Information on breeding and wintering Harlequin Ducks (Histrionicus histrionicus) in Greenland is very limited, and no firm estimates of breeding and wintering numbers are available. However, it is assumed that only a few thousand pairs breed in Greenland. Numbers of molting males are estimated at 5,000 to 10,000 birds. They comprise both Greenland and eastern Canadian breeders; their proportions are unknown, but the latter may very well constitute a significant part. If true, Greenland has a significant responsibility for the well being of the eastern Canadian breeding population. Presently, there are no immediate threats to the Harlequin Duck population in Greenland. Hunting (illegal) has only negligible effects and habitat destruction is extremely limited. However, oil spills from transport of oil and increased offshore oil exploration constitute a potentially serious threat to molting and wintering birds from the breeding populations in Greenland and in eastern Canada.

In Iceland, the Harlequin Duck (Histrionicus histrionicus) is generally distributed as a breeder on rapid clear-water streams, from sea level to about 700 m. Densities are generally low, except at the headwaters of streams draining productive lakes where blackflies (Simulium vittatum) are abundant. In northwest, north and east Iceland, molting Harlequin Ducks are found in similar localities as the wintering birds, but molting birds are largely absent from the southwest coast. During the molt, the birds often stay close to cliffs or rocks and are hard to find. Harlequin Ducks winter on exposed rocky coasts and densities are correlated with substrate as indicated by shore exposure. In 1998-2001, the total wintering population of Harlequin Ducks in Iceland was estimated at 14,000, with 95% confidence limits of 12,000 to 16,000. Assuming that migration in and out of Iceland is insignificant, the total Icelandic breeding population is 3,000-5,000 females, or three to five females (pairs) 100 km^-2. A decline in wintering numbers was found in southwest Iceland in 1961-2002, but numbers in the northeast have increased. This may be related to increased temperatures, but changes in exploitation may also have affected the population. The conservation of the Icelandic population of the Harlequin Duck is largely a matter of wise use of the coast and streams which should include preemptive measures, such as minimizing pollution and development. Small-scale river development for power production and salmon fishing pose potential threats to Harlequin Ducks, but direct exploitation of this population is not likely.

The distribution and abundance of the Harlequin Duck (Histrionicus histrionicus) is poorly known in Arctic Canada. The limited historical information on this species is summarized, and new field surveys and interviews with Inuit hunters in 1998-2002 were also conducted. Recent data confirmed that Harlequins still occur on Baffin Island and are breeding. The majority of new data on this duck come from the Kimmirut area of southeastern Baffin Island, where surveys in the 1930s also found the species. However, it occurs as far west as Cape Dorset, and as far north as Clyde River. Few birds were observed during recent field surveys, suggesting that the population on Baffin Island is small and dispersed. Future monitoring of Harlequin Ducks in Nunavut will probably be best accomplished using community-based surveys by hunters.

In the last decade, several survey and research efforts have been undertaken to better understand the ecology and distribution of Harlequin Ducks (Histrionicus histrionicus) in Québec. The results of these efforts are summarized and new survey, research and management needs are identified. Harlequin Ducks breed on rivers and streams of the Gaspé Peninsula, the Québec North Shore, and the immense Hudson Bay and Ungava Bay drainage basins. Québec is possibly the most important breeding area for Harlequin Ducks in eastern Canada, and there is every indication that a significant proportion of the birds wintering in south-western Greenland and in eastern North America breed in Québec. In spring, significant numbers of birds spend a few days or weeks along the Gaspé Peninsula before departing for their breeding or molting areas. The most important molting sites in Québec are Bonaventure Island, the Port-Daniel/Newport area, and the Pointe du Sud-Ouest/Jupiter River area on Anticosti Island. Known fall staging areas include the coastal waters of the Gaspé Peninsula, the Pointe du Sud-Ouest/Jupiter River area, and Brion Island, on the Magdalen Islands. The few Harlequin Ducks that overwinter in Québec are mostly in Baie des Chaleurs, on the south shore of the Gaspé Peninsula. There are no good data on trends of Harlequin Ducks in Québec. Major threats to population recovery/health include hydroelectric developments, illegal and subsistence harvests, coastal exploitation (aquaculture, fisheries, boating), and recreational activities on breeding rivers and streams. Survey, research and management needs include better estimates of breeding densities and distribution, characterization of spring, summer and fall coastal habitats, comprehensive genetic analysis, monitoring, and public education on the status of Harlequin Ducks.

