Lindera melissifolia (Walt.) Blume is a federally endangered southeastern endemic shrub, that is declining due to land clearing and conversion to agriculture or silviculture. In addition, gene flow is reduced due to habitat fragmentation, and new populations are not establishing. Reproduction by seeds is rare, and many populations are becoming male-biased.
The goal of this study was to determine favorable environmental conditions for reproduction via seeds (safe sites) as well as for sustaining equal sex-ratios under field conditions. Seedlings were grown under varied moisture and light conditions for varied lengths of time and then harvested to determine dry mass and leaf area. Growth rates were analyzed using a two-way factorial analysis of covariance (ANCOVA). Net assimilation rate and standard morphological ratios were calculated and compared using a two-way analysis of variance (ANOVA). Seedlings were clipped below and above root collars as a means of simulating two levels of disturbance and then assessed for mortality and resprout relative growth rates. Treatments were compared to a control using a one-way ANOVA. Adult stems of both sexes of Lindera melissifolia and co-occurring vascular flora were surveyed for percent cover within non-random plots centered on the highest densities of flowering stems. Flowering stems were counted and compared to transmittance and stem density using regression analysis. Multi Response Permutation Procedure (MRPP) and Indicator Species Analysis Indicator Values (IV) were used to compare species composition within plots with females present or absent.
A violation of equal regression slopes occurred in the two-way ANCOVA model, but when using a one-way ANCOVA grouped by light, lowered moisture had a negative effect under the highest (40%) light treatment (p=.032). Net assimilation rates were significantly different for light effects, but moisture effects were significantly different only under highest transmittance when analyzed by light, using a t-test (P=.0495). Morphological responses showed a higher amount of plasticity under varied light than under varied water.
Clipping of seedlings below the root collar increased mortality, while clipping of seedlings above the root collar did not. Relative growth rates of both clipping treatments were significantly lower than the control treatment (p=<.001).
Percent cover of Lindera melissifolia explained 52% of the variation in male flowering stems, but explained only 14% of the variation in female stems. Percent transmittance did not have a significant effect on flowering stems. Indicator species analysis revealed a strong association of male flowering stems with Pinus taeda (IV=59.8) and Vaccinium corymbosum sensu lato (IV=47.7), and a weak association of female flowering stems with Taxodium ascendens (IV=26.7).
Results confirm that hydrology is more important than light in creating safe sites and maintaining sex-ratios, but that increased light under dry conditions likely lowers competitive ability. Infrequent, low intensity disturbance may also prove to be a useful management tool for limiting competition from coastal plain woody species.