Pt tie eee 4 c NS “WS eh, eerie AS we by PAM AU Coan! waink ROS SINR ye see alae y Wok YAM Dts Uaty | o ; an Rw be : The Tyron et efae, ae Tete eiyye ee P68. vat aim Aeris Pa Paks es Oste Sony, hay, © 18 whrad. een "* Sing ate Dolo Beh ytbhisg Meek ad ok Loaner d the ta? rh a ita ant ie) Pee te ML Sy SYS awh, VV imi see SS enirs SD hes oe bolas erat vagy Sere My SM ea heey i. teeny ot - Dinmwihyen a ry ~ rend # VW datos Nid hws Te aaa on Sie SET ci Saves eae PRN ea nay 0 ie ee Vie eee 1 rai! * Ayia a 4j a) “ee The CANADIAN FIELD-NATURALIST Published by THE OTTAWA FIELD-NATURALISTS’ CLUB, Ottawa, Canada Volume 115, Number 1 January—March 2001 The Ottawa Field-Naturalists’ Club FOUNDED IN 1879 Patrons Her Excellency The Right Honourable Adrienne Clarkson, C.C., C.M.M., C.D. Governor General of Canada His Excellency John Ralston Saul, C.C. The objectives of this Club shall be to promote the appreciation, preservation and conservation of Canada’s natural heritage; to encourage investigation and publish the results of research in all fields of natural history and to diffuse infor- mation on these fields as widely as possible; to support and cooperate with organizations engaged in preserving, maintain- ing or restoring environments of high quality for living things. Honorary Members Edward L. Bousfield R. Yorke Edwards John A. Livingston Eugene G. Munroe Irwin M. Brodo Anthony J. Erskine Don E. McAllister Robert W. Nero William J. Cody W. Earl Godfrey Theodore Mosquin William O. Pruitt, Jr. Francis R. Cook C. Stuart Houston Stewart D. MacDonald Mary E. Stuart Ellaine Dickson George F. Ledingham Hue N. MacKenzie Sheila Thomson Bruce Di Labio 2001 Council President: Eleanor Zurbrigg Ronald E. Bedford Francis R. Cook Garry McNulty Vice-President: Roy John Rosanne Bishop Barbara Gaertner David W. Moore : Fenja Brodo Anthony Halliday Rita Morbia Recording Secretary: Ken Allison Irwin Brodo Terry Higgins Robert Roach Treasurer: Frank Pope John Cameron David Hobden Stanley Rosenbaum William J. Cody Beverly McBride David Smythe Dorothy Whyte To communicate with the Club, address postal correspondence to: The Ottawa Field-Naturalists’ Club, Box P.O. Box 35069, Westgate P.O. Ottawa, Canada K1Z 1A2,-or e-mail: ofnc @achilles.net. For information on Club activities telephone (613) 722-3050 or check http//www.achilles.net/ofnc/index.htm The Canadian Field-Naturalist The Canadian Field-Naturalist is published quarterly by The Ottawa Field-Naturalists’ Club. Opinions and ideas expressed in this journal do not necessarily reflect those of The Ottawa Field-Naturalists’ Club or any other agency. We acknowledge the financial support of the Government of Canada through the Publication Assistance Program (PAP), toward our mailing cost. Editor: Dr. Francis R. Cook, R.R. 3, North Augusta, Ontario KOG 1RO0; (613) 269-3211; e-mail: feook @achilles.net Copy Editor: Wanda J. Cook Business Manager: William J. Cody, P.O. Box 35069, Westgate P.O. Ottawa, Canada K1Z 1A2 (613) 759-1374 Book Review Editor: Dr. J. Wilson Eedy, R.R. 1, Moffat, Ontario LOP 1JO; e-mail: edith@netcom.ca Associate Editors: Robert R. Anderson Robert R. Campbell Brian W. Coad W. Earl Godfrey Charles D. Bird Paul M. Catling Anthony J. Erskine William O. Pruitt, Jr. Chairman, Publications Committee: Ronald E. Bedford All manuscripts intended for publication should be addressed to the Editor and sent by postal mail (no courier, no post requiring signature on delivery). Exception: book reviews should go directly to Book Review Editor. Subscriptions and Membership Subscription rates for individuals are $28 per calendar year. Libraries and other institutions may subscribe at the rate of $45 per year (volume). The Ottawa Field-Naturalists’ Club annual membership fee of $28 (individual) $30 (family) $50 (sustaining) and $500 (life) includes a subscription to The Canadian Field-Naturalist. All foreign subscribers and mem- bers (including USA) must add an additional $5.00 to cover postage. The club regional journal, Trail & Landscape, covers the Ottawa District and Local Club events. It is mailed to Ottawa area members, and available to those outside Ottawa on request: It is available to Libraries at $28 per year. Subscriptions, applications for membership, notices of changes of address, and undeliverable copies should be mailed to: The Ottawa Field-Naturalists’ Club, P.O. Box 35069, Westgate P.O. Ottawa, Canada K1Z 1A2. Canada Post Publications Mail Registration number 09477. Return Postage Guaranteed. Date of this issue: January—March 2001 (July 2001). Cover: Female Canada Goose, Branta canadensis, on its nest showing typical incubating low profile posture. The nesting habitat is the typical early spring low grass meadows of the Saint Lawrence River islands. Note the presence of last year’s dead vegetation with the new vegetation of the year. Taken on Ile aux Fermiers, Varennes, Quebec, by Martin Picard, Wildlife Biologist and Nature Photographer, 934 de la Rochelle, Boucherville, Quebec J4B 5M5 Canada. See: Establishment of a breeding population of Canada Geese, Branta canadensis, in southern Quebec by Jean-Francois Giroux, Josee Lefebvre, Luc Bélanger, Jean Rodrigue, and Stephane Lapointe, pages 75-81. THE CANADIAN FIELD-NATURALIST Volume 115 2001 Cag ed ‘" JUL 2 7 2001 LIBRARIES THE OTTAWA FIELD-NATURALISTS’ CLUB OTTAWA CANADA ft ’ x Py a A jo Boy as Onna iN pt poke The Canadian Field-Naturalist Volume 115, Number | January—March 2001 Merriam’s Shrew, Sorex merriami, and Preble’s Shrew, Sorex preblei: Two New Mammals for Canada. DAvip W. NAGORSEN!, GEOFFREY G. E. SCUDDER, DAVID J. HUGGARD3, HEATHER STEWARTI!, and NIcK PANTER! !Royal British Columbia Museum, PO Box 9815, Station Provincial Govt, Victoria, British Columbia V8W 9W2, Canada 2Zoology Department, University of British Columbia, Vancouver, BC V6T 1Z4, Canada 3Centre for Applied Conservation Biology, Forest Sciences, University of British Columbia, Vancouver, British Columbia V6T 1Z4, Canada Nagorsen, David W., Geoffrey G. E. Scudder, David J. Huggard, Heather Stewart, and Nick Panter. 2001. Merriam’s Shrew, Sorex merriami, and Preble’s Shrew, Sorex prebli: two new mammals for Canada. Canadian Field- Naturalist 115(1): 1-8. A specimen of Merriam’s Shrew (Sorex merriami) and three specimens of Preble’s Shrew (Sorex preblei) were captured in pitfall traps set for terrestrial arthropods in the southern Okanagan Valley of British Columbia from 1995-1998. They rep- resent the first captures of these mammals in Canada. The records are 130-300 km north of the nearest populations in Washington, but this distributional pattern probably reflects inadequate sampling. Both species were captured in shrub- steppe grassland habitats. Five species of shrews occur in the southern Okanagan Valley. To what extent they are syntopic in grassland habitats is unknown. Key Words: Merriam’s Shrew (Sorex merriami), Preble’s Shrew (Sorex preblei), pitfall traps, sagebrush-steppe, distribu- tion, Okanagan Valley, British Columbia, Washington. The shrew fauna of Canada has been reviewed nationally (van Zyll de Jong 1983) and regionally (e.g., Nagorsen 1996). Nevertheless, most distribu- tional data for these mammals are derived from gener- al museum collecting or broadly focussed ecological inventories using conventional snap or live-traps. Williams and Braun (1983) and Corn and Bury (1987) demonstrated that pitfall traps are more effective for capturing insectivores and inventories based solely on snap or live-traps may fail to detect the small or rare shrew species in a community. Consquently, our understanding of the geographic distribution and habi- tat associations of shrews is rudimentary especially for the ecologically diverse Cordilleran region of western Canada. Van Zyll de Jong (1983) speculated that Merriam’s Shrew (Sorex merriami), Preble’s Shrew (Sorex pre- blei), and the Dwarf Shrew (Sorex nanus), three species found in the western United States near the international boundary, could occur in western Can- ada. Recently, one of us (G. Scudder) captured one S. merriami and three S. preblei specimens in arthropod traps set in the Okanagan region of British Columbia. They represent the first captures of these species in Canada and extend their distributions considerably north of the nearest populations in Washington. Herein we describe the specimens, their habitat, and the biogeographic implications of these new records. Study Area and Methods Our study area was in the southern Okanagan region of British Columbia near the international boundary with Washington State. The area is bound- ed by the Cascade Mountains to the west and the Columbia Mountains to the east. Two major water- sheds, the Okanagan River and Similkameen River, form narrow, low elevation (300 m) valleys separat- ed by intervening rolling hills and low mountains (Figure 1). Situated in the rain-shadow of the coastal mountain ranges, the climate is dry and continental with wet springs, hot dry summers and variable win- ters (Cannings et al. 1987). Vegetation in the low valleys consists of shrub-steppe grassland and nar- row Strips of riparian habitat bordering lakes and rivers. Hills and lower slopes of mountains support open parkland habitats with scattered Ponderosa Pine (Pinus ponderosa) and Douglas-fir (Pseudo- tsuga menziesii). The region has been affected by cattle grazing, irrigation for agriculture, and urban development (Cannings et. al. 1987;* Bryan 1996*). *See Documents Cited Section D THE CANADIAN FIELD-NATURALIST The new shrew records were part of a sample of small mammals captured incidentally in pitfall traps for capturing terrestrial arthropods that were set in 20 study sites in the south Okanagan Valley from 1993-1998. Rings of five traps were placed at 50 m intervals along vegetational transects to sample spe- cific biophysical habitats. Each trap ring was 10 m in diameter; the five traps were placed in a pentagonal pattern at the periphery of the ring. The traps con- sisted of two plastic beakers (beer mugs) with a top diameter of 8.5 cm and depth of 11 cm. One beaker was set in the ground with the top level to the soil surface; the other beaker, snug-fitting into the first, was one-third filled with 50% propylene glycol. We checked and emptied traps at monthly intervals from April to October but we left traps unchecked from October to April. As a result, the precise capture date could not be determined for shrews. A total of 24 shrews were captured in 20 different traps from 14 study sites. We identified shrew specimens from dental and cranial traits using the keys and diagnostic traits given by Diersing and Hoffmeister (1977), Junge and Hoffmann (1981), Carraway (1995), and Nagorsen (1996). We measured condylobasal length with Helios dial calipers; other cranial measurements were taken with an ocular micrometer. For the four specimens suspected to be S. merriami or S. preblei, we extracted and cleaned their skulls. Their carcass- es and skulls are stored as voucher specimens in the collections of the Royal British Columbia Museum (RBCM). We confirmed our identification of S. mer- riami from reference specimens borrowed from the Connor Museum, Washington State University. The three specimens identified as S. preblei were sent to Lesley Carraway at Oregon State University for veri- fication. For the S. merriami and S. preblei sites, we recorded the dominant plant species, slope, latitude, longitude, elevation, percent shrub cover, distance to nearest tree, and distance to standing water at each site. Percent shrub cover estimates were based on 10m X 10m plots using the methods described by the British Columbia Ministry of Forests and British Columbia Ministry of Environment, Lands and Parks (1998*). Slope was calculated using the percent scale of a SUUNTO® clinometre. Aspect was recorded as the orientation of the slope measured in degrees by compass. The nearest tree to each trap site was measured with a 100-m steel tape. We esti- mated distance to nearest standing water from topo- graphic maps. We calculated coordinates and eleva- tion (x, y, and z) with a CMT-GPS® Global Positioning System device. The biogeoclimatic zone ecosystem classification for sites was described by Meidinger and Pojar (1991*); ecosystems were based on habitat polygons mapped by the British Columbia Ministry of Environment, Lands and Parks. Plant names follow Douglas et al. (1994*). Vol. 115 Results Sorex merriami RBCM 19982 (Table 1), was positively identified as S. merriami by the absence of medial tines on the first pair of upper incisor teeth. Other diagnostic traits included the third upper unicuspid tooth larger than the fourth and the presence of a postmandibular fora- men on the rami of the mandibles. Measurements are consistent with those described by Diersing and Hoffmeister (1977) for a sample of 53 individuals from Washington State (Table 2). The specimen was an adult female captured in summer (14 July—16 August 1996). The trap site (KL1-1) was about 0.5 km NW of Kilpoola Lake south of Richter Pass (Figures 1, 2; Table 1) in sagebrush rangeland. The biogeoclimatic zone ecosystem classification for this site is: Ponderosa Pine zone, very dry hot subzone (PPxh1). Although now protected as a special management area with cattle excluded by fencing, the area was heavily grazed until the early 1990s. Shrub cover was sparse (5%) consisting equally of Big Sagebrush (Artemisia tridentata) and Threetip Sagebrush (Artemisia tripar- tita) less than 1 m high. Grasses were dominant (> 50%) at the site and included: Fow! Bluegrass (Poa palustris), Japanese Brome (Bromus japonicus), and Columbian Needlegrass (Stipa nelsonii var doreii). Forbs included: Timber Milkvetch (Astragalus miser), Parsnip-flowered Buckwheat (Eriogonum hera- cleioides), Yarrow (Achillea millefolium), and Meadow Death Camas (Zigadenus venenosus). There was no tree overstory; a stand of Trembling Aspen (Populus tremuloides) was situated about 75 m from the site. Nearest standing water was about 0.5 km away at Kilpoola Lake. No other shrews were cap- tured at this site. Sorex preblei RBCM 19993, 20005, and 2006 were clearly members of the Sorex cinereus group with medial tines present on the edge of their first upper incisors and their third upper unicuspid teeth equal to or larg- er in size than the fourth unicuspids. We identified them as S. preblei from their short skulls, mandibles, and rostrums; and flat braincases. Measurements (Table 3) are consistent with those described by Cornely et al. (1992) and Carraway (1995). RBCM 19993 was an old adult of unknown sex captured 9 July—10 August 1995. RBCM 20005 and 2006 were captured between 5 October 1994-9 April 1995. RBCM 20005 was a subadult female; RBCM 2006 was a young adult male. The three Sorex preblei were taken from three sepa- rate sites: K3-2, Y5-2, and Z3-1 (Table 1). Trap site K3-2 (Figures 1, 3; Table 1) was on the south slope of Mt. Kobau, north of Richter Pass. The biogeoclimatic zone ecosystem classification for this site is: Interior Douglas-fir zone, wetter subzone (IDFdk1). The area was moderately grazed by cattle. The trap station was 2001 NAGORSEN, SCUDDER, HUGGARD, STEWART, AND PANTER: SHREWS 3 TABLE |. Habitat and geographic data for capture sites of three Preble’s Shrews (Sorex preblei) and a Merriam’s Shrew (Sorex merriami) from the southern Okanagan Valley of British Columbia. Shrub Species RBCM# Site Location Latitude Longitude Elevation Slope Aspect Cover S. merriami 19982 KL1-1 Kilpoola Lake 49°02’0.31’’N 119°33’56.90’W 827m -8% 123° 5% S. preblei 19993 K3-2. Mt. Kobau 49°06°12.37"°N 119°40°22.69"W 1724m ~-35% 148° 80% S. preblei 20006 Y5-2. Vaseux Creek 49°165.10"N = 119°30°55.43°W 343m -26% 312° 40% S. preblei 20005 23-1 Vaseux Creek 49°15'38.98’°N 119°30’32.39°N 452m e702 210-/ Vaio in a patch of open grassland surrounded by scattered stands of Douglas-fir. Big Sagebrush (Artemisia tri- dentata) and scattered Snowberry (Symphiocarpus albus) about 1 m in height formed a dense shrub cover (80%). Under this dense cover, grasses were dominant and included Bluebunch Wheatgrass (Elymus spica- ta), Idaho Fescue (Festuca idahoensis), Columbian Needlegrass (Stipa nelsonii var doreii) and Junegrass (Koeleria micrantha). Forbs included Nettle-leaved Giant Hyssop (Agastache utriculfolia), Parsnip- flowered Buckwheat (Erigonum heracleoides), Old Man’s Whiskers (Geum triflorum) and Fern-leaved Desert-parsley (Lomatium dissectum), Yellow Penstemon (Penstemon confertus). There was no tree overstory but scattered Douglas-fir trees were 13 m from the trap site. Nearest known standing water was at least 2.3 km from the site. The Dusky Shrew (Sorex monticolus) and Vagrant Shrew (Sorex vagrans) were also taken at this trap station and several other trap sites on Mt. Kobau. Trap site Y5-2 (Figure 1, Table 1) was on a bench about 0.9 km southeast of Vaseux Lake in the Vaseux-Bighorn National Wildlife Area. The biogeo- climatic zone ecosystem classification for this site is: Bunchgrass zone, very dry hot subzone (BGxh1). According to Krannitz (1997) the site was grazed by cattle and horses until 1996; the area is still grazed by various wildlife species including Bighorn Sheep (Ovis canadensis). Antelope Bush (Purshia tridenta- ta) about 2 to 2.2 m high formed a dense shrub cover (40%). Cheatgrass (Bromus tectorum) is the most widespread herbaceous plant at the site (Krannitz, unpublished data) with Diffuse Knapweed (Centaurea diffusa) the second most common plant species. Other grasses include Needle-and-thread grass (Stipa comata) and Sand Dropseed (Sporobolus cryptandrus). Other forbs included Annual Jacob’s Ladder (Polemonium micranthum), Small-flowered Blue-eyed Mary (Collinsia parviflora) and Brittle Prickly-Pear Cactus (Opuntia fragilis). There was no tree overstory but scattered Ponderosa Pine occurred 15 m from the trap site. Nearest standing water was at Vaseux Lake 0.9 km from the site. No other shrews were captured in this study site. Trap site Z3-1 (Figures 1, 4; Table 1) was on a bench about 1.7 southeast of Vaseux Lake. Habitat and the biogeoclimatic zone classification at this site was similar to trap site Y5-2. Although grazed by wildlife species, historically the site has had little grazing from domestic livestock (Krannitz 1997). Antelope Bush about 1.5—1.7 m high formed a dense shrub cover (30%). Cheatgrass was the dominant plant cover with Blue Forget-me-not (Myosotis stric- ta) almost as common. Other grasses included Needle and thread grass, Six-weeks Fescue (Vulpia octoflora) and Sandberg’s Bluegrass (Poa secunda). There was no tree overstory and the nearest trees were scattered Ponderosa Pine about 40 m from the site. Nearest standing water was at Vaseux Creek, about 350 m from the trap site. No other shrews were captured in this study site. Discussion Associated with the arid grasslands of western TABLE 2. Measurements (mean, range) for Merriam’s Shrews (Sorex merriami) from Washington State and the specimen from British Columbia. Measurement Tail length Hind foot length Skull length Braincase breadth Unicuspid toothrow length Breadth across 2™4 unicuspids Breadth across 34 upper molars Breadth across 4 upper premolars RBCM 19982 Washington@ 33 36 (30-43) 14 12 (11-14) 16.1 157 (1SA=16.1) 8.9 8.3 (7.88.3) i 2.0 (1.8—2.2) 2.0 2.1 (2.0—2.4) 4.3 4.4 (4.14.7) 4.4 4.3 (4.0-4.6) based on a sample of 53 specimens taken from Diersing and Hoffmeister (1977). 4 THE CANADIAN FIELD-NATURALIST nh Washington 012345 Kilometers ee VoLans BC Min. Env. Lands, Parks "Southern Intenor Region" Feb 25, 2000 Ficure 1. The southern Okanagan Valley region of British Columbia-and locations of the shrew collecting sites. North America, S. merriami is one of the most xeric- adapted of all North American shrews (Junge and Hoffmann 1981). In Washington where this species inhabits the Columbia Basin, it is mainly associated with Big Sage and Bunchgrass habitats (Johnson and Clanton 1954; Johnson and Cassidy 1997*). The British Columbian record is in similar habitat, and the Kilpoola Lake site is part of one of the largest tracts of sagebrush grassland remaining in the south- ern Okanagan Valley (Bryan 1996*). Although Johnson and Clanton (1954) and Johnson and Cassidy (1997*) reported that S$. merriami is always found in runways and burrows of the Sagebrush Vole (Lemmiscus curtatus), this vole is absent from British Columbia. However, in British Columbia S. merriami may use runways of the Montane Vole (Microtus montanus), the most common arvicoline rodent in the intermontane grasslands of southern British Columbia. Nagorsen (1995*) reported M. montanus in grassland habitat near the S. merriami site. Irregularly distributed throughout the Columbia Plateau, Great Basin, and Great Plains, S. preblei is associated with a wider variety of habitats than S. merriami. In the Blue Mountains of Washington, S. preblei was found at 1525-1830 m elevation in dense forests of Lodgepole Pine (Pinus contorta), Subalpine Fir (Abies lasiocarpa)-Lodgepole Pine, and Grand Fir (Abies grandis)-Englemann Spruce (Picea engelmanii) (Armstrong 1957). However, these habitats are atypical. In other regions of the western United States S. preblei inhabits arid steppe- grassland, montane forest, riparian areas, and sea- sonally wet sagebrush communities (Ports and George 1990; Cornely et al. 1992). Sagebrush and Antelope Bush are often mentioned in habitat de- scriptions for this species. The three S. preblei from British Columbia were taken in two distinct commu- 2001 NAGORSEN, SCUDDER, HUGGARD, STEWART, AND PANTER: SHREWS 5 FiGuRE 2. Merriam’s Shrew (Sorex merriami) capture site (KL1-1) at Kilpoola Lake, British Columbia. Photo taken 11 July 1999 by D. W. Nagorsen. nities: a low elevation Bunchgrass-Antelope Bush community and a montane Sagebrush- Interior Douglas-fir community. These communities shared the common habitat attributes of a dense shrub cover and no tree overstory. The S$. merriami and S. preblei records from Brit- ish Columbia appear to be isolated from the nearest populations in Washington, but this distributional pattern may largely reflect inadequate sampling par- ticularly in regions adjacent to the international bor- der. Because both species are rare and usually cap- tured in pitfall traps, they may escape detection in general mammal surveys. The nearest known popu- lation of S. merriami is 130 km south in the Colum- bia River basin. According to Johnson and Cassidy (1997*), the species is unknown from the west side of the Columbia River north of Ellensburg or Van- tage in Washington. Nevertheless, only three shrew specimens have been taken from Okanagan County (Johnson, personal communication), the county adja- cent to the Okanagan Valley of British Columbia. With suitable bunchgrass-sagebrush habitat found throughout the Okanagan River valley in Wash- ington, it seems likely that the British Columbian population of S$. merriami is linked to the Columbia Basin population via intervening populations. The TABLE 3. Measurements (mean, range) for Preble’s Shrew (Sorex preblei) from Oregon and three specimens from British Columbia. RBCM Measurement 19993 Total length - Tail length _ Skull length 14.0 Maxillary breadth 4.1 Maxillary toothrow length 5.5 Dentary length 6.3 Length of mandibular toothrow 4.1 RBCM RBCM 20006 20005 Oregon? i 76 91 (75-99) 33 32 36 (29-40) 14.3 14.5 14.4 (13.2-14.8) 4.2 4.3 4.0 (3.84.2) 5.4 3.8) 5.1 (4.7-5.4) 6.1 5.8 6.2 (5.7-6.8) 4.2 4.3 4.0 (3.84.1) abased on a sample of 15 males from Verts and Carraway (1998) 6 THE CANADIAN FIELD-NATURALIST 1999 by D. W. Nagorsen. distribution of S. preblei is more puzzling. The near- est population to British Columbia is more than 300 km south in the Blue Mountains of eastern Washington. Johnson and Cassidy (1997*) devel- oped a model for this species’ distribution based on GIS analysis, but they emphasized that the status of S. preblei in Washington was essentially unknown. Verts and Carraway (1998) suspected that the few records for this species probably result from sam- pling effort. Clearly comprehensive inventories employing pitfall traps are required to delimit this species’ distribution in Washington and British Columbia. We also note that there are records of S. merriami and S. preblei from northern Montana (Diersing and Hoffmeister 1977; Cornely et al. 1992) and both species could occur in southern Alberta. Although sagebrush-steppe habitats generally sup- port low population densities of shrews, species rich- ness may be high in these habitats. As many as five shrew species coexisted in some of the sagebrush- steppe communities studied by Kirkland et al. (1997). Shrew assemblages associated with the inter- montane grasslands of British Columbia are poorly known. Four species were recovered in our pitfall traps: S. merriami, S. monticolus, S. preblei, and S. vagrans with S. monticolus and S. vagrans account- Vols FIGURE 3. Preble’s Shrew (Sorex preblei) capture site (K3-2) at Mt. Kobau, British Columbia. Photo taken 9 September ing for 83% of the 24 shrew captures. Curiously, we did not trap the Common Shrew (Sorex cinereus), yet Kirkland et al. (1997) found it to be the dominant shrew in sagebrush-steppe habitats in Wyoming. There are historical museum records of S. cinereus from Penticton, Summerland, and Okanagan Land- ing in the Okanagan Valley of British Columbia (Nagorsen 1996) but they lack habitat data. To what extent these five species are syntopic in southern British Columbia is not clear. We trapped only one shrew species at most of our study sites, but S. pre- blei, S. monticolus, and S. vagrans were taken in the same trap station at one of the Mt. Kobau study sites. Our results should be viewed with caution as we used only a few pitfall traps in each study site. Kirkland et al. (1997) employed arrays of three trap lines of 25 pitfall traps each for a total sampling of 514 060 trap days. A similar sampling effort will be required to delimit shrew communities associated with the grasslands of the southern Okanagan Valley. Acknowledgments The British Columbia Ministry of Environment, Lands and Parks and the Canadian Wildlife Service permitted access to our study sites. We thank J. Jarrett for his assistance in the field, P. Krannitz for sharing 2001 NAGORSEN, SCUDDER, HUGGARD, STEWART, AND PANTER: SHREWS 7 Photo taken 9 September 1999 by D. W. Nagorsen. her vegetational data for the Vaseux Creek sites, F. Sachedina for preparing the map with GIS software, L. Carraway for confirming our identifications of the three S. preblei, and R. Johnson for providing a loan of S. merriami specimens and his unpublished data on Washington shrews. We are particularly grateful to O. Dyer for his support and assistance in various aspects of the study. Our research was supported by the Royal British Columbia Museum and by grants to G. G. E. Scudder from Forest Renewal British Columbia and the Natural Sciences and Engineering Research Council of Canada. Documents Cited British Columbia Ministry of Forests and British Colum- bia Ministry of Environment, Lands and Parks. 1998. Field manual for describing terrestrial ecosystems. Land Management Handbook 25, Crown Publications, Victoria. Bryan, A. 1996. Kilpoola-Kobau-Chopaka habitat manage- ment plan. British Columbia Ministry of Environment, Penticton. 95 pages. Douglas, G. W., G. B. Straley, and D. Meidinger. 1989. The vascular plants of British Columbia. Parts 1-4. Special Report Series. British Columbia Ministry of For- ests, Research Branch, Victoria. Johnson, R.E., and K.M. Cassidy. 1997. Terrestrial mammals of Washington Sate: Location data and pre- dicted distributions. Volume 3 in Washington State Gap Analysis-Final Report. Edited by K.M. Cassidy, C. E. Grue, M.R. Smith, and K. M. Dvornich. Washington Cooperative Fish and Wildlife Research Unit, University of Washington, Seattle, Washington. 304 pages. Meidinger, D., and J. Pojar, Editors. 1991. Ecosystems of British Columbia. Special Report Series, Number 6. British Columbia Ministry of Forests, Victoria. 330 pages. Nagorsen, D. W. 1995. Status of the western harvest mouse in British Columbia. Wildlife Working Report Number WR-71. British Columbia Ministry of Envi- ronment, Wildlife Branch, Victoria. 23 pages. Literature Cited Armstrong, F.H. 1957. Notes on Sorex preblei in Wash- ington State. The Murrelet 38: 6. Bury, R.B., and P.S. Corn. 1987. Evaluation of pitfall trapping in northwestern forests: trap arrays with drift fences. Journal of Wildlife Management 51: 112-119. Cannings, R.A., R. J. Cannings, and S. G. Cannings. 1987. Birds of the Okanagan Valley, British Columbia. Royal British Columbia Museum, Victoria. 420 pages. Carraway, L.N. 1995. A key to the Recent Soricidae of the western United States and Canada based primarily on dentaries. University of Kansas, Museum of Natural History, Occasional Papers 175: 1-49. Cornely, J. E., L.N. Carraway, and B. J. Verts. 1992. Sorex preblei. Mammalian Species 416: 1-3. Diersing, V. E., and D. F. Hoffmeister. 1977. Revision of the shrews Sorex merriami and a description of a new 8 THE CANADIAN FIELD-NATURALIST species of the subgenus Sorex. Journal of Mammalogy 58: 321-333. Junge, J. A., and R.S. Hoffmann. 1981. An annotated key to the long-tailed shrews (Genus Sorex) of the United States and Canada, with notes on Middle American Sorex. University of Kansas, Museum of Natural History, Occasional Papers 94: 148. Kirkland, G., Jr., R. R. Parmenter, and R. E. Skoog. 1997. A five-species assemblage of shrews from the sagebrush-steppe of Wyoming. Journal of Mammalogy 78: 83-89. Krannitz, P. G. 1997. Seed weight variability of Ante- lope Bitterbrush (Purshia tridentata: Rosaceae). The American Midland Naturalist 138: 306-321. Nagorsen, D. W. 1996. Opossums, shrews and moles of British Columbia. Royal British Columbia Museum Handbook. University of British Columbia Press, Van- couver. 168 pages. Volsiits Ports, M. A., and S. B. George. 1990. Sorex preblei in the northern Great Basin. The Great Basin Naturalist 50: 93-95. van Zyll de Jong, C.G. 1983. Handbook of Canadian mammals. 1. Marsupials and insectivores. National Museums of Canada, National Museum of Natural Sciences, Ottawa. 210 pages. Verts, V. J., and L. N. Carraway. 1998. Land mammals of Oregon. University of California Press, Berkeley, California. 668 pages. Williams, D. F., and S. E. Braun. 1983. Comparison of pitfall and conventional traps for sampling small mam- mal populations. Journal of Wildlife Management 47: 841-845. Received 4 April 2000 Accepted 31 January 2001 Effects of Beaver, Castor canadensis, Herbivory on Streamside Vegetation in a Northern Ontario Watershed D. M. BARNES! and A. U. MALLIK? 'Department of Biology/Faculty of Forestry and the Forest Environment, Lakehead University, Thunder Bay, Ontario P7B 5E1, Canada *Department of Biology, Lakehead University, Thunder Bay, Ontario P7B 5E1, Canada Barnes, D. M., and A. U. Mallik. 2001. Effects of Beaver, Castor canadensis, herbivory on streamside vegetation in a northern Ontario watershed. Canadian Field-Naturalist 115(1): 9-21. Based on the life-form and their utility to Beavers (Castor canadensis), we classified the riparian plants around Beaver impoundments into five categories: Alder (Alnus spp.) — dam construction; Trembling Aspen (Populus tremuloides) — primary food; White Birch (Betula papyrifera) and Willows (Salix spp.) — secondary foods; shrubs — occasionally used for food and dam construction; and conifers — occasionally used for dam construction. To study the influence of Beaver herbivory on shoreline woody plants, we randomly chose eight recently active (<5 years since establishment), seven recently abandoned (< 12 years since abandonment), and eight old abandoned (> 12 years since abandonment) dam sites. We found that Beavers concentrated the majority of their herbivory to within 20 m of the impoundment edge. Three expla- nations are proposed: (1) thermoregulatory restrictions; (2) directionally leaning trees within riparian habitats; and (3) reduction of predation risk. After 12+ years of abandonment, we found the Beaver's preferred food tree, the Trembling Aspen, had not re-established along streams adjacent to abandoned dam sites. In contrast, less desirable food trees and conifers showed increased recovery. If this selective herbivory continues unchecked, the riparian habitat along the Swanson River will become dominated by conifers and thus unsuitable for sustaining Beavers. We believe that fire is need- ed to rejuvenate the failing Trembling Aspen stocks. Key Words: Beaver, boreal forest, Castor canadensis, foraging distance, woody vegetation, predation, Ontario. Johnston and Naiman (1990) suggested that of all the North American herbivores, Beavers (Castor canadensis) have the greatest potential to affect plant communities. They provided three reasons why Beaver herbivory may have such a large impact: (1) Beavers cut down trees and shrubs, thus affecting the canopy and sub-canopy light levels of forests; (2) Beavers restrict their cutting activity to narrow, riparian habitats, thus concentrating the impact along the shorelines; and (3) Beavers harvest trees in excess of their nutritional needs. As central-place foragers, Beavers focus their for- aging activities around a central water-based colony (Basey and Jenkins 1995). In northern Ontario, Beavers rely heavily on riparian plants for food acquisition and dam construction and maintenance. Trembling Aspen (Populus tremuloides) is consid- ered to be their primary food, with Willow (Salix spp.) and White Birch (Betula papyrifera) being of secondary importance. To a lesser degree Beavers will also make use of deciduous shrubs, such as Pin Cherry (Prunus pensylvanica), Mountain Maple (Acer spicatum), Red Osier Dogwood (Cornus stolonifera), Mountain Ash (Sorbus americana), Choke Cherry (Prunus virginiana), Serviceberry (Amelanchier spp.), Beaked Hazel (Corylus cornu- ta), and River Birch (Betula glandulosa). Beavers, also, use riparian plants for dam construction, most notably Alder (Alnus spp.) although deciduous shrubs and conifer species are occasionally used (Barnes and Mallik 1996). Most Beaver research has concentrated on food foraging (Chabreck 1958; Hall 1960; Brenner 1962; Nixon and Ely 1969; Northcott 1971; Jenkins 1980; Belovsky 1984; McGinley and Whitham 1985; Fryxell 1992). Research dealing with the overall effects of Beaver herbivory on plant communities has largely been restricted to a few studies: western Wisconsin (Barnes and Dibble 1988); northern Minnesota (Johnston and Naiman 1990); and eastern Ontario (Donkor and Fryxell 1999, 2000). With the exception of Donkor and Fryxell (1999, 2000), who studied one 3-year-old abandoned dam, these studies largely based their analysis on data collected at active beaver colonies. Despite the many studies that have been devoted to abandoned Beaver dams, no study has quantitatively assessed the recovery of woody plants adjacent to abandoned impoundments over an extended period of time. For the most part this research has focused on two main aspects of abandoned dams: (1) the plant regeneration within the abandoned basin of impound- ments (Morgan 1868; Rudemann and Schoonmaker 1938; Ives 1942; Wilde et al. 1950; Knudsen 1954; Neff 1957; Naiman et al. 1988; Terwillger and Pastor 1999); and (2) the effects on wildlife (Beard 1953; Neff 1957; Rutherford 1964; Hodkinson 1975; Novak 1987; Terwillger and Pastor 1999). 10 THE CANADIAN FIELD-NATURALIST To address these research needs, our present study focused on the Swanson River, an entire watershed area within the Chapleau Crown Game Preserve of northern Ontario. The specific objectives of the study were to: (1) determine the effect of Beaver herbivory on shoreline woody plants at recently active dam sites; (2) to assess the recovery of shore- line woody plants at abandoned Beaver dams; and (3) to determine if prolonged foraging by Beavers had any effect on species richness, diversity, and evenness of streamside woody plant communities. Study Area The study was conducted within the Swanson River drainage basin of the Chapleau Crown Game Preserve (48°05’N, 83°20'W; elevation range 348 — 510 m asl [above sea-level]; area of 700 000 ha) of northern Ontario (Figure 1). The Swanson River has a 200 km network of streams covering an area of 228 km? (Figure 1). Alder dominated riparian habi- tats. The forests were dominated by Jack Pine (Pinus banksiana) and Black Spruce (Picea mariana) inter- spersed with mixed stands of White Spruce (Picea glauca), Balsam Fir (Abies balsamea), White Birch and Trembling Aspen. Associated with these forest trees were numerous understory trees and shrubs such as Willow, Pin Cherry, Mountain Maple, Dogwood, Mountain Ash, Choke Cherry, Black Ash (Fraxinus nigra), Serviceberry, Beaked Hazel, and River Birch. Since its inception in 1925, there has been a ban on all hunting and trapping in the preserve (Anony- mous 1985). Although there have been notable exceptions of this policy (i.e. road-side removals of nuisance beavers and translocation of Beavers to rehabilitate depleted populations throughout Ontario), no Beavers were trapped during the spring and summer of 1992 and 1993 (C. Todesco, Ontario Ministry of Natural Resources, Chapleau ON, per- sonal communication). As a further precaution, no active dam was selected for study if that dam bor- dered on a road. Logging has been an ever-present activity within the Preserve. Between 1910 and 1940, timber- harvesting operations were concentrated around rail- way communities. By the early 1940s most of the easily accessible timber was harvested and transport- ed by river or rail (Anonymous 1985). To access inland timber resources, road based operations were begun and have persisted to present day. Between 1948 and 1958, salvage operations were started on a number of large fires near Racine Lake (45 000 ha) (Anonymous 1985). Since the 1950s, the Swanson River watershed has received very little harvesting activity. From 1975 to 1995, all cutting was done on the north, east, and south peripheries of the Swanson River watershed (B. Riche, Ontario Ministry of Natural Resources at Chapleau, Ontario, personal communication; Figure 1). We observed no cutting Vol. 115 activity around any of the active Beaver dams ana- lyzed. Based on trapping and logging history, we are confident that the Swanson River watershed provid- ed a natural forest setting where Beavers were stud- ied without major human disturbance. Methods The Swanson River system was sub-divided into 200 1-km sections (Barnes and Mallik 1997). A total of 40 stream sections were classed as active (at least one active dammed colony), 85 as abandoned (at least one abandoned dammed colony with no evi- dence of active dams), and 75 as no-dam sites (no dam building activity evident). Beaver impound- ments were located by aerial survey (Cessna 180) in the autumn of 1991 (Bergerud and Miller 1977) and a stereoscopic examination of 1980 (R:F-1:15840) and 1992 (R:F-1:8000) aerial photographs. From these 200 sites, 15 active, 15 abandoned, and 12 no- dam sites were randomly chosen. Since some of the sample sites were remote, the selection process in- volved assessing the site’s accessibility. To ensure that the data could be completed by September, we chose only those sites in which data could be collect- ed in one day (Barnes and Mallik 1997). An active Beaver dam that has been active for an extended period of time can differ significantly from a newly formed site. To address this concern, we used only recently established dams (< 5 years since creation). Eight active sites were classed as "recently active" based on the longevity of flooded Alder. Swank (1949) found that Alder survived only two to four years after being flooded. Using this technique, we chose only those active impoundments in which the flooded Alder were alive. Further, we sub-divided the abandoned dam sites into seven recently aban- doned (< 12 years since abandonment), and eight old abandoned (> 12 years since abandonment) sites (Figure 1). Using aerial photography, time since abandonment was arbitrarily assessed as being recent or old. An impoundment was classed as recently abandoned if there was no evidence of the im- poundment on the 1980 aerial photos and if the impoundment viewed in the 1992 aerial photos showed evidence of being abandoned; i.e., broken dams, presence of Beaver meadows (Wilde et al. 1950), presence of dead trees, and a reduced water level in the impoundment. Old abandoned dam sites were those, which appeared in the 1980 aerial pho- tos as being abandoned. To assess the effect of Beaver herbivory on shoreline woody plant communities, we collected utilization/availability data at the eight recently active impoundment sites. To select which of the two impoundment shorelines to analyze, we placed two pieces of paper representing each shoreline into a cap and randomly chose one. Along the chosen shoreline, a sample plot was established with a 2001 83°20’ Racine Lake BARNES AND MALLIK: EFFECTS OF BEAVER HERBIVORY 11 FiGuRE 1. The Swanson River drainage basin study area with the location of recently active (A), recently abandoned (R), and old abandoned (O) Beaver impoundments sampled. Location of the Chapleau Crown Game Preserve study area in Ontario is shown in the inset. Also included is the network of gravel roads (solid dark lines) and areas logged between 1975 — 1985, Band 1985 — 1995,Z. width and length of 40 m and 170 m, respectively (Barnes and Mallik 1997) (Figure 2). We randomly chose three shoreline locations within the sampling plot. At each of these locations, a 1 m X 40 m belt transect was established perpendicular to the impoundment’s edge. Each transect was sub-divided into 40 - 1m X 1 m sub-plots (Figure 2). We mea- sured all Beaver-cut stumps and uncut stems with diameters > 0.5 cm along the 24 — belt transects of the eight recently active impoundments. Diameter measurements of Beaver stumps were measured at the point cutting was initiated. For standing trees and shrubs, diameters were taken 30 cm above the ground (Johnston and Naiman 1990; Barnes and Mallik 1996). Because of the low number of cut stems encountered in each transect, the total stem 12 Impoundment Plots Stream Flow THE CANADIAN FIELD-NATURALIST Vol. 115 Stream Flow \\ B Is) A Vv 38 R \\ D A M \\ FIGURE 2. Sampling design for studying the impact of Beaver herbivory on shrubs and trees at eight recently active (< 5 years of occupancy), seven recently abandoned (< 12 years since abandonment), and eight old abandoned (> 12 years since abandonment) impoundments and the downstream comparison plots in the Swanson River basin of the Chapleau Crown Game Preserve, 1993. number (combining all the transects) were used to determine Beaver herbivory. No data was collected at recent and old abandoned impoundment sites, because the decayed state of cut stumps made species identification unreliable. To assess the recovery of shoreline woody plants after Beaver herbivory, we established sample plots along the shoreline of the seven recently and eight old abandoned impoundments. For comparison, we chose vegetation plots downstream of these aban- doned impoundments. These plots were established where the stream returned to its original width (Figure 2). A straight-line measurement was taken from mid-dam to the impoundment shoreline. Using this distance, we were able to offset the downstream plot location away from the water’s edge, thus ensur- ing the sampling of a similar forest sector as that of the impoundment plot (Figure 2). Employing the same methodology used for recently active impound- ments, we recorded shrub and tree diameters and later calculated their basal areas and density. To evaluate differences among woody plant basal areas and density associated with recently abandoned and old abandoned impoundments and their respective downstream comparison plots, we performed a split- split-plot analysis of variance followed by Least Significant Differences Tests (Velleman 1992). At each of the recently abandoned and old aban- doned impoundments and their respective down- stream control sites, we measured the changes in the woody plant community using three ecological indices; species richness, diversity, and evenness. Species richness of shoreline woody vegetation was determined by a direct count of the number of species within the sampling plots (Ludwig and Reynolds 1988). Although most ecologists have used the Shannon’s Index, i.e. - [2(n/n)/Ln(n/n)] where n, = ith of S species in the sample and n = total number of individuals in the sample, to determine species diversity, this estimator is biased, as the total number of species found within a community will most likely be greater than the num- ber of species observed in any sample (Ludwig and Reynolds 1988). To overcome this bias, Hill’s Diversity Number 1: Ni= e ‘Shannon's dex Ell 197.33 DeJong 1975) was used to measure species diversity. There have been several indices developed to measure species evenness. In each case, if all species in a sample are equally abundant, then the evenness index should have a maximum value. Conversely, lower values mean a divergence away from even- ness, i.e. dominance. Based on Ludwig and Reynolds (1988), we used Evenness Index 5: E5 = N2-1/N1-1. where N1 = Hill’s Diversity Number 1 and N2 = Hill’s Diversity Number 2, N2 = 1/Simpson’s Index, where Simpson’s Index = X(n,(n,-1)/ n(n-1), n, = ith of S species in the sample and n = total number of individuals in the sample. E5 values approach zero when species become dom- inant and one when evenness is a maximum (Alatalo 1981). Once again, to evaluate differences among ecolog- ical indices associated with recently abandoned and old abandoned impoundments and their respective downstream vegetation plots, we performed a split- 2001 BARNES AND MALLIK: EFFECTS OF BEAVER HERBIVORY % cut / uncut PO B/W CN 0 - 10.1 m from water edge % cut / uncut PO B/W CN 10.1 - 20 m from water edge = = S) = = — bd = oe 3s PO B/W CN 20.1 - 30 m from water edge % cut / uncut PO B/W CN 30.1 - 40 m from water edge DISTANCE FROM THE IMPOUNDMENT EDGE Figure 3. Proportion of Beaver cut Hf and uncut L] woody plants at 10 m intervals within 40 m of the impoundment edge of eight recently active dams (< 5 years of occupan- cy) in the Chapleau Crown Game Preserve, 1993. 14 THE CANADIAN FIELD-NATURALIST T. ASPEN 16 10.1 - 20 m 10.1 -20m 30.1 - 40 m DISTANCE THE BIRCH/WILLOW 30.1 - 40 m IMPOUNDMENT Volniis CONIFERS SHRUBS 10.1 -20m =) a2 & 30.1 - 40m EDGE FiGuRE 4. The basal area (m7/ha) fl and SE L] of woody plants at 10 m intervals within 40 m of the impounded edge of seven recently abandoned (RAb, < 12 years since abandonment), and eight old abandoned (OAb, > 12 years since abandonment) impoundments and the downstream comparison plots in the Swanson River basin of the Chapleau Crown Game Preserve, 1993. split-plot analysis of variance followed by Least Significant Differences Tests (Velleman 1992). We determined the power of our 2 X 2 X 4 split- split-plot factorial ANOVA by the using a Table of Minimal Detectable Differences A, for a fixed factor or interaction in a general design, a = 0.05, b = 0.10 (Appendix 12 — Lorenzen and Anderson 1993). If A is less than 0.5, then the experiment is capable of detecting extremely small differences among the fac- tor levels. If D is between 0.5 and 1.5, then the experiment is capable of detecting small differences among the factors. Results Herbivory at Recently Active Beaver Impoundments Beavers concentrated their cutting activity within 20 m of the shoreline of the recently active impound- ments, 12% of available stems (41 cut, 390 avail- 2001 able) within 10 m of the impoundment edge and 9% between 10.1 — 20 m (Figure 3). At distances 20.1 — 30 m and 30.1 — 40 m, Beavers only harvested 0.4% (4 cut, 493 available) and 1% (3 cut, 247 available) of available stems (Figure 3). In addition, we found 49 % of all cut stems within 10 m of shore and 91% within 20.1 m (Figure 3). More specifically, Beavers concentrated their for- aging for food items in close proximity to water. Within 10 m of the water’s edge 71% (5 cut, 7 avail- able) of Trembling Aspen stems were harvested and 54% (7 cut, 13 available) at distances between 10.1 and 20 m (Figure 3). Only 7% (3 cut, 41 available) of Trembling Aspen plants were cut at a distance greater than 20 m (Figure 3). The forage of sec- ondary food trees showed a similar pattern; at dis- tances 0 — 10 m and 10.1 — 20 m, Beavers cut 36% (5 cut, 14 available) and 14% (4 cut, 28 available) of White Birch and Willow stems, respectively (Figure 3), and at distances 20 — 40 m, only 3% (1 cut, 34 available). All Alder stems cut were within 20 m of impound- ment edges, despite a similar number of available stems at 0 — 20 m (80) and 20.1 — 40 m (83) (Figure 3). The Beaver’s cutting pattern did not vary greatly throughout the first 20 m, i.e. 24% (11 cut, 46 avail- able) at distances between 0 — 10 m and 21% (7 cut, 34 available) between 10.1 — 20 m. BARNES AND MALLIK: EFFECTS OF BEAVER HERBIVORY 15 Beavers only foraged for other hardwood shrub species within 20 m of water; 4% (1 cut, 23 avail- able) between 0 — 10 m and 2% (1 cut, 6lavailable) between 10.1 — 20 m. This result occurred despite shrub densities being 1.6 times greater at distances greater than 20 m from water. No conifers were cut even though they were readily available throughout the plot width (164 available between 0 — 20 m and 197 between 20.1 — 40 m) (Figure 3). Woody Plant Recovery at Abandoned Impoundments Power Analysis. Using the technique developed by Lorenzen and Anderson (1993), we found that the power of our experimental design was appropriate. All three factors involved in the experiment had Minimal Detectable Difference A values of less than 0.5 and therefore capable of detecting extremely small differences (i.e. Condition — 0.47; Type — 0.47; Distance — 0.43). The experiment’s factor interac- tions had A values of between 0.5 and 1.5, indicating the ability to detect small differences (Condition * Type — 0.67; Condition * Distance — 0.61; Type * Distance — 0.61; Condition * Type * Distance — 0.87). Basal Areas. At both recently and old abandoned impoundments, we noted that the basal area of Alder and White Birch/Willow stems were not significant- ly different from basal areas found in the down- TABLE |. The significant results of a Split-Split-Plot Analysis of Variance of woody plant basal area (m2/ha) found within 40 m of the shoreline of recently abandoned (RAb, < 12 years), and old abandoned (OAb, > 12 years) impoundments (Impd) and their respective downstream (DStr) plots in the Chapleau Crown Game Preserve. Woody Least Square Plant Main Interaction Differences Categories! Effects P Effects (LSD) P Alder Dist? 0.03 0-10 vs 20.1-30 m 0.06 0-10 vs 30.140 m 0.01 10.1—20 vs 30.140 m 0.02 Trembling Cond} * Type* *Dist 0.008 RAb/Impd/ 0-10 m vs RAb/DStr/0-10m_ 0.01 Aspen White Birch Cond * Dist 0.003 RAb/ 0-10 m vs RAb/20.1—30 m 0.002 and RAb/10.1—20 m vs RAb/20.1-30 m 0.005 Willow RAb/20.1—30 m vs OAb/20.1-—30 m 0.0004 OAb/10.1—20 m vs OAb/20.1—30 m 0.04 Conifer No significance noted amongst the Factors and their Interactions Shrubs Type 0.03 Type * Dist 0.05 Impd/ 0-10mvsDStr/ 0-10m 0.01 Impd/10.1—20 m vs DStr/ 10.1-20 m 0.002 Impd/30.1—40 m vs DStr/ 30.140 m 0.01 Impd/ 0-10 m vs Impd/20.1—30 m 0.0008 Impd/20.1- 30 m vs Impd/30.1—40 m 0.03 'Transformations with normal probability plots (Velleman 1992). *Dist — Distance indicates where the sample area is located in relation to the impoundment's shoreline. 3Cond — Condition indicates whether the sample area is associated with a recently abandoned or old abandoned impound- ment. 4Type indicates whether the sample area is associated with an impoundment or a downstream site. 16 THE CANADIAN FIELD-NATURALIST TREMBLING ASPEN 16000 1000 2800 12000 750 2100 8000 500 1400 4000 250 700 0.1-20m DISTANCE FROM THE WHITE BIRCH / WILLOW IMPOUNDMENT Vol. 115 CONIFERS SHRUBS 10000 7500 5000 2500 EDGE FIGURE 5. The density (number of stems/ha) Hf and SE | of woody plants at 10 m intervals within 40 m of the impounded edge of seven recently abandoned (RAD, < 12 years since abandonment), and eight old abandoned (OAb, > 12 years since abandonment) impoundments and the downstream comparison plots in the Swanson River basin of the Chapleau Crown Game Preserve, 1993. stream comparison sites (Figure 4 and Table 1). Alder and White Birch/Willow woody plants showed a general recovery to original levels. Within 10 m of the recently abandoned impound- ment shoreline, we found Trembling Aspen basal area was significantly less than their downstream comparison plot, indicating that these plants did not recover from Beaver herbivory. Between 10.1 and 40 m from shore, there were no significant differ- ences. At old abandoned dam sites, we could not find evidence of Trembling Aspen within 30 m of shore (Figure 4 and Table 1). At recently abandoned impoundments, we found that there was no significant difference in the basal area of conifers when compared to downstream com- parison sites. At old impoundments, although conifer basal areas did not show any significant differences, on average there was a 60% increase in their basal area within 10 m of shore, 44% increase between 20.1 - 30 m, and 25% between 30.1 - 40 m (Figure 4 and Table 1). -At abandoned sites (recently or old), the basal areas of shrubs were significantly reduced at 0 — 10 m, 10.1 — 20 m, and 30.1 — 40 m when compared to their corresponding downstream sites. Shrubs, then, did not recover to original basal area values. Density. At both recently and old abandoned impoundments, Alder and White Birch/Willow stems were not significantly different from the downstream comparsion sites, indicating that Alder and White Birch/Willow plants recovered to their original densities (Figure 5 and Table 2). At recently abandoned dam sites, we recorded a significant decrease in Trembling Aspen density within 20 m of shore. In addition, when we com- pared the density within 10 m of shore with those at 30.1 — 40 m, we noted a significant increase in Trembling Aspen stems (Figure 5 and Table 2). At old abandoned sites, we did not detect the presence of any Trembling Aspen. Even at distances 30.1 - 40 m from shore, we discovered that the density was significantly reduced (Figure 5 and Table 2). There was no significant difference found in conifer densities at recently and old abandoned dam sites (Figure 5 and Table 2). Generally, at both the recently and old abandoned dams, the density of 2001 shrubs was found to be significantly less than origi- nal densities recorded in the downstream plots (Figure 5 and Table 2). Ecological Indices. We found that there was no significant change in species richness, diversity, and evenness within 40 m of either recently and old abandoned impoundments when compared to their downstream sites (Figure 6). Discussion Herbivory at Recently Active Beaver Impoundments Within the Swanson River watershed of the Chapleau Crown Game Preserve, we found that most of the foraging by Beavers was within 20 m of the impoundment edge. This narrowing of the Beaver’s foraging range is interesting, considering they have the ability to haul stems 200 m overland (Novak 1987). We present three possible explanations for this phenomenon. Our first explanation is based on thermoregulation in Beavers. Belovsky (1984) working on Isle Royale showed that Beaver summer activity was limited to BARNES AND MALLIK: EFFECTS OF BEAVER HERBIVORY Ly only 4.5 hours, the peak being in the evening/night and secondary peak in early afternoon. Part of the explanation for this reduced feeding time was the thermal stress. The Beaver is a very clumsy and slow moving animal and as a result expends a lot of ener- gy moving cut stems overland. Belovsky (1984) and Fryxell (personal communication) found no cut stems larger than 2.5 and 4.2 cm diameter, respec- tively, could be hauled intact by Beavers. Therefore, acquisition of a forage tree requires that the Beaver cuts down the tree, cuts it up into stems of manage- able size, and then hauls it to the impoundment edge. Steen (1965), Miller (1967), and Coles (1969, 1970) concluded that the tail of the Beaver had an impor- tant function as an organ for controlled heat dissipa- tion. Thus, Beavers may have chosen to forage close to water to facilitate tail wetting and control against hyperthermic increase in body temperature. A second reason why Beavers may be cutting for- age trees, (i.e. Trembling Aspen, White Birch and Willow) in close proximity to water may be due to how these trees grow in a riparian setting. Because TABLE 2. The significant results of a Split-Split-Plot Analysis of Variance of woody plant density (number of stems/ha) found within 40 m of the shoreline of recently abandoned (RAb, < 12 years), and old abandoned (OAb, > 12 years) impoundments (Impd) and their respective downstream (DStr) plots in the Chapleau Crown Game Preserve. Woody Least Square Plant Main Interaction Differences Categories’ Effects P Effects P (LSD) P Alder Dist 0.02 0-10 vs 30.140 m 0.003 10.1—20 vs 30.140 m 0.01 Trembling Cond * Type’ *Dist 0.05 RAb/Impd/ 0-10 m vs RAb/DStr / 0-10 m 0.0004 Aspen RAb/Impd/10.1—20 m vs RAb/DStr /10.1—20 m 0.07 RAb/Impd/30.1—40 m vs RAb/DStr /30.1—40 m 0.01 RAb/Impd/ 0-10 m vs RAb/Impd/30.1—40 m 0.003 White Birch Dist 0.02 0-10 vs 20.1-30 m 0.005 and 10.1—20 vs 20.1-30 m 0.03 Willow 0-10 vs 30.140 m 0.04 Cond * Dist 0.02 RAb/ 0-10 m vs RAb/20.1—30 m 0.0009 RAb/10.1—20 m vs RAb/20.1—30 m 0.002 RAb/ 0— 10 m vs RAb/30.1—40 m 0.006 Conifer No significance noted amongst the Factors and their Interactions Shrubs Type 0.02 ‘Transformations with normal probability plots (Velleman 1992). ‘Dist — Distance indicates where the sample area is located in relation to the impoundment's shoreline. ‘Cond — Condition indicates whether the sample area is associated with a recently abandoned or old abandoned impound- ment. "Type indicates whether the sample area is associated with an impoundment or a downstream site. 18 THE CANADIAN FIELD-NATURALIST SPECIES RICHNESS 1.25 DIVERSITY Vol. 115 EVENNESS ScoN BO oe) SS So in nn mn S S BS AX ee et ks RAb CRAb A RAb 2 2 2 3 2 ete e) (o) [a4 io4 o) oe) ie) O O -10m 0-10m aes ee Sp eee aL Wet cS nee a4 o io) e) ie) e) a4 os oO O O O O - s) 10.1 - 20m 10.1 - 20m 10.1-20m (—) N & “A loo} = ee) me) Ss ao.wNM nN nN n Sas aie i) ax ~I (—} mn nN nN — 2 2 2 2 2 2 < t < < < < a4 ~ oe) oe) a4 a4 O O O 20.1-30m 2 2 po} 2 2 Q < < BS < < < o) ie) [e4 o) (o) ©) O O 30m 20.1 - 30m o,N BO = vd yy YY — S S Ww oS SS onN inn = 2 2 2 2 2 2 2 2 2 = < < < < < < < < < < < ia lad io) fo) fa fa fo) lo) [aa a4 lo) eo) O oO O O 1S) O 30.1 - 40 m 30.1 - 40 m 30.1 - 40m DISTANCE FROM THE IMPOUNDMENT EDGE FIGURE 6. Species richness, diversity, and evenness index values Hf and SE [|] for the woody plants at 10 m intervals within 40 m of the impounded edge of recently abandoned (RAb, < 12 years since abandonment), and eight old abandoned OAb, > 12 years since abandonment) impoundments and the downstream comparison plots in the Swanson River basin of the Chapleau Crown Game Preserve, 1993. of reduced canopy and more sunlight, riparian trees tend to lean towards the water or have more foliage in that direction (Johnson 1983). Since Beavers can- not directionally fall trees (Novak 1976), we propose that Beavers over evolutionary time may have select- ed to cut trees close to the water as they are more apt to fall towards or into the water because of this dis- proportionate growth on their stream side. The third reason involves the risk of predation. The trade-off between foraging activity and preda- tor risk has long been regarded as one of the great ecological conflicts experienced by animals (Milinski and Heller 1978; Abrams 1984; Werner and Gillian 1984; McNamara and Houston 1987; Lima and Dill 1990; Anholt and Werner 1995; Werner and Anholt 1996; Eklov and Halvarsson 2000). The importance of predation during evolu- tionary time is clear, but growing evidence suggests that animals also have the ability to assess and behaviourally influence their risk of being preyed upon in ecological time (i.e., during their lifetime) (Lima and Dill 1990). 2001 Beavers are classed as central-place foragers and as such move out from a central location to select food that may be eaten immediately or transported back to the central location for later use (Basey and Jenkins 1995). Researchers have long suggested that the Beaver’s foraging behaviour has been affected by the risk of predation (Jenkins 1980; Belovsky 1984). Recently, Basey and Jenkins (1995) showed experimentally that Beavers were “trading off maxi- mization of profitability against minimization of pre- dation” by Coyotes (Canis latrans). Since Timber Wolves (Canis lupus) are very effective predators of Beavers (Pimlott et al. 1969; Frenzel 1974; Voigt et al. 1976; Theberge et al. 1978; Potvin et al. 1992) and commonly prey on Beavers within our study area (Carbyn 1987; MacDonell 1993; Elliott 1997), we suggest that these northern Beavers may be adjusting their foraging patterns to minimize capture by large predators such as Timber Wolves. To date, there have been no field studies that demonstrate experimentally that Beavers reduce their foraging range in response to predator pressure. However, we believe some evidence of this predator- induced reduction trend can be seen when we com- pare Beaver foraging studies done in areas where Timber Wolf populations are high with those with- out these large predators. On Isle Royale, in northern Minnesota, and southern Algonquin Provincial Park, areas known to have high Wolf population levels, studies by Belovsky (1984), Johnston and Naiman (1990) and Donkor and Fryxell (1999, 2000), have documented Beaver feeding activity is largely con- centrated within 20 m of water. In contrast to this, Beaver studies done in Maine (Hodgdon and Hunt 1953) and Michigan (Bradt 1947), areas where no Timber Wolves reside, show that the foraging range is greatly expanded. Alder plays a crucial role in minimizing predation by Timber Wolves. Across the northern landscape, the ubiquity of streamside Alder provides Beavers with a ready source of preferred dam-building mate- rial in close proximity of the water’s edge (Barnes and Mallik 1996). Since Beavers are able to trans- port whole Alder stems overland with no appreciable energy cost (Barnes and Mallik 1996), they are able to construct dams with high efficiency. Beavers within our study area did not have to venture further than 20 m to obtain Alder stems (Figure 3). This sit- uation ensures the speedy establishment and mainte- nance of a dam, thus providing a protective water environment against predation. Woody Plant Recovery at Abandoned Impoundments Donkor and Fryxell (1999, 2000) found that Beavers had a great effect on species richness and diversity three years after a dam site was abandoned. Our findings are contradictory. We found that there were no significant changes in these two ecological indices over an extended period of time (12+ years). BARNES AND MALLIK: EFFECTS OF BEAVER HERBIVORY 19 A possible explanation for this discrepancy could be found in their choice of sample size. In their study, Donkor and Fryxell (1999, 2000) used one aban- doned site. We found that there was much variation from one site to another and found it necessary to use 15 sites to account for these fluctuations in measurements. Over time, we found that the Trembling Aspen did not show any signs of recovery. Indeed, their presence after 12 years was not detected by our sampling tech- niques. Since we had no details concerning the colo- nization history of these dam sites, Beavers may have occupied these sites several times in the intervening years and their harvesting pressure may be spread over the time period, not giving Trembling Aspen enough time to recover. Similarly, Gese and Shadle (1943) and Graham et al. (1963) found that regenera- tion of Trembling Aspen was greatly hampered by the continual cutting by herbivores, such as Beavers and White-tailed Deer (Odocoileus virginianus). Johnston and Naiman (1990) showed that Beaver herbivory is especially significant in the boreal for- est, because it affects the overstory causing increases in light, temperature, and nutrients. In our study, we found that after 12 years of abandonment, there was an increase in the basal area of conifers adjacent to abandoned Beaver impoundments. Lawrence (1952) working in Michigan found that after 10 years of abandonment, there was a substantial increase in the growth of shoreline conifers. Based on their work in northern Minnesota, Naiman et al. (1988) and Johnston and Naiman (1990) noted that selective cutting by Beavers could cause a shift towards conifer-dominated forests. Bryant and Chapin (1986) and Pastor et al. (1988) showed that selective brows- ing by other forest herbivores had a similar result. Slough and Sadleir (1977) believed that Trembling Aspen was an important factor in main- taining healthy Beaver populations in northern Canada. Since the Trembling Aspen is an early seral stage plant, fire is needed to open up new coloniza- tion areas (Graham et al. 1963). Because of this Slough and Sadleir (1977) proposed using fire as a silvicultural tool to increase high quality Beaver for- age. At that time, the use of fire was not widely prac- ticed (Slough and Sadleir 1977). Fire, however, is now considered one of the most significant distur- bances in northern forests. In recognition of its importance, the Ontario Ministry of Natural Resources has recently produced draft forest man- agement guidelines for retaining forest ecosystem structure and function using fire disturbance patterns as a template (Anonymous 2000). We believe that as more of these novel, more ecologically sensitive, forest management strategies are implemented across the forest landscape, dis- turbance-reliant species such as the Beaver will flourish. 20 THE CANADIAN FIELD-NATURALIST Acknowledgments We acknowledge the financial assistance of Ducks Unlimited, the Ontario Ministry of Universities and Colleges, and Environmental Youth Corps. We are indebted to C. Todesco of the Ontario Ministry of Natural Resources for logistic support throughout the project and K. Brown for his statistical expertise. We are grateful to F. DeGagne, V. Piercey, M. Sinclair, and M. Osika for technical assistance in the collec- tion and processing field data. In addition, we thank B. Barnes for her proof-reading skills and C. Chapin for producing the study area map. Literature Cited Anonymous. 1985. The Chapleau Crown Game Preserve. Ontario Ministry of Natural Resources Leaflet. Anonymous. 2000. Forest Management Guidelines for Retaining Forest Ecosystem Structure and Function: The Fire Simulation Guidelines (Draft). Ontario Ministry of Natural Resources. Abrams, P. A. 1984. Foraging time optimization and interactions in food webs. American Naturalist 124: 80-96. Alatalo, R. V. 1981. Problems in the measurement of evenness in ecology. Oikos 37: 199-204. Anholt, B. R., and E. E. Werner. 1995. Interaction between food availability and predation mortality medi- ated by adaptive behaviour. Ecology 76: 2230-2234. Barnes, D. M., and A. U. Mallik. 1996. Use of woody plants in construction of beaver dams in northern Ontario. Canadian Journal of Zoology 74: 1781-1786. Barnes, D. M., and A. U. Mallik. 1997. Habitat factors influencing Beaver dam establishment in a northern Ontario watershed. Journal of Wildlife Management 61: 1371-1377. Barnes, W. J., and F. Dibble. 1988. The effects of beaver in riverbank forest succession. Canadian Journal of Botany 66: 40-44. Basey, J. M., and S. H. Jenkins. 1995. Influences of pre- dation risk and energy maximization on food selection by beavers (Castor canadensis). Canadian Journal of Zoology 73: 2197-2208. Beard, E. R. 1953. The importance of beaver in water- fowl management at the Seney National Wildlife Refuge. Journal of Wildlife Management 17: 398-436. Belovsky, G. E. 1984. Summer diet optimization by beavers. American Midland Naturalist 111: 209-222. Bergerud, A. T., and D. R. Miller. 1977. Population dynamics of Newfoundland beaver. Canadian Journal of Zoology 55: 1480-1492. Bradt, G. W. 1947. Michigan beaver management. Michigan Department of Conservation, Game Division. 56 pages. Brenner, F. J. 1962. Foods consumed by beavers in Crayford County, Pennsylvania. Journal of Wildlife Management 26: 104-107. Bryant, J. P., and F. S. Chapin III. 1986. Browsing- woody plant interactions during boreal forest plant suc- cession. Pages 213-225 in Forest ecosystems in the Alaskan Taiga: a synthesis of structure and function. Edited by K. Van Cleave, F. C. Chapin III, P. W. ’ Flanagan, L. A. Viereck, and C. T. Dyrness. Springer- Verlag, New York. Vol. 115 Carbyn, L. N. 1987. Gray wolf and red woif. Pages 359-376 in Wildlife Furbearer Management and Conservation in North America. Edited by M. Novak, J. A. Baker, M. E. Obbard, and B. Mallock. Ashton-Potter Limited, Concord, Ontario. 1150 pages. Chabreck, R. H. 1958. Beaver-forest relationship in St. Tammany Parish, Louisiana. Journal of Wildlife Management 22: 179-183. Coles, R. W. 1969. Thermoregulatory function of the Beaver tail. American Zoologist 9: 1092 (Abstract). Coles, R. W. 1970. The beaver tail: anatomical adapta- tions for thermal function. American Zoologist 10: 556-557 (Abstract). DeJong, T. M. 1975. A comparison of three diversity indices based on their components of richness and even- ness. Oikos 26: 222-227. Donkor, N. T., and J. M. Fryxell. 1999. Impact of beaver foraging on structure of lowland boreal forests of Algonquin Provincial Park, Ontario. Forest Ecology and Management 118: 83-92. Donkor, N. T., and J. M. Fryxell. 2000. Lowland boreal forests characterization in Algonquin Provincial Park relative to beaver (Castor canadensis) foraging and edaphic factors. Plant Ecology 148: 1-12. Eklov, P., and C. Halvarsson. 2000. The trade-off between foraging activity and predation risk for Rana temporaria in different food environments. Canadian Journal of Zoology 78: 734-739. Elliott, B. J. 1997. The non-winter food habits of wolves in the Chapleau Crown Game Preserve. HBScF. Undergraduate thesis, Lakehead University. 66 pages. Frenzel, L. D. 1974. Occurrence of moose in food of wolves as revealed by scat analysis: a review of North American studies. Le Naturalist canadien 101: 467-479. Fryxell, J. 1992. Space use by beavers in relation to resource abundance. Oikos 64: 474-478. Gese, E. C., and A. R. Shadle. 1943. Reforestation of aspen after complete cutting by beavers. Journal of Wildlife Management 7: 223-228. Graham, S. A.,_R. P. Harrison, Jr., and C. E. Westell, Jr. 1963. Aspens-phoenix trees of the Great Lakes Region. University of Michigan Press, Ann Arbor. Hall, J. G. 1960. Willow and aspen in the ecology of beaver on Sagenhen Creek, California. Ecology 41: 484494. Hill, M. O. 1973. Diversity and evenness: a unifying notation and its consequences. Ecology 61: 225-236. Hodgdon, K. W., and J. H. Hunt. 1953. Beaver manage- ment in Maine. Maine Department of Inland Fisheries and Game, Game Division Bulletin Number 3. 102 pages. Hodkinson, I. D. 1975. A community analysis of the ben- thic insect fauna of an abandoned beaver pond. Journal of Animal Ecology 44: 533-552. Ives, R. L. 1942. The beaver-meadow complex. Journal of Geomorphology 5: 191-203. Jenkins, S. H. 1980. A size-distance relation in food selection by beavers. Ecology 61: 740-746. Johnson, R. B. 1983. Aspen utilization by beavers (Castor canadensis) in northern Wisconsin. Wisconsin Academy of Sciences, Arts and Letters 71 Part 2: 82-86. Johnston, C. A., and R. J. Naiman. 1990. Browse selec- tion by beaver: effects on riparian forest composition. Journal of Forest Research 20: 1036-1043. 2001 Knudsen, G. J. 1954. Beaver damage to forests and forest reproduction. Wisconsin Wildlife Research 13: 121-129. Lawrence, W. H. 1952. Evidence of the age of beaver ponds. Journal of Wildlife Management 16: 69-79. Lima, S. L., and L. M. Dill. 1990. Behaviour decisions made under the risk of predation: a review and perspec- tus. Canadian Journal of Zoology 68: 619-640. Lorenzen, T. J., and V. L. Anderson. 1993. Design of experiments: A no-name approach. Marcel Dekker, Inc. New York. 414 pages. Ludwig, J. A., and J. F. Reynolds. 1988. Statistical Ecology — A primer on methods and computing. John Wiley & Sons, New York. 337 pages. MacDonell, M. R. 1993. A summer food habit study of timber wolves (Canis lupus L.) in the boreal forest of northeastern Ontario. HBScF. Undergraduate thesis. Lakehead University. 59 pages. McGinley, M. A., and T. G. Whitham. 1985. Central place forging by beavers (Castor canadensis): a test of foraging predications and the impact of selective feeding on the growth of cottonwoods (Populus fremontii). Oecologia 66: 558-562. McNamara, J. M., and A. I. Houston. 1987. Starvation and predation as factors limiting population size. Ecology 68: 1515-1519. Milinski, M., and R. Heller. 1978. Influence of predator on the optimal foraging behaviour of sticklebacks (Gasterosterus aculeatus L.). Nature (London) 275: 642-644. Miller, K. L. 1967. Caudal nerve function as related to temperature in some Alaska mammals. Comparative Biochemistry and Physiology. 21: 679-686. Morgan, L. H. 1868. The American beaver and his works. J. B. Lippincott. Philadelphia, Pennsylvania. Naiman, R. J., C. A. Johnston, and J. C. Kelley. 1988. Alterations of North American streams by beavers. BioScience 38: 753-762. Neff, D. J. 1957. Ecological effects of beaver habitat abandonment in the Colorado Rockies. Journal of Wildlife Management 21: 80-84. Nixon, C. M., and J. Ely. 1969. Foods eaten by a beaver colony in southeast Ohio. Ohio Journal of Science 69: 313-319. Northcott, T. H. 1971. Feeding habits of beaver in Newfoundland. Oikos 22: 407-410. Novak, M. 1976. The beaver in Ontario. Ontario Ministry of Natural Resources. 20 pages. Novak, M. 1987. Beaver, Pages 282-313 in Wildlife Furbearer Management and Conservation in North America. Edited by M. Novak, J. A. Baker, M. E. Obbard, and B. Malloch. Ashton-Potter Limited, Concord, Ontario. 1150 pages. Pastor, J., R. J. Naiman, B. Dewey, and P. McInnes. 1988. Moose, microbes, and the boreal forest. Bio- Science 38: 770-777. BARNES AND MALLIK: EFFECTS OF BEAVER HERBIVORY 21 Pimlott, D. H., J. A. Shannon, and G. B. Kolenosky. 1969. The ecology of the timber wolf in Algonquin Provincial Park. Ontario Department of Lands and Forests, Research Wildlife Report 87. Potvin, F., L. Breton, and M. MacQuart. 1992. Impact of an experimental wolf reduction on beaver in Papineau- LaBelle Reserve, Quebec. Canadian Journal of Zoology 70: 180-183. Rudemann, R., and W. J. Schoonmaker. 1938. Beaver dams as geologic agents. Science 88: 523-525. Rutherford, W. H. 1964. The Beaver in Colorado: Its biology, ecology, management, and economics. Colorado Game, Fish, and Parks Department, Game Research Division, Technical Publication Number 17. 49 pages. Slough, B. G., and R. M. F. S. Sadleir. 1977. A land capability classification for beaver (Castor canadensis Kuhl). Canadian Journal of Zoology 55: 1324-1335. Steen, I. 1965. Thermoregulatory importance of the beaver’s tail. Comparative Biochemistry and Physiology 15: 267-270. Swank, W. G. 1949. Beaver ecology and management in West Virginia. Conservation Commission of West Virginia, Division of Game Management Bulletin Number |. 65 pages. Terwilliger, J., and J. Pastor. 1999. Small mammals, ectomycorrhizae, and conifer succession in beaver meadows. Oikos 85: 83-94. Theberge, J. B., S. M. Oosenbury, and D. H. Pimlott. 1978. Site and seasonal variations in food of wolves, Algonquin Park, Ontario. Canadian Field-Naturalist 92: 91-94. Velleman, P. 1992. Data desk4 handbook. Data Descrip- tion Inc. 613 pages. Voigt, D. R., G. B. Kolenosky, and D. H. Pimlott. 1976. Changes in summer foods of wolves in central Ontario. Journal of Wildlife Management 40: 663-668. Werner, E. E., and J. F. Gillian. 1984. The ontogenetic niche and species interactions in size-structured popula- tions. Annual Reviews of Ecology and Systematics 15: 393-425. Werner, E. E., and B. R. Anholt. 1996. Predator-induced behavioral indirect effects: consequences to competitive interactions in anuran larvae. Ecology 77: 157-169. Wilde, S. E., C. T. Youngberg, and J. H. Hovind. 1950. Changes in composition of ground water, soil fertility, and forest growth produced by the construction and removal of beaver dams. Journal of Wildlife Manage- ment 14: 123-128. Received 17 May 1997 Accepted 19 December 2000 Comparison of Parental Roles in Male and Female Red Foxes, Vulpes vulpes, in Southern Ontario VALERIA VERGARA Trent University, Biology Department, Peterborough, Ontario K9J 7B8, Canada Present address: Department of Renewable Resources, University of Alberta, Edmonton, Alberta, ToG 2E9 Canada Vergara, Valeria. 2000. Comparison of parental roles in male and female Red Foxes, Vulpes vulpes, in southern Ontario. Canadian Field-Naturalist 115(1): 22-33. The Red Fox (Vulpes vulpes) is normally monogamous but often deviates to polygyny. Variability in paternal behaviour may be just one part of its general social system flexibility. I evaluated the extent of male participation in parental care in eight Red Fox families in one suburban and seven rural habitats of southern Ontario, by examining the relative contribution of males and females in rearing the young. I observed a wide range of strategies and considerable variation amongst males’ involvement in raising the kits. The observed variation occurred within a relatively small geographical area and, except for an urban den, the den sites were in similar rural habitats. Despite this variability, several behavioral patterns emerged. Females visited the den more frequently and for longer periods of time than males, they spent a greater proportion of time in behaviors that indicate parental attentiveness (excluding nursing), and they stayed closer to the young. Males spent sig- nificantly more time than females in vigilant behaviour. Key Words: Red Fox, Vulpes vulpes, parental care, paternal care, monogamy, helpers, Ontario. The most common mating system among canids is “obligate monogamy” (Kleiman 1977). Corres- pondingly, some form of male care is widespread in canids (Malcom 1985). Male care has been observed in over half (21/36) of the species in the family Canidae (Malcom 1985; Asa and Valdespino 1998). Data on the rest of the canid species are not available. Several studies of parental behaviour in canids have documented the distribution of care-giving behaviour between the sexes (Wolves, Canis lupus: Harrington and Mech 1982, Fentress and Ryon 1982; Jackals, Canis mesomelas and canis aureus: Moehlman 1983; Red Foxes: Macdonald 1979; African Wild Dogs, Lycaon pictus: Malcom and Marten 1982; Maned Wolves, Chrysocyon brachyu- rus: Rasmussen and Tilson 1984; for a review see Malcom 1985). An examination of such studies reveals a substantial amount of interspecific varia- tion within the family in terms of the contribution of the male to raising the young. In some species the male feeds the young as much as the female (e.g., Arctic Foxes, Alopex lagopus: Garrott 1980; jackals: Moehlman 1986) and in others there is no indication that the male provides food either to the female or to the cubs (e.g., Blanford’s fox, Vulpes cana: Geffen and MacDonald 1992; Maned Wolf: Veado 1998; see Asa and Valdespino 1998, for review of canid reproductive biology). In order to identify the subtle reasons for changes in the reproductive strategies of canid species, it is important to quantify parental care. Any such attempts should compare actual amounts of paternal and maternal care, and investigate how male care varies among years and between different popula- 22 tions or individuals of a particular species. The Red Fox (Vulpes vulpes) is remarkably well suited for this type of study. It has the largest geographical range of any living carnivore (Henry 1993), and its social system is exceptionally flexible. Pronounced differences in the behaviour of Red Foxes between urban, rural, suburban, agricultural, and woodland habitats have been reported (e.g., Macdonald 1980; Doncaster and Macdonald 1991; Voigt and Macdonald 1984; Voigt 1987; see Cavallini 1997 for review). While monogamous pairs seem to be the norm in some habitats (Storm et al. 1976; Voigt 1987) other habitats sustain coopera- tive groups where polygyny (Macdonald 1979) and polyandry (Baker et al. 1998) are common. Even in the same area in different years the same fox popula- tion may exhibit alternate social strategies in response to variations in food supply (von Schantz 1984; Henry 1986; Zabel and Taggart 1989). Female foxes may tolerate a polygynous mating system where the male is not devoted to only one female and her young, provided that the costs of sharing the male’s help may be outweighed by the advantage of mating with a superior male in a more suitable area (polygyny threshold model: Verner 1964; Orians 1969). Consequently, the contributions by each sex to parental care may be unequal. Trivers (1972) suggested that each sex attempts to maximize the care its mate provides while minimizing its own contribution. This study aims to understand how this inherent conflict between the sexes applies to poten- tially monogamous, but sometimes polygynous, Red Foxes. Some studies suggest that the male may sup- ply the majority of food for the kits (Henry 1986; 2001 Macdonald 1979), while others indicate that males share little of the responsibility of looking after the kits (Burchfield 1979). This variability in paternal behaviour may be just one part of a general social flexibility, or it could be simply a reflection of a lack of studies dealing specifically with the role of the male in the family group. In this study I evaluate the extent of male partici- pation in kit-rearing in eight Red Fox families in suburban and rural habitats of southern Ontario. Specifically, how do males contribute to rearing the young and how does this compare to the female’s contribution in rearing the young? VERGARA: PARENTAL ROLES IN RED FOXES 23 Study Area and Methods The study was conducted during April — June 1994 and 1995. Seven rural habitats and one suburban habitat of southern Ontario (all within a 90 km radius of the city of Peterborough) were included in the study (see Table 1). The area is characterized by warm summers and cool winters. Temperatures aver- age, for a thirty year period, 16.6° C in june, 19.7° C in July, -9° C in January and -7.8° C in February. The 1994 study year was colder than normal both in January (-15.9°C) and February (-11.7° C), warmer than normal in June (18.0° C) and close to normal in July (19.8° C). The 1995 study year was warmer than TABLE |. Composition of the studied families, hours of observation and den characteristics. Litter Hours of Family F M H Size: Observation Gaines ¥ ¥ - 4 ip? Keene ag y¥ ¥ 6 75 Tweed fy? iy = 7 88? Douro ne Y. _ 6 105 Railway Y. ¥ - = 139 Pogue a iY. - 9 eid Knox vi Vd Y 5 76 Highway Ye - - 2 384 Den site Characteristics Fenced field behind dog-food factory, suburban habitat, two distinct den areas 200 m apart simultaneously in use, multiple entrances. Open field, 300 m from barn, rural habitat, two distinct den areas 110 m apart in use, multiple entrances. Fenced hay and alfalfa field, rural habitat, two distinct den areas 60 m apart in use, multiple entrances. Location City of Coubourg 43°58’ N, 78°10’ W Otonabee Township 44°15’ N, 78°08’ W Huntingdon Township 44°28’ N, 77°25’ W Douro Township 44°23’ N, 78°12’ W Seymour Township 44°17’ N, 77°50’ W Verulam Township 44°33’ N, 78°40’ W Otonabee Township 44°18’ N, 78°14’ W Otonabee Township 44°15’ N, 78°09’ Open hay field, 300 m from farm house, rural habitat, multiple entrances. Open grassy hill side, 100 m from old railway track, rural habitat, two distinct den areas 50 m apart simultaneously in use, multiple entrances. Open hay field, 50 m from farm house, rural habitat, multiple entrances. Open grassy hill side, rural habitat, four distinct den areas 30 m apart in use. Fenced grass field, rural habitat, two den areas 27 m apart simultaneously in use, multiple entrances, one den only 2 m from Highway. F = female; M = male; H = helper; Y = present. 'Number of kits in each family at the beginning of the observation period. ?The Tweed male was only present until the kits were seven weeks old. °44 out of the 88 hours were observations made from blinds by John and Janet Foster (film makers and owners of the property) during the making of their documentary on Red Fox behavior, two weeks prior to my arrival to the study site. Their notes were kindly provided to me. *Observations at the highway family den were made in four sessions throughout a period of over one week. During this time I only saw two kits and the female. The male was never seen. I found the carcasses of two other kits lying at the side of the road; they had been killed by cars. 24 THE CANADIAN FIELD-NATURALIST normal in January (-5.6° C) and colder in February (-9.9° C), warmer in June (18.3°C) and close to nor- mal in July (19.5° C). Precipitation is rather uniform throughout the region and year varying from about 28 cm to 39 cm. The seven rural habitats in this study were typical of the southern Ontario landscape, with a mixture of cropland (principally hay and corn), uncultivated fields, cow pastures, and inter- mittent mixed forests of evergreen and deciduous trees. My study was carried out on a total of eight differ- ent Red Fox families (four each season) located with the help of interested rural residents. All observa- tions were made at the den sites, from a small vehi- cle, a tent, or a building. Since all eight den sites were in open areas, they could be observed from a distance of over 100 m with the aid of a spotting scope and wide-angle binoculars. At three of the den sites observations were made from a building, so the foxes could not see the observer (Gaines, Tweed, and Pogue dens). At the other five sites the foxes became habituated to the presence of the research vehicle or the tent within one to two days. Each adult was identified by the pattern of black pigmentation on the muzzle, the extent of white on the chest and neck, the shade of the fur (ranging from a light, sandy colour to an intense orange), and the shape and extent of white on the tip of the tail. It also soon became possible to identify adult male and female foxes based on differences in their behaviours. In fact, gender identification by behavioural differences has also been reported in other observational natal den studies (Swift Fox, Vulpes velox: Pruss 1994). In total, 724 hours of observations were conducted on the eight dens. A recorded hour is defined as an hour observing the den site, regardless of whether the adults were present or the kits were above or underground. Dens were observed primarily during hours of daylight, but did include 20 hours of noctur- nal observations using an ambient light image-inten- sifier scope. I adopted the working definition proposed by Woodroffe and Vincent (1994), and considered male care as any behaviour that benefits the young and that the male would not carry out if young were not present. Such direct parental behaviour includes: vis- iting and providing food to the cubs, resting and hud- dling with them, grooming, carrying, retrieving, actively defending, babysitting, and playing or socializing with the kits. I also looked at one form of indirect care: vigilant behaviour, and at adult-young distances as a measure of attentiveness which has been used in previous studies (Walters 1984). Instantaneous scan samples (see Altmann 1974) were taken at 30-second intervals. The percent of time is estimated from the percent of samples or intervals in which a given activity was recorded (Altmann 1974). This method of data collection can Vol. 115 be used to obtain data from more than one group member, by observing each in turn. If the behaviour of all visible group members (in this case adults and kits) is sampled within a very short time period the record approaches a simultaneous sample on all indi- viduals. Thus, this technique allowed to take full advantage of those instances in which the whole family unit was seen at once. Every 30 seconds, the activity of the adult and the kits and the distance of each adult to the nearest kit were recorded. I defined my behavioural categories based on observations made during den watching in a 1993 pilot study, and on the forms of direct care listed by Kleiman and Malcom (1981) and Woodroffe and Vincent (1994). Also, the number and duration of each adult visit to the den and the number of trips when an adult brought a prey item to the kits (and the type of prey) were noted. For each family the number of visits to the den made by each adult was divided by the total number of hours of observation of that family. Thus, the fre- quency of visits per hour (visiting rates) was calcu- lated for each adult. To estimate the distances, I considered one adult fox body length to equate 1 m. This method was veri- fied with the research assistants that were helping at three den sites (Gaines, Keene and Railway), by prac- ticing estimating a few distances between chosen fea- tures of the terrain, and then measuring the actual dis- tances to confirm that they fell within the estimates. Families studied and times of observation I conducted my observations from the time the kits were 5 weeks of age (kits emerge from the den at about 4 weeks), until approximately 11 weeks of age (see Figure 1). The composition and location of each family and the total number of hours of obser- vation at each den-site are summarized in Table | (further details on each family are given in Vergara 1996). The end of my observations of each family SECOND MONTH FIRST MONTH THIRD MONTH | Sr eee EE EO ae Oe | ACTATION KITS UNDERGROUND KITS FEED ON SOLID FOOD - | GAINES 1994 KEENE 1994 TWEED 1994 DOURO 1994 RAILWAY 1995 POGUE 1995 HIGHWAY 1995 KNOX 1995 FIGURE 1. Observation times for each family in relation to the major events in the rearing cycle. Dashes (Tweed) indicate that the family was observed dur- ing that time period, but data were recorded in a dif- ferent fashion (ad libitium). 2001 was not arbitrary but triggered by the sudden move of the foxes to a new unknown or unobservable den. Figure | shows the times at which the different fami- lies were observed in relation to the general develop- ment of the cubs (diagram design borrowed from Malcom and Marten 1982). The Highway den was only observed for 38 hours (four sessions, between 8 and 10 hours in duration each, distributed in one and a half weeks), and the male was never seen. The lack of male visitation at this den might be a reflection of the relatively low number of hours of observation of this family com- pared to the other families (average of 98 h). Similarly I observed the Tweed family for 44 hours (four 12-hour sessions distributed in two weeks), and obtained additional 44 hours of recorded observa- tions (a total of 88 hours) by documentary makers John and Janet Foster (see Table 1). I never observed the male, but he visited the den three times during the 44 hours prior to my arrival (John and Janet Foster, personal communication). Thus, statements concerning male visitation rates and feeding rates throughout this paper will be based on the remaining six families. Results Visiting Rates Any adult fox within 0 to 100 m from the den site was recorded as visiting the den. In general, adults visited during all hours of the day, but predominately in early morning (up to 3 hours after sunrise) and evening (2 or 3 hours before dusk). Although visits in the middle of the day were not infrequent, they were usually shorter in duration. Figure 2 shows the frequency of visits per hour for each adult. A Mann-Whitney U test revealed signifi- cantly more visits/hour by females than males (W = 84.0 p < 0.02). Females averaged 0.2 visits per hour (S.E.= 0.03) and males averaged 0.1 visits per hour (S.E.= 0.027). There is no significant correlation between the visiting rates of males and females (r = - 0.041, df = 5, p > 0.05). VISITS / HOUR KEENE TWEED DOURO VERGARA: PARENTAL ROLES IN RED FOXES 25 Excluding the Tweed and Highway families, male visitation rates varied from 0.06 visits/hour at the Railway den, to 0.24 visits/hour at Gaines (the adult male in this family visited the kits more often than the female did). Duration of the Visits For each family, the total number of visits for which a duration was known was used to calculate the mean time/visit for each adult (Figure 3). Males usually made shorter visits than females. The overall weighted average for females was 45 minutes, and for males 22 minutes. Thus, females stayed, on aver- age, twice as long as the males did when they visited the den. Yet, as shown by the standard errors, the visits varied considerably in length. The Railway female, for instance, stayed for a minimum of 4 min- utes, and a maximum of 171 minutes. Pupsitting by males Dog foxes spent a large proportion of their time at the den site in the absence of the vixens (Figure 4). On average, 78 % of the time spent by males at the den site they were there alone (SE = 8.7 %). This is referred to as “pupsitting” or “den guarding” in other studies (e.g., Moehlman 1983, for Silverbacked and Golden Jackals; Malcom and Marten 1982, for African Wild Dogs; Pruss 1994, for Swift Fox). Feeding the kits Number of prey items per trip The average number of prey items per trip was calculated by dividing the total number of prey items each parent brought to the young by the observed number of visits of that parent (Figure 5). There was no significant difference between males and females (W = 55.0 P > 0.1, Mann-Whitney U-test), however five of the six males have higher values than females, reflecting the fact that when males visited the den, they usually did bring food. Relative feeding contributions Figure 6 shows the relative contributions of each adult to feeding the kits in the two-parent families HIGHWAY FIGURE 2. Frequency of visits/ hour by each adult in each family. 26 THE CANADIAN FIELD-NATURALIST 14 80 — 10%) 70 LSUIL A Les 3 died, WY LILLE VLE 12 3 00 OV 1 EEE LMLLLLLL SIITLITL LL Mean time / visit (minutes) § ) Vol. 115 pp N LLL LL, SIISTIPLLSI GGA, Yh \O MMMM S SN 7 VNN N N. N NI IN BNE BLE Vi VN\N WE WY \ NI A oN} VNN Nh Vi \ N\A Vi VNN N] NI} NE NIX Ni NNN Ni NI WG NJ I GAINES KEENE TWEED DOURO RAILWAY POGUE KNOX HIGHWAY FiGuRE 3. Mean duration of visits (in minutes) for adults in each family. Vertical bars represent standard errors. The numbers above each bar represent the total number of visits to the den by that individual. The means are based on the number of visits for which the actual duration was known. (Highway and Tweed families: male absent during my observation period), as a proportion of the total number of prey items that the kits received. The Gaines and Knox males were the most attentive fathers, contributing 64 % and 46 % of the total number of food items respectively. The males of three families, Douro, Railway and Pogue, provided only 20 % of the food items. The third adult (a helper, see below) in the Keene family contributed 41 % of the total food items, more than the female or the male in the family (each provided 29 % of the food items). Figure 7 shows the types of food items that males and females were observed to bring to the cubs. Small rodents constituted the bulk of the food items for both males and females. More groundhogs were brought to the den by males than females (13 % vs. 1 %). Only the Tweed vixen was observed bringing a groundhog to the kits. Behavioral comparisons between males and females - Figure 8 shows the average proportion of time that males and females spent in different behaviours. Aside from nursing, the main difference between males and females seems to be in time spent in vigi- lant behaviour (females 35 %, males 60 %), and grooming the kits (females 12 %, males 2 %). Four of the six males groomed the young (Gaines, Keene, Douro and Knox), and did so a lower proportion of time than the respective females (Gaines: 1 % vs. 5 %: Keene: 4 % vs. 12 %; Douro: 4 % vs. 30%; Knox: 1% vs. 10 %). Vigilant behaviour was consid- ered an indirect form of care, represented by two behavioural categories: casual scan and very alert scan. When these categories were combined into one, males spent a significantly higher proportion of time than females in such alert behaviour (p < 0.02, Mann-Whitney U-test). The following behaviours were combined to pro- vide a single value for parental attentiveness (exclud- ing nursing) to test for significant differences between the sexes: (a) picking up and carrying prey items at the den site, caching them, dropping them at the den mound or feeding them directly to the kits; (b) groom- ing or cleaning the kits; (c) playing with the kits, (d) 2001 422 min Percentage of Time GAINES KEENE DOURO RAILWAY POGUE VERGARA: PARENTAL ROLES IN RED FOXES 27 124 min 411 min 44 min . KNOX Males in each family FIGURE 4. Percentage of the total time spent by males at the den site in the absence of the female. The 100 % level represents the total time (in minutes) that each male attended the den. The relatively small amounts of time for the Douro, Pogue, and Railway families reflect the fact that the visits of these males were particularly brief (usually to bring food to the den and leave). making muzzle to muzzle contact with the kits, usual- ly when the kits are soliciting food (i.e., “being solicit- ed’’); (e) being harassed or followed by the kits, which refers to all those instances when the adults passively tolerated kits that persisted in following the adult, bit- ing its legs and tail, climbing over it, or pouncing on it; and (f) checking the den entrances, which refers to an adult sticking its head and upper body in a den (if the cubs are inside the den, this behaviour often sum- mons them out). Overall, females spent a significantly greater proportion of time (27.6% on average) than males (16.1 % on average) in behaviours that indicate i Wl PREY ITEMS / TRIP YUN Witt ULM Wi, ease Wn with YM “2 2 < o | iu parental attentiveness (Mann-Whitney U-test, W = 78 p< 0.02). Females interacted with the kits in their visits sig- nificantly more than males (chi square goodness of fit for two samples, y° = 45.8) di = 1, P< 01001). Males often came to the den site bringing a food item without approaching the kits or interacting with them in any way (28 % of the visits were non-inter- active, n = 59), dropping the item near the main den entrance and immediately moving away from the mound. Females interacted with the young in 96 % of their visits (n = 156). WM, CMLL * WLLL WM hs Eee] 2 UMM Hu LLL POGUE HIGHWAY FIGURE 5. Number of prey items/visit for each adult in each of the eight study families. 28 THE CANADIAN FIELD-NATURALIST Helper(41%) Female(80%) Helper(0%) Male(46%) Female(54%) Vol. 115 Male(64%) Female(36%) 1e(31%) FIGURE 6. The relative contribution of each adult to feeding the kits in six families. Adult distances from the kits As shown in Figure 9, females spent most of their time at the den in contact with the kits (35 %), with- in 1 m (25%) and between | to 10 m (25 %) of the kits. Males, on the other hand, spent most of their time between | to 10 m (35 %), and between 10 to 20 m (19%) from the nearest kit. Males spent a higher proportion of their time than females at dis- tances of 20 to 50 m (females 5 %, males 13 %) and 50 to 100 m (females 2 %, males 14 %). For the seven distance categories in Figure 9, a chi-square test of independence provides strong evidence that females remain closer to kits than do males (x? = 163.238, dt = 6) P<0:0)) Patterns of parental care: presence of helpers There were differences in parental behaviour between males and females, as well as between females of different families and males of different families (Vergara 1996 provides behavioural details for each individual). Furthermore, two fam- ilies (Knox and Keene) had quite attentive fathers and an additional non-breeding adult (helper) that contributed to raising the young. The Keene helper was a female. Her gender was confirmed when the animal rolled over on its back or sat on its haunch- es. Also a squatting urination was observed. She visited the den site as often as the breeding female (0.17 visits/hour, Figure 2), showed the same behaviours than the adult female (excluding nurs- ing) indicative of parental attentiveness, and con- tributed 41 % of the food items brought to the kits. The Knox third adult was only observed once, and 2001 VERGARA: PARENTAL ROLES IN RED FOXES 29 Females (n = 75) Unidentified( 11%) Placenta chunks(9%) ) Squirrel(3%) Rabbit( 1%) Groundhog( 1%) Muskrat(5%) | ~ VSO ELE: Large domestic bird(3%) ¥ Small bird (chickadee size)(3%) Medium Bird (Robin size)(8%) Small rodent (vole sizc)(28%) Medium Rodent (squirrel size)(9%) ‘Small / medium mammal (unident.)(19%) Males (n = 32) Unidentified(6%) Dead Fox Kit(3%) Medium Bird (Robin size)(3%) <& Large Grouse(3%) Small rodent (vole size)(34% ) Medium Rodent (squirrel size)(6%) Small / medium mammal (unident.)(13%) FIGURE 7. Food items observed being fed to kits by male and female Red Foxes in the eight study families (n = individual feedings). its gender was not identified. Since it was greeted by the kits, and it groomed them and guarded them for a period of two hours in the absence of the breeding pair, I assumed it was indeed a helper (see Vergara 1996 for behavioral details on these helpers). I cannot relate patterns of care (e.g., presence of helpers) to offspring survival since I did not observe the young until the age of independence. Discussion In general, this study shows that Red Fox males tend to contribute less direct care than females in raising the young. Apart from lactating, females were generally more interactive with the young, and stayed closer to the young than males. Females also visited the den more frequently and for longer peri- ods of time than males. The number of trips/hour that one parent made to the den was independent of 30 45% 40% 35% 30% (Dd) = % 25% o s |p aed (Bb) (au, 15% i SSS: 10% \ \ « N N \ N N N ZRrREES Se EO eae wa ai |! Ce ae ee ce Sv tere SF Ges AS x < ke? SO) ibe? a = aa o Ors = x 3 aj A THE CANADIAN FIELD-NATURALIST SEE EN 1 Casual Scan | LLAALLcLLLLLLLLELLLLLLELLLLTE Vol. 115 FEMALES (N = 8) [] MALES (N=6) U ey SS: Ne su eg NT N ap fom oS oe) Soap > ob 16 Sees So Se eS eS 7) fost ay wo oe on > a a = a v FIGURE 8. Average proportion of time that males and females spent in different behaviours, based on a total of 165 females visits and 58 male visits to the den. Vertical bars rep- resent standard errors. the number of trips/hour made by the other parent. Males did exhibit, to a lesser extent, the same direct parental care patterns as females. The difference was particularly marked when looking at behaviours that involved physical contact, such as grooming the kits. The combination of greater direct care and the ener- getic costs of gestation and lactation results in a greater total investment in direct care by female than by male Red Foxes. To make conclusive statements about food provid- ed to the kits, variations in the gross energetic con- tent of the prey items should be taken into account. The size of the prey items brought to the den influ- ences the energy intake of the young. Furthermore, some prey items may take more time and energy to hunt than others. The only real difference seems to be in the number of groundhogs brought to the den. However the numbers are relatively small (four ‘groundhogs brought by males vs. one brought by a female), and the relative percentages of the other prey types do not seem to differ between males and females. Further studies should incorporate an analy- sis of the energetic aspects such as energy supply by different prey items, estimations of prey relative abundance and ease of capture. There was some individual variation in parental behaviour among females (particularly in terms of grooming, nursing and contact rates with the young), and considerable variation among the males. There were three males that rarely visited the den (Douro, Railway, and Pogue), and provided less food items than females to the young. Three other males visited regularly (Keene, Gaines, and Knox) but showed dif- ferent degrees of close interaction with the young. The Knox male barely interacted with the young and kept his distance from them, whereas the Gaines and Keene males were indeed quite affectionate to the young. In my study, only one father (Gaines) provid- ed more food items to the kits than the mother (two thirds of the food items brought to the den). 2001 30% Percentage of Time b x LS 1-10 m 0% ¥ g lm 10-20 m VERGARA: PARENTAL ROLES IN RED FOXES 3] [_] FEMALES (N = 8) MALES (N = 6) Me TWN CWS 20-50m 50-100m >100m FIGURE 9. Average proportion of time that males and females spent at different distances from the nearest kit, based on 165 female visits and 58 male visits. Vertical bars rep- resent standard errors. However, dog foxes providing more food than vix- ens have also been reported in other studies. For instance, Macdonald (1979) found that the father of a captive Red Fox family group provided 41.1% of the food items to the kits, more than either the moth- er (21.2%), or the two helpers (17 % and 20 %). Finally, another aspect of the observed variability in parental care patterns is the presence of a helper in two of the families (Keene and Knox). In Ontario, evidence of several vixens within a family group, helping, and communal denning is sparse (Voigt and Macdonald 1984). Intraspecific variation in paternal care is not unique to foxes. Male care in Coyotes (Canis latrans) was reported absent in some parts of their range (e.g., Minnesota, Berg and Chesness 1978) and present in others (e.g. Wyoming, Camenzind 1978). In fact, intraspecific variation in patterns of parental care and social organization in contrasting habitats may be as pronounced as interspecific differences (Macdonald 1980, and Moehlman 1989 for canids; Bekoff et al. 1984, for carnivores, citing examples of Coyotes, Wolves, Kalahari Lions, Panthera leo, Brown Hyenas, Hyaena brunnea, Striped Hyenas, Hyaena vulgaris, Golden Jackals, Raccoons, Procyon lotor, African Wild Dogs and Red Foxes; Moehlman 1998, for Feral Asses, Equus africanus, Roberts et al. 1998, for Prairie Voles, Microtus ochrogaster, Yamagiwa and Hill 1998, for Japanese Macaques, Macaca fus- cata; Blumstein and Armitage 1999, for several species of Marmot, Marmota spp. ). Variability in the reproductive strategies of male foxes may be related to food, as it is well known that differences in food availability may be primarily responsible for intraspecific variation in carnivore social organization (Bekoff et al. 1984; Macdonald and Moehlman 1982). Examples in canids are preva- lent (e.g. Coyotes: Bekoff and Wells 1982, Bowen 1981; Red Foxes: Englund 1980, Macdonald 1979, Zabel and Taggart 1989; Jackals: Ferguson et al. 1983, Moehlman 1983; African Wild Dogs: Frame et al. 1979). However, in this study I did not quantify and compare the resources available around the dif- ferent den sites. Thus, any proposed relationship between the level of paternal care found in this study and the availability of food will remain a specula- tion, and invites further research. The individual variability I observed could be simply explained as idiosyncratic differences. Such individual differences have been reported in other canid studies. For instance, Venkataraman (1998) noted significant differences across adult males in two Dhole, Cuon alpinus, packs in the frequency of pup-raising behaviours displayed at the den. Indirect care An important consideration when evaluating the extent of male care is the possibility that some indi- viduals make most of their investment indirectly. My study indicates that dog foxes do invest less than vixens in direct parental effort. However, my results clearly show that males tend to spend significantly more time in sentinel behaviour than females. Vigilant behaviour is often considered a form of indirect male parental care. Indirect care may include activities which a male would perform regardless of the presence or absence of the young, but that may aD, THE CANADIAN FIELD-NATURALIST increase the young’s survivorship (Kleiman and Malcom 1981). The frequencies or proportions of these behaviours (such as very alert or casual scan in this study) may alter in response to the existence of young, even if direct care is not seen (Kleiman and Malcom 1981). It would be useful, but difficult, to compare vigilant behaviour in males and females when both are away from the young, in order to determine if the higher proportion of time spent scanning by males at the den sites is associated with the presence of young. The importance of indirect forms of care should not be underestimated. In the present study, although males provided less direct care than females, they consistently scanned the area more than females. This behaviour may help to detect danger, such as Humans, Coyotes or strange foxes. An additional indirect form of male care is shelter construction and maintenance (Kleiman and Malcom 1981). Few carnivore males are reported to take a major role in burrow construction (Kleiman and Malcom 1981). In my study, the Knox male con- tributed to the excavation of the woods den, where the family subsequently moved. This insures that the necessary resources (in this case, space) are available to the young. This study supports the notion that for polygyny to occur, the female must be able to provide most of the offspring’s needs. In Red Foxes, which exhibit a potentially monogamous, occasionally polygynous system, males contribute less than females to the direct care of the young, and their contribution is also qualitatively different. The present study also corroborates the great flex- ibility in the behaviour of Red Foxes. Elucidating the ecological circumstances that promote this variation in both social groupings and patterns of parental care in southern Ontario should be an objective of further studies. Understanding the social structure, mating system, and patterns of parental care of any fox pop- ulation requires an understanding of food dispersion and availability in such area. Thus, further studies of variation in paternal care should quantify and com- pare the resources available around the different den sites. Behavioural variation in male paternal care and the variation in female responses to different degrees of care by the male should be an important research priority. Acknowledgments This research was entirely funded by a Trent University Research Committee Grant to my super- visor, Michael Berrill. I thank him for his assistance and encouragement in every stage of this study. I am also grateful to Joe Cebek, David Lasenby, Jim Sutcliffe, Erica Nol and Dennis Voigt for their help- ful assistance during the analysis of my data and for field equipment. The comments and suggestions of Shelley Pruss and three anonymous reviewers great- Vol. 115 ly improved this manuscript. Many thanks to the owners of the properties where the dens were locat- ed for permission to work on their lands, and to John and Janet Foster for providing footage and data on the Tweed den. I am grateful to Reka Anthony, Cam Collier, Heather Lee, and Lise Macgillivray for their help in the field. Finally, I must thank the foxes for allowing me the privilege of sharing their world with them. Literature Cited Altmann, J. 1974. Observational study of behaviour: sampling methods. Behaviour 48: 1-41 Asa, C.S., and C. Valdespino. 1998. Canid reproductive biology: an integration of proximate mechanisms and ultimate causes. American Zoologist 38: 251-259. Baker, P. J., C. P. J. Robertson, S. M. Funk, and S. Harris. 1998. Potential fitness benefits of group living in the red fox, Vulpes vulpes. Animal Behaviour 56: 1411-1424. Bekoff, M., J. T. Daniels., and J. L. Gittleman. 1984. Life history patterns and the comparative social ecology of carnivores. Annual Review of Ecology and Syste- matics 15: 191-232. Bekoff, M., and M. C. Wells. 1981. Behavioral budgeting by wild coyotes: the influence of food resources and social organization. Animal Behaviour 29: 794-801. Berg, W. E., and R. A. Chesness. 1978. Ecology of coy- otes in Northern Minnesota. Pages 299-247 in Coyotes. Edited by M. Bekoff. Academic Press, New York. Blumstein, D. T., and K. B. Armitage. 1999. Coopera- tive breeding in marmots. Oikos 84: 369-382. Bowen, W.D. 1981. Variation in coyote social organiza- tion: the influence of prey size. Canadian Journal of Zoology 59: 639-652. Burchfield, G. L. 1979. The ecology of the red fox, Vulpes vulpes, in south-western N.S.W. M.Sc.thesis. University of Sydney, Sydney, Australia. Camenzind, F. J. 1978. Behavioral ecology of coyotes (Canis latrans) on the National Elk Refuge, Jackson, Wyoming. Ph.D. thesis. University of Wyoming, Laramie. 97 pages. Cavallini, P. 1997. Variation in the social system of the red fox. Ethology, Ecology and Evolution 8: 323-342. Doncaster, C. P., and D. W. Macdonald. 1991. Drifting territoriality in the red fox Vulpes vulpes. Journal of Animal Ecology 60: 423-439. Englund, J. 1980. Population dynamics of the red fox (Vulpes vulpes) in Sweden. Pages 107-121 in The red fox. Edited by E. Zimen. Biogeographica. Volume 18. The Hague, The Netherlands. Ferguson, J. W. H., J. A. Nel, and M. J. DeWet. 1983. Social organization and movement patterns of black- backed jackals, Canis mesomelas, in South Africa. Journal of Zoology (London) 199: 487-502. Fentress, J. C., and J. Ryon. 1982. A long term study of distributed pup feeding in captive wolves. Pages 239-261 in Wolves of the world, perspectives of be- haviour, ecology and conservation. Edited by F. H. Harrington and P. C. Paquet. Noyes Publications, Park Ridge, New Jersey. 474 pages. Frame, L. H., J. R. Malcom, G. W. Frame, and H. Van Lawick. 1979. Social organization of African wild 2001 dogs (Lycaon pictus) on the Serengeti Plains, Tanzania, 1967-1978. Zeitschrift fuer Tierpsychologie 50: 225-249. Garrott, R. A. 1980. Den characteristics, productivity, food habits and behaviour of arctic foxes in northern Alaska. M. Sc. Thesis, Pennsylvania State University. Pennsylavania. 95 pages. Geffen, E., and D. W. Macdonald. 1992. Small size and monogamy: spatial organization of Blanford’s foxes, Vulpes cana. Animal Behaviour 44: 1123-1130. Harrington, F. H., and L.D. Mech. 1982. Patterns of homesite attendance in two Minnesota wolf packs. Pages 81-105 in Wolves of the world, perspectives of behaviour, ecology and conservation. Edited by F. H. Harrington and P. C. Paquet. Noyes Publications, Park Ridge, New Jersey. 474 pages. Henry, D. J. 1986. Red Fox. The catlike canine. Smith- sonian Institution Press. Washington, D.C. 174 pages. Henry, D. J. 1993. How to spot a fox. Chapters Publish- ing Ltd., Shelburne, Vermont. 110 pages. Kleiman, D. G. 1977. Monogamy in Mammals. The Quarterly Review of Biology 52: 39-65. Kleiman, D. G., and J. R. Malcom. 1981. The evolution of male parental investment in mammals. Pages 347-387 in Parental Care in Mammals. Edited by D. J. Gubernick and P. H. Kopfer. Plenum Press. New York and London. Macdonald, D. W. 1979. Helpers in fox society. Nature 282: 68-71. Macdonald, D. W. 1980. Rabies and Wildlife. A biolo- gist’s perspective. Oxford University Press. 151 pages. Macdonald, D. W., and P. D. Moehlman. 1982. Cooperation, altruism, and restraint in the reproduction of carnivores. Pages 433-67 in Perspectives in Ethology: Ontogeny. Volume 5. Edited by P. P. G. Bateson and P. H. Klopfer. Plenum Press. New York. 519 pages. Malcom, J. R., and K. Marten. 1982. Natural selection and the communal rearing of pups in African wild dogs (Lycaon pictus). Behavioral Ecology and Sociobiology 10: 1-13. Malcom, J. R. 1985. Paternal care in canids. American Zoologist 25: 857-859. Moehlman, P. D. 1983. Socioecology of silverbacked and golden jackals (Canis mesomelas and Canis aureus). Pages 423-453 in Recent Advances in the Study of Mammalian Behaviour. Edited by J. F. Eisenberg and D. G. Kleiman. Special Publication 7, American Society of Mammalogists. 753 pages. Moehliman, P. D. 1986. Ecology of cooperation in canids. Pages 64—87 in Ecological aspects of social evolution. Edited by D. I. Rubenstein and R. W. Wrangham. Prince- ton University Press, Princeton, New Jersey. 551 pages. Moehlman, P. D. 1989. Intraspecific Variation in Canid Social Systems. Pages 143-163 in Carnivore Behavior, Ecology and Evolution. Edited by J. L. Gittleman. Chapman and Hall, London. Moehlman, P. D. 1998. Feral asses (Equus africanus): Intraspecific variation in social organization in arid and mesic habitats. Applied Animal Behaviour Science 60: 171-195. VERGARA: PARENTAL ROLES IN RED FOXES 33 Orians, G. H. 1969. On the evolution of mating systems in birds and mammals. American Naturalist 103: 589-603. Pruss, S. D. 1994. An observational natal den study of wild swift fox (Vulpes velox) on the Canadian prairie. M. Sc. thesis. University of Alberta. Edmonton. 213 pages. Rasmussen, J. L., and R. L. Tilson. 1984. Food provi- sioning by adult maned wolves (Chrysocyon brachyu- rus). Zeitschrift fuer Tierpshychologie 65: 346-352. Roberts, R. L., J. R. Williams, A. K. Wang, and C.S. Carter. 1998. Cooperative breeding and monogamy in prairie voles: influence of the sire and geographical var- iation. Animal Behaviour 55: 1131-1140. Storm, G. L., R. D. Andrews, R. L. Phillips, R. A. Bishop, D. B. Siniff, and J. R. Tester. 1976. Morphology, reproduction, dispersal, and mortality of midwestern red fox populations. Wildlife Monographs 49: 1-82. Trivers, R. L. 1972. Parental investment and sexual selec- tion. Pages 136-179 in Sexual selection and the descent of man. Edited by B. Campbell. Aldine, Chicago. Veado, B. V. 1998. Parental behaviour in maned wolf Chrysocyon brachyurus at Belo Horizonte Zoo. Interna- tional Zoo Yearbook 35: 279-286. Venkataraman, A. B. 1998. Male-biased adult sex ratios and their significance for cooperative breeding in dhole, Cuon alpinus, packs. Ethology 104: 671-684. Vergara, V. 1996. Comparison of parental roles in male and female red foxes (Vulpes vulpes) in Southern Ontario, Canada. M. Sc. thesis. Trent University, Peter- borough, Ontario. 115 pages. Verner, J. 1964. Evolution of polygamy in the long-billed marsh wren. Evolution 18: 252-261. Voigt, D. R. 1987. Red Fox. Pages 378-392 in Wild Fur- bearer Management and Conservation in North Amer- ica. Edited by M. Novak, J. Baker, M. E. Obbard, and B. Malloch. Ontario Ministry of Natural Resources, Toronto. 1150 pages. Voigt, D. R., and D. W. Macdonald. 1984. Variation in the spatial and social behaviour of the red fox, Vulpes vulpes. Acta Zoologica Fennica 171: 261-265. von Schantz, T. 1984. Non-breeders in the red fox Vulpes vulpes: a case of resource surplus. Oikos 42: 59-65. Walters, J. R. 1984. The evolution of parental behaviour and clutch size in shorebirds. Pages 243-287 in Behavior of Marine Mammals. Volume 5. Edited by J. Burger and B. L . Olla. Plenum Press, New York. Woodroffe, R., and A. Vincent. 1994. Mother’s little helpers: patterns of male care in mammals. Trends in Ecology and Evolution 9: 294-297. Yamagiwa, J., and D. A. Hill. 1998. Intraspecific varia- tion in the social organization of Japanese macaques: past and present scope of field studies in natural habi- tats. Primates 39: 257-273. Zabel, C. J., and S. J. Taggart. 1989. Shift in red fox, Vulpes vulpes, mating system associated with E] Nino in the Bering Sea. Animal Behaviour 38: 830-838. Received 27 September 1999 Accepted 12 May 2000 Biodiversité microfongique du Fagus grandifolia dans une forét ancienne: bioindicateurs et structure mycosociologique VLADIMIR VUJANOVIC!? et JACQUES BRISSON! ‘Institut de recherche en biologie végétale, Université de Montréal, 4101 est, rue Sherbrooke, Montréal, Québec H1X 2B2 Canada "Département de Sciences biologiques, Université de Montréal, C.P. 6128, Succursale « Centre-ville », Montréal, Québec, H3C 3J7 Canada Vujanovic, Vladimir, et Jacques Brisson. 2001. Biodiversité microfongique du Fagus grandifolia dans une forét ancienne : bioindicateurs et structure mycosociologique. Canadian Field-Naturalist 115(1): 34-42. En Amérique du Nord, I’étude de la biodiversité des micromycétes dans les foréts anciennes représente un potentiel incon- tournable et largement inexploré. Le présent inventaire nous a permis de mettre en évidence la biodiversité microfongique, les bioindicateurs et la structure mycosociologique associée au hétre a grandes feuilles (Fagus grandifolia) dans une forét anci- enne de l’est du Canada. La méthodologie utilisée s’appuie sur la stratégie proposée par le All-Taxa Biodiversity Inventory of Fungi (ATBIB). Plusieurs espéces s’avérent des nouvelles mentions sur un hdte du genre Fagus. C’est le cas pour tous les membres de la sous-famille Phialostromatinae (Deuteromycota, Coelomycetes), soit quatre genres et cing espéces, dont cer- taines pourraient étre des espéces trés rares ou menacées. Quatre associations, chacune étant associée a une partie spécifique de l’arbre, ont été déterminées. De plus, deux autres associations sont caractéristiques d’habitats particuliers dans la forét. Notre étude supporte l’hypothése selon laquelle les forets anciennes sont des écosystemes en homéostasie qui constituent des réservoirs d’espéces microfongiques d’une grande diversité génétique. Outre la valeur intrinséques de cette diversité excep- tionnelle, les micromycétes ont également une valeur pratique potentielle, comme dans le contexte de la lutte biologique. Mots clés: All Taxa Biodiversity Inventory (ATBD), biodiversité, micromycétes, bioindicateurs, coelomycetes, endophytes, hétre, forét ancienne, Fagus grandifolia, Québec. Les micromycétes entretiennent des relations de ation de la distribution spatiale de micromycétes parasitisme ou de mutualisme trés intimes avec les colonisateurs (Chapela et Bodddy 1988). Le plantes, insectes, animaux ou autres micro-organ- paradigme de la colonisation des arbres vivants par ismes suite au processus de coévolution (Pirozynski des micromycétes est déterminé majoritairement par et Hawksworth 1988). Ce groupe taxonomique est la position des niches écologiques; la lumieére, la pourtant fort important, car on estime qu’il y a beau- _—disponibilité de la chaleur et 1’ humidité (Boddy et coup plus d’espéces de micromycétes dans la nature Rayner 1983; Oshima et al.1997). que de plantes hdtes (Hawksworth 1991). Ces Le retard a combler dans la connaissance de la bio- espéces microphongiques sont étroitement associées _diversité fongique, l’augmentation de pathogénes vir- a des substrats particuliers, soit un hdte spécifique, a ulents (CFS, 1997) et une diminution drastique de une partie d’un h6te (écorce, feuille, graines, etc.) ou plusieurs de ces organismes utiles di a l’influence de encore a un stade de développement particulier |’activité humaine (Roberts et Gilliam 1995) ont (Lisiewska 1992). Ainsi, dans une biocénose forte- |mené au développement d’une nouvelle stratégie sci- ment organisée comme dans une forét agée, le plus _ entifique, le Al/-Taxa Biodiversity Inventory of Fungi grand nombre de niches potentielles peut résulter en (ATBIF), une composante majeure du All-Taxa une plus grande biodiversité microfongique (Amolds Biodiversity Inventory (ATBI) (Rossman 1994). Le 1988). Les micromycétes sont davantage susceptibles ATBIF propose une méthodologie de réalisation de se comporter comme des indicateurs particuliers d’inventaire qui tient compte de la variété des sub- de l’Age de la forét, mais aussi des changements envi- strats (hétes) et de la variation d’abondance des ronnementaux (Horn 1985; Miller 1989; Vujanovic champignons dans l’espace et dans le temps. Cette 1994). stratégie s’avére particuliérement appropriée pour La forét ancienne est reconnue par des arbres 4gés, _|’étude de la biodiversité des micromycetes dans un de grandes variétés de milieux, sa complexité et sa _ habitat, et sur une superficie limitée (Vujanovic et al. richesse biologique ainsi que ses communautés d’Or- 1997). ganismes vivants (Carroll 1995; Stone et al. 1996). La réserve écologique du Boisé-des-Muir est une La communauté microfongique est une composante _forét ancienne dominée par |’érable a sucre, le hétre et intégrale de ces milieux (McCutcheon et al. 1993) qui___la pruche. Avec ses arbres tricentenaires, elle serait interagissent avec d’autres organismes selon la phys- une des derniéres représentantes de la grande forét ionomie, la structure et de la couverture des arbres _décidue de 1’ Amérique du Nord, telle qu’elle existait a (Chapela 1989). Comme résultat, on détecte une vari- _|’arrivée des premiers colons européens (Brisson et al. 34 —— ee eee ae 2001 1992). Conséquemment, elle est susceptible de révéler une biodiversité microfongique particuliére incluant des espéces trés rares et associées a des niches écologiques étroites. Notre étude préliminaire a per- mis la découverte de deux espéces de micromycétes nouvelles pour la science, le Diarimella laurentidae Vujanovic, St-Arnaud, Neumann & Fortin (Vujanovic et al. 1998) et le Polynema muirii Vujanovic, St- Arnaud, Neumann, Fortin & Brisson (Vujanovic et al. 1999a). Il est généralement reconnu qu’une biodiversité spécifique est associée aux foréts en santé (Aplet et al.1993). Il y a done une certaine urgence 4 mieux connaitre la biodiversité du Boisé-des-Muir si on veut pleinement profiter des bénéfices de sa spécificité, car son intégrité écologique est menacée par | ’intro- duction récente d’une maladie fongique exotique, la maladie corticale du hétre (Vujanovic et Brisson 1999b). La maladie corticale menace de causer la dis- parition de certains micromycétes associés directe- ment ou indirectement au hétre, et 1’augmentation d’espéces pathogénes ou de décomposeurs qui prof- iteront de l’affaiblissement du hétre. A plus longue échéance, les changements climatiques et |’ éventual- ité de plus en plus grande que des événements clima- tiques extremes, comme le récent verglas, viennent bouleverser |’équilibre de la forét, favoriseront vrai- semblablement une augmentation des maladies fongi- ques (Coakley 1995). L’objectif principal de cette recherche est d’exam- iner la biodiversité microfongique, les bioindicateurs et la structure mycosociologique associée au parties aériennes du hétre a grandes feuilles (Fagus grandi- folia Ehbrh.) dans la réserve écologique du Boisé-des- Muir. Méthodologie La méthodologie utilisée s’appuie sur la stratégie développée par ATBIF (Rossman 1994). Etant donné que la diversité microfongique peut varier selon les situations édaphiques et phytosociologiques locales (Vujanovic 1995; Vujanovic et Vuckovic 1994), l’échantillonnage a été stratifié 4 l’aide des cartes de distribution des arbres de la réserve (Brisson et al. 1989) afin de couvrir tout le spectre des conditions locales. Echantillonnage L’échantillonnage a été effectué 2 fois, en automne 1996 et 1997. Un échantillonnage systématique a été réalisé en sélectionnant 45 arbres dont 15 dans cha- cune des catégories suivantes: étage supérieur, étage médian et sous-bois. Aussi, des parties vivantes et mortes ont été récoltées afin de couvrir a la fois les champignons pathogénes et décomposeurs. Au total, 1911 échantillons d’écorce ont été récoltés sur dif- férentes parties de l’arbre: tronc (morceau d’écorce de 36 mm”), branches (diamétre > 3 cm ; portions de branche échantillonnée de 10 a 20 cm de longueur), VUJANOVIC ET BRISSON: BIODIVERSITE MICROFONGIQUE 35 rameaux (diamétre = 0,5 a 3 cm; portions échantillon- nées de 5 a 10 cm) et ramilles (diamétre < 0,5 cm por- tions échantillonnées de 1 45 cm). Notons qu’étant donné le court mandat de cette étude, l’échantillonnage ne couvre qu’une partie de la biodiversité microfongique associée au hétre car elle ne permet pas de couvrir les micromycétes apparents a d’autres moments de |’année (I’hiver, le printemps, l’été) ou sur d’autres parties de la plante (fruits, feuilles, graines). Les échantillons récoltés ont été placés dans des boites ou sacs stériles. Au moment du prélevement, nous avons noté les traces de symp- tomes, la date de prélévement et la localisation dans le boisé (Myren et al. 1994). Isolation des micromyceétes sur milieux de culture Les échantillons ont été conservés dans les sacs stériles en plastique a l’obscurité dans un incubateur a 4°C, jusqu’a leur traitement. La procédure de | ’isola- tion consiste : (1) stérilisation du surfaces des échantil- lons d’écorce et de rameau a l’aide de bain d’hypo- chlorite de sodium (0.5%); (11) la préparation des milieux de culture (potato dextrose agar, czapeck agar, malt agar); (111) l’inoculation d’une piéce d’écorce (8 mm/?) sur chacun des milieux, en trois répétitions; (iv) le repiquage et la purification des iso- lats et (v) la conservation des isolats sur milieux nutri- tifs selon la procédure développée par Dhingra and Sinclair (1987) et Booth (1971). Identification Les identifications de champignons ont été faites selon les caractéres morphologiques et culturaux a Vaide des clés d’identification : Microchampignons selon Ellis et Ellis (1985) et Lanier et al. (1978); Coelomycetes selon Nag Raj (1993) et Sutton (1980); Ascomycota selon Hanlin (1990), Sivanesan (1984), Breintenbach et Kranzlin (1981), Dennis (1978), et Barr (1978); Deuteromycota selon Barnett et Hunter (1987); Hyphomycétes selon Ellis (1971). La nomen- clature et la phylogénie adoptée est celle de Hawks- worth et al. (1995). Etant donné la complexité souvent encore inexplorée associée au groupe des micro- mycétes, plusieurs des identifications réalisées se limiteront au stade du genre, et certains champignons demeureront méme non identifiés. Les spécimens microfongiques sur |’écorce portant des fructifications des micromycétes ont été déposés a l’ Herbier Marie-Victorin (IRBV, MT: 10311-13260). Statut des espéces trouvées La biodiversité microfongique est si mal connue qu’il est souvent difficile de statuer sur la rareté des taxons identifiés. Les connaissances actuelles concer- nant le statut des espéces en Amérique du Nord en est a peine a l|’étape de |’inventaire. La liste des micro- myceétes du Boisé-des-Muir a été comparée avec l’inventaire le plus récent des espéces trouvées au Québec (SPPQ 1996) au Canada (Ginns 1986), aux Etats-Unis (Farr et al. 1989). Enfin, nous avons quali- 36 THE CANADIAN FIELD-NATURALIST fié la rareté des taxons a |’échelle du Boisé-des-Muir selon leur représentation dans les échantillons récoltés. Résultats et Discussion Biodiversité microfongique sur l’hdte du hétre Au total, 60 genres comprenant 76 espéces appar- tenant au Ascomycota, Deuteromycota (= Mitosporic Fungi) et au Zygomycota ont été isolés (Figure 1 et Tableau 1). Parmi ces espéces, 65 % des Hypho- mycetes, 75% de Coelomycetes et 39% des Ascomy- cotina représentent les nouvelles mentions comme colonisateurs du hétre au Canada (Figure 1). Pour le Québec, on parle de 84%, de 100%, et de 96%; et pour les Etats-Unis, de 58%, de 45% et suivi de 26%, respectivement. D’ailleurs, plusieurs espéces trou- vées au Boisé-des-Muir n’avaient encore jamais été mentionnées sur un héte du genre Fagus. C’est nota- mment le cas pour tous les membres de la sous- famille Phialostromatinae (classe Coelomycetes) trouvés au boisé (Figure 2), soit quatre genres 1) Diarimella, 11) Dinemasporium, iii) Polynema, et iv) Pseudolachnea, comme le confirme la mise a jour de ce groupe de microchampignons réalisée par Nag Raj CS83) Biodiversité microfongique selon la position dans l’arbre Plusieurs espéces s’attaquent spécifiquement a une portion d’arbre, alors que d’autres s’aveérent ubiquistes (Tableau 1 et Figure 3). I. Association : Ustulineto-Nectrietum. Au total 20 espéces caractérisent cette mycocénose. Ustulina deusta agent de la carie blanche et plusieurs Nectria spp. responsables de chancre nectrien colonisent la Vol. 115 partie basse du tronc des vieux arbres, jusqu’a une hauteur de 3 m; II. Association : Hypoxylonetum. Au total 9 espéces caractérisent cette mycocénose, dont celles du com- plexe de l’Hypoxylon spp. (chancre hypoxylonien). Elles dominent sur les premiéres branches de la couronne et sur la partie supérieure du tronc; III. Association : Quaternato-Diatrypetum. La communauté microfongique associée aux rameaux (0.5-3 cm diam.) est représentée par les espéces pathogénes du complexe de Dyatrype spp. (dépérisse- ment diatripéen). La biodiversité de cette mycocénose compte au total 28 espéces récoltées ; IV. Association : Valsetum. Sur les ramilles (<0.5 cm. diam.), les espéces du complexe Valsa spp., ana- morphe: Cytospora spp., (chancre cytosporéen ou dépérissement cytosporéen) dominent. Dix-sept especes caractérisent cette mycocénose. Parmi les genres ubiquistes, on retrouve Verti- cillium, Trichoderma et Cladosporum. Quelques-unes de ces espéces présentent un potentiel intéressant comme agent de lutte biologique, contre certaines maladies fongiques dont la maladie corticale du hétre (Lonsdale 1982). Dispersion horizontale des communautés microfongiques du Boisé-desMuir Un traitement complémentaire a été fait dans V optique d’enrichir l’analyse de la diversité par une information mycosociologique basée sur les bioindi- cateurs des habitats étudiées (Tableau 2). A,«, Association : Ascodichaenetum. L’ apparition de l’espéce différentielle Ascodichaena rugosa et les quatre autres espéces (Diatrypae, Botryosphaeria, Microdiplodia, Valsa) est indicateur plus contribu- Québec Canada E-U Boisé-des-Muir a Ascom ycetes =] Coelomycetes Hyphomycetes 0 5 10 15 Quantité de nouvelles espéces sur le Fagus Peet: 20 25 30 35 FIGURE 1. Niveau de connaissance de la biodiversité d’espéces des microchampignons du hétre en Amérique du Nord 37 4 E MICROFONGIQUE BIODIVERSIT VUJANOVIC ET BRISSON 2001 vooAWiosKZ, *7, ‘sajaaduoyddy — eyookwo1gnaqj :H{ { Sajaauoja0y7 — eyooKkwio1gnaq :— ‘ejooAwIoosy — vy ‘umnykyg — d : ION H ‘Sold (WUTZ) MUDIa] WNYINAAaA V ‘ds psjpA H ‘Iqv'S1og appiia DULAapOYy IIA V “I (1 VSiog) suaigup psjpA H TERY WNUDZIADY DULAAPOYII J, V ‘ds pisapydsojpwas J F ‘ds sndoziyy ®) ‘ds wntodsoao0og H ‘ds njjaippj2oucyy V INOIYD (‘s19g) DIDUAaJONb vispUsaJONG V ‘ds piodsoajg ®) ‘ds payovjopnas gq WS2 UMOIG H (‘W]OH) snsouvf saxtuopiaad ®) “ABLL (WIsaq) Dsuojgo sisdowoyd H ‘ds wnypouad @) ‘ds pwoyd “ADIN VY id (iy:epoy) viuapydsida piujoan =O (Iq:'Siag) pydiowdjod piupjKX D ‘Od 3 ‘UNS (SAA) XYI1Y DIUOSLapuayoan Ff ‘uasaly SNSOWAIDA AOINA OV yenog (14) visnap puynisy V "I (Iq ‘opo) punvgqnuuld v14jaaN H ‘ds wniydvasojday VY ds piuapydsojpuaa J V ‘ds visio H ‘ds piuosiapuayy so "SoG DUASI]IDS DIAJIAN ) ‘Usaq vauisvf[DIJaagiT VV ‘ds psjpA THA HE H ‘ds wnippjpoyy = ‘ds p1ujoan V (Tue) wnjouupu uoj fe) = @ = Z rs) @ Oo x 5 © j= Température horaire moyenne (°C) 01 23 45 6 7 8 9 10 114 12 13 14 15 16 17 18 19 20 21 22 23 lever du soleil coucher du soleil Période de non reproduction (nombre de passages moyen) Taux d'activité moyen @ Température horaire moyenne (°C) 0123 45 67 8 9 1011 12 13 14 15 16 17 18 19 20 21 22 23 a lever du soleil coucher du soleil FIGURE 2. Cycle d’activité journalier du li¢évre d’ Amérique et température moyens selon la saison de reproduction. 2001 25 20 15 Taux d’activité moyen an Aube Jour Crépuscule Nuit Période FiGuRE 3. Taux d’activité moyen du liévre d’ Amérique selon les périodes de la journée: |’aube (a), le jour (j), le crépuscule (c) et la nuit (n). Résultats d’une Anova de Conover 4 un facteur suivie d’un test de comparaisons multiples selon la méthode de Bon- ferroni (**p<0.01 et *p<0.05). mente pour atteindre son maximum juste avant le coucher du soleil. Niveaux d’activité selon la période de la journée Les résultats présentés a la figure 3 confirment les observations effectuées pour le cycle journalier d’activité. Les taux d’activité moyens a l’aube, au crépuscule et au milieu de la nuit sont plus élevés qu’au milieu de la journée. La nuit, le niveau d’activité modéré observé sur 24 heures se traduit par une moyenne significativement inférieure aux périodes de lever et de coucher du soleil, qui elles, ne présentent pas de différences entre elles. Les don- nées d’activité basées sur les séquences de 3 heures dans la journée semblent donc représentatives de P activité sur 24 heures puisqu’on observe les mémes tendances dans les deux échantillonnages. La prise de données par séquence nous a permis d’effectuer un plus grand nombre d’observations et d’avoir une idée plus précise de l’influence des paramétres cli- matiques. Influence des parametres climatiques Comme on peut le voir a la figure 4, les liévres présentent deux niveaux d’activité selon la tempéra- ture. A des températures inférieures a 5°C, les taux d’activité sont significativement plus élevés qu’a des températures plus chaudes. Cette tendance est égale- ment observée par rapport aux variations de tem- pérature dans chaque sous-période de 3 heures. Les individus sont, en effet, plus actifs lorsque la tem- pérature est en diminution plut6t qu’en augmenta- tion. Cependant, cette tendance est moins nette pour les valeurs extrémes. En ce qui concerne |’humidité relative, les indi- vidus semblent plus sensibles aux variations durant la sous-période (écart d’humidité relative), qu’a la THEAU ET FERRON: INFLUENCE SUR LES PATRONS D’ ACTIVITE DU LIEVRE 47 valeur absolue de ce paramétre. En effet, on observe une activité plus élevée a partir de la classe [-3 Of, c’est-a-dire lorsque |’humidité demeure relativement stable ou qu’elle augmente. Lorsqu’il y a baisse d’humidité relative, les individus semblent moins actifs. Cette tendance est moins nette lorsqu’on met en relation |’activité avec les valeurs ponctuelles d’humidité relative. Seules les valeurs extrémes présentent des différences significatives, les indi- vidus étant nettement moins actifs a tres faible humidité et plus actifs a tres haute humidité. L’écart de pression barométrique a |’intérieur d’une sous-période de trois heures semble également avoir un effet sur le niveau d’activité. Lorsque la pression est en diminution, les taux d’activité sont nettement plus faibles que lorsqu’elle augmente. Deux classes correspondant a des diminutions de pression présentent des différences significatives avec 4 classes correspondant a des augmentations de pression. Nos résultats ne mettent pas en évidence de dif- férences significatives entre les taux d’activité et la vitesse du vent. On observe cependant une tendance négative entre les deux. A V’échelle saisonniére, la figure 5 présente les relations entre les parametres climatiques et |’activ- ité. Au printemps et en été, l’axe 1 est créé par |’ op- position entre la température (T) et l’humidité (H), Vactivité étant fortement corrélée négativement avec le premier et positivement avec le second. La dif- férence de pression barométrique (DP) et la vitesse du vent (V) caractérisent également cet axe. L’axe 2, créé par l’opposition entre l’écart de température (DT) et d’humidité relative (DH) présente moins d’influence sur I’ activité pendant ces deux saisons. En automne et en hiver, |’axe 1 est créé par |’ oppo- sition entre l’écart de température (DT) et d’humidité relative (DH) qui présentent, respectivement, de fortes corrélations négative et positive avec I’ activité. La différence de pression barométrique s’ajoute a la formation de l’axe 1 en automne. Durant, cette derniére saison, la température (T) et l’-humidité (H) sont corrélées a l’activité avec une importance plus grande pour la température. Par contre, c’est en hiver que la situation est la plus différente. En effet, |’ activ- ité semble étre en relation plus grande avec les dif- férences de température (DT) et d’humidité relative (DH) qu’avec les autres facteurs. Discussion Patrons journaliers d’ activité Nos résultats montrent que le liévre d’ Amérique est principalement actif au crépuscule et a l’aube et que sa principale période d’activité est concentrée durant la nuit, entre ces deux moments de la journée et cela pendant toute l’année. Ce patron d’activité a déja été décrit chez cette espéce (Mech et al. 1966; Keith 1974), ainsi que chez d’autres lagomorphes comme le lapin de garenne (Villafuerte et al. 1993), 48 THE CANADIAN FIELD-NATURALIST Vol. 115 40 40 ire) 35 2s - 30 30 o oS 2 25 = 2g = 20 20 8 as x 15 = S 15 5 2 = w = = Ee o @ Es 10 e8 10 S = Ee 5 8 5 7 © Bb) bed bel be ke 0 nes 0 = beg ses UN ae ; Sp & “ss “Wo “ 08 fo Os 50 Na 3 Rag Bete Paige Bilt, %y ye % ST YDS % e idité relative (9 Température ( C) Humiditée relative (%) © > fa} £ a) 2 o & re) x< 3 fe “85 “$5 & O53. & <5 (0 25 e, s Ss ‘S, 0 jo-5] }§-10] >10 y Bey Be Oy OG We Ig ay Vitesse du vent (km/h) fe) Ecart de température dans la sous-période ( C) 25 © 4 aa So 5 2 = a, RES 2 P 3 = 8 b* 15 © ne) 5 10 © fe me he by, &9 fo 0 ,@ » @ lo Os Oe a As = br a M5 , 20,4 ig 385% Og ne Cr Tae Bi % Ree 29 7 Oe GY FH SH FY Ecart d'humidité relative dans la sous-période (%) Ecart de pression barométrique dans la sous-période (hPa) FIGURE 4. Taux d’activité du liévre d’ Amérique en fonction de différents facteurs climatiques. Résultats des Anova a un facteur avec l’approche de Conover et des tests de contraste selon la méthode de Bonferroni (**p<0.01 et * p<0.05). 2001 1.0 S S x 0.5 NI S & N ~N = H a 3g 0.0 ACTIVITE s e 2 = — row o 05 = & g PRINTEMPS -1.0 s -1.0 -0.5 0.0 0.5 1.0 -1.0 -0.5 Axe principal 1 (31,65 %) ACTIVITE Axe principal 2 (26,89 %) 0 -1.0 -0.5 0.0 0.5 1.0 Axe principal 1 (30,69 %) Axe principal 1 (34,93 %) THEAU ET FERRON: INFLUENCE SUR LES PATRONS D’ ACTIVITE DU LIEVRE 49 1.0 ° a ACTIVITE H Axe principal 2 (22,37 %) oO oO ACTIVITE AUTOMNE .O 1.0 -1.0 -0.5 0.0 0.5 1.0 Axe principal 1 (27,11 %) 0.0 0.5 T: Température H: Humidité relative V: Vitesse du vent DT: Ecart de température dans la sous-période DH: Ecart d’humidité relative dans la sous-période DP: Ecart de pression barométrique dans la sous-période (%): Pourcentage de variance expliquée FiGURE 5. Analyses en composantes principales saisonniéres des facteurs climatiques et des taux d’activité du liévre d’ Amérique. le li¢vre d’ Europe (Homolka 1986; Pépin et Cargnelutti 1994) et le liévre variable européen (Lemnell et Lindlof 1981). Nos résultats apportent cependant certaines nuances par rapport aux travaux effectués par Keith (1964) qui ont décrit l’ activité du liévre d’ Amérique selon une courbe symétrique sur 24 heures et dont le pic se situe 4 23 heures. Ces résultats provenaient de données prises pendant un été seulement. Nos résultats permettent également d’apporter des nuances aux travaux effectués par Foresman et Pearson (1999) qui ont décrit |’activité du livre comme étant exclusivement nocturne du- rant ’hiver. Nos observations, effectuées sur une année entiére, indiquent que |’activité du liévre est effectivement concentrée entre le coucher et le lever du soleil mais qu’elle est davantage manifeste a V’aube et au crépuscule, et cela durant toute |’ année. Le patron journalier d’activité d’un grand nombre d’espéces de proies est influencé par celui de leurs prédateurs (Curio 1976). Ces espéces synchronisent leur rythme d’activité inversement a celui de leurs prédateurs. Chez le lévre d’Amérique, un niveau d’activité accru au coucher et au lever du soleil per- mettrait de limiter les contacts avec les prédateurs diurnes et nocturnes qui cessent ou débutent leur période d’activité 4 ces moments de la journée. Le trés faible niveau d’activité observé pendant le jour serait également lié aux risques de prédation. Pendant cette période, les individus sont en effet plus vulnérables puisqu’ils sont exposés aux prédateurs chassant a vue. Les facteurs expliquant |’ observation de patrons d’activité a deux pics ne font cependant pas l’unanimité. Aschoff (1966) mentionne que méme si des stimuli environnementaux tels que l’aube ou le crépuscule peuvent accentuer ce type de patron, leur origine demeure cependant endogéne. Cet auteur se base sur des expérimentations effec- tuées sur de nombreuses espéces animales qui mon- trent que méme en |’absence de stimuli externes (ex : conditions de luminosité ou de température con- stantes), le patron circadien d’activité 4 deux pics reste présent jusqu’a une période d’un an chez une espéce de pinson (Fringilla coelebs). L’ observation de ce type de patron chez un grand nombre d’espéces est donc probablement lié a la combinai- son de facteurs externes et endogénes. Méme si ce patron a deux pics reste constant tout au long de l’année chez le liévre, nos résultats montrent que selon les saisons, certaines modifications du niveau d’ activité se produisent et semblent, en partie, reliées a la température. En effet, en superposant la température moyenne horaire aux taux d’activité, il apparait qu’en période de non reproduction, les individus favorisent les heures les plus chaudes de la journée. Lors de cette période, qui correspond aux mois froids de l’année, 50 THE CANADIAN FIELD-NATURALIST les températures les plus hautes sont observées au coucher du soleil et correspondent aux taux d’ activité les plus élevés de la journée. En réduisant leur niveau d’ activité lors de la partie la plus froide de la journée, les individus limiteraient leurs pertes d’énergie. Inversement, lors de la période de reproduction, qui correspond au printemps et a 1’été, les individus sem- blent éviter les périodes les plus chaudes de la journée puisqu’ils sont principalement actifs autour du lever du soleil. [ls limiteraient ainsi leurs pertes d’eau par évaporation qui peuvent étre importantes lorsque les températures sont élevées (Hart et al. 1965). Villafuerte et al. (1993) ont également observé ces mémes changements de niveaux d’activité entre les saisons a l’intérieur de la journée chez le lapin de garenne. Le harcélement par les insectes, au print- emps et en été, pourrait également expliquer une activité accrue lors des périodes les plus froides de la journée puisque le dérangement est plus important lorsqu’il fait chaud. Cet effet a été documenté chez le caribou (Rangifer tarandus). Morschel et Klein (1997) ont en effet montré que la présence d’insectes a des températures élevées affecte les caribous en causant une augmentation du temps passé debout et une diminution de |’ alimentation. Impact des facteurs climatiques Le liévre d’ Amérique adapte ses niveaux d’activ- ité entre les saisons et il bénéficie d’un pelage vari- able au cours de l’année qui lui assure une isolation thermique relativement efficace (Hart et al. 1965). Cependant, il emmagasine trés peu de réserves énergétiques et doit se nourrir réguli¢rement sur de courtes périodes. Malgré ces adaptations, il doit donc S’exposer aux conditions climatiques rigoureuses et adopter des stratégies d’activité a plus court terme lui permettant de maintenir sa balance énergétique. Ces stratégies se reflétent dans nos résultats par une activité plus élevée lorsque les températures sont basses ou en diminution et lorsque I’humidité relative est élevée ou en augmentation. Il est difficile de dis- socier l’effet de la température et de ’humidité car ces deux facteurs sont fortement corrélés, toute l’année. Cependant, au printemps et en été, les liévres semblent plus sensibles aux niveaux d’humidité et de température, alors que durant |’automne et I’hiver, ce sont les variations de ces facteurs a l’intérieur des périodes qui occupent un grand rdéle dans les patrons d’activité observés. Sur toute l’année, lors de condi- tions froides ou de refroidissement, nos résultats d’ activité refléteraient une augmentation de |’ alimen- tation puisqu’une étude connexe (Théau et Ferron 2000) a montré une relation inverse entre la fréquence d’alimentation et l’écart et le niveau de température. Dans cette étude, le comportement d’ ali- mentation était trés fortement relié au déplacement, qui constitue notre indicateur d’ activité. Les besoins énergétiques accrus lors de périodes froides ou de refroidissements rapides peuvent expliquer ces Volainis patrons d’activité. De fagon générale, les tendances d’activité observées par rapport aux autres facteurs climatiques supportent également les résultats obtenus par Théau et Ferron (2000), en ce qui con- cerne les comportements d’alimentation et de déplacement. Les li¢vres semblent notamment réagir aux chutes de pression, qui accompagnent le mauvais temps, en diminuant leur niveau d’activité et en laugmentant lorsque la pression barométrique est en augmentation. Il est cependant difficile d’isoler et d’interpréter l’impact de la pression car elle est cor- rélée avec de nombreux autres facteurs (Marten 1973) et méme si son effet a été mis en évidence chez d’autres espéces (Ables 1969; Marten 1973; Beltran et Delibes 1994), les mécanismes physiologiques reliés a Sa perception et a son mode d’action sur le comportement ne sont pas connus. Seule la vitesse du vent ne présente pas d’influence significative sur V’activité, bien que la tendance observée montre un impact négatif. Comme la vitesse du vent réduit la fréquence des comportements de toilettage, de déplacement et d’alimentation du liévre (Théau et Ferron 2000), ceci suggére également que ce facteur climatique influence négativement l’activité en général. Les variations saisonniéres de l’influence de ce facteur pourraient cependant expliquer |’ absence de relation significative sur une base annuelle. En effet, contrairement a la situation rencontrée au prin- temps, en automne et en hiver, |’activité semble favorisée par temps venteux, en été. La diminution du harcélement causé par les insectes pourrait expliquer ce patron d’ activité. Nos résultats confirment donc que les liévres sont principalement actifs a l’aurore et au crépuscule et qu’ils sont nocturnes. De plus, ils adaptent leurs niveaux d’activité au cours de l’année en réagissant différemment aux facteurs climatiques entre les saisons, afin notamment de bénéficier des conditions de température journaliére favorables. A 1’ échelle journaliére, ils réagissent également a trés court terme aux changements de conditions climatiques en modifiant leurs niveaux d’activité afin de limiter leurs pertes énergétiques. Remerciements Nous remercions Véronique St-Louis, Mathieu Coté, Marie-Claude Richer, Pierre Etcheverry, Francis Bélisle, Yves Lemay et Jean-Marc Duguay pour leur aide sur le terrain ainsi que Pierre Bélanger pour ses conseils en informatique et Alain Caron pour son aide dans le traitement statistique des données. Nous remercions également M. Proulx pour avoir autorisé la construction de l’enclos sur son terrain et pour en avoir facilité l’accés tout au long de l’étude. Nous sommes également recon- naissants envers et les correcteurs anonymes qui ont commenté des versions antérieures du manuscrit ainsi qu’envers Jean-Yves Gautier, a la mémoire de qui nous dédrons cet article. 2001 Littérature citée Ables, E. D. 1969. Activity studies of red foxes in south- ern Wisconsin. Journal of Wildlife Management 33: 145-153. Alain, G. 1986. Plan tactique : le li¢vre d’ Amérique. Mini- stére du Loisir, de la Chasse et de la Péche, Direction générale de la faune, Québec. 40 pages. Aschoff, J. 1966. Circadian activity pattern with two peaks. Ecology 47: 657-662. Beltran, J. F., et M. Delibes. 1994. Environmental deter- minants of circadian activity of free-ranging Iberian Lynxes. Journal of Mammalogy 75: 382-393. Bider, J.R. 1961. An ecological study of the hare Lepus americanus. Canadian Journal of Zoology 39: 81—103. Colgan, P. W. 1978. Quantitative Ethology. John Wiley & Sons, New York. 364 pages. Conover W. J. 1980. Practical nonparametric statistics. 2 edition. John Wiley and Sons, Toronto. Curio, E. 1976. The ethology of predation. Springer- Verlag, New-York. 250 pages. Drew, G.S., et J. A. Bissonette. 1996. Winter activity pat- terns of American martens (Martes americana) rejection of the hypothesis of thermal-cost minimization. Canadian Journal of Zoology 75: 812-816. Ferron, J. 1993. How do population density and food supply influence social behaviour in the snowshoe hare (Lepus americanus)?. Canadian Journal of Zoology 71: 1084-1089. Foresman, K.R., et D.E. Pearson. 1999. Activity pat- terns of American martens, Martes americana, snowshoe hares, Lepus americanus, and red squirrels, Tamiasciurus hudsonicus, in Westcentral Montana. Canadian Field- Naturalist 113: 386-389. Gilbert, B. S., et S. Boutin. 1991. Effect of moonlight on winter activity of snowshoe hares. Arctic and Alpine Research 23: 61-65. Gray, D.R. 1993. Behavioural adaptations to arctic win- ter : shelter seeking by arctic hare (Lepus arcticus). Arctic 46; 340-353. Hart, J.S., H. Pohl, et J.S. Tener. 1965. Seasonal acclimatization in varying hare (Lepus americanus). Canadian Journal of Zoology 43: 731-744. Homolka, M. 1986. Daily activity pattern of the European hare (Lepus europaeus). Folia Zoologica 35: 33-42. Keith, L. B. 1964. Daily activity pattern of snowshoe hares. Journal of Mammalogy 45: 626-627. Keith, L. B. 1974. Le liévre d’ Amérique. Environnement Canada Service de la faune, Ottawa. 4 pages. THEAU ET FERRON: INFLUENCE SUR LES PATRONS D’ ACTIVITE DU LIEVRE 51 Kline, P.D. 1965. Factors influencing roadside counts of cottontails. Journal of Wildlife Management 29: 665-671. Lemnell, P. A., et B. Lindléf. 1981. Diurnal and seasonal activity pattern in the moutain hare. Pages 349-356 in World lagomorph conference. Edited by K. Myers et C. D. McInnes, Guelph, Ontario. Marten, G.G. 1973. Time patterns of Peromyscus activi- ty and their correlations with weather. Journal of Mam- malogy 54: 169-187. Matuszewski, G. 1981. Circadian activity of European hares in spring, on the kampinos forest border. Pages 357-365 in World lagomorph conference. Edited by K. Myers et C. D. McInnes, Guelph, Ontario. Mech, L.D., K. L. Heezen, et D. B. Siniff. 1966. Onset and cessation of activity in cottontail rabbits and snow- shoe hares in relation to sunset and sunrise. Animal Behavior 14: 410-413. Moreno, S., R. Villafuerte, et M. Delibes. 1996. Cover is safe during the day but dangerous at night: the use of vegetation by European wild rabbits. Canadian Journal of Zoology 74: 1656-1660. Morschel, F. M., et D. R. Klein. 1997. Effects of weather and parasitic insects on behaviour and group dynamics of caribou of the Delta herd, Alaska. Canadian Journal of Zoology 75: 1659-1670. Pépin, D., et B. Cargnelutti. 1994. Individual variations of daily activity patterns in radiotracked European hares during winter. Acta Theriologica 39: 399-409. Seabloom, R. W., R. G. Schwab, et S. C. Loeb. 1994. Seasonal activity and feeding patterns in a population of deer mice (Peromyscus maniculatus gambelii). Canadian Journal of Zoology 72: 2238-2243. Théau, J., et J. Ferron. 2000. Influence des conditions climatiques sur le comportement du liévre d’ Amérique (Lepus americanus) en semi-liberté. Canadian Journal of Zoology 78: 1126-1136. Vickery, W.L., et J. R. Bider. 1981. The influence of weather on rodent activity. Journal of Mammalogy 62: 140-145. Villafuerte, R., M. B. Kufner, M. Delibes, et S. Moreno. 1993. Environmental factors influencing the seasonal daily activity of the European rabbit (Oryctolagus cuni- culus) in a Mediterranean area. Mammalia 57: 341-347. Wilkinson, L. 1996. SYSTAT 6.0 for Windows. SPSS Inc. Chicago, Illinois. Received 29 December 1999 Accepted | June 2000 Fall Food Habits and Reproductive Condition of Fishers, Martes pennanti, in Vermont KYLE R. VAN Why and WILLIAM M. GIULIANO! Wildlife Biology Program, Department of Biological and Environmental Sciences, California University of Pennsylvania, California, Pennsylvania 15419, USA ‘Present address: Louis Calder Center — Biological Field Station, Fordham University, 53 Whippoorwill Road., P.O. Box K, Armonk, New York 10504, USA Van Why, Kyle R., and William M. Giuliano. 2001. Fall food habits and reproductive condition of Fishers, Martes pennan- ti, in Vermont. Canadian Field-Naturalist 115(1): 52-56. We examined 197 Fishers harvested in Vermont during December 1996, to determine (1) differences in diet among Fishers of different age, sex, mass, fat reserves, and female reproductive condition, and (2) differences in female reproductive con- dition among Fishers of different age, mass, and fat reserves. Fishers had a diverse diet, consuming 31 different items. Most of the diet was mammalian (72%), with avian prey (15%) and fruit (10%) of secondary importance. Diets were simi- lar (P > 0.10; Morisita’s index >0.80) among Fishers of different ages Guvenile, yearling, adult), sex, body mass (< 2, 2-3.5, >3.5 kg), and fat reserves. Reproductive condition (i.e., number of corpora lutea) of females increased with age (P = 0.001), but did not depend on level of fat reserves (P = 0.297), body mass (P = 0.719), or diet (P = 0.665). Key Words: Fisher, Martes pennanti, age, diet, fat reserves, reproductive condition, sex, Vermont. Fishers (Martes pennanti) are secretive, primarily from trappers by Vermont Department of Fish and carnivorous mammals long prized for their pelage Wildlife personnel during the 1996 trapping season (Douglas and Strickland 1987). As a predator, Fishers (6-21 December). Animals were harvested through- can play an important role in community dynamics out the state. Each carcass was weighed, sexed, and (Douglas and Strickland 1987; Nudds 1987; Powell and _ assigned an ocular visceral-fat rank, as an index to the Zielinski 1994), and may be indicators of community level of fat reserves (1 = no fat to 4 = large fat health in mature forest systems (Powell and Zielinski deposits; Giuliano et al. 1989). We determined age by 1994). Historically, Fisher populations have fluctuated examining cementum annuli of a lower canine tooth in response to a variety of factors, including changing (Johnston et al. 1987). To index reproductive condi- furbearer harvest regulations, fur prices, and land-use tion of females, we removed the right ovary, pre- practices such as timber harvesting (Allen 1987; Novak served it in formalin (10%) for 24 hours, and then 1987; Obbard et al. 1987; Powell and Zielinski 1994). transferred it to ETOH (10%) until processing. Ten As fur harvest regulations become more restrictive and _ cross-sections of each ovary were mounted on slides fur prices decline, changes in land-use may be most — to determine the number of corpora lutea, which we important to Fisher population dynamics. These — used as an index to reproductive condition (Douglas changes in habitat may negatively impact Fisher repro- and Strickland 1987). ; duction and survival, by altering cover and prey avail- We removed stomachs and intestines from each ability (Douglas and Strickland 1987). carcass and rinsed contents through a 0.5 mm sieve. Little is known about Fishers in Vermont, where Prey remains such as bones, hair, seeds, etc. were farm abandonment, timber harvest, and urbanization identified macroscopically by comparison to refer- are changing the landscape. Thus, increased knowl- ence collections. We identified trace amounts of hair edge of Fisher ecology is important to its conserva- to species by microscopic identification of guard hair tion. Our purpose was to better understand Fisher food pigmentation and comparison to reference specimens habits and reproduction in Vermont to improve the (Stains 1958). scientific basis for conservation of the species. Speci- We compared body mass between sexes and fically, we determined differences in diet among among ages (juvenile [<1 yr], yearling [1 yr], adult Fishers of different age, sex, mass, level of fat [22 yrs]) by analysis of variance (ANOVA) and, reserves, and female reproductive condition, and when necessary, a Fisher’s LSD test. Differences in examined differences in female reproductive condi- number of corpora lutea were compared among ages tion among Fishers of different age, mass, and physi- using an ANOVA followed by a Fisher’s LSD test. cal condition in December. We compared body mass between females with cor- pora lutea to those without using ANOVA. Methods Differences in the level of fat reserves were com- We obtained 197 skinned Fisher carcasses collected pared between sexes, adult females with corpora 52 2001 lutea to those without, and among ages using a con- tingency table and log-likelihood ratio statistic (G). We used Morisita’s index (C,; Morisita 1959) to examine dietary overlap between sexes, ages, body mass classes (<2, 2—3.5, > 3.5 kg), fat reserve class- es (1 = no fat to 4 = large fat deposits), and female reproductive condition classes (adult females with and without corpora lutea). Morisita’s index varies from 0 (no overlap) to | (identical diets). To examine the use of major food groups and dif- ferent-sized food items, we divided foods into several categories. Mammalian prey was partitioned into size classes: small (Peromyscus spp., Short-tailed Shrew [Blarina brevicauda], Sorex spp., Star-nosed Mole [Condylura cristata]), medium (Rattus spp., Gray Squirrel [Sciurus carolinensis], Red Squirrel [Tami- asciurus hudsonicus], Eastern Chipmunk [Tamias striatus]), large (Virginia Opossum [Didelphis virgini- ana|, Mink [Mustela vison], Muskrat [Ondatra zibethicus|, Snowshoe Hare [Lepus americanus], and Sylvilagus spp.), and very large (Raccoon [Procyon lotor|, Striped Skunk [Mephitis mephitis], Beaver [Castor canadensis], Porcupine [Erethizon dorsat- um]). White-tailed Deer (Odocoileus virginianus) was a separate category of mammalian prey because the majority of its consumption was carrion or bait (Coulter 1966; Powell 1993; Giuliano et al. 1989). Avian prey, which consisted primarily of passerines (Passeriformes) and Ruffed Grouse (Bonasa umbellus), were pooled. Fruits, primarily apple (Malus spp.), were pooled. We used a contingency table and log-likelihood ratio statistic (G) to examine differences in occurrence of major food groups among Fishers of different age, sex, mass, fat reserves, and reproductive condition. For all ANOVA, we confirmed homogeneous variances using Levene’s test, and normality by plot- ting data against the expected normal distribution. All tests were performed using SYSTAT software (SYSTAT 1992), and considered statistically signifi- cant at P< 0.10. This value was used, rather than the more common P < 0.05, to reduce the likelihood of making a Type II error (Zar 1999). VAN WHY AND GIULIANO: FISHERS IN VERMONT 53 Results Male Fishers were heavier (P < 0.001) and had greater fat reserves than females (P < 0.001; Table 1). Adults were heavier than juveniles among males (P = 0.001), but not among females (P = 0.134; Table 1). Level of fat reserves was similar among differ- ent-aged males (P= 0.607) and females (P = 0.118; Table 1). Corpora lutea counts ranged from 0-3, and were present only in >1 year old females (P < 0.001; Table 1). Mean count for females containing >1 cor- porus luteum was 1.83 + 0.31 (x + SE). Adult females with corpora lutea had similar fat reserves (2.17 + 0.14; n = 6) to those without corpora lutea (2.65 + 0.23; n = 19; P = 0.297). Adult females with corpora lutea (1.90 + 0.04 kg) weighed the same as those without (1.83 + 0.03 kg; P = 0.719). Vermont Fishers consumed 31 different food items. Mammalian food items accounted for 72% of the diet, with 15% avian, and 10% fruit. There were eight food items which were of major importance (>5% occurrence) to either male or female Fishers. Less commonly consumed foods included: Virginia Opossums (Didelphis virginiana), Beaver (Castor canadensis; assumed to be bait), Mink (Mustela vison), moles (Condylura cristata and Parascalops breweri), voles (Clethrionomys gapperi and Micro- tus pennsylvanicus), Woodland Jumping Mice (Napaeozapus insignis), lagomorphs (Sylvilagus spp. and Lepus americanus), Rattus spp., Raccoons (Procyon lotor), Striped Skunks (Mephitis mephi- tis), invertebrates, fish, reptiles, and amphibians (Table 2). Diets overlapped considerably between sexes, ages, body mass classes, and fat reserve classes (C, > 0.90), with deer, small mammals, birds, and fruit being the most common food groups (Tables 2, 3, and 4). Use of major food groups did not differ between sexes (P = 0.883), ages (P = 0.836), body mass classes (P = 0.640; Table 3), or fat reserve classes (P = 0.153). Diets overlapped considerably (Table 4), and use of major food groups was similar (P = 0.121) between adult females with and without corpora lutea. TABLE 1. Body mass (kg), fat reserves, and reproductive condition (i.e., number of corpora lutea) of Fishers harvested in Vermont, December 1996 (x + SE). Male Age n Body Mass ___ Fat Reserves™ Juvenile 43 3.18+ 0.06 SZ = O12 Yearling St -3.82'20:07° 5.00 + 0.13? Adult ad 427 2007" Sts 2OA2* All 105 3.76 + 0.07 3.10 + 0.07 “number of Fishers sampled. “ocular visceral-fat rank: 1 no fat - 4 large fat deposits. “corpora lutea counted in right ovary. Female n Body Mass Fat Reserves’ Corpora Lutea”™™ 34 L772 0.07% 2.64 + 0.12? 0+0* 14 L772 O01 2.19 0.19" 0+ 0 25 1.97 + 0.08? 22 OLS" 0.48 + 0.19° The: 1.84+0.15 2.43 + 0.09 0.13 + 0.06 abewithin a column, values followed by the same letter are similar (P > 0.10). 54 THE CANADIAN FIELD-NATURALIST Vol. 115 TABLE 2. Percent occurrence of food items in Fishers harvested in Vermont, December 1996. Males Females Both Food Item (a=) (S32) (n = 197) White-tailed Deer (Odocoileus virginianus) 18.1 15.6 Ga Avian (mostly passerines) 13:8 15.6 14.6 Peromyscus spp. 9.6 13:3 Hil Apple (Malus spp.) 4.8 9.4 On Muskrats (Ondatra zibethicus) 6.4 6.3 6.3 Shrews (Blarina brevicauda and Sorex spp.) 6.9 a9) Bi Squirrels (Sciurus carolinensis, Tamiasciurus hudsonicus, and Tamias striatus) 5.8 47 pal Porcupines (Erethizon dorsatum) 53 3.9 4.7 Virginia Oppossums (Didelphis virginiana) Sal Dip} 32 Voles (Clethrionomys gapperi and Microtus pennsylvanicus) Ded Sali 2.8 Lagomorphs (Lepus americanus and Sylvilagus spp.) Sih 1.6 2.8 Fish 1.6 1.6 1.6 Beavers (Castor canadensis) 1.6 OF, les Mink (Mustela vison) ial 1.6 1.3 Invertebrates 0.5 1.6 0.9 Moles (Condylura cristata and Parascalops breweri) 0.5 0.8 0.6 Woodland Jumping Mice (Napaeozapus insignis) 05 0.8 0.6 Reptiles and Amphibians ill 0.0 0.6 Raccoons (Procyon lotor) 0.5 0.0 0.3 Striped Skunks (Mephitis mephitis) 0.0 0.8 0.3 Rattus spp. 0.0 0.8 0.3 Discussion Powell (1993), and Raine (1987), who found We found Fishers to have a diverse diet with the most utilized dietary items during late fall being White-tailed Deer, birds, and small mammals. These findings are similar to those of other studies of Fisher food habits in eastern North America (see Douglas and Strickland 1987; Martin 1994; Powell and Zielinski 1994 for reviews) that found Fishers to be opportunistic feeders. Consumption of deer was probably the result of scavenging or use of baits, and appears to be an important dietary component of Fishers in northeastern North America (Coulter 1966; Kelly 1977; Rego 1984; Giuliano et al. 1989). However, unlike Coulter (1966), Leonard (1980), Snowshoe Hare to be a frequently used food, we found Snowshoe Hare to contribute little to Fisher diets. Further, in contrast to Fishers in Manitoba (Raine 1987), our results suggest that birds and small mammals are important prey, as did numerous other studies (see Douglas and Strickland 1987; Martin 1994; Powell and Zielinski 1994 for reviews). Where their ranges overlap, Porcupines are often a major prey of Fishers, composing as much as 34% of the diet in some studies (deVos 1952; Rego 1984; Powell 1993). Although Porcupines were consumed in Vermont, they were consumed less than reported levels. The differences in findings among studies TABLE 3. Percent occurrence of major food groups in Fishers harvested in Vermont, December 1996. Juvenile Food Item (Gave 7/7) White-tailed Deer (Odocoileus virginianus) 24.7 Small mammal? 