Mate desertion by male Great Reed Warblers Acrocephalus arundinaceus at the end of the breeding season

Male Great Reed Warblers usually take part in the care of offspring as nest defenders and by feeding young, but at the end of the breeding season they desert their mates with eggs or nestlings. Deserted females continue offspring care. Desertion does not depend on the male's mated status (polygynous or monogamous) nor on his past breeding success. Deserted females compensate for the loss of their partners by increasing the frequency of food-bringing, resulting in a reduction in the amount of time the nestlings are brooded. Although desertion may lead to increased rates of offspring mortality through predation, breeding success of deserted females was high, especially if the male assisted during the early stages. Deserters pay costs by giving up the chance of additional matings and by lowering the reproductive success of existing mates. Male warblers reduce the former cost by choosing the season of desertion and the latter is lowered by the female's high parental ability. A deserter was found to start moulting while his mate was still feeding his nestlings, and an earlier start to the moult may be the primary benefit that he gains.     

Mate desertion by male Great Reed Warblers Acrocephalus arundinaceus at the end of the breeding season

Male Great Reed Warblers usually take part in the care of offspring as nest defenders and by feeding young, but at the end of the breeding season they desert their mates with eggs or nestlings. Deserted females continue offspring care. Desertion does not depend on the male's mated status (polygynous or monogamous) nor on his past breeding success. Deserted females compensate for the loss of their partners by increasing the frequency of food-bringing, resulting in a reduction in the amount of time the nestlings are brooded. Although desertion may lead to increased rates of offspring mortality through predation, breeding success of deserted females was high, especially if the male assisted during the early stages. Deserters pay costs by giving up the chance of additional matings and by lowering the reproductive success of existing mates. Male warblers reduce the former cost by choosing the season of desertion and the latter is lowered by the female's high parental ability. A deserter was found to start moulting while his mate was still feeding his nestlings, and an earlier start to the moult may be the primary benefit that he gains. Male Great Reed Warblers usually take part in the care of offspring as nest defenders and by feeding young, but at the end of the breeding season they desert their mates with eggs or nestlings. Deserted females continue offspring care. Desertion does not depend on the male's mated status (polygynous or monogamous) nor on his past breeding success. Deserted females compensate for the loss of their partners by increasing the frequency of food-bringing, resulting in a reduction in the amount of time the nestlings are brooded. Although desertion may lead to increased rates of offspring mortality through predation, breeding success of deserted females was high, especially if the male assisted during the early stages. Deserters pay costs by giving up the chance of additional matings and by lowering the reproductive success of existing mates. Male warblers reduce the former cost by choosing the season of desertion and the latter is lowered by the female's high parental ability. A deserter was found to start moulting while his mate was still feeding his nestlings, and an earlier start to the moult may be the primary benefit that he gains. Male Great Reed Warblers usually take part in the care of offspring as nest defenders and by feeding young, but at the end of the breeding season they desert their mates with eggs or nestlings. Deserted females continue offspring care. Desertion does not depend on the male's mated status (polygynous or monogamous) nor on his past breeding success. Deserted females compensate for the loss of their partners by increasing the frequency of food-bringing, resulting in a reduction in the amount of time the nestlings are brooded. Although desertion may lead to increased rates of offspring mortality through predation, breeding success of deserted females was high, especially if the male assisted during the early stages. Deserters pay costs by giving up the chance of additional matings and by lowering the reproductive success of existing mates. Male warblers reduce the former cost by choosing the season of desertion and the latter is lowered by the female's high parental ability. A deserter was found to start moulting while his mate was still feeding his nestlings, and an earlier start to the moult may be the primary benefit that he gains. Male Great Reed Warblers usually take part in the care of offspring as nest defenders and by feeding young, but at the end of the breeding season they desert their mates with eggs or nestlings. Deserted females continue offspring care. Desertion does not depend on the male's mated status (polygynous or monogamous) nor on his past breeding success. Deserted females compensate for the loss of their partners by increasing the frequency of food-bringing, resulting in a reduction in the amount of time the nestlings are brooded. Although desertion may lead to increased rates of offspring mortality through predation, breeding success of deserted females was high, especially if the male assisted during the early stages. Deserters pay costs by giving up the chance of additional matings and by lowering the reproductive success of existing mates. Male warblers reduce the former cost by choosing the season of desertion and the latter is lowered by the female's high parental ability. A deserter was found to start moulting while his mate was still feeding his nestlings, and an earlier start to the moult may be the primary benefit that he gains. Male Great Reed Warblers usually take part in the care of offspring as nest defenders and by feeding young, but at the end of the breeding season they desert their mates with eggs or nestlings. Deserted females continue offspring care. Desertion does not depend on the male's mated status (polygynous or monogamous) nor on his past breeding success. Deserted females compensate for the loss of their partners by increasing the frequency of food-bringing, resulting in a reduction in the amount of time the nestlings are brooded. Although desertion may lead to increased rates of offspring mortality through predation, breeding success of deserted females was high, especially if the male assisted during the early stages. Deserters pay costs by giving up the chance of additional matings and by lowering the reproductive success of existing mates. Male warblers reduce the former cost by choosing the season of desertion and the latter is lowered by the female's high parental ability. A deserter was found to start moulting while his mate was still feeding his nestlings, and an earlier start to the moult may be the primary benefit that he gains.     

