I describe a new species of Micrastur (forest-falcon) from the rainforests of Brazil and adjacent northeastern Bolivia. Initial detection of the cryptic new taxon was enabled through hearing its distinctive voice, notably different from any of its congeners. Several specimens of the undescribed species subsequently were located in several museums; because the new species closely resembles M. gilvicollis (Lined Forest-Falcon), these specimens had remained unrecognized for more than a century. The new taxon not only has a vocal repertoire that differs from those of congeners, M. gilvicollis, M. plumbeous (Plumbeous Forest-Falcon), and M. ruficollis (Barred Forest-Falcon), but it also exhibits subtle yet consistent morphological distinctions that distinguish it from all other forest-falcons. This elusive raptor inhabits humid terra firme forest in southeastern Amazonia, and a disjunct population exists in the Atlantic rainforests of eastern Brazil (the latter known only from historic specimens) and merits great conservation concern. I summarize current information on the new species’ distribution, offer insight into its natural history and conservation, and discuss its systematic relationships within the genus Micrastur.

Canyon Wrens (Catherpes mexicanus) occur throughout the semiarid regions of the western United States in habitats dominated by canyons and steep rock formations. Their inaccessible habitat has made them one of the least studied among North American birds. We studied a population of Canyon Wrens in the Front Range of Colorado, documenting many aspects of their breeding biology. We report on territory density and size, nest site characteristics, nesting phenology, nesting success, and nestling development.

We studied orientation of nest sites relative to nearby vegetation for dabbling ducks (Cinnamon Teal, Anas cyanoptera; Blue-winged Teal, A. discors; Gadwall, A. strepera; Mallard, A. platyrhynchos; and Northern Shoveler, A. clypeata) and Short-eared Owls (Asio flammeus) in ungrazed grassland habitat during 1995–1997 in westcentral Montana. We estimated an index of vegetation height and density in intercardinal directions (NE, SE, SW, NW) immediately around nests. All species oriented nests with the least vegetation to the southeast and the most vegetation to either the southwest or northwest. Furthermore, maximum vegetation around nests shifted from the southwest to the northwest with increasing nest initiation date, apparently as a response of individuals tracking seasonal change in the afternoon solar path. Thus, nests were relatively exposed to solar insolation during cool morning hours but were shaded from intense insolation in the afternoon throughout the breeding season. We suggest that nest microhabitat was selected in part to moderate the thermal environment.

We analyzed landscape characteristics surrounding Mexican Spotted Owl (Strix occidentalis lucida) nest and roost trees in Arizona at three spatial scales: one circular plot of 201 ha (800-m radius) and two 400-m-wide “ring” plots between 800 m and 1,600 m from each nest or roost tree. The percentages of vegetation types were significantly different between 51 owl and 51 random areas only within the 201-ha circular plots. Owls selected both mature and young mixed conifer forests that had high canopy closure (≥55%) more than expected based on availability. Owls selected pine (Pinus spp.) and pine-oak (Quercus spp.) forests in proportion to availability. Forty-one percent of all nests and roosts were located in mixed conifer forests, even though this forest type covered only 5% of the study area. Pine and pine-oak forests covered 78% of the study area, and 59% of nests and roosts were located in these forest types. The only forest type in which we did not locate nests and roosts was mature open canopy ponderosa pine (P. ponderosa) forest. Owls occupied areas of predominantly younger forests, but only if residual large (≥45.7 cm dbh) trees were present. Indices of landscape structure did not differ significantly between owl and random areas. Future management of Mexican Spotted Owls in areas of moderate topographic relief should focus on retention of mature forests, especially mixed conifer stands with canopy closure ≥55%. Residual large trees, especially Gambel oaks (Q. gambelii), are important microhabitat components in younger forests.

