1899.] THE MICE or ST. KILDA. 79 



the form of M. st/lvaficus described by my friend Mr. W. E. de 

 Winton^ from the Isle of Lewis, Outer Hebrides. The adult 

 female from St. Kilda (which is in spirit) may possibly not be so 

 stout in foot nor so small in ear as the Hebridean Mice, but the 

 two forms are very close to each other, and there can be no doubt 

 that the St. Kilda Mice belong to the Hebridean type, although 

 their rufous belly has carried them a little further along the same 

 line of development in which Mus hebridensis deviates from 

 typical sylvaticus. In this respect I find that the most rufous 

 skin of all is the first one collected by Mr. J. Steele Elliott. In 

 it there is no perceptible line of demarcation between the colours 

 of the upper and under surfaces, the transition from the one to 

 the other being, as stated above, quite gradual. As regards 

 the specimens obtained for me by Mr. Evans, the colour of the 

 belly of the adult female, which is in spirit, agrees with that of 

 Mr. Steele ElUott's specimen ; but in the male, which has been made 

 into a skin, the belly is slightly hghter, the median broad buff 

 belly-line of Mus hebridensis is more evident, and there is a just 

 perceptible line of demarcation between the colours of the two 

 surfaces. The colour of the upper surface of the body of all the 

 specimens is also, as in Mus hebridensis, more evenly distributed 

 than in typical sylvaticus, there being less tendency to the develop- 

 ment of a dark dorsal line. 



It is exceedingly interesting to find this graduating series, and 

 to have the gap between Mus sylvaticus and the St. Kilda Mouse 

 partially bridged over by the occurrence of Mus hebridensis on the 

 intervening islands. 



This slight variation of the St. Kilda specimens in regard to the 

 colour of the belly, the white colour of which is so extremely 

 constant in and characteristic of Mus sylvaticus, is worthy of note, 

 being exactly what we should expect to find in a comparatively new 

 species which has not yet finally settled down into its new groove 

 of development. We thus find that while in the colour of the 

 belly some of the St. Kilda Mice may vary in the direction of 

 Mus hebridensis, it is in this very respect that the latter form may 

 vary in the direction of Mus sylvaticus. Indeed, in this regard 

 Mus hebridensis is very variable, and I have examined some Isle of 

 Lewis specimens, especially those from the eastern coast, which 

 come very close to Mus sylvaticus in the colour of the underside. 



In addition to the above mice, Mr. Evans also procured for me 

 five specimens of the House-Mouse of St, Kilda, of which the 

 Museum already possessed five specimens collected on previous 

 occasions and now preserved in spirit. These mice are, if possible, 

 of even greater interest than the Mus sylvaticus-Mke species, since 

 they are characterized by the possession of a buff-coloured 

 underside clearly marked off from the colour of the upperside by 

 a distinct line of demarcation, and are thus very different from 

 the ordinary almost uniformly smoky-brown-coloured House-Mice 



1 Zool. 1895, p. 369. 



