194 BOTANICAL GAZETTE [SEPTEMBER 
The photograph for this series is given in jig. 20, and the curves 
in fig. 21. The three criteria arrange the cultures in the same order, 
although by transpiration the difference between nos. 2 and 3 is 
emphasized. Transpiration per unit area increases somewhat with 
the fertility of the solution; while 
that per gram is practically uni- 
form throughout the series. The 
former of these ratios thus appears 
to vary in the opposite direction 
from that shown in certain of the 
soil series already described. It 
may be that the presence of a 
superabundance of water about the 
roots raises the transpiration per 
s————+ gram and per unit area above what 
it would be if the roots were in soil. 
CONCLUSION. 
From the experiments which have been described the conclusion 
seems evident that total transpiration of wheat plants grown in vari- 
ous media is as good a criterion for comparing the relative growth in 
these media as is the weight of the plants. That these two criteria 
vary generally with the weight and area of the leaves gives the expla- 
nation for this conclusion. The facts are made clear that, for the 
types of media investigated at least, the amount of transpiration is 
practically a simple function of the leaf surface; and that this latter 
varies quite uniformly with the leaf weight, which in turn varies with 
the weight of the entire tops. Thus total transpiration appears 
be a measure for the growth of the plant. With some series there 
seems to be a slight variation in the ratios of transpiration 0 
weight and to area respectively, according to the nature of the 
medium; but these variations are so small when compared 
with those of their component terms, and lack uniformity in 
the different series to such a degree that they are practically 
negligible in the comparison of the cultures. It thus appeals 
that the nature of the soil or solution in which the roots ay 
grown has little or no influence on those structural and physt- 
