136 UNITED STATES NATIONAL MUSEUM BULLETIN 2 76 



selection, this sequence of events could go on indefinitely until genes of 

 high altitude populations were eventually introduced into low altitude 

 populations, and vice versa. 



This does not appear to be an effective means of exchange, however, 

 because, if it were, populations such as incurvum would not be able to 

 maintain their distinctive characteristics. Several factors could act to 

 reduce this exchange: 



1. Natural selection undoubtedly favors some gene combinations, 

 and it must be assumed that in most cases the gene combinations of the 

 native population are better adapted to the native habitat than most 

 new combinations that result from the introduction of different genes. 



2. The number of females which fly to higher or lower localities is 

 probably quite small, resulting in a relatively low number of genes 

 being introduced into the native population. 



3. Mating preferences for members of their own population could be 

 present, further reducing the number of genes introduced into the 

 native population. 



4. The number of genes from the original population carried by a 

 "hybrid" female to the next higher or lower level would probably be 

 reduced with each succeeding generation, resulting in relatively few 

 genes eventually being introduced from a low elevation population 

 into a high elevation population or vice versa. In addition to these 

 factors, there are such things as possible host preferences which could 

 prevent successful movement to new areas where the preferred host 

 did not grow, and gaps in the distribution of suitable hosts that could 

 be difficult to cross. 



Granting that it is possible for some gene exchange to take place be- 

 tween incurvum populations and populations of californicum which occur 

 at higher elevations, this exchange appears to be quite low in areas 

 that have been studied. It may be greater in other areas, especially 

 Mexico, but, even if this turns out to be true, it probably still will be 

 desirable to recognize some populations as the species incurvum for the 

 reasons already given. 



Specimens Examined. — 6,362 (Museum specimens — 1,41 1 males, 499 females, 

 806 larvae; reared specimens — 2,106 males, 1,540 females plus numerous larvae 

 and egg masses). See Appendix I for data on collections made during this 

 study (page 259). 



United States. — Arizona: Alpine, Apache Indian Res., White Moun- 

 tains; Baboquivari Mountains (Pima Co.); Bear Wallow Trail, Santa Catalina 

 Mountains; Bright Angel Point, Grand Canyon; Catalinas; Catalina Springs; 

 Chiricahua Mountains; Christopher Creek, Mogollon Rim (Gila Co.); Cibola 

 (Yuma Co.); Cochise Co.; Concho; Dewey; Fairview; Flagstaff; Greer; 

 Kohl's Ranch, Mogollon Rim (Gila Co.) ; Palmerlee, Miller Canyon, Huachuca 



