1910. | On Vegetable Assimilation and Respiration. 387 
very soundly established, as it is derived from the mean of six concordant 
hours of assimilation. 
Each experiment of this kind is a whole day’s work, but a considerable 
number have been put together in the years 1905, 1907, and 1908. The 
series dealing with the relation between assimilation and the CO>:-content 
of the liquid is fairly complete, and the results will be published in the next 
paper of this series. 
(d) Blank Expervments—We may conclude this section with a reference 
to one of the blank experiments made to test the accuracy and uniformity 
of the estimations. The course of this experiment is represented graphically 
in fig. 2. 
"MOTYNOS JO *9°D ONT 
jo queyvatnbe [OH OT/N ‘0°0 
iOam. Il noon | 2 3 4 5 6 7 p.m. 
The procedure was exactly like that just described, but the chamber was 
set up empty, with no plant in it. The COs-solution prepared for the 
experiment was about quarter saturated, of such a strength that 200 c.c. of 
it equalled about 40°6 cc. of N/10 HCl. The current was started at 
9 o'clock (see fig. 2), and an estimation of the effluent was made every hour 
till 7 p.m. At the moment of the end of each collection of the effluent 
a sample of the affluent also was taken and analysed. Both sets of results 
are plotted in the diagram. Here the ordinates directly represent the 
equivalent in cubic centimetres of N/10 acid. The series of heavy horizontal 
lines represents the effluent collected in the pipettes D or Hi; as each 
contains 200 c.c., or the flow during 40 minutes, the lines drawn are of 
a corresponding time-length. The small crosses are the analyses of the 
affluent placed at the times when they were withdrawn in the pipette W. 
In this experiment the CO: solution was presumably not well mixed or the 
CO, generator at fault, for the strength of the solution rose steadily from 
9 AM. till noon. From noon to 5 P.M. uniformity was maintained, after 
2F 2 
