Understanding maternal effects in long-lived plant species to improve genotype selection for conservation

Background/Question/Methods

As the climate changes and invasive species continue to spread, proactive management may be needed to conserve native plant populations. Selecting appropriate plant material for restoration or other actions that will sustain populations is an integral part of any such plan and must take into account local genetic differentiation to limit maladaptation. Common gardens are used to determine the genetic basis of trait variation among populations from different geographic sources. However, maternal effects, the effect of environment on the mother plant during offspring development, can also affect offspring performance, complicating interpretation of these studies. Using limber pine (Pinus flexilis) and Rocky Mountain bristlecone pine (Pinus aristata) as model species, we explored the contribution of maternal effects to early seedling growth among populations in a greenhouse common garden study. We grew offspring sourced over multiple years from the same mother trees, comparing growth traits between source years. Additionally, we collected five twig clippings from the upper canopy of each mother tree and measured characteristics indicative of the relative vigor of the tree during each seed source year. These data provide information on the impact of the environment during seed development on seed characteristics and early seedling growth.

Results/Conclusions

Principal component analysis of the twig characteristics revealed significant factors explaining variation in mother tree growth. For both species, PCA axes 1 and 2 explained moderate amounts of variation in twig characteristics, and these axes were significantly related to offspring seed mass (p<0.01). This indicates that seed traits within a family are influenced by the vigor of the mother tree during seed maturation. These maternal effects influenced early seedling growth; larger seeds produced larger seedlings for the first two to three months before growth stabilized among all families. An initial negative relationship (p<0.0001), with low predictive power, between seed mass and relative growth rate also diminished over time. These data support the hypotheses that maternal effects are evident in both P. flexilis and P. aristata and these effects translate into variation in the first several months of seedling growth. Therefore, to correctly assess genetic differentiation among seed sources when determining appropriate genotypes for conservation, we recommend that common garden studies extend for at least seven months to minimize the confounding influence of maternal effects.