Comparison of isozymes and quantitative traits for evaluating patterns of genetic variation in purple needlegrass (Nassella pulchra)

Decimation of the California (U.S.A.) prairie has increased interest ill the conservation and ecological restoration of grasslands composed of native perennial species. Widespread plantings are being attempted, and the potential for contaminating existing remnant populations with maladapted germplasm has led to concerns about The use of nonlocal seed sources. We evaluated regional patterns of isozyme and quantitative trait variation in the native perennial grass Nassella pulchra for the purpose of developing recommendations about the spatial scales over which seed can be translocated Seed was collected from 10 remnant native populations Progeny from all 10 populations were scored for isozyme variation with 11 stains, and progeny from 8 populations were planted in a common garden to evaluate variation for 11 quantitative (polygenic) traits. The correspondence between isozyme and quantitative traits variation and the relationships of both types of variation to geographic distance and climate were explored by Mantel test regressions. Populations were strongly differentiated for both isozymes and quantitative traits, but cluster analysis based on each type of data did not result in the same population groupings. This lack of congruence was further demonstrated by the nonsignificance of the regression of Hedrick's distances for isozymes on Mahalanobis distances for quantitative traits Quantitative trait variation was strongly, associated with climatic variables, whereas isozyme variation was not. This suggests that the relative importance of genetic drift and selection in shaping patterns of genetic differentiation may depend upon the type of trait evaluated. Quantitative traits are potentially better indicators of adaptation to regional and local environmental variation, and thus the usefulness of isozymes for making recommendations about the spatial settles over which seed of N. pulchra can be translocated may be limited. Unfortunately, obtaining data on patterns of quantitative traits variation is often time- and labor-intensive. The close association of quantitative trait variation with regional climatic variables indicates that an index based on readily obtainable climatic information might aid restorationists in making rapid decisions about appropriate spatial scales for translocating native grasses.