Price et al. [(1999). J. Bacteriol. 181:2358–2362] conducted a genetic study of the toxin transport protein (PA) of Bacillus anthracis, the bacterium that causes anthrax in humans. Within the 2294-nucleotide gene in 26 strains they identified five point mutations—two missense and three synonyms—among different isolates. Necropsy samples from an anthrax outbreak in 1979 revealed a novel missense mutation and five unique nucleotide changes among ten victims. The authors concluded that these data indicate little or no horizontal transfer between different B. anthracis strains.

On what basis did the authors conclude that evidence of horizontal transfer is absent from their data?