Only one estimate of sex ratio at hatch exists for Greater Sage-Grouse (Centrocercus urophasianus). Managers typically assume a ratio at the population level of approximately 2:1 (female:male), primarily on the basis of sex ratio in the harvest. We determined the sex of newly hatched young and unhatched Greater Sage-Grouse by amplifying a portion of the sex-linked CHD gene. Sex ratio for Greater Sage-Grouse in east-central Nevada was 0.51 ± 0.03 (SE; n = 272). We found no substantial difference in size between eggs that produced male chicks and those that produced females (44.5 ± 0.2 mm³ vs. 44.3 ± 0.3 mm³) or between the masses of male and female chicks (25.8 ± 0.3 g vs. 26.3 ± 0.3 g), which suggests that energetic cost investments by females were similar between offspring of different sexes. We also found no effect of female condition on differential investment in male versus female offspring. Given that adult survival does not differ substantially between the sexes in our study population (J. S. Sedinger unpubl. data), we suggest that this population may not contain 2 adult females to 1 adult male and that any bias in adult sex ratio is likely attributable to differential survival from hatch to first breeding.

The Haida Gwaii archipelago (formerly the Queen Charlotte Islands) off northern British Columbia hosts many endemic taxa of plants and animals, including 9 endemic subspecies of birds. Genetic studies suggest that some of these taxa originated in a glacial refugium in the area, whereas other taxa apparently arose after the recession of the glaciers. The Orange-crowned Warbler (Vermivora celata) is a Nearctic—Neotropic migrant passerine broadly distributed across North America. To gain insight into the importance of historical fragmentation versus recent differentiation in diversification of the Haida Gwaii avifauna, we analyzed geographic variation in 8 microsatellite loci and the mitochondrial control region among Orange-crowned Warblers from 7 breeding localities, including 2 of the 4 currently recognized subspecies (V.c. celata and V. c. lutescens). We found modest genetic differentiation among populations, with an isolation-by-distance pattern among western populations and significant differentiation between subspecies. Results from several analyses indicate that differences between Haida Gwaii and other western populations arose through restrictions in gene flow after the recession of the glaciers. Thus, patterns of genetic diversity in western Orange-crowned Warbler populations do not provide evidence that the species persisted in a Pleistocene refugium in the area of Haida Gwaii.

The functional significance of bird egg color and patterning is a continuing subject of interest and debate. Extreme polymorphism in eggshell appearance is often thought to be maintained by frequency-dependent selection operating within populations. However, variation could also be explained by small-scale differentiation combined with limited migration. Here, we report the existence of extreme variation in egg color in a population of Mexican Jays (Aphelocoma ultramarina) inhabiting a steep elevation and habitat gradient within a single mountain range, the Sierra del Carmen of Coahuila, Mexico. We quantified egg color of 143 eggs from 54 nests throughout the mountain range, using digital photos. Color was also quantified for a subset of these eggs, using a spectrometer. Results from both methods support the conclusion that egg color has diverged at a remarkably small spatial scale of 3–15 km. Photo color quantification indicated that eggs at high elevation were greener than those at low elevation. Spectrometer results supported this conclusion, with more-pigmented (i.e., less reflectant) eggs occurring at high elevation. Unlike in other species, differences in condition do not seem to drive these divergence patterns in egg color. Further study is needed to determine to what extent differences result from heritable genetic change attributable to divergent selection pressures or from an environmentally induced phenotypic response.

The tail morphology of many birds is shaped by both natural and sexual selection. Models of tail aerodynamics predict that functions related to flight naturally select for moderately forked tail shapes, constraining the tendency of sexual selection to generate interspecific morphological diversity. Moreover, models predict that small birds will have low variation in tail morphology. Although hummingbirds (family Trochilidae) depend exclusively on flight for locomotion, the diversity of their tail morphology is among the greatest in all bird families. Hummingbird tail length exhibits positive allometry, scaling as approximately mass^0.5, which causes tail surface area to scale as mass^1.0. Forked tail morphology arises at least 26 times in the clade, and forked taxa tend to be sexually dimorphic, whereas species with graduated tails tend to be monomorphic. Small hummingbird species exhibit higher variation in tail morphology than large species. These results suggest that moderately forked tail morphology tends to arise via sexual selection in hummingbirds and that large species are more constrained in tail morphology than small species. Both patterns are inconsistent with current models of how bird tails function. This suggests that, in terms of aerodynamics, hummingbird tails function differently from the tails of other birds.

