MAMMALS OF THE MEXICAN BOUNDARY. 411 



Luis Mountains and scorched plains -impassable harriers to it. This 

 mouse is probably of Mexican origin, and, south of the United States 

 border, these two species (P. torniUo and P. arizonse) may come to- 

 gether and intergrade, in which case P. torniMo would become a sub- 

 species of l } . arizonse. The size of the two species is about the same, 

 though P. tomillo appears to he the stoutest, and has a smaller ear, 

 more massive skull, and larger teeth (see figs. 85 and 87), with a pecu- 

 liarity of the anterior upper molar, in which the marginal intermediate 

 subsidiary cusp is obsolete; and the coloration is much paler. 



1 have compared the specimens of this species in our collection w iih 

 the type of Sitomys americanus arizonx of Allen, and find that they 

 arc unquestionably the same. 



Habits and Inail distribution.— In 1885 Doctor Merriam sent me a 

 "delusion" mouse trap, which I set under some driftage beside the 

 Verde River, Arizona, on September 23. The morning after it was 

 missing, a curious raccoon having carried it off; but I followed the 

 tracks of the raccoon and luckily recovered the 

 trap, in which I caught two Arizona wood mice 

 the next day. This was the beginning of my 

 systematic mammal trapping and of my ac- 

 quaintance with the present species, which I sub- 

 sequently caught quite often in the cottonwood 

 and willow groves bordering the Verde River. 

 Some were also caught in the post hospital at FlG " 87 ._ PEROMYSCUS " ARI . 

 Fort Verde. Two species of wild mice infested zon.e. crowns of molab 

 the hospital— the present speciesand Peromyscus ^ee's™ SEMES; 

 in miens anthonyi. Both were taken at the same 



time, in this same trap, set in a closet, in the hospital. On October 

 10, 1885, I found a nest of Peromyscus arizonx in the hospital dis- 

 pensary, in a drawer that was opened many times each day. The 

 aesl was composed of a small quantity of cotton and shreds torn 

 from muslin bandages. It was soft, nicely felted, and altogether a 

 snug little dwelling, in which I found a single mouse, just born, of 

 bright pink color, and destitute of hair. This is the common mouse 

 of the Verde Valley, ranging between 3,000 and 4,000 feel abovesea 

 level, in which range it associates with Peromyscus eremicus anthonyi. 

 The higher mountains of this region are the habitat of Peromyscus 

 sonoriensis rufinus. 1 did not find any form of the Peromyscus boylii 

 group north of the Gila River, in Arizona. 



At the Morris ranch, on (dear Creek, near Fort Verde, on February 

 3, 1887, I discovered a nest of this mouse under a big stone in an alfalfa 

 held. 1 put them in a box, together with some pocket -gophers 

 (Thomomys). The gophers ate them all up, beginning al their tails. 

 On May 11. 1887, I saw a mouse running about under the cotton- 



