220 



KNOWLEDGE. 



[September 1, 1897. 



gives the same simultaneous exposure in all three cases, 

 develops the resultant negatives in the same manner, and 

 then proceeds to compare the results as to general 

 appearance, and also by counting the number of stars 

 shown on the three plates. Either of these methods of 

 comparison must under the conditions be unfair to the 

 larger or the smaller instruments. For, the relative 

 exposure times being very different in the three cases, the 

 method of development best suited for one would be 

 entirely inappropriate for the others, particularly for 

 bringing out nebular details. For nebular details the 

 nlatire exposures would be about in the ratio of one, one 

 hundred and four, and twenty-six respectively.* It is no 

 wonder that the plates taken with the smaller lens were 

 " badly fogged " by the development given to the others. 

 And under such circumstances the plates taken with the 

 portrait lenses would show less detail in the nebular 

 structure, even if the very much smaller scale of the 

 photographs was not taken into consideration at all. The 

 comparison of the iiuinber of stars shown on the dififerent 

 plates is equally misleading, for the obvious reasons 

 pointed out by Prof. Barnard! (ditierence in scale of 

 photographs, and difl'erence in the comparative richness 

 of different portions of the sky iu stars of different mag- 

 nitude). 



The intensity of the star images at the focus of the three 



. . . ,, • iu i- (20)' (:!-5)' (5)' 



mstruments would vary m the ratio ,g^.„ : ,q., .„ : (-mTo^j- 



or about in the ratio of 10 : 1 : 1-1. 



Considering this, and considering also that the five-inch 

 lens was considered to show from two and a half to three 

 times as manij stars as the three and a half inch for a 

 photographic effect less ten per cent, greater, it is only 

 surprising that the numh, r of stars shown by the reflector 

 was not very much greater than actually was the case. 

 Probably part of the star images on the plate taken with 

 the three and a half inch lens were blotted out in the 

 development. 



There is another point of great importance in this con- 

 nection that must not be lost sight of, and that is the 

 practical effect of using very short focus (large angular 

 aperture) combinations in increasing the general illumina- 

 tion of the field. The shorter the foaus the greater the 

 curvature of the lenses, and consequently the greater the 

 quantity of light scattered by surface reflection and 

 diffusion. For this reason the use of a larger ratio than 

 one to five in refractors is never desirable. Dr. Roberts 

 arrives at a similar conclusion from a different point 

 of view, however — viz., that with larger angular aper- 

 tures the stellar images are not as good ; and in view 

 of this it is rather surprising that he should consider 

 the results of his comparisons as at all conclusive, leaving 

 aside entirely the question of relative linear apertures. 

 A comparison of some of his photographs with those of 

 the same region of the sky obtained with the Bruce 

 photographic telescope, which has the same angular 

 and very nearly the same linear aperture as the twenty- 

 inch reflector, would be much fairer and more con- 

 clusive as to the relative efdciency of the two classes of 

 instruments. 



On one point there can be no doubt (since it has been 

 proven over and over again by experiment), i.e., that up 

 to a certain aperture (probably not less than eighteen to 



* See the comparison of Barnard's results with the " lantern " 

 and tEe'Willard lens, and also with Pickering's results with the 

 Voightlilnder lens, already described. 



f " On tlie Comparison of Reflector and Portrait Lens Photo- 

 graphs," -T/.iP., November, 1896, p. 10. 



twenty inches) the light-grasping power of a photographic 

 objective is greater than that of a silver on glass reflector.'' 

 In photographic work, moreover, the iraij in which the 

 light is lost is quite as important as the quantity lost. In 

 the refractor the larger proportion is lost by absorption 

 (which does not produce any general illumination of the 

 field) ; in the case of the reflector, diffusion and irregular 

 reflection due to minute scratches on the film, dust on the 

 silvered surfaces, etc., etc., play the more important part. 

 The field of the reflector is, therefore, as a rule considerably 

 brighter than that of the refractor. This general 

 illumination of the field by this cause may, in certain cases, 

 be even greater than that due to the sky, in which cases, 

 of course, the conclusions which have been drawn respect- 

 ing the influence of the latter in determining contrast effect 

 with different apertures may require modification. They 

 would not hold in their entirety, for example, in the case 

 of such extremely large angular apertures as ^■] ^. Such 

 lenses would be likely to soon fog the plates by light 

 scattered from the lenses alone, no matter how small the 

 aperture. 



One advantage of the reflector which is made much oi 

 is that of a more perfect concentration of the light of 

 differing wave-lengths into a single image. It must be 

 pointed out first of all that the image is not properly a 

 single image — even in the case of a perfectly corrected 

 reflector — but a number of superposed images, each of 

 which, in the case of a star or other point source, is a 

 diffraction pattern whose "width" (first minima) varies 

 directly as the wave-length, and whose intensity at the 

 centre varies inversely as the square of the wave-length. 

 P'or this reason, and because of the limited range in 

 sensitiveness of photographic plates, only that light of wave- 

 length near that corresponding to maximum sensitiveness 

 contributes to any great extent to the effective photographic 

 intensity of the image. A refractor (particularly a photo- 

 graphic doublet) can now be constructed so as to unite all 

 these effective rays almost perfectly, and at the same time 

 give a comparatively large field. 



No one is a more enthusiastic advocate than I have been 

 of the great advantage of the reflector over the refractor 

 for astrophysical — particularly spectroscopic — investiga- 

 tions ; t but this, I hope, does not prevent my recognizmg 

 its obvious disadvantages in other lines of work. It seems 

 to me that not only on the ground of the much larger field 

 obtainable, but also on account of its greater freedom 

 from diffused light, an i its greater stabihty during long 

 exposures, the photographic refractor must be considered 

 as unquestionably superior to the reflector for general 

 work — although for particular purposes, where an unusually 

 large linear or angular aperture and only a small field is 

 desired (as in photographing exceedingly faint individual 

 stars), the latter instrument may be used with advantage. 



* .See paper by Prof. Hale "On the Comparative Value of Refract- 

 ing and Reflecting Telescopes for Astrophysical Investigation," 

 Astrophysical Journal, February, 1897, p. 119. According to the 

 table (page 127) accompanying this paper, which is based on careful 

 experiments by Vogel on the absorption of light by different varieties 

 of optical glass, on cxjjeriments by Langley on the reflective power of 

 a silver surface, and on direct coiupanson of the refractor with the 

 reflector by Dr. Gill, Prof. Pritchard, and others, the light-grasping 

 power of the two forms of objective becomes equal for au aperture of 

 about eighty-seven centimetres (about thirty-four incliesi. In the 

 case of a photographic doublet more light is lost than in the cas; of 

 an ordinary objective ; hence I have assumed a correspondingly lower 

 limit. < ., J 



t See various papers in the Phil. Maj., Vol. XXVIII., p. 137 j 

 Vol. XXVIII., p. 337; Astronomy and Astrophysics, Vol. XIII., 

 p. 835 ; Astrophysical Journal, Vol. I., p. 52; Vol. I., p. 232 ; Vol. 

 III., p. 169 («ee particularly pp. 182 and 183) ; Vol. III., p. 317 ; etc. 



