REVIEW AND DISCUSSION. loo 
the most varied character. The giant cactus, in its close association 
with the palo verde and other deep-rooted perennials, affords a good illus- 
tration of the mutual accommodation of two or more different plants 
which are capable of growing advantageously side by side by reason of 
differences in distribution of their root-systems, while other cases that 
have been cited (pp. 52-56) indicate something of the extent to which 
competition prevails between annuals and perennials, between different 
species of annuals, and between individuals of the same species. It 
would be difficult to imagine a case of greater complexity than is pre- 
sented by the mutual relations of the plants of the Laboratory domain, 
one in which generalization is less safe, or in which the necessity of study- 
ing every species by itself is more imperative. 
The associations here represented are in unstable equilibrium, not only 
as to their constituents, but also as to their place. The inevitable topo- 
graphical shifting observed in a region where erosion is both rapid and 
long-continued necessarily results in advance or retreat, or both, on the 
part of all the association groups. As the process of base-leveling pro- 
ceeds, the associations of the hill slowly give way before the upward 
movement of the creosote-bush association as it advances step by step, 
corresponding with the upward progress of the long slope to which it 
belongs. Most plainly, too, is seen the forward movement of the palo 
verde-catclaw association as the wash and its branches work their way 
slowly toward the mountains. Between the slope and the flood-plain is 
a line along which the relative positions of the associations on either side 
are determined by the preponderance of erosion or deposition, the creosote- 
bush characteristically belonging, as we have seen, to areas of erosion, the 
mesquite to those of deposition. 
Thus the movements of the associations as they are taking place at 
the present time are comparatively simple. The tendency of the topo- 
graphic changes that have been described is manifestly the building up 
of the long slopes at the expense of the hills and mountains, which now 
rise above them, and with this is necessarily associated the succession 
indicated in which the plants of the slopes become more and more domi- 
nant and the creosote-bush association ultimately represents for them 
the climax type. It is to be borne in mind, however, that in this region 
profound alteration of topographic features may at any time follow a 
new uplift or other phase of geologic activity, and accordingly we are in 
no position to predict changes which may greatly modify or retard the 
orderly succession which has been described and which is now actually 
going on in our sight. 
It is not possible, with present knowledge, to reproduce with certainty 
the history to this time of successive associations on Tumamoc Hill itself. 
The present stage exhibits phases that are more or less transitory, but 
it seems highly probable that the essential features of its vegetation have 
