Cooperative Breeding and Group Structure in the Lake Tanganyika Cichlid Neolamprologus savoryi

As yet, cooperative breeding has been described only for some fish species. However, evidence is accumulating that it is widespread among Lake Tanganyika cichlids. We studied the cooperative breeding system of the substrate breeding cichlid Neolamprologus savoryi. Breeding groups typically consisted of a large breeding male with one to four breeding females and three to 33 helpers (mean group size: 14.3 members). Group size was significantly related to breeding male and female body sizes, and larger males had more breeding females and larger sized male helpers. The size of the largest female in the group was positively related to the number and sizes of secondary breeding females and female helpers. In case of multiple breeding females, these females usually divided the group's territory into sub‐territories, each with its own helpers (subgroups). Interspersed between groups, independent fish were detected defending an individual shelter (4.4% of all fish). In 9% of the groups no breeding female was present. All group members participated in territory defence and maintenance, and showed submissive behaviours to larger group members. As expected, the level of between‐subgroup conflicts was high compared with the level of within‐subgroup conflicts. We compare these results with data available from other cooperatively breeding fishes.     

The Roles of Stimuli from Young, Previous Breeding Experience, and Prolactin in Regulating Parental Behavior in Ring Doves (Streptopelia risoria)

In addition to stimulating crop "milk" formation in ring doves, prolactin (PRL) may promote the parental regurgitation behavior that transfers this "milk" to the young at the time of hatching. Although earlier studies suggest that previous breeding experience is an important modulator of PRL-induced parental regurgitation behavior in ring doves, the ways in which experience, hormones, and stimuli from young interact to promote parental behavior have not been well characterized in this species. In the first study, untreated, nonbreeding female doves with and without previous breeding experience were given 10 daily parental behavior tests (2 h/day) with a hungry 5- to 10-day-old foster squab. Experienced females exhibited a higher incidence of regurgitation behavior, defensive behavior, and crouching or sitting in the nest than did inexperienced females. In a second study, nonbreeding females were given 10 daily tests for parental behavior while they received sc injections of ovine PRL or vehicle. Prolactin reduced squab-directed aggression and increased the incidence of regurgitation feeding behavior of foster squabs in both experienced and inexperienced females. However, the average number of regurgitation feeding acts displayed by those PRL-treated females that showed the behavior was over eight times higher in experienced females than in inexperienced females. Previous experience also enhanced the stimulatory effects of PRL on defensive behavior and crouching or sitting in the nest. The parental behavior exhibited by nonbreeding, PRL-treated experienced females was qualitatively and quantitatively similar to that observed in normally breeding females during a single test with their own hungry 5- to 10-day-old squabs. These findings indicate that PRL and previous breeding experience both enhance the parental responsiveness of nonbreeding female doves and that under optimal hormonal, experiential, and squab exposure conditions, nonbreeding doves exhibit levels of parental activity that rival those of normally breeding parents.     

Agonistic Interactions in Juvenile Sticklebacks (Gasterosteus aculeatus) in Relation to Local Predation Risk

Size-matched groups of three-spined stickleback (Gasterosteus aculeatus) fry were established using fish from two sites, one with abundant piscivorous predators (the high-risk site) and the other with very few heterospecific predators (the low-risk site). Over a period of two months, social interactions within groups were monitored regularly and mean lengths and coefficients of variance in length were tracked. Agonistic behaviour was observed in 45% of observation periods, but was more common in the groups of fry from the low-risk site, where it was more likely to take the form of direct attack rather than displacement without contact. In both categories of fish, the great majority of agonistic acts were directed to fish of a similar size or smaller. Mean size increased rapidly in all groups. At the same time, the coefficient of variation also increased, approaching 15% after two months in both categories of fish, reflecting differing growth rates among individuals. After three months, the two largest and the two smallest fish from each group were observed individually during an encounter with a model predatory fish. Large and small fish from both sites behaved identically in this test, suggesting that in sticklebacks (unlike guppies, for example) early experience of aggressive attacks from conspecific companions may not influence the development of anti-predator responses. As has been reported previously, sticklebacks from the high-risk site responded much more strongly than low-risk fish.     