48.1 Medium mammal? 6.5 Large mammal* 23.4 Very large mammal? ey Fruit (mostly apple [Malus spp.]) 14.3 Avian (mostly passerines) OM hs) Fisher Age Fisher Body Mass (kg) Yearling Adult <2 2-3.5 > 3D (n = 44) (M = 59) (n'='65))- @=56) Gi) PH 8) 30.9 15.4 15.4 © 19.4 47.7 41.8 39.0 28.2 SS) 13.6 Onl 6.3 SI 4.6 18.2 18.2 11.4 12.0 8.3 4.6 i227 4.6 6.8 eo 11.4 2.0 8.0 15.4 7.4 9 PIS) 18.2 12.8 139 aPeromyscus spp., Blarina brevicauda, Sorex spp., and Condylura cristata. Rattus spp., Sciurus carolensis, Tamiasciurus hudsonicus, and Tamias striatus. ‘°Didelphis virginiana, Ondatra zibethicus, Lepus americanus, Sylvilagus spp., and Mustela vison. ¢Procyon lotor, Mephitis mephitis, Castor canadensis, and Erethizon dorsatum. 2001 TABLE 4. Dietary overlap (C,) of Fishers of different sex, age, body mass, fat reserves, and between females with and without copora lutea, harvested in Vermont, December 1996. eh Male vs. Female 1.00 Juvenile vs. Yearling 0.97 Juvenile vs. Adult 0.98 Yearling vs. Adult 0.97 Body Mass: <2 kg vs. 2-3.5 kg 0.98 Body Mass: <2 kg vs. >3.5 kg 0.93 Body Mass: 2-3.5 kg vs. >3.5 kg 1.00 Fat Reserve Class*™* 1 vs. 2 0.92 Fat Reserve Class | vs. 3 0.95 Fat Reserve Class 1 vs. 4 0.98 Fat Reserve Class 2 vs. 3 0.98 Fat Reserve Class 2 vs. 4 0.96 Fat Reserve Class 3 vs. 4 0.99 Females With Corpora Lutea vs. Females Without Corpora Lutea 1.00 “Morisita’s index (C,) varies from 0 (no overlap) to | (iden- tical diets) “Ocular visceral-fat rank: 1 no fat — 4 large fat deposits. were likely the result of differences in prey species availability, diversity, and quality (1.e., caloric and nutrient value) among regions and periods (Leonard 1980; Raine 1987; Giuliano et al. 1989; Thompson and Colgan 1990). However, data to address these possibilities were not available. Our finding that male and female Fishers had highly similar diets was similar to that of Coulter (1966), Kelly (1977), Giuliano et al. (1989), and Powell (1993), but contrary to the general prediction of Brown and Lasiewski (1972; supported by Powell and Leonard 1983) who suggested that mustelids evolved intersexual differences in body size to reduce intraspecific competition for food. Use of dif- ferent foods and foods of different size by Fishers of different ages and size also could reduce intraspecif- ic competition. However, we found that fall diets did not differ by Fisher age or size. In nearby New Hampshire, Giuliano et al. (1989) found that large males consumed more White-tailed Deer and apples than did females. They suggested this was because males used larger home-ranges than females, increasing their encounters with these foods, and because males denied smaller females access to these concentrated foods. However, neither our data nor that of Kelly (1977) support this conclusion. Only adult (22 yrs) females had corpora lutea, which probably reflects increased fertility of Fishers with age (Leonard 1986). Females with corpora lutea had fewer (mean = 1.8) than have been found for females in other studies (range 2.7—3.7; Eadie and Hamilton 1958; Wright and Coulter 1967; Shea et al. 1985; Leonard 1986). Other studies have also docu- mented females with corpora lutea as yearlings VAN WHY AND GIULIANO: FISHERS IN VERMONT 55 (Eadie and Hamilton 1958; Shea et al. 1985; Leonard 1986). The reason for lower corpora lutea counts in Vermont Fishers is not clear. Adult Fishers with and without corpora lutea had the same level of fat, suggesting that condition was not a limiting fac- tor. Further, we found that reproductive condition was not a function of diet or body mass among adults during late fall. It is possible that during late winter, when food resources are presumably more limiting, and during spring, when the energy demands of reproduction increase dramatically (Powell and Leonard 1983), that female Fisher reproductive condition and fat reserves may be influ- enced by diet. Land management practices that promote a diver- sity of habitats, and thus prey, should be advocated as a conservation strategy for species such as Fisher, capable of utilizing a variety of prey. Of particular concern are anti-logging initiatives and the suppres- sion of disturbance factors (e.g., fire, pest outbreaks) that reduce the availability of early successional habitats important to numerous prey (e.g., deer, lago- morphs, grouse, and many small mammals). Fishers in Vermont and elsewhere exhibit age-specific fecundity, with older animals having greater fecundi- ty. Thus, if older females are more vulnerable to har- vest or otherwise underrepresented in the population, management to reverse this trend may be needed. Acknowledgments This research was supported by the Vermont Department of Fish and Wildlife and the Wildlife Biology Program, Department of Biological and Environmental Sciences, California University of Pennsylvania. We thank K. Royers for help with car- cass collection, E. Zuchelkowski for assistance with ovary preparation, and C. Elliott, C. A. Miller, G. Roloff, and J. Tirpak for editorial comments. Literature Cited Allen, A. W. 1987. The relationship between habitat and furbearer management. Pages 164-179 in Wild furbearer management and conservation in North America. Edited by M. Novak, J. A. Baker, M. E. Obbard, and B. Malloch. Ministry of Natural Resources, Ontario. 1150 pages. Brown, J.H., and R. C. Lasiewski. 1972. Metabolism of weasels: the cost of being long and thin. Ecology 53: 939-943. Coulter, M.W. 1966. Ecology and management of fish- ers in Maine. Ph.D. dissertation, Syracuse University, Syracuse. 183 pages. deVos, A. 1952. Ecology and management of fisher and marten in Ontario. Technical Bulletin, Ontario Depart- ment of Lands and Forests. 90 pages. Douglas, M. A., and M. A. Strickland. 1987. Fisher. Pages 510-529 in Wild furbearer management and con- servation in North America. Edited by M. Novak, J. A. Baker, M. E. Obbard, and B. Malloch. Ministry of Natural Resources, Ontario. 1150 pages. 56 THE CANADIAN FIELD-NATURALIST Eadie, R. W., and W. J. Hamilton Jr. 1958. Repro- duction in the fisher in New York. New York Fish and Game Journal 5: 77-83. Giuliano, W. M., J. A. Litvaitis, and C. L. Stevens. 1989. Prey selection in relation to sexual dimorphism of fish- ers (Martes pennanti) in New Hampshire. Journal of Mammalogy 70: 639-641. Johnston, D. H., D.G. Joachim, P. Bachmann, K. V. Kardong, R. E. A. Stewart, L.M. Dix, M. A. Strick- land, and I.D. Watt. 1987. Aging furbearers using tooth structure and biomarkers. Pages 228-243 in Wild furbearer management and conservation in North Amer- ica. Edited by M. Novak, J. A. Baker, M. E. Obbard, and B. Malloch. Ministry of Natural Resources, Ontario. 1150 pages. Kelly, G.M. 1977. Fisher (Martes pennanti) biology in the White Mountain National Forest and adjacent areas. Ph.D. dissertation, University of Massachusetts, Am- herst. 178 pages. Leonard, R.D. 1980. Winter activity and movements, winter diet and breeding biology of fisher (Martes pen- nanti) in southeastern Manitoba. M.S. thesis, University of Manitoba, Winnipeg. 181 pages. Leonard, R. D. 1986. Aspects of reproduction of the fish- er, Martes pennanti, in Manitoba. Canadian Field- Naturalist 100: 32-44. Martin, S.K. 1994. Feeding ecology of American martens and fishers. Pages 297-315 in Martens, sables, and fishers: biology and conservation. Edited by S. W. Buskirk, A. Harestad, and M. Raphael. Cornell Univer- sity Press, Ithaca. 512 pages. Morisita, M. 1959. Measuring of the dispersion of indi- viduals and similarity between communities. Memorial Faculty of Science, Kyushu University, Series E (Biol- ogy) 3: 65-80. Novak, M. 1987. The future of trapping. Pages 89-100 in Wild furbearer management and conservation in North America. Edited by M. Novak, J. A. Baker, M. E. Obbard, and B. Malloch. Ministry of Natural Resources, Ontario. 1150 pages. Nudds, T. 1987. The prudent predator: applying ecology and anthropology to renewable resource management. Pages 113-118 in Wild furbearer management and con- servation in North America. Edited by M. Novak, J. A. Baker, M. E. Obbard, and B. Malloch. Ministry of Natural Resources, Ontario. 1150 pages. Vol. 115 Obbard, M. E., J. G. Jones, R. Newman, A. Booth, A. J. Satterthwaite, and G. Linscombe. 1987. Furbearer harvest in North America. Pages 1007-1038 in Wild furbearer management and conservation in North America. Edited by M. Novak, J. A. Baker, M. E. Obbard, and B. Malloch. Ministry of Natural Resources, Ontario. 1150 pages. Powell, R. A. 1993. The fisher: life history, ecology, and behavior. Second Edition. University of Minnesota Press, Minneapolis. 237 pages. Powell, R. A., and R. D. Leonard. 1983. Sexual dimor- phism and energy expenditure for reproduction in female fisher Martes pennanti. Oikos 40: 166-174. Powell, R. A., and W. J. Zielinski. 1994. Fisher. Pages 38-73 in The scientific basis for conserving forest carni- vores: American marten, fisher, lynx, and wolverine in the western United States. Edited by L. F. Ruggiero, K. B. Aubry, S. W. Buskirk, L. J. Lyon, and W. J. Zielinski. USDA Forest Service General Technical Report RM 254. Raine, R. M. 1987. Winter food habits and foraging behaviour of fishers (Martes pennanti) and martens (Martes americana) in southeastern Manitoba. Canadian Journal of Zoology 65: 745-747. Rego, P. W. 1984. Factors influencing harvest levels of fisher in southcentral and southeastern Maine. M.S. the- sis, University of Maine, Orono. 54 pages. Shea, M.E., N. L. Rollins, R. T. Bowyer, and A. G. Clark. 1985. Corpora lutea number as related to fisher age and distribution in Maine. Journal of Wildlife Management 49: 37-40. Stains, H. J. 1958. Field key to guard hairs of middle western furbearers. Journal of Wildlife Management 22: 95-97. SYSTAT. 1992. SYSTAT for Windows. SYSTAT Inc., Evanston, Illinois. 750 pages. Thompson, I. D., and P. W. Colgan. 1990. Prey choice by marten during a decline in prey abundance. Oeco- logia 83: 443-451. Wright, P. L., and M. W. Coulter. 1967. Reproduction and growth in Maine fishers. Journal of Wildlife Man- agement 31: 70-87. Zar, J. H. 1999. Biostatistical analysis. Fourth edition. Prentice Hall, Upper Saddle River. 663 pages. Received 18 October 1999 Accepted 16 November 2000 Relative Abundances of Forest Birds of Prey in Western Newfoundland JoHN W. Gosse!, and WILLIAM A. MONTEVECCHI?? 'Terra Nova National Park, Glovertown, Newfoundland, AOG 2L0, Canada *Biopsychology Programme, Psychology and Biology Departments, Memorial University of Newfoundland, St. John’s, Newfoundland, A1B 3X9, Canada Gosse, John W., and William A. Montevecchi. 2001. Relative abundances of forest birds of prey in western Newfound- land. Canadian Field-Naturalist 115(1): 57-63. We surveyed woodland birds of prey in different-aged Balsam Fir (Abies balsamea) forests in western Newfoundland. Nine species of birds of prey were recorded: Merlin, American Kestrel, Osprey, Rough-legged Hawk, Sharp-shinned Hawk, Northern Goshawk, Boreal Owl, Great Horned Owl and Northern Hawk-Owl. Playback broadcasts generated low response rates but did facilitate most detections of nocturnal owls and rarer birds of prey. Numbers of birds of prey were low but were highest and most diverse in old-growth forests. Clearcuts provided habitat for open-nesting raptors (hawk- owls, kestrels and in one year Rough-legged Hawks). Owing to rarity of woodland raptors in western Newfoundland, we recommend that large expanses of late-successional forest be set aside for the conservation of birds of prey and other wildlife. Key Words: Birds of prey, Newfoundland, broadcast surveys, habitat associations, relative abundance. Extensive research describing habitat associations ranges of many North American woodland raptors. of woodland birds of prey has been conducted in The requirement for research on forest birds of prey North America and Europe (Titus and Mosher 1981; in this region is necessary since this landscape has Armstrong and Euler 1982; Reynolds et al. 1982; been drastically altered by forest harvesting resulting Hayward et al. 1993; Solonen 1994). Though patterns in extensive fragmentation and a shift in the age- of habitat use have emerged, results are often site- class distribution of the forest. Habitat loss from tim- specific and vary greatly across the geographic _ ber harvesting is recognized as a serious threat to ranges of some species. Breeding densities of birds of | populations of some woodland raptors (Armstrong prey are generally assumed to be limited by either an and Euler 1982; Carey et al. 1990; Crocker-Bedford adequate food source, the number and distribution of | 1990), yet on insular Newfoundland, guidelines for suitable nesting sites, or some combination of these managing woodland raptors are inadequate and, factors (Newton 1979). Furthermore, these factors prior to this research, the consequences of forest likely vary within and between species and across _ removal on raptor ecology had not been considered. different temporal and spatial scales. Monitoring the Large scale commercial forest harvesting is intensive densities of wildlife populations (number of individu- in Newfoundland (Thompson et al. 1999) thus, it is als or pairs/area) is of interest to ecologists and man- important to investigate the species composition and agers when assessing the potential environmental _ relative abundances of birds of prey in different for- impacts of land-use practices. Habitat quality can be est habitats to assess how forestry practices may partially assessed by determining the number of indi- influence raptor diversity and abundance. In this viduals that inhabit an area, though the validity of | study we surveyed the diversity, distribution, and using breeding density as the only or most important abundance of woodland birds of prey in different- indicator of habitat quality has been questioned (Van _aged balsam fir forests in western Newfoundland. Horne 1983). Though numerous raptor surveys have been conducted across North America, surveys are Study Area and Methods typically localized and feature more easily detected The Western Newfoundland Model Forest species in open habitats (Schmutz 1984; Andersen (WNMEF) study area (707 000 ha) occurs within the and Rongstad 1989). Density estimates are difficult Corner Brook subregion of the Western Forest to obtain for woodland species during the breeding Ecoregion of insular Newfoundland (Damman season because they are typically wide-ranging, 1983). This subregion is characterized by hilly ter- secretive and often nest in inaccessible areas, particu- rain up to 600 m above sea level. Forests are primar- larly in northern boreal forests (Fuller and Mosher ily Balsam Fir with occasional with White Spruce 1987). (Picea glauca), Black Spruce (P. mariana) and The boreal forest of insular Newfoundland consti- White Birch (Betula papyrifera; Newfoundland tutes the extreme eastern and southern limits of the Forest Service 1992). The humid climate minimizes a 58 THE CANADIAN FIELD-NATURALIST fires (Damman 1983), and under natural conditions infestations of Hemlock Looper (Lambdina fiscellar- ia) and Spruce Budworm (Choristoneura fumifer- ana) are responsible for forest renewal in western Newfoundland (Bazukis and Hansen 1965). De- foliated patches of forest and clear-cuts from timber removal for pulp have resulted in a very fragmented landscape in this region. We classified “uncut old- growth” as forest greater than 80 years old that has never been commercially harvested. Small forest openings (~100 X 100 m) resulting from insect defoliation, and an abundance of snags and coarse woody debris typify these sites. Tree heights often reach 20—24 m (Newfoundland Forest Service1992). Second-growth forests in western Newfoundland are AO to 60 year old regenerated stands that originated following timber harvesting earlier in the 20th centu- ry. Compared with uncut older forests, these stands are characterized by smaller tree diameters, more stems/hectare, higher shrub diversity and less woody debris (Thompson and Curran 1995). Clear-cut sites resulting from forest harvesting have had most of the wood volume removed, though some deciduous and otherwise unmerchantable trees remain on these sites. Herbaceous ground vegetation consisting of raspberry (Rubus spp.) and alder (Alnus spp.) is dense; however some clear-cuts are devoid of most ground vegetation owing to treatment with herbi- cides. Clear-cuts range from 5 tol5 years old. Pre- commercially thinned areas are found throughout the study area and range from 10-30 years old. These regenerating sites have been artificially thinned to promote more efficient tree growth and are com- posed mainly of Balsam Fir. Study sites within the WNMEF were established in uncut old-growth, second-growth, pre-commercially thinned areas and clear-cuts, and were located near Little Grand Lake, George’s Lake and Cook’s Pond, respectively (Figure 1). In 1993, seven transects pass- ing through uncut old-growth, second-growth, clear- cuts and pre-commercially thinned areas were estab- lished along forest roads. In areas inaccessible by truck, an all-terrain vehicle or boat was used to access the transects. However, because of the relative inac- cessibility of old-growth sites and the time constraints for surveying during the breeding season, habitats were not sampled relative to their availability. Each transect was surveyed three times for the target species during the breeding season (Table 1). Survey stations were set up at 800 m intervals along the tran- sects. At each survey station, territorial vocalizations were broadcast to elicit responses from territorial adults that could occur in the area (Mosher et al. 1990). In 1993, the target species were Sharp-shinned Hawks (Accipiter striatus), Northern Goshawks (A. gentilis), Merlins (Falco columbarius), Boreal Owls (Aegolius funereus) and Great Horned Owls (Bubo ‘virginianus). A battery powered Realistic vsc-2001 cassette recorder and two Realistic portable Minimus- Vol iis Legend: 1 Little Grand Lake 2 George's Lake 3 Cook's Pond FIGURE 1. The western Newfoundland Model Forest Study area (shaded region). Study sites are indicated by numbers 1-3. 0.6 speakers (83 db/1m) were used for the playbacks. Recordings were obtained from the Cornell Laboratory of Ornithology and from the Peterson Field Guide Series (Myer and Peterson 1990). Surveys were not conducted during periods of inclement weather; i.e. heavy fog, prolonged rain, or winds greater than Beaufort 3 (13-19 km/h; Mosher et al. 1990). Upon arriving at each survey station, observers would listen and look for birds of prey for 1 min. Then, a series of six 20-sec playbacks of a single species’ vocalizations, separated by 30 sec silent intervals, were made during a 5-min period. Three vocalization segments were broadcast toward an arbitrarily selected side of the road (determined by coin toss), followed by three broadcast segments to the other side. The observer would then remain at the site for an additional 5 min to listen and look for birds of prey. Diurnal species were identified from vocalizations, flight characteristics and plumage pat- terns, and nocturnal owls from auditory cues. Birds were considered territorial adults if repeatedly recorded at the same site, if an aggressive response was elicited by the observer, or if fledglings were seen late in the breeding season. Unconfirmed sight- ings of raptors were not included in the analyses. Only one target species was surveyed on any morn- ing or night. Playbacks for the remaining target species were broadcast on successive days. Nocturnal surveys were conducted from 22:30 to 02:00, and morning surveys from 06:00 to 10:00. The survey period in 1993 was from 24 May to 14 August. Birds observed between survey stations 2001 GOSSE AND MONTEVECCHI: ABUNDANCES OF FOREST BIRDS OF PREY 59 TABLE |. Survey transect specifications for the four forest types in western Newfoundland, 1993-1994. Number of transects Forest Habitat Type 1993 Uncut old-growth Second growth Thinned Clearcut Total 1994 Uncut old-growth Second growth Clearcut Total “SIND NM RK “SIND NB W were recorded to the nearest station from the point of contact. In 1994, survey routes were re-established to ensure that distances surveyed in each habitat type were comparable. The distances surveyed in uncut old-growth, second growth and clear-cut areas were 18, 16 and 16 km respectively, and an equal number of surveys (n = 3) were conducted along each route. Pre-commercially thinned areas were not surveyed in 1994 because the total area of this habitat type in the WNME is negligible and therefore of less interest. Playbacks for Merlins were not broadcast in 1994, because results from 1993 indicated that they did not respond to broadcast vocalizations. Broadcast sur- veys using vocal recordings of Sharp-shinned Hawks, Northern Goshawks, Boreal Owls, Great Horned Owls and Northern Hawk-Owls (Surnia ulula) were run from 1 June to 15 August 1994. In contrast to the methods used in 1993, individual sur- veys involved playbacks for multiple species on each outing. That is, nocturnal surveys involved broad- casting playbacks for each owl species at every sec- ond 800-m interval along the survey route. Lack of response due to habituation to the playbacks was considered minimal since only three visits were made to each point during the season. Playbacks for individual diurnal species were broadcast at every third 800-m interval along survey routes. Broad- casting for multiple species on each survey con- trolled for variations in weather conditions that might otherwise influence detectability when indi- vidual species broadcasts were run on successive days. Further, surveys were initiated at alternating ends of survey routes on successive visits to balance the potential effect of varying broadcast times. We also ensured that vocalizations of the target species were broadcasted at each survey station during the breeding season. In March 1995, 26 km of uncut old- growth forest was surveyed for Boreal and Great Horned Owls using playbacks. Broadcast stations were 800 m apart and access was by snowmobile. Number of Surveyed Transect broadcast area lengths (km) stations (km2) 15 19 18.0 9 13 10.8 18 22 28.8 16 20 25.6 58 72 83.2 18 22 21.6 16 20 19.2 16 20 25.6 50 62 66.4 The area surveyed in each habitat type was calcu- lated by multiplying the km driven on forest roads and lake shorelines by the estimated distance on each side of the route to which broadcasts were audible. Mosher et al. (1990) found that broadcasts with a comparable db output were audible to humans at 750 m away from the source in a hardwood stand in Maryland. In this study, the distance that broadcasts were still audible to researchers was estimated to be 600 m in forest habitats and 800 m in clearcuts and pre-commercially thinned areas. The numbers of each species in each habitat were compiled and compared using a non-parametric ran- domization technique (Manly 1991). The randomiza- tion test is a re-sampling procedure which creates its own frequency distribution based on the original abundance data, thus eliminating the assumptions of normality (Adams and Anthony 1996). The measure used for the among-habitat comparisons was the num- ber of birds/km averaged over all visits to that station. In the initial step of the randomization test, the differ- ence between the observed mean values of raptors/ broadcast station between the two habitats being com- pared was calculated. A frequency distribution of 3000 possible outcomes of differences between the mean values was then randomly generated with replacement from the original data. Habitats are sig- nificantly different with respect to the number of birds/km if the observed difference of mean values between the two habitats lies outside the 95% confi- dence interval set around the distribution of 3000 pos- sible mean outcomes. Results Effectiveness of vocalization broadcasts Seven transects totalling 58 and 50 km (habitats combined) were surveyed for woodland birds of prey in 1993 and 1994, respectively (see Table 1). Although responses of raptors to the broadcast vocalizations were elicited on occasion, most sight- ings occurred while driving between broadcast sta- 60 tions or before the calls were broadcast. In 1993, only three of 94 birds detected (3.2 %) were in re- sponse to broadcast vocalizations. Two of these were Boreal Owls and the other a Sharp-shinned Hawk. In 1994, 8 of 105 raptors detected (7.6 %) were in response to broadcast vocalizations, these included Sharp-shinned Hawks (2), Great Horned Owls (2), Northern Hawk-Owls (3), and a Boreal Owl. It is notable that when responses to broadcasts did occur, they were elicited during the 5-min broadcasting period. Numbers for both years were not combined because of differences in the survey methodologies used during each breeding season. Species composition Nine species of raptors were recorded within the study area during 1993 and 1994: Merlin, American Kestrel (Falco sparverius), Osprey (Pandion haliae- tus), Rough-legged Hawk (Buteo lagopus), Sharp- shinned Hawk, Northern Goshawk, Boreal Owl, Great Horned Owl, and Northern Hawk-Owl (see Table 2). Relative abundances and habitat associations of birds of prey Uncut old-growth Balsam Fir forests near Little Grand Lake were used by Sharp-shinned Hawks, Merlins, Boreal Owls and Ospreys in 1993 and 1994, and a Northern Goshawk in 1994. The num- ber of territorial adults and the number of birds/km for each species by habitat and year are given in Table 3. Although numbers are low, indices of rela- tive abundances are derived. In 1993, Sharp-shinned Hawks and Boreal Owls were the most abundant species (birds/km for these species were 0.27 and 0.33, respectively). Merlins and Ospreys were less abundant along survey routes through this habitat at 0.07 and 0.13 birds/km, respectively. In 1994, Sharp-shinned Hawks were only found in old- growth forest at a density of 0.22 birds/km. A com- parison of Merlin numbers between years was not THE CANADIAN FIELD-NATURALIST Vol. 115 made because of differences in the surveying methodology used in each season. Only one Boreal Owl was detected in 1994 compared with five the previous year. In addition, the only goshawk record- ed during this study occurred in uncut old-growth forest in 1994. In 1993, 9 km of this habitat was surveyed result- ing in only one sighting of each of these species. In 1994, vocalizations broadcast along 16 km of sec- ond-growth forest near George’s Lake resulted in four Merlin and two Great Horned Owl detections (Table 3). Three species of woodland raptors were identified along survey routes transecting second- growth forests near Victoria Lake and George’s Lake: Sharp-shinned Hawk, Merlin, and Great Horned Owl. In addition to the focal species, sec- ond-growth forests in various successional stages were utilized by an estimated 20 nesting pairs of Osprey in the Stephenville Crossing area. Clear-cuts provided nesting and foraging habitat for both Northern Hawk-Owls (two nests) and American Kestrels (two nests) as well as foraging sites for Merlins and Rough-legged Hawks. In 1993, birds/km ranged from 0.13 for Rough-legged Hawks to 0.19 birds/km for Merlins and Northern Hawk- Owls. In 1994, Rough-legged Hawks were not sighted in clear-cuts; however, two American Kestrels were recorded (0.13 birds/km). Merlin were recorded at a density of 0.19 birds/km (Table 3). In 1993, surveys along 18 km of pre-commer- cially thinned areas indicated that only Rough- legged Hawks (0.17 birds/km) and Merlins (0.06 birds/km) used these areas for foraging (Table 3). No species were known to utilize thinned areas for nesting. Owls were not detected along 26 km of uncut old-growth forest in March 1995. Old-growth Balsam Fir forests were utilized by more species than the other forest types, though habitats did not differ significantly with respect to the mean number of birds/km recorded along the survey routes. TABLE 2. Sightings and habitat associations of adult birds of prey in western Newfoundland, 1993-1994. Numbers also include birds of prey recorded in areas outside of survey routes but in the WNMF region. Total Balsam Fir habitat type 47 (30) Clear-cuts, old and young second growth, uncut old-growth 13\@2) Clear-cuts 40 (28) Young second growth and uncut old-growth near large water bodies 22 (18) Barren, clear-cuts, cliffs 28 (14) Uncut old-growth, old second growth 4 (1) Uncut old-growth 9 (6) Uncut old-growth 10 (6) Old second growth 24 (12) Clear-cuts Number of sightings (estimated number of individuals) Species 1993 1994 Merlin 25 (14) 22 (16) American Kestrel 9 (8) 4 (4) Osprey 5 (4) 35 (24) Rough-legged Hawk 22 (18) 0 (0) Sharp-shinned Hawk 13 (6) 15 (8) Northern Goshawk 0 (0) 4 (1) Boreal Owl ep) 1 (1) _ Great Horned Owl Di) 8 (4) Northern Hawk-Owl 10 (5) 14 (7) 2001 GOSSE AND MONTEVECCHI: ABUNDANCES OF FOREST BIRDS OF PREY 61 TABLE 3. Number of birds of prey/km detected along survey routes in different Balsam Fir forest types in western Newfoundland, 1993-1994. Numbers of individuals in parentheses do not include repeat sightings. Species name abbrevia- tions: M= Merlin, K=American Kestrel, O= Osprey, RLH=Rough-legged Hawk, SSH=Sharp-shinned Hawk, GH=Northern Goshawk, B = Boreal Owl, GHO=Great Horned Owl, NHO= Northern Hawk-owl. Habitat type Km 1993 surveyed M K Oo REA Uncut old-growth 15 ROTO! O05 0 (1) (2) Second-growth 9 Hie ov 0 0 (1) Clear-cut 16 D190 0 0.13 (3) (2) Thinned 18 06-10 0 0.17 (1) (3) 1994 Uncut old-growth 18 Otay “a0 AOE | 0 (3) (2) Second-growth 16 2a 0 0 0 (4) Clear-cut 16 Oats 0.13). 0 0 G2) Discussion Woodland birds of prey generally occur in low den- sities, are secretive, wide-ranging, and therefore diffi- cult to census (Fuller and Mosher 1981). In light of this, broadcasting taped calls as a means of improving detection rates, and hence estimating bird densities, has become an increasingly prevalent technique for detecting forest raptors. This method has been used to increase detection rates of Spotted Owls (Strix occi- dentalis) (Forsman et al. 1977), Red-shouldered Hawks (Buteo lineatus), Cooper’s Hawks (Accipiter cooperii), Barred Owls (Strix varia) (Mosher et al. 1990) and Northern Goshawks (Kennedy and Stahlecker 1993). Four species of birds of prey responded to taped broadcasts in our study area. Although not highly effective in the present study, (only 3 and 8 % of the total detections of woodland raptors in 1993 and 1994, respectively), playbacks accounted for two of three Great Horned Owl detec- tions, three of six Boreal Owl and Northern Hawk- owl detections, and one of nine Sharp-shinned Hawk detections. We speculated that responses to vocaliza- tions and hence the abundance estimates of Boreal and Great Horned Owls might have been higher in late winter or early spring (February—April) when ter- ritories are being established (Morrell et al. 1991; Hayward et al. 1993). To test this, we conducted sur- veys for these species in uncut old-growth forest dur- ing 15-21 March 1995; however, no owls were detected. Lundberg (1979) found that “territorial and breeding pairs of Boreal Owls were more silent than non-territorial individuals,” so censuses using play- backs may give biased estimates of owl abundances and may not be sensitive for detecting breeding pairs. Hayward et al. (1993) contended that a lack of under- Birds/km # of SSH GH BO GHO NHO Overall species 0.27 0 0.33 0 0 0.80 4 (4) (5) (12) 0.11 0 0 0.11 0 0.33 3 (1) (1) (3) 0 0 0 0 0.19 0.51 3 (3) (8) 0 0 0 0 0 0.23 2 (4) 0:22-0:06". “0:06 0 0 0.62 5 (4) Ce) (11) 0 0 0.13 0 0.38 2 (2) (6) 0 0 0 0 0.19 0.51 3 (3) (8) standing of the factors that affect Boreal Owl singing rates makes the vocalization method an inappropriate monitoring tool. Hence, censusing birds of prey in forested habitats remains problematic, and research is needed to develop more reliable and robust censusing techniques. Surveys conducted during the 1993 and 1994 breeding seasons indicated that uncut old-growth forest contained the most individuals and the most species. Forests in advanced stages of natural succes- sion in western Newfoundland are typified by exten- sive snag retention and stands are of various ages, thus allowing a potentially wider range of species to inhabit this habitat type. Boreal Owls, for example, require tree cavities for nesting (Hayward et al. 1993) and were restricted to uncut old-growth Balsam Fir forests. Sharp-shinned Hawks prefer a dense canopy cover for nesting (Moore and Henny 1983) and a relatively open understory for hunting, and were found mainly in old-growth forests. Second-growth forests in Newfoundland are younger even-aged forests (Thompson and Curran 1995) with less structural diversity than uncut old- growth forests. They supported fewer species than uncut old-growth forests. Surveys in second-growth forests indicated the presence of only three species of birds of prey, one of which, Great-Horned Owl, is considered a habitat generalist with a wide ecologi- cal tolerance (Bosakowski et al. 1989). Clear-cuts provided breeding and feeding habitat for Northern Hawk-Owls and American Kestrels. Large hardwood snags have been left intact through- out these areas providing both nesting structures and perching sites for foraging. Merlins were also fre- quently observed hunting in this habitat as were 62 THE CANADIAN FIELD-NATURALIST Rough-legged Hawks in 1993. The latter species is an aerial predator which prefers open ground, although typically at more northerly tundra habitats (Poole and Bromley 1988; Whitaker et al. 1996). Recommendations for the management of birds of prey in western Newfoundland Birds of prey have received increasing attention with respect to land-use decisions in recent decades. Human development has led to a general decline in birds of prey on a global scale. Three main factors have been identified as causing declines (or limiting numbers): destruction and degradation of habitat, per- secution by humans, and contamination by toxic chemicals (Newton 1979). The standard method of timber harvesting in western Newfoundland is clear- cutting. This method may be the most economical in terms of obtaining large amounts of timber, however it creates large areas of unsuitable habitat for forest birds of prey that are dependent on older age classes. Alternative harvesting techniques should be explored to ensure the continued existence of wildlife popula- tions in regions with intensive forest utilization. Different silvicultural techniques may influence raptor populations in various ways thus it is critical to deter- mine which species of birds of prey and other species are to be given priority in forestry management plans. Though none of the woodland birds of prey observed in this study are listed by the Committee for the Status of Endangered Wildlife in Canada [COSEWIC] (Kirk and Hyslop 1998), we propose that species primarily associated with mature and uncut old-growth Balsam Fir forest (Northern Goshawks, Boreal Owls and Sharp-shinned Hawks), be featured in management programs because the distribution of this age class in western Newfoundland has been greatly reduced, and further habitat loss would likely have negative impacts on the resident species (Thompson et al. 1999; Setterington et al. 2000). Woodland raptors occur in low numbers, and it is necessary to preserve large expanses of adequate habitat in order to provide for viable populations. Conservation of older forests would also benefit other wildlife species such as American Marten (Martes americana atrata) and some forest birds (see Whitaker and Montevecchi 1997, 1999; Thompson et al. 1999). Forest harvesting may also have inadvertently increased the amount of habitat available to certain species that prefer open habitats and consequently resulted in an expansion of their ranges or an increase in their populations. For example, Northern Hawk- Owls and American Kestrels utilized clearcuts for nesting and foraging. The American Kestrel’s range has expanded into Newfoundland since the 1940s, or at least its numbers appear to have increased (Montevecchi and Tuck 1987). Foraging raptors that use pause-travel-search tactics (i.e., Northern Hawk- . Owls) have been documented to use clearcuts with perches significantly more frequently than clear-cuts Vol. 115 lacking perches (Widen 1994). Habitats for cavity- dependent species such as Boreal Owls and American Kestrels may also be potentially expanded or improved by establishing nest-box programs, espe- cially in younger successional forests where suitable nesting cavities are lacking. Such programs would be relatively inexpensive and could also provide valu- able and much-needed public education concerning forest birds of prey and wildlife conservation. Acknowledgments We thank D. Whitaker and M. Mayo for valuable assistance in the field, and L. Mayo, B. Greene, J. Brazil, C. Knox, M. Setterington, B. Mactavish and D.Butler for further logistical support. D. Schneider and R. Knoechel provided statistical advice through- out this study, and G. Bortolotti, D. Steele and I. Warkentin provided further helpful comments. Funding for this research was provided by a contract to W. Montevecchi from the Western Newfoundland Model Forest Corporation for which we acknowl- edge our gratitude. Literature Cited Adams, D.C., and C.D. Anthony. 1996. Using random- — ization techniques to analyse behavioural data. Animal Behaviour 51: 733-738. Andersen, D. E., and O. J. Rongstad. 1989. Surveys for wintering birds of prey in southeastern Colorado: 1983-1988. Journal of Raptor Research 23: 152-156. Armstrong, E., and D. Euler. 1982. Habitat usage of two woodland Buteo species in Central Ontario. Canadian Field-Naturalist 97: 200-207. Bazukis, E. V., and H. L. Hansen. 1965. Balsam fir: a monographic review. University of Minnesota Press, Minneapolis. Bosakowski, T., R. Speiser, and D. G. Smith. 1989. Nesting ecology of forest-dwelling Great Horned Owls, Bubo virginianus, in the eastern deciduous biome. Canadian Field-Naturalist 103: 65-69. - Carey, A. B., J. A. Reid, and S. P. Horton. 1990. Spotted owl home range and habitat use in southern Oregon Coast Ranges. Journal of Wildlife Management 54: 11-17. Crocker-Bedford, D. C. 1990. Goshawk reproduction and forest management. Wildlife Society Bulletin 18: 262-269. Damman, A. M. H. 1983. An ecological subdivision of the island of Newfoundland. Pages 163-206 in Bio- geography and ecology of the island of Newfoundland. Edited by G. R. South. Junk Publishers, The Hague. Forsman, E. D., E. C. Meslow, and M. J. Strub. 1977. Spotted owl abundance in young versus old-growth forests, Oregon. Wildlife Society Bulletin 5: 43-47. Fuller, M.R., and J. A. Mosher. 1981. Methods of detecting and counting raptors: a review. Studies in Avian Biology 6: 235-246. Fuller, M.R., and J. A. Mosher. 1987. Raptor survey techniques. Pages 37-65 in Raptor Management Tech- niques Manual. Edited by B. A. Giron Pendleton, B. A., Millsap, K. W., Cline, and D. M. Bird. National Wildlife Federation, Washington D.C. 2001 Hayward, G. D., P. H. Hayward, and E. O. Garton. 1993. Ecology of boreal owls in the northern Rocky Mountains, USA. Wildlife Monographs 124: 1-59, Kennedy, P.L., and D.W. Stahlecker. 1993. Responsiveness of nesting northern goshawks to taped broadcasts of three conspecific calls. Journal of Wildlife Management 57: 249-257. Kirk, D. A., and C. Hyslop. 1998. Population status and recent trends in Canadian raptors: A review. Biological Conservation 83: 91-118. Lundberg, A. 1979. Residency, migration and a compro- mise: adaptations to nest site scarcity and food special- izations in three Fennoscandian owl species. Oecologia (Berlin) 41: 273-281. Manly, B. F. J. 1991. Randomization and Monte Carlo methods in biology. London. Chapman and Hall. Montevecchi, W.A., and L. M. Tuck. 1987. Newfound- land birds: Exploitation, study, conservation. Nuttall Ornithological Club number 21, Cambridge, Massa- chusetts. Moore, K. R., and C. J. Henny. 1983. Nest site charac- teristics of three coexisting Accipiter Hawks in North- eastern Oregon. Journal of Raptor Research 17: 65-76. Morrell, T. E., R. H. Yahner, and W. L. Harkness. 1991. Factors affecting detection of great horned owls by using broadcast vocalizations. Wildlife Society Bul- letin 19: 481-488. Mosher, J. A., M. R. Fuller, and M. Kopeny. 1990. Surveying woodland raptors by broadcast of conspecific vocalizations. Journal of Field Ornithology 61: 453-461. Myer, S. D., and R. T. Peterson. 1990. A field guide to the bird songs of eastern and central North America 3rd edition (sound recording). Peterson Field Guide Series. Houghton Mifflin, Boston. Newfoundland Forest Service. 1992. 20 year forestry development plan 1990-2009. Department of Forestry and Agriculture. February 1992. 133 pages. Newton, I. 1979. Population ecology of raptors. Buteo Books, Vermillion, South Dakota. Poole, K. G., and R. G. Bromley. 1988. Interrelationships within a raptor guild in the central Canadian Arctic. Canadian Journal of Zoology 66: 2275-2282. Reynolds, R. T., E. C. Meslow, and H. M. Wight. 1982. GOSSE AND MONTEVECCHI: ABUNDANCES OF FOREST BIRDS OF PREY 63 Nesting habitat of coexisting Accipiters in Oregon. Journal of Wildlife Management 46: 124-138. Schmutz, J. M. 1984. Ferruginous and Swainson’s hawk abundance and distribution in relation to land use in southeastern Alberta. Journal of Wildlife Management 48: 1180-1187. Setterington, M.A., I.D. Thompson, and W. A. Montevecchi. 2000. Woodpecker abundance and habitat use in mature and balsam fir forests in Newfoundland. Journal of Wildlife Management 64: 335-345. Solonen, T. 1994. Factors affecting the structure of Fin- nish birds of prey communities. Ornis Fennica 71: 156-169. Thompson, I. D., H. A. Hogan, and W. A. Montevecchi. 1999. Avian communities of mature balsam fir forests in Newfoundland: Age dependence and implications for timber harvesting. Condor 101: 311-323. Thompson, I. D., and W. J. Curran. 1995. Habitat suit- ability for marten of second-growth balsam fir forests in Newfoundland. Canadian Journal of Zoology 73: 2059-2064. ; Titus, K., and J. A. Mosher. 1981. Nest-sites selected by woodland hawks in the central Appalachians. Auk 98: 270-281. Van Horne, B. 1983. Density as a misleading indicator of habitat quality. Journal of Wildlife Management 47: 893-901. Whitaker, D. M., and W. A. Montevecchi. 1997. Effects of riparian and nonriparian edges on breeding bird assemblages in balsam fir forests in Newfoundland. Canadian Journal of Forestry Research 27: 1159-1167. Whitaker, D. M., and W. A. Montevecchi. 1999. Breed- ing bird assemblages inhabiting riparian buffer strips in Newfoundland, Canada. Journal of Wildlife Manage- ment 63: 167-179. Whitaker, D. M., W. A. Montevecchi, and J. W. Gosse. 1996. Breeding season irruptions of Rough-legged Hawks (Buteo lagopus) on insular Newfoundland. Arctic 49: 306-310. Widen, P. 1994. Habitat quality for raptors: a field experi- ment. Journal of Avian Biology 25: 219-223. Received 8 November 1999 Accepted 17 October 2000 Distances Moved by Small Woodland Rodents within Large Trapping Grids JEFF BOWMAN! 3, GRAHAM J. FORBES?, and Tim G. DILwortTu! 1Department of Biology, University of New Brunswick, P. O. Box 45111, Fredericton, New Brunswick E3B 6E1, Canada 2New Brunswick Cooperative Fish and Wildlife Research Unit, P. O. Box 44555, Fredericton, New Brunswick E3B 6C2, Canada 3Present address: Department of Biology, Carleton University, 1125 Colonel By Drive, Ottawa, Ontario K1S 5B6, Canada Bowman, Jeff, Graham J. Forbes, and Tim G. Dilworth. 2000. Distances moved by small woodland rodents within large trapping grids. Canadian Field-Naturalist 115(1): 64-67. During a four-year study in New Brunswick, Canada, we documented long-distance movements (> 125 m) for three small- mammal species. Individuals from every species studied made long-distance movements, but relative to abundance, more Woodland Jumping Mice moved than any other species (9.4% of captures). Mean straight-line distances moved were: 370 m (Deer Mice, Peromyscus maniculatus; N = 44), 225m (Woodland Jumping Mice, Napaeozapus insignis; N = 33), and 224 m (Red-backed Voles, Clethrionomys gapperi; N = 23). Frequency of movement decreased with distance for all species. The study demonstrated that long-distance movements were not uncommon, and as such, they may be an impor- tant component of the population dynamics of small mammals. Key Words: Red-backed Vole, Clethrionomys gapperi, Deer Mice, Peromyscus maniculatus, Woodland Jumping Mouse, Napaeozapus insignis, movement, scale dispersal, New Brunswick. Movement contributes to temporal and spatial opportunity to assess long-distance movements for structure of populations (Wiens et al. 1993). three species. Here, we describe the magnitude, vari- Empirical descriptions of movements for small ability, and seasonality of long-distance movements mammals are rare, in part, because of the methods made by: Peromyscus maniculatus, Napaeozapus used to sample populations (Howard 1960; Clark et insignis, and Red-backed Voles, Clethrionomys gap- al. 1988). Trapping grids frequently are too small in peri. spatial extent to detect long-distance movements (e.g., Burt 1940; Smith et al. 1975; Wegner and Methods Merriam 1990; Merriam 1995). There are, however, The study took place in the private industrial forest scattered reports of long-distance movements for a of Fraser Papers Inc., in the Appalachian forest of variety of small-mammal species, for example: Deer northwestern New Brunswick (47°N, 67°W). Upland Mice, Peromyscus maniculatus (Howard 1960; _ sites were dominated by an overstory of Sugar Maple Bowman et al. 1999), White-footed Mice, Pero- (Acer saccharum), Yellow Birch (Betula alleghanien- myscus leucopus (Krohne et al. 1984; Wegner and _ sis), and American Beech (Fagus grandifolia). Merriam 1990), Woodland Jumping Mice, Napaeo- Lowland sites were dominated by Black Spruce zapus insignis (Ovaska and Herman 1988), (Picea mariana), White Spruce (Picea glauca), and Stephens’ Kangaroo Rats, Dipodomys stephensi Balsam Fir (Abies balsamea). (Price et al. 1994), Western Harvest Mice, The study design was described in detail by Reithrodontomys megalotis (Clark et al. 1988), and Bowman et al. (2000). Two 4900-ha forested land- observations of several species described in scapes with contrasting management intensities Kozakiewicz and Szacki (1995). These reports arean (>50% clearcuts or plantations < 15 yrs old vs important source of data which can help ecologists to <15% clearcuts or plantations < 15 years old) were understand the process of dispersal (e.g., systematically live-trapped using a set of nested Kozakiewicz and Szacki 1995; Sutherland et al. grids. The two largest grids (one per landscape) had 2000). For example, dispersal distance is frequently grains (or distance between sampling points) of 1000 a component of spatially-explicit simulation models m and areal extents of 4900 ha (8 X 8; 64 points in (e.g., Pulliam et al. 1992; Wilson et al. 1993; With each grid). Nested within each of the large grids was and Crist 1995), and such models can be parame- a smaller grid with a grain of 250 m and an extent of terised with empirical dispersal data. 310 ha (8 X 8; 64 points in each grid), and on the During a four-year study of the spatial structure of less intensively-managed landscape, there was a small-mammal populations in a managed forest in third grid with a grain of 125m and an extent of New Brunswick, Canada, the study design involved 3l1ha(5 xX 6; 30 points). A total of 260 sample large (4900 ha) live-trapping grids and provided an points were spread systematically across the two 64 2001 landscapes within these grids. At each point an array of five Victor Tincat multiple-capture live traps (Woodstream Corp., Lititz, Pennsylvania, USA) was used to sample small-mammal populations. Traps were placed at point centre and 35-m from the centre on each compass ordinal; each trap was placed in a “most likely runway” position. Traps were prebaited for three days with oats and sunflower hearts and then set for four consecutive nights. The trapping protocol was carried out twice per annum in spring (May—June) and fall (August-September). Captured animals were weighed, identified to species and gen- der, checked for reproductive condition, marked with a l-g monel ear tag (National Band and Tag Co., Newport, Kansas, USA) and released. Our protocol was approved by the University of New Brunswick Animal Care Committee. We considered that an animal had moved if it was recaptured (i.e., with an ear tag) at a sample point where it had not been captured previously. The finest grain in our study was 125 m (1.e., the space between sample points on the 31 ha trapping grid) so 125m was our definition of a long-distance movement. Shorter movements were not long enough to move an animal between sample points, and were not con- sidered in our analysis. Movement distances were calculated as straight-line distances between the cen- tres of the capture and recapture sample points. Calculations were made using digital maps of the study area and a Geographic Information System (GIS; Arc/View). We used nonparametric analyses to compare dis- tances moved by different species, and where sample sizes were suitable, to compare distances moved by sex and age groups within species. We calculated the relative frequencies of long-distance movements by different species, and within-species seasonal differ- ences in the number of captures after long-distance movements. Results The three most abundant rodents captured were Red-backed Voles (9.40 captures/100 trap nights), Deer Mice (7.66 captures/100 tn), and Woodland Jumping Mice (2.78 captures/100 tn). These three were the only species for which long-distance move- ments were observed. Relative to abundance and BOWMAN, FORBES, DILWORTH: DISTANCES MOVED BY RODENTS 65 combining seasons, more Woodland Jumping Mice moved between sampling points than any other species: 9.4% of captures occurred after long- distance movements, compared to 4.2% and 1.8% for Deer Mice and Red-backed Voles, respectively (Table 1). These proportions were calculated by omitting captures on the two largest (1000-m grain) grids, because no movements were ever recorded at this largest scale. There were no differences in distances moved by males, females, or juveniles within any species. Thus, we combined within-species sex and age class- es to compare distances moved by different species and found that Deer Mice moved farther than either of the other species (x? = 5.9, df = 2, P = 0.048) (Figure 1, Table 1). Sample sizes were too small to assess within-species, age- and sex-related statistical patterns in seasonal movements. However, by com- bining ages and sexes and controlling for trap effort, we found that Deer Mice (x2 = 4.5, df = 1, P= 0.035) and Red-backed Voles (x? = 7.4, df = 1, P=0.007) were more frequently captured in fall than in spring, after having moved a long distance. Conversely, Woodland Jumping Mice were more frequently cap- tured in spring than in fall, after having moved a long distance (x2 = 8.76, df = 2, P = 0.003) (Table I): Discussion Individuals from all three of the abundant species in our study area moved distances in excess of 125 m. The frequency of movements generally decreased with distance, which is consistent with studies of dis- persal across a range of taxa (Taylor 1980; Sutherland et al. 2000). Deer Mice made longer movements than either Red-backed Voles or Woodland Jumping Mice — the longest of these (1768 m) was reported by Bowman et al. (1999). Other authors have recog- nised the ability of Deer Mice (and closely-related White-footed Mice) to move long distances. Howard (1960) recorded a movement of 1000 m for P. man- iculatus, and Wegner and Merriam (1990) speculat- ed about movements by P. leucopus of > 1000 m. While Ovaska and Herman (1988) demonstrated a movement by a Woodland Jumping Mouse of > 800 m, we are unaware of other studies reporting TABLE 1. Distances (m) moved by small woodland rodents within large trapping grids in New Brunswick, Canada, during 1997-1999. Species Mean SE Peromyscus maniculatus 370 5 Napaeozapus insignis 225 22 Clethrionomys gapperi 224 24 ATotal number (number in spring, number in fall) BPercentage of captures that had moved > 125 m. N (S, F)4 Rate (%)8 Jc BD ME 44 (15, 29) 4.2 23 7 14 607 33 (25, 8) 9.4 3 10 20 494 23(5~ 18) 1.8 13 l 9 CNumber of juveniles and subadults DNumber of adult females ENumber of adult males 66 THE CANADIAN FIELD-NATURALIST 25 20 = = 15 & 4S} as Zhi B : Z 5 ; : 250 31D 2500 Vol. 115 1.0 ea) Fa fae} Q. Qu = — Cy 0.5 i) = cD) te) = S com D) > Bb a cD} : z al 0 625, J50 22150 Distance (m) FiGuRE 1. Distribution of distances moved by woodland rodents within large trapping grids in New Brunswick, Canada. White bars are Peromyscus maniculatus, solid bars are Napaeozapus insignis, and hatched bars are Clethrionomys gappervi. The relative num- ber of pairs of traps within each distance class is indicated by the solid line. long-distance movements by either N. insignis or C. gapperi, with the exception of homing studies. For example, Bovet (1980) recorded successful homing by Red-backed Voles from as far as 600 m. Seasonal differences in captures between Woodland Jumping Mice and the other species were not surprising, as jumping mice are true hibernators whereas Red-backed Voles and Deer Mice are win- ter active (Whitaker and Wrigley 1972). Our data were consistent with patterns of increased movement in late summer by murids and in early summer by zapodids. These seasonal patterns should be inter- preted carefully however, as the analysis included both movements made within a trapping season, and between seasons. Thus, we actually measured the number of captures made after long-distance move- ments per season, rather than the season when the movement was actually made (which was often unknown). Two additional sources of error must be consid- ered when interpreting these data. First, we have not corrected for the uneven distribution of trap pairs within different distance classes. The number of trap pairs varied with distance (Figure 1) and in fact, appeared to bias the distribution of distances toward longer movements: there were fewer trap pairs at the shortest distances. This was further confounded by a -second bias: the density of traps varied throughout the study grids as a result both of geometry and the nested trapping design. Rather than make question- able corrections against these two biases, we present unmodified data and caution the reader against over interpretation. Even with the inherent problems, these data are of value because of the scarcity of information on small-mammal movements (e.g., Wegner and Merriam 1990; Kozakiewicz and Szacki 1995; Merriam 1995). We expect that many of the movements which we have operationally-defined as “long-distance” were actually dispersal movements. Such movements like- ly have important, albeit poorly understood, effects on small-mammal populations. We are particularly interested in the relationship between dispersal and spatial population structure. For example, Deer Mouse, Red-backed Vole, and Woodland Jumping Mouse populations exhibit heterogeneity in abun- dance at distances of 133 - 350 m, on the same land- scapes where the present movement data was col- lected (Bowman et al. 2000). The spatial extent of the heterogeneity is consistent with the distribution of dispersal distances. This supports speculation that dispersal distance in small mammals is related to a scale of population heterogeneity (e.g., Krohne and Burgin 1990). Acknowledgments The authors acknowledge financial support from a grant to Tony Diamond by the Sustainable Forest 2001 Management Network , and grant and scholarship support from Fraser Papers Inc., NSERC, The Sir James Dunn Wildlife Research Centre, and EarthWorks NB. We collaborated with Tony Diamond, Mark Edwards, John Gunn, M.-A. Villard, Julie Bourque, Stan Boutin, Crissy Corkum, Jason Fisher, Sue Hannon, Rich Moses, and many assis- tants: we thank them all. Comments from John Bissonette, Kringen Henein, Jeff Holland, Matt Litvak, Dan Quiring, M.-A. Villard, and two anony- mous reviewers improved the manuscript. Literature Cited Bovet, J. 1980. Homing behaviour and orientation in the red-backed vole, Clethrionomys gapperi. Canadian Journal of Zoology 58: 754-760. Bowman, J., G. Forbes, and T. Dilworth. 2000. The spa- tial scale of variability in small-mammal populations. Ecography 23: 328-334. Bowman, J., M. Edwards, L. Sheppard, and G. Forbes. 1999. Record distance for a non-homing movement by a deer mouse, Peromyscus maniculatus. Canadian Field- Naturalist 113: 292-293. Burt, W. H. 1940. Territorial behaviour and populations of some small mammals in southern Michigan. Miscel- laneous Publications of the Museum of Zoology, Uni- versity of Michigan, Number 45. 61 pages. Clark, B. K., D. W. Kaufman, G. A. Kaufman, E. J. Finck, and S.S. Hand. 1988. Long-distance move- ments by Reithrodontomys megalotis in a tallgrass prairie. American Midland Naturalist 120: 276-281. Howard, W. E. 1960. Innate and environmental dispersal of individual vertebrates. American Midland Naturalist 63: 152-161. Kozakiewicz, M., and J. Szacki. 1995. Movement of small mammals in a landscape: patch restriction or nomadism? Pages 78—94 in Landscape approaches in mammalian ecology and conservation. Edited by W. Z. Lidicker, Jr. University of Minnesota Press, Minneapolis. Krohne, D. T., and A. B. Burgin. 1990. The scale of demographic heterogeneity in a population of Pero- myscus leucopus. Oecologia 82: 97-101. Krohne, D. T., B. A. Dubbs, and R. Baccus. 1984. An analysis of dispersal in an unmanipulated population of Peromyscus leucopus. American Midland Naturalist 112: 146-156. BOWMAN, FoRBES, DILWORTH: DISTANCES MOVED BY RODENTS 67 Merriam, G. 1995. Movement in spatially divided popu- lations: responses to landscape structure. Pages 64-77 in Landscape approaches in mammalian ecology and con- servation. Edited by W. Z. Lidicker, Jr. University of Minnesota Press, Minneapolis. Ovaska, K., and T. B. Herman. 1988. Life history char- acteristics and movements of the woodland jumping mouse, Napaeozapus insignis, in Nova Scotia. Canadian Journal of Zoology 66: 1752-1762. Price, M. V., P. A. Kelly, and R. L. Goldingay. 1994. Distances moved by Stephens’ kangaroo rat (Dipodomys stephensi Merriam) and implications for conservation. Journal of Mammalogy 75: 929-939. Pulliam, H. R., J. B. Dunning, Jr., and J. Liu. 1992. Population dynamics in complex landscapes: a case study. Ecological Applications 2: 165-177. Smith, M.H., R.H. Gardner, J.B. Gentry, D. W. Kaufman, and M. H. O’Farrell. 1975. Density estima- tions of small mammal populations. Pages 25-33 in Small mammals: their productivity and population dynamics. Edited by F. B. Golley, K. Petrusewicz, and L. Ryszkowski. Cambridge University Press, Cam- bridge. Sutherland, G. D., A. S. Harestad, K. Price, and K. P. Lertzman. 2000. Scaling of natal dispersal distances in terrestrial birds and mammals. Conservation Ecology 4(1): 16. [online] URL: hAttp://www.consecol.org/vol4/ issl/artl6. Taylor, R. A. J. 1980. A family of regression equations describing the density distribution of dispersing organ- isms. Nature 286: 53-55. Wegner, J., and G. Merriam. 1990. Use of spatial ele- ments in a farmland mosaic by a woodland rodent. Biological Conservation 54: 263-276. Whitaker, J. O., Jr., and R.E. Wrigley. 1972. Napeao- zapus insignis. Mammalian Species 14: 1-6. Wiens, J. A., N.C. Stenseth, B. Van Horne, and R.A. Ims. 1993. Ecological mechanisms and landscape ecol- ogy. Oikos 66: 369-380. Wilson, W.G., A. M. de Roos, and E. McCauley. 1993. Spatial instabilities within the diffusive Lotka-Volterra system: individual-based simulation results. Theoretical Population Biology 43: 91-127. With, K., and T.O. Crist. 1995. Critical thresholds in species’ response to landscape structure. Ecology 76: 2446-2459. Received 13 December 1999 Accepted 28 September 2000 Abundance of Stream Invertebrates in Winter: Seasonal Changes and Effects of River Ice MELANIE D. MARTIN!, RICHARD S. BROWN’, DAviID R. BARTON, and GEOFF POWER Department of Biology, University of Waterloo, Waterloo, Ontario N2L 3G1, Canada 1Robin Drive, Apartment 6, Elmire, Ontario N3B 1X1, Canada 2Battele, Pacific Northwest National Laboratory, Ecology Group, MSIN K6-85, 902 Battelle Boulevard, P.O Box 999, Richland, Washington 99352, USA Martin, Melanie D., Richard S. Brown, David R. Barton, and Geoff Power. 2000. Abundance of stream invertebrates in winter: Seasonal changes and effects of river ice. Canadian Field-Naturalist 115(1): 68-74. The association between anchor ice and stream invertebrate density in the drift and on the substrate were studied in the Grand River, Ontario, over two winters. Under certain climatic and geomorphological conditions, anchor ice can form a thick blanket on the bottom of rivers and streams. There were almost thirty times more aquatic insects in the drift on morn- ings after anchor ice developed than when anchor ice did not occur. Both Diptera (mostly chironomids) and Trichoptera were more abundant in the drift when frazil slush was present. During both winters the total density of insects in the drift decreased as the seasons progressed, regardless of the presence or absence of anchor ice on the day of sampling. Within drift samples taken after anchor ice events this relationship was seen in the orders of Diptera and Trichoptera. There was also a decline in the total number of insects in the surface layer of substrate as the winter progressed, particularly in the families of Chironomidae and Simuliidae. Anchor ice and frazil slush appear to be significant factors in invertebrate disper- sal and their role in the winter ecology of aquatic invertebrates deserves much more study. Key Words: Frazil slush, anchor ice, aquatic invertebrate, fish, dispersal, drift, Ontario. During fall and winter, stream invertebrates are isms, although, the overall depletion of benthic fauna exposed to much different conditions than during as a result of this was negligible. Brown et al. (1953) warmer seasons. Colder temperatures result in a viewed aquatic organisms floating alongside dense lower metabolism, and ice formations can cause masses of slush in the West Gallatin River, Montana, drastic changes to habitats. During cold weather, thus supporting the hypothesis that anchor ice may subsurface ice commonly forms in areas with moder- temporarily increase the number of bottom organ- ate to high turbulence. In turbulent areas not covered _isms in the drift. Similar to Benson (1955) however, with surface ice, water temperatures can drop below _ their results also showed no decrease in the overall 0.0°C; this is known as supercooling (Tsang 1982). abundance of bottom invertebrates during the winter While the water is supercooled, small ice crystals months. (called frazil ice) form in the water column. When Other authors suggest that anchor ice, and the these ice crystals adhere to the substrate they form scouring that occurs as a result, not only frees aquat- anchor ice. Anchor ice can form thick blankets on ic invertebrates from the substrate into the water col- the bottoms of rivers and streams (Tsang 1982). umn but continuously changes the benthic composi- When stream temperatures increase, or anchor ice tion of the stream bed over the winter months becomes thick enough to become buoyant, it lifts off (Maciolek and Needham 1952; O’Donnell and of the stream bottom, carrying with it pieces of sub- Churchill 1954; Reimers 1957). Reimers (1957) stat- strate, macrophytes, and aquatic invertebrates. This ed that areas where anchor ice commonly occurred ice, called frazil slush, then becomes incorporated experienced a gradual depletion in benthic fauna due into the flow of the river and transports invertebrates to repeated scouring of the streambed. During downstream. Reimers’ study at Convict Creek, California, there Due to the lack of equipment specialized for win- | was a considerably lower food intake by trout mid- ter sampling, and to harsh working conditions, few way through the winter in these habitats, which sug- papers have been published on winter invertebrate gests lower availability of food after repeated anchor drift and the effect of anchor ice on invertebrate ice events. transport. Some authors suggest that anchor ice has These contradictory results make it unclear whether no permanent effect on the removal of invertebrates the severity and frequency of anchor ice events from the substrate (Brown et al. 1953; Benson 1955). throughout the winter can change the forage available During observations at the Pigeon River, Michigan, — to fish or progressively reduce the benthic population Benson (1955) concluded that the daily release of through repeated disturbance. When we observed that anchor ice from the stream bottom served asa mech- masses of frazil slush were transporting sediment and anism for the downstream dispersal of bottom organ- debris downstream, we began collecting samples of 68 2001 the frazil slush for examination and followed up with a benthic survey. Our objective was to determine whether anchor ice events are associated with changes in the density of stream invertebrates in the drift and on the substrate. To achieve this goal, both drift and substrate samples were collected during periods when anchor ice occurred and when it was absent. Study Area and Methods The study was conducted along a 10 km reach of the Grand River (43°38'-43°4'N, 80°26'-80°2'W), Ontario in the Elora Gorge Conservation Area. The river channel in the study area has a relatively low gradient (mean 1.4 m km!) and flows in an open valley. Invertebrates in the drift were collected during periods when anchor ice was present and when it was not present over two winters (21 January—20 March 1998 and 16 December 1998-23 February, 1999; Table 1). For drift samples with frazil slush, the amount of drift sampled (m*) was determined by multiplying the amount of frazil slush collected (20-23 litres) by the estimated fraction of frazil in the drift and converting the units to cubic metres. The fraction of frazil slush in the drift was estimated by taking samples of the water column with a gradu- ated cylinder and recording the ratio of frazil slush to water. Repeated samples were taken and the mean was used to estimate the fraction of frazil slush in the drift. The drift net (opening of 0.05 m?, 2 mm mesh size) was deployed for a maximum of 3-5 minutes for each sample, depending on the amount of frazil slush in the drift. For drift samples without frazil slush, the amount of water sampled (m3) was mea- sured by multiplying the area of the net opening (0.09 m2) by the water velocity (mes-!) and the amount of time (s) that the net was in the flow. Dual drift nets, each with a net opening of 0.09 m2 and mesh size of 180 micrometers collected drift for 30-60 minutes per sample. Drift samples were col- lected between 09:00 and 13:00. All samples were filtered through 180-micrometer mesh, then pre- served in 85% ethanol. Samples with frazil slush were melted before being filtered. Organisms were counted and identified to family, when possible, using McCafferty (1981). Some organisms were missing appendages and/or were physically damaged to the point where identification to family was unre- liable. In such cases, they were identified only to order. Data from both winters (1997-1998 and 1998- 1999) were used in comparing the number of organ- isms in drift samples with and without frazil slush. To determine if there was a difference in the number of insects in the drift when frazil slush was present and when not present, a Mann-Whitney U test was performed since data were not normal (determined with a Lilliefors test). Frazil slush samples collected in 1997-1998 were used to determine invertebrate MARTIN, BROWN, BARTON, AND POWER: STREAM INVERTEBRATES 69 density trends over the winter in the drift when frazil slush was present. Similarly, the drift samples col- lected in 1998-1999 were used in determining sea- sonal trends for drift without frazil slush. To deter- mine these relationships, regression analysis was performed on the abundance of invertebrates per day in the drift when frazil slush was present and when frazil slush was not present. For the purposes of this analysis (and the analysis of invertebrates on the substrate), the earliest day on which samples were first collected in both years, 16 December, was used as the starting date of winter. Since not all relation- ships between invertebrate abundance and day of winter were linear, the type of regression analysis applied (linear or polynomial) was that which pro- vided the most appropriate degree of fit. Drift samples with frazil slush were collected in three different locations: at transect #1, 4.9 km down- stream from transect #1 and 8.4 km downstream from transect #1. Drift samples without frazil slush were collected at transect #2 (0.6 km downstream from transect #1). There were no barriers to drift in the study area. Fifty-two substrate (rock) samples (mean = 5 rocks per day) were collected between 16 December 1998 and 8 March 1999 (Table 1). Rocks (mean diameter = 8.68 cm; SD = 3.03) were selected in rif- fles along transect #1 and transect #2 (approximately 600 m apart) in the Grand River, Ontario. A small dip net (180-micrometer mesh) was used to trap invertebrates that had come loose off the rock when lifted out of the water. Each rock was put into a small container and rinsed with 85% ethanol. Insects on the net were rinsed into the container with 85% ethanol. All of the rock samples were scrubbed with a soft brush and handpicked with forceps to free the invertebrates from the rock surface. The sample was then filtered through the 180-micrometer screen and the material retained was stored in 85% ethanol until identified. The surface area of each rock was esti- mated by measuring the three largest diameters of the rock (Dall 1978). Both transects were marked so that rock samples and flow measurements could be taken at the same location on subsequent dates. Water velocities (mean and bottom) were measured using a Sigma Doppler flow meter. At a 5 m3 s! discharge rate (approximate winter base flow) transect one had a mean depth of 23 cm (N = 6, SD = 0.06) and mean velocity of 0.21 m*! (N = 6, SD = 0.11). Transect two had a mean depth of 24 cm (N = 7, SD = 0.10) and mean velocity of 0.68 ms! (N = 7, SD = 0.45). Substrate in both tran- sects was predominantly large cobble. Since there were no significant differences (p > 0.05) between depths or water velocities (Mann-Whitney U test since data were non-normal), data for both transects were grouped. Regression analysis was used to determine if there was a trend in the overall insect abundance on the substrate over the winter. ANCO- 70 THE CANADIAN FIELD-NATURALIST TABLE 1. The number of invertebrate samples taken from the drift and substrate when frazil ice was present or absent in the Grand River, Ontario. Number of samples Drift with Drift without Substrate Date frazil ice frazil ice samples 21 January 1998 3 - - 26 January 1998 yy - - 6 February 1998 3 - - 16 February 1998 2, - - 11 March 1998 3) - - 12 March 1998 4 - - 13 March 1998 3 - 16 March 1998 3 - - 17 March 1998 3 - - 16 December 1998 - 1 6 18 December 1998 - 1 6 23 December 1998 - 3 3 31 December 1998 1 - 6 6 January 1999 - i 3 13 January 1999 - 2 6 20 January 1999 - Dy 6 3 February 1999 - De 6 8 February 1999 - 2 6 21 February 1999 1 - - 23 February 1999 4 - 5 8 March 1999 - - 4 VA was used to determine if a consistent relation- ship existed between the day of the winter and abun- dance of aquatic invertebrates in both the drift (with- out frazil slush) and on the substrate. To determine if the density of invertebrates in the drift was influenced by water discharge, water dis- charge on sample days when frazil slush was not present was compared to water discharge on days when frazil slush was present. This was done using a Mann-Whitney U test since data were not normal. Subsequently, regression analysis was used to deter- mine if there was any relationship between water discharge and invertebrate density. Results Drift There was a significantly (p < 0.05) larger number of drifting invertebrates on mornings after anchor ice events (mean 118.0 organisms/m? drift, SE = 18.6) than when anchor ice events did not occur (mean 4.2 organisms/m?3 drift, SE = 1.8). Both Diptera and Trichoptera were significantly (p < 0.05) more abun- dant when frazil slush was present, but there were no significant (p > 0.05) differences in densities of Ephemeroptera and Annelida. Within the order Diptera, there was a significantly (p<0.05) larger number of chironomids in the drift when frazil slush was present (40.5/m3) than when it was not present (1.8/m3). There were no significant (p > 0.05) differ- ences found in the drift within the families of the Vol. 115 order Trichoptera. Nearly all of the Trichoptera in the drift samples were species of the family Hydro- psychidae (90%) in 1999 and the family Lepidosto- matidae (75%) in 1998. One factor that may have influenced these results is that river discharge was significantly (p< 0.05) higher on days when frazil slush was present (mean 9.5 m3s!, SD = 7.3, range 1.5—28.0) than when frazil slush was not present (mean 2.4 m3s!, SD = 1.5, range 1.6—5.0). However, no significant (p>0.05; r* = 0.03) relationship between water discharge and invertebrate density was found in samples when frazil slush was present. Total changes over winter in drift During the winters of 1997-1998 and 1998-1999 the density of animals in the drift decreased signifi- cantly (p<0.05) as the seasons progressed, regard- less of the presence or absence of frazil slush on each sampling day (Figure 1). In drift samples taken after anchor ice events (1997-1998 field season), both Diptera and Trichoptera (especially Lepido- stomatidae) showed significant (p<0.05) negative relationships between drift density and day of the winter. There was no significant (p> 0.05) trend in either Annelida (Class Oligochaeta) or Ephemer- optera. No significant (p> 0.05) relationships were found among individual taxonomic groups sampled when no anchor ice was present (1998-1999 field season). Benthos The total number of insects in the surface layer of substrate declined significantly (p < 0.05) as the win- ter progressed (Figure 2). There was a rebound in numbers following a period when surface ice cov- ered most of the river and anchor ice did not form. Invertebrate densities decreased again when the sur- face ice cover was gone and nightly anchor ice for- mations again occurred. : When analyzed by order, only Diptera declined significantly (p< 0.05) over the winter (Figure 3). This significant (p < 0.05) negative trend was seen in the families Chironomidae and Simuliidae, when analyzed separately, but not in Empididae, Tipuli- dae, Athericidae or Ephydridae. The relationship between day of winter and insect abundance paral- leled the same relationship found between day of winter and insect abundance in the drift (with frazil slush). There was however, no significant (p > 0.05) relationship between water temperature and abun- dance of aquatic invertebrates on the substrate over the winter. Discussion There are many variables in the natural riverine environment that influence aquatic invertebrate dis- tribution from year to year. Natural disturbances such as summer flooding, winter freezing and spring thawing, moving ice, and rapidly fluctuating temper- ee 2001 400 300 = S Number of aquatic invertebrates / m3 100 0 20 40 60 80 Day of winter FIGURE 1. (a) Regression plot of the number of aquatic invertebrates collected in the drift on days when frazil ice was present from 21 January to 17 March 1998 and from 31 December 1998 to 23 February 1999 (shown as triangles) in the Grand River, Ontario. The regression analysis includes only data from the first winter. (b) Regression plot of the number of aquatic invertebrates collected in the drift on days when frazil slush was absent from 16 December, 1998 to 8 March 1999 in the Grand River, Ontario. 16 December was designated as Day 1 of the winter for both seasons. Note that the scale of the y axis varies between panels. atures play important roles in the distribution and survival of aquatic invertebrates (Miller and Stout 1989). Anchor ice and frazil slush were the main focus of this study and appear to be important factors in invertebrate dispersal. Since the greatest abun- dance of immature aquatic insects occurs during the winter season (Maciolek and Needham 1952; Reimers 1957), any conclusions about the dispersal or removal of aquatic invertebrates by anchor ice are MARTIN, BROWN, BARTON, AND POWER: STREAM INVERTEBRATES 71 ‘E oS 12 ~~ — 2 r2 = 0.60 =] p<0.05 7 iS 10 ms n = aur n 2 "4 i) Xa) oO 5 [ > 6 NY & 2 =] =} 4 ion ne Ice covered period ° = oO 2 21 Om _“— ¢ =) = ¢ = 0 0 20 40 60 80 Day of winter FIGURE 2. Regression plot of mean (+ SD) number of aquatic invertebrates found on the substrate (organ- isms / cm?) on different days from 16 December 1998 to 8 March 1999 in the Grand River, Ontario. Presence of frazil slush and anchor ice at the sam- pling site (diamonds) and presence of a surface ice cover (a solid line) in the pool upstream of the sam- pling site are also indicated. important in understanding invertebrate and fish ecology. Anchor Ice The importance of anchor ice to benthic organisms is not due to low water temperatures (near 0.0°C) as many aquatic invertebrates are capable of surviving temporary freezing (Olsson 1981; Andrews and Rigler 1985) but due to its mechanical force on the substrate. As frazil ice crystals adhere to the substrate overnight or as the anchor ice releases, benthic organ- isms are at risk of physical damage and entrapment. Some studies have shown that invertebrates are able to survive entrapment in the ice with little or no mor- tality (Brown et al. 1953; Benson 1955). We also observed living invertebrates in the melted frazil slush samples. More importantly, our results show that sequential anchor ice events can change the local benthic population density. On days when anchor ice was released from the streambed, the density of invertebrates in the drift was much larger than when no anchor ice was released. The frequency of frazil and anchor ice formations depend on climate. For these formations to occur, sub-freezing air temperatures are required but a solid surface ice cover must not be present. Freshwater ice cover in Canada is highly variable, from periodic skims in southerly temperature regions to mean thicknesses over 2 m on high-latitude rivers (Prowse qe THE CANADIAN FIELD-NATURALIST Volt . @ Diptera Y = -0.051 X + 7.06 r=0.11 p<0.05 ® Y = -0.003 X + 0.43 Trichoptera 1 =0.05 p>0.05 Y = -0.0003 X + 0.02 2 12 =0.01 Coleoptera 7>0.05 Number of aquatic invertebrates on substrate / cm2 0 10 20 30 40 50 60 70 80 90 Day of winter Y =-0.002 X +0.23 e r = 0.06 Ephemeroptera os Y = -0.004 X + 0.33 4 12 = 0.03 Annelida p>0.05 Y = -0.001 X + 0.10 12 = 0.02 p>0.05 | Hydracarina 0 10 20 30 40 50 60 70 80 90 Day of winter FiGurE 3. Regression plot of the number of aquatic invertebrates in six different orders found on the substrate (organisms / cm?) on different days from 16 December 1998 to 8 March 1999 in the Grand River, Ontario. 1990). Mean freeze-over dates, mean ice thickness, and mean ice-free dates are provided by Allen (1977) and reviewed by Prowse (1990). In many of the colder parts of Canada, streams have a short freeze-up period followed by a fairly stable period of surface ice cover until spring break-up (Prowse 1995). This occurs in the northern part of the coun- try, and much of the prairie provinces (Beltaos 1997). In more moderate areas of Canada, such as the southern parts of Ontario, Quebec, and British 2001 Columbia, and the Atlantic provinces, ice covers remain for a shorter period of time and both spring and mid-winter ice break-ups are common (Beltaos 1997). The Grand River is representative of the win- ter environments expected in moderate areas of Canada. Thus, frazil and anchor ice formations can occur frequently during the periods when a solid sur- face ice cover is not present. The conditions in the Grand River were favourable for anchor ice forma- tion (water temperatures at or near 0.0°C and air temperatures below -5°C) on at least 42 nights dur- ing the winter of 1997-1998 and 57 nights during the winter of 1998-1999. The increased insect abundance in the drift may temporarily elevate the amount of food available for fish. Reimers (1957) observed trout feeding aggres- sively among floating frazil slush, suggesting that the organisms were being transported downstream with the frazil slush. It should not be inferred, how- ever, that this would increase the amount of forage available for fish because the organisms may be trapped in the floating frazil slush and not available for fish consumption. Benthic composition Most of the aquatic invertebrates found in our study area in the drift and on the substrate were Diptera, with Chironomidae, Tipulidae, and Simuli- idae, respectively, being the most numerous fami- lies. These results agree with other winter findings in temperate streams and rivers (Clifford 1978; Colbo 1979; Mills et al. 1981), although some stud- ies show Trichoptera as the most abundant order followed by Diptera (Brown et al. 1953; Benson 1955). In Newfoundland streams, Colbo (1979) observed that simuliid population density in a stream normally increases between December and April. However, in our study, the number of simuli- ids on the substrate decreased as the winter pro- gressed. Colbo (1950) also noted that certain Simuliidae larvae were absent or rare in zones where anchor ice occurs. This would compare favourably with our observations that the release of anchor ice removed larvae from the streambed and increased the number in the drift on days when frazil slush was present. In comparison, densities of Chironomidae, (another common family of the order Diptera) may be highest in December and January, then gradually decrease until May (Clifford 1978). Conversely, our results show that the number of Chironomidae decreased in the substrate throughout December and January. Although Clifford (1978) suggests that population density trends can be attributed to vari- ous phenomena such as the appearance of new gen- erations, delayed hatching, overlapping generations and available habitat, our observations indicate that this decrease in Chironomidae may be exacerbated by anchor ice. If so, this abrupt change in drifting MARTIN, BROWN, BARTON, AND POWER: STREAM INVERTEBRATES sb) density could be very important to fish species which prey on these invertebrates, particularly Chironomidae, which normally show a distinct peri- odic drifting pattern (McCafferty 1981). Further work to confirm these findings is advised, however, since a large amount of influence on the regression is due to the low invertebrate abundance on the sub- strate during the last winter sample. This study provides evidence that anchor ice events correlate with changes in the local density of stream invertebrates and may play an important role in the dispersal of aquatic invertebrates in the Grand River. Our finding that invertebrates were more abundant in the drift when frazil slush was present than when it was absent suggests that the anchor ice removed invertebrates from the streambed causing them to be carried off by the drift. In addition, the number of stream invertebrates in the drift and on the stream bottom was gradually reduced over the winter. This effect may not occur in all rivers. Anchor ice forms in turbulent areas which are not readily covered by surface ice. If such areas are dis- tributed along a river, invertebrates dislodged will resettle downstream and the overall effect on the stream benthos will be minimal. The largest impact from the removal of invertebrates may be areas directly below reservoirs, where invertebrates fall from the drift, preventing recolonization of areas which are depopulated by anchor ice. Warm water discharged from the dam precludes formation of anchor ice for a short distance downstream (depend- ing on air temperatures) reducing the supply of replacement organisms. Depletion of the benthic fauna over the course of the winter may be a fairly widespread phenomenon as most rivers throughout the world have impoundments. In future studies, it is recommended that inverte- brate densities and distribution in the substrate be sampled prior to the formation and immediately after the release of anchor ice. This type of distur- bance study (Underwood 1994) may provide more insight into the immediate influence of anchor ice. Additionally, the restrictions of winter sampling make it difficult to control each of the variables that usually influence invertebrate density. Comparative sampling of neighbouring streams, one with anchor ice and one without anchor ice (possibly below a hydroelectric dam) which have similar structure, flow, temperature, etc., would be advantageous. One ecological condition that has not been considered during this research, but that may prove significant, is how feeding habits of fish change in association with anchor ice events or throughout the winter. Feeding habits of fish could be monitored and any analogous changes in benthic composition could be identified. During winter, aquatic invertebrates are clearly exposed to highly variable environments which may have large impacts on their population 74 THE CANADIAN FIELD-NATURALIST densities. Thus this area of ecology deserves much more research. Acknowledgments This work was financed in part by a Natural Sciences and Engineering Research Council of Canada grant to Geoff Power. We thank Jason Darisi, Roy Wilhelm and Rick Doucett for their assistance with field work and identification of invertebrates. We also appreciate the cooperation of the Grand River Conservation Authority. Literature Cited Allen, W. T. R. 1977. Freeze-up, break-up and ice thick- ness in Canada. Report number CLI-1-77. Atmospheric Environment, Fisheries and Environment Canada, Downsview, Ontario. 185 pages. Andrews, D., and F. H. Rigler. 1985. The effects of an Arctic winter on benthic invertebrates in the littoral zone of Char Lake, Northwest Territories. Canadian Journal of Zoology 63: 2825-2834. Beltaos, S. 1997. Effects of climate on river ice jams. Proceedings of the 9th Workshop on River Ice, Fredericton, New Brunswick, 24 September 1997: 225-240. . Benson, N. G. 1955. Observations on anchor ice in a Michigan trout stream. Ecology 36: 529-530. — Brown, C. J. D., W. D. Clothier, and W. Alvord. 1953. Observations on ice conditions in the West Gallatin River, Montana. Proceedings of the Montana Academy of Sciences 13: 21-27. Butler, R. L., and V. M. Hawthorne. 1979. Anchor ice, its formation and effects on aquatic life. Science in Agriculture 26: 2. Clifford, H. F. 1978. Descriptive phenology and seasonali- ty of a Canadian brown-water stream. Hydrobiologia 58: 213-231. Colbo, M. H. 1979. Development of winter-developing Simuliidae (Diptera) in Eastern Newfoundland. Canadian Journal of Zoology 57: 2143-2152. Dall, P. C. 1979. A sampling technique for littoral stone dwelling organisms. Oikos 33: 106-112. Danks, H. V. 1971. Overwintering of some north temper- Vol} dis ate and arctic Chironomidae. II. Chironomid biology. Canadian Entomologist 103: 1875-1910. Logan, S. M. 1963. Winter observations on bottom organ- isms and trout in Bridger Creek, Montana. Transactions of the American Fisheries Society 92: 140-145. Maciolek, J. A., and P. R. Needham. 1952. Ecological effects of winter conditions on trout and trout foods in Convict Creek, California, 1951. Transactions of the American Fisheries Society 81: 202-217. McCafferty, W. P. 1981. Aquatic Entomology. Science Books International, Boston. 488 pages. Miller, M. C., and J. R. Stout. 1989. Variability of macro- invertebrate community composition in an arctic and subarctic stream. Hydrobiologia 172: 111-127. Mills, E. L., S. B. Smith, and J. L. Forney. 1981. The St. Lawrence River in winter: population structure, biomass, and pattern of its primary and secondary food web com- ponents. Hydrobiologia 79: 65-75. O’Donnell, D. J., and W.S. Churchill. 1954. Certain physical, chemical and biological aspects of the Brule River, Douglas County, Wisconsin. Brule River Survey Report number 11. Wisconsin Academy of Science, Arts and Letters, Transactions 43: 201-255. Olsson, T.1I. 1981. Overwintering of benthic macroinver- tebrates in ice and frozen sediment in a north Swedish river. Holarctic Ecology 4: 161-166. Prowse, T. D. 1990. Northern hydrology: an overview. Northern Hydrology: Canadian Perspectives (Editors T. D. Prowse and C. S. L. Omanney), NHRI Science report number 1. National Hydrology Research Institute, Environment Canada, Saskatoon, Saskatchewan. 36 pages. Reimers, N. 1957. Some aspects of the relation between stream foods and trout survival. California Fish and Game 43: 43-69. Tsang, G. 1982. Frazil and anchor ice - a monograph. National Research Council of Canada, Subcommittee on Hydraulics of Ice Covered Rivers, Ottawa. 90 pages. Underwood, A. J. 1994. On beyond BACI: Sampling designs that might reliably detect environmental distur- bances. Ecological Applications 4: 3-15. Received 9 December 1999 Accepted 4 July 2000 Establishment of a Breeding Population of Canada Geese in Southern Quebect JEAN-FRANCOIS GIROUX!, JOSEE LEFEBVRE!, LUC BELANGER’, JEAN RODRIGUE? and STEPHANE LAPOINTE! ‘Université du Québec 4 Montréal, Département des Sciences biologiques, C.P. 8888, Succursale Centre-ville, Montréal, Québec H3C 3P8, Canada ' *Canadian Wildlife Service, 1141, route de l’Eglise, C.P. 10 100, 9° étage, Ste-Foy, Québec G1 V 4H5, Canada Giroux, Jean-Francois, Josée Lefebvre, Luc Bélanger, Jean Rodrigue and Stéphane Lapointe. 2001. Establishment of a breeding population of Canada Geese in southern Quebec. Canadian Field-Naturalist 115(1): 75-81. We documented the natural establishment of a breeding population of Canada Geese (Branta canadensis) on four islands located in the St. Lawrence River near Montreal. Number of nests increased from 3 in 1992 to 50 in 2000. We estimated that the population is currently growing at 36-46% per year. Over the years, nest success was high with 80% of the nests hatching at least one egg giving a Mayfield estimate of 66%. Morphometric measurements of geese nesting and molting in southern Quebec conform to those of Giant Canada Geese (B. c. maxima). Resightings of banded and neck-collared geese revealed that two birds marked in New York and one in Michigan, as part of resident flocks, have colonized the islands. Moreover, natal-fidelity of a female hatched on one of the islands indicates that the breeding population will continue growing. The colonization of the islands may result from the expansion of the resident populations of southern Ontario and eastern United States or from escapes from captive flocks. We also hypothesize that the establishment could be related to the use of the area by molt migrant resident geese that may return the following spring to breed. We recommend that man- agement actions be taken to prevent establishment of a larger population of summer resident geese in southern Quebec. Otherwise, we predict that problems like those caused by resident Canada Geese established in urban and suburban settings elsewhere will occur. Climatic changes in a near future may also incite some Canada Geese to winter in southern Quebec and this could exacerbate potential problems. Key Words: Canada Goose, Branta canadensis, population, control, breeding, molt migration, Quebec. Recent establishment of breeding populations of of small waterbodies and the lack of predators Canada Geese in southern Canada and central and ___ (including hunters whose activities are often limited eastern United States have resulted from the release in urban and suburban settings) are contributing to of captive birds or transplantation programs _ the establishment of breeding populations of Canada (Blandin and Heusmann 1974; Lumsden 1981; Geese (Conover and Chasko 1985; Conover and Hindman and Ferrigno 1990). These introductions Kania 1991; Converse and Kennelly 1994). In the have been very successful as demonstrated in south- Atlantic Flyway, Canada Geese breeding south of ern Ontario where numbers of Giant Canada Geese 47° N of latitude are considered resident and consist (B. c. maxima) had reached 350 000 birds in fall mainly of Giant and Western Canada Geese (B. c. 1998 (Dennis et al. in press). The same situation moffitti) (Hindman and Ferrigno 1990). Their popu- occurred in Europe (Madsen and Andersson 1990; lation was estimated at 1.1 million birds in spring Allan et al. 1995). Reasons for these introductions 1999 (Anonymous 1999°). Although referred to as included a desire to restore an extirpated sub- residents, non-productive Giant Canada Geese (sub- species like in southern Ontario, to increase hunting adults, non- breeders and failed breeders) may opportunities and to add wildlife to adorn parks and migrate north to molt (Abraham et al. 1999; J. country estates (Lumsden 1981; Madsen and _ Rodrigue, unpublished data). Andersson 1990). As well, attempts to reduce local Recent increases in the number of resident goose problems in one area have involved transplan- Canada Geese have given rise to nuisance problems tation of Canada Geese to other areas (Hindman and ___in many areas. Geese foul beaches, parks and golf Ferrigno 1990). Attractive feeding sites such as_ courses with droppings, overgraze lawns, cause lawns in parks or golf courses, where the presence agriculture damage and constitute aircraft hazards (Blokpoel 1976; Conover 1985; Conover and Chasko 1985; Hindman and Ferrigno 1990; Allan et tIdeas and recommendations presented in this paper reflect al. 1995; Ankney 1996; Anonymous 1999"). Human the personal point of view of the authors and do not repre- pathogens have also been isolated in Canada Goose sent the position of the Canadian Wildlife Service of feces (Graczyk et al. 1998). Although the contami- Environment Canada. nation by these pathogens is not easily diagnosed, *See Documents Cited section. the risk of infection is probably low (Anonymous 75 716 THE CANADIAN FIELD-NATURALIST Vol. 115 FiGuRE 1. Map showing location of the four Varennes islands (1), Les Tles de Contrecoeur National Wildlife Area (2) and the Dow Chemical plant at Varennes (3), Quebec. 1999*). There is also some concern that competition for food between molt-migrant Giant Canada Geese and the migratory sub-species (B. c. interior and B. c. canadensis) on the northern breeding grounds could affect the reproductive output of migratory populations (Ankney 1996; Abraham et al. 1999). Finally, genetic integrity of the different stocks could be threatened by pairing of resident and migratory geese (Wendt and Boyd 1990). Little is known about the natural establishment of Canada Geese outside their traditional breeding range. In this paper, we document establishment of a breeding flock of Canada Geese near Montreal in southern Quebec. It is unlikely that pre-settlement characteristics of southern Quebec resemble the prairie-like conditions characterizing Giant Canada Geese breeding habitat (Hanson 1965). We also report the origin and morphometric data of a sample of birds and characterize the molt migration of resi- dent Canada Geese to Quebec. We finally propose possible mechanisms by which this establishment occurred, discuss potential consequences, and sug- gest management strategies. Study Area and Methods We conducted our study on four islands near Varennes (45°40’ N, 73°27’ W) located within the St. Lawrence River, 15 km northeast of Montreal (Figure 1). The four islands included Grande ile (57.4 ha), Masta (9.4 ha), St-Patrice (16.6 ha) and fle-aux-Fermiers (28.1 ha) for a total of 111.5 ha. About 100 cows are pastured on the islands between June and November each year (Lapointe et al. 2000). A rotational grazing system was established in 1992: some pastures (36.8 ha) were improved by seeding Timothy (Phleum pratense), Yellow Sweet-clover (Meiilotus officinalis), Brome (Bromus inermis) and clover (Trifolium spp.). Other portions of the islands were converted as dense nesting cover (DNC) for ducks (20 ha) by seeding Reed Canary Grass (Phalaris arundinacea), Western Wheatgrass (Agropyron smithii) and Crested Wheatgrass (Agropyron cristatum). Finally, some portions were left as idle fields with no cows (39.1 ha) or as unim- proved pastures (15.6 ha), both dominated by Tufted Vetch (Vicia cracca), Redtop (Agrostis alba), Red Fescue (Festuca rubra) and Kentucky Bluegrass (Poa pratensis). No trees or shrubs are found on the islands as a result of recurrent grazing by cattle. There are two inland marshes (<5 ha) that are per- manent and a few temporary ponds that become dry in summer. Emergent vegetation in marshes consists 2001 primarily of Narrow-leaved Cattails (Typha angusti- folia) and scattered clumps of Giant Bur-reed (Sparganium eurycarpum) and_ arrowhead (Sagittaria spp.). More details about the study area and the management practices are given by Lapointe et al. (2000). All portions of the islands were visited during the breeding period every 10-15 days in 1992-1994 and 1999-2000 and every 3-5 days in 1996-1998. No nest search was conducted in 1995. A nest was defined as a bow! with one or more eggs or with fresh down. Each nest was marked at 5 m with a plastic pole and located on aerial photos (1:10 000). The presence of the female, the number of eggs and their incubation stage, determined by flotation (Walter and Rusch 1997), were noted when the nest was first located. During our regular visits to the islands, each nest was inspected when the pair was absent or after the expected hatching date. A nest was considered successful if at least one egg hatched. The identity and origin of some birds nesting on the islands was established by approaching the nest and using a spot- ting scope and either reading the code on neck collars (Hestbeck 1995) or the number on the USFWS alu- minum bands. In July 1999, flightless geese were captured by driving them toward a funnel-shaped trap on the Varennes islands and at a nearby molting area on the already fenced grounds of the Dow Chemical plant at Varennes (Figure 1). The birds were aged (adult and juvenile), sexed by cloacal examination and leg banded. Morphometric measurements taken by the same observer (JR) included skull, culmen and total tarsus length (Dzubin and Cooch 1992). We used unpublished discriminant functions developed by T. J. Moser (US Fish and Wildlife Service, personal communication) for differentiating Canada Geese of the Atlantic Population that breed between Ungava Bay and the east coast of Hudson Bay from resident Canada Geese of the Atlantic Flyway. The equations for males and females were Y = -87.961 + 0.463 cul- men +0.499 skull and Y = -86.328 + 0.581 culmen +0.476 skull, respectively. Values greater than 0 for Y represent resident B. c. maxima. The chronology of molt migration of 1130 Canada Geese in southern Quebec was established in the summer of 1991 at Les Iles de Contrecoeur National Wildlife Area, 25 km downstream from Varennes GIROUX, LEFEBVRE, BELANGER, RODRIGUE AND LAPOINTE: BREEDING CANADA GEESE Th (Figure 1). Each day, five observers recorded the number of geese flying over the area or staging on water. Results We found 136 nests during our study: 3 in 1992, 4 in 1993, 5 in 1994, 10 in 1996, 17 in 1997, 21 in 1998, 26 in 1999 and 50 in 2000. Using the exponen- tial model N, = N,) e", we estimated a growth rate OQeu=e* sot Al with ar95 %. ClLiof 1.36-1.46 (R?= 0.99; F, ,= 498; P < 0.0001). Over the years, Canada Geese initiated nests between 25 March and 17 May with 70% of them between 10 and 25 April. Clutch size averaged 5.5 + 0.14 (SE) with a mode and median of 6 eggs (n = 125). Apparent success of 129 nests with known fate was 80% with a Mayfield nest success of 66%. Desertion and predation occurred at 5 and 12 nests, respectively. All preyed-upon nests were emptied with no apparent sign of predators, although one Red Fox (Vulpes vulpes) was observed on the islands in 1999. Human disturbance, including egg removal, may have caused some predation and desertion. Three nests were flooded in 2000 during an unusual rise of the St. Lawrence River water levels in early May. Finally, infertile eggs had been laid in six nests that were abandoned after the normal incubation period. Since 1996, 64 (53%) nests were established in the idle portions of the islands, 29 (24%) in dense nesting cover, 13 (11%) in unimproved pastures and 15 (12%) in improved pastures. Number of nests in the four habitats differed from expected based on the area covered by each habitat (x* = 29.898; df = 3; P< 0.01). There were proportionally more nests in idle fields and less in improved pastures. Nine previously-marked birds were identified at Varennes. Two birds had been banded in July as adults (After Hatching Year) when molting at Wilson Hill Wildlife Management Area near Mas- sena, New York, with resident geese. One of these birds banded in 1991 nested at Varennes every year of our study since 1994. One female, hatched on or near the Varennes Islands and banded as a local bird in 1993, nested successfully in 1997 and 1998. Another female that also nested for four consecutive years (1997-2000) was banded in winter 1995 as an adult near Albany, New York, where both resident TABLE 1. Mean morphological measurements (mm) of Canada Geese captured in July 1999 in southern Quebec. Culmen Sex x SE Male (n = 41) 60.3 0.47 Female (n = 41) SRS) 0:37 Total tarsus Skull % SE x SE jes) 0.71 IS. 0.69 108.5 0.60 124.4 0.49 78 THE CANADIAN FIELD-NATURALIST and migrant geese winter. This goose was later observed in February 1996 at Long Island, New York (J. Hestbeck, Massachusetts Cooperative Fish and Wildlife Research Unit, personal communica- tion). A female that initiated a nest during two suc- cessive years had been banded as a local in 1992 ina flock of resident Giant Canada Geese at Lower Brace Lake near Marshall, Michigan (G. Belyea, Michigan Department of Natural Resources, person- al communication). Finally, four geese (2 males and 2 females) banded in July 1999 at Varennes were observed breeding on the islands in 2000. In 1991, more than 95% of molt-migrant geese were recorded during the first three weeks of June. This pattern was subsequently observed each year in the Contrecoeur/Varennes area (J.-F. Giroux and J. Lefebvre, unpublished data). Moreover, since about 1995, about 50—200 geese spent part of the summer and early fall only 8 km from the Varennes islands (N. Rouette, Dow Chemical Canada Inc., personal communication). These birds molt within the fenced grounds of the Dow Chemical plant surrounded by a lawn with a pond. There were no differences in any of the three mea- surements between adult birds captured in 1999 on the nesting islands and those at the nearby molting area (P > 0.05). The mean measurements obtained on these birds (Table 1) conform to the measurements given for B. c. maxima for the Atlantic Flyway resi- dent population (T. J. Moser, personal communica- tion) and for resident birds from southern Ontario (Merendino et al. 1994). Using discriminant func- tions developed by T. J. Moser, only two birds (2.4%) did not fit the measurements of Giant Canada Geese. These were two males, both conforming to Atlantic Population migrant birds. The three previ- ously-banded birds that nested on the Varennes islands in 1999 were captured and identified as B. c. maxima. Discussion There are approximately 275 islands in the St. Lawrence River and in its tributaries near Montreal (Figure 1). Canada Geese became established on other islands near Varennes because 11 nests were found during sporadic visits there since 1992 (L. Bélanger and J. Lefebvre, unpublished data). Moreover two broods were observed in 1999 at Les Iles de Contrecoeur National Wildlife Area (Figure 1; J. Lefebvre, unpublished data). Three Canada Goose nests were also initiated in 1998 and 1999 on the grounds of the Dow Chemical plant (N. Rouette, per- sonal communication). The breeding population of resident Canada Geese in the St. Lawrence River low- lands of southern Quebec (approximately 30 540 km? excluding the river itself) was estimated at 710 + 162 indicated breeding pairs during helicopter surveys conducted in 1998-1999 (D. Bordage, Canadian Vol. 115 Wildlife Service, personal communication). In addi- tion, 195 pairs were estimated on the islands and the shore of the St. Lawrence River (approximately 1300 km? between Cornwall and Matane) in 1990-1992 (D. Bordage, personal communication). This gives a minimum estimate of 900 pairs for south- em Quebec. The carrying capacity of southern Quebec for Canada Geese is unknown, but many safe nesting islands and farm ponds, availability of feeding sites and limited hunting opportunities, in suburban set- tings, are ideal conditions for growth of a breeding population of resident Canada Geese (Conover and Chasko 1985; Conover and Kania 1991; Converse and Kennelly 1994). The 36-46% annual growth rate observed on the Varennes islands may not be repre- sentative for the entire southern Quebec population because islands are attractive nesting sites for Canada Geese. Nevertheless, high natality resulting from large clutch size and high nesting success, faithfulness of females to the nesting islands, and natal-site fidelity will undoubtedly contribute to pop- ulation expansion (Allan et al. 1995). Based on Ettl’s (1993) data for 15 urban goose populations, which showed a good fit (R*>0.5) to an exponential growth function, we calculated an average annual growth rate (A) of 18 + 2% which means that these populations double every 5—6 years. Ankney (1996) reported that the resident population of Canada Geese in southern Ontario also doubles every five years. The Canada Goose population of southern Quebec is increasing at a comparable or even greater rate. Given an 18% growth rate, the population would reach > 20 000 breeding pairs in 20 years. The observation of two nesting birds that had been banded in New York, 140 km from Varennes, and one from Michigan, indicates that the significant increase in numbers of resident geese in southern Ontario and eastern United States could have lead to the extension of breeding populations into southern Quebec. Birds escaped from captive flocks are also a potential source for establishment of breeding flocks. There are about 300 permit holders of captive birds in Quebec who reported owning nearly 3000 Canada Geese in 1995 (G. Gagnon, Environment Canada, personal communication). Most migratory Canada Geese have left southern Quebec by mid-May (Cazelais 1992; Chabot and St- Hilaire 1996*). Therefore, birds passing through the area during the first weeks of June are molt migrants (Abraham et al. 1999). Measurements of geese nest- ing and molting at Varennes confirm that all but a few belong to B. c. maxima. Giant Canada Geese that previously came or passed through southern Quebec during their molt migration may have returned there to found the newly established breeding population. This hypothesis, however, needs to be confirmed by the marking of molt migrant geese in Quebec. 2001 Band recoveries and recaptures of birds, banded during summer in southern Ontario and northeastern United States, confirm the presence of molt migrants along the Saguenay River near St. Fulgence (48°27'N, 70°54’ W) and in northern Quebec between 52 and 62° N (J. Rodrigue, unpublished data; Hughes and Reed 1999*). If our hypothesis on the establishment of breeding populations by molt migrants is valid, colonization by breeding Giant Canada Geese of these more northern areas could be possible. The so-called resident Canada Geese may then become fully migratory and could jeopardize genetic integrity of migrant flocks of B. c. interior and B. c. canadensis (Wendt and Boyd 1990). The increase in breeding Canada Geese in south- ern Quebec will likely augment the number of molt migrants to northern Quebec. Competition for food with migratory Canada Geese, especially during the brood-rearing period, could affect reproductive out- put of B. c. interior and B. c. canadensis (Ankney 1996; Abraham et al. 1999). Many islands of the St. Lawrence River are used for cattle grazing (Bélanger and Lehoux 1995). On the Varennes Islands, however, cattle do not affect Canada Geese because most of the nests have hatched before cows are brought to the islands in late May/early June. Restricting cattle to portions of the islands results in idle fields that seem favored by Canada Geese. Lapointe et al. (2000) found more residual cover (dead vegetation) in idle fields and DNC than in other treatments. On the other hand, idle fields had less standing (live) vegetation than DNC, which would provide better visibility for nest- ing geese. Abandonment of these management prac- tices to decrease the attractiveness of the islands for geese would not be justified considering that their benefits for nesting ducks (Lapointe et al. 2000) and songbirds (Lavallée 1998) far outweigh the current problems caused by Canada Geese. Although most Canada Geese are not able to win- ter in southern Quebec because of the rigorous cli- mate, the situation might change in the future with global warming. Nevertheless, we anticipate that the population expansion will cause problems such as those seen in eastern and central United States and southern Ontario (Conover 1985; Conover and Chasko 1985; Hindman and Ferrigno 1990; Ettl 1993; Ankney 1996). Golf courses, parks, beaches and agricultural areas are found all along the St. Lawrence River in southern Quebec. Only a few complaints about nuisance resident Canada Geese have so far been filed but the number increased in 1999 (G. Paquin, Environment Canada, personal communication). Waiting to manage local popula- tions until numbers reach nuisance levels is not desirable (Ettl 1993). Reactive management plans are less likely to be cost-effective and politically acceptable than are preventive ones. We therefore GIROUX, LEFEBVRE, BELANGER, RODRIGUE AND LAPOINTE: BREEDING CANADA GEESE 719 recommend that management measures be taken to prevent further expansion of the population to avoid problems often associated with resident Canada Geese. This supports the recommendations of the Atlantic Flyway Council for Québec (Anonymous 1999*),. Further surveys should be conducted to locate pre- cisely concentrations of birds and to follow the demographic evolution of the resident Canada Geese. This will help to identify areas where inter- ventions should be directed. Consultation should be initiated with local stakeholders such as hunters, bird watchers, conservationists and municipal representa- tives to discuss population objectives and manage- ment strategies. Several methods have been developed to restrain population growth such as relocation, harassment, egg spraying, male sterilization and scaring with dogs (Conover and Chasko 1985; Aguilera et al. 1991; Converse and Kennelly 1994; Christens et al. 1995; Smith et al. 1999; Castelli and Sleggs 2000). Simulation models, however, have revealed that har- vesting a large proportion of geese every year (e.g. culling molting birds) is the most efficient technique to reduce numbers of nuisance Canada Geese (Ettl 1993). This method does not usually receive general approbation by the public and less so by animal rights groups, but public opinion may be changing (Loker et al. 1999). For southern Quebec, we first suggest liberalizing the current early hunting season (6—24 September) restricted to Canada Geese in agricultural lands. Increasing bag limits and season length and allowing hunting in more areas should maximize harvest of molt migrants and resident breeding birds while a judicious timing of this season should minimize mortality of the Atlantic Flyway population of migrants. This approach, however, will remain limit- ed within urban and suburban areas for public safety reasons. A second suggestion is to prevent Canada Geese from breeding successfully by spraying eggs with mineral oil or addling eggs. Nest sites can be easily located in spring because males often remain close to incubating females. Preventing hatching may be more politically acceptable than direct control of the birds. It might also be cost-effective for the St. Lawrence River islands because the population is still at a manageable level. However, considering that 80% of indicated breeding pairs are located else- where, the effectiveness of reducing the breeding success of geese on islands may be limited, and we may require a more extensive and expensive control program. In conclusion, there is not a single strategy that will be totally effective in controlling the number of geese that will be economically feasible and accept- able by the majority of stakeholders. Reducing 80 THE CANADIAN FIELD-NATURALIST reproductive output of the summer resident popula- tion of Canada Geese coupled with a more liberal early hunting season on both molt migrants and resi- dents may maintain numbers at an acceptable level and minimize nuisance problems on public and pri- vate property. If these strategies fail, other alterna- tives including direct control of molt migrants and breeding residents (ex. food shelf programs) would have to be considered. Acknowledgments We are grateful to F. St-Pierre for his continuous assistance with fieldwork and to R. Angers, A. Béchet, P. Brousseau, F. Demers, P. Messier and J. Olson for the banding operation. N. Rouette gave us access to the molting ground on the Dow Chemical Canada property. This study was financially support- ed by the Canadian Wildlife Service through the Eastern Habitat Joint Venture and by the Université du Québec a Montréal. We thank J. Hestbeck and G. Belyea for their data on the banded geese and T. J. Moser for measurements of Atlantic Flyway Canada Geese. G. Paquin provided information on com- plaints received by Environment Canada about nui- sance Canada Geese. We finally thank D. Bordage, P. Castelli, A. J. Erskine, J. Hughes and M. Laperle for their comments on the manuscript. Documents Cited (marked * after date in text) Anonymous. 1999. Atlantic Flyway resident Canada Goose Management Plan. Report prepared by the Can- ada Goose Committee of the Atlantic Flyway Technical Section. Atlantic Flyway Council. 42 pages. Chabot, D., and D. St-Hilaire. 1996. Quelques aspects de la halte migratoire printaniére de la Bernache du Canada dans |’Outaouais. Gouvernement du Québec, Ministére de l’Environnement et de la Faune, Direction régionale de l’Outaouais, Service de l’aménagement et de |’ exploi- tation de la faune, Hull. 29 pages. Hughes, J., and A. Reed. 1999. Breeding ground banding of Atlantic Population Canada Geese in northern Quebec — 1998. Unpublished report submitted to the Atlantic Flyway Council and Technical Section, 10 pages. Literature Cited Abraham, K. F., J. O. Leafloor, and D. H. Rusch. 1999. Molt migrant Canada Geese in northern Ontario and western James Bay. Journal of Wildlife Management 63: 649-655. Aguilera, E., R. L. Knight, and J. L. Cummings. 1991. An evaluation of two hazing methods for urban Canada Geese. Wildlife Society Bulletin 19: 32-35. Allan, J.R., J. S. Kirby, and C. J. Feare. 1995. The biology of Canada Geese Branta canadensis in relation to the management of feral populations. Wildlife Biology 1: 129-143. Ankney, C. D. 1996. An embarrassment of riches: too many geese. Journal of Wildlife Management 60: 217-223. Bélanger, L., and D. Lehoux. 1995. Use of various habi- ‘tat types by nesting ducks on islands between Montreal Vol. 115 and Trois-Rivieres. Canadian Wildlife Service Occa- sional Paper 87: 1-26. Blandin, W. W., and H. W. Heusmann. 1974. Establish- ment of Canada goose populations through urban gos- ling transplants. Transactions of the Northeast Section of the Wildlife Society 31: 83-100. Blokpoel, H. 1976. Bird hazards to aircraft. Clarke, Irwin & Company Limited. Toronto, Ontario. Castelli, P. M., and S. E. Sleggs. 2000. Efficacy of border collies to control nuisance Canada geese. Wildlife Society Bulletin 28: 385-3972. Cazelais, S. 1992. Ecologie de la Bernache du Canada lors de sa halte migratoire printaniére dans la région du Lac St-Pierre, Québec. M.Sc. thesis, Université du Québec 4 Montréal, Montréal, Quebec. 86 pages. Christens, E., H. Blokpoel, G. Rason, and S. W. D. Jarvie. 1995. Spraying white mineral oil on Canada goose eggs to prevent hatching. Wildlife Society Bulletin 23: 228-230. Conover, M. R. 1985. Alleviating nuisance Canada goose problems through methiocarb-induced aversive condi- tioning. Journal of Wildlife Management 49: 631-636. Conover, M. R., and Chasko G. G. 1985. Nuisance Can- ada goose problems in the eastern United States. Wildlife Society Bulletin 13: 228-233. Conover, M. R., and G. S. Kania. 1991. Characteristics of feeding sites used by urban-suburban flocks of Can- ada Geese in Connecticut. Wildlife Society Bulletin19: 36-38. Converse, K. A., and J. J. Kennelly. 1994. Evaluation of Canada Geese sterilization for population control. Wildlife Society Bulletin 22: 265-269. Dennis, D. G., N. R. North, and H. G. Lumsden. In press. Range expansion and population growth of giant Canada Geese in southern Ontario: benefits, drawbacks, and man- agement techniques. Jn Towards Conservation of the Diversity of Canada Geese (Branta canadensis). Edited by K.M. Dickson. Canadian Wildlife Service Occasional Paper Number 103. Dzubin, A., and E. G. Cooch. 1992. Measurements of geese: general field methods. California Waterfowl As- sociation, Sacramento. 20 pages. Ettl, G. J. 1993. A management model for urban Canada goose populations. Northwest Environmental Journal 9: 73-84. Graczyk, T. K., R. Fayer. J. M. Trout, E. J. Lewis, C. A. Farley, I. Sulaiman, and A. A. Lal. 1998. Giardia sp. cysts and infectious Cryptosporidium parvum oocysts in the feces of migratory Canada Geese (Branta canadensis). Applied and Environmental Microbiology 64: 2736-2738. Hanson, H. C. 1965. The Giant Canada Goose. Southern Illinois University Press, Carbondale. 226 pages. Hestbeck, J. B. 1995. Population study and management of Atlantic Flyway Canada Geese. Journal of Applied Statistics 22: 877-890. ; Hindman, L. J., and F. Ferrigno. 1990. Atlantic Flyway Goose Population: status and management. Transactions of the North American Wildlife and Natural Resources Conference 55: 293-311. Lapointe, S., J.-F. Giroux, L. Bélanger, and B. Filion. 2000. Benefits of rotational grazing and dense nesting cover for island-nesting waterfowl in southern Quebec. Agriculture, Ecosystems and Environment 78: 261-272. Lavallée, J. 1998. Effets de l’aménagement du couvert végétal pour la nidification de la sauvagine sur la repro- 2001 duction et l’alimentation des oisillons du Goglu des prés dans les files de Varennes, Québec. M.Sc. thesis, Université du Québec 4 Montréal, Montreal, Quebec. 92 pages. Loker, C. A., D. J. Decker, and S. J. Schwager. 1999. Social acceptability of wildlife management actions in suburban areas: 3 cases from New York. Wildlife Society Bulletin 27: 152-159. Lumsden, H. G. 1981. History of breeding Canada Geese (Branta canadensis) in southwestern Ontario. Ontario Field Biologist 35: 49-56. Madsen J., and A. Andersson. 1990. Status and manage- ment of Branta canadensis in Europe. [WRB Special Publication 12: 66-69. Merendino, M. T., C. D. Ankney, D. G. Dennis, and J. O. Leafloor. 1994. Morphometric discrimination of GIROUX, LEFEBVRE, BELANGER, RODRIGUE AND LAPOINTE: BREEDING CANADA GEESE 81 Giant and Akimiski Island Canada Geese. Wildlife Society Bulletin 22: 14-19. Smith, A. E., S.R. Craven, and P.D. Curtis. 1999. Managing Canada Geese in urban environments. Jack Berryman Institute Publication 16, and Cornell Uni- versity Cooperative Extension, Ithaca, New York. 42 pages. Walter, S. E., and D. H. Rusch. 1997. Accuracy of egg flotation in determining age of Canada goose nests. Wildlife Society Bulletin 25: 854-857. Wendt, J. S., and H. Boyd. 1990. Goose management in Canada. Transactions of the North American Wildlife and Natural Resources Conference 55: 333-337. Received 24 January 2000 Accepted 20 July 2000 Edge-related Nest Predation Associated With the Retention of Residual Trees in Harvested Hardwood Stands RICHARD H. YAHNER!, AMANDA D. RODEWALD2, and SUSAN C. TALBoTT23 ‘School of Forest Resources and Intercollege Graduate Degree Program in Ecology, The Pennsylvania State University, University Park, Pennsylvania 16802, USA 2Intercollege Graduate Degree Program in Ecology, The Pennsylvania State University, University Park, Pennsylvania 16802, USA 3Present address: Eastern Neck National Wildlife Refuge, 1730 Eastern Neck Road, Rock Hall, Maryland 21661, USA Yahner, Richard H., Amanda D. Rodewald, and Susan C. Talbott. 2001. Edge-related nest predation associated with the retention of residual trees in harvested hardwood stands. Canadian Field-Naturalist 115(1): 82-87. We examined edge-related predation in June—August 1998 on artificial ground nests associated with the retention of residu- al trees in harvested oak (Quercus spp.)-hickory (Carya spp.) stands in a state in Pennsylvania. We tested the hypothesis that predation rates on nests did not vary with distance from edge in retention stands and adjacent unharvested forest stands. Artificial nests were placed during five time periods (trials) in three representative retention stands in Rothrock State Forest along random transects at four distances perpendicular to the edge: 50 m into the retention stand, at the edge, 50 m into the adjacent forest, and 150 m into the adjacent forest. Artificial nests consisted of three large brown chicken (Gallus gallus) eggs to simulate nests of Ruffed Grouse (Bonasa umbellus) and Wild Turkey (Meleagris gallopavo). Twenty-three % (82/358) of the nests were disturbed during the five trials combined. Nest fate was not influenced by the distance of nests from the edge (P = 0.88). We also found an interaction between trial and site (P = 0.01); in particular, one site (Cooper’s Gap) had much higher predation rates during trials 3-5 when abundances of American Crows (Corvus brachyrhynchos) were higher at the site. Predation rates observed in our study were comparable to those found elsewhere in other harvested forests of Pennsylvania, where the abundance of crows is known to be relatively low. Nest predation rates in retention stands probably did not vary with distance from edge because nests were well concealed by a dense layer of ground-level vegetation regardless of distance from an edge. We hypothesize that variation in predation rates among sites may be explained by differences in the degree of heterogeneity, i.e., extent of edges and amount of habitat contrast, within and among adjoining habitats. We conclude that habitat conditions created by retention stands do not have a major negative effect on the nesting success of common, ground-nesting gallinaceous birds. Key Words: artificial nest, bird, clearcutting, edge, even-aged management, forest, landscape, Pennsylvania, predation, retention stands. Edge-related nest predation in fragmented land- has become a controversial practice for ecological scapes has been well studied (see review by Paton and aesthetic reasons in much of the eastern decidu- 1994). Predation rates in forested tracts surrounded by — ous forest (e.g., King et al. 1998; Yahner 2000). agricultural landscapes are generally higher on nests Harvesting practices, which retain individual trees, located near edges compared to those nests located snags and small patches of trees within stands (termed farther from edges; however, relationships between __variable-retention systems) are increasingly used nest predation rates and distance from edge are much across North America in an effort to meet forestry and less likely to occur in managed forested landscapes, ecological objectives (Franklin et al. 1997). perhaps because of differences in predator communi- In 1992, the Pennsylvania Bureau of Forestry initi- ties between these two landscapes (Andrén 1995; ated a new practice to replace clearcutting on state Donovan et al. 1997; Hartley and Hunter 1998; forests, which is termed “even-aged reproduction with Yahner 2000; but see King et al. 1998). Some studies reservation guidelines” (hereafter termed retention in forested landscapes have shown that incidences of | stands, Boardman and Yahner 1999; Rodewald and nest predation are associated with the degree of het- | Yahner 2000). This new practice permanently retains erogeneity; i.e., extent of edges or amount of habitat relatively high densities of live trees and snages in contrast, within and among adjoining habitats in the multiple crown and size classes of both commercially local landscape rather than distance from edge and non-commercially important tree species. In addi- (Yahner and Scott 1988; Yahner et al. 1989; Sargent tion, the retention of trees produces gradual rather et al. 1998). than abrupt edges (after Suarez et al. 1997) at the Harvesting practices, such as clearcutting, not only _ interface of the cut and the adjacent unharvested forest create edge but also markedly change vegetative (R.H. Yahner, personal observation). Hence, state structure compared to unharvested adjacent habitat in forests in Pennsylvania provide an excellent opportu- the local landscape (e.g., Yahner 2000). Complete nity to examine how the retention of residual trees removal of overstory trees via clearcutting, however, affects nesting birds. 82 2001 Because the extent of state forest land in Pen- nsylvania is substantial (850 000 ha) and represents some of the largest remaining forested tracts in the eastern United States, an examination of edge-relat- ed nesting success in retention stands is needed. Furthermore, given the projected wide application of variable-retention systems in North America (Frank- lin et al. 1997), and understanding of their edge- related effects is important for avian conservation efforts. In our study, we tested the hypothesis that predation rates on artificial nests simulating those of common ground-nesting species of gallinaceous birds did not vary with distance from edge in reten- tion stands in harvested forested landscapes of Pennsylvania. Study Area and Methods Our study was conducted in harvested oak-hickory stands in Rothrock State Forest, Huntingdon County, Pennsylvania. This state forest is located in the Valley and Ridge Province, which contains forest stands from 70—90 years old (Brooks and Birch 1988). Principal overstory trees are Northern Red Oak (Quercus rubra), White Oak (Q. alba), Chestnut Oak (Q. prinus), hickories (Carya spp.), Red Maple (Acer rubrum), Black Cherry (Prunus serotina), and Black Gum (Nyssa sylvatica) (Rodewald and Yahner 2000). Retention stands have been established in Rothrock State Forest in lieu of clearcutting since 1992 (Boardman and Yahner 1999; Rodewald and Yahner 2000; Yahner 2000). This practice has the following guidelines: (1) retain an average basal area of at least 23-46 m2/ha over the entire treatment area and > 12 trees/ha in the dominant, codominant, and/or interme- diate class; (2) retain tree species representative of the pre-harvest stand; (3) retain all crown classes; (4) maintain natural spacing and consider leaving small groups of residual trees; and (5) select residual trees based on structural and species diversity. We chose three retention stands that were represen- tative of those created by the Bureau of Forestry in terms of size, shape, and age (Boardman and Yahner 1999; Rodewald and Yahner 2000). Turkey Hill and Cooper’s Gap stands were 38 ha each, and the Spruce Mountain stand was 28 ha; each stand was oblong in shape, was typically about 300 m wide and at least 900 m long, and followed the contour of the adjacent ridge. Turkey Hill, Cooper’s Gap, and Spruce Mountain were cut in 1993, 1994, and 1995, respec- tively. Stands were 2 to 15 km from each other, and amount of forest cover within 1.0km of each cut ranged from 80 to 82% (A.D. Rodewald and R. H. Yahner, unpublished data). The remaining cover types within this 1.0-km radius consisted of early-succes- sional and young forest, unimproved access roads, and scattered rustic cabins. There was no agricultural lands (cropland or pasture) were within the 1-km radius; distance of retention stands from agricultural lands varied from 1.9 to 3.1 km. YAHNER, RODEWALD, AND TALBOTT: EDGE-RELATED NEST PREDATION 83 We selected three random transects per retention stand for nest placement during each trial; distances between adjacent transects were at least 100 m. These transects were selected from random points located at 100-m intervals along the boundary of each retention stand; a different transect was selected for each trial. Transects were oriented perpendicular to the retention stand-unharvested forest interface. Each transect extended 50 m into the center of reten- tion stands but only 150 m into the adjacent forest because elevation, vegetation, and microclimate changed dramatically near the ridgeline in this area (Yahner and Smith 1990). Dense herbaceous vegeta- tion (69-87% coverage; S.C. Talbott and R.H. Yahner, unpublished data) near ground level at the retention stand-adjacent forest interface and tree retention produced a gradual edge at the interface. Thus, we determined the exact location of the edge for placement of transects on the basis of paint mark- ings remaining on large trees, which had served to delineate the boundary of the retention stand prior to its selective harvest. We placed artificial ground nests during five time periods (trials) from early June through early August 1998 (Yahner and Wright 1985; Yahner and Mahan 1996a). Artificial ground nests consisted of three fresh, brown Domestic Chicken (Gallus gallus) eggs placed in a shallow depression near a tree, log, stump, or rock (e.g., Yahner and Wright 1985; Yahner and Mahan 1996a). We wore rubber-soled boots and latex gloves to minimize human scent when placing nests (Nol and Brooks 1982). Large brown chicken eggs (light brown to buffy in color, 52 X 40 mm) were used to simulate nests of Ruffed Grouse (Bonasa umbellus; buffy, 39 X 30mm) and Wild Turkey (Meleagris gallopavo; pale buff or buffy white, 63 X 45 mm) (Harrison 1975; Yahner and Mahan 1996b). Ruffed Grouse and Wild Turkey are common, ground-nesting gallinaceous species within and near retention stands in Rothrock State Forest (Boardman and Yahner 1999). Trial length followed those of previous nest stud- ies using artificial ground nests (e.g., Yahner and Wright 1985; Yahner and Mahan 1996a), with a nest exposure of 6 days and 8 days elapsing between tri- als. During each trial, a nest was placed 15 m to each side of a transect at four distances from the edge: 50 m into the retention stand, at the edge, 50 into the adjacent forest, and 150 m into the adjacent forest. This experimental design gave 24 nests/stand in each of the five trials (n = 358 total nests; 2 nests were not relocated in trial 2). We determined the fate (undisturbed, disturbed by avian predators, or disturbed by an unknown preda- tor) after each trial. A disturbed nest had at least one broken or missing eggs. Nest appearance and mode of disturbance were used to classify predator type, e.g., peck holes suggested an avian predator (Rearden 1951; Yahner and Scott 1988; Hernandez et al. 84 1997), and other types of marks were attributed to unknown predators. Eggs and eggshell fragments were removed at the end of a trial. The relative abun- dance of potential avian nest predators, i.e., American Crows (Corvus brachyrhynchos) and Blue Jays (Cyanocitta cristata), were noted within 200 m of artificial nests at the time of placement and removal (Yahner and Scott 1988); time spent at a visit to a nest was about 5 minutes. A variety of other potential nest predators occurred in vicinity of the stands, e.g., included snakes, squirrels (Sciurus spp.), Eastern Chipmunks (Tamias striatus), Common Raccoons (Procyon lotor), Striped Skunks (Mephitis mephitis), and Black Bears (Ursus americanus) (R. H. Yahner, personal observation). We examined the dependency of nest fate (undis- turbed or disturbed) on distance from an edge (50 m into retention stand, edge [0 m], 50m into adjacent forest, or 150 m into adjacent forest), trial (1-5), and site (Turkey Hill, Cooper’s Gap, or Spruce Moun- tain) using a four-way test-of-independence (Dixon 1990). Likelihood ratios (G2) were used to examine interactions between nest fate and the other three variables based on log-linear models (Sokal and Rohlf 1995). Results Twenty-three % (82/358) of the artificial ground nests were disturbed during the five trials combined (Table 1). Only 7% (6/82) of the disturbed nests had evidence of avian predation, based on the presence of peck holes in eggs. Disturbance rates by American Crows, however, may have been underes- timated because crows are capable of carrying eggs away from an artificial nest (Montevecchi 1976); for instance, 50% (41/82) of the disturbed nests had at least one missing eggs, many of which may have been removed by crows. The relative abundance of American Crows ranged from 0.0/visit at Turkey Hill and 0.1/visit at Spruce Mountain to 1.7/visit at THE CANADIAN FIELD-NATURALIST Vole Cooper’s Gap, whereas the relative abundance of Blue Jays was 3.7/visit at Turkey Hill, 1.1/visit at Spruce Mountain, and 0.2/visit at Cooper’s Gap. Nest fate was not dependent on the distance of nests from an edge (G = 0.7, df = 3, P = 0.88); dis- turbance rates ranged from 20% at nests 150 m into the adjacent forest to 24% at nests at both the edge and 50 m into the forest (Table 1). Nest fate was associated with trial (G = 31.0, df = 4, P = 0.0000), being lowest in trial 2 (7% of total nests disturbed) and highest in trials 3 (42%) and 4 (29%) for all sites combined. Nest fate also was related to site (G = 24.4, df = 2, P = 0.0000); disturbance rates were much higher at Cooper’s Gap (38%) than at either Spruce Mountain (13%) or Turkey Hill (18%). Furthermore, disturbance rates were dependent on both trial and site considered concurrently (G = 22.1, df = 10, P = 0.01). In particular, disturbance rates at Cooper’s Gap averaged 54% (range = 33-71%) dur- ing trials 3-5 compared to an average of 15% (range = 0—38%) during other trial-site combinations. Discussion Nest Predation Rates and Predator Communities Predation rates observed in our study (23%) were comparable to those found elsewhere in other man- aged forests of Pennsylvania (22-24%) where abun- dance of avian predators, particularly American Crows, is low (Yahner et al. 1993; Yahner and Mahan 1996a). Moreover, predation rates in our study were similar to nest losses reported in studies of other gallinaceous birds, e.g., Spruce Grouse (Fal- cipennis canadensis franklinii; 17%) in boreal forests of Canada (Redmond et al. 1982). In contrast, when crow abundance is high, predation of artificial nests in forested landscapes is considerably higher (40-68%; Yahner and Scott 1988; Hogrefe et al. 1998). Hence, we attribute the greater predation rates at Cooper’s Gap partially to higher abundances of crows. For example, during trials 3 to 5, 2.0 TABLE 1. Number of disturbed artificial ground nests in relation to distance from edge, time period (trial), and site in Rothrock State Forest, Huntingdon County, Pennsylvania (U.S.A.), 1998. Variables Levels n % Distance from edge 50 m into retention stand 20 a2 at edge (0 m) 22 24 50 m into adjacent forest 2D 24 150 m into adjacent forest 18 20 Time period Trial 1 11 15 Trial 2 5 7 Trial 3 30 42 Trial 4 21 US) Trial 5 15 21 Site Cooper’s Gap 45 38 Turkey Hill 16 ig Spruce Mountain 21 18 2001 crows/visit were recorded at Cooper’s Gap compared to only 0.17 and 0.00 crows/visit at Spruce Mountain and Turkey Hill, respectively. A greater abundance of American Crows at Cooper’s Gap in July-early August (trials 3-5) may be the result of crows becoming more gregarious, with family groups using this particular retention stand as a foraging site in summer subsequent to fledging of young (Gross 1946; Yahner and Wright 1985). Predation rates in our study may have been differ- ent if our nests were baited with smaller eggs (e.g., Roper 1992; Haskell 1995). However, the experi- mental design of our study was to simulate nests of ground-nesting gallinaceous birds (Ruffed Grouse and Wild Turkey) in the region, thereby making brown chicken eggs a suitable substitute for actual eggs of these species (Yahner and Mahan 1996b). Use of these larger eggs likely precluded predation by smaller predators that might prey on smaller eggs of other ground-nesting birds in the retention stands, such as Eastern Towhee (Pipilo erythrophthalmus) (Haskell 1995; Blight et al. 1999). Nest Predation Rates in Relation to Edge As in most studies of artificial nests in forested landscapes (Yahner and Wright 1985; Rudnicky and Hunter 1993; Andrén 1995; Hartley and Hunter 1998), rates of nest predation in retention stands did not vary with distances from edges. Artificial ground nests in our study, regardless of distance from an edge, were well concealed by a dense layer of ground-level vegetation (usually > 70%, S.C. Talbott and R.H. Yahner, unpublished data), which likely reduced the foraging efficiency of predators and made these nests less susceptible to predation than aboveground (i.e., shrub) nests or those located in habitats with less dense vegetation near ground level (Bowman and Harris 1980; Sugden and Beyersbergen 1986; Yahner and Cypher 1987). Gradual edges created by ground-level vegetation and retention of trees of different size in retention stands probably reduced predation rates compared to higher rates that often occur at abrupt edges in forested landscapes (e.g., Suarez et al. 1997). Although predation rates on ground nests did not differ with distances from edges per se, rates per- haps varied with differences in the degree of hetero- geneity, i.e., extent of edges and amount of habitat contrast, within and among adjoining habitats in the local landscape. For instance, the Cooper’s Gap site, with the highest predation rates in our study, was characterized by considerable induced and inherent edges (after Forman and Godron 1976) compared to the other two sites. The extent of induced edge at the Cooper’s Gap site was increased by the retention of two 50-m wide, wooded corridors, which may have served as travel lanes or perching areas for for- aging predators (Bergin et al. 1997; Yahner and Mahan 1997). An induced edge also was created by Y AHNER, RODEWALD, AND TALBOTT: EDGE-RELATED NEST PREDATION 85 thinning in the adjacent forest along 10% of the downslope border of this retention stand. The inher- ent edge consisted of visible tree mortality caused by Gypsy Moth (Lymnatria dispar) defoliation along 20% of its upslope border prior to our study and a dense stand of Eastern Hemlock (Tsuga canadensis) along 5% of its downslope border. In contrast, Spruce Mountain and Turkey Hill stands had no wooded corridors and were proximal to rela- tively homogeneous stands of unharvested decidu- ous forest. Support for our contention that nest predation rates within and adjacent to retention stands were associated with the degree of heterogeneity comes from other artificial nest studies. Higher predation rates were noted on nests located in forested land- scapes affected by extensive clearcutting and con- taining considerable edge habitat compared to forest- ed landscapes with less clearcutting and, hence, reduced edge habitat (Yahner and Scott 1988; Yahner and Mahan 1996a). Higher predation rates on nests also were found in forested fragments sur- rounded by agricultural habitats (i.e., appreciable habitat contrast) than in fragments surrounded by unharvested pine [Pinus spp.] stands (i.e., relatively less habitat contrast; Sargeant et al. 1998). Further- more, an increasing number of studies are beginning to show that both localized and landscape-level fac- tors can have a profound effect on avian predation rates (e.g., Bayne and Hobson 1997; Donovan et al. 1997; Rodewald 2000). In particular, forest stnds within fragmented agricultural landscapes have a greater number of generalist predators (e.g., corvids) than within more forested landscapes (Andrén 1992; Haskell 1995; Bayne and Hobson 1997). A greater degree of heterogeneity in the land- scape surrounding retention stands, which results from more extensive edge habitat or contrast with adjoining habitats, may enhance the abundance of some nest predators. American Crows, for example, are adapted to edge habitats (Whitcomb et al. 1981), and their abundance would be expected to be higher in forested habitat containing considerable edge habitat (e.g., Yahner et al. 1993). Furthermore, abundance of mammalian predators, like Raccoon, typically increase with greater amounts of edge and habitat diversity in the landscape (Heske 1995). Management Implications Our study suggests that rates of nest predation on ground-nesting gallinaceous bird species (e.g., grouse and turkey) in variable retention harvests of hardwood stands may not be related to distance of nests from edges per se but rather be a function of habitat heterogeneity in the surrounding landscape. Thus, we recommend future research on the effects of both wooded corridors and habitat heterogeneity on nest predation rates in these local landscapes. We caution, however, that we have presented no data on 86 THE CANADIAN FIELD-NATURALIST predation rates of actual nests in retention stands to compare with our artificial nests. Although this study represents the first published study of nest predation in retention stands, our data suggest that habitat conditions created by these stands may not have a major negative effect on the nesting success of common, ground-nesting gallinaceous birds. Additional additional artificial-nest studies associated with variable-retention systems in North America are needed, especially those with similar experimental design in order to better detect temporal and spatial trends across treatments or over time (Yahner 1996; Sargeant et al. 1998; Wilson et al. 1998). Acknowledgments This study was funded by the Pennsylvania Experiment Station and the Pennsylvania Wild Resource Conservation Fund. We thank W. Cooper, H. Endy, J. Giocomo, A. Harpster, B. Harris, J. Kelligher, R. Kosoff, J. Mravintz, and M. Stitzer for field assistance and P. Rodewald for reviewing the manuscript. Literature Cited . Andrén, H. 1992. Corvid density and nest predation in relation to forest fragmentation: a landscape perspective. Ecology 73: 794-804. Andrén, H. 1995. Effects of landscape composition on predation rates at habitat edges. Pages 225-255 in Mosaic landscapes and ecological processes. Edited by L. Hanson, L. Fahrig, and G. Merriam. Chapman & Hall, London. Bayne, E.M., and K.A. Hobson. 1997. Comparing the effects of landscape fragmentation by forestry and agri- culture on predation of artificial nests. Conservation Biology 11: 1418-1429. Bergin, T. M., L. B. Best, and K. E. Freemark. 1997. An experimental study of predation on artificial nests in roadsides adjacent to agricultural habitats in Iowa. Wilson Bulletin 109: 437-438. Blight, L. K., J. L. Ryder, and D.F. Bertram. 1999. Predation on Rhinoceros Auklet eggs by a native poula- tion of Peromyscus. Condor 101: 871-876. Boardman, L. A., and R. H. Yahner. 1999. Wildlife com- munities associated with even-aged reproduction stands in two state forests of Pennsylvania. Northern Journal of Forestry 16: 89-95. Bowman, G. B., and L. D. Harris. 1980. Effect of spatial heterogeneity on ground-nest predation. Journal of Wildlife Management 44: 806-813. Brooks, R. P., and T. W. Birch. 1988. Changes in New England forests and forest owners: implications for wildlife habitat resources and management. Transactions of the North American Wildlife and Natural Resource Conference 53: 78-87. Dixon, W. J. Editor. 1990. BMDP statistical software. University of California Press, Berkeley. 1385 pages. Donovan, T.M., P. W. Jones, E. M. Annand, and F.R. Thompson III. 1997. Variation in local-scale edge effects: mechanisms and landscape context. Ecology 78: 2064-2075. Vol. 115 Forman, R.T.T., and M. Godron. 1976. Landscape ecology. John Wiley & Sons, New York. 619 pages. Franklin, J. F., D. R. Berg, D. A. Thornburgh, and J. C. Dappeiner. 1997. Alternative silvicultural approaches to timber harvesting: variable retention systems. Pages 111-139 in Creating a forestry for the 21st century — the science of ecosystem management. Edited by K. A. Kohn and J. F. Franklin. Island Press, Washington, D.C. Gross, A. O. 1946. Eastern Crow. Pages 226-259 in Life histories of North American jays, crows, and titmice. Edited by A.C. Bent. United States National Museum Bulletin, Number 191. Washington, D.C. Harrison, H. H. 1975. A field guide to birds’ nests of the eastern United States. Houghton Mifflin Co., Boston, Massachusetts. 257 pages. Hartley, M. J., and M. L. Hunter, Jr. 1998. Meta-analysis of forest cover, edge effects, and artificial nest predation rates. Conservation Biology 12: 465-469. Haskell, D. G. 1995. Forest fragmentation and nest preda- tion: are experiments with Japanese Quail eggs mislead- ing? Auk 112: 767-770. Hernandez, F., D. Rollins, and R. Cantu. 1997. Evalu- ating evidence to identify ground-nest predators in west Texas. Wildlife Society Bulletin 25: 826-831. Heske, E. J. 1995. Mammalian abundances on forest-farm edges versus forest interiors in southern Illinois: is there an edge effect? Journal of Mammalogy 76: 562-568. Hogrefe, T. C., R. H. Yahner, and N. H. Piergallini. 1998. Depredation of artificial ground nests in a suburban ver- sus a rural landscape. Journal of the Pennsylvania Acad- emy of Science 72: 3-7. King, D.I., R.M. DeGraaf, and C.R. Griffin. 1998. Edge-related nest predation in clearcut and groupcut stands. Conservation Biology 12: 1412-1415. Montevecchi, W. A. 1976. Egg size and the egg predatory behavior of crows. Behaviour 57: 307-320. Nol, E., and R. J. Brooks. 1982. Effects of predator exclosures on nesting outcome of Killdeer. Journal of Field Ornithology 53: 263-268. Paton, W.C. 1994. The effect of edge on avian nesting success: how strong is the evidence? Conservation Biology 8: 17-26. Rearden, J.D. 1951. Identification of waterfowl nest predators. Journal of Wildlife Management 15: 386-395. Redmond, G. W., D. M. Kepie, and P. W. Herzog. 1982. Vegetative structure, concealment, and success at two races of Spruce Grouse. Canadian Journal of Zoology 60: 670-675. Rodewald, A.D. 2000. The influence of landscape com- position on forest bird communities. Ph.D. Dissertation, The Pennsylvania State University, University Park. 147 pages. Rodewald, A. D., and R.H. Yahner. 2000. Breeding-bird communities associated with retention of residual trees in harvested hardwood stands. Journal of Wildlife Manage- ment 64: 924-932. Roper, J.J. 1992. Nest predation experiments with quail eggs: too much to swallow? Oikos 65: 528-530. Rudnicky, T. C., and M. L. Hunter, Jr. 1993. Avian nest predation in clearcuts, forests, and edges in a forest- dominated landscape. Journal of Wildlife Management 57: 358-364. Sargeant, R. A., J. C. Kilgo, B. R. Chapman, and K. V. Miller. 1998. Predation of artificial nests in hardwood fragments enclosed by pine and agricultural habitats. Journal of Wildlife Management 62: 1438-1442. 2001 Sokal, R.R., and F. J. Rohlf. 1995. Biometry. Third edi- tion. W. H. Freeman, San Francisco. 882 pages. Suarez, A. V., K.S. Pfenning, and S. K. Robinson. 1997. Nesting success of a disturbance-dependent songbird on different kinds of edges. Conservation Biology 11: 928-935. Sugden, L. G., and G. W. Beyersbergen. 1986. Effect of density and concealment on American Crow predation of simulated duck nests. Journal of Wildlife Manage- ment 50: 9-14. Whitcomb, R.R., C.S. Robbins, J. F. Lynch, B. L. Whitcomb, M. K. Klimkiewicz, and D. Bystrak. 1981. Effects of forest fragmentation on avifauna of the east- ern deciduous forest. Pages 125-205 in Forest island dynamics in man-dominated landscapes. Edited by R. L. Burgess and D. M. Sharpe. Springer-Verlag, New York, New York. Wilson, G.R., M. C. Brittingham, and L. J. Goodrich. 1998. How well do artificial nests simulate success of real nests? Condor 100: 357-364. Yahner, R.H. 1996. Forest fragmentation, artificial nests, and predator abundance. Conservation Biology 10: 672-673. Yahner, R.H. 2000. Eastern deciduous forest: ecology and wildlife conservation. 2nd edition. University of Minnesota Press, Minneapolis. 295 pages. Yahner, R.H., and B. L. Cypher. 1987. Effects of nest location on depredation of artificial nests. Journal of Wildlife Management 51: 178-181. YAHNER, RODEWALD, AND TALBOTT: EDGE-RELATED NEST PREDATION 87 Yahner, R. H., and C.G. Mahan. 1996a. Depredation of artificial ground nests in a managedforested landscape. Conservation Biology 10: 285-288. Yahner, R.H., and C.G. Mahan. 1996b. Effects of egg type on depredation of artificial ground nests. Wilson Bulletin 108: 129-136. Yahner, R.H., and C.G. Mahan. 1997. Effects of log- ging roads on depredation of artificial ground nests in a forested landscape. Wildlife Society Bulletin 25: 158-162. Yahner, R. H., C.G. Mahan, and C. A. DeLong. 1993. Dynamics of depredation on artificial ground nests in habitat managed for Ruffed Grouse. Wilson Bulletin 105: 172-179. Yahner, R.H., T. E. Morrell, and J.S. Rachael. 1989. Effects of edge contrast on depredation of artificial avian nests. Journal of Wildlife Management 53: 1135-1138. Yahner, R.H., and D. P. Scott. 1988. Effects of forest fragmentation on depredation of artificial nests. Journal of Wildlife Management 52: 158-161. Yahner, R. H., and H. R. Smith. 1990. Avian community structure and habitat relationships in central Pennsyl- vania. Journal of the Pennsylvania Academy of Science 64: 3-7. Yahner, R.H., and A. L. Wright. 1985. Depredation on artificial ground nests: effects of edge and plot age. Jour- nal of Wildlife Management 49: 508-513. Received 22 November 1999 Accepted 25 April 2000 The Vascular Flora of Akimiski Island, Nunavut Territory, Canada C. S. BLANEY! and P. M. KoTANEN? Department of Botany, Erindale College, University of Toronto, 3359 Mississauga Road North, Mississauga, Ontario LSL 1C6, Canada ‘Current address: Atlantic Canada Conservation Data Centre, PO Box 6416, Sackville, New Brunswick, E4C 4N1, Canada *Correspondence author. Blaney, C. S., and P. M. Kotanen. 2000. The vascular flora of Akimiski Island, Nunavut Territory, Canada. Canadian Field-Naturalist 115(1): 88-98. The James Bay region is floristically little known. In this paper, we substantially expand and revise the flora of the largest of the James Bay islands: Akimiski Island, Nunavut Territory. We have added 76 taxa in 33 families, for a total of 276 native taxa and 5 exotics. In addition, we have updated taxonomy and have provided information on rarity. A large fraction of the plants on Akimiski are classed as uncommon or rare in Ontario; in many cases, these represent either coastal species, or species near their geographic range limits. Some may be threatened by climate change and habitat degradation. Key Words: Akimiski Island, biodiversity, flora, global change, James Bay, Nunavut, Lesser Snow Geese. Little detailed floristic information exists describ- future threats to the biodiversity of the Hudson Bay ing the coastal Hudson Bay Lowlands. Most of this Lowlands. region lies north or west of the area considered by In this paper, we focus on the vascular plants of a the major eastern North American vascular floras single location: Akimiski Island, Nunavut Territory (Gleason 1952; Fernald 1970; Gleason and _ (including associated islets and shoals). This flora Cronquist 1991; Marie-Victorin 1995), and south or previously has been described by Riley (1981); we east of the area primarily covered by Porsild and hope that our additions and revisions will make this Cody (1980). The problem is acute for the Ontario one of the most complete floras for this region. Our coast and adjacent areas. The only flora to cover this work is significant not only because it adds substan- region is Scoggan (1978), which contains little local tially to existing knowledge, but also because the detail. The lists of Morton and Venn (1990) and limited area and isolation of offshore islands limits Newmaster et al. (1998) do include records from this rates of colonization and increases the risk of region, but contain little or no information on geo- extinctions (e.g., MacArthur and Wilson 1967; graphic distributions, while Argus et al. (1982-1987) | Brown 1995). Our paper constitutes a record of cur- and Oldham (1999) include some distributional rent floristic diversity on this island, but this flora is information but are solely concerned with rare likely to change substantially in future; we intend to species. The most detailed floristic publications create a record so that species lost or added in re- specifically devoted to the James Bay region are sponse to environmental changes may be recognized. Dutilly et al. (1958) and Riley and McKay (1980); more complete lists of Hudson Bay Lowlands plants Methods with distributional maps also have been compiled by Study Site Riley (1980). Akimiski Island (53°N; 81°W; 2300 km?) is the James Bay and Hudson Bay together represent a largest island in James Bay. It is separated from the biogeographic region very distinct from the remain- Ontario mainland by the Akimiski Strait, which is der of North America. Their low-latitude seacoasts about 20 km wide, but contains numerous shoals are unique within the centre of the continent, and (Riley 1981; Martini 1984). The island is and its are isolated from the Hudson Strait, their connec- flora are geologically very young, having been tion to the Atlantic, by more than 10 degrees of lati- exposed by the recession of the Tyrrell Sea about tude. Because climate zones dip south around 3500-4000 years ago (Martini and Glooschenko Hudson Bay, this region contains habitats and 1984). This postglacial uplift is still occurring at a species typical of much higher latitudes; for exam- rate of about 0.7 cmyr! (Martini and Glooschenko ple, one of the southernmost mainland areas of sea- 1984); as a result, new land is continuously being level tundra in the northern hemisphere is located exposed, especially on the north and west sides of on Cape Henrietta Maria, at the junction of James the island. With the exception of a few small out- Bay and Hudson Bay. The paucity of published crops of Palaeozoic limestones, the surficial geology floristic information for this region not only repre- _ of the island is dominated by marine silts and clays, sents a gap in our knowledge of North America’s generally covered with a peat layer, and interrupted ecosystems, but also may hamper efforts to predict by numerous gravel beach ridges (Martini and 88 2001 Glooschenko 1984). The highest elevations are along the south coast, where the shore rapidly rises to about 60 m in elevation; in contrast, most other coastal areas are dominated by extensive, low-gradient marshes and mudflats (Martini and Glooschenko 1984). Sporadic permafrost exists in inland areas (Martini and Glooschenko 1984), where palsas and other permafrost features occur. Numerous large lakes are found near the south shore of the island, while linear marsh systems lie between the south shore and the first few beach ridges. Most rivers drain towards the north shore. Akimiski Island lies within the boreal forest region. Regional vegetation is well summarized by Riley and McKay (1980), Riley (1981), and Martini and Glooschenko (1984). Picea mariana, Picea glauca, and Larix laricina forest are common throughout the island on beach ridges and as “is- lands” in peatland vegetation. Widespread shrub fens (largely Salix spp. and Myrica gale) occur close to the coasts, while graminoid-dominated peatlands occupy much of the island’s interior. Dry tundra-like vegetation (e.g., Saxifraga tricuspidata, Dryas inte- grifolia) is found on exposed beach ridges, primarily along the north coast. Coastal salt marshes are most commonly dominated by Festuca rubra at higher elevations and Puccinellia phryganodes, Carex sub- spathacea, and Glaux maritima at lower sites, but in many areas these species are replaced by Salicornia borealis and Spergularia canadensis. A thematic vegetation map of the island has been prepared from LANDSAT TM data by Andrew Jano of the Ontario Ministry of Natural Resources (reference: AKIBAMP95). Beaver formerly inhabited the island but are now apparently scarce, though abandoned dams are com- mon. Resident populations of caribou and moose are absent. Parts of the north shore are used by a rapidly expanding breeding colony of Lesser Snow Geese; Canada Geese nest throughout the island (Abraham et al. 1999). Most of the island now falls within a federal Migratory Bird Sanctuary. Akimiski Island has a limited history of human occupancy, with most impacts confined to coastal areas. Several Cree families from Ontario historical- ly trapped and hunted on the island for extended periods of the year (K. Abraham, personal communi- cation 1999). There are no villages on the island itself, but the island still is regularly visited by Cree from the Ontario shore, and there currently are some hunting camps near the west end. The only other structures on the island are associated with a research camp near Houston Point on the north shore, where studies of goose biology have been conducted by the Ontario Ministry of Natural Resources since the 1970s. A small Hudson Bay Company post once existed on the south coast, but its location is unclear (Riley 1981). BLANEY AND KOTANEN: FLORA OF AKIMISKI ISLAND 89 Sources of Information Numerous collectors have visited the island, but typically for very short periods. The early history of botanical explorations is summarized by Riley (1981). Most of the collecting since that date has been associated with the Ontario Ministry of Natural Resources research camp on the north shore. In par- ticular, Ken Abraham, Jennifer Bull, and Carol Lamey have collected and identified numerous spec- imens in the course of goose-related work; we have re-examined all of these specimens. In addition, in 1998, we had the opportunity to spend three weeks botanizing on the island; this represents the longest botanical collecting expedition in the history of Akimiski. Some additional collecting was done in 1999 and 2000 in association with other research by Peter Kotanen. Most of our collecting was performed in coastal areas (intertidal and supratidal marshes), but we also spent significant periods of time in fresh- water habitats (mostly sedge and willow fens), and briefly visited remote areas in the interior and the east end of the island. Riley (1981) stated no “shal- low marshes, beaver meadows, post-fire ridge sys- tems, larch fens, black spruce swamp or thickets” had yet been sampled on Akimiski, and that “graminoid and lowshrub fens” had been very little sampled. We visited all of these habitats except spruce swamp, though we did examine several areas of spruce forest. Because of the high costs of heli- copter travel, it is unlikely that any comparable expedition will happen in the near future. Results Our observations and collections have substantial- ly increased the known flora of Akimiski Island, and have provided vouchers at the University of Toronto herbarium (TRT) for many new and previously reported taxa. We also have updated and revised the taxonomy of Riley’s (1981) list; see Morton and Venn (1990) and Newmaster et al. (1998) for sum- maries of recent synonymy. A revised checklist is provided in Appendix A; this list has changed sub- stantially from the 211 taxa in 51 families reported by Riley (1981). Taxonomic changes led to the removal of 6 taxa, which we consider to be synony- mous with others on Riley’s list, but our list also includes 76 additions in 33 families, for a new total flora of 281 species in 55 families. We also have provided additional information on rarity and distribution. A high proportion of Aki- miski taxa are classified as rare in Ontario (28 species: Newmaster et al. 1998; Oldham 1999); in many cases, these represent either coastal species or species near their southern range limits. As well, two Akimiski species currently are unknown from Ontario: Potentilla crantzii and Salicornia borealis. The taxonomically difficult Potentilla crantzii occurs not far to the north (Scoggan 1978; Porsild and Cody 1980); Salicornia borealis is found in 90 THE CANADIAN FIELD-NATURALIST coastal regions of Hudson Bay (Wolff and Jefferies 1987). Discussion Riley (1981) speculated that 70 species could be added to the flora of Akimiski Island. Although our checklist adds more than this number, there is little reason to believe that this list is now complete. Further explorations of less-visited areas of Akimiski (especially the interior, the south shore, and both ends of the island) would undoubtedly yield addi- tional species. Despite this, the current evidence sug- gests that diversity is lower on Akimiski than on the Ontario mainland. The Ontario coast south of the Attawapiskat River supports almost 450 vascular plant taxa (Riley 1980, 1981), and about 320 taxa are found north of Lake River on Cape Henrietta Maria (Riley and McKay 1980; Riley 1981), though diver- sity in the coastal areas closest to Akimiski is lower (Riley 1981). In contrast to the vascular flora, the lichen and bryophyte flora of Akimiski remain very poorly known, despite the great abundance of moss- es and lichens in many habits; it is to be hoped that future research will address this gap in floristic knowledge. Though smaller, the flora of Akimiski generally is similar to that of the Ontario mainland (Riley 1980; Riley and McKay 1980; Riley 1981). The large num- ber of Akimiski taxa recorded as rare in Ontario probably reflects both the limited provincial distribu- tion of tundra, treeline, and seacoast habitats, and the limited surveying of inland sites which has yet taken place; most are likely to occur in suitable mainland areas, such as Cape Henrietta Maria. As a result, the true degree of rarity of some of these species may be open to question. Similarly, few Akimiski species are likely to represent true range limits; instead, apparent distributional outliers often may represent species which occur in suitable habitats elsewhere within this region. As is true of most remote northern floras (though not of transportation centres like Churchill: Scott 1996), the flora of Akimiski is almost entirely native, with only five clear exceptions (Morton and Venn 1990). Four of these (Chrysanthemum leucanth- emum, Plantago major, Sonchus arvensis, Taraxacum officinale) are cosmopolitan weedy species; one (Carum carvi) probably was deliberate- ly introduced by traders stationed at the Hudson Bay Company post (Riley 1981). None of these species is abundant. Morton and Venn (1990) also consider Artemisia tilesii, Erysimum cheiranthoides, Poten- tilla norvegica, and Rhinanthus minor to be intro- duced to Ontario, but these species are widespread in northern Canada, and at least northern populations are not treated as exotics by many authorities (Scoggan 1978; Porsild and Cody 1980; Newmaster et al. 1998; Voss 1972-1996; Cody 1996). Vol. 115 Although few exotics have yet colonized the island, in future the Akimiski flora is likely to suffer from two other types of impact. First, there is a sig- nificant risk that species will be lost as a result of global warming (Chapin et al. 1992). Northern ecosystems may be especially sensitive to climate change because small increases in temperature are expected to have significant effects on such important variables as snow cover, thaw depth, and peatland water tables (Maxwell 1992; Harvey 1997). As a result, like past climatic variation, future changes are likely to produce large shifts in treeline location, community composition, and ecosystem characteris- tics (Stevens and Fox 1991; Gates 1993). Akimiski lies not far to the north of the southern edge of per- mafrost; as temperatures rise, palsas and other per- mafrost-influenced landforms may be reduced or lost entirely (Maxwell 1992), to the detriment of species requiring drier or elevated microhabitats within the otherwise monotonous sedge fens that dominate the interior of the island (e.g., Rubus chamaemorus, Cladina). As well, though Akimiski is nearly 500 km south of the continental treeline in Keewatin or Québec, much of its flora has arctic affinities; such plants may be threatened by small increases in regional temperature. Akimiski may be especially at risk since many tundra species are already restricted to limited microhabitats (e.g., beach ridges, palsas), and since isolation from the mainland may both pre- vent these species from migrating as their habitat declines, and slow their replacement by species suit- ed to a warmer climate. A second set of impacts relates to geese. Con- siderable areas of coastal wetlands in the Hudson Bay region have been degraded by the foraging of rapidly increasing populations of Lesser Snow Geese (Jefferies 1988; Abraham and Jefferies 1997). The north shore of Akimiski was formerly used by thousands of staging Lesser Snow Geese during their migration to more northerly colonies (Abraham et al. 1999), and since 1968 has support- ed a nesting colony, currently numbering about 4000 birds (Abraham et al. 1999). Impacts are localized to areas heavily used by geese, but in these areas, many kilometres of the intertidal Puccinellia phryganodes zone and much of the supratidal Festuca rubra zone have been severely degraded, and some areas contain high densities of Salicornia borealis and Spergularia canadensis, which are indicative of severe goose damage (Srivastava and Jefferies 1996). Nearby fresh-water marshes also show signs of damage similar to those observed elsewhere on the Hudson Bay coast (Kerbes et al. 1990; Kotanen and Jefferies 1997). If the Akimiski colony continues to grow, it is proba- ble that populations of sensitive coastal plant species will be locally reduced or lost, though most are likely to persist in less degraded areas. 2001 In summary, this paper represents the most com- plete vascular flora yet compiled for Akimiski Island, and one of the more complete lists available for the James Bay region. More species inevitably will be discovered; however, results to date support the view that Akimiski has relatively low diversity compared to the mainland, as might be expected for an off-shore island. Currently, the flora of Akimiski includes numerous coastal and tundra species considered rare in Ontario, in addition to the many species typical of more southerly areas, but only a few exotics; if floris- tic changes do occur in future, this record represents a necessary first step towards their detection and docu- mentation. Further work still is necessary to document the vascular flora of little-visited areas of the island, and to describe the diversity of other important taxa, particularly lichens and bryophytes. Acknowledgments This work would have been impossible without the financial and logistical assistance of the Hudson Bay Project, the Ontario Ministry of Natural Resources, Wildlife Habitat Canada, the Arctic Goose Joint Venture, the Northern Scientific Train- ing Program, and NSERC. Special thanks go to K. Abraham, J. Leafloor, and R. L. Jefferies for support of all kinds, and to P. Ball for help with the identifi- cation of many specimens. K. Abraham, P. Ball, R. L. Jefferies, M. Johnson, and J. Riley helpfully com- mented on the manuscript. This is a publication of the Hudson Bay Project. Literature Cited Abraham, K. F., and R. L Jefferies. 