Factors affecting the cost of polygynous breeding for female Great Reed Warblers Acrocephalus arundinaceus

The male's role in feeding young and his effect on the mortality of eggs and nestlings of the Great Reed Warbler Acrocephalus arundinaceus were studied in Kahokugata, central Japan, during eight breeding seasons. Three potential costs of polygynous breeding for females were examined:(i) a reduction of male parental care per brood, (ii) an attraction of predators due to concentration of nests, and (iii) an increase of unfertilized eggs due to the reduced frequency of copulation per female. Only the first cost was found to occur:polygynous males fed their young, especially of their second broods, less frequently than monogamous ones. The lowered paternal feeding frequency, however, did not increase the mortality by starvation in second-hatched broods. This was because the warm climate emancipated females from brooding and enabled them to compensate for the deficiency of feeding by males. The paper discusses possible roles of the staggering of the time of laying among harem mates and the shift of breeding status of females caused by nesting failure, in reducing potential costs of polygynous breeding for females.     

Effects of mate removal on the behaviour and reproductive success of Reed Warblers Acrocephalus scirpaceus

The Reed Warbler Acrocephalus scirpaceus is a largely monogamous insectivorous passerine in which males and females provide equal care by day to eggs and chicks. Polygyny occurs occasionally, with males leaving one female unaided. When females were temporarily removed from three recently-completed clutches their males deserted and resumed the high song levels typical of unmated males. Males may desert either because they are physically incapable of incubation or because the energy expenditure needed for a male to return to an equivalent stage in the breeding cycle is much lower than for a female to do so. Widowed females ( n = 7 ), however, continued the breeding attempt alone, with similar incubation levels but higher provisioning rates than those of control females. In three out of four mid-season broods raised by lone females all fertile eggs were reared to healthy fledglings (in the fourth brood the female died), while only one of four late-season nests produced any fledglings (which were underweight). Late-season control nests were as successful as earlier ones. Loss of male help led to starvation of chicks, but caused no adverse effects during incubation. This explains the small changes in widows' sitting levels during incubation, but much greater effects after hatching. Females may need male help to rear late broods (but not early broods) as days are shorter and food is scarcer. Males may normally help at nests, even those in the mid-season, because in stressful spells (even for a few days) such help is vital for successful breeding but in good periods it costs the male little.     

Sexual Differences in the Responses of Captive Great Reed Warblers (Acrocephalus arundinaceus) to Variation in Song Structure and Repertoire Size

The function and evolution of repertoires in songbirds is still not clearly understood. This study involves the great reed warbler Acrocephalus arundinaceus and playback of recorded song to both captive males and females under controlled laboratory conditions. Females were implanted with oestradiol and the copulation solicitation display used as an index of sexual response. Males were not implanted and crest-raising used as an index of aggressive response. Two types of songs were used (short and long) and two sizes of repertoire (high and low). Females only responded to long songs, and responded significantly more to high than low repertoires. It was concluded that long songs have a sexual function and that high repertoires are the result of intersexual selection through female choice. Males responded to short and long songs, but did not discriminate between high and low repertoires. Adding a live male to the experiment made no difference to female responses but males responded significantly more. When field studies are also taken into consideration the overall conclusion is that song quality and repertoire size is particularly important in female choice, and that short songs and a male presence are more important in aggressive behaviour. Integration of both field and laboratory studies on males and females should lead to more rigorous testing of hypotheses concerning the function and evolution of bird songs.     

