THE COSTS OF BEING LATE - CONSEQUENCES OF DELAYING GREAT TIT PARUS-MAJOR FIRST CLUTCHES

1. The hypothesis that late breeding entails reproductive costs to birds was tested by comparing breeding traits of control and delayed great tit Parus major clutches. 2. In our study area, great tits and house sparrows Passer domesticus compete for nesting boxes. Since early March, nesting boxes were checked twice a week and initiated sparrow nests removed, to allow great tits to breed. 'Delayed' great tit pairs were those which, having a territory, had the nesting where they eventually bred occupied by sparrows, and therefore did not start laying until these abandoned the nesting box. Delayed pairs were assumed to be a random sample of the population. Age composition (1-year vs. older birds) and tarsus length did not differ between control and delayed birds. 3. In 1990-92, a total of 40 pairs had their breeding delayed, while 56 acted as controls. The mean laying date of delayed pairs was at least 10 days later than controls. 4. Delayed first clutches were smaller and took longer to hatch than controls, but mean egg volume did not differ between them. Fledging success was lower in delayed nests. Young from delayed broods were lighter at day 15 and reached a lower asymptotic weight. The mean growth rate of the nestlings was negatively related to clutch size, and differed between years, but did not differ between control and delayed broods. 5. Control pairs were more likely to produce a second clutch, if the first one succeeded, or a replacement clutch, if the first one failed, than delayed pairs. The total production of fledglings per pair per season was higher in control pairs. Control pairs were more likely to recruit at least one young into the local breeding population than delayed pairs. However, it is not known whether late fledglings dispersed further. 6. Our results show that delayed pairs produced fewer fledglings, of lower quality, and were less likely to recruit them into the local breeding population than controls. Therefore, the causal relationship between poor breeding performance and late breeding is supported.