mination of the sun's course through the heavens has been to 
reduce to insignificance ita deviation from the plane of the 
Milky Way. This is an implicit testimony to their value. 



ita 

"I 




TRANSLATION OF THE SOLAE SYSTEM 313 



« "¥* 



is difficult to coDceiye that course preflcribed otherwise than 
the combined attractions of the galiictic myriads. The 
lost probable auppositiou aa to the situation of the centre of 
force ewaying our syatem is that it lies somewhere in the 
cloudy zone which bo enhances the mysberious beauty of our 
skiea. If the orbit we are puxauing be approximately circular, 
tlien ita centre must be distant by a quadrant of the sphere 
from the apex — it must lie 
somewhere on a great circle of 
which the apex and anti*apex 
ate the polea. Now tliis great 
circle cuta the Milky Way at 
two opposite points in Cassiopeia 
and Centaur, and there, accord- 
ingly) two alternative centres 
of the Bolar motion might be 
looked for. Argelander chose 
for its position the spot near 
Cassiopeia »iarked by the 
great cluster in the sword- 
handle of Perseus ; ' but the con- 
jecture made no pretension to 
scientific authority, and the 
postulate upon which it was 
based of the sun^s path being at all nearly circular is in truth 
of a highly precarious nature. 

We are even ignorant whether the ascertained translation 
of our great lumioaiy representa a primary or a secondary 
order of etellar revolution. It perhaps merely indicates the 
interstitial movement appertaining to the aun as a member of 
a restricted group of stars, the common transport of which 
proceeds undetected in a totaUy different direction. Hence 
the possibility, suggested by Herschel, of the presence of a 
higher kind of systematic parallax than that revealed in the 
drift of bright stars.^ And Professor Campbell wrote, after a 
full century had ripened experience : " Tlie motion of the solar 
system is a purely relative quantity. It refers to specified 
groups of stars. The results for various groups may differ 

1 Minuiru prAunC^, St. P^tersboiu^, L iiL p, 602. 
» Phil Trans, vol ksiiL pp. 1276-77. 



Pio. 39.— Tb« EartLii Uotian Id SpouA 



au 



THE SYSTEM OF THE STABS 



widely, and all be correct"^ So far. bowerer, no toi^ible 
evidei&oe has been adduced to ehovr that difierenoea of thi^ 
kind^ grounded in the nutm^ of things, really exist If they 
do, they esoape through the wide meebes of the neta wiib 
which we oapture grosser facta. 

Pro^resa is here only poeeible through careful and mina 
study of the residual movementa of the stars — of the mo 
ments, that is to say, which remain after the ge&eral 
live effect of the sun's motion liaa been subtracted, and wb 
fbetoag; accordingly, to their indiv^idua] selrea. The queedons 
connected with them which most immediately present thca^H 
selves are thc^ : Has the sun compunioDs ou its journey, c^l 
does it travel alone ? and. Are real stellar displacemeDte 
govemed by any obvious law 1 ^m 

The great multitude of the stars are, to all appcaranoi^l 
indifferent to the transport of our system. They have clearl/ 
no share in it. Just because they stand aloof, and act a? 
indicators of the way, its progress becomes eensible to us. 
For motion is not alone undiscoverable, it is even unimagin- 
able without some fixed point of reference. Yet we caQoot 
pronounce with certainty against the existence of a particular 
dynamical bond connecting the sun with some few of the stars, 
which form with it a company associated by subjection to 
identical influences, and engaged on the same journey thiough 
space. As to the criteria by which such associated stars, if 
present, can be discriminated from the rest, something will 
said in the next chapter. 

There, too, we will consider what answer should be given 
to our second query. A great deal depends upon it as regards 
our conception of the sidereal universe. Nay, the result of 
inquiry upon the point has a vital bearing upon the subject 
we have just attempted, however inadequately, to deal with. 
For the assumption that the absolute movements of the stars 
have no preference for one direction over another forma tb^fl 
basis of nearly all investigations hitherto conducted into th6^ 
translatory advance of the aolar system.* The Uttlc fabric of 

' AitTOph, Jimrn. vol. liii. p. 87. ^^ 

* BrBT^ig, it LB true, diaciird«d th« hypothesis of casual propor luotiona, bq^B 
hftd to flubatituta for it (^uasLiouiible aBsuniptions regmrding stellar mosses, dis- 
t&Dcea, and velocities. -Vaurna2 de LiouviiU, t. riii. p. 1S£. [Sen Kaptej 
AstT. NtKh. No. 3723.) 



1 




TRANSLATION OF THE SOLAR SYSTEM 315 

laborioosly acquired knowledge regarding it at once crumbles 
if that baaiB has to be removed. Profitable inquiiy into the 
circumstances of the sun's journey have b^n rendered 
feasible by the supposition that, for tiie purpose in hand, the 
movements of the stars may be treated as casual irregularities ; 
should they prove to be in any visible degree systematic, the 
mode of treatment adopted becomes invalid, and its results 
null and void. The point is then of singular interest ; and 
the evidence bearing upon it deserves our utmost attention. 



CHAPTER XXIV 



THE PROFKR MOTIONS Or THS STAJtS 



Wma^ the relative poaitiona of the Btarg are compared at 
cotialderable intervals of time, they are i& many- oases found 
to have uadergoue amatl, but unmistakable cbangea of a 
Beemmgly capricious chamcter. ^eae are termed * proper 
motions'* to distinguieh them from merely nominal ahiftings 
due to the slow variation of the points of reference which 
[serve to define the places of all the heavenly bodies as seeu 
projected oa the inner surfaoe of an imaginary concaye epb^^ 
Proper motions are by no means easy to get at. Only from 
the most delicate observations, and with stringent precautions 
for bringing those at distant dates nnder preciaely similar 
conditioua, can thoy bo elicited with satisfactory accnracj. 
Otherwise, some trifling systematic discrepancies in the com- 
pared catalogues, or oversights in corapufeition, might simulate 
genuine effects of movement, with disastrous influence upon 
Bideieal investigations. Hence, proper motions cannot gener- 
ally be regarded as established unless, in addition to the 
terminal observations showing a sufficiently marked change 
of place in the course of thirty, fifty, or one hundred years, 
at least one intermediate observation is at hand to prove that 
the suspected motion has proceeded uniformly in the same 
direction, and is accordingly not the creation of peEwmal 
instrumental inaccuracy. 

Although not one among the scores of millions of stars 
can, with any show of reason, be supposed at rest, only 
ten thousand of the stellar army have, up to the preaent, 
shown measurable and progressive displacements.* Many of 

^ DyaoD, OifMr%)atorfft vol> xiviii, p> 275< 
316 





THE PROPER MOTIONS OF THE STARS 317 



ieae, icdadiug nearly all the l^ieidce of the northern hemi- 
sphere, were observed by Bradley between 1750 and 1762 i in 
the aouthem, Lacaille's amiultaQeouB tabours aerve to authenti- 
cate the changea of some three-score of objects to which he 
devoted especial care- So that a large stock of highly accurate 
data already posseflses an antiquity of ooe and a half centuries ; 
the cataloguee of Piazzi^ Lalande, and Groombridge are of two- 
thiide that age ; while multiplied subsequent obaervationa 
afford a further supply from which fresh and well-detennined 
proper motions ai© continually being harvested from the seed 
planted by an earlier generation. 

The aspect of the heavens iy, to the unaided sense, virtually 
unchanging. The conetellations disclosed at the present time 
by the nightly withdrawal of the veil of twilight would be 
famiLar, cotdd they revive to survey them, to the watchers 
irom the towers of Babylon. And most of the star-alignmeDts 
given in our text-books might be as useful to students of 
celestial physiognomy a couple of thousand years hence as 
they are to-day. Every one of the indicated stars will indeed 
most probably, by that tin\e, have shifted ita place to the 
extent of many thotisands of millions of miles. Yet so over- 
whelmingly vast is the sidereal ecale that thousaude of nullions 
of miles measured upon it sink into insignificance. 

Stars advancing in a century as much as 30", or about ^ 
the width of the full moon, are counted rapid travellers ; 
and the swiftest class, with secular motions of 100'' and 
upwards now embraces about one hundred members. Each 
of these, were it bright enough for ordinary perception, 
would in a couple of millenniums become very sensibly 
displaced even to an unskilled observer. But only a smalt 
proportion of the quickest starB are visible to the naked 
eye, and only ten reach the fourth magnitude ; hence their 
shiftings make very little difference in the general e£fect of 
the starry skies. 

As might have been expected, the stars in most rapid 
apparent movement are among thoae nearest to the earth. 
Vicinity, in fact, and ang^dat velocity vary together. Die- 
placements on the sphere are large just in the proportion that 
the distances of the objects travelling identically in space are 
small. Were there any approach to uniformity in the real 



iia 



THE SYSTEM OF THE STABS 



velocities of the stars, we cauld then f&irly estimate, tMSn 

^ their seeming movemeDte, their relative aituatiotia as legudi 

' onrselvaB. But there is do auch approach to tmifonoi^. 

Boundless variety prevuihi hero, as in every other branch 

.of sidereal Btatistic& Stars with large proper motions ttt 

Bometimea enormously remote; and, if stars with large pftril' 

laxee and little or no movemeot have not been diacovercd, it 

is perhaps because they have not yet been looked for. Their 

occurrence, for a reason to be presently explained^ would be d 

great interest^ and is not unlikely to be certified by mewure- 

ments on photographic plates^ 

But, however great the range of variety, it aeems certain 
beforehand that, on the whole, the amount of visible motion 
in % given number of stars must decrease as their distance 
incre&ees. And since their brightneas falls oCT at the «aiue 
time, although mudi more rapidly, there appears no escmpt 
from the concluaiou that motion and magnitude must, on i 
wide average, vary together aoeording to a definite ratid 
From stars of the sixth photometric magnitude^ for inatAnce, 
we receive only oue-huudredth part of the light seat to us by 
atara of the first magnitude ; they must then, one with another, 
be ten times more remote. Otherwise, we should be driven to 
the unwarrantable aasumptiou of a systematic diflTerence of real 
lustre between apparently large aud apparently small stars. 
But, if the average distance of sixth -magnitude stars be ten 
times, then their mean motion should be only one-tenth that 
of stars of the first magnitude. Yet in point of fact thia 
is not BO, The proper movements of classes of stare diminish 
indeed very notably with their brilliancy, but not in the 
computed proportion. The discrepancy deserves attentive 
study. 

The average proper motion appertaining to the sixth 
magnitude, aa determined directly by M, Ludwig Struve 
from 64Y of Bradley's stars, is 8" in one hundred yeaiK* 

■ Ten times this quantity^ or 80", ought to be the average move- 

■ ment of stars of the first magnitude. But the mean derived 

■ from the actually observed shiftinga of the twenty brightest 
I stars in both hemispheres is only ^O". 
B Stars of the second magnitude are still more noticeably 

H ^ M^meir«4 tU St, POtrthourgt t, xixv. Vo. 3. p, S. 




THE PROPER MOTIONS OF THE STARS 319 

ert. They should be, on the photometric scale, 6"3 times 
to the earth than stars of the sixth icaguitude. This 

ovld give, for theix mean eecular motionj 8''x 6"3 = 50"'4. 
Twenty - two such stars, however, from Bradley's and the 
Pulkowa catalogTieSj show no more than 17". And even this 
low figure more than doubles that repreaeutmg the average 
movement of forty-two eouthem etars^ of 1"7 to 2'7 magni- 
tudes, forming a descending Bcquence with the ten of first 
magnitude. Nor is this average improved by considering only 
the first twenty on the list, from ^ Crucis of 1*7 to k Orionia 
of 2'2 magnitude. The swiftest of these (7 Crucis) travela 
only 20" a century; taken all round, they move S'^ or with 
exactly the speed of stars presumably more than six times as 
rem^ote J 

The anomaly of low apparent velocity is accentuated by 
the close agreement between M. Struve'a results for stars from 
the second to the fourtli rank inclusive. A glance at the 
accompanying Table from his Memoir will Ber%*e better than 
verbal explanation to make the matter intelligible. The object 
of its compilation was to exhibit the divergence between the 
proper motions actually determined and those computed from 
the basis of the mean secular displacement corresponding to 
the sixth magnitude. In the fourth column, however, we 
have substituted figures derived from strict photometric star- 
diBtancea for others depending upon a scale of distances in- 
volving dubious assumptions. 



Table of Stcvlar Mean Proper Motions of <dl BradleyU Stars 
differing hy not leas than dffht-tenths of a ttiagnitude. 



M&gbitudt. 


No. of Stan. 


Mmn MotiCNL 


Obwrrvd. 


OompnifKL 


1 

9 
8 

4 
E 

ft 
T 

8 


Q 

22 

61 

lOfl 

S18 

64? 

92 

11 


l7"-2 
ir-5 

16"'2 

S'-a 
r-8 


80" -0 
60'-4 

sr-a 

20*^8 

8"-0 

ro 

3*'2 



320 



THE SYSTEM OF THE STABS 



It will be obvorred thjit the Telodty of each order farigbWr 
thfta the mxth filU siunt of ita thMvetical ■mooofc, while tlMt 
of the Cftinter onieni exoeedfl it. We hflflten to add that {m 
IC StruYe pcanU oat) little or no dapendaaoe can be pUcsBil 
on the Above mean rate of eighth^mgnitiide motion dedaoad 
from measoremente of otdj elerea objecte. 

And DOW, what are we to think ? How can we aoconnt 
for the indicated deficieiM^ of proper motion in the brighier 
aUra t Three poesible expl&fiatiODs pneent themselves It e 
eonoeirable that atara, ray of the sixth and aevecthf axe leaUj 
mailer ot dimmer bodieB on an average than ataxs of the fiiBt 
and BBOond magnitodea. and ax9 coQaequeatfy lea retoote than 
they shotdd be o& the mote uatuial supposition of their 
equality. Their diminished distaace would then at oaw 
render their extra celerilj intelligible: Again, thero ma^ be 
a systematic Increaee of motion outward from the son, pro- 
ducing m the fainter ators preponderating latee of displace- 
meat. Or thirdly, there may exist a special cUs oi stna 
deficient in light^power, but trarelling with exceptional speed. 
by the inSueoce of which the balance of seeming ewiftaeaB ia 
Corned in favour of the less brilliant ciaases of stars. 

Of these alternatives, the second may be dismissed ai 
being at variance with our best iuformatiou. There remain 
the Srst auil tliifd, neither of which, certainly, represents the 
whole truth, although each not improbably corresponds to 
some p&rlial aspect of it A hannonious adjustment of their 
olaima to credibility is thus difficult to bring about amid the 
joBtling of mutually adverse inferences Professor Kapteyn 
boa efttabliahed, by diligent sorting and aifling of multi- 
tudinoa? data, the broad principle that proper motion varies 
in amount in the due proportion of di&tance. But he has 
also established that magnitude is no safe index to distance^ 
apart from the discrimination of spectral typ>ea Stars of 
Sirian and helium quality taken all round ore, in fact, 2*7 
times more remote than solar stars of the &ame pliotometric 
rank. The former accordingly shine with a seven -fold 
intensity comparatively to the latter. Hence, variations in 
the distribution of the spectral genera may greatly perturb 
the methodical progression of observed proper motions. Here 
is one obvious source of anomalies ; and the large relative 



THE PKOPEE MOTIONS OF THE STAES 321 



iber of " white " stare included among the four or five 
dozen most liisfcroufl gems of the firmament (aee Appendix, 
Table I.) should be noted in this connection. 

It muBt farther be inquired whether stars of all classes, 
and eituatad in all parte of the sidereal world, are equally 
mobile ? Kaptejn and Newcomb agree that systematic 
differoBcea are unapparent ; but line-of-sight measures fumitih 
evidence of a more direct kind than that examined by them. 
Now, in classifying the radial velocitiea of the 280 Btara 
employed in detennining the solar translation, Profeaeor 
Campbell found their rate to increase notably with faintnesB,^ 
Within the acope of his investigation, the result was assured ; 
but its scope was restricted. Individual peculiarities of the 
etara embraced by it may have been concerned. This requiraH a 
word of explanation. Measra, Frost and Adams have gathered 
strong indicationSj from their work with the Bruce spectrograph 
on the 40-incb Yerkes refractor, that helium stars really travel 
in space more slowly than other atara.^ Hence, if objects of 
this kind were represented more largely among the bright, 
than among the fainter atars on Professor Campbell's list, the 
discrepancies in their ratea of movement would be accounted 
for. That their true explanation must be sought in some 
such apecial circumstance rather than in any systematic 
diversity, is practically certified by exhaustive statistical 
inquiriea, and more particularly by Professor Comstock's 
study of a collection of atara ranging from the ninth to the 
twelfth magnitude.* Their average linear velocity, estimated 
by Kapteyu's unexoeptionable mefchoil, agreed almost precisely 
with that of CampbeH's stars, notwithstanding their photo- 
metric inferiority to them, in the mean, by no less than six 
magnitudes. Summing up, then, it appears that atars of the 
same spectral type, wherever situated, possees, in a general 
way, similar movements ; but that helium stars are perhaps 
genuinely slow in pace ; while Sirian stare seem more leisurely 
travellers than solar, merely because they show equally bright 
at a much greater distance. 

Yet the preponderance of small stars in every enumeration 

^ AtiToph. Joum. vol, xiii p. 85. 
* Ibid. ToL irii. p. 246 ; Deceit. PuW. toL ?iii. p. 100. ^ 
' AMtT, /own, Ko, 6&S. 



322 



THE ST81X1C OF TfiS SUJtS 



«f npid pflopor matiam m a cbaBsBng ^eu Its 
cspla^iiioB ■ that the «veBif» of iMllBr ue 
tritoo-fODi an few. Miaaiyw WBrnanmrn. BdmUj, iMiveM; 
fchsr dfificiflDCT >• in liciift mthfr Uuii itt aMM&. A nsik 
€f Um evidme g bouing on the point M cwllBrted in Table ?l 
of oar Ajfptmdix, which gifva the nugaifTuiiw aad motkai «( 
th« Ukiitf qvkkMt alan oC ov aeqvn&taaea. Move thaa ktt 
sfe iaviaCble to th^ aaked eje ; tha four H«^^"tg the Ixi hiR 
A mflHt *"*c"i*"^ of 7*3 ; no kai than fclercD XBDge fnm ^ 
«i^itfl to hdow the ninth, while oolj two stazB of the in^ 
n9n4 oC the aeooiMl mod one of the third magniiade are ss- 
clttdfld in the oooqwctittL The largest proper mocum yH 
detected bdoc^ to an 8'5 magnitade ataf ntoated io tkt 
•oathem ooiiftellatioD Pictor. Its diacoTerj in 1897, in 
which foptejTQ and IniuM co-opented,^ was one of the 
frnita of lidereal knowledge plocked during the 
of the Cape Photographic Dui^limuistenmg. It aupeneded? 
the chaaipion raoer^ aod far outfitripa 1830 Crooinhndge, ■> 
ijurigniiiuiint star in the Great Bear^ picked out b^ ArgeUudsr 
in 1 842 for a rate of progress which would carry it in 185,000 
ymn round tJie entire sphere, or in 265 over as much of it as 
the fiUQ*8 diameter covers The corresponding aunual advaace 
amounta to 7" ; a-nd it ia very nearly equalled by two etoaU 
southern stars observed by Gould during bis stay at Coxdoba. 
One is a 7' 2 magnitude star in the Southern Fish (Lacadle 
9362), the other, one of 82 in the conatellation Sculptor. 
Next on the roll comes 6 1 Cygni with a proper motion of 5**2 J 
and a Centauri, with Z"'?, h&s eleventh placa Double stare 
are Gr^uently conspicuous for rapid movement; and it is 
noticeable that three out of the four tirst- magnitude Btare 
with proper motions exceeding one second of arc yearly^ — 
namely, Siriua, Procyon, and a Centauri — are binaries. 
Struve's geuerul inference aa to the quicker translation of 
multiple than of simple objectSj has received much support 
fitim further experience.* 

The " proper moUoua " of stars include, as was explained 
in the la^t chapter, an apparent, as well as a real element ; 
they consist, in technical phraseology, of the motus parai- 

' Attr, Natfi. No. 8468 {Kipteyn) ; Observtiit>ry. vol. iiiL p, 99 (Gill}. 
■ Oftvin J. Burns, AUrvph. /oifrrw. toI. iviL p. 67. 



THE PROPEK MOTIONS OF THE STARS 323 



iaeticus, optically transferred to the whole stellar multitude 
from the single real motion of the suit, and the m^tus 
peculiaris, belonging to each individual Btar. The Beparalion 
of these two constituents, blended together on a cursory 
inspection, is a necessity for progress. This is what Kapteyn's 
mode of analysis aims at, and in a measure accomplishea 
For the motion of a star in a direction petpendicalar to that 
of the eun'a route is altogether its own ; it ia an objective 
fact ; and the comprehensive study of such components of 
stellar velocities, purified by a sort of " fractionation " from 
parallactic ingredients, afiforda undoubtedly the most promising 
means of gaining insight into the general plan of celestial 
revolutions. 

But the motvs pecidiaris itself ia only a projection apon 
the sphere of a line of ti-avel which may make any angle with 
the line of sight, Its amount then varies with direction no 
less than with distance sxkd actual velocity. A star may 
appear devoid of motion simply because the whole of it is 
" end-on " ; while the movements of others seem large because, 
lying square to the line of sight, they are completely effective 
for apparent displacement. Here, just where ordinary obgerva- 
tioD ia baffled, the prismatic method cornea to the rescue. The 
spectroscope " takes up the runniog " for the telescope. 

The alliance could not have been rendered effective but 
for the momentous improvements effected late in the nine- 
teentii century in the processes of celestial photography. The 
prerogatives of the cumera in this line of work are euormoua 
Not only do the worst mischiefs of atmospheric disturbance 
vanish with its employment, but the upshot of measure- 
ments executed upon one line can be cheeked or ratified 
by comparisons with other lines in the same spectrum, and on 
the same plate. Where motion is in question, all must be 
equally affected by it; hence perfect security against illusion 
is afforded. The full realisation of these advantages through 
Vogel's skilful use of the spectrographic apparatus erected 
by him at Potsdam in 1888^^ thus constituted an advance of 
_fiisb-rate importance in practical stellar astronomy. 

The example thus set haa been widely followed — in 
T&jnerica, at the Cape, in Europe at Cambridge, Pulkowa, and 
1 ManaUberieiitet BerLiQ, March Ifi, 188@, p. 367. 



zu 



THE STSTEM OF THS STAfiS 



thtfter; hm 




ttfloed, Mtd bAVtt beeD Adrerted to xa aa 
tbcj anna mortif is ""■■■"^■'^ giabntie^aad 
not w cActm ■• tbe applei of Atikate ID divertnig 
from on wltimiitft goal B aoe miie B mto 
an iwinrtiTTy of a iMaitiod cbttwter; Ukj &iMnd Ihsi 
IUHljff of d^a^ itrwt tiUdOB w w M^ f titT^ zadial tneiocitieB nnt bs 
mdend ix>inpanble in sbundancr to thaee mlremdj •ocamslited 
of the projected moremeDte of tbe «Ui% b«ibce ooch reseudMS 
Oia be proRQcuted virith entaro roccwa EvcataaDf, too, radial 
velocity can be made to aerro «a a teat of mean pajmUax. It 
tmy lairly be amimed that atan tx&Tal on && avenge with 
eqiUl speed along each of the three otMndiiutis of caldcil 
space. Comparing then the mean radial speed in miles per 
Mcond of a given group of atani with their mean displace- 
ments in aeoonds of arc along circles of declination or rigbl 
aacetiAion, we at once learn their mean distance, since we thus 
virtually tranaUte angol&r into linear displacements.^ It is 
true (oA can be seen &om an inspectioQ of Table TIL in onr 
Appendix) that much swifter rates of stellar transport have 
been determined in a tangenti^ direction than in the line of 
night ; but tbe ineqnsJity m&y be an accident of investigaUoQ 
which its further progress will redress. Meanwhile, there is 
much reasou to believe that no general disparity of the kind 
subiiiAtfl, Kitimates of the average total velocity of tbe stars 
in space were derived by Kapteyn and Kewcomb from their 
observed proper motions, and by Campbell from their measured 
ratM of upproacb nr recession. They agreed almost exactly. 
With reference to a fictitious atationary centre of tbe stellar 
asoumblage, a niedtutn star is displaced by 21 or 22 miles 
a Bocond, V2^ miles being the rate of our sun. We are 
accordingly borne along in the train of a somewhat sluggisb 
lumiuiiry." 

Table VIT., 2, gives a list of etara with thwartwise mov«>- 
tnotita exoeediiig 25 miles a second, They are sometimes 
ciiUi^d " runaways," because their headlong course seems hardly 
compatible witli entire subjection to the sway of gravity. 

The first of theee startling examples to become known 

1 KluiW, Jsir. ^<K}t. No. 3037, 
• Nowcomb, The Starts p, 3(M. 



THE PEOPEE MOTIONS OF THE STAES 325 



1830 Groombridg«> The large proper motion and email 
a.llax of this Stat compel the ascription to it of a speed — 
dng into account only that part of it lying square to oqr 
iew, of at leaat 150 milea a second — a speed uncontrollable, 
aeeording to Professor Kewcomb, by the combined attractive 
power of the entire sidereal tuuverae. For hia calculations 
show that the maximum velocity attainable by a body falling 
from infinity towards and through a system composed of 
100,000j000 orbe, each five times aa massive aa our sun, and 
diBtributed over a disc-like space 30,000 light-years in ex- 
tent, would be 25 milea a eeoond.* But 1830 Groombridge 
poeeesges more than six times this speed ; and because velocity 
varies with the square root of the attracting mass^ a world of 
stars of fully thirty-aix fold the potency of that assumed as 
probable woiUd be required to set this object moving as it 
does unquestionably move ! 

Kow the velocity producible by an attractive syatem is the 
limit of the velocity it can control — that is, bend into a closed 
curve. It is then certain that unless the stellar system 
possesses what we may call occult gravitational energies, the 
star in questiou cannot be one of its permanent members. 
Virtually in a straight line and without slackening, it will 
pursue its coui'se right across the starry stratum it entered 
ages ago on its unknown errand, and will quit ages hence to 
be swallowed up in the dusky void beyond. There is, how- 
ever, an alternative supposition. The star may be acted upon 
by unknown compulsive influences. 

Lord K^elvin has lately estimated the quantity of gravi- 
tating matter needed to produce velocities such as the 
generality of stars exhibit. He finds it to be that of a 
thousand million suns like our own, evenly distributed 
throughout a sphere wi th a radi ub of 3262 light - years,' 
A prodigious lapse of time would further be needed for the 
development, by continuous acceleration, of the observed 
higher rates of speed ; nor can we be quite sure that even 
these heroic measures would suffice to produce them. Yet 
they are patent, and not extremely scarce. Groombridge 
1830 is no longer the only "runaway" of our acquaintance- 

* Pop, Asttanomy, p. 4ti9. 
^ RtpoTt Brit, Aa. 1901, p. £63 ; Natwrt, vol hdv, p. 02«, 



326 THE SYSTEM OF THE STABS 

linear stellar speed, apart from that share of it directed 
along the line of sight, exceeds Newcomb's computed maximum 
of 25 miles a second in about one- third of the cases in which 
it has been ascertained, and the excess is here and thete 
very large. Arcturus, for instance, travels at the tremendoos 
rate of 257 miles a second, /& Cassiopeiie at 113 ; while fire 
southern stars progress by 60 to upwards of 80 miles per 
second. 

" Fljing stars " can then no longer be regarded as men 
intruders into stellar society. Whether or not belonging to it 
" for better for worse," they evidently at present form an im- 
portant part of it, and the problem they present cannot be 
excluded from a general consideration of sidereal mechanism. 
Indeed they furnish a most significant index to the working 
of its secret springs. They pursue their careers, so far as 
observation can yet tell, in right lines, and at a uniform speed 
Their high velocities would be otherwise less perplexing ; for 
they might plausibly be attributed to the powerful attraction 
of invisible bodies in their neighbourhood, representing, by 
analogy, tlie rush past the sun of highly eccentric comet& 
But the evidence is wholly against any such hj'pothesis. All 
pro}>er motions known to us — whether of single stars or 
of the centres of gravity of multiple stars — are sensibly 
rectilinear. The centres of curvature, presumably, of the 
imaginary lines traced out by them, are inconceivably remote. 
A straight line is only part of the circumference of a cirtle 
of infinite radius. 

The fact accordingly confronts us that not a few of the stars 
possess velocities transcending the power of government of the 
visible sidereal system. Is that system, then, threatened with 
dissolution, or must we suppose the chief part of its attractive 
energy to reside in bodies unseen, because destitute of the 
faculty of luminous radiation ? The presence of many such 
bodies is unquestionable ; and those of which we take in- 
ferential cognisance may be few compared to the multitude 
wholly beyond our ken. The power of the univei'se is certainly 
reinforced by the dark stars it includes ; but to what extent is 
unknown ; and the uncertainty helps to maintain inviolate 
the final secrets of our cosmical environment. 

Physical peculiarities are not, in any obvious way, related 



THE PROPEB MOTIOJfS OF THE STAKS 327 



excessively rapid movemeiits. Arctiiraa ifi a solar star, 
sowing promineut titanium-absorption, and apparently well- 
ivanced towards the Antarian stage. Its masa is presumably 
enormous, since its light-power must be equivalent to that of 
Cwelve to thirteen hundred suns! Its nearest competitors in 
swiftness, 1830 Groombridge and ^ CassiopeijE, are, on the 
other hand, comparatively unpretending orba ; and neither 
differs markedly in spectral quality fi-om our aan. Indeed, 
the eolar type appears to be more often asaociated with Iiigh 
velocities than any other. Stars with banded and gaseous 
spectra, and variables of all elates mostly exhibit bub slight 
signs of displacement ; but this ta^y be aji effect of remote- 
ness, rather than of genuine inertness. 

An unmistakable connection, however, exists between 
proper motion and sidereal locality. The late Mr. Proctor 
drew attention to the prevalence in certain regions of the sky 
of what he termed " etar-drift/' ^ Here and there, unanimity 
is, to some extent, substituted for tlie caprice superficially 
characteristic of the " peculiar " movements of the Btar& 
Amid seeming confusion, order and purpose by glimpses 
reveal themselves. Battalions of stars — ■" flying synods of 
worlds " — regardless, as it were, of the erratic flittings of the 
casual surrounding crowd, iDarch in widely extended ranks, by 
a concerted plan along a prescribed track, under orders sealed 
perhaps for ever to human intelligence. 

Among the stars situated between Aldebaiftn and the 
Pleiades, there is little relative movement. They all drift in 
company towards the east, by about 10" to 20" in a century. 
Not, it is true, along strictly parallel lines. A fresh investiga- 
tion by M. Weersma* of 66 members of the Hyadea group has 
brought to view considerable divergences. Dr. Downing's dis- 
ouaalon/ too, intimates very clearly the division of the main 
cluster into subordinate families, travelling to some e^teut on 
their own initiative. The general tendency of their course is 
no doubt largely due to the sun's oppositely directed progress. 

Five of the " Seven Stars" (sepicwt triones) forming the 
Plough (those excluded being the " Pointer " next the pole, 

' Froc. RoU' Society^ roU XTJii. p. 109. 
* OrfmiTigen Publications, No. 13, 11iK)4> 
* JoMrw^ Srit. Attr. Am. toI, xr. p. 2S. 