We measured productivity and vegetation parameters of habitat quality at 16 Sharp-shinned Hawk (Accipiter striatus) nests in and near the greater Montreal area in order to evaluate nesting habitat use and its possible relationship to reproductive success in an urban setting. Mean clutch size was 4.4 and hatching success was 3.8 eggs per nest. At least one egg hatched in 11 of 16 nests (68.8%), 10 (62.5%) pairs raised young to a bandable age (≥10 days old), and 8 (50%) pairs successfully produced at least one fledgling. Immature individuals comprised 33.3% of male and 38.5% of female breeders. Mean values in the habitat assessment included nest tree height, 14.0 m; tree density, 955/ha; total canopy cover, 88.1%; coniferous cover, 39.7%; mean dbh, 17.6 cm; and distance to the nearest forest opening, 19.7 m. Sharp-shinned Hawks nested in a range of forest types, from mature conifer plantations to young, almost purely deciduous stands, and this population exhibited considerable flexibility with respect to most of the habitat features that we measured. Their use of older stands with more deciduous cover than those used by conspecifics elsewhere may reflect regional differences in habitat availability as well as in the abundance of competitor species. Breeding in an urbanized area does not seem to be detrimental to Sharp-shinned Hawks, as evidenced by this population’s relatively large proportion of immature breeders and normal productivity, which appeared to be independent of all the assessed parameters of habitat quality.

The Early Bird Hypothesis predicts that males fledged early in the breeding season have an advantage over their later-fledged counterparts during competition for breeding sites. We tested this hypothesis by examining the sex ratio of 1,025 fledglings from 265 broods of American Kestrels (Falco sparverius) breeding in nest boxes in Wisconsin during the period 1968–1997. We found a seasonal shift in the sex ratio: the sex ratio of fledglings was biased toward males early in the breeding season, but became increasingly biased toward females as the season progressed. Our results provide support for the Early Bird Hypothesis and suggest that the steepness of this trend may decrease with increasing latitude.

The Broad-winged Hawk (Buteo platypterus) breeds in eastern and central Canada and the United States, and winters in Central America and northern and central South America. Birders and ornithologists count migrating Broad-winged Hawks at dozens of traditional watch sites throughout the northeastern United States. We modeled counts of migrating Broad-winged Hawks from two raptor migration watch sites: Montclair Hawk Lookout, New Jersey, and Hawk Mountain Sanctuary, Pennsylvania, to determine whether annual abundance and trend estimates from individual sites within the mid-Atlantic states are representative of the region as a whole. We restricted ourselves to counts made between 10:00 and 16:00 EST during September to standardize count effort between sites. We created one model set for annual counts and another model set for daily counts. When modeling daily counts we incorporated weather and identity of individual observers. Akaike’s Information Criteria were used to select the best model from an initial set of competing models. Annual counts declined at both sites during 1979–1998. Broad-winged Hawk migration began, peaked, and ended later at Montclair than at Hawk Mountain, even though Hawk Mountain is 155 km west-southwest of Montclair. Mean annual counts of hawks at Montclair were more than twice those at Hawk Mountain, but were not correlated. Broad-winged Hawks counted at Montclair may not be the same birds as those counted at Hawk Mountain. Rather, the two sites may be monitoring different regional subpopulations. Broad-winged Hawks counted at the two sites may use different migration tactics, with those counted at Hawk Mountain being more likely to slope soar, and those at Montclair more likely to use thermal soaring. A system of multiple watch sites is needed to monitor various breeding populations of this widely dispersed migrant.

The Western Sandpiper (Calidris mauri) is the most numerous shorebird on the Pacific coast of North America during spring migration. Twenty-six sandpipers carrying miniature radio transmitters on northward migration across the Fraser River delta, British Columbia, remained on or near to the section of beach where they were first detected through their entire stay. They traveled about 4–6 km each day, following the tide up and down the beach. Although some Western Sandpipers were present on all beaches of the delta, our results suggest that individuals do not roam widely once they settle. The widespread distribution of sandpipers in mud, marsh, and sand habitats suggests that all regions of the Fraser River delta are required to support the large number of Western Sandpipers during spring migration.

