Body size response to warming: time of the season matters in a tephritid fly

Whether shrinking body size is a universal response to climate change remains controversial. Moreover, the mechanisms underlying body size shifts are poorly understood. Here, assuming that life history traits evolve to maximize fitness according to life history plasticity theory, we hypothesized that under global warming temperate multivoltine insects should emerge earlier with a smaller body mass in the early growing season, but emerge later with a larger body mass in the late season. We tested this hypothesis by conducting two field artificial warming experiments in an alpine meadow: 1) with one pre‐dispersal seed predator species (tephritid flies, Tephritis femoralis) and its two host‐plant species flowering in early and late growing seasons, respectively, and 2) with the tephritid flies and one host species with a flowering season that occupies parts of both the early and late growing seasons. These experiments were complemented by a microcosm chamber warming experiment, in which increasing and decreasing temperature trends were set to simulate temperature variation pattern in early and late growing seasons, respectively, but photoperiod was held constant. Warming generally advanced the adult emergence and decreased the body size (adult body mass) in the early season but delayed the emergence and increased the size in the late season in both field experiments, indicating that the seasonally different effects of warming on the fly body size was constant regardless of host‐plant identity. The chamber warming resulted in consistent responses of emerging timing and body size to the simulated seasonal warming, demonstrating that the temperature increase per se and its interaction with direction of temperature change, but not other correlated effects, should be primarily responsible for the observed contrasting shifts of body size at different times of the season. Our results indicate that taking into account temperate seasonal patterns of temperature variation could be of general importance for predicting animal body size changes in the warmed future.     

Thermal acclimation of metabolic rate may be seasonally dependent in the subtropical anuran Latouche's frog (Rana latouchii, Boulenger)

We tested the hypothesis that the lack of metabolic thermal acclimation ability in tropical and subtropical amphibians is dependent on season and investigated the effects of body size, sex, time of day, and season on metabolic rates in Rana latouchii. The males were acclimated at 15 degrees, 20 degrees, and 25 degrees C, and their oxygen consumption was measured at 15 degrees, 20 degrees, 25 degrees, and 30 degrees C in all four seasons, with the exception that we did not measure oxygen consumption at 30 degrees C in winter frogs. We also acclimated the males at 30 degrees C in summer for investigating diel variation of metabolic rate. The females were acclimated at 20 degrees and 25 degrees C, and their oxygen consumption was measured at 15 degrees , 20 degrees , 25 degrees , and 30 degrees C in summer. Our results showed that metabolic rates of R. latouchii differed by time of day, season, and acclimation temperature but did not differ by sex if the results were adjusted for differences in body mass. Summer males exhibited a 26%-48% increase in metabolic rates from the lowest values in the seasons. There was a trend of increased oxygen consumption in cold-acclimated males, but it was significant only at 15 degrees and 25 degrees C in summer, autumn, and winter. These results support the hypothesis that thermal acclimation of metabolism is seasonally dependent, which has not been reported in other tropical and subtropical amphibians.     

Effect of winter high temperatures on reproduction and circannual rhythms in hibernating ground squirrels

We tested whether prevention of hibernation in ground squirrels by midwinter exposure to high ambient temperatures influenced timing of the spring phase of reproductive maturation and the phase and period of subsequent circannual rhythms of reproduction and body mass. Exposing hibernating adult male Spermophilus lateralis to 30 degrees C for 6 weeks beginning December 4 advanced the timing of testicular recrudescence by 4-5 weeks, compared to controls left at 4 degrees C. Males exposed to 30 degrees C for 6 weeks beginning at the average time of spontaneous end of hibernation (January 15) reached reproductive maturation at a time intermediate to those of controls and of the December 4 experimental group. However, neither the date of the subsequent fall's body mass peak, the date of the next year's reproductive maturation, nor the periods of circannual rhythms of body mass and reproduction differed among groups. Premature interruption of hibernation appears to allow early expression of reproduction, but does not affect the underlying timing mechanism.     

Sex ratio in ixodid tick collections from Leningrad Province

The analysis of results of many-year regular collections of ixodid ticks has shown that females (56%) were dominant in I. persulcatus while in I. ricinus were dominant males (52%). The ratio between sexes in both species changed noticeably as in individual years so during the imago activity season. In summer collections of I. lividus females were dominant every year, the share of which accounted on the average for 70% and varied from 69 to 75% within different seasons and from 60 to 83% within imago activity season. During hibernation and in spring the share of males and females was nearly adequate.     

