Complete sequences of genome segment 7 (S7) from six isolates of epizootic hemorrhagic disease virus serotype 1 (EHDV-1) and 37 isolates of serotype 2 (EHDV-2) were determined. These isolates were made between 1978 and 2001 from the southeast, mid-Atlantic, Midwest and intermountain United States. Analysis of the S7 sequence similarities showed 98.1% identity among the EHDV-1 isolates and 91.0% identity among the EHDV-2 isolates. Comparison of the deduced amino acid similarities showed an even greater degree of similarity among the isolates (100% among the EHDV-1 isolates and 98.9% identity among the EHDV-2 isolates). There was only 75.8% identity between the EHDV-1 and EHDV-2 isolates at the nucleic acid level; however, there was 93.7% identity between the two groups at the amino acid level. The ratio of non-synonymous to synonymous nucleotide indicates a strong selection for silent substitutions. There was no evidence for reassortment between EHDV-1 and EHDV-2 isolates. The high degree of conservation of S7 gene codons and the VP7 protein, suggests that little variation is allowed in preserving the function of this protein. The high degree of conservation also validates the use of diagnostic tests for EHDV based on S7 and VP7.