323 



THE SYSTEM OF THE STABS 




aud 7} at the extremity of the handle), wct& regarded by Ul 
Proctor m members of a vast united group ftdvancing witlt, 
though outpacing the sun. This remarkable inferanoe hw 
been endorsed by ProfeaBor Newcomb ; * and the reality d 
the visible flow of movement was substantiated by specoo- 
graphic evidence obtained at Potadam, showijog that the fltaiS 
in {jueetiDn are approaching the earth with a common veloeLiy 
of about 18 miles a aeconi* The system they form has been 
roughly estimated by Dr. Hoffler * to bo eitn&ted &t & distance 
of 200 Ught-yeara, whence it followa that each must greatly 
exceed our sun in ladiative splendour. One of these linked 
orbe^ moreover, carries with it, as our readers are aware, three 
dependent atars, namely, the Eider- star Alcor* a slowly re- 
volviog telescopic attendant, and a more intimate associate 
apectroecopically revealed. 

It ia scarcely likely that the combination is self- 
centred. Concordant motion does not necessarily imply the 
mutual revolution of the objects to which it belongs. What 
it does imply is their dynamical connection. But that 
connection need not be of the kind exemplifed cloee at 
hand by the earth and moon. It may rather be such as 
prevails between the earth and Venus, or between Jupiter 
and Saturn. The group in Ursa Major, it is safe to aaeert, 
includes examples of both kinds of relationship, Of the 
movementfl of two satellites, Mizar (if Ursfe) is the un- 
doubted mainsprmg. The statue of AJcor is dubious. Its 
path at present appears strictly rectilinear; but latent curva- 
ture relative to the large adjacent star may in time become 
sensibla About the personal independence, however, of all 
the rest of the company there can be no questioD. Although 
dominated by the same influence, they advance each on its own 
account ; nor can their relative situations be looked upon as 
beyond the reach of change. Ultimately, the bond of union 
between them will perhaps even cease to be traceable. Slight 
ine<iualiti©B betraying differences in the period of revolution 
round the same remote centre may eaaily co-exist with what 
is known as common proper motion. Such discrepancies can 
alone hold the stars affected by them aloof from binary com- 



1 Th* Star$, p. 80. 



* H. C. Vagal, P^lietUiown, Bd. rii Th. L p. 164. 
» Aatr, Naeh, No. 34B6. 




THE PEOPER MOTIONS OF THE STARS 329 



binatioD. Whila travelling along parallel lines, they have 
BtiU a relative velocity exceeding^ at their distance apart^ the 
power of their mutual gravitation to Bway into an ellipse. 
One must hence fall very slowly behind the other, as Saturn 
falls behind Jupiter after conjunction. Evidence of their 
afi&nity ia then only temporarily acceaaible to ua. After 
many ages it will evade recognition. There may be, 
probably are, in distant parts of the sky, stars revolving in 
boundle&aly spacious orbits round the same focus of attraction 
with the stars of the Plough ; bat we have no means of 
identifying them. 

" Partial systems," governed presumably from without, are 
of tolerably frequent occurreoce. The first to become known 
was discovered by Bessel in 1818.' It is composed of a fifth 
and a seventh magnitude star, known respectively as 36 A 
Ophiuchi and 30 Scorpii, thirteen minutes of arc apart, yet 
endowed with an accordant movement of l"-25 yearly. The 
former star has a close attendant ; and an intermediate minute 
object also forms part of the company.^ Another interesting 
quadruple group was detected by Flammarion in IS??." Two 
couples in the Swan, one revolving^ the other in appearance 
fixed, separated by an interval of 15', drift together slowly 
Bouthward in a direction nearly perpendicular to the Una of 
march of the sun. Their movement is hence " proper " to 
themselves, perspective effects being unconcerned with it. 
The stationary pair is the fifth -magnitude yellow star, 17 
Cygnij with its bluish satellite at 26"; the circulating pnir 
consists of two eighth-magnitude stars at 3'', numbered 2676 
in Struve'a great Catalogue. 

A curious instance of concerted movement is afforded by 
two ninth-magnitude stars in Libra, discovered by Schonfeld 
in 1881 to progress across the sphere at the exceptionally 
quick rate of 3"*7 annually.* Notwithstanding the wide 
interval (6') separating them, their advance seems perfectly 
harmoniouBL They flit side by side, as if rigidly connected 
across a chasm probably some thousands of millions of miles 

^ Fiittdamentit Astf<»wmia, p. 811. 

^ Flannnanon, C^mptcM Bendw, t Ixxxr. p. 783 j Cat, dit ttsikt Do^Ma^ 
p. 106 ; luiies, Rt/ertnci Cat. p. 170 A, 
' Comptea Htmdm, t lixiv. jj, BIO. 
* SiUun0sb«richte Niederrhtiniicht Ot$. Bonn. 18S1, p. 173. 



8S0 



THB STSTEIC OF THE STABS 



jo width. A sttU wider sjrstaii* noted by lit. Ibih^ 
falPMl bj two stAT-pain m Toiicazi.' la neither cMe 
mauHree for parallax jet been executed. 

To the queetioii^ — Ha» the van anj osBOciate in his , 
through space? only a prorisonal answer can be 
lXtfo6 are known, bat inreetigHtioDS on the point are lanlj 
tiAnent. The pecoliaritieB which we should expect befarebAud 
to attend such campamon-stars^ are comparative proxiouij 
Md l«latiT« immobililTf. They should hare sensible paraflawi 
and be deroid both of radial and tangential velocity. Xeitlur 
apectroaoopie nor telescopic eridenoe of motion should Iib 
derivable from tbem. No star up to thia thoroughly ^^awiinM 
combines thef>« characters -, but tben they could not pom^j 
be found in the " proper motion '* stars chosen by prefsrenoe 
00 the Gubjects of parallactic obs^rvationfl. Only in one of 
the Btellar points employed for comparison with 61 Cygni h^ 
their presenoe been suspected. Dr. iSchur measured^ in 1899, 
what seemed like a parallactic «hift of this nameless star tfi 
the extent of 0"'6 j* and Mr. Grorameliu* having deduced far 
it an evanescent proper motion, threw out the suggestion that 
it might belong to the sun's bodyguard.* The possibility, 
however, depended upon the verification of the parallax* which 
ifl not known to have been effected. When mora has been 
done io photographically registering Une-of-sight movements, 
Btars may perhaps be discovered sensibly fixed as regards the 
aun, because borne along with him at the same tr&nslatoiy 
speed The construction of Buch a group, and the diBtinctive 
charactorisdtiou of its members, might open up a fascinating 
branch of inquiry. But its methods cannot well be eatabl iabed 
until its subject-matter is rendered less evasive. 

If the system formed by the stars be destined to 
nenoe in its present shape, some general law of movement 
must be obeyed by them. Even if its state be one of pro- 
gressive modificatioHj a definite mode of change ought to 
become apparent. Local irregularities, however, so effectnally 
di^^se the fundamental harmony that its prevalence may long 
continue a matter of speculative belief. 

The assumption is indeed indispensable^ as Dr Schonfeld_ 

1 Se» mU4, p. 97. ^ Aatr. Naek Ko. 36ftO, 

■ Obtervatory, voL riii. p^ S76. 



ibli^e^i 




THE PEOPEE MOTIONS OF THE STARS 331 

minted out in 1883,^ that the motions of the etars are some- 
related to the plane in which the vast majority of them 
are diepoaecL For otherwise their actual conhgurfttiou would 
be a wildly improbable accident of the time in which we liTe. 
The Milky Way, to put it otherwise, should he regarded as an 
evanescent phtiiiomenon, unsufltained by any persistently 
acting forces, the outcome of a hundred millions of casual 
conjunctions. If this be jucredible (as it surely is) then we 
are constrained to admit a preference^ in the long run, among 
stellar displacements for the grand level of stellar aggregation. 
The Milky Way must be, in some true sense, what Lambert 
called it one and a half centuries ago — the " ecliptic of the 
star^/' 

Sir John Herschel imagined the law of harmony to consist 
in a general parallelism of stellar motions, involving a kind 
of systematic circulation, as of a solid body round an axis per- 
pendicular to the galactic plane. Innumerable exceptions to 
any such rule are of course to be found, but they were 
assumed, in the upshot, to be mutually destructive, the main 
•* stream of tendency " 6owing on irrespectively of them. But 
it ie difiScult to conceive a physical basig for a quasi-rotational 
system wholly without warrant from experience. More 
plausible 18 M, Ludwtg Struve'a view that the main part of 
the revolutions of the stars round theii* common centre of 
gravity situated in the Milky Way, are performed in planes 
slightly inclined to that of the zone towards which they are 
concentrated.^ His attempt, indeed, to elicit a "rotation- 
component" from the secular movements of Bradley's stars, 
proved unavailing. Yet this is not decisive against the truth 
of au hypothesis compatible with a btilanced stellar circulation 
pursued in opposite senaes. An apparent drifting movement, 
detected by Sir David GUI in 1902.^ of the brighter stars 
witliin a southern zone 12^ wide relatively to their fainter 
associates, involves considerations of a different kind. It haa 
yet to be substantiated. Every imaginable precaution was 
indeed taken against insidious errors ; yet such a phenomenon, 
if genuine, should almost necessarily be universal; and Pro- 



» V. J. S. Aair. G«s, Jiihrg. xvii, p. 2B&. 

* Mtmaires d« Si. FHirabotiTg, t. xxxt. No. 3, pp. C, 19. 

» Attr. Nach. No. S9i>0. 



332 



THE SYSTEM OF THE STABS 



feseor Turner^ failed to derive from the Oxford zone-plates 
convincing evidence of iba extension to northern gtars, 

M. Bancken dealt in 18S3 with a strictly selected list of 
stars.' He admitted only those within thirty d^reefi on 
either side of the Milky Way^ and possesaed of annu&l proper 
motions not exceeding a quarter of a second. The solution of 
his eqtiationa showed these movements to include a common 
element of very alow progressive increase of galactic longitude 
That is to say, the 106 stars considered were being graduallj 
swept along the Milky Way in the direction from Aquila up- 
ward towardiB Cygnus and Ca^iopeia, and down past Opella 
through the Club of Orion towards the Ship. The reality aod 
extent of this flow of displacement will be a matter for future 
investigation. Should the one be confirmed^ and the other aacer- 
taiued, something like a clae to the labyrinth of stellar mo vements 
will have been provided. Perhaps the restricted nature of the 
inquiry contributed to its success. For the exclujsion of large 
proper motions was the most effective mode of aiftiug oat 
stars of the solar type, to which they in the main belong. 
And since this class of bodies are not perceptibly condensed 
towards the Milky "Wiij, they are unlikely to obey any law of 
subordination to it in their revolutions. In this line of 
research, accordingly ^ as in most others connected with sidereal 
structure, the discrimination of spectral types is prescribed 
nnder penalty of hopelessly confusing the issue. 

' JforttWy Koticf:^, Tola. Lriii p. 66, Ixiv, p, 3, 
' Jgtr. Nadi, No. 24831, 



CHAPTER XXV 



THE MILKY WAY 



The Milky Way showa to the naked eye as a vast, zone- 
ehaped nebula; but is resolved, with very slight optical 
assistance, into innumerable small stars. Its stellar con- 
stitution, already conjectured by Democritus, was one of 
Galileo's earliest telescopic discoveries. The general course 
of the formation, however, can only be traced through the 
perception of its cloudy effect ; and thia is impaired by the 
appLcatiou even of an opera - glass. Eendered the more 
arduous by thie very circumfltanoe, its detailed study demands 
exceptional eyesight improved by assiduous practice in catching 
fine gradations of light Our situation, too, close to the 
galactic plane is the most dieadvuntageoua possible for pnr- 
p<jses of survey. Groups behind groups, systems upon 
eystems, streams^ sheets, lines, knots of stars, indefinitely far 
apart in Bpace, may all be projected without diBtinction upon 
the same sky -ground. TJnawarea, our visual ray sounds end- 
less depths, and brings back only simultaneous information 
about the successive objects met with. We are thus presented 
with a flat picture totally devoid of perspective- indications. 
Only by a long series of inductions (if at all} can we hope to 
arrange the features of the landscape according to their proper 
relations. 

To the uncritical imagination ^ the Milky Way represents a 
sort of glorified track through the skies — 

A brand tmd ample rood, whoee dust it gp>\d 
And pavement stara, oa ston to tbec appear, 
Been in the ^bxy, tbat milky way, 
Which nightly aa a circhng zone thou eeest, 
Powdered with stare. 
333 



334 



THE SYSTEM OF THE STABS 



Id Aineric*!!- Indian fftnoy a mystenons " pAth of aavh," 
ito popular Oerman n^tcae, "die JakobsstiBflB^" raealh tk 
Cime when it stood as & celeetial figure of the wm; cf 
pUgtimage to CompoeteUA. Siimlarly. in medifleral ICwgWmi 
it got the title of " WftTfflnghAin W^y ** \jj asBOcdAtion with 
Our Uidy s Norfolk sanctuaiy ; while the Dutch dftHkatwl it 
to St Hilda, and the Finns aet it ap&rt for tha fiittog 
procasaiotiB of birds. More anciently, in the ocder ol 
ideas iuitiated hy the Accadians, it represented ibe mystic 
Snake-riyer of the abyss ' — the Homeric Ocean-stnAin navi- 
gated by Odysseus in his voyage to HadeSu Superficial 
iinprossionfi of homogeneity are. however, replaced, on closer 
inspection, by an aspect resembling rather that of a mgged 
trunk marked by strange cavities and excreaceDces, and sending 
out bmnchea in all directiona. 

The medial line of the Galaxy is scarcely distingnl^iable 
from a great circle* although Professor Jfewcomb's later 
invefltigationa indicate for it a southward displacement of 
nearly two degrees.' This would imply our position to be 
somewhat north of the main level towards Coma Berenices; 
but whether it i& central or eccentric there ie nothing to show 
decisively. Tlie movements of the earth bear uo obvious rela- 
tion to the etarry coUixstion around it. Jfeither the equator nor 
the ecliptic manifeats any trace of conformity to its plane. Tlj^ 
great circle of the Galaxy is inclined about sixty-three degilj^H 
to the celestial equator, which it intersects in the contst^Ia- 
tions Monoceros and Aquita. It passes in Cassiopeia witliitt 
twen^-seven degrees of the north pole of the heavens, in 
Crux, as near to the south pole, while its own poles are 
located respectively iu Coma Berenices* and Cetus. Over 
two-thirds of the celestial circuit, the general tinity of this 
stupendous structure is preeerved. Broken, however, near 
a Centaun by the interposition of a great fissure, it is only 
regained, after an interval of some 120% through the re- 
onion, in the neighbourhood of e Cygnij of the separated 
portions. Involuntarily, the image presents itself of a great 



' iL Q, Allen, Star Sanies and thfir Mranimgs, pp. 37i-79< 
* Gould, (Tranometria Arffentinaf p, 370. 

' AVfiCuw, ToL lix. p. 308. 
* B.A. lib 44» Dec. +2fl' 48' (Kewcomb). 



THE MILKY WAT 



335 



river, forced into a double channel by an encounter with a 
powerful obstacle, the removal of which lower down permits 
its waters to flow together again. The intervening long strip 
of islanded rock and gravel might stand for the great rift 
between the branqhcB of the sidereal stratum, which, although 
to the eye, owing to the effect of contrast with the " candid 
way " on either side, darker than the general aky, is in reality 
nowhere quite free from nebulous gUmtneringa. It ia en- 
croached upon by fringes, effusions, and filaments, spanned by 
bridges of light, and here and there it is half filled up by 
long, narrow, disconnected maaeea, or luminous pools, lying 
parallel to the general flow of the stream. One such 
" brilliant and tortuous streak '' ^ extends, in almost complete 
isolation, over nearly 20'', from the tail of Serpens across a 
corner of the Shield of Sobieski. Moreover^ only the more 
easterly of the two principal branches — that traversing AquLla 
and the bow of Sagittarius — is continuous. The other, after 
covering part of Scorpio " with a complicated system of inter- 
lajced streaks and masses," ^ dies out in Ophiuchus, about 
fifteen degrees south-west of the termination, just at the 
equator, of the arm sent out to meet it through Cygnus. The 
gap is, nevertlielese, partially veiled by a faint luminous 
extension from the souths and shows as absolute only over 
some five degrees of the sphere. 

This ia not the sole interruption to the course of the 
Milky Way. Another, visually, though not photographically 
apparent,^ cuts sheer across the imdivided stream in Argo. 
Here, at south declination 33°, the formation^ Sir John 
Herschel says, "opens out into a wide, fan -like expanse, 
nearly 20° in breadth^ formed of interlacing branches, all 
which terminate abruptly in a line drawn nearly through X 
and 7 Aig^H,"* On the opposite, or eastern side of a 
moderately broad blank space, a similar assemblage of branches 
converges upon the variable star rj Cariase. There is an 
obvious correlation of structure on either aide of the chasm ; 
subdivisions mutually correspond ; the broken series on one 
margin is resumed on the other. The impression is strongly 

'i VrwL. ArffoU. p. S81. * Hersohel, Ovitijua, art. 780. 

■ a. C. RuBBeU, Monthly NoUces, vol. IL p. 498. 

* Outlimi, art. 787. 




336 THE SYSTEM OF THK STAltS 

conveyed that star-strata oucd united, have here jieUed 
to the influeboe of some imknowti dis|)etsive force or foroofi, 
peiliApB etiU ID operation. Vet we can acarcelj' hope evet 
to omuoand the meaua of testing the conjecture. For the 
Iffoper motioQA of the faint telescopic stais near the edges of 
the gap^ aro do doubt of auch exceaaiTe minuteneee tbat 
oentariea, nay, uiilleummua may pass before they can became 
perceptible. 

The repTesentJttiOQ of the Milky Way aa a imiform atony 
afcream is purely eoDTODtioDol, Its real texture ia of a curdled 
or flaky description.' Between PerseuB and Sagittariiia» Sir 
William HerHchel counted eighteen luminous patches, *' re- 
sembling the telescopic appearunce of largi^, easily teaolred 
nebula" ;* and hi» eon perceived the lucid ramifications in 
Sagittarius to bo made up of " great cirrous maaaoa and 
streaks," the appearance, as his telescope moved, being "that 
of clouds passing in a scud, aa the aailora call it/' Furthei 
on, he remarks : " tho Milky Way is hkt) sa^d, not strewed 
evenly as with a sieve, but as if flung down by handfuls (and 
both hands at once), leaving dark intervals, and all consistii^ 
of 14tb, 16th, and 20th magnitudea,^ down to nebuloaity, in a 
most astonishing manner/** 

The bright spao^ of the galactic zone are commonly sur- 
rounded and set off by dark winding channels, and the rapid 
alternation of amazingly rich with poor, or almost vacant 
patches of sky, is a constantly recurring phenomenon.* 
associated by Mr. Maunder*^ with slow processes of stellar 
agglomeration. The most remarkable instance occura in the 
Southern Cross, the brilliant gems of which emblazon a 
broad galactic mass very singularly interrupted by a pear- 
shaped black opening eight dt^greea long by five wide, named 
by early navigators the " Coal-sack." This yawning excava- 
tion figures in Australian folk-lore as the embodiment of evil 
in the shape of an Emu, who lies in wait at the foot of a tree 

1 EouEwu, Uranom^rii OiniraUt p. Iff, 1ST8 ; Kloin, WocKtMehrifl /Or 
A»tr. 1867. p. 2B8. 

* FMl. Tratu. vol civ, p. 262. 

* Herschel'^ "20tli magnitude" QorrMponcU (pproxinuitelj with the Htb 
OQ the pliototii«tric £c«>Ie. 

* Cope ObafTvaiion*. p, 388. " HerBohol, Outtiiut, uts. 7W, 7ft7. 
' Knoicledffe, tqI, xviii. p. ^6. 



^ 



THE MILKY WAY 



337 



ipteaenteci by th& atara of the Cross, for an opossum drivea 

his peTsecntioDS to take refuge among its branches.^ The 

igend reads almost like a Cbriatian parable. The denudation 

the Coal-8ack is, however, shown bj Mr, H, C. Russell's 

otographs to be complete only towards its northera end* 

o the south, a considerable invasion of small stars modi£ea 

e contrasting darkness. 

Partial galactic vacuities, evidently of the same nature 

'th the southern Coal- sack, occur elsewhere, notably in 

lygnus; but they are inconspicuous to casual observers. A 

larkable doubly perforated star-cloud la exhibited in 

late XIX. from a photograph taken by Profe&sor Barnard 

in 1892. It makes part of the vivid scenery of the MiUcy 

Way in Sagittarius. 

An admirable deliueation of the formation, ao far as it 
ia visible in the northern hemisphere, was completed at 
PaTBonetown in 1889 by Dr. Otto Boeddicker, after five years of 
labour, amid climatic conditions of the lenst propitious sort. 
The general effiect may be best described ag that of a thick 
stem of light, closely set with curvilinear ramifications ; the 
stem itself being riddled with dusky convolutions, intricate 
panares, and '^ hoise-ehoe " or " key-hole " apertures, separated 
by lustrooa wisps and nebulous " pointed arches." The 
circumstance that " feelers are thrown out towards nebulEB 
and clusters,^ is of profound interest. Thus a feeble branch 
starting from a Casaiopeise terminutea at the Andromeda 
nebula. The Pleiades stand at the peaked summit of a 
dim vault springing, on one side, from near p Tauri, on the 
other, from e AurigiE, The Hyadea are separately involved ; 
Prsesepe is all but reached by a long streamer iasiting from the 
vicinity of ^ Cauis Minoria ; while a thin sinuoua eifuaiou, 
perhaps of a spiral nature, includes the great nebula in itft 
sweep through Orion. The subject was next prosecuted by 
M. Easton of Kotterdam. His galactic charts * are of high 
authority; and, following up the initiatory efforts of M, 
Houzeau, he has lately studied, with instructive results, the 
I intimate structure and sidereal relatione of the great zone by 

^1 * Mml 



■ MorUht^ A^oiKn, toI. li. p. 40. " Ibid, vol 1. }i. 12. 

« La Voie LaeUe, Farv, 189S, 



338 THE SYSTEM OF THE 3TAES 

means of " isophotal curves" ' or coDtour-UneB of lonunostx 
YiGual recorcU, however^ of bardj pensepUble deUOa ounot 
be entirely satisfiictoiy, their agreement being impaired )tj 
iadividual diversities both of purpose* and of faeoltf; lod 
they have been lar^ty, though not wholly Btiperaedod, hj tbe 
pn>mptly-8e«ured impreasions of the impartial and iHHtiiiBUr' 
sensitive pUte. 

The Milky Way Tanes greatly in latctral asctflol A 
brilliant stream no more than three or four d^fraM vide 
where it enters the Ciobb, it expands to AUlj twonty-twu 
degrees in tbe bifureated section stretching from Ophitu^us 
to AquiR At some apota, too, the nebulous effect Ui the eyr 
fadea away imperceptibly along tbe margins ; * at othflrs. Uje 
line erf demarcation is so sharp that a telescope may have aDe- 
half of its field crowded with galactic stars, while the other 
half is welt-uigb blank.' A de&nite semi -circular bounduji 
for instance, liiaits the formation near 2^ Aquila? ; its aoatfaen 
edge in Ophiuchxus was remarked by Sir John Herscbd, v 
" terminated by an irregular nebulous fringe aa if lacerated " ; * 
and marginal projectiomi, knobs, and briBtling outliers azv easitj 
perceptible elsewhere. The prevalent rule aeems to be (bftt 
the smaller the gtars considered, the more abrupt ia the 
commencement of the Milky Way ; while a mote and mort 
gradual condensatiou aooompaniee each step upward in 
brightness.* 

Sir William Herschel waa perfectly satisfied Chat, with 
his 20-foot reflector (equivalent to a modem reiiactor about 
14 inches in aperture), the Milky Way was, in general, 
" fathomable." The stars composing it, that is to say, wct? 
of definite numbers, and appeared projected upon a perfectly 
block sky. But this was not so everywhere ; certain jmuU 
completely baffled the penetrative faculty of his instrument 
One such was met with in Cepheus, where he found the eooall 
stars to become "gradually leas till they escape the eye, so 
that appearances here favour the idea of a succeedingj more 

' La Dittribuiian tU la Lumiire GaltKtiqut^ AlQsUcduSj 1903. 

' Rid. p. 7* 

' W, Horwhd. Phil. Tran$. vol. civ. jx 283. 

* Proctor, UnivtTM of Start, p. 86. 

* Klflid, Wodtenaehri/t, U67, p. 2S6 ; J, Hefichel. CojH SttuUt, p. 88». 

* Celoria, Memorie del M. J^iituio LfmibardOt t xit. pi e37> 




^^^^^^^^^^^^^^^^^^^^^^^^^f XIX. 


1 




* 






1 ■>^--^'^j:.>yy^^yvv,;.;,i;^-,.^-.-^:v^ 




1 ..:: r- ■'■■.■■ ■ -..-.- 


1 


Bv^'^^" '- '■■--■^ , .; .V.--.: .:. ..,' ..i. -■'.■■■;, -■■ 


^H Perfonitwl Oatactic Gi-oup in Sagittarius.' Photo({ra|^^led by K. K. IlmuonJ, 


J 



THE MILKY WAY 



KS» 



,nt clustering part." And he remarked, in exploring 
^^*^tween Sagitta and Aquilaj that " the end of the stratum 
^^^nnot bf Been." * Again, in the galactic branch traversing 
*<^^phiuchua. Sir John Herschel encountered " lai^e milky 
bulous iiTegulai patchea and banks, with few stars of visible 
^toagnitudea " ; he described " a very laige epace " of the 
^Kt jlky Way in Sagittarius as " completely nebulous like the 
diffused nebulosity of the Magelliinic Cloud " ; ^ and observed 
^ similar spot in Scorpio, " where, through the hollows and 
deep recesses of its complicated structure, we behold what has 
a.U the appearance of a wide and indefinitely prolonged area 
strewed over with diecontinaous masses aad clouds of stars 
which the telescope at length refuses to analyse." * 

Even with the best telescopes of recent construction, this 
perplexing and indeterminate aspect cannot be altogether got 
rid of. Professor Holden could obtain with the 36 -inch 
Lick achromatic directed to the Milky Way " no final resolu- 
tion of its finOT parts into stats. There is always the back- 
V ground of unresolved nebulosity on which hundreds and 
thousands of stars are studded^ — each a bright, sharp, separate 
point"* The lingering nebulosity was strongly indicated to 
be of stellar nature, but whether it was due to the presence of 
inniimerable small stars mixed up in the same region with 
larger ones, or to the indefinite extension into outer space of 
galactic agglomerations could not be pronounced off-hand. 

The explanations attempted of these complicated pheno- 
mena may be divided into disc- theories, ring -theories, and 
spiral -theories. The " disc - theory " of the Milky Way was 
first propoimded by Thomas Wright of Durham in 1750. 
He supposed all the stars to be distributed in a comparatively 
fihallovv layer, producing iin annular effect by its enormous 
lateral spread. Irregularities, he thought, were partly due to 
our eccentric position within the stratum, partly to " the 
diversity of motion that may naturally he conceived amongst 
the stars themselves, which may, here and there, in difierent 
parts of the heavens, occasion a cloudy knot of stars." * 



' PhiL 2'rajLs. 7ol. evil. p. 328, 

- Capt Obaervalwns, p. 389. ' OuUinet of Aatr, vt 78B. 

* AW. Mtst. August 1S8S, jr. 2&S. 

* An Original Theory of the Univem, p. 03. 




THB nvnU OF TBS SiABS 



m 



To that view Mr Williua HendKl | 
mMilji by the ■ypHertMa 

MdHd of " itg-ff^w" 
wntHitMiwwy WNiM wnb lua gnM 
iflCtha Ay,fcarfiiiwii t^tlfc»p 
«poD tte Awtta of tibe gn^ 
lb* Mflkjr yfmj. In ita ncig^ibovbooa, « 
mw%f InxB U, Umj «w» w ymadf ^iMnbolad Aad tldi faf i 
ngtdu yro^nmknct 6Batitf tnm tfca 0UK*ic polv to tbt 
gBLM^tic ee^uator, tb« UUer regioa Wag oo an avaage thmj 
tiiBflB riebar than tbt fotiMK. Ka«r, if wm «cn to admil. m 
Haidid did, a nenlT- eqaable acstteriiig ol ^ns in apMee. tJun 
woflld bt ao altemalive bat lo soppoae tbe adareal ajitfiat ex- 
lendad io otij direction proportiooately u> the comber of itm 
attn io thai diroctioiL Their crowdiiig Bfaoold^Mk Ukmi hypUkmt, 
be purely optical — the eSbct of the iniipfinita x^wesding out 
in tho line of ligbt of their erenlj aemed raoka. Soondiji^ 
the itar-d«pthfl upon tht« principle, Herwihd meaBnred tfas 
IflDgth of hiM line bj^ their Hceming popnlpa^aeae;, and OOB- 
•truotod, ftvm tbo PUmericaJ data tbos obtained^ the " dorvB 
diao " iDode]> bog aQoept«d «a repTeMtntlQg tbe croe fona of 
Ibti BtolUr unirtirae. 

But hiii own obnervattOQa at the vety moment of enoimciiig 
thifl theory, fiitdlly undermined it. Already^ in 1735, be 
romnrkfxl thut twrj or thrce hundred ^' beginnings or gaihefing 
chiHtem," might t>e pointed out in the galactic aystem, and be 
sunuiiKNJ iu cveutnul sopiirtttion, " after numbers of ages," into 
no nmny diHtinut '• nebuk;."* " Equable scattering" then, was 
nn iili^al Htatis of things long since abolished by th@ " ra^v&gea 
of tiniu." The couvit;tion that euch was the case grew with 
hie experienoe. " The immense starry aggregation " of the 
MilVy Way, hn wrote in 1802^ " ia by no means uniform. 
Th*i HtarH nf wliii;]i it ih (lompoBfid are very unequally scattered, 
and (thow ovidont marks of clustering together into many 
Mparato allotinoutH." Nor did he fail to perceive, from the 
^adiml iucroaMe of brightneaa towards the ceuLrea of the^e 
" allotmoutM/' thut they tended to assume a spherical form, 
and thuB auggested " the breuking-up of the Milky "Waj, in 
all it» minuto partH, aa the unuvoidable conaeqiaence of the^ 

^ PKil, TVvM. vol. Uxv. p. 20&. * Ibid, vol xcil 




THE MILKY WAY 



341 



lostering power arialug out of those prepondorating attractions 
rhich have been shown to be everywhere existing in its com- 
I." ^ The formal announcement o£ his conviction " that 
le Milky Way itself consists of stars very (litferently scjittered 
those which are immediately about us," * amounted to a 
antation of the principle of atar-j^uging. 
With it disappeared from Herachel'e mind the conception 
an optically -produced galaxy, In his ultimate opinion the 
tual corresponded very closely with the apparent structure : 
was composed, that is to say, mainly, if not wholly, of real 
louda of stars. Credit was thus restored to the early im- 
pression of Galileo, who in 1610 described the Milky Way 
" nothing else but a mass of innumerable stars planted 
aether in clusters," '' 

Wilhelm Struve's* effort towards the reorganisation of 
ie stratum-theory, though aided hy all the resources of hie 
great ability and address, could scarcely be counted as a step 
in advance Substituting for the hypothesis of equable dis- 
tribution that of concentration in parallel planes^ he imagined 
the avenge interval of space between the stars to dimixush 
regularly with approach to the central horizon of the system. 
The swarming aspect of the MUky Way was hence interpreted 
as agreeing with factj but the annular appearance as being 
illusory. Of illimitable dimensionB, the system was conceived 
to stretch away, still preserving its specific character, to an 
infinite, or at least unimaginable remoteness, comparatively 
narrow visual bounds being set to it by a supposed extinction 
of light. 

