f we had been working with a stable population. In
most instances the spans of our records represent only small parts of
the life spans of the individuals involved. Nevertheless, our records
emphasize the potentially greater longevity of the woodrat as contrasted
with the various smaller rodents living in the same area. Of several
thousand individuals of the genera _Mus_, _Zapus_, _Reithrodontomys_,
_Peromyscus_, _Sigmodon_, and especially _Microtus_, none is known to
have survived so long as two years, and only a few individuals are known
to have survived so long as one year after being marked.

_Summary_

Plant succession resulting from land use practices created habitat
conditions especially favorable for woodrats in the late nineteen
forties in northeastern Kansas, and particularly on the University of
Kansas Natural History Reservation. With protection from prairie fires,
woody vegetation had encroached onto areas that were formerly grassland,
and, later, fencing against livestock permitted dense thickets of
undergrowth to develop. In this region the woodrat usually lives in a
forest habitat, and requires for its house sites places that are
especially well sheltered, as in matted thickets of undergrowth, root
tangles exposed along eroded gully banks, hollow stumps or tree trunks,
bases of thorny trees with multiple trunks for support, thick tops of
fallen trees, or, especially, rock outcrops with deep crevices.

At the time of their maximum population density in or about 1947,
woodrats probably averaged several per acre on the woodland parts of the
Reservation. In the autumn of 1948, 17 were caught on the ten-acre tract
of woodland that was live-trapped most intensively. By then, however,
the population had already undergone drastic reduction, as shown by the
fact that there were many unoccupied and disintegrating houses
throughout the woodland. While the time and manner of mortality was not
definitely determined, circumstantial evidence suggests that the
downward trend began in early March, 1948, when record low temperatures
and unusually heavy snowfall coincided with the time when parturition
normally occurs. The rigorous weather conditions then may have been
injurious, not only to the newborn litters but to the females comprising
the breeding stock. Nevertheless, the population remained moderately
high through 1948, but by early spring of 1949 more than three-fourths
of the adults and subadults present in la