Breeding behaviour of a tropical bird: a study of the blue-throated Bee-eater (Merops viridis) using relational database and DNA fingerprinting.

Abstract

The breeding behaviour of the Blue-throated Bee-eater was studied at two colonies in Peninsula Malaysia during 3 breeding seasons, with particular emphasis on pair behaviour, mixed reproductive strategies and nestling competition. This is the first study of vertebrate social behaviour and ecology to contain the documentation of a relational database. This was designed to store and manipulate all data obtained from regular captures and biometric measurements of adults and nestlings and from observations of adults. DNA fingerprinting was used to establish the true genetic relationships between nestlings and their social parents: most nestlings were genetic offspring (72%). Nestlings were classified as illegitimate offspring using 95% confidence intervals of the band sharing coefficient and number of unexplained nestling bands as criteria. Very few if any nestlings were sired by an extra-pair male (fewer than 5%). Behavioural evidence of strong cooperation between pair members throughout the breeding season supports the DNA fingerprinting results of no confirmed case of offspring fathered by extra-pair males (extra-pair offspring; EPO). The Blue-throated Bee-eater probably has a near monogamous mating system. Most illegitimate nestlings had been 'dumped'. They were either the result of intra-specific nest parasitism (INP; 7%) or of 'quasi' parasitism (the offspring of the pair-male and an extra-pair female; 7-12%). INP by relatives of the hosts could have explained some intermediate band sharing coefficients. Anti-INP behaviour was demonstrated when experimentally 'dumped' eggs were almost always expelled before the onset of laying, but never afterwards. DNA fingerprinting showed that relatives may roost together and that related males may nest close together. Compared with other colonial Bee-eaters, M. viridis had low levels of helping-at-the-nest and EPO, but similar or higher levels of INP. The high nestling mortality in Blue-throated Bee-eaters was explained by a combination of three hypotheses, some of which were tested by experiment. (1) Insurance: extra-eggs are needed to counter hatch failure. (2) Brood reduction (including resource tracking): in times of food constraint, the laterhatched nestlings in asynchrously hatched broods starve. (3) Anti-INP hypothesis: these later-hatched nestlings are eliminated because they are likely to be illegitimate. Hatching failure was about 1 in 3 eggs overall. Help from the male allows an early onset of incubation which results in asynchronous hatching. Nestling hunger was shown to be a proximate factor affecting runt mortality both directly through competition and indirectly through nestling aggression. The demise of runts was delayed when conditions improved. Blue-throated Bee-eater broods are severely limited by food. Under this severe brood size constraint, breeding females may increase reproductive output by 'dumping' their last egg. This leads to the high frequency of INP observed in Blue-throated Bee-eaters. An early onset of incubation also gives the first-laid egg(s) a temporal developmental advantage over subsequently 'dumped' parasitic eggs. The 'dumped' nestlings are eliminated by starvation and siblicide, which may itself be an adaptation to INP to eliminate of unrelated nestlings.