Theoretical models of the evolution of resource allocation patterns to
male and female function make the assumption that there are inherent
trade-offs between the two. Here we use a quantitative genetic approac
h to quantify trade-offs between male and female function and to deter
mine whether plant populations could readily respond to natural select
ion by quantifying the amount of genetic variation for pollen and ovul
e production. Both intra-and interspecific crossing designs were appli
ed to two populations of the predominantly outcrossing Mimulus guttatu
s and two populations of the highly selfing congener, M. micranthus. T
he only significant correlations observed among pollen number, pollen
size and ovule number were positive. Positive genetic correlations amo
ng the traits were sometimes reduced after removing the effect of flow
er size but still no significant negative correlations were detected.
These results suggest that positive correlations between pollen and ov
ule production may be due to the joint positive correlation of these c
haracters with the resource pool available for pollen and ovule produc
tion, as reflected by flower size. Heritabilities were moderate to hig
h for ovule production but low for pollen number and pollen size and s
uggest that responses to selection would differ between the two traits
. Crosses between the species revealed that there are additional genet
ic factors contributing to differences between the two species for cor
olla width, vs. pollen:ovule ratio. This is consistent with the hypoth
esis that genetic variation for resource acquisition may in part be re
sponsible for the overall lack of a negative correlation between polle
n and ovule production and provides a genetic explanation for little e
vidence of trade-offs between sexual functions in Mimuhus.