But the quasi-geometrical regularity of Struve's galaxy is 
belied by innumerable details of the original. The swell of 
the tide of stars towards the j^lactic plane is neither uniformly 
progressive,^ nor does it proceed without couspitiuous inter- 
ruptions. Thus, the region near the horns of Taurus, although 
close to the MUky Way, is absolutely the poorest in the 
northern hemisphere ; * and it is matched in the southern ^ by 

' FhiL Trant. vol. civ, p, 282, ' Ibid, vol. xcU, p, 480. 

' iSVdtfnvJ Nuneius, ii^txs. by E. S. Cw\em, p. 4.2, 

* Mudes d'Astronomii Stellairt, IS47. 

* C. 8, Peirce^ Sarvard Annals^ vol, ix. p. 174^ 

' ArgciIaudeT, Simntr Stcb. R<1, v. Einkiiung ; Pto&tor, U-mwrst of Start, 
p. 32, "^ Thomej Conkita Dur^mttstentngt IntroductiDii. 




■fj l w — « 
Sir JobB niHiihJ fay Ui G^» f^ffwifwrn, 
Ui Mtaal idadMoe to diift &r cw»j fiw 
tiM poiitioB origiatttf taliM up t^ ha teliec ne dMp 
bgrUaiar tte |pk^ wm UmI of "« fiat mg,^ 
otlMria-wtfldig foia «f iwnaMB aad iniggmlglpiwM 
•od UttdaMHL'" BfpawW inditfnitelx fllaog tl» oootal 
]]>UiM, tike o«v Bodil mmfy diflbed ten the old exocfi ia 
no for u Itie idtt of boakoipaooai oonotmction wm given iqk 
T\m (Ii«c rom&iiied, bat with ita oeubre Kooped oqI The oolar 
nyfltflm WM locatod ia on odotbuhis ^ao^ of relative Tucoi^. 

The Milky Wajr, thtw reguxted* sppeafed to cMuist of an 
indit^tiiVo nantber of atellar coUecttone "brought by projectioii 
into Dearly the lame viaoal line" — to repreepot the fore- 
ftborittied tfTect (movft especially at a pamcular spot in 
HagiUariuA) of '* a vast and illimitable area scattered over with 
diiK;rjtiLitiuauH nuiiioa and Jiggregatea of stars in the manner of 
tho 4:LUiiiilt of a inadceral-sky/'' Bat in an aaaemblage of 
thiit iiiitnro neon edgewieej a " Coal-sack " would be a pheno- 
Tminoii fiH nnoimtlotm an in a uuiform stratum ; nor could it, 
wilJifniL viuluiit improbability, be conceived of aa rent by the 
«i»!iiHNal hvuituroH dividinj^ the actual Milky Way in Argo and 
UphiucUiia. 

t Frcnior, S\>c. eit> j*. in. * Outliiut, Art. ?BS. 

> Cap* Obitrmiiiotu, p. 380. 



THE MILKY WAY 



343 



To remedy these utconveniences. Professor Stephen Alex- 
ader devised in 1852/ upon the model of the wheel-shaped 
aebula in Virgo (M Q9), a spiral galaxy witli four curvilinear 
aches diverging from a central cluster formed hy the aun 
id lucid stars. By properly adjusting the mode of projection 
these radiating star-streams, the effects of rifts and coal- 
cks were duly produced ; but the arrangement, however 
lired for ingenuity, gave no persuasion of reality, and 
|uick]y dropped out of remembrance. E^entially diJTerent^ 
Ithough with some featuree in common, was that by which 
Proctor replaced it in 1869.^ Rather than a " spiral " 
adeed, the new desigu resembled a bent and broken ring, 
lib long, riband-like ends, looped back on either aide of an 
ening, accommodated to the shape of the gap in the visible 
acture in Argo, One of these loops, by the apparent 
intercrossing of its near with it& remoter branch, was sup- 
posed to generate the Coal-aack in Crux; while the other 
end, trailing lengthily backward^ afforded a deceptive effect 
of bifurcation. Excessive distance was invoked, as in Profeeaor 
Alexander's scheme, to explain the cessation of nebulous light 
in Ophiuchua. 

Of the manifold objections to which this hypothesis is 
liable*^ only two need here be mentioned. In the first place, 
it involves a wholly inadmissihle rationale of the openings seen 
in the Milky Way. If these were due to the interlacing by 
perspective of branches really far apart in space, the enclosing 
luminous formation should be markedly fainter on one side 
than on the other. But this is not eo. The borders of the 
southern Coal-sack are approximately of the same brightness 
all round. A single vivid moss has obviously been the scene 
of what, in the absence of better knowledge, may be described 
aiS an excavatoiy process. 

Again, on the spiral theory, the ^at rift in the Milky 
Way should represent the interval between branches mutually 
disconnected except through the optical effect of projection. 
But their mutual dependence is manifest. They strike apart 



^ Aiir. Joum. vol. iL p. 101. 

* Mimihljf NoiiceMj vol. xxx. p. 50, 

* See Mr. J. tL Suttoa's retoarks, lUuttratid Siienei Mcnthi^, tdI. ii. pp. 
63, 109 ; Knawled^St vol, xiv. p. 41 ; Eoetou, ibid, voL ixi. p. 13. 



344 



THE SYSTEM OF THE STABS 



gradually', und aa the result of changing conditions. Pre- 
monitory caritieg seem to Jinnounce their impending eepara- 
tioQ ; and even after it haa become definitive, abortiTe efforts 
towards reunion are indicated by the correspondence of oppo- 
site projections and by the occasional bridging of the fissure. 
The bifuication is beyond question a physical reality. 

The prevalence of spiral forms among the nebulge, em- 
phasised by Keeler's photographic explorations, adds weight to 
the ai^uTxientfi for assigning a more or less similar structure to 
the great firmmnental zone. M. Eoaton has illustrated tbe 



"' a y--;^ 



»H, 



-%, 



^AA 



-T> 



-*-^ 



AS 



\' 



1 



ajr^. 



S^^ 






^^ 



no. n.^Tb« 6*Uctic Splrmt (Euton). 

subject by a plan-sketch, copied in Fig, 3 7. It does not profess 
to give " even an approximate representation of the Milky 
Way, seen from a point in space situated on its axis/' ^ but 
merely to indicate a possible mode of distribution of stellar 
accumulations by which the observed phenomenon might be 
produced. A remarkable bright patch of the Milky Way in 
Cygnufl serves in it as the nucleus of the convolutions. The 
sun's position, though central, is detached from them. 

Some such design strongly tempts thought. It is recom- 
mended by analogy ; it is recommended alBo by its adaptability. 
A multitude of facta, at first sight incongruous, are combined 

' Astroph, Joum. Toh xii. p. }&7, 



THE MILKY WAY 



345 



by it into plausible unity. The ring-theory may be said 
to have broken down when Celoria' found it necesaary to 
establish a double annular eyetem. Hia researches, together 
with thoae of Plasamann, Seeliger, and Easton, leave no 
doubt that the Tarious sections of the stariy girdle differ 
prodigiously in distance. Moreover^ it seenjB impossible eo 
Ui place the bud in such an enclosing structure ae to 
get a reasonable explanation of these diffeteucea without 
having recourse to further complexities of artificial arrange' 
ment. True, no deeign corresponding in any degrea with 
what appeaig, caa be other than intricate. Were it permissible 
to adopt the opinion that the Milky Way is very much what 
it a^tmi, we should deecribe it as a ring with atreaming 
appendages extending from the main body in all possible 
directions, some nearly straight towards, or away from ub, 
others at every imaginable angle with our line of aigkt. The 
resulta in perspective foreshortening would evidently, under 
theae circamstattces, be highly complex ; the eye being pre- 
sented with groups and streams of etars, immensely various in 
remoteness* but all projected indiscriminately upon the same 
zone of the heavens. Thus, while some branches, pursued 
along their outward c-oiLree, fade at last into dim nebulosity, 
other Milky Way groups may be distinguished as bright 
separate stars, because much nearer to us than the generality 
of their associates. Closed ringSj however, are beginning to 
appear alien to the cosmic plan of structure. Nebulae pre- 
tienting that aspect are, perhaps without exception, resolvable 
into helical or spiral figures. And it would be hazardous to 
assert that the Milky Way lies outside the mysterious law 
imposed upon minor aggregations. 

Its internal organiHatiou is of baffling intricacy. It collects 
within its ample round, there is every reason to suppose, an 
absolutely endless variety of separate systems, A multi- 
tudinous aggregate of individual clusters, it is composed more- 
over as a whole very much like one single cluster on a colossal 
scale. Its fringed edges, its rifts and vacuities, are, as we 
have Been, reproduced in miniature in nimiberleae star-groups. 
" Einge," and "sprays,*' and " streams" of stars are unmistak- 
ably common to the two orders of formation ; and the stellar 
^ Hemorie dei R. IitUato Lomhardo, L xir. [k 62. 




T m 

hndd 

-f The twtiii ■ «9vvakMt te tka 

What k tfav «vKig» dii- 

I How far 
hftw lo KbtcI bdbre isdiDg onmiwm mUmkOj in tfae 
ci Um CRKwdMlobiKto paodMo^to 
" nllkr " dfeet «f • Debolooi atntaB ? 

Certuzi ktiomibA^ in thk mmttcr ii not W> be had; inliar 
tKMU sad profaabb Tititi»wt-ni bare to tak» ila phea 
Vairoomb, *■ Uie vpaboi of a disciunn enbnong all tbe 
sTsiiable fiute, arriveB at the oooohtnoa tliat the Milky Waj, 
in iDOtt of ita aedioiia, it no Dearer to ns Uksa mxdd be 
Jiignified by a pai^lax of otie-Uunuandth of a aecww],* oarz^- 
vpandin^ to a ligbt-joumej of S 200 yeai&. Now oar son, if 
thu» uaimaginably removed into space, wotild ahnnk to a sur 
of tb« fiftaeotb magciitiide; it would seem just one of the 
grainii of Bhining aand coagiilated into heaps out near Um 
oonfincM of the sidareal worhL PreBumablj, tb^n, a large pro- 
portion of tba loatrouii speclu formings in tLeir mmumbered 
agffregate, the nebtiloua arch amid the confitellations are really 
Mutw on tin* mixh of our own. They are, however, pretty 
avidenUy intermmgbd with many amaller globes, and with 
HOiiid vaMtly larger. M. Eaaton has nucoeeded,* by detailed 
coriijHFiriHoaa, in efitabliabiug a correlation between galactic 
KtrtuliircB tiiid tliu nUiTti, from about the sixth to the fifteenth 
niJiynitiidtw, cnuTiiefatt^d by Argelander and otbets. The in- 
fereuw jm thua n*ndcryd oompulsory that a percentage of these 
oomparativcLy bi-ight orba are geuutne cgtistituents of the clustera 

1 Th4 Stitrg, p. »ir. 
• ViMtriimtiim dt la iMmUn OatacligM, p. 24 ; Newoomb, Tk« Stan, p. 273. 




THE MILKY WAV 



347 



with which they are colliueated. There are be&idea strong 
grounds for the belief that manj*, if not moat, of the bluiBh 
brilliants giving helium apectra domioate these comprehensive 
groups, Mr. Kanyard pointed out in 1891/ from the evidence 
of M>, H, C, Ra83eU*8 photographs, that the chief luminary 
of the Southern Cross is centrally situated in a curiously 
symmetrical little cluster of excessively faint stare ; and the 
irreaistible concluaion of its being physically related to them 
may be safely applied to a number of other lustroiwi objects of 
the same spectral class. The range of actual size and splendour, 
then, among the components of galactic atar-drifta is astonish- 
ingly great — much greater, in Professor Newcomb's opinion, 
than in ordinary detatilied clusters, such m the Pleiades, or 
the double Sword-handle groups The helium Htars of the zone 
arej indeed^ veritable tritons amidst shoals of minnows ; 
they are frequently of a magnitude transcending the powers 
of imagination to realise, while their dwarfed aasociatee may 

L. weU be of average sun-like stature. 

^B The Milky Way clouds are not condensed from the general 

" contents of the sidereal heavens ; they are markedly distinct, 

I Their spectral peculiarities make thia clear. They are built 
up, essentially and fundamentally, out of Sirian stars ; * those 
of the solar and Antarian types seem to be totally absent 
from them. They include, however, nearly all the helium and 
bright-line stars that exist ; but they are relatively few ; they 
scarcely count as ingredients ; they are rari nantes in gv.rgite 
vctsto. Many galactic tracts, too, are suffused with a phos- 

• phorescent glow ; they harbour nebulous formations which 
only the photographic camera, through ite faculty of persistent 
gazing, has been able to actualise and define. Thia surprising 
characteristic affords an additional proof that cosmic conditions 
of a special kind prevail in the enigmatical girdle wluch 

L enclasps the mystery of the universe. 

^H From a most careful study of the Milky Way at Cordoba, 
where it was seen to peculiar advantage. Dr. Gould inclined 
to regard it as the product of two or more superposed galaxies,' 
The fact of the two narrowest and brightest, and the two 

^ KndioUdgtj voL xir. p. 112. 
■ PickeriDg, Harvard AmutUf vol. M, p. 2£, 
* Utw\. Arijenti'nai p. 381- 



348 



THE SYSTEM OF THE STARS 



moat diffused parts lying in pairs opposite to each other 
is certainly remarkaWef and lends gome countenance to 
the Burmifle that the " necka " in Cassiopeia and Cnix really 
repn^Bont the intersections of the two croesed zings visiblj 
divergiag in Ophiuchns. H. Celaria^ too, as we have seen, 
adopted the hypothesis of a compound Milky Way, but of 
eucli a form aa to allow the possibility of one of its oonstitueot 
annuli being comprehended by the other. The transition to 
a true gpiral shape was thence easily effected, and a wider 
range of facts was rendered capable of theoretical accommoda- 
tion. Whiit is unmsstukable is that the entire formation, 
ningle or compound, wliile individual and specific, is yet no 
isolated phenomenon. The contents of the firmament aw 
arranged mtiinly with teferenee to it. It ia a lai^ part of & 
larger il^ign exceeding the compasa of finite minda to gnep 
in its entirety. 



CHAPTER XXVI 



STATUS OF THE NEBUL« 



question whether nebulae are external galaxies hardly 
jy longer ne^ds diflcuaeion. It has been answered by the 
rogresB of research. No competent thinker, with the whole 
of bhe available evidence before him, can now, it is Hafo to eay, 
maintain any single nebula to be a atar system of co-ordLnate 
rank with the Milky Way. A practical certainty haa been 
attained that the entire contents, stellar and nebidar, of the 
sphere belong to one mighty aggregation, and stand in ordered 
mutual relations within the limits of one all-embracing scheme. 
All-embracing, that is to say, bo far as our eapacities of know- 
ledge extend. With the infiiute poasibilities beyond, science 
has no concern. 

The chief reasons justifying the assertion that the etatua 
of the nebulee is intra-galactic are of three kinds. They 
depend^ first, upon the nature of the bodies themselves; 
secondly, upon their individual stellar associations ; thirdly, 
upon their systematic arrangement as compared with the 
systematic arrangement of the stars. 

The detection of gjiseous nebulae not only directly demon- 
Btrated the non- stellar nature of a large number of theae 
objects, but afforded a rational presumption that the others, 
however composed, were on a commensurate scale of size, and 
situated at commensurable distances. It may indeed turn out 
that gaseous and non-gaseous nehulee form an unbroken aeries 
rather than two distinct classes separated by an impassable 
barrier. Their spectra have perhaps more in common than would, 
at first sights be supposed. For the vivid rays of green nebul» 
are superposed upon a gauzy background of continuous light, 

34^ 




850 THE SYSTEM OF THE STABS 

which appears to be resolvable into a mulUcude of bright lines 
in juxtn{H>aitioa [^ and the spectra of white nebulie show, wA 
a emooth, priamatic gradation, but alight inequalities in the 
flow of light, indicating effecta of abaorption, of emiasioD, or 
of both combined Before indeed anj settled opinion can be 
formed as to whether these analogiea have reallj the tiaoa- 
tional meaning we might be inclined to attribute to them, 
nebular spectroBcopy oiust be a good deal further advanced 
than it is at present. But apart from this question, rektioo- 
ehip between the various orders of nebula is manifest. The 
tendency of aU to assume spiral forms demonetrates, in itself 
their cloae affinity ; bo that to admit some to membership of 
the sidereal system while excluding others would be a palpable 
abaurdity. And unce those of a gaseous constitution must be 
so admitted, the rest follow inevitably. 

Of the physical connection of nebulse with particular etars, 
fresh and incontrovertible proofs accumulate day by day. 
Nothing can be more certain than that objects of each kind 
coexist in the same parts of space, and are bound together by 
most intimate mutual ties. To argue the matter seems, as the 
French say, like " battering in an open door." We need only 
recall tlie stars of the Pleiades, photographically shown to be 
intermixc^il with nebulae, and those in Orion still bearing in 
their spectra traces of their recent origin from the curdling 
maasea around. The nuclear poeitions ao frequently occupied 
In nebulae by stars single and multiple, reiterate the same 
assertion of kinship, emphasised still further by the phenomena 
of stellar outbursts in nebulee. The scenes of these must, as 
the late Mr. Proctor insisted, lie within the circuit of the 
MiUcy Way, unless we are prepared to assume the occurrence!, 
in eitra-siderffiil space, of conflagrations on a scale outraging 
all probability. It has been calculated that if the Andromeda 
nebula were a universe apart of the same real extent as the 
Galuxy, it should tie situated, in order to r^uce it to its 
present apparent dimensions, at a minimum distance of 
twenty-five galactic diameters,* And a galactic diameter being 
estimated by the same authority at thirteen thousand light- 
years, it follows that, on the supposition in questioi 

1 Pftlmer, Lick Bullttin, No. U. 
" Woaae. Sdtr^/ten Witner VtrnnSy Bd. t. p. 318. 




STATUS OF THE NEBULA 



351 



Duld require 325,000 years to reach us from the nebula. 
ie seventh -magnitude star then which suddenly shone out 
the midst of it iu August 1885 should have been an 
aluteiy portenfcouB orb. In real light it sliould have been 
^uiviilent to 762,000 stars like Siriua, or to sixteen million 
ach suns as our own J But even this extravagant result 
inadequately represents the real improbability of Ihu hypo- 
thesis it depends upou; since the Andromeda nebula, if un 
external galaxy, would almost certainly be at a far greater 
remoteness from a sister-galaxy than would be represented by 
twenty-five of its own diameters. 

Just as the Milky Way might be described as a great com- 
pound cluster made up of inuumerable subordinate clusters^ so 
the greater Magellanic Cloud seems to be a gigantic nebula 
combining into some kind of systemic unity multitudes of 
separate nebulie. To the naked eye it shows vaguely a 
brighter axis spreading at the extremities so as to produce h 
resemblance to the " Dumb-bell " nebula ; photographic ex- 
posures bring out unequivocal traces of a spiral conformation ; 
either way it shows signs of definite oi^anisation as a coherent 
whole ; and it includes, strangely enough, among its inmates, a 
miniaiure of itself (N.G.C 1978), but of much greater 
intensity and distinctness. Sir John Herschel's enumeration 
in 1847, of the contents of the "Cloud," gave conclusive 
evidence of the interstellar aitnatiou of nebulie — evidence, the 
full import of which Dr. WheweU was the first to periieive. 
Over an area of forty -two square degrees, 278 nebular objects 
(stare being copiously interBpersed) are distributed with the 
elsewhere unparalleled density of 6^ to the aq^uare degree. 
" The Nubecula Major," Herachel wrote, " like the Minor^ con- 
sists partly of large tracts and ill-defined patches of ineaolvable 
nebula and of nebulosity in every stage of resolution, up to 
perfectly resolved stars like the Milky Way, as also of regular 
and irregular nebulie properly so ealledj of globular clusters in 
every stage of reaolvability, and of cluetering groups sufficiently 
insulated and condensed to come under the designation of 
clusters of stars," ^ 

Here then we find — in a system certainly, as Herachel said, 
' $ui generis*' yet none the loss, on that account, instructive 
' Cape RMults, p. 14^. 




wd iwowwd (7 Dr. Max Wolf It k cwri— dty exaB|Aifed 
ia Clw rtarnlieQiided nidoi of the Aatnm B«baki' (■> 
Pbl« XVUL). which soons to hare abevfaed or npeOed haaa 
iti hmriftiatot nc^botu-hood the dtmm galacck ctetU impmA- 
big tomrdf IL Sunilarij, a eampaiBtiwely TMaiit itjgion 
aooDfDpUMi the Onoa nebula, within whkb. seveftbeloBr 
ttttt twuni Mftd flicker nfter the muDer of the componoatB of 
globohv ciiiitwi. Thfl i^laoement of stan hj n^mkna laattflr 
it flftfal ooofpiciioua in Plate XX., which reprodaoM, bj* Dr. 
Wolffl kiuil permiMion, a phot4^raph taken by hisn, July 10. 
1904, with an «xp06ure of four boom The depicted Dcbola^ 
which hod been discovered ten yeara previooalj. is about 10' in 

> /'Ml. Trtuu. vol IxxiT, p. U», 
(lUi*). 



^Mi 



STATUS OF THE NEBULA 



353 



diameter, of a round shape, and a complex Btructure. " It is 
placed centrally," Dr. Wolf Writes, " in a very fine lacuna ^t^id 
of faiitt stars, which aurrounda the luminous cloud like a 
trench." ^ Moreover, this negative " halo fonnfi the end of a 
long channel, running eastwai-d from the western nebulous 
clouds and their lacunfe^ to a length of more than two 
degrees.'* The coexietGnee in the same Bidereal district of 
nebulEe and stars could not well he asserted with stronger 
emphasis than by the clearing of a dark foaae for the accom- 
modation of the cocoon-like object in Plate XX. 

The larger plan of nebular distribution, as being the 
inverse to that of stars, partially revealed itself to the elder 
Herachel ; but Sir John first brought into clear view the 
distinct and striking division of the nebuhe into " two chief 
strata, separated by the Galaxy," Taking the circle of the 
Milky Way as a horizon, he remarked that the accumulation 
of them in Virgo and Coma Berenice© " forms, as it were, a 
canopy occupying the zenith, and descending thence to a con- 
aideiable distance on all aides, but chiefly on that towards 
which the (oelestifil) north pole lies.''^ 

Tliie crowding about the galactic pole is less marked in 
the southern hemisphere, though here too there is a ''chief 
nebular region " approximately corresponding to that in Virgo, 
The distributioD isj however, on the whole much mote uniform 
than in the northern hemisphere, or rather, more uniformly 
patchy, rich districts alternating with more or less ample 
vacuities. One of these exteuda about fifteen degrees all 
round the south pole, the Leeser Cloud marking its edge. 
The remarkable fact, too, was noticed by Sir John Hersehel 
that the larger nubecula seems "to terminate something 
approaching to a zone of connected patches of nebulce," reach- 
ing across Dorado, Eridanus, and Cetus to the equator, where 
it unites with the " nebular region of Pisces." A similar line 
of communication is less conspicuously kept open with the 
minor nubeculaj and this feature of "streams" of nebula? 
with terminal aggregationa was considered by Proctor to be 
distinctive of southern skies.* He adverted besides to the 
coincidence of two of them with stellar " streams " in Eridanus 

' Monthly Notiaif vol. liiv. p. 639. " Cap« Resiiita, p. IS7. 

1 Monthiy N^aiat, tqL xxix. p. 340. 



364 



THS SYSTEM OF THE STAES 



and Aquarius, and waa Htnick with a rignifioant dtAdeacj of 
bright atan over the iuterv^ between nebular groups. 

Ampler aoquiuixtanoe with tbia das4 of objects has, on tba 
whole, served to r&tifj earlier conclusions relative to their 
mad« of iiL'&tteriQg. Their overwhelmiiig teadeucy U* oon^ 
gate jibout the norUi galactic pole is accentuated by the nnllB 
of Dr. Mai Wolfs photographic surrey ; ' while the preseooe of 
« aecoudary focus of aggregation in Peiseus and Andromedi. 
remarked by WaterSj' was rendered unmietokable hj the sah- 
eequent investigations of StratonoE' M, £astoQ, diacaaiiig 
the subject with his accustomed thoroughness in 1 904. 
was especially struck with the disturbing influence of the 
Magellanic Clouds. Hence possibly eo fundamental a diveiaity 
m the modes of nebular difltribution on either eide of the 
Milky Way, that he coogidera it as " very probable that the 
structure of the southern galactic sky with regard to the 
nebulae differs entirely from that of the northern g&lactic aky."^ 

The leading facte of nebular distribution were correctly 
descrihod by Herbert Spencer in 1S54. "In that zone*** 
he wrote, " of celestial space where stars are exeeHsive^ 
abundantj nebuhe are rare ; while in the two opposite celeetul 
spaces tliat are furthest removed from this zone nebulse are 
abundant. Scarcely any nebulie lie near the g&Iactic cirde, 
and the great mass of them He round the galactic poles. Can 
thifi be mere coincidence I When to the fact that the generd 
maaa of nebulfie are antithetical in position to the general mass 
of stars^ we add the fact that local regions of nebulfc are regions 
where stars ore scarce, and the further fact that single nebuhe 
are habitually found in comparatively starless spots, does not 
the proof of a physical connection become overwhelming?*** 

Accompauying, but considerably overlapping the Milky 
Way aloug its entire round, is a " zone of nebular dispersion " 
(as Proctor called it) — a wide track of denudation, ao far 
as these objects are concerned. The nebular multitude shrinks, 
aa it were, from association with the congregated galactic octars. 
A relation of avoidance is strongly accentuated. But wjth- 

' Jicpm-t of Kani^flmhl OhKrwaary for 1902 | AUr. N<uK Ktt. 3812. 

* Monikly Noti^tf, vol. Im p. 527. 

* RilteupiLrt, K J. S. Adr. Gta. Jfthrg. xxxrii. p. 3&S, 

* Proc Aead. of Sdtme*, Am&Iord&tu, June 2£, 1901 ; Atir, Nach, Ko, SMft. 
> TM* Nftntiar ffypctft*su (with Addend*), p. 113. 



STATUS OF THE E'EBTTLjE 



355 



iwal implies recogaition. It implies Che Bubordination of 

and nebulae alike to a single idea embodied in a single 

Beme, The range of our possible acquaintance is aec^rd- 

5I7 restricted to one " island universe " — that within whose 

oundaries our temporal lot is cast, and from whose shores we 

wistfully into infinitude. 

Dismisaing, then, the grandiose but mi^eading notion that 
nebulae are systems of equal rank with the galaiy, we may 
turn our attention to the problems presented by the pecu- 
liarities of their interior situation. When these are subjected 
to a detailed examination, distinctions become evident between 
the di£ferent classes of nebuhe. Distinctions so marked as to 
lead almost to their separation. 

The " relation of avoidance " to the Milky Way juat adverted 
to prevaila on/y among the " unresolved " nebuhe. These, it 
ia true, are the great majority of the entire, so that the con- 
clusion of nebular crowding away from that zone remains 
unimpeachable. For certain classes of minor numerical, but 
high cosmical importance, the relation is precisely inverted. 
Over gaseous nebulae and clusters, the Milky Way seems to 
exercise an attractive infiuence equally strong with its repul- 
sive effect upon nebulse of other kinds. 

Forty out of one hundred and eleven globular cluBterg 
belong to the galactic zone,^ which is hence twice as richly 
furnished as the rest of the sky with these wonderful objects. 
And the excess rises to twenty - five times for irregular 
or nondescript clusters^ 434 out of 535 of which — that 
is, 81 per cent — are located in, or close to, the Milky 
Way. Many cluaters, indeed, obviously form an integral part 
of the formation itself; of others, it is diffiexJt to decide 
whether they should be ranked as distinct, or simply as 
inteneifications of ortliaary galactic star-groupings. To the 
latter categoi-y almost certainly belongs a collection (M 24) 
visible to the naked eye as a dim cloudlet near /x Sagittarii, 
and named by Father Secchi " Delle Caustiche/* from the 

' Taken as of ttie unifonn width of thirty degre«4, nml coTering ^ of the 
iphuTo. Mujor M*rk-wiolt {Joum. Lvv. Asir. ^Svc^ vol. Tii. ^. 182) Goda the pro- 
portloD&te area of the Milkf Way in the uorthfro hemispbero to bo ^, iaiho 
soutbcro i- Pickering (HaTWird Amials, yol. xlviil p. 165) ustisifttas Ibe 
galactic area at 1&,&12 squan degrees, or ^ of the sphere. 



356 



THE SYSTEM OF THE STABS 



peculiar armngemont of its staifl in t^jb^ arches^ caoaltt 
GarTBB^ and initirlwiued Bpirals. This again is included ia t 
great ovat condetiaution of galactic atars, obTioosly endoved 
with some degree of Btructural iDdependeiice. 

Oaseoufl aebuli^, like gaseous stara^ are nearly exdoai've in 
their galactic affinities.' Very few planetaiies caa be foawl 
at anj oonsidemble distance Irom the favoured xone; die 
Bpectroacopic senroh for stellar ttebulfe is &mtless unless within 

its borders ; and they embrace — with one exception all ibft 

irregular nebula. This single exception is a most significant 
one. It is that of the " great looped nebula " (see Pl^te XV.). 
an important constituent of the greater Magellanic Cloud 
PUiuly, then, the conditiona allowing primitive coemical 
matter to remain uucondensed in galactic regions, prevail alwi 
iu the nubecula. The individuality of its organisation has 
been strongly accentuated by the discovery, lately made at 
Harvard College, thut very many of the stars contaijied in it 
fluctuate mpidly in light.* Misa Lea\'itt'B examination af the 
Arequipa platea yielded at once a harvest of 152 variables; 
and more doubtless a^ait recognition. 

Gaseous and white nebulte meet on equal terms only in 
this comprelieusive aaaemblage. The Milky Way is more 
escclusive. It favours the former class largely at the expense 
of the latter Within its precincts only one in sixteen of those 
dim, often fautaatically-ahaped, objects is met with, the analysed 
light of which gives no indication of gaaeity, while their even 
texture, under the highest telescopic powers, suggests no 
approach to the stage of breaking-up into 9tar& What then 
is their nature I Is the difference separating them in appear- 
ance from the resolvable aggregations of star-dust, crowding 
the Milky Way, a difference of distance solely ? Are they 
too clufltora beyond the reach, through remoteness, of effective 
scrutiny ? There is nothing in their aspect to preclude this 
BUppOflition. So far aa observation can tell, they may be of 
stellar compoattiou. Only it is not easy to understand why 
nebula situated near the galactic poles should be immensely 
and consistently more distant than nebulas thronging the 
vicinity of the galactic equator. 

1 Bauacbinger, V. J, S. Aatr. QeM. Jahrg. xxlv. p. 4S. 
3 ffarvard Circular, TUq. 89 ; Attr. l^ath. No. 3»0e. 




STATUS OF THE NEBULA 



357 



Mr. CleveUnd Abbe* sought to overcome the difficulty 

t>7 imagining the nebulie to be equably diatributed over the 

rface of a " prolate ellipsoid," its loQger axis coinciding 

^proximately with the axis of the llilky Way ; and thia 

igetneut would undoubtedly give an appearance of crowd- 
ig in the observed directions, since to an eye placed near the 
Bnti-e of such an oval figure, objects imiformly scattered over 
its Burfoce would produce^ by perspective, the effect of nmning 
together near its pointed enda. But this highly artificial 
contrivance was scarcely realisable. Our minds demand 
from a theoiy not barely that it "cover the phenomena" 
but also that it show itself congruous with the general plan 
of operations upon which we can see that nature works. Be- 
sides, the local distribution of nebula is ao far from uniform, 
that antecedent probability is in favour of their general distri- 
bution being also marked by striking irregularitiea. The 
" canopy " of nebulee in Virgo is then, we may rest assured, 
as genuine an accumulation in its own way aa the spherical 
assemblage in the Magellanic Cloud. 