Stable isotope ratios in winter‐grown feathers of Great Reed Warblers Acrocephalus arundinaceus,Clamorous Reed Warblers A. stentoreus and their hybrids in a sympatric breeding population in Kazakhstan

Analyses of the stable isotope composition of feathers can provide significant insight into the spatial structure of bird migration. We collected feathers from Great Reed Warblers Acrocephalus arundinaceus, Clamorous Reed Warblers A. stentoreus and a small sample of their hybrids in a sympatric breeding population in Kazakhstan to assess natural variation in stable isotope signatures and delineate wintering sites. The Great Reed Warbler is a long‐distance migrant that overwinters in sub‐Saharan Africa, whereas the Clamorous Reed Warbler performs a short‐distance migration to the Indian sub‐continent. Carbon (δ¹³C), nitrogen (δ¹⁵N) and deuterium (δD) isotope signatures were obtained from winter‐grown feathers of adult birds. There were highly significant differences in δD and less significant differences in δ¹³C between Great and Clamorous Reed Warblers. Thus, our results show that the stable isotope technique, and in particular the deuterium (δD) signal, resolves continental variation in winter distribution between these closely related Acrocephalus species with sympatric natal origin. The isotope signatures of hybrid Great × Clamorous Reed Warblers clustered with those of the Great Reed Warblers. Hence, a parsimonious suggestion is that the hybrids undergo moult in Afrotropical wintering grounds, as do the Great Reed Warblers. The observed δD values fell within the range of expected values based on available precipitation data collected at precipitation stations across the wintering continents of each species. However, the power to predict the winter origin of birds in our study system using these data was weak as the expected values ranged widely at this broad continental scale.     

Breeding ecology of the Eastern Great Reed Warbler Acrocephalus arundinaceus orientalis at Lake Kasumigaura,central Japan

CapsuleTwo populations of the Eastern Great Reed Warblers, one close to settlements and one far away, differed mainly in nest losses which were higher at the latter site.

Aims To compare the breeding ecology of two Japanese populations, and with breeding data from Polish sites.

Methods We studied two sites in 1996. The Seimei site comprises a natural reed-bed (Phragmites australis) (2.0 ha) c. 2 km from residential areas. The Sakura site occurs as smaller reed-bed (0.8 ha) close to residential areas. Field study comprised mainly nest searching and nest monitoring.

Results Breeding density, nest height, number of stems supporting nest, clutch size and brood size were not significantly different between sites. The proportion of nests built on reed stems standing in the water (60%) was significantly higher at the Seimei site (P < 0.001), and the warblers in this habitat experienced significantly higher nesting failure (P < 0.05). Predation was a principal cause of failure at both sites. Significantly more nests suffered from predation at Seimei (53%) than at Sakura (16%). Breeding density was higher at both sites compared with European populations but it fell within ranges of other Japanese populations. Mean first-egg date of first brood does not differ between our sites and Polish ones.

Conclusion The differences between the two Japanese populations and between Japanese and Polish populations were relatively small and were seen mainly in nest losses, breeding density and proportion of nests built over water.     

Frequency of complete moult in adult and juvenile Great Reed Warblers(Acrocephalus arundinaceus) in Spain

Summary We examined a large data set of adult and juvenile Great Reed Warblers (Acrocephalus arundinaceus), caught over five years, to study the occurrence of complete moult in a Spanish population. A total of 27 adults and 5 juveniles were in active moult. The proportion of birds moulting each year varied between 0.03 and 0.19. Nearly half the adults were finishing moult. At least one bird was certainly of local origin, rather than on passage, and four more were moulting when retrapped in subsequent years, suggesting that they too were resident breeders in the area. Some other birds were probably resident also, because they were in moult when caught at the beginning of August. Two of the five juveniles finished their moult in the area. In conclusion, our data seem to show that complete moult before departure to the wintering quarters is a regular process undertaken by some of the breeding population of Great Reed Warblers in north-east Spain.

---

Das Vorkommen von Vollmauser bei Alt- und Jungvögeln des Drosselrohrsängers(Acrocephalus arundinaceus) in Spanien

Zusammenfassung Von 837 zwischen 1992 und 1996 im Ebro Delta gefangenen Drosselrohrsängern mauserten 27 Alt-und 5 Jungvögel. Der jährlichen Anteil mausernder Vögeln betrug zwischen 0,03 und 0,19%. Etwa die Hälfte der mausernden Altvögel und zwei der Jungvögel beendeten ihre Mauser im Untersuchungsgebiet. Diese Daten zeigen, daß Drosselrohrsänger in Spanien wenigstens teilweise auch schon vor Beginn des Wegzuges eine Vollmauser durchführen.