1997. High goose populations: causes, impacts and implications. Pages 7-72 in Arctic Ecosystems in Peril: Report of the Arctic Goose Habitat Working Group. Edited by B. D. J. Batt. Arctic Goose Joint Venture Special Publication. United States Fish and Wildlife Service, Washington, D.C., and Canadian Wildlife Service, Ottawa, Ontario. Abrahan, K. F., J. O. Leafloor, and H. G. Lumsden. 1999. Establishment and growth of the Lesser Snow Goose, (Chen caerulescens caerulescens), nesting colony on Akimiski Island, James Bay, Northwest Territories. Canadian Field-Naturalist 113: 245-250. Argus, G. W., K. M. Pryer, D. J. White, and C. J. Keddy. Editors. 1982-1987. Atlas of the rare vascular plants of Ontario. Four parts. National Museum of Natural Sciences, Ottawa. Brown, J. H. 1995. Macroecology. University of Chicago Press, Chicago. Chapin, F. S., II, R. L. Jefferies, J. F. Reynolds, G. R. Shaver, and J. Svoboda. 1992. Arctic ecosystems in a changing climate. Academic Press, San Diego. Cody, W. J. 1996. Flora of the Yukon Territory. NRC Research Press, Ottawa. Dutilly, A., E. Lepage, and M. Duman. 1958. Contribution 4 la flore des fles (TNO) et du versant ori- entale de la baie James. Contribution 7F of the Arctic Institute of the Catholic University of America, Wash- ington. BLANEY AND KOTANEN: FLORA OF AKIMISKI ISLAND 9] Fernald, M. L. 1970. Gray’s Manual of Botany (8th edi- tion). Van Nostrand, New York. Gates, D. M. 1993. Climate change and its biological consequences. Sinauer, Sunderland. Gleason, H. A. 1952. The New Britton and Brown Illustrated Flora of the northeastern United States and adjacent Canada. Macmillan Publishing, New York. Gleason, H. A., and A. Cronquist. 1991. Manual of the vascular plants of northeastern United States and adja- cent Canada (2nd edition). New York Botanical Garden, New York. Harvey, L. D. D. 1997. Climatic change. Pages 328-351 in The Surface Climates of Canada. Edited by W. G. Bailey, T. R. Oke, and W. R. Rouse. McGill-Queen’s University Press, Montreal and Kingston. Jefferies, R. L. 1988. Pattern and process in arctic coastal vegetation in response to foraging by Lesser Snow Geese. Pages 281-300 in Plant Form and Vegetation Structure. Edited by M. J. A. Wegner, P. J. M. van den Aart, H. J. During, and J. T. A. Verhoeven. SPB Aca- demic Publishing, the Hague. | Kerbes, R. H., P. M. Kotanen, and R. L. Jefferies. 1990. Destruction of wetland habitats by Lesser Snow Geese: a keystone species on the west coast of Hudson Bay. Journal of Applied Ecology 27: 242-258. Kotanen, P. M., and R. L. Jefferies. 1997. Long-term destruction of wetland vegetation by Lesser Snow Geese. Ecoscience 4: 1895-1898 MacArthur, R. H., and E. O. Wilson. The theory of island biogeography. Princeton University Press, Princeton. Marie-Victorin, frére. 1995. Flore Laurentienne (3rd edi- tion). Université de Montréal, Québec. Martini, I. P. 1984. Considerations on the sub-arctic coastal environments of the Akimiski Island area, N.W.T. and Ontario, Canada. Sedimentary Geology 37: 225-228. Martini, I. P., and W. A. Glooschenko. 1984. Emergent coasts of Akimiski Island, James Bay, Northwestern Territories, Canada: geology, geomorphology, and vege- tation. Sedimentary Geology 37: 229-250. Maxwell, B. 1992. Arctic climate: potential for change under global warming. Pages 11-34 in Arctic ecosys- tems in a changing climate. Edited by F. S. Chapin III, Ro L: Jefferies, J. FP: Reynolds, G. R: Shaver-and's. Svoboda. Academic Press, San Diego. Morton, J. K., and J. M. Venn. 1990. A checklist of the flora of Ontario: vascular plants. University of Waterloo Biology Series Number 34. Department of Biology, University of Waterloo, Waterloo. Newmaster, S. G., A. Lehela, P. W. C. Uhlig, S. McMurray, and M. J. Oldham. 1998. Ontario plant list. Forest Research Paper Number 123. Ontario Forest Research Institute, Ontario Ministry of Natural Resources, Sault Ste. Marie. Oldham, M. J. 1999. Natural heritage resources of Ontario: rare vascular plants (3rd edition). Natural Heritage Information Centre, Ontario Ministry of Natural Resources, Peterborough. Porsild, A. E., and W. J. Cody. 1980. Vascular plants of continental Northwest Territories, Canada. National Museum of Natural Sciences, Ottawa. Riley, J. L. 1980. The flora and phytogeography of the Hudson Bay Lowland. Masters thesis, Department of Botany, University of Toronto, Toronto. Riley, J. L. 1981. The vascular flora of Akimiski Island, 92 THE CANADIAN FIELD-NATURALIST Vol. 115 James Bay, N.W.T. Le Naturaliste canadien 108: Stevens, G. C., and J. F. Fox. 1991. The causes of tree- 229-235. line. Annual Review of Ecology and Systematics 22: Riley, J. L., and S. M. McKay. 1980. The vegetation and 177-191. phytogeography of coastal southwestern James Bay. Life Voss, E .G. 1972-1996. Michigan flora (3 vols.) Sciences Contributions 124. Royal Ontario Museum, Cranbrook Institute of Science and University of Toronto. Michigan Herbarium, Ann Arbor. Scoggan, H. J. 1978. The flora of Canada. National Wolff, S. L., and R. L. Jefferies. 1987. Taxonomic status Museum of Natural Sciences, Ottawa. of diploid Salicornia europaea (sensu lato) (Cheno- Scott, P. A. 1996. Flora of Churchill Manitoba (8th edi- podiaceae) in northeastern North America. Canadian tion) Department of Biological Sciences, University of Journal of Botany 65: 1420-1426. Alberta, Edmonton. Srivastava, D. S., and R. L. Jefferies. 1996. A positive feedback: herbivory, plant growth, salinity and the deserti- fication of an arctic salt marsh. Journal of Ecology 84: | Received 1 December 1999 31-42. Accepted 1 May 2000 Appendix A: Checklist of the Vascular Plants of Akimiski Island, Nunavut Territory This checklist is based on Riley (1981), and includes plants of Akimiski Island and associated islets and shoals. Nomenclature follows Morton and Venn (1990); family order follows Newmaster et al. (1998); these publications also pro- vide synonymy. Where names have changed from those of Riley (1981), the name used by Riley is given [in square brack- ets]. Additions from Riley (1980) are indicated by +; additions from Riley (personal communication 1990) by +; other additions are indicated by *. Taxa listed as extremely rare (S1), very rare (S2), or rare to uncommon (S3) in Ontario are indicated accordingly; "?" indicates questionable status (Newmaster et al. 1998; Oldham 1999). In addition, Potentilla crantzii and Salicornia borealis are not included in recent Ontario checklists (Morton and Venn 1990; Newmaster et al. 1998). Plants believed to be locally exotic are underlined. Taxa represented in the University of Toronto herbarium are indicated by “TRT”; those in the herbarium of the National Museum of Natural Sciences, Ottawa by “CCAN”’; those in the herbarium of the Plant Biosystematics Institute, Agriculture Canada, Ottawa by “DAO”. Specimens collected by Ken Abraham, Jennifer Bull, and Carol Lamey, and in the collection of Ken Abraham (Ontario Ministry of Natural Resources) are indicated by “K”. Marc Johnson contributed Galium trifidum and Goodyera repens. Specimens for which the authors have collected vouchers are indicated by “V”; these have been deposited with TRT. This checklist represents an ongoing project. Future revisions will be reported on the Akimiski Flora Web Page (http://www.erin.utoronto.ca/w3pkota/akiplants.html), maintained by P. M. Kotanen (Department of Botany, Erindale College, 3359 Mississauga Road North, Mississauga, Ontario, LSL 1C6, CANADA; pkotanen @credit.erin.utoronto.ca). CHECKLIST PTERIDOPHYTA Equisetaceae Equisetum arvense L. TRT,K,V : Equisetum fluviatile L. Vv Equisetum variegatum Schleicer ex Fried. Weber & Mohr TRT Ophioglossaceae Botrychium lunaria (L.) Sw. TRT,V Botrychium minganense Victorin K,V SPERMATOPHYTA Gymnospermae Cupressaceae Juniperus communis L. [var. montana Ait., var. depressa Pursh] TRT,V 7 Juniperus horizontalis Moench Pinaceae Larix laricina (Duroi) K. Koch TRT Picea glauca (Moench) Voss [var. glauca, var. porsildii Raup] TRT Picea mariana (Miller) Britton, Sterns & Pogg. £ Pinus banksiana Lambert (Continued) 2001 BLANEY AND KOTANEN: FLORA OF AKIMISKI ISLAND APPENDIX A: Checklist of the Vascular Plants of Akimiski Island, Nunavut Territory (Continued) Angiospermae Dicotyledons Ranunculaceae Actaea rubra (Aiton) Willd. Anemone canadensis L. Anemone multifida Poiret ex Lam 53 Anemone parviflora Michaux Caltha palustris L. % Ranunculus abortivus L. Ranunculus cymbalaria Pursh = Ranunculus gmelinii DC. [R. purshi Richards] 52 Ranunculus hyperboreus Rottb. Ranunculus macounii Britton S2 Ranunculus pedatifidus Smith ex Rees = Ranunculus reptans L. = Ranunculus subrigidus Drew Thalictrum venulosum Trel. Myricaceae Myrica gale L. Betulaceae Alnus incana (L.) Moench [A. rugosa (DuRoi)Spreng. ] Betula pumila L. [var. glandulifera Regel] Chenopodiaceae z Atriplex cf. subspicata (Nutt.) Rydb. e Chenopodium capitatum (L.) Asch. fi Chenopodium glaucum L. ssp. salinum (Standley) Aellen Salicornia borealis S.L. Wolff & Jefferies [S. europaea L.] SZ Suaeda calceoliformis (Hook.) Mogq. Caryophyllaceae Honkenya peploides (L.) Ehrh. Minuartia dawsonensis (Britton) House Moehringia lateriflora (L.) Fenzl [Arenaria laterifolia L.] = Sagina nodosa (L.) Fenzl S2 Spergularia canadensis (Pers.) D. Don Stellaria borealis ssp. borealis Bigelow {S. calycantha (Ledeb.) Bong.] S$2S3 Stellaria humifusa Rottb. Stellaria longifolia Muhlenb. ex Willd. Stellaria longipes Goldie (s.1.) Polygonaceae Polygonum fowleri Robinson Polygonum viviparum L. Rumex occidentalis S. Watson Sarraceniaceae * Sarracenia purpurea L. Droseraceae Drosera anglica Hudson ms Drosera rotundifolia L. Violaceae fs Viola nephrophylla E. Greene Viola renifolia A. Gray Salicaceae Populus balsamifera L. E Populus tremuloides Michaux 3 Salix arctophila Cockrell ex Heller Salix bebbiana Sarg. Salix brachycarpa Nutt. Salix candida Fliiegge ex Willd. 93 TRT,DAO,V TRT TRT,K TRT,K TRT,K,V TRT,K,V TRT,K,V TRT,K,V (Continued) 94 THE CANADIAN FIELD-NATURALIST Vol. 115 APPENDIX A: Checklist of the Vascular Plants of Akimiski Island, Nunavut Territory (Continued) Salix glauca L. ssp. callicarpaea (Trautv.) Bocher [S. glauca L. var. callicarpaea (Trautv.) Bocher] TRT,K S27, Salix myricoides Muhlenb. [S. glaucophylloides Fern.] TRT,V Salix myrtillifolia Andersson K,V : Salix pedicellaris Pursh Vv ; Salix planifolia Pursh K,V e Salix reticulata L. V 3 Salix serissima (L Bailey) Fern. V Brassicaceae $2? Arabis arenicola (Richardson ex Hook.) Gelert K,V SISO) Cardamine pratensis L. ssp. angustifolia (Hook.) O. Schulz K Draba aurea M. Vahl ex Hornem. TRT,K,V Draba glabella Pursh CAN,K,V Draba incana L. TRT,K,V % Erysimum cheiranthoides L. V Rorippa palustris (L.) Besser ssp. fernaldiana (Butters & Abbe) Jonsell [R. palustris (L.) Besser ssp. glabra (Schulz) Stuckey var. glabrata (Lunell) Stuckey] Empetraceae Empetrum nigrum L. TRT,V Ericaceae Andromeda polifolia L. ssp. glaucophylla (Link) Hultén [A. glaucophylla Link] TRT Andromeda polifolia L. ssp. polifolia [A. polifolia L.| TRARKEN Arctostaphylos alpina (L.) Sprengel [A. alpina (L.) Spreng. var. alpina] TRT Arctostaphylos rubra (Rehder & E. Wilson) Fern. [A. alpina (L.) Spreng. var. rubra (Rehd. & Wils.) Bean] K Arctostaphylos uva-ursi (L.) Sprengel [A. uvaursi (L.) Spreng. ssp. uvaursi, ssp. adenotricha (Fern. & Macbr.) Calder & Taylor] TRT,V Chamaedaphne calyculata (L.) Moench K Gaultheria hispidula (L.) Muhlenb. ex Bigelow TRT,V 3 Kalmia polifolia Wangenh. Vv Ledum groenlandicum Oeder TRT,K,V “3 Vaccinium oxycoccos L. K,V Vaccinium uliginosum L. TRT,V Vaccinium vitis-idaea L. ssp. minus (Lodd.) Hultén TRT,K,V Pyrolaceae Moneses uniflora (L.) A. Gray K Orthilia secunda (L.) House TRT,K Pyrola asarifolia Michaux TRT,K,V Pyrola chlorantha Sw. Pyrola grandiflora Radius [P. rotundifolia L. ssp. grandiflora (Rad.) Andres. ] Primulaceae Androsace septentrionalis L. TRT,CAN,K,V $3? Glaux maritima L. ssp. obtusifolia (Fern.) B. Boivin TRT,K,V Primula egaliksensis Wormsk. TRT,CAN,K,V Primula stricta Hornem. TRT,K,V Trientalis borealis Raf. TRT Grossulariaceae Ribes oxyacanthoides L. TRT,V Ribes triste Pall. Saxifragaceae = Parnassia palustris L. K,V Parnassia parviflora DC. TRT,V. Saxifraga hirculus L. TRT Saxifraga tricuspidata Rottb. TRT,K,V Rosaceae Dryas integrifolia M. Vahl TRT,K,V Fragaria virginiana Miller TRT,K,V Geum macrophyllum Willd. [var. perincisium Rydb. Raup.] Vv Potentilla anserina L. [var. anserina, var. groenlandica Tratt.] TRT,V (Continued) 2001 BLANEY AND KOTANEN: FLORA OF AKIMISKI ISLAND APPENDIX A: Checklist of the Vascular Plants of Akimiski Island, Nunavut Territory (Continued) Potentilla crantzii (Crantz) G. Beck. Potentilla fruticosa L. S2 Potentilla multifida L. 82 Potentilla nivea L. = Potentilla norvegica L. . Potentilla palustris (L.) Scop. Potentilla pensylvanica L. [var. pectinata (Raf.) Lepage] 52 Potentilla pulchella R. Br. Rubus acaulis Michaux Rubus chamaemorus L. Fabaceae Astragalus alpinus L. Hedysarum alpinum L. [var. americanum Michx.] Hedysarum boreale Nutt. ssp. mackenzii (Richardson) Welsh [H. mackenzii Rich. |] Lathyrus japonicus Willd. [L. martimus Willd. var. pellitus (Fern.) Gl.] Lathyrus ochroleucus Hook. Lathyrus palustris L. $3 Oxytropis viscida Nutt. [var. hudsonica (Greene) Barneby] Elaeagnaceae Shepherdia canadensis (L.) Nutt. Haloragaceae s Myriophyllum sibiricum Komarov 2 Myriophyllum verticillatum L. Onagraceae Epilobium angustifolium L. - Epilobium ciliatum Raf. Epilobium palustre L. Cornaceae Cornus canadensis L. Santalaceae Geocaulon lividum (Richardson) Fern. [Comandra livida Rich.] Linaceae 82 Linum lewisii Pursh ssp. lepagei (B. Boivin) Mosq. Apiaceae Si Carum carvi L. - Cicuta bulbifera L. = Cicuta maculata L. . Cicuta virosa L. S3 Ligusticum scoticum L. Sium suave Walter Gentianaceae Gentianella amarella (L.) Borner ssp. acuta (Michaux) J.M. Gillett = Gentianopsis detonsa (Rottb.) Ma s.1. Lomatogonium rotatum (L.) Fries ex Nyman Menyanthaceae Menyanthes trifoliata L. [var. minor Raf.] Boraginaceae Mertensia maritima (L.) Gray Hippuridaceae nS Hippuris tetraphylla L.f. Hippuris vulgaris L. Plantaginaceae Plantago major L. Plantago maritima L. Scrophulariaceae Bartsia alpina L. Castilleja raupii Pennell Euphrasia hudsoniana Fern & Wieg. [E. arctica Lange] 95 TRT TRT,K CAN Vv K,V TRT,CAN,V TRT,K TRT,V TRI5V TRT,V CAN TRT,CAN,K,V V K,V TRT,K,V K,V TRT,K,V TRT TRT,CAN,K TRT TRT,K K,V (Continued) 96 THE CANADIAN FIELD-NATURALIST APPENDIX A: Checklist of the Vascular Plants of Akimiski Island, Nunavut Territory (Continued) Pedicularis groenlandica Retz. Pedicularis parviflora Smith ex Rees Pedicularis sudetica Willd. S3 Rhinanthus minor L. [R. cristagalli L. (s.1.)] Lentibulariaceae Pinguicula vulgaris L. Utricularia intermedia Hayne : Utricularia vulgaris L. Campanulaceae Campanula rotundifolia L. - Lobelia kalmii L. Rubiaceae Galium labradoricum (Wieg.) Wieg. = Galium trifidum L. Caprifoliaceae Linnaea borealis L. ssp. longiflora (Torrey) Hultén [L. borealis L. var americana (Forbes) Rehd.] Valerianaceae Valeriana dioica L. ssp. sylvatica (Sol. ex Rich.) F. Meyer [V. septentrionalis Rehd.] Asteraceae Achillea millefolium L. ssp. borealis (Bong.) Breitung [A. millefolium L. var. nigrescens Mey.] Achillea millefolium L. ssp. lanulosa (Nutt.) Piper [A. millefolium L. var. occidentalis DC. | Antennaria pulcherrima (Hook.) E. Greene Sl Antennaria rosea E. Greene Artemisia campestris L. ssp. borealis (Pallas) H.M. Hall & Clements [A. borealis Pall.; Artemisia canadensis Michx.] S2 Artemisia tilesii Ledeb. ssp. elatior (Torrey & A. Gray) Hultén a Aster borealis (Torrey & A. Gray) Prov. S3? Aster brachyactis S.F. Blake Aster longifolius Lam. [A. johannensis Fern.] s3 Chrysanthemum arcticum L. ssp. polare Hultén Chrysanthemum leucanthemum L. Erigeron acris L. [var. asteroides (Andrz. ex Bess.) DC.] Erigeron lonchophyllus Hook. $3? Matricaria maritima L. ssp. phaeocephala (Rupr.) Rauschert [Tripleurospermum phaeocephalum (Rupt.) Pobed.] Petasites sagittatus (Banks ex Pursh) A. Gray Senecio congestus (R. Br.) DC. [var. palustris (L.) Fern.] Senecio indecorus E. Greene Senecio pauperculus Michaux ey Solidago multiradiata Aiton Solidago uliginosa Nutt. : Sonchus arvensis L. ssp. uliginosus (M. Bieb.) Nyman Taraxacum ceratophorum (Ledeb.) DC. (s.1.) Taraxacum lacerum E. Greene Taraxacum officinale G. Weber Monocotyledons Scheuchzeriaceae ‘3 Scheuchzeria palustris L. Juncaginaceae Triglochin maritimum L. Triglochin palustre L. Potamogetonaceae Potamogeton alpinus Balbis Potamogeton filiformis Pers. Volos (Continued) 2001 BLANEY AND KOTANEN: FLORA OF AKIMISKI ISLAND APPENDIX A: Checklist of the Vascular Plants of Akimiski Island, Nunavut Territory (Continued) Potamogeton gramineus L. 3 Potamogeton richardsonii (A. Bennett) Rydb. Zannichelliaceae Zannichellia palustris L. Zosteraceae $2 Zostera marina L. Juncaceae Juncus alpinoarticulatus Chaix [J. alpinus Vill.) Juncus balticus Willd. [var. littoralis Engelm.] x Juncus bufonius L. #537 Juncus castaneus Smith Juncus triglumis L. ssp. albescens (Lange) Hultén Luzula parviflora (Ehrh.) Desv. Cyperaceae $3 Blysmus rufus (Hudson) Link [Scirpus rufus (Huds.) Schrader var. neogaeus Fern. | Carex aquatilis Wahlenb. Carex aurea Nutt. Carex bigelowii Torr. ex Schwein Carex brunnescens (Pers.) Poiret ex Lam. Carex capillaris L. " Carex capitata L. Carex chordorrhiza Ehrh. ex L.f. Carex concinna R.Br. Carex diandra Schrank = Carex flava L. : Carex garberi Fern. Carex glareosa Schk. ex Wahlenb. Carex gynocrates Wormskj. ex Drejer Carex interior L Bailey Carex leptalea Wahlenb. Carex limosa L. Carex livida (Wahlenb.) Willd. Carex mackenziei V. Krecz. i Carex magellanica Lam. Carex microglochin Wahlenb. Carex norvegica Retz. [C. media R. Br.] - Carex oligosperma Michaux Carex paleacea Schreber ex Wahlenb. S27 Carex praticola Rydb. cS Carex rariflora (Wahlenb.) Smith = Carex rostrata Stokes - Carex saxatilis L. Carex subspathacea Wormsk. ‘ Carex trisperma Dewey var. trisperma Carex utriculata F. Boott Carex vaginata Tausch Carex viridula Michaux 5 Eleocharis acicularis (L.) Roemer & Schultes = Eleocharis smallii Britton Eriophorum angustifolium Honck. Eriophorum russeolum Fries ex Hartman 2 Eriophorum tenellum Nutt. Eriophorum vaginatum L. Eriophorum viridi-carinatum (Engelm.) Fern. Kobresia simpliciuscula (Wahlenb.) Mackenzie 3 Scirpus cespitosus L. i Scirpus hudsonianus (Michaux) Fern. Poaceae " Calamagrostis canadensis (Michaux) P. Beauv. 97 KV TRT,CAN,K,V V K,V V Ni V V TRT,CAN,V TRIY K K TRT,V TRT Vv V Vv (Continued) 98 THE CANADIAN FIELD-NATURALIST APPENDIX A: Checklist of the Vascular Plants of Akimiski Island, Nunavut Territory (Concluded) 15 Calamagrostis stricta (Timm) Koeler ssp. inexpansa (A. Gray) C.W. Greene Calamagrostis stricta (Timm) Koeler ssp. stricta [C. stricta (Timm) Koeler] i Deschampsia cespitosa (L.) P. Beauv. Dupontia fisheri R.Br. ssp. psilosantha (Rupr.) Hultén Elymus trachycaulus (Link) Gould in Shinn. ssp. trachycaulus [Agropyron trachycaulum (Link) Malte var. trachycaulum, var. novae-angliae (Scribn.) Fern. ] Festuca brachyphylla Schultes & Schultes f. Festuca rubra L. ssp. rubra Festuca saximontana Rydb. * Glyceria striata (Lam.) A. Hitchc. Hierochloe odorata (L.) P. Beauv. Hordeum jubatum L. Leymus mollis (Trin.) Pilger [Elymus mollis Trin. ssp. mollis] Poa alpina L. S22 Poa arctica R. Br. ssp. arctica Poa glauca M. Vahl Poa palustris L. Poa pratensis L. ssp. alpigena (Blytt) Hit. Puccinellia lucida Fern. & Weath. Puccinellia phryganodes (Trin.) Scribn. & Merr. Trisetum spicatum (L.) Richter [var. spicatum] Sparganiaceae Sparganium angustifolium Michaux 2 Sparganium natans L. Typhaceae “3 Typha latifolia L. Liliaceae Maianthemum trifolium (L.) Sloboda [Smilacina trifolia (L.) Desf.] Tofieldia glutinosa (Michaux) Pers. Tofieldia pusilla (Michaux) Pers. Iridaceae Tris versicolor L. Orchidaceae = Amerorchis rotundifolia (Banks ex Pursh) Hultén Calypso bulbosa (L.) Oakes [var. americana (R.Br.) Luer] Corallorhiza trifida Chatel Cypripedium passerinum Richardson ‘i Goodyera repens (L.) R.Br. S2 Listera borealis Morong * Listera cordata (L.) R.Br. Platanthera dilatata (Pursh) Lindley ex Beck Platanthera hyperborea (L.) Lindley Platanthera obtusata (Banks ex Pursh) Lindley Spiranthes romanzoffiana Cham. Vol. 115 K,V TRT,K,V Ni TRT,K,V TRT,V TRT,K,V TREE TRT, V Vv TRT,V TRT,K,V TREV, CAN,V CAN,K,V TREY TRT TRT,DAO TRT,CAN,K,V TRT,K,V Effects of Reopening Hunting on Survival of White-tailed Deer, Odocoileus virginianus, in the Bas-Saint-Laurent region, Québec JEAN LAMOUREUX!, MICHEL CRETE? and MATHIEU BELANGER®? 'Société de la faune et des parcs du Québec, 212 avenue Belzile, Rimouski, Quebec, G5SL 3C3, Canada *Société de la faune et des parcs du Québec, Direction de la recherche sur la faune, 675 boulevard René-Lévesque Est (BP 92), Québec, Québec, GIR 5V7, and Département de biologie, Université du Québec 4 Rimouski, 300 Allée des Ursulines, Rimouski, Québec, GSL 3A1, and Département de biologie, Université Laval, Sainte-Foy, Québec, GIK 7P4, Canada 3Société de la faune et des parcs du Québec, 506 rue Lafontaine, Riviére-du-Loup, Québec, GSR 3C4, Canada Lamoureux, Jean, Michel Créte, and Mathieu Bélanger. 2001. Effects of reopening hunting on survival of White-tailed Deer, Odocoileus virginianus, in the Bas-Saint-Laurent region, Québec. Canadian Field-Naturalist 115(1): 99-105. Between 1996 and 1998, 59 White-tailed Deer (Odocoileus virginianus) in two major wintering areas in the Bas-Saint- Laurent region, Québec, were fitted with radio collars and followed using telemetry. The study period coincided with the reopening of hunting following three years of closure. Diminished deer numbers had been linked to severe winters intensi- fied by Coyote predation. During the hunting moratorium, starvation, predation and vehicle collisions were the principal causes of deer mortality. Following the moratorium, hunting became the main cause of adult mortality (39% of known deaths) while starvation mostly affected fawns (13% of known deaths). Contrary to expectations, annual survival of adult females (0.73) did not differ statistically from that of adult males (0.66), despite firearm hunting being permitted only on adult males. It seems that accidental hunting and poaching of females in addition to bow hunting harvest lowered female survival rate and limited the possibility of population growth. Eastern Québec provides marginal habitat for deer due to severe winters. We thus conclude that maintenance of White-tailed Deer populations in this region requires strict adherence by hunters to hunting regulations, in addition to intensive management of wintering areas and supplementary feeding dur- ing harsh winters. Key Words: Canis latrans, Coyote, Odocoileus virginianus, White-tailed Deer, hunting, mortality, poaching, population growth rate, survival, vehicle collision, Québec. Entre 1996 et 1998, 59 cerfs de Virginie ont été munis de colliers émetteurs et suivis au moyen de la télémétrie dans deux des plus importantes aires de confinement de la région du Bas-Saint-Laurent au Québec. La période d’étude a coincidé avec la réouverture de la chasse qui suivait un moratoire de trois ans rendu nécessaire par une diminution des effectifs reliée a des hivers rigoureux et aggravés par la prédation du coyote. Durant le moratoire sur la chasse, |’inanition, la préda- tion et les collisions automobiles avaient représenté les principaux facteurs de mortalité des cerfs. Aprés le moratoire, la chasse a constitué la principale cause de mortalité des cerfs adultes (39% des décés constatés) alors que l’inanition a surtout affecté les faons (13%). Contrairement aux résultats attendus, le taux de survie annuelle des biches (0,73) n’était pas statistiquement différent de celui des males adultes (0,66), malgré des modalités de chasse a4 l’arme a feu orientées seulement vers la récolte des males adultes. Conjugué a la récolte des archers, |’abattage des femelles, par erreur durant la saison de chasse a l’arme a feu, et par le braconnage a vraisemblablement diminué le taux de survie des biches et limité la possibilité d’accroissement de cette population. L’Est du Québec offre des conditions marginales d’habitat aux cerfs de Virginie a cause des hivers rigoureux. Nous concluons que le maintien de la population de cerfs dans la région nécessite un respect strict de la réglementation par les chasseurs en plus de |’aménagement intensif des aires d’hivernage et d’une ali- mentation d’urgence lors des hivers rigoureux. Mots clefs: Canis latrans, Coyote, Odocoileus virginianus, Cerf de Virginie, braconnage, chasse, inanition, mortalité, survie, taux d’accroissement, Québec. Winter harshness is the principal factor limiting White-tailed Deer (Odocoileus virginianus) popula- tions occurring at the northern extremes of their dis- tribution (Lamontagne and Potvin 1994; Dumont et al. 2000). During particularly severe winters in the early 1970s, mortality by starvation reached 40% in the Pohénégamook wintering area, Québec (Potvin et al. 1981). On Anticosti Island, winter mortality of adult females varied from 3 to 38% according to snow hardness (Potvin et al. 1997). Hunting is a sec- ondary limiting factor in southeastern Québec because females have not been legally hunted with firearms since 1974 and deer populations can sustain an annual harvest of 5 to 10% of adult females and still increase in the absence of severe winters (Lamontagne and Potvin 1994). The White-tailed Deer population of Québec’s Bas-Saint-Laurent region declined rapidly in the early 1990s, following a series of severe winters. These harsh winters increased the influences of Coyote (Canis latrans) predation and loss of conifer canopy due to Spruce Budworm (Choristoneura fumiferana) in the traditional wintering areas (Dumont et al. 2000). The reduced deer numbers led ze 100 iliac deat 6S" I0W THE CANADIAN FIELD-NATURALIST Vol. 115 Gaspé Peninsula WINTERING AREA SUMMER RANGE 68°30'W FIGURE 1. Map showing the two wintering areas and summer ranges for deer studied in the Bas-Saint-Laurent region, Québec. to the closure of hunting in the region in 1993 and the institution of a recovery plan (MLCP 1992) which involved Coyote control in wintering areas, management of winter habitat and emergency feed- ing during harsh winters. At the same time, a research program was initiated to understand better the dynamics of this deer population. White-tailed Deer were thus fitted with radio collars and followed from the winter of 1994 in two of the largest winter- ing areas in eastern Québec, at Pohénégamook and Lac Témiscouata. Between 1994 and 1996, starvation, Coyote preda- tion and road accidents were the principal causes of deer mortality in the two wintering areas (Dumont et al. 2000). Competition for food appeared to regulate deer numbers at Pohénégamook while numbers of deer increased at Lac Témiscouata. The population as a whole recovered rapidly over the three years and hunting was reinstated in the fall of 1996. Our objective in this study was to measure the effect of sport hunting on survival of deer at the northern limit of their distribution following reopen- ing of the hunt by monitoring marked individuals for three years. Study Area The study took place in the Pohénégamook and ‘Lac Témiscouata wintering areas, located approxi- mately 40 and 55 km southeast of Riviere-du-Loup, respectively (Figure 1). Pohénégamook wintering area (PWA) (47°29'3” N, 69°15'5” W) covers 25 km? and is located approximately 40 km southeast of Lac Témiscouata wintering area (LTWA) (47°44'5” N, 68°48'5” W), which covers 221 km?. Numbers of deer in the PWA in 1996 were estimated at 520 + 62 deer (a = 0,10) using an aerial survey with double- count technique (Potvin et al. 1992), resulting in an estimated density of 20.7 + 2.5 deer/ km”. Numbers of deer in the LTWA in 1997 were estimated at 2166 + 606 deer (a = 0,10), for an estimated density of 9.8 + 2.7 deer/ km? (Dumont et al. 2000). The two wintering areas occur in the mixed Balsam Fir/Yellow Birch forest type. Dominant tree species are Balsam Fir (Abies balsamea), White Spruce (Picea glauca), Eastern White Cedar (Thuja occidentalis), Trembling Aspen (Populus tremu- loides) and Yellow Birch (Betula alleghaniensis). Due to the most recent Spruce Budworm outbreak in the late 1970s, the proportion of coniferous trees has been reduced (Dumont et al. 2000). Deer use all for- est types during early winter but concentrate in mixed forests dominated by conifers (between 50 and 80 % coniferous canopy closure) during periods of heavy snow (Dumont et al. 1998). Both wintering areas are actively managed to optimize forest habitat for deer (MEF and MRN 1994, 1995). 2001 Snow accumulation and sinking-depth conditions in the study area are among the most extreme for White-tailed Deer in Québec. The cumulative sink- ing-depth index (Potvin and Breton 1992) varied sig- nificantly over the study period for the two wintering areas. At PWA, winter conditions were mild in 1996 (2975 cm-days of sinking-depth), severe in 1997 (6380 cm-days) and near normal in 1998 (4443 cm- days). At LTWA, winter conditions were mild in 1996 (2242 cm-days) but severe in both 1997 (7291 cm-days) and 1998 (5257 cm-days). The study area is located within Québec hunting zone number 2 and deer hunting regulations are rela- tively conservative. Bow hunting extends over 14 days and indiscriminate harvest of adult males, adult females and fawns is permitted. Bow hunters har- vested 287, 316 and 321 deer in 1996, 1997 and 1998, respectively. The bow harvest consisted of 20% adult males, 42% adult females and 38% fawns. Firearm hunting extends over 16 days but only males with antlers of at least 7 cm in length may be taken. Total deer harvest, including bow hunting, was 1999, 1419 and 1354 deer in 1996, 1997 and 1998, respec- tively. Hunting was completely closed for three years between 1993 and 1995. Total harvest for the last year prior to the hunting moratorium was 352 deer in the fall of 1992. Methods White-tailed Deer were captured in January from 1994 to 1997 using modified Stephenson traps (Pichette and Samson 1975; Rongstab and McCabe 1984). Traps were placed near deer trails and baited with cedar twigs and commercial cattle feed. Captured deer were fitted with Holohil radio collars and followed from late September, 1996 to early December, 1998. Of the total 59 deer, 15 were fawns, 16 were adult (> | year old) males and 28 were adult females at time of capture. Forty-four deer were captured at PWA and 15 at LTWA. We located deer weekly during the study period using a Cessna 206 aircraft equipped with two Yagi anten- nae. For the period 1 May to 1 October, 1997, we located deer three times a week. When a collar emit- ted a stationary signal, we undertook a ground search to locate the animal and determine the cause of mor- TABLE |. Probability of various factors affecting survival rate of White-tailed Deer fitted with radio collars in the Bas-Saint-Laurent region in Québec, 1996-1998. Factor Univariate Test* Multivariate Test” Year 0.001 0.3001 Season 0.0002 0.0016 Age/Sex 0.047 0.0235 Area 0.6459 0.8188 “log-rank test of LIFETEST procedure (SAS 1988) >Jikelihood-ratio test of GENMOD procedure (SAS 1988) LAMOUREUX, CRETE, AND BELANGER: SURVIVAL OF WHITE-TAILED DEER 101 tality, if the collar did not prove to have been lost. Using circumstantial evidence in the field (Dumont et al. 2000), we attributed death to one of the follow- ing causes : hunting, poaching, predation, starvation or road accident. Deer were considered poached if accidentally killed during the hunting season or ille- gally killed outside the season. Collars from legally harvested deer were taken by hunters to big game registration stations. Statistical analysis We calculated the daily probability of death (DPD) of White-tailed Deer using the GENMOD procedure of SAS (1988), assuming an exponential model. First we examined whether year, season, age/ sex status (fawns, males > | year, females > | year) and wintering area affected survival rate of deer using both univariate and multivariate tests. The uni- variate analysis utilized a nonparametric log-rank test (LIFETEST procedure; SAS 1988) which has the advantage of not requiring a normal distribution; however, the log-rank test does not take into account other variables that could affect deer survival rate. The multivariate analysis was parametric; it utilized a likelihood-ratio test assuming the exponential model (GENMOD procedure; SAS 1988), which accounts for other variables when testing the effect of a specific variable. We tested whether the expo- nential model was appropriate by comparing it to the Weibull’s model using a Lagrange multiple chi- squared test (LIFEREG procedure; SAS 1988). The exponential model does not take into account indi- vidual variation, as it assumes that all surviving ani- mals have the same probability of dying at the begin- ning of the next time period. We divided the year into four periods : winter (136 days; 6 December to 20 April), spring migra- tion (10 days; 21 to 30 April), summer (approxi- mately 149 days; 1 May to the beginning of the bow hunting season) and fall (approximately 70 days; beginning of the bow hunting season to 5 December). Bow hunting commenced on 28, 27 and 26 September in 1996, 1997 and 1998, respectively. Annual survival rate (ASR) was estimated by com- bining the four seasonal survival rates according to the following formula (Dumont et al. 2000): ASR:=(1-DPD, )?°% (1-DPD,,,)'° ° G=DPD! 4)" @-DPD Oe The standard error of ASR can be estimated according to the formula (Agresti 1990): SE(ASR)=ASR *¢ [136*SE(X, )/(1-X,) + 10°*SE(X,)/(1—X,) + 149°SE(X,)/CU1—-X,) + T0°SE(X,)/(1-X,)] where X,, X,, X, and X, represent DPD DPD,,,, 20d DPD,.,, respectively. DPD win’ spr’ fal? Mortality rate due to hunting was estimated as the proportion of marked animals alive at the beginning 102 THE CANADIAN FIELD-NATURALIST Vol. 115 TABLE 2. Daily probability of death (DPD) and cause of mortality of radio-collared White-tailed Deer of differing age / sex status and at different seasons in the Bas-Saint-Laurent region in Québec, 1996-1998. Cause of Mortality Predation Hunting Poaching Starvation Collision Other‘ Standard DPD Error Age/Sex Status Season n@ x08 x10° Fawns Winter 9 3.58 1.79 Adult Males Winter 14 0.45 0.30 Spring i) 4.01 Sale Summer 15 0.59 0.35 Fall 19 3.3 123 Adult Females Winter 25 0.34 0.21 Spring 26 3.06 2.28 Summer 24 0.45 0.25 Fall 30 DiS?) 0.78 ‘number of marked deer firearm hunting “bow hunting dindeterminate causes of fall that was harvested. Confidence intervals were calculated using approximations generated from the binomial distribution and corrected for small sample size (Dixon and Massey 1969). DPD of adult males and adult females before and after the reopening of the hunt and of fawns and adults during winter were compared using Z- tests (Dixon and Massey 1969). We calculated the finite population rate of increase (A) of each of the wintering areas using matrix algebra according to the following formula: yee (Zee ie Sra orient where A, is the annual survival rate of adult females measured in this study, A, is the annual survival rate of fawns (PWA = 0.34; LTWA = 0.48 : Dumont et al. 2000) and f, is the number of female offspring produced by yearling and older females (0.79 fawns/ adult female; Dumont et al. 2000). Results Univariate analysis revealed that year, season and age/sex status, but not wintering area, significantly affected the DPD of deer fitted with radio collars (Table 1). Multivariate analysis showed the same trends with the exception that year was no longer a significant factor (Table 1). Season and age/sex sta- tus clearly affected survival. Since the univariate analysis showed an annual variation in deer survival, we examined the data to find that survival was espe- cially affected in the fall, a season for which we had three years of data. Because there were only two years of data for the other seasons, the sensitivity of the multivariate analysis was consequently reduced. We thus combined years in subsequent analyses. The exponential model was found to fit the data (Lagrange x” = 0.098, P = 0.75) and utilized to cal- culate the DPD. 3 1 1 i 7 NO KR Re Ds + 1 DPD of radio-collared adult deer was seven to nine times higher during spring migration and fall than during winter and summer (Table 2). There was no significant difference between sexes in daily mor- tality rates for any season (two-tailed Z-tests, P> 0.05), especially im falli(Z,— 054s ese) even though hunting mostly targetted adult males. For fawns, mortality was high during winter, although no data were available for other seasons. DPD of fawns during winter was significantly higher than that of adult males (Z = 1.70, one-tailed P = 0.04) and that of adult females (Z = 1.79, one- tailed P = 0.03). Overall winter survival rate of fawns was 0.61 (+ 0.15) compared to 0.94 (+ 0.04) for adult males and 0.95 (+ 0.03) for adult females. During the study period, hunting was the principal cause of mortality of adult deer (39% or 12 of 31 known deaths; see Table 2). Poaching accounted for 16% or 5 of 31 known deaths and 4-of these cases were adult females accidentally shot during the firearm hunting season. Starvation killed 13% (4/31) of marked animals, mostly fawns. Vehicle collisions 1994 01995 i 1996-1998 Ore Annual survival rate (=) Adult females Adult males FiGuRE 2. Annual survival rates of radio-collared adult males and adult females in the Bas-Saint-Laurent region, Québec, 1994-1998. * Dumont et al. 2000. 2001 1,2 0,99° 0,99° 1 Fall survival rate Adult males LAMOUREUX, CRETE, AND BELANGER: SURVIVAL OF WHITE-TAILED DEER 103 0,99" 0,99° 11996 | Adult females FiGuRE 3. Fall survival rates of adult males and adult female White-tailed Deer in the Bas- Saint-Laurent region, Québec, 1994-1998. * Dumont et al. 2000. (6%) and predation (3%) accounted for the remain- ing known deaths while 22% of mortality was of indeterminate cause (Table 2). Mortality rate due to hunting, including wounding and errors, was 0.17 (0.11 — 0.27, a = 0.05) for both sexes com- bined. During the same period, the harvest rate of adult males was 0.23 (0.10 — 0.42, a = 0.05). We calculated annual survival rate for adult males and adult females by combining measured survival rates for each season. Since the multivariate analysis did not reveal differences between years or areas, we calculated an overall mean for each sex. Annual sur- vival rate from 1996 to 1998 was 0.66 (+ 0.14) for adult males and 0.73 (+ 0.10) for adult females. These rates are much lower than those measured between 1994 and 1995 (Dumont et al. 2000), when hunting was closed (see Figure 2). Fall survival rate dropped sharply with the reopening of hunting from 0.99 to 0.70 for adult males and from 0.99 to 0.76 for adult females (see Figure 3). Using measured survival rates of deer, we estimat- ed finite population rate of increase (1) between 1996 and 1998 to be 1.00 at PWA and 1.08 at LTWA. The value found for the PWA is similar to those recorded during the hunting moratorium while the hunted population growth rate found for the LTWA was lower than those reported before hunting reopening (Dumont et al. 2000). Discussion Hunting was the principal cause of mortality of radio-collared adult deer in the Bas-Saint-Laurent region following reopening of the hunt. It was note- worthy that hunting and poaching significantly reduced annual survival rate of adult females because firearm hunting was only permitted on adult males. In addition, the fall DPD did not differ signif- icantly between the two sexes over three years. Under similar hunting restrictions in New Brunswick and Michigan, hunting and poaching mortality rates varied between 1.6 and 4% for adult females (Whitlaw et al. 1998; Van Deelen et al. 1997). Mortality rates due to hunting and poaching reached 16 % in our study. This level of additive mortality seemed to reduce adult female survival to such an extent that the population growth rate was curbed to 1.00 at PWA and 1.08 at LTWA, despite relatively mild winters. Numbers of adult males harvested with firearms within the entire hunting zone did not vary much from 1997 (1102) to 1998 (1032), which also suggests a relatively stable population. The annual survival rate of adult males in our study (0.66) was slightly higher than the 0.57 report- ed for northern New Brunswick between 1994 and 1997 (Whitlaw et al. 1998) and the 0.51 for Nova Scotia between 1994 and 1998 (Patterson 1999). Adult male survival rates can be as low as 0.22 in heavily hunted populations (Van Deelen et al. 1997) and thus our findings for males do not cause con- cern. The annual survival rate of adult females of 0.73, however, was lower than that found in other studies, e.g., 0.85 in New Brunswick and 0.80 in Nova Scotia, and resulted from identification errors during the firearm hunting season and from poach- ing. Illegal killing of female deer accounted for as much mortality as did legal bow hunting (Table 2). [lligal killing also represented a major mortality fac- tor for adult female Mule Deer (Odocoileus hemionus) in western USA (McCorquodale 1999). Overwinter fawn survival averaged 61% during our study, in comparison to 52 — 84% for the same two populations during the moratorium on hunting (Dumont et al. 2000). Winter fawn survival slightly exceeded 60% during the same period in New 104 Brunswick and varied between 51 and 61% in Nova Scotia (Ballard et al. 1999; Patterson 1999). Fall sur- vival rates tended to increase over the three years for both sexes of adults, but not significantly (Figure 3). This may be because the animals marked at the beginning of our study had aged and gained experi- ence that would reduce the risk of being killed dur- ing the hunt. In addition, it is possible that deer mod- ified their behaviour to lessen their vulnerability to hunting, as previously reported elsewhere (Kilgo et al. 1998). Besides hunting, starvation, collision and preda- tion remained the major mortality factors of deer during our study, as Dumont et al. (2000) previously observed. This contrasted with mortality causes of White-tailed Deer in New Brunswick and Nova Scotia where Coyote predation strongly dominated and starvation passed almost unobserved (Whitlaw et al. 1998; Ballard et al. 1999; Patterson 1999). Methodological differences in identifying mortality causes may partly be involved in such differences (see Dumont et al. 2000 for the criteria we used), but other factors likely explained this discrepancy. The reduced occurrence of Coyote predation in our study area could result from lower Coyote densities in south-eastern Québec than in New Brunswick and Nova Scotia. Coyote density averaged 0.5 animal/10 km? in the forest landscape of our study area, where most deer stayed year-round, in comparison to 2,7 coyotes/10 km? in the rural landscape (Richer 2000). No figures on Coyote density existed for the Maritimes, but a greater proportion of cultivated land in New Brunswick or Nova Scotia would likely result in higher Coyote density; milder winters in the Maritimes could also increase the carrying capacity for coyotes if they facilitated Coyote survival or fecundity. Depressed Coyote density in Québec could otherwise originate from high harvest rates by trappers, which varied between ~ 40-60% during the last decade (Créte and Lemieux 1996; unpublished). Lower incidence of Coyote predation in Québec could finally depend on deer density in winter and on location of deer home range within wintering areas; generally, deer wintering in small groups or at the periphery of wintering areas are more vulnerable to Coyote predation than those occupying large win- tering areas (Messier and Barrette 1985; Brundige 1993; Patterson and Messier 2000). Reopening of sport hunting led to a sharp decline in annual survival of adult female White-tailed Deer in the Bas-Saint-Laurent region, even though only the harvest of antlered males was permitted during the firearm season. Accidental or deliberate killing of adult females was sufficient to limit the popula- tion growth rate in this region situated at the north- ern limit of the species’ distribution. It is therefore important that hunters be made aware of the negative consequences of illegal hunting. In the long term, THE CANADIAN FIELD-NATURALIST Vol. 115 managers should also assess the effect of harvesting adult females through bow hunting, given the increasing popularity of this activity in Québec. Bow hunting represented 24% of the total regional! harvest in 1998 compared to 8% in 1990, while fully 42% of deer taken by bow hunters in recent years were adult females. These results lead us to conclude that the White-tailed Deer population in eastern Québec requires strict adherence of hunters to hunting regu- lations, in addition to supplemental feeding during harsh winters and intensive management of winter- ing areas. Acknowledgments | We wish to thank all those who participated in any way in the capture and marking of White-tailed Deer, especially C. Daigle and R. Lemieux. We also thank G. Daigle of the Service de consultation statis- tique at Université Laval for statistical analyses. Financial support was received from the Société de la faune et des parcs du Québec, Fonds FCAR (Québec), NSERC and the Fédération québécoise de la faune. Literature Cited Agresti, A. 1990. Categorical data analysis. John Wiley and Sons, New York. Ballard, W. B., H. A. Whitlaw, S. J. Young, R. A. Jenkins, and J. Forbes. 1999. Predation and survival of White-tailed Deer fawns in northcentral New Brunswick. Journal of Wildlife Management 63: 574-579. | Brundige, C. C. 1993. Predation ecology of the eastern Coyote (Canis latrans) in the Adirondacks, New York. Ph.D. thesis, State University of New York, Syracuse. Créte, M., and R. Lemieux. 1996. Population dynamics of coyotes colonizing the boreal forest of southeastern Québec. Journal of Wildlife Research 1: 99-105. Dixon, W. J., and F. D. Massey. 1969. Introduction to Statistical Analysis. McGraw-Hill, New York. Dumont, A., J.-P. Ouellet, M. Créte, and J. Huot. 1998. Caractéristiques des peuplements forestiers recherchés par le cerf de Virginie en hiver a la limite nord de son aire de répartition. Canadian Journal of Zoology 76: 1024-1036. Dumont, A., M. Créte, J.-P. Ouellet, J. Huot, and J. Lamoureux. 2000. Population dynamics of northern White-tailed Deer during mild winters: evidence of regu- lation by food competition. Canadian Journal of Zoology 78: 764-776. Kilgo, J. C., R. F. Labisky, and D. E. Fritzen. 1998. Influences of hunting on the behavior of White-tailed Deer: Implications for conservation of the Florida Panther. Conservation Biology 12: 1359-1364. Lamontagne, G., and F. Potvin. 1994. Plan de gestion du cerf de Virginie au Québec. 1995-1999. L’espéce, son habitat et sa gestion. Ministére de l’Environnement et de la Faune, Direction de la faune et des habitats, Québec. 114 pages. Report number 94-2501-11. McCorquodale, S. M. 1999. Movements, survival, and mortality of Black-tailed Deer in the Klickitat basin of Washington. Journal of Wildlife Management 63: 861-871. 2001 Messier, F., and C. Barrette. 1985. The efficiency of yard- ing behavior by White-tailed Deer as an anti-predator strategy. Canadian Journal of Zoology 63: 785-789. Ministére de l’Environnement et de la Faune and Mini- stére des Ressources Naturelles. 1994. Plan d’amé- nagement du ravage du Lac Pohénégamook : plan d’ intervention 1994-1999. Direction régionale du Bas- Saint-Laurent. 36 pages. Ministére de "Environnement et de la Faune and Mini- stere des Ressources Naturelles. 1995. Plan d’amé- nagement du ravage du Lac Témiscouata — bloc C: plan d’intervention 1995-2000. Direction régionale du Bas- - Saint-Laurent. 60 pages. Ministére du Loisir, de la Chasse et de la Péche. 1992. Plan de redressement de la population de cerfs de Virginie (Odocoileus virginianus) de la Gaspésie. Plan d’action pour 1992-1997. 16 pages. Patterson, B. R. 1999. The effects of prey distribution and abundance on eastern Coyote life history and predation on White-tailed Deer. Ph.D. thesis, University of Saskatchewan, Saskatoon. Patterson, B. R., and F. Messier. 2000. Factors influenc- ing killing rates of White-tailed Deer by coyotes in east- ern Canada. Journal of Wildlife Management 64: 721-732. Pichette, C., and N. Samson. 1975. Projet de marquage du cerf de Virginie du ravage de Armstrong. Ministére du Tourisme, de la Chasse et de la Péche, Service de la recherche biologique. 21 pages. Potvin, F., and L. Breton. 1992. Rigueur de |’hiver pour le cerf au Québec : description de l’indicateur prévisionnel NIVA et présentation d’un logiciel approprié. Ministére du Loisir, de la Chasse et de la Péche, Direction de la gestion des espéces et des habitats. 38 pages. Report number SP 1936-07-92. LAMOUREUX, CRETE, AND BELANGER: SURVIVAL OF WHITE-TAILED DEER 105 Potvin, F., J. Huot, and F. Duchesneau. 1981. Deer mor- tality in the Pohénégamook wintering area, Québec. Canadian Field-Naturalist 95: 81-84. Potvin, F., L. Breton, and A. Gingras. 1997. Déplace- ments et survie des biches d’ Anticosti de 1986 a 1990: une étude télémétrique. Ministére de |’ Environnement et de la Faune, Direction de la faune et des habitats. 43 pages. Report number 3643-97-08. Potvin, F., L. Breton, L.-P. Rivest, and A. Gingras. 1992. Application of a double-count aerial survey technique for White-tailed Deer, Odocoileus virginianus, on Anticosti Island, Quebec. Canadian Field-Naturalist 106: 435-442. Richer, M.-C. 2000. Evaluation des ressources alimen- taires estivales des coyotes (Canis latrans) du sud-est du Québec. M.Sc. thesis, Université du Québec 4 Rimouski, Rimouski. Rongstab, O. J., and R. A. McCabe. 1984. Capture tech- niques. Pages 655-676 in White-tailed Deer Ecology and Management. Edited by L. K. Halls. Stackpole Books, Harrisburg. SAS Institute, Inc. 1988. SAS/STAT User’s Guide. SAS Institute Inc., Cary, North Carolina. 1028 pages. Van Deelen, T. R., H. Campa III, J. B. Haufler, and P. D. Thompson. 1997. Mortality patterns of White-tailed Deer in Michigan’s upper peninsula. Journal of Wildlife Management 61: 903-910. Whitlaw, H. A., W. B. Ballard, D. L. Sabine, S. J. Young, R. A. Jenkins, and G. J. Forbes. 1998. Survival and cause-specific mortality rates of adult White-tailed Deer in New Brunswick. Journal of Wildlife Management 62: 1335-1341. Received 31 January 2000 Accepted 12 June 2000 Slumping Activity and Forest Vegetation Along the Northeastern Shore of Waskesiu Lake, Prince Albert National Park, Saskatchewan D. H. DE BOkER and O. W. ARCHIBOLD Department of Geography, University of Saskatchewan, 9 Campus Drive, Saskatoon, Saskatchewan, S7N 5A5, Canada De Boer, D. H., and O. W. Archibold. 2001. Slumping activity and forest vegetation along the northeastern shore of Waskesiu Lake, Prince Albert National Park, Saskatchewan. Canadian Field-Naturalist 115(1): 106—114. Changes in forest cover along the shore of Waskesiu Lake, Saskatchewan, can be related to slumping. This study evaluates the impact of slumping along the lakeshore on the structure of the vegetation. The oldest trees in the undisturbed forest were approximately 100 years old, whereas the trees which had established on recent scars were typically less than 10 years old. Tree rings were used to determine the ages of the reaction wood in bent trees that had survived the disturbances; this indicated that recent slumping events occurred in 1973, 1979, and 1984. These years either received higher than nor- mal precipitation or had above normal lake levels. Recent slope disturbances were characteristically colonized by a sparse cover of herbaceous plants and tree seedlings which contrasted with the adjacent slopes which supported a mixed forest cover dominated by White Spruce, Balsam Fir, and Paper Birch. Herbaceous ground cover on the slumps was composed of 35 herb and grass species compared to 14 species on the intervening slopes. Similarly, the shrub cover was more diverse on the slumps (9 species) than on the adjacent slopes (4 species). The greatest density of herbs and grasses occurred near the base of the slumps, where soil moisture conditions were more suitable for plant growth. Slumping is most likely caused by subsurface seepage, with annual fluctuations in lake level and seasonal wave activity possibly contributing to the process. As well as causing higher lake levels, wet years can also promote slope instability by increasing the water content of the surficial materials. However, there is no simple relationship between precipitation, temperature, lake level and the occur- rence of slumping. Key Words: White Spruce, Picea glauca, dendrochronology, reaction wood, shoreline instability, slumping, Waskesiu Lake, Saskatchewan. Vegetation can provide useful evidence of the fre- quency of occurrence and rates of geomorphic dis- turbance. Avalanche tracks, for example, are readily apparent in mountains because of the local absence of large trees on otherwise forested slopes. Estimating avalanche frequency was one of the earli- est applications of dendrochronology (Sherzer 1905). Other forms of mass wasting, such as debris flows and landslides, also have been dated in this way (Shroder 1978; Yamaguchi and Hoblitt 1995; Yoshida et al. 1997). Rates of river channel migra- tion (Gottesfeld and Gottesfeld 1990) and river flood frequencies (Everitt 1968) have been inferred from intra-ring abnormalities which appear as zones of enlarged vessels (Yanosky 1983). Similarly, chronologies of shoreline processes resulting from ice-push activity, wave action and lake level fluctua- tions induced by variations in precipitation have been established from tree growth patterns (Alestalo 1971; Begin and Payette 1988; Lepage and Begin 1996). Slumping and earth flows frequently create char- acteristic features that can be recognized by their vegetation cover which contrasts with that of the sur- roundings. The extent of the slump is well defined initially, but becomes progressively masked as vege- tation reestablishes at the site. Nevertheless, compar- atively recent slope activity in forested areas may still be detected by differences in the age and struc- ture of the plant cover. In addition, slope instability can cause trees to tilt. Reaction wood with distinct cellular structure is subsequently produced in the trunk in an attempt to restore the erect habit (Braam et al. 1987). Reaction wood in conifers forms on the compression side of the bend and favors the develop- ment of wider annual rings on that side of the trunk, which results in eccentric cross sections. Reaction wood is often darker than normal wood. Such defor- mities can provide dendrochronological evidence of slumping. Tilted trees are found at various points along the northeastern shore of Waskesiu Lake, Saskatchewan in conjunction with slope failures (Figure 1). The purpose of the present study was to evaluate the impact of slumping along the lakeshore on the com- munity structure of the vegetation, and to relate slumping events to lake level fluctuations using tree ring data. Study Area Waskesiu Lake is located in the east central part of Prince Albert National Park in northern Saskatchewan (53°57'N, 106°15'’W). This area was heavily glaciated and ice retreat occurred approxi- mately 11500 BP (Padbury et al. 1978). The lake basin and surrounding area consist of Wisconsin age tills and associated deposits at least 90 m thick. The tills are typically soft and unjointed, and consist of 106 2001 DE BOER AND ARCHIBOLD: SLUMPING ACTIVITY AND FOREST VEGETATION 107 FiGure 1. Characteristic bent trunks in White Spruce trees growing in slope failure sites along the northeast shore of /Waskesiu Lake. non-stratified deposits of boulders, gravels, sands, silts and clays. Veneers of stratified deposits of glaciofluvial, fluvial, and lacustrine origin are locally present. The drainage area of the Waskesiu Lake water- shed is approximately 500 km? and the surface area of lake is 68.3 km? (Gimbarzevsky 1973). Annual precipitation averages 455.7 mm, of which 32% falls as snow from October to April (Atmospheric Environment Service 1993). Rapid snowmelt, which normally occurs over a two-week period in late April, is an important feature of the hydrologic regime of the area. Water levels typically peak in June and July in response to snowmelt and summer thunderstorms. Humans have affected water levels of Waskesiu Lake in various ways since European set- tlement. Diversion of Beartrap Creek, an influent styeam, occurred as early as 1915. In 1936 a weir was constructed on the Kingsmere River where it enters Waskesiu Lake; this was designed to raise the water level in a section of rapids to facilitate boating in the National Park which was established in 1927. In 1938, the Beartrap Creek diversion was removed in an attempt to raise water levels following the drought years of the 1930s. The following year a dam was built near the head of the Waskesiu River where it exits Waskesiu Lake to improve boat access to the town site wharf (Anions and Beaven 1988). By 1941 water levels were so low that the dam was not functional (Rogers 1941), but in 1942 water lev- els had risen by as much as 0.75 m (Kooyman 1980). In 1962-1963 a new marina was constructed. To accommodate this facility, the dam on the Waskesiu River was upgraded and the lake level increased to an elevation of 532.18 m above sea level (Environment Canada 1989) compared to the pre- impoundment elevation of Waskesiu Lake of 530.3 m above sea level (Rawson 1936). There has been no set policy governing regulation of water levels in the park, and for the period of record (1954-1993) lake level has fluctuated by approximately 1 m (Figure 2). A weir on the Kingsmere River midway between Kingsmere Lake and Waskesiu Lake was removed in 1999. The site is located in the Mixedwood Forest Section of the boreal forest (Rowe 1972). The char- acteristic vegetation is mixed stands of White Spruce (Picea glauca) and Trembling Aspen (Populus tremuloides), with White Birch (Betula papyrifera), Balsam Poplar (Populus balsimifera) and Balsam Fir (Abies balsamea) locally present. The dominant soils in the area are shallow, medium-textured Luvisols with clayey B-horizons overlying sandy parent mate- rials. Along the northeastern section of the lakeshore the land rises abruptly with slopes of 20° to 40° ter- ’ 108 THE CANADIAN FIELD-NATURALIST Vol. 115 700 600 500 400 annual precipitation (mm) 300 1950 1955 1960 1965 1970 1975 1980 1985 1990 1995 year (AD) 532.75 532.50 532.25 June - July lake level (m) 532.00 1950 1955 1960 1965 1970 1975 1980 1985 1990 1995 year (AD) FicureE 2. Annual precipitation at Waskesiu Lake from 1966 to 1991 (upper) and mean water level in June for Waskesiu Lake for the period 1954 to 1983 (lower). 2001 DE BOER AND ARCHIBOLD: SLUMPING ACTIVITY AND FOREST VEGETATION 109 TABLE |. Circumference and age of trees in forest and on the perimeter of slump sites along the shore of Waskesiu Lake. Circumference (cm) Mean S:D; Max Undisturbed forest Balsam Fir 36.5 12.3 54 White Spruce 65.2 29.3 117 White Birch 5521 21.0 94 Perimeter of slumps Balsam Fir 34.8 15.5 56 White Spruce 54.6 34.9 133 White Birch 60.8 25.4 117 Trembling Aspen 54.0 $Y 64 minating in a pronounced convexity at the upper end of the slopes. The difference in elevation between the lake surface and the crest of the slopes ranged from 10 to 25 m and the horizontal distance from the shoreline to the crest varied from 25 to 55 m. Further inland a boggy depression has formed which acts a cistern for water draining through the surface materi- al towards the lake. Wave action has produced a low ridge of coarse boulders eroded from the glacial till along the base of the slopes, approximately 2.5 m wide and up to 1.5 m high. At the shoreline, cone- shaped slumps have developed, varying from 10.3 m to 39.8 m in width, with a sparsely colonized, sandy substrate that contrasts with the adjacent forested slopes. Methods Eleven recent slumps and four intervening, stable sites were selected for detailed investigation. Transects were located along the length of the slopes at right angles to the shoreline. The tree cover was assessed from 5 m-wide belt transects in which all individuals were noted for species and trunk circum- ference. Tree core and trunk diameter data were col- lected from 250 trees growing on the slumps and in the undisturbed forest beyond the influence of slumping. The rings were counted under a low- power binocular microscope and used in conjunction with tree diameter to establish a general age-size relationship that could be applied to the different species growing in the study area. Cores from selected White Spruce trees were ana- lyzed in detail to evaluate the temporal changes in ring width. White Spruce was chosen because it was ubiquitous in the field area, and because its rings were easily visible and well preserved throughout the whole length of the core. Ring widths were mea- sured for 28 White Spruce growing in the forest; the oldest of these provided a growth record back to 1897. An additional 41 White Spruce growing at the perimeter of the slumps also were measured and pro- vided a growth record back to 1904. Age (y) Min Mean SA. Max Min 17 46.7 9.4 67 21 24 72.9 22:3 98 44 27 80.6 Lich 98 67 15 30.2 11.6 46 13 15 38.3 216 88 10 17 76.4 16.7 89 30 45 72.8 9.7 81 58 At one site with recent slumping, five trees with marked trunk curvature were felled and sectioned into 1 m lengths for detailed ring analysis including assessment of annual ring widths and eccentricity of growth. The sample size was restricted by Parks Canada in order to minimize disturbance of the lakeshore forest. The age at which reaction wood developed was noted, and is assumed to be the date at which slumping occurred. Tree ring chronologies were subsequently compared to lake level and annu- al precipitation records. Results and Discussion The oldest White Spruce trees in the mature forest beyond the crest of the slope were approximately 100 years old. The oldest White Birch were of a sim- ilar age, although Balsam Fir was typically less than 50 years old. This area was burned about 100 years ago, and the age of the tree cover is consistent with the fire history and ecology of the dominant species. The circumferences and ages of the conifers growing in the forest are presented in Table 1. Circum- ferences ranged from 36.5 + 12.3 cm for Balsam Fir to 65.2 + 29.3 cm for White Spruce; this correspond- ed to a mean age of 47 + 9 years and 73 + 22 years, respectively. For White Birch mean circumference was 55.1 +21.0 cm giving a mean age of 81 +12 years. The average tree density in the forest was 3160 individuals ha". On the intervening slopes between the slumps, average tree density ranged from 1950 + 174 indi- viduals ha’! at the shoreline to 4300 + 240 individu- als ha! at 30-40 m from the lake. Balsam Fir was the most abundant species on these slopes and accounted for 54% of the tree cover compared to 32% for White Spruce, 8% for White Birch and 6% for Trembling Aspen. The oldest White Birch on the intervening slopes was aged at 89 years, compared to 88 years for White Spruce, 81 years for Trembling Aspen, and 46 years for Balsam Fir. Species com- mon to the boreal forest, such as Hazel (Corylus cor- nuta), Twinflower (Linnaea borealis) and Two- 110 THE CANADIAN FIELD-NATURALIST Vol. 115 TABLE 2. Density (individuals/m7) of ground cover species on recent slumps and intervening slopes. Om 10m 20m Herbs and grasses Achillea sibirica 0.5 Anemone canadensis ae Apocynum androsamifolium 0.1 Arabis holboeillii 0.4 0.1 Aralia nudicaulis 1.8 13) 0.6 Artemisia abronatum 0.5 0.4 0.3 Artemisia absinthium 1.8 DES Dll Aster ciliolatus 0.4 Dek 0.1 Aster conspicuus eS 108) 0.5 Aster erocoides 0.1 Aster hesperius 4.3 0.1 Campanula rotundifolia 0.8 1.0 126 Cicuta maculata 2.3 Cirsium arvense 0.8 Cornus canadensis 0.1 Elymus innovatus 0.8 3.0 3.8 Equisetum arvense DIS} Dl Equisetum hyemale 18) Erigeron philadelphicus 0.1 Fragaria vesca 0.7 Fragaria virginiana 0.5 Galium borealie JMO 4.2 14.7 Lathyrus ochroleucus NEO eg 3.4 Lathyrus venosus 0.8 0.4 Linnaea borealis Maianthemum canadense 0.8 0.4 1D) Mentha arvensis Bed, Parnassia palustris 2.0 Solidago candensis 5.8 DS 0.8 Solidago sp. is) 0.1 2-3 Sonchus arvensis 6.0 Stachys palustris 0.2 Thalictrum venulosum 0.7 0.8 Urtica dioica 0.6 0.1 Vicia americana 1.6 4.7 4.8 total herbs 67.2 31.0 40.5 Shrubs Actaea rubra 0.1 0.6 1.8 Alnus rugosa 0.1 0.7 0.2 Amelanchier alnifolia 0.2 10) DD, Corylus cornuta 0.4 Ribes americanum 0.7 0.1 Rosa acicularis 123 6.6 6.5 Rubus strigosus 0.9 0.1 Symphoricarpos albus 0.3 25 0.3 Symphoricarpos occidentalis 0.6 0.2 total shrubs B35 Wee? 11.6 leaved Solomon’s seal (Maianthemum canadense), as well as those typically associated with open or disturbed sites, for example Canada thistle (Cirsium arvense) and Rose (Rosa acicularis) were present in the understory (Table 2). The groundcover at these sites was composed of 14 species of herbs and grass- es and 4 species of shrubs. slumps intervening slopes 30m 40m 50m total total 0.5 ey 0.1 0.5 32 303) 1.6 iL ilatss 8.5 ill 0.4 o9 ai 8.3 4.3 5 ies) Heal 4.5 0.1 0.1 4.5 8.0 0.1 S55) 25 0.8 1.8 ie) 0.9 2.5 0.5 1.6 ool} 25.0 ty) 0.1 0.3 oy 4.2 4.8 0.5 9.4 2.4 32.4 355 WD 2.8 2 12.4 0.8 0.9 2M 0.1 0.1 6.5 6.5 aol 3.8 18.0 5.8 3.7 2.8 2.0 4.3 8.9 O)S) 3:6 6.0 GO 15 0.5 ed DD 1.4 14.8 0.5 34.6 22:1 7.0 0.5 0.4 a) 0.8 1.0 0.4 Bn, ie 0.4 is 10.3 1.0 1.8 Sal 2.6 20.6 6.3 1.4 Dep 4.6 0.9 4.0 PES) 0.8 8.0 6.7 Vegetation cover was extremely variable on the slumps. Average tree density adjacent to the shore- line was 2441 + 1922 individuals ha™', and decreased progressively away from the shoreline to 1195 + 668 individuals ha‘! at a distance of 10-20 m and 672 + 498 individuals ha‘! at the head of the scars at a dis- tance of 30-40 m. White Spruce was the most abun- 2001 DE BOER AND ARCHIBOLD: SLUMPING ACTIVITY AND FOREST VEGETATION 11] TABLE 3. Percent composition, circumference and age of trees growing on the recent slump sites 0-10 m 10-20 m Number of individuals Balsam Fir 6 8 Balsam Poplar s 6 Trembling Aspen 0 7 White Birch 13 5 White Spruce x 3 Percent composition Baisam Fir 7.8 10.4 Balsam Poplar 3.9 7.8 Trembling Aspen 9 White Birch 16.9 6.5 White Spruce 71.4 16.8 Trunk circumference (cm) Balsam Fir NS Tl 11.4+4.1 Balsam Poplar 38.0 + 46.6 5.4+3.] Trembling Aspen 4.6+1.8 White Birch 65.5 226.0 S83 L.6 White Spruce OS S24 5.4 213.1 *one individual dant tree species on the scars and accounted for 54% of the individuals compared to 13% for Trembling Aspen, 12% for Balsam Fir, 11% for White Birch and 10% for Balsam Poplar (Table 3). A total of 35 herb and grass species was present on the slumps. Northern Bedstraw (Galium borealie) was the most abundant (32.4 individuals/m7’) with Horsetail (Equisetum arvense), Two-Leaved Solomon’s Seal, American Vetch (Vicia americana) and Wild Sasaparilla (Aralia nudicaulis) also common (Table 20-30 m 30-40 m Total 6 | 21 6 2 17 3 14 24 | 0 19 12 16 96 7.8 1.3 11.9 7.8 2.6 9.6 3.9 18.2 13.6 [fee 0 10.7 ss 20.7 54.2 14.5 + 11.6 4.5* 4.1+0.9 25+ 0.7 2.9'= 1.6 67.7 = 103 60* 4.1.0.7 2. OT 2). The greatest density of herbs and grasses occurred near the base of the slumps (67.2 individu- als/m?) and density decreased to 7 individuals/m/? at the slump heads. The general increase in herbaceous cover on the lower part of the slumps is related to moisture seeping into the exposed sandy substrate and the near-surface water table along the shoreline. Nine species of shrubs were growing on the slumps. Prickly rose (Rosa acicularis) was the dominant shrub (20.6 individuals/m7). Other species, including FIGURE 3. Average annual wind direction at Waskesiu Lake for the period 1977 to 1982 (left) and lake fetch for the north- east shore of Waskesiu Lake (right) [after Cote 1984]. 112 THE CANADIAN FIELD-NATURALIST Vol. 115 5 FI 5 P1 £ £ E E = 3 £ 3 ie} 2} 4 = ia 2 Dm 2 = Cc ‘ec ‘= 1 1 0 0 1900 1920 1940 1960 1980 2000 1900 1920 1940 1960 1980 2000 year (AD) year (AD) 5 F2 5 P2 ee yt — 4 £ £ £ £ Se 3 = 3 as) 6 4 = oD) 2 5) 2 (= Cc oc ‘cc 1 1 0 0 1900 1920 1940 1960 1980 2000 1900 1920 1940 1960 1980 2000 year (AD) year (AD) 5 F3 5-P3 = = E £ le 3 fe 3 6 5 S S cee op) 2 (= Cc = 41> 1 1 0 0 1900 1920 1940 1960 1980 2000 1900 1920 1940 1960 1980 2000 year (AD) year (AD) FicureE 4. Annual tree ring widths for selected White Spruce ring trees growing in undisturbed forest and around the perimeter of slope failures. Raspberry (Rubus strigosus) and Snowberry (Symphoricarpos albus) were represented at densi- ties ranging from 0.8 to 4.6 individuals/m7). The shrubs were most abundant in the lower part of the slumps with maximum density (12.1 individuals/m7) occurring 10 m from the shoreline. The trees at the base of the slumps were generally larger than those further upslope. Mean circumfer- ence for White Spruce ranged from 33.7 cm near the shoreline compared to 2.5 cm at the top of the scars. Similarly, mean circumference for Balsam Poplar decreased from 38.0 cm to 2.5 cm away from the shoreline. Some large White Birch and Trembling Aspen (> 60 cm circumference, age approximately 80 years) occurred on the scars and presumably survived the slumping events. Trees with stems typically less than 20 cm in circumference appeared to have estab- lished within the past 10 years. The older trees which were occasionally present at the base of some of the slumps typically exhibited marked curvature of the 2001 trunk in response to slope instability. In some indi- viduals the stems were bent near the top so that the lower trunk tended to grow perpendicular to the slope while the upper trunk had regained its vertical orien- tation. In other trees, the trunks were curved at the base which suggests that disturbance occurred when they were relatively young. The ages of the five White Spruce trees that were sampled ranged from 56 to 102 years: the curved part of the trunks developed 10, 15 and 21 years prior to felling which indicated slumping had occurred in 1973, 1979 and 1984. Weather records and lake level data for Waskesiu Lake are incomplete. However, precipitation in 1973 and 1984 totaled 534 mm and 523 mm, respectively, which is well above the long-term normal of 398.9 mm. In 1978, one year prior to the 1979 slope distur- bance, 526 mm was received (Figure 2). Similarly, the surface lake level in June 1973 was 532.6 m, the highest of any year in the 1970s, and in 1979 it was also high at 532.5 m. The configuration of the shore- line in this part of the lake provides a fetch of approximately 7.5 km to the south and southeast. This extent of open water coupled with the frequen- cy of winds from this quadrant (Figure 3) could pro- duce significant wave action along the base of the slopes Consequently, slumping in this environment is most likely triggered by high moisture contents of the surficial materials caused by subsurface seepage from the boggy area inland supplemented by wave action. Many of the tree seedlings found on the slumps exhibited bent stems caused by burial of the young tissue. This suggests that slope adjustment continues for several years following the initial slope failure. Such instability can be attributed to ground- water seepage, rain splash and the activity of ants, all of which were observed to dislodge the loosely con- solidated mineral soil particles on these poorly vege- tated slopes. Characteristic ring width patterns for trees in undisturbed forest and those growing adjacent to the slumps are presented in Figure 4. The older forest trees (F1, Figure 4) typically exhibited curvilinear g¢owth patterns in which the annual ring width grad- ually decreased with age. For this individual, early growth may have been stimulated by a tree fall gap, but average annual growth declined to less than 1 mm y! with subsequent canopy closure. Incremental growth rates for younger trees (F2, Figure 4) that have been suppressed by the overstory were typical- ly about 0.5 mm y"!. Many of the older trees growing along the perimeter of the slumps exhibit accelerated growth phases. For example, early growth rings for specimen P1 (Figure 4) exceeded 4 mm y"! in some years, decreased sharply in the 1930s, and exceeded 3 mm y7! in the 1950s and 1960s. The growth pattern suggests that the tree has survived periodic slump- ing. Initial establishment and rapid early growth probably coincides with a reduction in competition DE BOER AND ARCHIBOLD: SLUMPING ACTIVITY AND FOREST VEGETATION 113 at the site while the period of depressed activity may reflect root disturbance associated with a subsequent slump. This tree survived and again benefited from the disturbance. A corresponding period of accelerat- ed growth during the early 1960s can be seen in younger individuals (P2, Figure 4) growing along this active shoreline, while trees which established soon after these slumps occurred (P3, Figure 4) have grown more rapidly than their counterparts in the forest. Tree growth rates generally decline with age and there is a corresponding decrease in annual ring widths. However, in some forest trees (e.g. F3, Figure 4) increased growth was noted in the mid 1970s. This may be linked to higher than normal pre- cipitation during this decade (Figure 2). Conclusions Slope failures along the northeastern shoreline of Waskesiu Lake can be identified by distinctive vege- tation patterns. Recently disturbed slopes were char- acteristically colonized by a sparse cover of herba- ceous plants and tree seedlings, and contrasted with the adjacent slopes which supported a mixed forest cover dominated by White Spruce, Balsam Fir and Paper Birch. On the slumps, 35 herb and grass species were found. Northern Bedstraw (Galium borealie) was the dominant species, followed by Horsetail (Equisetum arvense), Two-Leaved Solomon’s Seal (Maianthemum canadense), American Vetch (Vicia americana) and Wild Sasaparilla (Aralia nudicaulis). The greatest density of herbs and grasses occurred near the base of the slumps, where soil moisture conditions were the most beneficial for plant growth. Slumping is most likely caused by subsurface seepage, with annual fluctuations in lake level and seasonal wave activity possibly contributing to the process. Tree ring data indicated that slumping occurs periodically with the most recent events dated at 1973, 1979 and 1984. These years either received higher than normal precipitation or had above normal lake levels (Figure 2). As well as contributing to higher lake levels, wet years can also promote slope instability by increasing the water content of the sur- ficial materials, thereby decreasing their shear strength. Statistical analysis indicated that there is no simple relationship between precipitation, tempera- ture, lake level and the occurrence of slumping. The impact of slumping on the forest vegetation structure, however, is substantial. Consequently, lake level changes caused by changes in precipitation and tem- perature and by human activities such as the removal of the Kingsmere River weir in 1999 may initiate fur- ther changes in the shoreline forest vegetation. Acknowledgments Funding for this study was provided by Prince Albert National Park. We thank Pamela Nelson and 114 Dave Bretell for carrying out the fieldwork, Jeff Weir for assisting in sampling the trees, and Gary Crosby for the tree ring analyses. Literature Cited Alestalo, J. 1971. Dendrochronological interpretation of geomorphic processes. Fennia 105: 1-140. Anions, D. W., and L. W. Beaven. 1988. Water control structures and aquatic resources in Prince Albert National Park. Jn Use and Management of Aquatic Resources in Canada’s National Parks. Edited by D. C. Harvey, S. J. Woodley and A. R. Haworth. Heritage Resources Centre, University of Waterloo, Occasional Paper 11, Waterloo, Ontario, pages 180-197. Atmospheric Environment Service. 1993. Canadian Climate Normals 1961-90. Prairie Provinces. Minister of Supply and Services, Ottawa, Canada. Braam, R. R., E. E. J. Weiss, and P. A. Burrough. 1987. Dendrogeomorphological analysis of mass movement: a technical note on the research method. Catena 14: 585-589. Begin, Y., and S. Payette. 1968. Dendroecological evidence of lake-level changes during the last three centuries in subarctic Quebec. Quaternary Research 30: 210-220. Cote, M. 1984. Climate of Prince Alberta National Park. Jn Prince Albert National Park Resource Description and Analysis. Environment Canada, Parks — Prairie and Northern Region, Winnipeg. 86 pages. Environment Canada. 1989. Prince Albert National Park Aquatic Resource Management Plan. Resource Management Report 89-1 P.A., Canadian Parks Service, Natural Resource Conservation, Prince Albert National Park, Waskesiu Lake, Saskatchewan. 154 pages. Everitt, B. L. 1968. Use of the cottonwood in an investiga- tion of the recent history of a flood plain. American Journal of Science 266: 417-439. Gimbarzevsky, P. 1973. Prince Albert National Park natu- ral resources. Forest Management Institute, Canadian Forestry Service, Department of the Environment, Ottawa, Ontario. Gottesfeld, A. S., and L. M. Johnson Gottesfeld. 1990. THE CANADIAN FIELD-NATURALIST Vol. 115 Floodplain dynamics of a wandering river, den- drochronology of the Morice River, British Columbia. Geomorphology 3: 159-179. Kooyman, A. H. 1980. The Walleye population of Waskesiu Lake, Prince Albert National Park, 1975-— 1978. Canadian Wildlife Service, Winnipeg. 123 pages. Lepage, H., and Y. Begin. 1996. Tree-ring dating of extreme water level events at Lake Bienville, subarctic Quebec, Canada. Arctic and Alpine Research 28: 77-84. Padbury, G. A., W. K. Head, and W. E. Souster. 1978. Biophysical Resource Inventory of the Prince Alberta National Park, Saskatchewan. Saskatchewan Institute of Pedology Publication S185. 560 pages. Rawson, D. S. 1936. Physical and chemical studies in lakes of the Prince Albert National Park, Saskatchewan. Journal of The Biological Board of Canada 2: 227-283. Rogers, H. M. 1941. Maintenance of water levels in Lake Waskesiu. National Park Branch, Calgary. 3 pages. Rowe, J. S. 1972. Forest Regions of Canada. Canadian Forestry Service. Department of the Environment Publication Number 1300. Sherzer, W. H. 1905. Glacial studies in the Canadian Rockies and Selkirks. Smithsonian Miscellaneous Collections 47: 453-496. Shroder, J. F. 1978. Dendrogeomorphological analysis of mass movement on Table Cliffs Plateau, Utah. Quater- nary Research 9: 168-185. Yamaguchi, D. K. and R. P. Hoblitt. 1995. Tree-ring dat- ing of pre-1980 volcanic flowage deposits at Mount St. Helens, Washington. Geological Society of America Bulletin 107: 1077-1093. Yanosky, T. M. 1983. Evidence of floods on the Potomac River from anatomical abnormalities in the wood of flood-plain trees. United States Geological Survey Professional Paper 1296: 1-42. Yoshida, K., S. Kikuchi, F. Nakamura, and M. Noda. 1997. Dendrochronological analysis of debris flow dis- turbance on Rishiri Island. Geomorphology 20: 135-145. Received 31 January 2000 Accepted 30 October 2000 Rare and Endangered Fishes and Marine Mammals of Canada: COSEWIC Fish and Marine Mammal Subcommittee Status Reports: XIII R. R. CAMPBELL 131 Dahlia Avenue, Ottawa, Ontario K1H 6G1, Canada Campbell, R. R. Editor. 2001. Rare and endangered fish and marine mammals of Canada: COSEWIC Fish and Marine Mammal Subcommittee Status Reports: XIII. Canadian Field-Naturalist 115(1): 115-117. Seven status reports representing the 1998 fish and marine mammal status assignments have been prepared for publication. Committee (COSEWIC) and Subcommittee (Fish and Marine Mammals) activities are briefly discussed. Sept rapports de statut relatifs aux poissons et aux mammiféres marins auxquels ont été attribués un statut en 1998 ont été préparés pour publication. Les activités du Comité (CSEMDC) et du sous-comité (des poissons et des mammiféres marins) sont briévement discutées. Key Words: Rare and Endangered species, fish, marine mammals, COSEWIC. As indicated in previous submissions (Campbell 1984 through 1998), the intent of the Subcommittee on Fish and Marine Mammals is to publish (as fund- ing permits) the status reports (on those species of fish and marine mammals) which the Committee on the Status of Endangered Wildlife in Canada (COSEWIC) has reviewed, approved and used as a basis of assigning status to species in jeopardy in Canada. The group of reports presented here repre- sent seven of the nine fish and marine mammals con- sidered by COSEWIC which were assigned status in 1998 (see COSEWIC 1998). In addition, reports on the Redfin Pickerel (Esox americanus americanus) and the Atlantic Cod (Gadus morhua) were also con- sidered (the former was found to be Not At Risk, and the latter Vulnerable), but revised manuscripts have not as yet been received from the authors. In the meantime, summaries of these (and any and all reports) are available from the COSEWIC Secretariat (Canadian Wildlife Service, Environment Canada, Ottawa, Ontario K1A 0H3). It is hoped that we will be able to find continuing financial support to offer, in succeeding volumes, those reports reviewed in future years, as well as those not received in time for this publication. Progress COSEWIC has undertaken to make available to all Canadians supporting information on each species classified (see Cook and Muir 1984; Shank 1999). The Fish and Marine Mammal Subcommittee has been able to use this journal as one step in achieving the goal. A series of reports have appeared in various volumes and numbers from 1984 through 1998 [see Canadian Field-Naturalist 98(1): 63-133; 99(3): 404-450; 101(2): 165-309; 102(1): 81-176 and 102(2): 270-398; 103(2): 147-220; 104(1): 1-145; 105(2): 151-250; 106(1): 1-72; 107(4) 395-546; 110(3): 462-532; 111(2): 249-307; 112(1): 94-157]. As of April 1998, COSEWIC has reviewed the status of 97 fish species, two marine invertebrates, and 43 marine mammals (see COSEWIC 1998). Of the 155 species (or discrete populations) investigated eight are indeterminate (six fish, two marine mam- mals), 63 (29 fish, 33 marine mammals, one marine invertebrate) have been found not to require status designation and another 47 (41 fish, six marine mammals) have been designated as vulnerable, mainly due to natural rarity; leaving 27 species (18 fish, nine marine mammals) of immediate concern (threatened and endangered), and 10 species (six fish, three marine mammals, and one marine mol- lusc) extinct or extirpated. As of April 1998 there are 50 status reports on fish species (includes 22 updates), one marine mollusc (update), and 11 on marine mammal species (six updates) under review or in preparation (Table 1; Campbell 1998: Table 1). Several of these were to be presented to the Committee for status assignment in 12>) As well, some 73 additional species of fish (plus 21 to be updated), one marine mammal update, and 12 marine invertebrates have been identified as being worthy of consideration (Table 2; Campbell 1998: Table 2). A few may be found to not require status designation, but the process serves to bring together the information necessary to make the appropriate determination and satisfy the need to fill those knowledge gaps. Although some of these may be of no immediate concern, the Subcommittee will, as opportunity allows, attempt to document the sta- tus of these species to determine their status in Canada. 115 116 THE CANADIAN FIELD-NATURALIST TABLE 1. New (since Campbell 1998: Table 1) fish and marine mammal species updates and reports which are in preparation or under review as of April 1998. Species FISH Morrison Creek Lamprey* Aurora Trout°* Shortnose Cisco° Shortjaw Cisco® Redside Dace® Silver Shiner® Greater Redhorse Hadley Lake Stickleback Species Pair* Balkwell Lake Stickleback Species Pair* Emily Lake Stickleback Species Pair* Priest Lake Stickleback Species Pair* MARINE MAMMALS Fin Whale° Humpback Whale® Northwest Atlantic Northeast Pacific Harbour Porpoise Northwest Atlantic® Northwest Atlantic® *Endemic to Canada °Updated Status Report Scientific Name Lampetra richardsoni Salvelinus fontinalis timagamiensis Coregenus reighardi Coregonus zenithicus Clinsotomus elongatus Notropis photogenis Moxostoma valenciennesi Gasterosteus spp. Gasterosteus spp. Gasterosteus spp. Gasterosteus spp. Balaenoptera physalus Megaptera novaeangliae Phocoena phonoeca Proposed Status Threatened Endangered April 1987 Threatened April 1985 Threatened April 1987 Vulnerable April 1987 Vulnerable April 1987 Vulnerable Extinct Endangered Endangered Endangered Vulnerable April 1987 Vulnerable April 1985 Threatened April 1982 Threatened April 1990 Threatened April 1990 Vol. 115 review process initiated in 1993 (Table 2; Campbell 1998: Table 2) for those species which had not already received further examination following the initial assignment of status. In addition to soliciting further status reports on species of concern, the Subcommittee continues to obtain updates on the status of selected species as new information becomes available, or in the 10 year TABLE 2. New (since Campbell 1998: Table 2) fish and marine mammal species of Possible interest to COSEWIC. (not listed by priority). Species Scientific Name Proposed Status FIsH Blackfin Cisco Coregonus nigripinnis Threatened April 1988 Kiyi Coregonus kiyi Vulnerable April 1988 Umatilla Dace Rhinichthys umatilla Vulnerable April 1988 Bigmouth Buffalo Ictiobus cyprinellus Vulnerable April 1989 Black Buffalo Ictiobus niger Vulnerable April 1989 Black Redhorse Moxostoma dusquesnei Threatened April 1988 Margined Madtom Noturus insignis Threatened April 1989 Redbreast Sunfish Lepomis auritus Vulnerable April 1989 Orangespotted Sunfish Lepomis humilis Vulnerable April 1989 Banded Killifish Fundulus diaphanus Vulnerable April 1989 Enos Lake Stickleback* Gasterosteus sp. Threatened April 1988 Fourhorn Sculpin Myoxocephalus quadricornis Vulnerable April 1989 Blackline Prickleback Acantholumpenus mackayi Vulnerable April 1989 Bering Wolffish Anarichus orientalis Vulnerable April 1989 MARINE MAMMALS Sowerby's Beaked Whale Mesoplodon bidens Vulnerable April 1989 *Endemic to Canada 2001 Concluding Remarks The seven reports included in the following series are reports on the status of the respective species in Canada. Status was assigned by consensus of the COSEWIC Committee based on these reports which are published under the name(s) of the original author(s). The reports have undergone minor editing to provide a brief introduction and some degree of consistency in format and presentation. Acknowledgments The members of COSEWIC and the Fish and Marine Mammal Subcommittee would like to extend their thanks to the various authors who have so gen- erously contributed their time and talents in support of COSEWIC. The Committee also wishes to acknowledge the members of the Subcommittee for their unstinting efforts in reviewing the reports and their helpful comments. COSEWIC is grateful to the Canadian Wildlife Federation, World Wildlife Fund Canada, the Canadian Wildlife Service, the Canadian Museum of Nature and the Royal Ontario Museum for assistance provided in cash and kind. A special mention to Francis Cook and The Canadian Field-Naturalist for assistance in publication and editing and to all mem- bers of COSEWIC for their dedication and interest in the future of Canada's flora and fauna. We gratefully acknowledge the financial support provided by Environment Canada which permitted the contracting of several new reports and publication of this series. Special thanks are due C. Renaud, Canadian Museum of Nature for editing French abstracts. Literature Cited Campbell, R. R. 1984. Rare and endangered fish of Canada: The Committee on the Status of Endangered Wildlife in Canada (COSEWIC) Fish and Marine Mammal Subcommittee. Canadian Field-Naturalist 98(1): 71-74. Campbell, R. R. 1985. Rare and endangered fish and marine mammals of Canada: COSEWIC Fish and Marine Mammals Subcommittee Status Reports: II. Canadian Field-Naturalist 99(3): 404408. Campbell, R. R. 1987. Rare and endangered fish and marine mammals of Canada: COSEWIC Fish and CAMPBELL: FISHES AND MARINE MAMMALS OF CANADA STATUS REPORTS XIII Pa7 Marine Mammal Subcommittee Status Reports: III. Canadian Field-Naturalist 101(2): 165-170. Campbell, R. R. 1988. Rare and endangered fish and marine mammals of Canada: COSEWIC Fish and Marine Mammal Subcommittee Status Reports: IV. Canadian Field-Naturalist 102(1): 81-86. Campbell, R. R. 1989. Rare and endangered fish and marine mammals of Canada: COSEWIC Fish and Marine Mammal Subcommittee Status Reports: V. Canadian Field-Naturalist 103(2): 147—157. Campbell, R. R. 1990. Rare and endangered fish and marine mammals of Canada: COSEWIC Fish and Marine Mammal Subcommittee Status Reports: VI. Canadian Field-Naturalist 104(1): 1-6. Campbell, R. R. 1991. Rare and endangered fish and marine mammals of Canada: COSEWIC Fish and Marine Mammal Subcommittee Status Reports: VII. Canadian Field-Naturalist 105(2): 151-156. Campbell, R. R. 1992. Rare and endangered fish and marine mammals of Canada: COSEWIC Fish and Marine Mammal Subcommittee Status Reports: VIII. Canadian Field-Naturalist 106(1): 1-6. Campbell, R. R. 1993. Rare and endangered fish and marine mammals of Canada: COSEWIC Fish and Marine Mammal Subcommittee Status Reports: IX. Canadian Field-Naturalist 107(4): 395-401. Campbell, R. R. 1996. Rare and endangered fish and marine mammals of Canada: COSEWIC Fish and Marine Mammal Subcommittee Status Reports: X. Canadian Field-Naturalist 110(3): 454461. Campbell, R. R. Editor. 1997. Rare and endangered fish and marine mammals of Canada: COSEWIC Fish and Marine Mammal Subcommittee Status Reports: XI. Canadian Field-Naturalist 111(2): 249-257. Campbell, R. R. Editor. 1998. Rare and endangered fish and marine mammals of Canada: COSEWIC Fish and Marine Mammal Subcommittee Status Reports: XII. Canadian Field-Naturalist 112(1): 94-97. Cook, F. R. and D. Muir. 1984. The Committee on the Status of Endangered Wildlife in Canada (COSEWIC): History and progress. Canadian Field-Naturalist 98(1): 63-70. COSEWIC. 1998. Canadian species at risk: April 1998. COSEWIC Secretariat, c/o Canadian Wildlife Service, Environment Canada, Ottawa Ontario K1A 0H3. Shank, Christopher C. 1999. The Committee on the Status of Endangered Wildlife in Canada (COSEWIC): a 21-year retrospective. Canadian Field-Naturalist 113(2): 318-341. Accepted 29 May 2000 Status of the White-Beaked Dolphin, Lagenorhynchus albirostris, in Canada’ JON LIEN!, DAWN NELSON?, and DONG JIN HAP !Whale Research Group, Memorial University of Newfoundland, St. John's, Newfoundland AIC 5S7, Canada 2College of the Atlantic, Bar Harbour, Maine, USA 3Institute of Oceanology, Academia Sinica, 7 Nan Hai Road, Qingdao, Shandong, China Lien, Jon, Dawn Nelson, and Dong Jin Hai. 2001. Status of the White-beaked Dolphin, Lagenorhynchus albirostris, in Canada. Canadian Field-Naturalist 115(1): 118-126. White-beaked Dolphins are common and abundant in the north Atlantic. Abundance appears greater in northeastern areas compared to the northwestern Atlantic where its range extends from Davis Strait to the Hague Line on George’s Bank, and into U.S. waters. Relationships between populations observed in different areas are not known. Total or local populations are not known to be reduced or depleted. The biology and ecology of this species has been incompletely studied. It occurs in social groups ranging from dozens to hundreds of individuals; there is some indication that groups may segregate based on age/maturity. Sexual maturity occurs in females at lengths from 1.74 to 2.49 m; 50% are mature at 2.42 m. Males are larger at sexual maturity (2.51 to 2.57 m). Factors limiting northwestern Atlantic White-beaked Dolphins may include hunt- ing, incidental entrapments in fishing gear, ice entrapments and, possibly, pollution. Such factors should be monitored carefully but are not, at present, believed to have substantial impacts on the species, although there may be reduction of some local populations. Factors such as ice entrapments, parasites, infections, mass strandings and predation are recog- nized as naturally occurring mortality and are not presumed to normally limit the species. There is no evidence of any major threat to this species in zones under Canadian jurisdiction. However, careful monitoring of hunting activities in Labrador is recommended and fisheries by-catches should be carefully monitored. Key Words: Cetacea, Odontoceti, toothed whales, White-beaked Dolphin, Dauphin a nez blanc, Lagenorhynchus albi- rostris. The White-beaked Dolphin, Lagenorhynchus albi- rostris Gray 1846 (Figure 1), is one of two species of Lagenorhynchus found in the North Atlantic, the other being the Atlantic White-sided Dolphin, Lagenorhynchus acutus. On occasion, groups may contain both species (Haase 1987). Lagenorhynchus albirostris can be differentiated from the White-sided Dolphin by its more diffuse colouration, especially just behind the dorsal fin, and fewer teeth (Mercer 1973; Leatherwood et al. 1976; Reeves et al. 1996*). White-beaked Dolphins generally have a more northerly distribution than White-sided Dolphins. In the literature Lagenorhynchus albirostris is, on occasion, called the Whitenose Dolphin (Amundin and Amundin 1975). Along the Labrador coast they are most commonly known locally as “jumpers” or “squidhounds”’, but are also inappropriately called “herring hogs”. The latter name is more precisely applied to Minke Whales, Balaenoptera acutorostrata (Reeves et al. 1996*). The name “grampus” is fre- quently applied indiscriminately to all dolphin spe- cies, and to Minke Whales in some areas of New- foundland and Labrador (Lien et al. 1985). Reeves et al. (1996*) gives common names for White-beaked Dolphins in European waters. *Reviewed by COSEWIC April 1986, Not At Risk. *See Documents Cited section. 118 Although sightings of Lagenorhynchus albirostris have been reported as common, and it is often found stranded on both sides of the Atlantic, little is known about its biology or population ecology (Klinowska 1991). The species is commonly ice entrapped along the Newfoundland coast, occasionally caught inci- dentally in fishing gear (Dong et al. 1996; Lien 1988), and opportunistically hunted (Alling and Whitehead 1987; Brice-Bennet et al. 1977). The purpose of this paper is to review the current status of Lagenorhyn- chus albirostris in the northwest Atlantic in waters under Canadian jurisdiction. Appearance ) Lagenorhynchus albirostris (Figure 1) is generally black or dark grey, with one patch of light grey behind the dark dorsal fin and another which extends along the side (Jonsgard 1962; Mercer 1973; Leatherwood et al 1976). The “white beak”, from Which the species derives its common name, can in fact be dark grey or mottled (Jonsgard 1962; Corbet and Southern 1977; Leatherwood and Reeves 1983). Anterior to the dorsal fin in some animals there is a grey chevron which can extend into an irregular strip along the animal’s sides, and into the dorsal patch. The dorsal patch behind the dorsal fin may vary from white to shades of grey and is a useful field trait for distinguishing Lagenor- hynchus albirostris from Lagenorhynchus acutus. The underside of White-beaked Dolphins is white; the flip- 2001 LIEN, NELSON, AND HAI: STATUS OF WHITE-BEAKED DOLPHIN IN CANADA 119 FIGURE |. White-beaked dolphin (Lagenorhynchus albirostris). Drawn by Dawn Nelson. pers and fluke are black. There is considerable varia- tion in colouration among individuals (Reeves et al. 1996*), some of which may be attributable to age (Jonsgard 1962; Mercer 1973; Corbet and Southern 1977; DeBoer 1989). Distribution The White-beaked Dolphin is found only in cold temperate and subarctic waters of the north Atlantic (Leatherwood and Reeves 1983; Katona et al. 1993); it appears more common in eastern than western waters. In the northwest Atlantic (Figure 2), Lagen- orhynchus albirostris occurs from eastern Greenland (Jonsgard 1962; Jonsgard and Christensen 1968; Christensen 1972; Benjaminsen et al. 1976), through the Davis Strait and south to Massachusetts (Mercer 1973; Leatherwood et al. 1976; Leatherwood and Reeves 1983; Boles 1980*; Winn and Edel 1980; Alling and Whitehead 1987). In the east, the species has been observed off Iceland (Mercer 1973; Andrésson 1978), the Norwegian Sea (Haug et al. 1981; Leatherwood and Reeves 1983; @ritsland et al. 1989), the Barents Sea (Christensen 1972; Ben- jaminsen et al. 1976; Leatherwood and Reeves 1983), occasionally in the Baltic Sea (Aguayo 1978; Leatherwood and Reeves 1983), and in the North Sea, including waters near Denmark, the Nether- lands, England, and Scotland (Husson and Van Bree 1976; Corbet and Southern 1977; Evans 1980; Leatherwood and Reeves 1983; McBrearty et al. 1986; Baptist 1987; Smeenk 1989; Bakker and Smeenk 1990; Northridge et al. 1995). It is more rarely observed in the Irish sea (Evans 1980; Egan 1984; Jones 1984). There is some evidence to suggest that Lagenor- hynchus albirostris may be extending its range south in the eastern, but not in the western, Atlantic. Sightings south of 55 degrees latitude appear to have become more common in recent years. White- beaked Dolphins were first reported from France in 1981 and stranding records have continued since then (Collet and Duguy 1981; Duguy 1982, 1987). France appears to be the most southerly area fre- quented by this species as it has been reported only once from Portugal and not at all from Spain (Casi- nos and Vericad 1976). Strandings are typically of single individuals. On the coasts of Britain and the Netherlands, strandings may have increased in recent years (Evans 1980). Bakker and Smeek (1990), studying records for the Dutch coast between 1900 and 1988, indicate that there may have been a slight increase in strandings during the 1960s. However, it is not possible to determine whether such fluctuations are actually trends in abundance, changes in distribution, or due to effort. Protection All whaling, except aboriginal, is prohibited in Canadian waters under the Marine Mammal Regulations of the Fisheries Act of 1867 (as amended to date). Some coastal Labrador residents might qual- ify for exemption to take White-beaked Dolphins but hunting there is conducted opportunistically without licences. In United States waters deliberate taking of cetaceans is prohibited by the Marine Mammal Protection Act of 1972. Lagenorhynchus albirostris is listed in Appendix II of CITES, and Appendix II of the Berne Convention (Klinowska 1991); such regu- lations would be applied to the import or export of live captures or tissues. The North and Baltic Seas populations are also included in Appendix II of the Bonn Convention (Klinowska 1991). Population Size and Trends Little is known about the actual abundance of this species. Although it is generally regarded as com- mon, there is little substantive information on popu- lation sizes or trends. There is little study at present that could change this situation in the near future. Western Atlantic. A survey, designed specifically for small cetaceans, was conducted in 1987 covering the 200 meter depth contour between St. Anthony, 120 THE CANADIAN FIELD-NATURALIST Vol 15 FIGURE 2. Range (shaded area) of the White-beaked dolphin in the Northwest Atlantic. Newfoundland, and Nain, Labrador. It resulted in an estimate of 3486 (C.I. 2001-4971) Lagenorhynchus albirostris (Alling and Whitehead 1987). Hay (1982), in earlier surveys primarily focused on large cetaceans in adjacent areas, also made estimates con- sistent with this population size. There are not other numerical estimates. White-beaked Dolphins appear off Labrador as ice recedes in June and occur in greatest abundance in August-September (Boles 1980*). Long-term historical trends in the utilization of this species for food, which could be related to northern coastal abundance, are not evident in Labrador (Brice-Bennet et al. 1977; MacKay and Orr 1987). Jonsgard and Christensen (1968) reported the species as common off southeast Greenland, while Leatherwood et al. (1976) indicated this species was more numerous in the northern portion of its western range. There are no trends evident in sightings collected by the Lighthousekeepers Sighting Network (1979 to 1997) operated by Memorial University of New- foundland (Lynch 1987; Lien, unpublished data). No trends exist in the incidental entrapments of White- beaked Dolphins in fishing gear in Newfoundland (Lien 1988; Lien, unpublished data). Sightings at Mt. _ Desert Rock in Maine are much less common now than they were prior to the late 1970s; they have also declined around Cape Cod during the same period (Katona et al. 1993). Eastern Atlantic. In an aerial survey near Iceland Gunnlaugsson et al. (1988) recorded 942 White- beaked Dolphins which if extrapolated suggest a total population of at least a few thousand in Ice- landic waters (Reeves et al. 1996*). Between 1974 and 1975, a total of 2500 to 3000 White-beaked Dol- phins were sighted in the Barents Sea (Benjaminsen et al. 1976). Oien (1990) made a rough estimate of 132 000 (c.v. 0.266) dolphins, primarily White- beaked, in the Barents Sea based on sightings made during ship surveys. It should be noted that these surveys were designed for large cetaceans and may underestimate small cetaceans which were not the target of survey designs. Although exact numbers are not known this spe- cies is also considered to be numerous in the North Sea with the greatest concentrations in the central and northern sectors (Brown 1975; Corbet and Southern 1977; Massey 1978; Evans 1980; Leatherwood and Reeves 1983; Bakker and Smeenk 1990; Reeves et al. 1996*). It has been reported as common near Sweden, northern Norway, Iceland, and around the Faroe Islands (Jonsgard 1962; 2001 Mercer 1973; Aguayo 1978; Leatherwood and Reeves 1983). It is difficult to estimate a population size given the local nature of most surveys and that they were primarily designed for other species existing data. The total population of White-beaked Dolphins in the north Atlantic has been estimated in the low hun- dreds of thousands (Reeves et al. 1996*). Habitat Generally White-beaked Dolphins are pelagic and found both on and off the coastal shelf. Specific habitat requirements for this species are unknown. Sightings are most common in nearshore waters of Newfoundland and Labrador but this may reflect the relatively greater effort which occurs there. They also occur on the Southeast Shoal (Whitehead and Glass 1985), an offshore area of the Grand Banks with nearshore habitat characteristics. Abundance is known to increase in Newfoundland and offshore with the presence of spawning Capelin, Mallotus vil- losus. The northward shift in abundance of Lagenor- hynchus albirostris that occurs during summer months appears to follow the northward progression of spawning concentrations of Capelin (Carscadden et al. 1989). Capelin form spawning concentrations at fairly uniform temperatures (Shackell et al. 1994), however, sightings of White-beaked Dolphins around Newfoundland occur at all times of year in water temperatures from -1°C to + 15°C, in both win- ter and summer. Sightings of White-beaked Dolphins in December in the Barents sea near Svalbard occurred in association with concentrations of Capelin (J. Lien and I. Christensen, unpublished observations). General Biology / Reproductive Capacity: Foetal lengths from preg- nant females that died in an ice entrapment around Newfoundland in March ranged from 79 to 94 cm in length (Dong et al. 1996). Length at birth is reported to be 1.20 m, while maximum adult length and weight may be up to 3.10 m and over 305 kg respec- tively (Fraser 1974; Corbet and Southern 1977; Leatherwood and Reeves 1983; Perrin and Reilly 1984). Males are reported to be sexually mature at 2.50 to 2.60 m in length and females at 2.50 m (Perrin and Reilly 1984), however a pregnant female measuring 1.70 m in length has been observed (Dong et al. 1996). A sample from Newfoundland showed that male animals reach physical maturity, as indicated by completion of phalanges fusion, at a length of 2.80 m and an estimated age of 13 Growth layer Groups (GLGs), while most females become physically mature at a length of 2.60 m and an esti- mated age of 16 GLGs (Dong et al. 1996). Com- pletion of phalanges fusion in the flippers was at 2.60 m (15 GLGs) in females and at 2.80 m (13 GLGs) in males (Dong et al. 1996). LIEN, NELSON, AND HAI: STATUS OF WHITE-BEAKED DOLPHIN IN CANADA 121 In European waters about half of females reach maturity at 2.42 m (Reeves et al. 1996). The smallest sexually mature male was 2.51 m and the largest sex- ually immature male was 2.57 m (Reeves et al. 1996*). Evans (1980) suggested that the main breed- ing season of White-beaked Dolphins around Britain occurred between July and September based on the fact that all new-born and very young animals found stranded between 1913 and 1966 occurred between these months. Movements: Migrations are poorly understood. White-beaked Dolphins are found in northern seas between spring and late autumn, although they appear to winter further south and some may remain there until the summer months (Leatherwood and Reeves 1983; Klinowska 1991). In the northwestern Atlantic the Lighthousekeepers Sighting Network in Newfoundland report year round sightings of White-beaked Dolphins (Lien, unpub- lished data). Ice entrapments of this species on the northeast coast of Newfoundland occur from February to April (Dong et al. 1996). According to fishermen, this species is first seen off Labrador in late June, while the majority move north mainly in August (Alling and Whitehead 1987). The southern migration past Labrador apparently peaks in September (Alling and Whitehead 1987). Other sightings and strandings indicate that Lagenorhynchus albirostris are generally found along the Labrador coast in the summer months (Sergeant and Fisher 1957; Sergeant et al. 1970; Alling and Whitehead 1987). They have also been sighted on the Southeast Shoal of the Grand Banks in June and July, with numbers decreasing toward the end of this period (Whitehead and Glass 1985). In the northeastern Atlantic strandings of White- beaked Dolphins on the Dutch coast are slightly lower in summer months (Bakker and Smeenk 1990). Sightings of White-sided Dolphins are common in British waters throughout the summer months, although the peak in sightings occurs between October and November for southwest England (Evans 1980). White-beaked Dolphins appear to be largely absent from the coastal waters of Britain between December and May (Evans 1980). Feeding: White-beaked Dolphins are known to feed on Whiting (Merlangius merlangius), Bib (Trisopterus luscus), Poor-cod (Trisopterus minutus), Atlantic Cod (Gadus morhua), Herring (Clupea harengus), Haddock (Melanogrammus aeglefinus), squid, octopus, and sometimes benthic crustaceans (Sergeant and Fisher 1957; Corbet and Southern 1977; Evans 1980; Leatherwood and Reeves 1983; Smeenk 1986; Dong et al. 1996; Reeves et al. 1996*). Jonsgard and Christensen (1968) reported White-beaked Dolphins taking Capelin out of the mouths of feeding Fin Whales (Balaenoptera physalus). Stomach samples from ice entrapped animals on the southwest coast of 122 Newfoundland primarily contained Atlantic Cod, but some had remnants of crabs, Chionoecetes opilio, and seaweed; 20% of the stomachs con- tained fish hooks used on cod longlines (Dong et al. 1996). Ostrom et al. (1993) in a study of feeding in nine cetacean species found nitrogen isotope val- ues, which are good indicators of the trophic level of feeding, that were consistent with a piscivorous diet in White-beaked Dolphins. Social Behaviour: There is little information on social behaviours and social organization in Lagenorhynchus albirostris. White-beaked Dolphins can occur in large schools of several hundred, espe- cially in the northeastern Atlantic (Corbet and South- ern 1977; Evans 1980; Leatherwood and Reeves 1983; Haase 1987). Much smaller groups are typical- ly encountered in the northwest Atlantic with reports of groups numbering 10 to 20 individuals being common (Sergeant and Fisher 1957; Amundin and Amundin 1975; Evans 1980; Anonymous 1983; Baptist 1987; Riddiford 1987; DeBoer 1989; Harvey 1990). Large schools may be associated with feeding or migration (Evans 1980). Some groups are separated by age/maturity. Small groups have been observed to contain both adults and young (Evans 1980; Baptist 1987) although this is not always the case (Jonsgard 1962). Ice en- trapped groups which have been examined in Newfoundland have been primarily composed young animals 1.89 to 2.41 m in length and an estimated 2 to 7 years of age. In such groups most animals were estimated to be 2 years of age; none were pregnant. Other groups were composed of large animals, esti- mated to be older than six years, with several preg- nancies; other animals in this pod were very young animals estimated to me no more than one year of age (Dong et al. 1966). In mass strandings, strong group cohesiveness can be displayed. Animals that are unbeached remain in areas where stranded animals are alive and vocaliz- ing remain (Lien 1996). Instances of bow riding by White-beaked Dol- phins are common (Leatherwood et al. 1976; Jones 1984; Alling and Whitehead 1987; DeBoer 1989). Other behaviours observed in this species include high-speed travelling with shallow leaps, jumping and spinning on a longitudinal axis, high vertical leaps of up to five metres, and “foraging in a line” (Baptist 1987; Haase 1987). White-beaked Dolphins sometimes associate with Common Dolphins (Delphinus delphis) (Haug et al. 1981; Jones 1984), White-sided Dolphins (Haase 1987; DeBoer 1989), Bottlenose Dolphins (Tursiops truncatus) (Evans 1980), and Fin Whales (Jonsgard and Christensen 1968). Humpbacks, Megaptera novaeangliae, have been observed tail slapping as groups of White-beaked Dolphins - approach an area where the humpbacks were coop- THE CANADIAN FIELD-NATURALIST Vol. 115 eratively feeding on capelin (Lien, unpublished observations). White-beaked Dolphins produce several distinct types of sounds. Burst-pulses, presumably used for echolocation, had energy above 325 kHz with pulse rates from 100 to 750 pps and source levels over 200 dB re | UPa at | m (Mitson 1990). “Squeals”, pre- . sumably used for social communication, range from 6.5 to 15 kHz (Schevill and Watkins 1962). These vocalizations are frequently produced by animals during confinement by ice entrapment or while stranded (J. Lien, unpublished observations). Limiting Factors A number of conditions could limit this species, but there is not firm evidence of significant impacts by any given factor. Parasites: Nematodes, Anisakis spp., are the only parasites which have been reported in the stomachs of White-beaked Dolphins (Sergeant and Fisher 1957; Bratty and Clark 1992; Dong et al. 1996). Parasites are common in cetaceans generally and normally are not a serious limiting factor. Infections: Buck and Spotte (1986) describe in detail the microbiology of several individuals of © White-beaked Dolphins captured in Newfoundland during an ice entrapment and removed to a U.S. aquarium. All died from Erysipelothrix rhusiopathi- ae infections. Recent epidemic die-offs of dolphins in some areas due to viral infections, such as the morbillivirus (Duignan et al. 1996), could occur in White-beaked Dolphins. In stranded animals, disease is common in the jaws and teeth (Harmer 1927; Fraser 1946; Van Bree 1977). In some stranded individuals teeth are miss- ing (Reeves et al. 1996); in some instances, tooth loss may occur during process of beaching (Lien 1996*). Such dental deterioration could reflect aging. Pollutants: Organochlorines and heavy metals have been documented in the tissues of white-beaked dolphins from the coast of Newfoundland (Muir et al. 1988) in amounts substantially higher than found in other cetacean species (Muir and Nordstrom 1993). A variety of contaminants, including PCB’s and heptachlor epoxide have been found in animals from the North Sea (Kerkhoff et al. 1981; Law et al. 1991). There are few data available for species com- parisons and “safe” levels of most pollutants are unknown. White-beaked Dolphins in Newfoundland waters exhibited high values of contamination by DDTs and PCBs (4.5 to 89.0 ppm wet weight; and 4.7 to 182.0 ppm wet weight respectively)[Muir et al. 1988] compared to those found by Borrel (1983) in Faroe Island animals (0.57 to 29.3 ppm wet weight and 4.1 to 60.1 ppm wet weight respectively). These values compare with DDT accumulation in Beluga 2001 Whales (Delphinapterus leucas) in the St. Lawrence Estuary of 1.16 to 225.6 ppm wet weight (Martineau et al. 1987) and PCBs from 0.22 to 576.0 ppm wet weight (Muir et al. 1990). Contaminants are believed to be having a detrimental effect on immune hor- monal and reproductive functions in Beluga in the St. Lawrence Estuary population (Beland et al. 1991*: Martineau et al. 1994). There have been actions to prohibit and restrict point source dis- charges of PCBs. DDT use has been prohibited in Canada. This has reduced such introductions into the marine environment, but contamination persists by leachate from dump sites and recirculation of con- taminants present in benthic sediments. Because of bioaccumulation of these compounds in White- beaked Dolphins levels will likely remain high and persist for some time. Monitoring of contamination levels should continue on a regular basis and strand- ed animals should be examined for medical impacts. Incidental Entrapments in Fishing Gear: Inciden- tal entrapments of White-beaked Dolphins in fishing gear are under-reported (Piatt and Nettleship 1987; Lien et al. 1988) and generally depend on obtaining log book or other data from the fishers themselves. Fishers have little motivation to report catches since entrapments of this species typically causes little gear damage (Leatherwood and Reeves 1983; Klinowska 1991; Lien et al. 1985; Dong et al. 1996). In some cases, incidentally caught individuals are used for human or animal food. Survey methods which require cooperation of fishers to accurately report catches may be seriously effected by the fish- ers views’ of such by-catch and their assessment of its management implications (Lien et al. 1994). Because so little is know of the extent of incidental catches, the impact of by-catch on this species is almost completely unknown. As fisheries by-catch is known to effect other small cetaceans, such as the Harbour Porpoise, Phocoena phocoena, (Gaskin 1992), in northwestern Atlantic waters it should be carefully monitored. Ice Entrapments: White-sided Dolphins regularly become caught in ice off the coast of Newfoundland during years of heavy pack ice (Sergeant and Fisher 1957; Muir et al. 1988; Dong et al. 1996). A total of 21 ice entrapments involving approximately 350 ani- mals were reported in Newfoundland from 1979 to 1990; known mortality as a result of entrapment was about 55% (Dong et al. 1993). There have been fewer numbers of ice-entrapment events in more recent years (Lien, unpublished observations). Some animals that escape entrapments quite likely sustain injury and would likely show subsequent scars; some injuries may be serious and debilitating (Dong et al. 1996). Mass Strandings: Mass strandings are less com- mon in Lagenorhynchus albirostris than in Lagen- orhynchus acutus. In Newfoundland there have been LIEN, NELSON, AND HAI: STATUS OF WHITE-BEAKED DOLPHIN IN CANADA 123 only three mass strandings from 1979 through 1996 (Dong et al. 1996; Lien 1996*) involving a total of 21 animals; mortality was 19%. Predation: Reeves et al. (1996*) report Killer Whales (Orcinus orca) swimming with White-beaked Dolphins but are unsure if predation was occurring. Lien (unpublished observation) has observed tail slap- ping by Humpbacks as White-beaked Dolphins approached closely. This was presumed to be threat or aggressive behaviour, not predation; no injury or mor- tality was observed. Hunting: Hunting of white-beaked Dolphins occurred historically in several areas of eastern Canada (Reeves et al. 1996. In more recent times hunting has been conducted opportunistically in sey- eral locations of Newfoundland and Labrador (Brice- Bennett et al. 1977; Mackey and Orr 1987; Sergeant and Fisher 1957; Alling and Whitehead 1987), Greenland (Heide-Jorgensen 1990), Iceland (Reeves et al. 1996*), Norway (Jonsgard 1962), and the Faroe Islands (Bloch and Joensen 1985). Generally hunting which now occurs in Canadian waters is Opportunistic and in remote regions of Labrador where enforcement of regulations is minimal. Numerical estimates of the extent of captures in a local population are given by Alling and Whitehead (1987) for southern Labrador. Here it was estimated, in 1982, that 10% of a population estimated to be 3486 individuals was killed, with an estimated 25 to 50 additional individuals struck but lost. Additional monitoring was done in the summers of 1985 and 1986 by interviewing Labrador fishers from L’ Anse- au Clair to Cartwright (Lien, unpublished data) which found historical levels of catch near those estimated by Alling and Whitehead (1987), but estimated sub- stantially under 100 animals for those years (1985 and 1986). In interviews, fishers frequently did not dis- criminate between Harbour Porpoise and several dol- phin species; this tendency could seriously bias the numbers of White-beaked Dolphins estimated to be taken by hunting. Special Significance of the Species White-beaked Dolphins are a favourite species tar- geted by commercial whale-watching cruises in some areas. The White-beaked Dolphin is the toothed cetacean which is studied and used to introduce the Odontoceti, associated species and the ocean environ- ment in school curricula in Newfoundland and Labrador. “Jumpers”, including White-beaked Dolphins, are a dietary component in coastal peoples of Labrador (Alling and Whitehead 1987; Mckay and Orr 1984). Although the number of animals harvest- ed for food is highly variable, the dietary importance of this hunt is thought to be reduced in recent years due to the availability of store-supplied foods. However, such hunting could impact local popula- tions; additional monitoring is required. 124 Evaluation Although White-beaked Dolphins are considered to be relatively common in the northwest Atlantic there are several factors potentially limiting popula- tions including hunting, incidental entrapments in fishing gear, ice entrapments and possibly pollution. Such factors should be monitored carefully but are not, at present, believed to have substantial impacts on the species although there may be some reduction of local populations. Careful monitoring of hunting activities in Labrador is recommended. Factors such as parasites, infections, mass strandings, and preda- tion are recognized as naturally occurring mortality and are not normally presumed to limit the species. Recent epidemics of disease in some dolphin popula- tions may be exacerbated by anthropogenic condi- tions. Monitoring of fisheries by-catches is important as these removals impact local groups or the total population. There are no major threats identified for this species in the northwestern Atlantic in waters under Canadian jurisdiction. Acknowledgments We would like to thank Kristina Curren for assis- tance and reviewing this manuscript. Dong Jin hai’s interest in the species was great encouragement for us. We would like to thank Robert Campbell for his encouragement; a more patient editor there is not. World Wildlife Fund (Canada) provided financial assistance for D. Nelson for her work on this project. Several students assisted in working on White- beaked Dolphins in our laboratory and helped com- plete this paper; these include Catherine Hood, Christoph Richter, Fernada Marques and Rosie Seton. Thanks are also given to two anonymous reviewers who substantially improved our work. Documents Cited [marked * in text] Beland, P., S. De Guise, and R. Plante. 1991. Toxicol- ogy of St. Lawrence Marine Mammals. Final Report, Wildlife Toxicology Fund, St. Lawrence National Insti- tute of Ecotoxicology, World Wildlife Fund, Toronto, Ontario. 95 pages. Boles, B. K. 1980. Data report listing sightings of ceta- ceans in the Labrador Sea. OLBS, Report by Atlantic Biological Services Ltd., St. John’s, Newfoundland for Eastcan Exploration Ltd., Calgary, Alberta. Lien, J. 1996. Entrapments of cetaceans in fishing gear and strandings reported during 1996 in Newfoundland and Labrador. A preliminary report to the Department of Fisheries, St. John’s, Newfoundland. 22 pages. Reeves, R. R., R. L. Brownell, C. C. Kinze, C. Smeenk, and J. Lien. 1996. White-beaked dolphin (Lagenorhyn- chus albirostris Gray, 1846). Manuscript for Handbook of Marine Mammals. Edited by S. Ridgeway, R. Har- rison. In Preparation. Literature Cited Aguayo, L. A. 1978. Smaller cetaceans in the Baltic Sea. Reports of the International Whaling Commission 28: 131-146. THE CANADIAN FIELD-NATURALIST Vol. 115 Alling, A. K., and H. P. Whitehead. 1987. A preliminary study of the status of White-beaked Dolphins, Lageno- rhynchus albirostris, and other small cetaceans off the coast of Labrador Canada. Canadian Field-Naturalist 101: 131-135. Amundin, B., and M. Amundin. 1975. Sightings of whitenose dolphins Lagenorhynchus albirostris (Gray) in Danish waters. Zeitschrift fiir Sdugetierkunde 40: 58-59. Andrésson, G. 1978. Chasing the elusive dolphin. Atlan- tica and Iceland Review 16: 28-31. Anonymous. 1983. Cetacean records in 1983. Fair Island Bird Observatory Report (36): 58. Bakker, J., and C. Smeenk. 1990. Dolphins on the Dutch coast, an analysis of stranding records. Lutra 33(2): 190-191. Baptist, H. J. M. 1987. Marine mammals in the Dutch part of the North Sea. Lutra 30: 93-104. Benjaminsen, T., J. Berlund, D. Christensen, I. Chris- tensen, I. Huse, and O. Sandnes. 1976. Marking, sightings and behaviour studies of whales in the Barents Sea and at Svalbard in 1974 and 1975. Fisken-og-havet 2: 9-23. Block, D., and J.S. Joensen. 1985. Progress report for Denmark. Reports of the International Whaling Com- mission 35: 165-166. Borrel, A. 1993. PCBs and DDTs in blubber of cetaceans from the northeastern North Atlantic. Marine Pollution Bulletin 26: 146-151. Brattey, J., and K. Clark. 1992. Effect of temperature on egg hatching and survival of larvae of Anisakis simplex B. Canadian Journal of Zoology 70: 274-279. Brice-Bennett, C., A. Cooke, and N. Davis. Editors. 1977. Our foot prints are everywhere. Inuit land use and occupancy in Labrador. Labrador Inuit Association, Nain, Labrador. Brown, S. G. 1975. Relation between stranding mortality and population abundance of smaller cetacea in the Northeast Atlantic Ocean. Journal of the Fisheries Re- search Board of Canada 32: 1095-1100. Buck, J. D., and S. Spotte. 1986. Microbiology of captive white-beaked dolphins (Lagenorhynchus albirostris) with comments on epizootics. Zoo Biology 5: 321-330. Carscadden, J. E., K. T. Frank, and D.S. Miller. 1989. Capelin spawning on the Southeast Shoal: influence of physical factors past and present. Canadian Journal of Fisheries and Aquatic Sciences 46: 1743-1754. Casinos, A., and J. R. Vericad. 1976. The cetaceans of the Spanish coasts: a survey. Mammalia 40: 267-290. Christensen, I. 1972. Minke whale investigations in the Spitsbergen-Barents Sea area (May—June 1972). (1 Norske) Fiskets Gang 48: 961-965. Collet, A., and R. Duguy. 1981. Lagenorhynchus albi- rostris (Cetacea, Odontoceti): new species for the fauna of France. Mammalia 45: 387-388. Corbet, G. B., and H. N. Southern. 1977. The Handbook of British Mammals. Blackwell Scientific Publications, Oxford. 520 pages. DeBoer, R. 1989. A mixed herd of white-sided dolphins Lagenorhynchus acutus and white-beaked dolphins Lag- enorhynchus albirostris in the central North Sea. Lutra 32: 181-184. Dong, J. H., J. Lien, D. Nelson, and K. Curren. 1996. A contribution to the biology of the White-beaked Dolphin, Lagenorhynchus albirostris, in waters off Newfoundland. Canadian Field-Naturalist 110: 278-287. tii pees 2001 Duignan, P. J., C. House, D. K. Odell, L. J. Hansen, M. T. Walsh, D. J. St. Aubin, B. K. Rima, and J. R. Geraci. 1996. Morbillivirus infection in bottlenose dol- phins: evidence for recurrent epizootics in the Western Atlantic and Gulf of Mexico. Marine Mammal Science 12: 499-S]5. Duguy, R. 1982. Annual report on cetaceans and pin- nipeds found off the coasts of France. Annales. Societe des Sciences Naturelles de la Charente-Maritime 6: 969-988. Duguy, R. 1987. Annual report on cetacea and pinnipedia found on the coasts of France. Annales. Societe des Sciences Naturelles de la Charente-Maritime 7: 617—640. Egan, D. 1984. White-beaked dolphin Lagenorhynchus albirostris Gray. Irish Naturalist’s Journal 21: 361. Evans, P.G.H. 1980. Cetaceans in British waters. Mam- mal Review 10: 1-52. Fraser, F.C. 1946. Report on cetacea stranded on the British coasts from 1933 to 1937. (12) British Museum, Lohdon. Fraser, F.C. 1974. Report on cetacea stranded on the British coasts from 1948 to 1966. (14) British Musuem, London. Gaskin, D. E. 1992. Status of the Harbour Porpoise, Pho- coena phocoena, in Canada. Canadian Field-Naturalist 106: 36-54. Gunnilaugsson, T., J. Sigurjonsson, and G. P. Donovan. 1988. Aerial survey of cetaceans in the coastal waters of Iceland, June—July, 1986. Reports of the International Whaling Commission 38: 489-500. Haase, B. J. M. 1987. A mixed herd of white-beaked dol- phins Lagenorhynchus albirostris and white-sided dol- phins Lagenorhynchus acutus in the southern North Sea. Lutra 30: 105-106. Harmer, B. J. M. 1927. Report on cetacea stranded on the British coast from 1913 to 1926. (10) British Museum, London. Harvey, P. 1990. Cetacean records in 1990. Fair Isle Bird Observatory Report 43: 68. Haug, T., B. Gullinksen, and I. Christensen. 1981. Obser- vations of the common dolphin, Delphinus delphis, in North Norway. Fauna 34: 97-100. Hay, K. 1982. Aerial line-transect estimates of abundance of humpback, fin, and long-finned pilot whales in the Newfoundland-Labrador area. Reports of the Interna- tional Whaling Commission 32: 475-486. Heidi-Jorgensen, M-P. 1990. Small cetaceans in Green- land: Hunting and biology. Jn Whaling Communities. Edited by E. Vestergaard North Atlantic Studies 2: 55-58. Husson, A. M., and P. J. H. Van Bree. 1976. On strand- ings of cetaceans on the Dutch coast in 1974 and 1975. Lutra 18: 25-32. Jones, P. H. 1984. Cetaceans seen in the Irish Sea and ap- proaches, late summer 1983. Nature in Wales 3: 62-64. Jonsgard, A. 1962. On the species of dolphins found on the coast of northern Norway and in adjacent waters. Norsk Hvalfangst-Tidende 51: 1-13. Jonsgard, A., and I. Christensen. 1968. A preliminary report on the Hargybuen cruise in 1968. Norsk Hvalfangst-Tidende 57: 174-175. - Katona, S. K., V. Rough, and D. T. Richardson. 1993. A field guide to whales, porpoises, and seals from Cape Cod to Newfoundland. 4th edition. Smithsonian Institu- tion Press, Washington. 316 pages. LIEN, NELSON, AND HAI: STATUS OF WHITE-BEAKED DOLPHIN IN CANADA 125 Kerkhoff, M., J. DeBoer, and J. Geerdes. 1981. Hepta- chlor epoxide in marine mammals. Science of the Total Environment 19: 1-50. Klinowska, M. 1991. Dolphins, Porpoises and Whales of the World. The IUCN Red Data Book. IUCN, Gland, Switzerland and Cambridge, United Kingdom. 429 pages. Lay, R. J., C. F. Fileman, A. D. Hopkins, J. R. Baker, J. Harwood, D. B. Jackson, S. Kennedy, A. R. Martin, and R. J. Morris. 1991. Concentrations of trace metals in the livers of marine mammals (seals, porpoises and dolphins) from waters around the British Isles. Marine Pollution Bulletin (22): 183-191. Leatherwood, S., D. K. Caldwell, and H. E. Winn. 1976. Whales, Dolphins, and Porpoises of the Western North Atlantic, A guide to their identification. NOAA Tech- nical Report NMFS CIRC-396. Leatherwood, S., and R. R. Reeves. 1983. The Sierra Club handbook of whales and dolphins. Sierra Club Books, San Francisco. 302 pages. Lien, J., L. Fawcett, and S. Staniforth. 1985. Wet and Fat: Whales and seals of Newfoundland and Labrador, Breakwater Books, St. John’s, Newfoundland. 136 pages. Lien, J., G. B. Stenson, and I. H. Ni. 1988. A review of incidental entrapment of seabirds, seals and whales in inshore fishing gear in Newfoundland and Labrador: A problem for fishermen and fishing gear designers. Pages 67—71 in Proceedings of a World Symposium Fish Gear Fish Vessel Destruction. Marine Institute, St. John’s, Newfoundland. Lien, J., G. B. Stenson, and J. Chardine. 1994. How many did you catch? The effects of methodology on bycatch reports obtained from fishermen. Pages 535 to 540 in Gillnets and Cetaceans. Edited by W. F. Perrin, G. P. Donovan, and J. Barlow. Reports of the Interna- tional Whaling Commission Special Issue 15. Lynch, K. 1987. Humpback, finback, minke and pilot whale distributions in Newfoundland and Labrador 1976-1983. M.Sc. thesis, Queen Elizabeth II Library, Memorial University of Newfoundland, St. John’s, New- foundland. 195 pages. Mackey, M.G. A., and R. D. Orr. 1987. An evaluation of household country food use in Makkovik, Labrador, July 1980—June 1981. Arctic (40): 60-65. Marineau, D., P. Beland, C. Desjardins, and A. Lagace. 1987. Levels of organochlorine chemicals in tissues of beluga whales (Delphinapterus leucas) from the St. Lawrence Estuary, Quebec, Canada. Archives of Envi- ronmental Contamination and Toxicology 16: 137-147. Martineau, D., S. De Guise, M. Fournier, L. Shugart, C. Girard, A. Lagace, and P. Beland. 1994. Pathology and toxicology of beluga whales from the St. Lawrence Estuary, Quebec, Canada: Past, present and future. Sci- ence of the Total Environment 154: 201-215. Massey, C.I. 1978. Stranded dolphins on the Yorkshire coast. Naturalist 103(946): 108. McBrearty, D. A., M. A. Message, and G. A. King. 1986. Observations on small cetaceans in the north-east Atlantic Ocean and the Mediterranean Sea: 1978-1982. Pages 225-249 in Research on Dolphins. Edited by M. M. Bryden and R. Harrison. Clarendon Press, Oxford. Mercer, M. C. 1973. Observations on distribution and intraspecific variation in pigmentation patterns of odon- tocete Cetacea in the western North Atlantic. Journal of the Fisheries Research Board of Canada 30: 1111-1130. 126 Mitson, R. B. 1990. Very high frequency acoustic emis- sions from the white-beaked dolphin (Lagenorhynchus albirostris). Pages 283—294 in Sensory abilities of ceta- ceans: Labratory and field evidence. Edited by J. A. Thomas, and R. A Kastelein. Plenum Press, New York. Muir, D. C.G., and R. J. Nordstrom. 1993. Marine mam- mals as indicators of environmental contamination by PCBs and Dioxins/Furans. Proceedings of the 17th Annual Aquatic Toxicity Workshop. Canadian Fisheries and Aquatic Sciences Technical Report (1774): 820-826. Muir, D.C.G., C. A. Ford, R.E. A. Stewart, T.G. Smith, R. F. Addison, M.E. Zink, and P. Beland. 1990. Organochlorine contaminants in belugas, Delphi- napterus leucas, from Canadian waters. Canadian Bulletin of Aquatic Sciences 224: 165-190. Muir, D.C. G., R. Wagemann, N. P. Grift, R. J. Nor- strom, M. Simon, and J. Lien. 1988. Organochlorine chemical and heavy metal contaminants in white-beaked dolphins (Lagenorhynchus albirostris), and pilot whales (Globicephala melaena), from the coast of Newfound- land Canada. Archives of Environmental Contamination and Toxicology 17: 613-630. Northridge, S. P., M. L. Tasker, A. Webb, and J. M. Williams. 1995. Distribution and relative abundance of harbour porpoises (Phocoena phocoena L.), white- beaked dolphins (Lagenorhynchus albirostris Grey) and minke whales (Balaenoptera acutorostrata Lacepede) around the British Isles. ICES Journal of Marine Science 52: 55-66. Oien, N. 1990. Sightings surveys in the northeast Atlantic in July 1988: Distribution and abundance of cetaceans. Reports of the International Whaling Commission 40: 499-511. @ritsland, T., N. Oien, J. Calambokidis, I. Christensen, J.C. Cubbage, S. Hartvedt, P. M. Jensen, G. G. Joyce, K. Tellnes, and B. L. Troutman. 1989. Norwe- gian whale sightings surveys in the North Atlantic, 1987. Reports of the International Whaling Commission 39: 411-415. Ostrom, P. H., J. Lien, and S. A., Macko. 1993. Evalua- tion of the diet of Sowerby’s beaked whale, Mesoplodon bidens, based on isotopic comparisons among northwest- ern Atlantic cetaceans. Canadian Journal of Zoology 71: 858-861. THE CANADIAN FIELD-NATURALIST Voloiels Perrin, W. F., and S. B. Reilly. 1984. Reproductive para- meters of dolphins and small whales of the family Del- phinidae. Reports of the International Whaling Com- mission. Special Issue. (6): 97-133. Piatt, J. F., and D. N. Nettleship. 1987. Taeidental catch of marine birds and mammals in fishing nets off New- foundland, Canada. Marine Pollution Bulletin 18(6B): 349-354. Riddiford, N. 1987. Cetacean records in 1986. Fair Island Bird Observatory Report 39: 65. Schevill, W. E., and W. A. Watkins. 1962. Whale and porpoise voices. A phonograph record. Woods Hole Oceanographic Institution, Woods Hole, Massachusetts. Sergeant, D. E., and H. D. Fisher. 1957. The smaller cetacea of eastern Canadian waters. Journal of the Fish- eries Research Board of Canada 14: 83-115. Sergeant, D. E., A. W. Mansfield, and B. Beck. 1970. Inshore records of Cetacea for Eastern Canada, 1949-68. Journal of the Fisheries Research Board of Canada 27: 1903-1915. Shackell, N. L., J. E. Carscadden, and D.S. Miller. 1994. Migration of pre-spawning capelin Mallotus villo- sus as related to temperature on Northern Grand Bank, Newfoundland. ICES Journal of Marine Science 51: 107-114. Smeenk, C. 1986. Cetaceans stranded on the Dutch coast in 1981, 1982, and 1983. Lutra 29: 268-281. Smeenk, C. 1989. Cetaceans stranded on the Netherlands coast in 1984, 1985, and 1986. Lutra 32: 164-180. Van Bree, P. J. H. 1977. On former and recent strandings of cetaceans on the coast of the Netherlands. Zeitschrift fiir Sdugetierkunde 42: 101-107. Winn, H. E., and R. K. Edel. 1982. A characterization of marine mammals and turtles in the mid- and north- Atlantic areas of the U.S. outer continental shelf. CETAP Annual Report for 1980. Cetacean and Turtle Assessment Program, University of Rhode Island, Con- tract AA551-CT8-48. Whitehead, H., and C. Glass. 1985. The significance of the Southeast Shoal of the Grand Bank to humpback whales and other cetacean species. Canadian Journal of Zoology 63: 2617-2625. Accepted 29 May 2000 Updated Status of the Vancouver Island Lake Lamprey, Lampetra macrostoma, 1n Canada‘ R. J. BEAMISH Department of Fisheries and Oceans, Pacific Biological Station, 3190 Hammond Bay Road, Nanaimo, British Columbia VOR 5K6, Canada. Beamish, R. J. 2001. Updated status of the Vancouver Island Lake Lamprey, Lampetra macrostoma, in Canada. Canadian Field-Naturalist 115(1): 127-130. The Vancouver Lamprey was originally described in 1982 from a restricted area on Vancouver Island. It is a derivative of Lampetra tridentata (Pacific Lamprey), but has a larger disc and is able to survive and feed in freshwater. It feeds heavily on freshwater salmonids and juvenile anadromous Coho Salmon, Oncorhynchus kisutch, and would be a predator of other salmonids if it is introduced into other systems. Key Words: Vancouver Island Lake Lamprey, Lamproie a grand disque, Lampetra macrostoma, parasitic lampreys, British Columbia, rare and endangered species, Petromyzontiformes. The Vancouver Island Lake Lamprey, Lampetra macrostoma, (Figure 1) is endemic to Vancouver Island and is the only known Lampetra tridentata, Pacific Lamprey, derivative in Canada. It is also the only freshwater parasitic lamprey known from British Columbia. Originally described by Beamish (1982), the species was designated by the Committee on the Status of Endangered Wildlife in Canada (COSEWIC) as Vulnerable in 1986 (Beamish 1987). Distribution The Vancouver Island Lake Lamprey has been found only in two adjacent lakes, Cowichan Lake and Mesachie Lake, on Vancouver Island, British Columbia (Figure 3). Recently, two other popula- Aions of freshwater, parasitic Lampetra tridentata like lamprey have been reported from Village Bay Lake, Quadra Island and West Lake, Nelson Island, British Columbia (Figure 3), but their relationship with Lampetra macrostoma has not been studied. The larger lake is the 34 km long Cowichan Lake, which has a mean depth of 51 m and a maximum depth of 150 m. The much smaller lake, Mesachie Lake is 2 km to 7 km long and has a surface area of approximately 59 hectares and a generally uniform depth of 33 m. Protection There is no proposed protection or control for this species. If the distribution is confined to the two known lakes, then the species is rare. As local fish- ermen do not “like” this species, it may become endangered unless its importance as an isolated ‘Vulnerable status originally assigned by COSEWIC April 1986, reaffirmed April 1998. species is appreciated. It is also a potentially serious source of salmonid mortality, if introduced to other lakes. The Conservation Data Centre ranks the species as Gl, S1, ie., both globally and provincially criti- cally imperiled and the Province (British Columbia) as “Red”, i.e., threatened. Population Size and Trend No population estimates exist. The incidence of wounding and scarring on prey suggests that the population fluctuates. There is no indication that the population is in danger of extinction, but research on this species should be controlled. There has been no research since the mid-1980s. Further study is needed to identify biological and taxonomic rela- tionships with its close relative the Pacific Lamprey and the two other reported Lampetra tridentata like specimens. There has been a recent decline in wild Coho, Oncorhynchus kisutch, (DFO 1998) and this may affect the survival or prey preference. Habitat The known distribution is extremely restricted. It is probable that its distribution results from anoma- lous patterns of glaciation during the last glacial period. There is no indication that habitat loss is reducing distribution, in fact, increasing siltation of lakes and rivers may be increasing habitat for ammocoetes. The species is a lake spawner and would be affected by loss of shallow water gravel areas used for spawning. General Biology Like all lake lampreys, the Vancouver Island Lake Lamprey reproduces only once. It is not possi- ble to age an lamprey accurately, however, an “edu- 127 128 THE CANADIAN FIELD-NATURALIST Vol iis FiGuRE 1. Head and tail regions of a 22.8 cm Lampetra macrostoma photographed live and captured in Lake Cowichan, November 1980. cated guess” is that the species is about 8 years old when it reproduces (6 years as a larva and 2 years as a young adult and adult). Lampreys appear to be able to increase in abun- dance relatively rapidly, indicating a relatively high rate of juvenile survivai at low population levels. Also, lampreys are known to be at least 300 million years old, hence they are one of our most primitive and successful fishes. This indicates that this species and lamprey in general may be adaptable and capa- ble of adjusting to a variety of pressures if allowed to behave naturally. However, this species is not found outside of the two lakes described and it is captured and killed by sport fishermen. Thus, the biology of the Vancouver Island Lake Lamprey has not been well studied. It is known that spawning occurs from May to August, and during this period, there is an aggregation of males and females on the gravel based deltas. Spawning has been observed at the mouth of several creeks. Spawning areas varied from 20 cm to more than 2 m deep. Spawning occurred in the shallow water, but we could not determine if it occurred in water deep- er than 15 cm. Larval lamprey are found in the lake in the vicinity of creeks and occasionally in the creeks, but always close to the lake. No Vancouver Island Lake Lamprey have been found more than a few hundred meters upstream of the lake in the few creeks that have been examined, and there were no lamprey in the creek connecting Mesachie Lake and Cowichan lake. This suggests that this species spawns in each lake and young remain in the lake. Spawning behaviour as observed in the laboratory is similar to the behaviour reported for Lampetra richardsoni and Lampetra tridentata by Pletcher (1963). It is only during the spawning period that adult lamprey are easily captured. Behaviour of young adults at other times is largely unknown and _unstudied. Feeding juveniles have been caught by anglers when they have towed live fish on “stringers” behind their boat. While it is rare to land live lamprey in this manner, one fisherman landed three live Vancouver Island Lake Lamprey that had attached themselves to a live Cutthroat Trout (Oncorhynchus clarki). . The Vancouver Island Lake Lamprey metamor- phoses into a young adult from July to October. From October to spring of the following year the young adult probably remains in the gravel. In spring, the young adult begins feeding and attacks large numbers of young salmonids, including age 1 and 2 Coho Salmon. Feeding juveniles readily attack resident fishes. Carl (1953) reported that 8 out of 10 fish examined from Cowichan Lake had evidence of lamprey attacks. Beamish (1982) observed that up to 50% of the salmonids captured in Mesachie Lake throughout the years had evidence of lamprey feeds without killing its hosts. Since salmonids remain in the lake and the lake is still popular for sport fishermen, it appears that the Vancouver Island Lake Lamprey coexists with its hosts in a manner similar to the landlocked Sea Lamprey, Petromyzon marinus in Cayuga Lake, New York, USA (Wigley 1959). Beamish (1982) reported that in one sample of 221 salmonids, 15% had wounds that penetrated deeply into the body FIGURE 2. Coho showing fresh scar from a Lake Lamprey. The salmon was 19 cm in length and found dead on the bottom of Mesachie Lake. Notice the size of the scar relative to the size of the fish and the healed wound posterior to the fresh wound. 2001 BEAMISH: UPDATED STATUS OF VANCOUVER ISLAND LAKE LAMPREY 129 So". NELSON CY §: nee “ae Ne ta Mesachie —__~ patch) a : Lake FiGure 3. Distribution of the Vancouver Island Lake Lamprey, Lampetra macrostoma. (Figure 3) and could cause mortality, thus some mortality does occur. The species feeds primarily on Coho Salmon and Cutthroat Trout. However, other salmonids such as Dolly Varden (Salvelinus malma) are attacked. No Pacific Lamprey adults of ammocoetes have been found in the same area as the Vancouver Island Lake Lamprey. Pacific Lamprey do exist a few kilo- meters from the lake in the river flowing out of Cowichan Lake, but it is unknown if they occur in the lake. The timing and length of spawning period of the Vancouver Lamprey is different from the Pacific Lamprey that occur in the same area of British Columbia (Beamish 1980). The lake spawn- ing habitat and the late spawning period could offer reproductive isolation from the Pacific Lamprey. It is important to note that this lamprey readily sur- vives and feeds as a young adult in freshwater, whereas its close relative, the Pacific lamprey, can- not be kept as a young adult in freshwater (Beamish 1980; and unpublished data). Limiting Factors Unknown. Special Significance of the Species The Vancouver Island Lake Lamprey is endemic to Canada and is only known to exist on Vancouver Island. It is of considerable scientific interest, but it is unlikely that some sectors of the general public would understand or support its preservation and protection. Its occurrence in Cowichan Lake is of particular interest to the understanding of glacial history of Vancouver Island. Confirming the closest 130 relatives and the molecular genetic dating of the time it split off from a common ancestor would con- tribute to the study of glacial refugia on Vancouver Island. It is probable that this species will also pro- vide important information about evolutionary pro- cesses in lampreys. Evaluation Comments from fishermen about the incidence of observed lamprey wounds suggest that the popula- tion fluctuates. Although no population estimates exist there has been no indication that there is a long term decline in numbers. There has not been an assessment of this species since the mid-1980s, but its restricted distribution indicates that it will proba- bly remain confined to Cowichan Lake and Mesachie Lake. The abundance could be guessed to be between one or two thousand adults in both lakes. It is also possible that the recent decline in coho abundance has affected the survival or feeding habits, or both, but there have been no studies. Acknowledgments I thank COSEWIC for the opportunity to present this manuscript. THE CANADIAN FIELD-NATURALIST Vol. 115 Literature Cited Beamish, R. J. 1980. The adult biology of the river lam- prey (Lampetra ayresi) and the Pacific lamprey (Lam- petra tridentata) from the Pacific coast of Canada. Cana- dian Journal of Fisheries and Aquatic Sciences 37: 1906-1923. Beamish, R. J. 1982. Lampetra macrostoma, a new spe- cies of freshwater parasitic lamprey from the west coast of Canada. Canadian Journal Fisheries and Aquatic Sci- ences 39: 736-747. Beamish, R. J. 1987. Status of the Lake Lamprey, Lampetra macrostoma, in Canada. Canadian Field- Naturalist 101: 186-189. Carl, G.C. 1953. Limnobiology of Cowichan Lake, Brit- ish Columbia. Journal of the Fisheries Research Board of Canada 9: 417-449. Department of fisheries and Oceans (DFO). 1998. Coho salmon final report. Coho Response Team, Fisheries and Oceans Canada, Pacific Region, Vancouver, British Columbia. 508 pages. Pletcher, F. I. 1963. The life history and distribution of lampreys in the Salmon and certain other rivers in British Columbia, Canada. M.Sc. thesis. University of British Columbia, Vancouver, British Columbia. 195 pages. Wigley, R. L. 1959. Life history and distribution of lam- preys of Cayuga Lake, New York. U.S. Fish and Wildlife Service Fisheries Bulletin 154: 561-617. Accepted 29 May 2000 Status of the Sympatric Smelt (Genus Osmerus) Populations of Lake Utopia, New Brunswick’ Eric B. TAYLOR Department of Zoology and Native Fish Research Group, University of British Columbia, Vancouver, British Columbia V6T 1Z4, Canada Taylor, Eric B. 2001. Status of the sympatric smelt (genus Osmerus) populations of Lake Utopia, New Brunswick. Canadian Field-Naturalist 115(1): 131-137. Lake Utopia, southwestern New Brunswick, contains phenotypically and genetically distinct populations of smelt, genus Osmerus. One form, “Dwarf Smelt”, matures at between 100 to 120 mm total length, has 35 to 36 gill rakers, and spawns in small streams in May. The other form, “ Normal Smelt”, matures at 150 to 250 mm total length, has 30 to 32 gill rakers and spawns in larger, lake-headed streams in early April. The two forms are genetically distinct as assessed by differences in the frequencies of mitochondrial and nuclear DNA restriction fragment length polymorphisms. Quantitative genetic dif- ferences between the forms are suggested by inter-annually stable differences in morphology and spawning time and site. The two forms in Lake Utopia: (i) are reproductively isolated in sympatry and behave as distinct species, (ii) appear to have originated independently from sympatric forms in other lakes, and (111) may have resulted from speciation within Lake Utopia. Both forms are subject to annual dip-net recreational fisheries. The Normal Smelt appears to be in no immediate danger, but Dwarf Smelt may be especially vulnerable owing to its apparently lower abundance and limited spawning dis- tribution in small, easily disturbed streams. Il y a deux populations d'Eperlan (genre Osmerus) dans le Lac Utopia dans le sud-ouest du Nouveau-Brunswick, qui peu- vent étre distinguées par leur génotype et leur phénotype. Le type “normal” atteint une longueur de 150 a 250 mm a matu- rité, posséde 30 a 32 branchicténies et fraie au début avril, dans les ruisseaux qui aboutissent dans le lac. Le type nain atteint une longueur de 100 4 120 mm a maturité, posséde 35 a 36 branchicténies et fraie au mois de mai dans les petits cours d'eau. Les deux types se distinguent génétiquement par des différences dans la fréquence des fragments de restric- tion de l'ADN des mitochondries. La stabilité morphologique de chaque phénotype quant l'endroit et la période de frai suggérent qu'il y aurait des différences génétiques quantitatives. Les deux types du Lac Utopia (i) sont isolés au point de vue reproductif en sympatrie et fonctionnent tel deux espéces, (11) semblent avoir une origine indépendante des types en sympatrie dans les autres lacs, et (111) pouraient étre le résultat de spéciation dans ce lac. Les deux types sont péchés au moyen d’épuisette chaque année, par des pécheurs de loisir. Le type “normal” ne semble pas en danger immédiat, mais le type “nain” pourrait l'étre car il parait étre moins abondant et ne fraie que dans quelques cours d'eaux qui sont facilement perturbés. Key Words: Rainbow Smelt, Osmerus, speciation, Osmeridae, New Brunswick, parallel evolution. The Rainbow Smelt, Osmerus mordax (Mitchill piscivorous (see Taylor and Bentzen 1993a). There A814), is a euryhaline osmerid fish native to water- sheds tributary to the western North Atlantic Ocean from Long Island to Lake Melville on the Labrador coast (Scott and Scott 1988; Nellbring 1989). Throughout this native range, Rainbow Smelt may be anadromous, growing to maturity in nearshore marine environments before returning to freshwater streams to spawn in the spring, or they may reside permanent- ly in freshwater. Like many north temperate freshwa- ter fish, lake-dwelling Rainbow Smelt may exist as “dwarf-” or “normal-sized” individuals, maturing at 70 mm and 250 mm total length, respectively (Taylor and Bentzen 1993a). Dwarf smelt are usually morpho- logically specialized for plankton feeding while nor- mal smelt appear to be macrophagous and are often ‘The dwarf form was assigned a Threatened status by COSEWIC, April 1998. are at least five well-documented cases of lake popu- lations consisting of sympatric dwarf- and normal- sized forms: Lac Heney, Québec; Green and Onawa lakes, Maine; Lochaber Lake, Nova Scotia; and Lake Utopia, New Brunswick (Lanteigne and McAllister 1983; Taylor and Bentzen 1993a). Sympatric forms differ in size-at-maturity, morphol- ogy, and feeding and reproductive ecology (Delisle 1969; Langteigne and McAllister 1983; Taylor and Bentzen 1993a). Lanteigne and McAllister (1983) considered the dwarf form of Osmerus in northeastern North Amer- ica to be a distinct species, the Pygmy Smelt (Osmer- us spectrum) first described by Cope (1870) from specimens in Maine. Taylor and Bentzen (1993a), however, examined relationships among 19 anadro- mous, and dwarf- and normal-sized freshwater popu- lations by assaying mitochondrial DNA (mtDNA) restriction site variation. The mtDNA data refuted the idea of monophyly of dwarf smelt and cast doubt on lfaM! 132 the validity of the designation Osmerus spectrum. There were no restriction sites that were diagnostic for dwarf or normal smelt, and dwarf-sized smelt populations did not cluster together and separately from normal-sized or anadromous populations. For instance, dwarf-sized smelt from Lake Utopia and Green Lake (previously diagnosed as Osmerus spec- trum) as well as dwarf smelt from Lochaber and Onawa lakes were not monophyletic and were more similar to geographically proximate normal-sized or anadromous populations than to each other (Taylor and Benzten 1993a). Consequently, Taylor and Bentzen (1993a) concluded that dwarf Osmerus were polyphyletic and had diverged repeatedly, perhaps sympatrically, from geographically proximate normal or anadromous populations following deglaciation and that the various populations perhaps comprise a complex of undescribed biological species. Although casting doubt on the validity of Osmerus spectrum as envisioned by Langteigne and McAllister (1983), the data of Taylor and Bentzen (1993a,b) established that in at least two lakes where they occurred sympatrically (Lake Utopia, New Brunswick and Lochaber Lake, Nova Scotia), dwarf and normal Osmerus were genetically distinct and, hence, repro- ductively isolated. Reproductive isolation in sympa- try, coupled with interannually stable morphological and ecological differences between the forms (Figure 1) led Taylor and Bentzen (1993a,b) to conclude that dwarf and normal smelt in Lake Utopia fulfilled the criteria of distinct biological species (sensu Mayr THE CANADIAN FIELD-NATURALIST Vol. iis 1963). Both forms in Lake Utopia, however, fall well within the morphological and molecular “boundaries” of Osmerus mordax relative to European Smelt (Osmerus eperlanus) and Arctic Rainbow Smelt (Osmerus dentex) (cf. Taylor and Dodson 1994) which complicates the taxonomic recognition of either or both of dwarf and normal smelt as distinct from Osmerus mordax. On average, however, the mtDNA data indicated that dwarf Osmerus from Lake Utopia were slightly more divergent from Osmerus mordax found outside Lake Utopia than were normal smelt (Taylor and Bentzen 1993a). Further, the divergence that has resulted in genetically distinct populations of dwarf and normal smelt in Lochaber Lake, Nova Scotia has occurred independently from that in Lake Utopia (Taylor and Bentzen 1993a). In all other lake systems where they co-exist, therefore, dwarf- and normal-sized smelt probably have arisen independent- ly and the forms do not represent distinct lineages in northeastern North America (Taylor and Bentzen 1993a). Consequently, the status of dwarf and normal smelt from each lake where they are sympatric needs to be evaluated individually. As a first step in this pro- cess, this report summarizes what is known of the biology of the Lake Utopia smelt populations, outlines their scientific importance, and discusses issues rele- vant to their conservation. Distribution Lake Utopia (45° 10' N, 66° 47' W) is part of the Magaguadavic River drainage, Charlotte Co., south- 1cm (= Ficure 1. Illustration of Lake Utopia Normal (upper) and Dwarf (lower) Smelt (Osmerus). Both specimens were mature males. i i eat 2001 western New Brunswick, and is located about 100 km SW of Saint John (Figure 2). Dwarf smelt from Lake Utopia were introduced into Meech Lake, Québec in 1924 (Dymond 1939) where they appear to have established a self-sustaining population because mature individuals have been collected as recently as 1991 (Delisle and Veilleux 1969; E. B. Taylor and N. Alfonso, unpublished data). Genetically distinct dwarf and normal smelt also occur in Lochaber Lake, Nova Scotia (Guysborough County) and these forms appear to have arisen independently from those in Lake Utopia (Taylor and Bentzen 1993a). Putative sym- patric populations of dwarf and normal smelt (i.e., reproductive isolation between forms has not been confirmed by genetic means) also occur in Lac Heney, Gatineau River drainage, Gatineau Co., Québec (Delisle 1969); Green and Onawa lakes, Maine; Lake Champlain (Québec, Vermont, and New York); and Lake Kénogami, Québec (Delisle and Veilleux 1969). Current data indicates that sympatric smelt populations in northeastern North America have had independent origins (Taylor and Bentzen 1993a); the status of these other putative sympatric popula- tions, therefore, will need to be evaluated indepen- dently from the Lake Utopia populations. Protection Exploitation of the smelt populations of Lake Utopia is limited to local dip-net fisheries during the spring spawning periods (April and May). Most of Trout Lake ¥ Stream Atlantic Area of Ocean Enlargement FiGuRE 2. Lake Utopia and surrounding watershed includ- ing Dwarf (“A”) and Normal Smelt (“B”) spawning streams. TAYLOR: STATUS OF SYMPATRIC SMELT IN LAKE UTopIA, NEW BRUNSWICK 133 the effort appears to be focused on the Normal Smelt populations in Mill Lake Stream and Trout Lake Stream (Figure 2) owing to the ease of public access and larger spawning runs (E. B. Taylor, personal observations). Legal methods of fishing include gill nets, bag nets, box nets, dip-nets, and angling, but a closed season for the first three methods extends from | April through 14 October. Gill, box, and bag nets require a licence from the federal Department of Fisheries and Oceans, but none of these permits have been issued since the early 1960s (P. Cronin, New Brunswick Department of Natural Resources and Energy, R.R. 6, Fredericton, New Brunswick, person- al communication). The dip-net fishery in Lake Utopia and its tributaries is open from 15 April to 31 May, with a daily catch and possession limit of 60 fish. There are no size limits in effect. Possession is permitted for domestic consumption only and ice fishing is not permitted on Lake Utopia (Anonymous 1996). A private Atlantic Salmon (Salmo salar) smolt farm is operated at the northeast end of the lake. Although the farm is land-based and freshwater- resident salmon are native to Lake Utopia (T. Vickers, N.B. Department of Natural Resources and Energy, personal communication), any accidental release of smolts into Lake Utopia could increase predation mortality on dwarf smelt and juveniles of both forms. A pulp and paper mill is also located near the southeast part of the lake, but its effluent does not enter Lake Utopia (Vickers, personal communica- tion). Some protection of spawning streams is afford- ed indirectly by a game refuge at the northeast end of the lake that includes Mill Lake and its outlet stream. Population Size and Trends No formal population census has been conducted in Lake Utopia for either dwarf or normal smelt. Personal observations by the author, however, over several nights during the spawning runs were made in 1991. Normal smelt in Mill Lake and Trout Lake streams appeared to number in the order of several thousand individuals. By contrast, numbers of dwarf smelt spawning in two streams at the northwest end of the lake appeared to be much less numerous; less than two hundred individuals were observed spawn- ing over two nights. These observations are qualita- tive at best, but left the impression that in terms of maintaining viable smelt populations in Lake Utopia, dwarf smelt appear to be the form at greater risk. The exploitation of normal smelt in Lake Utopia in the annual spring dip net fishery is considered to be “low to moderate” (Vickers, personal communica- tion). The extent of the dip-net fishery on dwarf smelt, however, is less well known owing, in part, to the relative remoteness of their spawning streams. Habitat Lake Utopia is a relatively shallow lake with aver- age and maximum depths of 11.1 and 25.6 m, 134 respectively. The lake is 7.2 km long and covers an area of approximately 1370 hectares. The morpho- edaphic index is 0.94 and pH typically ranges from 7.0 at the surface to 6.4 at 25 m. The mid-summer (July) thermocline was at 10-15 m (27 August 1969) and temperatures ranging from 19°C (surface) to 7.8°€ (25 my) were recorded on) 3 July, 9699 The lake is frozen from early December until the first or second week in April (Vickers, personal communi- cation). Spawning habitat of dwarf and normal smelt from Lake Utopia appears to be largely confined to four small streams at head of the northern half of the lake (Figure 2). Normal smelt spawn in Mill Lake and Trout Lake streams, both of which are lake outlet streams. During spawning, these streams had high to moderate flows (up to 1 m/s) and are 2—5 m in width. Dwarf smelt spawned in two much smaller (1-2 m wide), slower-flowing streams (< 10 cm/s), both of which are not lake-headed. All of the land surround- ing Lake Utopia or its tributary streams is privately- owned, and there is considerable cottage develop- ment, particularly in the south half of the lake. General biology Reproduction Spawning normal smelt were collected in both Mill Lake Stream and Trout Lake Stream on 7 April 1980 at a water temperature of 4°C while Dwarf Smelt were collected from “Mill Brook” on 12 May 1980 (Lanteigne and McAllister 1983). Bridges and Delisle (1974) reported dwarf smelt spawning in Lake Utopia on 22 May 1972. Collections of dwarf and normal smelt made in 1990 and 1991 also indicated a large difference in spawning time between the forms; nor- mal smelt were collected and observed to be ripe on 8-13 April 1991, whereas spawning dwarf smelt were collected on 7 May, 1990 and 10-12 May, 1991 (Taylor and Bentzen 1993b). Dwarf and normal smelt were both observed to enter the tributary streams at dusk with the greatest number ascending the streams and spawning between 21:00 and 04:00 (E. B. Taylor, personal observations). Interestingly, the smelt from Meech Lake that originated from a trans- plant of dwarf smelt from Lake Utopia were observed to spawn at the same time as the donor population: 8 May 1963 and 19 May 1971 (Bridges and Delisle 1974; Lanteigne and McAllister 1983). Feeding Ecology Dwarf smelt of Lake Utopia appear to be largely zooplanktivorous, Bajkov (1936) reported stomach contents to consist of Diaptomous, Cyclops, Lepto- dora, Daphnia, Epischura, and Bosmina. By con- trast, normal smelt (at least once they surpass the average size of dwarf smelt) are macrophagous, often consuming juvenile smelt (presumably of both forms) (Bajkov 1936; E. B. Taylor, personal obser- _vation). Bajkov (1936) also reported that normal THE CANADIAN FIELD-NATURALIST Vol. 115 smelt may be taken by angling with artificial flies at the surface. Size, Age, Growth, and Morphology Dwarf and normal smelt collected from Lake Utopia in 1981 were reported to average (+SE) 96.9 (1.5) mm and 177.9 (4.1) mm in standard length, respectively (Lanteigne and McAllister 1983). Dwarf smelt collected in 1990 and 1991 averaged 142.9 (2.34) and 112.9 (1.84) mm total length, respectively (see Taylor and Bentzen 1993a). Normal smelt col- lected in 1991 averaged 226.0 (2.89) mm total length (Taylor and Benzten 1993a). There are no data con- cerning age or growth of the forms in Lake Utopia. Morphological characterization of Lake Utopia dwarf and normal smelt was summarized by Taylor and Bentzen (1993a). In general, dwarf smelt have more gill rakers, larger eyes, but smaller mouths than normal smelt. The greatest differences are apparent in gill raker counts. Lanteigne and McAllister (1983) reported mean total gill raker counts of 34.1 (0.11) and 31.1 (0.30) in dwarf and normal smelt, respective- ly. Similar differences between the forms were report- ed by Taylor and Bentzen (1993a,b); gill raker counts for dwarf smelt collected in 1990 and 1991 averaged 36.2 (0.21) and 35.4 (0.20), respectively, and 32.8 . (0.15) in normal smelt collected in 1991. Interannual stability of differences in gill raker counts between the forms is consistent with a genetic basis to trophic morphological differentiation between dwarf and nor- mal smelt. Further, high gill raker counts of dwarf smelt are retained when fish are transplanted to differ- ent environments; total gill raker counts of Lake Utopia dwarf smelt collected in 1980 were not signifi- cantly different from those characterizing Meech Lake dwarf-sized fish (derived from Lake Utopia dwarf smelt) collected in 1963 (NMC71-0651, Copeman and McAllister 1978; Lanteigne and McAllister 1933): Limiting factors Production of Lake Utopia smelt is presumably regulated to some extent by the level of primary pro- ductivity of the lake. In addition, Lake Utopia con- tains populations of Atlantic Salmon and Brook Trout (Salvelinus fontinalis), both of which have been recorded to prey on smelt in Lake Utopia and else- where (Nellbring 1989; Sayers et al. 1989). Of five Brook Trout sampled from Lake Utopia during 1996, three were examined for stomach contents and all contained smelt. The Brook Trout ranged in size (TL) from 38.2 to 42.6 cm and the smelt ranged in size from 18.0 to 21.9 cm (Vickers, personal communica- tion). The production of dwarf and normal smelt in Lake Utopia is potentially limited by predation pres- sure from salmonids, particularly on smaller sized juveniles and dwarf smelt. Furthermore, large smelt are often piscivorous and may include smaller smelt in their diet (E. B. Taylor, personal observations). It oo 2001 is likely that a balance exists between smelt and their predators which may be a factor limiting smelt pro- duction in Lake Utopia. It is also possible that salmonid enhancement efforts in the lake could upset this balance and have negative impacts on production of juvenile stages of both forms and adult dwarf smelt. At present, enhancement of Lake Utopia salmonids is very limited; every other year Atlantic Salmon are planted into the lake at | fish for every 2 ha of lake surface area (Vickers, personal communi- cation). Perhaps the most important factor limiting smelt populations in Lake Utopia is the size and number of spawning streams. Although precise data are not available, the majority of spawning by normal smelt ‘occurs in two streams at the northeast end of the lake (Figure 2) while dwarf smelt appear largely restrict- ed to two much smaller streams at the northwest end of the lake. Although Trout Lake and Mill Lake streams are lake-headed and appear relatively stable in terms of long term persistence, the dwarf smelt spawning streams appeared to be more ephemeral and susceptable to habitat degradation owing to their small size. Special Significance of the Sympatric Smelt The significance of the Lake Utopia smelt popula- tions comprises at least two areas concerned with the recognition of biological diversity. First, Lake Utopia smelt are part of a general phenomenon char- acteristic of many north temperate freshwater fish faunas where genetically distinct and reproductively isolated populations are sympatric and differentiated in morphological and ecological traits (e.g., Svardson 1961; McPhail 1984, 1992; Hindar et al. 1986; Foote et al. 1989; Verspoor and Cole 1989; Bernatchez and Dodson 1990; Ferguson and Taggart i eeaylor et al. 1996; Taylor et al. 1997). Because the forms are reproductively isolated and ecologically distinct in sympatry, Taylor and Bentzen (1993a,b) argued that dwarf and normal smelt in Lake Utopia fulfilled the criteria for recog- nition as biological species (sensu Mayr 1963). The understanding of processes involved in speci- ation, the splitting of a single lineage into two or more reproductively isolated lineages, is a central theme of evolutionary biology. Such understanding is difficult to obtain directly because while one can observe the products of speciation (the species them- selves), the processes involved are historical and, thus, impossible to observe. Notwithstanding this limitation, the principal model of speciation envi- sions genetic divergence during long term geograph- ic partitioning of a lineage; i.e., allopatry. By con- trast, the Lake Utopia smelt populations provide strong evidence of reproductive isolation developing postglacially (e.g. < 12 000 years — see Hughes et al. 1985) and in the absence of obvious geographic TAYLOR: STATUS OF SYMPATRIC SMELT IN LAKE UTopiA, NEW BRUNSWICK 135 separation (Taylor and Bentzen 1993a,b). The data and arguments in favour of such relatively rapid, sympatric divergence in Lake Utopia have contribut- ed to an increased appreciation of sympatric specia- tion as a legitimate and perhaps common mechanism of speciation (Bush 1994). The argument that sym- patric speciation, while possible, is a “minor” or “rare” mechanism relative to allopatric speciation (see Gibbons 1996, page 1499), simply heightens the importance of the Lake Utopia populations where sympatric divergence has been postulated to occur. Some perspective of the scientific importance of the Lake Utopia smelt can be gained by appreciating that similar evidence for sympatric divergence put forth for Lake Utopia smelt has also been suggested to explain the explosive speciation of cichlid “species flocks” in Great Rift and crater lakes of Africa (Meyer et al. 1990; Schliewen et al. 1994). Further, because the divergence of Lake Utopia smelt popula- tions appears to have occurred independently of divergence between smelt populations in other lakes (see Taylor and Bentzen 1993a) they provide an example of parallel evolution and the possible importance of deterministic processes, such a natural selection, in speciation (Schluter 1996). Second, dwarf and normal smelt in Lake Utopia are, at present, not distinguished from one another taxonomically and such designations are not neces- sarily a simple matter (see discussion by Taylor and Bentzen 1993b). The uncertainty concerning their taxonomic recognition, however, cannot obscure the reality of their evolutionary and ecological distinc- tions and illustrates that considerable biodiversity is “hidden” and taxonomically unrecognized, and per- haps is unrecognizable using our current system of latin binomials (Taylor 1999). Situations such as the smelt of Lake Utopia, therefore, present a challenge to conservation philosophies and programmes focused on current taxonomic procedures. Evaluation The smelt populations in Lake Utopia are apparent- ly stable and in no immediate danger. The major spawning streams of the normal smelt are located in a game reserve and fishing pressure appears not to be excessive. A major data limitation relevant to their evaluation is the lack of information on the population trends, spawning habitat area and protection, and fish- ing pressure on the dwarf smelt. As with many other fishes, one of the greatest potential dangers facing the smelt populations are introductions of exotic species (e.g., Allendorf 1991) or enhancement of native game fish which are predators of smelt. Artificial increases in Atlantic Salmon or Brook Trout in the lake could upset any predator-prey equilibrium that may exist in the lake and may be particularly important for the dwarf smelt which are within the size range of salmonid prey for throughout their life. 136 Acknowledgments My interest in Lake Utopia smelt populations was stimulated by discussions with D. M. Blouw and D. E. McAllister. I have greatly appreciated the many helpful discussions concerning the evolution of Lake Utopia's smelt populations that I have had with J. D. McPhail and D. Schluter. P. Cronin and T. Vickers (New Brunswick Department of Natural Resources and Energy) provided advice and information con- cerning the biology and exploitation of Lake Utopia smelt. I appreciate comments made on an earlier ver- sion of this report by J. S. Nelson, E. J. Crossman, and A. Dextrase. The illustrations of smelt (Figure 1) were made by D. McPhail and were facilitated by the generous loan of specimens by D. McAlpine of the New Brunswick Museum (Natural History Division). Partial funding for this report was provided by NSERC operating grant 586196 awarded to EBT and by COSEWIC. Literature Cited Allendorf, F. W. 1991. Ecological and genetic effects of fish introductions: synthesis and recommendations. Can- adian Journal of Fisheries and Aquatic Sciences 48 (Sup- plement 1): 178-181. Anonymous. 