Piping Plover (Charadrius melodus) is a migratory shorebird that is listed as endangered in Canada and the U.S. Great Lakes and as threatened throughout the rest of its breeding and winter range. We undertook a comprehensive molecular-genetic investigation to (1) address subspecific taxonomy, (2) characterize population genetic structure, and (3) infer past bottlenecks and demographic processes in this species. Analyses included individuals from 23 U.S. states and Canadian provinces and were based on mitochondrial DNA sequences (580 base pairs, n = 245) and 8 nuclear microsatellite loci (n = 229). Our findings provide support for separate Atlantic and Interior subspecies (C. m. melodus and C. m. circumcinctus, respectively). Birds from the Great Lakes region were allied with the Interior subspecies and should be referred to as C. m. circumcinctus. Population genetic analyses illustrated stronger genetic structure among Atlantic than among Interior birds, which may reflect reduced natal- and breeding-site fidelity of Interior individuals. Furthermore, analyses suggested that Interior birds previously experienced genetic bottlenecks, whereas there was no evidence of such patterns in the Atlantic subspecies. We interpret these results in light of 25 years of range-wide census data. Overall, differences between Interior and Atlantic Piping Plovers may reflect differences in spatiotemporal stability of nesting habitat between regions.

Many studies of nutrient allocation to egg production in birds use stable isotope ratios of egg yolk to identify the origin of nutrients. Dry egg yolk contains >50% lipids, which are known to be depleted in ¹³C. Currently, researchers remove lipids from egg yolk using a chemical lipid-extraction procedure before analyzing the isotopic composition of protein in egg yolk. We examined the effects of chemical lipid extraction on δ¹³C, δ¹⁵N, and δ³⁴S of avian egg yolk and explored the utility of an arithmetic lipid correction model to adjust whole yolk δ¹³C for lipid content. We analyzed the dried yolk of 15 captive Spectacled Eider (Somateria fischeri) and 20 wild King Eider (S. spectabilis) eggs, both as whole yolk and after lipid extraction with a 2:1 chloroform:methanol solution. We found that chemical lipid extraction leads to an increase of (mean ± SD) 3.3 ± 1.1‰ in δ¹³C, 1.1 ± 0.5‰ in δ¹⁵N, and 2.3 ± 1.1‰ in δ³⁴S. Arithmetic lipid correction provided accurate values for lipid-extracted δ¹³C in captive Spectacled Eiders fed on a homogeneous high-quality diet. However, arithmetic lipid correction was unreliable for wild King Eiders, likely because of their differential incorporation of macronutrients from isotopically distinct environments during migration. For that reason, we caution against applying arithmetic lipid correction to the whole yolk δ¹³C of migratory birds, because these methods assume that all egg macronutrients are derived from the same dietary sources.

In the East Usambara Mountains in northeast Tanzania, research on the effects of forest fragmentation and disturbance on nest survival in understory birds resulted in the accumulation of 1,002 nest records between 2003 and 2008 for 8 poorly studied species. Because information on the length of the incubation and nestling stages in these species is nonexistent or sparse, our objectives in this study were (1) to estimate the length of the incubation and nestling stage and (2) to compute nest survival using these estimates in combination with calculated daily survival probability. Because our data were interval censored, we developed and applied two new statistical methods to estimate stage length. In the 8 species studied, the incubation stage lasted 9.6–21.8 days and the nestling stage 13.9–21.2 days. Combining these results with estimates of daily survival probability, we found that nest survival ranged from 6.0% to 12.5%. We conclude that our methodology for estimating stage lengths from interval-censored nest records is a reasonable and practical approach in the presence of interval-censored data.