Labrador represents a significant portion of the breeding range for the Northwest Atlantic population of the Harlequin Duck (Histrionicus histrionicus). The region comprises almost 300,000 km of wilderness area with habitat for breeding, molting and staging activity. The wilderness quality of Labrador has also posed several challenges in terms of understanding the status of this species. Expensive and logistically-challenging survey programs have only recently (i.e., last 20 years) targeted Harlequin Ducks but due to the co-ordination amongst agencies and proponents, an appreciation of the range, numbers, and timing of movements for this species in Labrador has been developed. Over 800 helicopter survey hours have been completed on at least 111 river systems/sections during late May-early June (July in northern-most areas) from 1987-2008. Presence of the species was widespread and confirmed on approximately 67% of areas examined. It is estimated that 395 breeding pairs occur on rivers where presence has been confirmed and the majority of potential habitat for this species has been examined in Labrador. The actual number of breeding pairs, their success and other population limiting factors remain unclear. Regardless, Harlequin Ducks are common in Labrador and the population appears to be stable or increasing.

Relatively low numbers of Harlequin Ducks (Histrionicus histrionicus) are present year-round in Newfoundland, with breeding, molting and wintering at specific areas. Breeding is largely restricted to the Great Northern Peninsula, but there is some evidence of breeding in remote areas of the eastern part of the island. A large molting concentration is present on the Grey Islands, east of the Northern Peninsula. Harlequin Ducks winter on the south coast of Newfoundland, with the largest concentration at Cape St. Mary's. Counts from Cape St. Mary's indicated Harlequin Ducks decreased through the 1980s, but have been increasing rapidly since the early 1990s. Although anecdotal, information from early sources suggests populations of Harlequin Ducks in Newfoundland are much reduced. Chronic oiling and misidentification of Harlequin Ducks by hunters remain as threats, while forestry and hydroelectric development may be potential sources of habitat reduction.

Harlequin Ducks (Histrionicus histrionicus) are a relatively rare species in the Canadian Maritime Provinces. Recent confirmed breeding records are restricted to northern New Brunswick. Sightings of small numbers of birds during the molting and staging periods occur in all three provinces. It is estimated that New Brunswick supports as many as 200 wintering Harlequin Ducks, and almost 600 winter off the coast of Nova Scotia, although due to the extent of coast and occasional sightings of Harlequin Ducks from previously unknown locations, it is likely that these are minimum estimates, particularly for Nova Scotia. Few trend data are available in the Maritimes, but available information suggests an increase in the last five years. Hunting, both intentional and unintentional, still occurs; hunter education and enforcement to stop the take of this species could be worthwhile. Coastal development, including vacation home building and aquaculture sites, are occurring in places where Harlequin Ducks winter. There is a general need to monitor the impacts of these developments on local Harlequin Duck populations.

