124 Proceedings of Royal Society of Edinburgh. [sess. 
! 
Another difference between the trichosanthes colouring matter and 
chlorophyll is that when treated either with ammonia sulphide 
or hydrochloric acid the former ceases to be green in thin layers, 
while the latter ceases to be red in thick layers. The reason for 
this becomes at once plain when the spectra are examined. The 
question of the relation which the trichosanthes colouring matter 
bears to chlorophyll is one of considerable interest, for it is evident 
that there is some relation, since in the products of the action of 
hydrochloric acid there are three bands which are common to the 
two spectra. That they differ greatly is equally evident from the 
dissimilarity of the original spectra and of those got after acting on 
the solutions with ammonia sulphide. The trichosanthes colouring 
matter is found inside a thick opaque rind, and so has probably 
been formed in the dark, at least it cannot serve the purpose which 
chlorophyll serves in leaves. Chlorophyll, according to Sachs, is 
formed in complete darkness only in the cotyledons of some 
conifers and in the leaves of ferns ; but etiolin, which in darkness 
usually takes the place of chlorophyll, has a spectrum which does 
not materially differ from the chlorophyll spectrum, and is certainly 
not the same as that under discussion. Chlorophyll itself is held 
by many who have studied it to be a mixture of several distinct 
colouring matters. Professor Stokes * finds that the chlorophyll of 
land plants is a mixture of four substances, two green and two 
yellow, three of which are easily decomposed by acids. I cannot 
find any account of the experiments from which this conclusion is 
deduced, nor any description of the spectra of the several con- 
stituents, but it seems probable that the first band in the red is 
that due to the constituent which is not easily decomposed by 
acids. Mr Sorby f holds that the chlorophyll of land plants is a 
mixture of two substances, which he calls “ blue chlorophyll” and 
“yellow chlorophyll.” To each of these he ascribes a single 
absorption band in the red; but the diagram which accompanies his 
paper is so unsatisfactory that it is impossible to identify these 
with bands in the natural chlorophyll spectrum, though presum- 
ably the first band in the spectrum is that due to his “ blue 
chlorophyll.” But this first band is quite unrepresented in the 
trichosanthes spectrum, so that it seems a fair conclusion that the 
* Proc. Roy. Soc. xiii. 144 . + Proc. Roy. Soc., xxi. 451 . 