Declining populations of grassland breeding birds have led to increased efforts to assess habitat quality, typically by estimating density or relative abundance. Because some grassland habitats may function as ecological traps, a more appropriate metric for determining quality is breeding success, which is challenging to determine for many cryptic-nesting grassland birds. This difficulty led Vickery et al. (1992) to propose a reproductive index based on behavioral observations rather than nest fate. We rigorously evaluated the index for 2 years using a Savannah Sparrow (Passerculus sandwichensis) population in western New York and found a weak correlation in classification of the breeding stages of monitored territories among multiple observers (r = 0.398). We also discovered a large difference between overall territory and nest success rates independently estimated with the index (9.8% over the entire breeding cycle) and with nest searching and monitoring (41.7% of nests successfully fledged young). Most importantly, we made territory-level comparisons of index estimates with actual nest fate and found that the index correctly predicted fates for only 43% of the monitored nests. A Mayfield logistic regression analysis demonstrated that only index rank 4 (eggs hatched, but young failed to fledge) showed a strong positive correlation with nest success. Although the reproductive index may function as a coarse indicator of habitat suitability (e.g., documenting production in potential ecological traps), in our study the index exhibited neither internal consistency nor the ability to predict nest fate at the plot or territory level and functioned poorly as a substitute for nest searching and monitoring.

Survey methods that account for detection probability often require repeated detections of individual birds or repeated visits to a site to conduct counts or collect presence-absence data. Initial encounters with individual species or individuals of a species could influence detection probabilities for subsequent encounters. For example, observers may be more likely to redetect a species or individual once they are aware of the presence of that species or individual at a particular site. Not accounting for these effects could result in biased estimators of detection probability, abundance, and occupancy. We tested for effects of prior detections in three data sets that differed dramatically by species, geographic location, and method of counting birds. We found strong support (AIC weights from 83% to 100%) for models that allowed for the effects of prior detections. These models produced estimates of detection probability, abundance, and occupancy that differed substantially from those produced by models that ignored the effects of prior detections. We discuss the consequences of the effects of prior detections on estimation for several sampling methods and provide recommendations for avoiding these effects through survey design or by modeling them when they cannot be avoided.

Satellite transmitters and geographic-positioning-system devices often add substantial mass to birds to which they are attached. Studies on the effects of such instruments have focused on indirect measures, whereas the direct influence of extra mass on pelagic behavior is poorly known. We used 2.5-g geolocators to investigate the effect of extra mass on the pelagic behavior of Cory's Shearwaters (Calonectris diomedea) by comparing the traits of a single foraging trip among a group carrying 30-g weights, a group carrying 60-g weights, and a control group. The weights were attached to the birds' backs using typical techniques for attaching satellite transmitters to seabirds. The extra mass increased the duration of the birds' trips and decreased their foraging efficiency and mass gained at sea. These indirect effects may be related to foraging traits: weighted birds showed a greater search effort than control birds, traveled greater distances, covered a greater foraging area, and increased the maximum foraging range. Furthermore, the time spent on the sea surface at night was greater for weighted than for control groups, which showed that the extra mass also affected activity patterns. Our results underline the need to quantify the effects of monitoring equipment commonly used to study the pelagic behavior of seabirds. We suggest that geolocators can be used to obtain control data on foraging-trip movements and activity patterns.

Fuel deposition rates of migrating birds may indicate the quality of habitat at stopover sites, yet little is known about how diet habits and food availability affect fat and protein metabolism in free-living songbirds at stopover sites. We compared plasma indicators of fat deposition (triglyceride), fat catabolism (B-hydroxybutyrate), and protein catabolism (uric acid) among passerine species that are frugivorous to a variable degree during autumn stopover on Block Island, Rhode Island. We also compared plasma lipid metabolites from 3 of these species that were captured at 2 stopover sites with different fruit abundance. The more frugivorous Hermit Thrushes (Catharus guttatus) had the highest plasma triglyceride, and uric acid was highest in the least frugivorous species sampled on Block Island, but other differences among species were not clearly related to diet. B-hydroxybutyrate was more variable among the species sampled on Block Island. Plasma triglyceride was significantly higher in Hermit Thrushes captured on Block Island, where fruit resources were abundant, than in Hermit Thrushes captured at a mainland site in southern Rhode Island, where less fruit was available. Our results suggest that diet habits may influence fat and protein metabolism in migrating passerines, but careful study design and statistical analyses are necessary to control for or minimize the effects of the many influential factors that affect plasma metabolites so they can be used to assess fuel deposition in free-living birds and to compare the quality of migration stopover sites.