Energetic and endocrine mediation of natal dispersal behavior in Belding's ground squirrels

Natal dispersal, the permanent departure of an individual from its birth site, is sex biased in most mammals, with males dispersing at higher rates or over greater distances than do female conspecifics. Because dispersal movements may be energetically expensive, their occurrence should theoretically be influenced by energy availability. Moreover, the male bias typical of mammalian dispersal suggests that this behavior might be mediated by gonadal androgens. Using free-living Belding's ground squirrels (Spermophilus beldingi) as subjects, we provisioned juveniles with extra food to evaluate energetic influences on male dispersal behavior. Provisioning increased body mass and body fat of juvenile males and caused them to disperse at younger ages, but did not affect blood glucose levels. Dispersing males were fatter than same-aged males that had not yet dispersed. Moreover, body fat of provisioned and unprovisioned males did not differ when evaluated relative to the week during which they dispersed, suggesting that there may be a fat threshold for dispersal. In a second experiment, we measured plasma concentrations of testosterone (T) in provisioned and unprovisioned, free-living juveniles to evaluate the hypothesis that male dispersal behavior is activated by concurrent high levels of T. We observed no increase in plasma T associated with dispersal by juvenile males, no sex differences in circulating T among juveniles, and no effects of food provisioning on juvenile T levels. In a third experiment with free-living S. beldingi, we concurrently altered early androgen exposure by treating females with T at birth and manipulated energy availability by food provisioning. Perinatal T-treatment increased the likelihood of dispersal among juvenile females. Provisioning increased body mass and body fat of juveniles and caused males and T-treated females to disperse at significantly younger ages than either their unprovisioned counterparts or the few provisioned control females that dispersed. These results suggest that early T exposure in this species determines the probability of dispersal, whereas the amount of energy an individual has stored as fat strongly influences the timing of dispersal. Early T exposure also appears to cause the timing of dispersal to respond to energy availability and body fat in a male-typical way, possibly by organizing masculine mass and fat thresholds for dispersal.     

Age-class differences in the pattern of hibernation in yellow-bellied marmots,Marmota flaviventris

Summary Age-related differences in the patterns of body temperature regulation during hibernation were found in yellow-bellied marmots. The timing of all entrances into and arousals from torpor was determined from continuous records of thermocouples mounted in each animal's nest box. Older marmots spent more time at high body temperatures following periodic arousals from torpor than did juveniles undergoing their first season of hibernation. In addition, older marmots spontaneously terminated their hibernation seasons in the spring, whereas most juveniles continued to hibernate until either they were emaciated from starvation or they were fed. These two patterns of hibernation reflect age- and size-related differences in the degree to which the animals are constrained energetically and the probability that they can successfully reproduce in spring. The patterns also are consistent with age-related differences in the timing of dormancy in nature.     

Regulation of gut function varies with life-history traits in chuckwallas (Sauromalus obesus: Iguanidae)

We examined the effects of hibernation and fasting on intestinal glucose and proline uptake rates of chuckwallas (Sauromalus obesus) and on the size of organs directly or indirectly related to digestion. These lizards show geographic variation in body size and growth rate that parallels an elevational gradient in our study area. At low elevation, food is available only for a short time during the spring; at high elevation, food may also be available during summer and autumn, depending on rainfall conditions in a given year. We hypothesized that low-elevation lizards with a short season of food availability would show more pronounced regulation of gut size and function than high-elevation lizards with prolonged or bimodal food availability. Hibernating lizards from both elevations had significantly lower uptake rates per milligram intestine for both nutrients, and lower small intestine mass, than active lizards. The combination of these two effects resulted in significantly lower total nutrient uptake in hibernating animals compared to active ones. The stomach, large intestine, and cecum showed lower masses in hibernators, but these results were not statistically significant. The heart, kidney, and liver showed no difference in mass between hibernating and nonhibernating animals. Lizards from low elevations with a short growing season also showed a greater increase in both uptake rates and small intestine mass from the hibernating to the active state, compared to those from high elevations with longer growing seasons. Thus, compared to those from long growing season areas, lizards from short growing season areas have equal uptake capacity during hibernation but much higher uptake capacity while active and feeding. This pattern of regulation of gut function may or may not be an adaptive response, but it is consistent with variation in life-history characteristics among populations. In areas with a short season, those lizards that can extract nutrients quickly and then reduce the gut will be favored; in areas where food may be available later in the year, those lizards that maintain an active gut would be favored. While other researchers have found much greater magnitudes of gut regulation when making comparisons among species, we find the different patterns of change in gut function between different populations of chuckwallas particularly intriguing because they occur within a single species.     