Harlequin Ducks (Histrionicus histrionicuS) arrive along the eastern coast of the United States at traditional wintering grounds from late September to December through early January. Counts in January and February best represent wintering bird numbers. Spring migrants depart from April to May, although from more southerly wintering areas, birds appear to depart in March. Breeding has never been confirmed in the eastern United States. Molting females have been documented in Maine during the fall, but the extent of molting has not been assessed. It is estimated 1,575 to 1,800 Harlequin Ducks wintered in the eastern United States from Maine to North Carolina between the winters of 1997-98 and 2001-02. Over 75% of the birds wintering in the eastern United States were in Maine, primarily in the vicinity of Isle au Haut. Smaller concentrations representing about 18% of the wintering birds in the eastern United States were at locations in Massachusetts and Rhode Island, primarily the Cape Anne region and Martha's Vineyard, Massachusetts and the Sachuest Point region, Rhode Island. South of Rhode Island, Barnegat, New Jersey is the only area that regularly reports counts of over 25 birds. In the Isle au Haut region of Maine, numbers along survey routes have not shown a linear change from 1989-2002, but show a pattern of decline from 1989 to 1993 followed by increases since then. In eastern Maine, population increases are greatest and birds are expanding their winter range into new areas. All other areas in the eastern United States south of Isle au Haut that regularly report over 50 birds show significant population increases since the 1980s, north-eastern US.

Mid-winter activity budgets and diving behaviors of Harlequin Ducks (Histrionicus histrionicus) at Isle au Haut, Maine were examined. Feeding was the most frequent activity and the number of hours during the day devoted to feeding increased from December to March as day length increased. Resting was not a frequent mid-winter activity, but birds rested more in the morning than in the afternoon and the duration of resting bouts increased in March, likely in relation to increasing day length. During foraging bouts, first-winter males had longer submergence times and longer pause times than adult males, although the resulting dive:pause ratio was similar. This may be related to feeding inefficiency of the young birds or lack of experience with local habitats or prey species. Mean dive durations and pause durations at Isle au Haut were greater than reported elsewhere for this species, and may be related to strong local currents associated with the wave washed ledges on Isle au Haut.

Foraging behavior of Harlequin Ducks (Histrionicus histrionicus) and its response to changing environmental conditions during winter was investigated at Cape St. Mary's, Newfoundland. Behavioral synchrony among individuals permitted continuous observations of flocks to be conducted, sometimes over entire days. Overall foraging effort of Harlequin Ducks was examined at two organizational levels: dive cycles (dive and surface pause) and foraging cycles (foraging bout and rest bout). Overall foraging effort decreased at greater tide depths as Harlequin Ducks decreased the duration of foraging bouts. Dive:pause ratios did not change within these shorter foraging bouts at high tides, however the duration of dives, pauses, and the total dive cycle all increased. Overall foraging effort decreased in response to increased wind/wave exposure due to a decrease in dive duration. Overall foraging effort did not change in response to decreasing ambient temperature; however the frequency of dive cycles decreased which could decrease energy expenditure associated with post-dive thermoregulatory costs. Overall foraging effort increased throughout the day, particularly in the last foraging bout before the overnight fasting period. Interestingly, this strategy was accomplished by decreasing dive durations but increasing foraging bout duration. These opposite results across levels of behavioral organization are interpreted in the context of intermittent exercise and locomotion whereby decreasing effort at one level of energy expenditure could allow for increased effort at another. Therefore, different (and sometimes opposite) responses to environmental conditions can occur at different levels of behavioral organization. Even when overall foraging effort remains unchanged, the strategy employed can differ in its temporal allocation or frequency, which could be important in balancing energy budgets under increased energetic costs and/or time constraints. These results have particularly important implications for interpreting behavioral responses investigated at only a single level of behavior, extrapolating data from brief observation periods to longer time scales, and foraging models which only consider single levels of behavior such as the dive cycle.

Female Harlequin Ducks (Histrionicus histrionicus) incubate eggs and rear broods, while males normally depart the breeding grounds at the onset of incubation. On 27 July 2000, a male Harlequin Duck was observed in association with a female and brood of five ducklings on the Ikadlivik River in northern Labrador. The male maintained some distance (∼100 m) from the female and brood; however, upon appearance of a researcher, the group amalgamated and the male appeared to influence the behavior of the female and brood. Possible explanations for early male departure and this exception are discussed.

