CALIFORNIA ACADEMY OF SCIENCES 



[Proc. 4th Ser. 



we had regarded as one subspecies, and called Pituophis caten- 

 ifer annectens, were not all alike, but represented several cen- 

 ters of geographic variation. It has seemed important to deter- 

 mine whether this actually is so, and whether conclusions more 

 satisfactory than those published could be attained by further 

 study of the group. 



Such study required additional data. The color patterns of 

 these snakes had not been considered in our former investiga- 

 tion for the reason that published records seemed to show so 

 much variation as to render them valueless as a means of clas- 

 sification. Investigation of them now, however, shows that 

 they do afford valuable data. Individual variation is great, but 

 so are the average differences found in several geographic 

 areas. The whole subject has been reopened, and the results of 

 additional study are given in the present paper. 



Reverting to the former paper, it may be recalled that it was 

 there shown that the gopher-snakes of western North America 

 may be divided into two groups by the number of their gastro- 

 steges. This is indicated in the following diagram, figure 1, in 

 which the upper curve represents the gastrostege counts in the 

 snakes we called Pituophis catenifer catenifer, while the lower 

 curve shows the same data concerning those regarded as Pituo- 

 phis catenifer annectens: 



,4ti> M/. . i» .itt *p! j.t» ui/t-jii^ jm iir.M'-UJ''^i^Mt.jj*..x3a'.xij..;a^.Mt 4jg'.ji«l3M...iW-2H..^»~JfB.Mhi:jff.ia-J^',jit»Mt-^trt^ 



Fig. 1 



As one proceeds south and east from the cool north coast 

 regions toward the wami south coast and interior desert 

 regions, the average counts gradually increase. Although the 

 transition is gradual, these gastrostege counts are of great use 

 in the separation of these snakes into the two groups which we 

 then regarded as two subspecies. 



This same difference and relationship are shown when the 

 combined gastrostege and urostege counts are charted for the 

 two groups, as has been done in figure 2. In this chart the 



