We measured the bones of extinct great auks Pinguinis impennis that were killed during recent centuries on Funk Island off the northeast coast of Newfoundland. Comparisons of these measurements with those taken elsewhere suggest that great auks from Funk Island, which is situated in a Low Arctic oceanographic region, were larger than conspecifics from Boreal oceanographic regions. This finding is supported by extant alcid species that inhabit Boreal, Low Arctic or Boreal through High Arctic ocean regions and tend to increase in body size with increasing latitude (generally decreasing sea surface temperature). We suggest that paleoecological sea surface temperatures and food webs may have favored oceanographic-related variation in body sizes of great auks. The variances of the bone sizes of great auks from Funk Island were not less than those of a sample of great auk bones collected from Scandinavian archaeological sites that cover an extensive geographic range and that span seven millenia. This finding is inconsistent with a previously suggested latitudinal cline in body size among great auks in Scandinavia. Research techniques and studies that could address questions of great auk feeding ecology and population genetics are considered.

Field and activity—specific metabolic rates of 20 free—living Northern Gannets (Sula bassanus; mean mass = 3.21 kg) rearing chicks at Funk Island, Newfoundland, were measured using doubly labeled water and activity timers. Field metabolism (FMR) averaged 4865 kJ/d or 6.6 x basal metabolism (BMR). Regression analyses indicated a metabolic rate of 144 kJ/h while at the nest or on water, 349 kJ/h during flight, and 250 kJ/h during diurnal time at sea. These metabolic rates are high, probably because of costs of thermoregulation and flapping flight. BMR was slightly lower than predicted. Statistical analyses of metabolic rates of free—living seabirds indicated that rates are elevated in seabirds in cold—water regions and in seabirds that use flapping flight, but do not vary with phylogenetic order. FMR was correlated strongly with both BMR and metabolism at the nest. Population energy models may gain accuracy if metabolic rates are either expressed as multiples of metabolism at the nest or calculated from allometric equations appropriate for the species' activity pattern, oceanographic regime, and foraging mode.

Northern gannets, Sula bassanus, and possibly other seabird species nesting on Funk Island off northeastern Newfoundland preyed on postsmolt Atlantic salmon, Salmo salar. Salmon comprised less than 1% of 2928 regurgitated food samples collected from gannets at the colony. Ten smolt tags were recovered in and near the gannetry during August or September in 1984 through 1986. The tags were from smolts released 3–4 mo earlier in the Penobscot River (Maine) (n = 7) and one each from the Saint John River (New Brunswick) and the Lower Clyde and LaHave rivers (Nova Scotia). These recoveries provide evidence that postsmolt Atlantic salmon from rivers in New England, the Bay of Fundy, and the Atlantic coast of Nova Scotia migrate off eastern Newfoundland This migratory pattern contrasts with that of postsmolts from the Gulf of St. Lawrence, which tend to move northwards along Newfoundland's west coast and through the Strait of Belle Isle.

An activity recorder is described that can be used to determine time spent by aquatic animals diving, at the surface of the water, and out of the water. Time spent flying also can be calculated for volant aquatic birds if time on land is measured by direct observation. Data are displayed on a digital screen that can be read from a distance through binoculars or a telescope or directly from an animal in hand. The recorder is simpler to use than other nontransmitting activity recorders and has wide application in timeenergy studies of aquatic wildlife. J. WILDL. MANAGE. 51(2):395-399 The allocation of time to various activities is an important element of life history strategy and energy balance in animals. Several studies have measured activity budgets of aquatic birds, mammals, and reptiles through direct field observation (Titman 1981, Maxson and Bernstein 1984, Cooper 1985) and radio telemetry (Siniff et al. 1971, Obbard and Brooks 1981, Melquist and Hornocker 1983). Direct observation is unsuitable for cryptic, nocturnal, or widely mobile species, and determination of activity budgets from telemetry may be hampered by limited transmitter range and difficulties of interpreting signal patterns (Harrison and Stoneburner 1981, Wanless et al. 1985). Recently, Wilson and Bain (1984) and Wilson and Achleitner (1985) developed auto-radiographic meters for marine birds and mammals, which can be used to calculate time budgets (Nagy et al. 1984), and Prince and Francis (1984) devised an immersion timer that uses electrolyte deposition to measure the time albatrosses (Diomedea spp.) spend on the water. These devices do not require that the animal remain within visual or transmission range of the observer, but they require careful calibration before use and laboratory equipment to decode results. We have developed a new type of activity recorder for aquatic wildlife that is inexpensive, gives directly readable results, operates for long periods without maintenance, and, in some circumstances, yields activity data without need to recapture the animal. This paper outlines design and construction of this recorder and describes its application to 2 marine birds, the common murre (Uria aalge) and the northern gannet (Sula bassanus). We are grateful to K. G. Adams, D. L. Cairns, and G. F. Pierce for essential background information. F. Anderka of Holohil Systems Ltd. gave valuable advice and suggestions. We thank R. MacLellan and the staff of Mem. Univ. Tech. Serv. for assistance with many construction details, and W. Jacobs of Mem. Univ. Faculty Eng. for testing electronic designs. Thanks are also due to S. Parent of the Montreal Aquarium for arranging timer trials on captive birds and to G. Chaisson of Mem. Univ. Mar. Sci. Res. Lab. for pressure-testing prototype timers. Fieldwork in the Witless Bay and Funk Island Ecol. Reserves was authorized by the Newfoundland and Labrador Wildl. Div. This research was funded by Can. Dep. Fish. and Oceans Subvention Grants (DKC and WAM), Natl. Sci. and Eng. Res. Counc. Can. Grants A0687 and E6828 (WAM), and the Newfoundland Inst. for Cold Ocean Sci. and Psych. Dep. Mem. Univ. This is Rep. 116 from the Newfoundland Inst. Cold Ocean Sci. DESIGN AND PRINCIPLE OF