Harlequin Ducks (Histrionicus histrionicus) and aquatic insects have been monitored on the River Laxá at Myvatn, from 1975 and 1977, respectively, up to the present. The area holds the densest breeding population of this species in Iceland. The relationship between food abundance, reproductive success and density of breeding Harlequin Ducks were examined on three sections (outlet, upper and lowland) of the Laxá. Spring density was highest on the lake outlet (2.1 ha^-1) and much lower on the upper (0.3 ha^-1) and lowland (0.4 ha^-1) sections. Numbers of Harlequin Ducks on the Laxá increased significantly during the study period, 1975-2002. The trend of increasing numbers could be attributed to either a general, as yet unexplained, increase of Harlequin Ducks in northern and northeastern Iceland, or perhaps to the successful eradication of American Mink (Mustela vison) breeding at the lake outlet (the main production centre of young Harlequin Ducks). Sex ratio in spring was 41% females at the outlet and on the upper Laxá but 36% on the lowland part. Yearling males were about 1% of all males on the upper river but 6-19% on the lowland part. Most young were produced at the outlet, while few females or broods were found further down the river. The mean total number of young produced annually at the outlet was 74 ± 11 (range = 3-235). Young per female present in spring at the outlet and upper part was 0.49 ± 0.09, or if only females at the outlet in spring are considered, 0.82 ± 0.17. Annual production of young was positively correlated with total yearly black fly abundance. We conclude that food limits the production of young in the Harlequin Duck. Changes in the densities of adults on the breeding grounds showed density dependence but were not significantly associated with food resources or previous production of young.

Harlequin Ducks (Histrionicus histrionicus) breeding on the upper Torrent River of the northern peninsula of Newfoundland were studied from 1993 to 2002. This reach of the watershed was proposed for hydro-electrical development in the 1990s. Harlequin Ducks arrive on the Torrent River as soon as ice-out permits in early to late May. Pairs concentrate in the upper watershed that drains the Long Range Barrens ecoregion. Nesting and brood rearing appear to take place primarily in proximity to spring-pair activity, although movements of more than ten km were observed. Young remain in the upper watershed until fledging in late August to mid-September. Counts of Harlequin Ducks breeding on the upper Torrent River increased throughout the 1990s, and the estimated population growth rate for the Torrent River birds was very similar to the growth rate for birds wintering at Cape St. Mary's, Newfoundland (1.14 vs. 1.13) during the same period. Paired females spent ∼ 40% of the day in feeding activities whereas paired males allocated less time to feeding (∼19%) and more time to alert or vigilant behaviors while females fed. Birds rested for 35-40% of the day. Harlequin Ducks exploited rapids, riffles and runs, and were especially associated with the boulder-strewn inlets and outlets of ponds. The large quantities of submerged and semi-submerged angular boulders provide increased surface area for attachments of larval insects, and movement of water through boulders assures high rates of oxygenation important to filter-feeding insects. Broods selected areas with flow rates of 1.27 ± 0.24 m/s. The family Chironomidae contained the most taxa and was the most numerous invertebrate group found in the Torrent River. Chironomidae may comprise the most important food items for pre-nesting Harlequin Ducks when considering volume of prey types consumed. A high rate of brood production in 1997 and 1998, compared to adjacent watersheds, suggests the possibility that the Torrent River system may behave as a source population for the general region of northern Newfoundland.

The number of Harlequin Ducks (Histrionicus histrionicus) wintering in eastern North American is small, and little was known about these birds, especially their breeding distribution. During the breeding period, 990 km of river and streams on Newfoundland's North Peninsula was surveyed where 69 Harlequin Ducks were sighted. Breeding densities varied among watersheds (range 0-0.090 males/km), were slightly lower than elsewhere in eastern North America, and lower than in western North America and Iceland. A total of 128 ± 45 males (indicated pairs) or 284 ± 89 birds (±95% CI) was estimated to occur on the North Peninsula during the breeding season. This may represent 20% of Harlequin Ducks wintering in eastern North America and highlights the importance of the Northern Peninsula as a breeding area for this rare duck.