     

Experimentally Simulating Paternity Uncertainty: Immediate and Long-Term Responses of Male and Female Reed Warblers Acrocephalus scirpaceus

In many socially monogamous species, both sexes seek copulation outside the pair bond in order to increase their reproductive success. In response, males adopt counter-strategies to combat the risk of losing paternity. However, no study so far has tried to experimentally prove the function of behaviour for paternity assurance. Introducing a potential extra-pair partner during the female fertile period provides a standardised method to examine how pair members respond immediately (e.g. increase mate guarding or copulation frequency) or long term (e.g. later parental investment and paternity uncertainty). In this study on a socially monogamous passerine species, we experimentally confronted pairs of reed warblers with a conspecific male (caged male simulating an intruder) during egg-laying. Our results revealed that occurrence of an intruder during that period triggered aggression against the intruder, depending on the presence of the female. The male territory owner also attacked the female partner to drive her away from the intruder. Thus territory defence in reed warblers also serves to protect paternity. The increase in paternity uncertainty did not affect later paternal investment. Paternal investment was also independent of the actual paternity losses. In females, the experiment elicited both, immediate and long-term responses. E.g. female copulation solicitations during the intruder experiment were only observed for females which later turned out to have extra-pair chicks in their nest. In relation to long term response females faced with an intruder invested later less in offspring feeding, and had less extra-pair chicks in their nests. Extra-pair paternity also seems to be affected by female quality (body size). In conclusion female reed warblers seem to seek extra-pair fertilizations but we could demonstrate that males adopt paternity assurance tactics which seems to efficiently help them to reduce paternity uncertainty.     

Longer is fatter: body mass changes of migrant Reed Warblers (Acrocephalus scirpaceus) staging at Eilat,Israel

Ecological barriers are the riskiest phases of the annual migrations for migratory birds. Comparatively little field data exists pertaining to the ability of migratory birds to prepare for the challenges of crossing ecological barriers, or their ability to recuperate afterward. Migrating Reed Warblers (Acrocephalus scirpaceus) were captured in Eilat, Israel, during their spring and autumn migrations. Data on spring and autumn body masses, their inter-annual variation, and the pattern of body mass increase were analysed. The birds show a significant inter-annual variation in their body mass and body condition index in both seasons, which is consistent with the data from other sites and for other passerine species. During stopovers, mass gain occurred in both seasons. Birds in poor initial condition, and those that stop over for a longer period of time, gained more body mass faster. In spring, but not in autumn, the progress of the season was also an important factor; late-arriving birds gained more fuel faster. The average rate of fuel gain was 0,157g·day^?1 ± 0.018 SE.     

Enemy Recognition of Reed Warblers (Acrocephalus scirpaceus): Threats and Reproductive Value Act Independently in Nest Defence Modulation

Organisms should respond more aggressively towards species perceived as a danger to their offspring, but intensity of defence may be gauged by the value of current offspring weighed against the value of future reproductive opportunities. We tested whether defensive responses of nesting reed warblers (Acrocephalus scirpaceus) are the result of an interaction effect between the type of stimulus confronted and the value of the warbler’s nesting attempt. We quantified the ability of reed warblers to discriminate among brood parasites, nestling predators and non‐threatening species at different stages of the breeding cycle. We also determined whether variables that influence the value of offspring, such as time of season, size and age of clutch or brood, and time of day and number of visits to the nest, explain variation in the intensity of defence recorded during the egg and nestling stages. Responses to the three stimuli differed significantly, as reed warblers consistently directed their mobbing calls and attacks towards parasites, whereas they were less conspicuous when confronted with models of predators. Reed warblers modulated their responses towards each stimulus in accordance with the threat each posed at a specific nesting stage, whereas they were not affected by other variables relative to their reproductive potential. The churr call, however, was uttered independently of the stimulus, as it was triggered by the mere presence of nestlings in the nest.     

Catabolic enzyme activities in the pectoralis muscle of premigratory and migratory juvenile Reed Warblers Acrocephalus scirpaceus (Herm.)

