The evolution of avian cooperative breeding

I am interested in avian social behavior and I ask questions from an evolutionary perspective.  I use a combination of field observations, experiments in natural populations and molecular genetic techniques in the laboratory. In the past 10 years I have had over 30 students participate in research my lab. These students have had the opportunity to study the evolution of social behavior and learn the molecular genetic techniques I use in my lab.

The focus of my research is the acorn woodpecker (Melanerpes formicivorus), which has one of the most complex social systems among vertebrate societies. Social groups consist of up to 15 individuals, including up to seven cobreeding males and four joint-nesting females that compete for the opportunity to mate. Groups also contain non-breeding helpers from prior breeding attempts who delay dispersal. All group members defend engage in territorial defense, feeding of nestlings and collecting acorns that they store in granaries. I collaborate in my research with Walter Koenig at Cornell University and Eric Walters at Old Dominion University. Acorn woodpeckers have been under study at Hastings Natural History Reservation in central coastal California since 1971.


To understand of the evolution and fitness benefits to individuals in complex social groups of acorn woodpeckers, we address the following questions:

  • How do individuals in social groups partition reproduction? Reproductive partitioning, which can range from egalitarian with each individual obtaining about equal reproductive success to high skew where a single individual of each sex monopolizes breeding, becomes an important component of each individual?s fitness.
  • Do theoretical models of reproductive partitioning that were developed mainly for social invertebrates work as well for vertebrates?
  • When and under what circumstances do individuals in social groups attempt to disperse to other social groups?
  • Is incest by helpers always avoided?
  • Do individuals that are genetic parents of offspring produced in a group contribute more to their care and protection?

Previous work with students from Gonzaga University has produced the development and optimization of over 20 microsatellite loci which we have used to genotype nearly 3000 individuals. My students have worked on relatively small sets of groups that are similar in structure (e.g. 2 potentially cobreeding males and a single female breeder) addressing specific hypotheses.

Current projects:

Project 1

Proximate mechanisms of sperm competition and reproductive partitioning. The goal of this project is to determine proximate mechanisms by which skew is maintained and paternity determined by genotyping the sperm present on the perivitelline membrane of eggs. The idea behind this relatively new technique is to identify an allelic record of sperm competition. It involves collecting eggs either prior to or immediately after the start of incubation and removing the perivitelline membrane. The technical hurdles involve eliminating the maternal and/or embryonic DNA attached to the membrane that, due to their higher concentration, can potentially confound PCR amplification of the sperm DNA, and thus matching the remaining DNA to potential sires. One of my students, Laura Seifert is starting working in the Fall of 2013 to optimize DNA extraction. In the case of acorn woodpeckers, the application of this technique over the next few years offers the opportunity to infer information about the mating history of females. In acorn woodpeckers, copulations are rarely observed and likely occur in roosting or nesting cavities, so observational data on mating patterns among group members can not be easily obtained. Genotype data from eggs will be particularly valuable in the following contexts: (a) Is sperm from both males in groups with 2 cobreeder males (and a single breeder female) generally present at fertilization, even though almost two-thirds of nests are sired by a single male? If not, then it would suggest that females copulate with only a single male for each clutch, in which case it is plausible that males could adjust their behavior according to copulatory access (e.g. for degree of effort in feeding nestlings). If sperm from both males is generally present, the question remains as to why one of the males is generally so much more successful at gaining paternity and has implications for breeders not knowing actual paternity. (b) Do sperm present differ between sequential nests involving the same coalition of males? If so, this would suggest that females copulate with a different male for each clutch, offering a mechanistic explanation for the observed switching of paternity between successive clutches. (c) In large, complex groups containing 2 joint-nesting females and/or 4+ cobreeder males this technique will be especially informative, particularly when such coalitions include younger sons that appear not to achieve their fair share of paternity (based on previously collected data). d) Do sperm present differ in sequential nests involving the same coalition of males and breeder females? If so, this would suggest that females are not consistent with which males they copulate with, again offering a mechanistic explanation for the observed switching of paternity between successive clutches.

Project 2:
Reproductive partitioning in complex groups.The other students working in my lab are continuing to investigate the patterns of reproductive bias in large male coalitions (one student) and among complex social groups containing joint-nesting females and large male coalitions (a second student). While only 12% of groups between 2004 and 2008 involved 3 or more cobreeder males, 27% of breeder males lived in coalitions of 3 or more and are thus important to estimating the direct and indirect fitness benefits of cobreeding. Similarly only 15% of groups contain joint-nesting females, but 23% of breeder females live in coalitions. Preliminary data from the complex social groups suggest that nests produced by joint-nesting females are more likely to be multiply-sired than nests of single females and that there is no difference in the frequency of multiple paternity between nests with small vs. large coalitions of cobreeder males. Both of these patterns will be addressed in these student projects.