A Melting Pot, or Just a Salad? Evaluating Population History Through Gene Flow

Gene flow is superficially a simple concept; we may liken this evolutionary mechanism to a melting pot. At a fundamental level, when individuals from two defined groups exchange genes, these populations have experienced gene flow. New gene(s) are introduced into one or both populations, and the population becomes more diverse through an increase in genetic variance. When teaching evolution at an introductory level, we generally conclude our explanations at this point.

Perhaps you’ve seen this image. Gene flow at its simplest - the brown beetle will mate with a green beetle, introducing new variation to the green beetle population. Beetles for everyone! (but maybe not in your soup pot)  From evolution.berkeley.edu.

Perhaps you’ve seen this image. Gene flow at its simplest - the brown beetle will mate with a green beetle, introducing new variation to the green beetle population. Beetles for everyone! (but maybe not in your soup pot) From evolution.berkeley.edu.

However, identifying gene flow in the past is considerably more difficult. This is particularly challenging in ancient human populations, which is the focus of my research. How do we know that gene flow has taken place, especially when we don’t have first-hand accounts of individuals from certain populations mating with each other? This is a question that has been discussed frequently from both a practical and a theoretical perspective in ecology and evolutionary biology, as well as among anthropologists. As we have come to realize, however, our ability to address gene flow in biological anthropology is often constrained in ways that other biological sciences do not experience.

Much of what we know about gene flow has come from biological fields working with living populations in both natural and laboratory settings. In living populations, gene flow can be documented through genomic sequencing. This has been done in a variety of genera, including examples from mammals (Cox et al., 2008; Buchalski et al., 2015; McLean et al., 2016; Kumar et al., 2017), reptiles (Thorpe & Richard, 2001; Bu et al., 2014, Schield et al., 2015), amphibians (Fontenot et al., 2011), birds (van Els et al., 2014; Harvey & Brumfield, 2015; Battey & Klicka, 2017), sharks (Li et al., 2015), mussels (Ouagajjou & Presa, 2015), moths (Saccheri et al., 2008) and of course the often-utilized Drosophila (Kang et al., 2017). 

While many of these studies may include geographically separate groups of the same species (e.g., studies of human populations are by definition limited to one species), often they are performed through the lens of past or present hybridization between two or more species. For example, Sanders et al. (2014) conclude that a high rate of genetic hybridization between several species of sea snake in Australia is indicative of fragile interspecies reproductive barriers–95% of their sample were hybrids, despite having the phenotype of a particular species. Studies like this and others (e.g., Olave et al., 2011) have used both molecular and morphological evidence to study hybridization. Ackermann et al. (2016) discuss both types of evidence in our own hominin lineage, suggesting that our species has archaic origins in hybridization.

Thanks to biologists in many other fields (and many researchers within our own), biological anthropologists have a good idea of what to look for in DNA when it comes to gene flow. Of course, these methods are not without problems; past and current models do have limitations (see Bossart & Prowell, 1998, for a synopsis). For instance, there are multiple explanations for genetic homogeneity, including gene flow, balancing selection, and shared ancestry. This is where a consideration of multiple potential interpretations is crucial. A salient example in human history is presented in a recent study by Hunley & Cabana (2017), suggesting that some genetic results previously attributed to localized gene flow among humans may actually be the product of other processes (at least in part).

In studies of ancient human populations, we sometimes have an additional hurdle to surmount. What if genomic data are not available? Frequently encountered scenarios include DNA degradation, constraints on the ability to perform destructive analysis, or lack of consent among living individuals to be involved in studies pertaining to their genetic ancestry. In such cases, quantitative genetics methodology proves to be a useful tool. These methods allow us to use morphological traits to address questions about evolutionary mechanisms (including gene flow) by observing trait variance, which is argued to be proportional to genetic variance.

From a quantitative genetic perspective, gene flow should have certain hallmarks. At the start of a gene flow occurrence, new variance may be introduced, resulting in the receiving population having greater than expected variance given the population size. If we observe a population with greater than expected variance, then, this may be evidence of gene flow. As time passes, this effect will become less pronounced. Gene flow has the effect of increasing variance within a population, but decreasing variation between populations. Effectively, consistent gene flow may bring populations to the same or similar allele frequencies unless counteracted by other forces such as genetic drift or natural selection (Futuyma, 2013).

