VARIATION IN CERTAIN TAXA OF SPHAGNUM FROM THE ATLANTIC COASTAL PLAIN

Abstract

  Taxonomic problems concerning poorly known and variable taxa of Sphagnum from the Atlantic Coastal Plain were investigated using methods drawn from biometry, experimental taxonomy, and genecology. The methods used in the study of character variation included intensive sampling within individual plants, measurement of microscopic characters, and plotting of histograms and scatter diagrams. Follow-up studies, tailored to the variation observed in each pair of taxa studied, included reciprocal transplants, analysis of herbarium specimens using an adaptation of the hybrid index and an index of dispersion, collections along a local environmental gradient and culture under uniform conditions.

  It was not the intent of this study to carry out systematic revisions of the selected taxa. However, certain specific implications for the taxonomy of the taxa studied were revealed. Variants of S. lescurii Sull. that resemble S. cyclophyllum Sull. & Lesq. could be separated from the latter. Sphagnum fitzgeraldii Ren. & Card. and S. torreyanum Sull. were found to be separated from S. cuspidatum Hoffm. No evidence, however, could be found to support the continued recognition of S. trinitense C. Müll. and S. cuspidatum as separate species. Based on evidence of overlapping characters within plants, erratic variation, and discordant variation in herbarium specimens, the two taxa should be synonymized. The two varieties of S. macrophyllum Brid., var macrophyllum and var. floridanum Aust. should be retained. Habitat forms of S. perichaetiale Hampe and the two varieties of S. imbricatum Russ. should not be recognized.

  An analysis of variance showed that in three taxa variation among individual branch leaves contributed significantly more to the overall variation observed in a clump than did variation among branches or among plants. The evidence needed to support different infraspecific categories in Sphagnum is discussed, and the potential for misuse of mixed collections as evidence in this regard is emphasized.