Summary Differences in catabolic capacity in the pectoralis major muscle of premigratory and migratory Reed Warblers were examined. The oxidative capacity was greater in the migratory birds, probably reflecting a training effect caused by the increased locomotion activity just prior to migration.Fatty acid oxidation was already high in the premigratory birds, possibly reflecting a saving of carbohydrate for anabolic purposes (feather growth) during the moult. Glycolytic capacity was slightly increased in the migratory birds, suggesting an increased carbohydrate oxidation during migration in spite of the great contribution of fatty acid oxidation.     

Philopatry of winter moult area in migratory Great Reed Warblers Acrocephalus arundinaceus demonstrated by stable isotope profiles

Stable carbon- (δ¹³C), nitrogen- (δ¹⁵N) and hydrogen (δD) isotope profiles in feathers of migratory Great Reed Warblers Acrocephalus arundinaceus recaptured for 2 or more years in 6 successive years were examined to test whether the isotope profiles of individual warblers appeared to be consistent between years. Similar isotopic signatures in successive years suggested that individual birds tended to return and grow their feathers in Afro-tropical wintering habitats that generate similar δ¹³C, δ¹⁵N and δD signatures. Previous studies have shown that Great Reed Warblers exhibit strong natal and breeding philopatry, with most of the surviving birds returning to the breeding site. The present study of feather δ¹³C, δ¹⁵N and δD isotopic values demonstrate the year-to-year fidelity might also include the African moulting sites in this migratory species.     

Genetic and phenotypic associations in morphological traits: a long term study of great reed warblers Acrocephalus arundinaceus

We present estimates of standardized selection (directional and quadratic) differentials via reproductive success on eight morphological traits in a newly founded long-term study population of great reed warblers Acrocephalus arundinaceus in Sweden. In order to predict the evolutionary response to selection in these traits we present estimates of heritabilities (h²), phenotypic (r_P) and genetic (r_A) correlations among the same traits. We also examined the extent of parental effects in the expression of the phenotypic traits. Overall, the population is subject to low levels of directional selection and higher levels of stabilizing selection. This makes us predict that the population is changing very little if anything, even though h² in many cases are considerable. Midparent-midoffspring h² ranged from 0.14 and 0.94 (mean 0.58) and were significant for seven of eight traits. We found indications of positive maternal effects in tarsus length. Phenotypic correlations between traits ranged from 0.02 to 0.43 (mean 0.15) and showed generally much lower values than the corresponding genetic correlations that ranged between 0.08 and 1.04 (mean 0.46). Overall, the correlation between r_P and r_A was significant, although moderate, but they tended to differ in magnitude, possibly due to overestimation of additive covariance between traits.     

三个东方大苇莺种群大杜鹃寄生率的变异

东方大苇莺Acrocephalus orientalis是大杜鹃Cuculus canorus的主要寄主之一。本文对东方大苇莺的3个地理种群进行杜鹃寄生率的时空比较。发现不同东方大苇莺种群被大杜鹃的巢寄生率在21.4%和65.5%之间,差异显著;不过,地理位置相近的两个种群在巢寄生率上则没有显著差异。另外,同一地理种群的巢寄生率年间也存在显著差异。     

Variation in mating system among birds: ecological basis revealed by hierarchical comparative analysis of mate desertion

Since most bird species are socially monogamous, variation among species in social mating systems is determined largely by variation in the frequency of mate desertion. Mate desertion is expected to occur when the benefits, in terms of additional reproductive opportunities, outweigh the costs, in terms of reduced reproductive success from the present brood. However, despite much research, the relative importance of costs and benefits in explaining mating system variation is not well understood. Here, we investigate this problem using a comparative method. We analyse changes in the frequency of mate desertion at different phylogenetic levels. Differences between orders and families in the frequency of desertion are negatively associated with changes in the potential costs of desertion, but are not associated with changes in the potential benefits of desertion. Conversely, differences among genera and species in the frequency of desertion are positively associated with increases in the potential benefits of desertion, but not with changes in the potential costs of desertion. Hence, we suggest that mate desertion in birds originates through a combination of evolutionary predisposition and ecological facilitation. In particular, ancient changes in life-history strategy determine the costs of desertion and predispose certain lineages to polygamy, while contemporary changes in the distribution of resources determine the benefits of desertion and thereby the likelihood that polygamy will be viable within these lineages. Thus, monogamy can arise via two very different evolutionary pathways. Groups such as albatrosses (Procellariidae) are constrained to social monogamy by the high cost to desertion, irrespective of the potential benefits. However, in groups such as the accentors (Prunellidae), which are predisposed to desertion, monogamy occurs only when the benefits of desertion are very limited. These conclusions emphasise the additional power which a hierarchical approach contributes to the modern comparative method.     

