The breeding range of Barrow's Goldeneye (Bucephala islandica) is largely restricted to northwestern North America, and little is known of the small population that winters in eastern Canada. Based on weak evidence, this eastern population was thought to nest mainly in northern Labrador. Our May 1990 to 1998 surveys identified a breeding area in the forest regions of the Quebec Laurentian Highlands. We observed Barrow's Goldeneyes on 137 lakes and 5 rivers, of which 95.2% were along the north shore of the St. Lawrence estuary and gulf. The species was found mainly on small lakes (≤ 10 ha) at greater than 500 m elevation. Most occupied lakes (96.5%) were within 100 km of the St. Lawrence River and 48.9% of them were headwater lakes. Four broods observed in 1998 represent the first eastern North American documentation of breeding. By means of satellite telemetry, 5 of 7 males captured on the wintering grounds were relocated on the north shore of the St. Lawrence River in May, 60–140 km inland from the estuary and gulf. Each male spent 34–50 days at its respective site, presumably with a mate. The north shore of the estuary and gulf may be the core breeding area for Barrow's Goldeneyes wintering along the St. Lawrence River.

We examined the winter diet of Steller's Eiders (Polysticta stelleri) in Varangerfjord, northern Norway, by analyzing the esophagus content of 29 individuals (12 juveniles and 17 adults). A total of 8389 prey items of 31 species were identified: 13 species of gastropods (making up 68.4% of total number of items); 4 species of bivalves (18.5%); 12 species of crustaceans (13%); and 2 species of echinoderms (<0.1%). In terms of percentage aggregate wet weight 31.4% was gastropods, 22.6% was bivalves, and 41.4% was crustaceans. Juvenile eiders ate more crustaceans (x̄ = 61% aggregate w.w.) than adults (x̄ = 26%, P < 0.05), possibly because they were in poor body condition and may have had higher energy requirements than adults. Adults tended to eat more gastropods (x̄ = 41% vs 22%) and Mytilus edulis (x̄ = 27% vs 12%) than juveniles. There were only small differences between sexes. Most of the prey items were of species known to be associated with kelp plants, especially Laminaria hyperborea, suggesting that Steller's Eiders obtain a large proportion of their prey directly from the vegetation.

We identified causes of nest failure and assessed the effectiveness of predator exclosures at Piping Plover (Charadrius melodus), Snowy Plover (C. alexandrinus), and Killdeer (C. vociferus) nests in southeastern Colorado during 1994–1995. Predation, nest abandonment, and weather were the primary causes of nest failure in all three species. For any of the three species we found no significant difference in the daily survival rate between nests that were protected by predator exclosures and nests that were unprotected. We reevaluated the experimental design and data analysis of previous predator exclosure studies and identified several confounding factors, including non-random assignment of exclosures, unbalanced sample sizes between protected and unprotected nests, data pooling across years, and inappropriate statistical analyses. We suggest ways to design (e.g., randomly allocate exclosures to nests and balance sample sizes between protected and unprotected nests) and analyze (e.g., use Mayfield method) future predator exclosure studies.

We monitored the movements of nine radio-tagged, adult, male Cooper's Hawks (Accipiter cooperii) in adjacent home ranges during the breeding seasons of 1996 or 1997 in Tucson, Arizona, to ascertain the sizes and degree of overlap of home ranges, and to assess habitat selection at two spatial scales. Size of home ranges differed among hawks (13.3–130.6 ha), but the average was small [65.5 ha ± 40.7 (SD)] compared to the size of home ranges reported for Cooper's Hawks in the literature. Home range size generally decreased with the number of years that a hawk had lived on its breeding territory. Only one pair of home ranges overlapped each other; overlap of one home range on the other in this pair was 14.2% and 10.6%. Proportions of land-use categories in home ranges varied widely among hawks, and suggested that the hawks did not select their home ranges on the basis of the categories we examined. Patterns of habitat use inside individual home ranges suggested that male hawks hunted primarily in the environments that surrounded their nests. Cooper's Hawks in Tucson feed primarily on doves [Mourning Doves (Zenaida macroura), Inca Doves (Columbina inca), and White-winged Doves (Zenaida asiatica)], and we speculate that the abundance of doves throughout Tucson allowed the hawks to hunt successfully in several urban environments. We also speculate that Cooper's Hawks in Tucson have relatively small home ranges because they do not need to range far from their nests to find food.