Using a Canadian weather surveillance radar (CWSR), we assessed the relationship between aural passerine counts and radar reflectivity during autumn migration on 16 nights. Reflectivity was positively correlated on all but 1 night with the number of birds detected aurally, but the correlation strength varied between -0.58 and 0.93 among nights (mean ± SD = 0.69 ± 0.42). Using linear mixed-effects models with aural counts nested within nights, we found that the number of birds detected by observers increased with reflectivity. The slope of this relationship did not vary between observers, nor was it affected by time since sunset, but the number of birds detected aurally tended to be lower when ambient noise levels were high. We know that the radar was relatively sensitive to low bird densities, because the intercept was slightly positive and its 95% confidence interval marginally included zero. However, the relationship between the number of birds detected aurally and reflectivity varied significantly among nights. Such variation was likely caused by a combination of (interacting) factors, including bird species and behavior (e.g., calling rate, flight altitude), influencing bird detectability by the observers and the radar. The weather radar network of the United States (NEXRAD) is already used for bird migration studies, and we conclude that the use of CWSR can extend NEXRAD's coverage farther north by hundreds of kilometers, thereby increasing our understanding of how birds use the North American landscapes during migration.

Birds use local environmental cues to fine-tune the timing of egg laying such that the nestling period normally coincides with the local peak in food availability. Ambient temperature, vegetation phenology, and insect phenology are often considered the most likely cues, but no previous studies have explicitly compared and partitioned their relative effects. We used confirmatory path analyses and a long-term study of Blue Tits (Cyanistes caeruleus) to identify and measure the relative weighting of the causal paths that link laying date to spring phenology and temperature in deciduous and evergreen oak forests on Corsica. Path analysis showed that the effects of temperature and vegetation phenology vary between forest types and season. In deciduous oak forest, where females lay eggs early in spring, phenology of vegetation or insects sets the laying date. In evergreen oak forest, where breeding occurs later in the season, females shift from a predominantly phenology-based cue system to a predominantly temperature-based cue system. This plasticity in the decision process allows birds to minimize the risk of mismatching breeding date with the optimal time window and may be critical in allowing birds to track human-induced environmental change.

Offspring provisioning constrains the foraging behavior of breeding seabirds temporally and spatially. In species whose foraging grounds overlap with commercial fisheries, quantifying the provisioning behavior of breeding adults throughout the season can illuminate the nature of interactions with the fisheries and, where seabird bycatch exists, contribute to development of mitigation measures. In 2004–2005, we studied the provisioning behavior of Flesh-footed Shearwaters (Puffinus carneipes) on Lord Howe Island, Australia, using the repeated-weighing technique. Specifically, we determined (1) meal size, feeding frequency, and chick growth rates; (2) attendance behavior of adults; and (3) estimated food consumption at the individual and colony levels. Incubation shift duration averaged 9.5 ± 1.6 (SD) days. Foraging-trip duration was bimodal, with both short trips (<3 days) and long trips (>3 days) recorded. Average meal mass was greater during the late chick-rearing period and was not influenced by the time interval between meals. Developing chicks had a food conversion efficiency of 27%, with a total individual food requirement of 2,337 g to fledge successfully. An increase in the interval between provisioning events from the early to the late chick-rearing period was associated with significantly more long foraging trips by parent birds; however, there was no variation in the proportion of study birds that returned each night.

Sugar-type preferences of nectarivores may be an important selective factor in the evolution of sugar composition in floral nectar. We investigated sugar preferences of the Malachite Sunbird (Nectarinia famosa), using experiments in which birds were offered paired choices between energetically equivalent solutions over a range of sugar concentrations. The birds preferred hexose at low (5%) concentration and sucrose at high (25%) concentration; they showed no preference at 10%, 15%, and 20% concentrations. The birds regulated energy intake by adjusting volumetric consumption, except on a 5% concentration diet, where they failed to maintain energy balance. They also exhibited a strong preference for concentrated solutions, given a choice between 10%, 15%, 20%, and 25% sucrose solutions. We discuss the significance of these results in terms of the nectar composition of sunbird-pollinated plants.

Mercury is a toxic heavy metal that can cause obvious physiological and reproductive problems in animals. Very little is known, however, about its subtle behavioral effects. We examined whether birds that inhabited mercury-contaminated sites exhibited differences in singing behavior compared with birds at uncontaminated reference sites nearby. We recorded the songs of 3 oscines, the Carolina Wren (Thryothorus ludovicianus), House Wren (Troglodytes aedon), and Song Sparrow (Melospiza melodia), and 1 suboscine, the Eastern Phoebe (Sayornis phoebe). Spectrographic analysis revealed that songs of oscines living on contaminated sites contained a lower diversity of note types and were sung at lower tonal frequencies than songs of birds on reference sites. Additionally, both species of wren tended to sing shorter songs. By contrast, the songs of Eastern Phoebes did not differ between contaminated and reference sites, suggesting that mercury may affect singing behavior only in species that learn their songs. Such alterations in song could have important implications for the fitness of songbirds in polluted areas. Our results highlight the importance of considering behaviors in evaluations of contaminant effects.

