Volume 59 – Issue 1 (March 1994)

Isoetes hyemalis, sp. nov., is described from deciduous and mixed forest swamps in the southeastern United States. Populations are known from Virginia, South Carolina, and North Carolina, with the greatest number occurring in the latter. The tetraploid I. hyemalis is found primarily on the Coastal Plain, with a few populations occurring in lower elevation Piedmont streams. Ecological, morphological, and geographic characteristics are described and discussed. Speculation on the origins and affinities of this new species is offered.

The vegetation, as recorded in the Bright Expedition linear survey of southern Middle Tennessee, was conducted during spring 1807 before white settlement. Thirty-five taxa (602 trees) were named in 502 miles of lines in a 1,205 square mile area of the southeastern Highland Rim and southern Central Basin. Agreement between surveyors was good for most species, but unexplained percentage differences occurred on lines across similar topography/soils/vegetation. Diameter distributions approximate those in the virgin Dick Cove nearby and median diameter exceeds those of commercial forests by four inches. Tree composition approximates that from modern inventories with mainly oak forests on the flat-rolling Rim, oak and mixed forests in the dissected Rim and outer Basin and oaks and redcedar prominent in the inner Basin. Composition of known upland community types is averaged in this small, linearly arranged sample set. Non-forest open areas, shrubby and areas of small growth occupied about 12 percent of the flat-rolling Rim; some of these were open barrens for which the area is well known.

Practically no information is available regarding the population size and habitat requirements of Clematis socialis Kral, a federally-listed endangered species. This study was conducted to document population size and habitat characteristics for each of the four known populations. Population mapping revealed that most (75%) of the 17,866 stems counted occurred at a single site, and no plants were found on one site that had been grazed by cattle. Soils associated with all sites were mapped as Conasauga or Firestone series. A detailed study of the largest population included analysis of soil characteristics, light level patterns, and associated plant species. Neither soil chemical traits, soil textural traits, nor light levels varied significantly between microsites containing and lacking C. socialis. Examination of associated plant species in the largest population revealed that Phloxgla glaberrima was an indicator of C. socialis habitat on both macro- and micro-scales. Other species showing increased abundance in the zone containing C. socialis included Cynodon dactylon, Duchesnea indica, and Aster sp. Vernonia noveboracensis was less abundant in plots containing C. socialis.

Long-term herb layer studies require nondestructive sampling, with species identified by above-ground characters. The purpose of this study was to determine the above-ground vegetative characters that best differentiate West Virginia members of Dentaria diphylla, D. heterophylla, and D. laciniata. Eleven morphological characters were measured on West Virginia herbarium specimens and data were subjected to canonical discriminant analysis and analysis of variance. Results showed that species fell into three distinct groups with length and width of basal and cauline leaflets as the most important characters for identification. Dentaria laciniata was identified by its lack of basal leaves and its long, narrow cauline leaflets. The basal leaflets of D. diphylla were similar to but larger than its cauline leaflets whereas the basal leaflets of D. heterophylla were much wider than its cauline leaflets. This study confirmed that these West Virginia members of the Dentaria complex may be identified using above-ground characters, and we provide a key to West Virginia Dentaria species using these characters.

Four species of herbaceous plants (Sanguinaria canadensis L., Jeffersonia diphylla (L.) Pers., Erythronium americanum Ker, and Trillium sessile L.) commonly found in the understory of mature deciduous forests in the eastern United States were subjected to partial defoliation in the spring of 1990 at the time of flowering. Data were gathered on the reproductive responses of control and defoliated plants in 1990 and again in 1991. Partial defoliation had little effect on reproduction in any of the species in 1990, but 50% defoliation significantly reduced multiple fruiting by J. diphylla in 1991, while 67% defoliation significantly diminished three measures of reproductive performance in T. sessile. Defoliation of 50% did not influence reproduction in E. americanum in either year. Experimental defoliation (50%) did not affect reproduction in S. canadensis in 1990, but complete defoliation by deer and ground hogs eliminated reproduction in 1990 and 1991. In these four species the negative consequences of herbivory on reproduction are often delayed, and may not be significant until the level of defoliation reaches at least 50%.

Phenology, population dynamics, survivorship, and sexual reproduction were studied in two populations (A and B) of Senecio antennariifolius from 1986 to 1988. Senecio antennariifolius flowered from mid-April to late May; by mid-June, all seeds were dispersed. Number of adults in population A decreased by 15% between 1986 and 1987, and by 49% between 1987 and 1988. However, due to a large number of seedlings in 1988 the total number of genets remained nearly constant over a three-year period. In population B, the number of adults decreased by 17% between 1987 and 1988, but the number of seedlings in 1988 more than replaced the number of adults that died. Adults produced flowers when they had 4 rosettes in population A, and 10 rosettes in population B. Mortality was greatest among small adults, and was caused by erosion, invasion of Lonicera japonica, and possibly failure to produce new basal rosettes.

Vegetation present in three abandoned rice fields of the Winyah Bay Estuary at the Belle W. Baruch Institute for Marine Biology and Coastal Research, Georgetown County, South Carolina, observed in 1968 was compared with the vegetation present in 1987-91. Species composition in the three marshes changed little in 20 years. The index of similarity is high (greater than 0.80) for each site. There has been a decline in the total number of species present in the Airport and Alderly Marshes. Ninety-four percent of the species observed in 1968 and 1991 are indigenous to the United States. Phragmites australis, an aggressive rhizomatous perennial, is rapidly colonizing the Airport and Alderly Marshes.

Noteworthy Collections: Arkansas and West Virginia

Those of us who boast about the dramatic advances of modern ecological science too often forget, as the saying goes, that we can see so far today only because we can stand on the shoulders of those who went before. James R. Troyer’s book on the life and work of B.W. Wells is a forcible reminder of this fact.