350 ANNUAL REPORT SMITHSONIAN INSTITUTION, 1926 
A puzzle, at first glance, is offered by the occurrence of the two 
forms, albatus in the delta silts below Salton Sea, and perpallidus 
on the floor of the northwestern end of the same (Colorado) desert. 
Both races are restricted to fine-textured soil in the vicinity of 
water or where at least some underground seepage permits the proper 
growth of salt grass and other plants whose stems or roots are 
sought by gophers. Adbatus follows the western distributaries of 
the Colorado River over the delta, and of late has found wonderfully 
favorable conditions for itself, with resulting enormous spread and 
multiplication of its numbers, on the irrigated lands of the Imperial 
Valley. Perpallidus occurs chiefly at the mouths of permanent 
streams coming down the cafions out of the San Jacinto and Santa 
Rosa Mountains onto the floor of the northwestern end of the 
Colorado Desert, known locally as the Cahuilla Valley. Why 
should these two races be as distinct as they are from one another 
when the floor of the general desert area they occupy is continuous, 
and only about 150 miles in greatest length ? 
Not long ago, even measuring in years, the northwestern arm 
of the Colorado Desert was occupied by “ Blake Sea,” of which 
the present Salton Sea is the residuum. More remotely yet, the 
Gulf of California extended continuously up from its present 
terminus clear through the Cahuilla Valley; and to-day the floor 
of that desert is in many places covered with shells of ocean- 
inhabiting mollusks, and shore lines at sea level are to be seen 
along the bases of the mountains which rise abruptly on either 
hand. The rapidly accumulating silts from the Colorado River 
filled in the depression opposite its mouth and cut off the basin 
of Biake Sea; and the arid climate resulted in the disappearance 
of the waters of that sea by evaporation. But completion of this 
cutting off of Blake Sea and the evaporation of its waters was of 
quite recent occurrence. We can, I think, look to the former long 
and complete separation of the gopher stocks as resulting in per- 
pallidus and albatus, respectively, during the period that the waters of 
Blake Sea, at sea level, lapped the steeply rising rocks on either side, 
impassable to gophers. Complete isolation was thus afforded for 
the initial bottom land stocks of gophers at the northwest and to 
the southeast. With the retraction of the shores of Blake Sea, 
there is now no barrier between perpallidus and albatus, save as 
is comprised in unwatered tracts; and these are getting smaller 
with the spread of irrigation. It will be interesting to see what 
happens when and where the two gopher populations meet. 
An additional case of isolation by desert conditions is that of 
operarius at Owens Lake, also a member of the perpallidus group. 
Operarius is a quite distinct form, so distinct with respect to shape of 