Reproductive performance in female common hamsters

We monitored individual reproductive timing and output in a common hamster population in Vienna over a 3-year period. Animals were live-trapped, weighed, individually marked, and reproductive status was determined at capture. Costs of reproduction were investigated by measuring body condition shortly before hibernation and overwinter survival. Our results indicated that early emerging females had more litters and weaned more offspring per season. Body mass throughout the active season did not differ significantly between females with high and low reproductive output. High reproductive output seemed to affect the duration of the active season. Successful females had a longer postreproduction period before hibernation than less successful ones, probably serving to balance the costs of reproductive effort by extended preparation for hibernation. Also, females that had emerged early in spring and had high reproductive success were more likely to survive the subsequent winter. Hence, we found female common hamsters to vary strongly in maternal investment capacity and to tailor reproductive strategies accordingly.     

Influence of energy stores on activation of reproductive function in male golden-mantled ground squirrels

Summary Male golden-mantled ground squirrels were captured in the field in late summer and placed on either unlimited or restricted (80% normal) food rations in the laboratory until each animal began to hibernate. At entrance into hibernation mean body mass was 274.5 g for the unlimited group and 224.5 g for the restricted group. Only six of 21 males subsequently underwent reproductive maturation during winter. Each of these six received the unlimited ration prior to hibernation, and these six included the five heaviest animals in the experiment. In addition to activating their reproductive systems, these heavier squirrels began hibernation later and ended hibernation earlier than the lighter squirrels. Squirrels that remained reproductively quiescent averaged 25.5% body mass as fat in early spring and thus were not severely limited in energy stores during winter.In another experiment a group of 13 male and female squirrels were housed together in a common outdoor enclosure in order to examine the possible relationship between reproductive condition and social standing of males. Although there appeared to be a social hierarchy among males, each male fattened substantially in fall and became reproductively active during winter.These results suggest that the level of energy stores accumulated as fat prior to hibernation in fall affects the potential of male ground squirrels to breed in spring. This dependence of reproductive development on energy stores may reflect the high energetic costs associated with breeding in males when they emerge from hibernation in spring when food availability is low.     

Annual cycle of energy and time expenditure in a golden-mantled ground squirrel population

Summary We have analyzed seasonal shifts of energy and time allocation in a population of golden-mantled ground squirrels (Spermophilus saturatus) by directly measuring total daily energy expenditure (DEE) with an isotopic technique (doubly labeled water=dlw), and by estimating components of total DEE through an integration of field behavioral observations with laboratory-measured rates of energy expenditure (oxygen consumption) associated with major behavioral and physiological states. Hibernation laster about 7 1/2 months, and the 4 1/2-month activity season consisted of mating, a 28-d gestation of 3–5 young, 5 1/2 weeks of postnatal growth building to a peak in lactation just before the young emerged above ground, an additional 2–3-week period of maternal care before dispersal, and finally restoration of body mass preceding hibernation. Although the hibernation season comprised nearly two-thirds of the year, it involved only 13–17% of annual energy expenditure, leaving about 85% of energy expenditure for the active season. Ground squirrels were actually present on the surface for only about 11% of the year's time, and the foraging time required to obtain the total annual energy supply amounted to only about 2% of the year's time. The squirrels fed mainly on herbs in the early season and hypogeous fungi later; both were used extensively during peak lactation when female energy expenditure and demand were maximal. Average daily foraging time increased steadily throughout the season to a maximum of 28% of aboveground time as availability of greens diminished and fungus predominated in the diet; time availability did not limit foraging since the animals sat on average for 65% of the daily surface time of about 7 h. Timing of reproduction is apparently optimized such that peak reproductive energy demands are matched with maximal food availability and moderate thermal conditions that minimize energy demand. Despite the greater body mass of males, the greatest total DEE (measured by dlw) of any squirrels at any time of year was that of females during peak lactation. For production of young and lactation through above-ground emergence of an average litter of 2.7, females required a total energy increase of 24% above annual nonreproductive metabolism. Yearling females all bred and performed similarly to older females, yet some costs were greater because the yearlings began and ended hibernation at smaller mass, compensated by giving birth later, and finally showed a greater absolute increase in body mass over the active season than older females. Annual metabolic energy expenditure of breeding males was about 18% greater than that of females, due to greater male body mass. Yet the annual energy intake requirement for both sexes was essentially identical (about 42MJ) due to the greater reproductive export by females in the form of newborn and milk. During the mating season males showed wide-ranging exploratory behavior and social interactions, including aggression, that involved considerable locomotory energy expenditures. Although we did not directly account for the energetics of these specific reproductive behaviors, they are critical to male reproductive success and on a daily basis they probably involved much greater energy expenditure than sperm production. Some yearling males avoided these costs by foregoing testicular development, yet they allocated four times as much energy to growth as older males, thereby increasing somatic condition for the future.     