The marine distribution of Northern Gannets in the southern Labrador Sea and western North Atlantic Ocean was studied by aerial surveys over a 7-month period in 1981. Gannets were seen on every biweekly aerial survey from mid-April to the end of October. Densities were highest within a 60 km radius of Funk Island, but few gannets were seen offshore outside this radius. Before July, most gannets were seen south of Labrador (52°N), but as the breeding season advanced, gannet densities increased in northern areas. Densities in most areas peaked during mid-July. From 1977 to 1982 we collected food samples from gannets in the colony on Funk Island, Newfoundland. A broad spectrum of prey are taken. Mackerel and herring, prey that are associated with warm water, appear to be preferred owing to their large sizes and high energy densities. Prey occurrence in gannet food samples was associated with the time of spawning (capelin) and migration near Funk Island (herring and mackerel). The northernmost region of warm and cold water mixing appears to be the main factor limiting the northern distribution of gannets in the western North Atlantic Ocean.

Tagging of 398 Atlantic wolffish, Anarhichas lupus, 81 spotted wolffish, A. minor, and 101 northern wolffish, A. denticulatus, with Petersen disc tags in the Newfoundland area during 1962-66 yielded 5.0, 7.4 and 3.0% returns respectively.Recapture information indicates that migrations are generally short but some rather long migrations were noted.The rate of recapture of Atlantic wolfish declined rapidly from 15.4% for coastal taggings to 6.7% for offshore taggings except Flemish Cap and to 1.1% for Flemish Cap tagging.Depths of recapture of Atlantic wolffish ranged from 20 to 476 m. ansen (1958) mentioned that Atlantic and spotted wolffishes were tagged off West Greenland, and Jonsson (1982) reported on extensive tagging of Atlantic wolffish off Iceland.c;l>stvedt (1963) described the migrations of spotted wolffish from tagging in the northern Norwegian and Barents Sea area, and Albikovskaya (1982) referred to tagging of spotted wolffish in the Barents Sea.These taggings showed usually short migrations but occasionally long ones of 200-500 naut.miles, the longest being especially evident in Jonsson's (1982) tagging of A. lupus in Icelandic waters.

Bryum knowltonii Barnes, known previously in North America only from northern Canadian stations, has been found in arctic-like tundra of the Mount Zirkel Wilderness in Colorado. Bryum knowltonii Barnes was collected August 1981 in the Mount Zirkel Wilderness, north-central Colorado. The habitat is flat to gently undulating tundra, usually wet, arcticlike rather than the typically drier alpine tundra in Colorado. This species is known in North America from Funk Island, Newfoundland, from Southampton Island, Northwest Territories, and from Fort Churchill, Manitoba. Andrews (in Grout 1936) reported it from a few localities in arctic America, south to Newfoundland. Determination was made by Elva Lawton and confirmed by A. C. Crundwell, with the assistance of F. J. Hermann. Diagnostic features include indistinctly bordered ovate leaves, revolute to the apex or nearly so, acute to shortly acuminate, with costa subpercurrent to slightly excurrent, synoicous reproduction, capsules broadly pyriform, pendent, with a short, distinct neck, and spores 16-27 um. Endostome segments are narrow and the cilia rudimentary. A helpful characteristic, pointed out by Dr. Crundwell, is that the capsules mature at different times instead of simultaneously as in most Brya. Crum and Anderson (1981) give a complete description. Grout's Moss Flora and Bryologia Europaea use the synonym B. lacustre. United States collection: COLORADO. JACKSON CO.: Rainbow Lakes Trail approaching the Continental Divide, just W of the Roxy Ann Lake turnoff, 1 km N of Mt. Ethel, 3400 m, 40?40'00N, 106'40'35W, Rolston 81106 (coLo; GL; NY; US; WTU). Bruch, P., W. P. Schimper & T. Giimbel. 1836-1855. Bryologia Europaea. 3 Vols. Reprint (1971). A. Asher, Amsterdam. Crum, Howard A. & Lewis E. Anderson. 1981. Mosses of Eastern North America. 2 Vols. Columbia University Press, New York. Grout, A. J. 1936. Moss Flora of North America North of Mexico. Bryaceae. Newfane, Vermont. 2: 219-220. 007-2745/82/420$0.25/0 This content downloaded from 157.55.39.45 on Fri, 02 Sep 2016 06:13:29 UTC All use subject to http://about.jstor.org/terms

Seabird Transects between Valleyfield and Funk Island, Newfoundland,Summer, 1969

The Gannets of Funk Island

Lucas on the Anatomy and History of the Great Auk The Expedition to the Funk Island, with Observations upon the History and Anatomy of the Great Auk Frederic A. Lucas