Field evidence for bi-directional sex change in the polygynous gobiid fish Trimma okinawae

The social condition of bi-directional sex change in the gobiid fish Trimma okinawae was investigated at Akamizu Beach, Kagoshima, Japan. Social groups of T. okinawae usually consisted of a large male and one or more smaller females. The number of females in the group was positively correlated with male body size and groups were usually separated from each other by 1–3 m. In total, 22 instances of female-to-male sex change and three instances of male-to-female sex change were observed during the 16 months that social groups were monitored. Two individuals changed sex twice: female to male and back to female. Female-to-male sex change occurred when the male disappeared from a group. Either the largest remaining female changed sex to male or a large female from another group immigrated and changed sex to male. Larger individuals appear to benefit from becoming male because they can monopolize the breeding opportunities with several females, as reported in other protogynous fishes. Sex change from male-to-female only occurred when a solitary male joined another group as a subordinate. Mortality rates are high in these small fish, therefore joining another group and reproducing as a female is likely to increase the reproductive value of a solitary male.     

A new perspective on size hierarchies in nature: patterns, causes, and consequences

Many plant and animal aggregations have size hierarchies within which a variety of sizes of individuals, from large to small, can be found. Size hierarchies are thought to indicate the existence of competition amongst individuals within the aggregation, but determining their exact cause is difficult. The key to understanding size hierarchies lies in first quantifying the pattern of size and growth of individuals. We conducted a quantitative investigation of pattern in the size hierarchy of the clown anemonefish Amphiprion percula, in Madang Lagoon, Papua New Guinea. Here, groups of A. percula occupy sea anemones (Heteractis magnifica) that provide protection from predators. Within each anemone there is a single group composed of a breeding pair and zero to four non-breeders. Within each group there is a single size hierarchy; the female is largest (rank 1), the male is second largest (rank 2), and the non-breeders get progressively smaller (ranks 3–6). We demonstrate that individuals adjacent in rank are separated by body size ratios whose distribution is significantly different from the distribution expected under a null model—the growth of individuals is regulated such that each dominant ends up being about 1.26 times the size of its immediate subordinate. We show that it is decisions about growth at the individual level that generate the size hierarchy at the group level, and thereby determine maximum group size and population size. This study provides a new perspective on the pattern, causes and consequences of size hierarchies.     

Perceived wintering latitude determines timing of song output in a migratory bird

Migratory bird populations frequently consist of individuals that overwinter variable distances from the breeding site. Seasonal changes in photoperiod, which varies with latitude, underlie seasonal changes in singing frequency in birds. Therefore, migratory populations that consist of individuals that overwinter at different latitudes with large overwintering ranges could experience within‐population variation in seasonal production of song. To test the influence of overwintering latitude on intrapopulation variance in song production in the spring, we subjected two groups of Eastern Song Sparrows (Melospiza melodia melodia) from the same partially migratory breeding population to different photoperiodic schedules associated with a 1,300‐km difference in overwintering location. One group remained on the natural photoperiodic schedule of the breeding site (resident group) while the other group experienced a nonbreeding photoperiod that mimicked a southern migration in the fall followed by a northern migration back to the breeding site in the spring (migratory group). We compared song output between the two groups in three different stages (nonbreeding, prebreeding, and breeding). Little singing occurred during nonbreeding stage sample dates (20 November, 6 December) for the resident group, and no singing occurred for the migrant group. During the prebreeding stage (27 January, 7 February), significantly more singing occurred in the resident group than in the migrant group. During the breeding stage (21 March, 4 April), after a simulated migration for the migrants, song output was similar in both groups. These results suggest that within‐population variation in wintering latitude may contribute to variation in seasonal changes in singing behavior, which may covary with readiness to breed. Studies utilizing confirmed migrants and residents, rather than merely simulated migrants and residents, are also needed to better understand these processes.     

Infant Carrying Behavior in Callitrichid Primates: Callithrix and Leontopithecus

Care of offspring by individuals other than the mother is ubiquitous in callitrichid primates. In spite of its widespread occurrence, however, there is considerable taxonomic variation in the timing and intensity of parental effort by breeding males and nonbreeding juvenile group members. These differences may be attributable to generic and specifies differences in the costs of reproduction for females or in ecological constraints on travel and foraging. We present data on patterns of infant carrying in social groups of two taxa of callitrichid primates (Callithrix and Leontopithecus) throughout the first 3 months of infant life. We evaluated patterns of care in small groups (two or fewer juvenile or subadult helpers) and in large groups (three or more helpers in addition to the breeding adult male and female). Group size had little effect on levels of maternal carrying effort in either marmosets or lion tamarins, and mothers ceased carrying infants by 3 months of age. Carrying efforts by fathers were significantly reduced in groups with many helpers relative to small groups. Helpers carried at consistent rates during the second and third months of infant life in Leontopithecus, while in Callithrix, carrying by helpers peaked during the second month. These results suggest that if helpers reduce energetic demands on lactating females, the mechanism by which helpers reduce these burdens is independent of maternal carrying effort.     

