We analyzed survival of breeding Greater Flamingos, Phoenicopterus ruber roseus, using the capture histories of 2000 breeding birds ringed as chicks and resighted at their natal colony in the Camargue, southern France. As found in previous analyses, recapture probability varied according to year, sex, and age of the bird, and annual survival was strongly affected by winter severity. However, by using a much larger data set than in earlier analyses, we detected previously nonsignificant effects. Indeed, for the first time, sex and age of the bird were found to influence annual survival probability. We tested the hypothesis that the observed sex-related difference in survival corresponded to asymmetric costs of reproduction. A model including a cost of first observed reproduction on survival in young females only provided the best fit to the data and explained the majority of the sex-related difference in survival of birds Ͻ7 yr old. Because a cost of reproduction may be partially masked by birds that have already bred undetected, we estimated the proportion of experienced females among those observed breeding for the first time. This proportion varied with the age of the birds and was used to calculate the expected cost of early recruitment. Such a cost of early reproduction may have contributed to the evolution of deferred breeding in females. Survival of experienced females was higher than that of males, with the difference being more pronounced in early age classes. Age had a significant positive effect on survival probability of birds.

We studied movements of individually marked greater flamingos (Phoenicopterus ruber roseus) born in the Camargue, southern France, between their two most important breeding colonies in the western Mediterranean (Camargue and Fuente de Piedra, Spain) from 1986 to 1992. The two sites differ in the frequency with which they offer suitable conditions for breeding. Flamingos have bred each year in the Camargue since 1974, but in only 12 of the past 22 years at Fuente de Piedra. Higher colony fidelity is thus expected in the less variable environment (Camargue), but if dispersal occurs competition might be an important factor causing this dispersal. Following years during which breeding birds in the Camrgue were disturbed (1988 and 1990) a higher proportion of adults changed colonies between breeding attempts (= breeding dispersal, 12.4%), while only 0.4% of flamingos breeding in the Camargue dispersed in the other years. As expected, flamingos breeding at Fuente de Piedra showed a higher rate of breeding dispersal (8.14%). No differences were observed between males and females. The importance of breeding failure as a factor causing breeding dispersal in flamingos was also confirmed by the movements of individual birds. The proportion of young flamingos that moved from their natal colony to start breeding at Fuente de Piedra (= natal dispersal) was independent of sex and age, but increased when breeding access to the Camargue colony was more difficult. However, natal dispersal was also higher in 1988 and 1990 (40.5%) than in the remaining years (1.2%), as was breeding dispersal. We discuss possible ways in which the increased natal dispersal among inexperienced birds could be linked with the increased breeding dispersal of adults in the same year.

Annual local survival of adult Greater Flamingos, Phoenicopterus ruber roseus, breeding at their natal colony in the Camargue (southern France), was studied using ringed birds observed at the colony from 1983 to 1991. Survival and probability of resighting were estimated separately, using capture—recapture models. Because all the birds we observed were nesting, differences in probability of resighting reflect differences in probability of breeding. Winting severity significantly affected survival. Annual local survival was 93% during normal years, but in 1984—1985, following a severe cold spell, survival fell to 76%. Sex and age (range 4 to 14 yr old) had no significant effect on survival. Resightin rates varied between 11% and 88% during the study period. There were significant additive effects of year, sex, and age on resighting. The probability of resighting increased with age in both males and females, reflecting the progressive access to regular reproductive status in this population. However, males had a higher probability of being resighted than females, probably due to their higher rates of mate switching and renesting following breeding failure

In many colonial waterbirds, reproductive success is affected by water levels around the colonies. In a study of the Greater Flamingo (Phoenicopterus ruber roseus) in the Camargue, southern France, we examined annual variation in water levels around the breeding colony site between 1984 and 1991 in relation to (1) the number of breeding pairs, (2) colony productivity (fledglings per breeding pair), and (3) physical condition of fledglings. The number of breeding pairs each year was highly variable and was positively correlated with high water levels during the March—July period. There was no significant relationship between the number of breeding pairs and the number of chicks fledged. Although the productivity of the colony was not affected by water levels during the chick—raising period, fledglings were heavier in years with high water levels. There was no relationship between colony productivity or the absolute number of chicks fledged and the average body condition of chicks in the creche. Differences between the 1984 and 1985 cohorts in the frequency of young birds making breeding attempts (i.e., ages 5 and 6 yr) suggested that body condition at fledging may have important life history consequences. Estimation of reproductive performance in colonial waterbirds, usually limited to the average nest productivity, could be improved by considering average body condition of chicks at fledging as well.

The development of the (24 hour) time-budgets of foals was studied as part of a broader investigation of the time-budgets of Camargue horses living in semi-liberty. Nine categories of activity were studied; resting postures (lateral and sternal recumbency, and standing resting), foraging, and postures related to movement and orientation (standing alert, walking, trotting, galloping and rolling). The process of development of the time-budgets as a whole was described by using a multivariate technique, correspondance analysis : variations in time-budgets were largely determined by age, but the environment intervened in mid-summer, when the process of development was slowed down. It was suggested, on the basis of other evidence, that the attacks by horseflies were probably responsible for the arrest in development. It was also found that for some individuals there was also a period shortly before weaning when development was slowed down. Foals developed time budgets closely similar to those of sub-adults (horses 1-2 years older than the foals) very soon (2-4 weeks) after weaning. The changes in the time-budgets involved considerable decreases in the time spent in the lying postures (e.g. lying flat occupied 15% after birth and 2% after weaning), and in movement and orientation (standing alert decreased from 31% to 5% in the same period) while foraging increased from 13% to 62%. It is suggested that the milk supply from their mothers allows the foals to invest time in sleep, and in exploration of their physical and social environment. After weaning, most of this time is spent foraging.