F
isher (1930)explained why sexually reproducing
diploid organisms should invest equally in male and
female offspring at the population level based on ge-
netic arguments. Given this general expectation of equal in-
vestment, deviations from unity usually are considered note-
worthy and beg explanation. Not surprisingly, this area of
inquiry has attracted attention from researchers working
with diverse species. Although numerous hypotheses that
incorporate demographic and life-history parameters have
been advanced to explain deviations from unity, consis-
tent support is lacking for most vertebrate taxa (Krackow
2002). Sex ratios are a key element of life history theory
(Stearns 1992) and well-deserving of attention because, as
West et al. (2002, p. 122) stated, “if we cannot understand
sex ratios, we cannot hope to understand most other life
history traits, whose evolution usually depends upon far
more complex trade-offs.”
Most reviews of sex ratio adjustment (Clutton-Brock and
Iason 1986; Bull and Charnov 1988; Frank 1990; Cock-
burn et al. 2002; Hardy 2002; Krackow 2002; West et al.
2002) note little consistency in data regarding available hy-
potheses. But perhaps our perception of the magnitude of
biased sex ratios is overblown. If sex ratios in mammals are
in fact normally distributed, then Type I statistical error
will still result in a certain percentage of samples that are ei-
ther female or male biased. If these male- or female-skewed
ratios, representing the tails of a distribution, are then re-
ported at a higher than expected frequency simply because
they deviate from parity (Festa-Bianchet 1996; Krackow
2002), our interpretation of the frequency of skewed sex
ratios becomes biased by the over-reported minority, and
we feel compelled to fit these skews with explanations.
Publication of sex ratio data that do not differ from unity
will help counter this tendency (Duquette and Millar
1998), but only if the data constitute tests of one of the
hypotheses for biased ratios. What types of studies, then,
constitute appropriate tests, and what types of data are nec-
essary if we are to make headway in eliminating alternative
explanations?
Rodents are an especially attractive group for evaluating
deviations of sex ratio because obtaining large samples is
often feasible and because many species make suitable lab-
oratory models for investigation of mechanisms of adjust-
ment. Rodents typically produce litters rather than single
offspring, so sex ratios potentially can be adjusted both at
the level of an individual litter as well as at the population
level. Further, as litter sizes for at least some taxa may be
adjusted by the mother to suit environmental conditions
(Slade et al. 1996), the potential exists for easy manipula-
tion of sex ratios as well, by favoring one sex over the other
within the context of litter size adjustments. This potential
may be countered, however, by the short life expectancies
of many species of small rodents in the wild, which in turn
limits future reproductive potential. In these cases, if ma-
nipulating sex ratios requires eliminating offspring, the lim-
ited future reproductive potential of small rodents may pro-
vide too little selective advantage to offset the reduction in
size of a current litter.