Date of Award


Degree Name

Biological Sciences


College of Science

Type of Degree


Document Type


First Advisor

Frank S. Gilliam, Co-Chairman

Second Advisor

Leonard J. Deutsch, Dean

Third Advisor

Jeffrey D. May, Co-Chairman

Fourth Advisor

Paul J. Harmon


Shale Barren Rock Cress (Arabis serotina Steele) was listed as a federally endangered species by the U.S. Fish and Wildlife Service in August 1989. As part of a cooperative agreement between the West Virginia Natural Heritage Program and the U.S. Navy, research was conducted in 1994 and 1995 at Little Fork Shale Barren (LFSB) to provide baseline information on the population dynamics of A. serotina and to study the vegetation and physical parameters associated with this shale barren community. Soil samples and plant community data were collected from 40 1-m2 circular plots for herbaceous vegetation and six 0.02-ha circular plots for overstory vegetation. Overstory vegetation was sampled on LFSB and on a mixed hardwood forest located on the north-northwest slope behind LFSB.

Carex pensylvanica dominated the herbaceous layer of the shale barren, whereas Quercus prinus dominated the overstory on both the shale barren and the adjacent forested slope. The dominance of Q. prinus on both slopes is attributable to the low annual precipitation (84 cm yr-1) on the leeward side of the Allegheny Mountains. Stem density of overstory vegetation on the forest slope was approximately double that found on the barren. Basal area was also much greater on the forest slope than on the barren. Seven herbaceous species sampled are considered to be endemic or near-endemic to the mid-Appalachian shale barrens, including A. serotina, Paronychia montana, Calystegia spithamaea ssp. Purshiana, Antennaria virginica, Scutellaria ovata, Allium oxyphilum, and Phlox subulata. Comparison of soils between LFSB and the forest slope showed significant differences in NO3-N and pH of the water extraction (both higher on LFSB) at p ≤ 0.05. LFSB also had significantly higher NH4-N and pH of the KC1 extraction at p ≤ 0.10.

Canonical correspondence analysis (CCA) showed no significant correlation between A. serotina and any of the measured soil variables. CCA did appear to indicate some response of total species composition to a soil NO3-N gradient. Gradients of soil factors appear to vary in a patchy fashion, perhaps dependent on historical and concurrent rates of litter accumulation rather than direct relation to an elevational gradient.

The A. serotina population on LFSB was shown to be extremely variable with respect to both spatial and temporal changes in population structure. The population declined from 880 individuals in 1992 to only 96 individuals in 1995. These population changes appear to result from the sensitivity of this species to high temperatures coupled with low precipitation early in the growing season, but herbivory, fungal infection, stochastic processes, and human disturbance may also be involved. In July 1995, 71% of rosettes and 91% of bolting plants showed evidence of some type of herbivory. Also, mortality was shown to be substantial within a single growing season for both rosettes and bolting plants. Because mortality appeared to be highest from July through August, recommendations are made to expand the August 15 to September 5 sampling period recommended by the A. serotina Recovery Task Force and the Shale Barren Protection Strategy Group.


Arabis – Ecology – West Virginia.

Plant ecology – West Virginia.