Dispersal - Genetic Diversity in Establishing Plant Populations: Founder Number and Geometry - page 184

occurring at rather high rates. For example, population O has a maximum
dispersal distance of 12 units, and since the minimum distance between
individuals of separated populations is 40 units, mixing of populations
should begin within two to three generations. However, 4.5% of the unique
alleles are lost for population O between generation 5 (795 individuals)
and generation 14 (6,632 individuals). Despite what would seem to be a
signifi cant amount of mixing of these populations, unique alleles are still
being lost at an appreciable rate.
For some species (e.g., tropical rain forest trees), more distant dispersal-
establishment of offspring may be common. To explore this issue, trials
replicating the above conditions were run, but with an increase of dispersal
distance of both offspring and pollen to 400, and then to 600, grid units.
When trial populations were slightly subdivided (the four subgroups 20
units from the preserve center), no signifi cant difference could be detected
for population growth rate, heterozygosity, F value, or unique allele
retention results compared to when the 172 founders were placed in one
undivided central square. However, when subgroups were placed with
one approximately 80 grid units from each of the four corners, population
number dropped below 172, inbreeding increased, and there was a greater
loss of heterozygosity and unique alleles (up to 60% greater loss) when
dispersal distances were set to 400 or 600 units (data not shown). Such
distant dispersal induced edge effects of the type discussed earlier, with
losses of offspring outside the preserve so great that they negatively affected
population establishment. Such negative boundary effects under this
greatest degree of subdivision were not apparent with offspring and pollen
dispersal distances at 30 grid units, which was in fact the most benefi cial
dispersal distance for population establishment in terms of population
growth and unique allele retention. The establishment of a species must be
carefully modeled over a range of potential life history characteristics values
to ensure that realistic possible increases or decreases in those values do
not have pronounced negative effects on population growth and diversity
retention.
These results indicate that several different factors must be assessed
when designing a species reintroduction plan. In one case, optimizing
rapid growth of the population might be deemed most important, calling
for greater subdivision or spacing between founders. If dispersal for a
species is known to be strongly leptokurtic, then a decision will have to be
made whether avoiding inbreeding and heterozygosity loss is of primary
interest (better to avoid subdivision), or population growth and unique
allele retention is more desirable (better to have subdivision). All of these
decisions must also be weighed against the costs of introduction: traveling
by foot a distance of 640 grid units total (8 in-out trips * 80 grid units) to
plant four corner plots of founders (assuming travel to the edge of the