24 
AMERICAN JOURNAL OF BOTANY 
[Vol. io, 
Considering first the absorption data presented in table 2, it will be 
seen that the amount of water absorbed or lost by transpiration from the 
water cultures is very much greater, with few exceptions, than it is from 
the sand cultures. It is interesting to note that the sand cultures tran¬ 
spiring more water than the water cultures are characterized by having a 
high proportion of Ca(H 2 P0 4 )2 with only a small proportion of KN0 3 and 
MgS 0 4 . The five water cultures showing the greatest absorption (marked 
H ) are not the same as the five sand cultures showing the greatest absorption, 
with the exception of culture R5S2. The five cultures showing the lowest 
absorption are the same for both sand and water cultures with but a single 
exception. In the case of the control solutions, the quantity absorbed is 
practically the same for both sand and water cultures. However, the total 
absorption for the control solutions is very much lower than that for the 
five sand and water cultures marked H. These transpirational data, 
therefore, show a marked superiority of type III solution over Shive’s 
“best” solution (R5C2—1.75 atm.). 
Three of the five water cultures showing the greatest absorption also 
have the highest fresh and dry weight of tops, while four of the five sand 
cultures marked H reveal this relation. Thus, in general these data uphold 
the conclusion of Livingston (6) and of other writers that the amount of 
transpiration appears to be as good a criterion as the final dry weight for 
judging the comparative growth obtained in different solutions. 
A comparison of the fresh and dry weight of tops produced by the 
sand and water cultures shows that the highest yields favor the water 
cultures with the exception of those cultures having high concentrations of 
Ca(H 2 P0 4 )2. But the sand cultures favor the greatest growth of roots, as 
shown by a comparison of the dry weight of roots grown in the sand and 
water cultures. By comparing the dry weight of tops produced by the 
high-yielding sand and water cultures with the average dry weight of tops 
produced by the control solution (Shive’s R5C2), it will be seen that the 
water culture IIIR2S1 is superior to Shive’s R5C2 by 144 percent, while 
the sand culture IIIR3S3 is superior to Shive’s R5C2 by 103 percent. 
When the dry weight of roots of IIIR3S2 of water cultures is compared 
with the dry weight of roots of Shive’s R5C2 there is a difference of 30 
percent. These differences are great enough to overbalance any possible 
errors in plant variation, and furnish conclusive evidence that the optimum 
proportion of salts in type III solution gives better growth for wheat than 
Shive’s (R5C2—1.75 atm.) solution. 
In comparing the data given in table 4, it is clear that there is very 
little difference in the total dry weight (tops and roots) of the sand and 
water cultures. On the total dry-weight basis, the average increased yield 
of the best sand and water cultures of type III solution over the average 
yield of Shive’s solution is 131 percent. 
As a means of facilitating comparison of the salt requirements of wheat 
