84 MAMMALS OF PENNSYLVANIA AND NEW JERSEY. 



of palusiris was only based on circumstantial evidence, so far as appears in 

 his account of the affair. — Rhoads. 



Habits, etc. — I have the following notes from a muskrat trapper near 

 Greenwich who procured some of these water-rats for me on the tide marshes 

 of Cohansey Creek ; " Found on all tide marshes in this locality. Uses same 

 runs as the muskrat and gets caught in the same traps. Perfect specimens 

 hard to get, as they devour each other when fast in traps. Have seen them 

 in nests up among the reeds but believe the nests were built by marsh wrens 

 and confiscated by the rats. They live in holes in the big muskrat houses, 

 which are mounds of reed and mud on our salt marshes where the tide rises 

 and falls 5 feet. I have never seen them on the upland, but frequently found 

 them on oyster boats tied to the banks. They will walk a line to get on board, 

 and once there, will gnaw holes in the sails and build a nest. I think they 

 are principally vegetarians, but when hungry will eat meat or anything that 

 comes handy. They do not live in colonies but are scattered over the 

 marshes. They swim like a duck but do not move about in daytime." 



Mr. Bangs says of the rice-field mouse of Georgia ( Oryzomys ofjzivorus 

 (Aud. and Bachm.) ), "while perhaps preferring fresh and salt-water marshes 

 as its abode, it is by no means confined to such places. I have caught it in 

 dry, old fields, heavy swamps and hummocks, and even on sand hills." 



Bachman says of it : " It burrows in the dykes or dams [of the Georgia 

 rice fields] a few inches above the line of the usual rise of the water. Its 

 burrow is seldom much beyond a foot in depth. It has a compact nest at the 

 extremity, where it produces its young in April. There are usually 4 or 5. A 

 singular part of the history of the rice mouse is the fact that in the extensive 

 salt marshes along the borders of Ashley and Cooper Rivers, this species is 

 frequently found a quarter of a mile from the dry ground. Its nest is sus- 

 pended on a bunch of interlaced marsh grass. In this situation we observed 

 one with five young. It has no disrelish to the small Crustacea and mollusks 

 that remain on the mud at the rising of the tide. In an attempt at capturing^ 

 some alive, they swam so actively and dived so far from us, that the majority 

 escaped." 



In the latter part of March, 1902, I visited the Cohansey Creek marshes, 

 and secured several of this interesting species. They were only found in the 

 tops of muskrat houses scattered over the salt marshes at the head of ordinary 

 high tide. These houses invariably had underground connection with a tide 

 ditch by which not only the muskrat, but other tenants of the house, viz.,, 

 Oryzomys, Microtus and Sorex, could escape when the house was attacked 

 from without. The runways of the water-rat and meadow-mouse often com- 

 pletely riddled the whole structure of the muskrat's house and descended into 

 the marsh itself, making connection with the waterway exit of their host» 

 When it was torn to pieces, the nests of the smaller tenants of the muskrat 