1996. New Brunswick Recreational Fishery Regulations. New Brunswick Department of Natural Resources and Energy, Fredericton, New Brunswick. Bajkov, A. D. 1936. Investigations on smelt in Chamcook Lake, N.B. Fisheries Research Board of Canada Manu- script Report 201A: 1-15. Bernatchez, L., and J. J. Dodson. 1990. Allopatric origin of sympatric populations of lake whitefish (Coregonus clupeaformis) as revealed by mitochondrial DNA re- striction analysis. Evolution 44: 1263-1271. Bridges, C. D., and C. Delisle. 1974. Postglacial evolu- tion of the visual pigments of the smelt, Osmerus eper- lanus mordax. Vision Research 14: 345-356. Bush, G. L. 1994. Sympatric speciation in animals: new wine in old bottles. Trends in Ecology and Evolution 9: 286-288. Cope, E. D. 1870. A partial synopsis of the fishes of the fresh waters of North Carolina. Proceedings of the Amer- ican Philosophical Society 11: 448-495. Copeman, D. G., and D. E. McAllister. 1978. Analysis of the effect of transplantation on morphological and meris- tic characters in lake populations of the rainbow smelt, Osmerus mordax (Mitchill). Environmental Biology of Fishes 3: 253-259. Delisle, C. 1969. Ecologie, crissance et comportement de l'eperlan du lac Heney, Comte de Gatineau ainsi que la repartition en eau douce au Québec. Ph.D. thesis. De- partment of Biology, University of Ottawa. Delisle, C., and C. Veilleux. 1969. Repartition geogra- phique d'eperlan arc-enciel Osmerus eperlanus mordax et de Glugea hertwigi (Sporozoa: Microsporidia) en eau douce au Québec. Naturaliste canadien 96: 337-358. Dymond, J. R. 1939. The fishes of the Ottawa region. Contributions of the Royal Ontario Museum 15: 1-43. Ferguson, A., and J. B. Taggart. 1991. Genetic differen- tiation among sympatric brown trout (Salmo trutta) pop- ulations of Lough Melvin, Ireland. Biological Journal of the Linnean Society 43: 221-237. THE CANADIAN FIELD-NATURALIST Vol. 115 Foote, C. J., C. C. Wood, and R. E. Withler. 1989. Bio- chemical genetic comparison of sockeye salmon and kokanee, the anadromous and nonanadromous forms of Oncorhynchus nerka. Canadian Journal of Fisheries and Aquatic Sciences 46: 149-158. Gibbons, A. 1996. On the many origins of species. Sci- ence 273: 1496-1499. . Hindar, K., N. Ryman, and G. Stahl. 1986. Genetic dif- ferentiation among local populations and morphotypes of Arctic charr (Salvelinus alpinus). Biological Journal of the Linnean Society 27: 269-285. Hughes, T., H. W. Borns, Jr., J. L. Fastook, M. R. Hyland, J. S. Kite, and T. V. Lovell. 1985. Models of glacial reconstruction and deglaciation applied to Mari- time Canada and New England. Geological Society of America Special Paper 197: 139-150. Lanteigne, J., and D. E. McAllister. 1983. The pygmy smelt, Osmerus spectrum, Cope 1870, a forgotten sibling species of eastern North American fish. Syllogeus 45: 1-32. Mayr, E. 1963. Animal Species and Evolution. The Belknap Press of Harvard University, Cambridge. McPhail, J. D. 1984. Ecology and evolution of sympatric sticklebacks (Gasterosteus): evidence for a species pair in Enos Lake, British Columbia. Canadian Journal of Zoology 70: 361-369. McPhail, J. D. 1992. Ecology and evolution of sympatric sticklebacks (Gasterosteus): evidence for a species pair in Paxton Lake, Texada Island, British Columbia. Cana- dian Journal of Zoology 70: 361-369. Meyer, A., T. D. Kocher, P. Basasibwaki, and A. C. Wil- son. 1990. Monophyletic origin of Lake Victoria cichild fishes suggested by mitochondrial DNA sequences. Nature 347: 550-553. Nellbring, S. 1989. The ecology of smelts (Osmerus): a literature review. Nordic Journal of Freshwater Research 65: 116-145. Sayers, R. E, J. R. Moring, P. R. Johnson, and S. A. Roy. 1989. Importance of rainbow smelt in the winter diet of Atlantic salmon in four Maine lakes. North American Journal of Fisheries Management 9: 298-302. Schliewen, U. K., D. Tautz, and S. Paabo. 1994. Sym- patric speciation suggested by monophyly of crater lake cichlids. Nature 368: 629-632. Schluter, D. 1996. Ecological speciation in postglacial fishes. Philosophical Transactions of the Royal Society of London 351: 807-814. Scott, W. B., and M.G. Scott. 1988. Atlantic fishes of Canada. Canadian Bulletin of Fisheries and Aquatic Sci- ences number 219. Svardson, G. 1961. Young sibling fish species in north- western Europe. Pages 498-513 in Vertebrate Speci- ation. Edited by W. F. Blair. University of Texas Press, Austin. Taylor, E. B. 1999. Species pairs of north temperate freshwater fishes: evolution, taxonomy, and conserva- tion. Reviews in Fish Biology and Fisheries 9: 299-324. Taylor, E. B., and P. Bentzen. 1993a. Evidence for multi- ple origins and sympatric divergence of trophic ecotypes of smelt (Osmerus) in northeastern North America. Evolution 47: 813-832. Taylor, E. B., and P. Bentzen. 1993b. Molecular genetic evidence for reproductive isolation between sympatric populations of smelt Osmerus in Lake Utopia, south- western New Brunswick, Canada. Molecular Ecology 2: 345-357. 2001 Taylor, E. B., and J. J. Dodson. 1994. A molecular anal- ysis of relationships and biogeography within a species complex of Holarctic fish (genus Osmerus). Molecular Ecology 3: 235-248. Taylor, E. B., C. J. Foote, and C. C. Wood. 1996. Molec- ular genetic evidence for parallel life-history evolution in a Pacific salmon (sockeye salmon and kokanee, Oncorhy- nchus nerka). Evolution 50: 401-416. Taylor, E. B., S. Harvey, S. Pollard, and J. Volpe. 1997. TAYLOR: STATUS OF SYMPATRIC SMELT IN LAKE UTOPIA, NEW BRUNSWICK 137 Postglacial genetic differentiation between reproductive ecotypes of kokanee Oncorhynchus nerka in Okanagan Lake, British Columbia. Molecular Ecology 6: 503-518. Verspoor, E., and L. J. Cole. 1989. Genetically distinct sympatric populations of resident and anadromous At- lantic salmon, Salmo salar, Canadian Journal of Zoology 67: 1453-1461. Accepted 29 May 2000 Updated Status of the Northern Madtom, Noturus stigmosus, 1n Canadat ERLING HoLM2 and NICHOLAS E. MANDRAK?3 2Centre for Biodiversity and Conservation Biology, Royal Ontario Museum, 100 Queens Park, Toronto, Ontario MSS 2C6, Canada 3Y oungstown State University, Youngstown, Ohio 44555 USA Holm, Erling, and Nicholas E. Mandrak. 2001. Updated status of the Northern Madtom, Noturus stigmosus, in Canada. Canadian Field-Naturalist 115(1): 138-144. The Northern Madtom, Noturus stigmosus, is a small member of the family Ictaluridae. It is disjunctly distributed in the Mississippi River and western Lake Erie drainage basins and has only recently been collected in Canada. There was insuf- ficient information to assign a status in the original status report on the Northern Madtom. Recent collections indicate that reproducing populations are present in the Canadian waters of Lake St. Clair, the Detroit River and the Thames River. However, the stability, size and range of the populations are unknown; therefore, it is recommended that the Northern Madtom be classified as Vulnerable in Canada. Key Words: Northern Madtom, chat-fou du nord, Noturus stigmosus, Vulnerable, Detroit River, Lake St. Clair, Thames River. The Northern Madtom, Noturus stigmosus Taylor 1969, (Figure 1) is one of 25 species in the genus Noturus of the bullhead catfish family Ictaluridae (Robins et al. 1991). Only one specimen was known from Canada at the time the original status report was written (Goodchild 1993); therefore no status was assigned. Species in the genus Noturus can be distinguished from other genera in Ictaluridae by an adipose fin that is a long, low, ridge-like extension of the cau- ‘Indeterminate status assigned by COSEWIC April 1993, reassessed April 1998 and Vulnerable status assigned. ‘ =) ("96 FiGuRE 1. Northern Madtom, Noturus stigmosus Taylor, 1969, 77 mm TL, Detroit River. © 1996, Joseph R. Tomelleri. dal fin (Scott and Crossman 1973; Page and Burr 1991). However, in the Northern Madtom, the adi- pose fin appears to be almost completely separated from the caudal fin by a deep notch. The Northern Madtom has an overall colour pat- tern that is mottled with three irregular dark saddles on the back located at the front of the dorsal fin, behind the dorsal fin and at the adipose fm. Unlike the Brindled Madtom, Noturus miurus, the dorsal and adipose fins have pale distal margins. There are three or four irregular crescent-shaped bars on the caudal fin; the middle bar usually extending across the upper and lower caudal rays and touching the caudal peduncle. Two pale spots about three- 138 2001 quarters the diameter of the eye are usually present just anterior to the dorsal fin. Maximum total length is 132 mm. In spawning males, the head flattens, dark pigment diffuses and conspicuous swellings develop behind the eyes, on the nape, and on the lips and cheeks. This description is a compilation of diagnostic characters based on observations of ROM specimens and on those given by Page and Burr (1991), Etnier and Starnes (1993), and Trautman (1981). Only five species of Noturus have been collected in Canadian waters (Coad 1995). One of these, the Margined Madtom (Noturus insignis), is likely not native to Canada (Mandrak and Crossman 1992) although some (McAllister and Coad 1974, Goodchild 1993) have suggested that it may be indigenous. Noturus miurus, and Noturus stigmosus differ from Noturus insignis, Noturus flavus, and Noturus gyrinus, in having a mottled pattern with saddles on the back instead of a more uniform brown or gray colour. The two mottled madtoms also differ from the plain-coloured madtoms in hav- ing the posterior edge of the pectoral spine strongly serrated instead of weakly serrated. Scott and Crossman (1973) indicated that specimens of Noturus stigmosus are very similar to Noturus miu- rus and should be watched for in collections from HOLM AND MANDRAK: UPDATED STATUS OF NORTHERN MADTOM IN CANADA 139 southwestern Ontario. The characters above distin- guish Noturus stigmosus from Noturus miurus which has a dark blotch at the tip of the dorsal fin and a dark bar which extends to the extreme upper edge of the adipose fin. All specimens labelled Noturus miurus (40) in the fish collection of the Royal Ontario Museum (ROM) were examined and none were Noturus stigmosus. None of the Noturus miurus specimens in the Canadian Museum of Nature appear to be Noturus stigmosus (Goodchild 1993). Prior to 1969 the Northern Madtom was consid- ered a synonym of the Carolina Madtom, Noturus furiosus. Taylor (1969) described Noturus stigmosus as a distinct species and included it in the subgenus Rabida in the “furiosus-group” which included three other species of Noturus (Noturus munitus, Noturus furiosus, and Noturus placidus). He suggested that it was most closely related to Noturus munitus. Recent analysis of morphological, allozymic, and chromoso- mal data corroborates a monophyletic relationship among members of an expanded “furiosus-group” consisting of seven species (Noturus furiosus, Noturus munitus, Noturus placidus, Noturus stigmo- sus, Noturus eleutherus, Noturus flavater, and Noturus flavipinnis). Relationships within the “furio- sus-group” are not clear and depend on the charac- FIGURE 2. The North American distribution of Northern Madtom. Adapted from Rohde (1980) and Page and Burr (1991). 140 A N 10 0 10 Kilometers Algonac State Park a 2 = : S ) Lake St. Clair SA Peche Island THE CANADIAN FIELD-NATURALIST St, c\air i Vol. 115 Lake deuce wn = Lake Erie FiGuRE 3. Canadian distribution of the Northern Madtom 1963—1999. ters analyzed and the method of analysis (Grady and Legrande 1992). No subspecies of Noturus stigmo- sus have been recognized, but Mayden et al.(1992) indicated that it might be polytypic which may war- rant its division into several species. Distribution The Northern Madtom is found in the Mississippi and western Lake Erie and Lake St. Clair basins (Figure 2). In the Mississippi drainage, it is found in several tributaries in Mississippi and Tennessee, in the main stem between Arkansas and Tennessee and throughout most of the Ohio River basin in Kentucky, Indiana, Ohio and restricted areas of Illinois, Pennsylvania and West Virginia. It is found in several western Lake Erie tributaries in Indiana, Michigan and Ohio; and, in the St. Clair River, Lake St. Clair and the Detroit River which form the border between Michigan and Ontario (Rohde 1980; Stauffer et al. 1982: Cincotta et al. 1986). In Canada, the Northern Madtom is known only from the Detroit River, Lake St. Clair, and two tribu- taries of Lake St. Clair, the Thames River are the Sydenham River (Figure 3). The Northern Madtom has been known from the American side of the Detroit River since 1903 (University of Michigan Museum of Zoology; UMMZ 132009). It was first recorded in Canada in Lake St. Clair near the origin of the Detroit River where a single specimen was col- lected in 1963 (Trautman 1981). Another specimen was collected in 1994 near the first capture site on the Canadian side of the Detroit River (ROM 68328). In 1996, approximately 50 specimens were either cap- tured or observed in the area around Peche Island. Also, in 1996, three juveniles were seined at night in Lake St. Clair at the mouth of Belle River approxi- mately 19 kilometres east of the Detroit River. The lack of Canadian records in the Detroit River and Lake St. Clair between 1963 and 1994 is likely the result of limited trawling and night seining, and incorrect field identification. Specimens captured in 1963 and in 1996 were collected primarily by trawl- ing, a collecting method not commonly used on the Canadian side of Lake St. Clair and the Detroit River (Don MacLennan, Ontario Ministry of Natural Resources (OMNR), Lake St. Clair Fisheries Management Unit, personal communication). If pre- viously collected, specimens might have been incor- rectly identified since the Northern Madtom is not included in taxonomic keys commonly used in Canada (e.g., Scott and Crossman 1973). In July 1991, an adult specimen was captured by the ROM in the Thames River near Wardsville. A juvenile specimen, captured in August 1997 at the site of capture of the adult, indicates that Northern Madtoms are established in the Thames River. Both specimens were captured by seining in the daytime in the direction of the current during periods when the water level was low enough to wade across the river. These records are approximately 90 kilometers from the nearest Lake St. Clair record at Belle River. There are no apparent barriers such as dams to pre- vent its dispersal from Lake St. Clair to the Thames River at Wardsville. 2001 It has been recently recorded and is apparently established on the Michigan side of the lower St. Clair River at Algonac State Park (see Figure 3) (D. Jude, University of Michigan, personal communica- tion). It is likely that this population dispersed there from the Detroit River via Lake St. Clair. Sampling by the ROM in 1996 (day and night trawls in 2-10 metres, and day and night seining in 0.1-1.3 metres) failed to capture it on the Canadian side of the St. Clair River. In 1999, a specimen was captured by a commercial fisherman in Lake St. Clair off Walpole Island (ROM 720938). Protection 7 No specific legal protection exists for the Northern Madtom in Canada. The original COSEWIC report on the Northern Madtom conclud- ed that insufficient scientific information was avail- able to assign a status designation (Goodchild 1993). The species and/or its habitat may be protected by the Canada Environmental Assessment Act, Canada Environmental Protection Act, Canada Fisheries Act, Canada Water Act, Canada Wildlife Act, Ontario Environmental Protection Act, Ontario Environmental Assessment Act, Ontario Game and Fish Act, Ontario Planning Act and Ontario Water Resources Act. A recent Natural Heritage Policy Statement, 2.3, of the Ontario Planning Act reduces protection for species classified as Vulnerable. Only habitats of Threatened and Endangered species are protected against development and site alteration (fan Buchanan, Ministry of Natural Resources, per- sonal communication). The population of Northern Madtom in the Detroit River is in one of 43 “Areas of Concern”. The Detroit River has been identified by the United States and Canada as having several beneficial uses which have become impaired. These include degrad- ed fish and wildlife populations and loss of fish and wildlife habitats (Hartig et al. 1996). A Remedial Action Plan has been initiated and, if fully imple- mented, is likely to improve water quality, increase amount of fish habitat, and improve prospects for the survival of the Northern Madtom. The Northern Madtom is listed as of Special Concern in the United States by Mayden et al. (1992). Johnson (1987) listed it as legally protected in Michigan and Ohio and as of Special Concern in Kentucky, Mississippi and West Virginia. It is listed as of Special Concern in Tennessee by Etnier and Starnes (1993). It was listed as Threatened in Kentucky by the Kentucky Nature Protection Commission, but Burr and Warren (1986) recom- mended downlisting it to Special Concern. Global and North American federal, state and provincial conservation status and ranks were obtained from the Eastern Regional Office of the Nature Conservancy, Boston, dated 9 June 1997. The HOLM AND MANDRAK: UPDATED STATUS OF NORTHERN MADTOM IN CANADA 14] ranks assigned to Northern Madtom indicate that it is rare to extremely rare throughout its range: Global Rank: G3 National Ranks: US: N3 Canada: N1 Regional Ranks: IL (SH-historic), IN(S1), KY (S1), MI (S1), MS (S3?), OH (S1S2), PA (S1), TN (S3), WV (S1), ON(S1) l=extremely rare, 2=very rare, 3=rare to uncom- mon, 4=common, 5=very common Population Size and Trends No studies examining population size and trends have been conducted on the Canadian populations of the Northern Madtom. Recent collections of the species in Canada (37 specimens from 14 sites) and observations of males guarding larvae (MacInnis 1998) indicate that sustainable reproducing popula- tions are established in the Detroit River, Lake St. Clair and the Thames River. The new records, including those from the St. Clair River on the American side, suggest that the species is undergo- ing a range expansion. However, these new records result from a more intensive sampling program of trawling and night seining specifically targetting the Northern Madtom. Therefore, it is uncertain that the population size is increasing. The Northern Madtom has not been captured in the Sydenham River since 1975 (NMC 75-1623) despite several sampling attempts in the 1990s. It is possible that it no longer occurs there. Habitat The preferred habitat of the Northern Madtom is clear to turbid water of large creeks to big rivers with moderate to swift current. It occurs on bottoms of sand, gravel and rocks occasionally with silt, detritus, and accumulated debris, and is sometimes associated with macrophytes (Taylor 1969; Smith 1979; Trautman 1981; Cooper 1983; Burr and Warren 1986; Robison and Buchanan 1988). In Ontario, it was trawled in the Detroit River at depths of 1-3m on smooth, firm bottoms often covered by macrophytes such as Chara. The surface waters were not turbid, but a gradient of increasing turbidi- ty with increasing depth is present in the Detroit River (B. Ray, University of Windsor, personal communication). The Northern Madtom has also been seined at night in Lake St. Clair near the outlet of the Detroit River and around Belle River on sandy substrate devoid of cover. Two specimens were seined in the highly turbid Thames River (sec- chi <0.2m) on a bottom consisting of sand, gravel and rubble from areas where the substrate was free from silt and clay. Current was moderate, maximum depth of capture was 1.2 m, water temperature was 23-26°C, conductivity was 666 wS, and pH was 7.9. In the St. Clair River, it has been collected in trawls at depths of 3-7 metres (D. Jude, University of Michigan, unpublished data). 142 Biology Reproductive Capability Nests are made under large rocks and in anthro- pogenic debris such as large submerged cans, milk bottles, and boxes. In Michigan, Noturus stigmosus reproduced slightly earlier than Noturus miurus, and clutch sizes were larger ranging from 61 to 141 eggs (Taylor 1969). MacInnis (1998) observed and video-taped nest- ing of 21 adult Northern Madtoms in Lake St. Clair during the summer of 1996 while conducting research on the Round Goby, Neogobius melanosto- mus. Gravid females and recently spawned eggs were observed on 24 July 1996 in artificial goby nests set near Peche Island (see Figure 3). The nests were set in gentle current on a sandy bottom sur- rounded by a thick bed of aquatic macrophytes (pri- marily Chara). Eggs were approximately 3 mm in diameter and clutch size was conservatively estimat- ed to range from 32 to 160. The male guarded both the eggs and newly hatched fry and, did not abandon the nest when disturbed. Larvae and juveniles about 9 mm total length were observed being guarded by males on 13 August. The temperature during this period was 23°C. A male Brindled Madtom was also observed nesting during this period but when dis- turbed, would abandon the nest. Diet Recent analyses of stomach contents indicate that the Northern Madtom has a varied diet and is likely an opportunistic feeder. The stomach contents of 11 specimens from the Detroit River and one specimen from the Thames River were identified. Diet of the Detroit River specimens consisted primarily of chi- ronomids, fish (Mimic Shiners, Notropis volucellus), mayflies, particularly Hexagenia bilineata and possi- bly H. limbata, crustaceans (Malacostraca, an ostra- cod, and an amphipod). In addition they contained smaller amounts of nematodes, Lepidoptera, and cad- disflies (such as Triaenodes aba, Hydropsyche scalaris, and probably Polycentropus). The Thames River specimen contained mostly caddisflies (primari- ly Potamyia flava but one Hydropsyche scalaris) and mayflies (Emphemerella and probably Stenonema). French and Jude (in press) described stomach con- tents of 25 adult Northern Madtoms (82-130 mm TL) captured inn the St. Clair River. These specimens were caught near Algonac State Park, Michigan, in May and June of 1994. They contained mostly mayflies (primarily Baetisca and occasionally Hexagenia and Baetis). Caddisflies (primarily repre- sented Phryganea, Banksiola and a few Hydropsyche) were also present in substantial numbers. Midges of the family Chironomidae and Stoneflies (Plecoptera) were also represented to a lesser extent, and three small Round Gobies, Neogobius melanostomus, were found in the stomach of one specimen. The guts of juveniles (31-37 mm TL) collected at the same site THE CANADIAN FIELD-NATURALIST Vol. 115 contained mostly Diptera as well as numbers of mayflies (Ephemeroptera). Species Movement There is no published information on the move- ments or migration of the Northern Madtom. Behaviour/Adaptability The Northern Madtom probably feeds (Goodchild 1993) and spawns (Coad 1995) during the night. During diving transects in the Detroit River and Lake St. Clair, a few Northern Madtoms were observed off Peche Island at night, but none were seen during the day (B. Ray, University of Windsor. personal communication). Limiting Factors The restricted distribution and low numbers of the Northern Madtom suggests that it has specific eco- logical requirements (Goodchild 1993). The apparent persistence of populations of the Northern Madtom in the Detroit River, one of the most heavily polluted areas of the Great Lakes, suggests that it is relatively tolerant to human disturbance. However, the appar- ent absence of the Northern Madtom on the more polluted Canadian side of the St. Clair River (Griffiths et al. 1991) as well as the more polluted American side of the Detroit River (David Jude, per- sonal communication) suggests that it avoids the poorer water quality of these rivers. Its northward dispersal may be limited by temperatures which are warm enough (e.g. 23°C) for spawning to occur. Global warming may allow the species to spread far- ther north. The rapidly expanding populations of Round Goby in Lake St. Clair and the Detroit River is threatening some native species such as the Mottled Sculpin, Cottus bairdi (Jude et al. 1996). Its impact on the Northern Madtom is not known. Special Significance Noturus species exhibit cryptic behaviour and pos- sess poison glands associated with the pectoral spines that are unique to the Canadian fish fauna (Scott and Crossman 1973). One of the four Noturus species native to Canada, the Brindled Madtom, is listed by COSEWIC as Vulnerable (Campbell 1995) and we recommend a Vulnerable status for the Northern Madtom. Therefore, the genetic diversity expressed by behaviour, ecology and morphology in the genus Noturus may be in jeopardy in Canada. Evaluation The Northern Madtom is at the northern limit of its range in Canada. It has been found recently only in the Detroit River, Lake St. Clair and a tributary of Lake St. Clair, the lower Thames River. Although additional sampling is required to determine the sta- bility, size and range of the species, it appears that reproducing populations are established in the 2001 Detroit River, Lake St. Clair and the Thames River. These populations should persist if its habitat is not significantly altered. There is no evidence of any other reproducing populations in Canada. Therefore, it should be rec- ognized that, if these populations were extirpated due to human activity, the only known established Canadian populations of Northern Madtom would be lost. It is recommended that Noturus stigmosus be classified as Vulnerable in Canada. Acknowledgments Financial support for fieldwork and preparation of this report was provided by the Royal Ontario Museum, World Wildlife Fund (Canada), Canadian Wildlife Service, and Section 25 of Human Resources Development Canada. A. MacInnis and B. Ray, University of Windsor, provided invaluable field assistance, unpublished data and distribution records. W. Ramshaw and and A. Salomon prepared the maps using ARC/VIEW GIS software provided by ESRI Canada. D. Boehm, K. Banks, S. Newton, M. Ciuk, K. Ditz, R. Guppy and W. Ramshaw assist- ed with fieldwork. P. MacCulloch identified inverte- brate stomach contents. We are grateful to D. Jude, University of Michigan, for providing unpublished data on collections of Northern Madtom in the United States. The manuscript was improved by comments from the reviewers: E. Crossman, A. Dextrase, A. Peden, and J. Nelson. This paper is Contribution number 92 of the Centre for Biodiversity and Conservation Biology, Royal Ontario Museum. Literature Cited Burr, B. M., and M. L. Warren, Jr. 1986. A distribu- tional atlas of Kentucky fishes. Kentucky Nature Preserves Commission, Scientific and Technical Series Number 4. Campbell, R. R. 1995. Rare and Endangered fishes and marine mammals of Canada: COSEWIC Fish and Marine Mammal Subcommittee Status Reports IX. Canadian Field-Naturalist 107: 395-401. Cincotta, D. A., R. L. Miles, M. E. Hoeft, and G. E. Lewis. 1986. Discovery of Noturus eleutherus, Noturus stigmosus, and Percina peltata, in West Virginia, with discussions of other additions and records of fishes. Brimleyana 12: 101-121. Coad, B. W. 1995. Encyclopedia of Canadian Fishes. Canadian Museum of Nature, Ottawa, and Canadian Sportfishing Productions Inc., Waterdown, Ontario. Cooper, E. L. 1983. Fishes of Pennsylvania and the northeastern United States. The Pennsylvania State University Press, University Park, Pennsylvania. Etnier, D. A., and W. C. Starnes. 1993. The fishes of Tennessee. The University of Tennessee Press, Knoxville, Tennessee. French, J. R. P., HI, and D. J. Jude. in press. Diets and diet overlap of nonindigeneous gobies and small benthic native fishes co-inhabiting the St. Clair River, Michigan. Journal of Great Lakes Research. HOLM AND MANDRAK: UPDATED STATUS OF NORTHERN MADTOM IN CANADA 143 Goodchild, C. D. 1993. Status of the Northern Madtom, Noturus stigmosus, in Canada. Canadian Field-Naturalist 107: 417-422. Grady, J. M., and W. H. LeGrande. 1992. Phylogenetic relationships, modes of speciation, and historical bio- geography of the madtom catfishes, genus Noturus Rafinesque (Siluriformes: Ictaluridae). Pages 747-777 in Systematics, historical ecology, and North American freshwater fishes. Edited by R. L. Mayden. Stanford University Press, Stanford, California. Griffiths, R. W., S. Thornley, and T. A. Edsall. 1991. Limnological aspects of the St. Clair River. Hydrobiologia 219: 97-123. Hartig, J. H., D. P. Dodge, D. Jester, J. Atkinson, R. Thomas, and K. Cullis. 1996. Toward integrating remedial action planning and fishery management plan- ning in Great Lakes Areas of Concern. Fisheries 21: 6-13. Hubbs, C. L., and K. F. Lagler. 1964. Fishes of the Great Lakes Region. The University of Michigan Press, Ann Arbor, Michigan. Johnson, J. E. 1987. Protected fishes of United States and Canada. American Fisheries Society, Bethesda, Maryland. Jude, D. J., J. Janssen, and G. Crawford. 1996. Ecology, distribution, and impact of the newly intro- duced round and tubenose gobies on the biota of the St. Clair and Detroit rivers. Jn The Lake Huron Ecosystem: Ecology, Fisheries, and management. Edited by M. Munawar, T. Edsall, and J. Leach. Ecovision World Monograph Series, S. P. B. Academic Publishing, The Netherlands. Mandrak, N. E., and E. J. Crossman. 1992. Checklist of the Fishes of Ontario, annotated with distribution maps. Royal Ontario Museum. Mayden, R. L., B. M. Burr, L. M. Page, and R. R. Miller. 1992. The native freshwater fishes of North America. Pages 827-863 in Systematics, historical ecol- ogy, and North American freshwater fishes. Edited by R. L. Mayden. Stanford University Press, Stanford, California. MacInnis, A. J. 1998. Reproductive biology of the Northern Madtom, Noturus stigmosus (Pisces: Ictaluridae) in Lake St. Clair. Canadian Field-Naturalist 112: 245-249. McAllister, D. E., and B. W. Coad. 1974. Fishes of Canada’s national capital region. Miscellaneous Special Publication 24. National Museum of Natural Sciences. Page, L. M., and B. M. Burr. 1991. A field guide to freshwater fishes. Houghton Mifflin Company, Boston, Massachusetts. Robins, C. R., R. M. Bailey, C. E. Bond, J. R. Brooker, E. A. Lachner, R. N. Lea, and W. B. Scott. 1991. Common and scientific names of fishes from the United States and Canada. American Fisheries Society Special Publication 20, Bethesda, Maryland. Robison, H. W., and T. M. Buchanan. 1988. Fishes of Arkansas. The University of Arkansas Press, Fayetteville, Arkansas. Rohde, F. C. 1980. Noturus stigmosus Taylor, Northern Madtom. Page 469 in Atlas of North American freshwa- ter fishes: Hdited by). Ss Lee: GC: R. Gilbert, C,H. Hocutt, R. A. Jenkins, D. E. McAllister, and J. R. Stauffer Jr. North Carolina State Museum of Natural History, North Carolina Biological Survey Number 1980-12. 144 THE CANADIAN FIELD-NATURALIST Vol. 115 Scott, W. B., and E. J. Crossman. 1973. Freshwater fish- Taylor, W. R. 1969. A revision of the catfish genus es of Canada. Fisheries Research Board of Canada Noturus Rafinesque with an analysis of higher groups in Bulletin 184. the Ictaluridae. U. S. National Museum bulletin 282: Smith, P. W. 1979. The Fishes of Illinois. University of 1-315. Illinois Press, Urbana, Illinois. Trautman, M. B. 1981. The fishes of Ohio. Ohio State Stauffer, J. R., Jr.. B. M. Burr, C. H. Hocutt, and R. E. University Press, Columbus, Ohio. Jenkins. 1982. Checklist of the fishes of the central and northern Appalachian Mountains. Proceedings of the Biological Society of Washington 95: 27-47. Accepted 29 May 2000 Status of the Bluntnose Minnow, Pimephales notatus, in Canada‘ J. HOUSTON 743 Fireside Drive, R.R.1 Woodlawn, Ontario KOA 3MO, Canada Houston, J. 2001. Status of the Bluntnose Minnow, Pimephales notatus, in Canada. Canadian Field-Naturalist 1 15(1): 145-151. The Bluntnose Minnow, Pimephales notatus, is a small cyprinid with a Canadian distribution from the upper St Lawrence drainage of southwestern Quebec west through southern Ontario to south eastern Lake Superior. It also occurs in the Lake- of-the-Woods region west of Lake Superior and other populations have been reported from two sites on the Winnipeg River and one site on the Red River of southeastern Manitoba. Although abundant where found in Ontario and Quebec it is rare in Manitoba. Le ventre-pourri, Pimephales notatus, est un petit cyprinidé dont l'aire de répartition au Canada s'étend depuis l'amont du bassin versant du Saint-Laurent, dans le secteur sud-ouest du Québec, jusqu'a l'extrémité sud-est du lac Supérieur et englobe le sud de l'Ontario. On peut aussi trouver l'espéce a |’ ouest du lac Supérieur dans la région du Lake-of-the-Woods et d'autres populations ont été signalées 4 deux sites dans la riviére Winnipeg et a un site dans la riviére Rouge, au sud-est du Manitoba. L'espéce est abondante, ot on la trouve, en Ontario et au Québec, mais elle est rare au Manitoba. Key Words: Cyprinidae, cyprinids, minnows, Bluntnose Minnow, ventre-pourri, Pimephales notatus, rare fishes. The Bluntnose Minnow (Figure 1), Pimephales notatus (Rafinesque, 1820), is a small cyprinid (~ 64 mm average length) occupying a broad range in central North America (Scott and Crossman 1973). It bears some resemblance to the Fathead Minnow (Pimephales promelas), Slim Minnow (Pimephales tenellus) and Bullhead Minnow (Pimephales vigi- lax) [Trautman 1957], the remaining three North American members of the genus (Smith 1979). However, only Pimephales notatus and Pimephales promelas are found in Canada (Scott and Crossman 1973). In Canada it is abundant and widely distributed from southwestern Quebec west through the Great Lakes to the eastern end of Lake Superior and north to Temagami (Scott and Crossman 1973). It is also found in the Lake-of-the-Woods area in Ontario, and in Manitoba where it is known only from three localities (Keleher and Kooyman 1957; Scott and Crossman 1973). This synopsis is a result of a request from offi- cials of the Manitoba Department of Natural Resources to have the species considered by the Committee on the Status of Endangered Wildlife in Canada (COSEWIC). Description The Bluntnose Minnow (Figure 1) is a small cyprinid averaging 64 mm in length with an elon- gated, tubular body. The head is small, broad and bluntly triangular with moderate eyes; the snout is rounded and blunt giving rise to the common name "Designated “Not at risk by COSEWIC April 1998. and slightly overhangs the upper lip. The mouth is inferior, overhung by the snout, the gape does not reach the front of the eye; pharyngeal tooth count is 0,4-4,0. The dorsal fin of eight rays originates slightly behind the origin of the pelvic fins which also have eight rays; the caudal fin is moderately forked and rounded; the anal fin has seven (some- times eight) rays and originates below the tip of the depressed dorsal; the pectoral fins are small with 15 rays but may have 14 to 17. The scales are cycloid, small and dense predorsally; the lateral line is complete with 42 to 50 scales. The peri- toneum is black, the intestine is elongate and the vertebrae number 37 or 38 (see Scott 1967; Scott and Crossman 1973). The fish are olive-green to brown dorsally, sil- very on the sides and silvery white ventrally. The scales are dark-edged which gives a cross-hatched appearance which is particularly noticeable in young fish. A dark lateral band extends from the eye to a conspicuous black spot at the base of the caudal fin (Scott and Crossman 1973), the band is usually present around the snout, but may be less noticeable (Trautman 1957). Specimens from clear waters or areas with vegetation are usually promi- nently marked, those from turbid waters may be pale and lack the band and cross hatching (Trautman 1957). The fins are transparent, in older fish there may be a yellowish (Scott and Crossman 1973) or olive tint and the dorsal fin usually has a dark blotch on the two or three anterior rays (Trautman 1957). Breeding males develop a thickened papilla like protuberance at each corner of the mouth, three rows of large, sharp nuptial tubercles across the 145 146 THE CANADIAN FIELD-NATURALIST Vol ails MIMICS Say “ye FiGureE |. Bluntnose Minnow, Pimephales notatus [from Scott and Crossman (1973) by per- mission]. snout (Scott and Crossman 1973) and a spongy pad on the flatter, predorsal portion of the back, some- times thick enough to hide the scales (Trautman 1957). The dorsal surfaces of the pectoral fins may also have tiny tubercles (Trautman 1957). The males become very dark, almost black during breeding, even the fins, particularly the dorsal fin are permeated with chromatophores (Scott and Crossman 1973). The Bluntnose Minnow differs from the Fathead Minnow in that the Fathead Minnow has a terminal, oblique mouth; an incomplete lateral line; and is more yellowish in colour (Trautman 1957). It dif- fers from other shiners with a black caudal spot in le NEPA S—-— S \ the crowded predorsal scales, elongate intestine and black peritoneum (Smith 1979). Distribution The species is native to central North America in the Mississippi and Great Lakes drainages (Figure 2). In the U.S., it ranges from eastern North Dakota east to New York, south to Virginia on the Atlantic slope, southwest to the Gulf States and north through the Mississippi basin to the Great Lakes (Scott and Crossman 1973; Lee and Shute 1980). In Canada, the species has a somewhat disjunct distribution (Figure 2) representative of post-glacial dispersion (Hocutt and Wiley 1985). It is widely i) HE ate ie FIGURE 2. Canadian distribution of the Bluntnose Minnow, Pimephales notatus [from Scott and Crossman (1973) by permission]. 2001 HOUSTON: STATUS OF BLUNTNOSE MINNOW IN CANADA 147 Pimephales notatus bluntnose minnow FIGURE 3. Distribution of the Bluntnose Minnow, Pimephales notatus, in Ontario [from Mandrak and Crossman (1992) by permission]. distributed from southwestern Quebec west through the Great Lakes to the Lake-of-the-Woods area in Ontario (Figure 3), and in Manitoba (Figure 4) where it is known only from three localities (Keleher and Kooyman 1957; Scott and Crossman 1973), two on the Winnipeg River, and one on the Red River. Crossman and McAllister (1985) discuss possible post-glacial dispersion of the species into the Hudson Bay watershed from a Mississippi- Refugium. Protection The Bluntnose Minnow is not subject to any spe- cific protection in Canada. The species has been considered to be of “special concern” in Manitoba (Johnson 1987), but has not as yet been considered for protection under the provincial Endangered Species Act. In the United States, Bluntnose Minnows have been considered to be of “special concern” only in North Carolina (Johnson 1987). Population Sizes and Trends The Bluntnose Minnow is generally common where suitable habitat exists and is widely distribut- ed in southwestern Quebec, southern Ontario and western Ontario (Lake-of-the-Woods), where it appears in most areas in sufficient numbers to be utilized as a bait fish (D. E. McAllister [Researcher Emeritus, Canadian Museum of Nature], Perth, Ontario K7H 3A0; personal communication). Although many streams have undergone habitat degradation similar to that responsible for the demise of other less sensitive cyprinids, the species may still be found at most sites where it has been known in this century and there is no evidence of a decline in the species in Ontario or Quebec [Collections at Canadian Museum of Nature, Ottawa; Royal Ontario Museum, Toronto; and Ontario Ministry of Natural Resources, Toronto; appendices of all records to 1990 have been deposited with COSEWIC at Canadian Wildlife Service, Ottawa). In Manitoba, Pimephales notatus occurs in small 148 roe FIGURE 4. Distribution of the Bluntnose Minnow, Pimephales notatus, in Manitoba (after sources cited in the text). numbers in the Winnipeg River above the Point du Bois Dam (50°18'N, 95°30’W) and vicinity, and Echo Lake, Whiteshell Provincial Park [ROM (Royal Ontario Museum) Number 14896]; as well as St. Andrew's Locks on the Red River (ROM 14896). Stewart (K. W. Stewart, Department of Zoology, University of Manitoba, Winnipeg, Manitoba; personal communication 1997) has examined the Red River specimens and confirms their validity. He feels that these were either waifs washed down the Winnipeg River, or more likely, released bait brought in from farther south. Apart from this anomalous collection Stewart (personal communication) indicated that there has been no downstream movement of Pimephales notatus in the Winnipeg River. Scott and Crossman (1973) reported that the species was rare in Manitoba. Stewart (personal communication) believes that the species is of concern in Manitoba only because of its extremely restricted range. Although not abun- dant in the Winnipeg River above the Point du Bois Dam it can be collected there with some pre- ' dictability. THE CANADIAN FIELD-NATURALIST Vol. 115 Habitat The Bluntnose Minnow is a resilient species occur- ring in almost ali types of waters within its natural range except the deeper waters of lakes and rivers (Trautman 1957). It prefers hard-bottomed, sandy or gravelly shallows of pools in creeks and small rivers, but may be found elsewhere, except (and as previ- ously noted) in swamps, heavily weeded areas and heavily silted ditches and ponds (Scott and Crossman 1973; Smith 1979). The Bluntnose Minnow appears to be able to survive in areas where there is high competition from a number of other species and can inhabit waters with gradients of 0 to 15 m/km, although its abundance is limited where the gradient is more than 9 m/km or where there is intense com- petition, especially with Fathead Minnows. It has a high tolerance to siltation and turbidity, and organic and inorganic pollutants, but is less abundant in excessive turbidity and siltation or extreme (high and low) pH values (Trautman 1957). Schlosser (1985) found that the species was more successful in stable flow regimes and was less abundant in periods of above normal flow conditions. Stauffer et al. (1984) found that the species could tolerate water temperatures in the range of € to 36°C with a preferred temperature of 26.3°C. Cherry et al. — (1977) found the preferred temperature to be 28.4°C in Virginia. Spawning usually occurs when water temperatures reach 20°C (Scott and Crossman 1973). It apparently will spawn just about anywhere in shallow waters and will migrate up the smallest temporary brook to spawn. Males build nests around spawning objects which may be just about anything including the underside of stones, boards, logs, tin cans leaves or other objects that are not easily silted over (Trautman 1957; Scott and Crossman 1973). Bluntnose Minnows can be successfully artificial- ly propagated in ponds, especially where artificial nesting sites are provided (Cooper 1936; Dobie et al. 1956; Trautman 1957; Scott and Crossman 1973); General Biology Reproduction Spawning commences in May to June when water temperatures reach 20°C and is prolonged (into August) in some areas (Toner 1943; Scott and Crossman 1973). Spawning ceases when water tem- peratures fall below this minimum and does not commence again until a few days after the minimum is reached. Higher temperatures also lead to cessa- tion of spawning (Gale 1983). Not all spawning occurs at one time. Nests may contain eggs in vari- ous stages of development and the ovaries of females contain 1700 to 2200 eggs in various stages of matu- ration (Westman 1938; Scott and Crossman 1973). Gale (1983) and Westman (1938) found the spawn- ing usually occurs at night, but it may occur during the day (Hubbs and Cooper 1936). Captive females 2001 deposited an average of 93 to 239 eggs per session, intervals between sessions lasting two to 14 days (Gale 1983). Minnows in the wild spawned at inter- vals of two to eight days (Westman 1938) laying 25 to 100 eggs per session, each session lasting 10 to 30 minutes (Hubbs and Cooper 1936). Gale (1983) reported that captive females spawned seven to 19 times over the spawning season (May to August) producing a total of 1100 to 4200 eggs; Westman (1938) noted that a pair in a 500 m? pool spawned 12 times and produced 2300 eggs. Gale (1983) surmised that fecundity was unrelated to body size. The eggs are large, 1.0 - 1.5 mm in diameter, adhesive and deposited in a nest on the undersurface of submerged or floating objects where they are guarded and aerated by the male (Scott and Crossman 1973; Smith 1979; Gale 1983). Westman (1938) found that the male's presence is essential to ensure a steady flow of water and oxygen over the eggs and to keep them free of silt by the movement of the water and by cleaning the eggs with the spongy tissue on the nape (Trautman 1957). The eggs hatch in 7 to 14 days depending on water tem- perature (Scott and Crossman 1973). Growth Fish (1932) provided a detailed description of the eggs and young at lengths of 5 mm (newly hatched), 6 mm (seven days old) and 12 mm [two weeks after hatching (Westman 1938)]. Growth depends on water temperature and food abundance (Mahon and Ferguson 1981). They may attain a length of 55 mm by the end of their first growing season in New Jersey (Westman 1838). Mahon and Ferguson found young of the year raised in a reservoir at Guelph, Ontario averaged 33.3 mm S.L. [Standard Length] as compared to 27.6 mm S.L. for those taken at the Speed River in Guelph at the end of the growing sea- son. Trautman (1957) reports that young of year in Ohio may reach lengths of 71 mm depending on date of hatching and year-old-fish 72 mm. Females reach maturity in their second year and males in their third year. Longevity is not known, but three-year-old fish have been taken. Males grow faster than females and may attain a length of 102 mm (New Jersey), while females reach 76 mm (Scott and Crossman 1973). Trautman (1957) noted greatest length in Ohio at 108 mm and Smith (1979), found a maximum of 90 mm in Illinois. Diet The Bluntnose Minnow is a generalized bottom feeder and the diet consists of organic detritus, algae, plankton, aquatic vegetation and insect larvae (Scott 1967; Scott and Crossman 1973; Smith 1979; Lee and Shute 1980). The inferior mouth is an adaptation to bottom feeding (Scott and Crossman 1973). Individuals from Kearney Lake, Algonquin Park, Ontario, fed mainly on chironomid larvae and algae (Scott and Crossman 1973), whereas Keast and HOUSTON: STATUS OF BLUNTNOSE MINNOW IN CANADA 149 Webb (1966) found that fish from Lake Opinicon, Ontario, relied on bottom ooze, chironomid larvae and Cladocera, in that order. Keast (1985) also found that the Bluntnose Minnows in Lake Opinicon were seasonal, specialized planktivores and that Chydorus sphaericus, was the main zooplanktor taken, regard- less of abundance. Morgan and Colgan (1988) found that naive school members begin to feed more quick- ly on a new food source when experienced school members are present. Behaviour Spawning The male selects a suitable stone or other object resting on the bottom in 0.15 to | m of water and excavates a depression beneath the object and cleans its undersurface with his tail and horny snout. The males do not seek out females and will only allow a female to enter the nest when it is completed. The adhesive eggs are deposited on the underside of the object and fertilized by the male (Trautman 1957; Scott and Crossman 1973; Smith 1979). The males are territorial and aggressive and guard the nest until after all eggs have hatched; they will permit addi- tional females to deposit eggs, but drive away other intruders (Smith 1979). Up to 5000 eggs have been found in a single nest (Hubbs and Cooper 1936). The nesting territory may be small when large objects are available; Hubbs and Cooper (1936) reported that long boards could have one nest per linear foot. The males also clean the eggs with the spongy tis- sue on the nape which keeps them from being suffo- cated by silt. Males also provide oxygen by moving water over the eggs with their tails (Trautman 1957). If the male is removed from the nest the eggs will die within 12 hours (Westman 1938). Ming and Noakes (1984) found that Bluntnose Minnow males dominate Fathead Minnows in selec- tion and defence of spawning sites because of their larger size and inherent aggressiveness where ade- quate spawning sites (near the substrate) are avail- able. If spawning sites are scarce, the Fathead Minnows may dominate Bluntnose Minnows. They also found that interspecific dominance was not a factor of size or maturity and that codominance may occur where there are plenty of spawning sites. Movement Bluntnose Minnows segregate into localized groups of similar-sized individuals (Becker 1983). In the presence of predators, the incidence of size seg- regation increases; smaller fish move to the centre of the school leaving larger fish on the outside (Theodorakis 1989), thereby reducing the risk of pre- dation to the smaller individuals. In spring, mature fish migrate into shallower waters (0.15 to 1 m) to spawn. Schlosser (1985) indi- cated the importance of spring flooding in second- order streams for spawning and Trautman (1957) 150 noted their habit of migrating up small, temporary streams to spawn. The adults and young return to deeper, permanent waters before the temporary streams dry up (Trautman 1957). Predators/Parasites The Bluntnose Minnow is an important forage fish for game fishes, particularly Yellow Perch (Perca flavescens), sunfishes (Lepomis spp.), Rock Bass (Ambloplites rupestris), Smallmouth Bass (Micropterus dolomieu) and Largemouth Bass (Micropterus salmoides), and other centrarchids (Scott and Crossman 1973). The species is cannibal- istic as young males will devour the eggs in unpro- tected nests (Scott and Crossman 1973). The trematodes, Uvullifer sp., Uvullifer vancleavi, and Lebouria cooperi; the cestode Ligula intestinalis; Protozoa and Mysosporidia have been found on Bluntnose Minnows from Lake Erie (Bangham and Hunter 1939). Specimens from Lake Huron were infected with the trematodes including Posthodiplostomum minimum, protozoans and the nematode Rhabdochona cascadilla (Bangham 1955). Hoffman (1967) lists many species of Protozoa, Trematoda, Cestoda, Nematoda, molluscan glochidia and the crustacean Lernaea cyprinacea as parasitic of the species in North America. Hockett (1988) found that parasite loads of metacercariae (Uvulifer sp.) had no effect on the growth or thermal tolerance of the species. Pot and Noakes (1985) found that individual Bluntnose Minnows could be recognized by means of the pattern of spots caused by trematode cysts. Limiting Factors Although it is not particularly sensitive to turbidi- ty and siltation (Trautman 1957), the species is least abundant in areas of excessive turbidity and with high rates of siltation; in waters of extreme pH val- ues, acid or alkaline; where the gradient is greater than 9 m/km high. It is also less abundant in areas where there are large concentrations of predators or competitors such as the Fathead Minnow (Trautman IW). Special Significance The Bluntnose Minnow is an important forage fish for game fishes, particularly centrarchids and uses a food source not exploited by predacious fishes. It is also used for live bait in Canada and the United States, but does not take well to crowding in a pail (Trautman 1957; Scott and Crossman 1973; Cooper 1983). Its broad distribution and abundance may be relat- ed to the fact that it occupies habitat not usually inhabited by other species, and can survive in low numbers under extreme competition. The species is tolerant of turbidity and siltation and does well in streams of very low gradient (Trautman 1957). THE CANADIAN FIELD-NATURALIST Vol. 115 The disjunct distribution and habitat requirements are of interest to science in relation to the zoogeo- graphic history and possible dispersal of fish subse- quent to the Wisconsin glaciation. Evaluation The Bluntnose Minnow is abundant and widely distributed in Canada from southwestern Quebec to the eastern end of Lake Superior. It is known only from three sites in Manitoba on the Red River, the Winnipeg River, and Echo Lake of the Winnipeg- Rainy River system. It is also found in the Lake-of the-Woods area in Ontario. The western distribution in Ontario is probably the result of postglacial dis- semination from a Mississippi Refugium. Its rarity and restricted range in Manitoba where it is known from only three locations has lead to it being consid- ered a species of concern there (Stewart, personal communication), but it appears to be abundant and not at risk elsewhere in Canada. Acknowledgments Funding for this report was made available through World Wildlife Fund (Canada) and the Department of Fisheries and Oceans. The author is grateful for the assistance of A. Derksen, Manitoba Department of Renewable Resources; Don E. McAllister, Canadian Museum of Nature; E. J. Crossman, Royal Ontario Museum; G. Gale, Ontario Ministry of Natural Resources and P. Hood, Quebec Ministry of The Environment and Wildlife, all of whom provided collection records and advice, and to COSEWIC for the opportunity to present the report. Literature Cited Bangham, R. V. 1955. Studies on fish parasites of Lake Huron and Manitoulin Island. American Midland Naturalist 53: 184-194. Bangham, R. V., and G. W. Hunter III. 1939. Studies on fish parasites of Lake Erie. Distribution studies. Zoologica 24: part 27: 385-448. Becker, G. C. 1983. Fishes of Wisconsin. University of Wisconsin Press, Madison, Wisconsin. Cherry, D.S., K. L. Dickson, J. Cairns Jr., and J. R. Stauffer, Jr. 1977. Preferred, avoided and lethal tem- peratures of fish during rising temperature conditions. Journal of the Fisheries Research Board of Canada 34: 239-246. Cooper, E. L. 1983. Blackchin shiner, Notropis het- erodon (Cope). Page 106 in Fishes of Pennsylvania. Pennsylvania State University Press, University Park, Pennsylvania. Cooper, G. P. 1936. Importance of forage fishes. Pages 305-310 in Proceedings first North American Wildlife Conference. U.S. Fish and Wildlife Service, Washington, D.C. Crossman, E.J., and D.E. McAllister. 1985. Zoogeography of freshwater fishes of the Hudson Bay drainage, Ungava Bay and the Arctic Archipelago. Pages 54-101 in The Zoogeography of North American fresh- water fishes. Edited by C.H. Hocutt and E. O. Wiley. John Wiley and Sons, New York, New York. 2001 Dobie, J., O. L. Meehean, S. F. Snieszko, and G.N. Washbaurn. 1956. Raising bait fishes. U.S. Fish and Wildlife Service Circular 35. Fish, M. P. 1932. Contributions to the early life histories of sixty-two species of fishes from Lake Erie and its tributary waters. Bulletin of the U.S. Bureau of Fisheries 10(47): 293-398. Gale, W. F. 1983. Fecundity and spawning frequency of caged Bluntnose Minnows — fractional spawners. Transactions of the American Fisheries Society 112: 398-402. Hockett, C. T. 1988. [Abstract] Effects of metacorcaria on growth and thermal tolerance of spotfin shiners (Notropis spilopterus), striped shiners (Notropis chrysocephalus) and bluntnose minnows (Pimephales notatus). Page 44 in April program abstracts. 97th annu- al Meeting, the Ohio Academy of Science, April 29-30, May 1, 1988. Ohio Journal of Science 88(2). Hocutt, C. H., and E. O. Wiley. Editors. 1985. The Zoogeography of North American freshwater fishes. John Wiley and Sons, New York, New York. Hoffman, G. L. 1967. Parasites of North American fresh- water fishes. University of California Press, Los Angeles, California. Hubbs, C.L., and G. P. Cooper. 1936. Minnows of Michigan. Cranbrook Institute of Science Bulletin 8: 1-95. Johnson, J. E. 1987. Protected fishes of the United States and Canada. American Fisheries Society, Bethesda, Maryland. Keast, A. 1985. Planktivory in a littoral-dwelling lake fish association: prey selection and seasonality. Journal of Fisheries and Aquatic Sciences 42: 1114-1126. Keast, A., and D. Webb. 1966. Mouth and body form rel- ative to feeding ecology in the fish fauna of a small lake. Lake Opinicon, Ontario. Journal of the Fisheries Research Board of Canada 23: 1845-1867. Keleher, J. J., and B. Kooyman. 1957. Supplement to Hinks “The fishes of Manitoba.” Manitoba Department of Mines and Natural Resources, Winnipeg, Manitoba. Lee, D. S., and J. R. Shute. 1980. Pimephales notatus Rafinesquew, Bluntnose Minnow. Page 340 in Atlas of North American freshwater fishes. Edited by D. S. Lee, @eEeeGuibert, CH. Hocutt, R. E. Jenkins, D. E. McAllister, and J. R. Stauffer Jr. North Carolina State Museum of Natural History, North Carolina Biological Survey Publication 1980-12. HOUSTON: STATUS OF BLUNTNOSE MINNOW IN CANADA 151 Mahon, R., and M. Ferguson. 1981. Invasion of a new reservoir by fishes: species composition, growth, and condition. Canadian Field-Naturalist 95: 272-275. Mandrak, N. E., and E. J. Crossman. 1992. Checklist of the Fishes of Ontario, annotated with distribution maps. Royal Ontario Museum, Contribution. Ming, F. W., and D. L.G. Noakes. 1984. Spawning site selection and competition in minnows (Pimephales nota- tus and P. Promelas) (Pisces, Cyprinidae). Biology of Behaviour 9: 227-234. Morgan, M. J., and P. W. Colgan. 1988. The role of experience in foraging shoals of bluntnose minnows (Pimphales notatus). Behavioural Processes 16: 87—93. Pot, W., and D. L. G. Noakes. 1985. Individual identifi- cation of bluntnose minnows (Pimephales notatus) by means of naturally acquired marks. Canadian Journal of Zoology 63: 363-365. Schlosser, I. J. 1985. Flow regime, juvenile abundance, and the assemblage structure; of stream fishes. Ecology 66: 1484-1490. Scott, W. B. 1967. Freshwater fishes of eastern Canada. Second Edition. University of Toronto Press, Toronto, Ontario. Scott, W. B., and E. J. Crossman. 1973. Freshwater fish- es of Canada. Fisheries Research Board of Canada, Bulletin 184. Smith, P. W. 1979. The fishes of Illinois. University of Illinois Press, Urbana, Illinois. Stauffer, J. R. Jr., E. L. Melisky, and C. H. Hocutt. 1984. Interrelationships among preferred, avoided, and lethal temperatures of three fish species. Archiv Fiir Hydrobiologie 100: 159-169. Theodorakis, C.W. 1989. Size segregation and the effects of oddity on predation risk in minnow schools. Animal Behaviour 38: 496-502. Toner, G. C. 1943. Ecological and geographical distribu- tion of fishes in eastern Canada. M.A. thesis. University of Toronto, Toronto, Ontario. Trautman, M. B. 1957. The fishes of Ohio. Ohio State University Press, Columbus, Ohio. Westman, J. R. 1938. Studies on the reproduction and growth of the blunt-nosed minnow, Hyborhynchus nota- tus (Rafinesque). Copeia 1938: 57-61. Accepted 29 May 2000 Status of the Texada Stickleback Species Pair, Gasterosteus spp., in Canada’ J. HOUSTON 743 Fireside Drive, R.R.1, Woodlawn, Ontario KOA 3MO, Canada Houston, J. 2001. Status of the Texada Stickleback Species Pair, Gasterosteus spp., in Canada. Canadian Field-Naturalist 1151): 152-156. A sympatric pair of stickleback species (Gasterosteus spp.) occur in Paxton Lake on Texada Island, British Columbia. These species are reproductively isolated and adapted to alternate trophic niches. One form (benthic) is a littoral, benthic forager, whereas the other (limnetic) is a pelagic plankton feeder. The benthic member is larger (90 mm Standard Length), with a short body, wide mouth relatively few, short gill rakers, and a reduced number of lateral plates and dorsal spines. The entire pelvic girdle is absent in about 80% of adults. The limnetic form is smaller, slim, small-mouthed, with long gill rakers and normal lateral plates, dorsal spines, and pelvic girdle. Crosses breed true, suggesting the forms represent distinct gene pools. Genotypic and phenotypic characteristics have remained stable for over 20 generations with less than 1% inci- dence of hybridization which indicates that gene flow between the forms is negligible and that the forms behave as biologi- cal species. Although both forms are numerous, they are restricted to one small (17 ha) lake which is used for a source of water for industry. In the short term the species could be threatened by the introduction of other fishes and in the long term by water draw down and further industrial development. On trouve dans le lac Paxton sur l'ile Texada, en Colombie-Britannique, deux espéces sympatriques d'épinoche (Gasterosteus spp.). Ces espéces sont isolées du point de vue reproductif et adaptées a des niches trophiques alternatives. La forme benthique est surtout littorale et se nourrit de benthos tandis que la forme limnétique est planctivore et vit dans la zone pélagique. La forme benthique est plus grande (90 mm de longueur standard), a un corps trapu, une bouche large, des branchicttures courtes et relativement peu nombreuses, et un nombre réduit de plaques latérales et d'épines dorsales. La ceinture pelvienne est absente chez 80% des adultes. La forme limnétique a un corps mince, une bouche étroite, de nom- breuses branchicténtes longues et des plaques latérales, des épines dorsales et une ceinture pelvienne normale. Dans les expériences de croisements, les formes ne donnent pas d'hybrides et cela indique qu'elies représentent des pools génétiques distincts. Les caractéres génotypiques et phénotypiques sont restés stables pendant plus de 20 générations avec moins de 1% d’hybrides. Ceci indique que le passage de génes entre les deux formes est négligeable, et que les deux formes se com- portent comme des espéces biologiques. Les populations des deux espéces sont élevées, mais elles sont confinés a un petit lac (17 ha) du lequel on préléve l'eau pour l'industrie. A court terme, les espéces pourraient étre menacées par I'introduction d'autres espéces de poissons, et a long terme par l'utilisation industrielle de l'eau et une augmentation du développement industrielle. Key Words: Gasterostidae, sticklebacks, Gasterosteus, Threespine Stickleback, Texada Island, d’épinoches de Texada, Paxton Lake, British Columbia, endangered fishes. The Threespine Stickleback (Gasterosteus aculea- tus Linnaeus, 1758) is widely distributed in temperate marine and fresh waters and cooler coastal waters throughout the northern hemisphere. Within this vast range there is a plethora of ecological forms and mor- phologically distinct geographic isolates (McPhail 1989, 1992a). The taxonomy of the genus is chaotic (Miller and Hubbs 1969; Hagen and McPhail 1970) and the only thing that is certain is that Gasterosteus aculeatus is not a single species, but a complex of species (McPhail and Lindsey 1970; Bell 1976). Description The nominate species, the Threespine Stickleback is a small fish [averaging 51 mm Standard Length ‘Threatened status assigned by COSEWIC April 1998. (S.L.)] with a laterally compressed, elongate body tapering to a slender acaudal peduncle. There are usu- ally three to four branchiostegal rays; a dorsal fin con- sisting of three isolated, short serrated spines, the last being shorter, preceding 11 to 13 soft rays. The pec- toral fins are large and located laterally behind the gill opening; the pelvics are thoracic and ventral, each with one spine and one soft ray. The body is plated with a variable number of vertical, oblong pates on each side. These may be numerous (30 or more) on marine specimens, but are usually few to none on freshwater forms. The lateral line is complete and high on the sides (see Scott and Crossman 1973). The colour of these forms varies, depending on habitat, from silvery green to grey, olive, greenish brown, to black. The sides are lighter with silvery reflections and the belly silvery. The fins are trans- parent, but may have a red tinge. During the breed- ing season mature males may be brilliantly coloured SZ of 2001 with red bellies and flanks, and blue eyes; females may assume pink tints on the throat and belly. The larvae are yellowish and young fish silvery (see Scott and Crossman 1973). There is tremendous morphological variability within the species (see Hagen and McPhail 1970). Marine forms have a complete series of bony plates, are silvery in colour, larger in size, and are anadro- mous. Freshwater forms are smaller, have reduced, but variable numbers of plates and spend their lives entirely in fresh water (Scott and Crossman 1973). A particularly remarkable diversity has evolved in the freshwater populations of British Columbia (and Washington) in relation to the isolation of the lakes in which they are found and the speed with which the isolates react to local selective regimes (see later discussion under Special Significance). In these lake populations there are inherited differ- ences in size, colour, spines, plate numbers, gill raker length and numbers, and mouth size (McPhail 1992a). Although the trophic morphology reflects an evolutionary response to a local foraging regime, McPhail (1989; 1992b), and others (Moodie and Reimchen 1976; Reimchen 1984a,b; Reimchen et al. 1985) have shown that at least some of these forms (e.g., Enos, Boulton, Rouge, Serendipity and Mayer lakes), including the species pair found in Paxton Lake, Texada Island, are distinct (but unnamed) species and not part of a large gene pool containing a foraging polymorphism. Recently McPhail (1992b) described the two sym- patric forms from Paxton Lake. They are reproduc- tively isolated, genetically discrete and adapted to exploit alternate trophic niches, filling all the require- ments of biological species (Mayer 1963). At present the least confusing way to refer to these animals is simply the Texada or Paxton Lake Stickleback species pair. A formal taxonomic designation must await a detailed examination of their relationship to other divergent forms of Gasterosteus. The benthic form (Figure 1) is larger (up to 90 mm S.L.) than the nominate (51 mm) or limnetic form which averages 45 mm S.L. Benthic fish are stouter, have wider mouths, fewer gill rakers and a reduced number of lateral plates and dorsal spines, in about 80% of adults the pelvic girdle is missing (see McPhail 1992b: his Tables 3,4). On the other hand, the limnetic fish (Figure 1) are more like the nomi- nate form, slim bodied, narrow mouth, numerous big gill rakers, and normal lateral plates and dorsal spines and a normal pelvic girdle (see McPhail 1992b: his Tables 3,4). Distribution The Texada Stickleback species pair is restricted to Paxton Lake (49°42'30"N, 124°31'30"W), Texada Island, British Columbia (Figure 2). Two other lakes (Myrtle and Cove) downstream of Paxton Lake and HOUSTON: STATUS OF TEXADA STICKLEBACK SPECIES PAIR IN CANADA 153 Rumbottle Creek contain sticklebacks similar to the benthic form and these may be derived from Paxton Lake benthics (McPhaill 1992b; Cannings 1993). Of the eight other lakes on the island containing stickle- backs, only Paxton Lake contains girdleless fish (McPhail 1992b). Protection Currently the fish are not subject to any protected status and could not qualify for general protection under Habitat Sections of the Fisheries Act (not commercial species). General protection could be afforded if required under British Columbia provin- cial wildlife and endangered species legislation. Population Sizes and Trends Both forms are numerous in the lake, the total population of each probably exceeding 10 000 indi- viduals (Cannings 1993), but there is no data on which to base trends in population size. The popula- tions are probably more or less stable at this time (McPhail 1992b), but probably fluctuate over time. McPhail (1992b) suggests this has happened in the past since 5% of the limnetic fish are girdleless, and postulates that this has arisen through gene flow from the benthics at a time when the density of parental forms was biased towards the benthic fish. Habitat Paxton Lake is a small (17 ha) lake with a maxi- mum depth of about 15 m (Figure 3). The lake is about 90 m above sea level and the only outlet, Rumbottle Creek, drops about 80 m in a series of small falls in its last two km before entering Malisipina Strait and thus isolating the upper por- tions of the creek and the lake from the sea. There is no permanent surface flow into the lake and the out- let was dammed in 1956 to provide water for a near- by iron mine. This increased the lake level by 1.5 m. Apparently, before damming of the outlet the lake consisted of two lakes joined by about 100 m of stream at the North Rockface (Figure 3). The lake now has two equal basins connected by a narrow channel (McPhail 1992b). The lake lies in limestone on postglacial marine sediments and the substrate, especially at the south end is marl (Larson 1976). The annual drawdown now appears stable at about | m (McPail 1992b), but from 1957 to 1979 was approximately 3 m (Larson 1976; McPail 1992b). Summer water temperature may reach 23°C with a thermocline between 2 to 5 m, hydrogen sulphide may be detected below 7 m in summer, Chara is the dominate vegetation in the littoral zone (Larson 1976). Benthic fish occur in open, mud-bottomed situa- tions, above the deoxygenated zone, and smaller fish (<50 mm) are usually to be found in shallower water. They prefer some cover and are often found 154 THE CANADIAN FIELD-NATURALIST Vol. 115 FIGURE 1. Paxton Lake Sticklebacks. (A) Limnetic male (45.3 mm S.L.). (B) Benthic male (45.9 mm S.L.). Reproduced from McPhail (1992b) by permission. around sunken logs. Spawning occurs in shallower waters of the littoral zone and nests are usually found under cover in aquatic vegetation (Larson 1976). The limnetics form large aggregations near the surface in the littoral zone during the day, usually in areas with surface cover (trees, leaves, etc.,) or in tall vegetation above Chara. They appear to migrate off- shore at dusk and in winter are found on the lake bottom. Spawning occurs in open areas in the littoral zone devoid of cover (Larson 1976). Biology ‘The only published account of the biology of the species pair is that of Larson (1976). Both species show the typical Gasterosteus life cycle and behav- iour (see Scott and Crossman 1973). They do differ in their feeding behaviour, spatial distribution and seasonal habits. The limnetic form is a pelagic plankton feeder concentrating on zooplankton, main- ly cladocera; whereas the benthic form is a benthos feeder, feeding mainly on amphipods (Gammarus sp.), midge larva (Chironomidae) and ostracods. Some small individuals feed partially on plankton (Larson 1976). During the sumer the limnetic species occurs near the lake surface and moves to deeper water in winter. The benthic species occupies the lit- toral zone in summer and disperses over the entire lake bottom in winter. The species pair are the only fish which breed in the lake. Cutthroat Trout (Oncorhynchus clarki) are also native to the lake, but have virtually disappeared since the outlet stream was dammed in 1956, there being no trout spawning site in the lake itself (McPhail 1992b). There are sticklebacks intermedi- ate between the two forms, but these are not abun- dant (1 to 2 %) and their morphology and behaviour probably puts them at a disadvantage relative to the parental forms (McPhail 1992b). The majority of sticklebacks in the lake can easily be separated by eye into benthic and limnetic types; the terms allude to the two major habitats used in the lake and do not necessarily imply a relationship with benthic or limnetic forms in other lakes. The two forms differ in morphology (number of dorsal spines and body plates, presence of pelvic girdles and spines, and number of gill rakers — see Description above and McPhail 1992b). Benthics are larger and more aggressive and frequently chase the limnetics forcing them to seek cover above the Chara. The sticklebacks of Paxton Lake have been under intermittent study since 1968. The differences in morphology and allelle frequencies between the two forms have remained stable during the period of study with about 1% incidence of hybridization McPahil 1992b). In laboratory crosses the two forms breed true indicating the two forms represent sepa- rate gene pools. Limiting Factors Water from the lake is used for industrial purpos- es. At present the annual drawdown is about | m, but in the past has been as much as 3m (McPahil 1992b). If larger drawdowns occur the populations 2001 PAXTON LAKE FiGURE 2. Paxton Lake, Texada Island, and the Strait of Georgia. Reproduced from McPhail (1992b) by per- mission. could be adversely affected by subsequent changes in the ecology of the lake. The only known predator is the Cutthroat Trout which is also native to the lake. These have all but disappeared since the outlet stream was dammed in 1956, there being no trout spawning site in the lake (McPhail 1992b). However, local people still occa- sionally introduce trout. Coho Salmon (Oncorhynchus kisutch) where introduced into the lake in 1968 and were significant predators. They did not breed however, and by 1973 were extirpated (McPail 1992b). A major threat to these small B.C. lakes is the introduction of exotic species, especially Catfish (Americanus nebulosus) and Pumpkinseed (Lepomis gibbosus). The introduction of such exotics has apparently led to the extinction of at least one species pair in Hadley Lake (R. Campbell, Chairman, COSEWIC Subcommittee Fish and Marine Mam- mals, Ottawa, Ontario; personal communication). Special Significance of the Species The sympatric populations of Threespine Sticklebacks in Paxton Lake provide excellent materi- al for the study of speciation. What is remarkable about the Paxton Lake pair is the rapidity of the event (less than 12 500 years, see McPahil 1992b) and the relatively short period of geographic isolation. The morphological, ecological and behavioural differences HOUSTON: STATUS OF TEXADA STICKLEBACK SPECIES PAIR IN CANADA 155 [it Keyes OAS arat ‘ RUMBOTTLE WS phate Ae! aoe a \ CREEK ~~===- (i7)-. DAM . ‘ pcteetS ‘ \ 5 2 Se Se eas ROCKFACE TRANSECT 100 m / ie Zs af i GAT) ---~—. FiGurRE 3. A bathymetric map of Paxton Lake, showing the two basins, and key (collection) sites. Reproduced from McPhail (1992b) by permission. that occur between members of the pair evolved with- in the lake. This in situ evolution and the rapidity of its occurrence makes the species pair unique. Evaluation The stickleback species pair of Paxton Lake is one of very few such pairs in the world. All of these are confined to the Strait of Georgia region of southwest- ern British Columbia. The Paxton Lake forms meet the criterion for biological species (Mayr 1942; McPahil 1992b), but final taxonomic recognition awaits a com- plete revision of the genus (McPhail 1989). This does not negate the responsibility to protect unique popula- tions where there is compelling evidence for more than one species such as in Paxton Lake. The species pair is restricted to one small lake, susceptible to exotic fish introductions, urban and industrial development and water drawdown for industrial use. They should be considered threatened as was done in the case of the Enos Lake species pair (McPhail 1989). Acknowledgments The author would like to thank R. Campbell, Subcommittee Chairman of the COSEWIC Fish and 156 Marine Mammals Subcommittee for helpful com- ments and advice in preparing this report. Thanks also to D. E. McPhail, Biodiversity Centre and Department of Zoology, University of British Columbia for his critical review and comments. Financial support was provided by World Wildlife Fund (Canada) and the Department of Fisheries and Oceans. Literature Cited Bell, M.S. 1976. Evolution of phenotypic diversity in Gasterosteus aculeatus subspecies on the Pacific coast of North America. Systematic Zoology 25: 211-227. Cannings, S.G. 1993. Rare freshwater fish of British Columbia. B.C. Conservation Data Centre Report to Fisheries Branch, B.C. Ministry of Environment, Lands and Parks, Victoria, British Columbia. Hagen, D. W., and J. D. McPhail. 1970. The species prob- lem within Gasterosteus aculeatus on the Pacific coast of North America. Journal of the Fisheries Research Board of Canada 27: 147-155. Larson, G. 1976. Social behaviour and feeding ability of two phenotypes of Gasterosteus aculeatus in relation to their spatial and trophic segregation in a temperate lake. Canadian Journal of Zoology 54: 107-121. Mayr, E. 1963. Animal species and evolution. Harvard University Press, Cambridge, Massachuetts. McPhail, J. D. 1989. Status of the Enos Lake stickleback species pair, Gasterosteus spp. Canadian Field- Naturalist 103: 216-219. McPhail, J.D. 1992a. Speciation and the evolution of reproductive isolation in sticklebacks (Gasterosteus) of southwestern British Columbia. Jn Evolution of the threespine stickleback. Edited by M. A. Bell and S. A. Foster. Oxford University Press, Oxford, England. THE CANADIAN FIELD-NATURALIST Vol. 115 McPhail, J.D. 1992b. Ecology and evolution of sym- patric sticklebacks (Gasterosteus): Evidence for a species pair in Paxton Lake, Texada Island, British Columbia. Canadian Journal of Zoology 70: 361-369. McPhail, J. D., and C. C. Lindsey. 1970. Freshwater fishes of northwestern Canada and Alaska. Fisheries Research Board of Canada Bulletin 173. Miller, R. R., and C. L. Hubbs. 1969. Systematics of Gasterosteus aculeatus, with particular reference to intergradation and introgression along the Pacific coast of North America: a commentary on a recent contribu- tion. Copeia 1969: 52-69. Moodie, G. E. E., and T.E. Reimchen. 1976. Phenetic variation and habitat differences in Gasterosteus popula- tions of the Queen Charlotte Islands. Systematic Zoology 25: 49-61. Reimchen, T. E. 1984a. Status of the Giant (Mayer Lake) Stickleback, Gasterosteus sp., on the Queen Charlotte Islands, British Columbia. Canadian Field-Naturalist 98: 115-119. Reimchen, T. E. 1984b. Status of Unarmoured and Spine- deficient populations (Charlotte Unarmoured Stickle- back), of Threespine Stickleback, Gasterosteus sp., on the Queen Charlotte Islands, British Columbia. Canadian Field-Naturalist 98: 120-126. Reimchen, T. E., E.M. Stenson, and J.S. Nelson. 1985. Multivariate differentiation of parapatric and allopatric populations of threespine sticklebacks in the Sargas River watershed, Queen Charlotte Islands. Canadian Journal of Zoology 63: 2944-2951. : Scott, W.B., and E. J. Crossman. 1973. Freshwater fish- es of Canada. Fisheries Research Board of Canada Bulletin 184. Accepted 29 May 2000 Updated Status of the Central Stoneroller, Campostoma anomalum, in Canada‘ E. HoLM and E. J. CROSSMAN Centre for Biodiversity and Conservation Biology, Royal Ontario Museum, 100 Queens Park, Toronto, Ontario M5S 2C6, Canada Holm, E., and E. J. Crossman. 2001. Updated status of the Central Stoneroller, Campostoma anomalum, in Canada. 5 Canadian Field-Naturalist 115(1): 157-167. The Central Stoneroller was known in Canada from only two areas in Ontario when it was classified as Vulnerable in 1985. It has continued to expand its range not only where it was originally found in the Thames River drainage, but also through- out southern Ontario. It is now known from the extreme southwestern end of Ontario near Kingsville, east to a tributary of the Rideau River near Perth, and north to a tributary of southern Georgian Bay near Midland. This ecologically important herbivore has the potential to make profound changes in the stream flora and fauna where it occurs in abundance. Moderately tolerant to stream degradation and abundant throughout most of its North American range, this large minnow should now be considered as Not at Risk in Canada. Key Words: Campostoma anomalum, Central Stoneroller, Cyprinidae, roule-caillou, Vulnerable. The Central Stoneroller, Campostoma anomalum (Rafinesque, 1820), was assigned a status of Rare (= Vulnerable) in Canada by the Committee on the Status of Endangered Wildlife in Canada (COSEWIC) in 1985 based on its limited and isolated distribution in the North Thames River in southwestern Ontario. Although a range expansion was evident in 1979, McAllister (1987) recommended that the species deserved some level of protection in Canada because the expansion may have been temporary and the streams in the agricultural areas where it occurred were “exposed to pesticide spills and other environ- mental disasters”. This report provides updated information on the continuing spread of the species both in the Thames River drainage and in many other river systems of southern Ontario. A summary of recent information from research on the ecology and biology of the Central Stoneroller is also pre- sented. Taxonomy and Description Campostoma anomalum (Figure 1) is the most common and widespread of four species currently recognized in the genus (see Page and Burr 1991). Careful examination of the lower jaw with a probe will reveal a distinctive white cartilaginous ridge which is an important diagnostic character of the species. It is also distinguished from other Canadian cyprinids by its extremely long gut that is coiled around the swim bladder. Spawning males of the Central Stoneroller are very distinctive, having ‘Reviewed and Approved by COSEWIC, April 1998, Not At Risk (formerly Vulnerable — April 1985). numerous tubercles on the head and back and a black band on the orange dorsal and anal fins (Figure 1, top). Since it was only relatively recently discovered in Canada, Campostoma anomalum, has probably been frequently misidentified and therefore missed in sur- veys of fishes. It was not included in the widely used key to minnows in Scott and Crossman (1973). If that key is used, specimens of the Central Stoneroller would probably key out to the Brassy Minnow (Hybognathus hankinsoni). It has also been known to be confused with Blacknose Dace (Rhinichthys atratulus), River Chub (Nocomis micropogon), Hornyhead Chub (Nocomis biguttatus), and juvenile suckers (Catostomidae) (personal observations; McAllister 1987; Gruchy et al. 1973). Currently three subspecies of Campostoma anomalum are recognized but the variation in the species is unclear (Burr 1980; Page and Burr 1991; Jenkins and Burkhead 1994). Campostoma anoma- lum michauxi from Virginia (Jenkins and Burkhead 1994) and the Carolinas lacks a black band on the anal fin of breeding males (Page and Burr 1991). The usual concept is that the individuals found in Ontario should be Campostoma anomalum pullum. Most character values of 24 Canadian specimens (Table 1) approximate those for Campostoma ano- malum pullum, although circumferential, dorsal cir- cumferential, and predorsal scale counts tend to be slightly lower than those described by Burr (1978) for Illinois populations. Larval characteristics and development of Campo- stoma anomalum from the Susquehanna drainage near Berwick, Pennsylvania were described by Buy- nak and Mohr (1980). However, identification of specimens less than 20—30 mm is unreliable. Si 158 $. AAP ARY SV See ee ped, +8 684: $€4¢ 07%," ees OEY Yy i! Gt? at "8 vy 4 VAG adds i THE CANADIAN FIELD-NATURALIST Vol. 115 FicurE |. Central Stoneroller Campostoma anomalum: upper, 6 99 mm TL; lower 2 91 mm TL (Copyright 1987, Joseph R. Tomelleri). Distribution Campostoma anomalum is widespread and ubiqui- tous throughout much of eastern and central United States (see Figure 2). Older records of this species from Alabama, Georgia and Tennessee recorded by Burr (1978) have been reidentified as the Largescale Stoneroller, Campostoma oligolepis (Page and Burr 1991; Boschung 1992; Brooks M. Burr, personal communication). Campostoma anomalum is found in the Red River in North Dakota and Minnesota (Burr 1980) and therefore would have access to streams in Manitoba. It would be expected to show up first in the Rat and Roseau rivers where some limited sam- pling has been done but no Central Stonerollers have been found (Ken Stewart, personal communication). McAllister (1987) reported the Central Stoneroller from only two areas of Ontario. There was a well- established population occupying a 25 X 45 km por- tion of the Thames River drainage between London and Mitchell and there were two records from the Niagara River around the mouth of Frenchman’s Creek (see open circles, Figures 3 and 4). Campo- stoma anomalum was first discovered in Canada in 1972 in the North Thames River approximately 125 km from the nearest populations in Michigan (Gruchy et al. 1973). McKee and Parker (1982) noted a considerable expansion of range between 1972 and 1979 in the North Thames River and the population was maintaining itself in 1982 (McAI- lister 1987). The Niagara River records consist of two immature specimens of Campostoma anomalum captured in 1977 in stream surveys conducted by the Ontario Ministry of Natural Resources (OMNR). No subsequent captures of the species have been made in the Niagara River or its tributaries. Records reported by McAllister (1987) appear as open circles in Figures 3 and 4 and new records are indicated by solid squares. In order to follow through time the spread of the species, the dates at which the species was collected from new locations were tabu- lated (Table 2). Table 2 and the distribution maps have been prepared from records (which are on file with the COSEWIC Secretariat, Canadian Wildlife Service, Ottawa, Ontario K1A 0H3) compiled from a variety of sources including the OMNR, Royal Ontario Museum (ROM), Canadian Museum of 2001 HOLM AND CROSSMAN: UPDATED STATUS OF THE CENTRAL STONEROLLER £59 TABLE 1. Characteristics of the two common subspecies of Campostoma anomalum and a sample of Canadian specimens. C. a. anomalum 15-16! 45-51 (x=48.2, n=51)? 34-41 (x=37.6, n=51)? 83-91 (x=85.8, n=51)? 15-173 18-23 (x=19.5)? 21-29 (x=24.9, n=14)? 44-58 (x=50)* Character pectoral rays lateral line scales (Ils) circumferential scales (cs) Ils+cs dorsal circumferential scales predorsal scales gill rakers gape width C. a. pullum 16-19! 46-63 (x=50.8, n=875)? 36-50 (x=41.9, n=876)? 85-108 (x=92.7, n=874)? 18-203 20-27 (x=23.0) 26-35 (x=30, n=31) 44-60 (x=55)4 Canadian specimens 17-19 (x=18, n=24) 48-56 (x=51, n=24) 36-43 (x=40, n=21) 87-98 (x=91.0, n=21) 17-19 (x=18, n=24) 19-24 (x=21, n=18) 24—34 (x=29, n=20) 48-65 (k=56.5, n=23) 'According to Page and Burr (1991) 2According to Burr (1976) for Illinois populations 3Dorsal circumferential scales (from Page and Burr 1991: 94): scales over the body from lateral line to lateral line at the dorsal origin (including lateral-line scales) 4Thousandths of SL according to Burr (1976) for Illinois Nature (NMC), Conservation Authorities, education- al institutes, and private consultants. The species was recorded in 1978 in a small Essex County stream (Mill Creek) near Kingsville. Fieldwork in 1996 con- firmed the existence of the population in Mill Creek and found the Central Stoneroller in two additional streams just to the west. These records represent the western limit of the species in Ontario. In the Ontario portion of the Lake Huron drainage, it was first recorded in 1980 and has recently been reported from municipal drains of the upper Maitland River. In 1984 the species was discovered in the Rouge River in Metropolitan Toronto and has since spread throughout that system and into neighbouring streams. Although we have seen only one specimen from the Lindsay area (from Sturgeon Lake), it is apparently well established in the Scugog River, and its tributaries Jenning’s Creek, McLaren Creek, and Mariposa Brook (Ted Warren, Sir Sandford Fleming College, personal communication). The species is also known from an area around Kaladar in central Ontario (see “?” symbol on Figure 3). A single spec- imen captured from a Tay River tributary in Lanark County represents the eastern limit of the known range of the species in Canada and specimens cap- tured from Stothart Creek (see Table 2) represent the northern limit of the species in Ontario. The range of the Central Stoneroller in the Thames drainage has continued to expand. It is now known from there as far downstream as Tates bridge 27 km SW of London and upstream to a Thames tributary near Woodstock, 42 km northeast of London (Figure 4). Since it is often misidentified, however, the range of the Central Stoneroller may be greater than our records indicate. Campostoma anomalum has been considered native to Ontario (Gruchy et al. 1973; Mandrak and Crossman 1992; and Litvak and Mandrak 1993). Its distribution in the southern half of the Michigan peninsula (Burr 1980) suggests that it could have used the Chicago, Grand Valley, or Fort Wayne out- lets to gain access to glacial lakes in Ontario (Mandrak and Crossman 1992). Its discovery in 1972-1980 in the North Thames River (open circles in Figures 3 and 4) so far from other populations suggests that the species was introduced probably from bait buckets. Alternatively, recent natural dis- persal from Michigan populations via Lake St. Clair and the Thames River is a distinct possibility. No barriers exist now or have existed in the past to pre- vent dispersal from Lake Saint Clair up to the North Thames River (Jack Robertson, Lower Thames Valley Conservation Authority, personal communi- cation). Dispersal up the North Thames River would, however, been prevented after 1950-1952 when the Fanshawe dam was being constructed. Populations in the Niagara River, St. Clair River, and Essex county streams are closer to United States popula- tions and it is more likely that these populations have resulted from natural dispersal. The Essex County records are 30-40 km from Michigan populations in Lake Erie tributaries. The St. Clair River records are about 35 km from records in Michigan tributaries of Lake St. Clair. The Niagara River records are located approximately 22 km northeast from specimens cap- tured near the mouth of Eighteenmile Creek, drain- ing into the south east end of Lake Erie, New York (Cornell University Catalogue number 20044). The Central Stoneroller has also been found in tributaries of the Niagara River in New York where it was recorded in several streams in the upper parts of Buffalo Creek from 1928 to 1949 (Greeley 1929, University of Michigan Museum of Zoology, and Cornell University collection records). Underhill (1957) suggested that the presence of Campostoma anomalum in the upper Mississippi River above the St. Anthony’s Falls in Minnesota is best explained by bait-bucket introduction. The Central Stoneroller has also been introduced into New Mexico (Burr 1980) and Connecticut (Nature Conservancy 1997). It has been suggested (McKee and Parker 1982; Litvak and Mandrak 1993) and it 160 seems likely that some Ontario populations especial- ly those in central and eastern Ontario have been introduced. It is considered to be one of the best bait species for bass (Micropterus spp.), Walleye (Stizostedion vitreum) and catfish (Ictaluridae) (McAllister 1987) and is known to be sold in Ottawa (Coad 1995), Toronto (Litvak and Mandrak 1993) and Parry Sound (E. McIntyre, OMNR, personal communication) to anglers who travel to a variety of places in southern Ontario. Protection The Federal Fisheries Act legislates against destruction of fish habitat but the existence of this legislation is not always successful in preventing this destruction. Classified as a Vulnerable species in 1985, the Central Stoneroller may have received some proac- tive protection from adverse environmental impacts of new developments through the Ontario Planning Act. However, as a result of amendments to this Act in 1996, this protection now applies only to Threatened and Endangered species (Jan Buchanan, OMNR, personal communication). There are, how- THE CANADIAN FIELD-NATURALIST ever, fish habitat protection provisions in the new Planning Act (Alan Dextrase, OMNR, personal com- munication). The species is classified globally as G5 (very common) and in the United States it is classi- fied as N5 (very common). Johnson (1987) listed the Central Stoneroller as Special Concern in North Dakota although its S-rank in that state is $4 (com- mon) according to the Nature Conservancy (1997*). In Louisiana, it has an S-rank of S2 (very rare). It is considered common (S4) or very common (S5) in 22 states and its S-rank in five other states is unknown (Nature Conservancy 1997*). Population Size and Trend Although the Central Stoneroller is considered a very common fish in many parts of the United States (Pflieger 1975; Smith 1979; Trautman 1981; Cooper 1983; Page and Burr 1991; Nature Conservancy 1997*), it may be rare or declining in Louisiana (S2), North Dakota (S3) (Nature Conservancy 1997*) and Wyoming (Baxter and Stone 1995). *See Documents Section 0 50 100 150 Kilometers ee ee _ eee =J FiGure 3. Canadian distribution of the Central Stoneroller, 1972 to 1996. Open circles represent records according to McAllister (1987). Solid squares represent records discovered since McAllister (1987). The question mark repre- sents The Kaladar area collections. McKee and Parker (1982) reported that it was fre- quently the most common fish in much of the upper Thames drainage. They gave density estimates for Ontario (0.37—3.72 fish m2, n=3) that are similar to those given by Mundahl and Ingersoll (1989) for Ohio (0.10-—3.91 fish m-2) and other parts of the United States (0.07—3.1 fish m-2). No population esti- mates have been conducted on Canadian popula- tions. Based on relative abundance to other species in catches, the Central Stoneroller is moderately abundant in several of the streams outside the Thames drainage. It comprised 0.4 to 5.1% (k=1.6, n=7) of the number of specimens in collections made in the Rouge River system and represented 0.6 to 16.5% (kx=6.1%, n=3) of the catch at the three col- lections made in 1996 in Essex County streams. A 1993 electrofishing survey of upper Thames streams in Oxford County captured a total of 32 Central Stonerollers at 7 of 59 sites sampled (ROM Accessions 6087, 6090, and 6107). The species was abundant at one site where it represented 55% of 22 specimens captured and moderately abundant at two sites where it comprised 3 of 15 and 8 of 52 specimens. A 1996 survey of creeks by the Grand River Con- servation Authority discovered a well-established population of Central Stonerollers in the upper Grand River drainage near Grand Valley and Luther Lake. It was found at 14 of 30 sites sampled. A total of 215 Campostoma anomalum were captured which represented 6.4% of all specimens collected. At three sites there were more stonerollers caught than any other species and at another three sites it was the sec- ond most abundant species in the catches (Grand River Conservation Authority, unpublished data). The cause of both the rapid increase in abundance and expansion of range in the Thames and subse- quently in the rest of southern Ontario is unknown. McAllister (1987) speculated that habitat could have improved as the result of climatic conditions. It is also possible that removal of riparian vegetation and heavy use of fertilizers in the agricultural areas have 162 THE CANADIAN FIELD-NATURALIST Volh iis 25 Kilometers RE FiGurE 4. Distribution of the Central Stoneroller, 1972 to 1996 in the Thames River drainage. Open circles represent records according to McAllister (1987). Solid squares represent records discovered since McAllister (1987). increased the food supply for Campostoma anoma- lum. A possible factor is increased awareness of fish collectors who have been more watchful for this species. However, the sudden appearance of the species in the North Thames in the 1970s does not appear to be the result of misidentification of the species in earlier collections. McKee and Parker (1982) noted that they found Campostoma anoma- lum abundant at many sites where they were not found by C.G. Gruchy in 1972. Collections were made in the North Thames drainage in 1936, 1946, 1953 and 1965 (ROM records). Specimens in the ROM collection, of species such as Hybognathus hankinsoni, Nocomis species, and Rhinichthys atrat- ulus which could have been confused with Campo- stoma, were examined but no Central Stonerollers were found suggesting that the North Thames popu- lation resulted from a recent introduction followed by a rapid increase in abundance. It is not known whether the two small specimens captured at the confluence of Frenchman’s Creek and the Niagara River in 1977 represented strays from nearby American creeks or an established pop- ulation. It is unlikely that the Central Stoneroller is established there now because surveys since 1977 have been unable to detect it. Sampling at that site conducted in 1979 by Parker and McKee (1980*), the OMNR (1979 surveys), and ROM in 1996 and 1997 have failed to capture any specimens. Habitat Although a few specimens have been captured in large rivers with little current, the Central Stoneroller thrives in small rivers and creeks of moderate gradi- ent (0.5-41 m/km) where habitat consists of riffles and adjacent pools with rocky substrate. McKee and Parker (1982) indicated that it was found in North Thames streams with gradients of 1.0—3.7 m/km but was absent in the Thames River system where the gradient decreased abruptly to 0.5 m/km. In Little Rouge Creek, where the Central Stoneroller has been captured from its mouth to near its headwaters, aver- age gradient is approximately 5 m/km. Average gra- dients vary from 2 to 6 m/km in Essex County streams. Lennon and Parker (1960) noted that the species was not present when gradients increased beyond 41m/km. Kraatz (1923) recorded a few spec- imens from lakes and it has also been recorded from one lake in Ontario. Mundahl and Ingersoll (1989) noted that popula- tion densities of Campostoma anomalum in a small Ohio stream were found to be higher in pools and 2001 HOLM AND CROSSMAN: UPDATED STATUS OF THE CENTRAL STONEROLLER 163 TABLE 2. New locations in Canada for Campostoma anomalum since previous status report (McAllister 1987). Location Year Mill Creek in Kingsville 1978 Creek in village of Bogies Beach 1980 Rouge River, Metro Toronto 1984 Deer Creek, Norfolk County 1984 North Sydenham River 1984 Kaladar area 1986 Lower Grand River 1986 Talbot River, Lake Simcoe 1987 Scugog River, Sturgeon Lake 1988 Duffins Creek, Metro Toronto 1989 Thames River at Tate’s Bridge 1989 Wye River 1990 Humber River, Metro Toronto 1992 St. Clair River 1993 Sally’s Creek, Woodstock 1993 Upper Maitland River drainage os Carroll Creek, upper Grand River 1995 Tay River tributary, SW of Perth 1995 Upper Grand River near Luther Lake 1996 St. Clair River, Baby’s Point 1996 Wigle Creek, 5 km NW of Kingsville 1996 Cedar Creek, 8 km NW of Kingsville 1996 Stothart Creek Haliburton County 1996 Silver Creek, 13 km S of Aylmer 1996 Wilson Creek, Lake Simcoe 1996 Pigeon River, 5 km SW of Omemee 1996 +Via T. Warren Sir Sanford Fleming College (SSFC) riffles with areas of open canopy, probably due to the higher primary productivity and therefore increased food supply. Campostoma anomalum appears to be a hardy species that can tolerate a wide range of environmen- tal conditions. It is frequently found in urban or agri- cultural locations with an assemblage of widespread and common species which are tolerant of high lev- els of turbidity, contaminants and extreme fluctua- tions in flow. The species has been described as being both “intolerant of silt” (Smith 1979) and “‘tol- erant of turbid, silty waters” (Burr 1980). McAllister (1987) documented that the species can tolerate oxy- gen levels as low as 3 mg/l and fluctuating turbidity levels in the North Thames River. Descriptions of bottom type for 42 ROM records indicate that the Central Stoneroller is found primari- ly in creeks and rivers that have some rocky compo- nents (80%) such as boulder, rubble, or gravel and less frequently (20%) over bottoms consisting of only one or more of clay, silt, sand, detritus, and muck. In 20 of the records that had a comment on vegetation 14 (70%) had some sort of aquatic vege- tation (usually submerged) and six recorded “NONE”. Of 37 records with notes on water current, only two of the streams (6%) had no flow, the remainder had at least a slight current. Nine (36%) of 25 locations had turbid water (secchi less than or Source of Record ROM 35374, OMNR: Ward and Patterson ROM 41606, OMNR: Maronets and Groom ROM Accession 4749: Holm and Boehm ROM 57374, OMNR: Colonello ROM 46164 Boehm, MacCulloch, Holm NMC 86-0206, NMC89-0206, NMC89-116 T. Warren, SSFC ROM 57984, OMNR: Shackleton, Ross, Galamb T. Warren, SSFC ROM 56735 U of Toronto: Wichert and Regier ROM 56950 ROM fieldwork: Holm et al. ROM 59416 Boehm ROM 64776 RSMI Consultants: Boyd ROM 67799 OMNR: Long and Masterson ROM Acc. 6087 OMNR Aylmer: Schraeder MNR: Malhiot 1996* ROM 70160 U of Waterloo: Fitzgerald ROM 69106 Beak Consultants: Meisenheimer ROM 70501-70511: Grand River Conservation Authority ROM 70496, ROM: Holm, Ramshaw, Guppy ROM 70396, ROM: Holm, Ramshaw, Guppy ROM 70393, ROM: Holm, Ramshaw, Guppy M. Brohm, SSFC+ G. Colgan, SSFC7 K. Good, SSFC+ A. Gibson, SSFC+ equal to 0.5m). Stream widths ranged from 3 to 60 m G23: n= 15): A study (OMNR), conducted on municipal drains in the drainages of the North Thames and Maitland rivers, found Campostoma anomalum in streams with widths of 0.7—3.5 m (k=2.1), depth of water 6—30cm (x=15.4), water flows of 19-757 litres-minute"! (k= 185), and temperatures up to 25°C (x=17.4). Bottoms consisted primarily of gravel (x = 29%), rubble (X=27%), boulder (X=18%) and sand (x=13%) with minor components of silt (x =7%), marl (kx=3%) and rarely clay (kx=1%) (Mal- hiot 1996*). Habitat data was collected for the Upper Grand River collections in late August 1996 by the Grand River Conservation Authority (ROM Accession 6456). Central Stonerollers were found in streams with an average width of 1.7—24m (x=8 m) and aver- age depths of 12-42 cm (x=24). Water temperatures varied from 18 to 23°C averaging 3°C lower than air temperature. Dissolved oxygen was 7.4—9.1 mg-I! (x = 8.2), pH was 8-9 (X= 8.2) and conductivity was 230-500 up mhas (x= 366). Aquatic macrophytes were usually present at the sites. The greatest number of Central Stonerollers (81) was captured at a site where the land use and terrestrial vegetation was described as follows: “cattle grazing — cattle access to stream — no buffer — severe erosion along banks”. Many of 164 the other sites where Central Stonerollers were cap- tured, were developed for residential and agricultural uses with some natural shoreline, meadow or woods. Bottom was primarily rocky: bedrock (0O-40%, x = 4.5), boulder (O—20%, k= 6.7), rubble (O-75%, x = 37.3), gravel (1-40%, x= 17), sand (3-20%, x = 16.4), silt (0-10%, x= 4.5), clay (0-10%, x= 0.9), muck (O-40%, x= 6.0) and detritus (0-60%, x= 6.8). A comparison of the 13 sites where Campostoma anomalum occurred with 14 sites where it was not captured in the upper Grand River indicated that the species was found more frequently on rocky, larger sized substrates (bedrock to gravel) than soft and smaller-sized substrates (sand, silt, clay, muck and detritus). The species also occurred at sites which were more open and had more riffles and pools. Small-sized gravel and nearby pools are necessary for spawning. Spawning areas in Missouri are often so shallow that the backs of males are exposed dur- ing pit construction (Pflieger 1975). Biology See McAllister (1987) for information on repro- ductive capability and age. Updated information on species movement and diet is summarized below. Species Movement The movements and migration of Canadian popu- lations of the Central Stoneroller have not been stud- ied. Lennon and Parker (1960) found relatively little movement occurring in the Great Smoky Mountains National Park population. Miller (1962) cited three studies which indicated that this species migrated to spawn but found that one New York population had localized spawning migrations confined to move- ments from pools to adjacent riffles. He suggested that the length and duration of migrations depend on habitat suitability. A study of the movements and home range of the species in a stream in Ohio in early fall indicated that the species had a small home range, averaging 35+14.1m (x+95% Cl), and moved a maximum of 135m (Mundahl and Ingersoll 1989). Rakocinski (1984) suggested that Campo- stoma anomalum pullum migrates upstream to spawn in a creek in Illinois. Mundahl and Ingersoll (1989) suggested that more examination is needed to deter- mine if the species exhibits a true spawning migra- tion, and if so, whether the fish return to their origi- nal home ranges following spawning. They also suggested that predators such as bass and large Creek Chub (Semotilus atromaculatus) will restrict movements of Campostoma and limit the home range size. Predators also influence emigration of stonerollers from pools (Matthews et al. 1987). Diet The Central Stoneroller is known for being highly herbivorous, feeding almost exclusively on periphy- ton. Inorganic matter such as sand and silt can account for up to 75% of the gut contents of adults. THE CANADIAN FIELD-NATURALIST Vol. 115 It is not known if the inorganic material is ingested accidentally while scraping attached algae or inten- tionally to help grind algal cells like stones in a bird’s gizzard. Small percentages of macroinverte- brates (primarily chironomids) are also reported in dietary studies of the Central Stoneroller. Johnson and Dropkin (1992) found that this species would feed on larval American shad (Alosa sapidissima) in a situation where the shad were abnormally high in abundance. Kraatz (1923) found that diatoms, partic- ularly Navicula and Nitzschia, were prominent in the diet of Central Stonerollers in Ohio. Other prominent food items were filamentous green algae such as Spirogyra. Small young had consumed not only a high percentage of diatoms but significant portions of zooplankton such as rotifers, copepods, and clado- cera. They also consumed less inorganic matter (15-25%), which may be a reflection of a lesser amount of bottom feeding. In a small stream in the Ozarks in northeastern Oklahoma, Campostoma ano- malum showed little seasonal variation in diet. Gut contents consisted of 38% non-filamentous algae, 28% sand, 21% detritus, 7% filamentous algae, 3% vascular plants, and 3% macroinvertebrates of which most were chironomids (McNeeley 1987). Fowler and Taber (1985) found that Campostoma anomalum in Missouri fed during daylight consuming 27% of their body weight per day. Foreguts contained detri- tus, diatoms (preferring non-motile over motile), inorganic matter and green and blue-green algae. McKee and Parker (1982) found that specimens from the North Thames drainage, Ontario, had 23% filamentous green algae with most of the remainder being inorganic material. Limiting Factors Potential limiting factors include availability of periphyton, excessive siltation and pollutants (McAllister 1987). Impoundment and channelization of streams has reduced its abundance in Virginia (Jenkins and Burkhead 1994) and Ontario (McAI- lister 1987; McKee and Parker 1982). The species appears to be moderately tolerant to siltation and tur- bidity but the species is generally found most often on rocky bottoms where the current has swept away silt. It seems to actively avoid clay bottoms which may explain why it has not shown up in Frenchman’s Creek since 1977. It will utilize sections of streams which dry up later in the summer in Illinois (Rakocinski 1984). Cold temperatures are probably not limiting because it has spawned in water as low as 11°C (Jenkins and Burkhead 1994). The commonly encountered black spot parasite (Uvulifer sp.) has been shown to adversely affect some species of fishes through increased mortality, reduction in weight gain, low condition coefficient and slow growth. McKee and Parker (1982) noted that most of the specimens they collected were infected with black spot, sometimes heavily. 2001 However, this is likely not a limiting factor for the species as Baker and Bulow (1985) demonstrated that the condition of Central Stonerollers was not adversely affected by parasite density. Even heavily infected individuals appeared healthy in a central Tennessee stream. Besides piscivorous fishes such as bass, the Central Stoneroller is preyed upon by avian predators such as the Belted Kingfisher (Ceryle alcyon) and the Great Blue Heron (Ardea herodias). The Belted King- fisher preferred Campostoma anomalum over other more common smaller minnows in one New York study (cited in Matthews et al. 1986). Central Stone- rollers would be particularly susceptible to terrestrial predators during the spawning season when they occur in shallow riffles. Both the Belted Kingfisher and the Great Blue Heron are common summer resi- dents in Ontario (James 1991). The Mink (Mustela vison), common in Ontario (Dobbyn 1994), is also a potential predator (Hamilton 1959). Special Significance of the Species Much has been written recently on the important ecological role played by Campostoma anomalum in the aquatic ecosystem. Sometimes referred to as a “stream cow’, it uses a resource (algae, periphyton) not commonly exploited by other species of Canadian fishes, converting the primary productivity of the stream into a form that can be used by piscivorous fishes and terrestrial predators such as kingfishers, herons and mink. Where abundant, it has a significant effect on the standing crop of algae in streams and the composition of the algal community. In field experi- ments in Oklahoma, after Campostoma anomalum was introduced into streams, the dominant algal over- story such as filamentous green algae (Spirogyra) was replaced with a faster-growing community of low-growth forms of diatoms (Navicula, Achnanthes, Gomphonema, Cymbella, Tabellaria, Fragilaria, Synedra, Cocconeis, Melosira) and blue-green algae (mostly Oscillatoria) (Gelwick and Matthews 1992). Invertebrate and bacteria populations are also affected. Through faecal production and mechanical fragmentation, the Central Stoneroller alters the size of the particulate organic matter in the sediment cre- ating a detrital food resource that is preferred by cer- tain types of invertebrates and bacteria. On the other hand, by reducing the algal overstory other types of invertebrates lose their cover and food source. In Oklahoma, non-tanypodine chironomids such as Dicrotendipes and Stictochironemus increased in grazed areas whereas amphipods (Hyalella azteca) decreased as the result of stoneroller grazing (Gelwick and Matthews 1992). Vaughn et al. (1993) demonstrated negative impacts of Campostoma graz- ing on crayfish (Orconectes) and beneficial effects on snails (Physella). Growth of crayfish was reduced by the removal of their food source, the dominant algal HOLM AND CROSSMAN: UPDATED STATUS OF THE CENTRAL STONEROLLER 165 overstory, and by prevention of day-time feeding of the crayfish. Snails grew larger and delayed repro- duction because grazing by Campostoma increased food available to the snails by making low growth forms of periphyton more accessible and increasing the bacteria in the sediment. Bacteria play a key role in nutrient cycling, con- version of non-living organic carbon into biomass and as a food source for invertebrates (Findley et al. 1986). By altering sediment particle size and remoy- ing the algal overstory, Campostoma anomalum also affect bacteria populations both positively and nega- tively. Bacteria associated with the healthy algal overstory declined in grazed areas (Gelwick and Matthews 1992). Gardner (1993) conducted experi- ments on artificial streams and found that grazed streams resulted in a higher proportion of fine benth- ic particulate organic matter (41—156 um) than the proportion in ungrazed streams. He suggested that this particle size alteration would result in increased bacterial activity and biomass in the sediment. Limited evidence suggests that high densities of Campostoma lowers diversity of small fishes, but Campostoma appears to improve feeding opportuni- ties of some species by displacing invertebrates from the bottom making them more available to insecti- vores (Matthews et al. 1987). Eddy and Underhill (1974) attributed little value to this species as a bait minnow and as a forage fish in Minnesota. However, it is important as a bait fish and sometimes preferred over trout as a food fish by fish- ermen in Virginia and Tennessee (Jenkins and Burkhead 1994). In Ontario, Litvak and Mandrak (1993) found it ranked 18 of 28 in frequency of occurrence in Toronto bait dealers’ tanks, just ahead of the Emerald Shiner. Coad (1995) also reported that the species is excellent bait and is sold by bait dealers in Ottawa which get them from a source in south cen- tral Ontario. Lennon and Parker (1960) showed that in Smoky Mountains National Park the Central Stoneroller can reduce numbers of Rainbow Trout (Oncorhynchus mykiss). It spawns in spring just after the trout and its activities destroy the redds before the trout eggs hatch. In Ontario streams, this would have an impact on the introduced Rainbow Trout but would not affect fall spawners such as the introduced Brown Trout (Salmo trutta), and Pacific salmons (Oncor- hynchus spp.) and the native Brook Trout (Salvelinus fontinalis). Evaluation The Central Stoneroller is undergoing a range expansion and should no longer be considered in jeopardy in Canada. The number of new records has more than quadrupled since the 36 records docu- mented by Parker and McKee (1980*). The species appears to have become established in many waters 166 of Ontario as a result of a combination of bait-bucket introductions and natural dispersal. Acknowledgments Financial support for fieldwork and preparation of this report has been provided by the Royal Ontario Museum, the World Wildlife Fund (Canada), and Human Resources Development Canada. ESRI donated GIS software. William Ramshaw prepared the maps, and along with Randy Guppy, assisted in the field. The following people provided data or information for this report: Brooks M. Burr, Alan Dextrase, Sylvie Laframboise, Mike Malhiot, Lori Richardson, Hal Schraeder, Ken Stewart, and Ted Warren. The manuscript was improved by comments from the reviewers (A. Dextrase, J.S. Nelson, and W.B. Scott) and M. Ciuk. This paper is Contribution number 91 of the Centre for Biodiversity and Con- servation Biology, Royal Ontario Museum. Documents Cited [marked * in text] Malhiot, M. 1996. Unpublished data on a 1995 Municipal Drain Survey. Ontario Ministry of Natural Resources, Wingham Area Office. Nature Conservancy. 1997. Global, national and S-ranks compiled by the Eastern Regional Office of the Nature Conservancy, Boston. Parker, B., and P. McKee. 1980. Rare, threatened, and endangered fish species of southern Ontario: status reports. Report submitted to Department of Supply and Services, Department of Fisheries and Oceans and National Museum of Natural Sciences by Beak Consul- tants Limited. 238 pages. Literature Cited Baker, S. C., and F. J. Bulow. 1985. Effects of black-spot disease on the condition of stonerollers Campostoma anomalum. American Midland Naturalist 114: 198-199. Baxter, G. T., and M.D. Stone. 1995. Fishes of Wyom- ing. Wyoming Game and Fish Department. Boschung, H. T. 1992. Catalogue of freshwater and marine fishes of Alabama. Alabama Museum of Natural History Bulletin 14: 1-266. Burr, B. M. 1976. Distribution and taxonomic status of the stoneroller, Campostoma anomalum, in Illinois. Chicago Academy of Sciences, Natural History Miscel- lanea, 194: 1-8. Burr, B. M. 1980. Campostoma anomalum (Rafinesque), Stoneroller. Pages 143-144 in Atlas of North American Freshwater Fishes. Edited by D.S. Lee, C.R. Gilbert, C.H. Hocutt, R. E. Jenkins, D. E. McAllister, and J. R. Stauffer. North Carolina State Museum Natural History, Publication Number 1980-12. Buynak, G.L., and H. W. Mohr. 1980. Larval develop- ment of stoneroller, cutlips minnow, and river chub with diagnostic keys, including four additional cyprinids. Progressive Fish-Culturist 42: 127-135. Coad, B. W. 1995. Encyclopedia of Canadian Fishes. Canadian Museum of Nature and Canadian Sportfishing Productions Inc. Cooper, E. L. 1983. Fishes of Pennsylvania and the north- eastern United States. The Pennsylvania State University Press. THE CANADIAN FIELD-NATURALIST Vol. 115 Dobbyn, J. 1994. Atlas of the mammals of Ontario. Fed- eration of Ontario Naturalists, Don Mills, Ontario. Eddy, S., and J. C. Underhill. 1974. Northern fishes with special reference to the upper Mississippi valley. Uni- versity of Minnesota Press. Minneapolis. Findley, S., J. L. Meyer, and R. Risley. 1986. Benthic bacterial biomass and production in two blackwater rivers. Canadian Journal of Fisheries and Aquatic Sciences 43: 1271-1276. Fowler, J. F., and C. A. Taber. 1985. Food habits and feeding periodicity in two sympatric stonerollers (Cypri- nidae). American Midland Naturalist 113: 217-224. Gardner, T. J. 1993. Grazing and the distribution of sedi- ment particle size in artificial stream systems. Hydro- biologia 252: 127-132. Gelwick, F. P., and W. J. Matthews. 1992. Effects of an algivorous minnow on temperate stream ecosystem properties. Ecology (Tempe) 73: 1630-1645. Greeley, J. R. 1929. VI. Fishes of the Erie-Niagara water- shed. Pages 150-179 in A biological survey of the Erie- Niagara system. Supplemental to Eighteenth Annual Report, 1928 of the State of New York Conservation Department, Albany. Gruchy, C. G., R.H. Bowen, and I.M. Gruchy. 1973. First records of the stoneroller (Campostoma anomalum) and the blackstripe topminnow (Fundulus notatus) from Canada. Journal of the Fisheries Research Board of Can- ada 30: 683-684. Hamilton, W. J. 1959. Foods of mink in New York. New York Fish and Game Journal 6: 77-85. James, R. D. 1991. Annotated checklist of the birds of Ontario. Royal Ontario Museum. Jenkins, R. E., and N. M. Burkhead. 1994. Freshwater fishes of Virginia. American Fisheries Society, Beth- esda, Maryland. Johnson, J. E. 1987. Protected fishes of United States and Canada. American Fisheries Society, Bethesda, Mary- land. Johnson, J. H., and D.S. Dropkin. 1992. Piscivory by the central stoneroller, Campostoma anomalum. Journal of the Pennsylvania Academy of Science 66: 90-91. Kraatz, W. C. 1923. A study of the food of the minnow, Campostoma anomalum. Ohio Journal of Science 23: 265-283. Lennon, R. E., and P. S. Parker. 1960. The stoneroller, Campostoma anemalum (Rafinesque), in Great Smoky Mountains National Park. Transactions of the American Fisheries Society 89: 263-270. Litvak, M., and N. E. Mandrak. 1993. Ecology of fresh- water baitfish use in Canada and the United States. Fish- eries 18(12): 6-13. Mandrak, N. E., and E. J. Crossman. 1992. A checklist of Ontario freshwater fishes, annotated with distribution maps. Royal Ontario Museum. Matthews, W. J., A. J. Stewart, and M.E. Power. 1987. Grazing fishes as components of North American stream ecosystems: effects of Campostoma anomalum. Pages 128-135 in Community and evolutionary ecology of North American stream fishes. Edited by W. J. Matthews and D.C. Heins. University of Oklahoma Press, Norman and London. Matthews, W. J., M.E. Power, and A. J. Stewart. 1986. Depth distribution of Campostoma grazing scars in an Ozark stream. Environmental Biology of Fishes 17: 291-297. 2001 McAllister, D. E. 1987. Status of the central stoneroller, Campostoma anomalum, in Canada. Canadian Field- Naturalist 101: 213-218. McKee, P. M., and B. J. Parker. 1982. The distribution, biology, and status of the fishes Campostoma anomalum, Clinostomus elongatus, Notropis photogenis (Cyprinidae), and Fundulus notatus (Cyprinodontidae) in Canada. Canadian Journal of Zoology 60: 1347-1358. McNeely, D.L. 1987. Niche relations within an Ozark stream cyprinid assemblage. Environmental Biology of Fishes 18: 195-208. Miller, R. H. 1962. Reproductive behavior of the stone- roller minnow, Campostoma anomalum pullum. Copeia 1962: 407-417. Mundahl, N. D., and C. G. Ingersoll. 1989. Home range, movements, and density of the central stoneroller, Campostoma anomalum, in a small Ohio stream. Envi- ronmental Biology of Fishes 24: 307-311. Page, L.M., and B.M. Burr. 1991. A field guide to freshwater fishes. Houghton Mifflin Company, Boston. Pflieger, W.L. 1975. The fishes of Missouri. Missouri Department of Conservation. HOLM AND CROSSMAN: UPDATED STATUS OF THE CENTRAL STONEROLLER 167 Rakocinski, C. F. 1984. Aspects of reproductive isolation between Campostoma oligolepis and Campostoma anomalum pullum (Cypriniformes: Cyprinidae) in north- ern Illinois. American Midland Naturalist 112: 138-145. Scott, W. B., and E. J. Crossman. 1973. Freshwater fish- es of Canada. Fisheries Research Board of Canada Bulletin 184. Smith, P. W. 1979. The Fishes of Illinois. University of Illinois Press, Urbana, Illinois. Trautman, M.B. 1981. The fishes of Ohio. Ohio State University Press. Columbus, Ohio. Underhill, J.C. 1957. The distribution of Minnesota min- nows and darters in relation to Pleistocene glaciation. Occasional papers 7, Minnesota Museum of Natural History. University of Minnesota Press, Minneapolis. 45 pages. Vaughn, C.C., F. P. Gelwick, and W. J. Matthews. 1993. Effects of algivorous minnows on production of grazing stream invertebrates. Oikos 66: 119-128. Accepted 29 May 2000 Notes Common Raven, Corvus corax, Observed Taking an Egg from a Common Loon, Gavia immer, Nest ROBERT ALVO! and PETER J. BLANCHER2 1GAVIA Biological Services, 58, rue des Parulines, Hull, Québec JOA 1Z2, Canada 2Canadian Wildlife Service, National Wildlife Research Centre, 100 Gamelin Blvd, Hull, Québec K1A OH3, Canada Alvo, Robert, and Peter J. Blancher. 2001. Common Raven, Corvus corax, observed taking an egg from a Common Loon, Gavia immer nest. Canadian Field-Naturalist 115(1): 168-169. We report an incident of a Common Raven (Corvus corax) taking an egg from an active Common Loon (Gavia immer) nest, and two other possible similar events. The fact that ravens can take whole eggs, leaving no trace of eggs on the nest, could explain a considerable proportion of the missing loon eggs that researchers have documented, particularly on lakes with few humans. Key Words: Common Loon, Gavia immer, Common Raven, Corvus corax, egg, nest, predation. Predation by Common Ravens (Corvus corax) on eggs can be a serious problem for waterfowl man- agement in some places (Braun et al. 1978). Ravens can also remove considerable numbers of eggs from the nests of colonial aquatic birds such as. gulls, mur- res and cormorants, concealing for future use what they cannot eat at once (Bent 1946). Montevecchi (1978) reported ravens frequently supplanting nest- ing Black-legged Kittiwakes (Rissa tridactyla) by suddenly approaching them closely while calling loudly, or by physically attacking them. He also watched a raven displace a kittiwake from a nest by pulling a tuft of dried grass from sod on a cliff face and dropping it on the kittiwake — the raven then jumped down and rummaged through the empty nest. Loons generally nest within 1-2 m of a freshwater pond, tarn or lake. Their eggs are susceptible to pre- dation from mammalian and avian predators. Known or suspected avian egg predators of the five loon species include Common Ravens, American Crows (Corvus brachyrhynchos), Hooded Crows (C. corone), Glaucous Gulls (Larus hyperboreus), Herring Gulls (L. argentatus), Great Black-backed Gulls (L. marinus), Lesser Black-backed Gulls (L. fuscus), Parasitic Jaegers (Stercorarius parasiticus), Long-tailed Jaegers (S. longicaudus), Sandhill Cranes (Grus canadensis), and Snowy Owls (Nyctea scandiaca). McIntyre (1977) and Titus and VanDruff (1981) have reported American Crows eat- ing and/or puncturing loon eggs. No one has report- ed seeing Common Ravens taking loon eggs. However, Common Loon eggshell fragments were found in a raven’s nest (Titus and VanDruff 1981). Here we report an incident of a raven taking a Common Loon egg from an active nest, and two other possible similar events. Each incident occurred in a different loon territory in or adjacent to Killarney Provincial Park (46°06'N, 81°13’W), | southwest of Sudbury, Ontario, in 1981 and 1982 during a study of the effects of acid precipitation on avian breeding success. In the first territory, in a stump-filled bay at the east end of Johnnie Lake (46°05’45” N, 81°12'45” W), PB discovered a loon nest with one egg 14 May 1981. The nest was on a mud platform in the centre of this large area of water, stumps and Myrica gale shrubs. On 23 May and 27 May, a loon was incubat- ing one egg, and a loon was incubating again on 30 May. On 3 June, as a loon incubated, a raven flew directly toward the nest from the direction of the main portion of the lake. The loon slipped off the nest without any resistance or sound as the raven approached, landed, picked up the egg whole and flew back toward the main part of the lake. Meanwhile, the loon swam several hundred metres toward the main part of the lake and joined a second loon, after which one gave a few long wails, then both flew away. No loons were observed on this ter- ritory on 7 June or 10 June, but a pair was seen on 14 June. A loon was incubating a new nest with two eggs in the same stump-filled bay on 28 June. A loon was incubating on 12 July and 15 July, and on 19 July a pair with two chicks was seen in the bay. In the second territory, on 29 May a loon was incubating in a stump-filled bay at the west end of Johnnie Lake (46°05'00” N, 81°14'45”" W), approxi- mately 3 km southwest of the first territory. The 168 2001 number of eggs was not determined. The nest had disappeared by 5 June, but a loon pair was seen at the entrance to the bay. One loon was seen very close to the old nest site on 9 June. On 13 June, a raven flew up from a small grass island about 100 m west of the first nest with what appeared to be a loon egg. The loon pair nearby in the water was silent. Upon inspection of the island, PB found an empty loon nest in good condition where the raven had been. On 17 June, a loon was incubating in a new nest at the site of the initial nest. Incubation of an unknown number of eggs was observed on 21 June, 25 June, 29 June, 2 July, 6 July and 10 July, despite the presence of ravens in the bay on 2 July and 10 July. The nest was empty on 15 July, and contained no eggshells. However, a single loon chick, young enough to be from this nest, was seen with its par- ents on 20 July, 23 July, 30 July and 9 August near the entrance to the bay. The third territory was in the north bay of Ruth- Roy Lake (46°05'30" N, 81°15’00" W), a fishless, acidic (pH 4.5) lake. A loon pair was seen on 15 May 1982 near what was apparently a nest in con- struction on the north side of a small island. The nest contained no eggs on 25 May, but a single loon was seen close by. A pair was seen near the nest on 29 May, and incubation was observed on 5 and 9 June. The number of eggs was not determined. On 13 and 17 June, the nest was empty and no loons were observed. On 21 June, on the west side of the lake’s north section (ca. 400 m from the first nest) a raven flew up from the back of a stump with something large in its bill, which could have been a loon egg, and flew directly away from PB while a pair of Eastern Kingbirds (Tyrannus tyrannus) attacked it (a nearby kingbird nest had lost 3 of 4 eggs since 17 June). A nearby loon uttered low, quiet wail-like “moaning” calls. Inspection of the stump from which the raven had flown revealed a loon nest with one intact egg. On 25 June, the loon and kingbird nests were both empty and no loons were seen. The lake was visited nine times from 25 June to 27 July, and the only loon sighting was on 15 July, when a pair was seen in the lake centre before flying off. It is clear that Common Ravens are loon nest predators, and it is now known that ravens may dis- place loons from their nests to take an egg, as they do to other birds. However, it is not clear to what extent this activity decreases loon breeding success, because the loons may renest successfully. In Wyoming, raven home ranges varied from 6.7 to 10.9 km?, which means that an individual could trav- el up to 3 km or more (Roy and Bombardier 1996). All three loon territories were within 2 km of the site NOTES 169 of a single raven nest in at least one year of the study, so it is possible that a single raven or raven family was responsible for all incidents reported here. The fact that ravens may remove whole loon eggs from nests is a critical point to bear in mind during loon breeding surveys. We suspect that many nests found empty after having previously been found with eggs in the Sudbury area (RA, personal obser- vations) were also the work of ravens. We consider the observations presented here to be very fortuitous because it is rare to observe such pre- dation events. During a three-year study of loon breeding success, RA never observed any predation events on loon nests, but he found numerous depre- dated nests (Alvo et al. 1988). Acknowledgments Field work in the Killarney area was financially supported by the Ontario Ministry of Natural Resources and the Canadian Wildlife Service. We thank Josée Nesdoly, A. J. Erskine and Jack Barr for reviewing the manuscript. Literature Cited Alvo, R., D. J. T. Hussell, and M. Berrill. 1988. The breeding success of common loons (Gavia immer) in relation to alkalinity and other lake characteristics in Ontario. Canadian Journal of Zoology 66: 746-752. Bent, A. C. 1946. Life histories of North American jays, crows and titmice. United States National Museum Bulletin Number 191, Part 1, Reprinted by Dover Publi- cations, Inc. 1964, New York. 214 pages + photos. Braun, C. E., K. W. Harmon, J. J. Jackson, and C. D. Littlefield. 1978. Management of national wildlife refuges in the United States: its impact on birds. Wilson Bulletin 90: 309-321. McIntyre, J. 1977. The Common Loon (Part 2): Identi- fication of potential predators on Common Loon nests. Loon 49: 96-99. Montevecchi, W. A. 1978. Corvids using objects to dis- place gulls from nests. Condor 80: 349. Roy, L. and M. Bombardier. 1996. Common Raven. Pages 730-733 in The Breeding birds of Québec : Atlas of the Breeding Birds of Southern Québec. Edited by J. Gauthier and Y. Aubry. Association québécoise des groupes d’ornithologues, Province of Québec Society for the Protection of Birds, Canadian Wildlife Service, Environment Canada, Québec Region, Montréal. 1302 pages. Titus, J. R., and L. W. VanDruff. 1981. Response of the Common Loon to recreational pressure in the Boundary Waters Canoe Area, northeastern Minnesota. Wildlife Monographs, Number 79. 59 pages. Received 31 December 1999 Accepted 5 May 2000 170 THE CANADIAN FIELD-NATURALIST Vol. 115 Two Cases of Infanticide in a Red Fox, ee vulpes, Family in Southern Ontario VALERIA VERGARA Trent University, Biology Department, Peterborough, Ontario K9H 7B8, Canada Vergara, Valeria. 2000. Two cases of infanticide in a Red Fox, Vulpes vulpes, family in southern Ontario. Canadian Field- Naturalist 115(1): 170-173. Two cases of directly observed infanticide in a Red Fox (Vulpes vulpes) family in southern Ontario are described. These observations were made in the context of a natal den study that was aimed at comparing male and female parental roles in wild Red Foxes. Key Words: Red Fox, Vulpes vulpes, infanticide, parental care, intraspecific predation, offspring mortality, Ontario. Infanticide is commonly defined as the killing of dependent young by a conspecific (e.g., Fosey 1984; Hayssen 1984; Hrdy 1979; vom Saal 1984). In- fanticide in mammals may be followed by cannibal- ism (Brooks 1984). This process of both killing and eating an individual of the same species is known as intraspecific predation (Polis 1981). In mammals, intraspecific predation involves adults preying on immature animals and cubs (1.e., infanticide) more often than adults preying on other adults (Polis 1981). Many descriptions of infanticide in mammals have been published (e.g., Brown Bears, Ursus actos: Olson 1993; Alpine Marmots, Marmota marmota: Coulon et al. 1995; Hanuman Langurs, Presbytis entellus: Borries 1997; Ground Squirrels, Spermo- philus beecheyi: Trulio 1996; Hippopotamus, Hyppo- potamus amphibius: Lewison 1998). In canid species, infanticide has occasionally been reported. Camenzind (1978) presents circumstantial evidence of infanticide in Coyotes, Canis latrans, by trespass- ing conspecifics. Maternal infanticide has been stud- ied in captive Silver Foxes, Vulpes vulpes (Braastad and Bakken 1993). However, I know of no specific example of infanticide in wild Red Foxes (Vulpes vulpes). 1 describe here two cases of directly observed infanticide in a Red Fox family, witnessed during a natal den study of eight Red Fox families aimed at comparing parental roles in male and female Red Foxes in southern Ontario (Vergara 1996). The fox family consisted of a large litter of nine kits, an adult female, and an adult male. I observed this family for a total of 131 hours, from 6 May, 1995 (nine kits, 6 weeks old) to 6 June, 1995 (five kits, 10 1/2 weeks old). Two dens, 18 m apart, both with multiple entrances, were in use simultaneously by this family. The whole den area (encompassing both dens with their multiple entrances) was located in the middle of a hay field, 50 m from a farm house, and was easily observed from the kitchen window (location: 20 km to the north-west of Bobcaygeon, Verulam Township, Victoria County: 44°33’, 78°40’). The driveway to the farmhouse separated the hay field from a cow pasture. On 8 May 1995, the adult female arrived at 1942 h carrying a small-size rodent that she fed to one of the kits. She nursed the kits for 2 minutes, then disen- gaged herself and trotted 25 m away from the den mound. The kits became very active. Some followed the female and began exploring the pasture, while others were involved in energetic play-chasing over the field where both dens were located. At 1950 h something frightened the female, who left the den site at a fast run, and was soon out of sight. Some kits went back to the den mound, and others remained rel- atively far from it (20-30 m), exploring and playing. At 1953 h an adult Red Fox of unknown sex and age, never recorded before, came trotting toward the den. It adopted an aggressive posture when it passed by the den with its back arched and head and neck low. It snarled at one kit, which crouched down flattening its ears and then went underground. The stranger con- tinued trotting towards the driveway where a single kit was sniffing around. The stranger immediately resumed its threatening posture, stalked the kit very briefly, dashed towards it, grabbed it by the neck and shook the kit-vigorously. It then scurried away carry- ing the limp body of the kit in its mouth. The remain- der of the litter stayed inside the den until dark. Only 3 days later, on 11 May 1995, I recorded a very similar event. At 2100 h the light had faded almost entirely. I used the ambient light image inten- sifier scope, which allowed to view the immediate den area since this was also dimly illuminated by the light coming from the farm house. The kits (now only seven, one lost to infanticide and one lost to unknown causes) were alone, outside the den, not too active, and not ranging from the den mound. At 2110 h an adult fox of unknown sex and age came to the den. The kits crouched down instead of running out to greet the fox, which suggested that the animal was a stranger. This was corroborated by the fox’s strange behaviour: it was running around with a very aggressive body posture (i.e. its back arched, its head and neck lowered). It grabbed a kit by the neck and shook it quite vigorously for a few seconds, as the remaining kits went in the den. It dropped the body 2001 of the kit when the adult female arrived and chased the stranger away. When I ended my observation session at 2120 h, no Fox was visible outside the den. The next day I counted only 6 kits, one less than the previous day, which suggests that the stranger had indeed killed the kit before the female chased it away. I surveyed the den area but did not find the carcass of the kit. The two cases of infanticide described above occurred in a litter that had a low male visitation rate, and was left unattended 80% of the time. During the 4 weeks that I observed this family, I only saw the adult male 8 times, compared to 42 visits from the vixen. The male brought food but interacted little with the kits: he never groomed them, played, or laid down with them, behaviors that were observed in the males of my other study families (see Vergara 1996 for details). His visits lasted between 4 and 8 min- utes, about half to two thirds the length of the visits by the males in my other study families (Vergara 1996). This low male den attendance may be consequen- tial, particularly when we take into account that vigi- lant males spent most of their time at the den site (78 % on average) in the absence of the female. This is referred to as “pupsitting” or “den guarding” in other studies (e.g., Moehlman 1983, for Silverbacked and Golden Jackals, Canis mesomelas and Canis aureus; Malcom and Marten 1982, for African Wild Dogs, Lycaon pictus). Pupsitting may help to detect danger such as Humans, Coyotes or strange foxes. This grants the speculation the fewer and shorter male visitations to the den site in this particular family may have had some influence on the vulnerability of this litter to dangers such as infanticide, a possibility that needs testing, and invites further study. Activities away from the den may also be important forms of indirect care that would go unobserved in this den-centered study. Nonetheless, a male that is patrolling the boundaries of the territory may miss an intruder at the den site, especially considering the reported large size of fox farmland territories in Ontario (900 ha on average, Voigt and Macdonald 1984). The phenomenon of killing of young by conspe- cific strangers, parents and siblings has been widely discussed in the context of a number of hypotheses (for review, see Parmigiani and vom Saal 1994). Most of them agree on the fact that the fitness of the killer is benefited. One of the possible immediate benefits of infanticide may be a nutritional one (Liips and Roper 1990). In the present case, I do not know if the infanticidal Fox ate the kits that it killed (i.e. if it was a predatory act). However, this is not an unlikely possibility, as Red Foxes do eat conspecific carcasses. I observed fox kits feeding on their dead littermates in two of my study families (see Vergara 1996). NOTES 171 The significance of the observations reported here rests on the important role that infanticide could play in the dynamics of Red Fox populations. There are reports of intraspecific predation on immature ani- mals or cubs substantially reducing populations of various species of carnivore (see Polis 1981 for extensive review). For example, adult Arctic Foxes, Alopex lagopus, may kill and eat other adults or pups during periods of low food availability, and this may be a factor in Arctic Fox population regulation (Chesemore 1975). Den-site behavioral studies would help to determine if infanticide in Red Foxes is more than incidental, and the degree to which it affects reproductive success. Only then we could start evaluating hypotheses addressing the adaptive basis and the significance of this behavior. Acknowledgments This research was entirely funded by a Trent University Research Committee Grant to my super- visor Michael Berrill. I thank him for his assistance and encouragement in every stage of this study. The comments and suggestions of Shelley Pruss and three anonymous reviewers greatly improved this manuscript. Many thanks to the owners of the prop- erties where the dens were located for permission to work on their lands. I am grateful to Reka Anthony, Cam Collier, Heather Lee, and Lise Macgillivray for their help in the field. Literature Cited Braastad, B., and M. Bakken. 1993. Maternal infanticide and periparturient behaviour in farmed silver foxes Vulpes vulpes. Applied Animal Behaviour Science 36: 347-361. Brooks, R. J. 1984. Causes and consequences of infanti- cide in populations of rodents. Pages 331-348 in Infanticide: comparative and evolutionary perspectives. Edited by G. Hausfater and S. B. Hrdy. New York: Aldine. Borries, C. 1997. Infanticide in seasonally breeding mul- timale of Hanuman langurs (Presbytis entellus) in Ram- nagar (South Nepal). Behavioral Ecology and Socio- biology 41: 139-150. Camenzind, F. J. 1978. Behavioral ecology of coyotes (Canis latrans) on the National Elk Refuge, Jackson, Wyoming. Ph.D. thesis. University of Wyoming, Lara- mie. 97 pages. Chesemore, D. L. 1975. Ecology of the Arctic Fox (Alopex lagopus) in North America — A review. Pages 143-163 in The wild canids: their systematics, behavi- oural ecology and evolution. Edited by M. W. Fox. Behavioral Science Series, Van Nostrand Reinhold Company, New York, Cincinnati, Toronto, London, Melbourne. Coulon, J., L. Graziani, D. Allaine, M. C. Bel, and S. Pouderoux. 1995. Infanticide in the Alpine Marmot (Marmota marmota). Ethology, Ecology and Evolution 7: 191-194. Fosey, D. 1984. Infanticide in Mountain Gorillas (Gorilla gorilla beringei) with comparative notes on Chimpan- WZ zees. Pages 217-235 in Infanticide: comparative and evolutionary perspectives. Edited by G. Hausfater and S. B. Hrdy. New York: Aldine. Hayssen, V. D. 1984. Mammalian reproduction: con- straints on the evolution of infanticide. Pages 105-123 in Infanticide: comparative and evolutionary perspec- tives. Edited by G. Hausfater and S B. Hrdy. New York: Aldine. Hrdy, S. B. 1979. Infanticide among animals: a review, classification and examination of the implications for the reproductive strategies of females. Ethology and Sociobiology 1: 13-40. Lewison, R. 1998. Infanticide in the hippopotamus: evi- dence for polygynous ungulates. Ethology, Ecology and Evolution 10: 277-286. Lips, P., and T. J. Roper. 1990. Cannibalism in a female badger (Meles meles): infanticide or predation? Journal of Zoology, London 221: 314-315. Malcom, J. R., and K. Marten. 1982. Natural selection and the communal rearing of pups in African Wild Dogs (Lycaon pictus). Behavioral Ecology and Socio- biology 10: 1-13. Moehlman, P. D. 1983. Socioecology of silverbacked and golden jackals (Canis mesomelas and Canis aureus). Pages 423-453 in Recent Advances in the Study of Mammalian Behaviour. Edited by J. F. Eisenberg and D. G. Kleiman. Special Publication, American Society of Mammalogists 7. Olson, T. 1993. Infanticide in brown bears, Ursus actos, THE CANADIAN FIELD-NATURALIST Vol. 115 at Brooks River, Alaska. Canadian Field Naturalist 107: 92-94. Parmigiani, S., and F.S. vom Saal. Editors. 1994. In- fanticide and Parental Care. Proceedings of a workshop held at the International School of Ethology, Ettore Majorana Centre for Scientific Culture, Italy, June 1990. Harwood Academic Publishers. 496 pages. Polis, G. A. 1981. The evolution and dynamics of intraspecific predation. Annual Reviews of Ecology and Systematics 12: 225-251. Trulio, L. A. 1996. The functional significance of infanti- cide in a population of California ground squirrels (Spermophilus beecheyi). Behavioral Ecology and Soci- obiology 38: 97-103. Vergara, V. 1996. Comparison of parental roles in male and female red foxes (Vulpes vulpes) in southern Ontario, Canada. M.Sc. thesis. Trent University, Peter- borough, Ontario. 120 pages. Voigt, D. R., and D. W. Macdonald. 1984. Variation in the spatial and social behaviour of the red fox, Vulpes vulpes. Acta Zoologica Fennica 171: 261-265. vom Saal, F.S. 1984. Proximate and ultimate causes of infanticide and parental behavior in male house mice. Pages 401-424 in Infanticide: comparative and evolu- tionary perspectives. Edited by G. Hausfater and S. B. Hrdy. New York: Aldine. Received 27 September 1999 Accepted 12 May 2000 An Unusual Record of a White-tailed Deer, Odocoileus virginianus, in the Northwest Territories ALASDAIR M. VEITCH Department of Resources, Wildlife & Economic Development, Government of the Northwest Territories, P.O. Box 130, Norman Wells, Northwest Territories, XOE OVO, Canada. Veitch, Alasdair M. 2001. An unusual record of a White-tailed Deer, Odocoileus virginianus, in the Northwest Territories. Canadian Field-Naturalist 115(1): 172-175. The most northern record of White-tailed Deer (Odocoileus virginianus), approximately 100 km south of the Arctic Circle in the Northwest Territories (N.W.T.), is described. This and other observations from the N.W.T. and southeastern Yukon extend the known northern limit for White-tailed Deer in North America. Key Words: White-tailed Deer, Odocoileus virginianus, range, Northwest Territories, Yukon, Arctic, parasite, global warming. White-tailed Deer, Odocoileus virginianus, are the most important big game species in North America (Banfield 1977; Halls 1978) and have proven adapt- able across their range (Hesselton and Hesselton 1982). The range of White-tailed Deer in Canada includes all provinces except the island of New- foundland and Prince Edward Island. Most range maps and descriptions do not include either the Northwest Territories (N.W.T) or the Yukon Territory (e.g., Rue 1968; Banfield 1977; Whitaker 1980; Hesselton and Hesselton 1982). However, the first published sight- ings of White-tailed Deer in the N.W.T. were by from 1965-1966 in the southern Fort Smith-Wood Buffalo National Park area near the Alberta-N.W.T. border (Kuyt 1966). The earliest and most northern record of White-tailed Deer was from a hunter that killed 4 of 9 Deer at Little Doctor Lake west of Fort Simpson (ca. 61°55'N; 123°30’W; Figure 1) in 1959 or 1960 (Scotter 1974). Wishart (1984) documented expansion of this species in northern Alberta and in western and northern British Columbia. More recently, Gainer (1995) provided additional records in British Columbia and the southern N.W.T. from Fort Simpson to north of Great Slave Lake (Figure 1). 2001 NOTES 173 Range of White-tailed Deer given in Wishart (1984) /\/ White-tailed Deer range extensions given in Gainer (1995) Kall location ofa 3-year-old female White-tailed Deer on 28 June, 1996 Reports of White-tailed Deer sighting (Watson Lake, September 1996) and kill location (Ross River, January-Febmary 1997) provided by Yukon Department of Renewable Resources Observations of White-tailed Deer given in Scotter (1974) 200 400 600 800 1000 Kilometers Seale 1:12,000,000 Jenuorz 2000 FiGuRE |. Range extensions of White-tailed Deer in western Canada. On 28 June 1996, Lawrence Jackson, a Sahtu§ Mackenzie River from Fort Good Hope to Norman Dene hunter from Fort Good Hope, N.W.T., and Wells. They encountered a deer swimming across three others were travelling upriver by boat on the the river from west to east at 65° 37’ N; 128° 16’ W, 174 approximately 100 km south of the Arctic Circle (Figure 1). Under the N.W.T. Act and the N.W.T. Wildlife Act, holders of aboriginal general hunting licences have the right to take non-endangered species of Cervidae for food during any season. The four hunters killed the deer and brought the head to the Department of Resources, Wildlife & Economic Development (DRWED) in Norman Wells for species identification. Gainer (1995) also reported that hunters killed White-tailed Deer that were swimming across the Mackenzie River near its out- flow from Great Slave Lake at Fort Providence. Mr. Jackson reported that the deer was in good shape and had good abdominal and back fat deposits. Head measurements (to the nearest 0.5 cm) were: neck circumference — 13.0 cm; zygomatic width — 12.5 cm; straight head length 11.5 cm; contour head length — 14.0 cm; left ear length — 15.5 cm. There were no antlers or antler pedicels and Mr. Jackson confirmed that it was a female. Resource Officers from DRWED and the author identified the specimen as a White-tailed Deer and took photographs for documentation (on file with DRWED, Norman Wells, N.W.T.). We extracted a lower incisor for age classification (Matson’s, Mill- town, Montana) and hair and tissue samples for species identification (Forensic Laboratory, Alberta Natural Resources Service, Edmonton, Alberta). A count of cementum annuli determined the animal was 3 + 0-years-old (Gary Matson, personal communica- tion; Matson’s file #AW120996-280). The tissue sample was analyzed by polyacrylamide gel elec- trophoresis, electrophoresis for phosphoglucose iso- merase (PGI), superoxide dismutase (SOD), and iso- electric focusing erythrocyte acid phosphatase (EAP) with known species standards. The PGI test ruled out Wapiti (Cervus elaphus), the SOD test ruled out Cari- bou (Rangifer tarandus), and the EAP ruled out Mule Deer (Odocoileus hemionus). The enzyme test results confirmed that the tissue came from a White-tailed Deer (Forensic Laboratory Report # 96-86, Alberta Natural Resources Service, Edmonton, Alberta). This record of a White-tailed Deer within 100 km of the Arctic Circle is the most northern record in North America. There are no previous records of this species in the Mackenzie River valley west of Great Bear Lake. The nearest records are from 300 km south of Norman Wells. Mr. Albert Moses of the Pehdzeh Ki First Nation in Wrigley reported that two White- tailed Deer were seen 20 km south of Wrigley on the east side of the Mackenzie River (ca. 63° 13’ N; 123°23’ W) in 1997 (Figure 1). Mr. Moses also reported that a local trapper shot and killed a White- tailed Deer on his trapline several years ago near Wrigley. On the evening of 23 September 1996, two White- tailed Deer were observed by two Yukon Department of Renewable Resources (DRR) Conservation THE CANADIAN FIELD-NATURALIST Voli dis Officers about 20 km south of the Yukon-British Columbia border (Yukon DRR Occurrence Report no. WH96-199; Figure 1). The Deer were feeding in a ditch along the Alaska Highway and were tentative- ly identified as an adult female and a yearling. On 5 February 1997 the carcass of a male White-tailed Deer, field aged at 3 or 4-years-old, was found about 2 km northeast of Ross River site (Kevin Johnstone, Yukon DRR, Ross River, Yukon, personal communi- cation). From available evidence at the kill site (ca. 62° 00’ N; 132° 00’ W), Yukon DRR personnel assume that this Deer had died sometime in late January or early February 1997. There are no records of White-tailed Deer occur- ring in Alaska north of the 60" parallel (A. Franz- mann, Soldotna, Alaska, and L. Adams, Anchorage, Alaska, personal communications). These recent and historical sightings of White- tailed Deer in the N.W.T. and Yukon indicate the northern limit of White-tailed Deer is poorly defined. The factors leading to an expansion could be milder weather, increased White-tailed Deer habitat as a result of human settlement, absence of ungulate competition, and scarcity of predators (Wishart 1984). The Mackenzie Basin experienced a warming trend of 1.5°C in the twentieth century and is expect- ed to warm by a further 4 to 5°C by the middle of the twenty-first century (Cohen 1997). However, the N.W.T. still has sparse human settlement (0.02 per- son/km?), populations of native ungulates — Wood Bison (Bison bison), Moose (Alces alces), Woodland Caribou (Rangifer tarandus caribou), and Barren- ground Caribou (R. t. groenlandicus), and a full complement of large predators — Gray Wolves (Canis lupus), Black Bears (Ursus americanus), and Grizzly Bears (U. arctos horribilis). Gainer (1995) suggests that seismic lines cleared for oil and gas exploration and extraction activities might serve as a conduit for northern excursions by White-tailed Deer. The White-tailed Deer found near Norman Wells may have traveled north along the 10 m-wide underground oil pipeline right-of-way that stretches along the east side of the Mackenzie River from Norman Wells to northern Alberta. The right- of-way is covered with early regeneration browse species such as Willow (Salix spp.) and Red-osier Dogwood (Cornus stolonifera). Periodic mainte- nance ensures that the right-of-way is continually covered with early regeneration species attractive to White-tailed Deer. The northern extension of White-tailed Deer also provides potential for a concurrent extension of the species’ parasites, such as the Meningeal Worm (Parelaphostrongylus tenuis) and Winter Tick (Der- macentor albipictus), that could be transmitted to local Moose or Caribou populations. Over the past three decades, the Meningeal Worm’s range has not extended north or west of the Manitoba-Saskatche- 2001 wan border (Bindernagel and Anderson 1972). Therefore, it probably does not currently parasitize White-tailed Deer in the N.W.T. (W. Samuel, University of Alberta, personal communication; Gainer 1995). Similarly, winter ticks currently occur only at low levels in the Yukon (Samuel 1989), do not naturally occur in Alaska (Zarnke et al. 1990), and are not likely to occur north of the 62° in the N.W.T. (Wilkinson 1967; Samuel 1989). In recent years, a few reports of winter ticks on moose in the vicinity of Fort Simpson have been received (J. Antoine, Fort Simpson, N.W.T., personal communi- cation). Global warming is expected to change the distri- bution of many wildlife species and their habitats in the N.W.T. over the next century (Cohen 1997). Increased fire activity and changing growing seasons as a result of global warming (Hartley and Marshall 1997) are most likely to enhance the ability of early- succession species, such as White-tailed Deer, to expand their range. Additionally, global warming may lead to increased capacity for agriculture in the N.W.T. (Cohen 1997), which would favour White- tailed Deer and Mule Deer over Moose, Caribou, and Bison. Acknowledgments I thank Lawrence Jackson for providing details of this event. Tom Packer and Bob McClymont of the Forensic Laboratory of Alberta Natural Resources Service, Edmonton, Alberta, provided positive iden- tification of a tissue sample from Mr. Jackson’s deer. Arianna Zimmer (DRWED, Norman Wells, N.W.T.) and Lana Robinson (Sahtu GIS Project, Norman Wells, N.W.T.) produced the figure. Albert Moses (Pehdzeh Ki First Nation, Wrigley, N.W.T.) and Kevin Johnstone (Yukon DRR, Ross River, Yukon) provided recent occurrences of White-tailed Deer in their areas. Jean Carey (Yukon DRR, Whitehorse, Yukon) provided a copy of the Yukon DRR occur- rence report for White-tailed Deer sighted near Watson Lake. Bill Samuel (University of Alberta, Edmonton, AB) provided insightful comments on Meningeal Worm and Winter Ticks, discussed the possible role of global warming on those parasite dis- tributions, and suggested relevant literature to review. Alison Welch (DRWED, Yellowknife, N.W.T.) provided selected references. Anne Gunn (DRWED, Yellowknife, N.W.T.), Michelle Wheat- ley (Sahtu Renewable Resources Board, Tulita, N.W.T.), and two anonymous reviewers provided useful suggestions and comments. NOTES I Be Literature Cited Banfield, A.W. F. 1977. The mammals of Canada. University. of Toronto Press, Toronto, Ontario. 438 pages. Bindernagel, J. A., and R.C. Anderson. 1972. Distri- bution of the meningeal worm in white-tailed deer in Canada. Journal of Wildlife Management 36: 1349-1353. Cohen, S. J. 1997. Results and reflections from the Mackenzie Basin Impact Study. Pages 25-42 in Mackenzie Basin Impact Study Final Report. Edited by S. J. Cohen, Environment Canada, Ottawa, Ontario. 372 pages. Gainer, R. 1995. Range extension of White-tailed Deer. Alberta Naturalist 25: 34-36. Halls, L. K. 1978. White-tailed Deer. Pages 43-65 in Big Game of North America. Edited by J. L. Schmidt and D. L. Gilbert, Stackpole Books, Harrisburg, Pennsyl- vania. 494 pages. Harteley, I., and P. Marshall. 1997. Modelling forest dynamics in the Mackenzie Basin under a changing cli- mate. Pages 146-156 in Mackenzie Basin Impact Study Final Report. Edited by S.J. Cohen, Environment Canada, Ottawa, Ontario. 372 pages. Hesselton, W.T., and R. M. Hesselton. 1982. White- tailed Deer. Pages 878-901 in Wild Mammals of North America. Edited by J. A. Chapman and G. A. Feld- hammer, Johns Hopkins Press, Baltimore, Maryland. 1147 pages. Kuyt, E. 1966. White-tailed deer near Fort Smith, N.W.T. The Bluejay 24: 194. Rue, L.L., II. 1968. Sportsman’s guide to game animals. Outdoor Life Books, New York, New York. 655 pages. Samuel, W. 1989. Locations of moose in northwestern Canada with hair loss probably caused by the Winter Tick, Dermacentor albipictus (Acari: Ixodidae). Journal of Wildlife Diseases 25: 436-439. Scotter, G. 1974. White-tailed deer and mule deer obser- vations in southwestern district of Mackenzie. Canadian Field-Naturalist 88: 478-489. Whitaker, J. O., Jr. 1980. The Audubon Society field guide to North American mammals. Alfred A. Knopf, New York, New York. 745 pages. Wilkinson, P.R. 1967. The distribution of Dermacentor ticks in Canada in relation to bioclimatic zones. Cana- dian Journal of Zoology 45: 517-537. Wishart, W.D. 1984. Western Canada. Pages 475-486 in White-tailed Deer: Ecology and Management. Edited by L.K. Halls, Wildlife Management Institute. Stackpole Books, Harrisburg, Pennsylvania. 639 pages. Zarnke, R. L., W. M. Samuel, A. W. Franzmann, and R. Barrett. 1990. Factors influencing the potential estab- lishment of the Winter Tick (Dermacentor albipictus) in Alaska. Journal of Wildlife Diseases 26: 412-415. Received 20 January 2000 Accepted 19 May 2000 176 THE CANADIAN FIELD-NATURALIST Vol. 115 Northern Gannet, Morus bassanus, Nesting on Whitehorse Island, New Brunswick SEAN CORRIGAN! and ANTONY W. DIAMOND2 1Biology Department, University of New Brunswick, Bag Service 45111, Fredericton, New Brunswick E3B 6E1, Canada 2Atlantic Cooperative Wildlife Ecology Research Network, University of New Brunswick, P.O. Box 45111, Fredericton, New Brunswick E3B 6E1, Canada Corrigan, Sean, and Antony W. Diamond. 2001. Northern Gannet Morus bassanus nesting on Whitehorse Island, New Brunswick. Canadian Field-Naturalist. 115(1): 176-177. In 1999 a pair of Northern Gannets raised a chick for several weeks among a colony of Double-crested Cormorants on Whitehorse Island, New Brunswick. This is the first confirmed breeing of this species in New Brunswick, and in the Bay of Fundy, since 1880. The chick had disappeared by early September (long before it could have fledged), and may have been taken by Bald Eagles. Previous attempts at nesting in the region since the early 1970s are documented. Key Words: Northern Gannet, Morus bassanus, breeding, Bay of Fundy, New Brunswick. Northern Gannets (hereafter “gannets”) have not nested in the Maritime Provinces since about 1880 (Tufts 1986) but have been prospecting potential breeding sites in the outer Bay of Fundy at least since the early 1970s (Huntingdon 1975*; Gaskin and Smith 1979). During the summer of 1999, S.C. frequently passed close to Whitehorse Island (44° 59’ N, 66° 52’W) by boat and observed. up to six adult gannets on the top of a cliff in a large colony of nesting Double-crested Cormorants Phalacro- corax auritus. A photograph he took on 11 June shows three adult gannets on three separate nests. On 22 July both authors landed on the island, accompanied by K. Mawhinney and P. Edwards, in order to verify the breeding status of the gannets. As we approached the cliff-top from the north, one adult flew off, but one remained on a nest and allowed approach to within 5 metres. At that point the bird stood up on the nest and revealed a single chick, naked but showing down feathers beginning to appear on the back; from the early stage of feath- ering the chick was no more than one week old (Nelson 1978, page 90). There was no sign of any other gannet nests nearby. On 22 August, S.C. landed on Whitehorse to observe the nest. As he approached (approximately 20 metres away) a single adult flew off leaving the chick unattended. At that time tail feathers were starting to appear, suggesting the chick was at least 5 weeks old (Nelson 1978). By the time S.C. returned to the boat, two adult gannets had returned to the nest. On 23 August, three adult gannets and one chick were observed at the same place from a boat. On 13 September, S.C. returned to Whitehorse to observe the chick. No birds were visible from below the cliff. About 12 cormorants and 3 immature Bald Eagles Haliaeetus leucocephalus were on the cliffs; *See Documents Cited section. the top of the island was completely vacant. The nest was empty, and several pieces of white down were found on nearby vegetation. Discussion Gannets evidently nested in the southwest Bay of Fundy when Europeans first explored the area; the account by Denys (1672), from his visits in the 1640s, echoed Champlain’s reference to gannets nesting on the Wolves Archipelago (a group of islands 12 km south-east of Whitehorse). The last breeding record is from Gannet Rock, New Brunswick (44° 30'N, 66° 47’ W) between 1830 (when the lighthouse was built) and 1865 (Nelson 1978). However gannets bred on the Nova Scotia side of the Bay of Fundy until about1880, when they were extirpated (by over-exploitation by fishermen) from the Gannet Rock in Nova Scotia, south of the Tusket Islands (Tufts 1986). Gannets have been observed in the Grand Manan area in circumstances suggesting attempts to breed since the early 1970s. Huntingdon (1975*) described a single gannet ‘defending a territory’ at the south end of Kent Island (44° 34’N, 66° 45’W) in 1972; two were present the next year, and in May 1974 two were observed courting, gathering nest material and building a nest, but had disappeared by mid June. On Whitehorse itself, “in 1977, 1978, and 1979, a pair of gannets has been nesting” (Gaskin and Smith 1979); no further details were given, and there are contradictory data in Canadian Wildlife Service (1979) and Lock (1982*) from a ground-count on Whitehorse on 31 May 1979 which found no gan- nets. Lock (1982*, page 25) specifically noted the absence of “gannets, which had been reported roost- ing in this colony”. In view of these inconsistencies in the 1979 observations, we view our record of nest- ing on Whitehorse Island in 1999 as the first fully documented record of Northern Gannets breeding in the Bay of Fundy since the 1880s. 2001 Acknowledgments Thanks to John Eldridge and the staff of Quoddy Link Marine for their participation in this research. Documents Cited [marked * after date in text] Huntingdon, C. E. 1975. Page 77 in Gull seminar pro- ceedings. Edited by A. Smith. Unpublished report, Canadian Wildlife Service, Sackville, New Brunswick. Lock, A. R. 1982. A survey of some marine bird colonies in southwestern New Brunswick. Unpublished report, Canadian Wildlife Service, Dartmouth, Nova Scotia. Literature Cited Canadian Wildlife Service. 1979. Summary of unpub- lished surveys of waterfowl and seabirds in Passama- quoddy Region, 1973-79. Pages 61-67 in Evaluation of recent data relative to potential oil spills in the Passama- quoddy area. Edited by D.J. Scarratt. Fisheries and Marine Services Technical Report Number 901, Depart- NOTES 177 ment of Fisheries and Oceans, St. Andrews Biological Station, New Brunswick. Denys, N. 1672. Description and natural history of the coasts of North America (Acadia). Translated and edited by W.F. Ganong, 1908. Facsimile edition, Greenwood Press, New York. 1968. Gaskin, D. E., and G. J. D Smith. 1979. Observations on marine mammals, birds and environmental conditions in the Head Harbour region of the Bay of Fundy. Pages 69-86 in Evaluation of recent data relative to potential oil spills in the Passamaquoddy area. Edited by D. J. Scarratt. Fisheries and Marine Services Technical Report Number 901, Department of Fisheries and Oceans, St. Andrews Biological Station, New Brunswick. Nelson, J. B. 1978. The Sulidae: gannets and boobies. Oxford University Press, Oxford, U.K. Tufts, R. W. 1986. Birds of Nova Scotia. 3rd edition. Nimbus Publishing, Halifax, Nova Scotia. Received 28 January 2000 Accepted 23 May 2000 Unusual Harlequin Duck, Histrionicus histrionicus, Nest Site Discovered in Central Labrador TONY E. CHUBBS!, BRUCE MACTAVISH2, KEITH ORAM3:4, PERRY G. TRIMPER?, KATHY KNOX2 and R. IAN GOUDIE> 'Department of National Defence, 5 Wing Goose Bay, Box 7002, Postal Station A, Happy Valley — Goose Bay, Labrador, Newfoundland AOP 1S0, Canada 2Jacques Whitford Environment, 607 Torbay Road, St. John's, Newfoundland AlA 4Y6, Canada 3Jacques Whitford Environment Limited, Box 274, Postal Station C, Happy Valley — Goose Bay, Labrador, Newfoundland AOP 1C0, Canada 4Present address: Box 782, Postal Station C, Happy Valley — Goose Bay, Labrador, Newfoundland AOP 1C0O, Canada SMemorial University of Newfoundland, Department of Biology, St. John's, Newfoundland A1B 3X9, Canada Chubbs, Tony E., Bruce Mactavish, Keith Oram, Perry G. Trimper, Kathy Knox, and R. Ian Goudie. 