NO. 597 RODENT ETHOLOGY — EISENBERG 31 



tlirougli estrus. Thus, as outlined in the previous section, there are 

 probably several physiological mechanisms that are affected adversely 

 in terms of reproductive success by social contacts. The relative 

 thresholds for these mechanisms varies from species to species de- 

 pending on their genetic makeup. 



The adverse effects of high population densities on the growth and 

 maturation of young rodents has been known for some time (Chitty, 

 1955), and the cessation of reproduction at high densities through a 

 failure of ovarian and testicular function has been studied from a 

 physiological standpoint by many workers (see Christian, 1963). 

 Recently Calhoun (1963b) has developed an elaborate theory of social 

 behavior based on the concepts of species-specific differences in social 

 tolerance and the evolution of social groups. Wynne-Edwards (1962) 

 has put forth an all-embracing theory on evolution of social groups 

 and the adaptive value of self-regulatory mechanisms whereby popu- 

 lations are kept below maximum numbers by such reproductive 

 failures as outlined previously. I do not wish to examine these 

 various theories from a critical standpoint but only wish to point 

 out that the evidence favors my interpretation of species differences 

 in social tolerance and, further, that these differences have resulted 

 from processes of natural selection to produce a social type adapted 

 to its particular niche. I have attempted to outline a methodology 

 whereby one can study social tolerance by means of several techniques 

 and arrive at some quantitatively based conclusions regarding species 

 differences in sociability. 



There remains a further consideration. Given the demonstration 

 that the rodents under study exhibit species-specific social tendencies, 

 what adaptive correlates can one discern? The arid-adapted sub- 

 family Gerbillinae spans the range from a relatively solitary form, 

 GerbiUus nanus, to the more tolerant, commmial Tatera indica. Jaculus; 

 orientalis and Allactaga elator are semitolerant; however, the family 

 Heteromyidae, whether forest-adapted or arid-adapted, appears to 

 be a solitary form. Within the genus Peromyscus, the desert species 

 F. crlnitus is intolerant whereas the equally xeric-adapted P. eremicus 

 is more tolerant. Adaptation to xeric habitats with dispersed food 

 supplies is not necessarily in and of itself conducive to selection for a 

 dispersed, solitary existence. Equally important are other aspects of 

 the species ecology including its mode of assembly of foodstuffs, its 

 shelter construction, and its reproductive rate. At this point it seems 

 safe to say that the Heteromyidae have retained the phylogenetically 

 ancient trait of solitary existence because of the adaptive advantage 

 accruing from its defense of cached food. It is not the case, however, 

 that this social trait is always a concomitant of caching. 



It is noteworthy that GerbiUus (Dipodillus) nanus resembles the 

 silky pocket mice of the heteromyid genus Perognathus in several 



