118 Mr. M. A. C. Hinton on 



possible, we were unahle to do more in our first paper than 

 refer all the material to A. sylvaticns and notice the ten- 

 dency of certain of the island forms to approach A. hebri- 

 densis of the Outer Hebrides. 



The skull-material obtained by the collectors named above 

 and a carefully chosen series from other British localities 

 have now been investigated. A long series of measurements 

 were taken with the utmost care. The data so collected 

 will, I hope, be published later elsewhere, but in the tal)le at 

 pp. 120-121 a summary consisting of the average relative 

 values of the various dimensions in each British race is given. 

 It is important to observe that in all cases, unless the contrary 

 be stated, the measurements refer to thoroughly adult skulls 

 in which the teeth are at least half- worn. 



It now appears that in practically every island of the 

 Hebrides differentiation from the parent stock has proceeded 

 so far that the most logical course would be to describe the 

 mice of each island as distinct subspecies. On the mainland 

 of Britain the same process is seen at work. Skulls from 

 the English plains taken here as being typical of A. s. sylva- 

 ticus seem to be distinguishable from those of Wales or 

 from the Scotch Highlauds ; those from the loAvlands of 

 Scotland in turn have their peculiar characters. But in 

 order to define these mainland forms it will be necessary to 

 have far more material of the right kind than is at present 

 available. To quote Barrett-Hamilton once more, the work 

 must be based upon averages, and not upon individuals, and 

 it will require the most careful procedure. 



Apodemus srjlvaticus, Linnaeus. 



A. Non-Hebridean British Races. 



A. sylvaticus sylvaticus, L. 



Twelve old skulls from Kent, Essex, Middlesex, Oxford- 

 shire, Gloucester, Leicester, and Suffolk, in which the 

 condylo-basal length ranges between 22 and 23 6 mm., are 

 treated as representative of this subspecies ; the average 

 values of the reduced dimensions, the condylo-basal length 

 taken as 100, are given in the table at pp. 120-J21. 



The Welsh skulls (c.-b. 1. 22-2-23-6) show a slight ten- 

 dency to become narrower and deeper, and more definite 

 indications of shortening of the postmolar and diastemal 

 lengths, accompanied by a decrease in the size of the bulhe. 



Two types seem to occur in Scotland. In one, represented 

 by skulls from Edinburgh, Lanai'k, and Dunphail, Elgin 