Of course, this isn’t always the case. The aforementioned scenario assumes fairly uniform initial variance among groups. However, as Williams-Blangero & Blangero (1990) point out in an anthropological application of this concept, an absence of initial variance homogeneity with subsequent high gene flow does not necessarily result in increased variance. Without significant time depth or historical knowledge, it can be difficult to tease out the actual population dynamics. But when we do have some of that background information, we can certainly try.

This is the idea behind Relethford & Blangero’s 1990 paper on gene flow using quantitative traits. Using a genetic model by Harpending & Ward (1982), they created a model using population variance statistics from quantitative traits. Their model assumes that variance in quantitative traits is proportional to heterozygosity, that the population is panmictic, and that traits are equally heritable across study groups. It compares expected variance in each population to actual phenotypic variance. As stated previously, if a population exceeds the expected variance, this may be an indication of gene flow.

Examples of contemporaneous pottery associated with Patayan, Hohokam, Mogollon, and Ancestral Pueblo Native Americans. Distinct material culture does not rule out relatedness among populations. Pottery courtesy of the BLM, Arizona State Museum, and Amerind Foundation Museum; photo from www.oldpueblo.org.

Examples of contemporaneous pottery associated with Patayan, Hohokam, Mogollon, and Ancestral Pueblo Native Americans. Distinct material culture does not rule out relatedness among populations. Pottery courtesy of the BLM, Arizona State Museum, and Amerind Foundation Museum; photo from www.oldpueblo.org.

Their methodology is a great way to get at our gene flow questions. The results, however, may prompt further questions about interpretation. This is an issue I encountered in a recent study I presented at the Annual Meeting of the American Association of Physical Anthropologists in New Orleans (the same meeting Kristen chronicled in her last blog post). In my study, I used dental measurements to explore possible gene flow in several populations in Arizona and Sonora, Mexico. In a number of these populations, I found that the biological variance did not match what we know about the archaeological record - or at least, how we’ve interpreted it thus far. While combining biological and archaeological data would make for a more holistic picture, sometimes they do come into conflict. Our interpretations need to be weighed very carefully, as noted by researchers such as Steadman (2001), Sassaman (2010), and Cabana & Clark (2011). 

As an example, if we have archaeological evidence of two distinct material cultures residing in an area, can we automatically assume that the practitioners of these cultures never mated? No, not really. Yet, we make a lot of these assumptions regarding both the biological and archaeological history of populations, and it is worth noting that they need to be constantly re-examined. This problem of incongruence among sources of evidence by which we interpret the past will be a topic I further consider in future papers (and posts to this blog). The problems we encounter in inferring gene flow do not make it a meaningless endeavor. We just have to ensure that we are cognizant of what our data are really saying, and subsequently the limitations of our interpretations.


Ackermann RR, Mackay A, Arnold ML. 2016. The Hybrid Origin of “Modern” Humans. Evol Biol 43(1):1-11.

Battey CJ, Klicka J. 2017. Cryptic Speciation and Gene Flow in a Migratory Songbird Species Complex: Insights from the Red-Eyed Vireo (Vireo olivaceus). Mol Phylogenet Evol 113:67-75.

Bossart JL, Prowell DP. 1998. Genetic estimates of population structure and gene flow: limitations, lessons and new directions. TREE 13(5):202-206.

Bu X, Liu L, Nie L. 2014. Genetic diversity and population differentiation of the Chinese soft-shelled turtle (Pelodiscus sinensis) in three geographical populations. Biochem Syst Ecol 54:279-284.

Buchalski MR, Navarro AY, Boyce WM, Vickers TW, Tobler MW, Nordstrom LA, García JA, Gille DA, Penedo MCT, Ryder OA, Ernest HB. 2015. Genetic population structure of Peninsular bighorn sheep (Ovis canadensis nelsoni) indicates substantial gene flow across US-Mexico border. Biol Conserv 184:218-228.

Cabana GS, Clark JJ. 2011. Migration in Anthropology: Where We Stand. In Rethinking Anthropological Perspectives on Migration. Cabana GS, Clark JJ, (eds.). Gainesville: University Press of Florida. Pp 3-15.

Cox MP, Woerner AE, Wall JD, Hammer MF. 2008 Intergenic DNA sequences from the human X chromosome reveal high rates of global gene flow. BMC Genet 9(76).

Fontenot BE, Makowsky R, Chippindale PT. 2011. Nuclear-mitochondrial discordance and gene flow in a recent radiation of toads. Molec Phylogenet Evol 59:66-80.

