Prof. Draper on the Chemical Action of Light. 387 



trum to have a positive result ? the want of effect may arise from 

 the intrinsic feebleness of the light, which on these points is just 

 ready to die away. That there is no weight in such a criticism 

 is very easily shown, by slightly modifying the conditions of the 

 erperiment, and admitting a feeble daylight or skylight to act 

 on the plate while the solar spectrum is at work, the action 

 being kept up long enough to permit such a light to stain the 

 plate uniformly all over. After submitting it to the action of 

 mercury, it will be seen that there are three points ; the extreme 

 red, the centre of the yellow, and the extreme violet, which have 

 wholly held the daylight in check ; and two intervening, where 

 the actions have conspired. On the former, the plate is in the 

 same condition as though it had never been exposed to the light, 

 and they are therefore represented as black in the impression ; 

 on the latter, the mercurial vapour readily condenses, and they 

 are therefore white or even blue. 



It thus appears that rays, the times of vibration of which are 

 1, 2, aid the daylight; but those whose times of vibration are 

 ^, li, 2A, interfere with it and destroy its effect. 



A slight modification of the foregoing experiment leads to a 

 very important result. Instead of having the daylight and the 

 sunlight act together, let us iirst expose the plate to a feeble 

 daylight, and then throw upon it the solar spectrum. The result 

 comes out essentially the same in all particulars as the foregoing, 

 but the conclusion is established that it is not necessary for the 

 two rays to act simultaneously ; they may work in succession ; 

 and the effect of one, as the case may be, will be destroyed or 

 increased by the other. 



It is to be understood, that the actions here spoken of as ne- 

 gative and positive are not of equal intensity in the different 

 points. The positive action in the indigo is far greater than that 

 in the red. On inspecting the curve of the effects, its ordinates 

 will be found negative before reaching the red, become positive 

 through that space, ai-e negative again in the yellow, positive 

 again in the indigo and violet, and negative beyond. They differ 

 not only in sign, but also in value ; being shorter in the red than 

 in the indigo when positive, and at the two extremes than in the 

 yellow when negative. 



One of the most striking instances of this positive and negative 

 action which rays can impress on a substance is exhibited by the 

 effect of an electric spark. I discovered the facts three years 

 ago, in investigating the intensity of the Leydcn spark as it 

 passes between metallic balls. Some have su])posed that its 

 brightness varies at different points of the path, the general idea 

 being tliat it is feeblest at the middle point. On causing its 

 image to be formed by a lens on a Daguerreotype plate, there 

 2 D .2 



