Genetic Variation and Its Management Applications in Eastern U.S. Feral Horses

Abstract

We used electrophoretic and immunologic techniques to analyze blood samples collected from feral horses (Equus caballus) on 4 eastern United States barrier islands. Genetic variation measured in the island herds was similar to that reported in domestic horse breeds. No deficiency in mean multilocus het- erozygosity at 10 protein loci (H = 0.319-0.400) was apparent. Mean number of alleles per locus ranged from 2.2 to 4.4 for 10 protein loci and from 2.7 to 4.8 for 6 red blood cell loci. Greatest genetic resemblance existed between Assateague Island, Virginia, and Cumberland Island, Georgia, horses (S = 0.830). The herd on Shackleford Banks, North Carolina, was least similar to the other feral herds (S = 0.745-0.793), but a small sample size (n = 4) could have affected our results. Overall, the feral herds did not appear to be genetically unique at the loci that we sampled. Local historic and economic values of feral horses might require maintenance of populations on some barrier islands. However, based on ecological and genetic criteria, we think that these populations should be reduced in size. We used models to estimate minimum effective population sizes for 3 island horse herds with different harem structures, age structures, or management regimes. We estimated that a minimum of 72 animals should be retained on Assateague Island National Seashore, Maryland; 122 animals on Cumberland Island, Georgia; and 155 animals on Chincoteague National Wildlife Refuge, Virginia, to limit genetic loss to <1% per generation.