246 University of California Publications in Zoology. (Vou. 5 
lead us to expect at least a slight peculiarity in the faunae of 
the various islands. From the geologic evidence now available 
it is certain that the islands of the Alexander Archipelago have 
been isolated since early Miocene time. Some small areas have 
been subjeet to depression and uplift since the isolation took 
place, but no extensive submergence has occurred. The whole 
region has been subject to a recent glacial epoch at which time 
much of the present topography was chiseled. A differentiation 
to a slight degree has taken place in some of the mammals on 
several of the islands, but in most eases it is so slight that it does 
not deserve recognition by a name. The species of limited dis- 
tribution arrange themselves in two sets on opposite sides of 
Chatham Strait which forms the water barrier between Admir- 
alty Island and the two closely associated islands of Baranof 
and Chichagof. This is doubtless one of the oldest channels in 
the archipelago, its depth being one hundred fathoms greater 
than any of the others. 
On Admiralty Island we find four species: Castor c. phaeus, 
Peromyscus hylaeus, Microtus admiraltiae, Mustela a. (actuosa?), 
which are not known to cross Chatham Strait. On Baranof and 
Chichagof islands we meet with Peromyscus sitkensis and with 
a very different form of Microtus, namely, sitkensis. The im- 
portance of Chatham Strait as a barrier is still further shown 
by the greater similarity of the fauna of Admiralty and the 
islands south of it, which also lhe east of the strait or its ex- 
tension. 
The mammal fauna of the three islands under consideration 
was derived chiefly from the coast and islands to the southeast. 
But a single form, Microtus sitkensis, has been received from 
the north and this form has already undergone some change. 
Microtus admiraltiae, Putorius c. alascensis and Sorex o. glacialis 
have doubtless been received from the Yukon Basin by way of 
the passes leading from the northeast. 
Generalizations based on the material at hand, however, can- 
not be considered conclusive, as the region is as yet only partially 
worked. This is well shown by the lack of complete duplication 
in the species taken by various collectors at the same localities 
throughout the region. 