A review of stopover behavior in birds revealed that most migrate as quickly as fuel reserves allowed (i.e., were time minimizers). This pattern stemmed from studies conducted almost exclusively on small-bodied species (< 50 g), which migrate primarily by flapping flight. However, body size has a significant effect on metabolic rate and costs of locomotion, two characteristics with direct relevance to migration and stopover behavior. I examined the autumn stopover behavior of a large-bodied species (4–5.5 kg) that migrates by soaring. I captured and radio-tagged 68 Bald Eagles (Haliaeetus leucocephalus) at Hauser Reservoir, Montana from 1992–1994. After capture, juveniles stayed at the stopover site longer than older eagles. Body condition was similar between the two age groups and did not correlate with post capture residence time. Consumption rates of eagles were not age-specific. Juvenile and 1.5-year old eagles had similar consumption rates and body condition, but differed in residence times, which suggested that experience modified stopover behavior. Stopover behavior of Bald Eagles supported the hypothesis that large species are less sensitive than small species to body condition and consumption rates during migration.

Structural variability of guttural calls of Blue-fronted Amazons (Amazona aestiva) was examined and the contexts in which they were given discussed relative to the role these vocalizations might play in short-range communication. Recordings were obtained at the Chancaní Reserve, Córdoba, Argentina, during breeding and non-breeding seasons. Guttural calls were given year round, but were more common during the non-breeding season when most of the individuals were interacting in flocks. Gutturals were detected when perched parrots contacted each other, during take-offs, landings, complex flock flying maneuvers, and feeding sessions. Guttural calls were subdivided into four types based on structure, but none of these types could be clearly assigned to a specific context. Structural variation in guttural calls was continuous rather than discrete, with the variability within and among individuals being similar. Many guttural notes graded into one another and were combined with other vocalizations. Gutturals were brief, had sudden onset, wide bandwidth, and low intensity. Gutturals could be regarded as short-range calls because of their large structural variability (fewer restrictions of sound attenuation and degradation) and low intensity (decreased attraction of predators). They were also produced by large flocks, probably as a means of maintaining contact, enhancing group spacing, and coordinating movements of individuals.

Throughout western North America, Red-naped Sapsucker (Sphyrapicus nuchalis) nests have been previously described primarily in trembling aspen (Populus tremuloides) with decay-softened wood. During 1974–1992, we located Red-naped Sapsucker nest trees (n = 125) in northwestern Montana old-growth coniferous forest that included widely scattered paper birch (Betula papyrifera). Sapsucker nests were in nine tree species (seven conifers). Most (68%) nest trees were live and 75% had broken tops. Western larch (Larix occidentalis) and birch were greatly over utilized compared to their availability. Larch nest trees (n = 84) were large [mean DBH = 69 ± 20.95 (SD) cm]. Mean DBH of birch nest trees (n = 30) was 37 ± 8.42 cm. All Mountain Chickadee (Poecile gambeli) nests (n = 36) and 12 of 23 Red-breasted Nuthatch (Sitta canadensis) nests were in old sapsucker excavated nest holes. Wood of larch and birch is inherently harder than that of aspen (specific gravity = 0.48, 0.48, and 0.35 respectively), posing a potential obstacle for relatively weak excavators such as sapsuckers. However, the entire inner wood column of birch is susceptible to decay fungi and the durable bark is thin. In larch sapsuckers mitigated the difficulty by selecting trees with extensive heartwood decay (old larch) and by excavating in the upper bole (mean cavity height = 21.5 m), where the bark is thinner. External evidence of heartwood decay was present in 87% of larch and 86% of birch. Decay incidence increases with age in western larch forests, amplifying their value as habitat for sapsuckers and many other species. Perpetuation of old-growth western larch is an important component in the conservation of biological diversity. Received 8 March 1999, accepted 24 August 1999.

