288 THE ALLOTROPISM OF CHLORINE. [MEMOIR XX. 



ascertaining the rate of decomposition in the reflected 

 beam, if the bulb be set in the direct sunshine the bub- 

 bles increase in number; the total quantity of oxygen 

 evolved becoming greater in the same space of time, an 

 effect obviously clue to the difference of intensity of the 

 reflected and incident beams. When a certain point is 

 gained, apparently no further increase of effect takes 

 place on increasing the brilliancy of the light, as I found 

 by employing a convex lens. 



With respect to the influence of temperature. If, 

 while one of the bulbs is actively evolving gas in the 

 sun-rays, it be warmed by the application of a spirit- 

 lamp, the amount of gas thrown off becomes very much 

 greater. A difference of a few degrees produces a strik- 

 ing effect. As an illustration of this I placed in the sun- 

 shine two bulbs which were nearly alike, except that 

 one of them was painted black with India-ink on that 

 portion which was farthest from the sun. The rays 

 coming through the transparent part had access to the 

 solution, and then, impinging on the dark side, raised its 

 temperature. On measuring the quantity of gas col- 

 lected, it was found 



In the transparent bulb 3.46 



In the half-blackened bulb C.19 



IV. The decomposition of water, once begun in the 

 sunbeams, goes on afterwards in the dark. 



1st. This very important fact may be established in a 

 variety of ways. Thus, if a bulb be removed from the 

 sunshine while it is actively evolving gas, and be placed 

 in the dark after all the gas has been turned out of it, a 

 slow evolution continuously goes on, the gas collecting 

 in the upper part of the bulb. 



2d. A bulb, A, Fig. 45, having a neck, b, the end of 

 which was bent at c upwards at an angle of about 45 

 degrees, was employed. After exposure to the sun, by 



