Cristy Gelling is Communications Director at the GSA, a science writer, and a lapsed yeast geneticist.
Detail from Orange Electrophoresis, watercolor painting by Michele Banks 2016.

The beautiful cover of the August issue of GENETICS was created by artist Michele Banks to commemorate the fiftieth anniversary of a pivotal moment in the history of evolutionary biology: the 1966 publication of a pair of GENETICS papers using protein electrophoresis to reveal that natural genetic diversity is bountiful. Thanks to a conversation between two specialists from different fields, these papers delivered a jolt of data to a stagnating debate. As part of the GENETICS Centennial celebrations, Brian Charlesworth, Deborah Charlesworth, Jerry Coyne, and Charles Langley discuss this work and its impact in “Hubby and Lewontin on Protein Variation in Natural Populations: When Molecular Genetics Came to the Rescue of Population Genetics

In 1966, two contrasting views of genetic variability held sway. The “classical” view posited that mutations were rare and harmful, and that most individuals in a population are homozygous for the wild type version of a gene. The “balance” view held that balancing selection often maintained multiple common versions of a gene in a population, with many individuals being heterozygous. But the debate was stuck. “Population genetics seemed doomed to a perpetual struggle between alternative interpretations of great masses of inevitably ambiguous data,” wrote Lewontin in 1991.

Part of the trouble was a lack of hard data. In the age before DNA sequencing, describing genetic variation in the wild was a piecemeal task, only possible in a few select cases. Geneticists could study the variability of specific traits with a clear genetic basis, such as snail shell patterns or human blood groups, or they could study major chromosomal variants, such as inversion polymorphisms in wild fruit flies. They could infer the extent of variation through statistical analysis of breeding experiments. But without the ability to isolate specific genes, there was no way to perform an unbiased survey of genome-wide variation.

As a student of pioneering evolutionary geneticist Theodosius Dobzhansky, Richard Lewontin was keenly aware of this problem, but could not find the right method to solve it. Then, a few years after graduating, he visited the University of Chicago and met a biochemist, John Hubby, who was analyzing proteins from fruit flies using electrophoresis. Hubby would grind up individual flies, run the protein extract on gels, and detect specific proteins through their enzymatic activity. Each protein migrated through an electric current applied to the gel at a characteristic speed. But if the same protein differed slightly between individuals —whether in charge, size, or shape—its migration pattern might be altered. Lewontin realized the biochemists already had the missing technique that population geneticists sorely needed:

“…not enough credit is given to the effect of talking to other people and dealing with other people. There’s too much emphasis on the great creative act of a great mind and it’s not like that. […] Here I was with a problem looking for a solution and here a guy was with a solution looking for a problem, and we got together. Many of the things we did in our lab in Chicago arose in the course of a conversation.”

Richard Lewontin, 2004

So convinced was Lewontin of the importance of this insight that he moved his lab to the University of Chicago, where he collaborated with Hubby. The pair quickly began surveying protein variability in Drosophila pseudoobscura, a wild cousin of the famous laboratory fruit fly beloved by geneticists. As Lewontin had hoped, they found electrophoretic variation in D. pseudoobscura proteins showed simple Mendelian inheritance, meaning the protein differences they were measuring corresponded directly to genetic variants. But to their surprise, around a third of the surveyed genes varied between individuals. To quantify this variation across five different D. pseudoobscura populations, they developed a simple summary of genetic diversity, “H” or expected heterozygosity, that is still widely used today. H provides an estimate of the chance that any two alleles at one gene are different, and with this measure, Hubby and Lewontin demonstrated substantial natural variability in all the sampled populations.

But was the richness of protein variation just a fruit fly quirk? The beauty of the new approach was it could be applied immediately to any organism, not just those, like Drosophila, with long histories in the lab:

“Here was a technique that could be learned easily by any moderately competent person, that was relatively cheap as compared with most physiological and biochemical methods, that gave instant gratification by revealing before one’s eyes the heritable variation in unambiguously scoreable characters, and most important, could be applied to any organism whether or not the organism could be genetically manipulated, artificially crossed, or even cultivated in the laboratory or greenhouse.”

Richard Lewontin, 1991

Charlesworth et al. write of the method’s immediate impact: “It triggered an explosion of “find ’em and grind ’em” studies of variability in natural populations of numerous different species, from bacteria to humans, which showed that the levels of variability originally found in Drosophila and humans were not unusual.”

It also turned out to be a boon for understanding the evolutionary histories of organisms without fossil records. In a parallel eruption of protein electrophoresis studies, geneticists combined the protein data with measures of “genetic distance” between species (such as Nei’s “D”) to estimate phylogenies and divergence times between populations and species.

Ironically, Lewontin was wrong in predicting the methods would break the deadlock of the classical/balance debate. Instead, the influx of data raised new questions and complications, and stimulated new avenues for debate. Hubby and Lewontin’s method is the direct ancestor of today’s studies of genetic variation using genome sequence data, but the field’s fundamental question remains active: What are the most important causes of genetic variation?

The solution to the problem of natural diversity may well emerge from another meeting between scientists from different fields, drawn into a new conversation.


Charlesworth, B., Charlesworth, D., Coyne, J. A., & Langley, C. H. (2016). Hubby and Lewontin on Protein Variation in Natural Populations: When Molecular Genetics Came to the Rescue of Population Genetics. Genetics,203(4), 1497-1503. DOI: 10.1534/genetics.115.185975

Hubby, J. L., & Lewontin, R. C. (1966). A molecular approach to the study of genic heterozygosity in natural populations. I. The number of alleles at different loci in Drosophila pseudoobscura. Genetics, 54(2), 577.

Lewontin, R. C., & Hubby, J. L. (1966). A molecular approach to the study of genic heterozygosity in natural populations. II. Amount of variation and degree of heterozygosity in natural populations of Drosophila pseudoobscura.Genetics, 54(2), 595.

Lewontin, R. C. (1991). Twenty-five years ago in Genetics: electrophoresis in the development of evolutionary genetics: milestone or millstone?. Genetics,128(4), 657.

    Leave a comment