Proceedings of the British Association. 399 



the agency of colored glasses, but which does not appear to have been 

 accepted by later authors, who have attributed it to the chemical rays. 

 There is but one way by which the question can be finally settled, and 

 that is by conducting the experiment in the prismatic spectrum itself. 

 When we consider the feebleness of effect which takes place, by rea- 

 son of the dispersion of the incident beam through the action of the 

 prism, and the great loss of light through reflection from its surface, it 

 would appear a difficult operation to effect the determination in this way. 

 Encouraged however by the purity of the skies in America, Dr. Draper 

 made the trial, and met with complete success. The process was as 

 follows : — a series of tubes, half an inch in diameter and six inches long, 

 were arranged so that the colored spaces of the spectrum fell on them. 

 In these tubes, water, impregnated with carbonic acid gas, and contain- 

 ing a few green leaves, (Poa annua,) was placed. It was expected, that 

 if the decomposition be due to the radiant heat, the tube occupying the 

 red space, or even the one in the extra-spectral red space, would, at the 

 close of the experiment, contain most gas. If it were the " chemical 

 rays," in the common acceptation of the term, we might look for the 

 effect in the blue, violet, or indigo spaces ; but if it were the light, the 

 gas should make its appearance in the yellow, with some in the green, 

 and some in the orange. I made the trial several times, (says Dr. D.) 

 and found it much more easy to accomplish than I had expected. The 

 results were briefly as follows : — In the tube that was in the red space 

 a minute bubble was sometimes found, but sometimes none at all. That 

 in the orange contained a more considerable quantity ; in the yellow ray 

 a very large amount, comparatively speaking ; in the green, a much 

 smaller quantity ; in the blue, the indigo, the violet, and the extra-spec- 

 tral space at the end, not a solitary bubble. From these facts, in con- 

 nection with some results obtained by the use of bichromate of potash, 

 as an absorptive medium, I conclude that it is the rays of light which 

 effect the decomposition, and that the rays of heat and the tithonic rays 

 have nothing to do with the phenomenon. The alkaline bicarbonates 

 are easily decomposed by elevation of temperature, yielding a portion 

 of their acid at the boiling point of water. Instead of using a solution 

 of carbonic acid I endeavored to effect the decomposition of these salts 

 by leaves in the sunlight, and found that it took place with facility. Nor 

 is the effect limited to the removal and decomposition of the second at- 

 om of the acid. It passed on to the first ; the neutral carbonate of soda 

 itself decomposing and yielding oxygen gas. In like manner, the ses- 

 quicarbonate of ammonia may be made to yield a very pure oxygen gas. 

 Dr. Draper, in conclusion, alluded to his method of multiplying the Da- 

 guerreotype pictures, as published in the Philosophical Magazine, and 

 mentioned a process of precipitating copper, after the picture has been 



