CAUSES AND EFFECTS OF DIVORCE IN THE BLUE TIT PARUS-CAERULEUS

1. Two hypotheses to explain why divorce in birds may be adaptive are the 'incompatibility hypothesis' and the 'better option hypothesis'. At least two more, non-adaptive, hypotheses exist: the 'accidental loss' and the 'forced divorce' hypothesis. We propose a third non-adaptive hypothesis 'the musical chairs hypothesis.' 2. After making predictions to make it possible to distinguish between these hypotheses we analysed five blue tit Parus caeruleus populations. Divorce rates varied between 8 and 85%. 3. Birds that divorced laid significantly later and tended to have raised fewer young in the year before the divorce compared to pairs that stayed together. These latter did better than the population mean. 4. Females, but not males, improve reproductive success after a divorce compared to birds that remained faithful. 5. Divorce rates decrease with female, but not with male age. This suggests that females divorce males, rather than the other way round and that divorce in blue tits can be explained through the better option hypothesis. 6. Benefits of divorce do not differ between study populations in the same way as divorce rates, implying that the same hypothesis can be invoked to explain divorce in all populations. 7. We suggest that differences in divorce rate are the result of differences in the costs of finding a new partner. These costs are probably mainly influenced by the social organization during the non-breeding season: in populations where most pairs remain on the territory search costs are high and divorce rates low; in populations where birds flock outside the breeding season search costs are low and divorce rates high.