POCKET GOPHERS—GRINNELL 349 
Another example, is the race amargosae, restricted to the imme- 
diate vicinity of the permanent springs in the otherwise dry and 
alkaline valley of the “Amargosa River,” at Shoshone, Inyo County. 
This quite distinct form, in the perpallidus group of gophers, may 
appropriately be looked upon as a relict form from earlier times 
when conditions of moisture much more generally prevailed in the 
Great Basin territory, and when the dependent fauna and flora were 
correspondingly widespread. Amargosae is not the only mammal 
at Shoshone dependent, directly or indirectly, on the presence of 
permanent water; for there is (or was a few years ago) a distinct 
race of meadow mouse (Microtus) occurring around the same 
springs. Then the springs themselves contain a unique species of 
fish, residuary of a stock which evidence shows occurred widely 
in the general region in former times. 
No pocket gopher whatsoever has been found in the depression of 
Death Valley, into which the “Amargosa River ” empties. The lowest 
parts of the valley are too intensely alkaline to support any vege- 
tation at all; and such water as there is around the margins is 
either too alkaline, too impermanent, or else too small in amount to 
have permitted the persistence of gophers there up until the present 
time. 
Even such general areas as that indicated on the small scale 
map for the race mohavensis are not at all continuously occupied 
by pocket gophers; and examination of representations from dif- 
ferent parts of such a general area shows minor differences char- 
acterizing the separated colonies. For example, those animals 
living in the bottom lands of the Mohave River differ slightly 
from those inhabiting the somewhat higher table-lands surrounding 
the Providence Mountains, in extreme eastern San Bernardino 
County. 
The Colorado River is significant in our study, in that it has 
evidently long served as an impassable obstruction to the transfer 
of pocket gopher populations in either direction. The race 
albatus, occupying the delta and “second bottom” on the western 
side of the lower course of the Colorado River, is distinctly different 
in numerous respects, chiefly relating to the skull, from the form 
chrysonotus of the mesa lands on the eastern or Arizona side of 
the river. The actual distance apart of the nearest populations of 
these two species is not more than 2 miles in places, yet the inter- 
vening river and its “first bottom,” of ancient existence and large 
and permenent volume, has acted effectively in preventing the inter- 
breeding of adjacent stocks. Complete isolation accompanied by 
even slight difference of environment accomplishes much, granted 
plenty of time. 
20837—27——_24 
