In this section discrepancies between the corridor simulations and predictions of one character quantitative genetic theory are partitioned into three factors: deviations from the predicted selection differential, the selection response and the asymptotic genetic variance.
Standard quantitative
genetic theory predicts that the rate of change due to directional
selection can be modeled as a two stage process. First selection acts
on the phenotypic distribution of the parental generation, which
produces the selection differential, and second the transmission to
the next generation. The selection differential is the difference
between the population mean value before 
and after
selection 
, 
. In the
simulations the selection intensity was measured by 
, the
standardized selection differential. Hence the realized selection
differential is
The expected selection
differential depends on the phenotypic variance of the parental
population and the strength of the directional selection. If the
phenotyic values are normally distributed, the strength of the
directional selection is given by 
, for a single character which is under
the influence of the fitness function 
. The expected selection differential therefore
is 
, and we
measure the relative realized selection differential by:
The selection response is the difference between the
character mean values before selection in the first generation and the
following generation, 
. The expected selection response is the product
of the selection differential and the heritability, 
. Note that in analysing the simulations,
both the heritability and the selection differential are observed
values. The prediction only concerns the functional relationship
between the selection response and the selection differential and the
heritability. The
Table A, 
Table B,
Table C, 
realized average selection response in our
simulation is measured as 
. Hence, we measure the relative response to selection
as
Finally
the asymptotic rate of evolution depends on how much genetic variation
for the character can be maintained under selection, mutation and
drift. For the case without pleiotropic effects the prediction of
Hill (1982) is 
. The relative amount of
genetic variation in
our simulations is calculated by
The relative rate of evolution can be reconstructed
from these three factors: 
. In table 6 the three factors influencing the relative rate of
evolution are listed. In the low mutation rate scenario it is obvious
that the selection differential has the smallest impact on the
relative rate of evolution. The realized selection differential is
between 88 and 99% of the expected. Clearly the stationary amount of
genetic variation has the strongest impact on the observed relative
rate of evolution. The relative magnitude of the selection response
is only important in cases where the stabilizing selection is strong
(
) and the directional selection is weak or moderate (
or 
). In the high mutation rate scenario, the only
significant factor influencing the relative rate of evolution is the
stationary amount of genetic variance.