As part of the Hydro-Québec Grande-Baleine (Great Whale) hydroelectric project feasibility studies, Harlequin Duck (Histrionicus histrionicus) surveys were conducted in 1990 and 1991 in the eastern Hudson Bay and James Bay drainage basins. A total of 142 and 420 Harlequin Ducks were counted in 1991 and 1992, respectively, of which 142 (1991) and 356 (1992) were found in the area surveyed both years. Most individuals were in pairs and the overall sex-ratio did not deviate significantly from 1:1. The highest numbers of Harlequin Ducks counted over the two years were found on the Little Whale River, Des Loups-Marins Lake, and Nastapoka, À l'Eau Claire and Boutin Rivers. Highest pair densities were observed in June 1992 on rivers located in tundra and forest tundra i.e., the lower Little Whale, À l'Eau Claire and Nastapoka Rivers, and near D'Iberville Lake. In 1992, pair densities varied between 0.003 and 0.093 pair/km, depending on the watershed, and followed a latitudinal gradient. Two broods were located in 1991 and three were found during a preliminary survey conducted in 1989. Broods were located on Boutin, Nastapoka, and Great Whale Rivers, as well as along the Hudson Bay coast. The difference in the number of Harlequin Ducks found in June 1991 and 1992 may have been related to weather and methodological factors. Considering the vastness of northern Québec and the limited area surveyed during this study, we suggest that Harlequin Ducks breeding in northern Québec may well number in the thousands, and represent a very high proportion of the Greenland molting and wintering populations.

In 1996, a study to locate and describe breeding sites, estimate pair densities and document brood movements of Harlequin Ducks (Histrionicus histrionicus) was conducted on selected rivers of the Gaspé Peninsula, Québec. Pair densities on the Port-Daniel, Sainte-Anne and Madeleine River systems were low: 0.32, 0.10 and 0.06 pairs/km, respectively. Six VHF radiotransmitters were implanted in nesting females on the Port-Daniel and Sainte-Anne Rivers. Two successful nests were located on cliff ledges and one unsuccessful nest was located on the ground on an island. These nests represented the first recorded for the eastern North American Harlequin Duck population. Of the marked hens, two successfully hatched ducklings. One was followed for a month until the female was killed by a Great Horned Owl (Bubo virginianus). The brood used a six km stretch of the Sainte-Anne River during that period. The other brood was tracked for two months and concentrated its activities along a three km stretch of the Port-Daniel River. The female was located in coastal waters near Newport on 24 August 1996. Two of the marked females from the Port-Daniel River and one from the Sainte-Anne River migrated to Bonaventure Island, where they likely molted. A female captured on the Port-Daniel River was killed by a Red-tailed Hawk (Buteo jamaicencis) before its nesting status was confirmed. A female from the Port-Daniel River was located at the Sally islands, Maine, on 5 December 1996. The study confirmed the importance of the Gaspé Peninsula Rivers for breeding Harlequin Ducks and the use of adjacent coastal molting sites by postbreeding females. It also established a link with the major Maine wintering area.

Between November 1994 and November 1995, regular boat surveys for Harlequin Ducks (Histrionicus histrionicus) were conducted on The Wolves archipelago in New Brunswick, Bay of Fundy. The birds arrived in early November and departed by mid-May, reaching a peak of 38 birds in March. Sex ratios tended to be close to unity, consistent with observations in Maine in the mid-1990s. Juvenile age ratios were noticeably higher than those observed at other wintering sites, suggesting that the juvenile birds may prefer The Wolves archipelago as a wintering site. Overall, these surveys confirm the importance of The Wolves archipelago as a non-breeding area for this species.

