September 1, 1891.] 



KNOWLEDGE. 



17^ 



supporting the picture in March number of Knowledge as 

 to the relative hrifiJitness of the seeeral /iiirts of the nebulous 

 light. I think, therefore, there can be no doubt that my 

 picture is right and Mr. Barnard's suffers from over- 

 development. I have adopted the same method of making 

 sure of the features in other parts of the Milky Way, and 

 I am sure it is necessary where exceedingly delicate details 

 have to be shown. When my photograph to which Mr. 

 Barnard refers was taken, your reproduction of his of the 

 same object was before me, and I saw the differences, which I 

 was then disposed to attribute to the great advantage Mount 

 Hamilton has in elevation, but Prof. Pickering's opinion of 

 the advantage of a mountain station — that it only enables 

 us to photograph stars J of a magnitude lower — seems to 

 make my supposition untenable, and to leave no alternative 

 but that the difference is due to development, and I think 

 over-development of the Mount Hamilton picture. Seed's 

 plates are undoubtedly very good — I have used some of 

 them, and I think highly of them — but they are not so 

 sensitive as the Ilford plates, which I used for the pictures 

 in question. 



]\Iy experience leads me to the conviction that where 

 extremely delicate details have to be caught, such as those 

 in the Nubecula Major, the nebula about Eta Argus, and 

 others, what is required is lengtli of exposure rather than 

 strong development. I have repeatedly tried the two 

 methods, i.e., long exposure and strong development, on 

 the same object, and have no hesitation in saying that for 

 delicate nebulous details the latter is to be decidedly 

 avoided. Yours very truly, 



Sydney Observatory, H. C. Russell. 



July Qtli, 1891. 



[A close examination of Mr. Barnard's picture of the 

 Sagittarius region referred to by Mr. Russell has convinced 

 me that the nebulosity shown upon it really corresponds to 

 areas of luminosity on the heavens, and is not a mere 

 photographic fogging due to over-rapid development. There 

 is an intimate connection between the dark areas on Mr. 

 Barnard's photograph and the disposition of the stars 

 which could not be due to chance, and the bright structures 

 are altogether diiJerent in form from the patches in which 

 fog shows itself on an over-developed picture. From my 

 experience in reproducing pictures of nebuhp, I can fully 

 confirm Mr. Barnard's statement that the density of the 

 nebulosity and fainter cloud forms is very dependent on 

 suitable development. It is to be regretted that circum- 

 stances have prevented Mr. Barnard from taking more 

 photographs of this extraordinary region of the Milky Way. 

 Mr. Russell has recently sent me some most interesting 

 photographs of the 'tj Argus Nebulse and the « Crucis 

 cluster, which I hope shortly to reproduce in Knowledge. 

 — A. C. Ranyard.1 



THE OBSEEVATION OP RED STARS. 

 To the Editor of Knowledge. 



Dear Sir, — I have been considerably interested by a 

 perusal of Miss Gierke's article, with the above heading, 

 in the August number of Knowledge. With much of 

 what is there stated I entirely agree ; though, with all 

 possible deference to the gifted authoress, I venture to 

 dissent from her conclusions on one very material point, 

 and that is — can the alleged variation of colour in certain 

 stars be accepted as a demonstrated faet "? A long experi- 

 ence of my own in observing star colours, as well as that 

 resulting from the superintendence of such work by other 

 observers, has made me extremely sceptical on this point ; 

 and I should requu'e stronger evidence than we now 

 possess to admit that such a conclusion was established. 



Casual star-gazers have no idea of the practical difficulties 

 that beset this subject. Eyes differ so widely in their 

 colour perceptions, that it is really not safe to accept the 

 uncoiToborated testimony of any one observer until it has 

 been ascertained that his colour faculty is fairly normal. 

 Neither can the differences arising fi-om various kinds and 

 apertures of telescopes be safely ignored : reflectors almost 

 invariably give deeper tints than refractors, whilst any 

 great diffei-ence in aperture has its corresponding effect. 

 Then there is the atmospheric influence to be added to 

 the account, and experienced observers know that this is a 

 potent factor. The altitude of the star, the degree of 

 darkness of the sky background, humidity and clearness 

 of the air, etc. — if apparently trivial each by itself— will, 

 combined, exercise some perceptible efl'ect on the resulting 

 colour. Therefore, I contend that the most satisfactory 

 method of eliminating these errors — personal, instrumental, 

 and atmospheric— is to take the mean of a large number 

 of independent contemporaneous estimates by different 

 observers, and accept that as the nearest probable approach 

 to the true result. I woiJd like to make a few remarks 

 on the objects cited as examples of variation in colour : — 



Espm-Birmingham, 352 (Schj. 148). The discrepancies 

 in its assigned tint of redness are quite within the usual 

 range of erroi-s of observation, the large dift'erence in the 

 apertures of the telescopes used being alone sufficient 

 to account for much of the discordance. The evidence 

 for variation in magnitude does not seem very strong ; 

 indeed, it might reasonably be inferred that Birmingham's 

 solitary observation of "no colour " belonged to some 

 other star. 



Es-Birm. 221 (Schj. 90). I think there is quite a 

 possibility that these irreconcilable estimates relate to 

 two different stars. According to the B.A. Catalogue 

 (which formed the basis of my working list in 1877-8), 

 there are two Gk mag. stars close together near this place 

 — B.A.C. 2365 and 2369, otherwise 44 and 45 Camelopardi. 

 Now the former of these agrees so closely with the position 

 given for Es-Birm. 221, that it seems as if they must be 

 synonymous ; this was the object I observed as " Gi mag., 

 orange red," with a 5 inch refractor. For the other one 

 {32s. following and 13' north), I could find no star larger 

 than 8 mag. in its place, and as no colour was mentioned, 

 it was presumably white. Possibly one or both of these 

 stars may be variable in magnitude. 



Birm. 118 (Schj. 63). This I noted in 1885, with 11^ 

 inch reflector, as "yellowish white." Perhaps there is 

 some reason to doubt its identity, judging from the foot- 

 note to the Dunsink observations, which suggests that 

 the estimate probably belonged to another star closely 

 preceding. 



Es-Birm. 544 (Schj. 214). It does look as if there was 

 some reason to suspect variation here, in magnitude if not 

 in colour. But a more extended series of observations are 

 recjuired before the latter assumption can be legitimately 

 entertained, as there is a gap of fourteen years between 

 Espin's observation and the last recorded one preceding. 



(■ Velorum does indeed appear to offer promising gi-ound 

 for further investigation, but, in the meantime, it would 

 be wise to suspend judgment. 



If suspected colour variables have been apparently 

 neglected, it is because the evidence for variation is so 

 slender that it finds but little acceptance in the astrono- 

 mical world. We iini.it be sure of the facts before proceeding 

 to draw deductions from them — and the facts are scarcely 

 established, as yet. In saying this, I am equally desirous 

 of further investigation in the fascinating field of stellar 

 chromatics, and think, with Miss Clerke. that the spectro- 

 scope is the most promising instrument to deal with the 