Induction of summer hibernation in the 13-lined ground squirrel shown by comparative serum transfusions from arctic mammals.

A “trigger” substance was again indicated to be present in sera of hibernating animals. Sera from the hibernating 13-lined ground squirrel, hibernating woodchuck, hibernating Arctic ground squirrel, and hibernating Arctic marmot were all capable of inducing the 13-lined ground squirrel to hibernate in the summer, a season when that species would normally be active. The hibernation trigger is thus not species specific. It is effective whether drawn from these two Arctic species of hibernators or drawn from these two species of hibernators from the midwestern states. The normothermic Arctic marmot appears to have an “anti-trigger” substance in its serum in the summer, which impedes fall hibernation in the transfused 13-lined ground squirrel. This is similar to the anti-trigger observed in the summer serum of active 13-lined ground squirrels and active woodchucks. With respect to hypothermia, it was induced in Artic marmots and in Arctic foxes at Point Barrow, Alaska, in summer. Though in such cases body temperatures fell significantly (as in hibernation), no trigger was recovered from their hypothermic sera that could be shown to be capable of inducing summer hibernation in the ground squirrel. Neither was anti-trigger found in the serum of hypothermic experimentals. These latter experiments thus suggest that the release of trigger into the blood during hibernation is dependent on a mechanism more complex than simply lowering body temperature.     

Predispersal reproductive biology of female Osyris quadripartita (Santalaceae), a hemiparasitic dioecious shrub of Mediterranean scrublands

Female Osyris quadripartita plants exhibit uninterrupted reproductive activity throughout the year, due to the long duration of successive stages in the cycle and marked within-crop developmental asynchrony. Cycles corresponding to the flowering seasons of consecutive years overlap in each individual. Flowering takes place in spring, and fruits develop in the dry summer season and ripen at any time of the year. Variation in flowering time explains a negligible proportion of variation in ripening time. The greatest reproductive losses are incurred in the phase extending from closed flowers through unripe fruits, mostly due to ovary abortion. Only 30% of closed flowers eventually reach this latter stage. In contrast, 75% of unripe fruits complete their development, with subsequent dispersal of seeds. The probability of the setting of ripe fruit steadily decreases from early to late season flowers, due to increased ovary abortion rates. Resource limitation in the dry summer season seems responsible for this pattern of selective fruit maturation.     

Body size and social dominance influence breeding dispersal in male Pachydiplax longipennis (Odonata)

1. Dispersal behaviour can be affected by an individual's phenotype, by the environmental or social context they experience, and by interactions between these factors. Differential dispersal propensities between individuals may also be an important modifier of functional connectivity between populations. To assess how a key trait, body size, affected both social interactions and dispersal behaviour, this study examined the relationship between body size, antagonistic interactions, and breeding dispersal in male dragonflies (Pachydiplax longipennis) across a seasonal decline in adult body size. 2. During a seasonal peak in male body size in this study, dispersers were smaller than non‐dispersers. Later in the season, the body size of dispersers and non‐dispersers did not differ. 3. Focal observations found that body size was related to competitive dominance, large males engaged in aggressive chases more often and smaller males were more frequently pursued. 4. These results indicate that when large males were present, small males were more likely to disperse suggesting that dispersal is a tactic adopted by social subordinates in this context. If breeding dispersal is typically undertaken by subordinate males, functional connectivity between populations may be less than estimated from absolute dispersal rates.     

