Chap, ii Photosynthesis 313 



being transformed into internal work ; to find out the 

 quantitative relation between the energy absorbed and the 

 work done — here lies the brilliant though perhaps arduous 

 problem, in attacking which modern physiologists ought 

 to unite all their forces.' 



The first outcome of the researches was to disprove the 

 view that there is an exact parallel between luminosity and 

 the work effected. Timiriazeff found that the experiments 

 of Draper, on which the idea had been based, had not the 

 value at first attributed to them owing to the fact that the 

 work had been done with a highly impure spectrum . He then 

 proceeded to plot out curves to represent respectively the 

 intensities of decomposition of carbon dioxide, of luminosity, 

 and of energy, and showed that a coincidence of chemical 

 effect with Fraunhofer's luminosity curve is out of the 

 question ; but that so far as the visible spectrum is con- 

 cerned, a relation with the energy curve can be seen. 

 Nor do the results coincide better with the distribution 

 of energy in the complete spectrum, the maximum of which 

 resides in the infra red. Cailletet showed that these rays, 

 which he isolated by passing the light through a solution 

 of iodine in carbon disulphide, have no effect in decom- 

 posing carbon dioxide. 



These researches directed attention to a law formulated 

 by Sir John Herschel, that a photo-chemical reaction may 

 be produced by those rays only that are absorbed by the 

 substance undergoing a change. The names concerned 

 with the application of this law to the action of chlorophyll 

 are those of Jamin, E. Becquerel, Lommel, and Timiriazeff, 

 though the credit of bringing it forward has been generally 

 given to Lommel. In his paper of 1871 (Ueber das Verhalten 

 des Chlorophylls zum Licht), published in Poggendorf's 

 Annalen, he put forward the opinion that the rays which 

 effect the decomposition are those which are absorbed by 

 the chlorophyll, and lays stress particularly on those corre- 