But if there be no systematic difference of distance between 
the nebular classes occupying contrasted situations as regards 
the Hues of galactic structure, there must be a systematio 
diffei-ence of constitution.^ The parts of those objects crowd- 
ing towards the poles must be comparatively small and close 
tc^ther. We have indeed already found reason to believe 
that clusters do, in point of fact, merge insensibly into nebulae 
— that groups of genuine suns at wide intervals stand at 
the smnmit of an unbroken gradation of systems with 
smaller and closer constituents, down to oocumulations of 
what ia ahnost literally '^star-dust," Eesolvahility is hence 
a question of constitution quite as much as of distance, and 
we are brought to the conclusion that, while galactic nebulte 
are of what we may roughly describe as stellar composition, 
noQ -galactic nebulae are more or less pulverulent. We 
cannot of course pretend to account for this remarkable 
distinction. All that can be said is that it appears to be 
actually existent. The irresolvable "polar" nebuhe perhaps 
escaped influences powerful over the " equatorial " ones. Their 

* Mi'nthly Noticts, vol, ixvii. p. 262. 
* Proctor, Monthly Noiia^ vol. iriz. p. 342. 




368 THE STSTKH OF THE STABS 

&gniopmmA, U anj nte, oaems to hsTe Uken « di&nsl 

eomxm. 

The iftectmcope iofenni qb that nebnlK pocBUB Eft&t 
wioGitJM of tbe MLine cHer «■ tbooe of eur&. And that 
^»paRnt fixity to wiaaal tkBEmtian nimniwlkMiaUlj imlti, 
in put from thor ei Lre uw nanoietteB, in {ivt finm tbitt^ 
fiT^WTnnmi of aoetumte nmsoremeoL Tbe fuogoUiitioi of 
iMbular ilJUrfhntiwi mttatd boides indirect erideticse of 
menL ** Streuninefli,' if it mean anjUiing^ implies that 
hodiwi affected bj it adrmnoe in oommoai towarda a csoniiBoe 
goaL Aggregatton at tbe end of a sboam pirompto tbe con- 
jeotvre ^at a rootion of advance was at a oertaia point, hf 
•ome aupervening attractioD, ewajed into a motaoo of feYolo- 
tion. A bint aa to the origin of the WagiJlanw* Clonds maj 
hence be derived. Thej represent in eome sort veaeets filled 
througb long pipee from a vast reservoir. And slqco the pipes 
are still there, the flow may he eoDCselred to be still in prognaa 
Wore it to cease, tbe ooimection of the nubecula? with th« 
main nebular bodj would eYentxiallj be interrupt^ and their 
inaulation would become complete. 

The fidelity with which gaseous nebula and clnsteis adhere 
to the Milkj Way as seen projected upon tbe sphere, wazranta 
the inference that their distribution in &pace is of a similar 
character. It would be lui reasonable to disconnect them from 
tt. forjuatton of which they ao closely follow the lines. We 
can Bcorcoly err in supposing that they lie in general vif&tA, 
not behind, or in front of it. Thus, the globular clusten 
richly strewn over the branch of tbe Milky Way from Scorpio 
to Opliiuchua, but withdrawn from tlie cotiterminoua dark rift, 
plainly belong to the cloud-like stellar masses, owing to tbe 
abaencti of which the fissure seema black, although den^y 
Htocked with stars to the tenth magnitude. Other compressed 
groups Btftud out from a curtain of apparently still more remote 
8tar6, repreaentiug poasibly a divergent galactic ramification. 

Such tamificatioufl must in many cases be greatly fore- 
shortened as viewed from our nearly central poaition ; in £ome, 
they may appear only as brilliant knots upon the " trunk " of 
the Milky Way. Possibly, the double cluster in Perseus may 
partake of this optical character. It may be the termination 
of a branch spreading inward, and seen nearly end-on. 



STATUS OF THE NEBULA 



359 



Like the Perseus clusteis, the OrioQ nebula gives indica- 
tions of greater proximity than the main galactic accumulation^ 
to which it is nevertheless beyond doubt etructuially related. 
For a winding nebuloius extension from the Milky Way can 
be traced past a Orionia through the belt and sword, the 
bright stars marking which are demonatrably associated with 
the nebula. The inference then presents itself that the whole 
mixed system, or series of systems, is placed upon an obliquely 
directed oElset from the galactic zone. Keasoning of the same 
kind may perhaps apply to the combined nebula and cluster 
M 8, It occurs as a premonitory outlier of the leading division 
of the fissured Milky Way, from which it Ilea a little apart ', 
and it seemed to Sir John Herachel only an " intent exaggera- 
tion " of the stellar collections in its neighbourhood.^ 

Summariaing our conclusions, we find the unity of the 
stellar and nebular systems to be fully ascertained. They are 
bound together by relations of agreement and contrast scarcely 
less visibly intimate than those severally connecting individual 
members of each order. The general plan of nebuW distribu- 
tion is into two vast assemblages, one on either side of the 
Milky Way ; but while this ia, comparatively speaking, avoided 
by the unresolved crowd, it is densely thronged with eluaters 
and gaseous nebuke. The conditions of aggregation witblu 
the zone may hence be inferred to differ from those prevailing 
outside it ; but how and why they differ remains inscrutable. 
As to the distancea of the nebulEe, we know nothing positive ; 
they no doubt vary extensively ; noi can either linenees of 
giain, or faintness of light (both of which may be inherent 
quahtiea) serve to distinguish between those nearest to, and 
thcKe further away from us. We may, however, plausibly 
conjecture that the Loodlike accumulation of nebulte in Coma 
Berenices is of the same approximate remoteness with the 
main galactic stream, and may thus be said to constitute the 

' cap of a sphere equatorially girdled by the Milky Way. 

'■ 0»p« MtntU&f p. 887. 



CHAPTEB XXVU 



Tin comnticTioN o» tue REMrwsa 



Wmi^u HJCRiiClliCL cDuceired it to be the sapxeow object 
'ei M lre uomy " to obtiUD ii knowledge of ihe ooortnictioik U 
%h% ht^ttat "; and thia, in his view, would be aooomplubad hf 
the " determination of the real place of evBzy oeiestial body 
iu ■pftce.'* * Thus limited, the problem would be completely 
Bolvdd oonld the abeolnte diatance bo aacertained of all Uie 
objdctt teleBcopicdllj or photographicallj di^certiible in tbe 
■ky. But even the attainment of this unattainable paint 
would never have satisfied Hciachers restlesa spirit. The 
imI loope of hifl inquiriee went far beyond it. They had oa 
hUtorical as well aa a statistical aim. "Looking before and 
after," thuy embraced the past and future, no lesa than tbe 
prtMDt of the OoHmoa. 

Modem inveatigators are of tbe same mind The heavens 
UO TOgarded by them from a 'physiological, rather than from a 
purely anatomical point of view. Mere knowledge of structure 
however ui-curute, will not content them. Tbe -vital functiona 
of the organiRm, the mutual dependence of its parts, the 
balanoG of the intemul forces t^^Qding towards destruction 
and preaervation, the dimly-apprehended aim of its divinely- 
sustained activity, engage their eager attention. The heavens 
livo luiil mov(\ mid t!ie laws of their life and motion involve 
the material dtistiny of man. It is impossible that he should 
be iudilTorent Uy them. 

Kvon, however, if our instinctive interest in the working 

of the machinB were less keen, we ghould be driven to search 

out the dynamical rulationa of its parta by the impossibility of 

' PML IVofH. voL cTiL p. 302. 

Ma 




THE COKSTRUCTION OF THE HEAVENS 361 



aerwise arriving at a true knowledge of their egometrical 
3.tton& Not only are these variable from one moment to 
lother, but acquaintance with them at any single momeut ia 
ot conceiTably acceeaible to us apart from previous acquaint- 
ice with modes and laws of motion. For our view of eidereal 
bjects iu not siinultaneous. Communication with them by 
of light takes time, and post-dates the scnaiUe impres- 
jas by which we are informed of their whereabouts, in the 
'direct proportion of their distances. We see the stars not 
where they are — not even where they were — at any one 
instant^ but where they were on a eliding scale of InatantB, 
The epoch corresponding to the apparent position of each 
is different, and the range of difference extends over some 
thousands of yeara The reduction of those poeitiona to a 
common epoch so aa to get a survey of the genuinely con- 
temporary relations in space of all sidereal objects — ideally 
feasible at best — could not bo much aa be thought of as 
pOBsible without a preliminary knowledge of their diaplace- 
ments during the centuries, or millenmuma elapsed since the 
ethereal vibrations they originate started ou their several 
joiu-neys hither. Thus the study of configurations blends 
with the study of movements and forces ; the reetiictiona 
placed upon thought by the effort to exclude all but a single 
aspect of phenomena fall away of themselves, and we are (con- 
fronted, whether we will or no, by the stupendous problem of 
be universe as a vital whole. 

As a whole ; but not necessarily ae the whole. The 
sidereal world presents us, to all appearance, with a finite 
system. Human reason would, indeed, otherwise be totally 
incompetent to deal with the subject of its organisation. There 
would be notliing for it but to lay down the arms of our 
underetauding before its transcendental and appalliug magni- 
tude. But the probability amounts almost to certainty that 
star-strewn spate is of measurable dimensions. For from 
innumerable stars, a limitless 3um*total of radiations should 
be derived, by which darkness would bo banished from our 
skies i and the '* intense inane," glowing with the mingled 
beams of suns individually indistinguishable, would bewilder 
our feeble senses with its monotonous splendour. This laying 
bare, so to speak, of the i-mpyrean would be the simple and 



362 



THE SYSTEM OF THE STARS 



certain resiilt oi the ockntlauaKice txd injtnitum af any 
ment of sktorasl objeote coiiip«mhUs with thAt prevailii^^ 
oar netghboorhood. Ualen, that is to say, light Boffer 
degree of eofeeblement in npace. If this be the case, thai' 
<rar raasooings are put to eikDce, and a reil ia drawn ia- 
peOfllTable to ficrutiny. But there is not a particle of evid 
tliAt any sncb toll is exacted *, contraij indications are 
and the awertiou that its pajiuent is inevitable depends np 
malogiee which may be wholly visioni^,^ Ethereal abmip* 
tioo and the actual inte-rceptiaa of stellar radiance by opaque 
mnnHTwi. &ti9 equally the creation of the speculative intellect 
K^ther mode of action ia vouched for by experience. We arc 
theiit for the present, entitled to disregard the problematioU 
elTeot of a more than dubious causa. The sidereal systeai 
cannot be regardtKl as in any true senae infinite. The scale 
of its construction, it \& true, strikes im^ination impotent; 
in the muUitudjnous splendour of ita component?, in the 
number and variety of the subordinate groups constituted by 
them, in the magnificenb play of forces it unfolds, in the dim 
prooesses of developmeut it suggests, it bears glorious witness 
to the power and wiadoin of the Almighty Designer ; yet it 
has limits, and for that reason it is a fit subject for the 
exercise of limited understandings. With further systems, 
" pinnacled deep *' out of our sight for ever, we have, properly 
speaking.no ecientific concern; we only know that "when a 
man hath done, then shall he begin " to declare the wonderful 
works of God. 

liegurding the visible world of stara and nebulee as aa 
isolated, though e^teeasively complei system, we may tiy to 
give the beat order we can to our ideas respecting ita consti- 
tution. Let us see what are the available data. The number 
of stars actually registered, inoludiug those in the Cape 
Photographic Durchmusterung, approaches a million, of which 
three-fifths, or thereabouts, are of magnitudes between the 
ninth and tenth, and the rest are brighter. Beyond the 
limits of this great censug, minute stars abound ; but to how 
many millions they would sum up if completely enumer- 
ated, can only be guessed very much ad lihituvi. Sir John 
Herschel estimated at five and a half millions the stars (to the 
1 Hiru, Conatituiiw de VStpaee CilaU, p, 207. 




rJbmteenth photometric magnitude) perceptible over the entire 
arity with his twenty-foot reflector; Struve ctilculated them at 
twenty millions; anJ it has been vaguely surmised that a 
hyniireil millioas could he shown by the most powerful modern 
r telescopes. The photographic International Chart should 
I contain, by M. Loewy'a reckoning in 1900, thirty million stars 
rto the fourteenth magnitude. Yet Urota the basis of the work 
I done in preparing it at GresEwich down to 1904, the number 
. was evaluated at only 13,880»000. Thus, we are still in great 
I measure ignorant on the point. Different parts of the sky 
vary extremely in richness. In some, telescopic stara literally 
' BWarm ; in others, they occur hut scantily. It has been com- 
puted by Mr. Qore,^ that if the whole heavens were ag thickly 
atrewn as the region of the Fleiadea, the number of stars to 
the seventeenth (nominal) magnitude, would be about thirty- 
thrcB milhona. But the method of distribution within a 
definite cluster evidently gives no clue to that prevailing out- 
side it. A fair specimen-field is, indeed, all but impossible to 
choose. Counts in the Milky Way, extended in the same 
proportion over the sphere, would vastly exaggerate the 
crowding of the stars; which would, in an equal degree, be 
underrated by counts executed outaide it. 

Reliable data on the subject might, one would have thought^ 
be collected with practical usefulness by the method of 
*' photographic star-gauging." Eeckonings of the stars in their 
light-ranks, upon plates exposed for various lengths of time, 
ought to tell with certainty how far the ideal law of augment- 
ing numbers holds good, and where " thinning-out " becomes 
apparent. In an equable stratum the stars must nearly 
quadruple at each descent of a magnitude, simply because the 
cubical apaee holding them is quadrupled.^ Should this i-ule 
be overthrown by excess^ a real crowding is indicated, at the 
di^itTice corresponding to the altered rate of increase ; if by 
defect, then obviously the supply of stars in the region 
examined is becoming exhausted, their scattering is sparser 
than in our nearer vicinity, and the termination of the aeries 
is at hand, if not already reached. Photographic soundings 

^P 1 Joitm, X«i. AstT. Soe. toL vii. p. 180. 

' For some reasoD atill TiDezpUiued, tli« obaervod ratio fttllfl cotiaiiteatly 
than of the theontickl ntiih 



364 



THE SYSTEM OF THE STARS 



h&ve. nevertheless, some special drawbacj£&* Theiz oaa % 
4bove all, hampered by wide disparities depending on subtle 
variations in atmospheric transparencj, which incalculably 
modify, from one night to the neit, the registering powers hi 
perfectly similar plates* 

But even Herschel'e Milkj Way gauges, although indis- 
criminate as regards magnitude, afforded distinct evidence o( a 
termmatlng series in the fact that the numbers of stais 
recorded by him amounted to only one-third of what might 



- g. * — * — ± — -It ^ - J. 

O O ~ ^ 1^ H 



^ — 3 — a — 3 — ■ 



, l<L--DI*tribu.aan of DM ataji within 1" of t^B Pols, uhowlDg tJw mOoof POmbew tt 
flpftcfl for <Acb hjilf-TiujrniUid«. 

have been anticipated from the penetrating power of his 
instrument appUed to an indefinitely extended system. And 
for a " mean sounding " at the northern galactic pole, M. 
Celoria, witU a refractor showing, at the utmost, eleventh' 
magnitude stars, obtained a number almost identical with that 
given by Herachel's great reflector. The larger instrument 
then here revealed no additional stars. Similar symptoms of 
exhaustion in the star*supplies may be found in Profesaor 
Pickering's photographic catalogue of 947 stars within one 
degree of the celestial north pole»* A single glance at the 

1 aftvib J, BuruB, Joum. £riL Ahr. Aa. toL xiL p. 75. 
* Uarvard Aniiahf vol. xviii. p. 203. 



THE CONSTRUCTION OF THE HEAVENS 365 



eynoptical table giving the nmnbera for each hftlf-magnitude, 
BUlBcea to show that the DumeTical i-epreBentation of the lower 
ranks is inadequate. The small stars are overwhelmingly too 
few for the apace they must occupy if of average hrighlneas ; 
and they are too few in a constantly increasing ratio. The 
accompanying diagram (Fig, 3S) represenbs graphically the 
decrease outward of density (or the proportion of numbers to 
epace), deducible from Professor Pickering's enumeration on 
the Bole supposition of the equal average lustre of each class 
of stars. Those of the ninth are the moat thickly strewn; 
the intervals between star and Btar widen rapidly and con- 
tinuously (for the sudden dip at 9 '5 magnitude in evidently 
accidental) down to 11'5 magnitude, when a slight recovery, 
lasting to the thirteenth magnitude, sets in. A result of a 
different character was obtained by Professor Neweomb from 
a count of 312 stars to the eleventh magnitude on plates 
taken at Potsdam for the International Catalogue. Although 
the region examined lay close to the galactic pole, he could 
detect no manifeafc falliag-off in the proportionate inerease of 
number with faintness-^ But until more comprehensive 
surveys of the same kind have been executed, it will be 
impossible to lay down any general rule for the thinning-out 
of sidereal strata. Nor, even if it were BuccASsfully formu- 
lated for the rest of the eky, could it apply to the Milky 
Way, where, "with both bands fuU/' veritable star -dust ia 
scattered. 

A far-reaching influence is exercised by this great zone of 
condensation oyer the scattering of the stara, taken in the 
gross ; but it grow^ more marked with their diminishing bright- 
ness. Seeliger'B ascription to it of infeiior efficacy in the 
southern hemisphere^ may now be tested with the help of the 
Cape DurchmuBterung. The prefei^ence of lucid stars for the 
Milky Way is slight yet unequivocal ; it appears from some 
careful statistics published by Mr. Gore," that even in them, 
the galactic zone is once and a half times richer than other 
parts of the sky. There is some evidence, however, that this 
crowding is towards a plane of condensation distinct from, 
though very close to that of the galaxy. 

* The Stan, p. 284. ' SiisutigsUrkftU, Munich, 1884, p. B21 ; 1856, p. 220. 
* Jovm. Liv. Astr, Stfc. voL rii. pp. 175, 1S2. 



366 



THE SYSTEM OF THE STABS 



A girdle of Urge stara epftoning the southern hemisphere 
WHS thought hj Sir John Hexscfael to be the projectioii of ' 4 
subordinate Rhoet, or stratam, deviating some tventj degree 
from paralleliKm to the Milky Waj." ^ The hint was 
developed hy Dr, Gould. " Few celestial phenomeoa ** he 
sidured to be " more pcUpable than th« existence of a stream or 
belt of bright stars," traceable *" with tolerable distinctziiMi 
through the entire circuit of the beaveoa, and formiag a gnat 
oinsle as well defined as that of the galaxy itself,*^* which it 
croflse« at an angle of about 20^ in Crux and Caseiopm 
Traversing in the southern hemisphere Orion, Oanis Major, 
Atgo, the Centaur, Lupus^ and Scorpio, it porsuee its way ia 
the northern through Taurus, Perseus, Cassiopeia, Cepheus, 
Cygnus, and Lyra, its line being lees obvioasly continued by 
the stars of Hereules and Ophiuchufli* Like the Milkj' Way 
it seema to bifurcate near a Gentauri, the branch there thrown 
off reuniting with the pareot stem in Andromeda. That tbe 
stars thus marked out, to the number of about five hundred, 
constitute with tbe sun a cluster ''of a flattened and somewhat 
bifid form," " distinct from the vaat organisation of the Milky 
Way," grew into a conviction with the progress of Dr. Gould's 
observations. 

The grounds upon which it was baaed have, neverthele^ 
been gradually undermined by close and varied research. 
Gould's star-belt continues to subsist; its reality is unquestioned; 
but its leLationa are dlffereut &om those which ho assigned to 
it. It has assuredly galactic, not solar proclivities. Mr. 
J, H. Sutton pointed out in 1891 that the tracts where it 
ooftleecea with the Milky Way are marked by profound dis- 
turbance of the nebulous streauL* But this could only result 
from an actual intermingling of the two formations. Again^ 
many^ perhaps most of the belt-stars give spectra of the 
helium type ; and the chosen habitat of helium stars is — we 
are led to believe — within the galactic aggregations. Nor is 
there the least justification for holding the sun to be a member 
of any specialised Btellar group. Some ostensible evidence 
derived from proper motions, implying the genuine existence 




^ Gap* SimitSj p. 9S5. 

Amer. Joum. of S^nct, vol. viii. p. 333 (1874J. 

' Uran, Argentina, p, 356. • XtuwUcigt, vol. xir, p. 121. 



J 



THE COIfSTEUCTION OF THE HKVVENS 367 

£L Bo]ar duster, is now known to have been misleading. 

rofessor Kapteju was justiSed, by the data placed at hia 

[>sal, in concluding that stars of the solar type are crowded 

the neighbourhood of our sun,^ bub he fully recognised later 
le defective nature of those data, and withdrew the inference 

ided on them.^ 

The wide separation of the belt-stars from our own 
ace in the uaiverae adds incalculably to the importance 

the phenomenon which they constitute. Their remoteness 
nficessarily enhances our estimate of their size and luminosity; 
the condensation of an extraneous stellar multitude towards 
the plane they affect testifies to their attractive power ; while 
their orderly arrangement, as if in a ring-like enclosure, still 
further perplexes the enigma presented by this surprising 
feature of the sidereal eyatem. Its natural effect in dis- 
arranging the theoretical progression of photometric ranks 
augments seriously the difficulty of searching out the laws of 
stellar distribution. Into those ranks close upon five hundred 
extra stars are intrudedj all above the seventh magnitude, 
which have no proper status among the denizens of intra- 
galactic space. There is accordingly an excess of bright stars 
with small proper motions — -sentinel-atars, one might call 
them, posted at the outskirts of the world, which might well 
serve to arrest runaways on the verge of breaking bounds 
through over -acceleration. Gould, in fact, showed that, 
deduction having been made of five himdred belt-stars, the 
remainder of those he had the means of enumerating, down to 
9^ magnitude, form a tolerably regular series, increasing in 
numbers nearly in the theoretical proportion of their diminish- 
ing light.^ And the conclusion that these five hundred 
superfiuous objects do in fact compose a group apart, is 
strengthened by the symmetrical arrangement, with regard to 
the " belt," of the bright gtai's outside it Their tendency 
to collect towards its central plane was thought by Dr. 
•Gould to be irrespective of the Milky Way, except in ao 



' Pnte. Amsirrdam Acad. t>/ Sdeiuit, April 2% 1AS2 ; Janaary 28, 1893, 
■ md, April 20, 1901, p, 6S9. 

' TUti " eiripiri<!Al ratio" (tLat reaaltlRj; from Actu&l euumerDtion) of multi* 
IpliuatioD of Qumbers per magijitud«, i^ Z'9\2 ; the theoretical ratio is 3'9&7. — 
Uran. ArgetUma^ p. 367. 




THE SYSTEM OF THE STABS 



for AS the two farmatioDfl oomctde bj tbo projoeCioii 

upon the other. 

It is worth QOtioc, U>o» that the present dijmitiaii ol ^ 
aoUr movement agrees with the general " li« '* of the tcj 
although other stars pve no sign of preference for ft&y I 
inoDUil j>lH.ae of revolutioiL Yet^ quite obviooftlj. Utt 
figuriLtiuD of moving bodies is the abrogate exproamoo otl 
motte of mavetneut- A ring of stars can be a sul 
reality oul^ if it be constructed on a dynamical fonDdatat 
IbH mombora niust travel along the JxttiQ of their di^tiibutio. 
Shoul*l they eventually be proved not to do so, thea G<itjl 
bolt will stand revealed as un illusory appearance. V 
question of eyatemiitic proper motions is evidently fundaisexi' 
to the rtrchi tectonics of the universe, and we are still a l<»g 
way from being able to answer it 

In a long series of researches^ Mr. MaxweU Hall of Jamaicft. 
attempted to fii the elements of the sidereal system, reganl*! 
as an undivided whole.* The plan of structure attributed U> 
it WiiA identical with that suggested as barely pos«ibIg to b« 
realised in globular clusters. Nothing could be simpler. 
There being no dominant central mass, attractive force and 
velocity increaee progressively outward, and ail the movements 
in the collection, in whatever plane pursued, and whether in 
orbits nearly circular or highly eccentric, are governed hj ft 
single period, or annus itiagnu^. But the scheme is practioally 
unworkable aa failiug to accommodate itself to the irregularitiea 
and veraatilibies of nature ; and its elaboration conveyed onI)r 
a warning that the great problem was, by such means, Im- 
pregnable. 

The subject is one which, for the present, can only be 
approached tentatively ; it would be highly undesirable that 
investigators should, for that reason, be discouraged from 
approaching it at all. The life of a science is in the thought 
that binds together its facts \ decadence has already set in 
when they come to be regarded as an end in themeelvea 
" Man is the interpreter of nature ; ** to draw up an inventory. 



i Mtmo'tTi R. AHt. Soeiity, vol. xliil. \k 1G7 ; MoKthl^ Notiets, vol. xxzix. 
p, 33fi; xldl. p. 621 ; Ivii. p, S&7 ; IviiL p. 473. In the ooiialudiiig iJAper* 
of the BWiea, Sir. Hall found it «dTi««ble to aubstitui* for a law of fot&i ruyiag 
directly m tho dUtauce one cobstaut at a!l distaocua Trom the ceutre. 



I 



THE CONSTEUCTION OF THE HEAVENS 369 



', is not to interpret. It is true that speculation is 

i^MPOiie to waader into devious ways ; but then " truth emerges 

.ore easily from error than from confuflion," And in sidereal 

ience especially, there i& danger lest inveatigatorB, seduced 

ty the wonderful facilities of novel methods, should eshaust 

eir energies upon the accumulation of data, and leave none 

Cor the higher work of marshalling them along the expanding 

lines of adequate theory. Mr, Hall's efforts had thus a value 

not to be measured by definite achievement. 

It ia scarcely probable that indications as to the 
general plan of the sidereal world euffidently definite for 
purposes of numerical calculation, can be gathered during the 
present era of human knowledge. A limitless field of fruitful 
Teeearch, however^ lies open even now in the systema of various 
^Leg^a of subordination, the federated combination of which 
■we may reasonably suppose to constitute the supreme unity 
of the coamoa. From double, triple, multiple groupings to 
knots, drifts, clusters, clouds of stars, an ascending scale of 
complex arrangement leads upwards to the unknown — perhaps 
beyond it to the unknowable. 

As the outcome of recent statistical inquiries, a distinction 
has been set up between the stars arrayed into more or less 
definite groups or masses, and the apparently incoherent 
multitude forming the ordinary population of sidereal space. 
Every cluster is strewn with interjtieenb stars, and projected 
upon a background of disconnected stars. Even the great 
■nlactic drifts are not only seen through a starry veil, but 
^Sui? possibly (ProfesBor Newcomb thinks) have a starry curtain 
drawn behind them. 

Hence Newcomb's inference that, if we could remove from 
the sky the swarms of stara constituting the cloud-forms of 
the Milky Way, as well as all local aggregations, " we should 
have left a scattered collection, constantly increasing in density 
towards the galactic belt." ^ Sweeping away the cry stall isat ion - 
products (to put it figuratively), wa should find remaining a 
saturated solution of stars. 

Again, Professor Pickering states that stellar distribution 
within and outside the Milky Way is so far identical, that the 
proportion of stars of a given magnitude to the total number 
' The Stan, p. 27a, 




TH£ SYSTEM OF THE SXAES 



in ft&y singU region is ereryvrbere tbe same. But -wiikh j 
the galactic oroa the numbt^rH ar« doubled Tben itr I 
twice as many Btars of each consecutive photoiaetiic nu ' 
down to the twelfth ma^itude. " The Milky Way." be adj- 
"coven about a third of the sky, aad contains about half tl^ 
BtatB. There ia no evidence of a liudt to thp faintneea of eti& 
although the proportionate increase in numbers becomee In 
for each auccesaiTe magnittide/' ^ 

These are remarkable facts; aud tbey acquire a dtept? 
meaning from their aaaociation with physical diveraitaa 
The varjuua spectral daasea are not alike in their mode U 
Bcatteriug. Each obeys laws and tendencies proper to it^ 
Thus, Sirian stars, which form the majority of the entL' 
are drawn towards the plane of the Milky Way. and sup^^ 
the material of its lucent thronga Solar and Antarian star?, 
on the other hand, display no galactic preferences ; they occur 
with equal profusion in all sections of the sphere. "Tbr 
Universe is thus ehown," Profesaor Pickering writes,* " to coo- 
siat of two portions ; first, the stars of ibe first type which, 
although frequent in all parts of the sky, predomicmte along t 
oertaiu plane^ thus forming the Milky Way. The aeoood 
portion conaiats of stars whose spectra are of the second and 
third types. They show no concentration in the Milky Way, 
but are, in general, uniformly distributed in all parts (^ 
the sky. These two portions should be treated separately 
in all discussions of the structure of the Universe, such 
as studies of piH^per motion, parallax, motion of the Sun 
in space, eto. The proportion of stars of the first type to 
the total number increases when fainter Btars aie included. 
while with the Orion stars the opposite seems to be the 
case," 

Our own sun, so far as present knowledge acquaints 
ua, belongs to the class of what HerscheJ called "inter* 
systematical stara"— stars, that iSj exempt from particular 
ties, and exhibiting in their moyements the net result of 
OOflmioal compulsion* How many of them there may be, we 
have no means of judging ; but it is easily seen that epecJAl 
association becomes more prevalent with increasing remoteneos 
from our pout of observation. For tkb. among other 

^ Sarwrd Annalt, roL xlviii. p. 166. ' Ibid. val. Ivl [x 36. 




THE CONSTRUCTION OF THE HEAVENS 371 

Baplacementa among the membera of the several groupa evade 
otice. When their nature develops, as it must with the efflui 
>ff time, it will perhaps occiision embarrassment The sub- 
OJBteace of a dyaamical equilibrium may not always be implied 
yj it Companionship, which on the sidereal time-acale might 

called transient, is even conceivable. One set of t-om- 
binations may be dissolved to give place to others ; a single 
itar may pass from one vast confederacy to the next, seeking 
its fortune, as it were, through space ; or, breaking away from 
the entire congeries of fiysfceins, rush out into the ethereal 
desert, to find itself, after milliards of ages, withiu the precincts 
of a strange galaxy beyond terrestrial ken with telescope or 
camera. 

The more attentively the configuration of the stars is 
studied, the more clearly do special phenomena of grouping 
come iuto view. Among the minute stars of the Milky Way, 
above all, a tendency towards the display of typical patterns 
is, in certain parts of the sky, almost as uiimielakahle as it 
would be in a ball-room crowded with dancers suddenly arrested 
in threading their way through the figures of a quadrille 
or a minuet Yet in the heavens, methodical distribution 
muflt always l>e to some extent masked by the projection upon 
the same surface of objects at totally different distances. 
That, under these circumstances, it should often be effaced is 
less remarkable than that it should occasionally become 
apparent. 

One of the " typical forms " in which stars seem to collect, is 
that of an ellipse, or circle seen in perspective.' Radiated struc- 
tures also occur; and Father Secchi, who early drew attention 
to this curious subject, regarded the presence of a large red star 
in a commanding situation among minor objects, as a common 
trait of physical arrangements,* The spectral r«lationB of the 
objects composing them was made the subject of a suggestive 
study by J. Maclair Boraston in 1893.' As specimens of 
a class of objects to which the " persevering student " coadd 
make large additions " with an increasing conviction as to 
the mutual interdependence '* of their constituents, Webb 

1 Century Magazine, Sept. 1889, p. 787. 

* Mti delV Acctid, Pont. t. vii p. 67. 
' Attr, and AArephys\e$y toL xiL p. 98. 




