580 Prof. R. W. Wood on Photographic 



developed, the spark-images would come out reversed and the 

 slit images not reversed. Before fogging it one end (the 

 upper in the print) was immersed for a few minutes in a very 

 dikite solution of bichromate of potash, slightly acidified with 

 nitric acid. It was then dried, exposed to candle-light and 

 developed. In the lower portion of the print we find the 

 sparks black and the slit-images white ; in the upper portion 

 both sets of images come out dark, the reversal of the 

 sparks being much stronger than on the untreated portion of 

 the plate. 



In the present paper I propose to discuss the Clavden 

 effect not only in connexion with the selective reversibility 

 hypothesis adopted by Professor Trowbridge to explain his 

 spectrum photographs, but also in relation to the time factor, 

 and radiations other than light, such as the Becquerel and 

 Pontgen ravs, which are quite different in their action from 

 light. 



I shall, in dealing with the subject, speak of the initial 

 exposure of brief duration as the light- shock. The subse- 

 quent illumination which causes the reversal of the impression 

 of the shock, I shall call the fogging exposure. 



The first subject investigated was the relation of the 

 phenomenon to the wave-length of the light. The light 

 shock in this case was administered by exposing the plate to 

 the spectrum of one or more sparks between cadmium elec- 

 trodes by means of a small quartz spectrograph. Even with 

 a small diaphragm the illumination by a single spark yielded 

 a developable image of the spectrum down to the extreme 

 ultra-violet. Six spectra were impressed on the same plate 

 with different sized diaphragms ; the plate was then exposed 

 to the light of a candle for a few seconds and developed, the 

 result being reproduced in Plate XXVI. fig. 1. It will be seen 

 that the lines and the continuous background come out posi- 

 tive in the two upper spectra, showing that if the shock is 

 too intense no reversal takes place, a circumstance in which 

 the Clayden effect differs essentially from ordinary reversal 

 due to over-exposure. Professor Trowbridge says that his 

 reversals occur where bright lines fall on a continuous back- 

 ground, and considers the reversing action proportional to 

 the product of the two effects. If his reversals are of this 

 nature this cannot be the case, for by making one factor (the 

 light shock) large, no trace of reversal appears. As I shall 

 show presently this statement requires some modification, for, 

 as we increase the intensity of the shock, we can by increasing 

 the fogging exposure still get reversal. As I said in my 

 previous note, it appears probable to me that in Professor 



