LOCAL DISTRIBUTION OF SPECIES. 55 
often stated that with desert plants the struggle for existence is with 
their physical environment rather than with each other, but it is certain 
that within the area under consideration severe competition prevails 
between many of the plants which constitute its vegetation. For the 
purpose of observing the results of this process upwards of 20 different 
areas, each including from 1 to 4 square meters, were staked off on the 
Laboratory domain in the spring of 1907, and the relations of the plants 
included in them carefully noted, but until both observation and experi- 
ment have been carried much farther it will be impossible to give a satis- 
factory account of competition as it takes place between the plants here 
represented. The following notes will serve to indicate certain obvious 
relations worthy of record: 
(1) On the flood-plain, in ground where there is abundance of water, 
there.is not the slightest doubt that certain weeds, notably Malva parvi- 
flora, have successfully competed with various other species. In places 
this plant covers the ground so completely with its rank growth that 
other plants fitted to the same habitat are entirely unable, in ground 
occupied by the Malva, to reach maturity. In this case the unsuccess- 
ful competitors are probably destroyed by the dense shade of the aggres- 
sive weed, though it is likely that they also suffer directly through lack 
of food materials that it has appropriated. 
(2) Annuals, which after the winter rains grow in great luxuriance 
on Tumamoc Hill, exhibit in a striking manner the effects of competition 
between individuals of the same species and between different species. 
The phenomena, stunted growth, failure to produce seeds, ete., the same 
here as elsewhere, are so familiar as to require no detailed description. 
(3) Everything in the position and mode of growth of Hilaria cench- 
rovdes and H. mutica on the hill goes to indicate that we have here a case 
in which a perennial advances relatively slowly into ground previously 
occupied from year to year by annuals and holds it against their future 
occupation for an indefinite time. From what is seen elsewhere there 
is no doubt that this will be the history also where the Bermuda grass 
has gained a foothold in a few places near the Laboratory. 
(4) Study of the root-system shows a complicated set of relations in 
which a given plant may be in direct competition with certain species 
and not with others. Thus Cannon (p. 64) finds that Cereus giganteus, 
growing with Larrea and Parkinsomaa, is, to a great extent, relieved of 
competition on their part by the deeper penetration of their roots, while 
it comes into direct competition with various annuals, the roots of which 
occupy the same or nearly the same horizon. 
(5) But while it is evident that competition has had much to do with 
determining the preponderance of species in the various associations, it 
is also clear that in general it here stands in an altogether subordinate 
relation to adaptation. As a striking illustration, the zone of creosote- 
