most often along rock outcrops of wooded hilltop edges in the type of habitat most favored by the rat. Most often pilot black snakes have attempted to escape into crevices of the outcrop. These snakes are also skillful climbers and often have escaped by climbing out of reach along branches or even vertical tree trunks. On several occasions these snakes have been found on or beside woodrat houses, or have escaped into them. Over a seven-year period 143 pilot black snakes have been recorded, 53 of which were adults.

On September, 1948, a large pilot black snake found basking on a rock ledge, distended by a recent meal, was palped and contained a subadult female woodrat. On June 19, 1953, one of us, approaching a live-trap set under an overhanging rock ledge, saw a four-foot pilot black snake on top of it. The snake struck repeatedly at the rat in this trap, but was unable to reach it. At each stroke the rat would dash about the trap frantically.

These snakes hunt by stealth, and might catch woodrats by entering their nests, or by lying in wait along their runways, but are not quick enough to catch them in actual pursuit. Young in the nest would seem to be especially susceptible to predation by the pilot black snake. These snakes hunt by active prowling, either by night or by day, and much of their food consists of the helpless young of birds and mammals found in the nests. While only well-grown or adult pilot black snakes would be able to swallow an adult woodrat, any but first-year young probably would be able to overcome and swallow the small young. The female woodrat's habit of dragging the young attached to her teats as she flees from the house at any alarm must save many litters from predation by the pilot black snake. First litters of young, born in early March, are already well grown, and past the age of greatest susceptibility to predation before the snakes emerge from hibernation in late April or early May.

The timber rattlesnake is another potentially destructive enemy, but on the Reservation, and throughout much of its original range it is now relatively scarce. The genus Neotoma largely coincides in its over-all distribution with the genus Crotalus, of the rattlesnakes. For most kinds of woodrats, the larger species of rattlesnakes are among the chief natural enemies.

The timber rattlesnake has habitat preferences similar to those of the eastern woodrat. Of 30 timber rattlesnakes recorded on the Reservation over an eight-year period, all but one were at or near hilltop rock ledges in woodland. The woodrat is probably one of the most important prey species for the timber rattlesnake. Like the woodrat, the rattlesnake is mostly nocturnal in its activity. Unlike the pilot black snake, it hunts by lying in wait, striking prey which comes within range, and waiting for it to die from the venomous bite, rather than by active prowling. Therefore, it is probably less of a hazard to young in the nest than is the pilot black snake. Even young rattlesnakes too small to eat woodrats are potentially dangerous to them, as they may strike and kill any that come within range.

Commensals

Rainey (1956) listed many kinds of small animals that use the houses of the eastern woodrat and live in more or less commensal relationships with these rodents.

A situation unusually favorable for observing woodrats and their associates was discovered on the Reservation where, in July, 1948, two old strips of sheet metal, each covering an area of approximately 25 square feet, were used as shelter by a lactating female with three young. This was on a brushy slope just below an old quarry site. A rock pile and remains of an old rock wall were nearby. Woodrats had carried many sticks back under the metal strips, filling the spaces beneath their edges. There was a nest and a system of runways beneath the strips. In the following seven years this site was seldom deserted for long and was used by a succession of individuals. The strips of metal could be easily raised and then lowered into place with little disturbance. Because the situation was not entirely natural, the findings may not be typical of other rat houses. Animals found over a period of years beneath these metal strips include: several dozen each of the ring-necked snake (Diadophis punctatus), five-lined skink (Eumeces fasciatus), and ant-eating toad (Gastrophryne olivacea); several individuals each of cottontail (Sylvilagus floridanus), white-footed mouse (Peromyscus leucopus), short-tailed shrew (Blarina brevicauda), least shrew (Cryptotis parva), American toad (Bufo americanus), Great Plains skink (Eumeces obsoletus), pilot black snake (Elaphe obsoleta); and one each of bull snake (Pituophis catenifer), spotted king snake (Lampropeltis calligaster), red milk snake (L. triangulum), and timber rattlesnake (Crotalus horridus). The snakes which were potential predators on the rats seemed to be merely utilizing the shelter in these instances, but they may have been lying in wait for prey there.

Among mammals, the cottontail and the white-footed mouse are the most persistent users of the woodrat houses, especially those that are no longer occupied by the rats. On one occasion five white-footed mice were caught simultaneously in a trap set beside a house at the base of an osage orange tree. Subsequent trapping showed that this house was no longer occupied by a rat, but that the mice lived in it. Occupancy of such an old woodrat house by white-footed mice may continue long after abandonment of the house by the rat, even after the house has partly decayed and settled to a small part of its original volume.

Cottontails often have their forms under the edges of houses, either occupied or deserted. These situations offer protection overhead and on three sides. Abandoned houses having one or more of the entrance holes enlarged, as by predators breaking through the side of the house to gain access to the nest, are especially well adapted for occupancy by the cottontail. The rabbit may make its form inside the house structure.

The opossum, also, finds the type of shelter that it requires in abandoned houses that have had the entrances sufficiently enlarged. On various occasions opossums or their remains have been found in such old houses, and opossums released from live-traps have been known to seek shelter in abandoned woodrat houses.

At the old quarry on the Reservation woodrat sign was especially abundant. A wooden bin approximately seven feet square, used to store crushed rock before quarrying operations were abandoned, was inhabited by one rat. At the base of a rock crusher on the top of a bank a few yards from the bin was an accumulation of sticks and other debris brought by woodrats. A rock wall at the top of the bank between the crusher and the bin had many crevices providing shelter for the rats, and projecting rocks were littered with their droppings. In the spring of 1949 the bin and rock crusher were removed, but at least one rat continued to live in the rock wall. In the summer of 1951 several tons of corn ruined in the flood were dumped on the top of the bank above the wall. By autumn, Norway rats, either brought in with the corn or attracted by it, had taken possession of the wall, evidently displacing the woodrats, which were no longer present. Although this Old World murid rat is much different from the woodrat in habits, it seemingly can compete with it and replace it where habitat conditions are otherwise favorable for both.

Movements

The woodrat is dependent on the stick houses that it constructs for shelter. For each individual the house constitutes a home base to which it is attached, and about which its movements revolve. The area within which