4 PROCEEDINGS OF THE NATIONAL MUSEUM vol. 122 



In a comparative study that spans 3 families and 34 species from 

 three continents, the problems of acquisition and maintenance are of 

 paramount importance. I should like to acknowledge the efforts of 

 my laboratory assistant, Mr. David Williams, who lent his sensitive 

 talents wholeheartedly to the work before us in Vancouver. Specimen 

 donations were also an integral part of this study. I am indebted to 

 Dr. D. Birkenholz of Illinois State University for 3 Heteromys des- 

 marestianus; to Mrs. W. Do\\ais of Yale University for 3 Gerbillus 

 gerbillus, 2 Jaculus jaculus, and 2 Jaculus orientalis; to Dr. T, Reed, 

 Director of the National Zoological Park, for 1 Pachyuromys duprassi; 

 and to Mr. W. Preston of the University of British Columbia for 12 

 Perognathus parvus. In addition, I should like to cite Mr. Ralph 

 Curtis of Miami, Fla., who was instrumental in arranging the im- 

 portation of a series of rodents from Pakistan that formed the basis for 

 our dipodid and gerbilline comparisons. 



The studies reported on in this paper were financed in part by the 

 following grants: E 350 and A 1371 from the National Research 

 Council of Canada, and a general research board grant from the 

 University of Maryland. The author is indebted to the Departments 

 of Zoology at the University of California, Berkeley, the University 

 of British Columbia, and the University of Maryland for the space and 

 facilities utilized in the 7-year course of this study. 



The Subjects 



The choice of the specimens utilized in this study was influenced 

 both by cu-cumstances and design. Initially the behavior of one 

 genus, Peromyscus, was selected for study. This North American 

 group includes over 40 recognized species (Hall and Kelson, 1959) that 

 have adapted to a myriad of habitats. As a genus it has lent itself 

 well to comparative ecological, physiological, and morphological 

 studies and offers a wealth of background studies that may be drawn 

 on by ethologists (King, 1967). Four species were selected by the 

 present author for behavioral studies, including two from the Transition 

 life zone and two from the Sonoran life zone (see Eisenberg, 1962a, 

 1963a). The beha^ior of this group is the subject of a recent review 

 (Eisenberg, 1967) and will not be redescribed here; however, relevant 

 data will be introduced since this group of four species represents an 

 example of the range of beha^^oral variation within one genus. 



Several questions were raised in the course of the Peromyscus study 

 concerning species-specific changes in social behavior resulting from 

 adaptation to desert habitats. For this reason a complete rodent 

 family, the Heteromyidae, was studied in order to correlate generic 

 differences with niche requirements as I compared moist forest-adapted 

 genera (e.g., Heteromys and Liomys) with desert or semidesert- 



