The first to describe phylogenetic maximum likelihood methods, according to Joe Felsenstein

(correction: my title is incomplete, it should include for molecular sequence data :)

    Theoretical work in Berkeley has been less widely noted. It started with the Jukes­Cantor distance and the stochastic model derived from it. Shortly afterward, Berkeley's famous statistician Jerzy Neyman became involved, owing to contact with Allan Wilson. He was the first to describe maximum likelihood phylogeny methods for molecular sequence data, a development for which I often mistakenly get the credit.

Taking Variation of Evolutionary Rates Between Sites into Account in Inferring Phylogenies -- Joseph Felsenstein at the Journal of Molecular Evolution, Volume 53, Numbers 4-5