The determinations made by Mindell, Shultz and Ewald regarding the anc
estral host for immunodeficiency retroviruses, and their conclusion th
at monkeys acquired their infections as a result of a host-switch from
humans, do not withstand rigorous scrutiny. Their hypothesis requires
the complete uniformativeness of third position transitions and of ga
pped regions in the alignment. When all of the data are permitted to c
orroborate or refute relationships, optimizing hosts on the viral phyl
ogeny renders either equivocal statements or an unequivocal simian anc
estry. However, merely optimizing hosts as characters on the viral phy
logeny is illogical. Not only does this treat hosts as dependent on th
e viruses (instead of the reverse) but it ignores 15 years of methodol
ogical developments specifically designed to answer questions regardin
g cospeciation or host-switching. (C) 1997 The Willi Hennig Society.