2001. Unusual Harlequin Duck, Histrionicus histrionicus, nest site discovered in central Labrador. Canadian Field-Naturalist 115(1): 177-179. During telemetry monitoring of adult Harlequin Ducks, Histrionicus histrionicus, in 1999, a female was radio-tracked to her nest site. The nest site, only the third recorded in Labrador, was unusual, as it was located 108 m from the nearest river, in open spruce-lichen boreal forest. Key Words: Harlequin Duck, Histrionicus histrionicus, nest, habitat, Labrador. In May 1999, we began telemetry monitoring of Harlequin Ducks (Histrionicus histrionicus) in cen- tral Labrador to investigate movement patterns of adult pairs in spring. The objective was to determine if spring aerial surveys provided adequate informa- tion on which to base measures to protect occupied breeding habitat from potential disturbance by low flying military aircraft, a commitment to protection identified by the Department of National Defence (DND 1994*). Little information is known on the *See Documents Cited section. breeding habitat of the eastern North American pop- ulation which was designated as endangered in 1990 (COSEWIC 1998*%*). On 26 May 1999, we captured an adult female Harlequin Duck near Fig River (53°04'N, 63°08'W), a tributary of the Churchill River in central Labrador. The river section around the capture site ranged in width from 50 to 150 m and consisted of a series of five pools interspersed with rapids and two small islands. We marked the female with a standard metal band, an alpha numeric, colored (1J, yellow) plastic leg band and attached a tail-mounted VHF transmitter (148.441 MHz). The transmitter signal 178 was heard on the river < 50 m from the capture site on 28 May but the bird was not observed. Telemetry locations on 29 May, 5 and 11 June, were approxi- mately 700 m downstream of the capture site, adja- cent to the nest location. Since the bird was not observed on the river during these three surveys, we assumed that she was nesting. She was located using radio-telemetry on 12 June 1999, and flushed when approached. A search of the area revealed a nest on the ground. This is the first Harlequin Duck nest dis- covered in central Labrador, whereas two others have been previously described on coastal rivers in northern Labrador (Rodway et. al 1998; Chubbs et. al 2000). The nest was located 108 m from the nearest edge of the river and 250 m from where the female was captured. Generally, Harlequin Ducks do not nest far (<5 m) from water (Robertson and Goudie 1999). The 22 cm diameter nest was concealed beneath the layered branches at the base of a 10m tall Black Spruce (Picea mariana) tree. Surrounding vegetation consisted of open spruce-lichen forest, a habitat quite dissimilar to others described both in Labrador (Rodway et al. 1998; Chubbs et al. 2000) and else- where in North America (Robertson and Goudie 1999). Black Spruce branches that grew into the ground formed a tent-like enclosure which provided 100% vertical nest cover; similar to many nest sites described for the western Harlequin Duck population (Bruner 1997; Robertson and Goudie 1999). The nest bowl was lined with down and dead Black Spruce twigs and contained five eggs. The opening to the nest site was 33 cm high and 15 cm wide. Ground cover within 5 m of the net was 95% caribou lichen (Cladina sp.) and 5% Sheep Laurel (Kalmia angusti- folia). The forest was 100% mature Black Spruce, with trees spaced approximately 5 m apart. The nest site was found abandoned when revisited on 19 June 1999. The clutch was missing and assumed depredat- ed. The fate of the female is unknown as she was not relocated or observed after 12 June 1999 and surveys were discontinued after 19 June 1999. However, she was resighted in 2000. Information on Harlequin Duck nesting habitat in Labrador is limited to two other nest sites, both on coastal rivers (Rodway et al. 1998; Chubbs et al. 2000). Our nest site was in a forested rather than a riparian habitat, and was relatively far from water. Although, the nesting site was dissimilar to that described by Rodway et al. (1998) and Chubbs et al. (2000), all nests found in Labrador have been near rivers with small islands and channels and relatively calm steadies. Other nesting records in North America indicate that Harlequin Ducks select a vari- ety of nesting sites including rock crevices, cliff ledges, woody debris, tree cavities, and islands (Brodeur et al. 1998; Rodway et al. 1998; Bruner 1997). It appears that the selection of favorable THE CANADIAN FIELD-NATURALIST Volwdals breeding locations by Harlequin Ducks in Labrador may be more dependent on the type and characteris- tics of the river rather than the microhabitat in the vicinity of the nest site. Based upon our telemetry locations and that five eggs had been laid by 12 June, we suspect that the female had initiated nesting at least as early as 29 May, resulting in our failure to locate her on the river. In addition, based on the relative proximity of other nest sites to water described in Labrador and else- where, our search strategy focussed on the shoreline. We speculate that the nest may have been so dis- tant from the shore due to extraordinary high spring water levels in this region, which may take two to three weeks to recede. Additionally, numerous observations of Mink (Mustela vison) along rivers in this region suggest that female Harlequin Ducks may be nesting far from rivers to avoid these predators. Acknowledgments This project was funded by the Department of National Defence, through the Goose Bay Office, Environmental Mitigation Program. Major G. Humphries was the Project Manager. Special thanks are extended to B. Turner and C. Smith for their advice on capture and handling techniques and to M. Robert who aided in the initial captures. G. Goodyear and Todd Brough of Universal Helicopters Newfoundland doubled as both pilots and team members during captures. We thank A. J. Erskine and G. J. Robertson for providing helpful comments on earlier versions of the manuscript. Documents Cited (marked * after date in text) COSEWIC (Committee on the Status of Endangered Wildlife in Canada). 1998. Canadian Species at Risk, April 1998, Ottawa, Canada. 21 pages. DND (Department of National Defence). 1994. EIS: military flight training — an environmental impact state- ment on military flying activities in Labrador and Qué- bec. Goose Bay Office, National Defence Headquarters, Ottawa, Canada. Literature Cited Brodeur, S., A. Bourget, P. Laporte, S. Marchant, G. Fitzgérald, M. Robert, and J.-P. L. Savard. 1998. Etude des déplacements du canard Harlequin (Histrioni- cus histrionicus) en Gaspésie, Québec. Canadian Wild- life Service Technical Report Series Number 331, Québec Region, Ste-Foy, Québec. Bruner, H. J. 1997. Habitat use and productivity of Harlequin Ducks in the central Cascade Range of Oregon. M.S. thesis, Oregon State University. Chubbs, T. E., B. Mactavish, and P. G. Trimper. 2000. Site Characteristics of a Repetitively used Harlequin Duck (Histrionicus histrionicus) nest in Northern Labra- dor. Canadian Field-Naturalist 114: 324-326. Montevecchi, W. A., A. Bourget, J. Brazil, R. I. Goudie, A. E. Hutchinson, B.C. Johnson, P. Kehoe, P. Laporte, M. A. McCollough, R. Milton, and N. Seymour. 1995. National recovery plan for the Harle- 2001 NOTES 179 Rodway, M.S., J. W. Gosse Jr., I. Fong, and W.A. Montevecchi. 1998. Discovery of a Harlequin Duck nest in eastern North America. Wilson Bulletin 110: 282-285. quin Duck in eastern North America. Report Number 12. Recovery of Nationally Endangered Wildlife Committee. Robertson, G. J., and R. I. Goudie. 1999. Harlequin Duck (Histrionicus histrionicus). In The Birds of North America. Number 466. Edited by A. Poole and F. Gill. The Birds of North America, Inc., Philadelphia, Penn- sylvania. Received 17 January 2000 Accepted 9 May 2000 “Standing Over” And “Hugging” in Wild Wolves, Canis lupus L. DAvID MEcuH! Biological Resources Division, U.S. Geological Survey, Northern Prairie, Wildlife Research Center, 8711 - 37th St., SE, Jamestown, North Dakota, 58401-7317, USA 'Mailing address: North Central Research Station, 1992 Folwell Avenue, St. Paul, Minnesota 55108, USA Mech, L. David. 2001. “Standing over” and “hugging” in wild Wolves, Canis lupus. The Canadian Field-Naturalist 115(1): 179-181. During six summers, I observed “standing over” (SO) and “hugging” in a pack of wild Wolves (Canis lupus) habituated to me. In SO, one Wolf positions its groin above a recumbent Wolf's nose. I observed SO among all yearling and older Wolves for 1-180 seconds (x = 69 + 46 S.D.; N = 16). SO appeared to be primarily female-oriented and may inform each Wolf of the reproductive status of the other. I observed “hugging” five times and only during years when food competition was minimal. Key Words: Wolf, Canis lupus, behavior, standing over, affection, reproduction. To help understand the nature of relationships among members of Wolf (Canis lupus) packs, the postures of interacting animals are important clues (Schenkel 1947; Zimen 1982; Goodman and Klinghammer 1985). However, because of the elu- siveness of Wolves, it is difficult to observe their behavior under natural conditions (Mech 1974). Even the few observational studies that have been conduct- ed of wild Wolves (Murie 1944; Clark 1971; Carbyn 1974; Haber 1977) have failed to quantify most Wolf postural behaviors, including “standing over” (Schenkel 1947) and “hugging” (Goodman and Klinghammer 1985). I attempt here to quantify and analyze these two behaviors in one wild Wolf pack. Methods Study Area This study was conducted during six summers from 1988 through 1996 on Ellesmere Island, Northwest Territories, Canada (80° N, 86° W). There, Wolves prey on Arctic Hares (Lepus arcti- cus), Muskoxen (Ovibos moschatus), and Peary Caribou (Rangifer tarandus pearyi), and live far enough from exploitation and persecution by humans that they are relatively unafraid of people (Mech 1988, 1995). During 1986, I habituated a pack of Wolves there to my presence and reinforced the habituation annually. *See Documents Cited section. The pack frequented the same area each summer and usually used the same den or nearby dens, but pack composition varied annually (Mech 1995). The habituation allowed an assistant and me to remain with the Wolves daily, to recognize them individual- ly, and to watch them regularly from as close as 1 m (Mech 1988, 1995; National Geographic 1988). During 1759 h of observation, we noted each time an individual Wolf interacted with another Wolf, except for pups, which were not distinguishable from each other. “Standing over” (SO) is a low-intensity display in which one Wolf casually approaches a recumbent Wolf and stands over or along side the recumbent Wolf so that the standing Wolf's groin is positioned above the recumbent Wolf's nose. I found no set pat- tern of behavior before or after SO by either the active Wolf or the recumbent Wolf — SO took place in a variety of contexts. In “hugging,” which I have never seen described in wild Wolves and only once in captives (Goodman and Klinghammer 1985), an individual Wolf puts its front legs around the head and neck of another while each lies on its side chest-to-chest, or on its haunches facing each other, or side-by-side on haunches with one placing front legs around the other's neck. Only a few instances of SO were observed each year. Therefore sample sizes were too small for sta- tistical comparisons by sex and age class within years. Pooling across years for statistical compar- isons was inappropriate because opportunities for 180 THE CANADIAN FIELD-NATURALIST Vol. 115 TABLE 1. Distribution of 35 observations of “Standing Over” amongst various dyads of Ellesmere Island Wolves during summer. (BM = breeding male; BF = breeding female; YF = young! female; YM = young! male; PF = post-reproductive female.) BM BM BM. BE BF BF YM YM YM YF YE YF 2BE PRO Gere BM Year BF YF YM -BM..YR) YM BE YF BM” (BNE (BE°? YM») PEs BE a viewer 1988 1 0 0 0 0 i 1 0 1990 199] HOO? 1994! 1996 Observed 0.00 0.00 1.00 8.00 3:00 1.00 0:00 0.00 1:00 2:00 1:00 “000° 100" 2200 200m Ese Expected? 5.25 175. 175. 5.257 V5 VIS eS Ss ea eS 7S, SITS i Sele ere 2 1 0 0 1 1 0 0 - - - - Seog 2 2 & i] ! i] i] mare NNN OO 1 1In 1988, the YF and YM roles were yearlings; in 1994, 2-year olds. 2Expected if interactions were proportional to opportunities based on years present. participation by Wolves of various sex and classes breeding female. The post-reproductive female was varied among years. the active Wolf next most often and was the recum- bent Wolf most often. It also seems important that Results during the first year (1990) that a three-year-old During eight summers from 1988-1998 (exclud- daughter replaced her mother as breeder (Mech ing 1989 and 1995) I observed SO 35 times during 1995), the daughter stood over the mother the most six of those summers (Table 1). During the same often of all possible dyads in all years (Table 1). I period, I saw dominance interactions 217 times observed hugging five times, four of which were in (Mech 1999). SO took place among all yearling and 1990 (Table 2). All possible dyads of the breeding older Wolves in both the standing and recumbent male, breeding female and post-reproductive female positions for 1-180 seconds (x = 69 + 46S.D.; N= _ engaged in hugging. 16), although not in all possible combinations (Table 1). In 14% of the observations, the lying Wolf Discussion sniffed at the groin or genitals of the standing Wolf. Behaviorists disagree on the meaning or signifi- The most intensive SO I ever observed was at cance of SO. Schenkel (1947, translation by F. H. 1647 hours on 2 July 1990, when the new breeding Harrington) saw SO in his captive Wolves only during female (three-year-old “Whitey”) stood over her “peaceful” times and did not seem to regard it as a post-reproductive mother (“Mom”) for 3 minutes: dominance-related posture. He believed that SO“... My field notes taken at the time state: “When Mom _ is probably derived from the presentation of the geni- moved her nose to one side or the other of Whitey's __talia by the young, behaviour that is released (stimulat- groin, Whitey would move her groin over Mom's _ ed) by the licking by the mother.” However, I never nose. Mom appeared nonchalant and disinterested, saw any genital licking associated with SO. Zimen although she did sniff both the inside and outside of (1982) listed SO as one of 48 postures he observed but Whitey’s legs casually. Whitey remained intent and _ stated nothing more about it except that it was a “neu- stiff-legged and persistent throughout the display.” tral” posture. Goodman and Klinghammer (1985) list- Standing over appeared to be primarily female- ed SO in the following categories of behavior: aggres- oriented, with the post-reproductive female involved _ sive-elicited; aggressive-food related; aggressive-sex most often. The active Wolf was most often the _ related; care-giving, care-solicitation; greeting; and TABLE 2. Description of “hugging” in members of the Ellesmere Island Wolf pack. Date Description 30 June 1990 Breeding male and post-reproductive female lie on side chest-to-chest, and each puts front legs over the other 14 July 1990 Post-reproductive female hugs breeding daughter from behind 3 times, sitting, chest to back but rumps were side-by-side 14 July 1990 Breeding female and breeding male lie down facing each other and put legs over each other's shoulders and nuzzle each other 17 July 1990 Breeding female and breeding male face each other lying down and female puts legs around male's neck 9 July 1992 Breeding female lies with breeding male and places both legs around his neck, but he jumps up 2001 play-agonistic. I did not see any hostility or aggres- siveness associated with SO, but my observations were made only in summer, the nadir of Wolf breeding physiology (Seal et al. 1979). R. O. Peterson (personal communication), during 28 winters observing Isle Royale Wolves, saw seven cases of SO, four of which were between breeders; he concluded that “SO came out as a pretty minor behavior, and depending on con- text appeared to have significance in dominance expression and courtship.” Derix et al. (1994) grouped SO with behaviors they considered sexual in their cap- tive pack. Three clues from my data (Table 1) about the sig- nificance of SO are (1) that any pack member could be an active or passive participant, (2) the breeding female, post-reproductive female, and breeding male were most involved, and (3) in at least some of the cases, genital or inguinal sniffing was involved. From these data, I propose that SO is a posture that simply informs each Wolf of the reproductive status of the other, i.e., gender and degree of reproductive maturity and readiness. Goodman and Klinghammer (1985) interpreted hugging as an agonistic, greeting, or courtship dis- play. However, in the context in which I observed the behavior during summer, it appeared more to be a deliberate display of friendliness and affection than to fit in any of the other categories. I cannot explain why I only observed hugging in 1990, and an abort- ed attempt in 1992 (Table 2), except that in 1990 there were three adult Wolves and only a single pup in the pack. Thus food competition was minimal. These observations greatly extend information about Wolf behavior, previously only described in captive situations, by placing the behaviors in their natural context. The hypotheses offered about their interpretation can be used to explore further their significance. Acknowledgments This project was supported by the National Geographic Society, the United States Fish and Wildlife Service, the Biological Resources Division of the United States Geological Survey, and the United States Department of Agriculture North Central Forest Experiment Station. Logistical help of the Polar Continental Shelf Project, Natural Resources Canada; Atmospheric Environment Services, Environment Canada; and High Arctic International are also greatly appreciated. Permits were granted by the Department of Renewable Resources and the Grise Fiord Hunter and Trapper NOTES 18] Association of the Northwest Territories. The logisti- cal help of one to two field assistants each year dur- ing the study is gratefully acknowledged. I also thank Glen Sargeant for statistical advice. This is PCSP paper 003198. Literature Cited Carbyn, L. N. 1974. Wolf predation and behavioral inter- actions with elk and other ungulates in an area of high prey density. Canadian Wildlife Service Department of Environment. 233 pages. Clark, K. R. F. 1971. Food habits and behavior of the tundra wolf on central Baffin island. Ph.D. dissertation, University of Toronto, Ontario. 223 pages. Derix, R. R. W. M., H. de Vries, and J. A. R. A. M. van Hooff. 1994. Relationships in wolves (Canis lupus) and African wild dogs (Lycaon pictus) in captivity. Pages 20-56 in The social organization of wolves and African wild dogs. Edited by R.R. W. M. Derix. Universiteit Utrecht, Utrecht, The Netherlands. Goodman, P. A., and E. Klinghammer. 1985. Wolf ethogram. North American Wildlife Park Foundation, Ethology Series Number 3. 31 pages. Haber, G. C. 1977. Socio-ecological dynamics of wolves and prey in a subarctic ecosystem. Ph.D. dissertation, University of British Columbia, Vancouver, Canada. 786 pages. Mech, L. D. 1974. Current techniques in the study of elu- sive wilderness carnivores. Proceedings of the 11th International Congress of Game Biologists 11: 315-322. Mech, L. D. 1988. The Arctic Wolf: Living with the pack. Voyageur Press, Stillwater, Minnesota. 128 pages. Mech, L. D. 1995. A ten-year history of the demography and productivity of an arctic wolf pack. Arctic 48: 329-332. Mech, L. D. 1999. Alpha status, dominance, and division of labor in wolf packs. Canadian Journal of Zoology 77: 1196-1203. Murie, A. 1944. The Wolves of Mount McKinley. National Parks Fauna Series Number 5. U.S. Government Printing Office, Washington, D.C. 238 pages. National Geographic. 1988. White wolf. National Geo- graphic Explorer video. National Geographic Society, Washington, D.C. Seal, U. S., E. D. Plotka, J. M. Packard, and L. D. Mech. 1979. Endocrine correlates of reproduction in the wolf. Biology of Reproduction 21: 1057-1066. Schenkel, R. 1947. Expression studies of wolves. Be- havior 1: 81-129. Zimen, E. 1982. A wolf pack sociogram. Pages 282-322 in Wolves of the world. Edited by F. H. Harrington, and P. C. Paquet. Noyes Publications, Park Ridge, New Jersey. 474 pages. Received 26 January 2000 Accepted 29 May 2000 182 THE CANADIAN FIELD-NATURALIST Vol. 115 Limb Mutilations in Snapping Turtles, Chelydra serpentina RAYMOND A. SAUMURE! 2 ‘Department of Natural Resource Sciences, McGill University, 21111 Lakeshore Road, Ste-Anne-de-Bellevue, Québec H9X 3V9, Canada Present address: Shark Reef at Mandalay Bay, 3950 Las Vegas Boulevard South, Las Vegas, Nevada 89119, USA Saumure, Raymond A. 2001. Limb mutilations in Snapping Turtles, Chelydra serpentina. Canadian Field-Naturalist 115(1): 182-184. Adult Snapping Turtles (Chelydra serpentina) with mutilated limbs were captured in Ontario and Québec. Although simi- lar injuries have been reported for several other species of North American turtles, this is the first documented report of such mutilations in this species. The implications of such injuries are discussed. Key Words: Snapping Turtle, Chelydra serpentina, mutilations, leech, Placobdella parasitica, Ontario, Québec. Limb mutilations have been reported for several species of North American turtles. Nemuras (1966) noted that Spotted Turtles (Clemmys guttata) often had limbs missing, but did not report exact figures. A later study on C. guttata reported that 5.8% of a population (n = 207) in Pennsylvania had missing limbs (Ernst 1976). A Musk Turtle (Sternotherus odoratus) population (n = 204) at the same Pennsylvania location exhibited a limb loss rate of 4.9% (Ernst 1986). Belinky and Belinky (1974) reported having encountered many Wood Turtles (Clemmys insculpta) with mutilated limbs. Harding (1985), however, was the first to quantify limb loss in this semi-terrestrial species. He found that 12.5% of Wood Turtles in a northern Michigan forest popu- lation (n = 337) had mutilated limbs. In most cases of limb mutilations in C. insculpta, the Raccoon (Procyon lotor) was observed and/or suspected. Subsequently, there have been numerous reports of limb loss in Wood Turtles. Foscarini (1994) reported a limb mutilation rate of 12.9% from an agricultural population (n = 270) in Ontario. Tuttle (1996) docu- mented a rate of 9% from an agricultural population (n = 82) in New Hampshire. Saumure and Bider (1998) found rates of 15.2% and 32.3% for Québec agricultural (n = 33) and forest (n = 31) population samples, respectively. Waller and Micucci (1997) reported that fewer than 13% of Geochelone chilen- sis (n = 59) in Argentina had missing limbs. The lowest percentage of mutilated turtles was reported by Meek (1989) for Testudo hermanni in Yugoslavia, where only two of 213 turtles (0.94%) had missing limbs. To date, there have been no pub- lished reports of limb loss in Snapping Turtles, Chelydra serpentina. Herein, I report two instances of limb mutilations in C. serpentina and discuss the implications of such injuries. On 23 June 1994, I encountered an adult female Chelydra serpentina nesting in a gravel pit located along an old logging road, approximately 30 m from a stream in a mixed-deciduous forest in Pontiac County, Québec (45°53'N; 76°12’W). This specimen had both anterior limbs amputated through the radius and ulna bones, at a point close to their junction with the humerus. The wounds were completely healed. On 13 June 1996, this turtle was recaptured in the same gravel pit, measured, and photographed (Figure 1). This specimen had a carapace length of 289.4 mm. Of note is the fact that this turtle produced clutches in at least two of three nesting seasons, despite being devoid of anterior limbs. This suggests that the foraging ability of this turtle was not serious- ly compromised by the mutilations. On 19 August 1995, an adult male Chelydra ser- pentina was captured as it crossed a road near Port Royal, Big Creek National Wildlife Area (BCNWA), Regional Municipality of Haldimand-Norfolk, Ontario (42°35'N; 80°31’W). The turtle had a cara- pace length of 248.2 mm. This particular turtle was noticed because it appeared to be limping, and because Snapping Turtles are frequently run over (x = 68/year) by motorized vehicles in this area (Ashley and Robinson 1996). The specimen was missing its right posterior limb, which had been sev- ered sufficiently high on the femur so as to leave the turtle without a stump to walk on (Figure 2). The point of amputation was completely healed, although the skin to the right of the posterior lobe of the plas- tron was chafed and bleeding. This chafing appears to have been the result of friction between this part of its body and the paved road. The turtle possessed three shallow bloodless gashes of unknown origin on the anterior plastron, only one of which had completely healed shut. Of 229 Snapping Turtles captured between 1992 and 1996 in the BCNWA, only this tur- tle (0.44 %) had a missing limb. Three adult Smooth Turtle Leeches (Placobdella parasitica) were attached to this turtle in the damaged limb socket. This is simi- lar to a report by Saumure and Bider (1996) of leeches parasitizing the injured areas of mutilated Wood Turtles (Clemmys insculpta). Hendricks et al. (1971) hypothesized that leeches could readily colonize sick or injured turtles because of the host’s inability to avoid or rid themselves of the parasites. 2001 NOTES 183 FiGuRE 1. Chelydra serpentina from Pontiac County, Québec, with mutilated anterior fore- limbs. The absence of previous reports of limb mutilation in Snapping Turtles is not surprising, as the species has a number of physical and behavioural character- istics which serve to dissuade potential predators. Wild adult male Chelydra can reach a maximum carapace length of 494 mm and weigh as much as 34 kg (Conant and Collins 1991: 41). When harassed, they can also exude a foul smelling musk from glands located on the shell bridge (Carr 1952: 65). Moreover, a Snapping Turtle will tilt and lower the portion of its shell under attack, while turning to face its aggressor (Dodd and Brodie 1975). Lastly, the Snapping Turtle has a distinctly belligerent nature which it backs up by lightning quick strikes at speeds exceeding 78 ms (Lauder and Prendergast 1992). Therefore, it is not surprising that encounters between Raccoons and nesting female Snapping Turtles do not lead to predation attempts on the tur- tles (Congdon et al. 1987). Known causes of limb loss in turtles include attacks by mammalian predators (e.g., Harding 1985) and agricultural mowers (Ernst 1976; Tuttle 1996). It is, however, improbable that an adult Raccoon could gnaw off both anterior limbs of a fully alert adult Snapping Turtle. Consequently, it is much more like- ly that the attack occurred when the turtle was small and relatively defenseless (e.g., Robinson 1989; Walley 1993), or while it was in a state of torpor dur- ing hibernation (e.g., Brooks et al. 1991). Due to the remote location of the Pontiac site, it is unlikely that an agricultural mower was the cause of the amputa- tions. However, limb loss due to agricultural mowers or predation cannot be ruled out for the BCNWA turtle. Brooks (personal communication) reports that the observed frequency of limb loss in Chelydra ser- pentina at his Algonquin Park, Hamilton Harbour, and Lake Erie research sites in Ontario was also < 1%, despite the various types of predators and anthropogenic disturbances at these sites. Acknowledgments This manuscript benefited from the comments of Catherine A. Saumure, Ronald J. Brooks, and one anonymous reviewer. I would also like to thank Todd Hunsinger for his help in obtaining references for the final version of this manuscript. Research was partially supported by a grant to the St. Lawrence Valley Natural History Society from the Québec Ministére de 1’Environnement et de la Faune. Logistic support was provided by the World Wildlife Fund, Canadian Wildlife Service, Ontario Ministry of Natural Resources, Biod6éme de Montréal, Royal Ontario Museum, and Canadian Museum of Nature. Literature Cited Ashley, E. P., and J. T. Robinson. 1996. Road mortality of amphibians, reptiles and other wildlife on the Long Point Causeway, Lake Erie, Ontario. Canadian Field Naturalist 110: 403-412. Belinky, C. R., and G. K. Belinky. 1974. Climbing, falling and depth perception in the wood turtle Part II. Bulletin of the New York Herpetological Society 11: 32-43. Brooks, R. J., G. P. Brown, and D. A. Galbraith. 1991. Effects of a sudden increase in natural mortality of adults on a population of the common snapping turtle (Chelydra serpentina). Canadian Journal of Zoology 69: 1314-1320. 184 a 4 4 THE CANADIAN FIELD-NATURALIST FIGURE 2. Chelydra serpentina from Regional Municipality of Haldimand-Norfolk, Ontario, Vol. 115 “” with amputated right posterior limb and two leeches (Placobdella parasitica) attached posterior to the injury. Carr, A. 1952. Handbook of turtles. Cornell University Press, Ithaca, New York. Conant, R., and J. T. Collins. 1991. A field guide to rep- tiles and amphibians. Eastern and central North America. 3" edition. Houghton Mifflin Company, Boston. Congdon, J. D., G. L. Breitenbach, R. C. van Loben Sels, and D. W. Tinkle. 1987. Reproduction and nesting ecol- ogy of snapping turtles (Chelydra serpentina) in south- eastern Michigan. Herpetologica 43: 39-54. Dodd, C. K., Jr., and E. D. Brodie. 1975. Notes on the defensive behavior of the snapping turtle, Chelydra ser- pentina. Herpetologica 31: 286-288. Ernst, C. H. 1976. Ecology of the spotted turtle, Clemmys guttata (Reptilia, Testudines, Testudinidae), in southeast- ern Pennsylvania. Journal of Herpetology 10: 25-33. Ernst, C. H. 1986. Ecology of the turtle, Sternotherus odor- atus, in southeastern Pennsylvania. Journal of Herpetology 20: 341-352. Foscarini, D. A. 1994. Demography of the wood turtle (Clemmys insculpta) and habitat selection in the Maitland River valley. M.Sc. thesis. University of Guelph. 108 pages. Harding, J. H. 1985. Clemmys insculpta (Wood Turtle). Predation-Mutilation. Herpetological Review 16: 30. Hendricks, A. C., J. T. Wyatt, and D. E. Henley. 1971. Infestation of a Texas red-eared turtle by leeches. Texas Journal of Science 22: 247. Lauder, G. V., and T. Prendergast. 1992. Kinematics of aquatic prey capture in the snapping turtle Chelydra ser- pentina. Journal of Experimental Biology 164: 55-78. Meek, R. 1989. The comparative population ecology of Hermann’s Tortoise, Festudo hermanni in Croatia and Montenegro, Yugoslavia. Herpetological Journal 1: 404-414. Nemuras, K. T. 1966. Genus Clemmys. International Turtle & Tortoise Society Journal 1: 26-27, 39, 44. Robinson, C. 1989. Orientation and survival of hatchlings and reproductive ecology of the common snapping turtle (Chelydra serpentina) in southern Québec. M.Sc. thesis, McGill University, Montréal, Québec. 80 pages. Saumure, R. A., and J. R. Bider. 1996. Clemmys insculpta (Wood Turtle). Ectoparasites. Herpetological Review 27: 197-198. Saumure, R. A., and J. R. Bider. 1998. Impact of agricul- tural development on a population of wood turtles (Clemmys insculpta) in southern Québec, Canada. Chelonian Conservation and Biology 3: 37-45. Tuttle, S. E. 1996. Ecology and natural history of the wood turtle (Clemmys insculpta) in southern New Hampshire. M.Sc. thesis. Antioch University. 238 pages. Waller, T., and P. A. Micucci. 1997. Land use and grazing in relation to the genus Geochelone in Argentina. Pages 2-9 in Proceedings: Conservation, restoration, and man- agement of tortoises and turtles - an international confer- ence. Edited by J. Van Abbema, and P. C. H. Pritchard. New York Turtle and Tortoise Society. Walley, H. D. 1993. Chelydra serpentina (Snapping Turtle). Predation. Herpetological Review 24: 148-149. Received 31 January 2000 Accepted 25 October 2000 News and Comment Notices American Birding Association Ludlow Griscom Award for Publications in Field Ornithology: W. Earl Godfrey, June 2000 On 29 June 2000, the Board of Directors of the American Birding Association presented the ABA Ludlow Griscom Award “recognizing professional excellence and achievements in field ornithology literature” to W. Earl Godfrey. The award included a framed Certificate of Appreciation, recognizing his book The Birds of Canada 1966, revised 1986 and a plaque. An additional plaque and a pair of state-of-the-art binoculars were presented by co- sponsor Bushnell Sports Optics. Dr. Godfrey has more than 200 publications, the result of field investigations and museum analysis of bird populations including a series of monographs on regional avifaunas of Canada produced by the (then) National Museum of Canada and others in The Canadian Field-Naturalist. He was curator of Ornithology at the National Museum from 1947, and later Chief of the Vertebrate Zoology Division, until he retired from these positions at the end of 1976, but has continued as a Curator Emeritus and Research Associate of the (now) Canadian Musuem of Nature to the present. He has long been an Associate Editor for The Candian Field-Naturalist and is an Honorary Member of the Ottawa Field-Naturalists’ Club. Froglog: Newsletter of the Declining Amphibian Populations Task Force (42) The December 2000 issue contains: A short note about the status and abundance of caecilian populations [Daniel Hofer] — A Malformed Dendrobates tinctorius from French Guiana [David Massemin and Christian Marty] — Amphibian declines in Ecuador: overview and first report of chytridiomycosis from South America [Santiago R. Ron and Andres Merino] (also in Spanish) — Scientific meeting raises awareness of amphibian declines in Asia [Vanc Vredenburg, Yuezhao Wang, and Gary Fellers — Froglog Shorts — Publications of Interest. Froglog is the bi-monthly newsletter of the Declining Amphibian Populations Task Force of The World Con- Marine Turtle Newsletter (91) The January 2001 issue, 24 pages, contains: GUEST EDITORIAL: Developing sea turtle ecotourism in French Guiana: Perils and practicalities — ARTICLES: Growth rates of juvinile Green Turtles (Chelonia mydas) from Atlantic coastal waters of S. Lucie County, Florida, USA — Helping people help the turtles: The work of Projeto TAMAR-IBAMA in Almofala, Brazil — Notes: A new kind of illegal trade of Marine Turtles in Uraguay — MEETING REPORTS —- ANNOUNCEMENTS — Book REVIEWS — NEWS & LEGAL BRIEFS, — RECENT PUBLICATIONS. The Marine Turtle Newsletter is edited by Brendan J. Canadian Species at Risk November 2000 The Committee on the Status of Endangered Wildlife in Canada (COSEWIC) has issued Canadian Species at Risk November 2000, 24 pages. The list gives (1) all species des- ignated and all status re-assessments, in five Risk cate- gories (Extinct, Extirpated, Endangered, Threatened and Special Concern), (2) Not at Risk and (3) Data Deficient. servation Union (IUCN)/Species Survival Commission (SSC) and is supported by The Open University, The World Congress of Herpetology, The Smithsonian Institution, and Harvard University. The newsletter is Edited by John W. Wilkinson, Department of Biological Sciences, The Open University, Walton Hall, Milton Keynes, MK7 6AA, United Kingdom; e-mail: daptf@ open.ac.uk. Funding for Froglog is underwritten by the Detroit Zoological Institute, P.O. Box 39, Michigan 48068- 0039, USA. Froglog can be accessed at http://www?2. open.ac.uk/biology/froglog/ Godley and Annette C. Broderick, Marine Turtle Research Group, School of Biological Sciences, University of Wales, Swansea, Singleton Park, Swansea SA2 8PP Wales, United Kingdom; e-mail MTN @swan.ac.uk; Fax +44 1792 295447. Subscriptions to the MTN and donations towards the production of MTN and its Spanish edition NTM [Noticiero de Tortugas Marinas] should be sent to Marine Turtle Newsletter c/o Chelonian Research Foundation, 168 Goodrich Street, Lunenburg, Massachusetts 01462 USA; e- mail RhodinCRF@aol.com; fax + 1 978 582 6279. MTN website is: It is available from COSEWIC Secretariat, Chief Coleen Hyslop, c/o Canadian Wildlife Service, Environment Canada, Ottawa, Ontario KIA O0H3. See Web site: http://www.cosewic.gc.ca 185 186 The Boreal Dip Net 5(1) The December 2000 issue of the Newsletter of the Canadian Amphibian and Conservation Network/ Reseau Canadien de Conservation des Amphibiens and Reptiles reports: CARCN/RCCAR 1999-2000 initiatives — Board of Directors Meetings — Thank You to Sponsors — Forest Stewartship Recognition Award — Field trip — Extended Abstracts from keynote speakes: James P. Gibbs, Betsie B. Rothermel and Raymond D. Semlitsch — Forest Harvesting and Amphibians — List of papers presented at THE CANADIAN FIELD-NATURALIST Vol. 115 CARCN/RCCAR Meeting [September 2000, Penticton, British Columbia] — Digital Frog International & CARCN/RCCAR Scholarship Winner — US DOI Amphibian Declines Program — Turtle nests created in Laurel Creek Conservation Area. For membership and other information on the CARCN/RCCAR contact Bruce Pauli, Canadian Wildlife Service, National Wildlife Reseach Centre, 100 Gamelin Boulevard, Hull, Quebec, Canada K1A OH3. Sea Wind: Bulletin of Ocean Voice International 14(4) The October-December 2000 issue, 40 pages, contains: Conserving Tanzanian Coral Reefs and Mangroves: A sup- port request — Beauty if the reefs — Kapis Shells: A diminishing part of Filipino heritage — Novel on commer- cial fishing: A request — Water a fundamental right, say Bolivian poor — World Fora of Fishers — Conference: Putting fisher’s knowledge to work — Then and Now — Quotes — Members Feedback — Sea News — Booke Nooke — Kids Korner — Please help us! Tiny cards to Recovery (17) The October 2000 issue contains: Landowners join recovery effort (Robert Wenting) — Working group addresses recovery issues — Migratory scientist retires [Charles (Chuck) Dauphine] — Conserving nature at regional and continental scales: a scientific program for North America (Michael E. Soule and John Terborgh — Communities conserve “habitat connections” (Jean Langlois) — COSEWIC Update: Aboriginal knowledge to improve process (Sara Goulet) — CITES Update: Criteria under review: Experts to consider recommendations this save big oceans. Sea Wind is a publication of Ocean Voice International and is edited by Donald E. McAllister (e-mail: mcall@superaje.com) and is available though subscription or membership from Voice International Inc., 2255 Carling Avenue, Suite 400, Ottawa, Ontario K2B 1A6, Canada; phone (613) 721-4541; fax (613) 721- 4562; Angela Jellett, Executive Director at: e-mail: ; home page: winter (Bertrand Von Arx) — Tracking the world’s largest reptile [Leatherback Turtle] (Kathleen Martin). Recovery is a free newsletter providing information on Canadian species at risk, Coordinated by the Canadian Wildlife Service, and edited and designed by West Hawk Associates, Inc. It is available in either english or french from Canadian Wildlife Service, Environment Canada, Ottawa, Ontario, K1A OH3 and is accessible at www.cws- scf.ec.gc.ca/es/recovery/archive.html Ontario Natural Heritage Information Centre Newsletter 6(1) The Winter 2000 issue contains: SCIENCE: The Big Picture Project: Developing a Natural Heritage Vision for Carolinian Canada — Community Ecology: Rare commen- tities of Ontario: Coastal marine beach ridges — Zoology: King Rail and Prothonotary Wabler added to Endangered Species Act; Eastern Sand Darter survey; Ontario Odonate surveys — Botany: National Status assessment of Canadian wild orchid speies; 2000 Botanical highlights; Double- crested Cormorant impacts on rare plants; Botanical inves- tigations on Lake Superior — Herpetology: Using the OHS database to identify sites for conservation; Herpetofaunal distribution maps posted on NHIC web — News AND Notes: NHIC involved in MNR Northern Boreal Initiative; Biological inventory of Kawartha Higlands OLL site; NHIC biologists conduct life science inventory in Polar Bear Prov. Park; Trent University — NHIC internships in conservation biology; NHIC is part of NMR’s species at risk program; NHIC Biologists author COSEWIC status reports; NHIC web site renovation; Point Pelee natural his- tory news; NHIC fieldwork in northwestern Ontario; The Association for Biodiversity Information launches NatureServe; Updates to the COSEWIC national species at risk list; NHIC completes hot-spot mapping for NCC — Book REvIEws — NHIC INFORMATION PRODUCTS — NHIC STAFF LIST. The Ontario Natural Heritage Information Centre Web Page is: http://www.mnr.gov.on.ca/MNR/nhic/nhic.html Mailing address: 300 Water Street, 2nd Floor, North Tower, P.O. Box 7000, Peterborough, Ontario K9J 8M5, Canada. Book Reviews ZOOLOGY The Birds of British Columbia Volume 3 By R. Wayne Campbell, Neil K. Dawe, Ian MacTaggart- Cowan, John M. Cooper, Gary W. Kaiser, Michael C. E. MeNall and G. E. John Smith. 1997. UBC Press, Vancouver. 693 pp., illus $50.00. The first two volumes of this series, published in 1990, provided a long overdue update on the non- passerine avifauna of British Columbia, as well as a brief history of the ornithology of the province and a detailed overview of its biogeography. Although the authors’ reputations led to high expectations for those volumes, they managed to exceed those expec- tations considerably (see review in Canadian Field- Naturalist 107: 547-548, 1993). Their third volume, covering tyrant flycatchers through vireos, is an even more thorough and splendid achivement, providing both an indispensible collection of scientific data and an easily-read overview of the British Columbia sta- tus and distribution of each species in 18 passerine families. In keeping with the authors’ policy of maximizing input, the text of volume 3 begins with a four-page section of acknowledgements to contributors to all aspects of production from collecting and contribut- ing data to writing, reviewing, and/or illustrating portions of the book to financing its production. This is followed by an introductory section on the book’s development, general aspects of the birds covered in this volume, and recent developments in avian tax- onomy, followed by a checklist of the 91 species of birds covered in the book. The bulk of the text con- sists of two to 12-page [usually four to eight-page] accounts of species of “regular” occurrence in the province. Thirteen additional one-half to one page accounts of “casual, accidental, extirpated, and extinct species” close the main text. Six appendices (on migration chronology, 1957-1993 Christmas Bird Count data, 1969-1994 Breeding Bird Survey data, a list of 6,498 (!) contributors of observations and two on computer data base details), a list of ref- erences cited, an index to species, and short biogra- phies of the authors conclude the tome. The species accounts are more thorough than those of most “Birds of’ works of recent years. The text of each species of regular occurrence includes sections on its overall range, its status in British Columbia, changes in status, nonbreeding and breed- ing distribution and habitat, nesting data (including nest-sites, nest structure, egg-laying dates, clutch size, brood dates, incubation periods, Brown-headed Cowbird parasitism, etc.), and “remarks” (various aspects of taxonomy, song, conservation status, habi- tat, and/or behaviour not covered in other sections). Information is included on different areas of the province, and data gaps are identified. The main text for each species is followed by a section on “note- worthy records,” arranged by season and geographi- cal portion of the province. Maps, graphs, and black- and-white and/or colour photographs illustrate each bird, its habitat, chronology of occurrence in differ- ent parts of the province, locations of nonbreeding and breeding records, and/or nests and eggs. The text of the 13 species of less regular occurrence concen- trates on usual range, records of occurrence in British Columbia, references to identification, and sometimes notes on erroneous and/or additional unsubstantiated records in the province. This book follows the taxonomy and nomencla- ture recognized by the American Ornithologists’ Union to 1995, except that Pacific-slope and Cordil- leran flycatchers are treated together as “Western Flycatcher complex.” Although the authors agree that those found along the coast are Pacific-slope Flycatchers, they feel that the identity of those in the interior is less certain, and exclude Cordilleran from their check-list. However, Campbell has included it in a subsequent provincial bird list (Wild Bird trust of British Columbia Wildlife Report Number 2, 1998). As attractive as the first two volumes were, this third volume reaches “coffee table” book quality in many of its photographs. The book also exceeds the previous volumes in the comprehensiveness of the information presented for each species. Information on the British Columbia distribution, status, and pop- ulation trends of the species covered is complete to December 1995 (with some significant subsequent records added in footnotes and addenda) and what- ever was known by then of life history details in British Columbia is summarized, with numerous lit- erature sources cited for follow-up. When details were not available from British Columbia, these are filled in from studies conducted elsewhere. In spite of the book’s length and comprehensiveness, it is relatively error-free, with most errors confined to minor grammatical matters, lapses in spelling of people’s names or in figure numbers, and failure to differentiate between same-year publications by a specific author cited. The latter problem no doubt results primarily from the authors’ efforts to keep the 187 188 information as current and accurate as possible until time to go to press. After a second same-year publi- cation by a given author is added, finding all the places that the first was cited previously would be a daunting task, although most should have been caught by a final, thorough, proof-reading. About 30 references cited in the text are not included in the reference list, at least under the date cited. Most of these probably represent mismatched dates, but at least three were omitted completely (Brewster 1893 cited on page 485; MacLean 1970 cited on page 128; and Southern 1958 cited on page 531). Authorship of The Birds of Alberta editions is garbled. The 1966 edition, revised by W. R. Salt, was published under the authorship of Salt and Wilk, not W. R. and J. R. Salt, while the 1976 book was authored by W. R. and J. R. Salt, not W. R. alone. I was able to detect only a few errors of a substantive nature. Although “most” accurately describes the British Columbia proportion of the breeding populations of most of the species listed on page 11, I believe that breeding populations in other provinces of Western Wood- Pewee, Warbling Vireo, House Finch, and Red Crossbill are sufficiently large that “most” would be somewhat overstated. The text statement (page 62) that Dusky Flycatcher has been reported only once on Vancouver Island contradicts the accompanying map (page 63) on which two Vancouver Island loca- tions are plotted, and the list of noteworthy records (page 67), where two records are included. Although Bewick’s Wren has nested in southern Ontario (page 310), nesting there is not regular and probably has not occurred since the late 1950s. Examination by National Museum of Canada [now Canadian Museum of Nature] personnel showed that the pur- ported Veery specimen collected on 20 May 1893 in Victoria (page 389) was in fact a Swainson’s Thrush (D. R. Gray in M. K. McNicholl, 1978. Murrelet 59: 102-104). The four-letter codes listed for each species are somewhat problematical. If such a list of codes is to A Guide to the Birds of the West Indies By H. Raffeale, J. Wiley, O. Garrida, A. Keith, and J. Raffeale. 2000. Princeton University Press, Princeton, New Jersey. 511 pp., illus. The real James Bond would be pleased. This book is an update to the original guide to the birds of the West Indies, published by Bond in 1936. Not only is it a revision of the Bond’s book, it goes beyond the traditional concept of a field guide and adds some new and interesting dimensions. THE CANADIAN FIELD-NATURALIST Vol. 115 be useful, the same codes should be used by all observers and the codes should be stable. Besides the obvious problem that some codes could logical- ly refer to two species (e.g. is BASW Barn Swallow or Bank Swallow?; is CEWA Cedar Waxwing or Cerulean Warbler?), some of the codes also change every time that the American Ornithologists’ Union publishes a list of name changes (approximately every two years). To test the usefulness of the codes, I compared those in this book with those for the same species published the same year by Peter Pyle in Identification Guide to North American Birds. Part I, Slate Creek Press, Bolinas, California. The codes in Pyle’s book are those considered official in 1997 by the Canadian and U.S. banding authorities. Although most codes in the two books are identical, those for 12 species (Western Wood-Pewee, Gray Kingbird, Tree Swallow, Bank Swallow, Barn Swallow, Gray Jay, Western Scrub-Jay, Canyon Wren, Cedar Wax- wing, Northern Shrike, Yellow Wagtail, and Black- backed Wagtail) differed. Moreover, the code used for Canyon Wren in the British Columbia book is used for Cactus Wren by Pyle and the banding authorities, and the code for Black-backed Wagtail (BBWG) in Campbell’s 1998 provincial list differs from the two codes used in the two books (BBWA and BWAG). Its few errors do not detract from the overall high quality of this book. Every professional and amateur ornithologist and “birder” in British Columbia and adjacent areas should have copies of all three vol- umes as a basic information source on what is and is not known about the avifauna of this large and diverse province. We all wait for the fourth and final volume with eager anticipation. MARTIN K. McCNICHOLL 4735 Canada Way, Burnaby, British Columbia V5G 1L3, Canada Looking first at this book as a field guide. It cov- ers 564 confirmed species, including five extinct birds. These animals are found on islands throughout the Caribbean, in a rough triangle from Grand Bahama to San Andrés and Granada. It does not include Trinidad, Tobago, Aruba, Cozumel, and a handful of smaller islands close to the South and Central American coasts. All the species are shown in full colour in a series of plates painted by seven artists. There are generally 2001 about eight to 12 species per page. As the book is slightly larger (16 X 24 cm or 6.25 X 9.5 inches) than a typical guide, the individual portraits are a respectable size. Opposite each plate are cryptic comments on the most significant field marks. The only black-and-white drawing is of bird topography. The 86 plates are followed by a one third to half page description of the key identification features, status, range, and other relevant comments. A 4 X 2.5 or 3cm range map accompanies each text section. The identification remarks cover all the key characteristics as needed for identifying a species in the field . Where appropriate this includes the differ- ence between males, females, and immatures. Similar species are noted along with their essential differences. This includes species that might occur as vagrants ( for example, American White Pelican and Eastern White and Pink-backed pelicans). When a species is distinctive this section is left out (e.g., Magnificent Frigatebird). Local common names are included for each island. This will help you identify at least the range of possibilities from information provided by local people. (“Gaulin” is used in sever- al places and for several species of heron-egret). The quality of the plates is splendid. I looked care- fully throughout the book at the shape, posture, and details of coloration of the birds depicted. I could find nothing of any consequence. Far more often I was impressed to the point where I began making a comparison to the great Thorburn. So as a field guide this book has it all: good up-to- date coverage, top quality plates, and crisp support- ing text. Now we can look at the additional material provided by the authors. The first, and most impressive, innovation is that they have repeated the illustrations of the island endemics on eight additional plates. So as well as BOOK REVIEWS 189 finding the Hispaniolan Woodpecker on the wood- pecker plate, you can also see a different rendition on the Hispaniolan endemics page. This brings a sharp focus to the species distribution on the individ- ual islands. In addition, there are 12 plates depicting a single endemic species, such as the Martinique Oriole. These paintings frequently include a full or partial rendition of the bird’s habitat. Such additions give the book an increased aesthetic appeal. There is a survey of the conservation status and efforts on 19 of the major islands or island groups. Not surprisingly the level-of-effort and the effective- ness of the conservation programs varies widely. What was a surprise is which islands have an effec- tive conservation effort and which need a greater effort. This section, as well as the introductory mate- rial, is well worth reading. So what could I find wrong with this book; not much really. The distribution is defined on the maps by encircling the area with a line. This is not as visu- ally effective as blacking in the land mass. My bias is to have the plates opposite the text in the style of the National Geographic guide. The limited descrip- tions opposite the plates are very cryptic. For exam- ple, for non-breeding Piping Plover it reads “Black bill”; for Semipalmated Plover it says “Dark bill.” You have to turn over 200 pages to get more infor- mation from the species account. As I said it is hard to find serious fault with this great new guide to a popular tourist area. I think 007 would also be happy. After all, it took five authors to replace the original Bond! Roy JOHN 2193 Emard Crescent, Gloucester, Ontario K1J 6K5, Canada A Guide to the Birds of India, Pakistan, Nepal, Bangladesh, Bhutan, Sri Lanka, and the Maldives Richard Grimmet, Carol Inskipp, and Tim Inskipp. 1999. Princeton University Press, Princeton, New Jersey. 888 pp., illus. U.S. $85. Each year brings more great new birding books, but rarely are they as long awaited and as valuable as the Birds of India. Not surprisingly, and as befits one of the world’s great centers of biological diversity, the Indian subcontinent is graced with a wealth of ornithological literature, including several useful field guides. However, up until now there has been no one book aimed at the bird watcher that pulls it all together in one neat package. Here it is. All 1300 species found in the seven countries that form the region are covered in an astonishing 153 plates crafted by 12 well-known bird artists. All major forms are illustrated including both sexes, immatures, and subspecies as appropriate. Where necessary, multiple views are included to illustrate birds in flight or at various angles. Despite the large number of illustrations, most of the plates escape overcrowding; some are works of art (the green pigeons on plate 35 are stunning). Each plate has a facing page with a short narrative description of the species covered. All of the illustrations are labeled and referenced in the facing text. The format of the book will be instantly familiar to anyone who owns any of the several regional 190 guides Princeton has released over the past few years; a short introductory section including a histo- ry of ornithology in the subcontinent, a description of the major habitat types, tips on how to use the book, followed by information on conservation. The species descriptions focus on identification and include sections on voice, habits, habitat, and breed- ing. Where appropriate, descriptions are broken down by sex, age, and subspecies. Each account includes a summary of the bird’s status in each of the seven countries covered by the book. The range maps indicate breeding, wintering, and passage ranges for each species. The exhaustive research that went into the book is demonstrated in high quality of the text, range maps, and extensive references. The fact that many range maps include question marks reflects the challenges inherent in studying such a massive area and will doubtless inspire further field- work. How good is this book? The plates and description of birds I know well are excellent; key features are accurately illustrated and described, and multiple views are provided of problematic species (e.g., rap- tors, lariids). The plates are on a par or surpass the quality of many top guides, for example, it has the best pictures of tropical terns I have seen in any (e.g., it pictures Sooty and Bridled Tern in heavy molt, and illustrates the pale secondary covert bar usually overlooked in drawings of Brown Noody). The recently released guide to the West Indies, also published by Princeton, has a total of three pictures of Sooty and Bridled Tern; The Birds of India has ten. Another good example: there is an excellent painting of a juvenile Temminck’s Stint, clearly illustrating the diagnostic sub-terminal “anchor” markings in the mantle feathers. This attention to detail is maintained throughout the plates, including in the myriad drawings of “little brown jobs” with which the subcontinent is amply blessed — enough obscure chats, warblers, flycatchers, and wheatears to satisfy the most discriminating birder’s palate. Are there problems? Any book of this nature must make compromises. The challenge of handling over one thousand often very similar species means that luxuries such as the stand alone “hawks in flight” plates featured in many guides are left aside. Also, some readers may find the small text a bit of a chal- THE CANADIAN FIELD-NATURALIST Vol. 115 lenge; occasionally the range maps are hard to make out, particularly the paler shading used to indicate wintering ranges. A few of the plates are overcrowd- ed, 45 pictures of psittacula type parrots on one page is overwhelming. Additionally, the arrangement of birds is occasionally confusing to the eye, and it is often necessary to cross check the numbers against the facing text. While the color separation and detail on the plates is exceptionally good, many are set against a rather gloomy gray background which detracts from their visual impact. Many of the facing pages to the plates have large patches of empty space, which is a waste given the severe space con- straints which faced the authors. I found no errors in the species accounts of birds, and the book is remarkably free of proof-reading errors. Some sort of index to the plates would be helpful, particularly to rapidly locate some of the more com- plex groups. It should be noted that this is perhaps the first major field guide to reflect some of the recent major shuffling of taxonomic groupings, thus birds tend to show up in odd places (e.g., Grebes and Loons, former front runners, are at plate 75, about halfway through). Should you buy this book? It is hard to imagine any birder who has caught the Asian birding bug not rushing to get it. But before you do, consider that it is a comprehensive reference work, not a field guide. At 888 pages and 2.5 kilos you cannot casually slip it into your pocket. Fortunately, and following a precedent set in earlier Princeton tomes, a slimmed down, low budget, field guide version consisting pri- marily of the plates is available, and I suspect that this is the version that most travelers will want to use. Is a bird book on the Indian subcontinent relevant to Canada? Oddly enough, yes. The treatment of Asiatic migrants is so good that is bound to become a critically important reference for anyone interested in vagrants. You don’t have to be a hard core birder to admire this handsome book; it would make a great addition to any natural history library. MARK GAWN Canadian High Commission, P.O. Box 404, Bridgetown, Barbados, West Indies 2001 Trogons and Quetzals of the World By Paul A. Johnsgard. 2000. Smithsonian Institution Press, Washington. xi + 223 pp., illus. U.S. $49.95. In the introduction, the author singles out one tro- gon, the Resplendent Quetzal as the “ultimate sym- bol of beauty.” This is how many birders and I think of this entire group of birds. He begins the actual text with “Trogons ... (are).. characterized by unique toe arrangement.” and, at first, this seems to me an odd way to view these colourful species. But, as you read through the taxonomy then this introduction makes sense. In fact this early statement sets the tone for this book. For while this is a book for all of those who are interested in tropical birds it is clearly writ- ten by a scientist with a scientific viewpoint. The author routinely uses terms such as rectrices and maxilla instead of the more common tail feathers or top of the bill. This is not a major problem if readers are willing to use the thoughtfully written glossary at the back of the book. The book covers the currently recognized 39 species of trogons. It begins with the comparative biology, which includes an absorbing account of tro- gon evolutionary development. The species accounts include a description of range, size statistics, and the plumage of males and females as adults and juve- niles. The notes on identification both in the field and the hand are helpful to birders and scientists. These accounts are rounded out with sections on ecology, behaviour, breeding biology, and conserva- tion. A distribution map is also included. Please note this is not, nor is it intended to be, a field guide. These species accounts are as detailed as possible given the current level of information. The author has rendered the birding and science communities a great service by pulling all this information in a coherent, readable form. What is very noticeable in these accounts is how many times there is little or no infor- mation available. Even such simple data as the typical number and colour of the eggs is unknown for several species. This means that amateurs and the growing number of ecotourists can make a singular contribu- tion if they find and carefully document any trogon nest, and then send this information to the author. The birds are primarily illustrated with plates by Gould. These plates were originally used in the last book on trogons published in about 1838. The illus- trations are delightfully Victorian, showing birds in a setting of orchids, bromeliads and other tropical veg- etation. As identifying tools they lack much of the clarity of current illustration techniques, but as this book is not meant to be a field guide I do not see this as an issue. Each distribution map is accompanied BOOK REVIEWS 19] by a line drawing of a male bird. While the drawings are sound, they lose much in being black-and-white. For example, the Violaceous (a blue-and-yellow bird) and Collared Trogons (a green-and-red bird) look identical in black-and-white whereas they are very different in reality. The author has added two sections I found most interesting. He has produced a key; much like those used in flower guides, for separating the species. This could be useful, especially with birds seen only partially because of the forest cover. He also defines the meaning and origin of the scientific names. To those of us who never studied Latin or Greek this adds a helpful clarification. I think this book is first an important contribution to bird literature, second an interesting and readable account of an astounding group of birds, and third, I hope, an inspiration to birders to record their obser- vations. However, I did find some silly errors. Figure 6, for example, is wrong. It is obviously Figure 5 repeated with the heading changed. I sus- pect the real Figure 6 lies buried in the printer's cup- board. The author states that Gould's illustrations, being painted from museum specimens, are not completely accurate. When I looked at the plate for the Blue- tailed Trogon, a species I have not seen, I decided to verify the bird's appearance against the description. Looking up the species in the index, I turned to the pages indicated and found the description was at odds with the plate. I searched my library and locat- ed another illustration that supported Gould's version of the bird. After a period of confusion, I realized the index was in error (It had led me to the Black-tailed Trogon). I did a sampling of the rest of the index and did not find another problem. Later, however, I dis- covered the Surucua Trogon is listed as plate 26 when it is plate 27. So be careful how you use the index. Despite this lapse of detail, this book is worth buying for several reasons. It will be valuable to the growing number of North Americans visiting South and Central America. The book will delight those who like attractive bird books. After all it will cost you about $20,000 U.S. to buy Gould's original. It opens a path for enthusiastic amateurs and profes- sionals to contribute to our knowledge. Hopefully it will also spur the process of conservation of the rapidly diminishing rain forest. Roy JOHN 2193 Emard Crescent, Gloucester, Ontario K1J 6K5, Canada 192 THE CANADIAN FIELD-NATURALIST Voli hs Research and Management of the Brown-headed Cowbird in Western Landscapes Edited by M. L. Morrison, L. S. Hall, S. K. Robinson, S. I. Rothstein, D. C. Hahn, and T. D. Rich. Studies in Avian Biology #18. Cooper Ornithological Society. 312 pp. U.S. $18.00. The Brown-headed Cowbird, once appropriately known as the Buffalo-bird, initially occupied a range similar to that of the American Bison. As agriculture advanced, for example as clearings appeared in forests, cowbirds expanded their range to cause adverse effects on many species including the Kirtland’s Warbler in Michigan. A conference held in Sacramento, California, 23-25 October 1997, involved almost all experts in the field, and included papers on cowbird ecology, cowbird management, and cowbird control. An overview essay precedes each of the three sec- tions. Scott Robinson, University of Illinois, summa- rizes the 12 papers reporting factors affecting the abundance and distribution of cowbirds. James N. M. Smith, University of British Columbia, reviews the 15 papers dealing with host selection, impact on hosts, and criteria for taking management action. Linnea S. Hall and Stephen I. Rothstein give an overview of the five papers dealing with the efficacy of control and proposed alternatives to standard con- trol practices. The reader quickly becomes aware of the prodi- gious amount of effort represented by these papers; several individual studies involved over 100 observers and innumerable hours of field studies and desk analysis. In monetary terms, a million dollars a year is spent annually trapping cowbirds in California. Habitat remains extremely important. Some wilderness areas in Idaho have no cowbirds at all; farther south, in the Sierra Nevada, cowbirds are pro- gressively less common at higher elevations. In Coastal Sage Scrub in California, only 3 of 217 nests were parasitized. Common Yellowthroats had lower parasitism and predation rates and higher nesting success when they nested in extensive cattail marsh- es. Western species, especially Sage Sparrows, Horned Larks, and Western Meadowlarks, largely avoid parasitism because cowbirds arrive after initia- tion of their nests. Scarcity of observation perches in shrub-steppe areas also tends to decrease parasitism. Cowbird abundance is strongly influenced by dis- tance to cattle; most cowbird flights are less than 3 km; cowbirds are usually absent 7 km from feeding areas. Rotation of livestock away from host breeding habitat thus becomes a promising management tool. Smith’s overview suggests that cowbird manage- ment programs are indicated when over 60% of nests are parasitized, and rarely indicated when nest para- sitism rates are below 30%. However, Lazuli Bunting populations are not sustainable when para- sitism rates exceed 20%, because 90% of Lazuli Bunting nests that fledged a cowbird chick did not fledge a single bunting chick. Similarly, 75% of southwestern Willow Flycatcher nests failed when parasitized. Up to 43% of nests of the endangered Least Bell’s Vireo were parasitized at least once; removal of cowbird eggs, a labour-intensive exer- cise, increased annual productivity by up to 44%. In British Columbia, where Song Sparrows experienced intense parasitism and frequent nest failure, cowbird egg removal sharply lowered sparrow nest failure rates. It is difficult to collect data on the fecundity of individual cowbirds, but they appear to lay between 1.72 and 8.16 eggs per female. Some cowbird females lay each egg in the nest of a different host species. When one prolongs the study period, adverse effects of cowbird parasitism are more severe than most short-term studies of nests alone would indicate. Survivorship studies to four weeks post-hatching found that hosts often restrict their feeding to the larger cowbird, allowing their own young to starve. Thus, Indigo Buntings are only 18% as likely to return the next year if they fledge from a nest that contains a cowbird. Survivors in other species may have lower fecundity later in life. What management tools are effective? Prescribed burning in southern Wisconsin lowered parasitism of Red-winged Blackbirds. In pastures with high cattle concentration at the Fort Hood military installation, Texas, initial Black-capped Vireo parasitism rates were 91%. Cowbird trapping, especially in pastures with high cowbird concentrations, reduced this rate. Next, removal of cattle from one area led to a decline in cowbird parasitism of: Biack-capped Vireos from 35% one year to zero the next year. With all control measures combined, this vireo para- sitism rate fell to below 9%. In central California, cowbird trapping decreased the parasitism rate on Willow Flycatchers from 65% to 39% and the num- ber of flycatchers fledged per female increased from 1.04 to 1.72. Items cited above are merely representative of the myriad of new and interesting facts gleaned from a well-planned, well-organized, well-reported, state- of-the-art conference. C. STUART HOUSTON 863 University Drive, Saskatoon, Saskatchewan S7N 0J8, Canada 2001 Owls: A Guide to the Owls of the World By Claus KG6nig, Friedhelm Weick, and Jan-Hendrik Becking. 1999. Yale University Press, New Haven CT. 462 pp., 64 colour plates. This is a beautiful, impressive book with major flaws. Apart from its aesthetic value, if you already own Burton’s Owls of the World (1973), do you need to buy this new book? With a combined wealth of experience, K6nig and Becking present 212 species of owls, worldwide, a remarkable increase from the 133 species listed by Burton in 1973. Many island taxa, considered sub- ~ species by previous authorities, especially James L. Peters in volume 4 of his Birds of the World, have been elevated by K6nig and Becking to full species status. Differences in vocalisations, even though some of these “may be barely distinguishable to the human ear,” have played a role in these decisions, which still await further elaboration in peer-reviewed journals. For example, in the genus Tyto this book lists 17 species versus 8 in Burton; in Otus, 67 ver- sus 34. The hero of this book is Weick, who has 402 indi- vidual paintings of owls in 64 colour plates, 7 to 16 per plate. Many species are also depicted in flight and in downy (mesoptile) plumage. In addition, Weick has contributed 49 appropriate and attractive black- and-whiie sketches throughout the introductory text. This is a prodigious achievement, far surpassing the 80 paintings in Burton’s book, — though Burton had coloured photographs and this book has none. There is a succinct 19-page overview of owls, their specialized anatomy, then Topography, Food, Hunting, Behavior, Breeding, Vocalisations, and Systematics and Taxonomy. As a somewhat uncon- nected interpolation, Michael Wink and Petra Heidrich offer a 19-page account of molecular evo- lution and systematics, too technical for the average reader and perhaps too radical for some specialists. Beyond the 64 plates, the remaining 240 pages are devoted to species accounts, roughly one page per taxon. Each species account is illustrated by one map, with helpful arrows pointing to individual islands where local forms exist. The strongest feature is the emphasis on vocalisations. Although the authors acknowledge assistance from experts in six major museums and universities in North America, I doubt that any of them read the species accounts, or advised on appropriate North BOOK REVIEWS 193 American reference sources, which are grossly under-represented. (This may represent poetic justice for generations of North American authors who have ignored important European sources). I will offer five examples, the first three concerning the Great Horned Owl: (1) The error of naming the pale, northern sub- species, Bubo virginianus wapacuthu, is perpetuated, although Browning and Banks have shown conclu- sively that the initial specimen given this name was a Snowy Owl. (2) The statement “weighs about 1000 g,” although even males average more than this throughout most of the range, and a female may reach 2500 g. (3) The contention, “has to be consid- ered endangered in some regions,” when in fact is it one of the most adaptable and successful of owls anywhere. (4) It is not true that “even the palest indi- viduals” of the Great Horned Owl are “much darker than” the Snowy Owl. (5) The Barred Owl is said to be “expanding its range on the Atlantic side of North America, where it now overlaps with that of the Spotted Owl,” whereas Pacific side is the case. For a text that claims to emphasize the importance of vocalisations and behaviour, I was disappointed in a number of accounts. For example, the repetitive, regularly repeated whistle of the Northern Saw-whet Owl is poorly described as a “toot,” and the remark- able injury-feigning of some Long-eared Ow! adults near the nest is not mentioned. Occasionally, the text contradicts the map (e.g., Western Screech Owl map and Oaxaca Screech Owl text). Some taxonomic decisions were admittedly made on remarkably little evidence (e.g., moving the Itombwe Owl into the genus Tyto). Available longevity records from bird banding have been ignored. In summary, the illustrations are the main reason for purchasing this book, even for some who already own the Burton book. However, those interested only in North American owls are advised to buy the more accurate and informative North American Owls by Johnsgard; I understand a new edition of Johnsgard is in preparation. C. STUART HOUSTON 863 University Drive, Saskatoon, Saskatchewan S7N 0J8, Canada 194 THE CANADIAN FIELD-NATURALIST Vol. 115 A Field Guide to Reptiles and Amphibians of Eastern and Central North America By Roger Conant and Joseph T. Collins. 1998. Third Edition, Expanded. Houghton Mifflin Company, Boston. Xvili + 616 pages. illus. $29.95. Distributed in Canada by Thomas Allen & Son, Markham, Ontario. This attractive printing of THE classic eastern North American amphibian and reptile guide has a brash new cover with a composite of illustrations from the book: Red Salamander, Barking Treefrog, Green Iguana (lizard), Milk and King snakes, and Western Painted Turtle, and a line drawing of a gen- eralized tadpole and its mouthparts, thus readily dis- tinguishing it from the original third edition which appeared in 1991 (reviewed in The Canadian Field- Naturalist 105(4): 608-610) with a cover featuring only a single Pine Barrens Treefrog. But do not be mislead by the redesigned exterior or the epithet “expanded” attached to this bulky new version. True, it is bigger, “expanded” from xx and 450 pages to 616 pages, but the reader would search in vain for new information. The accounts apparently remain word-for-word from those of 1991, and the introduc- tory sections and content of distribution maps are similarly unaltered. What makes it a “new” book to the publisher is the resetting of the type, the addition to some accounts of individual colour photographs, and the text placement of distribution maps in eye- arresting bold primary colours, red, blue, yellow, etc., to distinguish the ranges of subspecies instead of the variations of black-and-white hatching and spotting of the previous editions. ENVIRONMENT Acts of Balance: Profits, People and Place Grant Copeland. 1999. New Society Publishers, Gabriola Island. 163 pp., illus. $17.95. In Canada we are repeatedly being told future suc- cess relies on increased globalization. Also, in order for Canadians to compete bigger is better, or so our financial institutes argue. But is this true in all cases? Grant Copeland provides a look at mainly British Columbia cases with which he has been involved over the length of his career. These case studies sug- gest for communities, local and small are the more beneficial and successful enterprises. A balance between economics and ecology must be determined by society which according to Copeland can be achieved within the existing infrastructure only slowly and with great frustration. Copeland states in the first chapter that the main purpose of the book is to offer examples of how we can live more environmentally sound and socially appropriate and thus preserve our cultural and eco- logical diversity. The author divides the book into The classic comparative Peterson format, which stamped the series as unique and made it an out- standing success, grouped illustrations of similar species together in equivalent postures with diagnos- tic marks indicated by bars. Although these plates are retained together as a front section, they are de- emphasized by the addition of the largely redundant new text photographs of animals in varied postures. Also regrettable is that the ease of comparison of grouped maps of earlier editions is lost. Maps only appear with text on individual taxa. No explanations or justifications for these innovations are given nor any indication that either author, especially, the senior one with this long experience in the functional importance of this guide, had any hand or say in the new design. This, presumably, is a product of a pro- duction staff with more of an eye for picture books than field-useful tools. Field naturalists may still prefer to search out a copy of the 1991 edition. It has identical information in significantly less bulk, an important factor in a volume to be actually taken on outings. Armchair naturalists may, however, find more visual appeal in the new colourful “expanded” edition. FRANCIS R. COOK Researcher Emeritus, Canadian Museum of Nature, Ottawa, Ontario K1P 6P4, Canada six parts covering economics, quality of life, cultural diversity, integration of global with local, and gov- ernment. All sections of the book rely heavily on Copeland’s working experience. References are pro- vided. The majority of the references are reports from boards, government, and consultants cited within the text. The layout on the whole is logical and highly readable. This book, which would be of benefit to many in positions of authority, provides, if nothing else, an alternative to present economic thinking. Individuals fighting for community survival would do well to read the book to gain insight for potential solutions. I would also recommend the book to the general read- er as educational reading. My copy contains several highlighted points for future reference. M. P. SCHELLENBERG 434 4th Ave SE, Swift Current, Saskatchewan S9H 3M1, Canada 2001 MISCELLANEOUS Linnaeus: Nature and Nation By Lisbet Koerner. 1999. Harvard University Press, Cambridge, Massachusetts. 298 pp., illus. U.S.$39.95. Koerner combines her biography of Carl Linnaeus, 1707-1778, with “a case study of the rela- tion between natural knowledge and political econo- my.” Raised in a pastor’s home, Linnaeus committed the book of Genesis to memory. “He regarded nature as the revealed works of God in space.” Daily he read Latin, particularly Ovid and Virgil. His nature writings were “beautiful ... part rhapsody, part oral epic, part elegy, and part pericope exegesis.” As Erasmus Darwin said, Linnaeus “may be said to have formed a language...” Linnaeus built his botanical nomenclature upon the foundations of Aristotle, John Ray, Joseph Pitton de Tournefort, and Sébastien Vaillant. His legacy is the universally-used classification system of plants and animals. Koerner praises its practicality, simplic- ity, its appeal to both learned men and novices, and hence its usefulness. In 1738, Linnaeus began his career as a medical doctor in private practice in Stockholm, specializing in the treatment of venereal disease, especially com- mon among French courtiers. Such expertise may have helped his appointment as chief physician to the Swedish navy the next year. He was apparently the first to describe “miner’s lung,” contributed to knowledge about diet and nutrition, described bat- tered children in 1752, and noted the serenity of well-swaddled Sami infants. In 1741 he moved back to his alma mater, Uppsala University, as professor of medicine. A year later he traded chairs to become professor of botany. Linnaeus was a complex man. Given his achieve- ments, his egotism was warranted. “No one,” he said proudly, “has so totally reformed an entire science.” His moods swung from euphoria to melancholia. Today he would be labelled as at least a cyclothymic personality, perhaps as a full-blown manic depres- sive. His enthusiasm was infectious; up to 300 peo- ple would attend the public nature walks he led twice a week. “A man of charisma and drive,” he was noted for his intelligence and charm. Linnaeus supervised, and usually wrote, 186 stu- dent dissertations. These disciples collected speci- mens for him throughout the known world; their voyages are depicted graphically on a world map. Thirty of their lives are summarized in Appendix B; five of them died abroad, including the two consid- ered by Koerner to have been his most brilliant disci- ples, Pehr Lofling, who died in Venezuela, and Pehr Forsskal, who died in Yemen before he could take up his post as professor of botany. Among the sur- vivors, Daniel Solander and Anders Sparrman cir- BOOK REVIEWS £95 cumnavigated the world with Captain Cook on his first and second voyages, respectively; Pehr Kalm spent over three years in America; Johan Peter Falck explored Russia; Carl Peter Thunberg was the first naturalist to visit Japan, where his memory is still revered. Their collections allowed Linnaeus to develop latitudinal, global, and alpine acclimatiza- tion theories, and to conceptualize the earth as a self- regulating superorganism (a precursor of the Gaia theory!). Linnaeus hoped to reproduce the economy of empire and colony within Sweden by growing colo- nial cash crops in the north. Increased productivity would thus replace former territorial conquest as a means for Sweden to prosper. Such attempts to use science to stimulate the economy failed. He grossly underestimated the number of plants and animals present in the world. About 10,000 plants had been named; he guessed that only another 10,000 remained undescribed, and that most should be tabulated in his lifetime. Once his Species Plantarum was published in 1753, one could, as Linnaeus proudly said, name plants “as easily as one names a person ... easy to remember, easy to say and write.” Systema Naturae in 1758 did the same for the animal kingdom. His classification of humans as Homo sapiens (after toying with the alternate name of Homo diurnus) was “original in the realm of sci- entific thinking.” Somewhat ambivalent, he said that “humankind was put here as a guest,” yet that nature was created for man. Linnaeus wrote anonymous but glowing reviews of his own publications in Stockholm newspapers; he padded his expense accounts. But he was gener- ous with food to the hungry, especially starving mothers and children. Petrus Artedi, Linnaeus’ fellow botanist and best friend, drowned in an Amsterdam canal at age 30, after a party. Koerner omits mention that Linnaeus rescued the landmark ichthyology classifications of Artedi and published them three years posthumous- ly. Throughout much of this biography, she offers a somewhat off-putting stress on “cameralism,” (the theoretical elaboration of fiscal and economic gover- nance), the central theme of her thesis and the title of a 1909 economics history book. In the 1880s, Sweden was the source of mass emi- gration, losing one percent of its population each year to America. Associated with conservative think- ing, the Linnaeus name was almost forgotten by the post-war social democratic government in Sweden. The Linnaeus Day holiday was cancelled. The 250th anniversary of his birth went almost unnoticed. He dwindled into a local hero. Koerner’s fascinating book, based on her Harvard 196 University dissertation, is highly recommended. Appendix A, a chronology, is a useful reference source. Appendix B, biographical references, was to me worth the price of the book. There are 48 pages of detailed footnotes and 19 pages of references. Unfortunately the names mentioned in the biographi- NEw TITLES Zoology *Albatrosses. 2000. By W. L. N. Tickell. Yale University Press, New Haven. 448 pp., illus. + 52 plates. U.S. $60. Annotated bibliography of quarternary vertebrates of North America. 2001. By C. R. Harington. University of Toronto Press, Toronto. c360 pp., $120. The bird almanac: the ultimate guide to essential facts and figures of the world’s birds. 1999. By D. M. Bird. Firefly, Willowdale, Ontario. xvii + 460 pp., illus. $19.95. +The archaeology of animal bones. 2000. By T. O’Connor. Texas.A & M University Press, College Station. 244 pp., illus. U.S. $ 34.95. *Biology of plethodontid salamanders. 2000. Edited by B.C. Bruce, R-G: Jacger, “and LL.D. Houck: Kluwer/Academic Press/Plenum, New York. xiii + 485 pp.. illus. U.S. $195. 7Bird census techniques. 2000. By C. J. Bibby, N. D. Burgess, D. A. Hill, and S. H. Mustoe. 27 edition. Academic Press, San Diego. xvii + 302 pp., illus. *Birds of Europe. 2000. By K. Mullarney, L. Svensson, D. Zetterstrom, and P. J. Grant. Princeton University Press, Princeton. 400 pp., illus. Cloth U.S. $39.50: paper US. $29.95. «Birds of the World: a checklist. 2000. By J. F. Clements. Ibis Publishing, Vista, California. xx + 867 pp., U.S. $39.95. Birds of Nepal. 2000. By R. Grimmett, C. Inskipp, and T. Inskipp. Princeton Field Guides. Princeton University Press, Princeton. 288 pp., illus. U.S. $29.95. Bugs of Alberta. 2000. By J. Acron. Lone Pine Publishing, Edmonton. 160 pp., illus. $14.95. *+Checklist of the birds of northern South America. 2000. By C. Rodner, M. Lentino, and R. Restall. Yale University Press, New Haven. 136 pp., U.S. $23.50. *Conservation priorities for the amphibians and rep- tiles of Canada. 2000. By D. Seburn and C. Seburn. World Wildlife Fund Canada and Canadian Amphibian and Reptile Conservation Network, WWF, Toronto. 92 pp.. Free PDF copies available at www.wwfcanada.org _ (library) THE CANADIAN FIELD-NATURALIST Vol. 115 cal references are not cited in the otherwise useful index. C. STUART HOUSTON 863 University Drive, Saskatoon, Saskatchewan S7N 0J8 *Cuckoos, cowbirds, and other cheats. 2000. By N. B. Davies. Poyser, London, England. ix + 310 pp., illus. *A field guide to birds of the Indian subcontinent. 2000. By K. Kazmierezak. Yale University Press, New Haven. 352 pp.., illus. + map. U.S. $32.50. Fire ants. 2000. By S. W. Taber. Texas A & M University Press, College Station. 368 pp., illus. U.S. $29.95. *Fossil snakes of North America: origin, evolution, dis- tribution, paleoecology. 2000. By J. A. Holman. Indiana University Press, Bloomington, xi + 357 pp., illus. + plates. U.S. $ 69.95. *Gatherings of angels: migrating birds and their ecolo- gy. 1999. Edited by K. P. Able. Comstock Books, Corneil University Press, Ithaca. xi + 193 pp., illus. U.S. $29.95. *A field guide to the birds of India, Pakistan, Nepal, Bangladesh, Bhutan, Sri Lanka, and the Maldives. 1999. By R. Grimmett, C. Inskipp, and T. Inskipp. Princeton University Press, Princeton. 888 pp., illus. +Heron Conservation. 2000. Edited by J. A. Kushlan and H. Hafner. Academic Press, San Diego. xvi + 480 pp., illus. Islands of hope: lessons from North America’s great wildlife sanctuaries. 1999. By P. Manning. John F. Blair, Winston-Salem, North Carolina. 211 pp., illus. U.S. $15.95. *Kingbird highway. 2000. By K. Kaufman. Mariner Books, Houghton Mifflin, Wilmington, Massachusetts. 318 pp., illus. U.S. $13. Living on the wind: across the hemisphere with migra- tory birds. 1999. By S. Weidnesaui. North Point Press, New York. xii + 420 pp., illus. U.S. $26. *Manitoba birds. 2000. By A. Bezener and K. DeSmet. Lone Pine Publishing, Edmonton. 176 pp., illus. $17.95. +A manual for wildlife radio tagging. 2001. By R. E. Kenward. Academic Press, San Diego. x + 311 pp.., illus. *The nature of frogs: amphibians with attitude. 2000. By H. Parsons. Greystone Press, Vancouver. 112 pp., illus. Cloth $34.95; paper $24.95. 2001 +The nature of great apes: our next of kin. 2000. By M. A. Gilders. Greystone Press, Douglas and McIntire, Vancouver. vii + 104 pp., illus. $34.95. *Reproductive biology of bats. 2000. Edited by E. G. Crichton and P. H. Krutzsch. Academic Press, San Diego. xi + 510 pp., illus. +Research techniques in animal ecology: controversies and consequences. 2000. Edited by L. Boitani and T. K. Fuller. Columbia University Press, New York. xxxii + 442 pp., illus. U.S. $32. +The return of the wolf: reflections on the future of wolves in the northeast. 2000. Edited by J. Elder. University Press of New England, Hanover, New Hampshire. 175 pp. U.S. $24.95. Shrikes and bush-shrikes, including wood-shrikes, helmet-shrikes, flycatcher-shrikes, philentomas, batis- es, and wattle-eyes. 2000. By T. Harris. Princeton University Press, Princeton. 392 pp., illus. U.S. $49.50. *Snipes of the western palearctic. 2000. By R. Rouxel. Eveil Nature, Saint Yrieux sur Charente, France. 304 pp.., illus. English or French 215 FF. *+Threatened birds of the world. 2000. Edited by A. J. Stattersfield and D. R. Capper. Birdlife International and Lynx Edicions, Barcelona, Spain. xii + 852 pp., illus. U.S. $115. *Turtle conservation. 2000. Edited by M. W. Klemens. Smithsonian Institute Press, Washington. xv + 334 pp., illus. U.S: $ 57.75. Botany +Bryophyte biology. 2000. Edited by A. J. Shaw and B. Goffinet. Cambridge University Press, New York. x + 476 pp., illus. Cloth U.S. $100; paper U.S. $35.95. +Contemporary plant systematics. 2000. By D. W. Woodland. 3 edition. Andrews University Press, Berrien Springs, Michigan. xiv + 370 pp., illus. + photo CD. U.S. $64.95. *Food webs and container habitats: the natural history and ecology of phytotelmata. 2000. By R. L. Kitching. Cambridge University Press, New York. xiii + 431 pp., illus. U.S. $100. 7*Green plants: their origin and diversity. 2000. By P. R. Bell and A. R. Hemsley. 2"4 edition. Cambridge University Press, New York. 349 pp., illus. Cloth U.S. $90; paper U.S. $31.95. +Magical mushrooms, mischievous molds. 1998. By G. W. Hudler. Princeton University Press, Princeton. xv + 248 pp., illus. U.S. $14.95. *Phycology. 1999. By R. E. Lee. 3™ edition. Cambridge University Press, New York. x + 614 pp., illus. Cloth U.S. $ 100; paper U.S. $44.95. BOOK REVIEWS 197 *Spatial pattern analysis in plant ecology. 1999. By M. R. T. Dale. Cambridge University Press. New York. x + 326 pp., illus. Cloth U.S. $69.95; paper U.S. $35.95. Trees: their natural history. 2000. by P. Thomas. Cambridge University Press, New York. x + 286 pp., illus. Cloth U.S. $64.95; paper U.S. $24.95. F Environment Don’t breath the air: air pollution and U. S. environ- mental politics, 1945 — 1970. 2000. By S. H. Dewey. Texas A & M University Press, College Station. 336 pp. US. $39.95. *Fatal consumption: rethinking sustainable develop- ment. 2000. Edited by R. E. Woollard and A. S. Ostry. UBC Press, Vancouver. ix + 270 pp., illus. Cloth $75; paper $29.95. +The geometry of ecological interactions: simplifying spatial complexity. 2000. Edited by U. Dieckmann, R. Law, and J. A. J. Metz. Cambridge University Press, New York. xiv + 565 pp., illus. U.S. $74.95. +The global environment in the twenty-first century: prospects for international co-operation. 2000. Edited by P. S. Chasek. United Nations University Press, Tokyo. x + 465 pp., illus. U.S. $39.95. *Pandora’s poison: chlorine, health, and a new environ- mental strategy. 2000. By J. Thornton. MIT Press, Cambridge. xii + 599 pp. U.S. $34.95. *Reading the entrails: an Alberta ecohistory. 1999. By N. C. Conrad University of Calgary Press, Calgary. xviii + 197 pp. $19.95. *Seasons of the Arctic. 2000. Photographs by P. Nicklen, text by H. Brody. Greystone Books, Vancouver. 128 pp. $45. +Simulating ecology and evolutionary systems in C. 2000. By W. Wilson. Cambridge University Press, New York. xv + 301 pp., illus. Cloth U.S. $85; paper U.S. $29.95. Small creatures and ordinary places: essays on nature. 2000. By A. M Young. University of Wisconsin Press, Madison. 288. pp., illus. Cloth U.S. $50; paper U.S. $19.95. *Something new under the sun: an environmental histo- ry of the twentieth century world. 2000. By J. R. McNeill. W. W. Norton, New York. ix + 421 pp., illus. U.S. $29.95. *Voices for the watershed: environmental issues in the Great Lakes — St. Lawrence Drainage Basin. 2000. Edited by G. G. Beck and B. Littlejohn. McGill-Queen’s University Press, Montreal. xiii + 299 pp., illus. $39.95. + Wetland ecology: principles and conservation. 2000. By Paul A. Keddy. Cambridge University Press, New 198 York. xiv + 614 pp., illus. Cloth U.S. $140; paper U.S. $52.99; Miscellaneous Glacier ice. 2000. By A. Post and E. R. La Chapelle. University of Toronto Press, Toronto. 160 pp., illus. $40. Hidden Canada: an intimate travelogue. 2000. By N. Rauvin. Red Deer Press, Calgary. 224 pp., $18.95. *Yellowstone and the great west: journals, letters, and images from the 1871 Hayden expedition. 1999. University of Nebraska Press, Lincoln. xxix + 315 pp., illus. U.S. $29.95. Books for Young Naturalists The animal kingdom: a guide to vertebrate classifica- tion and diversity. 2000. By K. Whyman. Raintree Steck- Vaughn, Austin, Texas 48 pp., illus. U.S. $25.69. Animals in cold places; Animals in hot places; Animals in rivers and lakes; Animals in oceans; Animals in trees; and Animals on plains and prairies. 2000. By M. Butterfield. Rainbow, Austin, Texas. Each 32 pp., illus. Wis 7527.83: Arctic alphabet: exploring the north from A to Z. 1999. By W. Lynch. Firefly, Willowdale, Ontario. 32 pp., illus. Ecosystems and environment. 2000. By A. Fullick. Heinemann, Des Plaines, Illinois. 32 pp., illus. U.S. $24.22. Elephant. 2000. By W. Travers. Raintree Steck-Vaughn, Austin, Texas 48 pp., illus. U.S. $25.69. Eugenie Clark: adventures of a shark scientist. 2000. By E. R. Butts. Linnet Books, New Haven, Connecticut. x1 + 107 pp., illus. U.S. $19.50. THE CANADIAN FIELD-NATURALIST Vol. 115. Kangaroos. 2000. By D. Burt. Carolrhoda, Minneapolis. 48 pp., illus. U.S. $22.60. Killer whale. 2000. By M. Cawardine. Raintree Steck- Vaughn. 48 pp., illus. U.S. $25.69. Lion. 2000. By B. Jordan. Raintree Steck-Vaughn, Austin, Texas. 48 pp., illus. U.S. $25.69. Lions. 2000. By K. Darling. Carolrhoda, Minneapolis. 48 pp., illus. U.S. $22.60. My nature book. 2000. By A. Olmstead. Pajaro, Lafayette, California. 176 pp., illus. U.S. $17.95. Penguins. 1999. By W. Lynch. Greenwillow Books, New York 32 pp., illus. Cloth U.S. $19.95; paper U.S. $9.95. The plant kingdom: a guide to the plant classification and biodiversity. 2000. By T. Greenaway. Raintree Steck-Vaughn, Austin, Texas. 48 pp., illus. U.S. $25.69. Reptile rescue. 2000. By P. Thomas. Twenty-first Century Books, New York. 64 pp., illus. U.S. $23.90. Slugs. 2000. By A D. Fredericks. Lerner, Minneapolis. 48 pp., illus. U.S. $22.60. They walk the earth: the extraordinary travels of ani- mals on land. 2000. By S. Simon. Harcourt Brace, New York. 40 pp., illus. U.S. $17. Tiger. 2000. By V. Thapar. Raintree Steck-Vaughn, Austin, Texas. 48 pp., illus. U.S. $25.69. Wildlife winners: the peregrin falcon — endangered no more. 2000. By M. Priebe. Mindfull Publishing, Norwalk, Connecticut. 32 pp., illus. U.S. $15.95. *Book-review Assigned +Book available for review TABLE OF CONTENTS (concluded) Updated status of the Northern Madtom, Noturus stigmosus, in Canada : ERLING HOLM and NICHOLAS E. MANDRAK Status of the Bluntnose Minnow, Pimephales notatus, in Canada J. HOUSTON Status of the Texada Stickleback species pair, Gasterosteus spp., in Canada J. HOUSTON Updated status of the Central Stoneroller, Campostoma anomalum, in Canada E. HOLM and E. J. CROSSMAN Notes Common Raven, Corvus corax, observed taking egg from a Common Loon, Gavia immer, nest ROBERT ALVO and PETER J. BLANCHER Two cases of infanticide in a Red Fox, Vulpes vulpes, family in southern Ontario VALERIA VERGARA An unusual record of a White-tailed Deer, Odocoileus virginianus, in the Northwest Territories ALASDAIR M. VEITCH Northern Gannet, Morus bassanus, nesting on Whitehorse Island, New Brunswick SEAN CORRIGAN and ANTONY W. DIAMOND Unusual Harlequin Duck, Histrionicus histrionicus, nest site discovered in central Labrador Tony E. CHUBBS, BRUCE MACTAVISH, KEITH ORAM, PERRY G. TRIMPER, KATHY KNOX, and R. IAN GOUDIE ‘Standing over” and “hugging” in wild Wolves, Canis lupus L. DAviD MECH Limb mutilations in Snapping Turtles, Chelydra serpentina RAYMOND A. SAUMURE News and Comment Notices: American Birding Association Ludlow Griscom Award for Publications in Field Ornithology: W. Earl Godfrey, June 2000 — Froglog: Newsletter of the Declining Amphibian Populations Task Force (42) — Marine Turtle Newsletter (91) — Canadian Species at Risk November 2000 — The Boreal Dip Net 5(1) — Sea Wind: Bulletin of Ocean Voice International 14(4) — Recovery (17) — Ontario Natural Heritage Information Centre Newsletter 6(1) Book Reviews Zoology: The Birds of British Columbia Volume 3 — A Guide to the Birds of the West Indies — A Guide to the Birds of India, Pakistan, Nepal, Bangladesh, Bhutan, Sri Lanka, and the Maldives — Trogons and Quetzals of the World — Research and Management of the Brown-headed Cowbird in Western Landscapes —- Owls: A Guide to the Owls of the World — A Field Guide to Reptiles and Amphibians of Eastern and Central North America : Environment: Acts of Balance: Profits, People and Place Miscellaneous: Linnaeus: Nature and Nation New Titles Mailing date of the previous issue 114(4): 4 May 2001 138 145 152 157 168 170 172 176 177 17e 182 185 187 194 195 196 THE CANADIAN FIELD-NATURALIST Volume 115, Number 1 Articles Merriam’s Shrew, Sorex merriami, and Preble’s Shrew, Sorex preblei: two new mammals for Canada DAvID W. NAGORSEN, GEOFFREY G. W. SCUDDER, DAviD J. HUGGARD, HEATHER STEWART, and NICK PANTER Effects of Beaver, Castor canadensis, herbivory on streamside vegetation in a northern Ontario watershed D. M. BARNES and A. U. MALLIK Comparison of parental roles in male and female Red Foxes, Vulpes vulpes, in southern Ontario VALERIA VERGARA Biodiversite microfongique du Fagus grandifolia dans une forét ancienne: bioindicateurs et structure mycosociologique VLADIMIR VUJANOVIC et JACQUES BRISSON Influence de parametres climatiques sur les patrons d’activite saisonniers et journaliers du lievre d’ Amerique, Lepus americanus, en semi-liberte JEROME THEAU et JEAN FERRON Fall food habits and reproductive condition of Fishers, Martes pennanti, in Vermont KYLE VAN Why and WILLIAM M. GIULIANO Relative abundances of forest birds of prey in western Newfoundland JOHN W. GOSSE and WILLIAM A. MONTEVECCHI Distances moved by small woodland rodents within large trapping grids JEFF BOWMAN, GRAHAM J. FORBES, and Tim G. DILWORTH Abundance of stream invertebrates in winter: seasonal changes and effects of river ice MELANIE D. MARTIN, RICHARD S. BROWN, DAVID R. BARTON, and GEOFF POWER Establishment of a breeding population of Canada Geese, Branta canadensis, in southern Quebec JEAN-FRANCOIS GIROUX, JOSEE LEFEBVRE, LUC BELANGER, JEAN RODRIGUE, and STEPHANE LAPOINTE Edge-related nest predation associated with a retention of residual trees in harvested hardwood stands RICHARD H. YAHNER, AMANDA D. RODEWALD, and SUSAN C. TALBOTT The vascular flora of Akimiski Island, Nunavut Territory, Canada C. S. BLANEY and P. M. KOTANEN Effects of reopening hunting on survival of White-tailed Deer, Odocoileus virginianus, — in the Bas-Saint-Laurent region, Québec JEAN LAMOUREUX, MICHEL CRETE, and MATHIEU BELANGER Slumping activity and forest vegetation along the northeastern shore of Waskesiu Lake, Prince Albert National Park, Saskatchewan D. H. DE BOER and O. W. ARCHIBOLD Rare and endangered fishes and marine mammals of Canada: COSEWIC Fish and Marine Mammal Subcommittee Status Reports: XIII. R. R. CAMPBELL Status of the White-beaked Dolphin, Lagenorhynchus albirostris, in Canada . JON LIEN, DAWN NELSON, and DONG JIN HAI Updated status of the Vancouver Island Lake Lamprey, Lampetra macrostoma, in Canada R. J. BEAMISH Status of the sympatric smelt (genus Osmerus), populations of Lake Utopia, New Brunswick Eric B. TAYLOR (continued on inside back cover) — ISSN 0008-3550 20( 106 | 115 1188 127 | 1318 at : i ee et tS “ The CANADIAN FIELD-NATURALIST | Published by THE OTTAWA FIELD-NATURALISTS’ CLUB, Ottawa, Canada CE LIE AID Volume 115, Number 2 pA coe April-June 2001 The Ottawa Field-Naturalists’ Club FOUNDED IN 1879 Patrons Her Excellency The Right Honourable Adrienne Clarkson, C.C., C.M.M., C.D. Governor General of Canada His Excellency John Ralston Saul, C.C. The objectives of this Club shall be to promote the appreciation, preservation and conservation of Canada’s natural heritage; to encourage investigation and publish the results of research in all fields of natural history and to diffuse infor- mation on these fields as widely as possible; to support and cooperate with organizations engaged in preserving, maintain- ing or restoring environments of high quality for living things. Honorary Members Robert W. Nero Edward L. Bousfield Bruce Di Labio George F. Ledingham E. Franklin Pope Donald M. Britton R. Yorke Edwards John A. Livingston William O. Pruitt, Jr. Irwin M. Brodo Anthony J. Erskine Stewart D. MacDonald Joyce and Alan Reddoch William J. Cody John M. Gillett Hue N. MacKenzie Mary E. Stuart Francis R. Cook W. Earl Godfrey Theodore Mosquin Sheila Thomson Ellaine Dickson C. Stuart Houston Eugene G. Munroe 2001 Council President: Eleanor Zurbrigg Ronald E. Bedford Francis R. Cook Gary McNulty es . ‘ Rosanne Bishop Barbara Gaertner David W. Moore tice eee Rex tou Fenja Brodo Anthony Halliday Rita Morbia Recording Secretary: Ken Allison Irwin Brode Terry Higgins Robert Roach Treasurer: Frank Pope John Cameron David Hobden Stanley Rosenbaum William J. Cody Beverly McBride David Smythe Dorothy Whyte To communicate with the Club, address postal correspondence to: The Ottawa Field-Naturalists’ Club, Box P.O. Box 35069, Westgate P.O. Ottawa, Canada K1Z 1A2, or e-mail: ofnc @achilles.net. For information on Club activities telephone (613) 722-3050 or check http//www.achilles.net/ofnc/index.htm The Canadian Field-Naturalist The Canadian Field-Naturalist is published quarterly by The Ottawa Field-Naturalists’ Club. Opinions and ideas expressed in this journal do not necessarily reflect those of The Ottawa Field-Naturalists’ Club or any other agency. We acknowledge the financial support of the Government of Canada toward our mailing cost wee the Publication Assistance Program (PAP), Heritage number 09477. Editor: Dr. Francis R. Cook, R.R. 3, North Augusta, Ontario KOG 1RO; (613) 269-3211; e-mail: feook @achilles.net Copy Editor: Wanda J. Cook Business Manager: William J. Cody, P.O. Box 35069, Westgate P.O. Ottawa, Canada K1Z 1A2 (613) 759-1374 Book Review Editor: Dr. J. Wilson Eedy, R.R. 1, Moffat, Ontario LOP 1J0; e-mail: edith@netcom.ca Associate Editors: Robert R. Anderson Robert R. Campbell Brian W. Coad W. Earl Godfrey Charles D. Bird Paul M. Catling Anthony J. Erskine William O. Pruitt, Jr. Chairman, Publications Committee: Ronald E. Bedford All manuscripts intended for publication should be addressed to the Editor and sent by postal mail (no courier, no post requiring signature on delivery). Exception: book reviews should go directly to Book Review Editor. Subscriptions and Membership Subscription rates for individuals are $28 per calendar year. Libraries and other institutions may subscribe at the rate of $45 per year (volume). The Ottawa Field-Naturalists’ Club annual membership fee of $28 (individual) $30 (family) $50 (sustaining) and $500 (life) includes a subscription to The Canadian Field-Naturalist. All foreign subscribers and mem- bers (including USA) must add an additional $5.00 to cover postage. The club regional journal, Trail & Landscape, covers the Ottawa District and Local Club events. It is mailed to Ottawa area members, and available to those outside Ottawa on request. It is available to Libraries at $28 per year. Subscriptions, applications for membership, notices of changes of address, and undeliverable copies should be mailed to: The Ottawa Field-Naturalists’ Club, P.O. Box 35069, Westgate P.O. Ottawa, Canada K1Z 1A2. Canada Post Publications Mail Agreement number 40012317. Return Postage Guaranteed. Date of this issue: April-June 2001 (November 2001). Cover: Trillium ovatum Pursh variety hibbersonii (Taylor et Szczawinski) Douglas et Pojar, variety nova. See note by George Douglas and Jim Pojar, page 343. Photo courtesy of George W. Douglas. The Canadian Field-Naturalist Volume 115, Number 2 April—June 2001 Tiger Salamander, Ambystoma tigrinum, Movements and Mortality on the Trans-Canada Highway in Southwestern Alberta ANTHONY P. CLEVENGER!:2, MIKE McIvor3, DIANE McIvor?, BRYAN CHRUSZCZ* and KARI GUNSON?4 ‘Faculty of Environmental Design, University of Calgary, 2500 University Avenue, N.W., Calgary, Alberta T2N 1N4 Canada. 2Mailing address: 625 Fourth Street #3, Canmore, Alberta T1W 2G7 Canada 3Box 1693, Banff, Alberta TOL O0CO Canada 4Parks Canada, Box 900, Banff, Alberta TOL OCO Canada Clevenger, Anthony P., Mike Mclvor, Diane Mclvor, Bryan Chruszcz, and Kari Gunson. 2001. Tiger Salamander, Ambystoma tigrinum, movements and mortality on the Trans-Canada Highway in southwestern Alberta. Canadian Field-Naturalist 115(2): 199-204. We report on the mid-season movements and mortality of Tiger Salamanders (Ambystoma tigrinum) along the Trans- Canada highway in Kananaskis Country, Alberta. The highway was surveyed for road-killed animals between April and November during 1997, 1998, and 1999. Road-killed Tiger Salamanders were collected on one day in 1998 and eight days in 1999. A minimum of 183 Tiger Salamander mortalities were recorded on a 1.05 km section of highway. The mean snout-to-vent length of eight road-killed salamanders was 103.1 cm. Movement was concentrated in one 300 m section of highway, primarily in one direction, and related to heavy rainfall events and warm weather. It was not clear whether move- ments were pre-breeding or post-breeding or why there was a sudden eruption in movements away from Chilver Lake. Key Words: Ambystoma tigrinum, Tiger Salamander, migration, mortality, roads, Alberta. The Tiger Salamander (Ambystoma tigrinum) is the most widely distributed amphibian in North America (Stebbins 1985). Populations commonly designated Ambystoma t. melanostictum are found in Canada in the province of Alberta south of Edmon- ton, from the eastern slope foothills to southern Saskatchewan and east to Manitoba (Stebbins 1985; Hooper 1992; Russell and Bauer 2000). Tiger Salamanders breed in shallow lakes, ponds, and backwaters in a wide array of habitats and are found at elevations of up to 2800 m (Petranka 1998). Over its entire North American range, the species breeds in almost every month of the year, depending on subspecies, geographic location, latitude and eleva- tion (Petranka 1998), but mating in the northern lati- tudes generally occurs after early spring migration to breeding sites. Rarely seen in the open except during breeding season, Tiger Salamanders are primarily nocturnal and may be active from early spring to early autumn in Alberta. Movement patterns of ambystomatid salamanders are particularly unclear since much of their post- transformation life is spent in subterranean burrows and most congregate only briefly at aquatic breeding sites (Duellman and Trueb 1986). Seasonal patterns of migration from terrestrial environments to breed- ing areas are well documented (Sever and Dineen 1978; Semlitsch 1985; Sexton et al. 1990), yet move- ments after the breeding season are poorly under- stood (Hairston 1987; Whiteman et al. 1994). In this paper we report on the spatial and temporal pre- sumed post-breeding movements from road-killed Tiger Salamanders on a section of the Trans-Canada highway in Kananaskis Country, Alberta. Study area and methods The study area is in the Montane ecoregion and is composed of grasslands interspersed with forests of Lodgepole Pine (Pinus contorta), White Spruce (Picea glauca) and Aspen (Populus tremuloides). Situated in the Bow River Valley, the terrain has lit- tle topographic relief within the valley floor. A small pond and a lake (Chilver Lake) are located on the south side of the Trans-Canada highway, while another lake (Middle Lake) is on the north side of the highway (Figure 1). The distance between Chilver Lake and Middle Lake is approximately 1.7 km. Tiger Salamanders have been observed recently in all three water bodies (A. Clevenger, personal observation; H. Dempsey, personal communication). Earlier records of Tiger Salamander sightings in the Chilver Lake area were published by Salt (1979). 199 200 Between the months of April and November in 1997, 1998, and 1999, we systematically surveyed roads by vehicle in Banff National Park and adjacent Kananaskis Country, collecting data on road mortali- ty of small and medium sized vertebrates. One of two routes surveyed the Trans-Canada highway from the Banff townsite to the junction of Highway 40. Surveys were alternated each day, commenced less than 1 hr after sunrise, and were conducted by two observers, one driving 10 km/hr below the posted speed limit, while the other searched for road-killed salamanders and other vertebrates on the road pave- ment. We observed road-killed Tiger Salamanders on the Trans-Canada highway west of the town of Seebe (51°04’N, 115°04’W; Figure 1) and 45 km east of the town of Banff on the following dates: 29 July 1998 at 0729 h (1 hr, 28 mins after sunrise), 11 August 1999 at 0724 h (56 mins after sunrise), 12 August 1999, 13 August 1999 at 0754 h (1 hr, 30 mins after sunrise), 14 August 1999, 15 August 1999, 30 August 1999 at 0745 h (55 mins after sun- rise), 3 September 1999 at 0807 h (1 hr, 7 mins after sunrise), and 11 September 1999. The highway at the location of all road-kills was 43 m wide and consist- ’ ed of four lanes of traffic, an open grass-covered median. To assess whether road-killed salamanders were traveling in a specific direction we identified the high-kill concentrations on the eastbound and west- bound lanes. Using the westernmost extent of road- kills on eastbound and westbound lanes as reference points we plotted the location of the remaining sala- mander road-kills. All points were obtained using a global positioning system unit and differentially- corrected (Trimble Navigation Ltd., Sunnyvale, California, USA). The distance from each road- killed salamander to the origin was recorded for all of the kills on both sides of the highway. These dis- tances were used to add the salamander road-kills to an ArcView GIS point theme (Earth Systems Research Institute 1998). We described the spatial distribution of the salamander road-kills using spatial statistics software (Levine 1999). A cluster analysis was performed using the nearest-neighbor hierarchi- cal clustering algorithm to determine where the majority of kills occurred in the eastbound and west- bound lanes. An alpha value of 0.05 was used to locate clusters not produced by chance alone. We used a sample size of 92 (eastbound) and 19 (west- bound) road-kills to produce a cluster. The mean center of each cluster was calculated and we ran a line between the two centers to approximate the direction the salamanders traveled as they moved across the highway. Results and Discussion Mortality On the morning of 29 July 1998, we found and collected one road-killed Tiger Salamander in the THE CANADIAN FIELD-NATURALIST Vol. 115 westbound lane (north side) of the Trans-Canada highway west of the town of Seebe (Figure 1). The following year on the morning of 11 August 1999, we discovered a mass movement of Tiger Salamanders across the Trans-Canada highway at the same location as the single individual the year before. On this day we counted a total of 68 dead Tiger Salamanders on the highway. We collected only eight salamanders as most had been flattened and dismembered on the pavement. The total number of salamanders killed on the roadway was obviously much higher. While counting and removing the road- killed specimens from the pavement, we observed additional Tiger Salamanders approaching the road and unsuccessfully attempting to cross it. The fol- lowing day (12 August 1999) we did not survey the highway but returned to the site and found two live Tiger Salamanders and estimated at least 50 or more dead on the highway pavement. On 13 August 1999 we carried out the survey and encountered 23 Tiger Salamanders killed on the Trans-Canada highway at the same location and noted that the salamanders were approaching the road and trying to cross it. The next day (14 August 1999) we did not survey the road but returned to the site and found one live sala- mander and estimated 20 or more dead on the road. On 15 August 1999 we found two dead Tiger Salamanders on the Trans-Canada highway, four that were several days old on a frontage road 25 m south of the highway and we found no salamanders live or dead while driving a two-lane road in Bow Valley Provincial Park situated 300 m north of the highway and parallel to it. Two weeks later, on 30 August 1999, we counted a total of 18 Tiger Salamanders dead on the highway at the same location as previous road-kills. While dead specimens were being collected off the roadway live ones again were observed attempting to cross the busy highway. While conducting the survey on 3 September 1999 we counted one road-killed Tiger Salamander on the Trans-Canada highway at the same site as those recorded earlier. The final observa- tion of salamander movement on the highway was on 11 September 1999 when one unsuccessfully attempted to cross the highway northbound. The main pulses in Tiger Salamander movements across the Trans-Canada highway occurred during two distinct periods: 11-14 August 1999 (n= 163 mortalities) and on 30 August 1999 (n= 20). Overall we documented a minimum of 183 Tiger Sala- mander mortalities on the highway, all within a 1.05 km section, during a 30-day period. At a rate of 24—40 vehicles/hour, Kuhn (1987) found that 50% of a cohort of migrating Common Toads (Bufo bufo) were killed. At 26 vehicles/hour the estimated road crossing survival rate of Common Toads was zero in another study (Heine 1987). The high traffic volumes on this section of the Trans- Canada highway during summer (mean daily traffic 2001 CLEVENGER, MclIvor, McIvor, CHRUSZCZ, AND GUNSON: TIGER SALAMANDER MOVEMENTS Middle Lake oe ___ Trans-canao* A Chilver Lake Complex 201 Seebe Interchange LEGEND Salamander Roac-kill Cluster Center 500 Meters Presumed Direction of Travel FiGuRE 1. Location of Tiger Salamander road-kills along the Trans-Canada highway, Kananaskis Country, Alberta, Canada. volume = 21 450 vehicles/day; Parks Canada Highway Service Centre, Calgary, Alberta) coupled with the slow movements of ambystomatids would result in exceedingly few individuals surviving the highway traverse. We were unable to document how many or what proportion of salamanders approach- ing the highway actually crossed safely. We collected and measured the one specimen from 29 July 1998 and 7 of the 8 specimens collect- ed on 11 August 1999. One individual was dismem- bered such that accurate measurements could not be made. The mean snout-to-vent (SVL) length of the eight salamanders was 103.1 cm (SD = 8.8), ranging from 90.0 cm to 120.0 cm. The sex of the individuals could not be determined externally. Based on the SVL the collected specimens appeared to be newly- transformed adults but none were dissected to deter- mine either maturity or sex. We did not note any gill stubs that would have indicated transformation from larvae was just being completed as has been observed in some migrations elsewhere (F. R. Cook, personal communication). Direction of travel and mortality distribution In a similar report, Duellman (1954) found late- season eastern Tiger Salamanders moving randomly across a two-lane Michigan roadway; but he did not detect a definite migration in any one direction. The live Tiger Salamanders we observed approaching the highway were predominantly moving north- bound across the Trans-Canada highway. We base this on our observations of Tiger Salamander move- ment and the high number of road-killed salaman- ders on the south side of the highway as opposed to the north side. During the first pulse (11-14 August 1999), we counted 73 (79%) of 92 road-killed sala- manders on the south side, whereas 19 (21%) were picked up on the north side. Similarly, during the second pulse (30 August—3 September 1999) all 19 salamanders recovered were killed on the south side. Because we did not sample other areas periph- eral to Chilver Lake, we cannot assume that sala- manders only moved in the direction of the high- way; they could have dispersed in many different directions from the lake. Nonetheless, the majority 202 of salamander movements appeared to originate from Chilver Lake. Tiger Salamander kills were distributed on both sides of the highway for a distance of 1.05 km; how- ever, most kills were concentrated in a 300 m section (Figure 1). Movement was concentrated in one local- ized area and primarily in one direction suggesting that seasonal migrations breeding or otherwise were occurring. Previous studies have reported that breed- ing and non-breeding season movements of Tiger Salamanders are triggered by environmental factors, in particular precipitation and temperature (Petranka 1998; Whiteman et al. 1996; Duellman and Trueb 1986; Fowler 1935). Pre-breeding movements would normally take place several months earlier and would have been detected during previous road sur- veys. Mid-season movement between ponds could explain the phenomenon, although it is unclear what triggered the salamanders to move during this one year and not in previous years. The centers of road-kill clusters for east and west- bound lanes were offset, indicating that salamanders were apparently crossing the highway at an oblique angle and not straight across. In Figure 1, the arrow implying the general direction of travel shows that movement was in the direction of Middle Lake, but not directly aligned. It was noteworthy that the great- est concentration of salamander movements across the Trans-Canada highway occurred at the highest point in the road and where the highway passed clos- est to Chilver Lake. Weather Although there is universal agreement that particu- lar meteorological conditions elicit migratory behav- ior in Tiger Salamanders, few studies have noted what the specific ambient measures are, or tested their rela- tionships with movements (Semlitsch 1983a; Duellman 1954). In South Carolina the number of migrating adult Eastern Tiger Salamanders (Ambystoma tigrinum tigrinum) was consistently cor- related with total amount of daily precipitation, the amount of minimum daily rainfall amount being asso- THE CANADIAN FIELD-NATURALIST Vols ciated with movements was 13 mm (Semlitsch 1983a; Table 1). The minimum and maximum daily tempera- tures when more than 200 Eastern Tiger Salamanders were observed crossing a Michigan highway during autumn ranged from 5—9°C and 11—14°C, respectively (Duellman 1954). In our western study area, during the first and most important pulse in salamander movement across the Trans-Canada highway, 17 mm of precipitation fell in less than 24 hours and the mini- mum and maximum temperatures were 6.5°C and 17.0°C, respectively (University of Calgary Field Station, Kananaskis Country, Alberta). We obtained the meteorological records for the years we observed Tiger Salamander movements (1998, 1999) and months they are found active (April to September inclusive). To further investi- gate weather effects we subjectively assigned a mini- mum amount of daily precipitation for salamander movement to be greater than or equal to 15 mm and minimum and maximum daily temperatures of at least 6°C and greater than or equal to 12°C, respec- tively (Table 1). We found in 1998 there were eight days with rainfall greater than 15 mm, but only three of those days (all in June) where the minimum and maximum daily temperature criteria were met. In 1999, 11 days had rainfall greater than or equal to 15 mm; however, only 3 of the 11 days had daily temperatures that matched our criteria. Of those three days, two coincided with Tiger Salamander migrations in the study area (10-11 and 30-31 August). In 1999, we found Tiger Salamander movement to be related to heavy rainfall events and warm weath- er; however, the previous year there were favorable meteorological conditions for salamander move- ments but we did not detect any migration across the highway. First, although we assume here that the move- ments we observed were post-breeding we cannot rule out late pre-breeding movements. Generally, in Aiberta breeding occurs in early spring (Russell and Bauer 2000). However, at high elevation ponds in TABLE 1. Meteorological conditions for Tiger Salamander migrations reported in the litera- ture. Measurements in bold are minimal conditions in our study area when movements may be expected to occur. Temperature Precipitation (mm) minimum (°C) Mi 8.8 nr 5.0 13.4 nr 39.0 nr 16-18 5-8 215 26 *nr: not recorded. Temperature maximum (°C) Source 13.8 Duellman 1954 EUAT Duellman 1954 nr Semlitsch 1983b nr Semlitsch 1983b 11-13 This study >12 2001 CLEVENGER, McIvor, McIvor, CHRUSZCZ, AND GUNSON: TIGER SALAMANDER MOVEMENTS Colorado and Utah breeding may be as late as July or August (Whiteman et al. 1995; Wissinger and Whiteman 1992; Worthylake and Hovingh 1989). At our study area favorable conditions providing oppor- tunities for migrations in any given year or between years may be variable and few. Second, if the move- ments were post-breeding, the reason for the sudden eruption in movements away from Chilver Lake is unclear. They may have been moving to hibernation quarters (Hassinger et al. 1970; Fowler 1935) or moving between lakes as a result of density-depen- dent dispersal. There is some evidence to suggest that high density amphibian populations range wide- ly for resources and when populations reach a densi- ty threshold large-scale dispersal events can occur (Langton 1989). There is no detailed historical information pub- lished on Tiger Salamander movements in Alberta but Fowler (1935), writing from Aldersyde, noted local migration at the “harvest season”. Patch and Stewart (1924) described a mass movement of Tiger Salamanders in Manitoba during later summer and autumn rainfall and similar movements have been observed in Saskatchewan and Manitoba (F. R. Cook, personal communication.) Road mortality is thought to have significant neg- ative effects on density of local amphibians (Ashley and Robinson 1996; Fahrig et al. 1995; Langton 1989). However, the level of threat from traffic to an amphibian population depends on a number of fac- tors. The juxtaposition of ponds may or may not necessitate road crossing to access seasonal habitat. Some populations may be depressed through the added mortality presented by the road, while road mortality in other populations may have little impact (Langton 1989). In order to assess the potential impacts of the Trans-Canada highway on the persistence of Tiger Salamander populations in this area, the regional conservation value of the existing ponds and lakes must be determined. Information on the Tiger Salamander distribution and degree of habitat con- nectivity in the area will help in understanding Tiger Salamander metapopulation structure and human impacts on it. At a local level, it will aid in assessing the significance of Chilver and Middle lakes for sala- mander persistence. Proactive measures could be taken immediately to reduce road mortality of Tiger Salamanders in the study area by getting animals safely under the high- way and still allow natural movements patterns. A 0.75 m diameter metal culvert is favorably located under the highway less than 50 m from the road-kill cluster centers, and in its present state, with tempo- rary drift fencing, could be adapted seasonally for salamander passage. Tiger Salamanders usually are not philopatric to natal ponds and quickly colonize newly constructed ponds (Petranka 1998). Therefore 203 measures to reduce mortality by creating new habitat on the Chilver Lake side of the highway to discour- age cross-highway migrations might also prove effective at reducing salamander mortalities on the Trans-Canada highway. Acknowledgments The research was supported by Parks Canada and Public Works and Government Services Canada (under contract C8160-8-0010). We thank Terry McGuire of Parks Canada Highway Service Centre for securing funding the project. Mike Brumfit assisted us in the field, Cathy Hourigan helped out with the bibliography, while meteorological data were provided by Judy Buchanan-Mappin. Special thanks are due to Larry Powell for information on Tiger Salamanders in Alberta and a critical review of an early version of the manuscript. K. W. Stewart, F.R. Cook and an anonymous reviewer also provid- ed comments. Literature Cited Ashley, E. P., and J. T. Robinson. 1996. Road mortality of amphibians, reptiles and other wildlife on the Long Point Causeway, Lake Erie, Ontario. Canadian Field- Naturalist 110: 403-412. Douglas, M.E. 1979. Migration and sexual selection in Ambystoma jeffersonianum. Canadian Journal of Zool- ogy 57: 2303-2310. Duellman, W. E., and L. Trueb. 1986. Biology of amphib- ians. McGraw-Hill, New York. 670 pages. Duellman, W.E. 1954. Observations on autumn move- ments of the salamander Ambystoma tigrinum in south- eastern Michigan. Copeia, 1954: 156-157. Earth Systems Research Institute. 1998. ArcView GIS version 3.2. Redlands, California. Fahrig, L., J. H. Pedlar, S. E.Pope, P. D. Taylor, and J.F. Wegner. 1995. Effect of road traffic on amphib- ian density. Biological Conservation 74: 177-182. Fowler, R.L. 1935. A note on the migration of the tiger salamander, Ambystoma tigrinum. Canadian Field- Naturalist 49: 59-60. Hairston, N. A. 1987. Community ecology and salaman- der guilds. Cambridge University Press, Cambridge, England. 230 pages. Hassinger, D.D., J.D. Anderson, and G. H. Dalrymple. 1970. The early life history and ecology of Ambystoma tigrinum and Ambystoma opacum in New Jersey. Ameri- can Midland Naturalist 84: 474495. Heine, G. 1987. Einfache Meb- und Rechenmethode sur Ermittlumg der Uberlebenschance wandernder Amphib- ien beim Uberqueren von Straben. Beih. Veroff. Natur- schutz und Landschaftspflege in Baden-Wurttemberg 41: 473-479. Hooper, D. F. 1992. Turtles, snakes and salamanders of east-central Saskatchewan. Blue Jay 50: 72-77. Kuhn, J. 1987. Strabentod der Erdkrote (Bufo bufo L.): Verlustquoten und Verkehrsaufkommen, Verhalten auf der Strabe. Beih. Veroff. Naturschutz und Landschaftsp- flege in Baden-Wurttemberg 41: 175-186. Langton, T.E.S. Editor. 1989. Amphibians and roads. ACO Polymer Products Ltd., Bedfordshire, England. 199 pages. 204 Levine, N. 1999. Crimestat: a spatial statistics program for the analysis of crime incident locations. Ned Levine & Associates: Annandale, Virginia and National Insti- tute of Justice, Washington, D.C. Patch, C.L., and D. A. Stewart. 1924. The tiger sala- mander at Ninette, Manitoba. Canadian Field-Naturalist 38: 81-82. Petranka, J. W. 1998. Salamanders of the United States and Canada. Smithsonian Institution Press, Washington, D.C. 578 pages. Russell, A. P., and A. M. Bauer. 2000. The Amphibians and reptiles of Alberta. Second edition. University of Calgary Press, Calgary Alberta. 292 pages. Salt, J. R. 1979. Some elements of amphibian distribution and biology in the Alberta Rockies. Alberta Naturalist 9: 125-136. Semlitsch, R. D. 1983a. Structure and dynamics of two breeding populations of the eastern tiger salamander, Ambystoma tigrinum. Copeia 1983: 608-616. Semlitsch, R. D. 1983b. Terrestrial movements of an east- ern tiger salamander, Ambystoma tigrinum. Herpeto- logical Review 14: 112-113. Semlitsch, R. D. 1985. Analysis of climatic factors influ- encing migrations of the salamander Ambystoma talpo- ideum. Copeia 1985: 477-489. Sever, D. M., and C. F. Dineen. 1978. Reproductive ecol- ogy of the tiger salamander, Ambystoma tigrinum, in THE CANADIAN FIELD-NATURALIST Vols northern Indiana. Proceedings of the Indiana Academy of Science 87: 189-203. Sexton, O. J., C. Phillips, and J. E. Bramble. 1990. The effects of temperature and precipitation on the breeding migration of the spotted salamander (Ambystoma macu- latum). Copeia 1990: 781-787. Stebbins, R. C. 1985. A field guide to western reptiles and amphibians. Houghton Mifflin Company, Boston. 336 pages. Whiteman, H. H., S. A. Wissinger, and A. J. Bohonak. 1994. Seasonal movement patterns in a subalpine popu- lation of the tiger salamander, Ambystoma tigrinum neb- ulosum. Canadian Journal of Zoology 72: 1780-1787. Whiteman, H.H., R.D. Howard, and K. A. Whitten. 1995. Effects of pH on embryo tolerance and adult behavior in the tiger salamander. Canadian Journal of Zoology 73: 1529-1537. Wissinger, S.A., and H.H. Whiteman. 1992. Fluctu- ation in a Rocky Mountain population of tiger salaman- ders: anthropogenic acidification or natural variation? Journal of Herpetology 26: 377-391. Worthylake, K. M., and P. Hovingh. 1989. Mass mortali- ty of salamanders (Ambystoma tigrinum) by bacteria (Acinetobacter) in an oligotrophic seepage mountain lake. The Great Basin Naturalist 49: 364-372. Received 12 June 2000 Accepted 4 June 2001 Possible Microclimate Benefits of Roost Site Selection in the Red Bat, Lasiurus borealis, in Mixed Mesophytic Forests of Kentucky JEFFREY T. HUTCHINSON! and MICHAEL J. LACKI Department of Forestry, University of Kentucky, Lexington, Kentucky 40546 USA '!Present address: Florida Department of Environmental Protection, Bureau of Parks, Hobe Sound, Florida 33455 USA Hutchinson, Jeffrey T., and Michael J. Lacki. 2001. Possible microclimate benefits of roost site selection in the Red Bat, Lasiurus borealis, in mixed mesophytic forests of Kentucky. Canadian Field-Naturalist 115(2): 205-209. Avoidance of diurnal predators is considered a primary selection factor influencing the choice of roosting sites by Red Bats (Lasiurus borealis), as this species is cryptically colored and difficult to observe in diurnal roosts. Other benefits to Red Bats resulting from choice of roosting sites, such as lowered energy expenditures due to more stable microclimates, have not been explored. We compared the thermal environment within diurnal roosting sites of Red Bats to other locations in the habitat where the bats would be potentially subject to fluctuations in temperature extremes. Data indicated that roosting sites of Red Bats had significantly lower overall average temperatures, overall variance in temperatures, average diurnal temperatures, and variance in diurnal temperatures than other locations in the habitat, with no difference observed for any nocturnal temperature variable. These results suggest that Red Bats roosting in forest habitats may gain a thermal advan- tage by selecting diurnal roosting sites that ameliorate variability and extremes in temperature. Key Words: Red Bat, Lasiurus borealis, roosting habitat, temperature, Kentucky. Environmental selection pressures important to the choice of roosting sites by forest-dwelling bats remain poorly understood (Brigham and Barclay 1996; Kalcounis and Brigham 1998). This is espe- cially true for phytophilic bats; i.e., species that use the foliage of vegetation as roosting sites (Hutchinson and Lacki 2000). Bats that use ephemeral roosts, such as live trees and shrubs, switch roosts more fre- quently than species that occupy roosts that are pre- dictable in time and space; e.g., caves and mines (Kunz 1982). Reasons purported to explain roost switching in foliage-roosting bats include decreased commuting costs to foraging areas, reduced predation pressures, and selection of alternate microclimates (Lewis 1995). Regardless, little data are available on how microclimate influences choice of roosting sites by phytophilic bats. The Red Bat (Lasiurus borealis) selects roosting sites in the foliage of trees, shrubs, and vines (Barbour and Davis 1969; Shump and Shump 1982). Until recently, this species was believed to roost close to the ground along fencerows and forest edge (McClure 1942; Constantine 1959, 1966; Mumford 1973); however, more recent data obtained using radiotelemetric techniques demonstrated that this species also roosts high above the ground in the canopy of dominant trees in intact, second growth forest (Menzel et al. 1998; Hutchinson and Lacki 2000). As with other lasiurine species, the Red Bat is cryptically colored and blends in with furled, dead leaves. This trait has led several authors to postulate that cryptic coloration is an adaptation in Red Bats to minimize detection by diurnal predators while roost- ing in foliage (Orr 1950; McClure 1942; Constantine 1966). Hutchinson and Lacki (2000) discussed the relative merits of roost choice by Red Bats in rela- tion to vulnerability to diurnal predators, and sug- gested that diurnal roosts selected high above the ground and away from forest edge should result in lower predation pressures. Assuming predation pressure largely dictates diur- nal roost site selection in Red Bats, it is still likely that additional factors could be a minor influence on choice of roosting sites. Constantine (1958) noted the effects of weather on roosting lasiurine species, including the Red Bat, and Watkins and Shump (in Shump and Shump 1982) observed that preferred roosting sites of Red Bats were associated with low- ered evaporative water loss. Hutchinson and Lacki (2000) found Red Bats roosting exclusively in the shade in tree canopies, whereas others have observed Red Bats and other lasiurine species roosting in foliage but exposed to direct sunlight (Constantine 1958; Mumford 1973). This variability in roosting habits of Red Bats suggests that microclimate does influence roost site selection in Red Bats. In fact, based on his observations, Constantine (1966) listed vegetation that impedes wind currents, dust, and contributes to heat and humidity as an important fac- tor in roost site selection of Red Bats. In this study, we measured temperatures within known diurnal roosting sites of Red Bats. We tested the null hypothesis that there is no difference in temperature conditions within Red Bat roosts compared to other, more exposed, locations in the habitat. Methods Roosting sites of Red Bats were located in the 205 206 Cumberland Plateau physiographic region of eastern Kentucky (Hutchinson and Lacki 2000). The area ranges 200-500 m in elevation and covers 28 500 km? of rugged, forested terrain (McGrain 1983). Forests are classified as mixed mesophytic (Braun 1950). Roosting sites were in second-growth hardwood for- est that was largely undisturbed by silvicultural activ- ities and contained mature stands of timber with an average dbh > 25.4 cm (P. Kalisz, University of Kentucky, personal communication). Dominant for- est vegetation used as roosting sites by Red Bats were hickory (Carya spp.), Yellow-poplar (Liriodendron tulipifera), White Oak (Quercus alba), American Beech (Fagus grandifolia), and Chestnut Oak (Q. prinus), among others (Hutchinson and Lacki 2000). Red Bats were radiotagged and tracked to roost- ing sites from May through August, 1996-1997. Methodology used in tagging and tracking Red Bats is outlined in Hutchinson and Lacki (2000). We located diurnal roosts in the canopies of trees using binoculars, and by lying beneath the roost tree 20-30 min before sunset and watching the bat emerge that evening to feed. We placed temperature data loggers (Onset Com- puter Corporation, Pocasset, Massachusetts, USA) in eight roost trees, in locations observed to be used as roosting sites by Red Bats. A more detailed descrip- tion of roost sites of Red Bats examined in this study is available in Hutchinson and Lacki (2000). An additional eight data loggers were placed adjacent to or beneath the canopy of the roost trees in exposed settings for a paired comparison. An exposed setting was defined as a free air space > 10 m above ground that was not occupied or shielded by canopy foliage. We positioned data loggers in roosting sites and exposed sites using a slingshot, 18-kg fishing line, and a 56-85 g sinker. Once the line was in place, a data logger was tied to the end of the line and pulled up into position. This allowed data loggers to be directly situated in known roosting sites, while per- mitting data loggers in exposed sites to be placed as close to the same height off of the ground surface as possible. We periodically checked data loggers to ensure that their position was maintained. Kunz and Nagy (1988) recommended the use of taxidermic mounts to house temperature sensors to more accurately measure operative temperature (7); thus, these data cannot be used to infer T,. Because of the sample sizes involved and the likely damage to taxidermic mounts that could have occurred during this study, we were unable to make use of any existing museum collection. Regardless, given that taxidermic mounts still do not completely prevent bias in the data (Bakken and Kunz 1988), and that the approach we used required only a relative comparison, we suggest that the inferences made regarding differences in tem- perature conditions between roosting sites of Red Bats and exposed sites are valid. THE CANADIAN FIELD-NATURALIST Vol. 115 Data loggers were operated for 42 days, from 13 July to 23 August 1997. We set data loggers to record temperatures ca. every 1.3 hr. We download- ed the data into Quattro Pro spreadsheets and tested data sets using 1|-tailed, paired t-tests. We compared temperature conditions based on overall, diurnal, and nocturnal temperatures. In each case, we examined the average, minimum, maximum, and variance in temperature. To define diurnal and nocturnal temper- atures, we divided the 24-h period based on local sunrise and sunset times. We hypothesized a priori that temperatures within roosting sites would be lower in average temperature, higher in minimum temperature, lower in maximum temperature, and smaller in variance in temperature. Results We observed differences between roosting sites and exposed sites for overall average temperature (t=2.57, P = 0.025), average diurnal temperature (t= 2.57, P = 0.044), overall variance in temperature (t = 2.02, P = 0.014), and variance in diurnal temper- ature (t = 2.81, P = 0.013), with temperature condi- tions being less within roosting sites than in exposed sites (Table 1). We found no difference for any noc- turnal temperature variable. Temperatures within roosting sites peaked at a lower level than exposed sites in midday, then dropped to levels inseparable between locations at night (Figure 1). These data indicate that roosting sites appeared to offer protec- tion to Red Bats from the extremely high tempera- tures reached in midday, probably in part due to shading from direct solar radiation. Further, the low insulative capacity of vegetation likely explains the lack of a difference in temperatures within roosting sites and that of exposed sites during nocturnal hours. Discussion Red Bats in this area typically selected diurnal roosting sites high off of the ground in trees located on south-facing slopes on ridge tops of upland forests (Hutchinson and Lacki 2000). Hutchinson and Lacki (2000) argued that selection of roosting sites in these locations should result in lower expo- sure of Red Bats to diurnal predators. Regardless, these sites were more likely to receive a higher inci- dence of solar radiation than potential roosting habi- tats on north-facing slopes or at mid- and lower ele- vations. Red Bats did select locations within the roost trees under dense foliage near the outer can- opy; thus, the vegetation probably shielded the bats from direct solar radiation during at least part of the midday sun. No Red Bat was ever observed in a roost directly exposed to solar radiation; however, Constantine (1958) suggested that lasiurine bats pre- ferred west and southwest exposures to facilitate heating by the sun prior to emerging at dusk. 2001 HUTCHINSON AND LACKI: ROOST SITE SELECTION IN THE RED BAT 207 TABLE |. Means and standard errors (SE) of temperatures (°C) within roosting sites of Red Bats, Lasiurus borealis, and exposed sites in the habitat in mixed mesophytic forests, Kentucky, 1997. Roosting sites Exposed sites Temperature profile Mean + SE Mean + SE Overall 23.59 + 0.24 24.34 + 0.42 Minimum 8.44 + 0.48 8.20 + 0.40 Maximum 39°50 1:20 42.74 + 3.14 Variance 40.69 + 3.878 39:20 5.93 Diurnal 27.68 + 0.412 29.26 + 0.74 Minimum 1552 20:58 15.45 + 0.45 Maximum 39.50 1220 42.74 + 3.14 Variance 28.88 + 3.178 48.83 + 6.36 Nocturnal 19.28 + 0.19 19.23, .022 Minimum 8.44 + 0.48 8.20 + 0.40 Maximum oie Prey Memallay 2 S429 2.2 Variance 18.76 + 2.45 20.22 + 2.66 aMeans within rows are different (P < 0.05). Lewis (1995) proposed that foliage roosting bats have the benefit of alternative roosting sites and, because of greater roost site availability, have easy access to alternate roosting sites that offer different microclimates. Vaughan (1987) observed Yellow- winged Bats, Lavia frons, in Africa switching roost locations in accordance with temperatures, always choosing roosting sites in dense shade during the day in warm weather while moving to the open during cooler periods. Although the Red Bats in this area switched roosts frequently (about every 2.3 days; Hutchinson and Lacki 2000), they were most com- monly observed in diurnal roosts motionless with their wings folded around their body. Others Degrees Celsius TAB o \Sieammmesane 7/14 | 7/15 : Soin : VILA observed Red Bats in diurnal roosts in apparent tor- por and also stretching and grooming (McClure 1942; Orr 1950; Mumford 1973). Licht and Leitner (1967) suggested that reduced metabolic rates through inactivity appear to be important in reducing heat stress of bats in high tem- peratures. Tree bats are believed to maintain lower metabolism rates in order to conserve energy while in torpor when ambient temperatures are high (McNab1974). Torpor in bats was also demonstrated to be a strategy for water conservation, along with energy conservation (Herreid and Schmidt-Nielsen 1966). By maintaining themselves in a torpid-like state while in day roosts, we suggest that the Red 10 wn 7 © ~ co a [=] = = = = = N N = ~ KC ~ = = [_] Roost [_] Random FIGURE 1. Temperature profile within a roosting site of a male Red Bat and a ran- dom exposed site during a nine-day period from 13-21 July 1997, in mixed mesophytic forest, Kentucky. 208 Bats observed in this study were behaving in a man- ner consistent with the above patterns. Lasiurine bats are better insulated than other temperate-zone bats by having larger, thicker hair, and hair on the uropatagium. Greater insulation would enable these species to tolerate more air movement; 1.e., draftier conditions, without concomitant changes in water balance and thermoregulation (Shump and Shump 1980). Canopy surfaces act to absorb wind and wind currents are normally stronger in the upper canopy than the forest floor (Nadkarni 1994). All roosting sites of Red Bats we discovered were near the top of the canopy, where wind currents were likely stronger than at mid-canopy or near the forest floor in a forest with a high percentage of canopy closure (Hutchinson and Lacki 2000). Temperature and rela- tive humidity patterns vary inversely from the top of the canopy to the forest floor (Lee 1978; Parker 1995). During clear summer days, the temperature near the top of the canopy is higher and the relative humidity is lower, with these patterns reversed at the forest floor (Lee 1978). Our observations suggest that conditions under the canopy at ground level are stag- nant and humid during the summer months in the deciduous forests of eastern Kentucky. Such condi- tions would result in added heat stress and water loss for Red Bats roosting at mid-canopy or below (Licht and Leitner 1967). In fragmented habitats, such as urban areas and farmland, canopy closure is lower because the vegetation is more widely spaced; thus, Red Bats would be more likely to select roosting sites lower to the ground (Constantine 1958, 1959, 1966; Mumford 1973), where wind currents aid in evapora- tive cooling during periods of high temperatures. Most foliage roosting bats occur near the equator in tropical regions (Kunz 1982; Altringham 1997). Thus, the few species of foliage roosting bats that have distributions reaching into the Nearctic are like- ly to be subject to alternate selection pressures. One consequence of this is that most Nearctic foliage roosting bats are migratory, including the Red Bat (Shump and Shump 1982), to avoid climatic extremes in winter. We believe our data indicate that Red Bats are also sensitive to microclimate condi- tions during the summer maternity season. Acknowledgments Funding for this project was provided by the University of Kentucky College of Agriculture and the E.O. Robinson Trust Fund. Computer assistance was provided by D. Cremeans. Assistance in the field was provided by H. F. Yacek, Jr.. This research (KAES # 99-09-166) is connected with a project of the Kentucky Agricultural Experiment Station and is published with the approval of the Director. All methods used in connection with this project were approved by the University of Kentucky Animal Care and Use Committee (protocol # 96-0006A). THE CANADIAN FIELD-NATURALIST Volo mls Literature Cited Altringham, J.P. 1996. Bats: Biology and behavior. Oxford University Press, New York. 262 pages. Bakken, G. S., and T. H. Kunz. 1988. Microclimate methods. Pages 303-332 in Ecological and behavioral methods for the study of bats. Edited by T.H. Kunz. Smithsonian Institution Press, Washington, D.C. Barbour, R. W., and W.H. Davis. 1969. Bats of Amer- ica. University Press of Kentucky, Lexington. 286 pages. Braun, E.L. 1950. Deciduous forests of eastern North America. Hafner, New York. 569 pages. Brigham, R.M., and R.M.R. Barclay. 1996. Bats and forests. Pages XI-XIV in Bats and forests symposium. Edited by R. M. R. Barclay and R. M. Brigham. British Columbia Ministry of Forests, Victoria. Constantine, D.G. 1958. Ecological observations on lasi- urine bats in Georgia. Journal of Mammalogy 39: 64-70. Constantine, D.G. 1959. Ecological observations on lasi- urine bats in the North Bay area of California. Journal of Mammalogy 40: 13-15. Constantine, D. G. 1966. Ecological observations on Lasiurine bats in Iowa. Journal of Mammalogy 47: 3441. Herreid, C.F., II, and K. Schmidt-Nielsen. 1966. Oxygen consumption, temperature, and water loss in bats from different environments. American Journal of Physiology 211: 1108-1112. Hutchinson, J.T., and M. J. Lacki. 2000. Selection of day roosts by red bats in mixed mesophytic forests. Journal of Wildlife Management 64: 87-94. Kalcounis, M.C., and R.M. Brigham. 1998. Secondary use of aspen cavities by tree-roosting big brown bats. Journal of Wildlife Management 62: 603-611. Kunz, T.H. 1982. Roosting ecology of bats. Pages 1-55 in Ecology of bats. Edited by T. H. Kunz. Plenum Press, New York. Kunz, T.H., and K. A. Nagy. 1988. Methods of energy budget analysis. Pages 277-302 in Ecological and behavioral methods for the study of bats. Edited by T. H. Kunz. Smithsonian Institution Press, Washington, D.C. Lee, R. 1978. Forest microclimatology. Columbia Uni- versity Press, New York. 276 pages. Lewis, S.E. 1995. Roost fidelity of bats: a review. Journal of Mammalogy 76:481—496. Licht, P., and P. Leitner. 1967. Physiological responses to high environmental temperatures in three species of microchiropteran bats. Comparative Biochemistry and Physiology 22: 371-387. McClure, H.E. 1942. Summer activities of bats (genus Lasiurus) in lowa. Journal of Mammalogy 23: 430-434. McGrain, P. 1983. The geologic story of Kentucky. Special Publication 8, Series XI, Kentucky Geologic Survey, University of Kentucky, Lexington. 74 pages. McNab, B.K. 1974. The behavior of temperate cave bats in a subtropical environment. Ecology 55: 943-958. Menzel, M.A., T. C. Carter, B. R. Chapman, and J. Laerm. 1998. Quantitative comparison of tree roosts used by red bats (Lasiurus borealis) and Seminole (L. seminolus) bats. Canadian Journal of Zoology 76: 630-634. Mumford, R. E. 1973. Natural history of the red bat (Lasiurus borealis) in Indiana. Periodicals in Biology 75: 155-158. Nadkarni, N. M. 1994. Diversity of species and interac- tions in the upper tree canopy of forest ecosystems. American Zoologist 34: 70-78. 2001 Orr, R. T. 1950. Notes on the seasonal occurrence of red bats in San Francisco. Journal of Mammalogy 31: 457-458. Parker, G. G. 1995. Structure and microclimate of forest canopies. Pages 73-106 in Forest canopies. Edited by M. D. Lowman and N. M. Nadkarni. Academic Press, New York. Shump, K. A., Jr., and A. U. Shump. 1980. Comparative insulation in Vespertilionid bats. Comparative Biochem- istry and Physiology 66A: 351-354. HUTCHINSON AND LACKI: RoosT SITE SELECTION IN THE RED BAT 209 Shump, K. A., Jr., and A. U. Shump. 1982. Lasiurus borealis. Mammalian Species 183: 1-6. Vaughan, T. A. 1987. Behavioral thermoregulation in the African yellow-winged bat. Journal of Mammalogy 68: 376-378. Received 24 December 1999 Accepted 6 April 2001 Ectoparasites in Lekking Sharp-tailed Grouse, Tympanuchus phasianellus LEONARD J. S. Tsumt!, Jim D. KARAGATZIDES?, and GERARDO DEILuts? ‘Department of Environment and Resource Studies, University of Waterloo, Waterloo, Ontario N2L 3G1 Canada *Department of Geography, Queen’s University, Kingston, Ontario K7L 3N6 Canada >Department of Zoology, University of Toronto, Toronto, Ontario M5S 3G5 Canada Tsuji, Leonard J.S., Jim D. Karagatzides, and Gerardo Deiluiis. 2001. Ectoparasites in lekking Sharp-tailed Grouse, Tympanuchus phasianellus. Canadian Field-Naturalist 115(2): 210-213. To investigate the validity of assumptions based on the parasite avoidance model, fifty-two male Sharp-tailed Grouse (Tympanuchus phasianellus) collected in northern Ontario from six separate leks were examined. Ectoparasitic burdens were recorded along with traits (feather damage, hematomas, body condition, body mass) and cues (territorial position, age) which may be used by the female to discriminate among males for their parasite loads. Neither assumption of the transmission avoidance model. was met. Little variation in ectoparasite number existed in the breeding population as a whole or within individual leks. None of the traits or cues measured were correlated consistently with ectoparasitic bur- dens. These results suggest that randomly mating females would, on “average,” mate with lightly infested individuals. Therefore, benefit to a discriminating female Sharp- tailed Grouse would be negligible. Key Words: ectoparasites, lek, Sharp-tailed Grouse, Tympanuchus phasianellus. A lek can be defined as an area used for mating, where males establish territories to form a cluster. Females visit the lek for the sole purpose of mating; resources on the lek (e.g., food) are limited. Al- though females have the opportunity to select any one of the males present as a mate, they often show unanimity in their choice of mate, selecting males occupying centrally located territories on the lek (Lumsden 1965; Kermott 1982). In this simple mat- ing system, female mate-choice is theoretically based only on the benefits derived from the male himself. The “parasite” or “transmission avoidance” model suggests that females prefer to mate with males with low ectoparasite numbers to reduce the probability of becoming infested (Borgia and Collis 1990; Clayton 1990). Two important assumptions of this model are that enough variation in ectoparasite num- ber per male exists in the breeding population (and more specifically on individuals within a lek) to make mate choice worthwhile (Reynolds and Gross 1990), and that a trait or cue correlated with ectopar- asitic burden facilitates choice. In this study, we determined if these assumptions of the transmission avoidance model are valid by examining individuals from leks of the Sharp-tailed Grouse (Tympanuchus phasianellus). Methods Study sites were located in areas of muskeg near Fort Albany (52°15’N; 81°35’W) in the James Bay Region of northern Ontario. One researcher accom- panied several Native North Americans on their spring hunt for Sharp-tailed Grouse and examined birds shot before they were processed for food. Fifty-two males were examined from six separate leks during the 1990-1992 breeding seasons. The numbers of birds obtained and the total numbers of males at each lek were as follows: lek 1, 2 of 22; lek 2, 12 of 15; lek 3, 8 of 17; lek 4.:-Vof 1ID2lekes sligiok 19; and lek 6, 10 of 13. Leks 5 and 6 were observed for 12 days during the morning display period prior to collection to record copulations by individual males. Yearlings were differentiated from adults using the characteristic of feather-wear (Ammann 1944). Males were scored in the field, before they were collected, as occupying central vs. peripheral territories. Central territories were defined as those completely surrounded by other territories whereas peripheral territories were unbounded on at least one side (Kruijt and Hogan 1967). Upon being shot, individuals were placed separate- ly in plastic bags to prevent migration of ectoparasites between hosts (Eveleigh and Threlfall 1976). A feath- er-by-feather examination was conducted and ectopar- asites were placed in 70 % alcohol for subsequent identification (Ash 1960; Eveleigh and Threlfall 1976). Feather damage was assessed for each speci- men with individuals being placed in one of three categories: 0 (no damage to minimal); | (average damage); 2 (severe damage) [Clayton 1990]. Cervical apteria of each individual were assessed for presence or absence of hematomas (Johnson and Boyce 1989). Individuals were weighed to the nearest 1.0 g using a spring scale or triple-beam balance. The physical condition of each male was assessed using an index based on the shape of pectoral muscle from keel to sternum (Macdonald 1962). Condition of males was classified as follows: 0, in good condition 210 2001 (pectoral muscle distinctly convex and keel almost imperceptible); 1, in average condition (pectoral muscle flattened forming a planar surface between keel and thoracic ribs); 2, in poor condition (pectoral muscles concave and keel distinct). Spearman rank correlation analysis (r.) was used to investigate the relationship between ectoparasite number and body mass among individuals of all leks. Ectoparasite data, number of ectoparasites per individual, were pooled for all six leks and a fre- quency distribution was plotted. Frequency distribu- tions, number of ectoparasites per individual, were (a) 0 10 20 30 40 50 60 70 (c) eee ee oe ae od oe NUMBER OF INDIVIDUALS 0 10 20 30 40 50 60 70 TsusI, KARAGATZIDES, AND DEILUIIS: ECTOPARASITES IN GROUSE pH 6! constructed for individual leks at which more than 75% of males present on the lek were collected. Data for ectoparasites were subjected to a Mann- Whitney U test between central and peripheral males for leks at which more than 75% of the males were examined and for all leks combined. In addition, ectoparasite data for all leks were subjected to a Mann-Whitney U test to determine whether adults and yearlings differed. Results Of the 52 birds examined, 40 were classified as (b) oOo NWO er UO ~ CO WO SO 0 10 20 30 40 50 60 70 a> ob (d) (=) oN WS UO ~) COO 0 10 20 30 40 50 60 NUMBER OF ECTOPARASITES FIGURE 1. Frequency distributions of ectoparasites on lekking Sharp-tailed Grouse: a (all six leks combined), b (lek 2), c (lek 5), and d (lek 6). 22 TABLE 1. A comparison of ectoparasite number per indi- vidual male Sharp-tailed Grouse and its territorial location, peripheral or central on leks, using Mann-Whitney U tests. Ectoparasites per Individual N x+SD Ze All leks Peripheral males 26 6+8 =2 37 Central males 26 13+14 Lek 2 Peripheral males 7 13+14 -1.07"S Central males 5) 18+8 Lek 5 Peripheral males 8 4+3 -2.11° Central males 8 ees Lek 6 Peripheral males 4 #2 -0.54"s Central males 6 4+2 aSigmiticance of Z. *) P> 28: P< 0105 adults and 12 as yearlings. Males scored as occupy- ing central territories, 50% (N = 26) of all individu- - als, included only adult birds. The peripheral males, however, consisted of 46.2% (N = 14) adults with the remainder being yearlings. Only two copulations were recorded during this study, both by central males. One of those males was infested with nine ectoparasites and the other had five. Neither individual had the least or greatest number of ectoparasites in their respective leks; both had infestations within the range found for other birds at the same lek. None of the sampled males had more than mini- mal feather damage, even the most heavily infested individuals. No evidence of hematomas were found on any specimens examined. All lekking individuals were Classified as being in good condition. Spearman rank correlation analysis showed no significant rela- tionship between ectoparasite number and body mass (r= 0.04, P=0.76). Although the tick Haemaphysalis spp. (Baum- gartner 1939; Peterle 1954) and the louse Goniodes spp. (Snyder 1935; Emerson 1951) have been report- ed to infest Sharp-tailed Grouse, only the latter was found on the specimens examined. The frequency distribution for data from all the leks (Figure 1a), illustrates the characteristic “hollow” curve type in which most hosts have few parasites and most para- sites are on only a few hosts (Eveleigh and Threlfall 1976). Frequency distributions for lek 2 and 5 (Figure 1b,c) also showed the hollow curve type whereas data from lek 6 (Figure 1d) revealed low rates of infestation for all individuals sampled. Mann-Whitney U tests showed that individuals possessing central territories had significantly more ectoparasites than males occupying peripheral terri- tories in lek 5 and in all leks combined (Table 1). On THE CANADIAN FIELD-NATURALIST Vol. 115 leks 2 and 6, there was no significant difference in ectoparasitic burden between central and peripheral males (Table 1). Ectoparasite number did not differ significantly (P = 0.49) between adult and yearlings males for all leks combined (adults: N =40, x= 10:65, SD= 13.25; yearlings? N= SD =4.66). Also, it is noteworthy that peripheral males (independent of age) harboured fewer para- sites than central males. Discussion In the present study, it was found that during the mating season, there appeared to be neither enough variation in ectoparasites per male in the lekking population, and more importantly, between individu- als within a lek to benefit a discriminating female. If females can randomly choose, on average, a mate with few ectoparasites, female choice against ectoparasite-infested males is of no real benefit (Reynolds and Gross 1990). In other studies of lekking bird species, individual ectoparasitic burdens were also found to be low (e.g., Snyder 1935; Peterle 1954; Andersson 1992). Studies which have shown a mating advantage for non-infested males (e.g., Borgia and Collis 1989) were limited methodologi- cally because birds were not sacrificed for a subse- quent feather-by-feather examination, and there is a 10% probability that individuals scored as uninfested were really infested (Ash 1960). In other words, it is difficult, to quantify ectoparasite loads without sacri- ficing the birds, which is not always possible. No traits, related to ectoparasite loads, which could be used by discriminating females, were found in this study. Feather damage was minimal. Moreover, feather damage would not be a good trait for assess- ing ectoparasite burden because only basal and medi- al regions are consumed by lice; thus, feather damage is not readily visible (Clayton 1990). Only in the most severe cases do ectoparasites tatter plumage to the extent that water repellency and insulation are affected (Soulsby 1982). No cervical lesions were found in this study, al- though in Sage Grouse (Centrocercus urophasianus) lice have been associated with hematomas (e. g., Johnson and Boyce 1989) and mating success (Spurrier et al. 1989). However, in other studies of lekking grouse, ectoparasitic lesions have been report- ed as rare (e.g., Peterle 1954; Gibson et al. 1991). Lekking individuals in this study were all in good condition which is not surprising as good health is necessary in establishing and maintaining territories on the lek (Kermott 1982). Ectoparasitic load was not significantly related to body weight although such relationships have been reported in other stud- ies (Snyder 1935; Eveleigh and Threlfall 1976). Cues consistently correlated with ectoparasite numbers were not evident in the present study. Although occupancy of central territories in Sharp- 2001 tailed Grouse leks has often correlated with male mating success (e.g., Lumsden 1965; Kermott 1982), in the present study, pooled data revealed that central males had significantly larger ectoparasitic burdens compared to peripheral individuals. Within individu- al leks, only lek 5 showed a similar trend, with lek 2 and 6 showing no significant differences. Perhaps, the relatively large number of individuals examined from lek 5 had a disproportionate influence on the Mann-Whitney U test, when all leks were combined. Nevertheless, central males may sometimes show higher parasitic burdens compared to peripheral indi- viduals due to increased physical contact during the breeding season, as a consequence of small territory sizes (Lumsden 1965) increasing the chance of infes- tation (Ash 1960). Also, Borgia and Collis (1990) suggested that age-related grooming may be impor- tant in controlling louse populations in birds, decreasing loads being associated with increasing age. This was not true with the sharptails studied here as there was no significant difference between ectoparasitic burdens on adult and yearling lekking males. In overview, it does not appear that the benefit of reduced risk of ectoparasite transmission to female sharptails during mating is a significant factor in mate choice. However, females nonetheless should keep ectoparasites in check because parents can transfer lice to offspring and offspring may be more sensitive to ectoparasites than adults (Marshall 1981; Clayton 1990). This may be the reason why female grouse of lekking species shake vigorously and preen them- selves after successful copulation (e.g., Lumsden 1965), as it is known that lice can be transferred dur- ing copulation (Eveleigh and Threlfall 1976; Clayton 1990). Acknowledgments We thank A. S. Stephens, R. Gillies, and L. Gillies for allowing the salvaging of grouse remains; the laboratory of S.S. Desser for guidance in parasite collection, preservation, and identification; S.W. Cavanaugh and S. Hyndman for preparation of this manuscript; and comments from H. Rodd, A. J. Erskine, and D. Boag. Literature Cited Ammann, G. A. 1944. Determining the age of Pinnated and Sharp-tailed Grouses. Auk 8: 170-171. Andersson, S. 1992. Female preference for long tails in lekking Jackson’s widowbirds: experimental evidence. Animal Behaviour 43: 379-388. Ash. J.S. 1960. A study of the Mallophaga of birds with particular reference to their ecology. Ibis 102: 93-110. Baumgartner, F. M. 1939. Studies on the distribution and habits of the Sharptail Grouse in Michigan. TSUJI, KARAGATZIDES, AND DEILUIIS: ECTOPARASITES IN GROUSE 213 Transactions of the North American Wildlife Conference. 4: 485-490. Borgia, G., and K. Collis. 1989. Female choice for para- site free male satin bowerbirds and the evolution of bright male plumage Behavioral Ecology and Socio- biology 25: 445-454. Borgia, G., and K. Collis. 1990. Parasites and bright male plumage in the Satin Bowerbird (Ptilonorhynchus vio- laceus). American Zoologist 30: 279-285. Clayton, D.H. 1990. Mate choice in experimentally para- sitized Rock Doves: lousy males lose. American Zoologist 30: 251-262. Emerson, K. C. 1951. A list of Mallophaga from gallina- ceous birds of North America. Journal of Wildlife Man- agement 15: 193-195. Eveleigh, E.S., and W. Threlfall. 1976. Population dynamics of lice (Mallophaga) on auks (Alcidae) from Newfoundland. Canadian Journal of Zoology 54: 1694— L711. Gibson, R. M., J. W. Bradbury, and S.L. Vehrencamp. 1991. Mate choice in lekking sage grouse revisited: the roles of vocal display, female site fidelity, and copying. Behaviour Ecology 2: 165-180. Johnson, L. L., and M.S. Boyce. 1989. Female choice of males with low parasite loads in sage grouse. Pages 377-388 in Bird-parasite interactions. Edited by J. E. Loye and M. Zuk. Oxford University Press, Oxford, UK. Kermott, L. H. 1982. Breeding behavior in the Sharp- tailed Grouse. Ph.D. dissertation. University of Minne- sota, St. Paul, Minnesota, USA. Kruijt, J. P., and J.A. Hogan. 1967. Social behavior on the lek of the black grouse, Lyrurus tetrix tetrix (L.). Ardea 55: 203— 240. Lumsden, H.G. 1965. Displays of the Sharp-tailed Grouse. Research Report 66. Ontario Department of Lands and Forests, Research Branch, Toronto, Ontario, Canada. Macdonald, J. W. 1962. Mortality in wild birds with some observations on weights. Bird Study 9: 147-167. Marshall, G. 1981. The ecology of ectoparasitic insects. Academic Press, London, UK. Peterle, T. J. 1954. The Sharp-tailed grouse in the upper peninsula of Michigan. Ph.D. dissertation. University of Michigan, Ann Arbor, Michigan, USA. Reynolds, J. D., and M. R. Gross. 1990. Costs and bene- fits of female mate choice: is there a lek paradox? Amer- ican Naturalist 136: 230-243. Snyder, L.L. 1935. A study of the Sharp-tailed Grouse. University of Toronto Studies Biological Series 40: 5-66. Soulsby, E. J. L. 1982. Helminths, arthropods, and proto- zoa of domesticated animals. Bailliere Tindall, London, UK. Spurrier, M.F., M.S. Boyce, and B. F. J. Manly. 