Differences in Fecal Androgen Patterns of Breeding and Nonbreeding Kori Bustards (Ardeotis kori)

To better understand breeding conditions to promote reproduction in captive kori bustards, fundamental endocrine studies measuring fecal androgen metabolites in male and female kori bustards were conducted. Feces collected weekly from males and females were analyzed for testosterone using enzyme‐linked immunoassay. Results from adult males (n = 5), adult females (n = 10), immature males (n = 10), and immature females (n = 10) revealed seasonally elevated testosterone concentrations in fertile, but not nonfertile adult males and females (P > 0.05). Adult females that were not maintained in a breeding group, or that did not produce eggs, did not demonstrate increases in testosterone compared to egg laying counterparts. In males, but not females, seasonal testosterone increases were accompanied by weight gain. Peaks in male fecal androgen metabolites ranged from 10‐ to 22‐fold higher than nonbreeding season (181.5 ± 19.1 vs. 17.0 ± 0.94 ng/g; P < 0.05). Mean breeding season values for adult males were 83.6 ± 6.1 ng/g vs. nonbreeding season values of 12.3 ± 0.73 ng/g (P < 0.05). In females, average breeding season testosterone concentrations were approximately 4‐fold higher than nonbreeding season (55.9 ± 6.0 vs. 14.5 ± 1.8 ng/g), with peaks 10‐ to 30‐fold higher. Results show that noninvasive fecal androgen metabolite analysis can provide a means of predicting fertility potential of male and female kori bustards and might be utilized to assess effects of modifying captive environments to promote reproduction in this species. Zoo Biol. 32:54‐62, 2013. © 2012 Wiley Periodicals, Inc.     

Seasonal home range dynamics and sex differences in habitat use in a threatened,coastal marsh bird

A comprehensive understanding of spatiotemporal ecology is needed to develop conservation strategies for declining species. The king rail (Rallus elegans) is a secretive marsh bird whose range historically extended across the eastern United States. Inland migratory populations have been greatly reduced with most remaining populations inhabiting the coastal margins. Our objectives were to determine the migratory status of breeding king rails on the mid‐Atlantic coast and to characterize home range size, seasonal patterns of movement, and habitat use. Using radiotelemetry, we tracked individual king rails among seasons, and established that at least a segment of this breeding population is resident. Mean (±SE) home range size was 19.8 ± 5.0 ha (95% kernel density) or 2.5 ± 0.9 (50% kernel density). We detected seasonal variation and sex differences in home range size and habitat use. In the nonbreeding season, resident male home ranges coincided essentially with their breeding territories. Overwintering males were more likely than females to be found in natural emergent marsh with a greater area of open water. Females tended to have larger home ranges than males during the nonbreeding season. We report for the first time the use of wooded natural marsh by overwintering females. Brood‐rearing king rails led their young considerable distances away from their nests (average maximum distance: ~600 ± 200 m) and used both wooded natural and impounded marsh. King rails moved between natural marsh and managed impoundments during all life stages, but the proximity of these habitat types particularly benefitted brood‐rearing parents seeking foraging areas with shallower water in proximity to cover. Our results demonstrate the importance of interspersion of habitat types to support resident breeders. Summer draining of impounded wetlands that are seasonally flooded for wintering waterfowl allows regrowth of vegetation and provides additional habitat at a critical time for wading birds.     

Plural Breeding in the Splendid Fairy-wren,Malurus splendens (Aves: Maluridae), a Cooperative Breeder

