582 Prof. R. W. Wood on FhotogropMc 



dispose of the idea of selective reversibility, at least so far 

 as. the Clajden effect is concerned. 



InveHigation of the 2\me Factor. — To determine the maxi- 

 mum duration of time which the light-shock may have and 

 still reverse, the following method was used : — A disk of 

 cardboard 50 cms. in diameter was mounted on the shaft of 

 an electric motor, the speed of which could be determined by 

 the tracing of a tuning-fork on a smoked metal plate mounted 

 on the same shaft. Near the rim of disk a number of narrow 

 slits were cut, varying in width from 1 mm. to 5 mms. An 

 arc-light was focussed on the rim by means of a large con- 

 denslng-lens, the image of the crater being about half a 

 millimetre in diameter. By driving the disk at a high rate 

 of speed intermittent flashes of very brief duration were 

 obtained as the slits passed across the arc^s image. A short 

 distance behind the disk a rectangular metal tube was mounted 

 provided with a slit 1'5 mm. wide, immediately opposite the 

 point where the image of the arc fell on the disk. , Down this 

 tube the plate was dropped, receiving in its passage before 

 the ,slit light flashes of varying duration. The plate was 

 subsequently exposed to candle-light and developed. The 

 images of the slit in the case of the briefest flashes were 

 perfectly sharp, in other cases they were bioadened owing to 

 the rapid motion of the plate. This made the interpretation 

 of some of the records ditficult. and it was found better, when 

 working with flashes longer than 1/2000 of a second in dura- 

 tion, to lower the plate down the tube with a thread. A 

 print from one of these plates is shown in fig. 2, Plate XXVI. 

 In this case there were two 1 mm. slits on the rim of the 

 disk not very far apart, then a slit somewhat wider further 

 around, and after this a still wider one. It will be seen that 

 the slit images formed by the two very short flashes are com- 

 pletely reversed_, while the others are only reversed on their 

 edges. 



if the plate moves during the exposure, as was the case 

 in this photograph, it is obvious that the edges of the slit- 

 image will receive less exposure than the centre, which 

 accounts for the partial reversal. As the result of exposing 

 about two dozen plates, it w^as found that the duration of the 

 shock could be as long as 1/1000 of a second_, and still yield 

 reversals. It was only by carefully regulating the intensity 

 of the fogging-light and the duration of the development 

 that these reversals could be obtained. 



When the duration is less than 1/2000 sec. reversals could 

 be obtained without difficulty. Flashes varying in duration 

 from 1/15,000 sec. to 1/500 sec. were studied, and it was 