1989. Effects of parasites on mate choice by captive sage grouse. Pages 389-398 in Bird-parasite interactions. Edited by J.E. Loye and M. Zuk. Oxford University Press, Oxford, UK. Received 11 February 2000 Accepted 11 June 2001 Vascular Plants of a Successional Alvar Burn 100 Days After a Severe Fire and Their Mechanisms of Re-establishment PauL M. CaTLING!2, ADRIANNE SINCLAIR!, and Don Cuppy? ' Biology Department, University of Ottawa, 30 Marie Curie Street, P.O. Box 450, Station A, Ottawa, Ontario KIN 6N5 Canada * Agriculture and Agri-food Canada, Central Experimental Farm, Ottawa, Ontario K1A 0C6 Canada 3 Ontario Ministry of Natural Resources, Concession Road, Postal Bag 2002, Kemptville, Ontario KOG 1J0 Canada Catling, Paul M., Adrianne Sinclair, and Don Cuddy. 2001. Vascular plants of a successional alvar burn 100 days after a severe fire and their mechanisms of re-establishment. Canadian Field-Naturalist 115(2): 214-222. In order to describe the vascular plant community 100 days after a severe fire in alvar woodland near Ottawa, Ontario, we set out fifty one m square quadrats at 5 m intervals along transects at each of five sites and recorded presence and cover of vascular plant species in each quadrat. The woodlands that burned were dominated by Thuja occidentalis, Populus tremu- loides, Abies balsamea, Picea glauca, and Pinus strobus in approximate order of importance. The post-fire flora was diverse and mostly native. Although substantial variation occurred in presence, frequency and cover of species between the sites, there was remarkable similarity in a distinctive group of dominants including Populus tremuloides, Geranium bick- nellii, and Corydalis aurea. Rare species present in the burned woodland included Astragalus neglectus, Calystegia spi- thamaea, Carex richardsonii, Corydalis aurea, Muhlenbergia glomerata, Panicum flexile, Panicum philadelphicum, Petasites frigidus, Scutellaria parvula, and Viola nephrophylla. Of these, Corydalis aurea and Calystegia spithamaea were frequent. The development of vascular flora following the fire was a consequence of growth from roots, rhizomes, and root crowns that survived the fire, and seeds buried in the soil. Although abundant in the burned woodland, Corydalis aurea and Dracocephalum parviflorum had not been previously recorded at the site, suggesting that these species are adapted to early post-fire succession, surviving periods of up to 130 years between fires as seeds in a very large and widespread subter- ranean seed bank. Not only is post-fire succession well underway within a hundred days of a fire, but even in its earliest stages, it appears to serve as a specific niche for a distinctive group of species including some that are rare and restricted. A diverse native flora is involved indicating the importance of management involving removal of woody biomass. Key Words: alvar, fire, flora, species diversity, rare species, succession, management, biomass removal, prescribed burn, Great Lakes, Ontario, Canada. Periodically burned woodlands associated with alvar landscapes (1.e., landscapes with more or less drought-maintained open areas on thin soil over essentially flat limestone, dolostone or marble) have recently been recognized as an important part of the alvar ecosystem with regard to protection of rare species and biodiversity and have been termed “suc- cessional alvar burns” (Catling and Brownell 1998). Although their importance is generally accepted, the information upon which it is based is limited to com- mon sense, anecdotes, and a single evaluation in- cluding data analysis (Catling and Brownell 1998). On 23 June 1999, 152 hectares of mostly forested terrain in the Burnt Lands near Almonte (Figure 1) was burned. This was a major fire. A shower of ash rained down on streets in downtown Ottawa 40 km away and the fire moved at a rate of 15 m/min. The flames reached well over 30 m into the air and an area at least 1 km in length was severely burned. This area was last burned in 1870. Dry conditions in the relatively shallow soil over porous limestone rock led to a relatively slow return to a mixed boreal and fire prone forest type following the 1870 fire, and it seems likely that fire and post-fire succession has been going on in the area for many hundreds (or thousands) of years. Brunton (1986*) suggested that the Burnt Lands was a “fire-dependent environment of great provin- cial significance” that “will require continued renew- al by periodic burning if its important natural values are to be preserved.” The recent (1999) fire in the Burnt Lands provided an opportunity to gather infor- mation related to this suggestion. There are conflict- ing views about fire in alvar woodland that range from the creation of a long lasting desert to almost immediate return to the pre-fire floristic composi- tion. We agree fully with Brunton’s suggestion and further suggest that successional changes occur over decades providing high floristic and faunistic biodi- versity as well as spatial-temporal habitat for rare and restricted species. Information on the specific effects of fire is currently insufficient to allow informed choices between management options. The nature of succession in successional alvar burns is not documented with respect to timing of changes or floristic composition, and thus the importance of the succession to biodiversity protection is poorly under- stood. The compositional changes also relate to the required frequency of implementation of manage- ment actions. Finally, biomass removal by fire man- agement is potentially important, not just to the man- agement of biodiversity, but also to the protection of 214 2001 people and property (from catastrophic fire) where fire-prone vegetation exists in semi-urban areas. The present study was designed to document the earliest stages of succession thereby providing an indication of its rapidity and potential biodiversity significance including a consideration of whether or not there are species which are adapted to the very earliest stages of the succession. This information is needed for the protection and management of alvars, which are a globally imperilled ecosystem (Brownell 2000) confined in North America to a restricted por- tion of the Great Lakes region (Catling and Brownell 1995, 1999). Specifically, it is required for the man- agement of the Burnt Lands alvar landscape which has been designated as a provincially significant life science Area of Natural and Scientific Interest (ANSD, with intent to increase protection through designation of provincial crown lands and former federal properties as a provincial nature reserve (Brownell 2000). This requires increasingly compre- hensive management and monitoring of significant natural resources. Methods Fifty-one m* quadrats at 5 m intervals along tran- sects were set out in alvar woodland at each of five sites. The only criteria for determining placement of transect lines was complete burning of all vegetation, separation from each other by more than 300 m, and adequate representation of the burned area through coverage of the eastern, western, northern, southern and central portions. These regions were character- ized by different pre-fire tree composition as described below. For each quadrat the total cover (dm? estimated as the upper leaf surface area) of each plant species was recorded. Using an estimate of the photosynthetic surface area of the plants rather than the surface area of the quadrat covered by that species permits a more direct correlation between cover values and biomass. The total cover and total frequency for each species at each site was calculat- ed and total overall cover and total overall frequency were tabulated (Table 1). To determine the composi- tion of the original woodland, the tree in each of four quadrants closest to the quadrat centre was identified and measured (dbh-diameter at breast height). The data were then tabulated by species frequency and range of dbh for each site (Table 2). Field work was done on 1-7 October 1999, approximately 100 days after the fire and prior to the first severe frosts. There was no perceptible loss of post-fire vegetation due to dieback in the burned areas sampled. Status information on the species recorded is pro- vided on a spatial scale of rarity (N,P,R,S,O, see Table 1) with N = nationally rare (Brownell and Larson 1995*); P = provincially rare (Brownell and Larson *See Documents Cited section. CATLING, SINCLAIR, AND CUDDY: VASCULAR PLANTS OF AN ALVAR BURN 215 1995*); R = regionally rare (eastern Ontario in Brownell and Larson 1995*); S = rare in the St. Lawrence-Ottawa physiographic region (i.e., MNR 6—11 and 6-12, Brownell and Larson 1995*); and O = rare in Ottawa-Carleton (Brunton 1997*). Species list- ed were also distinguished as native or introduced. To determine the methods of colonization, plants were excavated and examined 50 days after the fire at site 1 and 100 days after the fire at all sites. Although not quantitative, the observations are con- sidered reliable and useful, and included here since they provide the only source of information on this aspect that is available. Voucher specimens for species recorded were deposited in the vascular plant herbarium of Agriculture and Agri-food Canada (DAO) in Ottawa. Results and Discussion General observations The woodlands that burned were dominated by Thuja occidentalis, Populus tremuloides, Abies bal- samea, Picea glauca, and Pinus strobus in approxi- mate order of importance (Table 2). Similar wood- lands are present on alvar landscapes elsewhere in Ontario but often also have Bur Oak (Quercus macrocarpa Michx.), and those on the Napanee Plain also have Eastern Red Cedar (Juniperus virginiana L.), and those on the Bruce Peninsula and Manitoulin Island also have Jack Pine (Pinus banksiana Lamb.) and Red Oak (Quercus rubra L.). Despite variation between sites there was remark- able similarity with respect to a distinctive group of dominants (Table 1). Approximately 100 species were recorded in the quadrats and only 16 of these were introduced. Of the total species, 19 exceeded 5% overall frequency with only 3 of the 19 being introductions. The majority of the plants and over 90% of the cover at each site was comprised of native species. Thus the flora developing after the fire was comprised largely of native species and it was relatively diverse (Table 1). Variation between sites The species which re-colonize a burn and the means of colonization can be expected to vary depending upon the intensity of the fire. There is nat- ural variation in the amount of heat generated, and fire-fighting efforts (such as water bombing) undoubtedly will locally reduce the penetration of heat into the soil. This may add to the fine scale vari- ation in patterns of re-colonization but is unlikely to alter the range of means observed over 50 quadrats. Consequently variation in post-fire colonization between sites is most likely to be related to factors other than fire intensity. Surface coverage of plants varied from approxi- mately 10-50% among the five sites and species with cover values exceeding 1% of quadrat surface varied from three to eight in number at a site. The THE CANADIAN FIELD-NATURALIST Volts 216 panuyuoo oT Or CE v0 (oe Or 0? Dib (Ge vv 8°83 v9 oF CC v9 9°¢ C6 8C 0) 8° 8°38 8V V9 0'Cl (ogy 801 9'CE CE VOL VCL [[PJ0AQ ues VI NN 9C (44 g 9 i, = a. r= C — 9 _ 8I et 7 (6, 9 = es Ol 9 1G 14 7; (6) Ol ng 9 8 (6 V — —, 474 — —s 9 9 14 Ol 9 = - 8 vi 8 9 bs Ol O£ - (G 76! = Ke, = 8C = 8 (6) z O07? 9¢ cl OV = 1G 87 (6 > 8 (G3 Of£ (G9 vl = 8C 8E 08 v6 8L v9 96 98 14 € iG Aosuonbeal % 9C VL leks CL 0°8 0'8 0'8 v8 ell 6 TT vst O'VI 6'VI cSI 081 8°07 8°0C 9'ET [Ve V's2 G97 SO£ (4 Cc 6v £09 (G6) v'e9 6 01 6 C9C 40) 7 C8CEI DEIEC [[212A9 [RIOL EX fe = 8 ie, 8 v8 = Ea 6 “2 ve i fay lth 10 te O'e LE 81 9 {L com (Gil = Oe 8 cll Ja | a el £6 8S 81 sv L’‘0 8°67 £7? = ag 8 = col v9 9°SS v0 8S ae 8°9V 8°8P C88 ad LVC 6861 CLS 6SSC 996CI Vv € SG 619 6SI [OSI O'L6C rely C (cWIp) 19A0D [RIOL SSPID WOE AVDAO"_ “I[NYOS 2 ‘Ws0y xo ‘Anvog ‘q (YT) pivoids piuoyjung “Jo oeUINS UOYseRg “T Dulydk) snyy Auoqdsey jremq “yey suaosaqnd ‘yo snqny WOM-s yor 319 “ds wnouad{y JOOMSIOILE, SUIQUIT[D “| SuapuDas snajsvjaD Aujoqdsey jremq “Jey suaosaqnd snqny JOOJOSOOD poarsy-o[depy “Jey (U0]) xayduns wnipodouay) ALOgMenyng uaueg “yer (XY) saplolMsvsf piulajspjvA (x) ULT[NJA, UOWOD ‘7] susdpys wndspqia, QUIQUINJOD PIAA ‘I Sisuapvuvd visajinby [jeqorey “YT pofipunjos pjnundupy (x) [Ao Bury snoone[yH “IIIA septoyfasoyid wnionsaiy] (x) OPSIYL [ING “Ue, (IAeG) aupsyna wnis.s1D [ospunolH weseg “xyo snjnosadnnd pigauasg (,.) Uolfapuegd UOWWIOD ‘Iago AA ‘D ajpuloiffo WNIPXDAD J, JOISV SPOIL “ [PUL] sMij0I719 4a1sy vliedesies pli “’T Synpoipnu viypiy (S) JOOJS}[O9-JO9MG “A (J) SnpIs1Af sapisvjag KAI-UOSIOd “‘QpAy Xd [[RUIS 134aqgpK4 snyy Jejdog wresjeg ‘viafiups)pg ‘dss ‘{ paafiuvsjog snjndog peoyuoseiq posoMojj-[[ews “ynyy wnsopfiasnd wnjpydas0a01q duwioy ueipuy “yy wnuiqnuuns wnudoody AlloqMous ‘oye _ ‘4S (7) snqjp sodapa1soyduks asoy APH ‘SIMIT “AA (UIOMYIS) 1ApS “dss “[Pul'] S14DjNIIID DSOY (x) YOIPSAL Woe “T Vuyjndny osvoipay JOISY Poarvoy-asie’y] “"T snyjydosovu saisy (QS) stfepAso_d usployH ‘vasnn ‘dss “pli A vaanv syppksoy UIOJ-UdyovIg UJO\se_ “IO[JOH “YW XO ‘MIJOPUL) “"T (-ASOC) WN]NISNYY] “eA uYyNy (TY) wnuapinby wnipiatd []}q-S oUBID S [JouUxog “OWE 17/aUxI1g WniUbsa_) uodsy SUI[quio1y, “XY Sapiojnuiasy snjndog OWIeNY OIJUIIOS ‘ssunuevyd sed pur s}so10y SUIPUNOLINS UL SIO1} JOP[O JO VDUISgK UO pase dIIS SIU) J PIONPOIUI 9q 0} PIAdI[Aq SI yey) so1oods DAIWeU B SI DUDISYUDA snug YY) VON ‘poonpo. jut oie (,.) YSUoJse UP YIM poxYIeU Sa1dadg “19A09 [[PIDAO JO JOPIO SUIPUDOSEp UT poyst] ae soroedg ‘osejooy | A[oyewrxoidde jo vase ue UIYIIM s}oosUR.) BUOTL INO PIL] oIIs YORs ye syespenb ,Wi QUO (VG UO paseq oIe eyep oYL, ‘OWekIUG ‘oJUOWTLY Je9U spure’] JUING 94) Ut oITy eB Joye SKep QO] Aloyeurrxosdde spuv;poom pouing sAtj UI (puNnosIs dAOGR UW | UL) sso) JOAR] GIy oY) UT sorsods juRId Je[NOseA JOF ((L66[ UOIUNIG) UOJo[IVD-ePMENC Ul oe = C ‘(GHG [[OUMOIG ‘71-9 pUv [1-9 ANI “'9'!) uorsar o1ydessorshyd ePMENO-ddOUdIIME’T IS OY) Ul 9IeI = § ‘(CHG [[OUMOIG Ul OLIRJUG UO -jsevo) o1et A[[eUOIseI = yY ‘(SHH [[OUMOIG) oIeI AT[eIOULAOId = J ‘(GHGH] [[OUMOIG) oIe1 A[[eUOTWeU = NY) sNyeys pue ‘AouoNba1y asevyudI0d ‘syespenb (¢ Ul (ZWIP) J9AOD [RIO], “| ATAV], 217 VASCULAR PLANTS OF AN ALVAR BURN CATLING, SINCLAIR, AND CUDDY 2001 panuyuoo OV Ol Ol = = = 81 fe 90 = = = (OS) ISPIg S UOspIeYoTY “Ig “Y MUuospavyIiA XID ZI t Z = = = 61 6'0 Ol - - - AIIDQME.NS JOPIVIS ‘YuNIUISAIA “ASS JOYA, VUDIUIBAIA DIUDSDA LY r0 = a es Z = 07 _ ~ - _ ‘6 saoyAssng “ds piupuuajuy v0 2 ry (a T C 3 9 Oe = ‘ ce za = (OSUd) 494A P2199[99N “prays “q (AvID ‘Vv 2 “LOL) snjaajsau snjpspajsy oT = = 8 = = O'€ = = € = = POOMSOG JOISO-poy “xXYOIAY Viafiuojojs snusoy (aS a = * OVS) a ie = rs ie = II te (OSU) psompurg MoT ‘papupyjids ‘dss ysing (1) vapunynids pisajscjv) VC 2 = cl = = its = = Ve = = ALIayD Ulg ‘J “T va1upajxsuad ‘Jo snunig OC 4 a 9 = i CE 70 7 € ; = ATT poo “TY wenoiydjapyjiyd wnyyT “Jo 09 C 14 07 = 174 6€ 10 0 (Sal = Z Ad][VA-9Y}-JO-ATIT PIL “JSoq asuapvuvs wnwayjuvivpy VC C (6 8 iz S OV 10 C0 ve = = (4) YIOA PON], “'T VIIA VIIIA 91 (4 = 9 iz =a ev € = CT = = SSBID SIU POARI]-MOLIRN “YSeN wnyOfiavaul] WNnI1UDg 89 7 8 VT (6 = vv a v'0 € I = aspes “ds xauwa v0 zs “i C = EF O'S = = ¢ ~ = poom-Moiy Aumog “yynYyd¢g wnuvinbsaulfps wnusngi,A v0 (é = x = = O's ¢ = = = — jeAorkuudg asjey ST wnv1yavAG VUASOYII J, v0 - - = = Z o's = = = = ¢ (x) YoIeA “ds D191A v0 e i= Ca f= fi O'S = a ¢ = = [Iseg PIL ‘’] a4p3jna wnipodouny VC = i x Ol C cs = * = ae C asol “ds psoy 87 = 9 Vv C (6 69 s v1 C C cl aps) “ds wnisu1d 8°0 (4 = (6 = = 09 C - v = F Auraqdsey poy PIM ‘Qyo0oJ (YOoIq) snispjouvjau ‘dss “"] snappi ‘Jo snqny VC 8 4 14 zs > LS L0 = ¢ = = porUap[OH powiuisajs-AvID ‘UOWY S1psOWau OsDp1/OS v9 # CT = Ol 8s 3 = ee = We JOJOIA Inds-poyooH “WIS voUNpD vjO1A oY Ol v v = Ol OL [LAS 70 c = LA SSPID ou aIeUTUINOY ‘WNIDUIUINID “IRA “MG WNoUlUNID WNnIIUD_ 8°0 ¢ = é = si 79 70 =~ 9 = - AuIayD ayoyD ‘vuvmis.iia ‘dss "] vudIUIs11A "JO snUNA VC cl = = Fe = 9°9 9°9 = = - = SUIQUINJOD PIAA “] Sisuapvuvd vasajinby v0 = = ( = = OL = = L = = apponsKouoy “ds p1aa1u0T [[249AQ ¢ v & (4 I [[249AQ ¢ v € (4 I ouleN SYHUSTIOS ueay Aouonbaly % [BIOL (zUIP) I9AO_ [RIO], Ppanuiuod “| ATEaV Volkiiis v0 aE G = = = 10 = 10 = = (x) WOU YON VOUWIOD “"] VIyMDYIDI snumDYyY “Jd lg} V9 7f ia > = 70 10 10 = = (S) deoynys [jews “xyop] vynasod v1upjjainos v0 = G = = a 10 es 10 = - (x) [[@-[eoH{ UourUOD ‘sz1ps8jna “dss "'] sipsjna DyJauNsg v0 G = % 3 = 10 10 = = = ‘ds wnoiupg 80 = v a = = c0 3 C0 = = pryoio “ds prayjunjuv)g v0 G = re = = c0 c0 = = = qniys pelyueprun V0 = C = = I 0 = £0 = -: SAJDAQ "JOOS XAIDD 80 4 Es = = iG £0 €0 = = = (x) OUIg your ““quiey] vupisyuvg snuig v0 a G = aa = £0 ‘7 €0 = = SseIQn euueyy “ds vi1a0k]}.H v0 (6 = 5 ae = V0 v0 = = = p ONSOI polynuoprun = 80 v = = = =e C0 C0 = = = (,.) WOM-S UYOr 1S UOUTWIOD “7 wnyosofiad wnoiadkyy = 70 = G _ és ps C0 = SO - - ASPs NUJsoYyD “qualye AA VAUDISDI XAIDD s v0 Z a = et S0 SO = — — (S) SseID UNS aeIOWIO[H ~ “UILL, (PIL) DIDAaWUIO]8 DIsAaqualynp < v0 aes ees c°0 = On = 2 DODO Z 80 v as a = = 9'0 9'0 = = — Arroqieag uourWod ““suaids (“"J) 1sun-pan sojkydvjsojo.y A 80 v ee is = = 90 90 = = = (SY) IOLA 30g WIAyWON ‘aude v7j] 50) were reported even when the wolf population numbered fewer than 25 animals (Michigan Department of Natural Resources, unpublished data). Except for the 1960s, there were six or fewer obser- vations of lynx occurrence per decade. The increase in occurrences in the 1960s was associated with an unusually large population irruption in Canada (Theil 1987; McKelvey et al. 1999a). McKelvey et al. (1999a), like Theil (1987), con- cluded that recent lynx dynamics in the Great Lakes States appears to be related to immigration from Canada. In addition, McKelvey et al. (1999a) looked for a lynx population response to large increases in Snowshoe Hare populations throughout the Great Lakes Region. They reasoned that a resident popula- tion should have responded to the periodic increases in the number of hares. No association between lynx occurrence data and Snowshoe Hare population indices was found. Similarly, there is no association between lynx occurrences and a Snowshoe Hare pop- ulation index in the Upper Peninsula (Michigan Department of Natural Resources, unpublished data). While McKelvey et al. (1999a) are correct in pointing out that these data do not prove whether or not a local population exists, they certainly do not provide sup- port for the existence of a resident population. In conclusion, we found no evidence to suggest that a resident lynx population might exist in Michigan’s Upper Peninsula. The evidence suggests that lynx dispersing from Canada are only occasion- ally present in the Upper Peninsula. Although the dynamics of lynx colonization are unknown, McKelvey et al. (1999b) provide a good theoretical discussion of the dynamics and probabilities of suc- cessful colonization. Acknowledgments Funding for this study was provided by Federal Aid in Wildlife Restoration, Pittman-Robertson Pro- ject W-127-R. We thank the U.S. Forest Service, U.S. Fish and Wildlife Service, U.S. Park Service, Mead Corporation, Champion International, Shelter Bay Forests, White Water Associates, Michigan State University Museum, University of Michigan Museum, and the Wisconsin Department of Natural Resources for providing information on Canada Lynx occurrences. Dottie Lewis assisted in data compila- tion. Ray Rustem and Henry Campa III, reviewed the manuscript and provided constructive comments. Literature Cited Aubry, K.B., G. M. Koehler, and J. R. Squires. 1999. Ecology of Canada lynx in southern boreal forests. Chapter 13 in Lynx Science Report. U.S. Forest Service General Technical Report RMRS — 30. 240 Baker, R.H. 1983. Michigan mammals. Michigan State University Press, Lansing, Michigan. Elton, C., and M. Nicholson. 1942. The ten-year cycle in numbers of the lynx in Canada. Journal of Animal Ecol- ogy 11: 215-244. Harger, E.M. 1965. The status of the Canada lynx in Michigan. The Jack-Pine Warbler 43: 150-153. Keith, L.B. 1963. Wildlife’s ten-year cycle. University of Wisconsin Press, Madison, Wisconsin. McCord, C.M., and J. E. Cardoza. 1982. Bobcat and lynx. Pages 728-766 in Wild Mammals of North Amer- ica. Edited by J. A. Chapman and G. A. Feldhamer. John Hopkins University Press, Baltimore, Maryland. McKelvey, K.S., K. B. Aubry, and Y. K. Ortega. 1999a. History and distribution of lynx in the contiguous United States. Chapter 8 in Lynx Science Report. U.S. Forest Service General Technical Report RMRS — 30. McKelvey, K.S., S. W. Buskirk, and C. J. Krebs. 1999b. Theoretical insights into the population viability of lynx. Chapter 2 in Lynx Science Report. U.S. Forest Service General Technical Report RMRS — 30. Mech, L.D. 1980. Age, sex, reproduction, and spatial organization of lynxes colonizing northeastern Minne- sota. Journal of Mammalogy 61: 261-267. Michigan Department of Conservation. 1938. a biennial report. Mowat, G., K. G. Poole, and M. O’Donoghue. 1999. Ecology oF lynx in Northern Canada and Alaska. Chapter 9 in Lynx Science Report. U.S. Forest Service General Technical Report RMRS — 30. Poole, K.G. 1997. Dispersal patterns of lynx in the Northwest Territories. Journal of Wildlife Management 61: 497-505. THE CANADIAN FIELD-NATURALIST Vol. 115 Pruitt, W. O., Jr. 1951. Mammals of the Chase S. Osborn Preserve, Sugar Island, Michigan. Journal of Mam- malogy 32: 470-472. Quinn, N. W.S., and G. Parker. 1987. Lynx. Pages 682-695 in Wild Furbearer Management and Conserva- tion in North America. Edited by M. Novak, J. A. Baker, M.E. Obbard, and B. Malloch. Ministry of Natural Resources, Ontario, Canada. Ruggerio, L. F., K.B. Aubry, S. W. Buskirk, G. M. Koehler, C.J. Krebs, K.S. McKelvey, and J.R. Squires. 1999. The scientific basis for lynx conservation: qualified insights. Chapter 16 in Lynx Science Report. U.S. Forest Service General Technical Report RMRS — 30. Slough, B. G., and G. Mowat. 1996. Population dynamics of lynx in a refuge and interactions between harvested and unharvested populations. Journal of Wildlife Man- agement 60: 946-961. Thiel, R. P. 1987. The status of Canada lynx in Wiscon- sin, 1865-1980. Wisconsin Academy of Sciences, Arts and Letters 75: 90-96. U.S. Fish and Wildlife Service. 1998. Proposal to list the contiguous United States distinct population segment of the Canada lynx: proposed rule. Federal Register 63: 36994-37023. U.S. Fish and Wildlife Service. 2000. Determination of Threatened Status for the Contiguous U.S. Distinct Pop- ulation Segment of the Canada Lynx and Related Rule; Final Rule. Federal Register 65: 16051-16086. Wood, N.A. 1917. Notes on the mammals of Alger County, Michigan. University of Michigan, Museum of Zoology Occasional Papers Number 36. Received 7 March 2000 Accepted 13 March 2001 Diet of the Prairie Rattlesnake, Crotalus viridis viridis, in Southeastern Alberta MARGARET M. A. HILL, G. LAWRENCE POWELL, and ANTHONY P. RUSSELL Department of Biological Sciences, University of Calgary, 2500 University Drive NW, Calgary, Alberta T2N 1N4 Canada Hill, Margaret M. A., G. Lawrence Powell, and Anthony P. Russell. 2001. Diet of the Prairie Rattlesnake, Crotalus viridis viridis, in southeastern Alberta. Canadian Field-Naturalist 115(2): 241-246. The diet of the Prairie Rattlesnake, Crotalus viridis viridis, in a multiple land use area in southeastern Alberta was investi- gated by examining the gut contents of 20 road-killed individuals and the composition of eight scats recovered from sequestered individuals. All individuals yielding dietary data were adults. No data on the composition of the local small mammal community are available. The number of prey items per snake varied (gut contents: 0-4 items; scat samples: 1-3 items). Frequency of occurrence (individual prey items) was: Sagebrush Vole (Lagurus curtatus) — 53% gut contents, 68% scats; Meadow Vole (Microtus pennsylvanicus) — 38% gut contents, 8% scats; Olive-Backed Pocket Mouse (Perognathus maniculatus) — 8% scats; Western Jumping Mouse (Zapus princeps) — 8% scats; Richardson’s Ground Squirrel (Spermophilus richardsonii) — 8% scats; unidentified passerines — 9% gut contents. Although individual rat- tlesnakes sometimes took more than one prey taxon (gut contents: 1—3 taxa; scat samples: 1—2 taxa), multiple prey items in a single gut were almost always of the same species, suggesting that individual rattlesnakes tend to exploit patches where colonial burrowing prey species are abundant. Key Words: Prairie Rattlesnake, Crotalus viridis viridis, diet, gut contents, scats, Alberta. The Prairie Rattlesnake (Crotalus viridis viridis) reaches its northernmost geographical range limit in southeastern and central Alberta and southwestern Saskatchewan (Pendlebury 1977; Gannon and Secoy 1984, 1985; Russell and Bauer 2000). Various aspects of its ecology in this part of its range have been described (Lewin 1963; Pendlebury 1977; Gannon 1978; Macartney and Weichel 1993; Gannon and Secoy 1984, 1985; Kissner et al. 1996*), but its diet in this portion of its range has received relatively little attention (Gannon and Secoy 1984). Here we describe observations on the diet of adult C. v. viridis from a population in southeastern Alberta. The data were derived from necropsies of dead adult specimens recovered from a road that crossed our study area, and from analysis of scats from sequestered wild-caught rattlesnakes, in the course of a radiotelemetric study of movement and habitat use in this population. Materials and Methods Study Site The study area (approximately 9000 ha in extent, centred on 50°10’30”N, 110°30’W; W4M, 14, 4—5) is located in an area of mixed range and cropland roughly 20km NW of the city of Medicine Hat, Alberta. It is rolling, with large coulees running to the east from the South Saskatchewan River’s channel, and much of the area supports the mixed grass prairie association (Coupland 1950; Strong 1992*), modified by the grazing of domestic cattle. Human uses of the *See Documents Cited section study area include cattle ranching, grain and hay cul- tivation, hunting, and natural gas exploration and extraction. Prairie Rattlesnakes are well-known from this area (Pendlebury 1977; Watson and Russell 1997). There is a diverse potential prey pool of small mammal species (Soper 1964; Smith 1993) and ground-nesting grassland passerines (Semenchuk 1992) available in the area. Dietary analysis For dead-on-the-road (DOR) rattlesnakes, date of collection, snout-vent length (SVL) and weight (when the state of the carcass permitted), location and sex of each individual were recorded prior to fixation in 10% neutral buffered formalin (as soon as possible after discovery) and subsequent preservation in 70% ethanol. Wild-caught adult rattlesnakes from the study area were also held in the laboratory, either for radio-transmitter implantation or specifically for scat collection, and their scats retained. Date and capture location of the donor were recorded for scats. Each DOR snake was radiographed and the posi- tions of gut contents (oesophagus, stomach, small intestine, and large intestine [including cloaca]) were recorded before they were removed from the body by dissection. Diagnostic bones or bone fragments (e.g., mandibles, femora, humeri etc.) were isolated, identi- fied and counted in order to determine the minimum number of identifiable mammalian prey items per species per snake. Samples of hair were also taken from each gut section and kept for identification pur- poses. Identification of mammalian prey items to species was made by comparing skulls and tooth cusp patterns of identified reference specimens (from the 241 242 University of Calgary Museum of Zoology) to those found in the gut, and by features of the external hair surface such as overall shape and cuticular scale pattern. Scats were softened in tap water for about 20 min- utes and then teased apart under a dissecting micro- scope using watchmaker’s forceps. All diagnostic bone fragments, and hair samples, were recovered and retained from each as described for the gut con- tents above. Due to the small size of the sample as a whole, differences in diet between the sexes were not tested, nor were differences between size classes, as there was little variation in SVL among the DOR snakes. The relationship between the number of prey items and the number of prey taxa in rattlesnake guts was tested by a Pearson’s correlation coefficient. The incidences of the main prey species in rattlesnake guts were tested for independence of occurrence by a G-test (Sokal and Rohlf 1995). Temporal patterns in the distributions of the main prey species through the study period are examined graphically; patterns of incidences and numbers of captures are compared with Sign tests, Wilcoxon Signed-ranks tests and Wald-Wolfowitz Runs tests, carried out with SYS- TAT (Wilkinson 1990). The statistical analysis of the gut data was intended to detect patterns of prey encounter. Crotalus v. viridis has elsewhere been shown to exploit patches of high prey density (Duvall et al. 1985, 1990), and a number of the potential prey species in this area occupy localized burrow systems (Soper 1964; Pattie and Hoffmann 1992*; Nietfeld and Roy 1993*). The data from the scat sample are treated separately from the gut sam- ple and are not analysed statistically; we have no data on how long faeces are retained in free-ranging rattlesnakes and cannot assume that an individual scat represents the remnants of a single feeding bout. Results Material Examined Twenty freshly-killed DOR snakes were collected during the 1997 field season (1 May—15 October) THE CANADIAN FIELD-NATURALIST Vol. 115 along RR 43 (5 females, 15 males — all DOR speci- mens were sexually mature). Eight scats were col- lected from wild-caught adult individuals temporari- ly held in the lab. Diet composition A total of 32 prey items (three avian, 29 mam- malian) were recovered from the 20 DOR individu- als (Table 1). In all three cases of avian remains, mammalian hair and bone fragments were found in the same region of the gut, intermixed with feathers. Mammalian remains only, representing 13 individual prey items, were retrieved from the 8 scats (Table 1). A few individual snakes had gut or scat contents at the upper range of dietary diversity, but snakes taking single prey taxa frequently took several items (Tables 2, 3). The modal number of items in the gut was | (range 0-4), and the modal number of taxa was | (range 0-3). For scats, the modal number of items was | (range 1-3) and the modal number of taxa | (range 1-2). There was a positive high corre- lation between the number of items and the number of prey taxa in the gut (Table 2); the pattern is less clear-cut in the scats (Table 3). When the numbers of Microtus pennsylvanicus and Lagurus curtatus (the two commonest prey species: Table 1) in guts were cross tabulated a pattern was evident; of the 17 guts containing these two species, they co-occurred only in three, in each case represented by one individual of each prey species, and the distributions of the numbers of each species were independent (Table 4); snakes tended to feed upon one or the other. Only one scat contained M. pennsylvanicus remains, and it contained no L. curtatus remains (Table 5). The dis- tributions of neither L. curtatus nor M. pennsylvani- cus in guts are significantly clustered in time through the study period (Figure 1; Wald-Wolfowitz Runs test: L. curtatus Z = -0.429; p = 0.668; M. pennsyl- vanicus Z = -1.097; p = 0.272). The temporal distri- butions of the two prey species in guts broadly over- lap over the study period (Figure 1) and there are no significant differences in either the number of days over the study period in which rattlesnakes preyed on either L. curtatus or M. pennsylvanicus (Sign test; TABLE |. Prey of adult Prairie Rattlesnakes (Crotalus viridis viridis) north of Medicine Hat, Alberta, from gut contents of road-killed snakes (20 individuals) and scats (8 individuals). See text for methods of prey species identification. Prey Species Lagurus curtatus (Sagebrush Vole) Microtus pennsylvanicus (Meadow Vole) Perognathus maniculatus (Olive-Backed Pocket Mouse) Zapus princeps (Western Jumping Mouse) Spermophilus richardsonii (Richardson’s Ground Squirrel) Unidentified Passerine Birds Totals Gut Contents Scats No. of Prey % of No. of Prey % of Items Total items Total 7 Soni 9 69.23 12 37.50 1 7.69 0 0.00 1 7.69 0 0.00 1 7.69 0 0.00 1 7.69 3 9.38 0) 0.00 SU is | | 2001 HILL, POWELL, AND RUSSELL: DIET OF THE PRAIRIE RATTLESNAKE 243 O 120 125 180 186 140 145 160 15 160 165 170 175 1680 186 190 106 200 206 210 215 220 228 280 285 240 246 250 256 O 120 126 180 185 140 145 150 165 160 185 170 175 180 185 190 186 200 206 210 216 220 225 280 285 240 246 260 256 FicurE |. Distributions of (above) the number of Crotalus viridis viridis with Lagurus curtatus in their guts over the study period; n = 12; and (below) the number of Crotalus viridis viridis with Microtus pennsylvanicus in their guts over the study period; n = 8. The verticle axis gives the number of rattlesnakes and the horizontal the Julian day in 1997 (counting day 1 as | January). 244 THE CANADIAN FIELD-NATURALIST Vol. 115 TABLE 2. Cross-tabulation of number of prey items per gut and number of prey taxa per gut for twenty DOR Crotalus viridis viridis from study area (— = not applicable). Pearson’s r = 0.722, p < 0.001. No Prey One Item Three Taxa - — Two Taxa — - One Taxon = i No Prey 3 p = 0.494), or in the number of prey of either species that a rattlesnake was likely to be carrying in its gut (Wilcoxon Signed-ranks test; Z = -0.835; p = 0.404). Within gut samples, items were much more abundant in the stomach and small intestine than in the large intestine, particularly multiple occurrences of L. cur- tatus and M. pennsylvanicus. Discussion All of the DOR snakes were collected along RR 43, a transect that runs directly through the activity ranges of at least part of the area’s summer rat- tlesnake population. The disturbed soil of the ditches flanking the roadbed may have represented a particu- larly favourable microhabitat for burrowing rodents and thus attracted a disproportionate number of rat- tlesnakes compared to the surrounding relatively undisturbed prairie. If this is the case our sample is biased, representing rattlesnakes preying upon rodent species that dwelt preferentially on disturbed ground. Without data on rodent distribution among micro- habitat types in this area, this possibility cannot be assessed. Only five of 20 DORs were female, which may be a function of the small sample size rather than a sam- pling bias. However, reproduction appears to be biennial for female Crotalus v. viridis at this latitude (Gannon and Secoy 1984; Macartney and Weichel TABLE 3. Cross-tabulation of number of prey items per scat and number of prey taxa per scat for eight Crotalus viridis viridis scats from study area (— = not applicable). One Item Two Items Three Items Two Taxa ~ 2 0) One Taxon 4 1 1 Two Items Three Items Four Items ;Nnuor 0 1 0 | © ws) | 1993; Kissner et al. 1996*), in which case we would expect a male:female ratio of approximately 13:7 in the wide-ranging fraction of the adult population, if all of the gravid females remained close to the hiber- nacula (Gannon and Secoy 1984) and if the natal sex ratio is 1:1. This ratio is close to what we found and none of the DOR females were carrying developed embryos. Crotalus viridis is a very widely distributed species (Klauber 1972), and shows considerable vari- ety in its diet over its range. In the more southerly portions, larger C. viridis have been observed to take subadults of any species of ground squirrel (Spermophilus) locally present, as well as subadult Black-tailed Prairie Dogs (Cynomys ludovicianus) and Nuttall’s Cottontail (Sylvilagus nuttallii); species of Dipodomys, Neotoma, Perognathus, Reithrodonto- mys, Peromyscus, and Microtus are also taken (Fitch and Twining 1946; Fitch 1949; Klauber 1972; Ludlow 1981; Duvall et al. 1985, 1990; Diller and Johnson 1988; Macartney 1989; Wallace and Diller 1990; Diller and Wallace 1996). Also, Lark Buntings (Calamospiza) and other ground-nesting bird species have been recorded as frequent prey; lizards and frogs are taken predominantly by subadult snakes (Klauber 1972). Duvall et al. (1985, 1990) showed that the annual movement patterns of the seasonally migratory adults in a high-elevation population of Crotalus v. viridis in southern Wyoming were strong- ly associated with the distribution across the land- scape of local concentrations of Peromyscus manicu- latus, which constituted the rattlesnake’s principal prey in this area. We found Crotalus v. viridis in this part of south- eastern Alberta to subsist primarily on Lagurus cur- tatus and Microtus pennsylvanicus, and to a much smaller extent upon three other terrestrial rodent species and an unknown passerine (Table 1). Results TABLE 4. Cross-tabulation of number of individual Lagurus curtatus per gut and number of individual Microtus pennsyl- vanicus per gut for twenty DOR Crotalus viridis viridis from study area. G = 9.17, 9 df; NS. No Lagurus One Lagurus Two Lagurus Three Lagurus No Microtus Dy D 3 i One Microtus 3 3 0 0 Two Microtus 0 0 0 0 Three Microtus y 0 0 0 2001 TABLE 5. Cross-tabulation of number of individual Lagurus curtatus per scat and number of individual Microtus penn- sylvanicus per scat for eight Crotalus viridis viridis scats from study area. No One Two Three Lagurus Lagurus Lagurus Lagurus No Microtus | + 1 1 One Microtus 1 0 0 0 of this study agree with observations in previous lit- erature, in that the diet consists primarily of small mammals (Klauber 1972; Mushinsky 1987). The rel- ative abundances of the prey species taken by our sample of adult Alberta C. viridis, in addition to the absence of Pocket Gophers (Thomomys bottae) and White-footed Mice (Peromyscus maniculatus), and the rarity of Richardson’s Ground Squirrels (Spermophilus richardsonii) in guts or scat (Table 1), are difficult to evaluate, since there are no quanti- tative data on the composition of the local small mammal community. The absence of P. maniculatus is particularly puzzling, as this species is generally abundant on the Alberta mixed-grass prairies (Soper 1964; Pattie and Hoffmann 1992*; Smith 1993) and is an important prey species for C. viridis elsewhere (Duvall et al. 1985, 1990; Macartney 1989). Spermophilus richardsonii, a visible species when present, was seldom seen in the study area during the summer of 1997, suggesting that local population densities were low. Both P. maniculatus and S. richardsonii were found in C. viridis guts in the vicinity of Leader, Saskatchewan, by Gannon and Secoy (1984), indicating that they are both dietary items in this part of its range. Of the prey species that did occur in our sample, both L. curtatus and Perognathus fasciatus are considered to be rare in Alberta (Smith 1993), although the former can apparently attain localized high densities (Soper 1931; Pattie and Hoffmann 1992*). Both Microtus pennsylvanicus and Lagurus curtatus form colonies marked by burrow and runway construction (Soper 1931, 1964; Pattie and Hoffmann 1992*), whereas Perognathus fasciatus and Zapus princeps are soli- tary burrowers (Soper 1964; Pattie and Hoffmann 1992*). Crotalus viridis, an ambush predator (Klauber 1972; Diller and Wallace 1996), favours waiting in burrows for rodents to enter them (Duvall et al. 1985, 1990), and all of these species would be vulnerable to such tactics. The relationship between the number of prey items and number of prey taxa per rattlesnake (Tables 2, 3) suggests that individual rattlesnakes in this area exploit patchily distributed rodent colonies. Most gut (Table 2) and scat (Table 3) samples yield- ed prey items of only one taxon, regardless of their number. This would be expected in a predator HILL, POWELL, AND RUSSELL: DIET OF THE PRAIRIE RATTLESNAKE 245 exploiting a patch of high prey density. The frequen- cy of rattlesnakes that had consumed more than one prey item (Tables 2, 3) also suggests exploitation of a patchily distributed resource, as does the tendency of multiple prey items in single guts to be restricted to the stomach and small intestine. Macartney (1989) seldom found more than one prey item or one prey taxon per snake in his study of an Okanagan Valley population of Crotalus v. oreganus, which did not display the movement patterns, apparently associated with prey distribution, typical of our population (Russell et al. unpublished data) or that studied by Duvall et al. (1985, 1990). The relationship between the presence of M. pennsylvanicus and that of L. cur- tatus (Tables 4, 5; Figure 1) in guts and scats is fur- ther evidence that runway systems of either colonial species were a particular target of the rattlesnakes in our sample, and that they tended to catch large num- bers of prey individuals in such systems. The lack of difference in the temporal distributions of these two prey species (Figure 1) in DOR rattlesnakes suggests further that high prey density was the criterion for patch choice, rather than prey species. Avian remains in C. viridis gut contents have been reported from other studies (Fitch 1949; Klauber 1972; Wallace and Diller 1990). Those found in our sample are most likely the fledglings of one of a number of ground-nesting species of passer- ine, the adults of which are known to frequently leave the immediate area of the nest, possibly in response to the threat of nest predation (Dorothy Hill, personal communication). It is probable that this occasional consumption of fledgling birds is opportunistic, because it is so rarely observed, and because, in the case of this study, avian remains were always found in association with mammalian remains in the gut. Acknowledgments We are obliged to Joanne Skilnick and Nancy O’Brien for their very capable work in the field, Jonathan Wright and Ed Ruff for invaluable help at critical times, Dale Eslinger and the rest of the staff of the Medicine Hat Fish and Wildlife Services office for the assistance which made this project pos- sible, Dr. Hal Canham of Medicine Hat General Hospital for medical advice, and Bob Mutch and Bonnie Leiteit of Medicine Hat College for lab facil- ities. We thank Bill Davies of the Davies Cattle Company for access to the study area. We are also grateful to Lisa McGregor for her help in the lab, Warren Fitch of the University of Calgary Museum of Zoology for providing reference specimens, and Dorothy Hill for some comments on ground-nesting passerines. An earlier version of this paper was greatly improved by the comments of two anony- mous reviewers. This study was funded partly by grants from Alberta Sports, Recreation, Parks and 246 Wildlife Foundation and from the Alberta Conser- vation Association to A. P. Russell, partly through a grant from the Friends of the Environment (Canada Trust) of Medicine Hat, and partly by Alberta Environmental Protection, Fish and Wildlife Ser- vices. All research and collection was undertaken under Alberta Environmental Protection collection licence 1668 and Alberta Environmental Protection research permit 0405 GP, both issued to A. P. Russell. Documents Cited (marked * in text) Kissner, K. J., D. M. Secoy, and M.R. Forbes. 1996. Assessing population size and den use of Prairie rat- tlesnakes (Crotalus viridis viridis) in southern Saskat- chewan. Grasslands National Park Annual Report Volume 1: 27-34. Nietfeld, M.T., and L.D. Roy. 1993. Alberta pocket gopher survey. Alberta Environmental Centre, Vegre- ville, Alberta. AECV93-R2. 28 pages. Pattie, D. L., and R.S. Hoffmann. 1992. Mammals of the North American Parks and Prairies. Edmonton, ISBN 0- 9694482-0-1 Strong, W.L. 1992. Ecoregions and Ecodistricts of Alberta. Volume |. Alberta Forestry, Lands and Wildlife, Land Information Services Division, 4 Floor, North Petroleum Plaza, 9945- 108 St., Edmonton, Alberta - T5K 2G6 Literature Cited Coupland, R.T. 1950. Ecology of mixed prairie in Can- ada. Ecological Monographs 20: 271-315. Diller, L. V., and D.R. Johnson. 1988. Food habits, con- sumption rates and predation rates of Western rattle- snakes and gopher snakes in south-western Idaho. Herpetologica 44: 228-233. Diller, L. V., and R. L. Wallace. 1996. Comparative ecol- ogy of two snake species (Crotalus viridis and Pituophis melanoleucus) in south-western Idaho. Herpetologica 52: 343-360. Duvall, D., M.B. King, and K. J. Gutzwiller. 1985. Behavioural ecology and ethology of the Prairie rat- tlesnake. National Geographic Research 1: 80-111. Duvall, D., M. J. Goode, W. K. Hayes, J. K. Leonhardt, and D.G. Brown. 1990. Prairie rattlesnake vernal migration: field experimental analyses and survival value. National Geographic Research 6: 457-469. Fitch, H.S. 1949. Study of snake populations in Central California. American Midland Naturalist 41: 513-579. Fitch, H.S., and H. Twining. 1946. Feeding habits of the Pacific rattlesnake. Copeia 1946: 64—71. Gannon, V.P. J. 1978. Factors limiting the distribution of the Prairie rattlesnake. Blue Jay 36: 142-144. Gannon, V.P.J., and D.M. Secoy. 1984. Growth and reproductive rates of a northern population of the Prairie rattlesnake, Crotalus v. viridis. Journal of Herpetology 18: 13-19. THE CANADIAN FIELD-NATURALIST Vol. 115 Gannon, V.P. J., and D. M. Secoy. 1985. Seasonal and daily activity patterns in a Canadian population of the Prairie rattlesnake, Crotalus viridis viridis. Canadian Journal of Zoology 63: 86-91. Klauber, L. M. 1972. Rattlesnakes: Their habits, life his- tories, and influence on mankind. Volume I. University of California Press, Berkeley, California. Lewin, V. 1963. The herpetofauna of southeastern Alberta. Canadian Field Naturalist 77: 203-214. Ludlow, M. E. 1981. Observations on Crotalus v. viridis (Rafinesque) and the herpetofauna of the Ken-Caryl Ranch, Jefferson County, Colorado. Herpetological Review 12: 50-52. Macartney, J.M. 1989. Diet of the Northern Pacific rat- tlesnake, Crotalus viridis oreganus, in British Colombia. Herpetologica 45: 299-304. Macartney, J. M., and B. Weichel. 1993. Status of the Prairie rattlesnake and the eastern yellow-bellied racer in Saskatchewan. Pages 291—299 in Proceedings of the Third Prairie Conservation and Endangered Species Workshop. Edited by G. L. Holroyd, H. L. Dickson, M. Regnier, and H. C. Smith. Provincial Museum of Alberta Natural History Occasional Paper Number 19. Mushinsky, H.R. 1987. Foraging Ecology. Pages 302-334 in Snakes: ecology and evolutionary biology. Edited by R. A. Seigel, J.T. Collins, and S. S. Novak. Macmillan Publishing Co., New York. Pendlebury, G. B. 1977. Distribution and abundance of the Prairie rattlesnake, Crotalus viridis viridis, in Can- ada. Canadian Field-Naturalist 91: 122-129. Russell, A. P., and A.M. Bauer. 2000. The amphibians and reptiles of Alberta. 2nd edition. University of Calgary Press, Calgary, Alberta. Semenchuk, G.P. 1992. The atlas of the breeding birds of Alberta. Federation of Alberta Naturalists, Edmonton, Alberta. Smith, H. C. 1993. Alberta Mammals. An Atlas and Guide. Provincial Museum of Alberta, Edmonton, Alberta. Sokal, R. R., and F. J. Rohlf. 1995. Biometry. The princi- ples and practice of statistics in biological research. 2nd Edition. W.H. Freeman and Company, New York, N.Y. Soper, J.D. 1931. Field notes on the Pallid meadow mouse, Lagurus pallidus (Merriam). Canadian Field- Naturalist 45: 209-214. Soper, J.D. 1964. The Mammals of Alberta. The Hambly Press Ltd., Edmonton, Alberta. Wallace, R. L., and L. V. Diller. 1990. Feeding ecology of the rattlesnake, Crotalus viridis oreganus, in northern Idaho. Journal of Herpetology 24: 246-253. Watson, S.M., and A. P. Russell. 1997. Status of the Prairie rattlesnake (Crotalus viridis viridis) in Alberta. Alberta Environmental Protection, Wildlife Management — Division, Wildlife Status Report Number 6, Edmonton, Alberta. 26 pages. : Wilkinson, L. 1990. SYSTAT: The System for Statistics. SYSTAT, Inc., Evanston, Illinois. Received 10 May 2000 Accepted 25 June 2001 Effects of Enclosed Large Ungulates on Small Mammals at Land Between The Lakes, Kentucky CLARE C. WEICKERT, JOSEPH C. WHITTAKER, and GEORGE A. FELDHAMER Department of Zoology, Southern Illinois University at Carbondale, Carbondale, Illinois 62901 USA Weickert, Clare C., Joseph C. Whittaker, and George A. Feldhamer. 2001. Effects of enclosed large ungulates on small mammals at Land Between The Lakes, Kentucky. Canadian Field-Naturalist 115(2): 247-250. Ungulates can affect primary production, plant and animal species composition, nutrient cycling, and soil properties. We conducted a study at the Tennessee Valley Authority's Land Between The Lakes to investigate the effects of introduced Elk (Cervus elaphus) and Bison (Bison bison) on small mammal fauna in a 324-ha enclosure. From June 1998 through May 1999, live traps were set for small mammals in open-canopy hardwood, closed-canopy hardwood, and pine sites both inside and outside the enclosure, for 7020 trap nights. Small mammals were captured significantly more often inside than outside the enclosure; habitat type was not a significant factor. Habitat management practices inside the ungulate enclosure, including burning, mowing, and fertilizing, may have contributed to the higher abundance of small mammals. Key Words: Bison, Bison bison, Elk, Cervus elaphus, Land Between The Lakes, Kentucky, small mammals, ungulates Small mammal populations provide a critical resource base for a variety of wildlife species at higher trophic levels. Ungulates can affect sym- patric small mammal populations both directly and indirectly. Ungulates may reduce forage available for herbivorous small mammals by consuming veg- etation and mast, or by altering local plant species composition (Horsley and Marquis 1983; Augustine and McNaughton 1998). Reduction or prevention of tree and shrub regeneration as a result of ungulate browsing in forest habitats (Crouch 1981; Putman 1986; Gill 1992) may result in long-term changes in canopy species composition and structure (Healy 1997; Singer et al. 1998). Feeding, trampling, and wallowing by ungulates also reduces protective veg- etation cover, potentially increasing vulnerability of small mammals to predation (Grant et al. 1982; Bock et al. 1984). Small mammal populations also may be influenced positively by ungulates. Ungulate defecation can contribute to accelerated nutrient cycling (McNaughton 1985) and seed dispersal (Janzen 1986; Malo and Suarez 1995). Also, antlers shed by cervids provide supplementary dietary min- erals for small mammals that consume them (Couturier 1995). Generally, however, cumulative abundance of small mammals is greater on areas from which ungulates are excluded (Bock et al. 1984; Heske and Campbell 1991; Hayward et al. 1997; Keesing 1998). Most previous studies on the effects of ungulates on small mammals have been conducted in desert or semi-desert habitats. An introduction of Elk (Cervus elaphus) and Bison (Bison bison) to a large enclo- sure provided an opportunity to examine the ecologi- cal impact of high ungulate density on relative abun- dance of small mammals in deciduous and pine forested habitats. Methods and Materials Our study was conducted in the Trigg County, Kentucky, portion of Land Between The Lakes (LBL), an approximately 70 000-ha natural area administered during the period of the study by the Tennessee Valley Authority in western Kentucky and Tennessee. LBL extends from approximately 36°30’ to 37°00'N, 87°50’ to 88°05’W (see Burbank and Smith [1966] for a description of habitat and hydrology of LBL). In January 1996, 29 wild Elk from Alberta, Canada, were released into a newly constructed, 324-ha drive-through wildlife viewing area enclosed with a 3-m-high deer-proof fence. During summer 1996, 44 Bison from an existing LBL herd were released into the enclosure. White- tailed Deer (Odocoileus virginianus) occurred natu- rally both inside and outside the enclosure. During the study, total ungulate density inside the enclosure, including Elk, Bison, and White-tailed Deer, was approximately 46.2 /km?. Elsewhere on LBL, the estimated density of White-tailed Deer was between 6.2 and 10.0 deer/km? (S. Bloemer, LBL, unpub- lished census data). There were no free-ranging Elk or Bison outside the enclosure. We sampled 12 forested sites for small mammal abundance during this study: six inside the enclosure (experimental) and six that were <10 km from the enclosure (control sites). Each group of six sites in- cluded two open-canopy hardwood sites (this habitat type comprised approximately 5.0% of the enclosure), two closed-canopy hardwood sites (this habitat type comprised approximately 40.8% of the enclosure), and two pine sites (this habitat type comprised approximately 4.8% of the enclosure). Hardwood sites were dominated by an overstory of Oak (Quercus spp.) and Hickory (Carya spp.). Common understory was Oak, Hickory, Persimmon (Diospyros virgini- 247 248 ana), Black Cherry (Prunus serotina), and Sassafras (Sassafras albidum). Pine sites were dominated by an overstory of Virginia Pine (Pinus virginiana), with a sparse understory of Oak and Ash (Fraxinus sp.). Stand age, most recent harvest date, and gross vegeta- tion structure (mean DBH, canopy closure, fallen logs) were similar for sites inside and outside the enclosure. In the enclosure, some forested area was cleared and planted to warm- and cool-season grasses, primarily Indiangrass (Sorghastrum nutans), Little Bluestem (Andropogon scoparium), and Switchgrass. (Panicum virgatum), as well as Bush Clover (Les- pedeza sp.). Grasslands, including recently cleared and originally existing areas, made up the remainder of the enclosure. We did not trap small mammals on grassland areas, however, because there were no grassland sites outside the enclosure for controls. Elk and Bison used all habitat types within the enclosure based on deposition of fecal material (Weickert, unpublished data). Within the enclosure, forested areas were burned on an irregular schedule, and grass- land patches were burned, fertilized, and mowed on an irregular schedule of approximately 12 months. Ungulates were supplementally fed with hay and, periodically, commercially produced pelletized feed during winter months. A 100-m transect with 10-m spacing between trap stations was set at each site. Fifteen small (5.1 x 6.4 x 16.5 cm) Sherman live traps (H. B. Sherman Co., Tallahassee, Florida) baited with cracked corn and sunflower seeds were used on each transect, with one or two traps at alternate stations. We sampled each site 13 times (at about four- to five-week inter- vals), for three nights each time, during the period from June 1998 through May 1999. Experimental and control sites were sampled simultaneously to alleviate influences such as weather or moonlight. Trapping effort totaled 585 trap nights per site, and a total of 7020 trap nights. We toe-clipped captured mammals and recorded their body mass, sex, and reproductive condition. Animals were released at the point of capture. Data on small mammal captures were transformed (log,,). Data were analyzed using THE CANADIAN FIELD-NATURALIST Vol. 115 two-way analysis of variance with enclosure and habitat type as factors. Analyses were considered significant at P< 0.05. Results We captured a total of 145 individuals of five species of small mammals inside the ungulate enclo- sure, compared to 96 individuals of three species on the control sites (Table 1). White-footed Mice (Peromyscus leucopus) comprised 89.7% of the indi- viduals inside the enclosure and 71.9% of individu- als outside the enclosure. Across all habitat types, there were more total small mammal captures (initial captures and recaptures) inside the enclosure than outside (F = 11.27, P=0.015). Habitat type was not a significant factor (F=4.07, P=0.076) nor was the interaction between habitat type and enclosure type (F=2.55, P0156), Discussion The greater abundance of small mammals cap- tured inside the high-ungulate-density enclosure was unexpected. Similar studies, as noted, have generally shown lower abundance of small mammals in areas of high ungulate density. Keesing (1998), working in East African savanna habitat, found that the African Pouched Rat (Saccostomus mearnsi), the most com- mon small mammal species in the area, responded to the exclusion of ungulates by increasing two-fold in density within one year (40% higher abundance than outside the exclosures). Similarly, Bock et al. (1984) and Heske and Campbell (1991) captured cumula- tively more rodents inside areas from which live- stock and Mule Deer (Odocoileus hemionus) had been excluded in Arizona. Hazebroek et al. (1995) captured significantly more small mammals inside than outside areas in hardwood and pine stands in the central Netherlands from which Red Deer (Cer- vus elaphus), Roe Deer (Capreolus capreolus), and Wild Boar (Sus scrofa) had been excluded for the six years prior to their study. Unfortunately, we could not determine the relative abundance of small mammals in the enclosure prior TABLE 1. Species and number of individual small mammals captured from June 1998 through May 1999 on three habitat types (P = pine forest; CC = closed-canopy forest; OC = open-canopy forest) inside and outside of an ungulate enclosure on Land Between The Lakes, Kentucky. Inside Enclosure Species P Ce Peromyscus leucopus 48 46 Blarina brevicauda 2 Microtus pinetorum 0 Ochrotomys nuttalli 0 Glaucomys volans 0 Totals 50 49 Outside Enclosure OC P Ce OC 36 20 19 30 5 16 5 1 1 0 1 4 4 0 0 0 0 0 0 0 46 36 yess) 39 2001 to the introduction of Elk and Bison. The site of the enclosure was chosen primarily because it was cen- trally located on LBL and contained a small, remnant portion of native prairie. There were no pre-existing differences known in any of the forest habitats that would have caused greater abundance of small mam- mals to occur prior to enclosure. Therefore, we expect that small mammal abundance was initially the same as on control sites. Several factors may have affected our results. The ungulate viewing enclosure on LBL was two years old at the start of the study. The resident ungulates may not yet have altered the forest habitats to the point of depleting food resources or protective herbaceous cover for sympatric small mammals. The approximately 5 x higher ungulate density inside the enclosure than outside has led to a proportionally higher rate of dung deposition. This may be benefit- ing the small mammals by allowing more rapid nutrient cycling through the system. Janzen (1986) found that some rodents mined and consumed seeds from ungulate dung. In North America, mice, specif- ically Peromyscus sp., and other small mammals are known to mine and consume seeds from the dung of bears (Ursus sp.; Bermejo et al. 1998) and Raccoons (Procyon lotor; Page et al. 1999). If they are also consuming seeds from ungulate dung, the dung inside the high-ungulate-density enclosure may pro- vide an additional food source. Herbivorous or gen- eralist small mammals may consume vegetation newly sprouted from those seeds. Also, insectivores and generalists may feed upon the accompanying invertebrate resource. The Elk and Bison dung prob- ably contains fewer seeds, however, than that of omnivorous bears and Raccoons. Aside from the high ungulate density, factors most strongly affecting the habitat inside the enclosure were the management activities designed to maxi- mize quality and quantity of forage for the ungulates — burning, fertilizing, and supplemental feeding. Burning and fertilizing may directly influence small mammals by promoting new forage available to them. Indirect benefits may result from supplemental hay by increasing available forage for ungulates and reducing their effects on the vegetation. Based on the results of previous studies, small mammal populations within the enclosure may decline if the high-density ungulate populations are permitted to effect more drastic changes in their habi- tat than have thus far been observed. Currently, how- ever, the expected negative impact of high ungulate density on the cumulative abundance of small mam- mals appears to have been outweighed by habitat management practices that promoted small mammal abundance. Introduction of free-ranging ungulates probably will not include these habitat management practices, and more “typical” negative effects on small mammal populations might be expected. WEICKERT, WHITTAKER, AND FELDHAMER: EFFECTS OF LARGE UNGULATES 249 Acknowledgments Funding was provided by a Senior Research Fel- lowship through the Center for Field Biology, Austin Peay State University. Housing was provided by the Wildlife Management Section, Land Between The Lakes, Tennessee Valley Authority. We thank J. Reeve, Southern Illinois University, for assistance with statistical analyses. Literature Cited Augustine, D. J., and S. J. McNaughton. 1998. Ungulate effects on the functional species composition of plant communities: herbivore selectivity and plant tolerance. Journal of Wildlife Management 62: 1165-1183. Bermejo, T., A. Traveset, and M.F. Willson. 1998. Post-dispersal seed predation in the temperate rainforest of southeast Alaska. Canadian Field-Naturalist 112: 510-512. Bock, C.E., J. H. Bock, W.R. Kenney, and V.M. Hawthorne. 1984. Responses of birds, rodents, and vegetation to livestock exclosure in a semidesert grass- land site. Journal of Range Management 37: 239-242. Burbank, J.H., and R.D. Smith. 1966. Wildlife habitat, forest resource, and hydrologic condition inventory at Land Between The Lakes. Proceedings of the South- eastern Association of Game and Fish Commissioners 20: 6-14. Couturier, D. 1995. Disappearance of shed white-tailed deer antlers in Leelanau and Chippewa Counties, Michi- gan. Michigan Academician 27: 386-387. Crouch, G. L. 1981. Part 3. Effects of deer on forest veg- etation. Pages 449-457 in Mule and black-tailed deer of North America. Edited by O.C. Wallmo. University of Nebraska Press, Lincoln. Gill, R.M. A. 1992. A review of damage by mammals in north temperate forests: 1. Deer. Forestry 65: 145-169. Grant, W. E., E. C. Birney, N. R. French, and D. M. Swift. 1982. Structure and productivity of grassland small mam- mal communities related to grazing-induced changes in vegetative cover. Journal of Mammalogy 63: 248-260. Hayward, B., E. J. Heske, and C. W. Painter. 1997. Effects of livestock grazing on small mammals at a desert cienaga. Journal of Wildlife Management 61: 123-129. Hazebroek, E., G. W. T. A. Groot Bruinderink, and J. B. Van Biezen. 1995. Changes in the occurrence of small mammals following the exclusion of red deer, roe deer and wild boar. Lutra 38: 50-59. Healy, W.M. 1997. Influence of deer on the structure and composition of oak forests in central Massachusetts. Pages 249-266 in The science of overabundance: deer ecology and population management. Edited by W.J. McShea, H. B. Underwood, and J.H. Rappole. Smith- sonian Institution Press, Washington. Heske, E. J., and M. Campbell. 1991. Effects of an 11- year livestock exclosure on rodent and ant numbers in the Chihuahuan Desert, southeastern Arizona. The South- western Naturalist 36: 89-93. Horsley, S. B., and D. A. Marquis. 1983. Interference by weeds and deer with Allegheny hardwood reproduction. Canadian Journal of Forest Research 13: 61-69. Janzen, D.H. 1986. Mice, big mammals, and seeds: it matters who defecates what where. Pages 251-271 in Frugivores and seed dispersal. Edited by A. Estrada and T. H. Fleming. W. Junk Publishers, Dordrecht. 250 Keesing, F. 1998. Impacts of ungulates on the demogra- phy and diversity of small mammals in central Kenya. Oecologia 116: 381-389. Malo, J. E., and F. Suarez. 1995. Herbivorous mammals as seed dispersers in a Mediterranean dehesa. Oecologia 104: 246-255. McNaughton, S. J. 1985. Ecology of a grazing ecosystem: the Serengeti. Ecological Monographs 55: 259— 294. Page, L. K., R. K. Swihart, and K.R. Kazacos. 1999. Implications of raccoon latrines in the epizootiology of Baylisascariasis. Journal of Wildlife Diseases 35: 474— 480. THE CANADIAN FIELD-NATURALIST Vol. 115 Putman, R. J. 1986. Competition and coexistence in a multi-species grazing system. Acta Theriologica 31: 271-291. Singer, F. J., L.C. Zeigenfuss, R.C. Cates, and D. T. Barnett. 1998. Elk, multiple factors, and persistence of willows in national parks. Wildlife Society Bulletin 26: 419-428. Received 8 June 2000 Accepted 25 June 2001 Thermal Habitat Use and Evidence of Seasonal Migration by Rocky Mountain Tailed Frogs, Ascaphus montanus, in Montana SUSAN B. ADAmMs! and CHRISTOPHER A. FRISSELL? Flathead Lake Biological Station, The University of Montana, Polson, Montana 59860 USA 'Present address and correspondence to: USDA Forest Service, Southern Research Station, 1000 Front Street, Oxford, Mississippi 38655 USA 2Present address: The Pacific Rivers Council, PMB 219, 1 Second Avenue E., Suite C, Polson, Montana 59860 USA Adams, Susan B., and Christopher A. Frissell. 2001. Thermal habitat use and evidence of seasonal migration by Rocky Mountain Tailed Frogs, Ascaphus montanus, in Montana. Canadian Field-Naturalist 115(2): 251-256. All life stages of Rocky Mountain Tailed Frogs (Ascaphus montanus) occurred in a reach of Moore Creek, Montana, where water temperatures exceeded those previously reported for Ascaphus in the wild. However, relative density of Ascaphus in the warmest reach, immediately downstream of a lake outlet, was lower than in cooler reaches downstream. Although we observed larvae and frogs in water temperatures up to 21°C, cold groundwater seeps contributed to a spatially complex thermal structure in the warmest stream reach. Frogs congregating near a cold seep and nesting in a groundwater-influ- enced site were likely using behavioral thermoregulation. At a stream weir in the warmest reach, we captured 32 Tailed Frogs moving downstream and none upstream, in September and October 1997. Because no migration was evident at five other weirs where summer water temperatures remained below 16°C, we propose that the frogs moving through upper Moore Creek migrated seasonally to avoid the high temperatures. The mature frogs may spend summers in the small, cold lake inlet streams, moving downstream in the fall to overwinter. Behavioral studies would be necessary to determine the extent to which individuals limit their overall thermal exposure in such spatially complex environments. Migration in response to local, seasonally changing habitat suitability could explain the diverse, and apparently contradictory, move- ment patterns (or lack thereof) among Ascaphus populations reported in the literature. Future studies of Ascaphus move- ments could benefit by accounting for seasonal changes in habitat suitability and by quantifying in-stream movements. Key Words: Ascaphus montanus, Tailed Frog, amphibian, water temperature, thermal complexity, habitat, movements, migration, behavior. Tailed Frogs (Ascaphus truei and A. montanus') live in cold, rocky streams in the Pacific Northwest and northern Rocky Mountains (Nussbaum et al. 1983) of the USA and in southwestern Canada. Although important to understanding the species’ habitat use, population dynamics, gene flow, and recolonization abilities, Ascaphus movements are not well documented. Mark-recapture studies directed at detecting movements of transformed Ascaphus have concluded that site fidelity is high among mature indi- viduals (Daugherty and Sheldon 1982a) or have been inconclusive (Metter 1964a). Three reports suggested that transformed frogs migrated seasonally; however, direct evidence of movement was lacking (Metter 1964a; Landreth and Ferguson 1967; Brown 1975). Such inconsistencies could reflect either shortcomings in the studies or spatial and temporal variations in movement patterns. Thermal tolerances and tolerance ranges of Ascaphus are lower than for any other anuran studied ‘Nielson et al. (2001) recommended that inland populations of Tailed Frogs be recognized as a distinct species (Asca- phus montanus). Minor inconsistencies between our text and the existence of two distinct Ascaphus species occur because our paper was already in press when we read Nielson et al. (2001). in North America (reviewed by Claussen 1973). Reports of Ascaphus occurrence are generally from streams with maximal temperatures not exceeding 16°C (Franz and Lee 1970; Welsh 1990). Laboratory experiments suggest that thermal tolerances vary among the life history stages. Critical examinations of whether temperature actually limits Ascaphus dis- tributions are lacking. In Rocky Mountain Ascaphus populations, individ- uals transform at age 4, first mature at age 8 and can live for 14 or more years (Daugherty and Sheldon 1982b). The frogs typically mate in the fall (but see Wernz 1969), and females retain sperm until the fol- lowing July when they lay eggs (Metter 1964b). Eggs usually hatch in late summer, but larvae apparently remain in the nest site until the following summer (Metter 1964a; Brown 1975). This implies that eggs and larvae experience the thermal regime at the nest site throughout an entire year. Incidental to a study of introduced Brook Trout (Salvelinus fontinalis) in two Montana streams (Adams 1999), we made new observations on Ascaphus sea- sonal movements and occurrence in warm water tem- peratures. We subsequently assessed Ascaphus relative abundances and water temperatures throughout one stream-lake network to determine how water tempera- ture was related to summer distributions of each life stage and to timing of downstream frog movements. pas) 232 THE CANADIAN FIELD-NATURALIST Volc iis Y "n, Canada _ “Y Montana Twelvemile \ Creek S.F. Little Joe Creek N inlet 1 inlet 2 4km inlet 3 te Moore Lake FIGURE 1. Locations of Moore and Twelvemile creeks in Mineral County, Montana, USA, and detail of Moore Creek showing locations of weirs (bars) and temperature recorders (stars). The lake weir was about 75 m downstream of the lake outlet and 5 m downstream of the upper temperature recorder. The lower weir was 120 m upstream of the confluence with South Fork Little Joe Creek. Materials and Methods consisted of two traps facing in opposite directions Moore Creek (St. Regis River drainage, Mineral County, Montana, 47°11’N, 115°15’W) is 3.2 km long and is fed by Moore Lake, a 16-m deep, 5.3 hectare, headwater lake at 1620 m elevation. About 320 m of shoreline separates the closest of three, small, lake-inlet streams from the outlet stream (Figure 1). During late summer low flows, average wetted stream widths ranged from 1.8 m near the lake outlet to 3.2 m near the mouth, channel slopes ranged from 10 to 19%, and average maximum pool depth near the lake outlet was 0.23 m. Riparian veg- etation was predominantly Western Redcedar (Thuja plicata) forest. Twelvemile Creek is larger, longer (22.6 km), and more moderately sloped than Moore Creek with wetted widths ranging from 1.8 to 9.6 m in summer and channel slopes from 1.3 to 8.1 %. We intermittently counted Tailed Frogs trapped at two weirs in Moore Creek (Figure 1) from 6 August to 9 October 1997 (dates shown in Figure 2) and at four weirs in Twelvemile Creek from July through late September 1997. We checked traps every 2 to 3 days during operation, and the longest period with- out counting frogs was 12 to 28 September, 1997. The weirs, constructed of 6.35 mm hardware cloth, and connected to each other and to shore by a fence (see Figure 1b in Gowan and Fausch 1996). An apron buried in the substrate prevented animals from easily passing under the weir. Each trap box was a 60 xX 60 X 60cm cube with a funnel extending almost to the back. Large rocks in the traps provided shelter and velocity refugia for captured animals, and lids minimized predation. After identification, ani- mals were released beyond the weir in the direction they were moving when trapped. We located Ascaphus via snorkeling and elec- trofishing targeting Brook Trout in Moore and Twelvemile creeks in 1997 and via visual survevs targeting Ascaphus in Moore Creek in 1998. In July 1998, students assisted with day and night searches for Ascaphus along sections of the lake inlet streams, the lakeshore between Moore Creek and the inlet, Moore Creek, and an unnamed creek near Moore Creek (Figure 1). We performed timed searches, turning over streambed rocks and visually scanning both the streambed and stream banks within 2 meters of the stream. The results are intended only for describing Ascaphus distribution and for a qualita- tive comparison of relative densities among reaches. 2001 The lakeshore, one inlet stream, and the uppermost Moore Creek reach were again searched during the night (and Moore Creek also during day) of 7-8 October 1998. Hobo-Temp® data loggers recorded stream tem- peratures from 31 July 1997 to 8 October 1998 (with several gaps) at three locations each in Moore (Figure 1) and Twelvemile creeks. Using a digital thermometer, we took a longitudinal temperature profile in Moore Creek downstream of the lake out- let on 28 July 1998 and took focal point tempera- tures at some Ascaphus locations. Results and Discussion Adult movements At the lake weir in Moore Creek, we trapped 32 frogs moving downstream and one moving up- stream. As the trapping period progressed, the num- ber of adult Ascaphus moving downstream into the trap increased from no frogs for the period of 7 August through 3 September 1997 to nine frogs on 6 October, three days before the weir was removed (Figure 2). The sex ratio of captured frogs was male biased (chi square = 6.1250, p = 0.0133). The timing of frog captures coincided with a drop in water temperature (Figure 2). The frogs began moving downstream into the trap when average daily stream temperatures dropped below 16°C, and most 24 NO NO (=) NO — Oo Water temperature (°C) eer =x NO ADAMS AND FRISSELL: HABITAT USE AND MIGRATION BY TAILED FROGS tt 253 were captured when average temperatures fell below 14°C (Figure 2). Whenever frogs were recorded in the trap, the maximum daily temperature had not exceeded 16.5°C during at least one day of the two- to-three-day trapping interval. The pattern of frog captures suggests that a direct- ed, seasonal migration was occurring in upper Moore Creek. The seasonality of the movement is clear from the complete absence of any frogs in the down- stream trap for at least the first month of operation. Although there was unquestionable directionality of capture in the traps, concluding that a downstream migration was actually occurring depends on two assumptions: (1) that the weir was not biased against capture or retention of frogs moving upstream, and (2) that the frogs were not moving upstream over land. We caught no frogs at the lower weir in Moore Creek or at any of the weirs in Twelvemile Creek, although the frogs and larvae were present through- out both creeks (Figure 3) (Franz 1970; S. Adams, personal observation). Thus, Ascaphus movement patterns can vary not only among, but also within, streams. The limited literature on Ascaphus move- ments also indicates that movement patterns may vary considerably among drainages. Daugherty and Sheldon (1982a) found no evidence of seasonal or directed movements by mature Tailed Frogs in r\ a — er ar rr ae Seng esl ies et aha i 7/30 8/4 8/9 8/14 8/19 8/24 8/29 9/3 9/8 9/13 9/18 9/23 9/28 10/3 10/8 Day FIGURE 2. Average (avg), maximum (max), and minimum (min) daily stream temperatures 5 m upstream of the “lake weir” in Moore Creek, Montana, 1997 and 1998. Vertical bars indicate dates in 1997 when frogs were counted in traps at the “lake weir’. Grey bars represent the absence and black bars the presence of frogs moving downstream. Each bar represents 2 to 3 days of trapping. Numbers of frogs counted are indicated above bars. 254 ° S) S) & (ep) © Individuals per minute searching ro) ine) 0 us mis ee dj oP De as os a) en o-= 35 ®-= 5 = o> = o> oes i Sain 28 - 5 25% 2 5 2. 2. =] 3 THE CANADIAN FIELD-NATURALIST Vol. 115 2 0 ® me) (op) ox oO oO = oe a 2 : = =] se ao gO =e Le =: Cr ees Qo = + O= s | © Cc 5296 526 = 5“ 8.25 wep = = 6h Sion all I up. Location and date, 1998 FIGURE 3. Numbers of Ascaphus larvae and adults per minute spent searching by each team during night surveys. Numbers above bars are the total numbers of individuals observed during each survey. Inlet 3 was surveyed only in October. u.S. = upstream; d.s. = downstream. Butler Creek (Montana) and concluded that the frogs exhibited extreme philopatry; however, their annual summer sampling period may have ended too early to detect an autumn migration. By contrast, based on seasonal changes in densities of adult Ascaphus, Metter (1964a) hypothesized that adults moved from the Touchet River (Washington) into more shaded tributaries in the late summer. In the Palouse River (Idaho) and tributaries during the same time period, though, he found no seasonal differences in frog densities. Brown (1975) observed aggregations of up to 20 females in “small, shallow tributaries” of Razor Hone Creek, Washington, in late July and sug- gested that females may move into the warmer tribu- taries to lay eggs. However, evidence of actual upstream movements was lacking. Landreth and Ferguson (1967) suspected that frogs moved down- stream out of small, intermittent creeks to the Lostine River, Oregon, to mate, but again, had no direct evidence of migration. If Tailed Frogs migrate in response to local conditions, diverse movements among sites would be expected. Stream temperatures and Ascaphus distribution The average August stream temperature at the upper Moore Creek temperature recording location exceed- ed 16°C both summers (Table 1). Average daily tem- peratures exceeded 18°C for at least 5 days in 1997 and 20 days in 1998 (Figure 2), and we likely missed recording many of the warmest days both years. Water temperatures taken during the day and night of 28 July 1998 were 5 to 10°C cooler in the lake inlet streams than in the lake outlet, a pattern that persisted into at least October. Significant water cooling occurred downstream of the lake at both within-reach and whole-stream scales. Water tem- peratures declined with distance downstream of the lake over 200 m on one warm afternoon (Figure 4), and average August water temperatures at the middle and downstream temperature recording sites were about 7°C cooler than at the site near the lake (Table Ib) Ascaphus larvae and adults were found in all stream sections surveyed except lake inlet 1, where only an adult was found (Figure 3), but none were found along the lake shore. In upper Moore Creek, we observed numerous larvae in water temperatures of 19.5—20.0°C, and several were found in tempera- tures of 21°C (focal point temperatures). Several TABLE |. Stream-scale changes in water temperatures along Moore Creek, Montana, during August 1997 (and 1998 in parentheses). Temperature recording sites were about 0.7, 1.4, and 3.2 km downstream of the lake outlet. Maximum Location Average Minimum Maximum daily range upstream 20.9(22.8) 16.9(18.4) 14.6(15.2) 3.9 (4.1) middle 12.6 yal Tsk pS) downstream 12.5 (13.9) 10.0(10.7) 75(7.8) 2.6(2.5) 2001 o at R* = 0.56 & 23 & P <0.02 5 22 H ® 21 e Sx ¢ § ake weir = 200 0 50 100 150 Distance downstream of lake outlet (m) FIGURE 4. Reach-scale changes in water temperatures with- in 200 m of the Moore Lake outlet. Temperatures were taken on 28 July 1998 between 1235 and 1410 hours, beginning at the outlet and proceeding down- stream. adult frogs also occupied water that was between 19 and 21°C, although most frogs we found were at least partially out of the water. Maximum stream temperatures in upper Moore Creek were 4.6°C higher than the highest tempera- ture (18.2°C) we found reported where adult Ascaphus occurred in nature (Landreth and Ferguson 1967). Reported tolerances of transformed Ascaphus in experimental settings are inconsistent but suggest that the frogs may tolerate temperatures of 20°C (Landreth and Ferguson 1967) to 24°C (Claussen 1973). However, low frog densities in the warmest stream reach relative to other reaches and movement through the warmest reach in the fall suggest that high summer temperatures near the lake outlet creat- ed an unfavorable summer environment for the frogs. Ascaphus larvae have been reported in sites with higher water temperatures than have transformed frogs, but not as high as those in upper Moore Creek. Low densities of Ascaphus larvae occurred in defor- ested watersheds of Mount Saint Helens (Washing- ton) where average maximum and mean stream tem- peratures were 19.5 and 14.4°C, respectively, in late summer (Hawkins et al. 1988). Although maximum summer temperatures in the Washington streams were nearly as high as in upper Moore Creek in 1997, average temperatures were considerably lower. In laboratory experiments larvae survived for a day at 22°C (Metter 1966) and sometimes occu- pied such temperatures in a thermal gradient (de Vlaming and Bury 1970). We found a deeply buried nest of Ascaphus eggs near the lake weir in Moore Creek on 30 July 1997. On 28 July 1998, we discovered a second nest shal- lowly buried in a riffle about 30 m downstream of the weir. The majority of embryos appeared viable in both nests. Water temperature in the middle of the pool containing the first nest was 17.4°C at the time of egg collection; however, cold ground water seep- ADAMS AND FRISSELL: HABITAT USE AND MIGRATION BY TAILED FROGS 235 ing through a bedrock fissure in the substrate less than | m upstream of the nest probably moderated temperatures at the nest microsite. When we found the second nest, water temperature was 20.2°C at the nest site and 20.9°C at the opposite side of the pool. Ascaphus embryos developed normally in laboratory experiments at temperatures between about 5.0 and 18.5°C, but not at higher temperatures (Brown 1975). Thermal Complexity We documented Ascaphus response to the spatial- ly complex temperature patterns in upper Moore Creek at both the reach and microhabitat scales. The relative density of both larvae and adults was much higher downstream of the lake weir, where water was cooler, than upstream during July 1998 (Figure 3). However, the relationship between water temper- ature and Ascaphus densities may be confounded by the higher density of Brook Trout upstream than downstream of the weir (unpublished data). The largest aggregation of adult frogs (6 frogs) found was at a cold water pocket (about 9 cm?) in a mossy bank adjacent to a stream pool. Water temperature was 20—21°C in the pool but was 5.3°C in the pock- et. No frogs were found in or along the warmer areas of the pool. The timing and location of frog captures in weirs suggests that many frogs avoided high water temper- atures by migrating. No frogs from upstream of the lake weir were captured until temperatures declined to levels more typically associated with Tailed Frog use. Although two frogs were found in the warm waters upstream of the weir in the summer, higher frog densities occurred in the cold, lake inlet streams (Figure 3). At the other five weirs, stream tempera- tures remained below 16°C throughout summer 1997, and we found no evidence of migration. We hypothesize that many of the frogs captured in the lake weir spend the summer in the cold, lake inlet streams (and perhaps in spring seeps around the lake shore), thereby avoiding localized, high summer temperatures in the outlet stream. When lake surface, outlet stream, and air temperatures cool in the fall, the frogs migrate through or around the lake and down the outlet creek to mate and overwinter. Because of their small size, the inlet streams may not provide sufficient overwinter or nesting habitats that are evidently available in the outlet stream. The Ascaphus that remain in the warmest reach during summer may persist due to the availability of cool refugia. Further behavioral studies would be necessary to determine if individuals rely on cool microsites to limit their overall thermal exposure in warm stream reaches; presently, we cannot conclude that occur- rence of individuals in warm water implies long- term tolerance of such temperatures. Behavioral responses to thermal heterogeneity and other vari- 256 able habitat features may help explain the diversity of movement patterns reported in the literature. Future studies of Ascaphus movements could benefit by accounting for seasonal changes in habitat suit- ability and by quantifying in-stream movements rather than focusing primarily on terrestrial move- ments. Acknowledgments S. Adams received partial funding from the USDA Forest Service, Rocky Mountain Research Station. W. Lowe, A. Sheldon, B. Rieman, and anonymous reviewers provided helpful suggestions on early versions of the manuscript. K. Keegan, A. Stephens, L. Rosenthal, and L. Steinbach assisted with weir operations. C.A.F.’s 1998 Ecology and Conservation of Aquatic Vertebrates class (U of Montana, Flathead Lake Biological Station) collect- ed distribution data: S. Brazier, A. Chilton, M. Cucchiara, H. Dean, M. Drum, R. Elliott, T. Mesa, M. Sireniawski, T. Van Roosmalen, and assistant T. Bansak. Literature Cited Adams, S. B. 1999. Mechanisms limiting a vertebrate invasion: brook trout in mountain streams of the north- western USA. Doctoral dissertation. The University of Montana. Missoula. Brown, H. A. 1975. Temperature and development of the tailed frog, Ascaphus truei. Comparative Biochemistry and Physiology 50A: 397-405. Claussen, D. L. 1973. The thermal relations of the tailed frog, Ascaphus truei, and the Pacific treefrog, Hyla regilla. Comparative Biochemistry and Physiology 44A: 137-153. Daugherty, C.H., and A. L. Sheldon. 1982a. Age-specific movement patterns of the frog Ascaphus truei. Herpeto- logica 38: 468-474. Daugherty, C. H., and A. L. Sheldon. 1982b. Age-deter- mination, growth, and life history of a Montana popula- tion of the tailed frog (Ascaphus truei). Herpetologica 38: 461-468. de Vlaming, V. L., and R. B. Bury. 1970. Thermal selec- tion in tadpoles of the tailed-frog, Ascaphus truei. Journal of Herpetology 4: 179-189. Franz, R. 1970. Egg development of the tailed frog under THE CANADIAN FIELD-NATURALIST Vol. 115 natural conditions. Bulletin Maryland Herpetological Society 6: 27—30. Franz, R., and D.S. Lee. 1970. The ecological and bio- geographical distribution of the tailed frog, Ascaphus truei, in the Flathead River drainage of northwestern Montana. Bulletin Maryland Herpetological Society 6: 62-73. Gowan, C., and K. D. Fausch. 1996. Mobile brook trout in two high-elevation Colorado streams: re-evaluating the concept of restricted movement. Canadian Journal of Fisheries and Aquatic Sciences 53: 1370-1381. Hawkins, C. P., L. J. Gottschalk, and S.S. Brown. 1988. Densities and habitat of tailed frog tadpoles in small streams near Mt. St. Helens following the 1980 eruption. Journal of the North American Benthological Society 7: 246-252. Landreth, H. F., and D. E. Ferguson. 1967. Movements and orientation of the tailed frog, Ascaphus truei. Herpe- tologica 23: 81-93. Metter, D. E. 1964a. A morphological and ecological comparison of two populations of the tailed frog, Asca- phus truei Stejneger. Copeia 1964: 181-195. Metter, D. E. 1964b. On breeding and sperm retention in Ascaphus. Copeia 1964: 710-711. Metter, D. E. 1966. Some temperature and salinity toler- ances of Ascaphus truei Stejneger. Journal of the Idaho Academy of Sciences 4: 44-47. Metter, D. E., and R. J. Pauken. 1969. An analysis of the reduction of gene flow in Ascaphus truei in the North- west U. S. since the Pleistocene. Copeia 1969: 301-307. Nielson, M., K. Lochman, and J. Sullivan. 2001. Phylo- geography of the tailed frog (Ascaphus truei): implica- tions for the biogeography of the Pacific Northwest. Evolution 55: 147-160. Noble, G. K., and P.G. Putnam. 1931. Observations on the life history of Ascaphus truei Stejneger. Copeia 1931: 97-101. Nussbaum, R. A., E. D. Brodie, Jr., and R. M. Storm. 1983. Amphibians and Reptiles of the Pacific Northwest. University of Idaho Press, Moscow, Idaho. Welsh, H. H. J. 1990. Relictual amphibians and old- growth forests. Conservation Biology 4: 309-319. Wernz, J.G. 1969. Spring mating of Ascaphus. Journal of Herpetology 3: 167-169. Received 1 August 2000 Accepted 27 April 2001 Updated 15 October 2001 Prey and Reproduction in a Metapopulation Decline Among Swainson’s Hawks, Buteo swainsoni JosEF K. ScumMutz,! C. STUART HOUSTON,” and SAMUEL J. BARRY? 'Centre for Studies in Agriculture, Law and the Environment (CSALE), University of Saskatchewan, 51 Campus Drive, Saskatoon, Saskatchewan S7N 5A8 Canada; e-mail: schmutzj@duke.usask.ca *863 University Drive, Saskatoon, Saskatchewan S7N 0J8 Canada; e-mail: houstons @duke.usask.ca 3Canadian Wildlife Service, Environment Canada, 4999-98 Avenue, Edmonton, Alberta TSB 2X3 Canada; e-mail: sam.barry @ec.gc.ca Schmutz, Josef K., C. Stuart Houston, and Samuel J. Barry. 2001. Prey and reproduction in a metapopulation decline among Swainson’s Hawks, Buteo swainsoni. Canadian Field-Naturalist 115(2): 257-273. Populations and productivity of Swainson’s Hawks (Buteo swainsoni) have declined for a decade at two study areas (Hanna, Alberta, and Kindersley — Alsask, Saskatchewan) near the northern edge of their Great Plains breeding range. Near Hanna, reproduction in successful nests appeared stable through 1990 and then declined by 31%, followed by a hawk population drop in 1995 and 1996. In Saskatchewan, reproduction began to decline in 1988 and reached unprecedented low levels in 1993 and 1996, followed by a noticeable drop in hawk populations beginning in 1994. In both areas, Richardson’s Ground Squirrels (Spermophilus richardsonii), the hawks’ main prey, have declined drastically in numbers, forcing the hawks to prey more often on ducklings, songbirds and corvids. Forty Swainson’s Hawks, marked as nestlings, bred on or near the Hanna study area, including five which moved from Saskatchewan. Recruitment of known-age breeders at Hanna was low at 4.4% of the breeding population, perhaps because the open prairie, which lacks geographic barriers, facilitates wide dispersal. The observed declines of this hawk and other species of grassland birds are of interest because they coin- cided with pervasive changes in the prairie ecosystem, especially an increasing use of biocides and fertilizers. Key Words: Swainson’s Hawk, Buteo swainsoni, Richardson’s Ground Squirrel, Spermophilus richardsonii, mixed-grass prairie, plant succession, population dynamics, predation, Saskatchewan, Alberta. Breeding bird surveys in western Canada have shown statistically and biologically significant declines in grassland bird species, including Sprague’s Pipit (Anthus spragueii), Loggerhead Shrike (Lanius ludovicianus), Northern Harrier (Circus cyaneus), Killdeer (Charadrius vociferus), Burrowing Owl (Athene cunicularia), Short-eared Owl (Asio flam- meus), and Western Meadowlark (Sturnella neglecta) (Houston and Schmutz 1999). The Burrowing Owl (Clayton and Schmutz 1999) and Greater Sage-grouse (Centrocercus urophasianus) (Connelly and Braun 1997) are in danger of following the Greater Prairie- Chicken (Tympanuchus cupido) on the path to extir- pation from the Canadian prairies. Recent unexplained declines in numbers of the Richardson’s Ground Squirrel (Spermophilus richardsonii), were quickly followed by a decline in productivity and then in numbers of the Swainson’s Hawk (Buteo swainsoni). We undertook an analysis of two large data sets of this quintessential grass- lands hawk and its main prey species, and examined the amount of grassland habitat, numbers of trees, and hawk dispersal. Data were accumulated in two different agricultural landscapes (ranching and cereal crop production) over 18 and 25 years, respectively. Breeding Swainson’s Hawks, but not Ferruginous Hawks (Buteo regalis), have adjusted well to as much as a 90% conversion of grasslands to cropland (Schmutz 1989). Once maintained by fire and by Bison (Bison bison) and Pronghorn (Antilocapra americana) graz- ing and browsing, and more recently by domestic grazers in an emerging ranching economy (Jameson 1986), much of the grassland in our study region has been lost to dry land crop production (Rowe 1987). The Canadian prairies have been called “one of the most altered habitats in the world,’ so much so that the sustainability of agricultural practices themselves has been questioned (Anderson et al. 1991). Study Areas Both study areas are in the dry mixed grass ecore- gion of the northern Great Plains (Figure 1; Strong and Leggat 1992). Native plant species in the gently undulating landscape are adapted to severe moisture deficits in mid- to late summer. Trees and shrubs used by the hawks for nesting include Trembling Aspen, Populus tremuloides, Chokecherry, Prunus virginiana, Saskatoon, Amelanchier alnifolia, Buffaloberry, Shepherdia argentea, Prickly Rose, Rosa acicularis, and willows, Salix spp., growing naturally since the control of prairie fires (Houston and Bechard 1983). The hawks also use introduced trees planted for shade and windbreaks (Manitoba Maple, Acer negundo, Green Ash, Fraxinus pennsyl- vanica, American Elm, Ulmus americana, Caragana, Caragana arborescens, and hybrid poplar, Populus Sp.). 257 258 THE CANADIAN FIELD-NATURALIST ; Vol. 115 O 41 km? Study Plots Native Vegetation Remaining Less than 10% 10% to 50% ALBERTA SASKATCHEWAN 40 km Ea More than 50% FIGURE 1. Percent of native vegetation in the two study areas. Boundaries of the completely searched Hanna study area are within a larger study perimeter. Of the 78-83 randomly selected study plots used, 30 lay within the region shown. The Saskatchewan area was searched opportunistically. Alberta: The town of Hanna (51° 38’N, 112° 33'W) is near the northwest corner of the study area, which encompassed 480 km? in all years except 1975 (southern 335-km/? portion) and 1983 (southern 326-km? portion). This ranching area, with approxi- mately 85% mixed grass remaining, was monitored intensively from 1975 through 1978 and from 1983 through 1996. During summers (May—August) in the mixed grass ecoregion of Alberta, the mean tempera- ture overall was 16.2°C (range 8.7—23.6°C). Average summer rainfall was 15.6 cm (8.8—23.3 = 25% & 75% percentiles; Strong and Leggat 1992). Here, at the southern edge of the southward-advancing aspen parkland, expansion of trees has been noticeable since the mid-1970s. Some Swainson’s Hawks nest- ed in P. tremuloides that grew during this study in previously treeless sites. Saskatchewan: Kindersley (51° 28'N, 109° 09'W) is the largest town in the study area, and Alsask (51° 21'N, 109° 59’ W), is at its western margin (Figure 1). This area was sampled mainly from 1972 through 1996. Near Kindersley, the land has been extensively altered for cereal crop production (< 10% native habitat), whereas Alsask was intermediate (10-50% native). On the large pastures near Kindersley, trees remaining in deserted farmstead shelterbelts contin- ue to die out. Mean July temperatures were 18.8°C and annual rainfall was 25.1 cm (Padbury and Acton 1994). | Methods In Alberta, a complete search for nests was carried out each year on the Hanna study area, and thrice on randomly selected study plots throughout southeastern Alberta (Table 1). Near Kindersley, Saskatchewan, Swainson’s Hawk nest finding was a by-product of Ferruginous Hawk banding in late June. Southeastern Alberta hawk population estimate In 1982, a count of nesting buteos for population monitoring was done on 80 randomly selected 41- km? study plots, a portion of which are depicted in Figure 1. The same protocol was followed in 1987, but in 1992 five new plots were added. The total number of plots searched was 80 in 1982 (Schmutz 1984), 78 in 1987 (Schmutz 1989), and 83 in 1992. Alberta (Hanna) Study Except in 1978 and 1983, JKS was present on the study area from late May or early June through early August. Each year, during 10 to 15 days in June, JKS searched by motorcycle all clumps of trees and shrubs, artificial nests, and eroded slopes. A 45-km? northern portion of the study area was searched in early July. 2001 TABLE |. Types of studies carried out on Swainson’s Hawks and Richardson’s Ground Squirrels on the study areas in Alberta and Saskatchewan, and their periods. 1h ST 1S 10 ie a, de | hee tee 89 90° ° OF 92 88 19 87 19 86 19 85 19 19 19 1o: = 19 9 19 [De OR SeHlO Oe TS = 19-80". Si 82 83 84 TS 1967-19 93 94.595" 06 74 Alberta (Hanna) Ground squirrel: SCHMUTZ, HOUSTON, AND BARRY: DECLINE AMONG SWAINSON’S HAWKS 259 mM P< P< ~< ~< xX ~< x< x ~< mM P< ~< ~< x XK ~*< ~< ~< X< x ~< Pm P< PX PM ~< <<< xX ~< rm P< ~< xX mK XK OX ~< rm PS PS »~< mM PS PS PM ~< mx PS PX »~< x xX ~< < ~< ~< ~< x< x< ~< PS PA PS PS PS m~< mS PS PS OX PS ~< ~< a 2 —_— DN SARS Pm oan, oe = Ye ue SESSOABE Foz a. Tse BSEE PSR Seo. Sis 6 eur OS “= © aaa] BS) Strychnine impact Alberta Pop. estimate Saskatchewan Banding: CWS bands colour 'By Richard W. Fyfe and colleagues From 1975 through 1977, we recorded nest con- tents in late May by climbing to the nest or using a mirror on a pole. Starting in 1983, all nests of Swainson’s Hawks on the study area were approached during the June search, but only until the occupant could be identified to species. In most cases nests were not closely inspected until the nestling period (July) to minimize chances of nest desertion (Houston 1974). For this reason, clutch size was often not known. To aid interpretation of a failed reproductive attempt, nests were carefully inspected. For a repro- ductive attempt to be counted, the nest base had to be complete, the cup lined, and the lining flattened as by a hawk’s body during laying or incubation. Often nests had been added to and sometimes even completed but not flattened. The presence of down feathers in the nest material was not a reliable indi- cator of laying or incubation. Nests in which young had died during the second half of the nestling peri- od generally showed a trampled nest rim, droppings, and a nest cup filled with prey remains, pellets, and sprigs of herbaceous plants. Nests in which young hatched but died soon thereafter showed the original deep cup. Often tiny egg fragments from pipping eggs remained lightly buried in the base of the nest cup. Eggs abandoned before hatching were opened and the approximate stage of the arrested develop- ment recorded. For brood counts and banding, nests were visited late in the nestling period in both study areas. We did not adjust the data based on nestling age and sur- vival probabilities (e.g., Mayfield 1975). Instead, a few nests were excluded when brood size was manipulated (e.g., Schmutz et al. 1980), or when nestlings in unusually late nests were less than half- grown (< 2% of nests). Near Hanna, searching the study area for nests and banding young was a priority in all but two years (Table 1). Aluminum bands were placed on 366 adults and 1968 young, and alpha-numeric colour bands were applied to 236 adults and 1240 young. Additional avenues of investigation varied (Table 1). We studied competition among buteo species in three summers, 1975 through 1977 (Schmutz et al. 1980), in an area studied by Richard W. Fyfe and co- workers between 1967 and 1974 (Schmutz et al. 1991). After a four-year hiatus, we resumed work on the Hanna study area from 1983 through 1996. In 1984, 1985, and 1986, we live-captured 196 adults (Bloom et al. 1992) for a survival and site- fidelity study; another 170 adults were captured dur- ing other years between 1975 and 1996. Reading colour bands, achieved on 510 occasions, was there- after a priority. The number of prey records, after young had hatched, varied between years, depending on the nature of other studies underway. In recent years, the 260 small prey sample in both study areas was primarily influenced by the decline in the number of nests, and presumably because the hawks were so food-stressed that their food was consumed quickly. In addition to species, sex and age, the portion of prey still avail- able for consumption was recorded in the field and converted to weight using prey weights — either taken from the literature or locally recorded. Ground squirrel weight varied by age and sex throughout the season (Schmutz 1977). For whole prey the consum- able portion was assumed to be 85%, and propor- tionately less for incomplete remains. At Hanna, we have three indices of ground squir- rel abundance, in addition to the number of ground squirrels found in nests. One was a mark-recapture ground squirrel study done to document dispersal and mortality in 1975 and 1976. The second was burrow counts. The third was a 25-year record of strychnine sold by the “Special Areas Board” (Gorman 1988). Strychnine was sold at a consistent low cost as a service to the farming and ranching community, hence cost did not affect the volume used. The poison data were independent of the ques- tions addressed here, representative of a large munic- ipality encompassing 76 686 km? and involving ove 100 ranchers. ; Strychnine was purchased in liquid form, Gopher Cop®, and used to coat oat bait. A person then dropped a spoonful of bait down each burrow entrance while walking or driving an all-terrain vehi- cle around field edges including the grassy borders, around garden plots, in corrals and through “calving pastures.” More poison was purchased in years of high ground squirrel density. In view of ground squirrel ecology and local poi- soning practices, demand for poison varied for two reasons. When survival during hibernation was high, the density of breeding females was high; males suf- fered a disproportionately high mortality and repre- sented only 19% of the adult population (Schmutz et al. 1979). Also, demand for poison could be high when the number of young produced was high. Beginning in early to mid-June, young disperse and THE CANADIAN FIELD-NATURALIST Volts dig new burrows or re-use formerly vacant burrows coinciding with crop in early stages of growth. We view the poison data as an index that is sensitive to — large and widespread changes in the number of ground squirrels (Schmutz and Hungile 1989). Saskatchewan Study CSH banded Swainson’s Hawks intermittently from 1944 through 1972. From 1973 on, he concen- trated most of his banding on and near large pastures near Kindersley, visiting twice each summer to apply leg bands and to monitor productivity. Most incubat- ing Swainson’s Hawk pairs were located during Ferruginous Hawk banding in June; all such nests were revisited for banding in mid- to late July. Additional Swainson’s Hawk nests were found en passant along intervening roads and trails, and by visiting occupied sites from preceding years. Prey remains were recorded during a single banding visit in late July. This study area was not completely searched and had no well-defined boundaries. The focus was on number of nests visited, not density per unit area. From 1988 through 1995, 684 double-rivet anodized alpha-numeric bands, manufactured by A- Craft Sign and Nameplate Co., 10342 59 Avenue, Edmonton, Alberta T6H 1E6, were applied on each nestling’s leg opposite that carrying the usual alu- minum band. CSH also studied, but less intensively and less regularly, Swainson’s Hawk nesting in parkland areas near Saskatoon. Results Southeastern Alberta hawk population survey Hawk distribution on the randomly selected 41- km? study plots was influenced by cultivation and tree availability (Schmutz 1989). Of 780 Swainson’s Hawk nests recorded on or off plots in the three sur- vey years, site descriptions were transcribed from field notes for 377 nests. For these, planted shelter- belts supported nearly as many nests (139) as single trees or clumps of aspens (145) and more than native shrubs (79; Table 2). TABLE 2. Nest substrates (n = 378) used by Swainson’s Hawks in the mixed, moist mixed and fescue grassland ecoregions in southeastern Alberta, in 1982, 1987 and 1992. Nests found in the perimeter of randomly selected study plots are included. “Natural” Cotton- Grassland Aspen wood Shrubs! Mixed 82 2 31 Moist mixed 39 6 30 Fescue 24 ) 18 Total 145 8 719 Cultural Shelter- Artificial Building/ Ground belt nest windmill 1 55) 2 1 0 37 2 0 0 49 0 [ 1 139 4 2 !Willow (Salix spp.), Chokecherry (Prunus virginia), Saskatoon (Amelanchier alnifolia), Buffaloberry (Shepherdia argen- tial), and Prickly Rose (Rosa asicularis). 2001 SCHMUTZ, HOUSTON, AND BARRY: DECLINE AMONG SWAINSON’S HAWKS 261 TABLE 3. Data from the Swainson’s Hawk population studied near Hanna, Alberta, 1975-1996. When the size of the study area was reduced, the northern portion was not searched. Nests Nests Abandoned Clutch Nests Brood Of Pioaeincom- Com- /100 Outcome Nest before after stage —— ss with «§ ——________ area (km?) plete pleted km’ known fell hatch hatch unkn. # Nests young # £Nests SD _ nests 1975 335 3 3811.34 34 2 a Z oy 250 24 2 aay 28° OGB | 1976 480 4 77~—«:16.04 72 3 8 3 2 2.45 29 36 © 194. der ORF Gg 1977 480 5 70 14.58 49 3 4 l Bo 200 16 18:7 200.5337; WOte. 19 1978 ~100 1.87 1S 0.74 1983 326 2 43 13.19 37 0 0 2 4 2.00 3 a? OF 40) (OST 2 1984 480 7 Tl: 14:79 70 3 + 7 8 =—-1.88 8 AS. 1.88. 348 “O76, 35 1985 480 8 84 17.50 81 2 5 7 bh. 2060) 5 96° .°2:06° 156) O68) 94 1986 480 11 104 21.67 102 11 6 8 5 2.46 13 12) e257~ VIB Bea 13 1987 480 8 Bret 2021 94 4 8 5 1 2.68 31 16. 2226.5, 470, “ao 22 1988 480 + 103 21.46 98 8 7 4 10 2.78 37 69 196 69 0.86 50 1989 480 21 64 13.33 63 2 3 + A Des 4 a0. 92" 2300s. 11 1990 480 9 84 = 17.50 81 8 7 11 Ay OTS 4 St 206 e835) O84" 13 199] 480 16 60 12.50 58 l 5 10 3) MSS 3 39) LBS 39° <0:62" 8 1992 480 8 69 14.38 65 3 8 5) 3) RS 4 AS | SLSO* 4G) O69 A 1993 480 15 ao 15521 ip? 7 8 23 Gi =, 231 13 28 28 0:31 8 1994 480 8 63).13.13 63 + 2: 11 10 2.60 3 36% 1.69) 3G) 09: ory 1995 480 18 33 «6.88 33 0 2 14 6 2,00 2 ih "b25 8 0.44 16 1996 480 16 a2 — 6.67 32 2 6 4 200 8 15 1.40 15 O51 7 1975-1985 29 383 -14.84 343 13. (24 22 DIE 92 85" 200. 12.03. Oo 078) BY 1986-1988 23 304 21.11 294 23; i 17 iG 2.67 Bt 217 2 AS, O84 a7 1989-1994 if} 413 14.34 402 25. 33 64 2 dp? 33. 2508) U6b 250) 75.531 1995-1996 34 65 6.78 65 v3} 8 18 le 592200) 10 260) 135 23 0.48 23 1975-1996 Hos), 1165. 14.82. 1104 6380 12] 85 241 209 749 1.91 745 0.80 401 Nest Densities ated until 1990, and then declined. The ground squir- Alberta (Hanna): Over 18 years of study by JKS spanning 22 years, nesting areas totalled 1730. Of these, 1165 nests were completed within the study area, and 163 nests were not completed (Table 3). Another 401 nests (1-94 per year) were monitored adjacent to the formal study area. As was found in the regional survey (Schmutz 1989), Swainson’s Hawks near Hanna also favoured cultivated areas. In the peak year, 1986, densities on the grassland block were 14.4 nests/100 km’, com- pared to 25.3 nests/100 km? in the surrounding mix of rangeland and cultivation. In the lowest density year, this trend persisted with 5.0 nests/100 km? in grassland and 7.5 nests/100 km? in mixed habitat. There were four distinct periods in the hawks’ pop- ulation dynamics; two periods of “moderate” numbers (1975-1985 and 1989-1994), interrupted by a period of “high” numbers (1986-1988), followed by a period of “low” numbers (1995-1996) (Figure 2). Litres of strychnine used varied in parallel with hawk densities during the peak prey years and before (Figure 2, Spearman rho = 0.631, P = 0.029). Overall, however, and without strychnine data for 1993-1996, _hawk density and strychnine were not significantly correlated (Spearman rho = 0.445, P = 0.107). The strychnine data suggested that ground squir- rels existed on the Hanna study area in moderate numbers before 1986, peaked in 1986-1988, fluctu- rel low continued beyond 1992, when the poison sales program ended. Saskatchewan: CSH recorded 1561 nests with young from 1973 on. As a result of increased search- ing effort, the number of nests increased up to 1987 (Figure 3). Although searching effort remained high and relatively constant in later years, the number of nests found declined in the 1990s. In the most con- sistently and intensively studied township, north of Kindersley, Jean Harris found 15.4 pairs/100 km? in 1990 and 1991; this dropped to 6.4/100 km? in both 1994 and 1996. To approximate the timing of a shift from an increase in number of successful nests to a decrease, we calculated a Spearman rank correlation coefficient for the increasing portion of the plot (Figure 3), sepa- rately from the decreasing portion, allowing a three- year overlap. We shifted these periods until the high- est combined absolute value of rho was obtained. We thus obtained a Spearman rho of 0.875, P < 0.001, for 1968-1988, and -0.692, P = 0.029, for 1986-1996. This suggests that the nest decline began in 1988. Dispersal, non-breeding floaters and recruitment Alberta (Hanna): We distinguished between sight- ings at nests and sightings > 300 m from nests. Of 1483 Swainson’s Hawks that were not individually marked and were more than 300 m from the nearest 262 —@®— Nests oe ——— Strychnine 20 2 § 20 (e) © g G = wo 15 Vj ; 4 (e) D Wy j 38 a GUA < za SG = © | £ 9 = £ > S 10 5 10 & ‘S v =) wad 5 5 sr / 1975 1985 1990 1995 FIGURE 2. Number of nesting pairs of Swainson’s Hawks for the Hanna, Alberta study area. Cross-hatched regions span two medium, one high and one low density period, and show 95% confidence intervals. Litres of strychnine poison used to poison ground squirrels are also plotted. nest, 140 were in close proximity to another and pre- sumed paired (70 pairs). One such pair is noteworthy because it shows that at least some satellite non- breeders pre-select nest sites for use in subsequent seasons. The female was radio-marked on 24 August 1995 (Schmutz et al. 1996). The female and the same male, judging from his plumage, returned to re- use this nest and lay eggs the following year. An additional 39 hawks existed in six flocks of 4, 4,5, 5, 10 and 11 individuals, with a mix of adults and subadults. These “satellite” hawks were also pre- sumed to be non-breeding, as few of the marked breeders were ever seen roosting as pairs more than 300 m from their nests (JKS, unpublished data) and communal hunting is not known in this species. All six flocks and 51 (73%) of the 70 non-breeding pairs were recorded during the high ground squirrel years of 1986-1988 (JKS, unpublished data). In July and August 1995, a seventh flock of 33 Swainson’s Hawks was noteworthy. These birds, in adult and sub-adult plumage (Clark and Wheeler 1987), were repeatedly seen feeding in cultivated but fallow fields. Their walking, stalking, and pecking behaviour indicated that they were feeding on insects. As noted in a local newspaper, local resi- dents had repeatedly seen large flocks of hawks that year. Thus, even in a year when conditions for repro- duction were poor, the area was sufficiently attrac- tive to hold non-breeders. THE CANADIAN FIELD-NATURALIST SYN Vol. Im sients. On the study area, we observed a total of six locally reared and individually marked hawks, | including five 1-yr-olds and one 5-yr-old, none of) them known to be nesting. Each was sighted only | once, even though we made a special effort to relo- | cate them on subsequent days. i Breeding hawks included both locally raised | hawks and immigrants. Of 39 adult Swainson’s Hawks of known origin that attended nests at Hanna | during 1977-1996, 35 had been locally marked as nestlings and four had immigrated 135 to 310 km | from near Kindersley, Saskatchewan. The greatest | number of recruits (n= 8) was sighted in 1986, when | these comprised 4.4% of 182 adults of known band- | ing status at nests on or off the Hanna study area. One five-yr-old local recruit attended an unfinished | nest in 1989, and one nine-yr-old was seen in copu- | lation but not recorded thereafter in 1996. The aver- age age of 33 breeders when first encountered was 4.5 yr (range 2-9 yr), from O to 45 km from their natal site. | | Many of the non-breeders appeared to be tran- | | } \ Saskatchewan: From 684 nestling Swainson’s | Hawks colour-marked in Saskatchewan, there were four sightings by a waterfowl] biologist within a | month after banding, all within 1 km of the nest. Although in each subsequent year up to 102 adults © were seen adequately to detect a colour band on a | | tibio-tarsus, there were no later local sightings of — 2001 2.7 _ U e e 0 O) OO e O) 2.2 “LF O @ O SP Oo N Y To o = ee = co 1972 1977 SCHMUTZ, HOUSTON, AND BARRY: DECLINE AMONG SWAINSON’S HAWKS 1982 263 120 — oO oO ye 8 a! 60 5 > OD e c d = . 7 40 S /2) J nD @ 20 = 0 1987 1992 1996 FIGURE 3. Yearly mean brood size and number of nests containing young Swainson’s hawks in mid and late July, in Saskatchewan. colour bands. However, one, 4-T, was read twice by telescope by JKS on the Hanna study area in Alberta, 135 km to the west, at three and seven years after fledging, and another colour-banded hawk was cap- tured and released by a research biologist on 4 June, when almost nine years old, at Caronport, Saskatchewan, 295 km ESE of where banded. Prey Items in Nests Swainson’s Hawks relied extensively on ground squirrel prey, 60% of items in Alberta and 38% in Saskatchewan (Appendix 1). Beyond ground squir- rels, these hawks exhibited a generalist strategy, using prey of at least 52 species in four taxonomic classes. Studies of prey items alone, without pellet analysis in the later years, and without time lapse photography, tend to underestimate small prey (e.g., Simmons et al. 1991). However, nest prey data can be useful to compare trends over time. The proportion of ground squirrels among prey declined in step-like fashion in both areas, beginning in 1983 in Saskatchewan, and in 1991 near Hanna. Alberta (Hanna): Before 1991, ground squirrels constituted more prey items than all other prey species together (Figure 4). Later, the proportion of ground squirrel prey declined (G = 61.95, df = 14, P <0.001). As ground squirrels declined, the smaller size of most substitute prey species resulted in a con- _ siderable reduction in prey biomass. Relative prey _ sizes are 73 to 443 g for all ground squirrel age and _ sex classes (Schmutz 1977), compared to mice and _ voles (17 to 26 g) and songbirds and corvids (20 to _ 450 g) or primarily juvenile waterfowl prey (39 g for juveniles, 565 g for adults taken infrequently). The correlation between ground squirrels present on vis- its to nests and the estimated grams of prey biomass still available for consumption was high because of the large size of ground squirrels relative to other prey (Spearman rho = 0.738, P = 0.003). When con- verted to biomass, 82% of nest prey and 69% of prey and pellets were ground squirrels, confirming the bulk importance of this prey (Schmutz et al. 1980). Saskatchewan: Through 1982, prey items were pre- sent in 35% of nests when visited; 50% of the prey items were Richardson’s Ground Squirrels and 1% Thirteen-lined Ground Squirrels. The proportion of ground squirrels found was significantly lower after 1982 (G = 28.6, P = < 0.0001, Figure 5). From 1983 through 1996, prey items were found in only 22% of nests, and only 30% of the prey were ground squirrels. With the obligatory switch in diet, ducklings, Horned Larks (Eremophila alpestris), and Western Meadowlarks (Sturnella neglecta), became important prey items. Some pairs, faced with insufficient food to raise more than one nestling, were feeding the youngest to the oldest (“cannibalism” or siblicide) at the time of the banding visit (Appendix 1). Additional prey included 181 other mammals of 11 species, 211 birds of 38 species, 3 amphibians and 2 reptiles. Hawk reproductive success Alberta (Hanna): There was a remarkably close agreement between amount of prey in nests (Figure 4) and brood size in Alberta (Figure 6), presumably representing cause-and-effect. 264 THE CANADIAN FIELD-NATURALIST Vol. 115 0.6 75 —@— Grams —QO— Richardson's ground squirrels aed —T}— Voles = : —<— Birds e 55 » 04 fa 2 DB > 45> 5 5 & 03 Q. WY = 35 o ro) a. ed ana pee Ree © Ww fone oO. = 0.1 Z 15 0 0.0 ol ere 1975 1985 1990 1995 FIGURE 4. Prey items in nests of Swainson’s Hawks on the Hanna study area. Shaded regions span the “moder- ate” and low ground squirrel years (see text), and represent 95% confidence intervals around the respec- tive means. Brood size departed little from a 2.06-young aver- age during the first 12 years, 1975-1978 and 1983-1990 (Figure 6). During the following six years, average brood size declined by 31% to only 1.42. The number of nests with 1 (n = 158), 2 N= 215) and 3-4 young (n = 172) was significantly dif- ferent during 1975-1990, compared to 1991-1996 (OI 32, 10) sespectively. G = 88:94 di = 25 0.001). The proportion of nests with young raised to near fledging was 78% from 1975-1990, and 58% from 1991-1996 (G = 42.3, P < 0.001). Nests that fell from trees (Table 3) were excluded from this analy- sis; this failure category was considered an event unrelated to food. The gap between brood size and a sampling of clutch size widened over time (Figure 6), suggesting that the decline in reproduction was due to a combination of smaller clutches and reduced embryo/nestling survival. The clutch/brood gap in the periods of medium, high, medium, and low den- sities (see Figure 2), was 0.29, 0.54, 0.46 and 0.65, respectively. Clutch size, too, was reduced in the low prey years 1991-1996, but brood size was reduced even more severely in those years (Figure 6). Expressing a total 209 clutches as either above or below the grand mean of 2.40, we found that 89 clutches were above and 85 below for 1975-1990, and 10 above and 26 below for 1991-1996 (G = 6.77, P = 0.009). Saskatchewan: Brood size was > 1.8 (except 1984) through 1987. During four years, 1972, 1976, 1978, and 1979, nests with three young were more common than those with two young. The percentage of nests with three young was 28% through 1987 and then dropped to 11% from 1988 through 1996, while the percentage with one young increased from 29% to 52% in the same periods (Houston and Schmutz 1995b). Brood sizes declined fairly steadily (by 29%) from-2.25 young per successful nest in the 1970s to 1.60 young in the 1990s (Table 4; Spearman rho = -0.868, P < 0.001; Figure 3). The least successful year was 1993, when the minimum nest failure rate between mid-June and late July was an all-time high of 61%, and the brood size was an all-time low of 1.27 (1.00 for the only eight success- ful nests near Alsask). Of 45 successful nests in 1993, adult hawks failed to appear during the band- ing visit at seven; in response to a decline in ground squirrels, the hawks may have been forced to feed farther from the nest and so did not notice the ban- der’s visit. We separated subsets for Kindersley and Alsask to show the differences in the timing of declines in brood size. Near Kindersley there was an unex- plained increase in brood size from 1976 to 1979 (Figure 7). In 1987, brood size began to decline near Kindersley, and in 1993 near Alsask (Figure 7). [In comparison, at Hanna, Alberta, this decline was first noticed in 1991 (Figure 6)]. The parkland (aspen bluff) areas of Saskatchewan, where Richardson’s Ground Squirrel numbers were 2001 SCHMUTZ, HOUSTON, AND BARRY: DECLINE AMONG SWAINSON’S HAWKS 265 100 —O— Richardson's ground squirrel 90 —e— Other species ig 80 70 H /\ . 60 Fe ry » & = at “e. 20 Co) i fo) ras @ 40 ) ; a ) ou " 20 10 0 a 1970 1975 1980 1985 1990 1995 FicurE 5. The proportion of Richardson’s Ground Squirrels in relation to other vertebrate prey in Saskatchewan. The shaded region represents means (50% and 32%) and standard deviation (17 and 11) for two periods of moderate and low ground squirrel use, respectively. maintained, produced strikingly different results from the more open grassland areas mentioned above. Near Saskatoon, from 1964 through 1987, Swainson’s Hawk brood size averaged 2.3 young per successful nest (154 young banded in 67 nests). Brood size dropped slightly to 1.8 (96 young in 52 nests) from 1988 to 1992, then returned to 2.2 (29 young in 13 nests) from 1993 to 1996. Ground Squirrel abundance Alberta (Hanna): A direct measure of ground squirrel abundance using mark-recapture was avail- able only for two years. On six 4-ha plots and sur- rounding areas, 1497 ground squirrels were individu- ally marked. Densities on these selected well-drained study plots ranged from 3.3 to 7.0 squirrels per ha in 1975, and 3.8 to 7.5 in 1976 (Schmutz et al. 1979). The shift in the 1990s away from ground squirrel prey (Figure 4) was independently supported by a second index of ground squirrel abundance, i.e., -counts of used burrows on study plots (Schmutz et al. 1979). In June and July 1975, 23 1-km transects, on which used burrows were counted within | meter of a slowly driven motorcycle, yielded on average 11.5 burrows/km. On 25 July 1996, a repeat single 1-km transect on each of the six plots yielded no used burrows, and no ground squirrels; although diminished and locally variable in numbers, ground squirrels were still present sparingly elsewhere on the study area during the 1990s. Jones (1993) also documented the extreme paucity of Richardson’s Ground Squirrels near Brooks and south of Edmonton, Alberta, in 1993. In stark con- trast to previous years, he observed extremely few live ground squirrels and no road-kills that year, and found none in nests of Swainson’s Hawks that year. Saskatchewan: Prior to 1988, young Richardson’s Ground Squirrels were abundant, with up to 10 encountered per km of prairie trail; road kills were encountered regularly. In contrast, on six days in June and four days in July 1993, CSH drove > 300 km per day in prime Swainson’s Hawk habitat, with sight- ings of only 0-3 ground squirrels per day, and not one road kill. On each of those days, more Red Foxes and Coyotes were seen than ground squirrels, a 100- fold reversal of the previous ratio of ground squirrels to either of these species. Richardson’s Ground Squirrels increased slightly over the next two years, though still remaining < 10% of former numbers. In 266 3.0 1.5 —O— "Clutch" size —®— Brood size VF eee nny, 1975 1985 THE CANADIAN FIELD-NATURALIST Vol. 115 (yun @ = 8 28 1990 1995 FIGURE 6. Number of young at banding (Brood) and eggs in late hatching nests (“Clutch”) of Swainson’s Hawks for the Hanna study area. Numerals represent number of nests involved. Shaded regions represent the 95% confidence interval around the respective means. The mean brood size and standard error for peri- ods prior to 1990, after 1990 and for all years were 2.05 + 0.13, 1.38 + 0.21 and 1.8 + 0.36, respectively. 1995, travels of > 300 km each day by CSH recorded, sequentially, 3, 9, 1, 29, 11, and 10 ground squirrels in June, and 2, 10, 3, 1, 1, 14, 2, 0 and O ground squirrels in July. From 1992 through 1996, few Richardson’s Ground Squirrels were encountered as prey in hawk nests, none at all in 1995. Discussion Dispersal Our study areas included grassland and cropland interspersed with permanent or semi-permanent wet- lands in mixed prairie. Both the Hanna, Alberta, and Saskatchewan study areas lie about 200 km south of the northern edge of a large region of western North America occupied by Swainson’s Hawks. Some colour-marked breeders at Hanna were recruited locally, but others moved onto the Hanna area from localities outside it, including five from the Kindersley area of Saskatchewan (Houston and Schmutz 1995a). Natal dispersal of Swainson’s Hawks across the Canadian prairies, with no moun- tain ranges or large bodies of water to serve as a bar- rier, was apparently sufficient that no subdivisions could be detected using DNA fingerprints from Swainson’s Hawks in 85 nests from Manitoba to Alberta (Portman 1997). It appears, therefore, that our hawks were part of an interacting metapopula- tion semi-isolated only through the species’ charac- teristic natal and breeding site fidelity. We suspect that the Saskatchewan and Alberta (Hanna) study areas were a source and a sink, respectively (Pulliam 1988). The source population may be indicative of | emigration in a declining population. In sharp contrast, a mountain range is apparently a barrier to dispersal for the population studied in the Butte Valley of northern California. The Butte Valley breeding population increased incrementally | from 12 pairs in 1984 to 83 pairs in 1994 (a density of 20 territories per 100 km?) and exhibited a high proportion of local recruits, while floaters or sub- adults were rare (Woodbridge et al. 1995). Hawk population dynamics In the Alberta (Hanna) study, a 69% increase occurred between pre-1986 and the three years of high population, 1986-1988. During these three years | Swainson’s Hawks responded quickly to the excellent food availability, the number of breeders increased, and both banded and unbanded non-breeders were common. Judging from the hawks’ quick response, they were able to judge the suitability of an area with — little delay, exemplifying Fretwell and Lucas’s (1970) “ideal distribution.” This response may have been mediated through the presence of a surplus of poten- tial breeders that could saturate those nesting areas made newly available through an increase in food. The stabilizing influence of dispersal on population — rate of change has been recognized in several studies (Hunter 1995). From 1986 through 1988, then, there was a surplus of Swainson’s Hawks capable of breed- _ ing. It is unlikely that resident breeding hawks merely © redistributed themselves from immediately outside the | study area, because a regional (74 686 km?) popula- — 2001 SCHMUTZ, HOUSTON, AND BARRY: DECLINE AMONG SWAINSON’S HAWKS 267 TABLE 4. Numbers of Swainson’s Hawk nests and reproduction are shown for all SW Saskatchewan nests combined, and separately for a cereal and pulse crop production area near Kindersley and a mixed farming/ranching area near Alsask, east of the Alberta/Saskatchewan boundary. SW Saskatchewan Successful pag Year Nests Size $.D. All 1944- 1967 30 2.30 0.65 1968 9 2.44 0.52 1969 24 2.67 0.69 1970 16 2.50 0.83 1971 18 2.06 0.86 1972 60 2.43 O75 2 1973 27 1.93 0.77 5 1974 34 2.41 0.71 6 1975 23 1.91 0.72 6 1976 8 2.38 0.53 | 1977 40 2.10 O73 10 1978 43 2.35 0.91 14 1979 43 2.26 0.74 15 1980 50 2.02 O71 9 1981 68 2.10 0:75 ee 1982 62 1.85 0.76 24 1983 84 2.05 0.87 52 1984 61 74 0.68 20 1985 66 2.05 0.83 32 1986 113 2.08 0.78 38 1987 115 1.83 1.00 49 1988 97 1.68 0.76 18 1989 a5 1.61 0.68 28 1990 113 1.70 0.74 34 1991 86 1.53 0.65 24 1992 9] 1.59 0.65 29 1993 45 V27 0.62 9 1994 95 1.83 0.81 DS 1995 81 1.49 0.62 30 1996 50 1.48 0.71 16 Total NPAT 1.92 493 tion estimate also showed an increase from an esti- mated 3879 nests in 1982 (Schmutz 1984) to 7 305 nests in 1987 (Schmutz 1989). Ferruginous Hawks responded similarly during these periods (Schmutz 1989). It is likely that at least in the moderate- and high-density years, the Hanna population had saturat- ed all or nearly all available territories. The observa- tion that the numbers of breeders within each of the three periods exhibited little variation in numbers is consistent with the notion of saturation at each level of prey availability. High nesting densities in compari- son to other studies, regular nest spacing (Schmutz et al. 1980), and frequent aerial attacks observed throughout this study further support this conclusion. _ The juxtaposition of stability and change in long-term hawk dynamics is striking. Near Hanna, reproduction in successful nests appeared stable through 1990 and then declined by 31%, followed by a hawk population drop in 1995 Kindersley Alsask Nee Brood nes Brood Failed Size All Failed Size 1 2.00 2 2.20 3 2.33 2) Deel | 3.00 5) 2.50 3 2.93 4 2553 6 201 3 1.94 2, 2.08 15 1.84 115) 1.80 6 2.09 8 1 2.00 3 DAS 13 6 1.85 13 1.63 21 2 DS 19 1.50 26 8 2.08 20 1.50 16 9 eS ih) 1.47 16 8 2.06 Ng |e) 20 9 1.80 18 1.38 6 14 1.67 14 22. 8 6 1.00 1 1.88 6 ill 1.50 5) 1.40 10 4 1.60 12 1.69 i Sy 1.14 206 1.80 1357 83 1.89 and 1996. In Saskatchewan, reproduction began to decline in 1988 (when Hanna numbers had peaked) and reached unprecedented low levels in 1993 and 1996, followed by a noticeable drop in hawk popula- tions beginning in 1994. Prey Alberta (Hanna): In 1986, there was a close cor- respondence between the rise in hawk and ground squirrel numbers (Figure 2). The hawk population dropped near Hanna, the area where habitat had been altered least, only in 1995-1996 (Figure 2), four years after the drop in brood size (Figure 6). Saskatchewan: Drastic ground squirrel prey and hawk brood size declines in Saskatchewan began in 1988 in the highly cultivated landscapes near Kin- dersley, Saskatchewan (Figure 3), earlier than in the rangelands near Hanna, Alberta. Alsask was interme- diate in vegetation (Figure 1), and in timing of the 268 3.0 2.5 ‘a o C4 4 fF Pes be. i i 2.0 Nestlings per Successful Nest 1.5 —O— Kindersley —-@— Alsask 1.0 1975 1980 THE CANADIAN FIELD-NATURALIST Vol. 115 1985 1990 1295 FIGURE 7. Annual mean number of nestlings reared by Swainson’s Hawks in two regions of Saskatchewan. The drought period reflects below average annual precipitation on a plot showing a 10-year moving average. declines. Record-setting heat during early June 1988, with accompanying drought and extensive loss of native vegetation at temporary ponds, may have con- tributed to the hawk brood size and ground squirrel declines. In Saskatchewan, brood size declines were followed six years later, in 1993, by the first sharp drop in nesting pairs. Some ponds were permanently drained and then cultivated. Pond loss is a widespread phenomenon on the prairies (Ignatiuk and Duncan 1995). Food niche On the Northern Great Plains, Swainson’s Hawks have a two-niche feeding strategy. Adults feed main- ly mammals to their young. But adults and fledglings rely heavily on insects, which are at their numerical peak after the young hawks have fledged. The Hanna study area experienced high densities of grasshoppers that overlapped with the high in ground squirrels. In the Hanna study, perhaps a quar- ter of the breeding adults captured during the nestling stage had “dirty feet,” as though they had been walking and feeding in fields, as non-breeding, insect-eating flocks are prone to do. Based on over 1500 habitat-stratified sampling sites about 10km apart in south-central and south- eastern Alberta, Johnson and Worobec (1988) described 1978 to 1982 as years of low grasshopper density followed by “severe” outbreaks from 1983 to 1986. Computer-generated maps showed Hanna area grasshopper densities of 0-2 (1982), 2-4 (1983), 4-6 (1984), 4-6 (1985), and 2-4 (1986) grasshoppers/m?, not quite reaching the 6—10/m/? that defines a severe infestation. The most common species of grasshop- per were Melanoplus sp. and Camnula pellucida. Among eight studies reviewed by England et al. (1997), the percent of insects by items (not by mass) in the diet of Swainson’s Hawks ranged from 0 to 55%. Only in New Mexico did insects comprise more than 20% of items; there, pellets were collected from nests and also on the ground (some of these pellets may have been shed by parent Swainson’s Hawks, increasing the apparent insect representation in the diet). Given the largely vertebrate complement of — other prey, these insects would comprise much less than 10% of biomass. No Swainson’s Hawk popula- tion has been known to raise its nestlings on insects. Therefore, the availability of grasshoppers and other insects may influence the dispersal of non- breeding hawks. The grasshoppers may have facili- tated the increase by attracting non-breeders, which — then filled new territories that became available dur- ing the peak three years of ground squirrel popula- tions. Where insect and vertebrate availability coin- — cide, breeding sites may be saturated and fledgling survival high, giving rise to a “source” population. Food habits after fledging may differ from pat- terns reported during the nestling period. One road- killed fledgling found near Alsask in August 1975 © had a crop engorged with insects. This reliance on insects results in part because adult male ground squirrels go into hibernation in June and adult females in July, leaving above ground only juve- niles, whose numbers are reduced by predation (Yeaton 1972; Schmutz et al. 1979). Potential song- | | | 2001 SCHMUTZ, HOUSTON, AND BARRY: DECLINE AMONG SWAINSON’S HAWKS 269 Plots FERS Yolo County 40 Calif. Alberta 30 Survey “3 nl = os . Hanna, AB q B high © g a 20 Sask. Butte V. ” CA ry 4 @ med. za _ SE New M. 10 San Joaquin E CA B low ii P.C. Colo. SE Idaho 0 (og 200 400 600 800 1000 2300 Size of Study Area (km*) FIGURE 8. Swainson’s Hawk densities based on study area size at Hanna (high, medium and low, Table 2), on the 1992 41-km? survey plots in SE Alberta (mean = 7.7 nests/100 km?), and on a township in the Kindersley, Saskatchewan area are compared to other recent studies: North Dakota (3 years; Ig] and Johnson 1997), Pinion Canyon, Colorado (4 years; Anderson 1995; SE New Mexico (3 years; Bednarz 1988; SE Idaho (3 years; Hansen and Flake 1995); Yolo County (5 years) and San Joaquin Valley (3 years), California, England et al. 1995; Butte Valley, California, (4 years; Woodbridge et al. 1995). bird prey are now mature and some already have departed during the dry late summer. Insects, in con- trast, are at their peak in August. Hawk densities In comparison with other parts of the species’ range (southeastern Idaho, Hansen and Flake, 1995; south- eastern Colorado, Andersen 1995; southeastern New Mexico, Bednarz 1988; Butte Valley, California, Woodbridge et al. 1995; Yolo Valley and San Joaquin Valley, California, England et al. 1995), our hawk densities are only slightly higher than average (Figure 8). The southeastern Alberta survey and that of Ig] et al. (1999) in North Dakota may be the only studies where Swainson’s Hawk population totals can be reli- ably extrapolated from samples representing 4.4% and 0.05%, respectively, of these two large regions. Swainson’s Hawks have adjusted remarkably well _ to cereal crop production, especially where a moder- ate mix of habitat persists. Our earlier analyses showed that Swainson’s Hawk densities are higher on plots with >30% cultivation than on plots with 0- 30% cultivation (Schmutz 1989). Elsewhere, nesting densities tend to be “...highest in areas with either a mixture of native habitat and agriculture or a high diversity of irrigated crops” (England et al. 1997). Trees Swainson’s Hawks require trees or shrubs for nest- ing. Throughout previous centuries, almost annual range fires kept the prairies almost treeless. Apart from willows on the edges of sloughs, both study sites were virtually devoid of trees until settlers brought most fires under control about 1910 (Houston and Bechard 1983; Houston and Schmutz 1999). The last major fire that burned over parts or all of the Hanna study area, eastward to near Alsask, occurred in 1909 (Gordon 1978). A much smaller fire there in 1978 burned some aspen clumps used by Swainson’s Hawks, but the trees resprouted and sup- ported Swainson’s Hawks again by 1985. Most farmers planted large shelter-belts, the trees supplied free from the Forest Nursery Station at Indian Head, Saskatchewan, founded in 1901 and by 1929 supplying four million trees per year (Cram 1984). Trees suitable for nesting by Swainson’s Hawks became more accessible about 1950 as increasing numbers of farms were abandoned. 270 Although many of these planted trees have been plowed down, and others are deteriorating from neglect, trees remain more numerous than they were when European settlement began roughly 100 years ago. For this reason alone, Swainson’s Hawks may now be more common on the northern Great Plains than they were in pre-settlement time. A feature of Swainson’s Hawk nesting ecology is the frequent use of low shrubs such as Willows, Chokecherry, Saskatoon, Buffaloberry, and Prickly Rose (Schmutz et al. 1980; Table 2), even when other trees are also available. Nest construction is well matched to a shrub nest base. Swainson’s Hawks, unlike sympatric congeners, use herbaceous plants and small branching twigs (Schmutz et al. 1980), easily anchored among willow stems, though not always secured for long. This nest adaptability of Swainson’s Hawks may reflect past selection pres- sures on tree-scarce plains. Shooting of any hawk on sight, common in the early days of settlement by Europeans, is now rare; some pairs have proved adaptable enough to nest in occupied farmsteads. Since the 1940s, many more pairs have been able to use trees in deserted farm- steads although the remaining trees are untended, rubbed by cattle, sprayed with aerial herbicides, and damaged by drought. Each year, fewer of these for- mer farmstead trees remain. Agricultural effects The “industrialization of agriculture” throughout most of the Great Plains has resulted in a reduced number of farms (Houston and Schmutz 1999), coin- cident with increasing use of fertilizer, pesticides, and herbicides; employment of larger machinery; bulldozing of abandoned shelterbelts; advent of rock-picking machines to allow breaking of marginal land; drainage of wetlands; cultivation of former railway rights-of-way and of the previous grassed and fenced 20.1 m wide road allowances, and use of chemical, untilled summer-fallow. Large areas of resulting treeless monocultures become unattractive to hawks and their prey species alike. The future of the Swainson’s Hawk rests largely with private ranchers and managers of government pastures. Ranchers, after all, are the stewards of the remaining grasslands. The Prairie Conservation Action Plan, launched in January 1986 at the first Endangered Species Workshop in Edmonton Alberta (Dyson 1996; Hummel 1987), is a promising devel- opment. More recently, the Saskatchewan Stock- growers Association (SGA) has led the Prairie Conservation Action Plan Initiative. On 4-10 June 2000, assisted by other agencies and Saskatchewan Minister of Environment and Resource Management Buckley Belanger, the SGA proclaimed Native Prairie Appreciation Week. Something deleterious is happening to grasslands, yet there is little evidence to suggest which changes in THE CANADIAN FIELD-NATURALIST Vol. 115 farming practice, alone or in combination, are respon- sible. With each decade the ever-smaller remnants of native habitat become an ever-scarcer component of the mosaic. It seems ironic that while the recent “¢mdustrialization/monoculturization” of agriculture is concentrated mainly on croplands, not grasslands, the continuing and seemingly inexorable bird species population declines affect grasslands disproportion- ately. Swainson’s Hawk productivity studies offer one measure of change in the grassland-cropland ecosys- tem; increased understanding may in future guide us to more appropriate management techniques. Acknowledgments Both studies have been assisted by numerous peo- ple whose kind and capable help we gratefully acknowledge. Jean Harris, Dean Francis, Sig Jordheim and Dan Zazelenchuk found many of the Saskatchewan nests, and numerous climbers assisted in banding. Michael A. Yates trapped the nine-year- old colour-banded hawk. The Hanna project was supported by Alberta Environmental Research Trust, Alberta Environmental Services, Alberta Sports, Recreation, Parks and Wildlife Foundation, Canadian Wildlife Service, Special Areas Board of Hanna, University of Saskatchewan and World Wildlife Fund Canada. Mary Gilliland and J. Frank Roy offered editorial criticism. Literature Cited Andersen, D. E. 1995. Productivity, food habits, and behavior of Swainson’s Hawks breeding in southeast Colorado. Journal of Raptor Research 29: 158-165. Anderson, D. W., C. J. Roppel, and R. M. Gray. 1991. Sustainability in Canadian agriculture. Science Council of Canada, Ottawa, Ontario, Canada. Bednarz, J.C. 1988. A comparative study of the breeding ecology of Harris’ and Swainson’s Hawks in southeast- ern New Mexico. Condor 90: 311-323. Bloom, P. H., J. L. Henckel, E. H. Henckel, J. K. Schmutz, B. Woodbridge, J. R. Bryan, R. L. Anderson, P. J. Detrich, T. L. Maechtle, J. O. McKinley, M. D. McCrary, K. Titus, and P. F. Schempf. 1992. The dho-gaza with Great Horned Owl lure: An analysis of its effectiveness in capturing raptors. Journal of Raptor Research 26: 167-178. Clark, W. S., and B. K. Wheeler. 1987. A field guide to hawks: North America. Houghton Mifflin, Boston, Massachusetts, USA. Clayton K. M., and J. K. Schmutz. 1999. Is the decline of burrowing owls Speotyto cunicularia in prairie Canada linked to changes in Great Plains ecosystems? Bird Conservation International 9: 163-185. Connelly, J. W., and C. E. Braun. 1997. Long-term changes in Sage Grouse Centrocercus urophasianus populations in western North America. Wildlife Biology 3: 229-234. Cram, W. H. 1984. History of the forest nursery station tree nursery. Pages 46-50 in History of Indian Head and District. Indian Head Historical Society, Indian Head, Saskatchewan, Canada. Dyson, I. W. 1996. Canada’s prairie conservation action 2001 plan. Pages 175-186 in Prairie conservation: Preserving North America’s most endangered ecosystem. Edited by F. B. Samson and F. L. Knopf. Island Press, Washington, D.C., USA. England, A. S., M. J. Bechard, and C. S. Houston. 1997. Swainson’s Hawk (Buteo swainsoni). In The birds of North America, Number 265. Edited by A. Poole and F. Gill. The Academy of Natural Sciences, Philadelphia, Pennsylvania, USA, and the American Ornithologists’ Union, Washington, D. C., USA. Fretwell, S. D., and H. L. Lucas, Jr. 1970. On territorial behavior and other factors influencing habitat distribu- tion in birds. Acta Biotheoretica 9: 16-36. Gordon, A., Editor. 1978. Roads to Rose Lynn: A history of the Rose Lynn, Halladay, Sheerness, Stanmore, Scotfield, Richdale and Berry Creek areas. Inter- Collegiate Press, Edmonton, Alberta, Canada. Gorman, J. 1988. A land reclaimed: A story of the Special Areas in Alberta. Gorman and Gorman, Hanna, Alberta, Canada. Hansen, R. W., and L. D. Flake. 1995. Ecological rela- tionships between nesting Swainson’s and Red-tailed Hawks in southeastern Idaho. Journal of Raptor Research 29: 166-171. Houston, C. S. 1974. Mortality in ringing: A personal viewpoint. The Ring 80: 157-161. Houston, C. S., and M. J. Bechard.- 1983. Trees and the Red-tailed Hawk in southern Saskatchewan. Blue Jay 41: 99-109. Houston, C. S., and J. K. Schmutz. 1995a. Swainson’s Hawk banding in North America to 1992. North American Bird Bander 20: 120-127. Houston, C. S., and J. K. Schmutz. 1995b. Declining reproduction among Swainson’s Hawks in prairie Canada. Journal of Raptor Research 29: 198-201. Houston, C. S., and J. K. Schmutz. 1999. Changes in bird populations on Canadian grasslands. Studies in Avian Biology 19: 87-94. Hummel, M. 1987. Prairie conservation. Pages 21-25 in Proceedings of the workshop on endangered species in the prairie provinces. Edited by G. L. Holroyd, W. B. McGillivray, P. H. R. Stepney, D. M. Ealey, G. C. Trottier, and K. E. Eberhart. Provincial Museum of Alberta Natural History Occasional Paper Number 9. Hunter, A. F. 1995. Ecology, life history, and phylogeny of outbreak and nonoutbreak species. Pages 41-64 in Population dynamics: New approaches and synthesis. Edited by N. Cappuccino and P. W. Price. Academic Press, San Diego, California, USA. Igl, L. D., D. H. Johnson, and H. A. Kantrud. 1999. Uncommon breeding birds in North Dakota: Population estimates and frequencies of occurrence. Canadian Field-Naturalist 113: 646-651. Ignatiuk, J., and D. C. Duncan. 1995. Wetland loss in aspen parkland of Saskatchewan. Blue Jay 53: 129-133. Jameson, S. S. 1986. The ranching industry of western Canada: Its initial epoch, 1873-1910. Prairie Forum 11: 229-242. _ Johnson, D. L., and A. Worobec. 1988. Spatial and tem- poral computer analysis of insects and weather: Grasshoppers and rainfall in Alberta. Memoirs of the Entomological Society of Canada 146: 33-48. Jones, E. 1993. Summer ‘93: A catastrophic year for SCHMUTZ, HOUSTON, AND BARRY: DECLINE AMONG SWAINSON’S HAWKS 27) Ferruginous and Swainson’s Hawk. Alberta Naturalist 23: 14. Mayfield, H. 1975. Suggestions for calculating nest suc- cess. Wilson Bulletin 87: 456-466. Padbury, G. A., and D. F. Acton. 1994. Ecoregions of Saskatchewan. Central Survey and Mapping Agency, Regina, Saskatchewan, Canada. Portman, J. 1997. Genetic diversity among Ferruginous and Swainson’s Hawks: An interdisciplinary interpreta- tion. M. Sc. thesis, Department of Biology, University of Saskatchewan, Saskatoon. Pulliam, H. R. 1988. Sources, sinks, and population regu- lation. American Naturalist 132: 652-661. Rowe, S. 1987. One hundred years of land use. Blue Jay 45: 127-139. Schmutz, J. K. 1977. Relationships between three species of the genus Buteo (Aves) coexisting in the prairie-park- land ecotone of southeastern Alberta. M.Sc. thesis, University of Alberta, Edmonton, Alberta, Canada. Schmutz, J. K. 1984. Ferruginous and Swainson’s Hawk abundance and distribution in relation to land use in southeastern Alberta. Journal of Wildlife Management 48: 1180-1187. Schmutz, J. K. 1989. Hawk occupancy of disturbed grass- lands in relation to models of habitat selection. Condor 91: 362-371. Schmutz, J. K., R. W. Fyfe, U. Banasch, and H. Armbruster. 1991. Routes and timing of migration of falcons banded in Canada. Wilson Bulletin 103: 44-58. 1989. Hawk occupancy of disturbed grasslands in relation to models of habitat selection. Condor 91: 362-371. Schmutz, J. K., C.S. Houston and G. L. Holroyd. 1996. Southward migration of Swainson’s Hawks: over 10,000 km in 54 days. Blue Jay 54: 70-76. Schmutz, J. K., and D. J. Hungle. 1989. Populations of Ferruginous and Swainson’s Hawks increase in syn- chrony with ground squirrels. Canadian Journal of Zoology 67: 2596-2601. Schmutz, J. K., S. M. Schmutz, and D. A. Boag. 1980. Coexistence of three species of hawks (Buteo spp.) in the prairie-parkland ecotone. Canadian Journal of Zoology 58: 1075-1089. Schmutz, S. M., D. A. Boag, and J. K. Schmutz. 1979. Causes of the unequal sex ratio in populations of adult Richardson’s ground squirrels. Canadian Journal of Zoology 57: 1849-1855. Simmons, R. E., D. M. Avery, and G. Avery. 1991. Biases in diets determined from pellets and remains: Correction factors for a mammal and bird-eating raptor. Journal of Raptor Research 25: 63-67. Strong, W. L., and K. R. Leggat. 1992. Ecoregions of Alberta. Publication Number T/245, Alberta Environmental Protection, Edmonton, Alberta, Canada. Woodbridge, B., K. K. Finley, and P. H. Bloom. 1995. Reproductive performance, age structure, and natal dis- persal of Swainson’s Hawks in the Butte Valley, California. Journal of Raptor Research 29: 187-192. Yeaton, R. I. 1972. Social behavior and social organiza- tion in Richardson’s ground squirrel (Spermophilus richardsonii) in Saskatchewan. Journal of Mammalogy 53: 139-147. Received 14 August 2000 Accepted 15 March 2001 DID THE CANADIAN FIELD-NATURALIST Vol. 115 APPENDIX |. Prey items found in nests of Swainson’s Hawks by region. The time period for the Saskatechewan study is split to allow comparisons between study areas. The number of items and their percentage by number and weight are reported. For prey near Hanna prior to 1983, see Schmutz et al. (1980). Saskatchewan 1966-1982 1983-1996 Alberta 1983-1996 # Jott =6ToWt # TH Wt # TH Wt Salamander Ambystoma tigrinum 1 0.05). Two traps/day at vidda area, 4 traps/day at edge area and 4 traps/day at woods area (except 5 on 31 July and 6 on 2 August). montana, Ay. nitififrons confiformis and Hy. sexfas- ciata. Tabanid populations at this inland site proba- bly peaked before yearly trapping began. Percentage of flies caught in apical collecting containers of traps versus numbers aspirated from interior panels and canopy of traps at shut down times The 22 samples in which the number of flies in the apical collection containers was compared with the number aspirated from the inner panels and canopy of the trap revealed that, for culicids, about '/4 (mean 22%, median 24%) of the catch was in the apical collecting container. For simuliids, about '/2 (mean 42%, median 47%) of the catch was in the apical collecting container. On 2 and 8 August 1985, 76% of 21 tabanids caught in 14 traps were in the apical collecting containers, and on 21 and 30 July 1987, all of 4 tabanids caught in 8 traps were in the apical collecting containers. Comparison of biting fly catches with oestrid catches on different days Knowledge of the percentages of biting flies in collecting container samples allowed estimates of total numbers caught on all days for comparison with the total number of oestrids caught on different days. Such comparisons revealed that CO,-baited traps caught large numbers of biting flies on days that were mostly unfavorable for oestrid host-seeking activity (Tables 3, 4). The correlation matrix (Table 5) and multidimensional scaling plot (Figure 2) show that the best catches of culicids and simuliids were positevely correlated, as were catches of the two oestrids. Similar comparisons for 1984 catches could not be made because almost all culicid and simuliid collections were destroyed in storage. In addition to the 1985 dates in Table 3, on 3 through 6 July 1985, six CO,-baited traps operated at area “B” caught a mean of 11723 + Ll6.0 culicids and’ 23.3 + 24-9 simuliids, but no oestrids or tabanids. After 10 August 1985 both the oestrid and biting fly popula- tions declined severely. Hypoderma tarandi declined to 1.1/trap on 14 August and less than 0.5/trap there- after; C. trompe remained at less than 0.5/trap after 10 August. In 10 traps, only one H. tarandi was caught on 18 August and one C. trompe on 21 August (Anderson and Nilssen 1996a). No oestrids were caught in 8 to 10 traps/day on 22 and 28 to 31 August. After 10 August 1985 total culicid catches varied from 30-—40/trap and simuliids from 40—50/trap through 22 August. Total culicid catches remained at 5-—10/trap and simuliids from 10—20/trap through 30 August. In addition to the eight days shown in Table 4, 12 CO,-baited traps operated on 24 and 27 July were estimated to have as many culicids and simuliids per trap as were caught on 21 July, but all traps caught only one C. trompe and one H. tarandi on 24 July, and only one H. tarandi and three tabanids (H. mon- tana) on 27 July. Climatic conditions associated with trap catches of parasitic flies Trapping days not included in Tables 3 and 4 expe- rienced some combination of being darkly overcast, cool (7—14°C) and windy (8-10 m/s), or cool and rainy. Temperature, light intensity and wind velocity affected the host-seeking activity of all flies captured (Table 6, and Anderson and Nilssen 1996a, for 1985 climatic conditions). Most oestrids were caught on warm (17—25°C), mostly sunny and calm days. On 27 trapping days during three years oestrids were not TABLE 2. Mean +SD number of tabanids caught per trap collecting container per day, 1984.° Date Woods Vidda 20 June 63 + 39.0 (4)* = 23 June LStiGl) —_ 27 June 191.34 49.9 (4) — 9 July 189.4 + 106.2 (5) 43 (1) 10 July (WO. 7) ae GB) 45.0 = 29113) 12 July 395.34 62.3 4) 52.0 + 31.5 4) 24 July Pps: = SNWO(A) 30.5+ 5.4 (4) “Numbers caught in apical collection container of a trap represent 76% of the total catch (aspirator samples on 4 days revealed that 24% of the tabanids in the trap were resting on inner surface of trap panels and the trap canopy). *Number of traps. For 23 June, several collections were inadvertently destroyed in storage. | | | 2001 ANDERSON, NILSSEN, AND HEMMINGSEN: HOST-MIMICKING TRAP CATCHES 279 TABLE 3. Comparative ranking of best CO,-baited trap catch days for culicids versus other taxa of parasitic flies, 1985” Culicid Simuliid H. tarandi Date rank rank rank 29 July I (484)* IV (266) ME 57) 9 August II (297) V (147) V (3.9) 2 August III (292) I (530) IV (4.1) 31 July IV (136) IT (341) VI (1.8) 8 August V (61) Ill (296) HELLS) 10 August VI (21) VI (83) II (6.8) C. trompe Tabanid Time of rank rank last collection IV (0.6) 1 (5.9) L720‘ RCL IV (0.6) VI (0.8) 1730 h (14.0) 1-0) Ill (2.3) 1921 h (15.3) W035) II (4.0) 1734 h (18.0) I (2.67) Vis) 1830 h (14.8) Ill (0.9) IV (1.6) 1836 h (19.0) “Catches of flies are from 6-10 traps operated on different days at woodland, vidda and woodland: vidda edge sites. *Mean number of flies/trap in parenthesis. For culicids, simuliids and tabanids mean numbers for 2 and 8 August are based on the actual trap totals (apical collection container catches plus flies aspirated from interior trap panels and trap canopy); mean numbers for all other dates are based on adjusted totals estimating that only 25% of all culicids, 50% of all simuli- ids, and 75% of all tabanids trapped were in apical collection containers. *Temperature (°C) at time of last collection; Time (NST). trapped when temperature was below 10°C; on dark, cloudy or rainy days (light intensity below 20 000 lux); or at wind velocities above 8 m/s. Most tabanids were caught on days having the same climatic condi- tions favoring oestrid activity, with only four caught on oestrid-negative days in 1985. Culicids and simuli- ids, however, were caught on all trapping days nega- tive for oestrids and tabanids, as well as on the days shown in Tables 3 & 4. Culicids and simuliids were not trapped when the temperature was less than 7.0°C or when wind velocities were 8—10 m/s. On some days, warm, sunny conditions were favorable for host- seeking activity by oestrids, tabanids and biting flies (e.g. Table 3, 29 July & 2 August; Table 4, 4, 9-10 August). On such days (Table 7) culicid and simuliid numbers peaked during the crepuscular period. Because of adverse climatic conditions while at field study sites traps were not operated on 11 days in 1984, 11 days in 1985 and 14 days in 1987. These days, plus the oestrid-negative trapping days, revealed that adverse climatic conditions prevented oestrid host-seeking activity on 21 of 35 days in 1984, 22 of 39 days in 1985, and 19 of 28 days in 1987. Catches of hematophagous flies at different diel times For trap catches collected twice a day in 1985 paired sample f-tests revealed that significantly fewer simuliids were caught in traps from 09:30 to 14:30 h than were caught in traps from 14:30 to 19:30 h (t= 4.01; df =54; P= 0.00019), but no sig- nificant difference in the numbers of culicids caught during the first versus the second trapping period (t = 0.97; df = 54; P = 0.34) nor the numbers of tabanids caught during the first versus the second trapping penodi— 0.026: di— 54. P' — O95): Temperature during crepuscular/overnight trap- ping periods had a marked affect on the number of culicids caught. Both the largest and smallest num- bers of culicids were caught when traps were operat- ed from early crepuscular hours through an overnight period (Table 7). At temperatures of 12.0°C to 14.0°C, large num- TABLE. 4. Comparative ranking of best CO,-baited trap catch days for culicids versus other taxa of parasitic flies, 1987*. Culicid Simuliid H. tarandi Date rank rank rank 31 July I (2034)? I (2369) VI (0.63) 30 July II (1768) Ill (1118) V (0.70) 21 July Ili (497) VIL (214) VII (0.50) 4 August IV (493) VI (499) IV (0.75) 10 August V (381) IV (756) II (4.4) 9 August VI (342) II (1529) I (6.2) 6 August VII (257) VIII (200) VII (0.0) 20 July VIll (118) ¥ (621) III (1.0) C. trompe Tabanid Time of rank rank last collection IV (0.13) V GeO) 2100h (14.0)* V (0.10) VI (0.0) T952bi( EB) VI (0.0) Ht (20) 1608h (13.6) Ill (0.25) V. (0) 2015h (16.9) II (0.22) EV (eS) 2000h (11.8) I (0.40) II (1.8) 2030h (10.7) VI (0.0) VII (0.0) 2057h (10.0) VI (0.0) LCL) 1733h (23.3) “Catches of flies are from two traps operated at woodland sites on 20 and 21 July, and from 4-8 traps operated at woodland and woodland:vidda sites on all other dates. *Mean number of flies/trap in parenthesis. For culicids and simuliids mean numbers for 21 and 30 July are based on total trap catches (apical collection container catches plus flies aspirated from interior panels and canopy of traps); mean num- bers for all other dates are based on adjusted totals estimating that only 25% of all culicids, 50% of all simuliids, and 75% of all tabanids trapped were in apical collection containers. *Temperature ( °C) at time of last collection; Time (NST). 280 THE CANADIAN FIELD-NATURALIST Vol. 115 TaBLE 5. Pearson correlation matix (r-values) from a correlation analysis based on mean catch data of the taxa shown in Tables 3 and 4, both years combined. Culicidae Simulidae Culicidae Sumulidae 0.76" Hi. tarandi 0.43 -0.21 C. trompe -0.36 0.27 Tabanidae -0.31 -0.13 “significant (p < 0.001) bers of culicids (A. communis, A. pionips and A. punctor) were collected from 19:00—20:20 h in power aspirator samples taken on 10 August 1987 while slowly walking around a CO_-baited Reindeer model and collecting flies landed on and swarming around the model (Table 8). For comparison with these numbers, the adjusted mean number of culicids caught/trap in nine traps set at similar woodland sites on this date was 381 (range = 296—624): for simulli- ids the number was 756/trap (range = 120 to 2904). The culicid index around a person in a woodland clearms was III at 19:00 h and IV at 20:00 h (see below) On the Reindeer model culicid and simuliud females crawled and probed areas of the head, as well as the Reindeer hide covering the body; H. tarandi females laid eggs on the hairs and C. trompe deposited larvae on the muzzle (Anderson and Nilssen 1990). MULTIDIMENSIONAL SCALING ~ 054 CULICIDAE 6 ‘ D e = SIMULIIDAE => D -0. 05 0 a5 1 Sa 2 DIMENSION i FaGure 2. Mulndimensional scaling (MDS) plot based on the correlation matrix m Table 5. Taxa close to one another show a positive correlation (abundance fluctuates mostly synchronously), whereas distant taxa show a high negative correlation. H. tarandi C. trompe Tabanidae 0.87 -0.05 -~0.12 Culicid swarming/landing indices at humans The 1987 swarming/landing indices of mosquitoes attracted to a human who had walked 50-100 m into a birch woodland are shown in Figure 3. Vidda indices were similar, except that on 20 July and on 1 and 5 August, winds of 5.0-8.5 m/sec negated culicid attrac- tion to humans throughout most of the day. Because of adverse climatic conditions on dates not shown, culicid activity was at the 0 to I level. The following swarm often included as many, or more, culicids than had landed on the observed human, as well as smaller numbers of simuliids. At a swarming/landing index of level IV on two dates, the mean catch of culicids in 20 figure 8 sweeps with an insect net was 59.5 + 13.4: for level III on two dates the mean catch was 33.5 + 2.3, and for level II on three dates the mean catch was 18.7 + 6.5. In these collections, culicids comprised 77.5% of the catch and simuliids 22.5%. Aedes communis, A. pionips and A. punctor were caught biting, and in sweep net collections. The number of attacking culicids fluctuated some- what with time of day and temperature. If the tem- perature was above 10.0°C in the moming, attack rates were at level II or Ill from about 09:00 to 12:00 h. At temperatures between 10.0°C and 20.0°C numbers attacking nearly always were lowest (level I or I) between 12:00 and 15:00 h, but they usually increased to level II or IV between 16:00 and 20:00 h. After 19:00—20:00 h, culicid biting activity decreased with decreasing temperature. At 7—10.0°C, it was always at level I between 20:00 and 00:30 h. On five dates, no culicids were observed at temperatures of 3.5 to 5.5°C (times were 07:00— 08:15, 22:20 and 24:00 h). Simuliids attracted to humans In both woodland and vidda habitat, simuliids also sporadically swarmed about humans, but as they rarely landed and attempted to bite, their numbers were not estimated. In 20 figure 8 imsect net sweeps taken when the culicid swarming/landing index was at level IV, the mean + SD catch of simuliids was 8.5 + 7.8: at culicid level II] the mean catch of simuliids was 16.0+5.7, and at culicid level Il the mean catch of simuliids was 8.0 + 7.8. Movements and reactions of Reindeer to flies At study area “A” from 19 June through 24 July "Onder of dates conforms to order of dates in Table 4. Cle ‘ the woodland Tapping site at stacy area “B™. 1984. Remdeer moved out of woodland habitat and up onto the treeless vidda prior to 07-00—08:00 h. They moved back down into the woodland between 19:00 and 19:30 h. Groups of 50—100 animals Served on the vidda from 10-00 to 13-00 h on two warm. sunny days foraged at a moderate walking ce for about S—10 mimutes and then suddenly tot- ed for 50 to several hundred meters, after which agai began walking and foraging. Individual animals observed with binoculars reacted with the types of anti-fly behaviors described by Anderson d Nilssen (1998). These observations and trap catches (Table 2) indicated that the animals were 2 to tabanids. Te chtreced from 19:30 to 22:00 h (after tabanid activity had ceased) slowly walked and for- aged through the woods. During 1—5 minute obser- vation periods about half of these animals responded ith one or two mild reactions consisting of 2 shak- ine lee kick. tail wag/ rump shake. ear flick. or 2 ead shake. All such responses are typical of host Teactions to culicids (Anderson and Nilssen 1998). The slow foraging pace through woodland. and mild Trapping Number hours of traps 18:20 - 08:20 I 18:10 - 07:50 i 18:35 - 08:15 4 1845 - 8:12 I 17245 - 08:15 I 19:35 - 09:05 3 20-15 - 09:30 3 22:15 - 08:15 + _ *Temperature at evening start time over overnight low. ANDERSON, NILSSEN. AND HEMMINGSEN: HOST-MIMECKING TRAP CATCHES 2381 6. Climatic conditions associated with daytime catches of parasitic flies mm Table 4 ee Temperatare* Wind velocity” : a5 cloud cover* min mean mia (m/s) 31 July 75.0 os o64 158 206400020 30 July 75.0 06 &5 142 304 V1020 21 July 100.0 82 146 22: 3.065.072.0225 August 375 84 158 210 3.00.5-2.0 ) August 375 02 108 1s2 15250011 August 125 -16 3 170 10200005 6 August 25.0 45 o6 146 405.072030 20 July 50.0 10.6 19.5 254 553453.440 i cover and temperature data are from the Norwegian weather station at Suolowsopam. located 25 im from the center ximum velocities at different trap sites over mimimum velocities. anti-fly reactions, mdicated litle or no harassment by culicids. A tame three-year-old cow aftacked by culeids af study area ~“B™ exhibited similar reactions as she walked, paused and foraged. When aitackms num- bers reached an estimated 40-50 culhcads the cow responded with a head/body shake or body shake. and then trotted to a new forasme site and resumed feedms. Culhcads were seen feedimg around the eyes and on the muzzie, antlers, and legs. as well as crawling and probing on the body. A fortustously encountered sick fawn sampled for attacking fhes on 20 July 1987 remained lying motionless while attacked by hundreds of culkcids. A 4-mim power aspirator sample taken at 19-30 h (20°C) contamed 468 cuhcads (A. communis, A. piomips and A. pumc- tor) and one unidentified Samudaumn spp- At study area “B”, the foraging activity of Reindeer rarely appeared to be even moderately affected by parasitic fies (Table 9). One exception occurred on 9 and 10 August 1985 when 40-50 Reindeer were observed standing and lyme on the highest mountain top m the area (659 m). At wind TABLE 7. Effect of temperature on crepuscular/ovemight catches of culicids’. Temperature* Number co of cuhads/tap* 122/104 320 16.074 2973 17.2465 1757 +3383 13.740 377 18.120 0 8554 2021.15 2eo+ 13.0230 10458 33235 “In addition to these data overnight catches of culicids in four traps operated each night on 14-15 August 1985 and 30 july 1987 were noted as being <10 as collecting containers were emptied and replaced on site. Temperatures af evening Start ' times were 10.0. 9.1 and 9.2. respectively. and overnight lows were 8.6, 64 and 3.6 respectively. _ “These apical container catches represent about 25% of the total culicids trapped. 282 TABLE 8. Number of culicids and simuliids collected in 5- minute vacuum samples while slowly walking around a CO,-baited Reindeer model; 10 August 1987*. Time No. of culicids No. of simuliids 19:00-19:05 291 80 19:15-19:20 265 32 19:30-19:35 320 20 19:45-19:50 DI 22 20:00-20:05 548 34 20:15-20:20 243 3 “Temperature dropped from 14.2°C at 19:00 to 12.0°C at 20:20. speeds of 6-8 m/s, 6-10 C. trompe males remained active at this mating site throughout the observation period (5 were netted). Some Reindeer exhibited oestrid avoidance behaviors (Table 9), but no female oestrids were seen. The most intense herd reaction to parasitic flies was observed at study area “B” on 27 July 1987, when a herd of about 500 animals was videotaped for 37 minutes between 10:44 and 13:20 h (17°C, 60% cloudy with intermittent sun- shine). After foraging periods during which animals exhibited only mild reactions to flies, on five occa- sions all animals rapidly merged tightly together and broke into a medium to fast trot for 3-5 min. After each bout of trotting, the herd slowed to a walk before starting to disperse and resume foraging. 41-100+) IV (l= 1-10; Il = 11-25; Ill = 26-40; IV Index of attraction 20 THE CANADIAN FIELD-NATURALIST Vol. 115 Zoom lens videotaping of individuals and small groups foraging within 10-50 m of the videotaper helped distinguish reactions to culicids and simuli- ids from those of oestrids. Within different 5- minute periods, 10 to 30% of the animals exhibited mild culicid/ simuliid responses, but as the animals continued to walk slowly and feed it was apparent that none of these intermittent reactions were seri- ous enough to interfere with foraging. As the herd foraged around and past the videotaper several H. tarandi were seen flying and landing on nearby vegetation and one was caught. No C. trompe were seen, but there were three instances of animal reac- tions to C. trompe. During the 37 minute video- taped period, the herd spent 27% of the time in fly- associated, non-foraging activities (i.e., trotting or standing in a tightly bunched formation). While dispersed animals were grazing, the videotaped sequence shows nine individuals reacting to H. tarandi. Total non-grazing times were not obtained for these animals because of intermittent videotap- ing of different parts of the herd. In contrast to the mild shakes in response to other flies, animals responded to H. tarandi with vigorous body shakes (sometimes with two or three in rapid succession), usually followed by trotting away for several meters and then assuming a stiff-legged, motionless stance for several minutes. During three 5 min videotaped sequences between 11:00 h and 13:00h on 28 July 1987 (8°C, and 0900-1200 h ea 1200-1500 h al 4600-2000 h August FiGuRE 3. Swarming/landing indices of culicids attracted to human in birch/willow woodland (I = 1-10; II = 11-25; Ill = 26-40; IV = 41-100+). Samples were not taken on cold, rainy days. 2001 ANDERSON, NILSSEN, AND HEMMINGSEN: HOST-MIMICKING TRAP CATCHES 283 _ TaBLe 9. Reactions of Reindeer to inferred attacks by parasitic flies, August 1985. Five 5 min group scans of 40-50 animals on mountaintop revealed one animal reacting to C. trompe and three to H. tarandi. Five animals showed reactions to None in two 5-minute observations of groups of five and 14 closely dispersed None for three animals observed for 19 minutes, and none for one animal None among 1|1 closely dispersed animals observed for 21 minutes, 14 observed for.8 minutes and 20 observed for 25 minutes. 24% of 8 to 20 closely dispersed animals observed in three 5-minute scans showing 1-2 reactions to culicids/simuliids. In ten 5-minute scans of about 100 closely dispersed animals, four reacted to H. tarandi and about 25% in each scan showed culicid/simuliid reactions. Five 5-minute observations of two animals showing no reactions. None among 19 animals observed for 11 minutes. In two 5-minute scans of 12 closely dispersed animals five exhibited 5-minute observation of 42 closely dispersed animals with nine exhibiting | Date Time Observed reactions® Oth 09:20 - 12:15 culicids/simuliids. ~ 10th 10:00 - 11:00 | animals.