August 8, 1889J 



NATURE 



345 



shining branches are nearly on a par with the dark rift 

 separating them as regards the distribution of stars even 

 fainter than the ninth magnitude. The nebulous effect 

 to the eye distinguishing the branches is, then, presum- 

 ably due to more remote collections. As to the further 

 limits of these, we know as yet nothing, except that 

 Herschel's gauge-numbers left it to be inferred that 

 "thinning-out" had become marked before the attain- 

 ment of fourteenth magnitude distance. On these, and 

 similar subjects, enlightenment may be hoped for through 

 the judicious use of means already at hand. 



For simple star-counts, we have only to substitute 

 star-counts by magnitudes over selected areas of the sky.^ 

 The relative numbers of the photometric ranks can hardly 

 fail to give highly valuable indications as to real distri- 

 bution ; provided only that the assumption of a general 

 uniformity in the brightness of the stars be valid. Not, 

 it need scarcely be said, of a uniformity such as to pre- 

 clude any extent of individual variety ; all that need be 

 supposed is, that the average size of a star remains 

 constant throughout sidereal space. This hypothesis has 

 far more probability in its favour than any other which 

 could be set up instead of it ; though it may receive 

 corrections as our inquiries advance. 



The photometric classification of small stars is one of 

 the many branches of sidereal science which will hence- 

 forth be prosecuted only with the assistance of the camera. 

 Visual methods are inadequate and insecure. Those by 

 photography, it is true, have also their difficulties, not yet 

 completely vanquished : they will, however, evidently 

 prove manageable. Prof. Pickering is tentatively esta- 

 blishing methods in photographic photometry which will 

 doubtless before long be brought to perfection. They 

 depend mainly upon comparisons of stellar impressions 

 upon any given plate, exposed under known conditions, 

 with standard impressions of standard stars obtained 

 with varied exposures or apertures. For the purpose we 

 have in view, accidental errors of estimation, even if very 

 large in amount, are of no importance. What is essential 

 is, that the integrity of the series should be preserved — 

 that the proportionate change of light from one magnitude 

 to the next should remain invariable from the first term 

 to the last. The realization of this aim, now virtually 

 attained, is one of the most weighty services rendered to 

 astronomy by the sensitive plate. 



We may nov/ describe the process of photographic 

 star-gauging. It consists in the enumeration, by mag- 

 nitudes or half-magnitudes, of the stars down, say, to the 

 fifteenth magnitude, self-pictured from distinctively situ- 

 ated patches of the sky. Each such area should be wide 

 enough to insure the elimination of minor irregularities in 

 distribution ; but a single large field would often suffice to 

 show the characteristic grouping of the smaller telescopic 

 stars. 



The Milky Way would naturally be the first subject of 

 inquiry ; and the comparison of several plates taken in 

 different sections of its course might be expected to yield 

 data of great significance as regards its constitution. 

 From simply calling over the muster-roll by orders of 

 brightness of the stars contained in them, answers may 

 be derived to the following questions : — 



(i) How far does the regular sequence of increasing 

 numbers extend ? That is, down to what grade of bright- 

 ness do the stars continue nearly to quadruple with each 

 additional magnitude ? 



(2) Is the progression interrupted by defect or excess, 

 or by each alternately.'' In other words, does the stellar 

 system embrace systematic vacancies, as well as system- 

 atic groupings ? 



(3) Supposing an accumulation of stars to set in at a 



This plan was first suggested by Prof. Holden in 1883, as a mode of 

 investigating the composition of star-groupings (" Washburn Publications," 

 vol. ii. p. 113). Counts with varied telescopic apertures gave him the num- 

 bers in the successive photometric ranks. We believe that a photographic 

 method of determining them has since been adopted by him. 



definite stage of space- penetration, where does it stop ? 

 Down to what magnitude is the augmented ratio of 

 increase maintained ^ 



(4) Are there symptoms of approaching total exhaustion 

 of the stellar supplies beyond ? 



These should be found in a concurrent decrease of 

 density with brightness, " density " being understood as 

 the proportion of the numbers present to the space 

 theoretically available for stars of a given magnitude. 

 For one of two things seems certain : either the thin- 

 ning fringe of stars is composed of really small ob- 

 jects interspersed among larger ones ; or of average stars 

 at average distances from us, but further and further apart 

 from each other. In the first case, the system ends ab- 

 ruptly ; in the second, it is, as it were, shielded by outliers 

 from the absolute void. 



Particular attention should be paid to the differences of 

 stellar distribution upon plates of the Milky Way proper, 

 and of the dark aperture between its cloven portionst 

 That this really forms an integral part of the galaxy is 

 shown by the far greater profusion of small stars there 

 than in the general sky at the outer margins of the 

 galactic branches — a fact in itself fatal to the '' spiral 

 theory," by which the rift was interpreted as a chink of 

 ordinary sky-background left by the interlacing, to the 

 eye, of two great streams of stars, one indefinitely more 

 remote than the other. From photographs we may now 

 hope to learn what is the nature of the distinction between 

 rift and branches — what are the magnitudes, relative 

 numbers, and presumable mean distances, of the cluster- 

 ing stars present in the latter, but absent from the former. 



Gauges taken in the neighbourhood of the southern 

 "coal-sack '■ ought to prove instructive as to the nature 

 of the nebulous stratum out of which it seems as if 

 scooped. If the Milky Way be there shallower than 

 elsewhere, a greater uniformity of lustre may be looked 

 for among the stars composing it. No background pro- 

 fusely stored with lessening ranks will come into view, 

 and stars below the average of those grouped in bright 

 masses, representing their genuine companions, will be 

 but scantily present. 



Outside the Milky Way, two points suggest themselves 

 as likely to be settled by photographic gauges. Arge- 

 lander found that the faintest stars in the Durch>nusterung 

 were everywhere in excess of their due proportion.^ Even 

 at the galactic pole, their increase, as compared with the 

 class next below, was sextuple instead of quadruple ; in 

 the undivided galactic stream it was 9J, in the rift 8| 

 times. If this semblance of crowding in all directions 

 at about the mean distance of a ninth magnitude star be 

 no accident of enumeration, then the Milky Way is only 

 the enhancement of a phenomenon universally present, 

 and the fundamental plan of the sidereal system must be 

 regarded as that of a sphere with superficial condensa- 

 tion intensified in an equatorial ring. The counts, to 

 settle this question, will have to extend over a consider- 

 able area. 



The second point for photographic investigation refers 

 to the limits of the system towards the galactic poles. 

 There is reason to believe them comparatively restricted. 

 M. Celoria, of the Milan Observatory, using a refractor 

 capable at the utmost of showing stars of eleventh mag- 

 nitude, obtained for a " mean sounding," at the north 

 pole of the Milky Way, almost identically the same num- 

 ber given by Herschel's great reflector.- That is to say, 

 no additiotial stars were revealed by the larger instru- 

 ment. Should this evidence be confirmed, the boundary 

 of the stellar scheme should here be placed at a maximum 

 remoteness of 3500 years of light-travel. 



As a specimen of a photographic gauge-field on a small 

 scale, we may take Prof. Pickering's Catalogue, from the 

 Harvard plates, of 947 stars within 1° of the north celes- 



' Bonner BeobachUmgen. Bd. v. " Einleitung." 

 ^ Metnoi ie dell' Istitiito Loiiibardo, t. xiv. p. 86. 



