340 



SCIENCE 



[Vol. XIX. No. 



action can be traced into the yellow as the result of very 

 long exposure, or even still further. For ordinary exposures, 

 however, we may consider that the action does not pass the 

 blue, particularly when photographing bright sources of 

 light, such as the stars, because the more refrangible rays 

 are so very much more powerful in their effect upon the 

 plate that they exert their full action before the others can 

 make a visible impression. To extend the time beyond that 

 point would result in a reversal of the effect sought for, a 

 change in the character ot the negative, and serious irradia- 

 tion or spreading of the light around the image, resulting in 

 impaired definition. With ordinary sensitive plates, there- 

 fore, the images we photograph are images made with blue, 

 ■violet, and ultra-violet rays, covering, indeed, a considerable 

 range in the spectrum, but excluding a large and important 

 portion of it. 



The differences in the character of star radiations are so 

 considerable that the blue is sometimes very strong and bril- 

 liant, even exceeding that of the sun relatively to the other 

 parts of the spectrum, as we find it in a Lyrae and in Sirius; 

 while in other stars the temperature is so low that there is 

 scarcely any blue, and line-absorption gives place to flutings, 

 or even to the bright lines of incandescence from comets and 

 nebulae. It is obvious, therefore, that one star not only 

 differs from another in glory as seen by the eye, but the 

 photographic plate, which takes no account of any colors 

 beyond its limited range of sensitiveness, tends to exaggerate 

 the difference, and give utterly false evidence of relative 

 brightness. For a red star may appear very bright to the 

 eye, while its image on the plate would be very faint or 

 perhaps scarcely discernable. 



With orthochromatic plates the result will be different, 

 provided the telescope itself is not at fault. We will assume 

 for the moment that the telescope is so constructed that the 

 "chemical" and visual foci exactly coincide, and that the 

 plates are equally sensitive to all the colors of the spectrum. 

 Then the negative will show exactly vvhat is seen by the eye, 

 and these are the only conditions under wliich such a result 

 can be perfectly attained. 



Doubtless such perfectly corrected telescopes, or perhaps I 

 should say such as are so corrected within the limits of the 

 optician's skill, are rarely available, and a very usual plan is 

 to make certain corrections for ordinary telescopes to adapt 

 them to photographic work. The effect of these corrections 

 now deserves consideration. 



The difference between the so-called "chemical" focus* 

 and the visual focus of a telescope may be little or it may be 

 half an inch. In either case the photographed image will be 

 decidedly out of focus if allowance for this difference be not 

 very carefully made. The usual means of doing this is to 

 change the position of the plate-holder, and find the place of 

 the sharpest definition by trials. By properly arranging the 

 ground-glass and the plate-holder, the plate will always be 

 in focus for the actinic rays when the image appears sharp 

 on the ground-glass. 



Having accomplished this result, we have succeeded in 

 doing precisely what we do not wish to do, viz., instead of 

 arranging the instrument to photograph what the eye can 

 see, by means of the extended and uniform sensitiveness of 

 an orthochromatic plate, we have arranged it to define only 

 with blue or violet rays, and have restricted its range to 

 stars that are specially characterized by highly refrangible 

 radiations, effectually cutting off the red and yellow stars, 

 and rendering the use of orthochromatic plates not only use- 

 ess but positively objectionable. 



As regards the red and yellow stars, the greater portion of 

 their light will be brought to a focus at the point of distinct 

 vision, not on the sensitive plate; and the feeble radiations 

 of higher refrangibility, being too weak to act strongly 

 upon the plate, such stars will be but faintly shown in 

 the negative. The rays not focussed on the plate wiH 

 tend to blur the images, and this effect will be more pro- 

 nounced and objectionable in proportion as the range of 

 sensitiveness of the plate to the different parts of the spec- 

 trum is increased. For this reason the most perfect pictures 

 would be produced, under the conditions described, by using 

 plates sensitive only to the particular rays that form the 

 image on the plate, or else by cutting off the other rays by a 

 screen, thus working with monochromatic light. 



It is possible that there may be some object in photograph- 

 ing stars with the different colors of the spectrum separately, 

 in which case orthochromatic plates can be so prepared that 

 they will select the particular light required, and such ob- 

 servations may be made with ordinary telescopes, correcting 

 them for each set of rays in turn, in the manner described. 

 But if I correctly understand the purpose of photographic 

 star-maps, they are intended not only to represent the dis- 

 tribution of stars and their relative positions, but also to show 

 their respective brightness, or, as we usually call it, magni- 

 tudes. Now magnitude measured by brightness is not the 

 same as the photographic action of the stars upon a plate of 

 restricted spectrum sensitiveness, such as all ordinary sensi- 

 tive plates, and this, although a self-evident proposition, has 

 not received in practice the attention it deserves. On the 

 other hand, orthochromatic plates will give perfectly truth- 

 ful representations of the starry heavens when used with 

 perfectly corrected telescopes, as already explained, and that 

 they will only do so under such conditions is, I believe, ob- 

 vious. 



If it is possible to make plates of uniform sensitiveness as 

 regards tests in the sensitometer, and also as regards all the 

 rays of the visible spectrum, and if such plates can be pro- 

 duced regularly in large quantities, we may consider the 

 problem of photographing the stars to be satisfactorily solved. 

 But much yet remains to be done before a plate that can be 

 regarded as standard can be adopted. The composition of 

 the emulsion, the manner of rendering it sensitive, the means 

 of testing the plates, including the standard of light to be- 

 used in the process, and the keeping qualities of the plates,, 

 must all be thoroughly investigated before it will be safe tO' 

 adopt a standard plate for universal use. Nevertheless, wei 

 are in a position now to begin practical work, and the re- 

 sults will be of permanent value if we act upon the proposi- 

 tion that with orthochromatic plates there is no distinction! 

 of chemical and visual rays, and that such plates can only 

 be advantageously employed when all the rays from red to. 

 ultra-violet are brought to a focus in a single plane. 



I do not venture upon any speculations as to the probably- 

 best method of preparing color sensitive plates for astronomi- 

 cal work, for the reason that new methods are constantly- 

 being tried. I will say, however, that I deem it not at alii 

 improbable that collodion will be found superior to gelatine- 

 as a vehicle for the emulsion, and although the gelatine plates-- 

 are at present more rapid than collodion emulsion plates,, 

 there is no obvious reason for this, further than that we not. 

 yet know how to make extremely rapid plates with collodion „ 

 But there are some objections to gelatine and none to collo- 

 dion. Gelatine swells in water, particularly in warm climates, 

 and, although this defect can be to some extent controlled, 

 it is really at times a serious trouble, which no "tropical "'" 



