142 BOTANICAL GAZETTE [FEBRUARY 
A very slight increase in number and length of branches is to be observed. 
Fig. 4, from a soil containing 10,000 parts per million of manure, shows 
numerous well-developed branches, while from a soil with 40,000 parts per 
million of manure, as shown in jig. 5, the branches have increased so 
markedly in number and extent as to make up by far the greater part of 
the system. 
From these facts it is seen that, for this series of soils the variation in 
growth of tops is correlated with the number and length of lateral roots. 
The water content of all the soils was the same, so that the variations in 
—Wheat roots grown in Ta- 3-—Roots apie in Takoma 
Fic. Fic. 
koma ot untreated. soil with poe p-p.m. manu 
growth cannot be related to this factor; therefore they must be connected 
with some unknown condition in the soil itself, a condition which is related 
_ to the amount of manure present. 
The comparative anatomy of these roots was investigated, both by 
hand and paraffin sections, with the result that in the poor soils the main 
roots have a strong tendency to swell by direct enlargement of the cortical 
cells, without increase in the number of these cells, while in the better soils 
this tendency is not nearly so marked. Very old wheat roots from autumm 
stubble in the field show this balloon-like enlargement of the cortical cells 
to a still greater degree. This is apparently a phenomenon of age, sug- 
gesting that roots in the poor soils age more rapidly than in the better 
ones. It was also found that the zone of root hairs, which normally has 
its lower limit s-1o™™ from the root apex, extends in the poor soils to 
within 1-3™™ of the tip. The outgrowth of root hairs from the piliferous 
layer may also be related to the age of the cells; as is well known, these 
