2(5 Prof. C. Timiriazeff. [A])r. 30, 



the sum of nearly all the rays of the spectrum, its limits only remaining 

 monochromatic. 



The next step was to prove that this coincidence of the two maxima 

 of luminosity and of the chemical effect did not in reality exist. 

 A simple inspection of these three curves (the curve of assimilation, 

 that of luminosity and the energy curve), taken from my first (and 

 last) German paper, proves with sufficient evidence that a coincidence 

 of the chemical effect with Fraunhofer's luminosity curve was out of 

 the question ; but, so far as the visible spectrum is concerned, there 

 exists a decided relation with the energy curve. 



In this first and preliminary research the more simple and con- 

 venient method of coloured liquid screens, introduced by Senebier 

 and applied in a more exact form by Daubeny, was adopted. I merely 

 introduced a more correct way of calculation and graphical representa- 

 tion of the results obtained. At the same time I applied the method 

 of gas analysis now in general use, though quite erroneously attributed 

 to Professor Pfeffer. 



But if these results were sufficient to dispose of the current ideas 

 on the importance of the luminosity, based as they were on Draper's 

 experiments, they were not sufficient for the building of another 

 theory. It was impossible, for instance, to consider the decomposition 

 of the carbon dioxide as simply an effect of the relative energy of the 

 radiation. The maximum of energy in a prismatic spectrum lies in 

 the infra-red, and a remarkable experiment of Cailletet (curiously too 

 often omitted in most of the historical sketches of the subject) gave 

 an unequivocal proof that the rays of light, filtered through Tyndall's 

 solution of iodine in carbon bisulphide, were incapable of producing 

 the reduction of the carbon dioxide in a green leaf. 



It was evident that some other principle had to be brought forward 

 in order to explain the absence of any chemical action in the invisible 

 part of the spectrum, as well as its distribution in the visible. The 

 principle that was appealed to was Sir John Herschel's law : that 

 a photo-chemical reaction may be induced by those rays only that are 

 -absorbed by the substance undergoing a change and consequently, as 

 a rule, presenting a complementary colour. It was supposed that, 

 applied to our case, Herschel's law would mean that the reduction 

 of the carbon dioxide must take place at the expense of those rays 

 of the spectrum which are absorbed by the green matter of the leaf- 

 that they must correspond to the absorption bands of chlorophyll. 



Lommel is generally credited with having been the first to enunciate 

 the idea that the reduction of the carbon dioxide may be considered 

 as a function of the energy of radiation and of the degree of its 

 absorption by chlorophyll. But this opinion, despite its being very 

 general, is none the less erroneous. Lommel himself admits that the 

 first half of the proposition, concerning the dependence on the energy 