Sea ducks have been captured with mist nets set across breeding streams and in drive traps at sea during the flightless period, but capture of flying birds on staging and wintering coastal areas presents a challenge. Here, we describe a highly successful technique for capturing Harlequin Ducks (Histrionicus histrionicus) at sea, modified from a mistnet set developed to capture Marbled Murrelets (Brachyramphus marmoratus) at sea. While the original method was passive, decoys were added to attract birds and occasionally birds were driven toward the nets using small boats. The capture technique proved to be safe and effective. Three hundred and seventy-eight birds were captured during 28 d of effort and no birds died in the nets or during handling. The technique has been further modified to capture molting Harlequin Ducks. The technique and its modified version could be easily used to capture other sea ducks on their staging, wintering and molting areas.

Capture-recapture data from a five-year field study of individually marked Harlequin Ducks (Histrionicus histrionicus) wintering at Isle au Haut, Maine was used to examine patterns in age- and sex-specific apparent survival and local movements. Adult females had lower annual apparent survival probabilities than adult males. Survival probabilities for adult females were lower during the summer season than the winter season. Adult males showed no differences in apparent survival between the summer and winter intervals and survival during the winter season was similar for adult males and females. There was little evidence to suggest differences in apparent survival between first winter males and females, although sample sizes, especially for first winter females, were small. Annual apparent survival rates were lower for first winter males than adult males and likely reflected a combination of greater dispersal and higher mortality. Adult males captured in April in the study area disappeared from the study area more than adult males captured in November and may represent spring dispersal of unpaired males searching for mates or individuals from other wintering sites gathering before spring migration. Greater dispersal of adult and first winter males to adjacent wintering sites in subsequent winters was noted than for adult females.

Survival rates are an essential parameter for understanding the dynamics and status of wild populations. Apparent survival (the combined probability of surviving and returning to a site) and capture rates were estimated for male Harlequin Ducks (Histrionicus histrionicus) (second-year, N = 30, adult, N = 83) that were captured and/or resighted on the Gannet Islands, Labrador, from 1999-2003. Most birds available for analysis were first captured locally (N = 74), although a significant number were originally banded on wintering grounds in Maine (N = 36), and the remaining three were banded on breeding grounds. Second-year male apparent survival rates were low (0.466 ± 0.118; 95% PLI: 0.256 - 0.689), and different from adult rates (0.744 ± 0.045; 0.647 - 0.822), likely due to increased dispersal of young birds. There was little evidence for annual variation in survival rates, but samples sizes were low. Apparent survival rates of male Harlequin Ducks were lower than found in most studies, suggesting that some males emigrate from the Gannet Islands molting site, but do confirm that there is some fidelity to molting sites, similar to observations of the Pacific population.

There is an increasing interest in using winter juvenile counts as indices of recruitment in species that show delayed plumage maturation, especially for species, such as dispersed breeders, for which it is difficult to obtain good productivity estimates. To date, however, the needed mathematical work to assist in interpreting these winter juvenile/adult ratios has not been conducted. A matrix-based population model is presented that is modified to allow the fecundity component to be measured in mid- to late winter. This model is simplified to a set of equations that relate adult survival, winter juvenile/adult ratios and population growth rates, allowing an assessment of population trend with only one survival rate and age ratio data. These models have the advantage of not requiring that age of first breeding be well known. As an example, age ratios and survival rates of Harlequin Duck (Histrionicus histrionicus) populations in British Columbia and Maine are presented. Models for both populations suggest recruitment of young is insufficient to compensate for adult mortality, which is contrary to observed trends. Although some further methodological work is needed, such as better estimates of true adult survival and a further understanding of how to obtain unbiased estimates of juvenile/adult ratios, these models may prove to be a useful tool to assess population trends when detailed demographic data are not available.