Groups of the cooperatively breeding splendid fairy-wren Malurus splendens may include more than one female. Previously this species has been described as singular breeding (only one female breeds). This paper describes the occurrence of plural breeding (PB) groups in 10% of group years, in which two females had separate nests. In all cases, the secondary female (Y) was related to the primary breeding female (X) and was generally a 2-year old female which had helped in the group during the previous breeding season. Plural breeding was correlated with an increase in population density and in the number of female helpers; PB groups were larger than singular-breeding groups. In most cases, the X female was occupied with her own nest or offspring when the Y female began to nest, and there was no aggression between them. Which birds helped the Y female to feed at her nest depended on the time between the hatching of the two nests. If the interval was small, some group members helped at each nest; with longer intervals, the group members began to feed at the earlier nest, and the other female was left to raise her brood alone. Female helpers were very active in feeding at single nestings, and the cost to an X female of a Y female breeding was mainly a loss of this assistance. The success of individual X nests was not affected. Effects on productivity were slight, but fewer X females in PB groups raised second broods than did experienced singular breeding females. Y females were less productive than X females, but no less productive than singular breeding novice females without helpers. It is not known whether Y females copulated with primary or secondary males within their group, or with males from outside the group. Certainly, they did not form an observable pairing with any male in the group. Plural breeding occurred in a minority of group years in response to extrinsic conditions and the current demographic situation, and shows the extreme plasticity of the mating system in M. splendens.     

Group composition affects male reproductive partitioning in a cooperatively breeding cichlid

Individuals within groups of cooperatively breeding species may partition reproduction, with the dominant pair often taking the largest share. The dominant's ability to reproductively control subordinates may depend on differences in competitive ability, due to, e.g. body size differences, but may also depend on the number of same‐sex competitors inside the group. We tested experimentally whether subordinates reproduce more when these subordinates are large or when a second subordinate of the same sex need to be controlled by the dominants, using the cooperatively breeding cichlid Neolamprologus pulcher. Dominant pairs were assisted by a large and a small unrelated subordinate; sexes of these fish were varied in a full‐factorial design (giving four treatments). Dominant males lost significantly more parentage to the large subordinate male when a small subordinate male was also present, compared to when a small subordinate female was present. However, subordinate paternity was generally low and did not significantly curb total dominant male reproductive output, which was more affected by the sizes and numbers of reproductive females present inside his group. Dominant female maternity, clutch sizes and total output did not depend on the treatments. Subordinate–subordinate reproduction was virtually absent (one out of 874 offspring). Female subordinates were more likely to provide care for their own broods. In contrast, male subordinates did not adjust their level of care to their parentage. Variability in female subordinate alloparental brood care was particularly high, with females showing more care than males in general. We also detected effects of growth rate and food ration on parentage independent of the treatments, most notably: (i) a trade‐off between dominant male growth rate and paternity; (ii) a decrease in dominant male paternity with increasing food ration; (iii) a positive effect of growth rate on paternity in small males. We conclude that dominant males should be sensitive to the number and sizes of subordinate males present in their group, particularly when these subordinates are not helpful or grow fast, and food is plentiful. Dominant females should be less sensitive, because female subordinates do not appear to impose reproductive costs and can be helpful through alloparental brood care.     

The role of prolactin in the regulation of brood care in the cooperatively breeding fish Neolamprologus pulcher

The hormone prolactin (PRL) is important for the regulation of parental care in many species of mammals, birds and fish, and for alloparental care (care directed at nondescendant young) in some mammals and birds. Its significance in alloparental brood care of cooperatively breeding fish has not yet been assessed. Here, we test the role of PRL in brood care behavior of the cooperatively breeding cichlid Neolamprologus pulcher. The expression of PRL mRNA was determined in the pituitary glands of breeders of both sexes, helpers that showed brood care behavior and nonbreeding fish as controls. In addition, PRL levels were experimentally manipulated in male breeders and helpers by intraperitoneal injections of ovine PRL, and the behavior of these test fish was recorded toward standardized clutches. Adult females had higher levels of PRL mRNA than adult males, which was true both for breeders and nonbreeders. Contrary to expectation, there was no positive correlation between PRL and brood care behavior in any category of test fish, and the experimental application of PRL did not change brood care propensity. Interestingly, brood-caring adult females had significantly lower levels of PRL mRNA than adult female nonbreeders, whereas there was no difference between helpers and similar-sized nonbreeding group members. PRL mRNA levels increased with body mass in juveniles, but decreased with body mass in adults. In conclusion, we found no evidence that elevated levels of PRL are directly involved in the regulation of brood care behavior in this species.     