We compared the use of body stores in breeding Barnacle Geese (Branta leucopsis) in traditional Arctic colonies in the Barents Sea with that in recently established temperate-zone breeding colonies in the Baltic Sea and North Sea by studying female body-mass loss and use of fat and protein stores during incubation. Average daily body-mass loss was almost identical in the 2 temperatebreeding populations (17.0 g and 16.5 g in Baltic Sea and North Sea, respectively), whereas Arctic-breeding females lost significantly less (10.6 g day^-1). Temperate-breeding females initiated incubation with body mass 125 g higher than that of Arctic breeders, but at the end of incubation, body mass was similar among the 3 populations, averaging 1,458 g. Body-mass loss during incubation amounted to 23% (North Sea), 22% (Baltic Sea), and 15% (Barents Sea). Fat mass, as measured by isotope dilution in a subsample of females, was consistently higher in North Sea than in Barents Sea birds, but both populations showed similar rates of fat-mass loss (9.4 g day^-1, on average). By contrast, loss of fat-free mass (assumed to represent wet protein) amounted to 9.3 g day^-1 in North Sea birds but only 1.5 g day^-1 in Barents Sea birds. Energy content of 1 g utilized body mass was 21.1 kJ (North Sea) and 34.9 kJ (Barents Sea), which equates to 376 kJ day^-1 and 415 kJ day^-1 drawn from stored energy, respectively. We suggest that differences in nest-attendance and postincubation demands are responsible for the differential use of body stores in temperate- and Arctic-breeding Barnacle Geese.

Recruitment is an essential component of the life history and population dynamics of bird species. We provide comprehensive information on the determinants of territorial recruitment in populations of the endangered Bonelli's Eagle (Aquila fasciata). Field work was based on a long-term study of two populations located in the northwest of this species' range, one in Catalonia (northeastern Spain) and the other in Provence and Languedoc-Roussillon (southeastern France). Nestlings were banded (n = 451 marked birds) and known territories were intensively monitored during the period 1980–2007. First, a global return rate of 9.97% (45 recruits) was calculated, with no significant differences between the two populations. Second, results showed that both the birth year and the breeding success of the birth territory had significant effects on recruitment probability: nestlings from territories with better breeding success were more likely to recruit. Third, seniority analyses based on capture—resighting techniques were used to estimate the age-specific probabilities that a territorial bird in a given year was inexperienced. The parameter estimates for this probability ranged from 0.985 to 0.999 for 2-year-olds and from 0.763 to 0.808 for 3-year-olds and then fell drastically to 0.066–0.272 for 4-year-olds and older. Fourth, females were found to disperse farther than males. Additionally, there was a significant interaction between sex and area of birth, in that females from Catalonia dispersed farther than females from France. Finally, previously occupied territories located in the highest-quality areas with the highest territorial density were found to be the most attractive to inexperienced individuals.

Many studies have examined the effects of forest fragmentation and management on songbird nesting success, but few have quantified postfledging survival, which is a critical component of population productivity. In 2005–2006, we estimated daily postfledging survival of Rose-breasted Grosbeaks (Pheucticus ludovicianus) by radiotracking 42 fledglings in forest fragments that had been managed by single-tree selection, by diameter-limit harvest, or as reference (not harvested for at least 25 years). Survival probability over the 3-week fledgling period was 0.62, and 86% of total fledgling mortality occurred during the first week out of the nest. Despite large differences in forest structure between forest management treatments, there was no effect of forest treatment on fledgling survival. Date of fledging, shrub cover, and patch size also had limited influence on fledgling survival. For all sites combined, females produced an estimated 0.23–0.37 recruiting daughters per year for the worst- and best-case scenarios of female fecundity and annual juvenile survival, which is lower than the expected annual mortality rate of breeding females (0.40–0.55). Even reference sites did not produce enough offspring to offset annual female mortality, which suggests that forest fragments in this region are population sinks.