•r M 



U-' I&e pn* 



W* 



ff 



It mm^ JBifcrrl. be wiMttBd ktet tfe wn 

gif«A tei^ cv b* p«t together by a rvrj otigfat stntch i 
fMf . II «M iton «<nwii ii wun y by Udy Hqggm 
tbe Ei^oai dh tf i btttMD of dote of ladiBB ink does 
mcladff the poaribility, to a |»edispoeed ere. of fanning 
Mo atBMSrt ftoj doind figanit- Ba( Ui« illastzntMm 
not^ nor «u it rinwgnmt to leod ns to set down w 
fnwginT7 tbe viidlile peenliaritieB of stellar 
Tldi mmld be an wlawnft qiute aa nuacfaieToiu as 
naqotlifled csednli^, flfnee there 15 camaUtiv« erideoeo^ 
the peculiarities in qoattiiMi aie, in m&nj cases, zeal 
algnificaot.* 

Mr Froctor expnaed hia conTiction that "star 1 
will eventually prore themflelvee genuine by the mxammlty 



* P^iMnUmt Wat fftndm Ohttrsat^rf, Tob. t iL 



THE CONSTItUCTlON OF THE HEAVENS 373 



."their proper motions.* AnU it is obvious that they can only 

Bubeist upon this condition. Their order would otherwise 

r bo a merely passing coincidence. Stars marching " in 

5 Indian file," are presumably swayed by an identical force 

jiactin,]^ upon Lbem from a very great distance. The group 

J that they form is not self-centred, but makes only a part of a 

larger organisation. Segregation, on the other hand« is the 

diatitiguishing " note " of true dusters. They might be 

described as autonomous demoeracies, each of their members 

obeying the united commands of all, while outside influences, 

although exerted upon them collectively, are without effect 

upon their internal regimen. 

A " streaming " can then be discriminated (at least 
ideally) from a " clustering " collection of stars, by the circum- 
Btance that the centre of movement of the one is external^ 
of the other internal It may possibly be found that the two 
plans of oi^nisation prevail respectively in different sections 
of the Milky Way; there is some appearance that they not 
unfifequently compete or combine within the same cluster, the 
streaming tendency working towards the diaaolution, the 
clustering tendency for the preservation of the system. This 
is not the only feature of sidereal construction which conveys 
a hint to us that the world of stars and nebxilse is in a state 
of transition. We see it in only one phase of its long develop- 
ment. To regard its condition as settled upon an uaalterable 
basis, would be to miaconetrue signe everywhere legible to 
attentive scnitiny. 

Since stars and nebula are undeniably united into a 
single scheme, our view of the universe must embrace both 
clashes. The distribution of nebulae is in fact complementary 
to the distribution of stars. Assemblages of the one kind fill 
in the outlinea left blank by the asaemblagea of the other. 
The Milky Way, so far as can be immediately discerned, is a 
rifted and irregular ring, furnished with innumerable tenta^ 
cular appendagea, and composed of stars in every stage of 
aggregation. This ring, however, has obvious geometrical 
relations with the rest of the sidereal etnicture. It marks the 
equator of a vast globe, of which the poles are canopied by 
le nebulae. Necessarily, too. of a rotating globe, since axial 
* UhiPtrie of Start, p. 84. 



374 



THB SYSTEM OF THE SXAJK8 



aovemaDt aione gms rise to the dwtmctjon bcl w mu poltti 

Iha opinicfti that the ab^te of the viable miiTas 1 1 
■pbericai^ or spbetoitkl. eatber than lenticular, wu fixprni i 
bj Rftdau/ Kkin, and F^b.' The poUr reUUoM a< di > 
ocbuhe to the pUne of the Milkj Waj admit indeed d m 
other iDterpretatioiL And Lheee rolfttums can ecaiteljWT* 
been determined otherwise than bj the rotation on ad txii d 
the coloaaal, uDdirided volnma The oooditioQ tliiis fratwHil 
aa primitivel/ existing m&y have become modified with tine; ' 
ftod. even if still prevalent, should be imreoogiiisftble bjr i& 
Hince the relative Bituatioua of the heavenlj bodies WQc]d«ii 
Falb remarked, be absolutely unaffected b^ it. 

All that we can see clearl/ is that an uniTersal moveaiMft 
of rotation had much to do with the present dis&ibatMii if 
matter ia sidereal space. Whether the forces which hiit 
brought it about axe 8till active, must remain an open queetiM- 
The opposite tendencies of Btars to gather in the ^quatoriil 
pkne^ and of nebnlse to stream towards the poles of the 
s^Btcui, may not even jet be exhausted ; but the decision of 
this point must be left to a dim futurity. The efficacy, in 
the past or at present, of contrnxy drifta would apparentlj 
imply an inherent diSerence in the qualities of the objects 
TOBpoL'tively swayed by them ; and would, so far, contribute to 
iuvaUdate the con;3ecrated hypotheaia of stellar developinent 
from nebulsG. 

Wti can, indeed^ heaitate to admit neither the fundamtf'nul 
identity of the material elementn of the universe, nor the 
Otfbuloufl origLD of stars. The transition from one to the 
other of tht) two great families of the sidereal kingdom is so 
gradual us to aifard a rational conviction that what we sae 
contemporaneously in dififerent objects haa been exhibited 
Buccassiyely by the same objects. Planetary nebuhe pass into 
gaseous stars on one side, into nebulous stars on the other ; 
the greater nebulEe into clusters. The present state of the 
Pleiades refers us inevitably to an antecedent condition closely 
resembling that of the Oriou nebula ; the Andromeda nebula 

^ Bitll AttT. t, il, p, as. 

■ Bandbueh der HimrMlftrtchHnungefi^ Tb. ii. p. 313; Sir(*ti, Biiude tu. 
p.. iO, nil. p. 19H. 




THE CONSTRUCTIOK OF THE HEAVENS 376 



may TepresecC the nascent stage of a splendid collection of suns. 
But even though stara without exception have Bprirng from 
nebuliK, it does not follow that nebulaa without exception 
grow into etara. The requisite conditions need not invariably 
fiubeist. Other ends than that of star production are 
perhaps met and promoted by the chief part of the present 
nebulous conteuts of the heavens. The contrast between 
stellar and nebular d^tribution is intelligible only as express- 
^ ing a definitive separation of the life-histories of the two 
L classes — a divergence destined to be perpetual between their 
lines of growth. 

Progress, then, is the law of the universe. From its 

present state we can obscurely argue a " has been " and a 

,flball be." The face of the skies is not cast in stereotype. 

a vesture Thou shalt change them, and they shall be 

.ged.'* They shall change, by no caprice of hazard, but 

subjection to laws unalterable in their essence although 

finitely various in their applications, divinely directed 

towards the continually more perfect embodiment of the 

unfolding Eternal Thought. 

But the glory of the heavenly bodies, it is asserted, must 
come to an end. It results from a merely transitory state of 
things. The radiations, by virtue of which they shine, are 
the outcome of what may be figuratively termed the effort of 
nature to establish a universal thermal equilibrium. This 
condition will be attained when the frigid " temperature of 
space " reigns in all the millions of bodies which once were 
suns, and will thenceforward revolvCj amid " darkness that 
may be felt," the mechanism of their movements unimpaired, 
but inert, lifeless, and invisible. Is this then the pre- 
destined end ? Science replies in the affirmative. That is to 
say, it knows no better. Yet there is much as to which it is 
ignorant. Matter rests upon a subsensible basis, into the 
arcana of which no inquiry has penetrated. The observation 
of phenomena leads, it may be said, to the shore of an all- 
ditfoaivo ocean of force, the existence of which is implicated 
in their occurrence. That is all we know ; at the brink of 
the ocean we pause, helpless to sound its depths, or number 
the modes of its manifestations, or predict the tasks of re- 
novation or preservation committed to it. We can only 




Sj« 



'-like • 



ft ■ 



D 


■ 


1 


■ 


^^H 


^^^ APPENDIX 


^^^^H 


^^H 


IP TABLE 


I 


^^^H 


^M Tne 40 


BRiaHTEST Stars in 


ORDER OF Magnitude^ ^H 


P 


THE 8UN. BTELLAR MAGNITUDE 


-26-6 




^^H Diialipii.tlon. 




IHKk 


M^- 


SpecCmiEL 


Rdnturlu. 




1 SiriuB 


h. m. 
6 40-7 


-16' 35' 


-1-58 


Sirian 


DtuuT ; iAtp)ltt<3 af vm. 


1 CdDOpilS , 


6 21-3 


-hTZV 


-o-se 


Sma,u-aDUr 


IrmnAlhlKi parHll^s. 




a OiitAUi-j 


14 32'a 


- 60" 25' 


+ 0-06 


SoUr 


Ciaia^liaiit»OfO'3i^Ulil I'CliA. 




Vega 


18 33-6 


+ 38M1' 


0-14 


Siriftn 






Ca]iclta 


6 &-3 


+ 45" 54' 


0-21 


Solar 


Spwtmcoi^cally eompCKmd.. 




ArctnruB . 


14 11-1 


+ 19' 42' 


0^24 


Solar 






Rig«l 


6 0'7 


- 8"!9' 


0-34 


Helium . 






PTocyofl . 


7*4'1 


+ S"29' 


0'4S 


SirtAD-SDlftr 


Ql&uy ; utftllftfl of ISn. 


^_ 


Aclienual. 


l$4-0 


-57' 44' 


0*80 


HuliLin 


RaH brlghtDSM 'SS tlm«i wUr. 


l^^^l 


y? Ceutauii 


13 Gfi'6 


-69" 53' 


0-86 


Heiiiun . 




^^H 


AlUir 


19 46 -d 


+ 8' 36' 


0-89 


Slnas 


BpKtmin hAxy. 




Bctelgeux 


5 ifl-e 


+ 7' 23' 


I'Oto 1-4 


AotariftD . 


Imguliriy TOTtoble. 




a Criicin . 


12 21-0 


- 62* 38' 


105 


Helium 


Otnnt^Jnei two tf\VMX starK. 




Aldeljwftrj 


4S0-2 


+ 16' 1»' 


1-00 


SoUr 






Pollux 


7 39-2 


+ 28" IS' 


1-21 


Solar 






Spica 


VA 19-9 


-lO'SS' 


1*21 


Uel ium . 


BpActTOftCople blniiry. 




Auures . 


16 'J&Z 


-26' 13' 


1-22 


Antarisn . 


Ccnnpanloa glvci t SIHaIi 




Fonmlhant 


22 52-1 


-30° 9* 


1-29 


SLrian 






a Cygnd . 


20 38-0 


+ 44° 65' 


1-S3 


Sirian 


InsenKlblB puvllax. 




10 3-0 


+ 12° 27' 


1-34 


Helium . 






CruoiB . 


12 41-9 


-6ft' »' 


1-60 


Helium . 


linClcui't "oxjKiin itu." 




OftStOT 


7 2S-2 


+ 83* 6' 


1-5S 


Sirian 


CompoDADtB '9'flk kwl X^afiB. 




1 y Cruoia . 


12 2&-d 


- B6' 83' 


1*60 


AtiturUo . 




^1 


e CAais Mnjaris 


6 54*7 


- 28° BO" 


1-es 


Helium 




^^^H 


* Urate Mnjoria 


12 4fi-6 


+ 66' 30' 


l-fiS 


diTiau. 




^H 


BeU^triz . 


6 19-8 


+ e'lfl' 


1-70 


Heliuia 






1 \ Rcorph . 


17 20-8 


-87' 2' 


t-71 


Hfllium 






4 Carina! . 


8 20'fi 


-5»"ir 


T74 


Solaf 


Spectrum cou]]iQ«ilA 




* Orianjfi . 


S8in 


' rw 


1-76 


Haliaiu 






■ ^ Tfturi 


&20-0 


+ 28' 31' 


1-78 


Helium 






^ Catwet . 


12-1 


-69' 18' 


1-80 


Sirtui 






A Triaaguli 


Ifl a8-i 


-68' 51' 


1^88 


SoUr 






Australia 
















' See ^arrartf A%luiU, 


ToL xlviii. t 


ro, <t 




^ 


377 




^ 





HPUP 


^^m^^^^^^^^^ 


^^H 




378 


THE SYSTEM OF THE STAES ^H 


^^H 
^^V 


DwUeuttoo, 


1900. 


IHH. 


1U«, 


Qpsctm^k 


1 

BivrkL , 


Or P^rani . 
t OriootB. t 
q Ur«e Mftjoria 
7 Oeminnrmn , 
0. VtBob Maoris 
r Sftgittuii 
£ Qaais Majona 
j9 CflDifl Majoris 


h.. 1A. 

$17 -a 

fiae-7 

6 31'fi 

len-s 

7 4-3 

8 18-S 


+49* 80' 

- 2" Q' 
+49' 49' 
+ 16" 29' 
+ «a:^17' 

- 26' U' 
'17°54' 


1*90 

1-ei 

I'Bfi 
l-«9 


3oUr 

HflHoffi . 
Helium 

SoliLt 

Sirian 
Solar 
HaliuiQ , 






TABLE II 


i 




EEPaESKNTATTVK ViBIi^LK STABS 1 






X. LOKO^PERIOD VARIABLES ' 




Nune, 


H.A. 




Mai-. 


P«r)«d la 
Pa/i. 




VtrA Oati 


h. m. 


- 3'*2«' 


1-7- e'6 


SS2 


J 
of rally- -20^ from t 


H 


V OrioniB 


5 40-9 


+20" 10'' 


ee-ia^s 


a75 


Symctrnta tbird tf 
brigh:^ H-Unu at □ 




S Qmna^, 


9 29-7 


-arai' 


4-5-10 


310 


Koberti liude aa ii 
Ij) th« periot] com 
■bQut Ss yean. 




H Kornue 


16 28-S 


- 49" 10' 


7 -11*6 


471 


Bi>coadu7 maxliiiia « 
nik 9b«ej-v«d by In 




X Cygni . 
V Del phi Ei 


19 46-7 

20 -13-2 


+ 32" 40' 
+ 18°58' 


45-18 & 

T-a-17'3 


404 
G40 


Period slowly leuKtb 
Bant?^ ol magnitude t 
lutQ<rD. 



2. IRRKOULAS VARIABLES 



SDoradOs 


6 18-9 


-69' 21' 


8-2- 9-8 




In cluster N.O.C. 191 
tnun first typo wi 

H-Usea. 


i;Gariiua . 


10 41-2 


-69^10' 


-1-0- +7-7 




At a constant minin 

IBM. 


R Goronffi 
Borealis 


16 44*4 


+28" 28' 


6 •6-10-1 


... 


8add«Q tUlorsB of U 
this star at nnoe 
tMTvala. 


u HercnliA 


1718-6 


+ 83" 12' 


4-6- 6-4 


... 


Helium apaetram ; 
lines. Badialraloc 
Tsriable. 


RScttti . 


18 42-2 


- 6° 49' 


4-8- 7-8 


... 


Spectnun aolar vlt 
Unea. Radial T«k 
stant at +42 1 
seootuL 


EWCygni 


20 26 2 


+ 39" 89' 


7-7-10-6 


... 


Spoctram probably 
of IbQXta tTps 
(BiplnX 



f^\.^ 



■ 4-:>^' 



^M 


^m APFEKDIX 


^^^ 379 1 


^■^ 


8. CEPHBID VARIABLES 




• 


^^H NuDfi. 


R, A. 

1 000, 


1H». 


Banga In 


PSTiDdlD 


WevuTkat 




^^V 


II m. 










^H Lepam 


4 53-0 


- 2r22' 


9-040'0 


13-9 


Rite very nplA, P«irlo4 




^" 












It Muacn . 


12 30-0 


-ea"52' 


6-5- 7 -a 


0-8 


SpMtrtua voUr typo. 




1? Bagittui « 


17 68-8 


- ar S5' 


4-8' 6-1 


7-6 


i[ntii"ltli1iH!it-chuiR&r Sub- 
*liliiiry [H-nijil of a84 linliv 




If Aquilie . 


19 47-4 


+ 0M6' 


3-7- 4-6 


71 


Diioyvvnd by Plgott, 17B1. 




T Vulpecolfe . 


20 47-2 


+ 27* 62' 


5-5' 6*6 


4-4 


SpHitnMOOfiic biBMxy (rniat}. 




i Cephfl) . 


22 25-4 


+ 57" 64' 


3-7- 4 8 


C-3 


BpaotroHoplG binarr. N'od- 












ecllptilDg. SpHUum wlmr 
type. 




^ 


4. GEMmiD VAEIABLES 




1 


{" Oemiuonun . 


6 68 -S 


+ 20*" 43* 


sa. 4-8 


10-2 


Non-eoUptttiig bUunr. Doable 
perlodlclliy indlutod (Ctatop. 




V Puppla . 


7 fi6-4 


- 46" 68' 


I'l- 4-8 


1'4 






8 AntUie 


a 27 -e 


-28Mr 


e-3- 6-8 


0-3 


SirlAii-HDlar spectrum. 




'W UrsFE Majoris 


36-7 


+ 68' 25' 


7 -ft- a-a 


0-17 


irim (UUllor Hid KMipf, 




^ Lyre . 


18 48-4 


+38-16' 


3-4- 4-L 


12-9 


BnliiuD BpecLnim witli nui&blD 
brltfbt llD«. Dottble pAHo- 
dliJty. 
















V Valpeculm . 


20 32-3 


+28' IB' 


a -3- 9-7 


767 


UEbt-dlKU« uulogoui to 






5. CLXT9TER VAHIABLES 




• 


Ko. 8 td Gentami 


ia20-8 


-48' 67* 


12*8.14 -3 


0'S2 


oftbQ^natwiit'berDcltistar. 
of Qh duraliun. 


J 


Ko. 7 Meui^r 5 


!ft l»-fi 


+ a'27' 


lS-6-U'» 


0-60 


Bltuat^v) In Uie globolAT cinih 
^InLibn. UtnimatulMta 
0-4 of iLotii parioi. 


4 


S Arffi 


17 61 -6 


-49" 26' 


9^6-lO-8 


0-45 


RLtu to iDftilmam In 1 h. lOm. 


^ 


YLyne , 


18 34-2 


+ 13' 62' 


11 -3-12 -3 


0-&1 


Incraufl qf ]fgbt 6ff<«U)<l In 


1 


' 










1 h, BOm- 


1 


14. 1904 Cygai 


20 1-3 


+ fi3* 4' 


10-7.11 -B 


13 


DfaOi:>V9r«d by kailima 


1 










CertHk! La IVOi. 


I 


Iprrcygm 


20 62'3 


+ 30° 3' 


9-6.10-4 0'5 


IUb« coDipleled In t b. 53 ut. 


1 


L 






i 





^^ 


i^^i^ni 


^^^880 


THE SYSTEM OF THE STABS ^M 




TABI.E 


III 


■ 






List of Ecupszng Stabs ^^B 




XkiMoratu-. 


K.X. 
L90a 


Ttw. 


Dv*. 


Fluaein 


— 




1l nk 












UOephet. 


0*3-4 


+81* «r 


a-fi 


7-0- »-2 


Daia«|gb «f itn 




2INfM4 . . 


sss*? 


+ 4I'8«* 


81 


9-4-12 






Al«ol 


a 17 


+ 40' 34' 


3-8 


21. 8-2 






RTF^rmi 


sie-7 


+*rir 


OSS 


9-5-11 D 


in 1»4. 




\ TADii 


S65*l 


+ia*iar 


a-« 


a-a- 4-2 


FliM* IhU 10 Ik I 




R C«ni« Uigorii 


7 11-9 


-wiy 


11 


6*7- 6*8 


Df*ii«tt7 euoKA «i 

tlMt D^ lit* flS 
RuweII> ^J 




TOuidopudftH 


7 27'fl 


+7e' 17' 


3-3 


»*a-u-9 


C*iukl Id liOoTf 




R* Papitti 


7 48-5 


-41' a* 


«-4 


I0-O.11-0 


QMtlt^ or «TsUm i 


1 


XCaruuc. 


a as 'I 


-6a'63' 


05 


7-9- 8-7 


Two tcfcht Hwn, 1 
Id l*% DtidKXQ 1 




SQufiri . 


999-2 


4i»'a4' 


0'4 


8-0-10-3 


Ptimtkra or tihu* llA 
mj-(hOae«i)ar(l 














1 


S Valorom 


S39-4 


-ii'M^ 


5-9 


7 '8- 911 


Ftan (KcajplM UK 
>0^ •oLu- (Ttoa^ 


1 














R» Veloram 


10 17 a 


-41*61' 


I '8 


lO-O-lO-ft 


Dnrmtion of pSum s| 




t Dnoatiis 


11 398 


*73°iS' 


1-4 


9'4-13'S 




' 












GwumiDiaoft. 


1 


fllabne . 


14 6S-6 


- »• 7' 


2-3 


CO. «-a 






V GoroiUB Bo- 


IC 14-1 


+ fi2' 1' 


3-4 


7-t- t-7 


DUBtflon of ph«ae 




r«&|u 










^0-1 wUe (RiiMll] 




R Am 


IB 81*4 


-M'48' 


4-4 


a -9- 8-0 






U Opbittobi 


17U-4 


+ \*W 


0-8 


a-0. 67 
















jean (Ctwndiec). 


1 


RV Ophiofihj . 


ITM-a 


+ 7'W 


» 


9-0.11 '2 


i 


' 


EHarouUk 


17 »^ 


+ 16* S' 


a-» 


71- 7-9 


Two UD«qnftl *^4B 


1 












ptaoe la Hch pH 
















■ 


RflSftgitUiii . 


18 11-0 


-34' 8' 


3^4 


57- a-B 


PMCfOd osBqwUT dir 

« HOOtodazT mlnJiiiui 














p 














[ 












L, 


^ 




^ 






•« est 3 c4 



? ? 





THE SYSTEM OF THE STABS 



111 






5ii£i 






P 
fe 






^ 'b ^ *» 



S 



w »« 



is SB 

^»2 g^ SB 






I. 















5 



g 






-c -9 « 



d 

^ 



o 








APPEKDIX 


383 ^^M 


' 


TABT.E 


V 


H 


List of Stars with sensible PAEAU.AXKa' 


^ 


014 flf fit^r. 


a. A. 


Dm. 
1900. 


M«ff. 


FimllftT, 


AaUiotitr. 


1 


mbndge 34 


3-3 


+ 68' 36' 


2^4 


0"-l6 


Piitdhord. Photogrsphy, 
An were. DiffBrences of R, A. 


127 


+ 43*27' 


s-o 


0*-39 




Dcaai 


U-& 


-e6"2S' 


4-3 


0*-138 


Elkin. HeliomettT. 


^^H 


^dri . 


20-5 


-77*49' 


SB 


0'*134 


OilL HelioiQQt«r, 


^^H 


eHiafwue 


34-8 


+ 55*fi9' 


3-6 


0''-04 


ft-itchard. Photography. 


^^1 


wiopeue . 


43-0 


+ 57" 17' 


3^ 


0*'1S 


Peter. HBliotnettr^ 


^^^^H 


sfiiopeUe . 


1 re 


+ E-l*2«' 


6-3 


0-*l3 


Peter. Heliometer. 


^^H 


ria 


1 22-e 


+ S8' le' 


21 


a"*05 


Pritchard, Photogr»phy, 




ftlhaat 


I 31 


-57 '46' 


0-8 


ifm 


Oill Heliomcter. 


^^^^^1 


tt 


1 39-4 


-l»*28' 


3^0 


(T-Sl 


Flint. Diffprencea ofRA. 




LdjiQi . 


3 IS'B 


^ 4r 27' 


4*8 


o*-ifi 


Ellcin, Heliometer. 




ridAni 


4 107 


+ 17' 48' 


4-5 


o'-ies 


Gill. Heliomt'tff. 


^^H 


^bt^ul 


4 30-2 


+ 18" 19' 


I'l 


0*'i07 


ElkiCL Holiomfltcr. 


^^1 


&bS4S 


B 6-7 


-46" 3' 


8-6 


0*-312 


Db Sitter. Helioiueter. 


^^H 


bUi . 


5 9*3 


+ 46' S4' 


0*2 


o"'oa 


ElkiD. Helioinetor. 


^^H 


»lgeux 


6 49-8 


+ 7' 23' 


1*0± 


C'OSS 


ElkiD. HoHomttiir. 


^^1 


mrigae 


39-5 


+ 43' 41' 


fi-S 


o"-n 


Schur. Micrometer. 


^^1 


Lia 


8 407 


-l6'aV 


-1*0 


0'-37 


Giil. Heliometer. 


^^^ 


ler. Oephei 


6 537 


+ 37" 12* 


0-2 


0''027 


Wagner. MeridiAa o!j»r- 

TatlQUS. 




fM . 


7 28-2 


+32" e' 


1-6 


o*-o& 


CnrtisE. Bpeotraacopic mea' 




syoa . 


7*34 -1 


+ 5'29' 


0-6 


0"'335 


ElkiD. Heliotndter. 




11X 


7 3ft-2 


+ 2SM6' 


1-a 


0--066 


Elkin. Heliometer, 




iadel&290 - 


7 47-2 


+ao"53' 


8-2 


o'-oa 


P*ter. Heliometer. 




tmt! Unjoris 


SM-a 


+ 42*11' 


4-1 


0"'02 


Wagner. Meridian ob«fir- 
TstiDns, 




mde 18116 . 


B 7-6 


+ 53" 7' 


7-B 


o^-ia 


Peter. Holiotneter. 




rsfe Maoris 


9 2S'2 


+53* 8' 


3-3 


0*'09 


Peter. Heliaineter. 




uide 19022 . 


fi 37 1 


+43no' 


8-0 


0"-06 


Kavt<^y^> DitfereDC&B of 

R,A, 
K4i't«ya. Difierenooa of 

E.A. 
Elkin. Heliometer. 




lOoBii Minoria 


9&&'S 


+82" 26' 


5-6 


0--06 




fllus . 


10 3 


+ 12" 27' 


1-3 


0^-022 




pmbridge . 
S13 

Dinbridf^ . 
340 


10 fi-3 


+ 49"S8' 


6-8 


0*'17 


Peter. Heliometer. 




10 21 -9 


+ 49' 19' 


6*5 


r-ii 


K&i>^yn. DilToTSDcaa of 
B.A. 




ombiidge . 

867 

uide 2118^ . 


10 277 


+ 49" 42' 


7-8 


0*-04S 


Eapteyu. Diflie rebcea of 




10 67-9 


+ 36*38' 


7-6 


0'-344 


H. N. RiuselL Pboto- 














gTApbiu measurea. 




utdt 21268 . 


11 0-6 


+ 44" t 


8'6 


0*24 


K*pt9yn, Auwen, and 

Kriiutir. 
D» B«ri. Uiarodeter. 




51« . 


11 S-fl 


+H' 1' 


7*0 


c-io 




In pTeparittg 


tliis liiit recourse h 


IS been ha 


d to BflV'QI 


'§,1 unthorities, MpociAlly to the 




MX colli'ctiaiu 


pubfished by Net 


rcomb ftn 


d KapU] 


n, in The Sian uad Oroninggn 


i&Uiong, So, 


9, r«ajJ>4ctively. 


_ 


_ 


-^ 



-^ 



SM 



TEX STBIXM OF XEK 8ZAB8 



Nimfr of staff. 


[t.A. t>«c 


«» 


PnriliBp 


Aothoiltf, 


A-Ofciim . 


11 14-8 j+e6'28' 


0-0 


O'-IO 


Franz. H«Uoiuettt 


nradWI&M . 


U2»a -ana' 


6-0 


(TilS 


niui DitfereooH 


ZlMl 


liaS'fl +*5M0' 


6-7 


c-oi 


E»pt«yii. Diffm 
B.A. 


ORHNnbridgo . 

1B2S 


iiio-a i+*aM4' 


BS> 


(T'OS 


Orooiu bridge . 

1S30 
Qroamfarid|^ . 

1S5A 


UiT-a +3S*2fl' 


0^ 


<r-i48 


EApteyn «ad othen 


13 4^ 


+40* 49' 


7-* 


y-fl? 


KaptayiL. Diffot 


a Crueu . 


13 31D 


-arss' 


1-0 


(T-flfi 


Oiil. Htliraneter. 


7 Vtrginia 


12 30*8 


- 0^66' 


2-9 


<r-(i7i 


H.N.Ibifoell. Phot 


j9 ComiQ 


13 7^2 


+ as* 2a' 


4'& 


or^ii 


P«tot. Heliametsi 


§ Ceqtann 


13 fie -s 


-69^63' 


09 


C-03 


GUI HHliometer. 


Arcturut . 


14 U 1 


+19^42' 


0-24 


(Toai 


Elkio, Heliometv 


a CentiuH 


14 S2-8 


-«0"25' 


0-OS 


O'^fi 


Om. Heliometpr. 


Pinnil XIV. 212 


14 dl -G 


-20" 68' 


0-3 


O"'107 


Do Sitter, Bclitm 


Uluida 3739S . 


14 G2-S 


+ 54' 4' 


7'fi 


O"-088 


F«ter. Heliometiei 


AnUree , 


leas-s 


-2fl*ia' 


1*2 


0*^121 


Fiolay. HelioDiQb 


If HernnliB 


le S9'6 


+89" r 


3-0 


ff'tO 


Wognar. Mcridui 


i Herculifl 


17 lO-i 


+ai'57' 


3-a 


0*106 


LoATflDWDlth. MIc 


T HerculiB 


17 11 


+ 3«*M' 


»'4 


o*-ii 


W^oer. UeridJu 


r' DncQiiia 


I7ft0'3 


+ M*16' 


4*9 


0--82 


Wagbsr. Meridiu 


A. Ob. 171 1£ . 


usz-o 


+66' ae' 


«*« 


0--35 


Kr^BT. H«Uoi&at 


70 Ophiucfai , 


18 0-4 


+ 2" 81' 


4-2 


0"-168 


Kr^ii. Hftliomet 


Teg» 


13 33 -e 


+38*41' 


O'U 


(T'osa 


Blkin. HQliometei 


Z 3398 . 


18 41-7 


+69"3»' 


8-2 


0--36 


livmy. Differences 1 

AttOD. 


SI AqnUe 


10 20*2 


+ ir44' 


5-3 


0*-O68 


Pet«r. Helioimeter 


r Draoouis 


IQ 32-41 


+ 69^29^ 


4-8 


0''176 


Fater. HBHometer. 


Altur 


1$4£-!> 


+ s'se' 


0-9 


er-oai 


Glkin. HeliameteT 


61 Cygni . 


21 2':t 


+38' IB' 


0-1 


0^-37 


Mv&n of photograpl 


3 Equukl 


21 9-6 


+ 9^38' 


4'7 


0'07 




■ Ibdi 


31 W7 


-57' 12- 


4^8 


0^-273 


GilUodElkin, Hel 


a Omia 


22 1-9 


-47" 27' 


2-2 


0"-0]5 


Gill, Ri^UDmetar. 


Kriiger (tO 


22 24'G 


+ 57^2' 


9-0 


0*278 


SchlcsingeT, Phot 


FoinalbAUt 


22 52-1 


-30' 9' 


1-3 


0'-130 


GUJL HeliotJietcr. 


Lftcailla 9353 . 


22 50-4 


- 36* 2fl' 


7-1 


€•■283 


GilL H^^liometet. 


Bradley 3077 . 


23 3-6 


+ 66" 37' 


a-0 


o"-isa 


Fflter. Heliometer 


85 Pegwi. 


2S67-a 


+ 26" 33' 


5-8 


0*'0H 


Briinnow. Microm 



■ 


1 


APPENDIX 


385 


1 


1 


TABT.R 


VI 




c OF Stars with Peopse Motions oi V''9 ako Upwards 




H of Stv. 


190O. 


Dm, 

1»Q. 


Hag. 


Anookl 
liDtJOO. 


jLtnuukt, 




1^243 


ta. m. 
5 7-7 


-46" 3' 


8-5 


r-7 


1>lieoY«T«d by RaptcTD uid Ihqob 
Ar]{Hl&&(lHri« "kfitif; star " 


i^dgelSSO 


11 47-2 


+ 38' 28' 


6-B 


T^-O 




lU 93&a , 


23 50-4 


- 36" 28' 


7-1 


r*-o 


DUUncMs:!!} Ilght^jttra. 




f»jts4ie . 


