Gravar-mail: Apples and oranges: avoiding different priors in Bayesian DNA sequence analysis