132 BOTANICAL GAZETTE [AUGUST 
were taken up from the mesa east of Tucson, January 13, 1904. 
The soil where they were dug, though rather light, is relatively deep, 
eee 
Fic. 5 —Seedlings of Covillea from the mesa 
east of Tucson, Arizona. 
The Position of this particular plant wit 
It Is instructive, and May account in part 
and it is noticeable that 
while the lateral roots 
had attained only a slight 
eveloy t, the tap-root 
had reached a depth of 
over 31°™ in the one case, 
and upwards of 53°™ in 
the other. A much older 
plant, taken up from the 
plain northward of the 
Desert Laboratory,where 
the soil is underlaid by 
rock, shows a strong 
development of secondary 
roots, and the tap-root, 
instead of continuing 
vertically downward, 
turns off at a small angle 
from the horizontal, but 
finally, at 80°™ distance 
from the main axis, turns 
directly downward. The 
lateral roots in their tum 
continue near the surface 
only a short distance, and 
then, in spite of the rocky 
nature of the substratum, 
turn downwards, reach- 
ing 40 to 45°™ in dep 
where they were broken 
off, though probably 
extending to a consider- 
ably greater depth (fig-6)- 
h respect to those aroun 
for the distribution of 16 
