106 BOTANICAL FEATURES OF NORTH AMERICAN DESERTS. 
would take up the succession, the process of adaptation continuing until 
the present succulent and spinose types were reached. On the other 
hand, the new conditions might be met by sports or mutants, offsprings 
of the existing forms, which diverged from the parental types very 
widely, and which occur in such manner that some would be born which 
would be able to survive and thrive under the modified conditions of 
soil and climate. Gradual modifications by which a long series of forms, 
each slightly different from its immediate progenitors, appear to have 
been found among animals, but with plants no such series has yet been 
brought to light. These organisms, on the contrary, exhibit sports or 
saltatory derivatives, which now have been seen and recognized in a 
number of species. Such mutants are now occurring and we may predi- 
cate with certainty that they have occurred with normal frequency dur- 
ing the formation of the deserts of southwestern America. That many of 
them have survived to become constituents of the present flora is a sup- 
position fully in accord with the facts, and it seems reasonable to assume 
that the greater number of the units of the plant population have arisen 
in this manner. 
The recession of an inland sea or the drying up of a lake might lay 
bare great areas in regions of relatively small precipitation, with the 
result that the desert would be extended. Such new desert basin would 
be populated with plants by invading movements from the contiguous 
territory. If the adjoining areas were of equal aridity, the occupation 
of the new desert would be a comparatively simple matter in the succes- 
sion of formations which would follow the receding waters, although a 
series of this kind has never been actually observed. The occupation of 
such a desert might, however, require great departures from the types of 
vegetation accessible to it, and in such instances we are confronted with 
the choice between possible modifications as direct adaptations to the 
newly presented environmental conditions or to fitting mutations. The 
time requirement would favor the latter in most instances, and there is 
yet no evidence at hand to prove the direct response of plants to any given 
environment by structural adaptations which would be suitable to that 
environment. Thus the armature of desert plants is often thoughtlessly 
cited as an adaptation by which these forms protect themselves against 
the ravages of animals. The presence of spines undoubtedly operates 
to prevent a plant from being eaten by animals, but the action of the 
animals has in no wise induced their formation by the plant. Asa matter 
of fact, the fatality among desert plants by injury from animals is greatest 
in the seedling stage. For every prickly-pear that survives, tens of 
thousands of seedlings are eaten by rodents, and these seedlings are as 
unarmed as those of any other type. The natural selection actually 
operative in such cases is one that chooses among forms offered, but does 
not in any sense exercise a directing or guiding action on the development 
of such forms. 
