classes. Taxodium sp. may depend on a drawdown of water levels for germination and
seedling establishment. Therefore, this type of distribution may be expected in natural
systems experiencing infrequent drawdowns. The majority of constructed wetlands in
this study have, as yet, not demonstrated recruitment of a new cohort of Taxodium sp.
saplings. Only in Parcel B is the probability of sampling Taxodium sp. in the smallest
size class equal to that of larger size classes. Limited regeneration may be occurring at
Parcel B and explain this variation in size class distribution. Regeneration may be
occurring at Parcel B, and not at other sites, for at least two reasons. First, a hydrological
drawdown may have occurred to facilitate germination and establishment or the
hydrology at this site may not be interacting with Taxodium sp. and limiting germination
and establishment. If regeneration is not occurring, then this skewed distribution could
be explained by slow growth rates in a greater than expected portion of the population.

 Fraxinus caroliniana demonstrated a hierarchical size class distribution in all
 study sites where it was present except for one. At this one site, the size class resembles
 that of an even aged stand. In a previous study (Miller 1983) of natural wetlands
 impacted by adjacent mining activity, size class distributions of Fraxinus caroliniana
 resembled that of an even aged stand.

 Tree basal area measured at breast height or above the buttressing of wetland
 species appears to reflect the environmental variability within a site, and community
 basal area integrates information about tree size class distributions. For these reasons,
 greater information about the self-organization of constructed forested wetlands may be
 available from the evaluation of community basal area (m2 ha') than from diameter at
 breast height of individual trees. Community basal area of trees greater than 5 cm in
 diameter at breast height in Florida wetlands range from 27.68 m2 ha-' in bayheads to
 38.89 m2 ha' in hardwood swamps (Davis 1991). Community basal area in this study
 ranged from 0 to 9 m2 ha'1. In these sites, tree height and canopy cover may reflect
 values typical of natural wetlands; the community basal area of trees is not yet indicative
 of natural wetlands.


 Subcanopy Tree Species

 The subcanopy component of constructed forested wetlands in this study are still
dominated by Myrica sp and Salix sp. Myrica sp. is considered a subcanopy or shrub
component of natural forested wetlands. Frequency of occurrence of Myrica sp. is high
in cypress domes (83%), but its importance value is low relative to other subcanopy
species (Davis and others 1991). Both, the frequency and importance value of Myrica sp
are lower in bayheads and hardwood swamps (Davis and others 1991).

 Relative frequency of Myrica sp.and Salix sp. in constructed forested wetlands
ranged from 0.03-0.94 and 0.15-1.0, respectively. The data suggest that Salix sp. is more
likely to dominate the subcanopy component in younger sites, while Myrica sp.
dominates in older sites. The transition in dominance occurs around 10-12 years.


7-86