The interaction of season length and development time alters size at maturity

An end-of-season penalty, with late-maturing individuals being smaller than early-maturing individuals, has been observed in a variety of univoltine terrestrial arthropods. The current study extends these observations, utilizing multiple populations of a single sexually dimorphic species to examine the ecological correlates and fitness consequences of late maturation at a small size. The orb-weaving spider, Nephila clavipes, inhabits a broad range of habitats that vary from mild to strong seasonality. Because males mature several instars earlier than females, they can reach maturity much earlier in the growing season. Within a cohort, I found that female size at maturity was negatively correlated with timing of maturation in strongly seasonal sites. At a less seasonal site, there was no correlation between female size and timing of maturation within a cohort. In most populations studied, male size was not correlated with the timing of maturation within a cohort. Within populations in strongly seasonal sites, late-maturing females had reduced fecundity. The probability of copulation, survivorship from maturity to first clutch, clutch size relative to female size, and the number of possible clutches were all reduced with delayed maturation. The probability of pre-reproductive death for late-maturing females was strongly affected by stochasticity in the timing of the end of the growing season. Received: 30 December 1998 / Accepted: 1 September 1999     

Sex-specific life history patterns in bluegill (Lepomis macrochirus): interacting mechanisms influence individual body size

The ultimate body size that an individual fish achieves can be a function both of direct effects of growth or indirect effects associated with the timing of sexual maturation (and associated energetic tradeoffs). These alternatives are often invoked to explain variation in body size within and among fish populations, but have rarely been considered simultaneously. We assessed how resource availability and timing of maturation interact to influence individual body size of bluegill (Lepomis macrochirus). Resource availability (high and low food) and the social structure of the population (presence or absence of large, mature males) were varied in experimental ponds. Food ration affected growth (larger fish in the high food treatments) and the social structure of the population affected timing of maturation (early maturation of males in the absence of large males). Treatment effects, however, were sex-specific; males responded to the social structure of the population and females were more responsive to resource availability. We also found individuals that became sexually mature were smaller than those that remained immature, although results were sex-specific and resource dependent. For males, individuals that matured were smaller when resources were limited; mature and immature females showed no difference in body size regardless of food ration. We show that both resource availability and the processes that control timing of maturation interact in sex-specific ways to influence body size of bluegill. These results suggest that a more robust explanation for variable body size requires consideration of sex-specific interactions between ecological (food and growth) and evolutionary (timing of maturation) mechanisms.     

Energy expenditure and testosterone in free-living male yellow-pine chipmunks

The onset of mating in yellow-pine chipmunks (Tamias amoenus) follows emergence from a prolonged period of energy conservation during hibernation. Energy expenditures are greatly accelerated to meet the demands of the reproductive season. When emerging from hibernation, typical male chipmunks (breeders) have enlarged testes and a high level of plasma testosterone (T). However, certain males that do not participate in reproduction (nonbreeders) maintain small testes and low plasma T levels and emerge several weeks later than the breeders. The timing of the terminal arousal from hibernation and onset of mating are associated with increased plasma T levels. Experimental elevation of T levels in T. amoenus outside the mating season has been associated with a decrease in body mass, further suggesting an effect of T on energy balance. To test this hypothesis, we measured daily energy expenditure (DEE) in free-living, nonbreeding male chipmunks in the presence and absence of a T-implant. We also measured DEE in breeding males when endogenous T levels were high. DEE of the nonbreeders was not affected by our manipulation of plasma T, and the DEE of breeding males did not differ from that of nonbreeders. We conclude that energy expenditure on a daily basis in male yellow-pine chipmunks is not influenced by levels of T. However, on a seasonal basis, the earlier emergence from hibernation by breeding males, which appears to be influenced by T, represents an overall seasonal energy expenditure that exceeds that of nonbreeding males.     

The influence of habitat heterogeneity on host-pathogen population dynamics

The influence of spatial heterogeneity on the population dynamics of a naturally occurring invertebrate host-pathogen system was experimentally investigated. At ten week intervals over a two year period, I quantified the spatial distribution of natural populations of the terrestrial isopod crustacean Porcellio scaber infected with the isopod iridescent virus (IIV). During the seasonally dry periods of summer and early fall in central California, isopod populations were highly aggregated and the degree of patchiness and distance between inhabited patches was greatest. Coincident with increased patchiness and patch spacing was an increase in isopod density within patches. During the wet seasons of winter and spring, isopod population patchiness, inter-patch spacing, and within-patch density was low. Seasonal changes in virus prevalence were negatively correlated with within-patch density, patchiness, and inter-patch spacing. The influence of the spatial distribution of isopods on virus prevalence was also tested in field experiments. The virus was introduced into arrays of artificial habitat patches colonized by isopods in which interpatch distance was varied. The prevalence of resulting infections was monitored at weekly intervals. In addition, dispersal rates between artificial patches and natural patches were quantified and compared. The results showed that isopods in treatments with the smallest inter-patch spacing had the highest virus prevalence, with generally lower prevalence among isopods in more widely spaced patches. The spacing of experimental patches significantly affected virus prevalence, although the experiments did not resolve a clear relationship between patch spacing and virus prevalence. Rates of dispersal between patches decreased with increased patch spacing, and these rates did not differ significantly from dispersal between natural patches. The results suggest that rates of dispersal between isopod subpopulations may be an important component of the infection dynamics in this system. I discuss the consequences of these findings for host-pathogen dynamics in fragmented habitats, and for other ecological interactions in spatially heterogeneous habitats.     