New information on the vocalizations and behavior of populations of the Pale-tipped Tyrannulet (Inezia subflava), along with analysis of biometric and plumage characters, reveal that it actually consists of two species-level groups: a northern/Guianan group, the member taxa of which are weakly differentiated from one another; and a group whose distribution is primarily Amazonian, the member taxa of which are morphologically distinct but vocally and behaviorally similar to one another. The two groups differ from one another in several vocal, plumage, and biometric characters. Tape playback experiments suggest that the vocal differences alone present effective mechanisms for reproductive isolation between the two groups. Analysis of vocalizations further suggests a possible link between the Amazonian representatives of Inezia subflava and the Yellow Tyrannulet (Capsiempis flaveola). Received 10 May 1999, accepted 9 Sept. 1999.

We compared recordings of the ‘gargle’ vocalization of Black-capped Chickadees (Poecile atricapillus) made in the field to recordings made in the laboratory of captive birds from the same populations. Individuals have repertoires of gargle calls, some of which are shared with conspecifics. Thus, a population sample of recordings reveals a variety of distinct gargle calls constituting the population repertoire. Captive birds were individually stimulated to vocalize by viewing their own image in a mirror. In the field, gargle calls were recorded during natural social interactions. We questioned whether the field recordings might reveal a greater variety of gargle calls than we found in the laboratory samples. We found that none of 901 field recorded gargle calls differed from those recorded from captive birds.

We investigated whether male parental assistance, reproductive output, and overwinter survival differed for female House Wrens (Troglodytes aedon) that did and did not have to share their mate's territories with a later-settling female during the nestling stage of breeding. During first breeding attempts of the season, when mate sharing occurred, primary females in polygynous trios and females from monogamous pairs fledged equal numbers of offspring and fledged offspring of similar mass. Rates at which males fed broods of primary females were not significantly different than rates at which males fed broods of monogamous females. Primary females did not make significantly more trips to nests to feed offspring than did monogamous females. In the first year of study, primary females were more likely than monogamous females to attempt a second brood. In the second year of the study, however, primary females tended to be less likely to attempt a second brood, took longer to start such broods, and tended to be less successful in fledging offspring from second broods. This suggests that in some years there may be a delayed cost of mate sharing. Females with primary and monogamous status for first breeding attempts in a particular year were equally likely to return to the study area the next year. Overall, our results indicate that mate sharing does not appear to affect the reproductive output of first mated females in our House Wren population in the first breeding attempts of the season, but may affect output in second attempts (and hence possibly annual and lifetime success), at least in some years.

We used inexpensive (<$30) cameras to document predators at active Wood Thrush (Hylocichla mustelina) nests in Great Smoky Mountains National Park. We observed such predators as black rat snakes (Elaphe obsoleta), American Crows (Corvus brachyrhynchos), southern flying squirrels (Glaucomys volans), and black bears (Ursus americanus) remove the contents of nests. Camera installation had no measurable effect on nest survival; daily nest survival was approximately 0.96 for nests with and without cameras. However, placement of an artificial egg trigger in the nest appeared to reduce hatching success. The immobile egg trigger might have interfered with the female Wood Thrush's ability to incubate her eggs. The variety of nest predators observed and the moderate daily survival rates recorded suggest that predation is an important constraint on Wood Thrushes nesting in large contiguous forests.

We quantified flocking behavior and examined the impact of social context (solitary, single-species flocks, and mixed-species flocks) on the foraging behavior of Prothonotary Warblers (Protonotaria citrea) wintering in a Costa Rican mangrove forest and surrounding habitats. Based on observations collected over two winters during 70 visits to four sites, 87% (483) of the 555 Prothonotary Warblers encountered moved in flocks and over 48% (271) of these individuals were in single-species flocks. Although the propensity to join flocks was 6% higher for Prothonotary Warblers in the second winter of the study, neither the average size of single-species flocks nor the average number of individuals or species in mixed-species flocks differed between years. Twenty-seven different species were identified in mixed-species flocks that had at least one Prothonotary Warbler, but Nearctic migrants dominated these flocks. Analyses of focal observations on 57 females and 93 males indicated that Prothonotary Warbler foraging behavior was largely independent of flock type and size. Foraging maneuver, substrate, and location did not differ significantly for individuals of either sex foraging alone, in single-species, or mixed-species flocks. The species is almost strictly insectivorous, gleaning made up 70% of 150 prey capture attempts observed and about half of all attempts (76 of 150) were directed towards leaf surfaces. Foraging generally occurred in the outer third of the tree, on branches less than 1 cm in diameter, in the bottom half of the canopy. Agonistic interactions among flock members that involved Prothonotary Warblers were uncommon and neither flock type nor size were useful predictors for rates of foraging, movement, preening, or vigilance.

