REVIEW AND DISCUSSION. 131 
that the creosote-bush is more sensitive than the mesquite to deficiency 
of aeration, and-accordingly shuns the flood-plain, and on the other hand 
that the mesquite is less capable than the creosote-bush of living with 
an insufficient water-supply, and on this account makes no advances 
upon the Larrea slope. That they occur together in places on Tumamoc 
Hill indicates that in such spots satisfactory conditions both of aeration 
and of water-supply are present. 
LOCAL MOVEMENTS. 
The discussion thus far has related chiefly to statical rather than 
dynamical phenomena. Desert plants have been referred to their places 
and associations very much as if these were permanent as regards place 
and composition; but here as elsewhere, it is necessary, as a matter of 
fact, to conceive of movements, both of associations and of individual 
plants, as continually taking place, resulting in a change of place of the 
association and a change of its composition through the introduction or 
loss of constituent species. 
Such movements are ordinarily too slow to easily admit of direct obser- 
_ vation, but they are readily demonstrable by suitable means. Com- 
parative observation, supplemented by exact mapping of selected areas, 
makes it possible to gain a satisfactory view of the course of movement 
at the present time, and the maps, as a permanent record, afford a body 
of reliable data by means of which, in the future, movements taking 
_ place in intervening years may be recognized. A review of a few cases 
will serve to illustrate. 
The most carefully observed case thus far is that of alfilaria (Erodiwm 
cicutartum), the path of which from California into southern Arizona has 
been clearly traced. It is plain, as its history (p. 52) and a reference 
to the map (plate 20) indicates, that it reached the northern part of 
the Laboratory domain by the way of the Santa Cruz Valley, and has 
advanced in the course of a few years to points some rods beyond the 
Laboratory building. ‘The same is true of foxtail (Hordeum murinum), 
which seems to have had an essentially identical history, but to have 
entered the domain more recently, and, though well established, has 
spread less widely (plate 22). Both have followed wagon tracks to a 
great extent, as if brought by teamsters, and both are already occupying 
some areas so thickly that it is plain from simple inspection of adjacent 
areas that they take possession of ground from year to year that but 
for their presence would be occupied by other annuals. 
Very different are the cases of some of the perennials that have recently 
found a place on Tumamoc Hill. A single individual of Cercedrum 
torreyanum stands in the gulch near the Laboratory, where, in all prob- 
ability, the seed was brought not long ago from the wash, less than a 
quarter of a mile away. The ironwood (Olneya tesota), two or three 