23 59-5 


-37*38' 


8-2 


a* -3 


" OouH-* rtkr " in Sculptor. 




de 21185 . 


SI 2-4 


+ 38M5' 


6-1 


6' -a 


CobtKNiuitA 31** n|Mrt. 




10 67fl 


+ ar 38' 


3-6 


4'-8 








21 557 


-&7M2' 


4 -a 


4' -7 






d« 31298 ^ 


11 O'fi 


+44' 2' 


8-6 


4" -4 






dftni 


4 107 


- 7" 48' 


47 


4'! 


CMiif of ■. trIpLft nfibCQ. 




■lopcin 


1 1*6 


+54" 28' 


5-4 


3" -8 






taori 


U 32*8 


- 60' 25' 


0-6 


3^7 


CompDDfwiU SS'Apftrt. 




. U31S . 


16 47 


-15*5fl' 


9-3 


|3''6J 


BUn Cf' aput. 




. 14320 


IB 47 


-15" 64' 


9-2 


IdonUcBl moliOD. 




lie 8760 . 


21 11*4 


- 39" IB' 


6-8 


3" -4 






l«Di. 


3 1S'» 


-43" 27' 


4-4 


a-i 






, 11677 . 


1! 14-» 


+ 86" 23' 


9 


S'-O 






i*iii. 


3 23-2 


" 9^48' 


88 


S-'O 




^^1 


:i bridge Si 


127 


+ 43' 27' 


7-9 


2" -8 


Duubia at 40* lubninj. 


^^1 


II. 123 . 


2 30-ti 


+ 0'25' 


6 


2" -4 




^^1 


de 25372 . 


la 407 


+ 15*28' 


8-6 


2" -3 




^^1 


nis , 


14 ll-l 


+ l»-42' 


0-2 


2* -3 




^^1 


8 . . 


18 417 


+59*29' 


8-2 


2* -3 


Com^QPSBti lA* ilArL 


^^1 


M . 


20-5 


-6F.*2S' 


27 


2" -2 




^^1 


de 7443 . 


3 56-5 


+35* 2' 


8-B 


y-2 




^^1 


B V. bn , 


5 20 4 


- 3" 42' 


9-0 


r-2 




^^1 


)y 8077 . 


23 3*5 


+ 6^37' 


6>0 


ri 




^^1 


am . 


14-S 


-85"^ 28' 


4-1 


r-Q 




^^1 


ie 15290 . 


7 47-2 


+ 30*56' 


8 2 


m 




^^1 


XIV, aia 


14 Bi ^a 


-20-53' 


e-0 


2*-0 


DDtlbtr atu lit 13". A ■Inubtft 
pdoBA t obtWTvv4 by ItinOi, 


■ 














^ ^ 


1 30-4 


-13*28' 


37 


I'^es 




^^H 


le 681 


2 S-4 


^&1*19' 


a-5 


r-B 




^^1 


MinLH 


39 32-6 


+ 69" 29' 


4-8 


l*-9 




^^H 


-a A.G.C. 


13 40-2 


+ 19" 20' 


9-0 


r-9 




^^1 


9 












■ 


t 



u% 




TABU Tn 



8C&1B SI Swnt JjXMMM JfOVBUlB I 



L RADIAL TXLOOmiB 



#kOliODta 



GnMBund^ 
«Cluiii]li4<>ri> 

<L«porU 

If Crohei . 

« AnoTomeda . 

a Phoemeia 
fi Sagittarii 



1 Pegui . 
61 Gygni . 



AldebftT&n 
Capella . 

7 Leonis . 
X Draoonis 



44 

S-9 
4-0 
4-6 
2-4 

4-0 

4*2 

8-0 
6-1 



1-1 
0-2 

2-8 
87 



Sokv 



Antarfanj 

Solar 



Heiiam 



Probably 
solar 
Solar 



-«1 

-W 

-fM 
+69 
-54 

-62 

+48*6 
-48 



-48 

-44 

-34-6 



+ 34 
+ 21 

-20 
+ 20 



O iB| l J.Jl><|ik* 



Dotannlned hj Om, 1 

OunpbeU, loo. ett. Tb 
of UiB ftw hdiui 
known to be In i^iid 

aunpbelI.Ibe.oU. 

Btiopcdaky. innMlli 

., Total ▼el> 

latiTe to nut a c 

(TOQDlO. 

Vogel and OunpbaD. 
H. C. Lord. Aitrofk. 
zzl SIS. 

H. a Lewd. TUDalb 
Spectroeeopie biuuT 
WrightX 



APPENDIX 



387 



2. TANGENTUL VELOCITIES 



Nuna ot Stja. 


X>0. 


PanUax. 


VeLoclt:; in mUe* 
per HCODd. 


Kflii]ju'k& 


ArotaruB . 


0-3 


QTim 


267 


BaHiil nlDcity = -1 &iUh * 
HCODd (N«waU]|t 


lAUnde 1&29Q . 


8-3 


ff^'Oas 


30« 




Gmombridgfl 

laao 


6-6 


(f^ii 


IGO 


Total Tulodly in spau IM 
luileA a •ootHid. 


fi Cusiop«ia9 


5-4 


(T'loa 


108 




A. Oe. 11877 . 


9*0 


tf-'lO 


S8 




Z.C.V\243 


8fi 


a*-3i3 


82 




LftuillB 2B&7 ' 


6-0 


(T-Ofli 


78 




Ucaille 93^3 , 


7 1 


tr-asB 


73 




Ag Brid&tii 


4-5 


r-lflB 


72 




Groombridge 


a-0 


(T-oaa 


71 




1822 










e Gridani . 


4-4 


tf'-ug 


61 




Ulude 21266 . 


a-fi 


0"-23S 


64 




£ 15S1 . 


67 


(T'oas 


50 




j9 Hjdri . 

Bradley 3077 . 


2-7 


(r-134 


49 




6-0 


tf'-isa 


44 




j* TUUOK . 


4-1 


(rise 


43 




^ UrstR MajoriB 


3-* 


0*'078 


41 


A 15-5BI cooiai at C, klLana 
it9 proper niOtlo^a of l'''ia 
(Ilumliiaai). 


31 A{|Uil» 


fi-3 


(f'OflS 


41 




61 Cy^ni , 
a Omu 


6-1 


(r-37 


41 




2-1 


(T-OIB 


40 


Tbiff etar gives ninrlT M4 










tiiuMi tlit> ll^ht of 0T)r vmi. 










HallDm spflctnim. 


UUDde2I1S5. 


7-5 


0*-35l 


40 




Latande 27SeS . 


7-5 


o*"08a 


36 




La]&tid6 ISllfi . 


7-5 


o-'ias 


36 




PiiZKiXIVbaia 


6-3 


(r-id7 


3& 




Lalande 1S022. 


8-1 


0*-Clfl8 


S& 




PoUui 


1-2 


(r-066 


34 




aOLconiaHiiioiU 


1-3 


(f'oa* 


33 




5-a 


(r-o»8 


SO 




^ ConuQ 


4-5 


(T-llB 


£B 




Groombridge 34 


7-9 


(T-aw 


28 





INDEX 



hbbo, diEtribnlioD of nebuIsB, 357 
AbemiliDii of llg-bt, Anmiul, 284, 911 ; 

Hcular, 311-312 
Abiponw, Pltlad-aaoestry, 218 
Atmcj, Sir Wniiani, iolnl KiarVighi, 7 \ 

photagixyhic exposures, 23 
AbHorptioD^ atmaipheric, 21 ; in Stftnt, 

47, SO. 62» &7, 75, 136 
Aduns, detection of speotroecopic 

biiiirifls, IBS, 198, 222 
Airy* Sir (Jeorge, 5t*Uar pumllajt, 284 ; 

investigftttoii of (be aim'4 moTeotent, 

306, 307 
Airy, MJu, discflmnieiit a( twelra 

Plf^iades, 218 
AjtkflH, doublu j(at«]liU of Ktgtl, 195 ; 

(letneisU of f HydtK, 197 ; nvalution 

of » nebulous biiury, 204 
Alcor, compuuaiLilnp with MIzbt, 201, 

213, 328 
Alcyone, spectrum, 59 ; mpdeni cim- 

«pieQot]RneM, S17 ; ieuter of tb« 

Pleiwlea, 217, 218, 219, 222» 227 ; 

prop«r ERoUon, 219, 220, 221 1 actuii 

brilllBncr, 221 i adJKeat avbuJie, 224, 

226 
Aldfltujim, brighlii^Aft, S, 22 ; ipectra] 

type, 41, 4i; calonr, 51, 1 35» 140; 

■«t«lliK», 150 I puraUai, 2fl7 
Al«xiikt]derj ipital theory of Milky Way, 

343 
Al^l, A helium star, S7 ; ecUpeea, 117, 

127-128; nature of ayjitem, 125-130 ; 

pertuT^tiDQS, 129, 132 ^ pirolliu:, 

133 ; colour, 13G 
Al^l roriablea, 82, 126 ; perltjdH, 132 ; 

mean dfluoitj^ 133 ; di^tribotioQ, 134 ; 

npid biuriefl, 18Q; tvrO'foMeclLpsu, 

191 
Al SM, redncei of Algol, 136 ; of n 

HydKE, 136, HO; mngnitade of # 

BridEni, 130; lDngbliii:n of the 

I'teiidea, 217 ; ti6ti£« of Audroiiieda 

nebula, 257 
itair, duifiiatloii, 2; ma^iiailej 22^ 

cpectmnk 41 ; panllai^ 297 




America nebula, photofrraphed by Wolf, 

270 ; vicinity drained of ittare, 862 
Ames, bydrog«u-liuen in solar BptK:trui|i| 

44 
ADtlerRoa, diaooTQilea of tctnpaTary 

stiLra, 90, 92; fudiug of Eridaui, 

180 
Andromeds, >, co1citi7% HC ; triple 

ayatein, 195, 256 
Andromeda uebula, early obHrratioiu, 

lOi 257 ; Hpectnim, 69-70, 260 ■ star- 

blaw in, 88-30, 361 ; Btructnrff, 268- 

259 ; linked with the Milky Way, 

259, 337 ; conatitiition, 260 ; oliliqaa 

poaitionj 260-261 ; intra ynlactjc, 360- 

351 
AbtAjQS, third-type tipectrum, 60 ; doIout, 

57, 137, 140; grwn iiat«mte, 144, 

146. 169 
ADtariaa siaiB^ 35, 50 ; clieialatryf 50- 

61 ; vMintillity, 54 ; nou -galiutic, 

347, 370 
AbthelmuH^ dlAQOvery of a temporary 

star, 85 
AntliiF, 8, «har4£t«r of It^ht • cbaagei, 

121, 126, 127 
Autonladl, viflw of a pUmetary nebula^ 

248 
Apex, of the Ann's way, 1ndicB.t*d by 

Blellar rmpkcenianta, 303, 304, SOS; 

by radial ireio^ititMi, 310; uhigoed 

poaitiotia, 305, 303, 307, 308, S09 
Aquiloe, t), n variable «pcctrC4C0pia 

biuary. US 
Arn, K, an ei;llp«iii« Rtar, 124, 120 
Ane, S, a cluster'TariabK 120 
Arctnrua, oacient idflntificatiffD, 2 ; 

magnitude, 3, 22 ; Bi>ectral t-Jjw, 41 ; 

titauiDm Rb^rptiOD, 45, 327; a^nitifs, 

51 ; tberraal ftlKcacy, 72, 74, 80 ; 

cnlmir, 135. 136, 140 ; parallax. 397 -, 

linuar velocity, 326 
Argelaoder, utar entimentlons, 1, 6, U, 

34fl ; identity of Tycho's Not*. 84 ; 

periodicity of Mitil, 101 ; li^bt-cban;^ 

of ff Lyns. 1^ ; of S Cancri, 180 ; 



889 



SftO 



THE SYSTEM OF THE STABS 



ealov flf Aietan^ IM ; bri^towi of 

•obr kpM. M6; uirthod of IsMt 
1—im SM^ S07 ; ■ — toi of viMt 
otUI. lis ; MDtIm of 1830 Omm- 
teUfi^ttt 

AiMK % aiRDTCrr M A dMbk atar, 
ISO; alowvaii of rrralBtiawi, 1£6. 
171 ; T»ri«hmty, ISO 
Ajrbaua^ lottiboaity of B ebnla, 380 
Atla^ taalwttiv WufctMM, 219 ; pa». 
itU* daplkftr. m 

Allfin> «b l—lfcwii" thMX^Om of 

viiWMiiiy, io». in 

AaTCHk iMiottoii of Ncvn Soprpii, 8S ; 
flrWii ftf Srlm «v«l Procjon, IfiQ, 
1A7 ; star-pwtlLuM, iSS^ £91 ; »- 
dneUoQ of BrKd]«y't dhMmtMBS, 907 

BtHvy, itiscQTwj of dutv - TariiUflo, 
118 ; tb«ir Uglkt-ciima, 119 ; gnrapod 

Btiljr. uitlqn* rvoordt of t^a FIcUdM, 
lU; daoRptioo of * itar-cltuOert 
3SB 

BOl, Sir RobHV ^zallat of Nota 
C>vm. 99 ; of ei CjfKi, ass ; MKtch 
for lar;^ ptnlUni^ MO 

Balnatt. kw of hj^if wria, M 

biiriQW, 4i*ciip«i« of r Orlo^ «• 

UvMid. obMmttou of Som^ Sft. S8, 
«», HS; of olwtoMvtaUi^ 119; 
««u* iHlhln tn[iciim» 906 ; BeibvfeK 
ta And JTuuati ifa« FkfadM, Si6 ; 
I^Mtognpht of e1iist*n> fiS^ ISO ; of 
Mbolou itux 2U, 244 ; of diffiwd 
utelotUte. 271 ; of ■ p e iJmA B d 
gilutfo cknd, iS7 ; dMnctiar of 
^otnkr cilutant,, SMO ; d i wawiy cf * 
i^nl Bobok., 251 ; « nm » te d tow** 
BUnt o5 jiaig^mhaiM, 970 ; obnrrstiDa 
of ■ TuiAblfl Debuik, 277 ; r*"'**^ '"^ 
61 Cygm*£SS 

Bajrw, iiomQucliitQn of Matt, 3 ; de- 
^gnaltiiiiH, 90, lOS, 196y SSfl -, Aodra- 
Difldb nebolk usBotiosd bj^ 357 

IMlAtnj:, ■dlleoa ■ MbaarpHdaot 37 ; on- 
modLfied rmdiatiraig, 48 ; abHuioe of 
iTDO^IiiHB from ipeobma, 47 

B^opohky, TBnable tpfetxaaoopAc 
hiB$nea^ US, 131, 131 ; nnwoo vfm- 
puioa Of Cufeor, 190 ; apectroaeopk 
ponllu, SM 

Bcaaal, Bppt^ci^oa ofBradloj, IS ; ^0- 
tntbvd motioQ of Srioi. 1&0 ; vtoftr 
Aotioii of 61 Cygiil, IKS ; mMmna of 
Ui« Pl«i»ltti, 219, :^, 3S1 ; Lnniti- 
gatioM of ftoUMr pNiOlu. 2S4. 887- 
988, 239 : dtteetioD of * pArtul 
syitom, 329 



BetaJgvox, dowgp*tiott> 3 ; napHak 
23 : tpoctnutt. &l i caXvar, $7, 1*1, 
136, 140; nrlabiltty, 10» ^ pcnlht 
397; liffht.ponr, S96 

Binuj lUis, daOaed. 1£2. At Dfilk 
Sum. 

Birmnekwii, Qittlapio of IM 9ub,B, 
139; diaoovvry of Not* Odw^tt: 
olmrrMinu of red otna, llOi Id: 
th«ir Aggiv^tioQ bvtween Aqnikal 
Cy^n^ 142 

BoeddickBr, ■'■SradMHifin of ItiDc}' V», 
337 

BolUnmn, Tmlidity cf Stabs'l I'*, 
72 

Bond, 0. P^ nl»tin lulxv of Srinnl 
the cBtL. as ; Flio«ogmpliy of ddiUi 
it«% 165 ; rifts im AmdromedA tuM fc 
2S8-S59;azt<Btofth«farAiAtioii.Ma; 
textatsftf Qrioa aeba]U,^S; catalCBV 
of inclndod itan, 368 

BonstoD, «p«^n] eliAnsteiutJcf of ittf • 
groups, 371 

Bcw, dct^rrnirtatioa of tDlar apex, 907 

BoudUnd, period of Mirm, 99 ; notiw «I 
AndroQinlfr btbtilm, 2fiT 

BojB, mdkmwtHral eip«rimeiita cm itw- 
boat, 73 

Bnd]«7, sludjud ol»emtiaiu» 12, 307, 
317; aotie«B gf donbla atan, ]£0 ; ol 
Caxtor, 153, of 61 Cygni, IS-I, of 7 
Aiutia, 1S8; ttallar panllaz, 968, 
364 ; diacOTBTT' <tf tha AbamtiqB of 
light, 384 

Bnho^, Tydio, tempotaiT- stof, S3-&4; 
raoord of Orioiuj, ?05 ; lu^fatDflal 
ai Akjontt 21 7 ; Andromada QaMa 
ignored by, 2fi7 ; atollir pa^OvcSSSL 
283 

BtMiklune, partUai oif Dnoa pUocttf;, 
347 

Bnabana, nugnJtadB of ^ CminB, 106 ; 
Df AOntwri, 133 

Braoe apactzograph, S3, 45, 321 

Bnica talaioope, 34 

BrflDnoWt panllaj: of Dnco xilanatwj. 
347 ; nawLTt^M in ataUiir pafaUu* 
293 

Bnffham, dark laaia in HarctUoi di 

asa 

Bll^c^heIl. bn^tonlog 9f v Caring H 
BartoQ, !ip«<tnim of Looped nalnla, 
Bamhuo, i::htomi.tifl donbl* stan, 1 45 ; 
rejectloa of wida pain. 151 ; di^ 
coT«ri«^ 151 169, 171. 181. 192. 195, 
19S, 203, 204 ; moTtffia&ta of 61 
CjT)^ 1S5 ; obaervatian of ^rixa, 
wtaUitOf 1A6; critioianu of stellar 
ortHtal compwUtioUk 168, 169, 900 i 
qttjdmple atan, 203 ; oonataoGy in 
light gf tispaihuD atan^ SOfi ; doaa 



INDEX 



zn 



pairs in the Pleiades, S22 ; nebulotu 
double htMT, 244 ; msumrea of Elng 
nebala, 27ti -, vsjuhl* oebuls, 27* 

ilciam, in nan aad atara, 42-44, &2, G3, 

npbelJ, bydragen-linei in Mira, &Z\ 
ia ^ Centtturi and y Velomm, 61, fl3 j 
bydiviigcn-tiiivelope of a star, 64 ; spec- 
trum of Orion nebula, Q^^ 7'9 ; spoctro- 
Bcopic binarieidisftoven^l \iy, 121, 13^ 
l&i, 196; tt)«Lr prop>[>rtioDate number, 
1S8 ; tolar LrauilatioQ from iteUar 
radial velocitiM, 3t0 ; jnaport of remit, 
313 ; 5wlftni!»a of faidt lUrh 321 

Caoori, 3, Mgol-viiriabltj, l;]0 ; llgbt- 
eurvfi at mtnLmom, iSl 

Cancri. f, calonra, H7 ; qttftternary 
■j-stem. 19S-200 

C«m;oti» IdiiH, liiacaamau. of stalljar «peG- 
trogTanis, 2(^ 

Canapiifl, mognitadt, 3; ipMtnuiL, 41, 
41, 2&3 ; inseiLStble pftnlUx, 293, 297 ; 
nat brilliaocy, 293-294 

CapslU, apparent brightneaaj 3, 7 ; 
photometTic rauk, 22 ; spectral typa, 
41, 4% 44, 75 \ bibary cbatacter, ii, 
189 ; Qltn - riolpt apectrum, 77 ; 
panllnx, 190, 211 ; stage af deT«bp- 
loeitt, 212 

CarboQ, in aim and stars, 42, 51, fi4, &5 

Carbon atars, 35 ; character ot spectra^ 
M, 55 ; aSluitiea, 56 ; aamber ngu- 
t<red, 67 

CariiuEi, X, uatore of light-chaaga, 124, 
126 

Carime, ij, spectrum, &2, 96. 109 ; light- 
chADgaa, 103408 *, aitoAtioQ, 269 

Oarringtan, ditwAruiaenl of PUiduies, SIS 

OauiopeiB, B, 84. Set Nova Coasiopflite 

CauiopfliB, 7, gaspou.1 .tpflctpim, 69, 62^ 
87 J variability af Wight lities, flO 

Camiopein, 17, colours, 146 ; orbit, 16'6 ; 
masaas and light>pow«rDf compnaantB, 

176, ne, aio 

spactrftl type, a27 
Caslor, apectmm, 41 ; a doubJa star, 

160 : i>eriod of reTolntion, U3, 167 ; 

diaURt attecdaiit. \&9. 197 ; lumi- 

ncnitjr, 176 ; a double Kpectmaoopic 

binary, IftO; ^pectKWWplc parallax, 

1»0. 236 
Oeloria, doublis annoJar tbeory of Milky 

Way, 345; star-souadm^, 364 
Cantnuri, R", a ibort-period nri«M«, 

126 
Centatiri, a, mM^gtiituilti^ 3^ 22 ; fluo-like 

apdctmtiL, 46 ; obiHerred aa a dodblfl 

star, 15Q ; orbit, l&S, 174 ; joiut masi^ 
, 164 I Quu and luiumosity of com- 



paaents, 175, 210 ; «p«ctTO»opio 

parallaT, 137, 236 ; aQDual [Mrallax, 

287, 292 I pfoper motion, 322 
Cfinlauii, fL, HpfflGtragraro showing bright 

lino*. 60 
C«nt»nrij hi, pftTiAbilityfif dustered cnm- 

poaenta, 110. 120, 236, 237 
Cephei, Ui nature of eoUpsBS, ISl ; tima- 

ineqnalitiftt, 132 
Oephei, 9, abort period variable, 82, 117 ; 

a spectrDBcopic biuuy, 118 ; blue 

attendant, 146, 169 
Ceph^id Tariablas, 117, 118 
Ceroski, v»rtabilitf of U Cepbd, 131 
Curaaki, Madame, diacovery of variablvs 

ataris 121 
Cb&mberlin, Diigiu of spiral nebuls, 278 
Chamberfl, catalogue of red ttan, 159 
Cbondlar, catalogue of Toriahle atani, 97 ; 

perioda, 99 ; relation to Dolonr, 114 ; 

inequalities nf Algol, 129 ; raiiatiouj 

of Y Cygni, 132 
Chase, ponillaz nf Algol, 133 ; paraUactic 

Hurv»y, 30O 
Ci{»ro, Qolour of Sinus, 135 
Clark, AlvAUr relative Ittatre of Biriufl and 

the aun, 22 
Clark, AWan G., Bruce teleacope, 24 ; 

discovery of Sirian Mtellite, 166; staia 

TFtthln tlie trapezinm, SOS 
ClnattiTB, Drganio unity, 215, 233 ; mode 

of origin, 240 ; spectra,, 241 ; crowded 

in Milky Way, 356 
Clusters, globular, lystemlo constitutlut, 

233-235 ', raria'bLlity ot component 

atius, ^36-237 ; diRtribution of itara in, 

234 ; trariou^ty oi^uuMd, 236-238 ; 

bou-ndbnlQua, 240 
Ctiiat«ri, b-rv^liir, tnotlaa of confbrma- 

tlon, 227-^26 i blfld and perforated, 

229-230 : diTerelfled ahapefl, 231 ; no 

evidflncn of Intemal mobility, 232-233 ; 

aebulfir relatioDH, 240, 241, 26S, 350 

ClMt^jr-TAriablM, 117, 1 18-121, 23fl-237 j 
in ChrioD nebula, 268 

Coal-oaek in Milky Way, 336 ; photo- 
graphed, 337 ; hypothosU In explana- 
tion. 347 

Cocoou bebula, aitoatad in a Btar-laeuais 
363 

ColDur^ pbotogiapbic afFeets, 26 ; of 
tantporaTT itara, 84, 86, &7. 03, 96, 
141-142 ; of Incid star^ 135-136, 138, 
14{1 ; of double Stan, 143-Ha, 160; 
allc^gcd changM, 147-H3 ; of neibalAf 
249, 364 

Comaa Soli, Tariabllity of trapedum- 
atan, 206 

Camnion, ptiotograplia of ntbulEe, 28, 
31, &9, 252 ; nebnbv io the Plaiadoa, 
224 



m 



i» THK 8XAB3 



«<K1 



K»; 



^t 



, Z3i 



S»:tfi 



tatty <tf 

0|^ Ci; m i%J Miilihi. ISriBS 
QF^ T, fcfi rrt i rBi i ^ IM; W 

■< ii|iiiBUl4»-ll»; ■mwiry h 
> p«fa; IW ; 1^ rial iiiiBMiliM, 1» ; 

Omi. X. •!»£«»», HI 104 ; M^H. 
&BVm, lOS-lOft, 111 

0|S»1» M, dilMttM •• • pH^ 1«B; 
pnpv wOo^ IM, lU. SS; flffUMl 
tiMlaa^ lift ; iMdu. 387-SBak 3M; 

Dnqslv* dlHVfWT of Bii« be^K 2M 

I/jfciniC ilifcMUltirtll a( * iwi lUr, 141 ; 
tamianamtiim of " pdd dart" clBstar, 
^9 ; qMctram of a hIpbI^ 24fi ; 
|M^«r ibotlda of Dnw fTwuUiy, 
347 : nriilife iMboiab 377. 37B. 279 

DKTwfaB. Bcvma^aqsfllliriBaf 129^3(M; 
eutb noon tjsttn^ 20S. 212 

CrgDi. SS8 

DAwea, donbla OcsultAtioD of AnUro, 
1i4 ; colonn of double utmn, liJt 
lift ■ ol-MrrfttioD of 44 Bootift, I7ft 

Dmibawiki, coloitni of doable vtmn, 147, 
144 ; obMrTRtiotu of pfttn mfabl« in 
tlcbt.lS0*181 



.an 



■».■< ■ ■Ifc rtK% t»-tM; tf 

Ml mmm mm, «7. M7. S7V; rf 

jq. !«, i*iL in, iM 



^•Wi-^' 



IW 



rtf^Maltcrel^ 



DB^tentef^eDiaai, 143-1 4«, l$9-1«e: 
*■■ lii iwtafcifii^, 14«-ia ; «d7 
UO; rtrfrifw, l&li 
m, U3, 1S»>1«0 ; nWMi 
U6; fiw|MAi m ■■■■ itj , 
1C1 ; ■■■■■■ I rti, 192 I QMpvte- 
tMH. ICS-1C5 ; ffcnfc»ip1ih RMrii, 
1M-1S7 ; ortibd nJBtirM, 1C7-172» 
173-174; mmmm, I70>17»; te»^ 

S0» : variaUitr in l«hW ITft-lM ; 
vtolntion. aM-111 ; pnpv ■otigi, 
322 
t>Dable nebalB, Bnnmw oa imn mm, 

276 
Downing iiiiiiniiiwli of tlta ByxJes, 337 
Ih-aoo plsaHuT uhnla^ tiMiibl* pmxtd.- 

lAx. 347 ; IwUc*! coaftknutMo. ftM, 

25« ; rwlulTvladtr, 275 
Draper cKtala^e of jitell&r «pertn, 21 
Dreyer, obserratioa of V HydiK. 140 ; 

allef^ chto^jts in ii«bat*, 276, 27d 
Dumt'-bell deba]^, Lnd^tennuiftte cliarae- 

tei, m ; ■na.togoqi objwtk, S&6, S5l 
Ptm^i dmnn^ of fooitlk-typa speetn, 

H ; i&cqvAlitiet of T Cygni, 132^ ; 

obaarvAtianH of rod sUn, 140, 141 ; 

of doabia lUn, 179, 163 



INDEX 



unlop, obKrvbliDiu at doable eUn, 

autmuig, Hoan, d» 7, 11, Stf ; 
photomfltric, 21 ; Cap« Pboto^rapliic, 
2€, 126, 322.362.36^ 
jBoa and Tli«i:keray, ilatenniDiitioti of 
the ffiiu'i! ap«E from Slrias und solar 
Btai^ 308 

Bkston, iRophotjJ cnrres in MUky W*y) 

337 ; 9pur»l g«laxy> Si4, 34:i ; relation 
r of brigbt st&rs to gdactlc BtrDcttim, 

310 ; □ebiilBr<lutxibTitioD, S54 

arhArd, «po3trogT»phic atndy of x 

Cygnj, 54 ; movctiifitila vitliln Orida 

QebuJa, A& ; binuy aitnre of c Auri- 

gw.% 105 
Bberhsri! and Hartiaum, origin of cpark 

»p«utruu, 76 
Ebert, uLamiLlDUl dUpeniOU ia «t«Iliu: 

■.tmoflpheres, 62 
Bcosutiioity of stallar orbita, 170. 173, 

177t UB, l&b- QTolutioDArj sigoiQ. 

cuice, 209 
Ecli}i8es, o^vAn&^^le Ht&rs, ItS, 123, 125- 

133 ; duplJHizate, 125, 13!^ 
Elduon, experirucntfl od steHiLr beat, 73 
Etutn. flitctuftling bjightne^ 21 S ; rtal 

ipl«fiidoDr, 221 ; nebulous counectiaai, 

224 
ICt'uctric&l conditioDfi in st^ar atmD-> 

jipbvrefl, 41, 47-48, 75-76, 78 ; indi- 

cated by butded spBctn, 4fi ; Id 

nebtilje, 70» 280 
ElflmtliU of stflUar Dt-^it», 170, noU 
RLkin, hclioitieter employed b^, IQ^ S94 ; 

meMuremetitu of the Pleindw, a20, 

221 : pnrall[iXe» of sOlltlicra etam, 292- 

293, 297 i of utn-them itare, 294, 3i97- 

29& ; meiDi panllax, 2H, 307 
Ellartnann, «yftDog«i tiatuig^ io Antariui 

s\as«t 51 ; apectroji^rapbic iovesti^tloD 

i>f t:arbou star:!, 5b 
EIlipticAl iiebiiliij, 243, 257 ; foresliorteTi- 

iag, 254, 260 ; examples, 257-261 ; 

rirtod, 261-262 
Engelmauit, Ntm Aadromedse, 89 ; 

colour ot 3 Cygni, 143 
Enhanced llnw, 41, 47, 76 
EqaatoresJ tfll»ao|n, ifeacribed, 15 
£!qiiul«i, 3, period of rerolutioD, 16>8 ; 

■pCctTascapic puraUix, 16«, 1S7, 2&4; 

Bweutricity of ofbit, 177 ; opticftl 

coiDpAuion, 201 
E>]aal«i, e, triplo system, 107 
Eridaoi, o^ mau onA ImQinosity, I75« 

176 i triple BTttttn, 196 ; rapid ototion, 

293 
Kipin, detHtion of carbon ilars. 57 ; 

Bpwtrnjn of X Cygnit 104 ; Ught- 

period of 33 Cygui, 108 ; vamMllty 



of a red star, 110 ; of U E*uppifl^ 181 ; 

