443 Prof. C. Timiriuzeff. [Apr. 30, 



layer of the colouring matter of practically the same concentration. 

 If the absorption due to the glass and to the solvent be determined 

 at the same time, the difference of the two quantities will give us the 

 amount of absorption attributable to the colouring matter of the leaf. 

 All the measurements were made by means of a particular form of 

 thermopile devised by me in 1870 and somewhat like the one recently 

 adopted by Rubens. The bars of the two metals are soldered in a 

 zig-zag, so that the two rows of junctions are turned to the same side. 

 AVhen one row is exposed to the sunlight, the other being sheltered by 

 a screen, we measure the effect of the direct radiation. On inter- 

 posing the chlorophyll solution we ascertain the fraction of sun- 

 light that is filtered by the solution, and lastly, by interposing the 

 same glass cell with the solvent we find the quantity transmitted by 

 both these media. The difference of these two last quantities corre- 

 sponds to the absorption by chlorophyll. The ratio of this quantity 

 to the one obtained in the first operation gives the fraction of direct 

 sunlight absorbed by the chlorophyll of a leaf. These three operations 

 may be reduced to two the measurement of the direct radiation 

 and that of the difference of the two liquids in a double-glass cell. 

 Here is the apparatus arranged for me by M. Pellin in 1889 

 (Plate 21, A, B). 



At the bottom of the cylindrical mantle lies the thermopile, protected 

 by a whole range of double slits from sudden variations of tempera- 

 ture, such as might be caused by wind (Plate 21, A). The double-glass 

 cell, enclosed in a metallic box furnished with two shutters, slides in a 

 groove in front of the outer slits (Plate 21, B). When we performt he 

 first operation, one of the slits is shut by the metallic box (both 

 shutters, of course, being also shut) ; when performing the second 

 operation the box is made to slide so as to shut both the slits, the 

 shutters being open. 



I proposed to call this apparatus a Phi/toadinonwter, since it gives us 

 the direct measure of the fraction of solar energy available to the 

 plant in the most important of its functions.* Here are some of the 

 values for a single leaf and for three leaves. Three is the maximum 

 number of leaves that may be superimposed with any profit to the 

 plant, the light reaching the fourth leaf being, as has been proved by 

 Professor Miiller, already deprived of its chemical activity. 



* This simple thermoscopic method is in most case; preferable to the far more 

 troublesome method of spectrophotometrj. 