St. Matthew Island (60° 24′ N, 172° 42′ W) and its small nearby satellites, Hall Island and Pinnacle Rock, are isolated in the northcentral Bering Sea. This infrequently visited location occupies a geographic position with a deep Bering Land Bridge history and is in an area of interdigitation of the Old World, New World, and Beringian avifaunas. It is known for its three Beringian endemics, a bird (McKay’s Bunting, Plectrophenax hyperboreus), a small mammal, and a plant. This level of endemism is striking for a high-latitude island. The only previous summary of the avifauna of St. Matthew island (Hanna 1917) included 37 species. Our report considers more than 125 species and synthesizes data on presence and absence, abundance, and phenology. Because visits have been infrequent and concentrated during summer, our understanding of migration in this region remains poor, but the area is clearly affected by both the Old and New world migration systems. There is sufficient evidence to show that some profound changes among the island’s breeding birds have occurred during the past century. In particular, the breeding range of Glaucous-winged Gulls (Larus glaucescens) has been extended north to include St. Matthew, a change that is correlated with a northward shift in the extent of sea ice (Maslanik et al. 1996). King and Common eiders (Somateria spectabilis and S. mollissima) also have shown substantial changes in summer abundance. Other changes in the summer avifauna (e.g., among shorebirds) may reflect the dynamics of edge-of-range phenomena. Because of its central position in a region undergoing profound climate change and its demonstrated track record in showing avifaunal shifts, St. Matthew Island may represent an important bellwether for monitoring the biological effects of further climate change in the northern Bering Sea.

Many animals advertise the presence of a predator threat through vocal signals. Black-capped Chickadees (Poecile atricapilla) use their chick-a-dee call as a mobbing call when encountering a perched hawk or owl. This social signal appears to serve as an alert to other chickadees, causing them to rally to the vicinity of the predator and join in a chorus of calling. We asked the question: do chickadees vary the mobbing call in a manner that could convey the immediacy of threat from a potential predator? We examined the responses of chickadees to a taxidermic mount of an avian predator presented at distances of 1 m and 6 m from each subject. Vocal responses were recorded and analyzed for response latency, calling rate, and syllable composition of calls. During 5-min trials, the subjects responded more quickly and produced significantly more chick-a-dee calls for predator presentations at the 1-m distance than at the 6-m distance. Alterations of syllable composition of the call also were observed under the two treatments. These results suggest that information about the immediacy of threat or proximity of a predator may be signaled by alteration of the rate of calling, with possible additional information contained in proportional changes in the different syllable types of the call. Studies of referential (symbolic) communication in birds and mammals often have failed to consider the problem of response urgency separately from predator-type labeling in vocal signal design.

I describe the nest, egg, and nesting behavior of the Snowy Cotinga (Carpodectes nitidus) in La Selva Biological Station, Costa Rica. The nest was placed 7.5 m high on a fork formed by four branches of a leafless tree. The nest was a simple platform made of dry tendrils and lichens. It contained a single large egg, which took at least 27 days to hatch. The nest and chick were predated the day after hatching. Both nest and egg characteristics resemble those of other cotinga species.

Very little demographic data is available for rocky intertidal shorebirds, including the long-lived Black Oystercatcher (Haematopus bachmani). In this paper we report on Black Oystercatcher chick production from 1992 to 2000, age of first breeding and natal philopatry in Laskeek Bay, Queen Charlotte Islands, British Columbia. Five percent of birds banded as half-grown chicks returned to breed in the same area. Known-age birds were 5 years old when first found breeding. This constitutes the first published evidence of natal philopatry for this species.

We report observations of Ring-necked Pheasants (Phasianus colchicus) parasitizing Lesser Prairie-Chicken (Tympanuchus pallidicinctus) nests in native sand sagebrush (Artemisia filifolia) rangeland in southwestern Kansas. We found low incidence of interspecific nest parasitism as only 3 of 75 prairie-chicken nests were parasitized. Two of the three parasitized clutches hatched, but no Ring-necked Pheasant chicks were known to have survived.

We report the Turquoise-browed Motmot (Eumomota superciliosa) feeding in the evening on insects attracted to artificial lights. This behavior has been observed in few diurnal bird species, and has not been reported previously for the family Momotidae. Although more than one motmot engaged in this behavior at one site in El Salvador, and feeding success appeared favorable, we have not observed the behavior at other sites in El Salvador, suggesting that the behavior has not spread through the population.

We report Fan-tailed Warblers (Euthlypis lachrymosa) foraging in association with the nine-banded armadillo (Dasypus novemcinctus) in El Imposible National Park, El Salvador. Although the warbler is known to forage opportunistically at ant swarms, this is the first report of commensal feeding with a mammal.

Ospreys (Pandion haliaetus) over much of Montana incorporate large amounts of bailing twine in their nests and active nestlings sometimes become entangled and perish. We observed 12 occurrences of entanglement in 260 nests during three years of study. Disposing of the twine or cutting the twine into smaller pieces when removing it from hay bales could reduce or eliminate the problem.