* me) 1th 18:10 - 18:36 observed for 5 minutes. 13th 11:00 - 12:00 18:00 - 18:15 14th 10:00 - 12:00 | 20:15 - 20:50 15th 09:35 18:00 - 18:25 culicid/simuliid reactions. 22nd 10:15 - 10:20 culicid/simuliid reactions. 30th 09:30 - 10:00 Two 5-minute observations of five and six closely dispersed animals, with one in each group exhibiting culicid/simuliid reactions. _ *Cephenemyia trompe reaction = classic head shaking/lowering response described by Espmark (1961, 1968); Hypoderma tarandi reactions = vigorous body shaking followed by running and then standing motionless in stiff-legged position (some lying and standing animals also jumped); culicid/simuliid reactions = one or more of the following: tail wag/rump shake; leg shake or kick; head shake; ear flicks). - *Animals spaced 2 - 8 m apart. cloudy bright) no anti-fly behaviors were seen among two groups of 5 and 7 animals lying down, and only 8 of about 100 grazing animals exhibited a reaction to culicids/simuliids. Individual animals videotaped for 5-minute periods while foraging on the vidda between 15:00 h and 17:10 h on 3, 6, and 10 August 1987 (13.4-17.5°C, and mostly sunny) showed only one culicid/simuliid response in two, one and three 5- minute periods, respectively. On 6 August, one ani- mal among a group of five closely spaced animals exhibited a response to H. tarandi, and during one of the three 5-minute periods on 10 August, one animal also exhibited a response to H. tarandi. - Discussion Value of CO,-baited traps as Reindeer mimics Reports of insect-induced, panicked herd behavior resulting in reduced grazing time of Caribou/ Reindeer are common, but unfortunately, circum- _ Stances have not permitted concurrent collections of parasitic flies from animals to evaluate which species may be responsible for the observed behav- ior. In fact, because only a few researchers have col- lected parasitic flies from tame Reindeer (see Introduction) there are almost no data concerning the guild of parasitic flies associated with Caribou/ Reindeer in different geographical areas. For exam- ple, although there are at least 26 species of culicids, 38 species of simuliids and 29 species of tabanids known from Alaska, U.S.A. (Sommerman 1958; Gjullin et al. 1961; Burger 1995), with a similar fauna likely in northern Canada (Wood 1985), noth- ing is known about which, or how many, species of these flies attack Caribou at different times. In North America, CO,-baited traps and animal models have attracted and caught the same species of hematophagous flies caught concurrently feeding on Black-Tailed Deer (Odocoileus hemionus colum- bianus) (Anderson and Olkowski 1968; Anderson and Hoy 1972; Anderson et al. 1974), cattle (Roberts 1972; McCreadie et al. 1985; Mason and Kusters 1990) and horses (Anderson and Yee 1995). Moreover, these tools also have proven effective for assessing the seasonal abundance and diel periodici- ty of species attacking host animals (Anderson and Yee 1995). In Finland, Helle et al. (1992) reported that CO,-baited traps even caught the same hemato- phagous species in the same relative abundance as were concurrently caught attacking a tame Reindeer. 284 In this study CO,-baited traps caught the same three culicid species caught attacking Reindeer and humans, as well as several tabanid and simuliid species also believed to feed on Reindeer. We previ- ously reported that CO,-baited traps caught both oestrid parasites of Reindeer (Anderson and Nilssen 1996a) and the same tabanid (Hy. auripila) seen attacking Reindeer (Anderson and Nilssen 1998). Other tabanids caught in CO,-baited traps in this study include three species (Hy. montana, H. pluvi- alis and H. lundbecki) caught feeding on Reindeer by either Bergman (1915), Breev (1950) or Helle et al. (1992). We therefore conclude that the CO,-bait- ed traps functioned as effective Reindeer mimics in attracting parasitic flies. In situations where attacking flies cannot be sam- pled directly from Caribou/Reindeer, in future stud- ies to assess the impact of different parasitic flies on individual animals and on herd behavior of Caribou/Reindeer, host-mimicking CO,-baited traps and animal models would be useful in evalu- ating such factors as the seasonal and proportional abundance, diel activity and the potential negative impact of different flies. Furthermore, because no studies on parasitic flies associated with R. tarandus have spanned the entire fly “season”, there also is a need for further comparisons of the numbers and species of parasitic flies caught in baited traps (and from animal models) with concurrent collections of flies attacking tame animals in different habitats and in different geographical areas. Catches of flies at different diel times In apical collection container catches from 09:30 to 14:30 h versus those from 14:30 to 19:30 h there were significantly more simuliids caught in the sec- ond period, but there was no significant difference in the numbers of culicids caught during these periods. Both the diel attack indices for culicids at humans and the few late evening/overnight trap runs com- pleted, established that on warm, sunny days both culicids and simuliids reached peak attacking densi- ties from about 17:00 to 20:00 h, with large numbers of host-seeking culicids remaining active throughout the overnight period. In a related study (Anderson and Nilssen 1998), significantly fewer culicids and simuliids were caught on warm, sunny days from 09:30 to 19:20 h than from 19:20 to 07:00 h the fol- lowing day, with 94.5% of the culicids and 87.8% of the simuliids caught in the crepuscular/overnight period. By contrast, significantly more tabanids (Anderson and Nilssen 1998) and significantly more H. tarandi and C. trompe females were caught in first period trap catches than during the second peri- od (Anderson and Nilssen 1996a). This pattern of fly activity is similar to the feeding pattern of tabanids and culicids that attacked cattle in Manitoba, Canada (Ralley et al. 1993). THE CANADIAN FIELD-NATURALIST Vols Comparison of biting fly catches with oestrid catches on different days Climatic conditions greatly affected the host- seeking activity of different parasitic flies. Warm, sunny/partly sunny days (0 to 75% cloud cover) resulted in large numbers of all parasitic flies being caught in the host-mimicking traps (Tables 3,4) and, by inference, attacking Reindeer. Most oestrids and tabanids were caught on warm (17—25°C), mostly windless and mostly sunny days. These flies were not caught on dark, cloudy days or on days when the temperature was below 10°C. As for oestrids and tabanids, most culicids and simuliids also were caught during warm, sunny days, but primarily dur- ing the crepuscular period. However, culicids and simuliids also were caught in large numbers on the many cool, cloudy days when climatic conditions greatly reduced or prohibited the activity of oestrids and tabanids (Tables 3, 4). The culicid/simuliid numbers in Table 3 are smaller than in Table 4 because in 1985 we shut down the traps before the host-seeking activity reached its usual “crepuscular” peak. Anytime the temperature was above 7°C and wind velocities were less than 8 m/s, culicids and simuliids were caught in CO,-baited traps. From June through August, therefore, there were likely few days when culicids and simuliids did not attack Reindeer. Oestrid attacks, by contrast, were greatly limited by frequent adverse climatic conditions (e.g., darkly overcast, rainy, cold). During our stud- ies such conditions negated oestrid activity on 60% of the days in 1984, 56% of the days in 1985 and 68% of the days in 1987. During a 15-day trapping period near Prudhoe Bay in northern Alaska, Pollard et al. (1996) caught H. tarandi on 13 days and C. trompe on 8 days at one site, and on only 5 and 3 days, respectively, at another site. Although stationary traps caught few or no culi- cids and simuliids on windy days, a mobile tame Reindeer cow was observed being attacked by culi- cids when vidda wind velocities were 6-10 m/s (with intermittent gusts to 12 m/s). As she walked or trotted from one area to another while foraging, her presence in the new area activated culicids that flew up from the vegetation to attack her, and she intermittently responded with various defensive reactions associated with culicids. When caught and held, no culicids were attracted while the fast wind speeds persisted. This type of opportunistic ambush feeding by culicids in response to the pres- ence of a host that moves into an area where culi- cids are sheltering permits localized aggregations of culicids to attack Reindeer even on very windy days. The mild anti-culicid behaviors of foraging animals observed on windy day videotape sequen- ces were an indication that they were activating resting culicids as they paused to feed in different areas. 2001 Movements and reactions of reindeer to flies At study area “A”, we observed that, although Reindeer moved from vidda habitat into the woods at times when woods trap catches and the woodland swarming/landing index of culicids at humans were at their peaks, host reactions to flies were not suffi- cient to interfere with a slow, mobile foraging pace. These observations agree with those of Ardo (1958) who reported that in forested areas of northern Sweden, Reindeer exhibited no apparent avoidance reactions against mosquitoes, but that they tried to avoid tabanids. Because significantly. more tabanids were trapped in woodland habitat than on the vidda (Table 2), the daily movement of Reindeer in and out of woodland habitat at study area “A” may have been a response to avoid attacks by large numbers of tabanids. The 1984 mean of 41.5 tabanids/trap/day on the vidda is slightly greater than the 1987 mean of 36.5 tabanids/trap/day, but much less than the 1985 mean of 83.0 tabanids/trap/day (Anderson and Nilssen 1998). The only times during which we observed pan- icked, disrupted herd behavior, or observed animals concentrated on mountain tops were between 10:40 and 14:00 h on warm, sunny/partly sunny days. This was when large numbers of both species of oestrid parasites were trapped on such days (Tables 3, 4; Anderson and Nilssen 1996a). If culicids were caus- ing Reindeer to disperse to mountain tops this behav- ior should have occurred between 16:00 h and 20:00 h, when numbers of host-seeking culicids peaked on warm, sunny days. Although Reindeer on mountain tops may have been trying to avoid oestrids, it was not surprising that they were attacked here because both H. tarandi and C. trompe are strong fliers (Nilssen and Anderson 1995a). In fact, C. trompe mating sites were located on all the high mountain tops in study area “B” (Nilssen and Anderson 1995b), and males were present during periods when Reindeer were observed on mountain tops. Also, 72% of all H. tarandi females captured by Downes et al. (1986) were caught on upper slopes and ridgetops. Hughes et al. (1981) reported that horses standing on windy hilltops did not experience a reduction in the number of tabanids harassing them, but further studies are needed to compare the numbers of flies associated with Caribou/Reindeer on mountain tops versus vidda grazing areas. In Northern Norway female oestrids were the pri- mary cause of the panicked individual and herd behaviors of Reindeer (Anderson and Nilssen 1996a; current study). This conclusion agrees with that of Bergman (1917) in Sweden who noted that neither gnats (i.e., culicids, simuliids and ceratopogonids) nor Tabanidae were the cause of Reindeer becoming so excited on hot summer days and exhibiting the types of reactions and movements he described. He attributed the intense panicked behavioural respons- ANDERSON, NILSSEN, AND HEMMINGSEN: HOST-MIMICKING TRAP CATCHES 285 es of Reindeer as reactions to the oestrid parasites. More recently, Downes et al., (1986) and Mo6rschel and Klein (1997) also concluded that oestrids were the primary tormentors of Caribou. Under the climatic conditions encountered during our studies, and the number of flies caught in traps, we conclude that there were few days when attacks by parasitic flies were intense enough to cause annoyance levels sufficient to reduce overall diel for- aging times of Reindeer. As noted by Arnold and Dudzinski (1978) grazing animals usually compen- sate for reduced daytime grazing time by grazing dur- ing night time hours. Although large numbers of hematophagous flies such as simuliids, culicids and tabanids can reduce daytime grazing time of large vertebrates (e.g., Breev 1950; Fredeen 1969; Hughes et al. 1981; Downes et al. 1986; Ralley et al. 1993) the absence or great reductions of such parasitic flies during night time would provide the opportunity for many hours of compensatory grazing. Thus, even on the worst fly days for Reindeer, animals in our study areas had an overnight period of about 6—10 hours during which the small numbers of culicids active then would not interfere with foraging. Because of low temperature, and other climatic conditions unfa- vorable for flies, during most nights, and on many days, Reindeer in our study area experienced many hours of fly-free foraging time. On the Hardanger vidda in Norway, Thomson (1971) also observed that from 22:30—03:00 h in July (the month when harass- ment by flies was most serious), Reindeer herds spent 41% of the time grazing and 44% of the time lying After co-existing with large populations of many parasitic flies for about 2-3 months each year, ani- mals seen in late August/September were in robust, near prime condition (compared to their thin condi- tion in May/June). This indicates that, in our study area, such fly parasites had little or no negative impact on foraging time or weight gain. Acknowledgments We thank I. Folstad and J.O. Bustnes, University of Tromsg, Tromsg, Norway and A. Karter, University of California, Davis, California, U.S.A. for assisting with these studies, and Utviklingsfondet for Reindrift, Alta, Norway, for partial financial sup- port. We also thank J. F. Burger, University of New Hampshire, Durham, New Hampshire, USA, for identification of culicids, and J. A. Shemanchuk, Agriculture Canada, Lethbridge, Alberta, Canada, for providing the Breev translation. We thank A. Karter and J. Washburn for comments on the man- uscript. Literature Cited Anderson, J. R., and J. B. Hoy. 1972. Relationships between host attack rates and CO,-baited insect flight trap catches of certain Symphoromyia species. Journal of Medical Entomology 9: 373-393. 286 Anderson, J. R., and A. C. Nilssen. i990. The method by which Cephenemyia trompe (Modeer) larvae invade reindeer (Rangifer tarandus). Rangifer Special Issue 3: 291-297. Anderson, J. R., and A. C. Nilssen. 1996a. Trapping oestrid parasites of reindeer: the response of Cephenemyia trompe and Hypoderma tarandi to baited traps. Medical and Veterinary Entomology 10: 337-346. Anderson, J. R., and A. C. Nilssen. 1996b. Trapping oestrid parasites of reindeer: the relative age, fat body content and gonotrophic conditions of Cephenemyia trompe and Hypoderma tarandi females caught in baited insect flight traps. Medical and Veterinary Entomology 10: 347-353. Anderson, J. R., and A. C. Nilssen. 1998. Do reindeer aggregate on snow patches to reduce harassment by par- asitic flies or to thermoregulate? Rangifer 18: 3-17. Anderson, J. R., and W. Olkowski. 1968. Carbon dioxide as an attractant for host-seeking Cephenemyia females (Diptera: Oestridae). Nature 220: 190-191. Anderson, J. R., W. Olkowski, and J. B. Hoy. 1974. The response of tabanid species to CO,-baited insect flight traps in Northern California (Diptera:Tabanidae). Pan- Pacific Entomologist 50: 255-268. Anderson, J. R., and W. L. Yee. 1995. Trapping black flies (Diptera:Simuliidae) in Northern California. I. Species composition and seasonal abundance on horses, host models, and in insect flight traps. Journal of Vector Ecology 20: 7-25. Ardo, P. 1958. On the feeding habits of the Scandinavian mosquitoes. Opuscula entomologica 23: 171-191. Arnold, G. W., and M. L. Dudzinski. 1978. Ethology of free-ranging domestic animals. Elsevier Scientific Publishing Co., New York, New York. 198 pages. Bergman, A. M. 1917. Om renens oestrider (in Swedish; English translation). Entomologisk Tidoskrift 38: 1-32, 113-146. Breev, K. A. 1950. The behavior of blood-sucking Diptera and bot-flies when attacking reindeer and the responsive reactions of the reindeer. I. The behavior of blood-suck- ing Diptera and bot-flies when attacking reindeer. Parazitologiceskij Sbornik 12:167—198 (in Russian; English translation). Burger, J. F. 1995. Catalog of Tabanidae (Diptera) of North America North of Mexico. Contributions on Entomology, International 1: 1-100. Downes, C. M., J. B. Theberge and S. M. Smith. 1986. The influence of insects on the distribution, microhabitat choice, and behavior of the Burwash caribou herd. Canadian Journal of Zoology 64: 622-629. Espmark, Y. 1961. Behavior of reindeer attacked by rein- deer nostril and grub flies. [In Swedish]. Zoologisk Revy 23: 29-37. Espmark, Y. 1968. Observations of defence reactions to oestrid flies by semidomestic forest reindeer (Rangifer tarandus L.) in Swedish Lapland. Zoologische Beitrage 14: 155-167. Fredeen, F. J. H. 1969. Outbreaks of the black fly Simulium arcticum Malloch in Alberta. Quaestiones Entomologicae 5: 341-372. Gjullin, C. M., R. I. Sailer, A. Stone, and B. V. Travis. 1961. The mosquitoes of Alaska. United States Department of Agriculture, Agricultural Handbook Number 182, 98 pages. THE CANADIAN FIELD-NATURALIST Vol. 115 Helle, T., J. Aspi, K. Lempa, and E. Taskinen. 1992. Strategies to avoid biting flies by reindeer: field experi- ments with silhouette traps. Annales Zoologici Fennici 29: 69-74. Hughes, R. D., P. Duncan, and J. Dawson. 1981. Interactions between Camargue horses and horseflies (Diptera: Tabanidae). Bulletin of Entomological Research 71: 227-242. Low, W. A. 1976. Parasites of Woodland caribou in Tweedsmuir Provincial Park, British Columbia. Canadian Field-Naturalist 90: 189-191. Mason, P. G., and P. M. Kusters. 1990. Seasonal activity of female black flies (Diptera: Simuliidae) in pastures in northeastern Saskatchewan. Canadian Entomologist 122: 825-835. McCreadie, J. W., M. H. Colbo, and G. F. Bennett. 1985. The seasonal activity of hematophagous Diptera attack- ing cattle in insular Newfoundland. Canadian Entomologist 117: 995-1006. Meyer, R. P., W. K. Reisen, B. R. Hill, and V. M. Martinez. 1983. The “AFS sweeper”, a battery- powered backpack mechanical aspirator for collecting adult mosquitoes. Mosquito News 43: 346-350. Morschel, F. M., and D. R. Klein. 1997. Effects of weath- er and parasitic insects on behavior and group dynamics of caribou of the Delta Herd, Alaska. Canadian Journal of Zoology 75: 1659-1670. Nilssen, A. C., and J. R. Anderson. 1995a. Flight capacity of the reindeer warble fly, Hypoderma tarandi (L.), and the reindeer nose bot fly, Cephenemyia trompe (Modeer) (Diptera:Oestridae). Canadian Journal of Zoology 73: 1228-1238. Nilssen, A. C., and J. R. Anderson. 1995b. The mating sites of the reindeer nose bot fly: not a practical target for control. Rangifer 15: 55-61. Pollard, R. H., W. B. Ballard, L. E. Noel, and M. A. Cronin. 1996. Parasitic insect abundance and microcli- mate of gravel pads and tundra within the Prudhoe Bay Oil Field, Alaska, in relation to use by Caribou, Rangifer tarandus granti. Canadian Field-Naturalist 110: 649-658. Ralley, W. E., T. D. Galloway, and G. H. Crow. 1993. Individual and group behaviour of pastured cattle in response to attack by biting flies. Canadian Journal of Zoology 71: 725-734. Roberts, R. H. 1972. Relative attractiveness of CO, and a steer to Tabanidae, Culicidae and Stomoxys calcitrans (L.). Mosquito News 32: 208-211. Sommerman, K. M. 1958. Two new species of Alaskan Prosimulium with notes on closely related species. Pro- ceedings of the Entomological Society of Washington 60: 193-202. SYSTAT. 1992. Statistics, Version 5.2 Edition. Evanston, Illinois; SYSTAT, Incorporated. 724 Pages. © Thomson, B. R. 1971. Wild reindeer activity. Hardanger- vidda: July-December 1970. Statens Viltundersgkelser, Direktoratet for jakt, viltstell og ferskvannsfiske, Trond- heim, Norway, 1-83. Wood, D. M. 1985. Biting flies attacking man and live- stock in Canada. Agriculture Canada, Publication 1781/E: 38 Pages. Received 11 September 2000 Accepted 25 June 2001 Germination Potential, Updated Population Surveys and Floral, Seed and Seedling Morphology of Symphyotrichum laurentianum, the Gulf of St. Lawrence Aster, in the Prince Edward Island National Park SARAH E. STEWART and CHRISTIAN R. LACROIX! Department of Biology, University of Prince Edward Island, 550 University Avenue, Charlottetown, Prince Edward Island CIA 4P3 Canada 'Author to whom correspondence should be addressed. Stewart, Sarah E., and Christian R. Lacroix. 2001. Germination potential, updated population surveys and floral, seed and seedling morphology of Symphyotrichum laurentianum, the Gulf of St. Lawrence Aster, in the Prince Edward Island National Park. Canadian Field-Naturalist 115(2): 287-295. Symphyotrichum laurentianum, the Gulf of St. Lawrence Aster, is an endemic aster of Prince Edward Island, New Brunswick, and the Magdalene Islands. It is considered by the Committee on the Status of Endangered Wildlife in Canada (COSEWIC) to be of special concern in Canada and to be critically imperiled in Prince Edward Island. One goal of this study was to test germination to clarify the relationship between dormancy and germination in this species. Seeds germinat- ed in culture media had an overall mean percent germination of 68.37% whereas seeds germinated in soil had an overall mean percent germination of 17.38%. Kinetin supplements positively affected germination rates. However, these kinetin supplements eventually led to abnormal morphological development in S. Jaurentianum seedlings. Cold treatments had no significant effects on the percent germination of S. Jaurentianum seeds. Micropropagation of this type of explant and rein- troduction should therefore be considered by Prince Edward Island as potential techniques for the conservation of S. lau- rentianum on Prince Edward Island. A complementary goal of this study was to survey the Blooming Point and Covehead sites to update the population statuses of S. Jaurentianum within the Prince Edward Island National Park. The size of the S. laurentianum population at Covehead Pond during 1999 was very similar to the population size at this site in 1993. Population sizes showed more fluctuation at Blooming Point between 1993 and 1999. No asters were found in the East Marsh in 1999; however, there was a large increase in population numbers at the Dune Slack site. Seeds of S. laurentianum appear to respond well to tissue culture. Annual monitoring of S. /aurentianum populations should also occur to ensure appropriate management of this species on Prince Edward Island. Key Words: Symphyotrichum laurentianum, Gulf of St. Lawrence Aster, Aster laurentianus, seed germination, tissue culture, plant conservation, rare plant, kinetin, micropropagation, Prince Edward Island National Park. Symphyotrichum laurentianum, Gulf of St. Lawrence Aster, is an annual halophyte endemic to Prince Edward Island, New Brunswick and the Magdalene Islands. Growing in sandy, brackish soils in areas flooded by salt water, S. laurentianum is found in salt marshes, around ponds and within dune slacks along the Gulf of St. Lawrence (Fernald 1950; Houle 1988; Houle and Haber 1990; Guignion et al. 1995). The Gulf of St. Lawrence Aster was previously clas- sified as Aster laurentianus; however, this species was reclassified by Nesom in 1994. S. laurentianum is considered by COSEWIC (Committee on the Status of Endangered Wildlife in Canada) to be of special concern in Canada (Houle and Haber 1990; Guignion et al. 1995; Keith 1998; Shank 1999; Houle and Belleau 2000) and is considered to be critically imperiled within Prince Edward Island (Argus and Pryer 1990). In 1992 there were six known populations of S. laurentianum within the Prince Edward Island National Park (Guignion et al. 1995). These sites included: Covehead Pond, Long Pond, Campbell’s Pond, Blooming Point Western Wetland, Blooming Point East Marsh and Blooming Point Dune Slack. Population estimates for S. Jaurentianum at the Blooming Point Eastern Marsh in both 1992 and 1993 were 48 000- 60 000, while the Blooming Point Dune Slack site had an estimated population size of 15 000- 20 000. The four other populations of S. lau- rentianum within the Prince Edward Island National Park had much smaller population sizes than the populations occurring at Blooming Point. In 1993, there were 214 S. laurentianum plants in the Covehead Pond population, while only one plant was found at Long Pond. Three S$. laurentianum plants were found at Campbell’s Pond in 1993 while no plants were found at the Blooming Point Western Wetland, a site once reported to have Gulf of St. Lawrence Asters. Population sizes within some sites also greatly fluctuated between years (Guignion et ale 1995): One of the main factors affecting population size in vascular plants species, especially annuals, is the germination potential of the seeds they produce. Houle and Haber (1990) reported an 80% germina- tion for S. laurentianum seeds. A percent germina- tion of over 50 has also been reported for this species by Houle and Belleau (2000). It is well known that 287 288 before germination is triggered, seeds of many vas- cular plant species require a dormancy period. S. lau- rentianum seeds mature in October and do not ger- minate until the middle of the following June. This indicates that S. laurentianum seeds have some mechanism which either prevents germination or does not induce germination until that time. Cold treatments and treatments with growth hormones, such as gibberellic acid and kinetin, have been used to overcome dormancy periods in other vascular plant species (Baskin and Baskin 1998). The objectives of this study are: (1) to perform germination tests to clarify the relationship between dormancy and germination in S. laurentianum seeds by using cold treatments, and treatments with kinetin and gibberellic acid to trigger the highest germina- tion rates possible for this species and (2) to survey the Blooming Point and Covehead sites to update the population statuses of S. laurentianum within the Prince Edward Island National Park. Materials and Methods Population Studies Due to the low reported population numbers of S. laurentianum at Covehead Pond (46° 25’ 47” N; 63° 9’ 11” W) a total population count was conducted at this site. The height of each plant was measured as well as the distance of each plant from the pond water’s edge. A total count of the number of inflo- rescences per plant was also recorded. At Blooming Point, the Western Wetland (46°24’ 56” N; 62°58’44”"W; 46°24’56”N, 62°59’6”W), Dune Slack (46°24’52”N; 62°59’42”W) and Eastern Marsh sites coordinates were all surveyed for S. lauren- tianum plants. Due to the large numbers of individu- als in these populations, total population counts were not feasible. Instead, population estimates were car- ried out at these locations. Five 0.5 m? plots were ran- domly chosen in the Dune Slack site and in the Eastern Marsh. Population sizes were estimated using marsh areas calculated by a global positioning system (GPS) and the average number of plants found in the 0.5 m? plots. Longitudinal and latitudinal coordinates were also recorded using a GPS. Seed Collection and Treatment Symphyotrichum laurentianum seeds were ran- domly collected from Covehead Pond, Blooming Point Dune Slack and Blooming Point East Marsh sites in the Prince Edward Island National Park. Half of the seeds were randomly chosen to represent the control group which was kept at room temperature until germination tests were performed. The remain- ing half of the seeds were placed in a refrigerator maintained at 4°C until germination tests were per- formed. This time period ranged from one month to eight months depending on replication number. Seeds in Soil Fifty seeds from each of the two treatment groups THE CANADIAN FIELD-NATURALIST Vol. 115 were randomly selected and were planted in labelled pots filled with equal parts vermiculite, peat moss | and potting soil. A further 50 seeds from each of the | two treatment groups were randomly selected and | underwent a treatment in 30 ml of a 5mg/100ml gib- berellic acid solution for 25 hours. Pots were placed in a Conviron Model E 15 growth chamber set at | 25°C, 80% humidity and on a light/ dark schedule of 16/8 hours. Four trials of this procedure were repeat- ed. A modified version of this protocol was carried out two additional times to duplicate Houle and Belleau’s (2000) germination study. In this situation the pots were placed in a growth chamber set at 80% humidity, on a light/dark schedule of 14/10 hours and at light/dark temperatures of 19/13° C. Germin- ation was considered successful when seedlings emerged from the soil. Seeds Germinated in Petri Dish Fifty seeds were randomly chosen from the room temperature group and the cold treatment group. Seeds were placed in separate labelled petri plates, on two layers of filter paper and covered with dis- tilled water. Plates were placed in a growth chamber set at 80% humidity, on a light/dark schedule of 14/10 hours and at light/dark temperatures of 19/13° C. This procedure was repeated twice for a total of three trials. Successful germination was considered when the radicle emerged from the seed coat. Seeds in Growth Media Seeds from both treatment groups were randomly selected and were cultured on Murashige and Skoog Basal Salt Media (1962), at quarter strength, half strength and full strength concentrations, supple- mented with either no kinetin, 2 mg/l kinetin, or 4 mg/l kinetin. All media were also supplemented with standard basic additives: 0.02 g thiamine, 0.04 g myo- inositol, 30.0 g sucrose and 10 g gelrite. The growth medium was autoclaved for 30 minutes at 121° C and was poured into petri plates and sealed with Parafilm M to prevent contamination and dehy- dration. Prior to culturing, seeds were sterilized in a 5.25% sodium hypochlorite solution and Triton-X- 100 nonionic surfactant for 35 minutes. The seeds were then soaked in one wash of sterile distilled water for three minutes. Plates containing seeds were placed in a growth chamber at 25° C, 80% humidity and on a light/dark schedule of 16/8 hours. Germin- ation was considered successful when the radicle emerged from the seed. Morphological Analysis Seeds of S. laurentianum at various stages of dis- section were dehydrated in a graded ethanol series and critical point dried using CO, as a transitional fluid in a model 28000 LADD critical point dryer. Specimens were mounted on stubs, grounded with silver paint, then coated with 300 Angstroms of gold- palladium using a Denton Vacuum Desk II sputter- 2001 1 STEWART AND LACROIX: GERMINATION AND SURVEYS OF ST. LAWRENCE ASTER 289 AS . oe a. tre ag oe, , te Ae a | a WS GY 3 o FIGURES 1-2. Macroscopic view of Symphyotrichum laurentianum plant and inflorescence. (1) White capitulum (arrowhead) of Symphyotrichum laurentianum. (2) Multibranching specimen of Symphyotrichum laurentianum. Scale bars represent 5 mm (1) and 75 mm (2). coater. Specimens were examined with a Cambridge Stereoscan 604 scanning electron microscope, and thermal prints of the digital SEM images were acquired using SEMICAPS® software and produced using a Mitsubishi P67U video copy processor. An Olympus SZ40 microscope equipped with a Pixera Digital Camera system was used to photo- graph S. laurentianum plants grown in culture from germination to early stages of shoot growth. Statistical Analysis T-tests with Bonferonni correction were per- formed to compare mean percent germination between seeds in different treatment groups. Results Population Information A total of 243 S. laurentianum plants (Figure 2) were found at the Covehead Pond site in 1999. The height of S. laurentianum plants at this site ranged from 2.5 cm to 46.0 cm. The mean height of S. lau- rentianum plants at the site was 12.7 cm + 6.85 (n=243). The number of inflorescences per plant at the Covehead site ranged from 0 to 729 while the mean TABLE 1. Comparison of population sizes for Symphyo- trichum laurentianum in Prince Edward Island. Population 1992 1993 1999 Covehead Pond 168 214 243 Western Wetland 425 0 0 Dune Slack 15-20 000* 15-20000* 65 250* East Marsh 50-60 000* 50-60000* 25 000* *estimated. number of inflorescences per plant was 15.42. S. lau- rentianum plants at Covehead Pond were found any- where from 5 cm to 1010 cm from the edge of the pond. The mean distance between the pond’s water- line and S. laurentianum plants at Covehead was 644.02 cm + 311.16 (n=243). All of the aster inflores- cences at both the Covehead site and the Blooming Point sites were white (Figure 1). Refer to Table 1 for Blooming Point population size comparisons. Seed Germination in Soil S. laurentianum seeds planted in soil in the growth chamber had an overall mean percent germination of 17.38% + 9.51 (n=800). There were no significant differences between the mean percent germination of seeds in the control group and the mean percent ger- mination of seeds in any of the other three treatment groups (t-test; p> 0.05). Seeds treated with a cold treatment prior to planting had a significantly higher mean percent germination than seeds that underwent both the cold treatment and the treatment with gib- berellic acid (t-test; p< 0.05). Seeds that were left at room temperature and then were treated with gib- TABLE 2. Mean percent germination of Symphyotrichum laurentianum seeds planted in soil. Treatment Mean Percent Germination Control? 16.5 + 9.98 Cold Treatment>4 22.5+7.19 Control + GA, Treatment-4 20.0 + 12.44 Cold + GA, Treatment® 10.5+6.19 Note: Treatments with the same letter superscripts are sig- nificantly different from one another. 290 THE CANADIAN FIELD-NATURALIST berellic acid also had a significantly higher mean percent germination than seeds that were treated with both a cold treatment and a treatment with gib- berellic acid (t-test; p< 0.05)(Table 2). Seeds maintained in the growth chamber as speci- fied by Houle and Belleau (2000) had an overall mean percent germination of 36.0% +16.14 (n= 400). Seed germination in petri plates Seeds germinated in petri plates had an overall mean percent germination of 15.33% + 9.35% (n=300). There was no significant difference between the mean percent germination of seeds kept at room temperature and the mean percent germination of seeds subjected to a cold treatment (t-test; p> 0.05). Seed germination in culture The overall mean percent germination of all seeds in culture was 68.37% + 26.38% (n=821). The mean percent germination of seeds maintained at room temperature prior to culturing was 84.88% + 16.31% (n=369). The mean percent germination of seeds treated with a cold treatment prior to culturing was 51.85% + 24.21% (n=452). The difference between these two mean percent germinations was not statis- tically significant (t-test; p>0.05). Seeds grown in quarter strength basal salt media in both the room temperature and cold treatment groups had significantly higher mean percent germi- nations than seeds grown in full strength media (t- test; p<0.05). In the room temperature group, seeds grown in quarter strength media also had significant- ly higher mean percent germination than seeds grown in half strength media (t-test; p<0.05). Within the room temperature group in both full strength and half strength media, seeds grown in Vol. 115 media supplemented with 2 mg/l kinetin had signifi- cantly higher mean percent germination than seeds grown in media that was not supplemented with kinetin (t-test; p<0.05). Seeds grown in full strength media supplemented with 4 mg/l kinetin also had significantly higher mean percent germination than seeds grown in full strength media which did not contain a kinetin supplement (t-test; p<0.05). There were no statistically significant differences between the mean percent germination of any of the seeds grown in different media strengths or different kinetin strengths within the cold treatment group (t- test; p>0.05). See Table 3 for a summary of the tis- sue culture results. Morphology Seeds: Symphyotrichum laurentianum achenes (Figure 4) have a mean length of 2.09 mm + 0.15 (m =30). Mature embryos are surrounded by a clear seed coat, and the seeds are encased in a hard fruit wall. The achenes are attached to a pappus com- posed of white barbed bristles (Figure 3) which have a mean length of 4.93 mm + 0.57 (n=30). Two long, flat and closely appressed cotyledons cover the shoot tip of the seed embryo which is oriented towards the pappus. A shoot meristem is visible between the two cotyledons in the mature seed but no leaf primordia are visible (Figure 8). Florets: All inflorescences consist entirely of disc florets (Figure 5). Each disc floret is surrounded by a pappus which is longer than the floret. Mean floret length is 3.74 mm + 0.22 (n=30). The corolla of tube florets is five lobed and shorter than the style (Figure 6). Two stigmas containing pollen collecting papillae extend above the corolla of each tube floret. Anthers TABLE 3. Mean percent germination of Symphyotrichum laurentianum seeds in culture. Treatment Mean Media Group Strength Control? 84.9 + 16.3 Full MS‘ 1/2 MS4 1/4 MSo4 Cold» Silas ax VAL Full MS* 1/2 MS* 1/4 MS* Mean Kinetin Mean Supplement 74.6 + 22.3 no kinetin?" 50.7 + 18.8 2 mg/l kinetin® 83.1 + 15.0 4 mg/l kinetin” 90.0 + 10.0 85.6 + 10.2 no kinetin! TiS 9m 2 mg/I kinetini 93.3 + 11.5 4 mg/l kinetin! 86.0 + 175 94.4 + 6.2 no kinetin* O5 2a 2 mg/l kinetin! 96.4 + 3.2 4 mg/l kinetin™ ONG 6 SONS 25,5 no kinetin” 38.0) ilses 2 mg/l kinetin°® SIRGEEIS Ss 4 mg/l kinetin? 49.7+ 44.1 51.4 + 27.6 no kinetin 55.2) 130 2 mg/I kinetin" 41.3 + 44.4 4 mg/l kinetin‘® Wess 7/58) 64.4 + 12.3 no kinetin 58.8 + 7.0 2 mg/l kinetin" FO Ord, 4 mg/I kinetin’ 64.3 + 19.7 Note: Treatments with the same letter superscripts are significantly different from one another. STEWART AND LACROIX: Ficures 3-8. Floret and fruit of Symphyotrichum laurentianum. (3) Symphyotrichum laurentianum fruit showing achene (A), pappus (P) and wilted remains of disc floret (D). (4) Achene showing pubescent fruit coat and barbed pappus (arrowhead). (5) Stigmas (arrowheads) protruding from the disc floret (D). (6) Five continuous petals (arrowheads) make up the corolla of the disc floret. (7) Pollen with apertures (arrowhead) caught in the papillae of the stigma. (8) Embryo with one cotyledon removed revealing the apical meristem (arrowhead). C= cotyledon. Scale bars represent 750 um (3), 300 um (4), 75 um (5), 75 um (6), 15 um (7), 150 um (8). 292, are connate and introrse and surround the stigmas inside of the corolla tube. Pollen: Symphyotrichum laurentianum pollen grains are circular, with trizonoporate apertures and echinae sculpturing, which is typical of many members of the Asteraceae (Figure 7). Seedling: The reddish-brown fruit walls of Symphyotrichum laurentianum seeds become clear as seeds imbibe water. Cotyledons become greener and can be seen through the translucent fruit wall (Figure 9). These features can be observed less than 24 hours after seeds are placed in media. Following these initial stages, the hypocotyl of the embryo axis extends through the achene and cotyledons begin to separate causing the fruit wall to split and be shed (Figure 10). This stage takes place less than 48 hours after seeds are introduced in the media. During the following 24 to 72 hours, cotyledons separate further and the hypocotyl continues to elon- gate. After four to six days in media, root elongation takes place followed by the formation of many root hairs (Figure 12). As early as five days in media or as late as ten days, the first true foliage leaves are visible in some seedlings. The second true foliage leaf was not observed until after seedlings had been in media for at least 12 days (Figure 14). Seedling Abnormalities: Root elongation was sup- pressed in all seedlings grown in media which con- tained a kinetin supplement. However, root hairs did form on the radicle of these seedlings (Figure 11). Stem elongation was also suppressed in all seedlings grown in kinetin supplemented media (Figure 13). Discussion Population Information Populations of S. laurentianum within the Prince Edward Island National Park have fluctuated and changed significantly over the last eight to ten years. Covehead: The S. laurentianum population at Covehead Pond has been slightly increasing over the past ten years. This site consisted of 168 Sym- phyotrichum laurentianum plants in 1992, 214 plants in 1993 (Jacques Whitford 1994), and 243 plants in 1999. The Symphyotrichum laurentianum population at Covehead occupies an area of approximately 320 m’. The average density of Symphyotrichum lauren- tianum plants along the north and east sides of the pond is approximately 0.759 plants per m*. All of these $. laurentianum plants grow in an area fre- quently flooded by salt water. Although this S. laurentianum population is increasing, it is still well below the population size of 500 recommended by Given (1994) as stable and self sustaining for vascular plants species. Annual moni- toring of this site is recommended to ensure that the Covehead population of S. Jaurentianum is not lost. THE CANADIAN FIELD-NATURALIST Vol. 115 Blooming Point: The population of S. laurentianum at the Blooming Point Western Wetland was lost in 1993 due to drier conditions in this marsh (Jacques Whitford 1994). Further evidence that the marshes in the Blooming Point area are becoming drier is the disappearance of a stream which ran through the East Marsh in 1992 and 1993. A decrease in habitat size at the East Marsh has also occurred over the past few years. The original six pockets of S. lauren- tianum at this site in 1993 were reduced to two pock- ets in 1999. Drought stress has been shown to be a limiting factor for this species (Houle and Belleau 2000). Drier conditions in this marsh may be con- tributing to the decline of suitable habitats for S. lau- rentianum at Blooming Point. The estimated population size of 65 250 asters at the Blooming Point Dune Slack site in 1999 is much larger than the estimated population size of 15 000 - 20 000 plants at this site in 1993. The S. lauren- tianum population at this site is extremely dense with an average of 160 plants per square meter. This is a much higher density than that of the Covehead popu- lation. However the aster plants at Blooming Point were much smaller and had fewer branches than the plants at Covehead. The S. laurentianum populations at Blooming Point are located along a peninsula which is relatively secluded from public areas. Disturbance by humans is not believed to be a factor that may contribute to the decline of populations at this location. However, human activities in and surrounding Tracadie Harbour may have the potential to directly affect the populations of S. Jaurentianum along the Blooming Point peninsula especially if future developments in the Tracadie Harbour area are considered. Blooming Point is an extremely important site for S. lauren- tianum on Prince Edward Island. It is critical for the perpetuation of this species on Prince Edward Island that these populations remain viable. The Blooming Point peninsula must continue to be monitored and protected to ensure that these marshes remain suitable habitats for this vulnerable species. Based on the population information available to date, it is recommended that Symphyotrichum lau- rentianum remain classified as critically imperiled on Prince Edward Island. Only three populations of Symphyotrichum laurentianum have been document- ed on P.E.I. in recent years which clearly places the Gulf of St. Lawrence Aster within this risk category. Seed germination The overall germination test results from soil and culture media trials reported in this study are lower than those reported in other studies. For example, Houle (1988) reported an 80% mean percent germina- tion in S$. laurentianum seeds and Houle and Belleau (2000) were able to achieve a mean percent germina- tion of over 50%. Houle (1988) does not include the parameters surrounding the seed germination compo- STEWART AND LACROIX: GERMINATION AND SURVEYS OF ST. LAWRENCE ASTER Ficures 9-14. Developmental sequence of Symphyotrichum laurentianum seedlings grown in MS culture media. (9) Seed germinating in culture medium. Note the seed (arrowhead) inside the transparent, pubescent fruit coat (arrowhead). (10) Seedling after six days in culture showing the cotyledons beginning to separate. R= radicle. (11) Seedling after five days in culture medium containing a kinetin supplement. Root hairs are present on the radicle (R) but root elongation is inhibited. (12) Seedling after six days in culture media which did not contain a kinetin supple- ment. Root elongation is not inhibited (arrowheads). (13) Seedling after 74 days in culture medi- um containing a kinetin supplement. Stem elongation is inhibited and leaves are consequently crowded. (14) Seedling after 51 days in media which did not contain a kinetin supplement. Stem elongation is not inhibited in this seedling. Scale bars represent 0.5 mm (9), 0.7 mm (10, 11), 0.9 mm (12), and 10 mm (13, 14). 294 nent of her study making it difficult to duplicate her seed germination experiments. While all of the seeds in our study were collected from Prince Edward Island, seeds in Houle’s (1988) study and Houle and Belleau’s (2000) study were collected from the Magdalene Islands. It is possible that these two popu- lations have different reproductive capacities. Seeds grown in tissue culture had a significantly higher mean percent germination than seeds grown in soil. In tissue culture conditions seeds are com- pletely surrounded by growth media which provides a higher surface area for osmosis and a balanced abundance of essential minerals and nutrients. The higher mean percent germination of seeds in quarter and half strength MS media as opposed to mean per- cent germination of seeds in full strength MS media may also be linked to the role of osmosis in seed ger- mination. Half strength and quarter strength media have a lower solute/solvent ratio than full strength media. A low solute/solvent ratio in the media leads to a lower water potential in the seed than in the sur- rounding media, causing water to flow into the seed and allowing the process of imbibition to take place. For future micropropagation projects involving S. laurentianum it 1s recommended that half or quarter strength MS media be used. Seeds grown in MS media supplemented with kinetin had higher percent germination than seeds grown in media without kinetin. Kinetin has been used as an agent to break dormancy in seeds of many plant species (Phillips 1971; Webb et al. 1973; Bewley and Black 1985). Supplements of 2 mg/l and 4 mg/l of kinetin were used in this study to induce shoot formation. It was believed that the high con- centrations of the exogenous kinetin supplements in combination with the generally low endogenous lev- els of auxin present in the seed would induce the ini- tiation of shoot meristems. However, both shoot and root elongation were inhibited in A. /aurentianus seedlings growing in media containing kinetin sup- plements. High levels of plant growth hormones have been known to inhibit growth altogether in other species (Yeoman 1973). While kinetin levels of 2 mg/l and 4 mg/l were sufficient to induce shoot elongation in other species (Amo-Marco and Ibafiez 1998), it is possible that these levels are too high for Symphyotrichum laurentianum. Supplements of plant growth hormones have also produced abnormal morphological development in other plant species (Amo-Marco and Ibafiez 1998). Due to kinetin’s positive effect in increasing mean percent germina- tion in S. laurentianum it is recommended that kinetin be initially present in media to increase ger- mination rates in this species. However, immediately following germination, seeds should be transferred to a kinetin free medium to ensure that root and shoot elongation are uninhibited. Symphyotrichum laurentianum seeds do not THE CANADIAN FIELD-NATURALIST Vol. 115 appear to benefit from cold treatments or treatments with gibberellic acid. In natural conditions S. lauren- tianum seeds do not germinate until late June, so per- haps it is the warming conditions in the spring which induce germination in this species. We would there- fore recommend that future studies include a heat treatment in any germination tests involving the Gulf of St. Lawrence Aster. Tissue culture is a simple, inexpensive and poten- tially useful tool to mass propagate S. laurentianum from seed. Although other vegetative tissue sources may also be explored, seeds are the recommended source for germination programs as they carry the maximum genetic diversity to offspring (Fay 1991). Plants grown using micropropagation techniques could be reintroduced into wild populations in an attempt to stabilize vulnerable populations. Micro- propagation and reintroduction programs have been used in past projects involving plant conservation (Nandi et al. 1999; Nadeem et al. 2000). Micropropa- gation of Senecio hadrosomus, Atractylis arbuscula var. shizogynophylla, Artemisia granatensis and Psi- ada coronopus, all rare members of the Asteraceae, has been successful in several conservation projects (Fay 1991). Based on these conservation projects and our preliminary tissue culture data, there is potential that Symphyotrichum laurentianum may also be suc- cessfully conserved through a micropropagation and reintroduction program. Acknowledgments This research was funded by a University of Prince Edward Island Senate Research Grant. We would like to extend our thanks to Marina Silva for her advice on statistical analyses. A special thank you as well to Marina Silva and to Sara Purcell- MacDonald for their comments and suggestions on the manuscript. Documents Cited (marked * in text) Jacques Whitford Environmental Limited. 1994. Distri- bution and abundance of the Gulf of St. Lawrence Aster (Aster laurentianus) in Prince Edward Island National Park. Project Number 80077. Keith, T. 1998. Ecosystem Conservation Plan: Prince Edward Island National Park. Literature Cited Amo-Marco, J. B., and M. R. Ibafiez. 1998. Micropropa- gation of Limonium cavanillesii Erben, a threatened stat- ice, from inflorescence stems. Plant Growth Regulation 24: 49-54. Argus, G. W., and K. M. Pryer. 1990. Rare vascular plants in Canada: Our natural heritage. Canadian Museum of Nature, Ottawa. Baskin, C. C., and J. M. Baskin. 1998. Seeds: Ecology, biogeography, and evolution of dormancy and germina- tion, Academic Press, San Diego. Bewley, J. D., and M. Black. 1985. Seeds: Physiology of development and germination, Plenum Press, New York. 2001 Fay, M. F. 1991. Conservation of rare and endangered plants using in vitro methods. In Vitro Cellular Devel- opmental Biology 28: 1+. Fernald, M. L. 1950. Gray’s manual of botany, 8" edition. D. Van Norstrand, New York. Given, D. R. 1994. Principles and practises of plant con- servation. Timber Press, Oregon. Guignion, M., C. Ristau, and D. Lemon. 1995. The dis- tribution and abundance of the Gulf of St. Lawrence Aster, Aster laurentianus, in Prince Edward Island National Park. Canadian Field-Naturalist 109: 462-464. Houle, F. 1988. Etude biosystématique de la section Conyzopsis du genre Aster (Asteraceae). Thése de doc- torat. Départment de sciences biologiques, Université de Montréal. Houle, F., and E. Haber. 1990. Status of the Gulf of St. Lawrence Aster, Aster laurentianus (Asteraceae), in Canada. Canadian Field-Naturalist 104: 455-459. Houle, G., and A. Belleau. 2000. The effects of drought and waterlogging conditions on the performance of an endemic annual plant, Aster laurentianus. Canadian Journal of Botany 78: 40-46. Murashige, T., and F. Skoog. 1962. A revised medium for rapid growth and bio assays with tobacco tissue cul- tures. Physiologia Plantarum 15: 473-497. STEWART AND LACROIX: GERMINATION AND SURVEYS OF ST. LAWRENCE ASTER 295 Nandi, S. K., A. Kumar, and L. M. S. Palni. 1999. Role of plant tissue culture in biodiversity conservation and economic development. Current Science 77: 1229-1231. Nadeem, M., L. M. S. Palni, A. N. Purohit, H. Pandey and S. K. Nandi. 2000. Propagation and conservation of pdpphyllum hexandrum Royale: an important medici- nal herb. Biological Conservation 92: 121-129. Nesom, G. L. 1994. Review of the taxonomy of Aster sensu lato (Asteraceae: Astereae), emphasizing the new world species. Phytologia 77: 141-297. Phillips, I. D. J. 1971. Introduction to the biochemistry and physiology of plant growth hormones. McGraw- Hill, New York. Shank, C. C. 1999. The committee on the status of endan- gered wildlife in Canada (COSEWIC): A 21-year retro- spective. Canadian Field-Naturalist 113: 318-341. Webb, D. P., J. Van Staden, and P. F. Wareing. 1973. Seed dormancy in Acer. Journal of Experimental Botany 24: 105-116. Yeoman, M. M. 1973. Tissue (callus) cultures- tech- niques. /n Plant tissue and cell culture. Edited by H. E. Street, University of California Press, California. Received 27 September 2000 Accepted 6 July 2001 Mule, Odocoileus hemionus, and White-tailed, O. virginianus, Deer in the Yukon MANEFRED HOEFS Yukon Fish and Wildlife Branch, Box 2703, Whitehorse, Yukon Y1A 2C6 Canada Hoefs, Manfred. 2001. Mule, Odocoileus hemionus, and White-tailed, O. virginianus, deer in the Yukon. Canadian Field- Naturalist 115(2): 296-300. Both Mule Deer (Odocoileus hemionus) and White-tailed Deer (Odocoileus virginianus) have colonized the southern half of the Yukon in recent decades. Mule Deer have attained a continuous distribution in suitable habitats, White-tailed Deer have remained rare. Deer habitats are largely open, south-facing grassy slopes bordered by aspen, sites of recent forest fires, and cultivated hay fields. Many sightings have been reported by the general public. These have been supplemented by interviews of native elders and other long-time residents and a literature search. Mule Deer first appeared in the Yukon in the late 1930s to early 1940s and by the 1980s had reached the latitude of Dawson (64° N) and crossed into Alaska in the Ladue River drainage. A northern record for Mule Deer was established with a sighting near Chapman Lake along the Dempster Highway (64° 50’ N, 138° 25” W). White-tailed Deer are more recent, first observed near the British Columbia border (60° 10’ N) at Tagish Lake in 1975 and reaching Moose Creek along the Klondike Highway (63° 30’ N) in 1998. Key Words: Mule Deer, Odocoileus hemionus, White-tailed Deer, Odocoileus virginianus, Yukon, range extension, traffic mortalities. Deer are a new addition to Yukon’s large mammal fauna. They have invaded and colonized much of the southern half of this Territory over the past 50 to 60 years. Both Mule Deer (Odocoileus hemionus) and White-tailed Deer (Odocoileus virginianus) are pre- sent, but Mule Deer are more common. Little atten- tion has been paid to their presence within the Yukon nor outside. Recent scientific publications about deer (Whitaker 1980; Hesselton and Hesselton 1982; Mackie et al. 1982) do not include the Yukon as part of their range, and management efforts by govern- ment have been “low key”. The Yukon Territory has a small population and limited resources and has had to focus its efforts on economically important species, while deer are relatively rare and protected from hunting. The information collected about deer, there- fore, consists largely of voluntary contributions, either by government staff collecting data concurrent- ly with other field studies, or by members of the gen- eral public reporting deer sightings. Methods Information about deer in the Yukon was collected by: (1) ongoing cataloguing and mapping of deer sightings, (2) collecting records of road mortalities, (3) questionnaires and interviews, and (4) literature search. In the 1960s the newly established wildlife management agency began to record deer sightings made by its staff or reported by the general public. Independently, Conservation Officers, RCMP Officers, and highway maintenance staff recorded deer mortalities resulting from collisions with vehi- cles along Yukon’s major highways. Incidental sight- ings as a data source was improved over the years by incorporating questionnaires and public education, and by recording interviews with native elders and other long-term residents. Reports by biologists, geologists, and fur traders, as well as books published by big game hunters going back to the turn of the century were searched for references to deer. To veri- fy the presence of White-tailed Deer in addition to the more common Mule Deer, tissue samples of the former were submitted to the Forensic Laboratory, Alberta Natural Resources Service, Edmonton, for species identification by DNA analyses. Results Over the past 40 years, 391 records were filed by the Yukon Fish and Wildlife Branch reflecting 1101 deer sightings. About 80% of these were observa- tions of Mule Deer (Figure 1), 19 were documenta- tions of White-tailed Deer (Figure 2), and the remainder were non-specific. Most observations (N = 737) did not attempt a break-down into sex and age classes; those that did (N = 373) revealed the fol- lowing population structure: 105 bucks, 190 does, and 69 fawns. Single deer were most frequently observed (N = 136), followed by pairs (N=51). A negative linear correlation followed (R = -0.722) between increasing group size and decreasing fre- quency of observation, ranging from 35 observations of groups with three deer to two observation of groups with over 20 deer. The numbers of observa- tions reported fluctuated greatly between years. However, if they are lumped by five-year intervals or by decades a steady increase was evident: from only one observation prior to 1960 to over 500 during the most recent decade (Figure 3). 296 2001 Ze HokgFs: MULE AND WHITE-TAILED DEER IN YUKON 297 FiGuRE 1. Mule Deer buck along the Takhini Hotspring Road west of Whitehorse. This is a high-density deer area, presumably because of the local agricultural activity. The number of deer killed in collisions with vehi- cles showed a parallel trend. Only two mortalities were recorded prior to 1975, while 54 deer were killed over the past five years (1995-1999) (Figure 4). Deer were observed during every month, but most often during September. This was also the month when most were killed in accidents (Table 1). Interviews with old-timers corroborate this documen- FiGuRE 2. Skull of a male White-tailed Deer found north of Ross River by Ron Etzel in February 1997. tation. The late Johnny Johns (Carcross), native elder and big game guide, remembered seeing his first deer near Judas Creek in the early 1940s. A. VanBibber (Champagne) reported several sightings in the Tak- hini River area west of Whitehorse in the late 1940s and early 1950s. Clyde Blackjack (Carmacks) saw his first deer along the Yukon River between Lake Laberge and Big Salmon in summer 1945. The late Phil Temple (Kluane) remembered observing a deer at the Junction of the Little River with the Takhini River in 1953. Dan Nowlan (Watson Lake) reported a deer during a hunting trip 60 miles up the Liard River 600 NUMBER OF OBSERVATIONS wo 8 8 50-59 60-69 70-79 80-89 $0-S9 TIME PERIODS IN DECADES FiGurE 3. Number of deer sightings reported to the Yukon Wildlife Branch since record keeping began in the 1950s. 298 NUMBER OF DEER 20 70-74 75-79 80-84 85-89 90-94 95-99 5-YEAR INTERVALS FIGURE 4. Number of deer killed in collisions with vehicles along Yukon’s major highways since record keep- ing began in the early 1970s. in the early 1950s. Nowlan was accompanied by the late George Dalziel on this trip, who apparently saw three deer on this occasion. The first published reference to deer is found in Rand (1944), who reported second-hand information from the following localities along the newly con- structed South Alaska Highway: Bucking Horse River, Steamboat Mountain, Summit Pass, Lower Liard Crossing, and Coal River. The latter two loca- tions were close to the Yukon border. Clarke (1946*) assumed that deer had crossed into the Yukon in these areas. Clarke (1946*) also reported deer within the Yukon from the Nisutlin River and the Teslin area. Youngman (1975) summarized anecdotal infor- mation about deer to the early 1970s and listed records from Ross River, Carmacks and the Takhini River west of Whitehorse. Discussion Mule Deer have colonized most suitable habitats in the southern Yukon, with the latitude of Dawson (64°) constituting the northern limit of their current distribu- tion (Figure 5). The notable exception is the record of a Mule Deer doe and her fawn along the Dempster Highway near Chapman Lake (64° 50’N, 138° 25” W) by M. Villeneuve in summer 1998. This is the northernmost record for this species. To the west, Mule Deer have crossed the border into Alaska, most likely following the valley of the Ladue River upstream toward Tok during the 1980s. It is assumed THE CANADIAN FIELD-NATURALIST Vol. 115 that they first entered the Yukon from the Liard River drainage of northern British Columbia in the late 1930s to early 1940s. White-tailed Deer were first observed in 1975 by W. Ward at Tagish Lake, only a few miles north of the British Columbia border. The northern-most sighting, made in 1998, comes from Moose Creek near Stewart Crossing. Using straight-line distances their rate of colonization averaged 25 km/year, exceeding that of Mule Deer with 17 km/year. Most deer were observed in open areas such as south-facing grassy slopes with thin snow cover, sites with early successional vegetation such as recently burnt forests, agricultural fields seeded with forage crops, and re-vegetated road allowances. Numbers of deer observed (Figure 3) and numbers of deer lost in traffic accidents (Figure 4) have in- creased since record taking began in the 1960s and 1970s. However, this upward trend only in part reflects a growing population and an expanding dis- tribution; it also shows increasing interest by the public and a greater willingness to report sightings, as well as greater effort by government to collect this information. While the distribution map of deer (Figure 5) is fairly reliable, it overestimates the area actually inhabited by them, since it is based on game man- agement zones in which deer have been reported and not on point locations. A reliable estimation of num- bers was not possible with the information at hand, and it appears that these also fluctuate greatly between years, apparently reflecting winter severity. An educated guess places their number at 500 to 800, of which fewer than 100 may be White-tailed Deer (Hoefs 2000*). The population structure of 105 d: 190 ©: 69 fawns is within the reported range, but it is some- what distorted. For an non-hunted population, the number of bucks is low compared to number of does. Also the ratio of fawns to does is low, considering that twinning is frequent in deer. However, not too much reliability can be placed on these numbers, which are averages from 30 years of records and may not reflect current population dynamics. The annual periodicity of deer observations and traffic mortalities, which are significantly correlated (R= 0.792), are assumed to reflect deer activity as well as demographics (Table 1). In late winter obser- vations were few and traffic losses low. Deer move- Table 1. Annual peridicity of deer observations and traffic mortalities. Jan. Feb. March April May June July Aug. Sept. Oct. Nov. Dec Total Deer Observed N_ 17 12 11 De, 19 20 34 17 64 26 18 18 278 To!) OANA 4.0 Tes) 6:8"), SEZ le) ea ost 23,0) OPA Ne G2 ess) 100.0 Deer Killed Nn 1 6 4 4 9 7 D3) 19 11 iti! 98 oe 1.0 6.1 4.1 PIS SiS ee 100.0 2001 HogFS: MULE AND WHITE-TAILED DEER IN YUKON MN Carcross 7% Y Number of Mule Deer Reports by Game Mangement Subzone 1-3 Wie >? Number of White-tailed Deer Reports 1 3 4 Figure 5. Map showing the distribution of deer sightings. Reported sightings were allocated to game management zones, since many of them were not detailed enough for a more accurate recording of location. ment is restricted by thick snow, and numbers are then lowest in the annual cycle of births and mortali- ties. Snow disappears in April and May allowing deer to move about in search for preferred forages and observations and losses thus increase. Most fawns are born in June, but because of the deer’s habit of hiding them when they are young, they are not visible until July when they start following their mothers. The annual peak in observations and mor- talities is reached in the pre-rut and rutting period (September—November) when deer are very mobile as well as less cautious of humans and traffic. The range extensions observed in the Yukon have also been reported from neighboring jurisdictions. Mule Deer have been sighted in the Tok area of Alaska since the mid 1980s. Since the deer indige- nous to that State are Sitka or Coastal Black-tailed Deer (Odocoileus hemionus sitkensis), Mule Deer must have moved into Alaska from the Yukon. The first observations were made near Tok, and the sight- ing farthest north and west was made in March 1989 near Salcha (Bostian 1989). This location is about 60 km east of Fairbanks along the Tanana River and about 300 km northwest of the Yukon border. Based on the chronology of observations it is assumed that deer crossed into Alaska in the late 1970s to early 1980s following the valley of the Ladue River upstream (Hoefs 2000*). Independent of Mule Deer the Black-tailed Deer have also been observed north of their traditional range in Alaska (Roberson 1986), but these may be temporary dispersals and not per- manent range extensions. In the Northwest Territories both Mule Deer and White-tailed Deer are moving north, but here it is the White-tailed Deer that has made most progress. The first sighting of a White-tailed Deer was reported by Kuyt (1966) from the Fort Smith area, while a Mule Deer had already been reported for the vicinity of Fort Simpson 30 years earlier (Bethune 1937). Scotter (1974) lists a number of observations of both species made between 1955 and 1971 along the South Nahanni River and adjacent areas, which have since become part of Nahanni National Park. Veitch (2001) summarized White-tailed Deer observations from the southwestern N.W.T and neighboring regions and described a record of a White-tailed doe taken by native residents on the MacKenzie River about 100 km south of the Arctic Circle. This repre- sented the northernmost occurrence for this species in North America. 300 This northward spread of deer in the Yukon and N.W.T. is a continuation of the range extensions reported for British Columbia (Cowan and Guiguet 1965) and the Prairies (Wishart 1984; Gainer 1995). A number of factors have been suggested as possible causes: warming trend in climate, northward spread of agriculture, control of predators, lack of competi- tion from other ungulates, and protection (Wishart 1984; Veitch 2001; Hoefs 2000*). It is my opinion that this range extension is a pre- dictable and expected post-glacial phenomenon with the factors listed above only serving to accel- erate this trend. Much of the Yukon was covered by ice during the Wisconsin glaciation, which lasted until about 12 000 years ago (Valdez 1982). Many of the large mammals indigenous to the Yukon (caribou, thinhorn sheep, grizzly, wolf) had sur- vived the glacial periods in the Beringia refugium, which extended from central Yukon across Alaska into Siberia. These species colonized the glaciated portions of the Yukon relative quickly, because they were already in the region and simply had to invade newly developing habitat in the wake of retreating glaciers. Deer, on the other hand, along with other large mammals such as cougar, elk and coyote, survived the glacial periods south of the continental ice sheets in what is now the United States. They had to colonize all of eastern British Columbia and western Alberta first, before reaching the Yukon border. The deer’s colonization of the Yukon is a post- glacial trend also seen in other mammals, birds and plants that reflects the youthful status of this Territory in the context of faunal and floral evolution (Hoefs 2000*). A balance between current species assemblages with environmental conditions as seen in more southern regions has not yet been reached. No doubt warmer winters in recent decades, wide- spread wolf controls in the 1950s to 1970s, and the status of deer as an endangered species under the Yukon Act, protecting them from subsistence hunt- ing, has accelerated the deer’s range expansion across the Yukon. Acknowledgments Thanks are due to many members of the public, conservation officers, biologists, and staff of highway maintenance camps, who have contributed most of the sight records and information on deer mortalities in traffic collisions, on which this report is based. Marcus Waterreus drafted the figures and Stuart Alexander applied his GIS expertise to generate the map. To all these contributors I extend my most sin- cere gratitude. THE CANADIAN FIELD-NATURALIST Vol. 115 Documents Cited (marked * in text) | Clarke, C.H.D. 1946. Biological reconnaissance of lands adjacent to the Alaska Highway in Northern British Columbia and the Yukon Territory. Canadian Wildlife Service, Unpublished Report. 41 pages. Hoefs, M. 2000. Deer in the Yukon — Status Report with Management Recommendations. Yukon Fish and Wild- life Branch, Unpublished Report. 40 pages. Hoefs, M., Editor. 2000. The Status of Yukon’s Wildlife. Yukon Fish and Wildlife Branch, Unpublished Report. 46 pages. Literature Cited Bethune, W. C. 1937. Canada’s western Northland: its his- tory, resources, population, and administration. Depart- ment of Mines and Resources, Ottawa. 162 pages. Bostian, K. 1989. Deer spotted in Salcha area. Fairbanks Daily News Miner, March 15/89. Cowan, I. McT., and C. J. Guiguet. 1965. The Mammals of British Columbia. British Columbia Provincial Museum, Handbook Number 11. 414 pages. Gainer, R. 1995. Range extension of White-tailed Deer. Alberta Naturalist 25: 34-36. Hesselton, W. T., and R. M. Hesselton. 1982. White-tailed Deer. Pages 878-901 in Wild Mammals of North Amer- ica. Edited by J. A. Chapman and G. A. Feldhammer. John Hopkins Press, Baltimore, Maryland. 1147 pages. Mackie, R. J., K.L. Hamlin, and D. F. Pac. 1982. Mule Deer. Pages 862-877 in Wild Mammals of North Amer- ica. Edited by J. A. Chapman and G. A. Feldhammer. John Hopkins Press, Baltimore, Maryland. 1147 pages. Kuyt, E. 1966. White-tailed deer near Fort Smith, N.W.T. The Bluejay 24:194. Rand, A.L. 1944. The southern half of the Alaska Highway and its mammals. National Museum of Canada Bulletin 98. 50 pages. Roberson, K. 1986. Range extension of the Sitka Black- tailed deer (Odocoileus hemionus sitkensis) in Alaska. Canadian Field-Naturalist 100: 563-565. Scotter, G. 1974. White-tailed deer and mule deer obser- vations in southwestern district of MacKenzie, North- west Territories. Canadian Field-Naturalist 88: 487-489. Valdez, R. 1982. The wild sheep of the world. Wild Sheep and Goat International, Mesilla, New Mexico. 186 pages. Veitch, A.M. 2001. An unusual record of a White-tailed Deer (Odocoileus virginianus) in the Northwest Terri- tories. Canadian Field-Naturalist 15: 172-175. Wishart, W.D. 1984. Western Canada. Pages 475-486 in White-tailed Deer: Ecology and Management. Edited by L.K. Halls, Wildlife Management Institute. Stackpole Books, Harrisburg, Pennsylvania. 639 pages. Whitaker, J.O., Jr. 1980. The Audubon Society field guide to North American mammals. A. Knopf, New York. 745 pages. Youngman, P.M. 1975. Mammals in the Yukon Territory. National Museum of Canada, Publications in Zoology Number10. 192 pages. Received 15 September 2000 Accepted 19 March 2001 New Records of Vascular Plants in the Yukon Territory II WILLIAM J. Copy!, CATHERINE E. KENNEDY” AND BRUCE BENNETT? ‘Biological Resources Program, Eastern Cereal and Oilseed Research Centre, Agriculture and Agri-Food Canada, Central Experimental Farm, Ottawa, Ontario K1A 0C6, Canada *Department of Renewable Resources, Government of the Yukon, Box 2703, Whitehorse, Yukon Y1A 2C6, Canada Cody, William J., Catherine E. Kennedy, and Bruce Bennett. 2001. New records of vascular plants in the Yukon Territory III. Canadian Field-Naturalist 115(2): 301-322. Ten native taxa including Arabis boivinii, Carex aquatilis ssp. stans, Descurainia incisa var. incisa, Draba lonchocarpa var. vestita, Isoetes maritima, Lepidium densiflorum var. macrocarpum, Malaxis paludosa, Phyllodoce X intermedia, Scirpus acutus, Stipa hymenoides and Taraxacum carneocoloratum are reported new to the known flora of the Yukon Territory. Three introduced taxa including Centaurea cyanus, Lolium perenne ssp. multiflorum, Rheum rhaponticum and Sorbaria sorbifolia are also reported as new to the known flora. Significant range extensions within the Territory are reported for 127 native and 21 introduced taxa and comments are presented on four native taxa. Of these, 10 native and three introduced are new for the Territory. The total vascular flora is now 1163 species. Key Words: Vascular flora, range extensions, Yukon Territory. Since the writing of New Records of Vascular Plants in the Yukon Territory II (Cody et al. 2000), a considerable number of plant specimens have been submitted to Cody for identification and confirma- tion, in particular those from: Bruce Bennett, while working for Yukon Department of Renewable Resources as a wildlife viewing biologist and volun- teering on other projects in various areas in southern Yukon; Greg Brunner, while working with a survey party in the Peel River wetlands area in northeastern Yukon; and John Meikle while cruising down the Snake River near the Mackenzie-Northwest Territory border. In addition, Cody spent five weeks in 1999 in the Territory which included two weeks with his son Gordon surveying flora along the high- ways in the southern part of the Territory, a week with Kennedy and Scott Smith studying the vegeta- tion of Herschel Island and three days examining the vegetation in the northern portion of proposed Tombstone territorial park in the Ogilvie Mountains also with Kennedy and Scott Smith. This paper, along with other additional records recently published (Cody et al. 1998, 2000) further updates the Flora of the Yukon Territory (Cody 1996) and Flora of the Yukon Territory, Second Edition (Cody 2000). The floristic information pre- sented earlier and updated here is essential for bio- logical research and ongoing work relating to agri- culture, forestry, sustainable resource management and wildlife management. With additions of ten native and three introduced species reported here the flora now includes 1163 species. The new native species are all rare (as defined by Douglas et al. 1981) in the Territory. The taxa in this paper appear first in a synoptic list by Yukon status in alphabetical order. The taxa are then discussed in taxonomic order, as presented in the Flora of the Yukon Territory with citation of specimens and other pertinent information. Common names follow Cody (1996) and Douglas et al. (1984, 1990, 1991, 1994). Most of the specimens which were sent to the senior author for identification and verification have been incorporated into the vascular plant herbarium maintained by Agriculture and Agri- Food Canada at the Central Experimental Farm, Ottawa (DAO). Those not retained were photo- graphed and the photographs have been preserved in DAO. Synoptic List by Yukon Status Errata in New Records of Vascular Plants in the Yukon Territory IT: (2) Cerastium nutans Draba densifolia Native taxa new to the Yukon Territory: (10) Arabis boivinii Descurainia incisa var. incisa Draba lonchocarpa vat. vestita Isoetes maritima Lepidium densiflorum var. macrocarpum Malaxis paludosa Phyllodoce x intermedia Scirpus acutus Stipa hymenoides Taraxacum carneocoloratum Introduced taxa new to the Yukon Territory: (3) Centaurea cyanus Rheum rhaponticum Sorbaria sorbifolia Range extensions of native taxa within the Yukon Territory: (127) Androsace septentrionalis Angelica lucida Antennaria densifolia 301 302 THE CANADIAN FIELD-NATURALIST Antennaria pulcherrima Artemisia michauxiana Artemisia tilesii Arabis drummondii Arabis holboellii var. retrofracta Arabis holboellii var. secunda Arabis nuttallii Arctophila fulva Astragalus adsurgens ssp. robustior Boykinia richardsonii Braya humilis Carex aenea Carex albo-nigra Carex aquatilis ssp. stans Carex atratiformis ssp. raymondii Carex atrofusca Carex aurea Carex bicolor Carex capillaris ssp. capillaris Carex capillaris ssp. robustior Carex chordorrhiza Carex crawfordii Carex diandra Carex lachenalii Carex livida Carex obtusata Carex rariflora Carex stylosa Carex tenuiflora Cassiope tetragona ssp. saximontana Castilleja caudata Castilleja miniata Castilleja raupii Castilleja yukonis Cicuta virosa Coeloglossum viride ssp. bracteatum Crepis elegans Cryptogramma stelleri Descurainia pinnata ssp. nelsonii Descurainia sophioides Draba borealis Draba cinerea Draba nemorosa vat. leiocarpa Draba scotteri Draba stenopetala Elymus trachycaulus ssp. subsecundus Equisetum arvense Equisetum scirpoides Equisetum variegatum ssp. variegatum Erigeron acris ssp. politus Erigeron elatus Eriophorum gracile Eriophorum russeolum var. albidum Geum aleppicum ssp. strictum Juncus bufonius Kalmia polifolia Limosella aquatica Lycopodium clavatum var. monostachyon Maianthemum trifolium Mentha arvensis Minuartia biflora Moehringia lateriflora Montia fontana Myrica gale Oxytropis campestris ssp. jordalii Oxytropis deflexa ssp. foliolosa Oxytropis deflexa ssp. sericea Oxytropis scammaniana Parrya arctica Pedicularis macrodonta Pedicularis verticillata Petasites frigidus ssp. frigidus Phalaris arundinacea Pinguicula vulgaris ssp. vulgaris Platanthera aquilonis (P. hyperborea) Platanthera obtusata Polygonum alaskanum Polygonum viviparum Potamogeton alpinus ssp. tenuifolius Potamogeton foliosus var. macellus Potamogeton richardsonii Potamogeton zosteriformis Potentilla norvegica Potentilla rubricaulis Puccinellia andersonii Pyrola minor Ranunculus hyperboreus Rhodiola rosea ssp. integrifolia Rorippa palustris Rubus chamaemorus Rubus pubescens Rumex acetosa ssp. alpestris Rumex salicifolius ssp. triangulivalvis Sagina saginoides Salix alaxensis ssp. longistylis Salix arctophila Salix barrattiana Salix pedicellaris Salix pyrifolia Saxifraga adscendens ssp. oregonensis Saxifraga aizoides Saxifraga bronchialis ssp. funstonii Scheuchzeria palustris ssp. americana Scirpus caespitosus ssp. austriacus Scirpus rollandii Selaginella selaginoides Senecio hyperborealis Senecio lugens Silene uralensis ssp. ogilviensis — Sparganium minimum Sparganium multipedunculatum Spiranthes romanzoffiana Stellaria longipes Stipa nelsonii ssp. dorei Thalictrum alpinum Tofieldia coccinea Tofieldia glutinosa ssp. brevistyla Tofieldia pusilla Triglochin maritimum Typha latifolia Utricularia intermedia Utricularia vulgaris ssp. macrorhiza Woodsia glabella Range extensions of introduced taxa within the Yukon Territory: (21) Alopecurus pratensis Astragalus cicer Caragana arborescens Clematis tangutica Vol. 115 2001 Crepis tectorum Elymus repens Galeopsis tetrahit ssp. bifida Hordeum brachyantherum Lepidium ramosissimum Lolium perenne ssp. multiflorum Medicago falcata Medicago sativa Melilotus alba Melilotus officinalis Polygonum buxiforme Polygonum lapathifolium Sonchus arvensis ssp. uliginosus Stellaria media Taraxacum officinale Thlaspi arvense Vicia cracca Comments on native taxa in the Yukon Territory: (5) Adoxa moschatellina Isoetes echinospora Platanthera aquilonis Poa cusickii Scirpus validus Taxonomic List LYCOPODIACEAE Lycopodium clavatum L. var. monostachyon Hook. & Grev., Common Club-moss — YUKON: alpine dry ridge above talus steep rocky slope, Beavercrow Ridge, 60°14’N 124°32’W, B. Bennett 98-366, 15 Aug. 1998 (DAO). The specimen cited above is an extension of the known range in the Yukon Territory of about 180 kilometers from a site west of longitude 127°W (Cody 1996). SELAGINELLACEAE Selaginella selaginoides (L.) Link, Northern Selaginella - YUKON: rich tall willow site, middle of Seela Pass, Tombstone Range, Ogilvie Mountains, 64°42.58’N 139°51.304’W, W. J. Cody 36401, 20-22 July 1999 (DAO). Cody (1996) knew this species in the Yukon Territory mainly from south of latitude 62°N and then disjunct to the northern end of the Canol Road and two sites north of lati- tude 64°N east of the Dempster Highway. The specimen cited above is an extension of the known range to the west of about 75 kilometers from the northernmost site adjacent to the Dempster Highway. ISOETACEAE Isoetes echinospora Dur. The northern dot on the distribution map of this species in the Flora of the Yukon (Cody 1996) should be removed. The specimen on which it was based has been revised to J. maritima — see below. Isoetes maritima Underw., Maritime Quillwort — YUKON: in shallow water, south end of Sheldon Lake, opposite Mile 219 Canol Road, Porsild & Breitung 11502, 17 Aug. 1944 (CAN); fen, Peel River Wetland Study, Vittrewka Lake, 66°49°7.41°N 135°31°8.93”W, G. Brunner 39-99, Copy, KENNEDY, AND BENNETT: VASCULAR PLANTS IN YUKON III 303 29 June 1999 (Yukon Renewable Resources, DAO) (determined by D. M. Britton). Porsild (1951) originally determined the first specimen as Isoetes braunii (I. echinospora Dur.). Brunton and Britton (1999) revised it to /. maritima. The second speci- men is an extension to the known range in the Yukon Territory of about 575 kilometers to the northwest. Brunton and Britton published the following information: “Origin and dispersal At their closest, contemporary populations of /soetes maritima in coastal southern Alaska are separated by ca. 500 km and across several mountain ranges from the Sheldon Lake population and are not connected to it by river systems. Disjunct interior British Columbia /. mariti- ma, on the other hand, is less than 200 km from the species’ primary coastal range and is connected directly to those populations by major river systems (Britton and Brunton 1995). The apparently isolated populations in the Fairbanks areas of eastern interior Alaska are over 900 km west of the Sheldon Lake population but occur within the same drainage system, the Yukon River. Much of the Yukon River watershed was unglaciated during the Wisconsinan glaciation when the Cordilleran and Laurentide ice sheets covered much of northern North America (Dyke and Prest 1987). The unglaciated corridor across the Yukon River watershed constituted the eastern portion of Beringia, serving both as a glacial era refugium and a post-glacial floristic and faunal migration corridor (Hultén 1968; Schweger 1989). The Alaska-British Columbia coastal band along which /soetes maritima is common today, however, as well as the intervening interior mountains, continued to be covered in glacial ice sheets during this time (Hughes et al. 1989).” Isoetes maritima can be separated from I. echinospora as follows: A. Spines on megaspores long and acute, not reduced in size near the equator; microspores smooth or with very fine thread-like spines I. echinospora A!. Spines on megaspores blunt, sometimes confluent into ridges, reduced to small tubercules near equator; microspores echinate with coarse pronounced spines I. maritima Isoetes maritima should be added to the list of rare plants in the Yukon Territory (Douglas et al. 1981). EQUISETACEAE Equisetum arvense L., Field Horsetail - YUKON: edge of active floodplain in poorly developed soil amongst leaf litter, Snake River, 64°57.16’N 133°00.21’W, J. Meikle 99-001, 19 July 1999 (Yukon Renewable Resources, photo DAO); Snake River, 65°56’N 133°17’W, J. Meikle 99-002, 22 July 1999 (Yukon Renewable Resources, photo DAO). This species has not previously been recorded from the Snake River area in the Yukon Territory. The specimens cited above are about 100 kilometers from sites mapped by Cody (1996). Equisetum scirpoides Michx., Dwarf Scouring-rush — YUKON: growing in feathermoss in white spruce valley.) cionest,, wWiake- »River:..'64°37,. 16M 133°00.25’W, J. Meikle 003, 19 July 1999 (Yukon Renewable Resources, photo DAO). This species, which is common in the Yukon Territory 304 (Cody 1996) was not previously known from the Snake River area where it was collected on a site about 150 kilo- meters northeast from the vicinity of Mayo. Equisetum variegatum Schleich. ssp. variegatum, Variegated Horsetail - YUKON: poorly developed soil at edge of active flood plain; in willow leaf litter at forest edge, Snake River, 64°57.16’N 133°00.21’W, J. Meikle 99-004, 19 July 1999 (Yukon Renewable Resources, photo DAO). This species is widespread in the Yukon Territory (Cody 1996). The specimen cited above from the Snake River area is an extension of the known range in the Territory of about 100 kilometers northeast of a site in the Wernecke Mountains (Cody 1996). PTERIDACEAE Cryptogramma stelleri (Gmel.) Prantl, Slender Cliff- brake — YUKON: SW facing slope, small carbonate face 70-100 m above river about 1 km downstream from lower South Bend on east side of river, Snake River, 65°06.5’N 133°07’W, J. Meikle 99-005, 19 July 1999 (Yukon Renewable Resources, photo DAO). Douglas et al. (1981) considered this species rare in the Yukon Territory. The nearest sites known to Cody (1996) to the specimen cited above were from about 240 kilome-_ ters to the west adjacent to the Dempster Highway. ASPIDIACEAE Woodsia glabella R.Br., Smooth Woodsia — YUKON: SW facing slope, small carbonate face 70- 100 m above river about 1 km downstream from lower South Bend on east side of river, Snake River, 65°06.5’N 133°07’W, J. Meikle 99-006, 19 July 1999 (Yukon Renewable Resources, photo DAO). The specimen cited above is an extension of approxi- mately 115 kilometers northeast of a site in the Wernecke Mountains in the Yukon Territory (Cody 1996). TYPHACEAE Typha latifolia L., Common Cattail - YUKON: large patches, Faro Sewage Ponds, 60°13’N 133°22’W, B. Bennett 99-561, 28 Oct. 1999 (DAO). The specimen cited above is only the third known locali- ty in the Yukon Territory; it is intermediate between a site in the vicinity of Mayo (Douglas et al. 1981) and a site in the extreme southeast (Cody et al. 2000). SPARGANIACEAE Sparganium minimum (Hartm.) Fries, Small Bur- reed — YUKON: Turner Lake, 66°09’56”N 134°17°16”W, G. Brunner 52-99, 1 July 1999 (Yukon Renewable Resources, photo DAO); No Gold Creek, Horseshoe Slough, 63°26’N 135°06’W, D. Mossop s.n., 1 July 1999 (B. Bennett Herbarium, photo DAO). Cody (1996) knew this rare species from only three sites in the Yukon Territory. Cody et al. (1998, 2000) added additional sites in the southeast in the vicinities of the La Biche River, Liard Plateau and Upper Coal River. The specimens cited above are extensions of the known range to the northeast of about 350 kilometers from the vicinity of Dawson and 200 kilometers southeast of Dawson. THE CANADIAN FIELD-NATURALIST Vol. 115 Sparganium multipedunculatum (Morong) Rydb. — “YUKON: small pond by South McQuesten River near Elsa, 63°56.2’N 135°34.4’W, S. Withers 528, 10 Aug. 1999 (DAO). This species, although not frequent in the Yukon Territory, was not included in the rare plants of the Territory because it is widespread to the south. Cody et al. (1998, 2000) added new records in the southeast from the vicinities of Frances Lake and Beaver River. The specimen cited above is from a location about 150 kilometers east of a site south of the south end of the Dempster Highway and 200 kilometers north of Carmacks (Cody 1996). POTAMOGETONACEAE Potamogeton alpinus Balbis ssp. tenuifolius (Raf.) Hultén, Pondweed — YUKON: Turner Lake, G. Brunner 41-99, 2 July 1999 (Yukon Renewable Resources, photo DAO). The specimen cited above is an extension of the known range of about 230 kilometers to the northeast of locations adjacent to the Dempster Highway and 280 kilometers to the southeast from a site north of the Porcupine River mapped by Cody (1996). Potamogeton foliosus Raf. var. macellus Fern., Closed-leaved Pondweed — YUKON: in poois below hotsprings, Beaver River Hotsprings, 60°22.53’N 125°34.33’W, B. Bennett 95-603, 14 Aug. 1998 (DAO). This is a rare species in the Yukon Territory (Douglas et al. 1981) where Cody (1996) knew it from only five sites, all west of longitude 135°W. The specimen cited above is from a site about 675 kilometers to the southeast. Potamogeton richardsonii (Benn.) Rydb., Richardson’s Pondweed — YUKON: fen, Turner Lake, 66°10’30.71”N 134°14’49.04’W, G. Brunner 42-99, 2 July 1999 (Yukon Renewable Resources, photo DAO); shallow water, Taber Lake, 66°58’42.26"N 134°46’55.51”W, G. Brunner 1-99 (Yukon Renewable Resources, photo DAO). Specimens mapped by Cody (1996) were north to about latitude 64°50’N adjacent to the Dempster Highway and disjunct to north of the Porcupine River. The specimens cited above are from the easterly part of the intermittent area. Potamogeton zosteriformis Fern., Eel-grass Pondweed — YUKON: Nordenskiold River area, 60°41.8’N 131°47.22’W, M. Dennington 28, 14 Aug. 1984 (DAO). Cody (1994, 1996) knew this rare species in the Yukon Territory from only two localities west of longitude 135°W. Cody et al. (1998) extended the known range in the Territory to the Old Crow Flats north of latitude 68°N and Cody et al. (2000) reported an additional locality in the extreme southeast in the Beaver River area. The specimen cited above is from a site intermediate between the Beaver River site and a site southeast of Haines Junction. SCHEUCHZERIACEAE Scheuchzeria palustris L. ssp. americana (Fern.) Hultén, Scheuchzeria — YUKON: fen, Turner Lake, 66°10’30.71”"N 134°14’49.04”W, G. Brunner 43-99, 2001 2 July 1999 (Yukon Renewable Resources, photo Die). ten, Taber Lake, 66°56’29.96”"N 134°43°48.30"W, G. Brunner 9-99, 25 June 1999 (Yukon Renewable Resources, photo DAO). Douglas et al. (1981) knew this rare species in the Yukon Territory only in the Mayo area. Cody (1996) plotted an additional location to the south in the Pelly River area and Cody et al. (1998) recorded two additional sites in the Frances Lake and Watson Lake areas. The specimens cited above extend the known range in the Territory about 350 kilometers north from the Mayo area. Triglochin maritimum L. Seaside Arrow-grass — meme. Tourmer Lake, 66°10°30.71"N 134°14°49.04"W, G. Brunner 45-99, 2 July 1999 (Yukon Renewable Resources, photo DAO). The specimen cited above is an extension of the known range in the Yukon Territory (Cody 1996) of about 180 kilometers northeast of a site adjacent to the Dempster Highway and of about 250 kilometers to the southeast of a site north of the Porcupine River. POACEAE (GRAMINEAE) Alopecurus pratensis L., Meadow Foxtail — YUKON: reclaimed mining trench, Red Ridge, 60°21.5°N 135°04.2’W, S. Withers 99-106, 14 July 1999 (B. Bennett Herbarium, photo DAO); white flats below Elsa, 63°55’N 135°29’W, Cody & Cody 35786, 4 July 1999 (DAO). Cody (1996) knew this introduced species in the Yukon Territory from only three locations: Dawson, south of Haines Junction and northwest of Ross River. Arctophila fulva (Trin.) Rupr., Pendantgrass — Vittrewka Lake, 66°49°57.54”N 135°29’10.81”"W, G. Brunner 24-99, 27 June 1999 (Yukon Renewable Resources, photo DAO). This species is widespread in the Yukon Territory (Cody 1996) but was previously unknown from the Eagle Plains between about latitude 65°N adjacent to the Dempster Highway and the Porcupine River. Elymus repens (L.) Gould, Quack Grass - YUKON: disturbed roadside, Robert Campbell Highway, Km 467, B. Bennett 99-542, 19 Aug. 1999 (DAO); road- side between La Biche and Liard River, 60°03’06”N 123°59°35”W, B. Bennett 95-237, 15 June 1995 (B. Bennett Herbarium); La Biche Airstrip, 60°07’42”N 124°02’21”W, B. Bennett 98-581, 21 June 1998 (DAO). This introduced species was known in the Yukon Territory (Cody 1996) from the vicinities of Dawson, near the southern end of the Dempster Highway and near Mayo. The specimens cited extend the known range about 750 kilometers to the southeast of Mayo. Elymus trachycaulus (Link) Gould ssp. subsecundus (Link) Gould - YUKON: eroding riverbank, Upper La Biche River, 60°13’58”N 124°13’58”W, B. Bennett 98-093, 18 June 1998 (DAO). The specimen cited above is an extension of the known range in the Yukon Territory (Cody 1996) of about 475 Copy, KENNEDY, AND BENNETT: VASCULAR PLANTS IN YUKON III 305 kilometers east from sites adjacent to the Canol and Tagish Highways. Hordeum brachyantherum Nevski, Meadow Barley — YUKON: little silty island in centre of pond, Paddy’s Pond, Hillcrest, Whitehorse [60°43’N 135°03’W], B. Bennett 99-001, 2 May 1999 (DAO); single patch along lakeshore, Schwatka Lake, Whitehorse, 60°41.45’N 135°02.4’W, B. Bennett 99- 549, 27 Aug. 1999 (DAO). This species was previously known in the Yukon Territory (Cody 1996) from only two localities: Carcross, where it was collected by J. M. Gillett in 1949 and consid- ered native, and Dawson where it was collected by Baum and Bailey in 1977 and considered introduced. From the habitat of the specimens cited above, it is probably intro- duced. The history of Paddy’s Pond is that the surface materials were bulldozed off during World War II for fill. Schwatka Lake is a reservoir for the Whitehorse Rapids Dam, and the lakeshore is completely unnatural. Lolium perenne L. ssp. multiflorum (Lam.) Husnot, Italian Rye Grass - YUKON: reclaimed mining trench, Division Mountain near Nordenskiold River, 61°19°N 136°3.5’°W, S. Withers 309, 26 July 1999 (DAO); adjacent to old railway track at foot of steep clay slope, Whitehorse, 60°42’14.4”N 135°03’10.8"W, Cody & Cody 37443, 1 Aug. 2000 (DAO) (determined by S. Darbyshire). Cody (1996) knew this introduced taxon from only three localities: Dawson, Pelly River Ranch, and east of Watson Lake. The first specimen cited above is from about 100 kilometers southeast of the Pelly River Ranch area. The second specimen is from about 100 kilometers further southeast. Phalaris arundinacea L., Reed Canary Grass — YUKON: reclaimed mining trench, Mechanic Creek, Big Creek Drainage, 62°20’N 137°19.5°W, S. Withers 203, 21 July 1999 (DAO). Douglas et al. (1981) considered this species rare in the Yukon Territory. The specimen cited above is from an area about 125 kilometers northwest of a site adjacent to the Klondike Highway mapped by Cody (1996) but was germi- nated from an experimental agronomic seed mix planted in 1995 which included Phalaris arundinacea. Poa cusickii Vasey, Cusick’s Bluegrass - YUKON: Artemisia — graminoid meadow on glacial till adja- cent to Carcross Dunes, rare, 60°10’N 132°42’W, C. Parker & B. Bennett 8271, 29 Aug. 1998 (DAO). Douglas et al. (1981) considered a collection from north- west of Haines Junction near Bear Creek to be an introduc- tion; however, Cody (1994) stated that specimens collected by R. Florkiewicz adjacent to Stony Creek, west of Whitehorse supported the possibility that this species was native in the Yukon Territory. The specimen cited above would also appear to be native in this region. Puccinellia andersonii Swallen — YUKON: gravel shoreline and adjacent vegetation, Pauline Cove Settlement on Simpson Point, Herschel Island, 69°34’N 138°55’W, W. J. Cody 36002, 13 July 1999 (DAO). 306 This species was known to Cody (1996) in the Yukon Territory from only two sites on the arctic mainland coast. Stipa hymenoides Roem. & Schult. (Oryzopsis hymenoides (Roem. & Schult.) Ricker, Achnatherum hymenoides (Roem. & Schult.) Barkworth), Indian Ricegrass (Figure 1) - YUKON: widespread on sandy eroding slope beside hydro dam growing with Stipa comata, Elymus calderi and Penstemon gor- manii, Schwatka Lake Dam, Whitehorse, 60°41.75’N 135°02.34’W, B. Bennett 99-499, 26 Aug. 1999 (DAO, B. Bennett Herbarium). This is a North American species, which is common in southcentral and southeastern British Columbia and is found eastward to southwestern Manitoba and south to Texas, New Mexico, Arizona, California and Mexico. It is new to the Yukon Territory and may be introduced in this area because it is adjacent to the old tramway which passed by the Whitehorse Rapids. It is, however, growing beside species which are endemic or rare in the Yukon. Its assocation with a natural habitat suggests that it is native and there is a prece- dent for widespread southern prairie species on south-facing slopes in the Yukon Territory. Stipa hymenoides may be sep- arated from S. richardsonii as follows: A. Awns 16-60 mm long, persistent; lemmas 5-6 mm long, pubescent with hairs about 1 mm long S. richardsonii . A!. Awns 3-6 mm long, deciduous; lemmas subglobose to fusiform, 2.5-5.0 mm long, pubescent with hairs to 6 mm long S. hymenoides Stipa nelsonii Scribn. ssp. dorei Barkworth & Maze, Columbian Needlegrass —- YUKON: stabilized dunes, Carcross Dunes, 60°10.35’N 134°43.26’W, B. Bennett 99-355, 27 Aug. 1999 (DAO). Cody (1996) knew this species from only six locations in southern Yukon Territory. At the site reported above over 50 plants were observed. CYPERACEAE Carex aenea Fern., Bronze Sedge - YUKON: sandy silt riverbank, La Biche River, 60°13’58”N 124°13’°58’W, B. Bennett 98-113, 17 June 1998 (DAO). Cody (1996) knew this species in the Yukon Territory from scattered localities north to latitude 64°N. The speci- men cited above is an extension of the known range in the Territory of about 275 kilometers east of Watson Lake. Carex albo-nigra Mack., Two-toned Sedge — YUKON: in rock crack of dolomite bedrock on exposed ledge on edge of ridge, Beavercrow Ridge, 60°12.41’N 124°35.99°W, B. Bennett 98-630, 16 Aug. 1998 (DAO). This species which was considered rare in the Yukon Territory by Douglas et al. (1981) was known to Cody (1996) as far north as latitude 62°30’N. The specimen cited above is an extension of the known range of about 280 kilometers east of a site west of Watson Lake. Carex aquatilis Wahlenb. ssp. stans (Drej.) Hultén — YUKON: wet mesic flats, SW shore of outer Delta, Phillips Bay, 69°14’N 138°29’W, Dickson & Allen 5353, 28 July 1982 (DAO). THE CANADIAN FIELD-NATURALIST Vol. 115 Hultén (1968) mapped several Yukon Territory speci- mens under this high arctic subspecies but Cody (1996) included them in his treatment of ssp. aquatilis. Cody et al. (1998) reported a collection of ssp. stans from Stokes Point Lagoon. The specimen cited above is the second record from this area. 3 Carex atratiformis Britt. ssp. raymondii (Calder) A. E. Porsild, Black Sedge - YUKON: high subalpine scree, Kotaneelee Range, 60°14’N 124°07’'W, B. Bennett & R. Rosie 98-345, 19 June 1998 (B. Bennett Herbarium, photo DAO). Cody et al. (2000) extended the known range of this taxon in the Yukon Territory about 200 kilometers south- east from Frances Lake to the Beaver River area. The spec- imen cited above is a further extension of about 125 kilo- meters to the southeast. FIGURE 1. Stipa hymenoides (Drawn by Lee Mennell). 2001 Copy, KENNEDY, AND BENNETT Carex atrofusca Schk. — YUKON: subalpine moist herb meadow, Kotaneelee Range, 60°14.31’N 124°07.19° W, B. Bennett 98-058, 20 June 1998 (DAO). Cody et al. (2000) extended the known range in the Yukon Territory of this circumboreal species (which was not included in the rare plants of the Territory, Douglas et al. 1981) to the headwaters of the Bonnet Plume River. The specimen cited above is an extension of the known range of about 400 kilometers eastward from a site adjacent to lon- gitude 131°W. Carex aurea Nutt., Golden Sedge, Snake River, 65°12.32’N 133°19.70’W, J. Meikle 96-007, 20 July 1999 (Yukon Renewable Resources, photo DAO); riverbar, Wind River Camp #1, 64°40.39’N 134°35.96’W, B. Bennett 00-748, 2 July 2000 (DAO). The specimens cited above are an extension of the known range in the Yukon Territory (Cody 1996) of about 225 kilometers northeast of a site near Mayo and east of a site adjacent to the Dempster Highway. Carex bicolor All., Two-coloured Sedge — YUKON: Salix brachycarpa-Carex parryana community, Slim’s River Delta, 60°59.6’N 138°29.4’W, S. Withers 38B, 7 July 1999 (DAO). Cody (1996) considered this species rare south of lati- tude 63°30’N in the Yukon Territory. The specimen cited above is from about 75 kilometers southeast of a site in Kluane National Park. Carex capillaris L. ssp. capillaris, Hairlike Sedge — MESON: ten, Taber Lake, 66°59’40.22”N 134°44°51.14”"W, G. Brunner 5-99, 24 June 1999 (Yukon Renewable Resources, photo DAO); swamp, Vittrewka Lake, 66°49719.23”N 135°29’45.93”W, G. Brunner 22-99, 27 June 1999 (Yukon Renewable Resources, photo DAO). The specimens cited above are from sites about 70 kilo- meters east and northeast of sites mapped by Cody (1996) east of the Dempster Highway. Carex capillaris L. ssp. robustior (Drej. ex Lange) Bocher, Hairlike Sedge - YUKON: marl area near McLean Creek south of Whitehorse, 60°35’N 134°55’W, Cody & Cody 35638, 27 June 1999 (DAO). This subspecies was widely scattered in the Yukon Territory west of longitude 137°W according to Cody (1996). The specimen cited above is an extension eastward from Kluane National Park of about 140 kilometers. Carex chordorrhiza Ehrh. ex L.f., Cordroot Sedge — MeeON: fen, Turner Lake, 66°10°43.25”N 134°14’37.37°W, G. Brunner 60-99, 1 July 1999 (Yukon Renewable Resources, DAO). The specimen cited above is an extension of the known range in the Yukon Territory (Cody 1996) of about 225 kilometers northwest of a site adjacent to the Dempster Highway and 240 kilometers southwest of sites adjacent to the Porcupine River. Carex crawfordii Fern., Crawford’s Sedge — YUKON: old road to wetland west of bridge, La : VASCULAR PLANTS IN YUKON III 307 Biche River, 60°05.74’N 124°01.98’W, B. Bennett 98-634, 15 June 1998 (DAO). Douglas et al. (1981) considered this species rare in the Yukon Territory. The specimen cited above is an extension of the known range in the Territory of about 180 kilometers to the east of the easternmost site mapped by Cody (1996). Carex diandra Schrank, Lesser Panicled Sedge — YUKON: “Turner ‘Lake; -66°09°56.9°N 134°17°16.6"W, G. Brunner 53-99, 1 July 1999 (Yukon Renewable Resources, DAO). The nearest sites of this species mapped by Cody (1996) were from the vicinity of Dawson about 325 kilometers to the southwest and adjacent to the Porcupine River about 285 kilometers to the northwest. Carex lachenalii Schk., Two-parted Sedge - YUKON: subalpine Salix/Abies adjacent to Carex meadow in draw, Beavercrow Ridge, 60°12.865’N 124°35.79’W, B. Bennett 98-631, 16 Aug. 1998 (DAO). Cody et al. (1998) extended the known range of this species into the southeast of the Yukon Territory. The spec- imen cited above is a further extension of the range in the southeast of about 175 kilometers. Carex livida Willd., Livid Sedge —- YUKON: fen, Turner Lake, 66°10’43.25”"N 134°14’37.37"W, G. Brunner 55-99, 1 July 1999 (Yukon Renewable Resources, DAO). The nearest site of this rare species in the Yukon Territory known to Cody (1996) was adjacent to the Dempster Highway, about 200 kilometers to the southwest of the specimen cited above. Carex obtusata Liljeb., Blunt Sedge - YUKON: Herschel Island, near Pauline Cove Settlement, 69°34’N 138°55’W, W. J. Cody & C. E. Kennedy 36351, 18 July 1999 (DAO). In the Yukon Territory this species was known to Cody (1996) in the south from the South Canol Road to the west and north to about 64°15’N and then rare north to the Arctic Coast. The specimen cited above is new to Herschel Island and the nearest collection by C. Cain in 1962 was found on a lushly vegetated bench on plain back from the mainland coast at about 138°10’W near King Point (DAO). Carex rariflora (Wahlenb.) Sm., Loose-flower Alpine Sedge — YUKON: bog, Vittrewka Lake, 66°49’4.94"N 135°30°48.84’°W, G. Brunner 37-99, 29 June 1999 (Yukon Renewable Resources, photo DAO). This species, which was considered to be rare in the Yukon Territory in the extreme northeast and adjacent to the Dempster Highway by Douglas et al. (1981) was shown to be much more widespread north of the Porcupine River by Cody (1996). The specimen cited above is from a site about 250 kilometers northeast of the northernmost location known adjacent to the Dempster Highway and 80 kilome- ters south of a site adjacent to the Rat River. Carex stylosa C. A. Mey., Long-styled Sedge — YUKON: hummocky terrain, White Hat Hills, Tombstone Range, Ogilvie Mountains, 64°45.20’N 138°29.113’W, W. J. Cody 36484, 20-22 July 1999 (DAO). 308 Cody (1996) knew this species which was considered rare in the Yukon Territory (Douglas et al. 1981) only as far north as the vicinity of Dawson. The specimen cited above is an extension of the known range northward of about 60 kilometers from Dawson. Cody et al. (2000) did however report a specimen from the Arctic coast. Carex tenuiflora Wahlenb., Sparse-leaved Sedge — YUKON: open shrub/Ledum/lichen, Vittrewka Lake, 66°49°42.37°N 135°36’10.82”W, G. Brunner 27-99, 28 June 1999 (Yukon Renewable Resources, photo DAO). The specimen cited above is an extension of the known range in the Yukon Territory (Cody 1996) of about 175 kilometers northeast of a site in the Ogilvie Mountains at 65°48’N 138°35’W and 200 kilometers southeast of the Old Crow region adjacent to the Porcupine River. Eriophorum gracile Koch, Slender Cotton-grass — YUKON: fen, Turner Lake, 66°11’06.09”"N 134°16’45.14"W, G. Brunner 48-99, 2 July 1999 (Yukon Renewable Resources, photo DAO). This is a rare species in the Yukon Territory that was unknown to Douglas et al. (1981). Cody (1994) reported the first collection from the southeast in the Coal River Springs area and Cody et al. (1998) reported two additional sites from the Upper Coal River and Frances Lake areas. The specimen cited above is an extension of the known ~ range of about 630 kilometers northwest from the vicinity of Frances Lake. Eriophorum russeolum Fries var. albidum Nyl. — YUKON: fen, Vittrewka Lake, 66°49’6.56”N 135°30’°55.33”W, G. Brunner 38-99, 29 June 1999 (Yukon Renewable Resources, photo DAO). This species is only frequent in the Yukon Territory (Cody 1996) north of latitude 67°N. The specimen cited above from Vittrewka Lake is from about 90 kilometers north of Ursus Lake just north of latitude 66°N. Scirpus acutus Muhl. (Figure 2, Map 1) —- YUKON: specimens from the vicinities of Mayo, Klondike Highway, Ross River and south of Whitehorse treat- ed by Cody (1996) as S. validus have been revised to S. acutus. These are from the central part of the Yukon Territory south of latitude 64°N. Scirpus acu- tus which is new to the Yukon Territory can be sepa- rated from S. validus as follows: A. Fresh culms dark green, firm; aerenchymal lacunae less than 0.9 mm in diameter; inflorescence length usually less than 47 mm; number of primary rays less than 7; secondary rays absent; clusters of 5 or more spikelets present; scales dull, basically pale or whitish brown, the midrib not strongly contrasting, the margins often more copiously ciliate and the backs copiously flecked with shiny red dots, often puberulent; mature achenes ca. 2.2-2.7 mm long, including the apicules, completely hidden by the scales S. acutus . Fresh culms light green, rather soft; aerenchymal lacu- nae more than 1.5 mm in diameter; inflorescence length greater than 80 mm; number of primary rays greater than 13; secondary, tertiary, and occasionally quater- nary rays present; spikelets mostly borne in clusters 1- 3; scales more or less shiny, rich orange-brown, often with prominent greenish midrib, the margins ciliate but > THE CANADIAN FIELD-NATURALIST Vol 1S KILOMETRES 141° 139° 137° 135° 133° 131° Map 1. Scirpus acutus. 2001 the backs essentially glabrous (puberulence and swollen red flecks, if any, limited to the region of midrib); mature (dark gray or lead coloured) achenes ca 1.6-2.1 mm long, includ- ing apicules, barely covered by the scales S. validus Scirpus caespitosus L. ssp. austriacus (Pallas) Asch. & Graeb., Tufted Clubrush - YUKON: swamp, Taber Lake, 66°56’38.65”N 134°44’12.56’W, G. Brunner 11-99, 25 June 1999 (Yukon Renewable Resources, photo DAO). Cody (1996) considered this species in the Yukon Territory as occasional north to the southern part of the Dempster Highway and then disjunct to north of the Porcupine River where it is apparently rare. The specimen cited above from Taber Lake is from the eastern section of the intermediate area. Scirpus rollandii Fern., (Trichophorum pumilum (Vahl) Schinz. & Tehell.), Tufted Clubrush — YUKON: marl lowland near McLean Creek, south of Whitehorse, 60°35’N 134°55’W, Cody & Cody 35645, 27 June 1999 (DAO). Douglas et al. (1981) knew this rare species in the Yukon Territory from three localities, all close to Haines Junction. Cody (1996) mapped five sites northwest and east of Haines Junction plus one south of Whitehorse at Watson River at latitude 60°15’N and one north of Whitehorse adjacent to the Twin Lakes at latitude 61°42’N. The new collection cited above is intermediate between Twin Lakes and Watson River areas. Scirpus validus Vahl, Common Great Bulrush (Map 2) — YUKON: 61°32’N 133°55’W, C. A. McEwen 6501, 22 July 1982 (DAO); in wet bog, along Dempster Highway near Mile 92, R. T. Porsild 1587, Map 2 Scirpus validus. Copy, KENNEDY, AND BENNETT: VASCULAR PLANTS IN YUKON III 309 16 July 1968 (CAN); common locally in very wet bog margins along Dempster Highway near Mile 91, R. T. Porsild 1752, 2 Aug. 1968 (CAN); shallow ponds and bogs northeast of Mayo, 63°35’-37’N 135°45’-55’W, R. T. Porsild 1243, 28 Aug. 1967 (CAN); artificial wetland, Elsa, 63°55.5’N 135°30.4’W, S. Withers 521, 10 Aug. 1999 (DAO); sedge meadow at lake edge, Buffalo Lake, 61°25’22.7°N 137°017°44.19"°W, Staniforth & Rosie 98-77, 25 Aug. 1998 (DAO). The specimens cited above were treated as S. validus by Porsild (1975) and Cody (1996). Other specimens in the Yukon Territory mapped by Cody (1996) have now been revised to Scirpus acutus Muhl. (Map 1). Scirpus validus is now known from four areas in the Yukon Territory, and should still be considered as rare in the Territory (Douglas et al. 1981). JUNCACEAE Juncus bufonius L., Toad Rush — YUKON: in slough off river in silty mud, Upper Wolf River, 60°41.98’N 132°06.62’W, B. Bennett 99-386, 13 Aug. 1999 (DAO). With the exception of a single collection just north of Watson Lake, Cody (1996) knew this species only in central Yukon Territory. Cody et al. (2000) reported an additional site in the extreme southeast in the La Biche River area. The specimen cited above is also from the south, about 180 kilometers west of the Watson Lake collection. LILIACEAE Maianthemum trifolium (L.) Sloboda, Three-leaved Solomon’s Seal —- YUKON: swamp, Turner Lake, 66°10°30.71”"N 134°14’49.04°W, G. Brunner 44-99, 2 July 1999 (Yukon Renewable Resources, photo DAO); Sphagnum bog, Wolf Lake, 60°42.6’N 131°44.08’W, B. Bennett 99-391, 11 Aug. 1999 (DAO). This species was considered rare in the Yukon Territory by Douglas et al. (1981) based on three locations in the southeast. Cody (1996) mapped eight sites in the same gen- eral area north to Frances Lake. Cody et al. (1998) reported a site at Enoch Lake north of latitude 68°N and Cody et al. (2000) reported a site south of latitude 66°N. The Turner Lake specimen cited above is from about 70 kilometers northwest of the site south of latitude 66°N; the Wolf Lake site which is the westernmost yet found in the south, is about 100 kilometers northwest of the nearest site mapped by Cody (1996). Tofieldia coccinea Richards., Northern False Asphodel — YUKON: shallow soil over limestone, Snake River, 65°24’N 133°24’W, J. Meikle 99-008, 20 July 1999 (Yukon Renewable Resources, photo DAO). The specimen cited above is an extension of the known range in the Yukon Territory (Cody 1996) of about 75 kilo- meters southeast of a site adjacent to the northern Bonnet Plume River. Tofieldia glutinosa (Michx.) Pers. ssp. brevistyla Hitchc., Sticky False Asphodel — YUKON: fen, Turner Lake, 66°10’1.63”N 134°16’50.66”W, G. SiO) Brunner 50-99, 1 July 1999 (Yukon Renewable Resources, photo DAO). This species was considered rare in the Yukon Territory by Douglas et al. (1981). The specimen cited above is an extension of the known range in the Territory of about 280 kilometers north from Mayo. Tofieldia pusilla (Michx.) Pers., Common False Asphodel — YUKON: subalpine, moist herb mead- ow, common, Kotaneelee Range, 60°14’31”N 124°07’ 19° W, B. Bennett 98-594 (DAO). This is a common species throughout most of the Yukon Territory (Cody 1996). The specimen cited above however is an extension of the known range in the southeast of about 200 kilometers east of a site west of longitude 127°W. ORCHIDACEAE Coeloglossum viride (L.) Hartm. ssp. bracteatum (Muhl.) Hultén - YUKON: sorted gravel with poorly developed, soil, | Smake River. 165 72.5 74N| 133°19.70’W, J. Meikle 99-010, 20 July 1999 (Yukon Renewable Resources, photo DAO). This species was considered rare in the Yukon Territory by Douglas et al. (1981). Cody (1996) knew it from only two localities: adjacent to the Bonnet Plume River in the north and Little Atlin Lake in the south. The specimen cited above is from an area about 50 kilometers east of the Bonnet Plume River site. Malaxis paludosa (L.) Sw. — YUKON: fen, Turner Lake, 66°10’43.25”N 134°14’37.37’°W, G. Brunner 56-99 and 57-99, 1 July 1999 (Yukon Renewable Resources, photo DAO). Cody (1996) suggested that this species should be looked for in southern Yukon Territory because it is known in extreme southwestern District of Mackenzie and Alaska (Porsild and Cody 1980; Hultén 1968). It is an extension of the known range northwest of the District of Mackenzie site of about 925 kilometers. Malaxis paludosa is a circum- polar species which is apparently rare and local with very large gaps. It should be added to the Rare Plants of the Yukon Territory (Douglas et al. 1981). Platanthera aquilonis Sheviak (P. hyperborea (L.) Lindl.), Northern Green Orchid — YUKON: fen, Turner Lake, 66°09’56.9"N 134°17’°16.6”’W, G. Brunner 5la-99, 1 July 1999 (Yukon Renewable Resources, photo DAQ); in white spruce forest above active flood plain at lower south bend of river, Snake River, 65°06’N 133°06’W, J. Meikle 99-014, 19 July 1999 (Yukon Renewable Resources, photo DAO). C. J. Sheviah (1999) has recently demonstrated that the name Platanthera hyperborea has been misapplied by North American workers and has provided our plant with a new name, P. aquilonis. The first specimen cited above is only the second collection from north of latitude 66° in the Yukon Territory (Cody 1996). This one is from about 80 kilometers northeast of a site adjacent to the Bonnet Plume River (Cody et al. 2000). The second specimen is about 115 kilometers southeast of a site adjacent to the northern Bonnet Plume River and northeast of a site in the Wernecke Mountains. Platanthera obtusata (Pursh) Lindl., Northern Bog THE CANADIAN FIELD-NATURALIST Vol. 115 Orchid — YUKON: Snake River, 65°01’N 133°07’W, J. Meikle 90-015, 19 July 1999 (Yukon Renewable Resources, photo DAO). This species is frequent throughout much of the Yukon Territory (Cody 1996). The nearest site to the location cited above is about 150 kilometers to the southwest. Spiranthes romanzoffiana Cham. & Schlecht., Hooded Ladies’-tresses — YUKON: fen, Turner Lake, 66°09’56.9"N 134°17’16.6’W, G. Brunner 51b-99, 1 July 1999 (Yukon Renewable Resources, photo DAO); fen, Turner Lake, 66°10’43.25”N 134°14°37.37°W, G. Brunner 58-99, 1 July 1999 (Yukon Renewable Resources, photo DAO). The specimens cited above are an extension of the known range in the Yukon Territory (Cody 1996) of about 125 kilometers east of a site adjacent to the Dempster Highway. SALICACEAE Salix alaxensis (Anderss.) Cov. ssp. longistylis (Rydb.) Hultén, Feltleaf Willow — YUKON: riverbar with sandy silt, La Biche River, 60°13’58”N 124°13’58’W, B. Bennett 98-133, 17 June 1998 (DAO). This willow is widespread throughout the Yukon Territory. The specimen cited above, however, extends the known range in the Territory into the extreme southeast about 75 kilometers east of a site about longitude 125°30’W mapped by Cody (1996). Salix arctophila Cockerell, Northern Willow — YUKON: fluvial fan with moss and peat, Ogilvie Mountains, Patrol Range, middle of Seela Pass, 64°04.58’N 139°51.301’W, W. J. Cody 36367, 36368, 20-22 July 1999 (DAO) (determined by G. Argus). Argus (1973) knew this rare willow in the Yukon Territory from only two localities, one on the Arctic coast and the other from the Macmillan Pass area near the Canol Road. Cody (1996) mapped three additional sites on the Arctic coast and two in the northern Richardson Mountains. The specimen cited above is a new record for central Yukon Territory about 415 kilometers southwest of the southernmost site in the Richardson Mountains and about 500 kilometers northwest of Macmillan Pass. Salix barrattiana Hook., Barratt’s Willow — YUKON: alpine meadow, creek bed and adjacent turfy slopes, Richardson Mountains, 66°44’N 135°52’W, Cody & Ginns 30896, 8 July 1982 (DAO) (determined by G. Argus). Cody (1996) knew this cordilleran species in the Yukon Territory as far north as the Ogilvie and Wernecke Mountains and then disjunct to the British Mountains. The specimen cited above, which is from the eastern part of the intermediate area, was unfortunately misidentified as S. planifolia ssp. pulchra which is a common species in the Richardson Mountains. Salix pedicellaris Pursh, Bog Willow — YUKON: bog, Jackfish, Lake, *66—°49752-938"N 135-50" 48.25” W, G. Brunner 61-99, 3 July 1999 (Yukon Renewable Resources, photo DAO). 2001 Douglas et al. (1981) knew this rare species in the Yukon Territory only from the vicinity of Watson Lake. Cody et al. (1998) reported new sites from the vicinities of the Beaver River and Frances Lake. The specimen cited above is an extension of the known range in the Territory of about 700 kilometers to the northwest of Frances Lake. To the east in the Mackenzie River valley Porsild and Cody (1980) knew this species only as far north as the vicinity of Fort Norman. Salix pyrifolia Anderss., Balsam Willow — YUKON: graminoid/forb area, Vittrewka Lake, 66°49°33.48”"N 135°28’47.03”"W, G. Brunner 26-99, 28 June 1999 (Yukon Renewable Resources, photo DAO). 7 Douglas et al. (1981) knew this rare species in the Yukon Territory from a single location at Palmer Lake in the southern Richardson Mountains. Cody (1996) mapped an additional area in the vicinity of Watson Lake and Cody et al. (2000) reported a collection from the Beaver River area in the extreme southeast. The specimen cited above is an extension of the known range in the Territory northwards from the Palmer Lake area of about 80 kilometers. MYRICACEAE Myrica gale L., Sweet Gale - YUKON: Turner Lake, 66°09°56”N 134°17°18.31°W, G. Brunner 54- 99, 1 July 1999 (Yukon Renewable Resources, photo DAO). Cody (1996) stated that this species was infrequent in the Yukon Territory north to about latitude 65°30’N adjacent to the Dempster Highway. Cody et al. (2000) extended the range further northward to 67°04’N 137°17’W. The speci- men cited above is from about 190 kilometers northeast of the Dempster Highway collection. POLYGONACEAE Polygonum alaskanum Wight ex Hultén — YUKON: fossil bluff, active fine scree bluff undercut by river, Snake River, 65°56’N 133°17.06’W, J. Meikle 99- 016, 22 July 1999 (Yukon Renewable Resources, photo DAO). East of the Dempster Highway, the only site between lat- itudes 64°N and 66°N known to Cody (1996) was north of the Peel River. The specimen cited above is from a site about 110 kilometers to the east. Polygonum buxiforme Small, Eastern Knotweed — YUKON: disturbed area near airstrip, La Biche Airstrip, 60°07°42”N 124°02’21”W, B. Bennett 98- 610, (B. Bennett Herbarium, photo DAO). Cody et al. (2000) extended the known range of this species in the Yukon Territory eastward from the vicinity of Johnson’s Crossing to the vicinity of the Beaver River. The specimen cited above is a further extension of about 60 kilometers to the east where Bruce Bennett suggests that it may be introduced. Polygonum lapathifolium L., Willow Weed — YUKON: uncommon in silty mud beside beaver pond, Ottertail Creek, west of Mt. Martin, 60°07’N 124°15’°20”"W, B. Bennett 98-609, 16 June 1998 (DAO). Cody (1996) knew this introduced weedy species in the Yukon Territory only from west of longitude 135°W where infrequent. Copy, KENNEDY, AND BENNETT: VASCULAR PLANTS IN YUKON III 311 Polygonum viviparum L., Alpine Bistort - YUKON: Snake River, 65°01’N 133°07’W, J. Meikle 99-017, 19 July 1999 (Yukon Renewable Resources, photo DAO). This species is widespread throughout the Yukon Territory (Cody 1996). The nearest site in the Territory to the specimen cited above is adjacent to the upper Bonnet Plume River, about 100 kilometers to the southeast. Rheum rhaponticum L., Rhubarb - YUKON: grow- ing wild around old buildings, Silver City on east side of Kluane Lake, 61°05’N 138°24’W, Cody & Cody 35685, 29 June 1999 (DAO). This cultivated species has not previously been reported growing wild in the Yukon Territory, but Bruce Bennett noted its occurrence on 17 July 1997 at 17 Mile (old town location 61°30’N 134°40’40”W) on the Teslin River. Rumex acetosa L. ssp. alpestris (Scop.) A. Love, Green Sorrel - YUKON: by creek, Yakamaw Creek watershed east of Angelcomb Peak, Tombstone Range, Ogilvie Mountains, 64°36’N 138°14’W, W. J. Cody 36822, 20-22 July 1999 (DAO); ridge, head- waters of Chandindu River, Patrol Range, Ogilvie Mountains, 64°45’N 139°06’W, W. J. Cody 36544, 36573, 20-22 July 1999 (DAO). Cody (1996) knew this rare species in the Yukon Territory from only three widely separated areas: northern Richardson Mountains, west of northern Kluane Lake and southwest of Watson Lake. Rumex salicifolius Weinm. ssp. triangulivalvis Danser, Willow Dock — YUKON: gravel beside building, Keno, 63°55’N 135°18’W, Cody & Cody 35781, 4 July 1999 (DAO). This species has not previously been found in the Mayo area. The nearest previously known localities (Cody 1996) were adjacent to the Klondike Highway about 125 kilome- ters to the west and in the vicinity of Carmacks about 200 kilometers south- southwest. PORTULACACEAE Montia fontana L., Blinks, or Water Chickweed — YUKON: in moist sand at back of beach, Herschel Island, Adavlek Fan, 69°34.5’N 138°15’W, W. J. Cody 36266, 17 July 1999 (DAO) (determined by Cody & McNeill). This is a circumpolar and often widely spaced species which was previously only known in the Yukon Territory from Shingle Point on the Arctic coast (Douglas et al. 1981; Cody 1996). The collection cited above is a new record for Herschel Island and is about 100 kilometers northwest of Shingle Point. It was a small mass of tiny plants recently sprouted from seeds which were still encas- ing leaves. CARYOPHYLLACEAE Cerastium nutans Raf., Nodding Chickweed — The drawing of this species in Cody et al., New Records of Vascular Plants in the Yukon Territory IT, Canadian Field-Naturalist 114(3): 430, 2000 was by Lee Mennell, not Valerie Fulford. Minuartia biflora (L.) Schinz. & Thell., Mountain Sandwort — YUKON: steep rocky slope above 312 stream, Ogilvie Mountains, Patrol Range, east of Chandindu River, 64°45’N 139°06’W, W. J. Cody 36602, 20-22 July 1999 (DAO). Cody (1996) stated that this species was occasional north to latitude 64°30’N in the Yukon Territory and then dis- junct to the Barn Mountains to the north. The specimen cited above is an extension of the known range in the Ogilvie Mountains of about 75 kilometers northwest from a site adjacent to the Dempster Highway. Moehringia lateriflora (L.) Fenzl, Blunt-leaved Sandwort — YUKON: Snake River, 64°57.16’N 133°00.21’W, J. Meikle 99-019, 19 July 1999 (Yukon Renewable Resources, photo DAO). This species is widespread throughout much of the Yukon Territory (Cody 1996). The nearest site to that cited above, however, is about 200 kilometers to the southwest in the vicinity of Mayo. Sagina saginoides (L.) Karst, Arctic Pearlwort — YUKON: near junction of Kaskawulsh, Dezadeash and Alsek rivers, 60°30’N-60°45’N 137°45’ W- 138°W, A. M. Pearson 69-10, 30 June 1969 (DAO). Douglas et al. (1981) did not include this species in The Rare Plants of the Yukon because it was “too widely dis- tributed”. Cody (1996), however, mapped it from only seven areas in the Territory, the nearest of which was in the vicinity of Haines Junction. Silene uralensis (Rupr.) Bocquet ssp. ogilviensis (A. E. Porsild) Brunton — YUKON: rich tall willow site, middle of Seela Pass, Patrol Range, Ogilvie Mountains, 64°42.58’N 139°51.304’W, W. J. Cody 36404, 20-22 July 1999 (DAO); rocky slope, Richardson Mountains, 66°16’N 135°48’W, Cody & Ginns 30403, 6 July 1982 (DAO) (determined by J.K. Morton); river bank, Firth River, British Mountains, 68°48’N 140°39°W, W. J. Cody 27123, 7 July 1980 (DAO) (determined by J. K. Morton.) Cody (1996) knew this rare endemic taxon in the Yukon Territory from only two sites adjacent to the Dempster Highway. The first specimen cited above is an extension of the known range of about 80 kilometers to the west south- west. The Richardson Mountains specimen extends the known range in the Territory about 250 kilometers to the northeast while the British Mountains collection extends the known distribution about 430 kilometers to the north. Stellaria longipes Goldie, Long-stalked Starwort — YUKON: Jasper Canyon, Snake River, 65°12’N 133°19°W, J. Meikle 99-020, 20 July 1999 (Yukon Renewable Resources, photo DAO). This is a common species throughout much of the Yukon Territory (Cody 1996). The nearest site to the specimen cited above, however, is in the Wernecke Mountains about 160 kilometers to the west. Stellaria media (L.) Vill., Chickweed — YUKON: common invasive garden weed, Burns Road, Whitehorse, B. Bennett 99-221, 26 July 1999 (DAO). This species which is introduced in North America from Eurasia was known to Cody (1996) in the Yukon Territory from only four localities, the nearest of which was adjacent to the Canol Road about 125 kilometers to the east of THE CANADIAN FIELD-NATURALIST Vol. 115 Whitehorse. It is now a well known weed of gardens in the Whitehorse area. RANUNCULACEAE Clematis tangutica (Max.) Korsh., Golden Clematis — YUKON: roadside gravel, across Carcross road from service station side of Highway near Nares Lake, 60°10’N 134°42’W, B. Bennett 99-543, 29 Aug. 1999 (B. Bennett Herbarium, photo DAO). The specimen cited above (which was from the only patch seen) is from about 65 kilometers south of Whitehorse, the only other location known in the Yukon Territory (Cody 1996). Ranunculus hyperboreus Rottb., Arctic Buttercup — YUKON: in muck border of lagoon near settlement, Pauline Cove, Herschel Island, 69°34’N 138°55’W, W. J. Cody 36188, 16 July 1999 (DAO). This species has not previously been reported from Herschel Island. It is frequent in the southern part of the Yukon Territory north to about latitude 64°30’N adjacent to the Dempster Highway. North of that locality Cody (1996) knew it only from the vicinity of the Bell River and on the Arctic coast in the vicinities of Komakuk Beach and Shingle Point. Thalictrum alpinum L., Alpine Meadow Rue — YUKON: SW facing slope, small carbonate face 70- 100 m above river, | km downstream from lower South Bend on east side of river, Snake River, 65°06.5’N 133°07’W, J. Meikle 99-23, 19 July 1999 (Yukon Renewable Resources, photo DAO). The specimen cited above is about 125 kilometers from a site to the southeast adjacent to the upper Bonnet Plume River and the same distance southwest to a site in the Wernecke Mountains (Cody 1996). BRASSICACEAE (CRUCIFERAE) Arabis boivinii G. A. Mulligan (Figure 3) — YUKON: gravel by lake shore, Kathleen Lake adja- cent to Haines Highway, 60°35’N 137°18’W, Cody & Cody 35906, 8 July 1999 (DAO) (determined by G. A. Mulligan); in gravel in partial shade, Dezadeash Camp Site, 51.3 km S of Haines Junction, 60°24’N 137°02’W, Cody & Cody 35698, 30 June 1999 (DAO) (determined by G. A. Mulligan). This species is new to the flora of the Yukon Territory and should be added to the list of rare species in the Territory (Douglas et al. 1981). A description and key is provided below: Biennial or short-lived perennial with a simple, compact caudex; stems erect, usually single, simple to few- branched, 30-60 cm high; cauline leaves entire to rarely few-toothed, glabrous to pubescent, sessile, mostly strongly sagittate-clasping stems, narrowly lanceolate, slender petio- late; lower surfaces of caudex leaves with sparse to dense sessile to short-stalked, branched, 3- to 4-parted trichomes mostly 0.35 mm wide; inflorescences semisecund, open; fruiting pedicels arcuate-spreading to arcuate-descending to strongly descending, straight to slightly arcuate, 4.0-6.5 cm long, 1.5-2.0 mm wide; style rudimentary. This native species is found in widely separated areas in North America: Quebec, Saskatchewan, southwestern 2001 FiGurE 3. Arabis boivinii (Drawn by Lee Mennell). Yukon Territory, Montana and South Dakota. Arabis boivinii can be separated from A. divaricarpa var. divaricarpa as follows: A. Undersurfaces of caudex leaves with unbranched, ses- sile or nearly sessile, 3-rayed trichomes with rays appressed to leaf surface without numerous prominent branches; inflorescences symmetrical; sliques spreading to ascending A. divaricarpa vat. divaricarpa . Undersurfaces of caudex leaves with short-stalked, 3- rayed trichomes with rays elevated above the leaf sur- faces with numerous prominent branches; inflores- cences semisecund to secund; siliques slightly to strongly descending A. boivinii _ A Arabis drummondii Gray, Drummond’s Rockcress — YUKON: stream edge, valley north of Robert Service Mountains [64°25’N 138°11’W], S. Kojima s.n., 28 June 1973 (DAO); valley floor, very wet wil- low/sedge Equisetum basin, Upper Bonnet Plume River Drainage near site #106, 64°25’53.6”"N 132°15’°11”°W, J. Staniforth 00-032, 5 July 2000 (DAO) (determined by G. A. Mulligan). Copy, KENNEDY, AND BENNETT: VASCULAR PLANTS IN YUKON III 313 Brooke and Kojima (1985) cited the first specimen above as the northernmost known in the Yukon Territory. It was unfortunately overlooked by Cody (1996) in the writing of the Flora. The second specimen matches the first for the northernmost in the east. Arabis holboellii Hornem. var. retrofracta (Grah.) Rydb., Holboell’s Rockcress - YUKON: gravel clearing by road, Dempster Hwy. Km 71, 64°23’N L38°25° W, Cedy-& Cody 35728, 2 July 1999 (DAO). The specimen cited above is the northernmost yet found in the Yukon Territory (Cody 1996). It is an extension of the known range of about 70 kilometers from near the Klondike Highway-Dempster Highway intersection. Arabis holboellii Hornem. var. secunda (Howell) Jepson, Holboell’s Rockcress - YUKON: dry steep slope across Wolf River, Wolf Lake, 60°42.6’N 131°44.08’W, B. Bennett 99-570A, 11 Aug. 1999 (DAO) (determined by G. A. Mulligan). Cody (1996) knew this taxon in the Yukon Territory from only four sites west of longitude 135°W, north to the vicinity of Dawson. The specimen cited above is an exten- sion of the known range in the Territory of about 250 kilo- meters to the southeast of a site adjacent to the Klondike Highway. Arabis nuttallii Robins., Nuttall’s Rockcress — YUKON: border of salt flats east of Takhini River on north side of Alaska Highway, 60°51’N 135°41°W, Cody & Cody 35677, 28 June 1999 (DAO); waste area adjacent to Bedrock Hotel, Mayo, 63°36’N 135°54’W, Cody & Cody 35776, 4 July 1999 (DAO); white flats below Elsa, 63°55’N 135°29°W, Cody & Cody 35785, 4 July 1999 (DAO) (determined by G. A. Mulligan). On the basis of two collections in the vicinities of the Tagish Road and Whitehorse, Cody (1994) stated that this species should be added to the list of rare species of the Yukon Territory (Douglas et al. 1981). The specimens cited above extend the known distribution in the Territory west- ward adjacent to the Alaska Highway and northward about 350 kilometers. Braya humilis (C. A. Mey.) Robins., Dwarf Braya — YUKON: Jasper Canyon, Snake River, 65°12’N 133°19°W, J. Meikle 99-024, 20 July 1999 (Yukon Renewable Resources, photo DAO); silty-sandy soil, broad channelized terrace at confluence of Snake River and tributary stream, 65°42.54’N 133°21.80° W,- J. Meikle’ 99-025, 22 July 1999 (Yukon Renewable Resources, photo DAO) (deter- mined by G. A. Mulligan). This species is scattered throughout much of the Yukon Territory (Cody 1996). The first specimen cited above is about 100 kilometers northwest of a site north of the Upper Bonnet Plume River. The second specimen is from about 140 kilometers southeast of a site in the Richardson Mountains. Descurainia incisa (Engelm. ex Gray) Britton var. incisa, Tansy Mustard — YUKON: disturbed ground adjacent to highway, overlooking Crag Lake, Tagish 314 Road Km 38, 60°14’N 134°30’W, Cody & Cody 35888, 7 July 1999 (DAO) (determined by G. A. Mulligan); open site growing amongst Dryas on course gravel, shallow to no soil development on ter- race about 3 m above active river course, appears occasionally flooded, Snake River, 65°54.77’N 133°30.50’W, J. Meikle 99-026B, July 1999 (Yukon Renewable Resources, photo DAO); soil slide, east bank of stream entering Summit Lake, Richardson Mountains, 67°42’N 136°28’W, J.G. Packer 1500, 1 Aug. 1961 (ALTA, photo DAO) (determined by G. A. Mulligan). This species is new to the flora of the Yukon Territory and should be added to the list of rare species in the Territory (Douglas et al. 1981). A description and key is provided below: Annual; stems single, usually branched above but some- times near the base, moderately pubescent to subglabrous; basal leaves pinnately divided, obovate in outline, 5-10 cm long, soon withering; cauline leaves reduced and less divid- ed upward, primary lobes narrowly obovate, these usually deeply dentate to incised; inflorescences elongated, termi- nating each branch; sepals yellowish to greenish yellow ca. 1.5 mm long; petals spatulate, yellow, ca. 2 mm long; fruit- ing pedicels 6-12 mm long; siliques terete, narrow, | mm or less wide, 6-12 mm long, slightly incurved, glabrous. This native species is found in western North America from Alberta and British Columbia south to New Mexico and California and is disjunct to the Yukon Territory and central District of Mackenzie. It can be separated from Descurainea sophia as follows: A. Leaves 2- to 3-pinnate; septa of the siliques with 3 dis- tinct nerves; introduced weed D. sophia A!. Leaves simply pinnate but pinnae usually dentate or deeply incised; septa nerveless or with 1 distinct nerve; native species but often weedy D. incisa vat. incisa Descurainia pinnata (Walter) Britt. ssp. nelsonii (Rydb.) Detling, (D. pinnata (Walter) Britt. ssp. fil- ipes sensu Hultén (1968)), Western Tansymustard — YUKON: disturbed ground adjacent to highway, Tagish Road Km 38, 60°15’N 134°30’W, Cody & Cody 35885, 7 July 1999 (DAO); gravel in front of buildings, Johnson’s Crossing, 60°29’N 133°18’W, Cody & Cody 35864A, 7 July 1999 (DAO) (deter- mined by G. A. Mulligan). Hultén (1968) mapped three localities in the Yukon Territory (Mayo, Kluane Lake and Whitehorse) but vouch- er specimens were not found in any Canadian herbarium by Douglas et al. (1981) when preparing the rare plant volume. Cody (1996) mapped specimens collected by D. A. Mitchell and J. M. Gillett in 1949 at Whitehorse. The speci- mens cited above extend the known range in the Territory to the south and east. Descurainia sophioides (Fischer) O. E. Schulz, Northern Tanseymustard — YUKON: open site grow- ing amongst Dryas on coarse gravel, shallow to no soil development on terrace about 3 m above active river course, appears occasionally flooded, Snake River, 65°54.77°N 133°30.50’ W, J. Meikle 99-026A, July 1999 (Yukon Renewable Resources, photo DAO) (determined by G. A. Mulligan). THE CANADIAN FIELD-NATURALIST Vol. 115 The specimen cited above is an extension of the known range in the Yukon Territory (Cody 1996) of about 175 kilometers to the east from a site adjacent to the Dempster Highway. Draba borealis DC., Northern Draba — YUKON: wet tundra, Clarence Lagoon, 69°37’N 140°49’W, W. J. Cody 26817, 5 July 1980 (DAO); hummocky lowland near coast, Avadlek Fan, Herschel Island, 69°34.5°N 139°15’°W, W. J. Cody 32280C, 17 July 1999 (DAO) (determined by G. A. Mulligan). This species is rare in the Yukon Territory north of lati- tude 65°02’N adjacent to the Dempster Highway. The speci- mens cited above are a northern extension of the known range in the Territory of about 500 kilometers. Sites previ- ously mapped by Cody (1996) adjacent to the Peel and Firth rivers have been revised to other species by G. A. Mulligan. Draba cinerea Adams, Gray-leaved Draba — YUKON: fertile slope adjacent to fox den, approx. 1.5 km E of Pauline Cove, Herschel Island, 69°34’N 138°55’W, W. J. Cody 36094, 36114B, 14 July 1999 (DAO) (determined by G. A. Mulligan). Cody (1996) mapped the northernmost four sites in the Yukon Territory of this species between 68°11’N and 68°34’ W. The specimens cited above are a northern exten- sion of the known range in the Territory of about 110 kilo- meters. Draba densifolia Nutt., Nuttall’s Draba — The drawing of this species in Cody et al. (2000), New Records of Vascular Plants in the Yukon Ter- ritory II, Canadian Field-Naturalist 114(3): 433, was by Valerie Fulford, not Lee Mennell. Draba lonchocarpa Rydb. var. vestita O. E. Schulz — YUKON: Kotaneelee Range, 60°14.31’N 124°07.71°W, B. Bennett 98-598, 19 June 1998 (DAO) (determined by G. A. Mulligan). This taxon is-new to the flora of the Yukon Territory and is disjunct from coastal British Columbia and the Alaskan Seward Peninsula. It can be distinguished from var. lon- chocarpa as follows: stem and pedicels frequently pubescent; stems 3 to 15 cm tall, with one or two leaves usually more than 2 mm broad; stellate hairs on lower sur- faces of leaves mainly stalked, mainly with eight or less rays; petals white; silicles appressed to stem, glabrous. Draba nemorosa L. var. leiocarpa Lindbl., Wood Whitlow-grass — YUKON: gravel banks of highway, Blackstone River culvert, Dempster Hwy. Km 77.9, 64°50’N 138°20’W, Cody & Cody 35738, 2 July 1999 (DAO) (determined by G. A. Mulligan). This nearly circumpolar species is found only occasion- ally in the Yukon Territory. The specimen cited above is the northernmost yet found in the Territory, about 30 kilo- meters north of the northernmost site known to Cody (1996) adjacent to the Dempster Highway. Draba scotteri G. A. Mulligan, Scotter’s Whitlow- grass — YUKON: Kluane National Park: near junc- tion of Kaskawulsh, Dezadeash and Alsek rivers, 60°30’N-60°45’N 137°45’W-138°N, A. M. Pearson 69-10B, 30 June 1969 (DAO) (determined by G. A. Mulligan). 2001 Cody (1996) knew only three sites in the Yukon Ter- ritory. The specimen cited above is the third known from Kluane National Park. Draba stenopetala Trautv., Star-flower Draba. — YUKON: alpine scree slope, Mount Skookum, off Annie Lake Road, 60°12’N 135°29°W, M. Whitley 98-014, 6 June 1998 (B. Bennett Herbarium, photo DAO); on exposed alpine plateau, Montana Mountain, 60°03’N 134°41’W, B. Bennett 99-086, 20 June 1999 (DAO) (determined by G.A. Mulligan). Douglas et al. (1981) considered this species rare in the Yukon Territory. The specimens cited above are an exten- sion of the known range in the Territory (Cody 1996) of about 140 kilometers to the east from the nearest site in Kluane National Park. Lepidium densiflorum Schrad. var. macrocarpum G. A. Mulligan - YUKON: Klondike River bank oppo- site Dawson, 64°04’N 139°27’W, Cody & Ginns 29020, 6 Aug. 1980 (DAO); disturbed gravel, N side of Pelly River at Ross River, Canol Rd. (N), 61°59’°N 132°27' W, W. J. Cody 26283, 25 June 1980 (DAO); by building, Ross River Forestry Station, 61°59°N 132°26’W, Cody & Ginns 28846, 3 Aug. 1980 (DAO); gravel by regrowth, Pelly Crossing, 62°49°N 136°34’°W, Cody & Cody 35790, 4 July 1999 (DAO) (determined by G. A. Mulligan). This variety is new to the flora of the Yukon Territory where it is possibly introduced. Elsewhere in Canada it is found from New Brunswick to British Columbia and south in the United States to Montana, Wyoming and Washing- ton. It can be separated from var. densifolium and var. elon- gatum as follows: A. Siliques puberulent, at least on the margins var. elongatum A!. Siliques glabrous B. Siliques averaging 2.5 mm long; pedicels slightly flattened, crowded, more than 9 peduncles per cm var. densifolium B!. Siliques averaging 3 to 3.5 mm long; pedicels con- spicously flattened, less crowded, usually less than 9 pedicels per cm var. macrocarpum Lepidium ramosissimum L., Branched Peppergrass — YUKON: gravel shore of lake, Nunatuk Camp- ground, Frenchman Lake, Cody & Cody 35805, 5 July 1999 (DAO) (determined by G. A. Mulligan). This species which Cody (1996) presumed to be intro- duced in the Yukon Territory was known to him from only three localities. The specimen cited above is intermediate between two sites adjacent to the Klondike Highway. Parrya arctica R.Br. - YUKON: slope overlooking bay, Pauline Cove Settlement, Herschel Island, 69°34’N 138°55’W, W. J. Cody 36030, 13 July 1999 (DAO) (determined by G. A. Mulligan). The only other collection of this species in the Yukon Territory was also from Herschel Island where it was found by P.F. Cooper at top of cliffs just east of Boot Eating Creek on 9 May 1979. It is endemic to the Canadian Arctic Archipelago (Cody 1994) and should be added to the list of Rare Plants in the Yukon Territory (Douglas et al. 1981). Copy, KENNEDY, AND BENNETT: VASCULAR PLANTS IN YUKON III aS Rorippa palustris (L.) Besser, Marsh Yellow Cress — YUKON: marsh, Taber Lake, 66°58’42.36”N 134°46°55.51°W, G. Brunner 2-99, 23 June 1999 (Yukon Renewable Resources, photo DAO) (deter- mined by G. A. Mulligan). The specimen cited above from the Peel River Valley is an extension of the known range in the Yukon Territory of about 175 kilometers to the northeast from a site adjacent to the Dempster Highway and 90 kilometers to the southeast from a site adjacent to the Rat River (Cody 1996). Thlaspi arvense L., Field Pennycress - YUKON: stony bank by house, Keno, 63°55’N 135°18’W, Cody & Cody 35783, 4 July 1999 (DAO). This species is an occasional introduction as far north in the Yukon Territory as Dawson. The specimen cited above from Keno is about 150 kilometers east of a site at the south end of the Dempster Highway and 150 kilometers northeast of a site adjacent to the Klondike Highway (Cody 1996). CRASSULACEAE Rhodiola rosea L. ssp. integrifolia (Raf.) Hara, Roseroot — YUKON: just north of Albert Creek near the Little Rancheria River, ca 60°10’N 129°30’W, B. Bennett 97-53, July 1995 (DAO). The specimen cited above is an extension of the known range in the Territory of about 150 kilometers south of a site just north of the Robert Campbell Highway known to Cody (1996). SAXIFRAGACEAE Boykinia richardsonii (Hook.) A. Gray, Richard- son’s Boykinia —- YUKON: on sub peak and adjacent valley to Mt. McDonald, Snake River, 64°43’N 132°44°W, J. Meikle 99-027, 17 July 1999 (Yukon Renewable Resources, photo DAO). The specimen cited above is the easternmost yet found in the Yukon Territory (Cody 1996) and is near the southern limit in the Territory. Saxifraga adscendens L. ssp. oregonensis (Raf.) Breitung, Wedge-leaved Saxifrage - YUKON: steep rocky slope above stream, Ogilvie Mountains, Patrol Range, east of Chandindu River 64°45’N 139°06’W, W. J. Cody 36603B, 20-22 July 1999 (DAO). This cordilleran taxon is scattered in southern Yukon Territory and then disjunct to a site adjacent to the Canol Road near the District of Mackenzie border and two sites north of latitude 64°N (Cody 1996). The specimen cited above, which is the northernmost yet found in the Territory, is from about 55 kilometers northwest of a site adjacent to the Dempster Highway. Saxifraga aizoides L., Yellow Mountain Saxifrage — YUKON: Jasper Canyon, Snake River, 65°24’N 133°24'W, J. Meikle 99-028, 20 July 1999 (Yukon Renewable Resources, photo DAO); undulating spruce/shrub/lowshrub/forb/moss, Upper Bonnet Plume River Drainage Site #105, J. Staniforth 00- 012, 5 July 2000 (DAO). Douglas et al. (1981) considered this species rare in the Yukon Territory. The specimens cited above were found about 75 kilometers northeast of a site between the Wind 316 and Bonnet Plume rivers and about 65 kilometers east of a site adjacent to the Upper Wind River. Saxifraga bronchialis L. ssp. funstonii (Small) Hultén, Spotted Saxifrage - YUKON: Jasper Can- yon growing in moist crag in limestone, Snake River, 65°24’N 133°24’W, J. Meikle 99-029, 20 July 1999 (Yukon Renewable Resources, photo DAO). The specimen cited above is the easternmost of this species yet found in the Yukon Territory (Cody 1996). The nearest collections are from the Richardson Mountains about 160 kilometers to the northwest and adjacent to lati- tude 64°N, 275 kilometers to the southwest. ROSACEAE Geum aleppicum Jacq. ssp. strictum (Ait.) Clausen, Yellow Avens — YUKON: Galkeno wetland, Keno City, 63°55.1’N 135°20.3’W, S. Withers 500, 10 Aug. 1999 (B. Bennett Herbarium, photo DAO). The specimen cited above is an extension of the known range of this species in the Yukon Territory (Cody 1996) northward about 230 kilometers from a site adjacent to the Klondike Highway. This species was considered rare in the Territory by Douglas et al. (1981). Potentilla norvegica L., Norwegian Cinquefoil — YUKON: marsh, Taber Lake, 66°58’42.36”"N 134°46’55.51”W, G. Brunner 3-99, 23 June 1999 (Yukon Renewable Resources, photo DAO). The specimen cited above is an extension of the known range of about 175 kilometers northeast of a site adjacent to the Dempster Highway (Cody 1996). To the northwest there is a site just south of latitude 68°N, a distance of about 225 kilometers. Potentilla rubricaulis Lehmann — YUKON: steep dry exposed cliff face, Upper Wolf River, 60°41.8’N 131°47.22’W, B. Bennett 99-442, 12 Aug. 1999 (DAO). This species was uncommon in the Yukon Territory west of longitude 133°W (Cody 1996). The specimen cited above is an extension of the known range in the Territory of about 150 kilometers to the southeast of a site mapped adjacent to the Canol Road. Rubus chamaemorus L., Cloudberry — YUKON: Jasper Canyon, Snake River, 65°24’N 133°24’W, J. Meikle 99-031, 20 July 1999 (Yukon Renewable Resources, photo DAO). This is a widespread species in the Yukon Territory (Cody 1996). The nearest collections to the specimen cited above are from south of the Peel River about 140 kilome- ters to the northwest and adjacent to the Bonnet Plume River about 150 kilometers to the southeast. Rubus pubescens Raf., Dwarf Raspberry - YUKON: open tall Abies/ feathermoss forest, approx. 5 km W of Rancheria, 60°06’25”N 130°41’05’W, C. Zoladeski 98-25-4, 1 Aug. 1998 (DAO). This is a rare species in the Yukon Territory (Douglas et al. 1981). Cody et al. (1998) cited a number of new sites in the southeastern part of the Territory. The specimen cited above is an extension of the known range in the Territory of about 120 kilometers west from south of Upper Liard Village that is west of Watson Lake. THE CANADIAN FIELD-NATURALIST Vol. 115 Sorbaria sorbifolia A. Braun, False Spiraea (Figure 4) — YUKON: garden escape, Dawson, S.M. Landhausser s.n., 30 July 1992 (DAO); found grow- ing and spreading in an empty lot near Robert Service cabin, Dawson City, 64°03.23’N 139°25.8°W, B. Bennett 99-253, 1 Aug. 1999 (B. Bennett Herbarium, photo DAO). This shrub is introduced from eastern Asia. In Canada it has been found escaped from cultivation from Newfound- land to Alberta and is now new to the flora of the Yukon Territory. The genus Sorbaria can be distinguished from Spiraea as follows: ’ A. Leaves simple, stipules absent or caducous A'. Leaves compound, stipulate Spiraea Sorbaria FABACEAE (LEGUMINOSAE) Astragalus adsurgens Pall. ssp. robustior (Hook.) Welsh, Standing Milk-vetch - YUKON: gravel of roadstop, 19 km from Stewart Crossing on Silver Trail Hwy., Cody & Cody 35764, 3 July 1999 (DAO). This is a rare taxon in the Yukon Territory (Cody 1996). The specimen cited above is an extension of the known range of about 220 kilometers to the northwest from a site adjacent to Faro on the Campbell Highway. Astragalus cicer L., Chick-pea Milk-vetch — YUKON: roadside photograph, Haines Highway near Klukshu, +60°17’30”’N 137°W, B. Bennett s.n., Oct. 1995 (DAO). The only other record of this introduced species in the Yukon Territory is from a roadside adjacent to the La Biche River in the extreme southeast (Cody et al. 1998). FiGurE 4. Sorbaria sorbifolia (Drawn by Lee Menell). 2001 Caragana arborescens Lam., Common Caragana — YUKON: ornamental shrub, commonly escaping cultivation, Dawson, 64°03.23’N 139°25.8’W, B. Bennett 99-251, 1 Aug. 1999; commonly escaping shrub, Whitehorse, 60°42.84’N 135°02.95’W, B. Bennett 99-220, 26 July 1999 (B. Bennett Herbar- ium, photo DAO). Cody et al. (2000) reported the finding of this escaping shrub in the Dawson area in 1949. The specimens cited above suggest its continued escaping in Dawson and also in Whitehorse. Medicago falcata L., Yellow Lucerne — YUKON: roadside, north Klondike Hwy., Km 478, 63°02.76’N 136°25.72’W, B. Bennett 99-294, 31 July 1999 (DAO); Faro-Campbell Hwy., 62°14’N 133°20’W, Cody & Cody 35810a, 5 July 1999 (DAO); parking area adjacent to highway, Haines Hwy. Km 202.3, 60°30’N 137°03’W, Cody & Cody 35974, 9 July 1999 (DAO). Cody (1996) recorded only four locations where this introduced plant had been found. An additional three loca- tions are cited above. This is now widespread in southwest- ern Yukon. Medicago sativa L., Alfalfa - YUKON: numerous collections by B. Bennett of this species were made from adjacent to the Klondike, Campbell and Alaska highways between longitudes 134°W and 138°34’W during the summer of 1999 where presumably they were the result of roadside seeding. Prior to that it was only known from the vicinity of Whitehorse (Cody 1996) and in the La Biche area (Cody et al. 1998). Melilotus alba Desv., White Sweet-clover — YUKON: gravel bank by Klondike Hwy., 61°32’N 135°57’W, Cody & Cody 35713, 30 June 1999 (DAO); found on Front Street, Dawson City, 64°03.23’N 139°25.8’W, B. Bennett 99-252, 1 Aug. 1999 (DAO); small patch, Dempster Hwy. Km 22, 64°06.93’N 138°32.42’W, B. Bennett 99-272, | Aug. 1999 (DAO). Cody (1996) knew this introduced species as far north as about latitude 62°20’N adjacent to the Klondike Highway and then disjunct to the vicinity of Mayo. The specimens cited above extend the known distribution in the Territory north to Dawson and the southern Dempster Highway. Melilotus officinalis (L.) Lam., Yellow Sweet-clover — YUKON: roadside, North Klondike Hwy. Km 478, North of Pelly Crossing, 63°02.76’N 136°25.72’W, B. Bennett 99-293, 31 July 1999 (DAO); roadside, North Klondike Hwy. Km 666, Klondike Hill, 63°55.33’N 138°33.31’W, B. Bennett 99-255, 1 Aug. 1999 (DAO); Guggieville, Dawson City, 64°02.49"N 139°23.34’W, B. Bennett 99-254, 1 Aug. 1999 (DAO). Cody (1996) knew this introduced species as far north on the Klondike Highway as the vicinity of Carmacks and then disjunct to the vicinity of Mayo. The specimens cited above extend the known distribution in the Yukon Territory north- west along the Klondike Highway to Dawson. Copy, KENNEDY, AND BENNETT: VASCULAR PLANTS IN YUKON III S17 Oxytropis campestris (L.) DC. ssp. jordalii (A.E. Porsild) Hultén, Late Yellow Locoweed — YUKON: moist pockets in talus, Kotaneelee Range, 60°14’°31”"N 124°07°19"W, B. Bennett 98-607, 19 June 1998 (DAO). Cody (1996) knew this species in the Yukon Territory to be frequent north of latitude 64°N and rare in the south in the Carcross area. The specimen cited above is about 575 kilometers east of the Carcross region and 650 kilometers southeast of the nearest northern site. It is, however, known in the southern Mackenzie Mountains to the east. Oxytropis deflexa (Pall.) DC. ssp. foliolosa (Hook.) Cody, Pendant-pod Locoweed — YUKON: growing in sandy soil on broad semi-active floodplain, Snake River, 65°42.54’°N 133°21.80’W, J. Meikle 99-034, 22 July 1999. This species is widespread in the Yukon Territory (Cody 1996). The nearest site to the specimen cited above is about 140 kilometers to the northwest. Oxytropis deflexa (Pall.) DC. ssp. sericea (T.& G.) Cody, Pendant-pod Locoweed — YUKON: in road- side gravel on edge of river, Ogilvie Mountains near Ogilvie River, B. Bennett 99-032, 15 June 1999 (DAO). Cody (1996) knew this taxon in the Yukon Territory only from the vicinity of Dawson and south of latitude 64°N. The specimen cited above is an extension of about 150 kilometers north of the Dawson area adjacent to the Dempster Highway. Oxytropis scammaniana Hultén, Scamman’s Locoweed — YUKON: lower carboniferous gray shale, head of Kandik River, 65°41’N 140°28’W, O. Hughes s.n., 18 Aug. 1962 (DAO). Cody (1996) knew this species in the Yukon Territory only as far north as the southwestern Ogilvie Mountains. The specimen cited above is an extension of the known range in the Territory of about 150 kilometers to the north- west. Vicia cracca L., Tufted Vetch — YUKON: roadside, downtown 6" and Hansen, Whitehorse, 60°42.84’N 132°02.95’W, B. Bennett 99-112, 14 July 1999 (DAO). Cody (1996) knew this introduced species in the Yukon Territory only from a collection in the vicinity of Dawson. Cody et al. (2000) reported a second collection from Wye Lake south of Watson Lake. APIACEAE (UMBELLIFERAE) Angelica lucida L., Seacoast Angelica - YUKON: along creek, Haines Hwy. Km 145.8, !/. km N of British Columbia border, 60°05’N 136°51’W, Cody & Cody 35960, 9 July 1999 (DAO); low moist area below road, Haines Hwy. Km 158 south of Million Dollar Falls, 60°06.5’N 136°56’W, Cody & Cody 35957, 9 July 1999 (DAO); many small patches in stony soil at back of beach in partial shade of Salix and Populus, about 12 m back from lake shore and adjacent to gravel roadside in partial shade of Populus balsamifera, Dezadeash Camp Site, 60°24’N 137°02’W, Cody & Cody 35692, 29 June 318 1999 (DAO); in open area adjacent to Populus tremuloides, Haines Hwy. north of Dezadeash Lake, 60°34’N 137°08’W, Cody & Cody 35975, 9 July 1999 (DAO). Douglas et al. (1981) knew this rare species in the Yukon Territory only from southeastern Kluane National Park and Macmillan Pass. Cody (1994) and Cody et al. (1998, 2000) added three additional locations, including the extreme southeast of the Territory. As can be seen from the collec- tions cited above, this species is quite frequent in the terrain adjacent to the Haines Highway north of the British Columbia border to latitude 60°34’N. Cicuta virosa L., European Water-Hemlock — YUKON: bog; Taber Lake, 66°53 ;30,19°N 134°45’50.66”W, G. Brunner 7-99, 25 June 1999 (Yukon Renewable Resources, photo DAO); marsh, Vittrewka Lake, 66°49’8.68”N 135°29’10.32”W, G. Brunner 32-99, 29 June 1999 (Yukon Renewable Resources, photo DAO); wet creek bank with wil- low, Mica Creek, 62°48.1’N 136°34.2’W, S. Withers SWOO0-052, 14 July 2000 (DAO). Cody (1996) knew this species from scattered sites in the Yukon Territory north to Mayo and Dawson and then dis- junct to the western Porcupine River area. The first two specimens cited above from the Peel River wetlands are from sites about 375 kilometers north of Dawson and Mayo © and 200 kilometers southeast of the Porcupine River area. The third specimen is from a site about 100 kilometers southwest of the vicinity of Mayo. PYROLACEAE Pyrola minor L., Lesser Wintergreen — YUKON: well drained alluvial soil on bench 3 m above river, Snake River, 65°01’N 133°07’W, J. Meikle 99-039, 19 July 1999 (Yukon Renewable Resources, photo DAO); steep slope, Ogilvie Mountains, Patrol Range, east of Chandindu River, 64°45’N 139°06’W, W. J. Cody 36582, 20-22 July 1999 (DAO); Vittrewka Lake, 66°49’28.26"N 135°29°45.19"W, G. Brunner 18-99, 27 June 1999 (Yukon Renewable Resources, photo DAO). Cody (1996) knew this species in the Yukon Territory only as far north as about latitude 64°30’N from sites in the Wernecke Mountains, adjacent to the Dempster Highway, and west of Dawson adjacent to the Top of the World Highway. The specimens cited above are extensions of the known range about 70 kilometers northwest of the Dempster Highway site, 115 kilometers northeast of the Wernecke Mountains site and about 275 kilometers north of the Wernecke Mountains site. ERICACEAE Cassiope tetragona (L.) D. Don ssp. saximontana (Small) A. E. Porsild, Four-angled Mountain-heather — YUKON: hummocky terrain, White Hat Hills, Tombstone Range, Ogilvie Mountains, 64°45.20’N 138°29.113’W, W. J. Cody 36431, 20-22 July 1999 (DAO); open Picea glauca wooded slope, 10 miles north of Chapman Lake, Ogilvie Mountains, 64°58’N 138°2.5’°W, Cody & Ginns 33972, 8 July 1984 (DAO). Cody (1996) mapped two sites north of latitude 64°N in THE CANADIAN FIELD-NATURALIST Vol. 115 the Yukon Territory: west of Dawson adjacent to the 60 Mile Road (Top of the World Highway) and east of the Dempster Highway in the Wernecke Mountains. Unfortunately, the second site was incorrectly mapped and should have been the second locality cited above. The two locations cited above extend the known range in the Territory north of a site near the junction of the Klondike and Dempster highways about 120 kilometers. Kalmia polifolia Wang., Bog-laurel - YUKON: undulating slope, “Tree Arc” west of Chapman Lake, Tombstone Range, 64°51’N 138°30’W, W. J. Cody 36787, 20—22 July 1999 (DAO). Cody (1996) unfortunately did not map a location adja- cent to the Dempster Highway cited by Brooke and Kojima (1985). The specimen cited above is an extension of the known range in the Ogilvie area of about 80 kilometers to the north. Cody et al. (2000) did, however, report a site north of latitude 67°N in the Porcupine River area. Phyllodoce X intermedia (Hook.) Rydb. - YUKON: in moist woodland, Mile 1406 Haines Road (near Kathleen Lake), J. Y. Tsukamoto s.n., July 1961 (DAO). The specimen cited above was unfortunately missed dur- ing the preparation of the Flora of the Yukon Territory (Cody 1996). This hybrid between Phyllodoce empetri- formis and P. glanduliflora can be distinguished by its somewhat glandular calyx and the corolla pinkish and glabrous to slightly glandular. PRIMULACEAE Androsace septentrionalis L. — Fairy-candelabra — YUKON: in gravel on road that goes down to creek, Dempster Hwy. near Km 100, 64°49.144’°N 138°21.023’W, B. Bennett 99-006, 15 June 1999 (DAO). This species was known to Cody (1996) mostly south of latitude 64°N and then disjunct to the Porcupine River. The specimen cited above is an extension of the known range northward along the Dempster Highway of about 95 kilo- meters. LAMIACEAE (LABIATAE) Galeopsis tetrahit L. ssp. bifida (Boenn.) Fries, Hempnettle - YUKON: growing in garden, Granger neighbourhood, City of Whitehorse, 60°42’N 135°05’W, J. McIntyre s.n., 27 Sept. 1999 (DAO). This is the third locality in the Yukon Territory where this species that is introduced from Eurasia has been found (Cody 1996; Cody et al. 1998). Mentha arvensis L., Field Mint - YUKON: bog, Taber Lake, 66°59’20.87”"N 134°44’29.67”"W, G. Brunner 4-99, 24 June 1999 (Yukon Renewable Resources, photo DAO). Cody (1996) knew this species in the Yukon Territory north almost to latitude 64°N. The specimen cited above is an extension of the known range in the Territory of about 360 kilometers from a site north of Mayo. SCROPHULARIACEAE Castilleja caudata (Pennell) Rebr. - YUKON: coarse gravel in thin to little soil on high energy riverbank, Snake River, 65°06’N 133°05’W, J. 2001 Meikle 99-042, 19 July 1999 (Yukon Renewable Resources, photo DAO). The specimen cited above is an extension of the known range in the Yukon Territory (Cody 1996) of about 175 kilometers to the southeast from a site in the southern Richardson Mountains. Castilleja miniata Dougl. ex Hook., Scarlet or Common Red Paintbrush — YUKON: brush, gravelly roadbank, Top of the World Hwy., 29.9 mi W of W. Dawson ferry docking, M. Egger 432, 4 July 1991 (WTU, photo DAO). Scotter and Cody (1979) reported the first record of this rare species in the Yukon Territory from the vicinity of Larsen Creek in the extreme southeast. C. E. Kennedy col- lected another specimen nearby adjacent to the Coal River Springs in 1983 (DAO) and Douglas et al. (1981) mapped a collection from the vicinity of Atlin Lake about 750 kilome- ters to the west. The specimen cited above was reported by Egger (1992) who suggested that this range extension of over 400 kilometers to the north was probably the result of seed being carried on an automobile tire from the south. The spec- imen cited above was unfortunately missed during the prepa- ration of the Flora of the Yukon Territory (Cody 1996). Castilleja pallida (L.) Spreng. ssp. candata Pennell — YUKON: moist pockets in talus, Kotaneelee Range, 60°14°31”"N 124°07°19"W, B. Bennett 98-567, 19 June 1998 (B. Bennett Herbarium, photo DAO) (determined by M. Egger). This species is widespread in the Yukon Territory west of longitude 134°W (Cody 1996). The specimen cited above is an extension of the known range of about 575 kilometers to the extreme southeast. Castilleja raupii Pennell, Raup’s Paintbrush — YUKON: in sandy soil over gravel on riverbank on braided channel, Snake River, 64°45.57’N 132°35.81°W, J. Meikle 99-044, 18 July 1999 (Yukon Renewable Resources, photo DAO); river- Bae ind River, 65°05.97’N 135°05.23’W, B. Bennett 00-462, 4 July 2000 (B. Bennett Herbarium, photo DAO); open well washed riverbar, Peel River, 65°55.97°N 135°02.99’W, B. Bennett 00-383, 9 July 2000 (B. Bennett Herbarium, photo DAO); riverbar, Peel River Camp #8, 65°56.03’N 134°58.84’W, B. Bennett 00-773, 9 July 2000 (B. Bennett Herbarium, photo DAO). The specimens cited above are an extension of the known range in the Yukon Territory (Cody 1996) of about 200 kilometers southeast of a site in the southern Richardson Mountains. Limosella aquatica L., Water Mudwort — YUKON: wet ditch with Callitriche verna, La Biche Airstrip, 60°07°42”N 124°02’21”’W, B. Bennett 98-579, 14 June 1998 (DAO). Douglas et al. (1981) knew this rare species in the Yukon Territory only from the vicinity of Mount Sheldon adjacent to the Canol Road. Cody (1994), Cody et al. (1998, 2000) added additional sites at the Nisutlin River Delta, Liard River south of Watson Lake and the Old Crow Flats. The specimen cited above is an extension of the known range of about 240 kilometers east of the Liard River location. Copy, KENNEDY, AND BENNETT: VASCULAR PLANTS IN YUKON III ato Pedicularis macrodonta Richards., Small-flowered Lousewort' — YUKON: fen, Turner Lake, 66°11°06.09"N 134°16’45.14’°W, G. Brunner 49-99, 2 July 1999 (Yukon Renewable Resources, photo DAO). This species is listed as rare in Douglas et al. (1981). Cody (1996) knew it only in the Yukon Territory from two localities in the extreme south. The specimen cited above from the Peel River wetland area is an extension of the known range northward about 700 kilometers. Pedicularis verticillata L., Whorled Lousewort — YUKON: open white spruce forest on terrace elevat- ed enough to avoid seasonal flooding, Snake River, J. Meikle 99-046, 65°01’N 133°07’'W, 19 July 1999 (Yukon Renewable Resources, photo DAO). The specimen cited above is an extension of the known range in the Yukon Territory (Cody 1996) of about 200 kilometers east of a site in the mountains adjacent to the Hart River. LENTIBULARIACEAE Pinguicula vulgaris L. ssp. vulgaris, Common Butterwort - YUKON: growing on wet moss on limestone, Jasper Canyon, Snake River, 65°24’N 133°24-W, J. Meikle 99-041, 20 July 1999 (Yukon Renewable Resources, photo DAO). The specimen cited above is an extension of the known range in the Yukon Territory (Cody 1996) of about 135 kilometers north of a site near the headwaters of the Wind River. Utricularia intermedia Hayne, Flat-leaved Bladder- wort — YUKON: bog, Turner Lake, 66°11°06.09”"N 134°16°45.14"W, G. Brunner 46-99, 2 July 1999 (Yukon Renewable Resources, photo DAO). Douglas et al. (1981) considered this circumpolar species to be rare in the Yukon Territory. Cody (1996) knew scat- tered collections north to about latitude 64°N and then dis- junct to the Porcupine River valley. The specimen from the Peel River wetlands area cited above is about 250 kilome- ters north of a site just north of Mayo and 280 kilometers southeast of the Porcupine River area. Utricularia vulgaris L. ssp. macrorhiza (LeConte) Clausen, Greater Bladderwort — YUKON: marsh, Taber Lake, 68°59’40.05”N 134°47’22.90”W, G. Brunner 6-99, 24 June 1999 (Yukon Renewable Resources, photo DAO). The specimen cited above from the Peel River wetlands area is from a site about 115 kilometers east of a site in the Richardson Mountains reported by Cody et al. (2000). ADOXACEAE Adoxa moschatellina L., Moschatel Cody (1996) stated that this circumpolar species was monotypic in the genus Adoxa and that Adoxa was the only genus of the family Adoxaceae. It has recently been brought to our attention (Hong Qian in correspondence 1999) that Liang and Wu (1995) state that the family Adoxaceae is now divided into 3 genera and four species, all of which occur in China: A. moschatellina L., A. orien- talis Nepomn., Sinadoxa corydalifolia C. Y. Wu, Z. L. Wu and R. F. Huang and Tetradoxa omeiensis (Hara) C. Y. Wu. 320 ASTERACEAE (COMPOSITAE) Antennaria densifolia A. E. Porsild, Dense-leaved Pussytoes — YUKON: Jasper Canyon, Snake River, 65°24’N 133°24’W, J. Meikle 99-048, 20 July 1999 (Yukon Renewable Resources, photo DAO); Wind River, 64°40.39’N, 64°48.46’N, 64°51.83’N, 65°06.83’N, 65°22.89’N, 65°36.03’N, B. Bennett 00-754, 00-316, 00-821, 00-158, 00-191, 00-170, 00- 277 (DAO). The specimens cited above are the easternmost record of this endemic species in the Yukon Territory (Cody 1996). The nearest collections are just north of latitude 66°N about 125 kilometers to the northwest, and adjacent to the Hart River about 175 kilometers to the west. This species is, however, known in the Mackenzie Mountains to the east (Porsild and Cody 1980). Antennaria pulcherrima (Hook.) Greene, Showy Pussytoes — YUKON: riverbar, Wind River Camp #1, 64°40.39°N 134°35.96’W, B. Bennett 00-352, 2 July 2000 (DAO); Canyon, Snake River, 65°12.32’N 133°19.70’W, J. Meikle 99-049, 19 July 1999 (Yukon Renewable Resources, photo DAO). The specimens cited above are the northernmost yet recorded in the Yukon Territory (Cody 1996). They are extensions of the known range in the Territory of about 100 kilometers to the north and northwest from a site adjacent to the upper Bonnet Plurne River. Artemisia michauxiana Bess. in Hook., Michaux’s Mugwort — YUKON: sand beach, Carcross, 60°10’N 134°42’W, Cody & Cody 35890, 7 July 1999 (DAO). This species, which was considered rare in the Yukon Territory by Douglas et al. (1981) was previously known in the Territory only from the vicinity of Little Atlin Lake about 50 kilometers to the northeast of the specimen cited above. Artemisia tilesii Ledeb. sl., Aleutian Mugwort — YUKON: collected amongst Dryas — widely braided channel, seasonally flooded, Snake River, 65°24.83’N 133°24.60’W, J. Meikle 99-050, 21 July 1999 (Yukon Renewable Resources, photo DAO). This is a widespread species in the Yukon Territory (Cody 1996). The specimen cited above, however, is from a site about 140 kilometers northwest of the nearest location adjacent to the upper Bonnet Plume River. Centaurea cyanus L., Bachelor’s-button (Figure 5) — YUKON: along railway tracks at Yukon River, Whitehorse, S. M. Landhaiisser s.n., 1992 (DAO); Dawson Dike along Yukon River, S.M. Landhaiisser s.n., 1992 (DAO). Not previously reported as occurring in the Yukon Territory; Douglas et al. (1989) stated that this species, which is introduced from the Mediterranean region, is a fre- quent garden escape along roadsides in southern British Columbia but in our experience it is not persistent in the Yukon Territory. Crepis elegans Hook., Elegant Hawksbeard — YUKON: growing in sandy soil on broad semi- active flood plain, Snake River, 65°42.51’N THE CANADIAN FIELD-NATURALIST Vol. 115 133°21.80’°W, J. Meikle 99-052, 22 July 1999 (Yukon Renewable Resources, photo DAO). The specimen cited above is an extension of the known range in the Yukon Territory (Cody 1996) of about 75 kilo- meters southeast of a site adjacent to the Peel River. Crepis tectorum L., Annual Hawk’s-beard — YUKON: built up river bar, Mayo, 63°36’N 135°54’W, Cody & Cody 35775, 3 July 1999 (DAO). The nearest site of this introduced species known to Cody (1996) is adjacent to the Klondike Highway in the vicinity of Minto, about 100 kilometers to the southwest. To the northwest it has also been found in the vicinity of Dawson. Erigeron acris L. ssp. politus (Fries) Schinz & Keller, Bitter Fleabane — YUKON: Snake River, 65°42.54’N 133°21.80’'W, J. Meikle 99-053, 22 July 1999 (Yukon Renewable Resources, photo DAO). The specimen cited above is an extension of the known range in the Yukon Territory (Cody 1996) of about 150 kilometers southeast from a site in the southern Richardson Mountains. Erigeron elatus (Hook.) Greene — YUKON: McDonald Camp on gravelly soil in semi-active floodplain of river, Snake River, 64°45.57’N 132°35.81’W, J. Meikle 99-054, 18 July 1999 (Yukon Renewable Resources, photo DAO); in grav- els in open riverbar, Wind River, 65°12.49’N 135°13.17’W, B. Bennett 00-459, 5 July 2000 (B. Bennett Herbarium, photo DAO); open riverbar in gravels, Illytd Creek confluence with Wind River, 65°30.07°N 135°22.88’W, B. Bennett 00-105, 6 July 2000 (B. Bennett Herbarium, photo DAO). The specimens cited above are extensions of the known range in the Yukon Territory (Cody 1996) of about 130 kilometers east and north from a site in the Wernecke Mountains. Petasites frigidus (L.) Fries ssp. frigidus, Sweet Coltsfoot - YUKON: in open White Spruce forest on terrace 3 m above river, Snake River, 65°01’N 133°07 W, J. Meikle 99-055, 19 July 1999 (Yukon Renewable Resources, photo DAO). The specimen cited above is an extension of the known range in the Yukon Territory (Cody 1996) of about 100 kilometers northwest of a site adjacent to the upper Bonnet Plume River. Senecio hyperborealis Greenm., Boreal Groundsel — YUKON: open dolomite delta, Wind River, 64°48.46°N 134°41.34’W, B. Bennett 00-225, 00- 294, 00-320, 3 July 2000 (DAO); in talus at base of cliff, Wind River, 65°12.49°N 135°13.17 Wot Bennett 00-368, 5 July 2000 (DAO); loose talus at base of steep active shale scree slope, junction of Wind and Peel rivers Camp #7, 65°50.48’N 135°18.25’W, B. Bennett 00-785, 00-809, 8 July 2000 (DAO); river bank, fine silty soil, below active scree slope, on seasonal floodplain, Snake River, 65°56’N 133°17'W, J. Meikle 99-057, 22 July 1999 2001 TEINS FiGuRE 5. Centaurea cyanus (Drawn by Iljvars Steins). (Yukon Renewable Resources, photo DAO); undu- lating dry hummocky tundra alpine slope, Upper Bonnet Plume River Drainage Site #116, 64°24’28.7°N 132°07°54’W, J. Staniforth 00-082, 7 July 2000 (DAO). This species, endemic to arctic and subarctic northwest- Copy, KENNEDY, AND BENNETT: VASCULAR PLANTS IN YUKON III 321 ern North America was not included in the rare plants of the Yukon Territory (Douglas et al. 1981) because it was too widespread. The specimens cited above are an exten- sion of the known range in the Territory (Cody 1996) of about 100 kilometers east and 225 kilometers southeast of sites in the southern Richardson Mountains. Senecio lugens Richards., Black-tipped Groundsel — YUKON: in hummocky Picea/Salix forest near river, Wind River, 64°34.19’N 134°25.77’W, B. Bennett 00-347, 2 July 2000 (DAO); in White Spruce forest on moss, Snake River, 64°57.16’N 133°00.21’W, J. Meikle 99-058, 19 July 1999 (Yukon Renewable Resources, photo DAO); undu- lating upper slope, Upper Bonnet Plume River Drainage Site #125, 64°32’45”N 132°50’48’W, J. Staniforth 00-098, 8 July 2000 (DAO). This is a common species throughout much of the Yukon Territory (Cody 1996). The specimens cited above are an extension of the known range in the Territory of about 150 kilometers northeast of the vicinity of Mayo. Sonchus arvensis L. ssp. uliginosus (Bieb.) Nyman, Perennial Sow-thistle - YUKON: roadside overlook- ing Teslin Lake, near Brooks Brook, 60°25.607’N 133°11.942’W, B. Bennett 99-538, 16 Aug. 1999 (DAO). Cody (1996) knew this species, introduced from Europe, from only four localities in the southern part of the Yukon Territory. Cody et al. (1998, 2000) added new sites from Whitehorse and the La Biche areas. The specimen cited above is from about 125 kilometers southeast of Whitehorse. Taraxacum carneocoloratum A. Nels. - YUKON: on ridge top, North Fork Pass, Ogilvie Mountains, Dempster Highway, 64°36’N 138°20’W, C. Parker 1191, 8 July 1984 (ALA, photo DAO). This species was described from specimens collected in the McKinley park area between Fairbanks and Anchorage in Alaska and was reported as new to the Flora of the Yukon Territory by A. E. Porsild (1975) on the basis of specimens collected by his brother R. T. Porsild adjacent to the Dempster Highway, but was unfortunately overlooked by Cody (1996). It should be added to the list of rare plants in the Territory (Douglas et al. 1981). The following descrip- tion is from Hultén (1968): Low-growing; leaves oblong- lanceolate, with 3-5 pairs of triangular, somewhat acute lobes; terminal lobe comparatively large, ovate-triangular, blunt; petioles pale at base; heads often wooly at base; outer bracts broadly ovate, purplish, scarious-margined, indis- tinctly corniculate; ligules up to 18 mm long, about | mm broad, pink to flesh-coloured; achenes spinulose-muricate at tip, with beak about as long as achene. Taraxacum officinale Weber ex Wiggers, Common Dandelion — YUKON: common along Alaska Highway right-of-way up to Haines Junction and north to Kluane Lake, south to Dezadeash Camp- ground, 60°24’N 137°02’W, Cody & Cody 35690, 29 June 1999 (DAO); Tombstone Campground, 64°23’N 138°25’W, Cody & Cody 35729, 2 July 1999 (DAO); gravel roadside, Dempster Hwy. Km 74, 63°25’N 138°20’W, Cody & Cody 35742, 2 July 1999 (DAO); 322 roadside gravel, Dempster Hwy. Km. 109, 64°48. 