the species is suitable for
occupancy. In these mice, as in other microtines (Elton, 1942; Piper,
1909), there are seasonal, and irregularly multiannual fluctuations in
population density, which sometimes are extreme. Consequently the mice
at some times seem to be absent from suitable habitats, and at some
other times occur there in amazingly large numbers.

Because the species is broken up into partly isolated, or at times
completely isolated, colonies or local populations it may be supposed
that various evolutionary forces such as selection and random genetic
drift operate to foster variation. The degree to which racial
distinction is attained may depend upon these forces and the time
available. In _Microtus montanus_ in the eastern Rocky Mountains the
degree of subspecific distinction is not great.

The study here reported upon is based on 1,187 specimens of _Microtus
montanus_ from Wyoming, Colorado, Idaho and Montana, and on work in the
field. I spent approximately four months in the field in this area, in
the summers of 1950, 1951, and 1952. The specimens studied were arranged
according to localities and the larger series were compared
statistically. Each of two series, totaling 136 specimens, was studied
intensively to ascertain the kind and range of variation within single
populations. Twenty-seven measurements, various proportions based on
these measurements, and differences in color were analyzed. Fifteen
characters, judged to be most significant, were selected for use in
comparing all series. In addition, certain characters that can not be
expressed easily by measurements, such as inflation of the auditory
bullae and the curvature of the zygomatic arch, were observed. The
studies by A. B. Howell (1924) of variation in _Microtus montanus
yosemite_ Grinnell in California and those by O. B. Goin (1943) of
_Microtus pennsylvanicus pennsylvanicus_ (Ord) were useful. The external
measurements are from the collectors' field labels. The measurements of
the skull all were taken with dial calipers reading to a tenth of a
millimeter. The anteroposterior measurements of the skull all were taken
along the shortest line between the points specified below and are not
necessarily along a line parallel to the long axis of the skull. These
measurements were taken on the left side of the skull whenever possible.
Some of the skulls are damaged and therefore some measurements could not
be taken and are omitted in the computation