Futuyma DJ. 2013. Evolution. 3rd ed. Sunderland, MA: Sinauer.

Harpending HC, Ward R. 1982. Chemical systematics and human evolution. In Biochemical Aspects of Evolutionary Biology, Nitecki M (ed.). Chicago, IL: University of Chicago Press. Pp 213-256.

Harvey MG, Brumfield RT. 2015. Genomic variation in a widespread Neotropical bird (Xenops minutus) reveals divergence, population expansion, and gene flow. Molec Phylogenet Evol 83:305-316.

Hunley K, Cabana GS. 2017. Beyond Serial Founder Effects: The Impact of Admixture and Localized Gene Flow on Patterns of Regional Genetic Diversity. Hum Biol Open Access Pre-Prints.100. http://digitalcommons.wayne.edu/humbiol_preprints/100

Kang L, Garner HR, Price DK, Michalak P. 2017. A Test for Gene Flow among Sympatric and Allopatric Hawaiian Picture-Winged Drosophila. J Mol Evol. doi:10.1007/s00239-017-9795-7.

Kumar V, Lammers F, Bidon T, Pfenninger M, Kolter L, Nilsson MA, Janke A. 2017. The evolutionary history of bears is characterized by gene flow across species. Sci Rep 7:46487.

Li C, Corrigan S, Yang L, Straube N, Harris M, Hofreiter M, White WT, Naylor GJP. 2015. DNA capture reveals transoceanic gene flow in endangered river sharks. Proc Natl Acad Sci U S A 112(43):13302-13307.

McLean BS, Jackson DJ, Cook JA. 2016. Rapid divergence and gene flow at high latitudes shape the history of Holarctic ground squirrels (Urocitellus). Molec Phylogenet Evol 102:174-188.

Olave M, Martinez LE, Avila LJ, Sites Jr. JW, Morando M. 2011. Evidence of hybridization in the Argentinean lizards Liolaemus gracilis and Liolaemus bibronii (IGUANIA: LIOLAEMINI): An integrative approach based on genes and morphology. Molec Phylogenet Evol 61:381-391.

Ouagajjou Y, Presa P. 2015. The connectivity of Mytilus galloprovincialis in northern Morocco: A gene flow crossroads between continents. Estuar Coast Shelf Sci 152:1-10.

Relethford JH, Blangero J. 1990. Detection of Differential Gene Flow from Patterns of Quantitative Variation. Hum Biol 62(1):5-25.

Saccheri IJ, Rousset F, Watts PC, Brakefield PM, Cook LM. 2008. Selection and gene flow on a diminishing cline of melanic peppered moths. Proc Natl Acad Sci U S A 105(42):16212-16217.

Sanders KL, Rasmussen AR, Guinea ML. 2014. High rates of hybridisation reveal fragile reproductive barriers between endangered Australian sea snakes. Biol Conserv 171:200-208.

Sassaman KE. 2010. Bridging the Empirical Divide of Human Variation Research in the New World. In Human Variation in the Americas: The Integration of Archaeology and Biological Anthropology. Auerbach BM (ed.). Center for Archaeological Investigations, Occasional Paper No. 38. Carbondale, IL: Southern Illinois University.

Schield DR, Card DC, Adams RH, Jezkova T, Reyes-Velasco J, Proctor FN, Spencer CL, Herrmann H-W, Mackessy SP, Castoe TA. 2015. Incipient speciation with biased gene flow between two lineages of the Western Diamondback Rattlesnake (Crotalus atrox). Molec Phylogenet Evol 83:213-223.

Steadman DW. 2001. Mississippians in Motion? A Population Genetic Analysis of Interregional Gene Flow in West-Central Illinois. Am J Phys Anthropol 114:61-73.

Thorpe RS, Richard M. 2001. Evidence that ultraviolet markings are associated with patterns of molecular gene flow. Proc Natl Acad Sci U S A 98(7):3929-3934.

van Els P, Spellman GM, Smith BT, Klicka J. 2014. Extensive gene flow characterizes the phylogeography of a North American migrant bird: Black-headed Grosbeak (Pheucticus melanocephalus). Molec Phylogenet Evol 78:148-159.

Williams-Blangero S, Blangero J. 1990. Effects of Population Structure on Within-Group Variation in the Jirels of Nepal. Hum Biol 62(1):131-146.