26 



PROCEEDINGS OF THE NATIONAL MUSEUM 



Group formation through internal recruitment. — By allowing 

 a pail- to reproduce in a large cage, one can fm"ther evaluate social 

 tolerance. In general, reproduction will continue until social con- 

 tacts reach such a peak that the female fails to cycle or deserts her 

 litter. If one adult male can keep other adults away and reduce the 

 level of contacts to which the sexually reproducing females are 

 subjected, then reproduction wiU continue. This is true for communal 

 as well as solitary species, but the latter rely on overt aggression more 

 often than do the former and also the solitary species are much more 

 sensitive to social interference than is the case with the more social, 

 communal forms. Thus, a communal species in a confined space 

 generally will exhibit an area in the cage where several sexually active 

 females are nesting and rearing litters. A single sexually active male 

 wiU have access to these females while the remainder of the popula- 

 tion, adults and juveniles of both sexes, dwell communally and fail to 

 reproduce. (See also descriptions for Mvs musculus, Crowcroft and 

 Rowe, 1963; and Rattus norvegicus, Calhoun, 1963a.) Eventually 

 reproduction may cease altogether. Figiu'es 7 and 8 demonstrate the 



20 Number of 

 individuals 



• P gremicus 

 " Fi (;rinitu 5 





\ 



N D 

 1957 



F M 

 1958 



Figure 7. — Population growth for two species of Peromyscus (stars^number of individuals 

 showing wounds during the spring fighting). 



population growth of several species of Peromyscus. Note in several 

 cases the tendency to form a stable plateau with a failiu-e of reproduc- 

 tion. Note also, in the case of P. crinitus, the wounding caused by 

 male aggression. Although not illustrated in these particular graphs, 

 there is in some cases a slight trend to begin sexual activity at the 

 onset of spring after a fall and winter plateau; however, as shown 

 here the plateau holds. Table 22 includes the results of all population 

 growth experiments with eight rodent species. Note that the rela- 