Birmingliwn'B Ited SUra, 139 ; red 

«Un ib Porutti cluster, 23? 
Evershed, titabiubi In dolflr promliienciM, 

45 
GvoktioD, of aUn, 36-37, 40-41, 46-47, 

51, 56 ; coEirlittoned liy tempenitar^ 

77>73 ; of BtfiUar syBtcina, 208^214 ; 

of rtar clusters, 220, 240, 267 
Extinctions, method of, 20 

Fabricias, discovery of Mira C<!Li, 90 
Falb, nhape of Ibe viiiible uuivcne, 374 
Fallows, magnitmie of 7 Cftriqim lOG 
Farley, colour of BCuphei, 110 
Feu«tr map of ^lars in M«8si«r II, 230 
Feiiillcf, obHerTation of a Cflataurl, lEiS 
Finlay, magnitude of 1} Carinas, 107 
Fizcuu, iDoUon^tbifta of apMtral Uuw, 

135 
FlammarioQ, colonn of 70 OpliLucbi, 

147 ; Tftriflbllity of 7 Arietis, 180 ; 

anomalotis Bjovemant of f Cancri, 

1^9 ; star eysteiii ia CygnuB, 329 
FlaniHtecd, aIu- dcKignntiotis, 196, 201 ; 

aberrationaJ ghifting of tbe pole atATj 

233, 284 
Fleming, Mn., BpectTograpbto dlicoreriM^ 

29, 52, 57, 62, 64 ; of Nan, 90, 02 ; 

of M! ec)ip*Lng star, 182 
Flint, rcHArcbeHi in «teUiu' pArallnx, 800 

Komalbnut, Apectrmii, 4l ; p«rflp«ctira 

FqwI^t, tit&uium^flutingB in AntarUo 
Htars, 51 

Prankf, colonrs of 70 OpbiutiUL, J47 

Franz, pBrallaX of Noirtl AudrODiedl^ 
SO 

Fraunhoftir, O'bjective piisni, 2$ ; in- 
stntmeotal refmenient*, U66 ; S«nigi- 
b«rg beliometer, 239 

Frost, »peatrograKt of Nova Geminorum, 
05 ; bctium it&rs. aa clote bfniulM, 
169 ; radial 7Tiotion of f Cancri, 200 

Froal and Adanj?, sp«clrngT*m of 
Arctiima, 4Q ; Bp«cLroaCopic fiOtD' 
poneut of 8 Orionia. 264 ; low 
TQloatiea of helium ut&rs, 321 

Galilflo, observation of Nova Serpentarii, 
84 ; BtflUar parallax, 232 ; coiiipaaitiou 
of the Milky Way, 333, 341 

UsMcns BtATs, triple ippatrs. 58 ; (Irrt 
reeD^itiotis, 5B ; iipectral rariability, 
60 ; nature of AtmoAphero, 01 ; 
eETecta in, of dispernion and aoomalous 
refi-acU&ft, 61-02 ; priatqaLic ffwwepa 
ffir, 64 

Geminid variablefl, 117, 121, 125, 12fl 

GemiTjomm, f, a apeetroacoplc binary, 
121 ; dlaturbuu of motum-canre, 1(12 



Eaw 



THE SYSTEM OF THE STAES 



OwniaannB, % leng- period inulkl>l* wiil 
■pw Uw uo p to biuur, ISiaaa. 103, 
1»7 

QUI, sir DtTid, pbotogT^pMc n^tra- 
tioB of «lan, fi ; tb4 auo'^ ttellw 
mignftniffi S^ ; n model wjIat «Ur, 
4fi i •p«c|jiua of ■^ CiHdb. 83, 10S : 
lM«tngism o{ ft Pfaoniiiru, IBS ; 
[Motognpli of 47 Toooui, 337 ; eiu 
plormiat df Uie hAliouiBlftr, 289 : 
pumUftUi of ■oaibara sUn, S02-£»4, 
tt7 ; ftTsnga pftTnUax, 294 ; roUUoa 
of bright vtan, 391 

tifllu*. rnloar of fl Curins, 107 

(lla>i!Dapp, compatation q( iteUar orbits, 

107 
Ooldiohmid^ nobnloas oonditira of 

FlviadM, 224 

OcoilndEe, vsmbilitjr of p Ljms, 122 : 
edipffc tLeoo' of M^l, 127 

Qmv, ihe mn's fltdtar mx^itadct SS ; 
•tellftT orl^its, 147, 182; up«ot of 
8& P«irui, 171 ; iiiuibi%btD«tf of 
titiurles, 176 ; ealiiii&t«d number of 
Uw cUiH, S63, US 

Guthudf tp«ctr^ni of 7 Ciusiop«i4i^ 60 ; 
pbol^npb* of ucbulie, 250, 2&3 

Qoalil, UruiDiiicLry, 1 ; viLria.liiltt]r of 
R NoruuB, 104 ; obaerratioiu of nA 
Bian, 140,141 ; pbotogrApbicmttunre- 
mput of Ibfl Pleiades, 219 ; iirift 
Roathem ttmn, Z2'i ; luperpoaod 
gftluitB, 547 ; belt of bright it*rs. 
36S 

Oroombridge, ctnumpoUr catalogiui, 
308,817 

Groombrif^ 1S30. proper motion, 322; 
ft nuMWAj ffUr, 325 ; spectral type, 

«a7 

QtiWj, roocrd of Not* Audntmed*, 89 
Gutholck, periodicity of Mir», 100, 101 

Hkbn, central itftr in ring nebula^ 2&1 
Hale, double'slit mfttbod in soIat photO' 
g^^phy, 4? ; hoIa? flocculi^ 43 ; cyABO- 
gBD in Antwian rtini M ; tp«ctro- 
gnpblo izinftti^tdoti ol cubaii Btom, 
CS; their ftffltittl«o^ 6ft; r*v«rtaif; in 
metallic spettm, 59 
Hikll, A., pontltx of Nova A&idroiiitdiJL', 

89 ; cntalognH of Htars In Pmaape, SSd 

H»ll, Maxw^l, sprwd of Meropfl uebulu, 

S24 1 blemuto of ftideruJ eyit«m, 308- 

S89 
^lley, absorrrtiona of wbol*. 10; of 

ctufton, 230, 239 ; niiignitqd« of if 

Canoit, lOa 
Harding, biigbt and fuut aectloDj of 

ring n«bii)at 25A 
Hbrtung, outburst of Not* AndromfldjB, 

«9 ; callpwu Of UZ Cygni, 1S2 



Hamrd FhotoaMtriMt 31; ■paoBo- 
sr*phi« MumyA, 2S^29 

Hois, flAr-fittuni«tmUoti, 1 ; higb ni Iw 
tnvKUBB of Mir*. 100 

BeU«al Debnlo. '2i8 

HftltoiMter, Yal« Oollf«eb »0, SHW; 

*d&pt*UMi to phvUmUo iiih 

289, 296 i Kooigsbcfs, M(^ »l: 
Cap?, 293 

Helicm, ft terresttul cletnuu, X ; a^ 
jorptiou in Btsn, S7, 47 i tmjmm 
66 ; variaUo bright line, W ; io* 
nfrHti«9 power, 63 ; pi—il s 
DcbslK, 67 

Hethllii-slar& jreceoi diBcrtminitioiw 8C : 
Dflbalar affinities, 3«, 46, $& £0; 
mbvorptiria feAtima, 37 ; gmUcCic «»■ 
d«ti»tioii, 37, 347 ; eubanettd sptctat 
Uneft* 47 ; emiasiw spectrm, £9, 80 : « 
taaporary appmritioii^ 9ti a* fchfrn- 
Tftriablo9,I83; ■■apectratoopie bovrti^ 
1S9; TcmotmMB, 299, SW; omit 
linoar TvlocLties, 321 ; gBUeUfi em- 
dwuatioa. 347 

Hfttmert, catalo^e of stars In U 11. tSQ 

Hettdenob, phrtllax of a C«iUaH, 287 

H«lu7, Faal and Prosper, photcfn^it 
tdoHope, 25 : photography of doabli 
•tan, 18€; of tbe Fldadtt^ S19; 
dLscorery of Pleiades oebuke, 2H 

Herenlo^ cluster in, eentr^ eoiidjSUft- 
tiOD, 9^i ; dtBtnbiition arul ntimbtf of 
COCDpOOOIlt*, £3^ 2^9 ; Sulraiiaed dic- 
tancc, 210 ; spectruiu, 241 

HercuUa, a, flated ■peetnuo, iBS; 
TariabiUty in lii^bt, 109 ; gptoi oooi* 
pmuon, 145, 178 ; their fflwt*™ 
proper motlob, \6B ; ftpuriooa ptmllia, 
S9S 

Hvrcvlia, 95; colour obim^ 147; ft 
physical piur, IfiS ; nlatlTo fijd^, 
171 

Hfirachel, Corolinft, atteadant to Aiulm- 
mflcla Dtibula, 35^, 259 

Htjnscbel, &iJF JobTn^ stfUftr DomeiHslatart* 
1 ; oatbunt of 11 Cikriii^ 107 ; rvAaam 
of stftt*. IS7, 143 ; coloum ofy Leonifc 
148 ; of dfi HnculiA, 147 ; disturbed 
uiotioQ oF 70 OphJacbi, 155 ; re- 
rctatiolu of t Cuicri, 199 i nebolom 
doubJe Htara, 304 ; deteoitoQ of dxtb 
trftp<BimQ-»tar, 305 ; coafonnition of 
clQst«n, 229, 231, 235i, 337^ de- 
scriptions of clusten, 330, 2M ; 
mejuiurentcnt of jvirBtled cloftter, 332 ; 
laiimbct And dyntmieftl MUkditioo of 
gLobubu- cluat«r«, 233-235 ; varietiei 
of nebubB, 24&, 249. 251, 252, 2fil. 
262 ; aatelUt«-st«r«, 2^0, 278 ; Wbiri- 
pool nebula, 253 ; doublfr nebul*. 
2&4 : dumb-bell, 2S8 ; ecpoct of 



INDEX 



395 



Orion uebulii, 264 ; nebula in Ai^, 
S68, 279 ; Omega nebula, 2€D ; ligbt- 
cbacge in nobulae, 218, 2^0 ; stuUar 
pftrallikZ, 2SS ;. iiyBteniatic movements 
of star*, 331 ; the MiJky Way is 
Argo, 33.^ ; gDJoctic Ktrudure, 336, 
833, 3S£^, 34'i ^ nebular distribution, 
SftS ; galactic relations of a claQter, 
Sfi9 ^ nuetbt-r of tUe aUra, 362 ; belt 

of bright aUia, see 

HflrftcLel, Sir William, observation!) of 
naliulB, 11 ; iiUnetary aebulie, ti5, 
24S, 24fi I maxlDiam of Mira, 99 ; 
variability of a HsrculiA, IQO ; coloufH 
o( Htars, 137. U3, 145, 148, 1-17 ; 
diACoveriei of dopM« •tare, 151, JCfl, 
196. 1&7, 193 : t^triori of Cantor. 1^7 ; 
lnagnituiJ«y of 44 BooIlb, 179 ; donbU* 
trtSle star, 203 ; arrundteoeiit of stars 
in diuton, 229, 230, 23& ; cluster in 
Fenea^ 2S1 ', components of Her- 
cules cl'usber, 2i^ ; ntibnloua Btar», 
24S, 244. 245; riBg-Behyfw, 262; 
diffuMd DBbBloi^tit^. 267, 270 ; Omegn 
nebula, 261) ', touiBta and. Dtibulte, 27'2; 
Variable uflbulte, 278, 279 ; mv«it)- 
g:!Ltbn of the bqu's moTeoient, 304, 
»05. 306 : syBt«liiintLc parallax. 313 ; 
atmctim of tbe Millcy Way, SAfl. 13S. 
340; due-theory, 340-341; '^hole" 
in BoorpiOi 342 ; replAceiueut of stara 
by nibulie, 352 ; conatmction of the 
hMTBDs, 300 ; atar-j^augBS, S&4 ; Inter- 
iiyntematicAl Htars, 370 

HeTeliiia, d&iignation of Mira Ctti, 99 ; 
rccorii of R Cepliei, 110; abHrratioD 
of 9 Ortooia, 205 

Eliggena, coili]ur-i;hanga of 95 Euvolia^ 
147 

fiind, erlnuon tint of R Iieporia, 67, 
140; obHrratlanB of effati Norss, 84, 
S& ; diacovery of Nova Ophiuohi, 65 ; 
variationa of 8 CAnerU 130 j ohserva- 
tiDD of a colour- vnrtaliln, 141 ; de- 
tection of n varial>la nebula, 277 

Eipparehnsj star deterraisatifjia, 12 ; 
Mw btar, 62 ; duster in Persuu% 231 ; 
ao uDtkii of Antlmmeda oebula, 2&7 

Bp^Qt, diatauve of Atar-vytftem in Ursa 
MlJQT, 32S 

Hold«Ttr atudj of Orioa nebula, ZS ; 
■MloDftt oif double Btara, 144 ; atellar 
^roup in irifid nebulai, 207 ; beUcal 
bebulK, 243 i stmctura of Satum 
nebula, 250 ; stars in Orioti nebultt 
268; type-uurve of nebuliBt 2/4; 
tt*inil<iafl effetta in Milky Way, 339 ; 
atiLr pwtterDB, 372 

Holwarda, light-chnnge of Mira, 93 
^odke, twinkliug of «tars, 4 ; duplicity 
of y Arietiat 160 ; obserration of fifth 



trapeziDm-atar, 20& ; ccuut of tbe 

Fleiadesi, 218 
nniigh, (luatemary atettar HyAteuiA, 203 
Iloiizeaii, visual cnuuieratioi}, cratjkra, 1 ; 

dtudy or galactic Umiiitwity, 337 
Hug^ina, Lady, Bxtthcial i<tu-palt«rBS, 

372 
Huggina, Sir WiUiani, nieaaurement of 

radiai inotiDn, 18, 29, 194 ; q«« of 

dry pliit«.l, 23 ; Jaiutueas of bebulru, 
27 ; apoctTogTaphiti mbthoda^ 29 ; dis- 
covery of hydiogBD - Miiea Id wbit* 
fftars, 3S, 39, 40 ; spacti^nuu of 
Aldebaj-au, 45 » Apactra of AiidraUJifida 

n«bula, 70 ; of Nova Coroimi, 8& : of 
Kove AnilromedK, 90 ; of ilercul&s 
dtttter, 241 ; of Draco plmctAry, 
247 ; eiperinjent^j on Btelliu- faiJAt, 72 ; 
inventiou of ntiectmg elit, 148 

Elugginih Sir William &ttd Lady, Alluof 
Stellar Spectra, 29 ; calcittlQ In the 
iiun» 44, 76 ; deteiraiuatidn of Qhicf 
nebuLor lincj 6fi ; spectra of Orion 
and Andl-ooieila uelmliB, 67, 260, 264 ; 
nltra-TiuletstHllaripectra,??; apectro- 
grama of fi Cygui, 1 id ; spectroflopi'U 
oompoaent of 9 Orionin, 206 ; wolu- 
tioiLAry oS'ectB df gravity, 211 

Hnmboldl, first view of 47 ToucanJ, 237 

Hiitney, period and maas of S Equulei, 
166, 169 ; elementa of the AyKtoni, 
177 r apectroscopic paralbix, 137> £83 

Hnygenn, obaervation of B Orionia, IGO ; 
(leliaeation of Orion nebula, 864 

Hy&dea, diaKction Into driiliit^ giDUpi, 
327; rsHcbud by a galactic aide-itream 

S37 
Bydfogen, abeorptiTe actioD to atarii 
36, 37, 38,,4Q, 41, 42, 43, 44, 47, 68 ; 

In tbfl sun, 44 ; cosmic, 37, S9, 47, 

03 ; rliytlipjicftl Jipectral series, 38-39 ; 
eini!»ioi] iu vftriable^j, &2-5a, 79, 102, 
104 ; in helium stars, 59, 60*01 ; in 
Wolf-Baypt etara, 63 ; in nebulaj, B6, 
67, 79 : in Novffi. 85, 67, 9i, 9&. 109 ; 
low refracUvity, 62 

Innes, invi^tigatioD of variable itan, 
104, 105. 120, 126; del«ctkio of 
quntemwy eyatems, 20^ 203, 3A0 ; 
variabiltty of a nebula, 279 ; a swiftly 
uioviQ^ ntar, 322 

Iron, absjirption in aun, 47 ; in Amtmiiin 
and earbou stnrt, 52, SG ; emleiiaii ta 
Him Cet), Ui 

Irre^ilar nuhalw, 243, 2!i7, 270 ; 
capriciouA romut, 2132-264, 2G9 

Jacob, mnrementa of 61 Cygnl, I6G; 
apurioua parallax of « HenvIWi 39S 



396 



THE SYSTEM OF THE STARS 



Jtooby, photop«]dUc oomptilnoi of iht 

PlolailM. 210 
Jummh, daoMa-sIit incUiod Is photo- 

JubtlMD, BUgDitUcl« of q GuiuB* lOA 

KtlMT, ooltTOZMiliBiig* of Anitnnu; 196 
Kjqjiajn, qiutitr of gnUctio stara^ S6 ; 
D«liulMlty rcund Nov% pBrwt^ 91 ; 
datMlJon of T«mbte sUn, 126 ; 
meridlui obaarriHoiaa foi «te!lu' 
pumlliuc, !Z98 ; pftnLllAZ KnA migiii- 
tude, 290 ; tnaitiDeiit of proper 
itiQlioaa. 806k XS3 ; JioUr Telocity, 
309 i ■*'■**""— aiwl proper motioiLS, 
^VK i2X ; k Awift lUr. 322 i tatan 

KajHT, waTfl4eDgtlu of bydrogwi-luies, 
S9 

Kaiwlai, dMorltiSm of the r]ekd«s,SlS 

KavIflTi pltolofn^jliic nebuUr sorrey,, 11 ; 
photcgnphjc Df uebiil»,. 31 : thetr 
eblc^ RjxictnLl linciik 6& ; Iholr cleotriiml 
glow, 70 : prevJent splnUity, 2M, 
844; thtir rAduU maveiueutfi, &7fi< ; 
j^wiagrmph of Hlnd'a nebutu, 277 

SkiIb^ Lord, trrartutivfl power of tbe 
Odfltttl world. 3fl& 

Hepler, iUj of 1604, SS, 84 ; diflUnca 
of tli« ■tui, S32, 3^5 

100, 209 ; CDDrorniAtJaQ, U€3-26f^ ; 

•Mming; ch&Q^ '27^ 
Klreh, TMlAblHty of x CyB^i ^^'-^ J 

obMrr&tiou of M 1 Iv 2*29 
Eirchbori Law, 61 
Eleln, clircinuitic period of a T7m 

H^otLa, 136 ; ahap« of «dcre*3 vorld, 

374 
Klinkerftiea, rtellaf tiriAbillty^ 111 
Knorre, ■colour-peiueptinn, 137 
Knott. vambilUy of U C>-giii, 183 
Konboly, ip«H3trat flactUAtloos of y 

CauiopoiK, 60 
K'tivosligethy, colonr-chuiifi or a Ume 

Majoris, 136 
KnlgBT, cmUIo^a of »d at&m. 13I» ; 

pRraUues of raf^dly - mOTicg start, 

n 

Qle, Kutliem DfibtiliB, 10; tnaguL^ 

tode of q CiTUUB. 306 ; of ct Centauri, 

1£3 ; at&r osUlogue, 817 
LiUodB, list of red slara, 13B ; notica 

of tho Draco planeUry, 217 ; itm 

obffflrrod by, 317 
Umbflrt, UUky Way, 331 
Ltwt, Irw of qoolinff for gBawiiJ bodies 77 
LongeDbkob, teiDpoTature* relatiotu of 

liaO'SpactrA, 70 



iMH^Djr intUtip>l« atb m trifid a^bak. 

207 
Laplrbos, tUrk atan. Iff A 
LMeell, cluster in GeniiBi, 228 ; obMCi^ 

tiou of Dflbobc, 2<fi, 349, 398 
l^a and Wirtz, ooltmr-cfeiuiee of s V^ 

Hvoiix, 1S6 
Least iq^tana, principlfl of, 3D5-S46 
LuvitS, HIis. detoetioii of nrUble lAm 

e7» sea, am 

Ejb Qea tfl, Lttfrndut to AadiOBiit 

nebula, 253 
Ijeowitz, apocryi^ial apparttioDa of Ibfa 

CassiopouB, 34 
[jeiris, nlatrra mjiiuM of binary iln^ 

I7fi 
Ugbt-9qutioo In itaUmr orUta, ISS^SSI 
Light-pTounn, effect on Nora IVn^ M 
Ligbt-rmlio deSned, 19 
liindianr, obflerr^tioD of Novft fTjHiopiin 

8i 
livuingADdDewu, l£biper«tarecif tlcetik 

spark, 76 
LockyoTt 8tr Nonnon^ bellTun ia Stan* 

36 ; ea1iHic«d liset, 41, 74 ; Aataria 

stan, &0 ; nebulai' ^lectram, 45; 

nltn-virilAt Rtallar spaetK*, 7S; 

temperatares of th« sfaus; 78 ; nd- 

ablca n^ardod u Hiuuiei, 111, 1^; 

conititatjioik of Andromeda. Dcbnla, 

260 ; coTOAtUT aflbiltua of ncbnlsf 

272 
Ijixrwj^ estimatad moniber of djiartad 

stars, Zm 
LoliBS, J. G,, ttiap of Btttw nui Koti 

Cygni. 88 
LobacOHelamBntsof SirUnsyBtem, 156; 

uiTesttgation of clufitcr in Pfli-Beai^ 2Si 
Loomii» period tttnbat«d to ^ Oaiinjt, 

107 
Looped uebala, apectmm, 69, 363 ; 

upect, 262-263 
Lodendorff, poriodicity of « AuHgKv lOB, 

jn 

LauL, spectrogTani tit a. Fbcexiieta, 1S4 

LynB,^ vjmible bright lines in apectram, 
G9, 60. 87 ; ligbtcbangis, a2, 116, 
122-123, 126 ; aaalc^us ol^Vfa. 123 

Lyrm,€, qundmpleflliu', 201*203 ; origiBf 
214 

M<:Clutt, distribution «f belium itan, 

36 ; oiyg^xi in stars, 87, 64 j apwlw- 

gr&pbic appamtDS, 109 
MaJler, disturbed ciwulaticn of 70 

0]ibiuebJ, lEi5 ; l)ypoth«tLi»l ptLtatUz, 

172 
Haaitlin, oli«ierTn.tion of Notr Serpvn- 

tKrli 84 : duceniincrit of Flaindea, 21S 
Ma^ljuiic douda, Wolf>RAyet stitn oot- 

l»Ct«d io, 64 t v&rUbU $tu% 119, 134, 



INDEX 



397 



Bystemic unity, 351-S52, 3C7 ; 
ab'itlM' connectLDna, ^63 ; couditi'Oiu 

neflium, ■ caoatitaect of iUra, 38, 10, 
[C2, 60 ; gnben. flutiog ol, 65 
Bitude, of atan, 2^3, 19, 22 

ougnituilfl, 218 ; rul brlUiiuiCT, 
F'S!21 ; apoctroacopic untallite, 222 ; 
■ttftchMl UflbnLe, 224, 225, 2&2 
HaIb, pwnUlH of €1 Cygni, 2SS ; wlu 

Mariotte, blae itara, 143 

MArtua, hrst telescopic obserfAtioii of 

AtnLrooitd* ncbalK, 2&7 
Marktirick, iukxuua of UItk Oti, 100 
MuioD onrl Smith, obaervAtioua at tlm 

trtfld nebula, 250 
MftS^'brightaesSt of biimTiea, 178 
Mvsaas,, of binary tUra, bov detericinAd, 
170 ; ns-aita obtakted, 171-172 ; rela- 
tive, 1712-176; cpnditioDBof ucertain- 
Baout for spectroscopic pairs, 190-lBl 
Maimder, galactic clui^tcrisg, S36 
HAarolfA^a, obserratiQiu of Nova Caasio* 

peiffi, 81 
Maarjr, Mi^, iJUdlvioo of tteUv «pH- 
trograois, 29 ; «volat)onu7 aeritit, 97 ; 
couipofiito spfctrs, 148 i Apdctniacopic 
duplicity of ^ Aurigsa, 189 
Mayer >, C.^ colours of double sthis^ 1*3 ; 
debectlDa of atolUf uOtiplM, 106, 106, 
1»5, 202 
May»r, Tobia*, di™ion of fCancri, 198 
Me^lowi. parallax of a Cectauri, 237 
Meridi&u photometer, priuciple of, 20 
M^Tope, magattnde, 213 ; ailJ&uKttt 

nebulic, 223, 224. 225, 226, 265 
Mcuisr, obaerratioiu of uebtilns, 10, 291, 

26B 
Moteofitic bypotkenifs U, 111, 184 
UiisUelL, rod etm, 136 ; Dumber of tb« 

Pkiivlm, 318 
Micrometer, described, 15 
Milky Wny, a collectiau of stir-gronpi, 
9, 340-341, 351 ; qunlJty of liglit, 26, 
347; conceDtratioD io, of bt>liuin st«TB, 
37, 347 ', of carbon stw, 60 ; of 
Wolf'Kayet aUrs, M ; of gaaeoQu 
nebalK, 65, 3fi6, 35S, 359; of t«m- 
poraiy atara, 63, 90 ; of sbort-period 
rarUblfiH, 133-134 ; of cliuUre, 228, 
231.3G&,9S8-3&fl; unrelated todoubie- 
Btor orbits, 177 : proloo^ed towaidi] 
gnat nebulffi, 2£i9, 359 ; Btmctiire, 
2€8,835>33S,373; inflneucsupoDHUn's 
coorwi 312, 313 ; iipou Atellar proffer 
tnotiotH, 331^33!^; aspwt, 333 ; folk- 
Ion, 334 ; track acrou tbo sky, 334> 
335; dolineations, 337 ; irresolvability, 
33d-33&; strAtuni-tbeory,3S9-34l ; ftat 
ikg theory, 343 ', ■pinl theoma. 



343-344, 343 ; cotdpQUad fortaatioiit 
345, 347 i distance, ZA$ i relation to 
QAbiilar (iisttibutioQ, 353 357 ; tttar- 
doiLjity in, 364, 365, 369-370 
MiucUlo, iaeaflurt<m«ut of HteUar beat, 73 
Mini Oti, bydrog¥U-li»u.4 bright io, £2, 
102 ; specLrograpliiu inreatiigaticai^ 
S3-54, ]l>2; a typiual i-artable, A2, 
99; pericidicity. ^9-102, 112; propar 
tcotLou, 133, iS3 
Hitohel, divisLan of y Scotptir 202 
Mitchell, Hpectrani of Sirius. 38 
MizftT {£" Ursn MiJ"'^^^ ^^^ leader of 

Alcor, 201, 213. 328 
Moack, masfi-brightnefla of liinoritid, 17& ; 

remotcue&s of whito stant, 299 
Mi::tTiti^uy, acint illation of Hie at&ix, 4, tV 
MiiU«r Aud Kempf, photometrio obaerra- 

tious 21 
Multiple starn, varietlea of oomliUiatioii, 
194-195. 197, 213 ; iaviMiblii comiioi]- 
enta, 197^^00 ; ^lenipecUTiu attaodantci, 
AOl ; i]UAttiniar7 ^oupA. 201-203; 
214 ; aebulouarelaljonsliipfl, ^04-207 ; 
«Tdutioii, 208-2i4 

Nobuln, early obserrnttoiia, 10, 267 ; 
number, 11, 2&4 ; usociatloii vritU 
bc'Iiitm'.ilarH, 36, 46 ; gaseooa tp«c- 
truiP. 66-67, 245, '^47, 256, 273; 
aobtiiiuoiiit, ^7, 70, 263, 281 ; strati' 

acatiou, eS ; etieeiftd, 69, 243 ; t«m- 
pcrfttuw, 79^ 2S0-2S1 ; fttrnotuw. 242, 
246, 'l&i, 2S1, 344; colour, 249, 
204 ; GomotaryjLffluitiea, 272-274, 276 ; 
Hoemiag immobility^ 275-270 ; rndLat 
%'elocitle«, 275, 3S8 ; variability in 
liRbt, 277-2S0 ; toauity, 280 ; llitra- 
gataclic atatua, 349, 362-^53, 36^ i 
atellar relations, 350, 352-353, 369 ; 
dlatribation, 353^367, 359 ; variety In 
comitltntiou, 357 

Xebulie ill the Pleiadtts, 223-220 ; fomii. 
274 ; drift through space, 275 

Ndbulium, a goaeoua conatitnetit of 
nBbul;e, 65. 08, 245 ; jpectnim, 66, 
67 ', amission by taBiporory start, 87, 
92 

Neboloaity, roaod Nova Perwif, 93*90, 
241 ; aUut the Fletadaa, 226 ; diffuaed, 
270-271 

NebiiloUHi stars, flnft notice of, 243 J 
aUinitl^ji, -244 

NewoU, apectram of Kara Oncminonuu, 
95 

N^wcomlj, total light of the atafs, 7 ; 
nature of tlie pebuloaity rouod Nova 
Paraei, 95 ; cotnploiity of Btvlliir 
ayattms, 199 ; proper motion of 
Alcyone. 219 ; invMligatjoa of aclar 
motioD, SOd ; ATeragtt st«llu- relooitlei, 



iS96 



TllE SYSTEM OF THE STARS 



mi, 334 : runaway it«Uv«pw*i 3S& ; 
■ur <lrift LQ UrM Mijor, 328 ; mcdid 
lint of lbs Ullk; W*;, 334 : ptnlkt 
of pUAcLle lUn, 844 ; t«if* 4f gilHtic 
nufaitndw, S47 ; itistribnttoB c€ tlin, 
3«A, 369 
N*wUMi'* Uw, nnlTerul ralltUty, 170 

Kiohohi, hMt of Um sCani, 7^-7-1 

Kotfel* obwrvALian ^<f it Ckrtiw. 100 ; rvd- 

MB of 'r CniotK, 1 40 
Mow AndmuttUbk oatbnr^t. 83, sei ; 

■pKCrum, 8ft, 00 ; colour, 142 ; tdtn- 

giatictic (Litoatioi), 3&0-S51 
Hon, AurtgiBt epectnl nlstiouibips* 02, 

91, lOlf I discovery, tiO ; tKbolu 

tnuuformation, dl-92 
Kora OunopaiKt 83-91 ; soloar-cluuiga, 

l«l 
Von OukUati, 90 

Kera Ooro&Bv ^^-Sft; oolobr-chaugih 141 
Korlt GTfBi, SO-!)S ; coMr-ehange, 142 
Kova Oeamomtn^ #5 ; colour* 143 
Jivrt, Potaeli luminoua Jlucluatiau^, 92- 

9^ ; ooaiiHt«il i)«buLo«ity, 03-06 ; 

ooloar-chuign, 142 

Olben, period of x Cypai. lOS 
Omega nebula, piacemaiU dUt^crunieiit, 
260 ; photogmpti, 270 ; fuajHUctod 
lamtBoiu eluiD^cMi, 279 
Ophiuchi, 70^ tiulour-chaogB, 147 ; per- 
turbed revolution, 156 ; inasiu anil 
lJght-pow«T of companieDtA, 17^, iJ'Sf, 
210 
Ortiibi of binary ttan, apparent, 1 &3- 165, 
167 ; actual, 163 ; number computed, 
l$7i 160 ;, methods of [jfltermjniag 
dimaiui'Oti^ ltj& ;. elvmenta, 170, tiote ; 
nlativa and nparata at Dompanauta, 
lTS-174; floceatricity, 177 
Orion nebula, inophotal contouw, 27 ; 
pliotographa, 28, 25^, 267; i|iettrm», 
66, <17, flS ; ftit«i-iinl niavamsiits, dfi, 
276 ; variable Ktwt in, 134, 268 ; 
"Fi»h-mouth" aipwt, 242; ^occn- 
loiit teittiire, 264 ; plnn ot ^trueton, 
2«5-26t] ; » d«velopin^ cltuti^r, 267, 
360 ; recesHion from tbe iim, 275 ; 
ateadridt lights 279 ; gtilactic coii- 
ufKUons, 337, 369 ; paucity of aur- 
roujidin^f stara, 352 
OtioD Stan, :36. See HaliQio-etara 
OrioDLB, e, Bpectrum, 36, 37, 40 
OrioDii, 0, early apectral type, 37, 6S ; 
eWtiufs bright Une«, 67-SJ3 ; multiple 
cbaraoter, 1£>0; relation to ann-ound- 
tog nebula, 205, 264 ; aucceasive 
datactioiL or conipoTieuta, 305'206 ; 
genetio AtBuitiu, 213 
Orionlfl, ir« lnultipt« star, 143, 203 
Orthochromatic pUtaa, 2S, 30, 50 



Oitkoff, ntdDMB of atara, 100 
OEford UmBomatry. 20 
Osygva ui tUra, 37, 47, M, M ; te I 
■tm, 42, 4» 

Fklm^r, ■pectTOBimni of Nora pRtili, I 
spectfrwcopic parsliaz of s 
187 ; binary cliarftctvT ot « 
192 ; compom-nts of Hvrcnka dailB, 
236. 239 ; Debalax apcctroiiv 381 

Paialluu uitLtul, mode of ^SetensiMtlak 
14. IS, 266-i«7 ; pliotiYnphi^ A 
S»a-S0O • of KoTK* SO, 04 ; of iwMk 
9tu*, 133 : apHfavoDoplc. 169, 197. 
190, 286 '28ft : hfpoCbeticaloffaiauiM, 
172 ; of the Platadea. S31 ; of anebak 
217 ; vmallnosa, 282 ; B«»oc)at«d tflarti^ 
S33-2S4 ; by llglit-*q nation in ftcDu 
orbita, 285 ; reiulls of maaMajma^tUt 
287-ZOa ; of ^actic atwa^ S44 

Parallax, sMular, SOS 

Pegajni, ^&, ralatire maaaea of cob- 
ponanta, 175 ; optical attendant, 201 

Pegaai, k, p«rio4, 160 ; ajMctsivoiae 
and Tiyoal system, 201 

PtfguL V, analogy vlth j9 Lym, 12S 

ppjTiter, Bcintillatioa of Sirius^ 5 

IVrrlDEi, Bpectrogram of Nora Aurtfv, 
92 ; rpcctmin and nebula o! S<m 
Perset, 93 

PaiaeuB, doufcio oloatcEr |q, SSl-SSSj 
aaeamed centre of »:un'« orbity 311 
galactic relationship, 358 

Prrsotml «quatioii, d«Qiied, 14 ; iu do*t)| 
star nieasnreiueDt, 16fi 

Peters C. A. F., rtdlftr parallw, 290, ! 

