SCIURID.E— SCIURUS COLLI2EI. 739 



pure white; no lateral line. Tail long and very broad, vertebra? alone 

 nearly equal in length to the length of head and body; above, deep black, 

 broadly edged with whitish or mixed white, black, and pale fulvous ; beneath 

 with or without a broad central area of bright tawny, bordered by a broad 

 subterminal band of deep black, and broadly edged with white. Sometimes 

 the central bright tawny zone can be faintly seen through the hairs of 

 the dorsal surface, as in the Arizona specimens ; again it is merely pale ful- 

 vous, striped with narrow lines of dusky, the central fulvous area being visi- 

 ble only from the lower surface. 



This species is thoroughly distinct from 8. carolincnsis, which it some- 

 what resembles in color, as it also apparently is from every other North 

 American species of Sciurus. It was described by Richardson, in 1839, from 

 a specimen from San Bias, on the west coast of Mexico, in latitude 21° 34', 

 where Mr. Collie found it common. My Mazatlan specimens are from near 

 the same locality (about one hundred miles farther north), and agree with 

 Richardson's original description. Dr. Gray's S. collicei, "var. 2", with bright 

 rufous sides and limbs and white belly, I refer with little hesitation to S. 

 boo/hie, while his "var. 1'', from the west coast of South America, with "the 

 under surface yellow", he considers the same as Ogilby's S.variegatoides and 

 his 8. griseocaudatus, both of which I refer to the S. hypojiyrrhus of Wagler. 

 The " Macroxus collicei' 1 '' of Gray seems to be only in small part referable to 

 the S. collicei of Richardson. 



I refer to this species also the S. arizonensis of Coues, described origin- 

 ally from a single specimen obtained at Fort Whipple, Ariz. Two other 

 specimens from Arizona, collected later by Mr. F. Bischoff, agree essentially 

 with Dr. Coues's specimen, except that they are somewhat larger. One of 

 them, however, has the brownish dorsal area less strongly developed than in 

 the others, and has the lower surface considerably varied with irregular 

 patches and streaks of pale yellowish-rufous, thus showing a tendency to the 

 acquisition of a rufous belly, so common a feature among the Squirrels of 

 Mexico and Central and South America. Dr. Coues's specimen, though killed 

 in December, was evidently not full-grown, being, as described by him, of 

 about the size of S. carolinensis. The other specimens, one of them a female 

 that had recently nourished young, are much larger, and indicate a species 

 fully as large as 8. aberti, if not even larger. 



The coloration of & collicei, at first sight, seems to bear a close resem- 



