578 Prof. R. W. Wood oji PhotocjmpUc 



function o£ the wave-length, that the present investigation 

 was undertaken. 



As I showed several years ago, the Clayden effect, or the 

 type of reversal giving rise to the phenomenon of dark 

 lightning, results from the action of a light-shock on the plate 

 before its exposure to diffuse light. The effect of this light 

 shock, which must be of very brief duration, is to decrease 

 the sensibility of the plate, resulting in a less energetic action 

 during the subsequent illumination. I made no attempt at 

 the time to determine the maximum duration of the light 

 shock which would still give the Clayden reversal, bat ex- 

 pressed the opinion that it could not exceed 1/10,000 of a 

 second. This opinion was based on a single experiment, 

 and I have since found that by a suitable adjustment of the 

 conditions the duration maybe as great as 1/1000 of a second, 

 though only a very slight trace of reversal occurs under these 

 conditions. 



It appears to me now that there are at least four different 

 types of photographic reversal, or perhaps live if we allow 

 the chemical treatment of the plate between two exposures. 

 As any one of these four types is liable to occur in any 

 photographic work, when the proper conditions are fulfilled, 

 it may be well to enumerate them at the beginning. 



First Type. — The ordinary over-exposure reversal, which 

 occurs when the plate is given three or four hundred times 

 its normal exposure and then developed in the usual way. 



Second Type. — The reversals produced by developing the 

 plate in full lamp-light, the plate having been more or less 

 over-exposed to begin with. This type has been extensively 

 studied by Nipher, the results of his experiments being given 

 in the Proceedings of the St. Louis Academy of Science. 



Third Type. — This type must occur frequently, though I 

 never remember to have seen it described. It happens when 

 a normally or under-exposed plate is developed and then 

 exposed to light for a minute or two before the hypo bath. 

 The fogging, which is usually of a reddish-brown colour, 

 does not occur on the portions of the plate where there is a 

 developed image, and even if this image is very feeble it 

 remains clean and almost transparent. I first noticed this 

 ■effect in some photographs of spectra which showed strong 

 reversals along the edges where the illumination must have 

 been very feeble, and was unable to explain it. Further 

 experimenting showed that it bad resulted from turning up 

 the light before the plate had been thoroughl}^ fixed. Doubt- 

 less this effect has been described time and again in the 

 photographic journals, but it was new to me, and may be 