Responses of great reed warblers Acrocephalus arundinaceus to experimental brood parasitism: the effects of a cuckoo Cuculus canorus dummy and egg mimicry

Egg rejection behaviour towards parasitic eggs was studied in a great reed warbler Acrocephalus arundinaceus population in central Hungary, which was heavily (about 65%) parasitised by the common cuckoo Cuculus canorus . Clutches were experimentally parasitised during the egg-laying period with artificial, moderately mimetic cuckoo eggs or with conspecific eggs that were good mimics of the hosts' eggs. Great reed warblers rejected 76.2% of the artificial cuckoo eggs, mainly by ejection, but accepted most of the conspecific eggs (87.5%). Cuckoo eggs in naturally parasitised clutches were rejected at a lower rate (32%). When, in addition to the egg mimicry experiments, a stuffed cuckoo was placed near the nest, accompanied by the recording of a female cuckoo call, hosts' rejection rate of the artificial cuckoo egg increased from 76% to 96%. The sight of the cuckoo, on the other hand, did not influence host's rejection behaviour when a conspecific egg was used in the experiment. A stuffed collared dove Streptopelia decaocto , accompanied by its call, was used as a control, and did not cause any increased rejection. Great reed warblers were more aggressive towards the cuckoo than to the dove dummy. When the cuckoo eggs in naturally parasitised clutches were exchanged with artificial cuckoo eggs, we observed no increase in the rejection rate. We conclude that great reed warblers in our heavily parasitised population are capable of detecting brood parasitism in their clutch by identifying the parasitic egg. The efficiency of this identification depends mainly on the mimicry of the foreign egg. The sight of the cuckoo at the nest may increase rejection rate by stimulus summation, and this conditional effect is mainly affected by the degree of mimicry of the parasitic egg.     

Dynamics of parasitemia of malaria parasites in a naturally and experimentally infected migratory songbird, the great reed warbler Acrocephalus arundinaceus

Little is known about the development of infection of malaria parasites of the genus Plasmodium in wild birds. We used qPCR, targeting specific mitochondrial lineages of Plasmodium ashfordi (GRW2) and Plasmodium relictum (GRW4), to monitor changes in intensities of parasitemia in captive great reed warblers Acrocephalus arundinaceus from summer to spring. The study involved both naturally infected adults and experimentally infected juveniles. The experiment demonstrated that P. ashfordi and P. relictum lineages differ substantially in several life-history traits (e.g. prepatent period and dynamics of parasitemia) and that individual hosts show substantial differences in responses to these infections. The intensity of parasitemia of lineages in mixed infections co-varied positively, suggesting a control mechanism by the host that is general across the parasite lineages. The intensity of parasitemia for individual hosts was highly repeatable suggesting variation between the host individuals in their genetic or acquired control of the infections. In future studies, care must be taken to avoid mixed infections in wild caught donors, and when possible use mosquitoes for the experiments as inoculation of infectious blood ignores important initial stages of the contact between the bird and the parasite.     

Female choice and the benefits of mate guarding by male mallards

Pair formation and breeding in many species of waterfowl are separated both temporally and spatially. Most studies of female choice in this group have focused on male characteristics at the time of pairing, with less attention given to how mate choice affects breeding season outcomes. In this study I compared pairing success, male plasma testosterone level and mate-guarding ability of male mallards, Anas platyrhynchos, in two experiments. In the first experiment females and males were group housed with equal sex ratios, thus allowing all of these males to pair. At the same time, an equal number of males was housed in groups without access to females and remained unpaired. In this experiment testosterone levels of paired and unpaired males during autumn (baseline) and spring (breeding) did not differ, indicating that the process of pair formation and breeding does not cause elevated spring testosterone levels in males. However, testosterone did temporarily decrease in paired males during the winter (pair formation) season. In the second experiment groups were male biased, allowing only half of the males to pair. Here paired males had significantly higher testosterone levels than unpaired males during the breeding season, but not during the preceding autumn. Together the results of these experiments indicate that successful pair formation predicts but does not alter male testosterone level during the breeding season. I also found that females paired to males with high levels of testosterone were missing fewer feathers due to forced copulation attempts by nonmates, suggesting that females may choose males based on their mate-guarding abilities. Copyright 2002 The Association for the Study of Animal Behaviour. Published by Elsevier Science Ltd. All rights reserved.