Reproductive success in martins (Hirundinidae) : studies of the behaviour and ecology of individuals using DNA fingerprinting

Abstract

The aim of the study was to isolate some of the sources of individual variation in reproductive success in wild bird populations, with particular emphasis on the number of clutches produced per year, and the consanguinuity between parents and offspring. The main study species was the house martin, with some comparative work on the sand martin (Plate I). Both at an individual and a population level double-broodedness was implicated as the annual breeding strategy with the highest fitness benefits for house martins, in terms of total annual output of fledged young. Older females laid earlier than first year females, and in contrast with earlier work on the same species (Bryant 1979. 1988a) there was a hint that small size in males might be associated with increased annual reproductive success. Experimental manipulations of first brood size in house martins indicated that the interval between breeding attempts increased with first clutch size, and that nestlings in enlarged broods grew more slowly and suffered an increased rate of mortality. Enlarging or reducing the size of the first brood also had an effect on the probability of a second clutch in the same breeding season. Pairs with enlarged broods seemed less likely to produce a second clutch, whereas there was an increased chance that pairs with reduced first broods would lay again. Over the time period 1972-1989, house martins breeding at study colonies in Central Scotland have apparently undergone a decline in annual reproductive success. Variation in food abundance apparently had some effects on the timing and success of breeding, but could not fully account for the observed changes. It is debatable whether this decline is real, or a result of changes in food availability, nest site preference, and possibly population age structure of house martins in Central Scotland over the same period. There was no evidence of egg dumping or intraspecific brood parasitism (IBP) in either of the study species. DNA fingerprinting indicated that 38% of house martin broods contained at least one offspring that had been fathered by an extra-pair male, with 15% of all young being unrelated to their putative father. Preliminary results indicated a slightly higher incidence of extra-pair paternity in sand martins. Observations of behaviour indicated that male sand martins and house martins guard their males during the prelaying and lying periods, although the degree of guarding differed between the two species, probably due to differing risks of extra-pair copulations (EPCs). There was some indication that the risk of cuckoldry for male house martins was linked to the intensity of mate guarding. In the absence of the pair male, female house marlins apparently chose whether or not to accept EPCs from males intruding into their nests. House martin males that have been cuckolded do not reduce their parental effort in terms of rate of food delivery to the brood. Extra-pair fertilisations (EPFs) therefore represent fitness or fecundity costs of reproduction for at least some males in both of the study species, and former estimates of apparent male reproductive success in house martins (Bryant 1988a, 1989) must now be revised. Unfortunately the fathers of extra-pair offspring were not identified, but there were indications that male house martins which achieved full paternity in their own families might also be likely to increase their reproductive success through extra-pair fertilisations. This finding is in accordance with the basic assumption of Trivets (1972), that individuals should pursue a mixed reproductive strategy in order to increase their lifetime reproductive success, although if it is older males that increase their fecundity at the expense of younger males then realised lifetime reproductive success in house martins is unlikely to be very different from apparent success. In addition, the pursuit of EPCs may represent an alternative reproductive lactic for unmated house martin males. A preliminary review of consanguinuity in wild birds failed to isolate any consistent themes. For each species, even closely related species within the hirundine family, the observed rates of EPF and IBP seem to be a result of unique interactions between behavioural and ecological factors. Male guarding apparently varies in intensity and effectiveness within and between species, as does the response of females to attempted EPCs. A cross species comparison indicated that the occurence of EPFs and IBP was apparently unrelated to breeding dispersion and breeding system, although within species an increased density of breeding individuals may lead to a higher frequency of non-kin offspring. What is clear is that EPCs probably occur in the majority of bird species, even though in some they rarely if ever lead to EPFs. It la no longer possible to ignore the effects of sperm competition and IBP, and no study of individual reproductive success in wild bird populations can now be considered complete unless it incorporates DNA fingerprinting.