one axis in a 3-dimensional graph of social structure, with
the distance from the origin reflecting the strength of the se-
lective pressure exerted by each factor. By plotting popula-
tions or species on these axes, it should be possible to assess
the relative importance of the different factors promoting
sociality as well as to identify higher-order patterns of so-
cial structure, such as ecological distinctions between sin-
gular- and plural-breeding rodent societies.
Clearly, one challenge that must be met by future stud-
ies of rodent social systems is to document behavioral and
environmental variables in ways that allow quantitative
comparisons of the type intended by our three-dimensional
model. At present, efforts to develop a general conceptual
framework for sociality in rodents are hampered by a lack
of quantitative data relevant to comparisons of social struc-
tures or environmental conditions. Developing metrics for
behavioral and environmental variables that can be used
across taxa will substantially improve our ability to identify
general behavioral patterns, as will efforts to test alternative
adaptive hypotheses for each species studied. Although our
discussion of rodent sociality has focused largely on subter-
ranean species, we believe that the conceptual framework
and graphical model outlined here can be extended to in-
clude all rodents. Ideally, our discussion of these ideas will
encourage considerable further investigation of the ecology
of sociality in these animals.254 Chapter Twenty-One