An indirect test of inbreeding depression in the termites Reticulitermes flavipes and Reticulitermes virginicus

We analyzed tandem-running pairs of the termites Reticulitermes flavipes and Reticulitermes virginicus utilizing 13 and 12 microsatellite loci, respectively. Newly formed pairs in both species were significantly related to one another, but this average relatedness was considerably higher in R. flavipes (0.130 vs 0.060). These average relatedness levels resulted from some tandem pairs forming between nestmate termites: more than one quarter of all R. flavipes pairs (26.1%) met this criterion, while this was the case for only about one of every 20 R. virginicus pairs (5.1%). The likelihood that termites paired with siblings was inversely related to the inferred dispersal ability of the two species. F _ST, measured over identical spatial scales, was significantly higher in R. flavipes (0.034) than in R. virginicus (0.008). A comparison in R. flavipes of the observed proportion of nestmate pairs observed during tandem running vs the proportion found in established colonies revealed a significant excess of close relatives when pairs were first formed. There are two possible causes of this discrepancy: inbreeding depression (ID) may eliminate inbred colonies early in development, or related pairs may part late in the tandem-running phase or after it is completed. The latter explanation of inbreeding avoidance implies either historical or contemporary ID, and these results therefore suggest that, either directly or indirectly, ID could be a more potent force in the evolution of termite mating systems than is generally appreciated. This work was funded by grants from the United States Department of Agriculture National Research Initiative Competitive Grants Program (nos. 00-35302-9377 and 2002-35302-12490).