Tit'i'tn'saJs in SjH'ctnoii Pliotogniphs. 587 



IMiH'li more expcM-iinental work will have to bo done before 

 any detinite notion can be obtained as to the nature of the 

 cliano^es produced by the action of radiation of any sort, and 

 it is hoped that the experiments described in this paper may 

 prove sugorestive to others. Doubtless an exhaustive study 

 of the action of various chemical ag(»nts on the ])late between 

 the two exjjosures would throw much light on the cause of 

 the reversals. 



If I interpret the strain-theory correctly, the application 

 to these phenomena would be to assume that the light-shock 

 produces a negative strain, while lamp-light produces a posi- 

 tive strain, either of which yields an image on development. 

 The reversal in cases where the lamp- light follows the im- 

 pression of the light- shock could be explained by assuming 

 that the negative strain has to be undone before the positive 

 strain can begin, consequently these parts of the plate lag 

 behind the parts which have not received the light-shock. 

 We should then have to assume that the positive strain once 

 started can be continued by a stimulus which, acting first, 

 would have ]n'oduced a negative strain, in order to account 

 for the fact that exposures to lamp-light are not reversed by 

 light-shocks. Moreover, it is difhcu t to explain on the strain- 

 theory that two different stimuli acting in succession may 

 produce only the same effect as one of them acting alone. 

 It appears to me that the strain-theory would lead us to 

 suppose that the negative strain produced by the first stimulus 

 might be exactly neutralized by a stimulus which produces a 

 positive strain, the plate returning to its original condition, 

 i, e. not darkening on development. This is never the case. 



In cases where reversed lines appear in the spectrum, 

 which are suspected of being pho ographic in origin, i. e., 

 not true absorption-lines, the followmg precautions should 

 be taken. Repeat the exposure a number of times, using 

 successively smaller diaphragms before the prism. If the 

 reversal is ordinary solarization, due to over-exposure, it 

 should disappear when the intensity of the light is sufficiently 

 reduced. It seems to me that in such cases the reversed line 

 should be bordered by bright edges, w^hich does not seem to 

 be the case in the photograjihs published by Professor Trow- 

 bridge. If the Clayden effect is suspected, the source of 

 light should be examined with a revolving mirror, to deter- 

 mine whether a dual illumination is present. The speed of 

 the mirror should not be too great, otherwise the phospho- 

 rescence, if it exists, may be spread out to such an extent 

 that no trace of it appears. This may account for the failure 

 to obtain evidence of a dual illumination in the case of 

 heavy discbarges in quartz tubes. 