The two subspecies of Prairie Warblers (Dendroica discolor) differ in migratory behavior and habitat use. A concurrent study of mitochondrial DNA variation showed that behavioral differences between subspecies may persist because little gene flow occurs between subspecies. To determine whether geographic variation in morphology parallels variation in migratory behavior and mitochondrial DNA, I used measurements of museum specimens from throughout the Prairie Warbler's range. The subspecies differ in overall body size, with birds of the Florida subspecies (D. d. paludicola) significantly larger than the nominate race. Males of the subspecies differ in the extent of white coloration in the outer rectrices. These differences were apparent even among specimens collected in northern Florida and southern Georgia, where the subspecies' ranges are most proximate. Morphological differences coincide with differences in behavior and mitochondrial DNA and support the recognition of migratory and non-migratory forms of Prairie Warblers as subspecies and as potentially independent evolutionary lineages.

Habitat features found within 4.1 ha sample areas surrounding the song perches of 33 Indigo Buntings (Passerina cyanea) and 33 Painted Buntings (Passerina ciris) in northeast Texas were compared to determine whether these species segregated according to habitat. The species did not differ in the proportions of open habitat or the type and amounts of successional and mature woodland in the sample areas surrounding their song perches. Indigo Buntings were mostly associated with lower elevations where they occurred along the edges of successional and mature woodlands. Painted Buntings showed no elevation bias. Compared to Indigo Buntings, Painted Buntings were associated with smaller, more numerous, and more heterogeneous stands of trees. In general, Indigo Buntings typically occurred where there were open areas within otherwise wooded habitats whereas Painted Buntings tended to occur where there were wooded areas in otherwise open habitat.

We examined macro- and microhabitat characteristics of breeding Henslow's Sparrows (Ammodramus henslowii) on Fort Riley Military Reservation, Kansas during 1995 and 1996. Survey points were identified at the macrohabitat scale as either grassland, savanna, or woodland edge. A military disturbance index was used to quantify the severity of training disturbance to the vegetation at survey and bird use sites. At the large scale, Henslow's Sparrows were associated with grassland habitat last burned in 1993, two or three years previously. Microhabitat at Henslow's Sparrow use sites had lower tree density than random survey points, but neither shrub density nor military disturbance index differed between use sites and survey points during spring. In summer, the military track index was higher on Henslow's Sparrow's use sites. Habitat used by Henslow's Sparrows was consistently tall and dense vegetation with high litter cover during early spring, late spring, and summer whereas the vegetation of random survey points changed in response to vegetation growth. Characteristics of Henslow's Sparrow use sites included high cover by litter and dense, structurally homogeneous vegetation, whereas litter depth and standing dead vegetation, physiognomic diversity, and military disturbance did not differ from random survey points.

To investigate the use of high elevation fir forests by fall passage migrants, we conducted standardized mist-netting and banding at a 1150–1175 m elevation site on Mt. Mansfield in north-central Vermont during the autumns of 1995–1997. Overall, we captured 3024 individuals of 62 species in 10,048 cumulative net hours (30.1 birds/100 net hr). We divided species into 3 classes: (1) those breeding regularly on Mt. Mansfield above 916 m elevation (16 species), (2) those breeding only sporadically or at very low densities above 916 m (6 species), and (3) those occurring only as transients (40 species). Breeding species accounted for 68% of new captures, followed by transients (22%), and sporadic/low density breeders (10%). Of the ten most abundant species captured, only two, Black-throated Blue Warbler (Dendroica caerulescens) and Ovenbird (Seiurus aurocapillus), were transients, accounting for 50% and 9%, respectively, of all transient captures. Hatching-year birds accounted for 94% of known-age transients and 81% of known-age individuals among locally breeding species. Recapture rates of transients were extremely low (0.3%), while recapture rates among species known to breed locally (including presumed transient individuals) were higher (2.2%). Nearly 75% of all birds captured were very lean at first capture, and only 26% of recaptured individuals increased their fat scores between first and final captures. Weight changes of recaptured birds varied: 48% lost weight, 44% gained weight, and 8% maintained the same weight between first and final captures. Our data suggest that conditions on the Mt. Mansfield ridgeline are not conducive to prolonged migratory stopovers and that most migrants may be unable to meet their energetic requirements for continued migration. However, we believe that montane forest habitats may be preferentially selected by those migrants that use them for breeding. We further believe that montane fir forests may be an important postfledging dispersal habitat for Black-throated Blue Warblers and other low or mid-elevation breeding species, and that conservation planning for montane forest sites should carefully consider the needs of migrants outside the breeding season.

