Allozymes in evolutionary genetics: self-imposed burden or extraordinary tool?

T HE controversial history of allozyme studies: Alternative heritable forms of enzymes, differing in charge or shape, have been known since the 1940s; these may be alleles of one gene (allozymes) or products of distinct but related genes (isozymes). LEWONTIN and HUBBY (1966), finding an unexpected bonanza of allozyme variation in Drosophila, re-cast the existing debate about the evolutionary meaning of genetic variation in terms of allozymes. A torrent of like data followed; its interpretation was dominated at first by the notorious “neutralist-selectionist” debate. Population-genetic theory alone proved unable to resolve this debate; pure genetic-statistical analyses lacked power to test deviations from neutrality (EwENS and FELDMAN 1976), and neutralist and selectionist models predicted convergent distributions of allelic/genotypic frequencies (GILLESPIE 199 l). Too little biology was present in the debate, and studying the impacts of allozymes on biological mechanisms in the wild promised to help. Mechanistic study of allozymes has indeed ensued, and its practitioners are mostly optimistic. LEWONTIN (199 l) , in contrast, stigmatized allozyme study since 1966 as a “fardel” or frustrating burden. Some others share his skepticism. Such clashing views bespeak varying awareness of what has been found, or else paradigm differences or other communication barriers. Here, I summarize progress in mechanistic allozyme study, critique reservations about it and explore its promise for new research. What has been learned from mechanistic study of allozymes? A thorough review is impossible here. I illustrate points with a subset of well analyzed cases, apologizing to those whose important work is omitted or discussed cursorily. I often cite recent summaries rather than original references. Function of allozymes in metabolic context: Consider a