PbuaQidiv et, radial 'v«]wit]F, 186 

Phdio^raphlo pl&tai, 2S, 2d, 30, SO, 6B ; 
Ulaacopea. 24-26. aO-31 ; lBt«nuti«al 
cbart, 26, 863 ; irradiatton, 3& ; aUltar 
magnitudai, 26 ; c«talt9^M> S«% Mi, 
365 i Btar-gaugiitg, 363 

Photographs, of Dabola;!, 28, 81, 03-0 
224-226, 251, 25S, 256, 25»-260» ! " 
266, 267, 289, 270-271, 377; _ 
Btellar spectra, 39, 40. 41, 42, 43, 44, 
45, 52, 53, 54, 60, 6S, 136 ; of Debular 
a)vec<tra, 67, 70. 260 : of double atais 
165-167; of UiB Fleladaa, 319. 2^4- 
2£6 ; at varioua duHtort, 238, 230. 
sae, 236, SI I; of nebulons BtV7, 
241, 244 ; in the MUky Way, S37, 
347 

Plintograpljiy, etnploymeDt la lideiBal ra- 
March, 22-26, 31-33 ; photometric, 
26-27 1 apeetrofloopjc, 2S-20 ; radial 
motioa dete^m!nl^d by, 29, 1S6, 323^ 
discoFerieaby, 29, S3, 52,90.92. OS-!' 
97. 236, 266 ; rt«llAr parultu by, 
293 ^2»e; rapidity and treraatilit; 
32 ; investigation by, of nobalo 



INDEX 



399 



ntatiDnibips, 207, 234, 242, S&0-2fil, 

, 254, 308^67 

' olaiueter, poUrigiDg, SO i vedge, 20 

hotoinetry, stellar, liin and loape, 19 ; 

; metbotii. 20 ; resultA, 21-22 ; photo* 
graphiK. 2;5-27 ; nebular^ 27 

>ipili«f«, AID «sKntuU f«»tare al nm- 

riik« bodieA. 34 

i, fltar-cBtiilogui.% 317 
kerin^, E. C, Harrard photoiratryf 
I ; mtXaO^phtFK |.bsorpt.iDn, SI : atiu'i 
iialtaj- miigiiLtiuk, 22 ; iiLd<2r«Al photo- 
grat'ti j» 24 ; dintrihatian of the BtarH, 
25 ; in the Milky Way, 3fl9.370 ; atellar 
'fpectrogfftiliB, 3ft, C3 ; discovery of 
coflmic hydrogao, 37, 30, 66 i sweeps 
for g4««oqH itara, 64, St, 246 ; clouiii- 
utiOD of variflhJea, 82, 10&, 116. 126 ; 
datflc Lion of «p«ctTD«copic binnrifs, 
12S, 189 ■■, invutigatioa of Algol's 
eeiipMfi^ 1147, 12S; colours of 96 
Hercidis, 147 ; ouua - brightnaM of 
binarleu, 17B, 201 ; tlie lost Fleiid, 
218; paucity of tnia Plaiados, 220; 
their spectro, 223 ; countg of fltara 
la globular c1uf.ter«, 234 ; utatns of 
Looped DcbQls^ 263 ; photognphiQ 
catalogae, SAi, U6 

Pitlttriagt W, H., netralar photomatry, 
27 ; photoprapliic dSEtiloRurv of gTBit 
Biuke ucbulh £67 

PEctor, flying Bt&f in. 322 

Pigott, discovery of R CorOluB, 109 

PLaticki Stefan's law of ndlati^iD, 72 

Planetary nebulK, apectraui, 6&, 6& ; 
Btraiiture, 66^ 246, S&d, 274 ; dimiff- 
iut«d by HOTBchel, 24& ; Bpeciiuens, 
217 ■ 2G0 ; imwn^ible parAllax^ 247, 
aOO; ndial velouitie^, 275 

FlasaniantiH, tidal tlieory of Rtellu: varia- 
bility, 111 ; oecondar)' naininiBm of ^ 
Tanri, 131 ; iTregalaritiea of U Cephei, 
132 ; galactic isqairiea, 345 

Fifliadled, mythological foniu, 215-217 ; 
DUiQbtir visible, 218 ; cotnmou nDOVQ- 
ment, Sl£^ 220 ; ficale of ny^tem, 221- 
222; nebuloufi coaditian, 223-220, 
8fiO ; a typical invgtiLar claiter, 227, 
247 ; forma of induded nebulte, 274 \. 
connwatjon with the Milky Way, 337 

Fleione, bright-lme spectrum, 69 ; the 
but Pleiad, 213 

Plamnifr, eiitimates of star-liiglit, 7 ; 
ohMirvaCioua of Nova CaMiopetie, 81 

PogHon, DbftOTTattoQ of Nova Bcorpil, 88 ; 
periodicity of It Cephei, 10&, ItO 

Poincar^, QgureH of rotating caaaiiefl, 209 

Fol«-«tar, the " cyDOflur*," 2 ; niagnitiide, 
B, 22 i ataiidard brtghlben^ 20 ; spec- 
tnim, 41 ; tberratioui^ diftplacemanta, 
383,284 



Pond. Btellar paraUai, 28S, 294 ; wcolor 
abamsLtioa, 311 

FosltloD -angle of coap]e<l etam, IBS 

PhEsepe, meaaarea of componvtiL atar^ 
233 

Prey, nlatiTe maAaei of biaary stara, 175 

Pritchard, photometTic inrestigaltoiw^ 
20 ^ sballai' parallax by photography, 
30, 298 ; surFcy of the PleiadcB, 220 

Proctor, Htar-drift, 327-328 ; theory of 
Milky Way, 343 ; siellnr o'utbuTBts in 
uebule, 350 ; nebnlnr diatribnUoUt 
353. 354 ; star-alreoni^ 372 

Procyon, magnlLude, 22; ^peutral type, 
41, 293; iatellite, lf^7 ; liiminoitity, 
158, 175; ij*rallaa:, 297; propee 
mol^gii, 322 

P^per motioiui, defined, 14, 316 ; of Tori- 
«bka, 133 ; of double ^tars, 164, 1C8, 
160, 174, 322; diiiturl>t^l, 1&6, 157, 
1&8 ; of the FleiadM, 219. 220 ; a 
criterion of ricinity, 290-291, 317-318; 
common perspective elemtni, 304, 
SOS, 322-323 ; non-tn^tUodical cbar- 
actter, 305, 314-315; aacertainnient, 
307. 308, ai6-3l7; aiiomalie% 318- 
820 ; TsrifctioD with •pectnU type, 
321 -, Bwifieift kaowD, 322 ; unwu- 
troUabla rates, B24-327 ; concerted, 
327-330; snraii&ed harmoniM, 331- 
332 

Ptolemy, Htar-inagnitiidtiiB, 2 ; ncbttlofi«, 
10 ; ignor^ tf CariniE, 100 ; ni&gnltuda 

of Eridaal, 180 5 Pleiad quaitatta, 

217 
Fnppis, U, variable doable star, t81 
Pappis, V, Boppoaeil eclipses, 123 
Puppia, ^. ooscnio hydrogen - litiH in 

ipectrum, 38 

EUdan, form of fdilereal uniTerae, 372 
I^dial motion, of dtara, apetitroBcopleally 
dBtemiiued, IB, 185-186 ; variabloj 
66, 121, 128, 1S7. 182 ; of Novte. 91, 
93 ; in binary orbits, 285 ; indicative 
of Bolar translation, 310 ; of nebula^ 
275. 358 
Radio-activity, coarnical effocta^ 80 
Kaocken, galactic rotation of itan, 9^% 
Ramyatil, itructufea in Odoo nebnla, 265} 
266 ; tenuity of nehulic, 2S0 ; com- 
poflittau of galactic duaters, 347 
Red Btars, jpeotral vartfttwa, 49 ; itropg 
Mir-ab«^tIon, GO, 62, 57, 140; 
catiJt^iiss, 55, 138, 13d ; variability 
in light, no, 140 ; in colpnr, Ul ; 
eiampleg, 140 ; dUtri button, 142 j 
ocoumncQ in pain, ICO ; in Persvas 
oluBtor, 2S2 ; in (galactic grqupinga^ 
S71 
Keflecting telescopes, adTaotagM and 



400 



THE SYSTEM OF THE STARS 



•oloanilM 
Bmla^ Bi^Blltiik, 3 ; aatoUtt*, 146. 

1S8 
ftMplcH prfaHtk dbeto of MtptiUft. 

tto*. « ; tfiMtnuB of T V«tormiii. 63 
JttoaML dkvMab of f Un» BUJarim IM 
Btpb*Mt Am ftbaomtlm cf « Owttjori, 

150 

tut, %7 ; t|H)otnuD ■ai^Ml by inw' 
■fanrpliA, 47 ; onitit gf ultn-riokt 
ndlBtMCM, 7& : bltta atlMdnt, 14£, 
lU; pu*Uu.2»7 
Blag iMbalA, iiMctnin, 47 ; vfini con 
ronuUom, Sfi^ fi61. *JI^A ; juiilc^oos 
otijcato, 2&1-3A^ 256 ; kuhiwcUU move 

Bltchey, twbtilA photogTupbiS, 31, 380 
aipuLUOD of Harm Pwsei ticbdUi, 94 

BoberU. A. W., itclUr e«U[)Mlv ISS 
ISA ; iD«ui deiiiitx of AJgo) ttan, 
12S ; ctntrv of grtvUf of b CstiUtui. 
174 ; periodicity of x PiToiits, l^H 

Eoberbk Unftc, mulUpl« €ip««ores, 33 \. 
photognipbJ of bebulXt 31^ 253, 25S, 
2C0, 2«1, 262, 205. 266, 270 ; of tfae 
PlaiulB^ 224 ; of nrioiu cliuten, 
SSO, 232, 23S, 241 ; of a nbaLoiu 
■tv, 244 ; of ilifftued DAfaoloutia^ 
S« 1 ; arTVL^ement of «tui to. eliutari, 
229, 231 ; v»TiaU« gtus Jn Onon 
nebala, 269 

Bijwu. tidrd Ear]^ colour of ■. nUr, 141 ; 
■tneton of HUr-cltuten, 23B» ^0, 
SS5, aaS ; Anm of Sahun netiid*, 250 

Bo«Bv fonitb Etfl, double ubuU, 254 

Roaidt H. C.t ta«Bfli«ia«bt of eliutir 
&boat s Cracis, 232 ; cbimgo k K«y- 
bolfl bobula, ^9 ; pboto^nph of tfa« 
OoKl'Uck, 387 ; of Atin in Crnx, S4T 

BwkII, H. N., tcDait; of Algol TorubltjB, 
133 

Rnlberfonlt pbotogrspbt of the FleUdeo, 
219 ; of pTBKpe, 233 

K^dberg, principaJ sehu of bf drogea In 

8ul]or, double sttelUte of RI^I, 195 
Hftfjuiki identity of Nora Cassiopaie, 

S4 ; Unctiutiona of rtd stan, 110; 

micettaiRtied £□ oolour-pctveptioll, 137 
SfttTor^, satellite of Siriaa, 1G6 ; tvxton 

of OrioD iiebulau 2fl5 
Bfttara uebuln, colonr, 249 ; mibr, 2&0 ; 

r&dloJ velocity, 275 
Bavary, Gnt cooipntatioa of a itellar 

orbit, 169 ; ponllkiu of biiuunr stuv^ 

28& 
Sftwyarr miBimildi of R Corons, 110 
Sdbtebarle, diAcovery of Pnxfon'j ooni- 



puioii, IS7 : halioLl B*b4b M; 

pfaoUigrsplii of Ring tnl Thimli Ml 

tiebul«, 251, 256 
Bcbcibcr, apvctnuB flf CkpilH 44; 

rmtioaale of fan^rt-ISBB atautK ^mut. 

61; >p«aibqgr^of AAdfaiBabMkilfc 

70, t«0 
Schupftralli, dapUcity of « Bj4i^ 117; 

Mercpft setmbL. 224 
Scl^*U*>ru[i, c»ulGgrtt* of ml ttm, A 

13d, 140 : obaemtldfi o€ * ealio> 

Tkriftblt, lil 
SoMtBTiw, pufelUxaB of fiitet i4m^ 

Schnidl, dteonxT of Ndn Cygni 16; 
nduMa uul periods of Tmn«iile ote^ 
314 ; minituum of S C^ocri, IM; 
ootoar of AlffoU 136 ; hittory ctf T 
Vlrglid^ Ifti ; &*bQlar nrubOity, S79 

Sn^DfiMil, teixad of T Pnriw. M; 
oolottr of K Obphn. 110 ; obHrvatlffa 
of 3 C^Mri, 130; nxiafala aatofe 
278 ; eoDBioii drift of ■ ttn-pm^ 
3S9 i pTDpar QotiiKu rrftgw i to 4h 
Milky Way, 330 

Schal«r, botke of Tycbo's stw, 84 

Scbiir, flxJSy of itan in PnBU|Mv Stt; 
paimllmx of ■ comp&iiKiB'ftu'* SSO 

Sebofter, dispfnira-bypoibwa of 
bri^t Unw, 61 

Scbnb, pnriodidty of Mira^ 101 

Scintilljition, of white iwd ted «(««» 
of None, 4, 84 

Sdjitillonifrtt'r, 4 

Scorpii^ V, m qiut^maiy ajsloB, 203 

Skorpii, {, a tr{pl« aysleB. IVT ; ooa* 
plu iHRliMUsm, 21 S 

Swle, oebulooitj abouX AJteytti^ £B4 

8«o«hl, objMtlivB prion, 26 ; culUr 
typM» 35, 36, 49 ; cwbon fttn, £4, 
65, 67 ; gawooa itu«, 69 ; auu- 
colomv. 141, 143, 147, 146 ; niniity 
In H closter, 229 i raotralidUty of • 
Debali^ 24V ; TwtecB of intentdUi 
ipuo, 301 ; Kilaotio doadlat, 3SS ; 
fttsr-pittenia, 371 

Sea, uii^ient redn«M of Siriiu^ 13& ; dti' 
tarbuce of 70 Optuncfal, I££ ; pvth 
of 17 Cusiopcile, 1&6 ; comput^iazi of 
rtaUar orbJU, 163, 167, 197 ; oeoafi- 
triciticff, 177 ; qaiatuple stv, 195 ; 
tiddl fricdoti ib fit«]liir syiteou^ 206, 
209, 213 

Sediger, oebtLbi round Ifork Pond^ 94 ; 
lytfbem of f" Caocrit 3O0 ; gidaotk 
reHUCha, 345, 365 

Senoca, colour of Slrins^ 136 

Sideival aitronomy, canditioii« for pro i> » 
cutioQ, 13, 14, 81 ; owtbodi* 13-14 ; 
ftTiLil&bility of tbe camen, 22-23, 26, 
Sl-33 ; muliabla ot lifbt, S3 







INDEX 



401 



Sidereal phy&iu, origia, 14 ; takan pro- 
gress 1& 

SiderenJ ny&tem, graviUtiTi power, 325, 
325 ; periBUieiice, 330, 331 ; spectral 
relatiotuhipB, 332, 370 ; tioite dimcm- 
Biono, 361-362; org&niHation, 363- 
369, 371 ; comprebeiiii'ive ^^f stats and 
iicbuUt, 373 ; primitive rotation, 371 

SidgresTes, fipectrum of Nota ikurigK<, 
91 - of Xava rorui, 93 

SUicDo, in stars, 37, 41| &i ; in the iiui, 

Ftiri&n JitarB, 35, 40 ; nature of Hp^ctra, 
S7, 47t 48 ; pbotograpliic iuiuftiga- 
tion. 36 ; brtlTogeu -absorption is, 33, 
41 ; nffinitieA, 46 ; Absence at emu&ive 
fiymptoiQH, £8 ; tlfht-vaTiaLiork.s 117, 
133 ; tnteQa^lmiiliiDaity, 17S ; ivninte- 
DUft, 29&» 320 ; aioaU pr&per motjopp. 
831 ; guJActii: cQbdetiaatiaii;, 347, 370 

Sirtua, bright&«M, 3, ^2 ; twinkling, 3, 
6 ; jpectrDgrami, 33, 4Q ; hydrogen 
Ebaarptioo, 41 ; ultra-violet spectrum, 
75 ; aaciant TOflnaBS, 135 ; binary 
iyst«in, 156-167. 16$ > ma«» rolationt^, 
17fi ; p&rallax. 293 \ pmpcr motion, 
322 

Smyth, Admiral, dlNcriinliiatloii of atar 

eoloun, 137 ; UnU of y lfOoui» and 
116, 147 ; Qlnator in 



95 UercnliA, 
QoniiDL, 22S 

Smyth. Piazzi, 
HercuLin, 147 

Solar lUra, 3i 



colODT'cliaDge of 9& 



relativQ Qiimbara, 41 ; 
spactrm, 42-45, 48; relation shiju, 46, 
!^^ fifl ; brIgLt liiiBa nliaent from, 
68; l)gbt*varifttJon.s 117 ; comparative 
viciiitty, 299, 320 ; diiUclied I'rom 
galactk fonuations, 332, S47, 370 
^onth, colours of doublet Btan, 144) 147 ; 
observation of ^ nancri, 198 

^^IKutra. nehiilur, G'j-JQ, 79, 24fi, 247, 

253, 256, '2m, 264, 281 
Spwtn, stellar, typ«g) 36 ; of luelium- 

ftan, 36-37 ; of Sirlim stars, 37-41 ; 

t>{ solar lUrs, 41-45 i fluted, 49-QQ i 

of Antarian jtirt, 60-64 ; ot carbon- 
(itars, Dl-r;? : of goMoys staiw, &S-61 ; 
of Wolf-Eayet stars, 62-6S ; of loug- 
period yariablwi^ 52, 79, 102, U3 ; of 
■bort- period variableB, 116; of tero- 
pomry atan, 62 ; a]t«ratiotiB In, 60 i 
niatiObB to tomperature, 74-77, 78- 
70; at double atnrfi, 14S, 161, 17$, 
ISO, 181, 183, 210, 211 ; of the 
Pleiades, 223 ; of elustera, 241 ; eon. 
nection with proper niotioa, SOS 
Sp«ctrogTkphJc methods, 2S - 29, 3S *, 
application to stara, 38, 53-54, &&-56 ; 
to nsbulc, €6-67, 70, 260 ; to Nmm, 
91 



Bpectroccopic bioarias, diBcoYBries, 80, 

105, lis, 121, 128, 131, 139-1&2, 
J97» 203, 206-207. 222 ; relations to 
telcMcopIc binariei, 168, 209 ; condi- 
tions of movement, 187-188; abunilauL 
oocurreccft, I8S-1S9 ; dims detennina- 
tinna^ Idl -, nebulous, 207, 270 

SyBctruiu analyBi?, principlia, 16 ; 
method^ 17^1S ', Applied to tho stan, 
35 

Spnacflr, distribittion of neWlie, 351 

Spies, a »p«Gtroacopk bjniry, 190 ; nuuu 
of aysteiD, 191 ; panUlox, 297 

t^jiirol DebuliB, ilbcovetod at Fatsoiu- 
tovrn, 252; two-brwicbed, 253 ; fore- 
ahortened iuto '^'«piQdJ4»," HSi ; 
copious oeciUTcnce, 344 

Star-dria, 327-328 ; cbankcteriseH partial 
aysteniH, 323-330 

^tan, Dumber viHiblor 1 *, de^giiatloD^, 
1-2; magiiituil«L5, 2-3; aointiUaUon. 
3-6 ; t«lsuopio multitude, 6, 362 ; 
toUl light, 7-S, 361 ; individaal rtndy» 
S, 16; comViiiutioDB, 9; catblogU««, 
IS, 317^ 364 ; detemiLnalioii ofplaoBS, 
13; EuoTement^ 14, 303-304, 305, 
308, 314, 3I&*332; paraUojtes, 14, 
15, 282-360; faintuoM of ligbt, 17, 
33; aim -like bodias, 34; spectral 
typM. 3&-&7 ; gMWJUH, 68-65 ; 
cbfloiifltiy, 37. 40-41. 47, Sl-Sa, 55 ; 
tejapsratiiraH, 71-80 ; tclipses, 126- 
133; colour-thangw, 135-136, 141- 

142, 146-148 ; naturo and variety of 
tinU, 136-149; pliotognplifo chut^ 
Saa, 365 ; distribution, 363-365, 36»- 
370 ; belt-foTmatioa, 366-368 ; coUec- 
llon iuto riags and ttreoms, 371-378 

Ht«bl>iii8, KpcctTitm of Mira, 63, 54 ; 

llRht-cbange. 100 
Stefan'g law of cooling, 71, 73 
Stellar bftbuliB, nionoL:liromAtic ligbt, 

64 ; collated in Milky Way, 66 j «- 

moteDess, 66, ^46 ; imdlal T«loctty, 

S76 
Btokea, 3lr Oeorge, temperdture-mUtioui 

of Une-Bpectra, 78 
StoDe, mea8iiruni6ut of ^lUr hfiat, 72 ; 

CBotra of ^rarity of a Cestsuri, 174 
StrntoQoff, coastitntJOD of tba Pleinden, 

220 ; pbotograpliH of PLciadeR nehnliR, 

225 ; dliitribatlcJii of uebuln. 354 
Strove, F. O. W., observatiou of 4 ired 

itar^ 141 ; colours of iioiib|« atart, 

143, 144, 147, 148, 182 ; Msigued 
limit of Apparent diatancc, 151, Ib^ ; 
detections, 164, 181, 197: otwrrva^ 
tionSf 193, A'iB ; colour- ptroaption, 
146; mognitudea of 44 BoiltiS;, 179; 
fifth star in trnpezfum, 205 ; ntcllar 
paraUai, 284 ; criterU of wtflUar 




jMlillf»nfatirMl.»M 
•irlfU M bd— I •!«, 344; Mhn UM 
Mbote, 347 i Mr*ttM mM% 3M ; 
iUlUr ud mMw «alM,9U; d*- 
UU la Omtm wnkmh, m 



TufI, X, w ■■llpiiM biwr, 1>0-1» 
T«libatt» Ckdlag of ■ nd «tar. 141 1 

wliHU dovbto ftar, IM 
Ttlawppw, ■^^towd ittOMrt l at, IS,3S; 

Tmp*!, dUm In AadraiB*!* lub^ 89 ; 

dtaeavvy of lUvopa babuta, 223 
T«ni|i«ralan, of tlia no, 71 | ^ «Uri. 

7S-U, 77-7*. flO; of »bol». 79. 

3M-U1 
TmporHT itaim «i|fa ol apaBin, 03, 

SC M ; faullridtua ontbtuiti^ »S-«5 ; 

dlftrLbtiUoR, M-BO; aui of thoM 

utlu'^ 184 ; wttour-chftogw, Ul*143 
Ttaulty. uf mILimIiix «l«n^ ISA ; of 

lUtla^ ■tmcUri of flatod ipta^ IM); 

poled of Outor, 147 
Thoma, Cordotw Durchtauttcrniu, t 

of fUlUr gyiUma, 306-200 
'i'liHruict Hatoxy of AIrqI'm laaqtullUM, 

13B. lau 

TlUnliiiiu afadiorptloa-lliiM In t^lur m^ 

mU\\n.x HpvotrE. iA» (3, &£ i HutingB in 
JuiUrlu itAn, Al 




▼■Udt mM^ I iiMilii. X77-: 

y«i>life ita>^ ndMi^ 6S, 1 U ; dSi^lir 
df Uriffcl Ham ■• Bsiin^ fil. U^ 

lit : aivd ipvclnl typtb M ; 

vuiteBMi^ TVs illiHilBia I 
«l ; Av« «Imv eS : 

psiodii, ftMOS; bngvlw, 10&-U0: 
•r •kort tMrfosU 07, lltf-iSd ; <&sixi- 
fatillgni< pviodfl, 96s H' ; cxptuMatr 
trpotht— , 110-115; «eliiiiM^ Ifd- 
iMi mllKtod ID cnMpa, IM: 
MapUd. IM. 17a-184 
V«^ iippwBl brUbkH)}; 3, 23; kf- 
dn|;ai-Hri« In cp«tnm, 81^ it; 
fcR&pentare, 74 ; ii]t»>Tioi«t taSxk- 
Uoiu, 77 
Vfllonuii, R', nittqre of «djp««, iSIt, 181 
VilonzM. &, frclitMtt tmdergoae by, ISS 
VelonuD, r, bdinif of coloor, 141 
Yfltoram, 7, cnuadoo-spectram^ fi8» 07 • 

oxfgieD^bHrptioD, 01 
Tsry, ncbaloiitj roand Norm Psvd, 9£ 
ViJluKwui, pianJlAX of binarieiH 2S5 
Viigiiii», -y. duplicity, 1*0; o^\, 17*- 
174; reltUve muaos, 17S; Inmiiioii^, 
176; vn-UhLlily, 178; jpectrnjo. !?•; 
flpectroHcapic p&rallAX, 26S 

of mlial velticity, 29, U6, 833 i 
dwortiuiDBtioii ot belium-MlOs, Sd ; 
iDor«menti! within Oiioii iwbiili, 60, 



INDEX 



403 



I STS ; obuTvation of Kova Corosa, 86 ; 
I spectroscopic binnrit:!, 105, 190, 102 ; 

ecljp!«s of Algnl, 117). 12S ; spectra nf 
, 06 Herculip, 117 ; iiiveeligEition* <jf 
, elaalers, 232, 341 ; peculiruilies of 

plinetary nebtilm, 246. 249. 256 ; 

mi&bla nebula, 278 
liogvl, H. W., spBctrum of bydro^eiii 

87 

■Vatrd, Nova Androinc^ia:, 89 
Waters^ rliiitributlDii ^f nobuln!, 351 
j^^Wftbb, colour of U Cygni, 183 ; arrange- 
^K meDL of ctiutertiig BtaiK, 237, 233 ; 
^^P pljinetaxy nebula, 3Q0 ; st&T-pHttertLn, 
I 871 

W«ber, twinkling t>r tUn, 6 

WMnmA, mt^vAJuflots of the Hyiidi^ 3Q7 

I We(lQ7, pbotogTftphi of the PUijuiei, 

225 ; diti^^iQB of Andromeda and 

OHoD uebulrti, 25S, 259, 26S 

WliHwell, interBtflUar aitnatjan of nebulKi, 

361 
Whirlpool nebulft, typical of spiral clua, 
243, 25i ; 'ItfliKtinb at bratichcs, 2^5, 
250 ; repiiLiive actioTi l>etokoiieil, 374 
WiedeuiiDii, liiroinesecnM, 77 
Wiea'a law of spectral energy. 79 
WlUiamfi, investigatioiiB o( variable dtara, 

120 
WllslDiff, stellar Mpectrograma, 36 ; phajie* 
curve of Algol, 121^ ; perlurballDTif of 
61 Cjgni, IfiS 
Wilson, nebulosity round Nova Perseij 
04; photographof nn irregul&r nebnla, 
283 ; repulsive affucts in Whiripool 
nebula, '274 
Winloek, double wtellite of itepdiw, 16& 



WinnKlce, light-change of S Monocerotif^ 
ISl ; m^aBUrumeut of PneMpe, 333 ; 
Tariability uf a nebula^ 276 ; parallax 
nf a faint star, 290 

Wolf, C.t catalogue of tha Pleiades. 318, 
220 : uebuloua condition, 221 ; map 
of itars in Preaepe, 233 

WolC, Idax, diacovBriea of ucbulsi, 11 ; 
observation of Andromeda nebula, 
89; nebalcwity rmmil Nova Peraai, 
93; photf>gTaphB of D«bTile, 253, 259, 
270, 271, 362-353 ; "inakB" nebula, 
267 ; replac«inent of ntan by nebuLc, 
352-363 I nebulflT survey, Sfi4 

Wolf, B., period *ttributfld to t) Carina, 
107 ; solar auil fitdlftr p«riod)oity, 113 

Wolf-RAyet Btan, emiMloa Ihuuh, 56, 
62-fiS; amuitip.'', 56, 64, M, 9fi ; 
Bum'tier and dintribution, 64 ; Id- 
Beiiai>!le p[iJikllax«.H, 300 

WoodB, detection of a variable atar, 136 

Wright, lliomas, disc-lh&ory of MUkj 
Way, 359 

WriKbt, W. H^ ipectrograniR of Mir*. 
53 ; apectrosoopic panUlax of a Caa- 
tauri, S6€ 

Ye&[d(!ll, colour uid period* of vftrtLblee, 

114 ; ccUpsea of U Cepbei, 131 

Yoonp, oolcium-vaponr iu aun, 42, 4S ; 
diagniQ of auD'a path in spuiQ, 812 

Zaeh, red stsrs, 138 

Zeemati ttttct impercepUhlit id Dnted 

K].)ectra, 49 
Zodiac, tiigns of the, 2 
Zollner, polamlng photome'ter, 2] 



THB WXTD 



PrtHUtt i^ it. & k. U.AM]t, Umitsd, £diMimtg/L 



^': 



t 



JAN 15 1941