1915.] Allen, Review of the South American Sciuride. | | 239 
Specimens examined, 18.— Colombia: Medellin, 2 (Br. Mus. and Nat. 
Mus.); “New Granada,” 2, type and paratype of S. gerrardi (Br. Mus.); 
Baudo, 2 (Am. Mus.); Bagado, 1 (Am. Mus.); Juntus de Tamané, 1 (Am. 
Mus.); Sisto, 1 (Field Mus.); Rio Sipi, Chocé, 3 (Br. Mus.); “ Valdivia, 
Lower Rio Cauca,” 6 (Br. Mus.). [See Addenda, p. 308.] 
Remarks.— Sciurus gerrardi, with its subspecies, has a range which 
extends along the Pacific coast from central Ecuador north to Panama, and 
east in the coast region of northern Colombia to Lake Maracaibo. It has 
become differentiated into a number of local forms, the extremes of which 
are very unlike, but which completely intergrade. One extreme is S. ger- 
rardt morulus of Panama, with the upperparts grizzled ochraceous and 
black instead of red and black, and the black dorsal area obsolete or greatly 
reduced; the other is S. gerrardi zulie with the:sides bright red and the dark 
dorsal area very broad and glossy black. Both intergrade with true ger- 
rardv of the west coast region of Colombia. 
The nomenclature of this group is much involved, several names having 
been more or less current for the typical form. The earliest name that has 
been applied to squirrels of the gerrardi group is Sciurus variabilis Geoffroy 
(Guérin’s Mag. de Zool., X, 1832, Classe 17°, Mamm., pl. iv and text), 
which, based’ on three specimens from “ Amérique,” is not satisfactorily 
identifiable, and therefore not available. Most of the description applies 
very well to white-bellied specimens of typical gerrardz of the present paper, 
except that the tail is not “toujours beaucoup plus noire & la face postérieure 
qu’ a Vantéreure,” nor does the description agree with the colored plate. 
In the plate the lower back and rump are represented as red like the thighs 
and flanks, instead of being black as in the red and black forms of gerrardt. 
The species was based on specimens presented, with many others, by M. 
Plée, and presumed to have come from America, “car plusieurs de ces 
objects appartenaient 4 la Colombie, plusieurs aux Antilles et quelques- 
uns aux Etats-Unis.” As indicative that the squirrels may have come 
from Colombia, the same collection contained a monkey which was de- 
scribed by Geoffroy in the same volume (J. c., pl. vii and text) as Stentor 
chrysurus (= Alowatta seniculus), which he states is known to occur in 
Colombia, and to be common in the valley of the Rio Magdalena (“vallée 
de la Madeline’). Some years since I discussed at length (Bull. Amer. 
Mus. Nat. Hist., XII, 1899, p. 212, and cbid., XX, 1904, pp. 434, 435) 
Geoffroy’s Sciurus variabilis with reference to the large red-backed, white- 
bellied squirrel of the Santa Marta district, and reached the conclusion: 
“Indeed, it seems now safe to assume that the real type locality of S. 
variabilis is the Magdalena River of Colombia, at some point quite remote 
hs) 
from the coast.... 
