Magnolia virginiana L. is a native North American species that embodies the issues associated with restricted usage, despite being the first magnolia introduced into cultivation. Although there have been several cultivars developed, this species remains seldom cultivated outside of the United States. The cultivars differ in their ability to form adventitious roots from stem cuttings and the difficulty associated with producing the best clones have limited commercial availability. The species has been little utilized in interspecific hybrids, suggesting that further understanding of its pollination biology is needed. Additionally, there have been no attempts to document phenotypic or genetic variation in this species with the objective of improved plant selection. We conducted two experiments to explore the variation in this species.
Magnolia virginiana is widely used in landscapes, but these plants are seedlings that can display variability for growth traits. We defined two ideotypes that are better suited to modern landscapes: a shrub form and a single stemmed tree form, both with profuse flowering. To that end, the major portion of this work focused on the measurement of phenotypic variation to discern underlying genetic variation and to determine optimal selection strategies in this species. A series of controlled pollinations were performed between selected plants of Magnolia virginiana var. virginiana to generate a series of F1 half-sib families for evaluation. These families were evaluated on an accelerated container production system and significant differences (P ≤ 0.0001) were found between them for height, branching, and flowering characteristics. Narrow sense heritability (h2i), phenotypic family mean, and genetic correlations were estimated for these traits and used in the development of selection indexes. Data indicated that formation of two selection indexes could result in our desired ideotypes. The first index involves highly branched, shorter forms with increased flower production. The second index consists of tall plants with reduced lateral branching, increased late flowering, and limited collar sprout formation. Selection for our ideotypes may be possible, but the use of additional germplasm may be necessary to increase genetic variation for these traits.
A semi-hardwood stem cutting propagation experiment was performed to elucidate differences in rooting response between F1 half-sib families in genetic tests. Differences were found between half-sib families and data suggested that the use of IBA in conjunction with ascorbic acid will produce successful results ranging from 0 to 93% that require further experimentation to determine differences in rooting responses within and between varieties of the species. The work presented herein will guide the design of further experiments.
Overall, there is potential for development and cultivar selection of novel Magnolia virginiana forms. Initial tests were important for providing a baseline for the formation selection indexes and guiding future experiments for propagation and production of novel species forms.