Darwinian evolution is normally modeled as the outcome of the interact
ion between reproduction, variation and selection. These interactions-
may become very complex, especially when considering that selection ma
y act on several genes simultaneously. A numerical simulation model wa
s developed, based on a Jive-step stochastic evolutionary algorithm, w
hich-in addition to reproduction, variation and selection included exp
licitly mate selection and phenotypic transformations. The model graph
ically illustrates:the dynamic interactions affecting a multi-gene sys
tem. The results emphasize that biological evolution is an irreversibl
e process showing strong genetic drift. Simulations showed :: that ase
xual reproductive strategies were very stable in evolutionary competit
ions and consistently out-bred sexual ones; Bisexuality succeeded in r
eplacing asexuality only in changing environments and could be maintai
ned only if non-random mate selection mechanisms were at work. Simulat
ions showed that in trisexual organisms, mate selection mechanisms and
natural selection do not work efficiently. Thus, reproductive strateg
ies with more than two sexes are improbable. Polyploidy may improve ev
olutionary success among monosexuals as it reduces the exhaustion of t
he variability of the gene pool due to continuous selection pressure.
I postulate, that sex evolved under changing selection pressures, allo
wing the emergence of sexual selection, which allowed sex to stabilize
, thanks to the strong irreversible dynamics in the evolution of compl
ex organisms.