Voles of the orcadensis Group. 453 



Mr. Miller says that " The degrees of distinctness of the 

 Orkney voles among themselves appear to bear a direct 

 relation to the depth of water separating the islands, and 

 therefore presumably to the length of time that the different 

 colonies have been isolated. The numerous specimens of 

 3[. orcadensis examined from four islands of the southern 

 group show no indication of the existence of local forms " *. 



Since the ' Catalogue ' was printed Mr. YV. R. Ogilvie- 

 Grant has enriched the National Collection with a tine series 

 of specimens from Sanday and Rousay, including a good 

 many thoroughly adult skulls. I have had to study this 

 material carefully in connection with the Pleistocene M. 

 comerij and it now appears difficult to regard the differences 

 between orcadensis and sandayensis as of specific value. In 

 this paper, therefore, all the Orkney voles are referred to 

 M. orcadensis. This species is apparently a modified relict of 

 the Pleistocene fauna of western Europe such as we should 

 expect to encounter in the Oikneys. As an effect of long 

 segregation on small islands M. orcadensis is differentiated 

 into a number of local races or subspecies; with the possible 

 exception of those from Shapinshay (which I have not seen 

 recently), one is able to distinguish the voles of each island 

 from those of all the other islands. 



The characters which differentiate the skull and teeth of 

 M. orcadensis from those of less modified members of the 

 arvalis group seem to be correlated with an increased deve- 

 lopment of the temporal muscle f ; the beginnings of the 

 changes consequent upon such an increase are seen in several 

 arvaloid voles, and they culminate in orcadensis J. The 

 features which distinguish the skulls and teeth of the various 

 island races of the latter from each other in turn appear to 

 be consequential upon variations in the development of the 

 anterior and posterior portions of the muscle. To appre- 

 ciate this it is necessary to study the skulls of old and young 

 individuals comparatively ; in the table at pp. 458 & 459 those 

 measurements which throw light upon these points have been 

 taken from my full tables and summarized. 



In young skulls of M. o. orcadensis (Pomona, condylo-basal 

 length 24-25*4 mm.) the temporal ridges are, of course, very 

 feebly indicated and are widely separated in the interorbital 

 and interparietal regions. In many respects the brain-case 



* Op. cit. p. 696. 



t The large size of the temporal fo.sspe was pointed out by Frof. 

 Charnock Bradley (Ann. Scott. Nat. Hist. L905, p. 5). 



I Parallel changes are Been in other groups of voles, c. //.. to a lesser 

 degree in M. ay rest is and to a more marked one in Stenocranius. 



