1899.] THE iUCE OF ST. KILDA. 87 



tyjyicv.s in houses. It seeins to me, therefore, probable that 

 both Mus hactrianus and Mus musculus are developments of some 

 original parent form to suit particular conditions, and we may 

 perhaps look for the latter to some Central Asian species like 

 M. ivagneri. 



Some of the white-bellied forms which are found in a wild state 

 in Western Europe and in other countries where Mus musculus 

 tyjjicus occurs in houses may be cases of reversion from the latter, 

 which is no doubt almost certainly the origin of such races as are 

 found on islands, such as the Salvage Islands, where the Mice 

 must have been accidentally introduced. But it by no means 

 follows that this is the case with Mus spicilegits, the size and pro- 

 portions of which are so much finer than in true Mus museuhis and 

 the tail shorter. Mus spicilegus, indeed, might even be regarded as 

 a wild parent form of Mus musculus; hence it is not with it, but the 

 forms which are certainly reversions from true Mtis musculus, that 

 we must associate Mus muralis of St. Kilda, and it is interesting 

 to note that the similarly derived Mice of the Salvage Islands 

 resemble those of St. Kilda very closely in their robust form. 



That a wild race of Mus musculus can be rapidly evolved from 

 Common House-Mice when living in a wild state has been recently 

 shown by my friend ^ Mr. H. Lyster Jameson, who has clearly 

 made out his case for the formation of an incipient species of 

 Mouse on the Xorth Bull, Dublin Bay, Ireland, a tract of sand- 

 hills about three miles in length and almost completely isolated 

 from the mainland. 



It is known that this sand-bank has not been in existence for 

 more than about 100 years, so tliat the coloration described by 

 Mr. Jameson must have been evolved in at most a period of that 

 length. 



Mr, Jameson lays great stress on the value of the change to 

 these mice as a protective feature, and so he has not, I think, 

 given sufficient emphasis to the fact that we have here a clear 

 instance of the development of an incipient subspecies of Mouse 

 with an exact period laid down in which the change occurred, 

 and we may faii-ly, I think, use Mr. Jameson's results in dealing 

 with other species or subspecies of Mice. 



If we are to judge from the analogy of Mr. Jameson's mice, 

 we must conclude that the Mice of St. Kilda have inhabited that 

 island for a considerable time. Not only are they more distinct in 

 colour than any other local form of Mus inusculus with which I am 

 acquainted (and I have been through the whole of the specimens 

 in the British Museum Collection), but their line of development 

 seems to have become fixed, and is no longer, as in the case of 

 Mr. Jameson's mice, in a state of uncertain evolution. On the 

 North Bull sand-hills, indeed, Mr. Jameson found not only mice 

 which had progressed for a considerable distance along the path 

 of their new development, but also mice which showed every kind 



^ Journ. Linn. Soc, Zool. vol. xxvi. pp. 465-473 : " On a probable Case of 

 Protective Coloration in the House-Mouse {Mus miisculiis, Linn.)." 



