214 MB. R. I. POCOCK ON THE 



kindly arranged for two male examples of the species, preserved 

 in alcohol, to be sent to me *. From these two specimens I find 

 that Tullbei'g was perfectly correct, and had not, as I supposed, 

 made an error of identification. The glans penis in the Alaska 

 specimen is slender and tolerablj^ evenly tapering, with the 

 orifice terminal and just beneath the apex. The upper surface 

 has a median groove, and a pair of crests of minute serrula- 

 tions running along it, and the sides are finely striolate. The 

 distal portion below the bend is somewhat flattened, and provided 

 on each side with a narrow laminate expansion. In the New 

 York specimen, assigned to the subspecies loqimx, the general 

 form and sculpturing of the penis are much the same, except that 

 the striolse are coarser, the intervening spaces forming fine ridges, 

 a difference possibly due to greater contraction of the epithelium, 

 and the distal portion beyond the bend is not so evenly tapering, 

 but shows an excrescence on each side. Nevertheless the two 

 are of essentially the same type, and in both the distal portion 

 is soft and flexible without a vestige of baculum that I can detect. 

 The structure of the penis sharply difierentiates hudsonicus from 

 the Squirrels with the vulgaris type of penis and baculum, and 

 compels the admission of Tamiasciurus as a well-defined genus 

 without any near allies in the family. (Text-fig. 18, B.) 



On the other hand, although Thomas stated that the baculum 

 of the Bornean species Rheithrosciurus macrotis is like that of 

 Sciurus vulgaris, he tells me that he is so impressed by the 

 improbability, on geographical and other grounds, of the two 

 species being alike in that respect t, that he considers confirmation 

 of the fact necessary before definite acceptance be accorded it. 

 There is always, of course, the possibility of confusion between 

 the bones by taxidermists during the cleaning process. 



The bacula of the S. vulgaris group are relatively well 

 developed, but vary in size with the species, from about 14 mm. 

 in *S'. niger to 8 mm. in S. ingrami. 



The foregoing account shows that the Paltearctic and American 

 Squirrels fall into two categories by the structure of the jpenis 

 and baculum : — 



1. Sciurus, embracing the Palsearctic species, all the Nearctic 



species except the one quoted below, and the Neotropical 

 species. 



2. Tamiasciurus, represented by hudsonicus and possibly related 



species. 



* My best tbanlis are due to the autliorities of the Smithsonian Institution for the 

 generous loan of these two Squirrels. One, in winter pelage (Januarj'), was collected 

 at Nulato in Alaska by W. H. Dall, the other, in summer pelage (September), at 

 Tamarak Swamp, New York, by E. A. Mearns. 



f Although Rheithrosciufus, with its grooved incisors, has always been regarded 

 as an isolated genus, the discovery that it is really related to Sciurus (s. s.), on the 

 evidence of the baculum, would surprise me much less than did the discovery of the 

 wide divergence between Tamiasciurus and Sciurus in the structure of the glans 

 penis and the occurrence of the baculum. 



