SNAKES OF THE GENUS PITUOPHIS 177 



Dr. H. S. Brode (in litt., 1928) mentions specimens taken in the 

 Blue Mountains at an elevation of 2,500 feet where the rainfall was 

 between 20 and 25 inches, at Walla Walla in the arid Transition 

 Zone, and near the Columbia River where the rainfall is only 5 inches 

 and the elevation 300 feet. One in the latter locality was seen 

 imitating a rattlesnake. 



The writer kept several snakes of this form in captivity for some 

 time. They could be handled freely. Mice were eaten whenever 

 offered, and sometimes several would be taken in rapid succession. 

 The food is usually, although not always, swallowed head first. 

 The method of swallowing seems to be dependent upon which end 

 the snake grasps before constricting, and this is generally the head. 

 On one occasion two snakes of about the same size were observed 

 attempting to swallow the same mouse. One had a hold upon the 

 head, the other upon the back. First one and then the other was 

 forced to loosen its hold but would return for another try immedi- 

 ately, until one of the two finally gave up in despair and retired to a 

 corner of the cage to sulk. 



Affinities. — The derivation of catenij'er deseriicola from sayi affinis 

 is indicated by the pattern and scale characters as well as by the 

 ranges of the two forms (fig. 63). The pattern of deseriicola may be 

 derived readily from that of qffinis by a slight increase in number and 

 decrease in the size of the dorsal spots, accompanied by an intensi- 

 fication of color in the spots from reddish to dark brown, grayish 

 brown, or black. In the ranges of variation in most of the scale 

 characters affinis and deseriicola are almost identical, and the aver- 

 age proportionate tail length is only slightly higher in deseriicola 

 than in affinis. Occasional specimens of deseriicola also have the 

 rostral slightly longer than broad, as is typical in affinis, but is never 

 found in c. caienifer or c. annectens. 



The separate derivation of annectens and caienifer from deseriicola 

 is indicated by the fact that, although in the number of dorsal spots 

 there is a continuous increase from deseriicola to caienifer to anneciens, 

 most of the scale counts show a decrease from deseriicola to caienifer, 

 but not from deseriicola to annectens. 



The probable relationships of deseriicola and the adjacent forms 

 have been expressed by the diagram on page 133. 



Table 13 lists the specimens of this form that have been examined. 



