72 



BULLETIN 114, UNITED STATES NATIONAL MUSEUM. 



closely approximated the extremes and average of hoylii. This 

 range of variation in number of rings is not now altered throughout 

 the region occupied by hoylii, yumensis, and conjuncta. 



The decisive change in the configuration of the pattern as well as 

 in the number of rings or bands occurs also in southeastern Arizona, 

 but striking as this change is, its steps are all easily recognizable, 

 and are explained in detail in the discussion of the evolution of the 

 group. 



Its pattern differs from that of hoylii in that the white centers of 

 the dark scales, derived from splendida, do not cover the whole of 

 the scale, but are mainly limited to the distal portion. Northward 



//s- 



//p 



//s 



f/o 



Fig. 16,— Map showing locality kecords foe Lampropeltis getulus yumensis. 



this condition very soon changes into that typical of hoylii; that is, 

 the white spreads over all of the scale, producing rings of continuous 

 white. Westw^ard, however, the pattern of yumensis is retained. In 

 the vicinity of Yuma the light rings are often heavily suffused with 

 brown. A specimen from near the mouth of the Colorado River, on 

 the west side (Volcano Lake), is still typical of yumensis. 



It will be of interest, in connection with what is known of the 

 incomplete development of the pattern of the 3^oung of liolhroohi, and 

 its possible ancestral significance, to learn whether the young of 

 yumensis are entirely like the adults in the degree of basal shading 

 of the dark scales. Only three yoimg specimens are now available. 

 One from Tucson looks more like hoylii, than any of the adults, but 