Effects of Familiarity and Population Density on Competitive Interactions and Growth: An Experimental Study on a Territorial Salmonid Fish

The deleterious effect of competition for space and food in animals increases with increasing population density. In contrast, familiarity towards conspecifics can relax the intensity of interference competition. Here, we hypothesized that familiarity towards conspecifics mitigates the effect of density‐dependent growth and dispersal behaviour in territorial animals. To test this, wild‐captured juvenile brown trout were subjected to two consecutive laboratory experiments. First, growth and fin erosion were measured for 40 d in a 2 × 2 factorial design manipulating density and familiarity. The density was manipulated via size of experimental tanks, while per capita food abundance and fish number was constant. All fish were subsequently exposed to an emergence test, giving them the option to leave their group and disperse to a novel unoccupied environment. The results show that familiarity increases growth and decreases the level of fin erosion (i.e. proxy of intensity of aggressive interactions). We found no significant effect of population density on growth rate. However, there was a tendency towards higher fin erosion in fish kept under high density. The growth of individuals was also affected by their size rank within the group, with the largest individuals in each group growing disproportionally faster than the rest of the group, probably due to their high social rank. However, the second and third fish in the size rank did not grow significantly faster and tended to suffer higher mortality than the rest of the group. During the emergence test, the largest individuals in the familiar groups left the shelter either as the first (six of 12 groups) or last (five of 12 groups) individual in the group, while no such pattern was observed in unfamiliar groups. Our results suggest that individuals in familiar groups receive less aggression and stress (i.e. fin damage) and grow faster than fish in unfamiliar groups. The mechanisms indicated in this laboratory study may be especially important in highly fecund organisms like fish which undergo density‐dependent bottlenecks during early life.     

Shared parentage and incest avoidance in the cooperatively breeding acorn woodpecker

Social groups of acorn woodpeckers (Melanerpes formicivorus) range in size from unaided pairs to 15 adults. Behavioural indicators of mate guarding, assumed incest avoidance and observations of egg-laying indicate that social organization ranges from monogamous pairs to groups with up to seven male and three female putative cobreeders plus up to 10 nonbreeding helpers. In addition, groups occasionally lack a putative breeder throughout the breeding season. Here we report results from multilocus DNA fingerprinting of 372 nestlings from 123 nests in groups with putative cobreeders of one or both sexes. No extra-group fertilizations were found. Putative cobreeding males within social groups shared paternity. However, the most reproductively successful male was, on average, almost three times as successful as the next most successful and additional males only occasionally sired offspring. In contrast, cobreeding females shared parentage equally. Helpers never bred incestuously when their opposite-sex parent (or another relative, such as their uncle) held breeding status in the group. However, during breeding male vacancies, 14 nestlings were produced when helper males bred incestuously with their mother. Both male and female helpers usually became successful cobreeders with their same-sex parent following replacement of the opposite-sex breeder(s) by unrelated individuals.     

Reproduction in captive lion tamarins (Leontopithecus): Seasonality,infant survival,and sex ratios

Diversity in reproductive and social systems characterizes the primate family Callitrichidae. This paper contributes to our appreciation of this diversity by presenting the first detailed comparative analysis of captive breeding in three species of lion tamarins (Leontopithecus chrysomelas, L. chrysopygus, and L. rosalia) housed at the Centro de Primatologia do Rio de Janeiro. The annual pattern of reproduction in all three species of Leontopithecus was markedly seasonal, with births occurring during the spring, summer, and fall months from August through March. While modal number of litters produced per female per year was 1, approximately 20% of breeding females produced two litters per year. The onset of breeding activity in years when two litters are produced was significantly earlier than in years when only one litter was produced. The cumulative number of offspring surviving to 3 months of age did not differ between years with one vs. two breeding attempts. Like other callitrichids, postnatal mortality was highest during the first week of life, and there were pronounced species differences in offspring survival through 1 year, with significantly lower survivorship in L. chrysomelas. Infant survivorship was affected by a number of experiential factors. Survivorship up to 30 days of life was higher in groups in which the breeding female had previous experience with infants as a nonbreeding helper than in groups in which the female lacked previous helping experience. Likewise, survivorship to 30 days of life was higher for infants born to multiparous females than for infants born to primiparous females. When parity and previous helping experience were analyzed concurrently, the lowest survivorship was associated with offspring produced by inexperienced primiparous females. Genus-wide, there was no significant departure from a 50:50 sex ratio at any point during the first year of life, nor was there evidence for differential mortality for male and female infants. However, L. chrysopygus produced significantly more male infants at birth (65:44) and had male-biased litters (approximately 60% males) throughout the first year of life, while L. chrysomelas showed a nonsignificant tendency toward female-biased litters. © 1996 Wiley-Liss, Inc.     

