1903.] The Cosmical Function of the Green Plant. 431 



grams of 1895, but now I prefer Cadett's excellent spectrum plates. The 

 most convenient form of this wedge-shaped cell for photographic use is 

 the following (fig. 3, A). A cubical glass cell is divided by a glass parti- 

 tion into two prismatic compartments, one of which is filled with the 

 coloured solution, the other with the solvent. A horizontal slit is, on the 

 whole, preferable, since it permits us to alter easily the concentration of 

 the solution so as to obtain the desired spectrogram, and the swing-back 

 of the camera may be used for focussing the spectrum. In these photo- 

 spectrograms (fig. 3, B) we have, of course, the most convenient way of 

 representing absorption spectra, but we shall see later that in certain 

 cases we shall still be obliged to have recourse to another method the 

 spectrophotometric. 



Returning now to our chief problem, we may see how important it 

 was to know the real form of the absorption curve. All the spectra 

 generally represented being discontinuous, showing sharp bands 

 alternating with light intervals, it was rather puzzling to account for 

 the continuous effect, as for instance, in Draper's experiments. 



Armed with the necessary gasometric and spectroscopic methods of 

 research, I could at last repeat Draper's classical experiment without 

 the fear of repeating his error. The result of my experiments was 

 a strict confirmation of the applicability of Sir John Herschel's law 

 to our case. Not only do the two maxima coincide (fig. 4), but the 

 assimilation curve reproduces very nearly the absorption curve, so far 

 at least as the less refrangible part of the spectrum is concerned. It 

 is to be remarked that the secondary maxima of absorption seem to 

 have no effect, but it is subject to doubt, especially after the latest 

 researches of Schunck and Marchlevsky, whether they appertain to the 

 principal chlorophyll constituent or to some products of its decom- 

 position. 



The same results were obtained later by quite another method. 

 The dissociation of carbon dioxide represented by this curve (fig. 4), 

 the " photolysis," as it may be called, is the first and by far the most 

 important stage of the whole process, being directly dependent on the 

 external source of energy. But it is closely followed by the photo- 

 synthesis of organic matter. Though, thanks to the brilliant researches 

 of Horace Brown and Morris, we now know that starch is neither the 

 first nor by far the only product of this synthesis, still it is a product 

 that follows the reduction of the carbon dioxide at the short interval 

 of some minutes, and, what is of still greater importance, its presence 

 may be easily shown by the well-known iodine test. 



If the reduction of carbon dioxide be considered a function of those 

 waves of light which correspond to the chlorophyll absorption bands,, 

 and, on the other hand, the production of starch is the next stage of 

 the same photo-chemical process, it may be fairly presumed that this 

 production of starch in a spectrum will be strictly localised, restricted 



2 H 2 



