THE ASSIMILATION OF CARBON BY AUTOTROPHIC PLANTS. II 127 



This beam remains always in the same place, no matter how much and what 

 part of the spectrum is obliterated. If a scale be inserted at SS^, we can 

 measure on it the wave lengths by observing Fraunhofer's lines, and are able 

 to compare exactly equal parts of the spectrum with each other, by taking light 

 of wave length, X = yoo-640 fxfx in one experiment, A = 640-580 /^/x in a second, 

 and X = 580-520 /jL/x in a third. The apparatus was later (Bot. Ztg. 1885) con- 

 siderably improved, but no further experimental results have been published 

 as a result of the use of the improved form. 



In using the spectrum several modifications are adopted in individual cases. 

 The macroscopic spectrum maybe employed with the gas-bubble method to carry 

 out eudiometric researches or estimate the amount of starch formed ; a micro- 

 scopic spectrum may also be thrown on the slide and the behaviour of an alga 

 may be studied by the bacterium method. Under certain circumstances, when 

 bacteria are introduced beneath the cover glass, aggregation of these at certain 

 regions of the spectrum gives an indication where the maximum activity occurs ; 

 more exact quantitative values may be obtained if one and the same object be 

 examined successively with light of dif- . 

 ferent colours, and each time, by lessening .^^j^^;;!^ 

 the slit of the spectroscope, a minimum "v^^^^^^^^::;^ /9^ 



light intensity may be obtained, at which <? ^"^^^^^^^^^^^r-^/^;^ 

 movement of the bacteria may still be \ \3S\^^^^# 



observed. Naturally the most effective \ \^>s^^^^^^'^ 



light through the narrowest slit will still \ \ % /^y ^ 



suffice for assimilation, and vice versa. \ \ ^^^^ F\ 



A comparison of observations de- \ %-^^^%\ 



rived from all the researches which <w^^^-~..„.^js\ 



have been made brings out the follow- /7 ^ ^""~"^^ 



ing points : — </ ^ 



1. Only light of wave length be- ^' 



tween 770 /a/a and 390 /a/a is conducive Fig. 26. Diagrammatic sketch of a spectrophore (after 



to assimilating activity in green plants ; L"- P^'pr^'- T^s:^J^i \=lo t A^A aI 



these are approximately the same rays diaphragm for cutting off extraneous rap; 6;converg- 

 Which are visible to us. '"^ '""' ' ^' P°''"°° °^ "^" P'^"' expenmented on. 



2. The assimilatory effect of different rays is unequal, but still not in such a 

 way that some only are active whilst others lying beyond these are quite inactive. 



If we express the wave length by abscissae and the activity of assimilation 

 by ordinates, we obtain a curve which does not in the least correspond with 

 the curve expressing the energy of sunlight obtained by Langley. 



On the other hand, no unanimity has been arrived at as yet as to the form 

 of the curve, especially on the question as to whether there is one maximum or 

 two, nor as to the exact position of the first and generally recognizable maximum 

 point. This point lies in all cases somewhere in the red or yellow, the second 

 maximum appears to lie in the blue region of the spectrum (compare the curves 

 in Figs. 27 and 28). The importance of the second maximum is emphasized by 

 Engelmann, who discovered it by the bacterium method. Since PFEFFER(Phys. 

 I> 334> Fig- 53) could not convince himself of its existence by the use of the same 

 method, and since the reasons which Kohl (1897) has brought forward more 

 recently for its occurrence are by no means convincing, we must assume that 

 the curve of assimilation has only one apex, somewhat as Reinke (Fig. 27) 

 has described it, and that the maximum point lies without doubt in the less 

 refrangible part of the spectrum. This conclusion, at once unexpected and 

 important, was already drawn from a consideration of their own observations 

 by the older investigators. Since the strongly refrangible rays are chiefly con- 

 cerned in the decomposition of silver salts, e.g. in photography, it had been the 

 custom to regard these as the chemically active ones. The data which have 