In order to better delineate affiliations between breeding, molting, and staging areas of the small Harlequin Duck (Histrionicus histrionicus) population wintering in eastern North America, in April 2001 satellite transmitters were implanted in eight adult males at Isle au Haut, Maine, a major wintering area for this population. Two birds were confirmed breeding on rivers on the Gaspé Peninsula, Québec, two birds molted in northern Labrador, and four birds migrated to southwestern Greenland during the molting season, including one bird that bred in Québec. The four birds tracked to southwestern Greenland likely molted there, although molting could be confirmed for only one bird. All birds tracked to Greenland, to breeding areas, and to molting locations returned to wintering sites in Maine the following wintering season. This study is the first confirmation of wintering male Harlequin Ducks from eastern North America molting in Greenland.

Satellite telemetry was used to determine the migration patterns between breeding, staging, and molting areas of adult male Harlequin Ducks (Histrionicus histrionicus) breeding in central Labrador during 2001 (N = 5) and 2002 (N = 6). Male Harlequin Ducks remained near their capture locations for an average of 19-20 d. Males spent an average of 18 and 34 d on the Labrador coast in 2001 and 2002, respectively, and moved along the coast between several important staging areas prior to migration to Greenland. Male Harlequin Ducks may be completing a pre-basic molt prior to migrating to Greenland. Seven males migrated from the Labrador coast to three distinct molting areas in Greenland. All transmitters expired by September preventing a confirmation of whether these males wintered in Greenland. The absence of band sightings during winter in eastern Canada, especially Maine, indicates that some Harlequin Ducks breeding in central Labrador may also winter in eastern Greenland.

The recent use of abdominally-implanted satellite transmitters to track movements of waterfowl is rapidly filling gaps in our understanding of their population structure and affinities. However, premature loss of transmitter signals is of serious concern. Such loss occurred in 21 out of 25 satellite transmitters implanted in Harlequin Ducks (Histrionicus histrionicus) in 1996-1998. The tracking of eight birds captured while migrating along the shores of Forillon National Park, Gaspé Peninsula, Québec, and the resightings of some of these birds after transmitter failure is detailed. The birds were followed for three to 373 d (mean = 127 d, SD = 82 d) until transmitter signal was lost, but four birds were resighted 493 to 1,474 d after the surgical procedures. These results indicate that premature failure of the transmitter is a common cause of signal loss, even when the battery voltage is adequate at the time of loss. Harlequin Ducks drakes implanted with satellite radios are able to live long after the transmitter has failed.

The movements of a total of 948 Harlequin Ducks (Histrionicus histrionicus) banded in eastern North America and Greenland from 1996-2003 (313 breeding, 186 molting/staging, and 449 wintering) were examined. Regular resighting and recapture projects have been underway at the Gannet Islands, Labrador, as well as Jericho Bay, Maine. Less consistent observations have been conducted on the Gaspé Peninsula, Québec, and incidental observations were obtained from the public. Affiliations between breeding, molting and wintering sites were assessed in an effort to determine movement patterns and the level of geographic separation. Population affiliations indicated by past satellite telemetry studies were verified by the capture-recapture data. Absolute geographic separation between the eastern North American wintering population and the Greenland wintering population could not be determined.

Harlequin Ducks (Histrionicus histrionicus) were captured and measured at a variety of staging, breeding, molting and wintering sites across their Northwest Atlantic range from 1996-2002. The consistency in the body measurements taken on the same birds across time and sites, as indexed by repeatability, was adequate (0.4-0.75), with the notable exception of total tarsus, which showed poor repeatability. Correlations among morphological measurements were weak within each sex. Some differences among locations were detected, notably in wing length, with birds breeding in Labrador showing longer wings than birds breeding further south and those wintering in Maine. Overall, however, major differences among sites were not apparent. Recent satellite and banding information indicate that many of these locations are linked and these observations are supported by this morphological data. Harlequin Ducks are similar to other ducks (except eiders) in that they are wide ranging and have disjunct populations, but show, at best, weak morphological differentiation across their range.