Population and group structure of western lowland gorillas (Gorilla gorilla gorilla) at Lokoué, Republic of Congo

During a 17-month study at the Lokoué clearing in Odzala National Park, Republic of Congo, we identified 377 western lowland gorillas. This population included 31 solitary males, 37 breeding groups, and eight nonbreeding groups. Its age- and sex-class structure was similar to those observed at two other clearings in the same forest block. However, the size of breeding groups varied with site (either clearing or forest sites). At Lokoué, breeding groups (mean size: 8.2 gorillas; range: 3-15) included a single silverback male and, on average, 3.2 adult females. Nonbreeding groups (mean size: 5.5; range: 2-15) were devoid of adult females. Five of the nonbreeding groups were composed predominantly of blackbacks, subadult males, and juveniles, and thus fit the definition of all-male groups previously observed in mountain gorillas. Our study confirms that 1) one-male breeding groups are the norm in western gorillas, and 2) all-male groups occur in this species. Despite frequent changes in members due to migrations of the males, the persistence of these all-male groups indicates that they may play an important role in the life of migrating males. Variations in population structure, and group composition and type among gorilla populations are discussed. However, a further understanding of the evolution of group-living in gorillas requires detailed ecological studies conducted in parallel with studies of the population structure and dynamics of these groups.     

Seasonal and sexual dimorphisms in the green anole forebrain

During the breeding season, male green anole lizards extend a throat fan (dewlap) in courtship. This behavior is facilitated by testosterone (T). Females extend a much smaller dewlap less often, even with the same dose of T. During the nonbreeding season when T is low, dewlap extension is reduced. To determine if parallels exist between structure and function, we investigated neuron soma size and density in the preoptic area (POA) and ventromedial nucleus of the amygdala (AMY), which are involved in the display behavior, in breeding and nonbreeding males and females. Cells from breeding animals were larger than cells from nonbreeding animals, but they were not sexually dimorphic. No significant effects existed in neuron density. This experiment indicates that portions of the anole forebrain important for sexual behavior are plastic and might be influenced by seasonal changes in steroid hormones. To investigate whether T can reverse the seasonal difference in soma size in both sexes, gonadectomized nonbreeding anoles were implanted with an empty or T propionate-filled capsule; animals were also tested for male-typical courtship behavior. Males and females treated with T had higher rates of dewlap extension, but across treatment groups these rates were greater in males. Neuron soma size in the POA and AMY was larger in males than females, but no effects of treatment were detected. Taken together, the results indicate that T can stimulate behavior in the nonbreeding season and suggest that a dissociation exists between the regulation of the courtship display and soma size of relevant brain regions.     

Sexual behavior and extragroup copulations in a wild population of common marmosets (Callithrix jacchus).

Sexual behavior and mating patterns are described for 3 free-ranging groups of common marmosets living in a coastal forest in northeastern Brazil. Each group contained 2 breeding females. Within groups, sexual behavior was generally restricted to breeding females and a single behaviorally dominant male. Of 101 mounts and copulations, 24 involved pairings of individuals from 2 different groups. Extragroup sexual behavior was performed by both breeding and nonbreeding group members, and 65% of all adults mounted or copulated with an extragroup individual at least once. Sexual behavior occurred throughout the female reproductive cycle but was significantly more frequent during an 11-day 'conception period'. Thus, while female marmosets show no physical signs of estrus, both males and females likely do have some information about the timing of ovulation. Mating patterns in this population included both polygyny and monogamy and varied between groups and over time.     

COMPARISON OF THE DIET OF BREEDING AND NONBREEDING CAPE GANNETS MORUS CAPENSIS

Berruti, A. 1991. Comparison of the diet of breeding and nonbreeding Cape Gannets Morus capensis. Ostrich 62:8-12.

The effects of breeding status on the diet composition of Cape Gannets Morus capensis were tested at two colonies in the western Cape, South Africa. At both colonies, there were no significant differences in the species composition of the diet, but the mean energy content of regurgitations of breeders was significantly greater. Although the regurgitations of breeders were heavier and contained more fish at both colonies, only the number of fish per regurgitation from breeders at one colony was significantly greater. Birds feeding chicks at Malgas Island took significantly shorter fish than nonbreeders. The small effect of breeding status on prey species composition shows that temporal and geographical variation in prey availability is more important than breeding status in affecting diet cornposition of the Cape Gannet.