Biology Reference
In-Depth Information
(e.g., stress survival due to dormancy or small size). As that recovering
population develops, only the unique alleles that have been passed to that
youngest cohort will be available to it. Comparisons of
Figs. 15.1,
15.4,
and 15.8 demonstrate that an establishing population can reach relatively
large numbers of individuals and still be at risk of loss of a signifi cant
number of low-frequency founder unique alleles under such catastrophic
circumstances. For example, considering only the N and O trial populations
at age 49, in the entire population there are approximately 1,650 individuals
that bear slightly more than 2,060 unique alleles. The next cohort to be
produced numbers approximately 235 individuals. However, the number of
unique alleles present in that cohort 50 is only 1,830, or 88.6% of the unique
alleles in the entire population. Thus, if a catastrophe eliminates all but the
latest cohort at age 50, just over 11% of the unique alleles resident in the
destroyed individuals will be transmitted to the future, even though there
are more than 230 individuals that remain, more than the original number
of founders. NEWGARDEN modeling can thus be helpful in estimating the
number of offspring that need to be harvested to ensure that most of the
unique alleles resident in a source establishing population are transferred
to a new population to be founded by those offspring.
Another issue raised by the above analyses involves F values. As shown
in
Figs. 15.3
and
15.7
above, F values in later generations increase more
rapidly when, all else being equal, the 172 founders are subdivided into
four groups of 43 at each corner. What is this trend due to?
1. Is it the greater loss of unique alleles when offspring are dispersed
outside the preserve, increasing the degree to which copies of certain
alleles are transmitted and thus increasing inbreeding?
2. Is it predominantly or partly the Wahlund effect, in which there is an
apparent excess of homozygosity in the observed data for the more
subdivided trials? In the latter case, since F is calculated as:
(H
expected
- H
observed
) / H
expected
an increase in the difference between expected and observed
heterozygosity (which can occur because of the Wahlund effect of
sampling what are actually four subpopulations as though they are
one) will cause F to get larger.
3. Is it greater inbreeding yielding greater losses of observed
heterozygosity that might be occurring within smaller, more isolated
groups experiencing lower realized gene fl ow?
4. Is it some combination of the above?
To examine the relative contributions of these factors to the increased F
exhibited when the founders are divided into four smaller corner groups,
the following two additional trials were conducted. In both trials, conditions
Search WWH ::
Custom Search