104 THE GROWTH OF GROUPS 



although B is, on the whole, considerably larger than A, 

 the distance between the ridges is greater in A than B, 

 in the proportion of three to two. This of course is due 

 to the fact that B has larger temporal muscles than A. 



Let us now examine the posterior surfaces of the three 

 skulls. In A and C the foramen magnum (shown black) 

 is of the same type, somewhat like a gothic arch. This 

 type of foramen is constant in both genera. The skull 

 C is the only one out of some hundreds taken from various 

 districts that departs from this type, and it departs very 

 far from it, as shown in the diagram. 



The skull C is typical of the genus Gunomys as found 

 in the district. Eighty of the kind were examined and 

 the illustration would almost serve for each of them. 

 But skulls A and B, although belonging to the genus 

 Nesokia since they possess the most important character 

 of that genus, are not typical of Nesokia in the district. 

 The skull A is however typical of a local group of rats 

 (Nesokia) taken from a particular system of burrows near 

 a village named Atari. The members of this group were 

 distinguishable from the others of the district in being 

 unusually small and in possessing a skull type of their 

 own. The skull designated A was taken from a member 

 of this group, and the illustration would serve equally 

 well for any one of the four adult members of the group 

 which were examined. 



The skull B is also of an uncommon type. But only 

 one specimen was obtained. There is, however, no reason 

 why it should not have been typical of a local group. 



The foregoing descriptions may perhaps appear pur- 

 poseless, but they serve to illustrate somewhat the pro- 

 blems before systematic zoologists, and to explain why 