Avian breeding success generally declines with later laying because of seasonal reductions in food supply, late laying by less capable pairs, or both. To understand the direct fitness consequences of breeding time requires distinguishing between these two possibilities. We used egg removal and re-laying experiments to evaluate how date and parental quality affect breeding success in a zooplanktivorous seabird, Cassin's Auklet (Ptychoramphus aleuticus). Egg laying began at the same time in all 5 years of study at Triangle Island, British Columbia, but, compared with a cold-water year, the population laid later and less synchronously in 4 warm-water years in which prey populations peaked earlier. As a result, Cassin's Auklets were less successful in years in which they laid later. Within seasons, early-laying females whose breeding attempt we delayed did not follow the population-wide seasonal declines in hatching success. This indicates a strong role for parental quality at the egg stage, probably because early, high-quality birds maintained more constant incubation. By contrast, the experimental females followed the population-wide seasonal declines in nestling survival and fledging mass. This indicates a strong role for date at the offspring-provisioning stage, which accords well with a previous study that found that success while raising nestlings is largely determined by the degree of temporal (mis)matching with the copepod Neocalanus cristatus. Our results offer novel insight into the causes of seasonal declines in avian breeding success, indicating that date and parental effects can be differentially involved, depending on the stage of breeding.

In many songbirds, the nesting period for a breeding attempt is extremely short, often lasting only a few weeks. Breeding adults can shorten this period by decreasing the number of eggs laid or reducing the length of the nestling period. Nestling-period length has received little attention in the literature but could have profound effects on annual fecundity, because each day represents a risk of nest depredation. Consequently, we were interested in assessing the biotic and abiotic factors that govern the nestling period in the Ovenbird (Seiurus aurocapilla). We provide evidence that food availability, more than predation pressure and climatic factors, influences nestling-period length, with increases in food availability decreasing the nestling period. We suggest that the nestling period is dictated by physiological constraints, which may be influenced by food availability and, thus, the ability to provision young. However, the greatest variation in nestling period was individual variation among breeding pairs. Thus, we believe that large-scale variation in ecological and environmental factors may determine the physiological constraints of the nestling period but parental behavior and quality within this framework dictate the actual length of the nestling period.

Renesting is an important strategy for coping with nest loss in many species of birds. We investigated renesting behavior of radiomarked Mallards (Anas platyrhynchos) breeding in the Canadian Prairie Parklands and found that females were persistent renesters, replacing >57% of 4,112 destroyed nests. Renesting propensity was most affected by nest initiation date, with ∼90% of unsuccessful females renesting if destroyed clutches had been initiated in April but <10% renesting for clutches initiated after 20 June. Probability of renesting declined with successive number of nesting attempts, but this was primarily an effect of initiation date. Renesting propensity increased with female age and body condition and declined for birds that had invested more time tending their previous clutch, but the latter effect was pronounced only among late-nesting females. The amount of time required to produce a replacement clutch was primarily a function of whether females were engaged in rapid follicle growth when nest failure occurred: 44% of females (383 of 870) that lost nests during early laying renested within 4 days, whereas only 2% (12 of 639) that lost nests after clutch completion renested within 4 days. Although clutch size was smaller in renests, this was entirely an artifact of later laying dates. Our results suggest that seasonal timing had the most influence on renesting behavior. Female age, body condition, and prior nesting effort had smaller, but demonstrable, influence.

The risk of breeding failure often varies strongly over the breeding season. If birds optimize production of offspring accordingly, investment in reproduction should be lowest when food for young is scarce or nest predation risk is highest. Evidence for this hypothesis is mixed and stems mainly from studies on northern temperate birds, and it is unclear whether these patterns apply to southern temperate birds with “slow” life histories that resemble those of tropical birds. We studied seasonal variation in reproductive investment and nesting success in a Neotropical temperate suboscine, the Firewood-gatherer (Anumbius annumbi). Pairs produced an average of 2.8 clutches over the very long (August–April) breeding season. Clutch size (mean = 5.1 eggs, range: 3–8) was very variable between and within pairs and showed an initial increase followed by a decline, the typical pattern of variation for multibrooded temperate species. The main determinant of breeding success was predation. Although the probability of nest failure increased over the season from 70% to >90%, clutch size was not correlated with failure risk. Egg volume did not show consistent seasonal patterns. Our results suggest that reproductive investment is not optimally adjusted to the seasonal variation in nest predation risk, and more research is needed to assess the influence of food availability.