Perennials - Genetic Diversity in Establishing Plant Populations: Founder Number and Geometry - page 242

leading to more localized inbreeding and consequently more rapid
loss of heterozygosity. These two factors act together to give trial m
the greatest increase in difference between expected and observed
heterozygosity, and thus the greater values of F.
Focusing on this particular Lindera example, since increases in F are
driven by increased inbreeding in smaller populations, subdividing the
founders may cause increased loss of fi tness in species subject to strong
inbreeding depression. In species for which inbreeding depression and
slightly increased chances of unique allele loss are not major issues,
subdividing the founders remains an option in restoration planning. Since
unique allele loss drops dramatically once reproduction in the population
commences, these results suggest that more effort should be spent on
maintaining the original founders until full reproduction is underway.
Founders that are lost should be replaced as soon as possible. If fi eld-
planting of founders from seed results in increased losses of individuals,
then fi eld-planting glasshouse-raised founders only when they are at or
near sexual maturity is another option to consider. NEWGARDEN analyses
regarding standing developing populations in later phases can also be used
to estimate various aspects of the stand genetic diversity loss or retention
history from data derived from monitoring such populations, and thus can
also be used to model effective strategies for maintaining or supplementing
diversity at target levels.
All of the above considerations obviously also affect natural populations
that establish in different geometric patterns. Estimated and observed
heterozygosity, F, rates of unique allele loss, and other calculations and
comparisons involving them are thus infl uenced by both the number of
founders and the pattern of their introduction. This may make comparative
interpretation of these values for similar populations diffi cult unless the
founding details of the populations are known. Species or populations with
different life-history characteristics will yield different results.
Age-specifi c Mortality
In natural populations, different individuals die at different ages, and rates
of death for individuals of a given age are often generally specifi able as a
probability of dying at a particular age. NEWGARDEN allows the analyst
to assign different probabilities of dying at different ages. In each round of
reproduction, individuals in each age class are randomly chosen to die from
the pool of all individuals in a class, depending on the assigned probability
of death for that class. In the Lindera trials discussed above, while mortality
rates were designated, all trials had exactly the same rates of age-specifi c
mortality. The following set of trials demonstrates how differing age-specifi c