18 
shoots, the common heterophylly in the horseradish, sassafras, 
and the mulberries all show that plants must be studied through- 
out their various stages of development and through the seasons. 
Variation in garden plants (in leaf, in flower color, shape, and 
arrangement) are common and are probably more important than 
they are usually considered. From similar variations reported 
from widely distributed points or at widely separated intervals 
the conclusion is drawn that a new form, which has appeared 
at various times and which because of the nature of the variation 
is incapacitated from reproducing itself by seed, would from this 
very fact constitute an ideal illustration of repeated mutation, 
since а hybrid origin of the individuals which appeared later, 
is excluded.—J. B. 
There has long been the impression that desert plants must 
have very deeply penetrating root systems, quite oblivious of the 
fact that in most desert regions the soil water lies so far below 
the surface that many if not the majority of plants would be. 
quite unable to develop roots capable of reaching it. Dr. 
Cannon* in a recent paper has shown that there is a great diver- 
sity in the root distribution of such forms. Those which grow 
in the flood plains of the rivers, as for instance the mesquite, 
may indeed have fairly deep roots, for the water table in such 
localities is within reach even in a desert. Those, on the other 
hand, which grow on the detrital slopes are much more likely 
to have shallow root systems which extend over a large area. In 
even the larger cacti, for instance, the tap-root is a negligible 
quantity except perhaps for anchorage and the superficial 
laterals spread out for a long distance. The water which the 
plants avail themselves of is the surface moisture which comes 
from the seasonal though brief and scanty rains of the region. In 
Tueson, Arizona, where there are two short rainy seasons, one in 
winter, thé other in summer, the annuals show a difference in the 
development of their absorbing systems which is apparently due 
to the relative difference of air and soil temperatures at those 
* W. A. Cannon, Root Habits of Desert Plants. Carnegie Institution A^ Wash- 
ington. Publication 131, pages 1-96, Pl. 1—23; fig. in text 1-17, Mar. 2 I 