Temporal variation in density dependent body growth of a large herbivore

Temporal variation both due to density dependent and density independent processes affect performance and vital rates in large herbivores. Annual fluctuations in climate affect foraging conditions and thus body growth of large herbivores during the short growing season in alpine habitats. Also, high animal densities on summer ranges may increase competition for food and reduce body mass gain. Yet, little is known about interactive effects of density and climate on alpine summer ranges, and the time scales these processes operate on. In this fully replicated landscape‐scale experiment, we kept domestic sheep at high and low densities over nine grazing seasons in an alpine habitat, and tested the relative role of density and annual variation for lamb body mass gain during summer and whether effects of density and annual variation interacted. We found that lambs at high density gained less mass over the summer season than lambs at low density. At short time scales the density effect interacted with annual fluctuations in body growth. We documented a long‐term temporal trend in body mass, consistent with the hypothesis that grazing effects affect habitat differentially at high and low density over years. At high density lamb autumn body mass declined during the first three grazing seasons and then stabilized, whereas body mass slightly increased over years at low density. This long‐term trend suggests accumulative density dependent effects from a biomass or quality reduction, and hence delayed food competition at high density and possibly facilitation at low density. Our experiment provides new insight into how density dependent effects on performance of a large herbivore depend on temporal scale of observation.     

Seasonal trends in body size of adult male mosquitofish,Gambusia affinis,with evidence for their social control

Synopsis Collections from a natural population of mosquitofish,Gambusia affinis, in southern Indiana showed that males born early in the breeding season mature rapidly at a small body size, while males born later in the breeding season delay maturation and achieve larger body sizes. A field experiment, involving removal of males from pond populations, was conducted to test the hypothesis that delayed maturation by late summer males is, at least in part, under social control. Mean total length (TL) of adult and maturing males in late summer in ponds from which large numbers of males had previously been removed was significantly lower than mean TL of males in control ponds. These data support the hypothesis of social control and support Sohn's competition-predation hypothesis for control of adult male body size inGambusia.     

The timing of gonadal development and moult in three raptors with different breeding seasons: effects of gender, age and body condition

Differences between species in breeding seasons are thought to be mediated through differences in their reproductive physiology. Little is known about how the timing and duration of gonadal maturation varies between raptor species, how the timing of moult relates to the gonadal cycle, whether the timing and degree of sexual maturation varies between juveniles and adults or whether body condition has a significant effect. To address these questions, data on gonadal size and moult for adults and juveniles of both sexes of three raptor species were extracted from the Predatory Bird Monitoring Scheme (based on birds found dead by members of the public). The three species, Sparrowhawk Accipiter nisus, Kestrel Falco tinnunculus and Barn Owl Tyto alba, have different ecologies – diurnal bird predator, diurnal mammal predator and nocturnal mammal predator, respectively. All are single‐brooded but have different breeding seasons. The duration of gonadal maturation was markedly different between the species. Barn Owls showed the earliest maturation and the latest gonad regression, and Sparrowhawks the latest maturation and earliest gonad regression. Kestrels were intermediate. In males of all species, the testes remained fully mature throughout their respective breeding seasons. In females, the ovaries remained partially mature throughout the breeding season. Moult started slightly earlier in Sparrowhawks than in Kestrels and coincided with gonadal regression in the two species. Although females of the two species started to moult earlier than males, moult duration was similar between the sexes. Barn Owls showed no distinct annual pattern of moult. In juveniles of all three species, the gonads were smaller than in adults throughout spring and started to mature later. Gonad size in birds that had starved tended to be smaller than in birds dying from other causes, but did not influence the difference in gonad mass between adults and juveniles and between seasons. Body condition had no effect on moult. Whilst ecology has led to the evolution of different breeding seasons, differences between species, and between adults and juveniles, are mediated through adaptive differences in their reproductive physiology.