146 



KNOWLEDGE & SCIENTIFIC NEWS. 



[July, 1904. 



Photography. 



Pvire dLrvd Applied. 



By Chapman Jones, F.I.C, F.C.S., &c. 



At the recent con\-ersazione oi tlie Ro}al Society there 

 were many exhibits that owed tlieir orig-in, at least 

 partly, to photography, such as an optical bench for 

 testing- lenses, by Messrs. R. and J. Beck, three-colour 

 photographs projected by a lantern in which spectrum 

 colours were used instead of coloured screens, by Sir 

 W. de W. Abney, and photographs and photo-micro- 

 graphs of vario'us kinds. But the exhibits tliat seemed 

 to me the most striking- were the stereoscopic trans- 

 parencies, by Mr. Francis Fox, of the Simplon tunnel, 

 now in course of co^nstruction through the Alps, and of 

 the \'ictoria Falls, and also some three-colour lantern 

 slides by Mr. E. Sangei'-Shepherd. The first showed 

 how perfectly stereoscopic transparencies, well mounted, 

 convey an impression of the actual object. And when 

 such views are supplemented as these were by samples 

 of the rock taken from the tunnel, one obtains as good 

 an idea of the actual circumstances as is possible with- 

 out visiting^ the place itself. Indeed, it is doubtful 

 whether a visit would give much more information to 

 the ordinary observer. The three-colour lantern slides 

 were of various spectra, and produced with such a 

 degree ol fidelity that they might well be used for 

 lecture demonstrations when it is desired toi show 

 spectra rather than the means of producing them. It, 

 for example, it were wished to sho'W the spectra of the 

 rarer gaseous elements recently isolated, the use of 

 such slides would give quite as good, if not better, re- 

 sults than the production of the actual spectra, without 

 the cost and risk of employing tubes of the gases them- 

 selves, and the trouble of fitting up spectroscopic ap- 

 paratus. Moreover, the slides would probably give a 

 more representative effect, because thev would be made 

 under the most suitable conditions, instead of having 

 to get the result during- the exigencies of the lecture, 

 and the projection on a screen would be much 

 preferable to the necessity that often arises in such 

 ca.ses of pro'\-iding instruments for the direct eye ob- 

 servation of the few members of the audience that are 

 f(jrtun;ite enough to- gain access toi them. 



Rc'-crsa/. — The reversal of the image is one of the 

 must inleresting and n-ivslcrious of photogfraphic 

 phenomena. Professor K. \\'. \\'ood has catalogued 

 five kinds or types of reversal, ;ind it has been suggested 

 th.'it at least one more should be added to- these. 1 

 think, l-iowe\er, that these should be called methods 

 ratlier than kinds of reversal, and believe that there is 

 g-ood reason for considering that there are probably 

 only two kinds of reversal, which I suggest might well 

 be called progressive and retrogressive respectively. 

 That is one in which the effect of the light action is con- 

 tinued, and one in which it is undone. If the effect is 

 resrarded as a rotary one, then that part which rotates 

 would continue, in the first case, to rotate in the same 

 direction, while in the second case it woiild turn in the 

 backward direction, both arriving at, or tending to- 

 wards, a zerO' or p.seudo-zero point. It does not follow- 

 that retrogressive reversal would of necessitv leave the 

 sensitive substance in just the sanie state a.s it was be- 

 fore it h:id been affected by light. I shall probably 

 return to this subject on another occasion. 



Reversal in Shutter-Speed Tests. — My present intention 

 was not to theorise on reversal, but to refer to a practi- 

 cal result of the Clayden effect; that is, reversal of the 

 developable condition that has been produced by a short 

 exposure to. an intense light, by means of a subsequent 

 comparatively long- exposure to a feeble light, the latter 

 not producing reversal when it precedes the former. In 

 the use of Wynne's shutter-speed tester, a convex 

 polished metal button moves in front of a graduated 

 diagram, and it is photographed while the sun shines 

 upon the whole arrangement. The length of the streak 

 of light produced by the moving button indicates the 

 period of the exposure. By moving- upwards or down- 

 wards the pendulum that carries the bright button, 

 several tests may be made on the same plate, if the 

 camera is not moved. Under these circumstances the 

 comparatively feeble light from the diag-ram produces 

 its effect over the whole surface of the plate during 

 each exposure, including those parts where the bright 

 light from the button imping-es on the plate, and this 

 gives the superposed intense and feeble exposures which 

 are liable to give reversal. I made seven tests on each 

 of a few plates, and generally the first streak was re- 

 versed, the second and third were hardly discernible, 

 the exposure effect from the bright button being incom- 

 pletelv reversed and the result neither one thing nor 

 the other. 1 find that it is best tOi start with, the longest 

 exposure and to let the others (generally not more than 

 three) follow. When giving 8, 6, 1^7, and J hundredths 

 of a second in the order stated there was no' reversal. 

 But 20, 3, 2, I, and ? hundredths of a second gave 

 reversal for ij divisions and a feeble effort for the next 

 4 divisions (each division equivalent tO' tlie hundredth 

 of a second) of the beginning' of the first streak. Tlie 

 rest of the streak representing the first exposure, and 

 all the others were represented by good black lines on 

 the plate. With a series of 27, 11,7, and 2 hundredths, 

 4.5 divisions at the beginning of the first were reversed, 

 the next 10 were feeble, and the remainder good strong 

 images. Tlius the Clavden efiect interferes som.etimes 

 in ;i verv practical wav under vei-y coimmon-place 

 circumstances, and it is desirable tO' bear this in mind 

 when giving- multiple exposures, as is often done in 

 experimental work. In the particular case referred to, 

 if the sunlight were constant in its brilliancv, the streaks 

 on the plate would all have practically the same ex- 

 posure, the long-er exposures simply giving- longer 

 streaks. But the leng-th of the exposure would affect 

 the general illumination of the plate by the surface of 

 the diagram, and it seems that it is desirable tO' have as 

 little as possible of this tO' jnUovj or to> be superposed 

 u|-)on the exposure effect due to the brilliant light re- 

 flected from the polished button. 



S pirit Levels. — \o camera intended for general work 

 is complete without some means of showing; when the 

 sensitive surface is perpendicular, and in by far the 

 greater number of cases a circular spirit level is the most 

 convenient and effective apparatus for this purpose. 



Messrs. Taylor, Tavlor, and Hobson many years ag-o 

 introduced the level shown in section, and were there- 

 by successful in obviating the leakage which, before 