We report range extensions of Caprimulgus sericocaudatus, Automolus ochrolaemus, Dichrozona cincta, Poecilotriccus andrei, Ramphotrigon fuscicauda, Ramphotrigon megacephala, Turdus lawrencii, and Euphonia chrysopasta for Pinkaiti and Gorotire, east of the Xingu River, southeastern Pará State, Brazil. The distributional status of the following species in southeastern Amazonia is also discussed: Nonnula ruficapilla, Synallaxis cherriei, and Simoxenops ucayalae. The new records we report suggest that the patchy distribution of some birds in eastern Amazonia, such as species associated with bamboo, can be explained by a complex mosaic of habitats at the eastern and southern fringes of the Amazonian basin. The range extensions we report demonstrate that areas inventoried a long time ago, before an emphasis on vocal identification of species, and assumed to be “well known”, can in fact turn out to be ornithologically undersampled.

There is little information concerning differences in migration chronology between male and female Common Snipe (Gallinago gallinago) and virtually no accounts of sex-related differences in winter habitat use. We collected 372 Common Snipe in five different habitat types during the non-breeding period along the central Gulf Coast of Texas. Proportions of male and female snipe collected on wintering areas during the beginning of the fall period (i.e., between 6 and 21 October) indicated a tendency for females to arrive ahead of males. Sex ratios during the latter part of spring (16 March–10 April 1998) suggested male snipe leave wintering areas before females. During the winter period (14 November 1997–4 February 1998), female snipe were more common than males along the Texas Gulf Coast. Differences in sex ratios during winter are likely due to sex-related differences in habitat use. During winter, females were more common than males in heavily vegetated habitats (e.g., coastal marshes and cultivated rice fields). Conversely, males were more common in open habitats (e.g., mud flats). Male snipe may begin spring migration before females to establish territories on the breeding grounds. Sex-specific differences in winter habitat use may be related to reverse sexual size dimorphism of Common Snipe.

Although the American Dipper (Cinclus mexicanus) uses a variety of sites on the ground adjacent to streams for nocturnal roosts, I observed nocturnal roosting in a tree by this species, apparently the first reported case for any dipper species. A fledgling spent at least 8 hours between 20:06 and 04:30 MST sleeping 1.5 m high in a black cottonwood tree (Populus trichocarpa), at the tip of a branch overhanging a creek. Use of arboreal roost sites may reduce the probability of predation on fledgling dippers while they are sleeping.

The reproductive success of avian brood parasites depends, to a great extent, on their ability to locate host nests that are at the appropriate stages of the host laying cycle. Consequently, brood parasites are expected to possess elaborate mechanisms and search modes to locate potential host nests. Through observing a population of Song Sparrows (Melospiza melodia) parasitized by the Brown-headed Cowbird (Molothrus ater) we examined two specific factors that may influence cowbird parasitism of a ground nesting host. Proximity to potential perches was a significant predictor of cowbird parasitism, but overhead nest visibility, either classified dichotomously as visible or not, or measured as the absolute area of a nest visible to an observer, was not correlated with the likelihood of parasitism. Comparisons with previous studies suggest that female cowbirds use similar nest searching mechanisms in open habitats, irrespective of the height of host nests.

We observed male Dickcissels (Spiza americana) commonly feeding nestlings in Conservation Reserve Program (CRP) fields in 1997 in east-central Illinois. Male Dickcissels fed nestlings at six of the eight nests we observed, accounting for 37% of the total nest visits. Overall, females made significantly more nest visits than males. However, at the six male-assisted nests, the number of male and female nest visits did not differ significantly. Male Dickcissel feeding behavior may have been prompted by low food abundance. Males were not observed feeding nestlings in 1998, when overall nest success was higher and nestling starvation was less than in 1997.