1913] Grinnell-Swarth: Hints and Mammals of San Jacinto 393 



THE BEHAVIOR OP GEOGRAPHIC RACES OX THE 

 MARGINS OF THEIR HABITATS 



In explanation of the diversity in behavior of geographic 

 races on the margins of their habitats, as pointed out in the 

 preceding section of our report, we may suggest that the direc- 

 tion of invasion of strains, whether from the desert towards the 

 Pacific, or vice versa, may be dependent in part upon one or more 

 of the following contingencies: (1) Continuity of the proper 

 association from one side to the other. (2) Disproportionate 

 representation of said association on one side as compared with 

 the other, as for instance with the riparian song sparrow. (3) 

 Lack of a complementary type on the opposite side, which would 

 if present check invasion by competition; for example, the cases 

 of the song sparrow and Ammospermophilus . 4) Unbalanced 

 rate of dispersal in the representative races of the two abutting 

 faunas; thus Lepus c. deserticola may have spread westwardly, 

 through a period of time allowing for many generations, well into 

 the margin of the habitat of L. c. bennetti because of greater 

 " intra-specific vigor," the line of intergradation being Ihns 

 forced oxer into the area where bennetti might be supposed to be 

 best fitted to exist. According to this idea, in this and similar 

 cases, unequal rate of distribution may carry one geographic 

 race beyond the limit of its area of differentiation, and yet the 

 frontier individuals, even for a considerable series of generations, 

 maintain their original characters. The unequal "intra-specific 

 vigor" may result from augmenting rate of reproduction. 

 periodically reduced death rate (which would amount to the 

 same thing), or from development innately of a greater degree 

 of aggressiveness in disposition, as a specific trait — a proneness 

 to wander, on the part of individuals. l'< romyscus m. sonorit mis 

 may lie cited in the latter connection. 



An assumption upon which the last consideration depends is 

 contrary to the common conception among laboratory experimen- 

 talists, namely, that most if not all of the characters recognized 

 by the systematic student of birds and mammals are somatic, sub- 

 ject to modification in the life-time of the individual or at least 