592’N 138°21.438’W, B. Bennett 99-078, 16 June 1999 (B. Bennett Herbarium, photo DAO). This invasive weed has expanded greatly since I trav- elled the Haines, Alaska and Klondike highways in the early 1980’s. It now occupies extensive areas along the roadsides in the southwest of the Territory and appears to be spreading northward up the Dempster Highway. Acknowledgments We thank Greg Brunner and John Meikle for mak- ing their collections from the Peel and Snake river areas available for the senior author; Robert Brooke for providing a loan of Dempster Highway plant specimens from his herbarium collection in Simon Fraser Herbarium (now perserved in Yukon Renew- able Resources Herbarium); Gerald A. Mulligan for the identification of Brassicaceae (Cruciferae) speci- mens; George Argus for the identification of Salix specimens; the artists Lee Mennell, Marcel Jomphe and Iljvars Steins for their artwork; Gordon Cody for assisting the senior author with collections in the summer of 1999 in the southern Yukon; the Government of Yukon, Department of Renewable Resources for logistical support for surveys on Herschel Island and in Tombstone Proposed Park; and especially Leslie Durocher, for the many hours inputting this information on her computer. We would also like to thank Paul Catling for reviewing an earlier version of this manuscript. Literature Cited Argus, G. W. 1973. The genus Salix in Alaska and the Yukon. National Museum of Natural Sciences, Ottawa, Publications in Botany, number 2. 279 pages. Brooke, R. C., and S. Kojima. 1985. An Annotated Vas- cular Flora of Areas Adjacent to the Dempster Highway, Central Yukon Territory IH. Dicotyledonae. Contribu- tions to Natural Science. British Columbia Provincial Museum, Victoria, British Columbia. 19 pages. Britton, D. M., and D. F. Brunton. 1995. Isoetes X trun- cata: a newly considered pentaploid hybrid from western Canada. Canadian Journal of Botany 71: 1016-1025. Brunton D. F., and D. M. Britton. 1999. Maritime Quill- wort, [soetes maritima (Isoetaceae), in the Yukon Ter- ritory. Canadian Field-Naturalist 113: 641-645. Cody, W. J. 1994. The flora of the Yukon Territory: Additions, range extensions and comments. Canadian Field-Naturalist 108: 428-476. Cody, W. J. 1996. Flora of the Yukon Territory. National Research Council Press, Ottawa, Ontario, Canada. 643 pages. Cody, W. J. 2000. Flora of the Yukon Territory. Second Edition. National Research Council Press, Ottawa, Ontario, Canada. Cody, W. J., C. E. Kennedy, and B. Bennett. 1998. New records of vascular plants in the Yukon Territory. Can- adian Field-Naturalist 112: 289-328. Cody, W. J., C. E. Kennedy, and B. Bennett. 2000. New records of vascular plants in the Yukon Territory II. Canadian Field-Naturalist 114: 417-443. THE CANADIAN FIELD-NATURALIST Vol. 115 Douglas, G. W., G. W. Argus, H. L. Dickson, and D. F. Brunton. 1981. The rare vascular plants of the Yukon. Syllogeus 28: 1-96. Douglas, G. W., G. B. Straley, and D. Meidinger. 1989. The vascular plants of British Columbia: Part 1 - Gym- nosperms and dicotyledons (Aceraceae through Cucur- bitaceae). British Columbia Ministry of Forests, Special | Report Series 1. 208 pages. Douglas, G. W., G. B. Straley, and D. Meidinger. 1990. The vascular plants of British Columbia: Part 2 — Dico- tyledons (Diapenseaceae through Portulacaceae). British Columbia Ministry of Forests, Special Report Series 2. 158 pages. Douglas, G. W., G. B. Straley, and D. Meidinger. 1991. The vascular plants of British Columbia: Part 3 — Dico- tyledons (Primulaceae through Zygophyllaceae). British Columbia Ministry of Forests, Special Report Series 3. 177 pages. Douglas, G. W., G. B. Straley, and D. Meidinger. 1994. The vascular plants of British Columbia: Part 4 — Mono- cotyledons. British Columbia Ministry of Forests, Special Report Series 4. 257 pages. Dyke, A.S., and V. K. Prest. 1987. Paleogeography of northern North America, 18,000-5,000 years ago. Map 1703A, Sheet 1: 18 000-12 000 years BP. Geological Survey of Canada, Ottawa. Egger, M. 1992. Yukon Territory, Castilleja miniata Douglas ex Hook. (Scrophulariaceae). Madrono 39: 244. Hughes, O.L., N.W. Rutter, and J.J. Clague. 1989. Yukon Territory (Quarternary stratigraphy and history, Cordilleran Ice Sheet): Pages 58-62, chapter 1 in Quaternary Geology of Canada and Greenland. Edited by R. J. Fulton. Geological Survey of Canada, Ottawa. 839 pages. Hultén, E. 1968. Flora of Alaska and Neighboring Terri- tories. Stanford University Press, Stanford, California. 1008 pages. Liang, Han-Xing, and Zheng—-Yi Wu. 1995. On the taxo- nomic system, phylogeny and distribution in Adoxaceae. Acta Botanica Yunnanica 17(4): 380-390. Porsild, A. E. 1951. Botany of southeastern Yukon adja- cent to the Canol Road. National Museum of Canada Bulletin 121. 400 pages. Porsild, A. E. 1975. Materials for a flora of central Yukon Territory. National Museum of Natural Sciences Publi- cations in Botany, number 4. 77 pages. Porsild, A. E., and W. J. Cody. 1980. Vascular plants of continental Northwest Territories, Canada. National Museum of Natural Sciences, Ottawa. 667 pages. Schweger, C. E. 1989. Paleocology of the western Can- adian ice-free corridor chapter 1, in Quaternary Geology of Canada and Greenland. Edited by R. J. Fulton. Geological Survey of Canada, Ottawa. 839 pages. Scotter, G. W., and W. J. Cody. 1979. Interesting vascu- lar plants from southeastern Yukon Territory. Canadian Field-Naturalist 93: 163—170. Sheviak, C. J. 1999. The identities of Platanthera hyper- borea and P. huronensis, with the description of a new species from North America. Lindleyana 14: 193-203. Received 21 November 2000 Accepted 24 April 2001 Observations of Change in the Cover of Polargrass, Arctagrostis latifolia, and Arctic Lupine, Lupinus arcticus, in Upland Tundra on Herschel Island, Yukon Territory C. E. KENNEDY! , C. A. S. SMITH?, and D. A. COOLEY? 'Fish and Wildlife Branch, Yukon Department of Renewable Resources, Box 2703, Whitehorse, Yukon Territory YIA 2C6 Canada ?Pacific Agri-Food Research Centre, Agriculture and Agri-Food Canada, 4200 Hwy 97, Summerland, British Columbia VOH 1Z0 Canada 3Fish and Wildlife Branch, Yukon Department of Renewable Resources, Box 600, Dawson City, Yukon Territory YOB 1G0 Canada Kennedy, C.E., C. A.S. Smith, and D. A. Cooley. 2001. Observations of change in the cover of Polargrass, Arctagrostis latifolia, and Arctic Lupine, Lupinus arcticus, in upland tundra on Herschel Island, Yukon Territory. Canadian Field-Naturalist 115(2): 323-328. Herschel Island is located in the southern Beaufort Sea off the northern coast of the Yukon Territory and has acted as an excellent observatory of environmental change in the Canadian western Arctic. Between 1986 and 1999 the percentage cover of Polargrass (Arctagrostis latifolia (R.Br.) Griseb.) was observed to increase from 1% to >5% in a dominant upland tundra vegetation type on Herschel Island. For example, the Arctic Willow/Dryas-Vetch type which occurs exten- sively on undulating portions of the island’s upland and is associated with periglacial processes leading to patterned ground formation, has undergone a notable physiognomic change. Initially characterized by low shrubs and forbs, it is now characterized by grass. Similarly, the cover value of Arctic Lupine (Lupinus arcticus S. Wats.) increased markedly on many of the sites inventoried within this same vegetation type. Temperature records from Inuvik in the nearby Mackenzie Delta, suggest that the last 15 years have been warmer and somewhat drier than the 30 year normal in the region. Much of this increased cover of Arctagrostis latifolia and Lupinus arcticus is occurring on the bare patches of soil (frost boils or mud hummocks) that make up the patterned ground features of the vegetation type, an indication that these soil features generated and maintained by frost churning have stabilized to some degree. We speculate that periodic stabi- lization would be consistent with lower soil moisture levels in the active layer of underlying Cryosolic soils, resulting from warmer, drier climatic conditions. Key Words: Polargrass, Arctagrostis latifolia, Arctic Lupine,Lupinus arcticus, Arctic tundra, Herschel Island, Yukon Territory. During the summers of 1985 and 1986 a 1:20 000 scale soil and vegetation mapping survey was con- ducted on Herschel Island (Smith et al. 1989), as part of a multi-disciplinary resource inventory. The island was established as a territorial park in 1987, as a result of a commitment made in the Inuvialuit Final Agreement, signed in 1984. Herschel Island lies off the north coast of the Yukon Territory in the Beaufort Sea (69’30” North, 139’00” West) approxi- mately 250 km northwest of Inuvik, NWT (Figure 1). The island is composed of glacial ice-thrust marine sediments (Mackay 1959; Rampton 1982) with no bedrock control to the topography. The most recent estimate of the age of the glaciation that led to the formation of the island is approximately 30 ka (Duk-Rodkin 1999). Soils are composed primarily of silt and clay with minor sands and gravels. The upland soils of Herschel Island have an active layer that averages 30 to 50 cm thick overlying ice-rich permafrost. The tundra vegetation cover is primarily comprised of Cottongrass tussocks, prostrate shrubs, diverse forbs and other graminoids, whose distribu- tion is controlled by soil moisture regime, exposure and degree of solifluction disturbance to the site (Smith et al. 1989). Vegetation data were collected during a five-week field survey in July and August 1985 and a one-week verification field survey in July 1986. Plant species occurrence and cover data from approximately 125 plots were summarized by tabular analyses into 11 vegetation types based primarily on the dominance of species in each strata. The undulating to level upland portion of the island is dominated by three vegetation types. Level, uneroded surfaces are cov- ered by “tussock tundra” dominated by Cottongrass (Eriophorum vaginatum) and a well developed moss layer dominated by Sphagnum species. Gently undu- lating surfaces associated with non-sorted patterned ground are dominated by a community of Arctic Willow (Salix arctica), with a mat of Dryas (Dryas integrifolia) and Milk-Vetch (Astragalus umbel- latus). On gentle slopes where there is less patterned ground, willows (Salix arctica and Salix reticulata), mosses and a diverse forb coverage including Milk- Vetch (Astragalus umbellatus), Bistort (Polygonum viviparum), Saxifrage (Saxifraga hirculus) and 323 324 THE CANADIAN FIELD-NATURALIST ny ao ces a a : ( 4 7 Herschel island | \/ WY ~ Re i WA { l Y) y . Ptarmigan Bay a 4 : I | Yy ! e Stokes Pt. | yak, | 4 | oe p t Ae t | ~— ! Ry eel mie e ¢ : s o | qv ge S Y \ ; my : . ny NS At) ry - uA Gm yi ig WW 3 ; F: . ‘ o [a H Oy 4B ! |2 ' S < h o 5 | E: zj | : 0 20 40 60 Kilometres | Coltsfoot (Petasites spp) is present. Eight other veg- etation types (i.e. communities) were defined for the island that covered the wetland environments (e.g. low centre polygons, thermokarst ponds), and the marine foreshore environment as well as the actively eroding slopes (Smith et al. 1989) on the island. In the summer of 1999 we returned to the island to set up vegetation monitoring plots as part of the International Tundra Experiment (ITEX) and had the opportunity to make observations on the general character of vegetation communities on the island. It became clear that very observable physiognomic changes due to species cover changes had occurred since 1985/1986 within at least one of the most widespread communities on the upland of the island. At first glance, much of the upland of the island FiGureE |. Location of study area, showing Yukon Territory (inset), Herschel Island and the adjacent mainland. Vol. 115 appeared to be grassier than described in 1985 large- ly as a result of an apparent increase in the cover of Arctagrostis latifolia. Here we describe our initial observations of the nature of this change in cover. Methods In the 1985/1986 field survey, vegetation plots were selected within homogenous vegetation stands. All vascular and non-vascular plant species were recorded for occurrence and percent cover (in percent- age classes) within a circular plot, approximately 20 metres in diameter. To quantify our initial visual impressions of cover change in the brief time on the island between 12 and 18 July 1999, we attempted to re-visit 1985 plot sites based on location points recorded on 1:10 000 black-and-white aerial pho- 2001 tographs and site photographs from the original sur- vey that we had with us. With a fair degree of confi- dence we located three sites, distributed around the island on a range of landform conditions associated with this vegetation type. On each of these sites, species occurrence and cover values were recorded by the senior author using the same plot sampling tech- nique as used in the original survey. Results and Discussion The vegetation community showing the marked difference in cover, Arctic Willow/Dryas-Vetch, occurs on the undulating portions of the upland asso- ciated with patterned ground formation. A species list and their basal coverage based on 21 plot observations in 1985/1986 is given in Table 1. Dryas integrifolia was present on all sites and had an average cover value of 24%. Salix spp. coverage was 13% and pre- sent on most sites. Numerous additional forbs and graminoids were present in low to trace amounts. Sites classified as belonging to this type had 15 to 65% of bare soil produced by active non-sorted circle KENNEDY, SMITH, AND COOLEY: POLARGRASS AND ARCTIC LUPINE ON HERSCHEL ISLAND 325 (frost boils) and net patterned ground associated with Orthic Turbic Cryosol soil formation (Smith et al. 1991). Bryophytes occur in the depressions between circles and can have total cover values from 10 to >20%. Moss species identified within this vegetation type include Timmia austriaca, Drepanocladus unci- natus, Pogonatum alpinum, Hylocomium splendens and Philonotis fontana. Many of these species are associated with calcareous substrates. Frost action in the soil tends to move calcareous soil parent materials to the surface and into the depressions between cir- cles: The 1999 cover values for graminoid species and one conspicuous forb, Lupinus arcticus are presented in Table 2. While such a small sample size does not allow for statistical comparisons between plot data from the two sample dates, visible changes in the veg- etation community were observed. In each of the three sites, Arctagrostis latifolia cover value increased, in one case (site 004) from 1% cover to 5% cover, in a second case (site 014) from not present to 1% cover and in a third case (site 308) from 5% cover to 10% TABLE |. List of the shrubs, forbs, graminoids and lichens that make up the Arctic Willow/Dryas-Vetch type based on 21 plots in 1985/1986. Layer Shrubs: Species Salix arctica Salix reticulata Forbs: Dryas integrifolia Astragalus umbellatus Astragalus alpinus Lupinus arcticus Parrya nudicaulis Myosotis alpestris Pedicularis capitata Senecio lindstroemii Polgonum viviparum Papaver spp. Saxifraga spp. Saxifraga punctata Saussurea angustifolia Oxytropis nigrescens Pedicularis lanata Draba spp. Minuartia obtusiloba Polygonum bistorta Pedicularis spp. Carex spp. Luzula nivalis Arctagrostis latifolia Poa spp. Cetraria cucullata Thamnolia subuliformis Alectoria ochroleuca Cetraria islandica Centraria nivalis Dactylia arctica Alectoria nigricans Graminoids: Lichens: Percent Percent Occurrence Cover 75 9 66 4 100 24 95 1 61 2 85 2 90 tr 80 tr 66 tr 70 tr 61 tr 70 tr 66 tr 52 tr 47 tr a2 tr 52 it 60 tr 52 tr 42 tr 60 tr 66 tr 47 tr 47 1 42 tr 85 3 84 5 84 tr a7 tr Si tr 2, tr 47 tr 326 THE CANADIAN FIELD-NATURALIST Vol. 115 TABLE 2. Summary of observed cover changes in selected species from three sites belonging to the Arctic Willow/Dryas- Vetch vegetation community. Values in bold show increased species cover between 1985 and 1999 observations. % Species Cover! Site 4 Site 14 Site 308 69°33.3’N, 139°09" W 69°35.4’N,139°10.37W 69°37.2’N,139°01.6’W Plant Species Year 1985 1999 1985 1999 1985 1999 Graminoids Arctagrostis latifolia 1 5 - 1 5 10 Carex lugens tr tr - - ls 15 Luzula nivalis tr tr _ _ 1 1 Eriophorum angustifolium 1 1 _ — tr tr Forbs Lupinus arcticus = 1 - - 1 10 1 = not present, tr = trace (<< 1%) cover. In 1985, this species was present on only 47% of sites classified as this type and with a 1% mean cover value. Similarly, Lupinus arcticus cover value increased markedly on two of the three sites invento- ried. In one case it occurred where it did not exist in 1986 and on another increased in cover from 1% cover to approximately 10%. Other graminoids that occur within the plant community (sedge, cottongrass, bluegrass) did not appear to have changed cover val- ues. Based on a limited number of plots and traverse observations through the upland of the island, areas surrounding patterned ground exhibited the most dramatic increase of Polargrass cover which seems to have established ubiquitous cover within the Arctic Willow/Dryas-Vetch vegetation type. Most notable was the physiognomic change of a commu- nity initially characterized by prostrate shrubs and forbs, to one characterized by grass (Figure 2). Incidental observations were also made in other veg- etation types on the uplands, eroded slopes and the marine foreshore; although some changes were observed, apparent change in cover among plant species appears to be relatively minimal and/or inconsistent. The Arctic Willow/Dryas-Vetch vegetation type is the most widespread on the island where it is the dominant type on the gently sloping uplands that make up almost 50% of the island’s total area (Smith et al. 1989). A change in vegetation structure and composition over this extent of the island may have ecological consequences for the Caribou (Rangifer TABLE 3. Summary of temperature and precipitation deviations from Inuvik, NWT between 1985 and 1999 (data for 1995 and 1996 are missing). Based on the 1960-1990 normals, values for mean annual tempera- ture is -9.5°C, annual precipitation is 257 mm, summer (June, July, August) mean temperature is 11.6°C and summer precipitation is 100 mm (Environment Canada 1993). Deviation from 30 year normal! Annual mean Summer mean temperature Annual precip temperature Summer precip Year (°C) (mm) (AG) (mm) 1985 +0.4 -91 -1.3 -69 1986 -0.6 -28 -0.1 -17 1987 +1.3 -2 +1.0 -16 1988 +1.4 -32 +1.4 5) 1989 +2.0 +29 +3.2 +16 1990 +0.1 -20 +0.6 -16 199] +0.8 +39 -0.9 +23 1992 +0.5 +20 +1.0 -27 1993 missing -8 missing -25 1994 +1.3 +4] +2.9 +2 1997 +2.5 -6 +1.5 +18 1998 +4.9 -1 +3.1 +18 1999 +0.7 -34 +0.4 0 'Data supplied by William Miller, Environment Canada, Whitehorse, Yukon Territory. 2001 KENNEDY, SMITH, AND COOLEY: POLARGRASS AND ARCTIC LUPINE ON HERSCHEL ISLAND 327 . eee saa 5 iii bins Be esti Race A eet omarniadtc aids oo oa einen RRR IRRES I iMES e t it i FIGURE 2. Cover changes in the Arctic Willow/Dryas-Vetch vegetation type are illustrated from one site typical of the upland on Herschel Island. Figure 2a (above) was taken on 4 August 1985. Figure 2b (below) was taken from the same site on 16 July 1999. The graminoid cover in 1999 is composed predominantly of Polargrass (Arctagrostis latifolia) . 328 tarandus) and Muskoxen (Ovibos moschatus) that use the island. The numbers of Caribou and Muskoxen grazing on the island have varied over the last decade. It will be necessary to examine specific grazing behaviors in order to assess the extent to which faunal impacts on vegetation cover values are responsible for the observed changes. We suspect that climatic variations over the two decades may be the most likely cause driving vegeta- tion change. Unfortunately, there are no continuous weather data available for this period from the auto- mated weather station on Herschel Island. Based on records from Inuvik in the Mackenzie Delta some 250 km to the east, the last 15 years have been remarkably warmer in the region and precipitation has been somewhat lower (Table 3). All but one year since 1985 have been warmer than the 30 year nor- mal; 1998 recorded a deviation in annual mean tem- perature of almost 5.0°C and summer soil thaw was abnormally deep in the region (Wolfe et al. 2000). Nine of the 13 years for which data are available, show below normal annual precipitation. Responses of tundra vegetation to experimental and observed changes in climate on the north slope of Alaska indi- cate that elevated temperature can increase nutrient. mineralization in the soil with a subsequent increase in plant nutrient availability (Chapin et al. 1995). Grasses are known to respond quickly to increased nutrient availability. Changes in plant biomass between 1985 and 1995 led Chapin et al. (1995) to conclude that regional climatic warming may already be altering the species composition of tundra in the Alaskan north slope region. We also noted that bare ground (frost boils or mud hummocks) generated by cryoturbation pro- cesses in the soil, seems to be undergoing invasion by Polargrass and Arctic Lupine, an indication that these features have stabilized to some degree. We know that both the degree and intensity of cryotur- bation in soil varies over several orders of time (Zoltai et al. 1978; Mackay and MacKay 1976). Periodic stabilization would be consistent with lower soil moisture levels in the active layer result- ing from warmer, dryer climatic conditions and would favour the growth of opportunistic species like Arctagrostis latifolia and Lupinus arcticus. Therefore, while the environmental changes that seem to be taking place on Herschel Island may not be outside of the natural variability of soil and vege- tation dynamics, understanding this variability is necessary in establishing a baseline for future ITEX monitoring on the island and in interpreting those observations. In addition to this long term monitor- ing, in the coming field seasons we will follow-up this initial observation of vegetation change with additional detailed plot work to quantify plant species and soil properties changes and determine THE CANADIAN FIELD-NATURALIST Vol. 115 the geographic extent of this phenomenon on upland plant communities in the region. Acknowledgments We thank the Government of Yukon, Parks and Outdoor Recreation Branch, Department of Renew- able Resources particularly Chief Park Ranger, Sandy Koep for logistical support getting to and moving around on Qikiqtaruk (Herschel Island) Ter- ritorial Park. We thank W.J. Cody, Research Branch, Agriculture and Agri-Food Canada for iden- tification of vascular plant species and for reviewing an earlier version of this manuscript. Funding for this research was provided by the Regional Manage- ment Branch, Yukon Department of Renewable Resources, and the Pacific Agri-Food Research Centre of Agriculture and Agri-Food Canada. Literature Cited Chapin, F.S., lil, G.R. Shaver, A. E. Giblin, K. J. Nadelhoffer, and J. A. Laundre. 1995. Response of arctic tundra to experimental and observed changes in climate. Ecology 76: 694-711. Duk-Rodkin A. 1999. Glacial limits map of Yukon Ter- ritory; Geological Survey of Canada, Open File 3694, Indian and Northern Affairs Canada Geoscience Map 1999-2, scale 1:1,000,000. Environment Canada. 1993. Canadian Climate Normals 1961-1990. Yukon and Northwest Territories. Atmos- pheric Environment Services, Ottawa. 58 pages. Mackay, J.R. 1959. Glacier ice-thrust features of the Yukon coast. Geographical Bulletin 13: 5-21. Mackay, J. R., and D. K. MacKay. 1976. Cryostatic pres- sures in nonsorted circles (mud hummocks), Inuvik, Northwest Territories. Canadian Journal of Earth Science 13: 889-897. Rampton, V.N. 1982. Quaternary geology of the Yukon coastal plain. Geological Survey of Canada Bulletin 317. Ottawa. 49 pages. Smith, C.A.S., C. A. Fox, and A. E. Hargrave. 1991. Development of soil structure in some Turbic Cryosols in the Canadian low arctic. Canadian Journal of Soil Science 71: 11-29. Smith, C.A.S., C.E. Kennedy, A. E. Hargrave, and K. M. McKenna. 1989. Soil and vegetation of Herschel Island, Yukon Territory, Yukon Soil Survey Report 1, Land Resource Research Centre, Agriculture Canada, Ottawa. 101 pages + maps. Wolfe, S. A., E. Kotler, and F.M. Nixon. 2000. Recent warming impacts in the Mackenzie Delta, Northwest Territories and Northern Yukon Territory coastal areas. Geological Survey of Canada, Current Research 2000- B1. Pages 1-9. Zoltai, S.C., C. Tarnocai, and W. W. Pettapiece. 1978. Age of cryoturbated organic material in earth hummocks from the Canadian arctic. Pages 325-331 in Proceedings of the Third International Conference on Permafrost. Edmonton, Alberta. Received 11 December 2000 Accepted 14 June 2001 Survey of Freshwater Mussels in the Petitcodiac River Drainage, New Brunswick JOHN MARK HANSON and ANDREA LOCKE Gulf Fisheries Centre, Department of Fisheries and Oceans, P. O. Box 5030, Moncton, New Brunswick, E1C 9B6, Canada Hanson, John Mark, and Andrea Locke. 2001. Survey of freshwater mussels in the Petitcodiac River drainage, New Brunswick. Canadian Field-Naturalist 115(2): 329-340. Five species of freshwater mussels were collected in a survey of 66 sites in the Petitcodiac River system, New Brunswick. One tributary, the North River, was almost devoid of all species — apparently due to habitat degradation associated with poor agricultural practices. The Eastern Pearlshell (Margaritifera margaritifera Linnaeus 1758) was common in most of the running water portion of the river system and was the only species present at 22 sites. The Eastern Floater (Pyganodon cataracta Say 1817) was restricted to habitats with fine sediments in the main Petitcodiac River and the lower 2 km of the Little River, and in two reservoirs. The Eastern Elliptio (Elliptio complanata Lightfoot 1786) was common in areas of sand or fine gravel in slower-flowing sections of the main Petitcodiac River and the Anagance River but was absent from the Little and Pollett rivers and Turtle Creek. The Brook Floater (Alasmidonta varicosa Lamarck 1819) was locally common on sand and fine gravel substrates in the lower 6 km of the Little River and also occurred in widely scattered patches in the main Petitcodiac River. The Triangle Floater (A. undulata Say 1817) shared the same general distribution as the Brook Floater but we seldom found more than five individuals of it at any site. Key Words: Mollusca, Unionidae, abundance, distribution, habitats. North America has the highest diversity of fresh- water mussels in the world, yet these molluscs repre- sent the most threatened taxonomic group in North America (Williams et al. 1993; Primack 1998). There are 297 recognized species and subspecies in the USA and Canada, of which only 70 are consid- ered to have stable populations. Although 53 species have been recorded in Canada (Williams et al. 1993; Metcalfe-Smith et al. 1998a), the conservation status of most freshwater mussel species in Canada has not been directly assessed by the Committee on Status of Endangered Wildlife in Canada (COSEWIC). Halt- ing the decline in freshwater mussel populations requires knowledge of the current abundance and distribution as well as an understanding of habitat requirements of both the freshwater mussel species and the fish hosts necessary for completion of the life cycle. Assessing the status of freshwater mussel species in Canada is difficult because few compre- hensive surveys of river systems have ever been con- ducted. A notable exception is the Lower Great Lakes Basin in Ontario, which is the site of ongoing assessments of a number of freshwater mussel species whose Canadian distribution is limited to this area (e.g., Metcalfe-Smith et al. 1998a,b; Staton et al. 2000; West et al. 2000). The Maritime Provinces also contain freshwater mussel species not found elsewhere in Canada. As with many parts of Canada, the published information needed to assess the status of the freshwater mussel fauna of the Maritime Provinces is almost non-existent. Clarke (1981) listed 12 species of freshwater mus- sel in the Maritime Provinces. One of these, the Dwarf Wedgemussel (Alasmidonta heterodon) is now con- sidered extirpated in Canada (Hanson and Locke 2000). The conservation status in Canada of the remaining 11 species has not been assessed by COSEWIC. The American Fisheries Society has list- ed two of these 11 species as threatened (Brook Floater Alasmidonta varicosa, Yellow Lampmussel Lampsilis cariosa), and three are listed as of special concern (Triangle Floater Alasmidonta undulata, Eastern Pearlshell Margaritifera margaritifera, Tidewater Mucket Leptodea ochracea) (Williams et al. 1993). The other five unionid species found in the Maritimes are the Alewife Floater (Anodonta implica- ta), Eastern Lampmussel (Lampsilis radiata radiata), Eastern Floater (Pyganodon cataracta), Newfound- land Floater (Pyganodon fragilis), and Creeper (Strophitus undulatus). The Eastern Pearlshell is unique among North American freshwater mussels because it also occurs in Europe. In Europe, this species is listed as endangered or extirpated over much of its range (Cosgrove et al. 2000). The current distributions of freshwater mussel pop- ulations in Maritime waters likely differ substantially from those described by Clarke decades ago (1981). The data summarized by Clarke were obtained from preliminary surveys (mostly unpublished) conducted in the 1950s and early 1960s. Published information on distributions of freshwater mussels in New Brunswick is very rare (e.g., Athearn 1961) while a small number of published studies exist for rivers in Nova Scotia (Athearn and Clarke 1961; Clarke and Rick 1963). However, these surveyors seldom sam- pled more than two or three sites per river and conse- quently may have missed species that were present at low densities or in very specific habitats. We are not 329 330 aware of any studies of freshwater mussels on Prince Edward Island. Environmental conditions have also changed a great deal since the studies mentioned above. In the past 40 or 50 years, many dams and causeways have been built in the Maritimes (Wells 1999). Blockage of rivers or impoundment is one of the major causes of loss of freshwater mussel popula- tions as it usually results in drastic habitat changes above and below the dam and the inability of fish hosts to pass by the obstruction (Bogan 1993; Watters 1995; Vaughn and Taylor 1999). In addition, there is the possibility that new species of freshwater mussel have become established in Maritime rivers, as a result of numerous introductions of freshwater fishes for sport fishing, some of which may have car- ried glochidia. During a previous study aimed at assessing the conservation status of the Dwarf Wedgemussel (Hanson and Locke 2000), we conducted a thorough survey of the entire freshwater portion of the Petitco- diac River system. To the best of our knowledge, this was the first time an entire river system in the Maritime Provinces was extensively surveyed for freshwater mussels. This report provides a baseline study to document the relative abundance and distri- bution of freshwater mussels in the Petitcodiac River system based on surveys conducted during 1997-— 2000. Study Site The Petitcodiac River is located in southeastern New Brunswick at the head of the Bay of Fundy (Figure 1). In 1968, the Petitcodiac River estuary was blocked by a causeway (46° 04.13’ North; 64°48.62’ West) located 21 km below the former head of tide, and a long, narrow, freshwater headpond was formed. Our study area included the headpond and the drainage basin upstream of the causeway. The fresh- water river and its five main tributaries (Anagance, Little, North, and Pollett rivers and Turtle Creek) New Brunswick Nova Scotia Petitcodiac Atlantic Ocean FiGurRE 1. Location of the Petitcodiac River system in Canada. THE CANADIAN FIELD-NATURALIST Vol. 115 have a combined length of 175 km and drain a water- shed area of 1360 km2. Mean flow at the causeway is 27.3 m/s (range 0.35 to 729 m3/s) (Caissie 2000). Turtle Creek was dammed to form a municipal drink- ing-water reservoir 10-km upstream of the mouth in 1982, and no fish passage facilities were provided. Most of the small dams that were present in the fresh- water tributaries at various times during the past cen- tury have been removed, but several small, derelict, impoundments remain in the headwaters of the Little and Pollett rivers. The presence of the causeway has resulted in the loss of many of the anadromous fish species that may have served as hosts for the glochidial stages of fresh- water mussel species inhabiting the Petitcodiac River system. Two fishways that were constructed to permit passage of diadromous fishes past the causeway have been ineffective and virtually all species have been eliminated or severely diminished (Beaulieu 1970; Wells 1999). Spawning runs of Atlantic Salmon (Salmo salar), American Shad (Alosa sapidissima), Striped Bass (Morone saxatilis), and Atlantic Tomcod (Microgadus tomcod) no longer occur. A very small population of Atlantic Salmon has been maintained by stocking. The anadromous species that continue to reproduce in the river include Sea Lamprey (Petro- myzon marinus), Brook Trout (Salvelinus fontinalis), Rainbow Smelt (Osmerus mordax), Alewife (Alosa pseudoharengus), Blueback Herring (A. aestivalis), and Fourspine Stickleback (Apeltes quadracus). The fish community of the headpond and most of the river system is now dominated by White Sucker (Catosto- mus commersoni), American Eel (Anguilla rostrata), Fourspine Stickleback, Ninespine Stickleback (Pungitius pungitius), and at least six minnow species (Cyprinidae). White Perch (Vorone americanus) are becoming increasingly common in the headpond and in the running waters of the river system. In recent years, Chain Pickerel (Esox niger), Brown Bullhead (Ameirus nebulosus), and Smallmouth Bass (Micro- pterus dolomieui) have been illegally introduced into the headpond and its tributaries and their populations are increasing rapidly (Locke 1999). Methods We conducted a watershed-wide survey of fresh- water mussels in the Petitcodiac River and its tribu- taries from 23 July to 21 October 1997, 2 June to 28 August 1998, and 14 to 20 June 2000. We used visu- al searches (by wading) to investigate 52 lotic sites and two headwater reservoirs in the system. Locations surveyed in 1997 and 1998 are listed in Hanson and Locke (2000); those surveyed in 2000 are listed in Table 1. Visual searches were conducted by surveying the full width of the river with a team of two to three searchers. In 1997, we conducted timed searches (one to two hours, two searchers). All sites visited during 1997 were also surveyed during 2001 HANSON AND LOCKE: FRESHWATER MUSSELS IN NEW BRUNSWICK 33] TABLE 1. Summary of sites visually surveyed for freshwater mussels in the Petitcodiac River watershed, in 2000. Positions of sites were obtained from Energy, Mines and Resources Canada 1:50 000 scale topographic maps of Moncton (21-I/2, edition 4), Salisbury (21-I/3, edition 3), Petitcodiac (21-H/14, edition 3), and Hillsborough (21-H/15, edition 3). Sites sur- veyed in 1997-1998 are listed in Hanson and Locke (2000). Site (access point) Railroad Bridge Town of Anagance River Petitcodiac River Anagance River Turtle Creek Foot bridge Turtle Creek Jonah Road Turtle Creek Tower Road Little River Blackwood Lake Pollett River The Glades 1998 and our assessment of relative abundance was based on the 1998 and 2000 surveys. In 1998 and 2000 (usually three searchers), we used a 50 m tape measure or a metered thread-dispensing device (Chainman II, Chainman Inc., Vancouver, British Columbia) to measure the distance searched, and did not restrict the time spent at each location. The stream length searched ranged from 60 to 1500 m (median 518m; quartiles 300 and 750m; total stream length searched 29.3 km) and stream width varied between 3 and about 50m. All stream banks and sand or gravel bars were carefully searched for empty shells. We also carefully examined shells in all Muskrat (Ondatra zebithicus) middens encoun- tered in the watershed (Meike and Hanson 2000). We searched the shorelines of two small impound- ments in the headwaters of the Little River for shells left on muskrat middens and also searched the bot- tom to a depth of 1.5 m using the same visual search method as used in running water sites. Water clarity was excellent at all locations. All submersed habitat was searched visually and by digging in sand and gravel around rocks and boulders. Locations with large deposits of sand and fine gravel were sampled with a 30 cm wide push net (each sample represented about 0.25 m2, to a depth of 5-6 cm), and sieved through the 6mm mesh netting. This method was even more sensitive than timed searches for detect- ing rare species as indicated by the fact that very small juveniles (as small as 7mm shell length) of several freshwater mussel species were collected. These small specimens were not detectable by stan- dard timed-search methods. We defined relative abundance at each site as not found, rare (<0.01 unionids per site), scarce (0.01 to 0.1 unionids/m2), common (0.1 to 1 unionids/m2), or abundant (> 1 unionids/m2). In most cases, living animals were identified to species and replaced. Some specimens were retained for voucher purposes or length-fre- quency analyses. Twelve stations in the main Petitcodiac Headpond were sampled during August of 1997, 1998, and 1999. Sediments were collected with a 23 X 23 X Latitude Longitude Distance Searched (DD MM) (DD MM) (m) 45 58.36 65 07.95 500 45 52.46 65 1559 60 46 01.24 64 53.55 200 45 56.43 64 52.05 600 45 57.59 64 52.77 300 45 45.56 64 57.74 300 45 56.68 65 05.28 500 23 cm Ekman dredge, 10 dredge hauls per sample, three samples at each of 1, 2, 3, and >4m depths. Sediments were washed on a 6-mm sieve. All ani- mals collected were taken to the laboratory where, as part of a separate study, we recorded age, shell length, shell weight, and viscera weight. The common and scientific names of freshwater mussels used in this study followed Turgeon et al. (1998). Voucher specimens have been deposited at the New Brunswick Museum, Saint John, N.B. Results Five species of freshwater mussels were collected in the Petitcodiac River system: Eastern Pearlshell (Margaritifera margaritifera), Brook Floater (Alas- midonta varicosa), Triangle Floater (A. undulata), Eastern Elliptio (Elliptio complanata) and Eastern Floater (Pyganodon cataracta). Much of the North River was not suitable habitat for any freshwater mussel species principally due to poor agricultural practices. Above the agricultural areas of the North River, the two sites located furthest upstream did not appear to be degraded and the habitat closely resem- bled the middle portion of the Little River where the Eastern Pearlshell was common. Nevertheless, fresh- water mussels were not found at either of these sites. In addition, no freshwater mussels were found in one of the small headwater reservoirs on the Little River. Freshwater mussels were found at virtually all other sites. The Eastern Pearlshell was widespread in lotic waters of the system (Figure 2) but absent from the Petitcodiac headpond and the two small reservoirs on the Little River. It was common to abundant in parts of the main Petitcodiac River, lower Anagance River, lower Pollett River, Little River, and Turtle Creek above the Turtle Creek Reservoir. We found clear evidence of recent reproduction (individuals <25 mm long) in all but the Turtle Creek sites. The Eastern Pearlshell was scarce in most of the Pollett River and in Turtle Creek below the drinking-water reservoir. Single specimens were collected at three sites on the North River. The Eastern Pearlshell 352 "A ene ce Pollett River THE CANADIAN FIELD-NATURALIST Vol. 115 Petitcodiac Petitcodiac River Headpond Turtle Creek Little River 0 5 10 = Kilometers Prosser Brook O None @ Rare @ Scarce Bi Common MB Abundant FIGURE 2. Distribution and relative abundance of the Eastern Pearlshell (Margaritifera mar- garitifera) in the Petitcodiac River and tributaries upstream of the Petitcodiac cause- way. occurred in single species beds in 22 of the 42 loca- tions where it occurred, mainly in the Pollett River, Turtle Creek, and most of the Little River. The fresh- water mussel beds in the lower Anagance River usu- ally contained equal numbers of Eastern Elliptio and Eastern Pearlshell. The Brook Floater was the species most often found with Eastern Pearlshell in sand and gravel sites in the main Petitcodiac River and Little River. The Eastern Floater was associated with reservoirs and with lotic habitats of reduced current speed (Figure 3). It was the only species collected in the one headwater reservoir of the Little River that con- tained mussels, and it overwhelmingly dominated the freshwater mussel community of the Petitcodiac Headpond. Large numbers of Eastern Floater occur- red in the lower, more lentic, half of the Petitcodiac Headpond but it was rare in the narrow, steep sloped, upper half of this impoundment. The sediments in the areas of high Eastern Floater abundance in both reservoirs were typically composed of soft mud or soft mud overlying hard clay. Finally, small numbers of Eastern Floater were collected from mud or silt substrates in quiet backwaters of the main Peti- tcodiac and lower Little Rivers and in two sites of the North River. We found clear evidence of recent reproduction (individuals < 15 mm long) in all sites except those in the North River. This species was absent from the Pollett River, most of the Little River and Turtle Creek, and the Anagance River. The Eastern Elliptio was found in slow-running water on sand or silty sand bottoms in the main Peti- tcodiac and Anagance rivers (Figure 4), usually in dense beds near the banks. The Eastern Elliptio usu- ally represented over 95% of the freshwater mussels present in these beds. We found clear evidence of recent reproduction (individuals < 15 mm long) in all sites occupied by Eastern Elliptio in these two rivers. Small numbers of Eastern Elliptio were found at the sites supporting mussels in the North River, usually where a stream entered the river or in springs, but there was no indication of recent reproduction. The Eastern Elliptio was not present in either reservoir on the Little River but a few specimens were collected along with large numbers of Eastern Floater on three sandbars in the main Petitcodiac headpond. The Eastern Elliptio was absent from the Little and Pol- lett rivers, and Turtle Creek. The sediments in these rivers ranged from coarse sand to cobble and boulder with moderate to rapid water current. The Brook Floater was found only in running water sections of the Petitcodiac River system 2001 < "% * R\ \Y oo oo po Pollett River HANSON AND LOCKE: FRESHWATER MUSSELS IN NEW BRUNSWICK 333 Petitcodiac Petitcodiac River Headpond Causeway Kilometers Prosser Brook O None B® Rare @ Scarce Bi Abundant FiGurRE 3. Distribution and relative abundance of the Eastern Floater (Pyganodon cataracta) in the Petitcodiac River and tributaries upstream of the Petitcodiac causeway. (Figure 5). The Brook Floater was found in locally high abundance in beds located in backwaters of sand or gravel bars in the lower 6 km of the Little River and in small groups (5 to 20 animals) in pock- ets of coarse sand behind large boulders. Small groups of Brook Floater were collected in similar habitats in the main Petitcodiac River. This species was not found in the Pollett or Anagance rivers, most of the North River, or in Turtle Creek. We found clear evidence of recent reproduction (individuals <15 mm long) at all sites occupied by this species except those in the North River. The distribution of Triangle Floater was almost identical to that of the Brook Floater except that it was very rare at all locations (Figure 6). We never collected more than five living Triangle Floaters at any site. This species was very rare in the Petitcodiac River and was not found in the Pollett or Anagance rivers. Occasional specimens were collected in the lower 6-km section of the Little River. We found five living individuals in a spring in the North River and single specimens at four other sites. No speci- mens were collected in the running water portion of Turtle Creek but one living animal was collected on a sand bar in the flooded (as part of the main Petitco- diac Headpond) lower section of the creek. We found no evidence of recent reproduction (individu- als <15 mm long) by this species except at one site in the Little River. The Triangle Floater was rare throughout the drainage and it is possible that we overlooked small individuals or mistook them for very small Eastern Elliptio. Discussion Five species of freshwater mussels were collected in the Petitcodiac River system, but only two of these, the Eastern Elliptio and Eastern Floater, are considered by the American Fisheries Society to have stable conservation status (Williams et al. 1993). The Dwarf Wedgemussel was once common in the Petitcodiac and North Rivers but was eliminat- ed from the system following completion of the Petitcodiac causeway (Hanson and Locke 2000). The Canadian range of the species was restricted to the Petitcodiac River, and the Dwarf Wedgemussel is now considered extirpated from Canada. The conser- vation status in Canada of the five species remaining in the watershed has not been determined by COSEWIC. There is little published information on the mussel fauna of the Petitcodiac River system; hence, it is not clear whether the abundance and distribution of the species documented in this study represent a stable situation or one in transition. Each of the extant 334 oe ase? Pollett River THE CANADIAN FIELD-NATURALIST Vol. 115 Petitcodiac Petitcodiac River Headpond ry Causeway r Turtle Creek It By Little River 0 5 10 = Kilometers Prosser Brook 5 None @ Rare Mi Scarce Hi Common BE Abundant FiGurE 4. Distribution and relative abundance of the Eastern Elliptio (Elliptio complanata) in the Petitcodiac River and tributaries upstream of the Petitcodiac causeway. species in the Petitcodiac system, however, has unique life history characteristics and relatively well defined habitat distributions that can be used to eval- uate the risk to persistence of the species in the sys- tem. Under normal conditions, adult freshwater mus- sels have low natural mortality rates, although their sedentary lifestyle renders them susceptible to a number of anthropogenic threats. These threats include the physical effects of habitat degradation (e.g., conversion of riffle habitat into still water), over-harvesting, anoxia, metal contamination, and introduction of encrusting competitors (e.g., the Zebra Mussel, Driessena polymorpha) (Nalepa et al. 1991; Bogan 1993; Blalock and Sickel 1996; Ric- ciardi et al. 1998; Sparks and Strayer 1998). In the absence of these effects, events at early life stages largely determine the abundance and distribution of freshwater mussels in a water body. Although adult freshwater mussels release tens of thousands to mil- lions of glochidia, only a very small fraction of them ever successfully attach to fish and survive to the benthic juvenile stage (Young and Williams 1983; Jansen and Hanson 1991; Buddensiek 1995). The distribution of the freshwater mussel species is sub- sequently determined by (1) sediment characteristics appropriate for the settling larvae, and (2) access by the host fish to the habitat. If the host fish can not enter a stretch of water, the freshwater mussel species will not be found there (Watters 1992; Graf 1997; Haag and Warren 1998). This is why the blockage of streams and rivers is a serious threat to the persistence of freshwater mussel populations worldwide. The Eastern Pearlshell is listed as an endangered species in Europe (Bauer 1983; Cosgrove et al. 2000) and the American Fisheries Society lists it of special concern in Canada and the USA (Williams et al. 1993). The habitat requirements (clean running water over coarse sand, gravel, or cobble and gravel bottom) are relatively well studied for the Eastern Pearlshell (Bauer 1992; Hastie et al. 2000) and suitable habitat is widely available in the Petitcodiac River Drainage. Unfortunately, the populations in the Petitcodiac River Drainage appear destined for extermination due to reproductive failure. The principal fish hosts for the glochidial stage are Atlantic Salmon, Brown Trout and, possibly, Brook Trout (Athearn and Clarke 1961; Smith 1976; Bauer 1987; Cunjak and McGladdery 1991). Atlantic Salmon and Brook Trout occur in the Petitcodiac system. In the current study, there clearly had been recent reproduction of Eastern Pearlshell in the Little River and parts of the Pollett and Petitcodiac 2001 HANSON AND LOCKE: FRESHWATER MUSSELS IN NEW BRUNSWICK 335 Petitcodiac Headpond Little River 0 5 10 o_O Kilometers Prosser Brook O None @ Rare Mi Scarce fi Common FIGURE 5. Distribution and relative abundance of the Brook Floater (Alasmidonta varicosa) in the Petitcodiac River and tributaries upstream of the Petitcodiac causeway. rivers, most likely due to the large numbers of Atlantic Salmon parr and smolts stocked in the late 1980s and early 1990s. Areas of Turtle Creek that have not received plantings of salmon for 30 years, however, showed no signs of reproduction — all ani- mals that we examined during our June 2000 survey (> 1000 individuals) were at least 80 mm long and most were > 100mm long. Although Brook Trout were present in Turtle Creek, they either are not com- mon enough to permit reproduction or this particular population of Eastern Pearlshell does not use Brook Trout as a glochidial host. In the case of the Petitco- diac River drainage, it is clear from the numerous failed attempts at stocking that Atlantic Salmon will never re-establish breeding populations as long as the causeway acts as a barrier to migration. Stocking activities have now ceased. Consequently, the Eastern Pearlshell populations in the Petitcodiac River system (similar to many other populations in North America) are functionally extirpated because reproduction is no longer possible due to the disappearance of the glochidial host fish. Non-reproducing populations of Eastern Pearlshell will likely persist for decades, how- ever, because individuals can live about 100 years (Bauer 1983, 1992). The Eastern Floater is considered by the American Fisheries Society to have stable conservation status in Canada and the USA (Williams et al. 1993). The Eastern Floater populations in the Petitcodiac River system do not appear to be threatened by current water management practices. In fact, this species arguably has benefited from formation of reservoirs in the drainage basin. The preferred habitat of the Eastern Floater is lakes, ponds, and slow moving streams — usually on sand or mud bottoms, and this is often the only species found in soft muddy habitats (Athearn and Clarke 1961). Reservoir construction in the Petitcodiac drainage has provided this soft muddy habitat in at least three impoundments but freshwater mussels are only present in two. One headwater reser- voir has no means of fish passage and was devoid of freshwater mussels. In contrast, the second headwater reservoir has a functional fishway and supports a pop- ulation of Eastern Floater. Lastly, the main Petitcodiac Headpond has had a history of colonization and popu- lation extermination due to various unsuccessful attempts to provide passage for Atlantic Salmon through the causeway. As recently as 1988, the entire population of Eastern Floater was killed when the headpond filled with salt water at each high tide. Drawdowns in 1998 and 1999 exposed the entire lit- toral zone to air and killed all freshwater mussels in 336 THE CANADIAN FIELD-NATURALIST Vol. 115 Petitcodiac River Petitcodiac Headpond FP. i Little River 0 5 10 Se Kilometers Prosser Brook FIGURE 6. Distribution and relative abundance of the Triangle Floater (Alasmidonta undulata) in the Petitcodiac River and tributaries upstream of the Petitcodiac cause- way. the headpond except those living in the sublittoral zone (S. Richardson, J. M. Hanson, and A. Locke, unpublished data). The ability to colonize these impoundments requires a highly mobile host fish for the glochidia. The known hosts for Eastern Floater include Threespine Stickleback (Gasterosteus aculea- tus), White Sucker (Catostomus commersoni), Pump- kinseed (Lepomis gibbosus), and Common Carp (Cyprinus carpio) (Wiles 1975; Clarke 1981; Threlfall 1986). Threespine Stickleback and White Sucker are recorded in the drainage but only White Sucker is abundant at all sites where Eastern Floater was col- lected. Indeed, White Sucker is common throughout all of the drainage but habitat requirements probably limit the distribution of the Eastern Floater, which was absent from moderate- to fast-running water and coarse substrate. The American Fisheries Society also considers the Eastern Elliptio to have stable conservation status in Canada and the USA (Williams et al. 1993). The pop- ulations of Eastern Elliptio in the Petitcodiac River system do not appear to be threatened. The only known glochidial hosts of the Eastern Elliptio are Yellow Perch (Perca flavescens) and Banded Killi- fish (Fundulus diaphanus) (Matteson 1948; Wiles 1975; Paterson 1985). Yellow Perch do not occur in the Petitcodiac system while Banded Killifish were observed at all sites where Eastern Elliptio was found. Banded Killifish were also observed at many sites where Eastern Elliptio was absent, especially in the main Petitcodiac Headpond. Similar to the East- ern Floater, the availability of suitable substrate appears to be limiting distribution of Eastern Elliptio rather than the distribution of the fish host. The East- ern Elliptio is generally most common in slow or still waters on sand (plus silt) or sandy gravel whereas the bottom of the main headpond is mostly soft silt, which is not the preferred habitat of this species (Matteson 1948; Ghent et al. 1978; Amyot and Downing 1991). The Brook Floater is listed by the American Fish- eries Society as “threatened” in Canada and the USA (Williams et al. 1993) because it has disappeared from many locations where it formerly was found (Clarke 1986; Counts et al. 1991; Strayer and Fetter- man 1999). The conservation status in Canada has not been assessed by COSEWIC but this might be a candidate species for immediate assessment under the criteria listed by Metcalfe-Smith et al. (1998). In Canada, this species is not reported outside of New 2001 Brunswick and Nova Scotia, where it is rare. Within New Brunswick, the Brook Floater is listed as rare and only occurring in the Petitcodiac River (Clayden et al. 1984). This limited distribution is misleading because little survey information has been published for New Brunswick waters. Indeed, Athearn (1961) reports it was found in the Renous River, a tributary of the Miramichi River and additional surveys may show it is widespread within the province. The Brook Floater was collected at only four of over 180 loca- tions searched in Nova Scotia during the 1950s and early 1960s, and it was not common at any of them (Athearn and Clarke 1961; Clarke and Rick 1963). It is unknown whether any of these previously identi- fied populations persist because there have been no large-scale surveys conducted since then. The distri- bution of Brook Floater in the Petitcodiac River Drainage was patchy but their persistence does not seem to be threatened by reproductive failure because we observed specimens < 15 mm long at several locations. This species seems to have fairly well defined habitat requirements that limits its distribu- tion in the Petitcodiac River Drainage. It requires running water and sand or sandy gravel substrate (Athearn and Clarke 1961; Strayer and Ralley 1993) and these were the only areas where we found Brook Floater. Not all suitable habitat supported the species. Much of the upper Little River consists of superficial- ly suitable habitat but the only species found was Eastern Pearlshell. This suggests that the host fish for Brook Floater does not ascend into headwaters of rivers; however, the glochidial host for Brook Floater is unknown. Based on the distribution of Brook Floater in the Petitcodiac River drainage, the host species would appear to be an anadromous fish. The fish species that best match the distribution of Brook Floater in the Petitcodiac River system as well as across the whole species distribution are Alewife (original range) and Blueback Herring, and these are the species that we recommend be studied first as potential hosts. To date, both fish species persist in the Petitcodiac River drainage and it is habitat destruction due to poor agricultural practices that is threatening some unionid populations (but see Muskrat predation below). The conservation status of the Triangle Floater in the Canada and the USA is listed as of “special con- cern” by the American Fisheries Society (Williams et al. 1993). We are unable to conclude whether the Triangle Floater is likely to persist in the Petitcodiac River Drainage. Throughout its distribution it is described as uncommon in some areas and locally common in others (Athearn and Clarke 1961; Clarke 1986; Strayer and Ralley 1991). The host species for the Triangle Floater have not been determined. The Triangle Floater occurs in the Trent-Severn and Ottawa River systems in Ontario (Clarke 1981; Metcalfe-Smith et al. 1998a); therefore, at least one HANSON AND LOCKE: FRESHWATER MUSSELS IN NEW BRUNSWICK Bo) fish host is not anadromous. Although it clearly occurs in lakes (Metcalfe-Smith and Green 1992), the Triangle Floater was not found in any of the reservoirs in the Petitcodiac System. We encoun- tered so few Triangle Floaters in our survey that it might be questionable whether this species attained a sufficiently large local concentration to permit fertil- ization of eggs except that the species appears to be a functional hermaphrodite (Kat and Davis 1984). While we are confident that we detected all species currently present in the system we are puzzled by the absence of the Alewife Floater (Anodonta implicata). The Alewife Floater is usually found in low gradient coastal rivers and stable coastal ponds (Strayer and Ralley 1991). The Alewife Floater is widespread in New Brunswick and parts of Nova Scotia (Athearn 1961; Athearn and Clarke 1961; Clarke and Rick 1963; Kat and Davis 1984), and there are large num- bers of Alewife (the principal glochidial host) in the Petitcodiac system; hence, we expected it to be pre- sent. Whether it formerly was present and eliminated due to construction of the Petitcodiac Causeway can not be resolved due to a lack of published records on the mussel fauna present in the system prior to cause- way construction in 1968. Habitat degradation is a common threat to all unionid species in the Petitcodiac River system. The North River in particular has been strongly affected by poor agricultural practices. Habitat degradation was extensive. The sediment in some areas was black in colour and released methane and hydrogen sulfide gas when disturbed. Cultivation of fields extended to within 2 to 3 m of the bank and in many places there were drainage ditches and pipes emptying directly into the river. In other areas, fences were built across the river to allow cattle full access to the river, result- ing in destruction of the stream banks, elimination of shoreline vegetation, erosion, and deposition of fecal material directly into the river. Finally, macrophyte growth was excessive in many areas and the plants themselves and the substrate were thickly coated with algae, i.e., conditions under the plants were anoxic. Of the species present in the Petitcodiac River sys- tem, Eastern Elliptio is considered to be most tolerant of degraded environmental conditions (Strayer and Ralley 1991). Hence it is not surprising that, when small patches of freshwater mussels occurred in the North River, the Eastern Elliptio was the main spe- cies collected. Muskrats (Ondatra zibethicus) are the only mam- malian predator (other than man) that kills large num- bers of adult freshwater mussels (Hanson et al. 1989; Neves and Odom 1989; Tyrell and Hornbach 1998). This predation tends to be localized and does not threaten the existence of populations of endangered freshwater mussels unless an area of high muskrat feeding coincides with a remnant population of an endangered species (Neves and Odom 1989; Bruen- 338 derman and Neves 1993; Hoggarth et al. 1995). Muskrat predation is not currently widespread in the Petitcodiac watershed and tends to focus on the most abundant mussels of large body size. The species con- sumed in the Petitcodiac in 1997 and 1998 were Eastern Pearlshell and Eastern Floater (Meike and Hanson 2000). We have since discovered several Muskrat middens adjacent to an extensive mussel bed dominated by Eastern Elliptio, which was the only species eaten. In addition, we discovered a single Muskrat midden adjacent to the largest single concen- tration of Brook Floater in the watershed. This mid- den contained shells of over 90 Brook Floater but we counted over 300 living animals in the River. Never- theless, this observation supports earlier reports that Muskrats can be a threat when the predation corre- sponds with a location where a threatened or endan- gered species happens to be locally abundant. In summary, we detected five freshwater mussel species in the Petitcodiac River Drainage, and one species known to be present in the early 1960s is extirpated (Hanson and Locke 2000). The current study is the first to clearly document the relative abundance and distribution of the entire freshwater mussel fauna in any Maritime Province river. In order to protect the integrity of our freshwater eco- systems in the Maritimes, it is critical that we know what species are present now and conduct regular surveys on key water bodies to assess the health of the populations and their habitats, and to monitor for the arrival of invaders (e.g., Zebra Mussels). Without comprehensive surveys, it is impossible to determine whether distributions have changed or abundance declined over time in this the most threatened taxon in North America. Native freshwater mussels do not play a major commercial role in Canadian waters, federal endangered species legislation (at time of writing) is lacking, and the group is not “popular” or well known to the public. As a result, there is little incentive to study or conserve freshwater mussels in Canadian waters; consequently, further species extir- pations are inevitable. Acknowledgments This report benefited from discussions with D. McAlpine and comments by M. Elderkin, G. Klas- sen, and three anonymous referees improved the manuscript. Technical support was provided by R. Bernier, S. Richardson, R. Stevens, and M. Knockwood. Literature Cited Amyot, J.-P., and J. A. Downing. 1991. Endo- and epi- benthic distribution of the unionid mollusc Elliptio com- planata. Journal of the North American Benthological Society 10: 280-285. Athearn, H.D. 1961. Additions to the New Brunswick checklist. Sterkiana 4: 33-34. Athearn, H.D., and A. H. Clarke, Jr. 1961. The fresh- THE CANADIAN FIELD-NATURALIST Volhis water mussels of Nova Scotia. National Museums of Canada Bulletin 183: 11-41. Bauer, G. 1983. Age structure, age specific mortality rates and population trend of the freshwater pearl mussel (Margaritifera margaritifera) in North Bavaria. Archiv fur Hydrobiologia 98: 523-532. : Bauer, G. 1987. The parasitic stage of the freshwater pearl mussel (Margaritifera margaritifera L.) Il. Host relationships. Archiv fur Hydrobiologia Supplement 76: 413-423. Bauer, G. 1992. Variation in the lifespan and size of the freshwater pearl mussel. Journal of Animal Ecology 61: 425-436. Beaulieu, G. T. 1970. Report on the Petitcodiac River Causeway. Department of Fisheries and Forestry of Can- ada, Fisheries Service Manuscript Report number 70-11: 19 pages. Halifax, Nova Scotia Blalock, H. N/D., and J. B. Sickel. 1996. Changes in mussel (Bivalvia: Unionidae) fauna within the Kentucky portion of Lake Barkley since impoundment of the lower Cumberland River. American Malacological Bulletin 13: 111-116. Bogan, A.E. 1993. Freshwater bivalve extinctions (Mol- lusca, Unionida): a search for causes. American Zoologist 33: 599-609. Bruenderman, S. A., and R. J. Neves. 1993. Lite history of the endangered fine-rayed pigtoe Fusconaia cuneolus (Bivalvia: Unionidae) in the Clinch River, Virginia. American Malacological Bulletin 10: 83-91. Buddensiek, V. 1995. The culture of juvenile freshwater pearl mussels Margaritifera margaritifera L. in cages: a contribution to conservation programmes and the knowl- edge of habitat requirements. Biological Conservation 74: 33-40. Cassie, D. 2000. Hydrology of the Petitcodiac River Basin in New Brunswick. Canadian Technical Report of Fisheries and Aquatic Sciences 2301. 26 pages. Clarke, A.H. 1981. The freshwater molluscs of Canada. National Museum of Natural History, National Museums of Canada, Ottawa, Ontario. 446 pages. Clarke, A.H. 1986. Unionidae of the Upper Connecticut River, a vanishing resource. The Nautilus 100: 49-53. Clarke, A. H., Jr., and A. M. Rick. 1963. Supplementary records of Unionacea from Nova Scotia with discussion of the identity of Anodonta fragilis Lamarck. National Museums of Canada Bulletin 184: 15-27. Clayden, S.R., D. F. McAlpine, and C. Guidry. 1984. Rare and vulnerable species in New Brunswick. The New Brunswick Museum Publications in Natural Science num- ber2: 95 pages. Cosgrove, P. J.. M. R. Young, L. C. Hastie, M. Gaywood, and P. J. Boon. 2000. The status of the freshwater pearl mussel Margaritifera margaritifera Linn. in Scotland. Aquatic Conservation: Marine and Freshwater Ecosys- tems 10: 197-208. Counts, C.L., II, T.S. Handwerker, and R. V. Jesien. 1991. The Naiades (Bivalvia: Unionoidea) of the Del- marva Peninsula. American Malacological Bulletin 9: 27-37. Cunjak, R. A., and S. E. McGladdery. 1991. The para- site-host relationship of glochidia (Mollusca: Margariti- feridae) on the gills of young-of-the-year Atlantic sal- mon (Salmo salar). Canadian Journal of Zoology 69: 353-358. Ghent, A. W., R. Singer, and L. Johnson-Singer. 1978. 2001 Depth distributions determined with SCUBA, and asso- ciated studies of the freshwater unionid clams Elliptio complanta and Anodonta grandis from Lake Bernard, Ontario. Canadian Journal of Zoology 56: 1654-1663. Graf, D. L. 1997. Sympatric speciation of freshwater mus- sels (Bivalvia: Unionoidea): a model. American Malaco- logical Bulletin 14: 35-40. Haag, W.R., and M.L. Warren, Jr. 1998. Role of eco- logical factors and reproductive strategies in structuring freshwater mussel communities. Canadian Journal of Fisheries and Aquatic Sciences 55: 297-306. Hanson, J. M., and A. Locke. 2000. The status of the Dwarf Wedgemussel, Alasmidonta heterodon, in Canada. Canadian Field-Naturalist 114: 271-278. Hanson, J. M., W.C. Mackay, and E.E. Prepas. 1989. Effect of size-selective predation by muskrats (Ondatra zebithicus) on a population of unionid clams (Anodonta grandis simpsoniana). Journal of Animal Ecology 58: 15-28. Hastie, L. C., P. J. Boon, and M. R. Young. 2000. Physi- cal microhabitat requirements of freshwater pearl mus- sels, Margaritifera margaritifera (L.). Hydrobiologia 429: 59-71. Hoggarth, M.A., D.L. Rice, and D. M. Lee. 1995. Dis- covery of the federally endangered freshwater mussel, Epioblasma obliquata obliquata (Rafinesque, 1820) (Unionidae), in Ohio. Ohio Journal of Science 95: 298-299. Jansen, W.A., and J. M. Hanson. 1991. Estimates of the number of glochidia produced by clams (Anodonta gran- dis simpsoniana), attaching to yellow perch (Perca flavescens), and surviving to various ages in Narrow Lake, Alberta (Canada). Canadian Journal of Zoology 69: 973-977. Kat, P. W., and G. M. Davis. 1984. Molecular genetics of peripheral populations of Nova Scotian Unionidae (Mol- lusca: Bivalvia). Biological Journal of the Linnean Society 22: 158-185. Locke, A. 1999. Fish communities of the Petitcodiac River reservoir and tributaries: 1998 monitoring study. Pages 177-196 in Environmental Monitoring Working Group. Environmental monitoring of the Petitcodiac River sys- tem, 1998: Petitcodiac River Trial Gate Opening Project. Department of Fisheries and Oceans, Moncton, New Brunswick. Matteson, M.R. 1948. Life history of Elliptio complana- tus (Dillwyn, 1817). American Midland Naturalist 40: 690-723. Meike, E., and J. M. Hanson. 2000. Effects of muskrat predation on Najads. Pages 163-184 in Ecology and evolutionary biology of the freshwater mussels Union- oidea. Edited by G. Bauer and K. Wachtler. Ecological Studies Volume 145. Springer Verlag, Berlin. Metcalfe-Smith, J.L., and R.H. Green. 1992. Ageing studies on three species of freshwater mussels from a metal-polluted watershed in Nova Scotia, Canada. Can- adian Journal of Zoology 70: 1284-1291. Metcalfe-Smith, J. L., S. K. Staton, G. L. Mackie, and N.M. Lane. 1998a. Selection of candidate species of freshwater mussels (Bivalvia: Unionidae) to be consid- ered for national status designation by COSEWIC. Canadian Field-Naturalist 112: 425-440. Metcalfe-Smith, J. L., S. K. Staton, G. L. Mackie, and N.M. Lane. 1998b. Changes in the biodiversity of freshwater mussels in the Canadian waters of the Lower HANSON AND LOCKE: FRESHWATER MUSSELS IN NEW BRUNSWICK 339 Great Lakes Drainage Basin over the past 140 years. Journal of Great Lakes Research 24: 845-858. Nalepa, T. F., B. A. Manny, J.C. Roth, S.C. Mozley, and D. W. Schloesser. 1991. Long-term decline in freshwa- ter mussels (Bivalvia: Unionidae) of the western basin of Lake Erie. Journal of Great Lakes Research 17: 214— 219. Neves, R. J., and M.C. Odom. 1989. Muskrat predation on endangered freshwater mussels in Virginia. Journal of Wildlife Management 54: 934-941. Paterson, C.G. 1985. Biomass and production of the unionid, Elliptio complanata (Lightfoot) in an old reser- voir in New Brunswick, Canada. Freshwater Invertebrate Biology 4: 201-207. Primack, R. B. 1998. Essentials of conservation biology. 2nd edition. Sinauer Associates, Sunderland, Massachu- setts. 660 pages. Ricciardi, A., R. J. Neves, and J.B. Rasmussen. 1998. Impending extinctions of on North American freshwater mussels (Unionida) following the zebra mussel (Dreis- sena polymorpha) invasion. Journal of Animal Ecology 67: 613-619. Smith, D.G. 1976. Notes on the biology of Margaritifera margaritifera margaritifera (Lin.) in central Massachu- setts. American Midland Naturalist 96: 252-256. Sparks, B. L., and D. L. Strayer. 1998. Effects of low dis- solved oxygen on juvenile Elliptio complanata (Bivalvia: Unionidae). Journal of the North American Benthological Society 17: 129-134. Staton, S.K., J. L. Metcalfe-Smith, and E.L. West. 2000. Status of the Northern Riffleshell, Epioblasma torulosa rangiana (Bivalvia: Unionidae), in Ontario and Canada. Canadian Field-Naturalist 114: 224—235. Strayer, D.L., and A.R. Fetterman. 1999. Changes in the distribution of freshwater mussels (Unionidae) in the Upper Susquehanna River Basin, 1955-1965 to 1996— 1997. American Midland Naturalist 142: 328-339. Strayer, D.L., and J. Ralley. 1991. The freshwater mus- sels (Bivalvia: Unionoidea) of the upper Delaware River drainage. American Malacological Bulletin 9: 21-25. Strayer, D.L., and J. Ralley. 1993. Microhabitat use by an assemblage of stream-dwelling unionaceans (Bivalvia), including two rare species of Alasmidonta. Journal of the North American Benthological Society 12: 247-258. Threlfall, W. 1986. Seasonal occurrence of Anodonta cataracta Say, 1817, glochidia on three-spined stickle- backs, Gasterosteus aculeatus Linnaeus. The Veliger 29: 231-234. Turgeon, D.D., J. F. Quinn, Jr., A. E. Bogan. E. V. Coan, F. G. Hochberg, W. G. Lyons, P.M. Mikkelsen, R. J. Neves, C.F. E. Roper, G. Rosenberg, B. Roth, A. Scheltema, F.G. Thompson, M. Vecchione, and J. D. Williams. 1998. Common and scientific names of aquat- ic invertebrates from the United States and Canada: mol- lusks, 2nd edition. American Fisheries Society, Special Publication 26, Bethesda, Maryland. 526 pages. Tyrell, M., and D. J. Hornbach. 1998. Selective preda- tion by muskrats on freshwater mussels in two Minne- sota rivers. Journal of the North American Benthological Society 17: 301-310. Vaughn, C.C., and C.M. Taylor. 1999. Impoundments and the decline of freshwater mussels: a case study of an extinction gradient. Conservation Biology 13: 912-920. Watters, T.G. 1992. Unionids, fishes, and the species- area curve. Journal of Biogeography 19: 481-490. 340 Watters, T.G. 1995. Small dams as barriers to freshwater mussels (Bivalvia, Unionoida) and their hosts. Biologi- cal Conservation 75: 79-85. Wells, P.G. 1999. Environmental impacts of barriers on rivers entering the Bay of Fundy: Report of an ad hoc Environment Canada working group. Canadian Wildlife Service Technical Report Series number 334. 46 pages. West, E. L., J. L. Metcalfe-Smith, and S. K. Staton. 2000. Status of the Rayed Bean, Villosus fabalis (Bivalvia: Unionidae) in Ontario and Canada. Canadian Field- Naturalist 114: 248-258. Wiles, M. 1975. The glochidia of certain Unionidae (Mol- THE CANADIAN FIELD-NATURALIST Vol. 115 lusca) in Nova Scotia and their fish hosts. Canadian Journal of Zoology 53: 33-41. Williams, J.D., M.L. Warren, Jr., K.S. Cummings, J. L. Harris, and R. J. Neves. 1993. Conservation sta- tus of freshwater mussels of the United States and Canada. Fisheries 18 (9): 6—22. Young, M., and J. Williams. 1983. The status and con- servation of the freshwater pearl mussel Margaritifera margaritifera Linn. in Great Britain. Biological Conser- vation 25: 35-52. Received 18 December 2000 Accepted 2 June 2001 Notes Three New Taxa and a Summary of the Mustard Family, Brassicaceae (Cruciferae), in Canada and Alaska GERALD A. MULLIGAN! 1600 Apeldoorn Avenue, Ottawa, Ontario, K2C 1V5 Canada ‘Formerly Scientist and Institute Director, Research Branch, Agriculture Canada, Ottawa, Ontario K1A 0C6 Canada Mulligan, Gerald A. 2001. Three new taxa and a summary of the species in the mustard family, Brassicaceae (Cruciferae) in Canada and Alaska. Canadian Field-Naturalist 115(2): 341-342. Three taxa of the mustard family (Brassicaceae/Cruciferae) are validated: Descurainia incisa var. viscosa, Smelowskia johnsonii, and Smelowskia media. These taxa are among those already included in a Key to the Brassicaceae (Cruciferae) of Canada and Alaska, the genera of which are summarized. A key to the genus Smelowskia is given. Key Words: Brassicaceae, Cruciferae, key, taxa, genera, Descurainia incisa var. viscosa, Smelowskia johnsonii, Smelowskia media. A key to 248 taxa in 58 genera of the Brassicaceae (Cruciferae) of Canada and Alaska appears on the internet site http://res2.agr.ca /ecorc/cwmt/brasskey/ index.htm. It includes the type species of each genus, pertinent synonyms, general distribution, and infor- mation on the native or naturalized status of each taxon. The family is represented in Canada by the following genera (the number of taxa in each genus appears in brackets): Alliaria (1), Alyssum (A), Aphragmus (1), Arabidopsis (2), Arabis (36), Armoracia (2), Athysanus (1), Barbarea (4), Berteroa (1), Brassica (5), Braya (3), Bunias (1), Cakile (3), Camelina (3), Capsella (1), Cardamine (23), Cardaria (3), Chorispora (1), Cochlearia (3), Conringia (1), Coronopus (2), Crambe (1), Descurainia (8), Diplotaxis (3), Draba (48), Eruca (1), Erucastrum (1), Erysimum (9), Eutrema (1), Halimolobos (3), Hesperis (1), Hutchinsia (1), Iberis (2), Idahoa (1), Isatis (1), Lepidium (15), Lesquerella (5), Lobularia (1), Lunaria (1), Malcolmia (1), Matthiola (1), Myagrum (1), Nasturtium (3), Neslia (1), Parrya (2), Physaria (1), Raphanus (2), Rapistrum (2), Rorippa (11), Schoenocrambe (1), Sinapis (2), Sisymbrium (3), Smelowskia (7), Subularia (1), Teesdalia (1), Thelypodium (1), Thlaspi (2), Thysanocarpus (1). Validation of taxa The following three taxa included in the Key to the Brassicaceae of Canada and Alaska have not pre- viously been validated: Descurainia incisa (Engelm. ex A. Gray) Britton var. viscosa (Rydb.) G. A. Mulligan, comb. nov., based on Sophia viscosa Rydb., Bull. Torrey Bot. Club 29: 2338) 1902. Smelowskia johnsonii G. A. Mulligan, sp. nov.: Sme- lowskia johnsonii a certis Smelowskia praedites foliis caulinis late oblongis vel ovatis, integris vel minute denticulatis, densissimis pilis simplicibis albis-villosis. Smelowskia johnsonii differs from other Smelowskia by its oblong to ovate, entire or shallowly toothed cauline leaves that are densely covered by long, simple, white-villous trichomes. Holotype: Hills adjacent to Kukpuk River, Lat. 68°17’N, Long. 165°32’W, Alaska. Elevation 300- 1400 ft., steep slopes and ridges of Flint Mountain and surrounding hills. Talus slopes, A. W. Johnson, L. Viereck and H. Melchior 688, 15-16 Aug., 1959 (DAO). David F. Murray, University of Alaska Museum, Fairbanks, said, in a fax dated 18 November 1997, that the Museum also has a specimen of this taxon from the same general area, Viereck and Bucknell 4359. Smelowskia media (Drury & Rollins) G.A. Mulligan, sp. nov., based on Smelowskia calycina var. media Drury & Rollins, Rhodora 54: 100, 1952. Smelowskia media is only known to have the chromosome number of 2n=12: Alaska, Dawe & Murray (1981); Yukon, Calder 34286 & 34367, Department of Agriculture Herbarium, Ottawa (DAO), Mulligan (unpublished). 341 342 THE CANADIAN FIELD-NATURALIST Vol. 115 Species in the genus Smelowskia in Canada and Alaska ie 2a AnKRWYN b. Smelowskia borealis (Greene) Drury & Rollins (AK, YT, NT-M). Native. . Smelowskia calycina (Stephan ex Willd.) C. A. Mey. var. americana (Regel & Herder) Drury & Rollins (AB, BC). Native. Smelowskia calycina var. porsildii Drury & Rollins (AK). Native. Smelowskia johnsonii G. A. Mulligan (AK, Bering Strait). Native. Smelowskia media (Drury & Rollins) G. A. Mulligan (AK, YT, NT-M). Native. Smelowskia ovalis M. E. Jones (BC). Native. Smelowskia pyriformis Drury & Rollins (AK). Native. Key to species in the genus Smelowskia in Canada and Alaska 1. Caudex mostly branched, slender, stems usually simple, each arising from a separate caudex branch; PE pedicels divaricate to ascending... 6... a ee ene 2 A ac oe D . Caudex mostly simple, stout, more than 5 mm in diameter; stems branched from near base upward, becoming decumbent in fruit; pedicels widely diversent to arcuate jaye. e- 6 Mature siliques oblong, tapering at both ends; sepals caducous; basal leaf bases strongly ciliate with long acicular trichOMes: oe 55 ee Si Sa ales bbe wiega hs walter! era Sale ee Salant Reels SRS kena a 3 Mature siliques ovate to slightly oblong, truncate at base; sepals persistent; basal leaf bases not ciliate sink eles bee e tive: outa bya, Sp aetna RuMtsMa eM at fade Segue awa a ARIAS ce ee ate OG esd aa er 5. S. ovalis Basal leaves entire or very shallowly toothed at the tips; cauline leaves entire or very shallowly toothed ...4 Basal leaves pinnately lobed; cauline leaves pinnately lobed; caudex leaves broadly oblong to ovate ....5 Caudex leaves broadly oblong to ovate, densely covered with long, simple, white-villous THHEHOMIES is ae FESR AT Pe RD ah TERRIA cr a ace Lae la eg 3. S. johnsonii . Caudex leaves linear to narrowly spatulate, densely covered with short dendritic PEUCHOMIES 52 Ne UN Mie ee Ree) A ce Ci ade ea gO a a 2b. S. calycina vat. porsildii Pedicels widely divaricate, angle of divergence 50° to 80°; siliques broadest above the middle; cauline leaves few-lobed: 2 or 3.per Stemi! Hie AeA Rhea 8 cary ee nol eo eee eee de 4. S. media Pedicels ascending, angle of divergence less than 50°; siliques broadest at the middle and below; cauline leavesimany-lobeds 3 to Wppenstemmite sick mie its cc. te as te tes a clea ete eae ee 2a. S. calycina var. americana . Basal leaves pinnately cut to the midrib the full length of the blade, ca 9-lobed, petals white or cream-coloured; sepals tan, ca 1 mm long; siliques elongated, pear-shaped, Stor. Qin lon gaca:2 may Wide sii woh wel soe os wh ee apse ea ah AL ee ee geen 6. S. pyriformis Basal leaves palmately 3- to 5-lobed, lobes short; petals purple; sepals purple, ca 2.5 mm long; siliques broadly ovate to oblong or linear, 5 to 19 mm long, 3 to6mm wide ............. 1. S. borealis Literature Cited Dawe, J. C., and D. F. Murray. 1981. Chromosome numbers of selected Alaskan vascular plants. Canadian Journal of Re Botany 59: 1373-1381. ceived 17 May 2000 Accepted 31 May 2001 2001 NOTES 343 Trillium ovatum Pursh variety hibbersonii (Taylor et Szczawinski) Douglas et Pojar, variety nova GEORGE W. DOUGLAS! and JIM POJAR2 'Conservation Data Centre, British Columbia Ministry of Environment, Lands, and Parks, Resource Inventory Branch, Wildlife Inventory Section, P.O. Box 9344 Station Provincial Government, Victoria, British Columbia V8W 9M1 Canada 2British Columbia Forest Service, Postal Bag 5000, Smithers, British Columbia VOJ 2NO Canada Douglas, George W., and Jim Pojar. 2001. Dwarf Trillium, Trillium ovatum Pursh variety hibbersonii (Taylor et Szczawinski) Douglas et Pojar, variety nova. Canadian Field-Naturalist 115(2): 343. A new combination, Trillium ovatum Pursh var. hibbersonii (Taylor et Szczawinski) Douglas et Pojar, is made within the species Trillium ovatum in order to more accurately reflect current taxonomic opinion and provide a more readily traced name in connection with research and conservation work. Key Words. Dwarf Trillium, Trillium ovatum variety hibbersonii, new combination, British Columbia. The Dwarf Trillium, found in British Columbia, has long been of interest to avid rock gardeners and was formally named Trillium ovatum Pursh forma hibber- sonii Taylor & Szczawinski by Taylor and Szczawinski (1974). It is now apparent that this taxon is best treated at the varietal level since it differs from the var. ovatum in its consistently shorter stature (which it maintains in the garden), its initial flower color (pink versus white) and its restriction to moist mossy cliffs and river boulders on western Vancouver Island. Production of the Illustrated Flora of British Columbia (Douglas et al. 2001) requires a new combi- nation for this taxon at the varietal level because the flora does not recognize forma or trivial variations of plants. This change of rank will more accurately reflect current taxonomic opinion and provide a more readily traced name in connection with research and conservation work. The new combination is docu- mented below: Trillium ovatum Pursh variety hibbersonii (Taylor et Szczawinski) Douglas et Pojar, variety nova. Basionym: Trillium ovatum Pursh forma hibbersonii Taylor et Szczawinski. Syesis 7: 250. 1974. Literature Cited Douglas, G. W., D. Meidinger and J. Pojar. 2001. Illus- trated flora of British Columbia. Volume 6. Monocotyle- dons (Acoraceae through Najadaceae). Ministry of Envi- ronment, Lands and Parks, Ministry of Forests, Victoria. 361 pages. Taylor, T. M. C., and A. F. Szezawinski. 1974. Trillium ovatum Pursh forma hibbersonii Taylor et Szczawinski. Syesis 7: 250. Received 20 January 2000 Accepted 19 April 2001 Killing of a Bison, Bison bison, Calf by a Wolf, Canis lupus, and Four Coyotes, Canis latrans, in Yellowstone National Park DoucLas W. SmitH!, KERRY M. Murpuy!, and STAN MONGER? Yellowstone Center for Resources, P.O. Box 168, Yellowstone National Park, Wyoming 82190 USA Gallatin Flying Service, Bozeman, Montana 59715 USA Smith, Douglas W., Kerry M. Murphy, and Stan Monger. 2001. Killing of a Bison, Bison bison, calf by a Wolf, Canis lupus, and four Coyotes, Canis latrans, in Yellowstone National Park. Canadian Field-Naturalist. 115(2): 343-345. We describe a fatal attack on a Bison calf (Bison bison) in Yellowstone National Park by a Wolf (Canis lupus) and four Coyotes (Canis latrans). A lone adult Wolf bit the neck of a lone, malnourished Bison calf while four Coyotes, simultane- ously bit at the hindquarters. After the calf died, the Wolf was intolerant of Coyote proximity to the dead calf, and did not allow the Coyotes to feed or approach the carcass. Key Words: Wolf, Canis lupus, Bison, Bison bison, Coyote, Canis latrans, predation Wood Buffalo National Park, Canada (Carbyn and Trottier 1987, 1988; Carbyn et al. 1993) and are fair- Observations of Wolves (Canis lupus) killing Bison (Bison bison) until now have been limited to 344 ly uncommon. Most Wolf-Bison interactions involve more than one Wolf (e.g., a pack of Wolves pursuing Bison) and do not involve other carnivores (interspe- cific cooperation). Bison may be the largest and most difficult prey for Wolves to kill (Mech 1970; Carbyn et al. 1993), so accounts of a single Wolf attacking a Bison are of interest. Of equal interest is when multi- ple predator species cooperate in making kills. In this case, one Wolf and four Coyotes (Canis latrans) simultaneously attacked a Bison calf. Wolves and Coyotes typically do not cooperatively hunt (Mech 1966; Paquet 1992), and Wolves have consistently killed Coyotes since Wolves were reintroduced to Yellowstone National Park (YNP) in 1995 (Crabtree and Sheldon 1999; YNP unpublished data). Wolf-Bison interactions are becoming more com- mon in YNP as Wolves learn how to prey on Bison (Smith et al. 2000). Wolves have focused their attacks on weak Bison in late winter (Smith et al. 2000), but the observation reported on here is the only interaction where one Wolf was involved, and where Coyotes attacked the same Bison with a Wolf. The kill took place at Gibbon Meadows, west- central YNP, on 24 March 1999. Weather conditions there are characteristically harsh with long winters (~ 5 — 6 months) and deep snow (90 — 150 cm), even during average winters. Gibbon Meadows is large, approximately 100 ha, and surrounded by Lodgepole Pine (Pinus contorta) forests and provides habitat for wintering Bison and Elk (Cervus elaphus). Because forage is typically limited in winter due to snow, Bison are often weakened by March or April (Smith et al. 2000). Winter and early-spring losses of Bison to malnutrition are common, numbering 2 to 10 per winter (depending on winter severity) in the mead- ows out of a wintering population of about 40-80 Bison. We made our observation while counting Bison from a Supercub airplane 160-170 m above ground. At 0800, we observed a Wolf and four Coyotes attacking a Bison calf. Based on its size the Wolf was >1 year old and likely less than 3 years old, because it was progeny of Wolves reintroduced to YNP in 1995 and 1996 (Bangs and Fritts 1996; Phillips and Smith 1996; Bangs et al. 1999). The four Coyotes appeared to be adults based on their size, but could have been yearlings. When we first saw the interaction, the Bison calf was prone with its head up and the Wolf was biting the underside of its neck, while the four Coyotes simultaneously were biting the hindquarters. We assumed the bites to the neck were fatal, as we could see no damage from the bites to the rear. This first Supercub left the area, but contacted by radio anoth- er Supercub that was aerially tracking wolves and which arrived on the scene 30 min later. The second Supercub observed the Wolf feeding on the Bison calf, now dead, with four Coyotes bed- THE CANADIAN FIELD-NATURALIST Vol iis ded approximately 60 — 70 m away. The Wolf remained alone at the carcass while we circled for five minutes after which we departed. Although the calf was first observed down, we believe it was brought down by the Wolf, and not attacked while prone. We have observed other situa- tions in YNP with malnourished Bison calves, and they stand until almost dead and often survive for prolonged periods in poor condition (Smith et al. 2000). We do not know if the calf was abandoned by its mother, but at the time of the kill the meadow was the only snow-free area and other Bison were nearby. This is the first reported observation of a Wolf and Coyotes simultaneously attacking prey. However, we doubt that the two species of canids were forag- ing cooperatively. When Wolves and Coyotes inter- act in YNP, Wolves usually chase and often kill Coyotes (Allen 1979; Crabtree and Sheldon 1999). Several incidents in YNP have been observed where several Coyotes have harassed and chased away lone Wolves, but this has never been observed at a car- cass. The food incentive may have caused the Wolf to defend the carcass more vigorously, but we do not know how long the Wolf was able successfully to control the carcass. Acknowledgments We appreciate capable piloting of research aircraft by R. Stradley. We thank D. S. Guernsey for editori- al comments. L. D. Mech encouraged us to publish this account and also provided comments on the manuscript. Literature Cited Allen, D. L. 1979. The wolves of Minong: Their vital role in a wild community. Houghton Mifflin Company, Boston, Massachusetts. 499 pages. Bangs, E. E., and S. H. Fritts. 1996. Reintroducing the gray wolf to central Idaho and Yellowstone National Park. Wildlife Society Bulletin 24: 402-413. Bangs, E. E., S. H. Fritts, J. A. Fontaine, D. W. Smith, K. M. Murphy, C. M. Mack, and C. C. Niemeyer. 1998. Status of gray wolf restoration in Montana, Idaho, and Wyoming. Wildlife Society Bulletin 26: 785-798. Carbyn, L. N., and T. Trottier. 1987. Responses of bison on their calving grounds to predation by wolves in Wood Buffalo National Park. Canadian Journal of Zoology 65: 2072-2078. Carbyn, L. N., and T. Trottier. 1988. Descriptions of wolf attacks on bison calves in Wood Buffalo National Park. Arctic 41: 297-302. Carbyn, L. N., S. M. Oosenbrug, and D. W. Anions. 1993. Wolves, bison ... and the dynamics related to the Peace-Athabasca delta in Canada’s Wood Buffalo National Park. Circumpolar Research Series Number 4, Canadian Circumpolar Institute, University of Alberta. 270 pages. Crabtree, R. L., and J. W. Sheldon. 1999. Coyotes and canid coexistence in Yellowstone. Pages 127-163 in Carnivores in ecosytems: The Yellowstone experience. Edited by T. W. Clark, A. P. Curlee, S.C. Minta, and 2001 P.M. Kareiva. Yale University Press, New Haven and London. 429 pages. Mech, L. D. 1966. The wolves of Isle Royale. Fauna of the National Parks of the United States, Fauna Series 7. 210 pages. Mech, L. D. 1970. The wolf: The ecology and behavior of an endangered species. Natural History Press, Garden City, New York. 384 pages. Paquet, P. C. 1992. Prey use strategies of sympatric wolves and coyotes in Riding Mountain National Park, Manitoba. Journal of Mammalogy 73: 337-343. NOTES 345 Phillips, M. K., and D. W. Smith. 1996. The wolves of Yellowstone. Voyageur Press, Stillwater, Minnesota. 125 pages. Smith, D. W., L. D. Mech, M. Meagher, W. E. Clark, R. Jaffe, M. K. Phillips, and J. A. Mack. 2000. Wolf- bison interactions in Yellowstone National Park. Journal of Mammalogy 81: 1128-1135. Received 21 February 2000 Accepted 24 May 2001 Short-eared Owl, Asio flammeus, Attack on a Burrowing Owl, Athene cunicularia, in Suffield National Wildlife Area, Alberta SARAH D. CARNEGIE, ERIN J. URTON, and DAviIp L. GUMMER! Biology Department, University of Saskatchewan, 112 Science Place, Saskatoon, Saskatchewan S7N 5E2, Canada ‘Corresponding author, present address: Mammalogy Program, Provincial Museum of Alberta, 12845 - 102 Avenue, Edmonton Alberta TSN O0M6, Canada Carnegie, Sarah D., Erin J. Urton, and David L. Gummer. 2001. Short-eared Owl, Asio flammeus, attack on a Bur- rowing Owl, Athene cunicularia, in Suffield National Wildlife Area, Alberta. Canadian Field-Naturalist 115(2): 345-346. During nighttime surveys for Ord’s Kangaroo Rats in Suffield National Wildlife Area, Alberta, we observed a Short-eared Owl attacking a Burrowing Owl. The incident is of interest because aggressive interactions between these two species have apparently never been documented and because both species are considered at risk by the Committee On the Status of Endangered Wildlife In Canada. Presumably, aggression between these two species is rare; however, such interactions may play a more important role than previously realized in governing the owls’ territories and survival, especially in areas of high owl and/or prey densities. Key Words: Burrowing Owl, Athene cunicularia, Short-eared Owl, Asio flammeus, attack, attempted predation, aggression, interaction, territoriality, Alberta. Incidental to our field studies of Ord’s Kangaroo Rats (Dipodomys ordii) in southeastern Alberta, we recorded observations of Great Horned Owls (Bubo virginianus), Snowy Owls (Nyctea scandiaca), Burrowing Owls (Athene cunicularia), Long-eared Owls (Asio otus), and Short-eared Owls (A. flam- meus). This note recounts an aggressive interaction that we observed between a Short-eared Owl and a Burrowing Owl on 21 June 1999. We conducted nighttime surveys for Ord’s Kanga- roo Rats in Suffield National Wildlife Area (SNWA; 50° 35’ N, 110° 25’ W) on 45 nights during May and June 1999. SNWA occupies 460 km? in the eastern part of Canadian Forces Base Suffield, Alberta. Eolian sand dunes, stabilized by native grassland and prairie scrub vegetation, are the most conspicuous landforms in SNWA. Our surveys consisted of field personnel driving a vehicle (< 25 kmeh'!) along gravel roads, trails, and sandy firebreaks, or hiking on open sand dunes, with spotlights (10° candlepower, Brinkmann Corporation, Dallas, Texas) aimed at the edges of vegetation (Kaufman and Kaufman 1982; Gummer et al. 1997; Ralls and Eberhardt 1997). On 21 June 1999, we conducted a spotlight survey along a 15 m wide fireguard in SNWA. At 00:43, we crested a small hill and noticed a Short-eared Ow! in the process of attacking a Burrowing Owl on the ground in the centre of the fireguard. Using its talons, the Short-eared Owl had the Burrowing Owl pinned to the ground. The Burrowing Owl was on its back with its wings spread open and its talons raised in a defensive posture. We approached in our truck to a distance of approximately 5 m from the owls. The Short-eared Owl turned and looked at us but remained atop the Burrowing Owl until 00:48. The Short-eared Owl then fled the scene temporarily. The Burrowing Owl, its movements clearly hindered by its injuries, fluttered across the fireguard to the east bank. The Short-eared Owl suddenly returned (00:49) and resumed the attack briefly before depart- ing. The Burrowing Owl then dropped down from the bank (1 m elevation) onto the road. Its wings were outstretched and twisted and its right eye appeared to be injured. At 00:51 the wounded Burrowing Owl retreated into an abandoned Badger (Taxidea taxus) den that was 2 m away. 346 According to Earhart and Johnson (1970), Short- eared Owls are approximately twice the mass of Burrowing Owls (315 g versus 146 g, respectively). Hence it is not surprising that the larger owl dominat- ed the smaller owl. We do not know if the Burrowing Owl eventually died of its injuries, or if it would have been killed had we not interrupted the attack. We speculate that the interaction may have related to: (1) an attempted predation by the Short-eared Owl on the Burrowing Owl; (ii) conflict over a potential prey item in the vicinity; or (11) a territorial dispute. Short-eared Owls are not documented as predators of Burrowing Owls (Haug et al. 1993) whereas Cooper’s Hawks (Accipiter cooperii), Swainson’s Hawks (Buteo swainsoni), Red-tailed Hawks (B. jamaicensis), Ferruginous Hawks (B. regalis), Merlins (Falco columbarius), Peregrine Falcons (F. peregri- nus), Prairie Falcons (F. mexicanus), Great Horned Owls, and American Crows (Corvus brachyrhynchos) and are all known threats to Burrowing Owls (Wedgewood 1978*; Konrad and Gilmer 1984; Haug et al. 1993). Short-eared Owls are not known to kill any other raptors (Holt and Leasure 1993). Both Short-eared Owls and Burrowing Owls are considered at risk by the Committee On the Status of Endangered Wildlife In Canada (COSEWIC 2000*). Because of long-term declines in population esti- mates, Short-eared Owls are listed as “special con- cern” (Cadman and Page 1994*; COSEWIC 2000*) and Burrowing Owls are classed as “endangered” (Wedgewood 1978*; Haug and Didiuk 1991*; Wellicome and Haug 1995*; COSEWIC 2000%*). Aggression between these owl species is probably rare, although it may be more common in areas of unusually high densities of owls and/or prey (e.g., Ord’s Kangaroo Rats in SNWA). Interspecific ag- gression may play a more important role than previ- ously realized in determining these owls’ territories and survival. Novel interactions among owls may also become more common as patches of natural habitat and abundant prey decline in frequency and size. Thus, interaction between owls may be yet another factor to consider for conservation and man- agement of Short-eared Owls and Burrowing Owls. Acknowledgments We are grateful to the late Malcolm A. Ramsay for his support and encouragement of our field efforts. We thank the Canadian Wildlife Service (Prairie and Northern Region) and Department of National Defence (Canadian Forces Base Suffield) for logisti- cal support of our activities in SNWA. The Alberta Challenge Grants in Biodiversity provided funds for THE CANADIAN FIELD-NATURALIST Vol. 115 our Ord’s Kangaroo Rat research through a grant to DLG and the Natural Sciences and Engineering Research Council of Canada supported our research through an undergraduate scholarship to SDC, a post- graduate scholarship to DLG, and a research grant to Malcolm A. Ramsay. Documents Cited (marked * in text) Cadman, M.D., and A.M. Page. 1994. Status report on the short-eared owl, Asio flammeus, in Canada. Com- mittee On the Status of Endangered Wildlife In Canada. Ottawa. COSEWIC. 2000. Canadian species at risk. Committee On the Status of Endangered Wildlife In Canada. Ottawa. Haug, E. A., and A. B. Didiuk. 1991. Updated status report on the Burrowing Owl Athene cunicularia hypugaea in Canada. Committee On the Status of Endangered Wildlife In Canada. Ottawa. Wedgewood, J. A. 1978. The status of the Burrowing Owl in Canada. Committee On the Status of Endangered Wildlife In Canada. Ottawa. Wellicome, T. L., and E. A. Haug. 1995. Status report on the burrowing owl Speotyto cunicularia in Canada. Committee On the Status of Endangered Wildlife In Canada. Ottawa. Literature Cited Earhart, C.M., and N. K. Johnson. 1970. Size dimor- phism and food habits of North American owls. Condor 72: 251-264. Gummer, D. L., M. R. Forbes, D. J. Bender, and R. M. R. Barclay. 1997. Botfly (Diptera: Oestridae) parasitism of kangaroo rats (Dipodomys ordii) at Suffield National Wildlife Area, Alberta, Canada. Journal of Parasitology 83: 601-604. Haug, E.A., B.A. Millsap, and M.S. Martell. 1993. Burrowing Owl. Jn The Birds of North America, Num- ber 61. Edited by A. Poole, and F. Gill. The Academy of Natural Sciences, Philadelphia, and The American Ornithologists’ Union, Washington D.C. Holt, D. W., and S.M. Leasure. 1993. Short-eared Owl. In The Birds of North America, Number 62. Edited by A. Poole, and F. Gill. The Academy of Natural Sciences, Philadelphia, and The American Ornithologists’ Union, Washington D.C. Kaufman, D. W., and G. A. Kaufman. 1982. Effect of moonlight on activity and microhabitat use by Ord’s kangaroo rat (Dipodomys ordii). Journal of Mammalogy 63: 309-312. Konrad, P. M., and D.S. Gilmer. 1984. Observations on the nesting ecology of Burrowing Owls in central North Dakota. Prairie Naturalist 16: 129-130. Ralls, K., and L. L. Eberhardt. 1997. Assessment of abundance of San Joaquin kit foxes by spotlight surveys. Journal of Mammalogy 78: 65-73. Received 17 December 1999 Accepted 9 April 2001 2001 NOTES 347 Arboreal Courtship Behaviour by Eastern Garter Snakes, Thamnophis sirtalis sirtalis, in September in Bruce County, Ontario DAvip A. GALBRAITH Science Department, Royal Botanical Gardens, P.O. Box 399, Hamilton, Ontario L8N 3H8 Canada Galbraith, David A. 2001. Arboreal courtship behaviour by Eastern Garter Snakes, Thamnophis sirtalis sirtalis, in September in Bruce County, Ontario. Canadian Field-Naturalist 115(2): 347-348. Courtship behaviour involving an adult female and two male Eastern Garter Snakes (Thamnophis s. sirtalis) was observed near Lake Huron in Bruce County, Ontario, in September 1997. Courtship behaviour took place with all three animals located among the branches of a small Jack Pine (Pinus banksiana) about 1 m above ground level. Instances of arboreal courtship in Eastern Garter Snakes appear to be most frequent in late summer or early fall in Ontario, and may be related to thermoregulation through direct insolation. Key Words: Eastern Garter Snake, Thamnophis sirtalis sirtalis, arboreal, fall, courtship behaviour, Ontario, Canada. I observed Eastern Garter Snakes (Thamnophis sirtalis sirtalis) on private land in a cottage commu- nity on the coastal sand dune area of the eastern shore of Lake Huron (81°44’ W, 44°06’ N), approxi- mately 10 km south-west of the town of Kincardine, Bruce County, Ontario. The area has been developed gradually over the past 40 years from open sand dunes with some native poplars to its present mix of native and introduced trees, bushes and other plants. Eastern Garter Snakes have been seen on occasion in the immediate area in previous years. Three snakes were observed in courtship in a small Jack Pine (Pinus banksiana) approximately 1 m from ground level, beginning at 09:15 h, 14 September 1997 (all times reported are local, DST). The snakes were lying in a fairly exposed position on the eastern side of the tree, approximately 3 m from the front porch of a cottage. Although sunrise occurred at 07:05 h on 14 September 1997 at this location (United States Naval Observatory 1997), the proximity of the cottage kept the snakes’ location in shade until approximately 10:20 h. Air temperature on the shaded side of the cottage was approximately 17°C at 09:15 h. Observations were made from (a) behind a bush approximately 2 m from the snakes, and (b) from a garden bench approximately 4 m from the subjects. I began recording observations and taking pho- tographs at 09:22 h. I designated the snakes as F (female), M1 (male proximate to F) and M2 (male distant from F). When discovered, M1 was on F and M2 was about 20-25 cm distant. M1 appeared small- er than M2. The abdomen of F had a very flaccid appearance, and I conjecture that she had recently given birth. The female F appeared to be particularly large (estimated snout-vent length 700 mm). Between 09:22 h and 11:00 h the snakes moved very little. M1 actually moved off twice, circling through the branches of the Jack Pine, to return to rest in contact with the female. Sunlight began to hit the males at 10:20 h, and the female by 10:40 h, but remained occasional because of broken cloud cover. By 10:52 h, all three snakes were in continuous sun- light. At 11:00 h the air temperature had risen to approximately 20°C, both males started to move into contact with F, and for the first time probed the air with their tongues. Between 11:09 h and 11:50 h, all three snakes were occasionally in motion. At first, they formed a tight coil, but by 11:32 h looser body coils hung down from the tree. The large female was coiled with her tail and posterior 10 cm of trunk hanging loose. Both males wrapped their tails and lower trunks around the female and appeared to be seeking cloacal apposition, although none was observed (Figure 1). At 11:50 h, all three snakes moved quickly to ground level. The transition was rapid, and I do not know whether they fell or moved down of their own accord. Once on the ground, contact between the three snakes appeared to continue for approximately five minutes, after which one of the males moved away from the female, while the other remained in contact with her. I caught glimpses of the female for the next few minutes, but she gradually moved deep- er under ground cover and I was unable to make fur- ther observations. Neither copulation nor direct cloa- cal apposition were observed. The snakes were not captured following observation. Snakes in the genus Thamnophis most frequently copulate in spring but have occasionally been noted to copulate in the fall, immediately prior to entering hibernacula (Cook 1984). Thamnophis sirtalis may mate frequently in the fall, as sperm may overwinter in the oviduct prior to entering the seminal recepta- cles of the female in early spring (Fox 1956). The possibility of fall mating, followed by sperm storage over winter, has been noted previously in the Red- sided Garter Snake (Thamnophis sirtalis parietalis; Crews 1984; Crews et al. 1984; Whittier and Crews, 1986). 348 FIGURE 1. One female and two male Eastern Garter Snakes (Thamnophis sirtalis sirtalis) exhibiting courtship behaviours in a small Jack Pine (Pinus banksiana) on 14 September 1997 in Bruce County, Ontario. Arboreal mating behaviour by Thamnophis s. sir- talis was reported in April of 1992 in Massachusetts, when five snakes were discovered in a bird’s nest within a Black Willow tree (Salix nigra; Martinez 1992). Arboreal courtship and mating behaviour in the Red-sided Garter Snake has been reported in the Interlake Region of Manitoba (Gregory 1975), but in spring. Unpublished observations of fall courtship or cop- ulatory behaviour and arboreal activity by Thamnophis s. sirtalis in Ontario were found in the Ontario Herpetofaunal Summary (OHS) database, made between 1956 and 1994 (M. Oldham, personal communication). Twelve observations of copulatory behaviour were recorded in the database. Two instances of courtship or copulation were observed between August and October, and arboreal activity was reported four times. All four observations of arboreal activity also took place between late August and early October. Twice a single individual rested in a tree or bush, in one instance two snakes were observed together in a tree, and in the fourth instance a male and female were observed mating in a Juniper THE CANADIAN FIELD-NATURALIST Vol. 115 (Juniperus communis) on 14 September, 1992 in the Oakville area of Ontario (unpublished observation of D. Gregory; all OHS unpublished data: M. Oldham personal communication). From the instance reported here and those in the OHS database, it appears that arboreal activity by Thamnophis sirtalis sirtalis may be most common in or around September in Ontario. Arboreal activity may present Garter Snakes with an opportunity to increase body temperature through direct insolation in the cool fall months. Arboreal behaviour also may minimize contact with the ground, which may act as a heat-sink if it is colder than a snake’s body temper- ature. It is also possible that the warming effect of direct sunlight encourages dissemination of olfactory signals important for copulatory behaviour. Acknowledgments Thanks to M. Oldham, W. Weller and the many participants in the Ontario Herpetofaunal Summary project for the OHS observations reported here, and to D. Gregory for confirmation of his unpublished observations. Thanks to R. Brooks, F. R. Cook, L. Rye, M. Oldham and two anonymous referees for constructive comments on earlier drafts of this paper. Documents Cited United States Naval Observatory. 1997. Time service department, U.S. Naval Observatory. On-Line Document. (http://tycho.usno.navy.mil). Literature Cited Cook, F. R. 1984. Introduction to Canadian amphibians and reptiles. National Museum of Natural Sciences, Ottawa, Ontario. 200 pages. Crews, D. 1984. Gamete production, sex hormone secre- tion, and mating behavior uncoupled. Hormones and Behavior 18: 22-28. Crews, D., B. Camazine, M. Diamond, R. Mason, R. R. Tokarz and W. R. Garstka. 1984. Hormonal indepen- dence of courtship behavior in the male garter snake. Hormones and Behavior 18: 28-41. Fox, W. 1956. Seminal receptacles in snakes. Anatomical Record 124: 519-539. Gregory, P. T. 1975. Arboreal mating behavior in the Red-sided Garter Snake. Canadian Field-Naturalist 89: 461-462. Martinez, J. 1992. Arboreal mating behavior in Thamno- phis sirtalis sirtalis. Bulletin of the Chicago Herpeto- logical Society 27(4): 96. , Whittier, J. M., and D. Crews. 1986. Ovarian develop- ment in redsided garter snakes (Thamnophis sirtalis parietalis): relationship to mating. General and Com- parative Endocrinology 61: 5—12. Received 26 May 1998 Accepted 9 April 2001 2001 NOTES 349 Increases and Expansion of the New Brunswick Breeding Population of Black-legged Kittiwakes, Rissa tridactyla F. PATRICK KEHOE!:? and ANTONY W. DIAMOND? 'Rish and Wildlife Branch Department of Natural Resources and Energy, P.O. Box 6000, Fredericton, New Brunswick, E3B 5H1 Canada Present address: Ducks Unlimited Canada, Box 818, Brooks, Alberta T1R 2B7 Canada 3Atlantic Cooperative Wildlife Ecology Research Network, University of New Brunswick, P.O. Box 45111, Fredericton, New Brunswick E3B 6E1 Canada Kehoe, F. Patrick, and Antony W. Diamond. 2001. Increases and expansion of the New Brunswick breeding population of Black-legged Kittiwakes, Rissa tridactyla. Canadian Field-Naturalist. 115(2): 349-350. The New Brunswick breeding population of Black-legged Kittiwakes has increased to over 100 pairs over the past seven years. There are now colonies at two sites. Key Words: Black-legged Kittiwake, Rissa tridactyla, colony, breeding, New Brunswick. Black-legged Kittiwakes were reported to breed in New Brunswick among the islands in the Bay of Fundy in the 1800s (Audubon 1840; Chamberlain 1887). However, those early accounts came into question during the species’ absence as a breeder in the 1900s (Squires 1952; Palmer 1949). With the discovery of a small colony of 12 nesting pairs on South Wolf Island in the Bay of Fundy (44° 56’N, 66° 44’W) in 1992, Kehoe (1994) documented the first recent records of Black-legged Kittiwakes (hereafter “kittiwake’”’) breeding in New Brunswick. In 1993, there were 25 nests at South Wolf Island. Since that time the colony has continued to grow, and has supported up to 135 nests. However on 6 July 1996 an exceptional hurricane-induced storm caused the loss of one-third of the nests at this site (K. Mawhinney, unpublished). Kehoe (1994) specu- lated that Whitehorse Island (44° 59’ N, 66° 52’ W), 12 km north-west of South Wolf, could become a second colony. Kittiwake nests were first found by A.W.D. on Whitehorse on 20 July 1998, and it sup- ported at least 42 nesting pairs and 7 visible downy young on 21—22 July 1999. Some of the 1999 pairs may have relocated from South Wolf Island, as at least that many pairs disappeared from South Wolf some time after 5 May (Table 1). There is a previous record of 20 pairs of kitti- wakes nesting on “Whitehorse Islet” in Lock et al. (1994, Table 5.1). This record is erroneous. White- horse Islet is a small flat rock in the Grand Manan archipelago, and lacks the cliff habitat required by kittiwakes; evidently this record confused the locali- ty, as is clear from the coordinates. A. R. Lock (per- sonal communication) confirmed that the record was based on observations by P. Pearce who in turn (per- sonal communication) confirmed that the birds were roosting or prospecting but were not confirmed as breeding, and that the island in question was Whitehorse Island not Whitehorse Islet. Our records therefore are the first confirmation of breeding by kittiwakes on Whitehorse Island. | Kittiwakes have been counted at South Wolf, and Whitehorse checked for their presence, annual- ly since the initial colony was discovered in 1992. We have counted each of these sites independently most summers (Table 1). Both colonies are located on rock cliffs facing south (South Wolf) or south- west (Whtehorse) and nests are located 3-10 m above the high tide line. All surveys have been con- ducted by boat, from a distance of 15-20 m from the base of the nesting cliff, with one primary observer doing the count. Counts by F. P. K. have been entirely visual; A.W.D. also made visual counts but confirmed them from counts from pho- tographs taken by either himself, K. Mawhinney or F. Huettmann. Nest sites (equivalent to the “occu- pied sites” of Walsh et al. 1995) were identified by the presence of nesting material (seaweeds and grasses). Ledges that were occupied by a pair of kittiwakes without nest material present were not counted. Breeding was confirmed at each colony by the presence of downy young, but the fate of indi- vidual nests was not documented. Variation in counts within years can be explained by differential identification of nests by different primary obser- vers as well as the addition or loss of nests between counts. Counts in May tend to be higher than those in June or July because nest-building begins in May and it is harder to distinguish nests at that time than in July (so roosting birds might be wrongly counted as nesting); A.W.D., who made all the May counts, may also be less conservative than F.P.K. in decid- ing whether or not a site is occupied by nesting birds. Since 1992 the breeding population of Kittiwakes in New Brunswick has increased to exceed 100 pairs in most breeding seasons (Table 1). In 1999 the occupation of a second colony by at 350 THE CANADIAN FIELD-NATURALIST VolzitS TABLE 1. Numbers of Nests of Kittwakes counted at South Wolf and Whitehorse Islands, New Brunswick. Year Location Observer Date Nest Count 1992 South Wolf F.P.K. 8 June : 8 South Wolf EP:K. 20 June 1 Whitehorse F.P.K. 20 June 0 1993 South Wolf F.PLK. 10 May 0 South Wolf EO PaKe 9 July D5 Whitehorse F.P.K. 9 July 0 1994 South Wolf Papa 8 July 69 Whitehorse F.P.K. 8 July 0 1995 South Wolf FPLK 20 June 104 South Wolf Mawhinney and Sears 1996 July 132 Whitehorse EP 20 June 0 1996 South Wolf BP. Ke 22 July 54 Whitehorse EPA 22 July 0 1997 South Wolf A.W.D. 13 May 135 South Wolf EP AKe 9 June 89 South Wolf A.W.D. 25 July oD Whitehorse A.W.D. 13 May 0 1998 South Wolf A.W.D. 6 May 124 Whitehorse A.W.D. 20 July 4 1999 South Wolf A.W.D. 5 May 134 South Wolf A.W.D. 22 July 63 Whitehorse A.W.D. 5 May I) Whitehorse F.P.K. 21 July 42 Whitehorse A.W.D. 22 July 48 least 42 nesting pairs has further secured the re- establishment of this species to its former range in the Bay of Fundy. Other species of sea birds have also recently re- colonized these and nearby islands. Razorbills have been confirmed as breeding on South Wolf since 1995 (Mawhinney and Sears 1996) and Northern Gannets on Whitehorse in 1999 (Corrigan and Diamond unpublished); both these species had been -absent as breeders from these islands since the 1800s. It is clearly important that the seabird populations at these sites continue to be monitored. Protection from the encroachment of aquaculture and fisheries activity as well as other forms of development and disturbance would likely ensure the continued suc- cess of seabirds breeding at these sites. In particular, Whitehorse Island has become a stop for tourist boats that pass close to the colonies, as deep water allows even large boats close access to the base of the cliffs. However such disturbance can result in decreased breeding success through the loss of chicks to predators and inconsistent incubation. Both cliff faces are currently under crown ownership, but have no formal protection. This is ACWERN publication number UNB-21. Literature Cited Audubon, J. J. 1840. The birds of America, Volume 6. New York and Philadelphia. Chamberlain, M. 1887. A catalogue of Canadian birds with notes on the distribution of the species. J. and A. McMillan. St. John, New Brunswick. Corrigan, S., and A. W. Diamond. 2001. Northern Gannet, Morus bassanus, nesting on Whitehorse Island, New Brunswick. Canadian Field-Naturalist 115: 176-177. Kehoe, F. P. 1994. A New Brunswick Black-legged Kitti- wake, Rissa tridactyla, colony. Canadian Field-Naturalist 108: 375-376. Mawhinney, K., and D. Sears. 1996. First nesting of the Razorbill, Alca torda, in the Wolves Archipelago, New Brunswick. Canadian Field-Naturalist 110: 698-700. Palmer, R.S. 1949. Maine Birds. Bulletin of the Museum of Comparative Zoology of Harvard College, Cam- bridge, Massachusetts. Volume 102. Squires, W. A. 1952. The Birds of New Brunswick. The New Brunswick Museum, Saint John, Monograph number 2. Walsh, P. M., D. J. Halley, M. P. Harris, A. del Nevo, I.M. W. Sim, and M. L. Tasker. 1995. Seabird moni- toring handbook for Britain and Ireland. Joint Nature Conservancy Council, Royal Society for Protection of Birds, Institute for Terrestrial Ecology Seabird Group, Peterborough, U.K. Received 11 February 2000 Accepted 30 April 2001 2001 NOTES 35] A Yellow Wood Lily, Lilium philadelphicum, from Nantucket Island, Massachusetts, With Notes on its Occurrence in New England DENVER W. Hott! and WESLEY N. TIFFNEY2 ‘Owl Research Institute, P.O. Box 39, Charlo, Montana 59824 USA "University of Massachusetts, Nantucket Field Station, P.O. Box 756, Polpis Road, Nantucket, Massachusetts 02554 USA Holt, Denver W., and Wesley N. Tiffney. 2001. A Yellow Wood Lily, Lilium philadelphicum, from Nantucket Island, Massachusetts; with notes on its occurrence in New England. Canadian Field-Naturalist 115(2): 351-352. In July 1985, we discovered a pure yellow flowering morph of the Wood Lily (Lilium philadelphicum) L. forma flaviflo- rum Williams, on Nantucket Island, Nantucket County, Massachusetts. This appears to be the first formal report of this color variation on Nantucket Island since 1915, suggesting its rarity. Key Words: Wood Lily, Lilium philadelphicum f. flaviflorum, yellow, Nantucket Island, Massachusetts, New England. On 31 July 1985, we found a Wood Lily (Lilium philadelphicum) with a pure yellow, unspotted flower. It was noted between Miacomet Pond and Miacomet Road, near the south-central shore of Nantucket Island, Massachusetts. Nantucket Island (130 km?) is located in southeastern Massachusetts, and approxi- mately 32 km south of the “elbow” of Cape Cod and 65 km southeast of the New England coast. The island is geologically composed of moraines and out-wash plains derived from Pleistocene glaciation (Woods- worth and Wigglesworth 1934). The island is unique and noted for its composition of maritime heaths or “moorlands” (Tiffney and Eveleigh 1985). We identified this find as Lilium philadelphicum L. forma flaviflorum Williams (Williams 1913). The plant grew in a mixed grassy heath and low scrub community. It occurred alongside 72 other Wood Lilies, all sporting the typical orange flower decorat- ed with purple-black spots, characteristic of this species. We returned three days later and pho- tographed the Lily, still in full bloom, but lacking some luster to its color and physical integrity. In 1986, we returned to the site and found only a few typical Wood Lilies and none of the yellow color. Two color morphs of the Wood Lily have been noted by previous authors on Nantucket. Owen (1888) wrote under L. philadelphicum that “A pure orange flower without spots is sometimes found”, and Albertson and Hinchman (1921) mention “A yel- low form without spots has been occasionally found ... in their natural history description of the species. Shurrocks (1958) also mentions the yellow unspotted morph, and probably is referring to Albertson and Hinchman (1921). Neither of these authors elaborated on the number of finds, dates and locations, and this may reflect their unfamiliarity with this morph. The herbarium at the Maria Mitchell Science Center on Nantucket Island, contains one specimen of “L. philadelphicum L.” collected by U. P. Wehmer on 22 July 1915, from Squam, Nantucket. We could see no spots on this specimen (possibly faded), and on the label was the notation “dandelion yellow”. These are the only records for the Nantucket yellow color morph of the Wood Lily we could find, and all may be referring to the same specimen (Albertson and Hinchman 1921; Shurrocks 1958). No specimens of the yellow morph exist at the New England Botanical Club, Harvard University, Cambridge, Massachusetts or the Gray Herbarium, Warren, New Hampshire. Furthermore, we contacted other field botanists and checked with the Massachusetts Natural Heritage Program data base and could not locate any other records of this morph. Williams (1913) collected the first specimen of this color morph from East Warren, New Hampshire, and it was believed deposited at the Gray Herbarium. In his review of the literature, he found no other reference to this color morph. Williams did however, receive information from two individuals regarding three observations of this morph. He never saw or received photographs. Other than the original stations cited by Williams (1913) (Warren, New Hampshire; Cape Elizabeth and Bathe, Maine; Marblehead, Massa- chusetts), we have found no other published records of this yellow morph in New England through 2000. Although we may have not located all the reports, the yellow morph of Lilium philadelphicum f. flaviflo- rum appears to be very rare. Protection of the open heath lands on Nantucket Island where this morph has now been reported twice may ensure its continued existence there. Photographs of the specimen have been deposited at the Maria Mitchell Science Center Herbarium, Vestal Street, Nantucket, and the Massachusetts Natural Heritage Program (MNHP). Acknowledgments We thank Bruce Sorrie (MNHP), for searching the Boston Herbaria and input to the manuscript. Douglas Beattie, Susan F. Beegel, and Jay Sisson for help in field searches of the Wood Lily. Literature Cited Albertson, A.O. and A. Hinchman. 1921. Nantucket Wildflowers. G. P. Putnam’s Sons, New York. B52 Owen, M.L. 1888. A Catalogue of Plants Growing with- out Cultivation in the County of Nantucket, Mass. Gazette Printing Co., Northhampton, Massachusetts. Shurrocks, A. A. 1958. A Grain of Mustard Seed. Nan- tucket Maria Mitchell Association, Nantucket, Massa- chusetts. Tiffney, W. N. Jr. and D. E. Eveleigh. 1985. Nantucket’s Endangered Maritime Heaths. Pages 1093-1109 in Coastal Zone “85 Volume 1. Edited by O. T. Morgan, Maryland. THE CANADIAN FIELD-NATURALIST Vol. 115 Williams, E. F. 1913. A New Form of Lilium philadelph- icum. Rhodora 15:(180) 217-218. Woodsworth, J.B. and E. Wigglesworth. 1934. Geog- raphy and Geology of the Region including Cape Cod, the Elizabeth Islands, Nantucket, Martha’s Vineyard, No Man’s Land and Block Island. Memoir of the Museum of Comparative Zoology 52. Harvard University, Cambridge. Received 24 February 2000 Accepted 5 June 2001 Initial Movements of Juvenile Piping Plovers, Charadrius melodus, from Natal Sites in Northwestern North Dakota JEFFREY M. KNETTER!, ROBERT K. MurpuHy?’, and R. Scott Lutz! Department of Wildlife Ecology, University of Wisconsin, Madison, Wisconsin 53706, USA U.S. Fish and Wildlife Service, Des Lacs National Wildlife Refuge Complex, Kenmare, North Dakota 58746, USA Knetter, Jeffrey M., Robert K. Murphy, and R. Scott Lutz. 2001. Initial movements of juvenile Piping Plovers, Charadrius melodus, from natal sites in northwestern North Dakota. Canadian Field-Naturalist 115(2): 352-353. Post-fledging movements may influence the survival of juvenile Piping Plovers (Charadrius melodus) in North America’s northern Great Plains. During July 1998, four juvenile plovers fitted with radio-transmitters departed their natal areas, on alkali lakes, in northwestern North Dakota when 24—25 days old. Two of the plovers were not found again. The other two moved about 50 km southeast and south, respectively. They moved again by the following day and could not be relocated. Juvenile Piping Plovers reared on alkali lakes in the northern Great Plains may migrate significant distances from natal areas within only 3-4 days of attaining flight. Key Words: Piping Plover, Charadrius melodus, endangered species, behavior, fledging, migration, radio telemetry, North Dakota, Great Plains. The Great Plains breeding population of Piping Plovers (Charadrius melodus) is listed as threatened in the U.S. and endangered in Canada due mainly to poor recruitment of juveniles into the breeding popu- lation (Haig 1985*; Sidle 1985). Behavior and sur- vival of juvenile Piping Plovers have been document- ed through fledging (21-28 days of age [Prindiville Gaines and Ryan 1988]) but not during post-fledging and migration. Post-fledging survival may be affected by the timing and distance of movements, and by habitats used. During 1998, we used radio telemetry to monitor movements of juvenile Piping Plovers from natal sites on prairie alkali lakes associated with the Mis- souri Coteau glacial moraine (Bluemle 1980) in southern Burke and northern Mountrail counties (48°37°N, 102°27’W), northwestern North Dakota. Piping Plovers breed on open, salt-encrusted, gravely shores (Prindiville Gaines and Ryan 1988) of 25 alkali lakes in the area. Adjacent land use is mainly small grain farming, tame hay production, and beef ranching on mixed grass prairie. *See Documents Cited section We captured juvenile Piping Plovers by hand when they were about 20 days of age and weighed them to the nearest g, using a 100-g spring scale. We attached a 1.0-g radio transmitter (< 2.5% of body weight) to individuals that weighed 40-42 g. Each transmitter (150 MHz BD-2; Holohil Systems Limited, Carp, Ontario) was 15 mm X 7mm X 4 mm, and had a 10- cm whip antenna. We clipped lower back feathers 1.5—2.5 cm anterior to the uropygial gland, used water-resistant epoxy (Titan Corporation, Linwood, Washington) to attach the transmitter to plastic mesh equal in size to the transmitter base, then attached the transmitter and mesh, with epoxy, to the bird (Warnock and Warnock 1993). Juvenile plovers were released within 15 minutes of capture. We used a three-element Yagi antenna and scan- ning receiver (Model R2000, Advanced Telemetry Systems, Incorporated, Isanti, Minnesota) to search for plovers and monitor daily movements from vehi- cles or on foot. When plovers moved from their natal areas, we searched for them using an airplane (Cessna 170) equipped with a pair of four-element, Yagi antennas (Gilmer et al. 1981). The airplane was flown about 600 m above ground along transects spaced 12-15 km apart. We searched within about 35 km north, east, and west and 180 km south of the last 2001 known location of an individual bird, focusing the search on potential habitat along the Missouri Coteau _ and nearby (30-55 km west) Missouri River. Transmitters were affixed to 15 juvenile Piping Plovers but fell off 11 of these within 24 hours of | attachment. Transmitters remained affixed on the _ other four juvenile plovers through their initial move- ments from natal areas. These four juveniles were _ captured on a 34-ha alkali lake on 15 July. On the date _ of capture, three of the juveniles were 20 days old _ (determined from records of laying or hatching dates [Murphy et al. 1999]) and the fourth was approxi- _ mately 20 days old, based on its relative size and plumage development. At this age they could rapidly flap their wings and lift off the ground. They were observed flying short distances (< 10 m) 1-3 days later (i.e., when 21—23 days old). We were unable to detect all four juvenile Piping _ Plovers, either visually or by telemetry, at their natal lake when they were 25 days old (20 July). Two were relocated on 23 July. One had moved about 50 km southeast to a complex of alkali and freshwater lakes on the Missouri Coteau. The other had moved 50 km directly south to the north shore of a large reservoir of the Missouri River (“Lake Sakakawea’”’). Neither juvenile could be relocated the following day, despite search via aircraft another 100 km south. The remain- ing two plovers were not relocated. Our observations indicate Piping Plover fledg- lings can depart natal areas on Great Plains alkali lakes within a few days of attaining flight. Further- more, their initial movements may cover consider- able distances. Movements we documented were likely initial migration. They may be corroborated by a record of a colour-banded juvenile plover that migrated more than 2000 km in less than 5 days, from its natal area on an alkali lake in central North Dakota to its wintering range on the Gulf Coast of Texas (M. R. Ryan, unpublished data). Acknowledgments Casey Kruse, U.S. Army Corps of Engineers, kindly provided guidance on radio transmitter attach- NOTES 353 ment. Our work was supported by the U.S. Fish and Wildlife Service (USFWS), Des Lacs National Wildlife Refuge Complex and Ecological Services (Bismarck, North Dakota office). We thank Mike Olson, Karen Smith, Nell McPhillips, and Fred Giese of USFWS for administrative and logistical support. Chuck Hoovestol, Ducks Unlimited, helped capture and radio-track plovers. Comments by David R. Curson, Sonya J. Knetter, and Robyn A. Niver improved the manuscript. Documents Cited (marked* in text) Haig, S.M. 1985. The status of the piping plover in Canada. Committee on the Status of Endangered Wild- life in Canada, Canadian Wildlife Service, Ottawa. Literature Cited Bluemle, J.P. 1980. Guide to the geology of northwest- ern North Dakota. North Dakota Geologic Survey, Education Series Number 8. 38 pages. Gilmer, D.S., L.M. Cowardin, R. L. Duval, L. M. Mechlin, C. W. Shaiffer, and V.B. Kuechle. 1981. Procedures for the use of aircraft in wildlife biotelemetry studies. U.S. Fish and Wildlife Service Resource Publi- cation 140. 19 pages. Murphy, R. K., B. G. Root, P. M. Mayer, J. P. Goossen, and K. A. Smith. 1999. A draft protocol for assessing Piping Plover reproductive success on Great Plains alka- li lakes. Pages 90-107 in Proceedings, Piping Plovers and Least Terns of the Great Plains and nearby. Edited by K. F. Higgins, M. R. Brashier and C. D. Kruse. South Dakota State University, Brookings, South Dakota. 133 pages. Prindiville Gaines, E. M., and M.R. Ryan. 1988. Piping Plover habitat use and reproductive success in North Dakota. Journal of Wildlife Management 52: 266-273. Sidle, J. 1985. Piping plover proposed as an endangered and threatened species. United States Federal Register 49: 44712-44715. Warnock, N., and S. Warnock. 1993. Attachment of radio-transmitters to sandpipers: review and methods. Wader Study Group Bulletin 70: 28-30. Received 28 January 2000 Accepted 9 March 2001 354 THE CANADIAN FIELD-NATURALIST Voloits Evidence of an Indirect Dispersal Pathway for Spotted Knapweed, Centaurea maculosa, Seeds, via Deer Mice, Peromyscus maniculatus, and Great Horned Owls, Bubo virginianus DEAN E. PEARSON and YVETTE K. ORTEGA USDA Forest Service, Rocky Mountain Research Station, Forestry Sciences Laboratory, PO Box 8089, Missoula, Montana 59807 USA Pearson, Dean E., and Yvette K. Ortega. 2001. Evidence of an indirect dispersal pathway for Spotted Knapweed, Centaurea maculosa, seeds via Deer Mice, Peromyscus maniculatus, and Great Horned Owls, Bubo virginianus. Canadian Field-Naturalist 115(2): 354. Spotted Knapweed (Centaurea maculosa) seeds were found in the pellets of Great Horned Owls (Bubo virginianus). That apparently resulted from owls preying upon Deer Mice (Peromyscus maniculatus) which had incidentally consumed knap- weed seeds while foraging for the larvae of biological control agents within knapweed seedheads. Successful germination of 1% of the seeds shows that knapweed seeds recovered from owl pellets can be viable after being ingested by both species and suggest that Great Horned Owls can act as indirect dispersers of Spotted Knapweed seeds. Key Words: Centaurea maculosa, Spotted Knapweed, Peromyscus maniculatus, Deer Mouse, Great Horned Owl, Bubo virginianus, exotic plants, seed dispersal. Spotted Knapweed (Centaurea maculosa) is an aggressive, exotic plant that has invaded vast areas of grassland and savanna in the western United States and Canada. Spotted Knapweed is known to be dispersed by vehicles, humans, and livestock, but little is known about knapweed dispersal by native fauna. Pearson et al. (2000) documented that Deer Mice (Peromyscus maniculatus) incidentally ingest- ed whole knapweed seeds while foraging for the lar- vae of the gall fly (Urophora spp.), biological con- trol agents introduced to suppress knapweed seed production, thereby implicating Deer Mice as poten- tial dispersal agents. However, they did not germi- nate seeds found in Deer Mouse stomachs. During April, May, and July 1999, we collected four Great Horned Owl (Bubo virginianus) pellets on the Calf Creek Wildlife Management Area about 16 km east of Hamilton, Montana (46°17 N, 114°00’ W). The four intact pellets contained remains of at least six Deer Mice, nine voles (Microtus species), one shrew (Sorex species), and 102 Spotted Knapweed seeds. One of the 102 seeds germinated successfully (1%), and the plant grew to reproductive maturity. Direct consumption of so many Spotted Knap- weed seeds by Great Horned Owls is highly improb- able as Great Horned Owls are not known to forage on seeds and no remains of seedheads were found in the pellets. Knapweed seeds found in the owl pellets likely came from the small mammals therein. Of the small mammal species identified within the pellets, only Deer Mice are known to ingest Spotted Knap- weed seeds (Pearson 1999; Pearson et al. 2000). Pearson et al. (2000) have shown that Deer Mice incidentally consume whole Spotted Knapweed seeds while foraging for the larvae of the gall flies overwintering within knapweed seedheads. We therefore conclude that Great Horned Owls may function as indirect dispersers of knapweed seeds by preying on Deer Mice that have consumed the seeds. Moreover, because Deer Mice ingest knapweed seeds only incidentally while preying upon the lar- vae of gall flies, the indirect dispersal of knapweed seeds resulting from this series of ecological interac- tions is initiated by the very biological control agent introduced to reduce the spread of Spotted Knap- weed. Acknowledgments We thank Colin Henderson, Kevin McKelvey, Cathy Zabinski, and anonymous reviewers for help- ful reviews of earlier drafts of this manuscript. We thank Mike Thompson and the Montana Department of Fish, Wildlife, and Parks for providing access to Calf Creek Wildlife Management Area for related research purposes. Literature Cited Pearson, D. E. 1999. Deer mouse predation on the biolog- | ical control agent, Urophora spp., introduced to control spotted knapweed. Northwestern Naturalist 80: 26-29. Pearson, D. E., K.S. McKelvey, and L. F. Ruggiero. 2000. Non-target effects of an introduced biological control agent on deer mouse ecology. Oecologia 122: 121-128. Received 9 March 2000 Accepted 4 July 2001 | 2001 NOTES 355 Cooperative Foraging by Steller Sea Lions, Eumetopias jubatus Scott M. GEnpeE!, JAMIE N. WoMBLE?, MARY F. WILLSON?, and BRIAN H. MARSTON* _ 1USDA Forest Service, Pacific Northwest Research Station, 2770 Sherwood Lane Suite 2A, Juneau, Alaska 99801, USA _ Institute of Marine Science, 245 O’ Neill Building, University of Alaska, Fairbanks, Alaska 99775-7220, USA _ 35230 Terrace Place, Juneau, Alaska 99801, USA _ 4Alaska Department of Fish and Game, 333 Raspberry Road, Anchorage, Alaska 99518, USA Gende, Scott M., Jamie N. Womble, Mary F. Willson, and Brian H. Marston. 2001. Cooperative foraging by Steller Sea Lions, Eumetopias jubatus. Canadian Field-Naturalist 115(2): 355-356. Steller Sea Lions were observed cooperatively foraging for Eulachon (Thaleichthyes pacificus) and possibly Herring (Clupea pallasi) in Berners Bay, southeast Alaska in spring, 1996-1999. southeast Alaska. Predators may forage singly or in groups, and for- | agers in groups may cooperate when hunting prey. _ Among the marine mammals, at least five species of whales, both toothed and baleen, are known to for- age cooperatively (Clapham 1993; Wursig and Clark _ 1993; Wells et al. 1999). However, cooperative for- _ aging in pinnipeds has been suspected but not docu- mented. Here we report several observations of 75 to 300 Steller Sea Lions (Eumetopias jubatus) foraging - cooperatively. The behavior was observed in late April or early May 1996-1999 during a study of the numerical re- sponse of vertebrate predators to the spring spawning run of Eulachon (Thaleichthys pacificus), a small anadromous smelt that spawns primarily in mainland glacial rivers from California to Alaska (Hart et al. 1944). Observers were present for a week before the Eulachon entered fresh water and for several weeks during and after the spawning run. Observations were made from a skiff in Berners Bay, a relatively small (4.5km X 3km) sheltered saltwater bay, 65 km north of Juneau, Alaska (58°45’N, 135°00’ W), that is fed by one clearwater and two glacial rivers. Around 0800h on 6 May 1996, SMG and JNW noticed a large disturbance outside the bay approxi- mately 4 km away. The disturbance at first appeared to be the wake of a large boat but within several minutes it became clear that it was a large group of Steller Sea Lions moving synchronously toward the bay. Approximately 20 minutes later, a line of rapidly porpoising Sea Lions was observed moving into the bay toward our skiff. The line of Sea Lions, comprised of about 200 to 300 individuals, was per- pendicular to the shore and stretched nearly 0.75 km, with the nearest Sea Lion approximately 0.5 km _ offshore. Most individuals were no more than a few meters apart. All individuals in the line porpoised for 8 to 20 seconds before diving simultaneously for 4 to 9 minutes, after which the water became calm with no indication that any Sea Lions were in the area. Upon simultaneous reemergence in a different _ Key Words: Steller Sea Lion, Eumetopias jubatus, Eulachon, Thaleichthyes pacificus, Herring, cooperative foraging, section of the bay, the Sea Lions again porpoised for 6 to 15s before diving simultaneously. During the third reemergence, the line reappeared in another part of the bay approximately 1 km away. The behavior continued for over an hour, with the for- aging “line” moving into and out of different sec- tions of the bay. The number of individuals partici- pating gradually decreased and after 2 hours no more Sea Lions were seen foraging in this manner. For much of the rest of the day, the Sea Lions formed large “rafts” of 10 to 80 sleeping or resting individuals in the middle of the bay (see also Orr and Poulter 1967). The “line” did not reform that after- noon and we left the area around 1700 h. When we arrived at 0900 h the following morning, the cooper- ative foraging was already in progress. The line, con- sisting of approximately 200 individuals, moved throughout the bay for over an hour before breaking apart into large rafts. The behavior was not observed again that spring. We revisited Berners Bay in the springs of 1997, 1998, and 1999 and similar behavior of cooperative foraging, followed by the formation of large rafts, was observed. On 4 and 5 May, 1997, 100 to 300 Sea Lions participated in the foraging line, and up to 100 Sea Lions were observed cooperatively forag- ing during the last week of April in 1998. In all years, cooperative foraging was observed only just before, or just after, the first Eulachon entered fresh water. As a result of the rapid movement and unpre- dictable location of the foraging line, it was impos- sible to determine with accuracy the prey species targeted by the Sea Lions while cooperatively forag- ing. However, in 1996 and 1997 we believe that Sea Lions were foraging on Eulachon for several rea- sons. First, we were able to observe the rafts at close range and some sleeping individuals had partially consumed Eulachon bodies hanging from their mouths. Second, the cooperative foraging behavior consistently occurred just as the Eulachon run was 356 beginning. Very little is known about the schooling behavior of Eulachon but anecdotal reports suggest that they overwinter in deep water and, following some environmental cue, rapidly migrate to freshwa- ter spawning areas. Thus, Sea Lions may have been targeting schools of Eulachon as they moved into Berners Bay en route to the rivers where they spawn. Herring (Clupea pallasi) may also have been the prey pursued by the Sea Lions while cooperatively foraging. Herring spawned extensively along the eastern shore of Berners Bay in 1996 and small groups of Sea Lions (3 to 5 individuals) were seen on occasion foraging for Herring near the shore. However, Herring did not spawn in Berners Bay in 1997 and 1998, suggesting that Eulachon or some other forage fish may have been the prey species pursued during those years. Although it is difficult to distinguish adult fe- males and immature males while in the water, ani- mals of both sexes and presumably all age classes were observed in the rafts following the cooperative foraging behavior. Berners Bay is located approxi- mately 18 km from Benjamin Island, where 300 to 600 individuals “haul out” from September—May, although they are absent in the summer months (Alaska Department of Fish and Game, 802 Third St., Juneau, Alaska, unpublished data). The “haul- out” is used by adults and subadults of both sexes, including nursing females (JNW, personal observa- tion). Many of the Benjamin Island Sea Lions proba- bly participated in the cooperative foraging behavior, because the “haul-out” was reported to have few individuals during our observations (E. M. Anderson, personal communication), whereas hun- dreds of Sea Lions were present there the week prior to our observations. To our knowledge this is the first recorded obser- vation of Sea Lions foraging in this manner, although cooperative foraging has been suspected. Near Unimak Pass, Alaska, large groups (up to sev- eral thousand) of Sea Lions were observed simulta- neously leaving rookeries or “haul-outs”, swimming out to sea and, when schooling fishes were targeted, feeding in groups of up to 50 individuals before simultaneously returning to land in the late afternoon (Fiscus and Baines 1966). When large schools of fish were absent, group foraging was absent or reduced to a few individuals (Fiscus and Baines 1966). Off the coast of central California, Sea Lion rafts, consisting of up to several hundred individuals, were repeatedly observed to dive synchronously (Orr and Poulter 1967). Other observations have been made of Sea Lions foraging in groups (Loughlin and Nelson 1986; references in Hoover 1988) but no studies have indicated the cooperative foraging tactic that we observed. THE CANADIAN FIELD-NATURALIST Vol. 115 Sometimes Sea Lions foraged individually for Eulachon in the rivers and probably in the ocean, and we also observed singletons foraging for Herring. However, we know very little of the specific condi- tions that favor solitary vs. cooperative foraging by Sea Lions. Cooperative foraging may lead to higher per capita energy intake or reduce the risk of poor foraging success (Packer and Ruttan 1988) when prey are aggregated, but such hypotheses remain to be tested in this system. Acknowledgments We thank the Alaska Department of Fish and Game and Brendan Kelly for access to unpublished data. Ellen Anderson, 5230 Terrace Place, Juneau, Alaska 99801, USA, provided observations from Benjamin Island. Glen VanBlaricom, Ken Pitcher, and Sue Hills provided helpful comments on the manuscript. This project was funded by the USDA Forest Service, Pacific Northwest Research Station. Literature Cited Clapham, P. J. 1993. Social organization of humpback whales on a North Atlantic feeding ground. Symposia of the Zoological Society of London 66: 131-145. Fiscus, C. H., and G. A. Baines. 1966. Food and feeding behavior of Steller and California sea lions. Journal of Mammalogy 47: 195-200. Hart, J.L., and J. L. McHugh. 1944. The smelts (Osmeridae) of British Columbia. Bulletin of the Fisheries Research Board of Canada 64: 1-27. Hoover, A.A. 1988. Steller sea lion. Pages 150—193 in Selected Marine Mammals of Alaska: species accounts with research and management recommendations. Edited by J. W. Lentfer. Marine Mammal Commission, Washington, D.C. Loughlin, T. R., and R.N. Nelson. 1986. Incidental mor- tality of northern sea lions in Shelikof Strait, Alaska. Marine Mammal Science 2: 14—33. Orr, R. T., and T. C. Poulter. 1967. Some observations on reproduction, growth, and social behavior in the Steller sea lion. Proceedings of the California Academy of Sciences 35(10): 193-226. Packer, C., and L. Ruttan. 1988. The evolution of coop- erative hunting. American Naturalist 132: 159-198. Wells, R.S., D. J. Boness, and G. B. Rathburn. 1999. Behavior. Pages 324-422 in Biology of marine mam- mals. Edited by J.E. Reynolds III and S.A. Rommel. Smithsonian Institution Press, Washington, D. C. 578 pages. Wursig, B., and C. Clark. 1993. Behavior. Pages 157—199 in The bowhead whale. Edited by J. J. Burns, J. J. Montague, and C. J. Cowles. Special Publications of the Society for Marine Mammalogy 2. Society for Marine Mammalogy, Lawrence, Kansas. Received 14 June 2000 Accepted 25 May 2001 A Tribute to Douglas Barton Osbourne Savile, 1909-2000. J. Ginns! and STEPHEN DARBYSHIRE2 '1970 Sutherland Road, Penticton, British Columbia V2A 8T8 Canada 7ECORC, Wm Saunders Bldg #49, Central Experimental Farm, Ottawa, Ontario K1A 0C6 Canada Ginns, J., and Stephen Darbyshire. 2001. A tribute to Douglas Barton Osbourne Savile, 1909-2000. Canadian Field- Naturalist 115(2): 357-364. D. B. O. Savile, honorary member of the Ottawa Field-Naturalist’s Club, died in Ottawa on 1 August, 2000. He is survived by his wife of 61 years Con- stance (Cole), a daughter Elizabeth, a son Harold, and their families. Doug was an extraordinary individual with an unusually broad range of interests. He was a botanist, mycologist, plant pathologist, ornithologist, and all- round naturalist. Born in Dublin, Ireland on 19 July 1909, his early childhood was spent in western Kenya. At the age of about six he was sent to England for schooling, which was completed at Weymouth College. He moved to Canada in 1928 to take a course in agriculture and in 1933 received a B.Sc. in Agriculture from Macdonald College, Quebec. It was at Macdonald College that Doug’s interests in biology began to emerge. He went on to finish a M. Sc. from McGill University (1934) and a Ph.D. from the University of Michigan (1939) where he studied under the mycologist Professor E. B. Mains. He joined the research division of the Canada Department of Agriculture in 1932 as a student assis- tant on the Fireblight (a bacterial disease of fruit trees) project at Abbotsford, Quebec. From 1941 to 1945 Doug served in the Aero-Engineering Branch of the Royal Canadian Air Force, where he refined his interests in optics, especially pertaining to binoculars and microscopes. His career was spent in the Plant Research Institute (PRI), latterly reorganized as the Biosystematics Research Institute, Canadian Depart- ment of Agriculture, Ottawa. At the time of retire- ment he had achieved the highest scientific ranking, that of Principal Research Scientist. He was assistant curator of the National Mycological Herbarium (known internationally by its acronym DAOM [Department of Agriculture Ottawa Mycology]) from 1943 to 1953, and curator from 1954 to 1967. Although Doug officially retired in July 1974, he hardly broke stride. Appointed as an Honorary Research Associate, he was at his desk almost daily for years afterward, solving research problems, responding to queries and writing manuscripts. During this productive time he published more than 50 papers. Doug’s botanical contributions were often inte- grated into his mycological research and can be all too easily overlooked. There are, however, several clearly botanical studies of note. Jim Calder, a botanist in PRI, and Doug began a systematic cover- age of the flora of British Columbia in 1953. The 1957 Calder, Savile and Roy Taylor collecting trip to the Queen Charlotte Islands led to the Calder and Taylor book The flora of the Queen Charlotte Islands. Doug and Jim Calder published three papers on the taxonomy of the Saxifragaceae, they cooper- ated on a note on the flora of Chesterfield Inlet, and another on the phylogeny of Carex in the light of parasitism by the smut fungi. Doug published detailed descriptions of the splash-cup dispersal mechanisms in Chrysosplenium and Mitella and was interested in the convergent evolution of splash-cup dispersal mechanisms in both plants and fungi. His northern field trips resulted in papers titled: Addi- tional plant records from Spence Bay, Boothia Isthmus; The botany of Somerset Island, District of Franklin; The botany of the northwestern Queen Elizabeth Islands; General ecology and vascular plants of the Hazen Camp area; and, Microclimate and plant growth at Isachsen and Mould Bay. And his 1972 book Arctic adaptations in plants brought together his careful observations from the field trips to the Canadian arctic in 1950, 1958, 1959, 1960, and 1962. It has been termed a classic in arctic biolo- gy. Other field trips took Doug to Newfoundland, Nova Scotia, New Brunswick, Quebec, Ontario, Manitoba, Alberta, and British Columbia. In 1962 Doug published a handbook intended for amateur and professional botanists (including mycol- ogists) titled Collection and Care of Botanical Speci- mens. Doug’s thoroughness and attention to detail made this booklet an extremely useful aid for collec- tors, as well as curators. In its time the publication was one of the few places where consolidated infor- mation could be found on all aspects of the collec- tion and care of botanical specimens. This little book proved so useful and popular that it was revised, translated into French and re-published eleven years later. Mycologically, Doug worked primarily with the groups of parasitic fungi known as rusts and smuts. His research covered taxonomy, ecology, phylogeny, co-evolution of host plants and their parasites, use of S57 D. B. O. Savile (right) with R to L: J. Walton Groves, Ruth Macrae, Ibra Conners, Mildred Nobles and Clara Fritz, on a THE CANADIAN FIELD-NATURALIST ks Vol. ~ oa A “i : ES collecting trip to Gatineau Park, Quebec, on 12 August 1943. parasites to decipher host plant relationships, biogeo- graphic history of Canadian plants, and he was involved in developing the use of rust relationships as a guide to taxonomic relationships and compara- tive chronology of the various groups of grasses. From 1943 through 1953 Doug co-authored with I. L. Conners the annual reports of the Canadian Plant Disease Survey. Doug Savile was a consummate naturalist writ- ing papers on many subjects, such as meteorologi- cal phenomena, flight capabilities of Archeopteryx, flight mechanisms of swifts and hummingbirds, and the function and convergence of biogeography. Doug was an avid bird-watcher and student of bird biology. He published several papers on the occur- Bibliography of D. B. O. Savile Savile, D. B.O. 1936. The reflecting prism in micro- scopy. Stain Technology 11: 122. Savile, D. B. O. 1936. Simple aids to microscope illumina- tion. Stain Technology 11: 161-164. Savile, D. B. O., and H. N. Racicot. 1937. Bacterial wilt and rot of potatoes. Scientific Agriculture 17: 518-522. Racicot, H. N., and D. B. O. Savile. 1938. Bacterial wilt and rot of potatoes. Phytopathology 28: 18. Racicot, H.N., D. B. O. Savile, and I. L. Conners. 1938. Bacterial wilt and rot of potatoes - some suggestions for its detection, verification, and control. The American Potato Journal 15: 312-318. rence and behavior of birds in various parts of Canada. In 1980 The Ottawa Field-Naturalists’ Club awarded him an honorary membership, but Doug’s numerous and significant research contributions to botany and mycology both at the Canadian and inter- national level have also been acknowledged by other scientific and academic organizations . He was elect- ed a Fellow of the Royal Society of Canada (1966); awarded the prestigious George Lawson Medal by the Canadian Botanical Association (1976); received an honorary Doctor of Science degree from McGill University (1978); and elected a Distinguished Mycologist by The Mycological Society of America (1988). Savile, D. B. O. 1939. Nuclear structure and behavior in species of the Uredinales. American Journal of Botany 26: 585-609. Savile, D. B.O. 1941. An unusual halo display. Monthly Weather Review 69: 73-74. Savile, D. B. O. 1942. Alteration of potato starch grain structure under the influence of disease. American Jour- nal of Botany 29: 286-287. Conners, I. L., and D. B. O. Savile. 1943. Twenty-Second Annual Report of the Canadian Plant Disease Survey 1942. Savile, D. B. O. 1944. Ice-crystal haloes. Nature 153: 25. Conners, T.L., and D. B. O. Savile. 1944. Twenty-third 2001 GINNS AND DARBYSHIRE: Da aa Tus, ee AR si ad 2 ahs George P. White. Annual Report of the Canadian Plant disease Survey 1943. Conners, I. L., and D. B. O. Savile. 1945. Twenty-Fourth Annual Report of the Canadian Plant Disease Survey 1944. Savile, D. B.O. 1946. Entyloma fuscum and related smuts attacking Papaveraceae. Canadian Journal of Research C, 24: 109-114. Savile, D. B.O. 1946. A new species of Stagonospora on Ambrosia. Mycologia 38: 453-454. Savile, D. B. O. 1946. A rapid freehand sectioning method for leaves. Stain Technology 21: 99-102. TRIBUTE TO D. B. O. SAVILE a09 D. B. O. Savile, in 1979 or 1980 at work in the William Saunders Building, Central Experimental Farm, Ottawa. Photo by Conners, I. L., and D. B. O. Savile. 1946. Twenty-Fifth Annual Report of the Canadian Plant Disease Survey 1945. Conners, I. L., and D. B. O. Savile. 1947. Twenty-Sixth Annual Report of the Canadian Plant Disease Survey 1946. Savile, D. B. O. 1947. A study of the species of Entyloma on North American composites. Canadian Journal of Re- search C, 25: 105-120. Savile, D. B. O. 1948. Bird navigation in homing and in migration. Science 107: 596-597. Conners, I. L., and D. B. O. Savile. 1948. Twenty-Seventh 360 Annual Report of the Canadian Plant Disease Survey 1947. Conners, I. L., and D. B. O. Savile. 1949. Twenty-Eighth Annual Report of the Canadian Plant Disease Survey 1948. Savile, D. B. O. 1950. The flight mechanism of swifts and hummingbirds. The Auk 67: 499-504. (Also reprinted in Panorama of Science, 1951, Annual Supplement to Smithsonian Series for 1951. Series Publishers Inc., New York, 1951). Savile, D. B. O. 1950. North American species of Chryso- myxa. Canadian Journal of Research C, 28: 318-330. Savile, D. B. O. 1950. Bird notes from Great Whale River, Que. Canadian Field-Naturalist 64: 95-99. Savile, D. B.O. 1950. A new rust on Deschampsia. Myco- logia 42: 663-667. Savile, D. B.O. 1950. Bobolink at Cochrane, Ontario. Canadian Field-Naturalist 64: 154. Savile, D. B.O. 1950. Unusual behaviour of Wilson’s Snipe. Canadian Field-Naturalist 64: 155. Conners, I. L., and D. B. O. Savile. 1950. Twenty-Ninth Annual Report of the Canadian Plant Disease Survey 1949. Savile, D. B. O. 1951. Changes in grassland near Ottawa, Ontario, following prolonged flooding. Canadian Field- Naturalist 65: 42-45. Savile, D. B.O. 1951. The ring-billed gull at Ottawa, Ontario, and its field recognition. Canadian Field- Naturalist 65: 109-112. Savile, D. B.O. 1951. Peronospora stigmaticola in Can- ada. Mycologia 43: 114. Savile, D. B.O. 1951. The relationship of Puccinia prae- gracilis and P. connersii. Mycologia 43: 456-458. Savile, D. B. O., and I. L. Conners. 1951. The rusts of Armeria and Limonium in North America. Mycologia 43: 186-195. Savile, D. B.O. 1951. Two new smuts on Carex in Can- ada. Canadian Journal of Botany 29: 324-328. Savile, D. B. O. 1951. Bird observations at Chesterfield Inlet, Keewatin, in 1950. Canadian Field-Naturalist 65: 145-157. Savile, D. B. O. 1952. A study of the species of Cintractia on Carex, Kobresia and Scirpus in North America. Canadian Journal of Botany 30: 410-435. Savile, D. B.O. 1952. Common names for subspecies. Science 116: 513-514. Savile, D. B. O., and J. A. Calder. 1952. Notes on the flora of Chesterfield Inlet, Keewatin District, N.W.T. Canadian Field-Naturalist 66: 103—107. Conners, I. L., and D. B. O. Savile. 1952. Thirtieth An- nual Report of the Canadian Plant Disease Survey 1950. Conners, I. L., and D. B. O. Savile. 1952. Thirty-First Annual Report of the Canadian Plant Disease Survey [95K Savile, D. B.O. 1953. Taxonomy of the parasitic fungi as an aid to taxonomy of the flowering plants. Estratto dagli Atti Del Vi Congresso Internazionale Di Microbiolagia Roma 5, 14: 217. Conners, I. L., and D. B. O. Savile. 1953. Thirty-Second Annual Report of the Canadian Plant Disease Survey 1952. Savile, D. B. O., and J. A. Parmelee. 1953. Notes and brief articles, systemic infection in Cintractia junca. Mycologia 45: 788-790. Savile, D. B.O. 1953. Notes on boreal Ustilaginales. Canadian Journal of Botany 31: 663-674. THE CANADIAN FIELD-NATURALIST Vol. 115 Savile, D. B.O. 1953. Short-season adaptions in the rust fungi. Mycologia 45: 75-87. Savile, D. B.O. 1953. Splash-cup dispersal mechanism in Chrysosplenium and Mitella. Science 117: 250-251. Savile, D. B. O., and J. A. Calder. 1953. Phylogeny of Carex in the light of parasitism by the smut fungi. Cana- dian Journal of Botany 31: 164-174. Savile, D. B.O. 1954. Taxonomy, phylogeny, host rela- tionship, and phytogeography of the microcyclic rusts of Saxifragaceae. Canadian Journal of Botany 32: 400-425. Savile, D. B. O. 1954. The fungi as aids in the taxonomy of the flowering plants. Science 120: 583-585. Parmelee, J. A., and D. B. O. Savile. 1954. Life history and relationship of the rusts of Sparganium and Acorus. Mycologia 46: 823-836. Savile, D. B. O. 1954. Cellular mechanics, taxonomy and evolution in the Uredinales and Ustilaginales. Mycologia 46: 736-761. Savile, D. B. O. 1955. Chrysomyxa in North America- additions and corrections. Canadian Journal of Botany 33: 487-496. Savile, D. B.O. 1955. A phylogeny of the Basidiomy- cetes. Canadian Journal of Botany 33: 60-104. Savile, D. B. O., and J. A. Parmelee. 1956. Some fungal parasites of Portulacaceae. Mycologia 48: 573-590. Savile, D. B. O. 1956. Known dispersal rates and migrato- ry potentials as clues to the origin of the North American biota. American Midland Naturalist 56: 434-453. Savile, D. B.O. 1957. The primaries of Archaeopteryx. The Auk 74: 99-101. Savile, D. B.O. 1957. Additions to the parasitic fungi of Nova Scotia. Canadian Journal of Botany 35: 197-206. Savile, D. B. O. 1957. Notes on boreal Ustilaginales, II. Canadian Journal of Botany 35: 279-286. Savile, D. B. O. 1957. Some recent Ottawa bird records. Canadian Field-Naturalist 71: 32-33. Savile, D. B.O. 1957. Adaptive evolution in the avian wing. Evolution 11: 212-224. Savile, D. B.O. 1959. Additional plant records from Spence Bay, Boothia Isthmus. Canadian Field-Naturalist 73: 168-169. Savile, D. B. O. 1959. Notes and brief articles. Two little known Ascomycetes attacking Filicales. Mycologia 51: 296-298. Savile, D. B.O. 1959. Notes on Exobasidium. Canadian Journal of Botany 37: 641-656. Savile, D. B.O. 1959. The botany of Somerset Island, District of Franklin. Canadian Journal of Botany 37: 959-1002. Savile, D. B. O. 1959. Limited penetration of barriers as a factor in evolution. Evolution 13: 333-343. Calder, J. A., and D. B. O. Savile. 1959. Studies in the Saxifragaceae. I. The Heuchera cylindrica complex in and adjacent to British Columbia. Brittonia 11: 49-67. Calder, J. A., and D. B. O. Savile. 1959. Studies in Saxi- fragaceae. II. Saxifraga sect. Trachyphyllum in North America. Brittonia 11: 228-249. Savile, D. B. O. 1960. Limitations of the competitive ex- clusion principle. Science 132: 1761. Savile, D. B. O. 1960. The evolutionary significance of barrier penetration. Jn Evolution: its science and doctrine. Edited by T. W. Cameron. Royal Society of Canada. University of Toronto Press. pp. 110-114. Calder, J. A., and D. B. O. Savile. 1960. Studies in Saxi- 2001 fragaceae. III. Saxifraga odontoloma and lyallii and North American subspecies of S. punctata. Canadian Journal of Botany 38: 409-435. Savile, D. B.O. 1961. Some fungi from Spence Bay, Boothia Isthmus. Canadian Field-Naturalist 75: 69-71. Savile, D. B.O. 1961. Some fungal parasites of Liliaceae. Mycologia 53: 31-52. Savile, D. B. O. 1961. Evolution on plant biosystematics in the north Pacific region. Symposium on plant biosys- tematics in the Pacific basin. Tenth Pacific Science Con- gress. p. 135. Savile, D. B. O. 1961. Evolution of Saxifragaceae in the north Pacific region. Symposium on Plant Biosyste- matics in the Pacific basin. Tenth Pacific Science Congress. p. 135. Savile, D. B.O. 1961. The botany of the northwestern Queen Elizabeth Islands. Canadian Journal of Botany 39: 909-942. Savile, D. B.O. 1961. Bird and mammal observations on Ellef Ringnes Island in 1960. Natural History Papers, National Museum of Canada. 9: 1-6. Savile, D. B. O. 1961. Evolution of Saxifragaceae from a mycologist’s viewpoint. Recent Advances in Botany. From Lectures & Symposia presented to the IX Inter- national Botanical Congress. Montreal 1959. University of Toronto Press. 1: 169-172. Savile, D. B.O. 1961. Swainson’s Thrush on Meighan Island, Franklin District. Canadian Field-Naturalist 75: 256. Savile, D. B.O. 1962. Taxonomic disposition of Allium. Nature 196: 792. Savile, D. B.O. 1962. The collection and care of botani- cal specimens. Canadian Department of Agriculture Publication 1113. 124 pp. Savile, D. B. O. 1962. Tilletia controversa. Mycologia 54: 109-110. Savile, D. B. O. 1962. Some fungal parasites of Onagra- ceae. Canadian Journal of Botany 40: 1385-1398. Savile, D. B.O. 1962. Gliding and flight in the verte- brates. American Zoologist 2: 161—166. Savile, D. B.O. 1962. Septoria streptopodis and Cerco- spora steptopi. Mycologia 54: 321-322. Savile, D. B.O. 1962. Mycology in the Canadian arctic. Arctic 16: 17-25. Savile, D. B. O. 1963. A rigid swinging-arm stand for the dissecting microscope. Mycologia 55: 678-681. Savile, D. B.O. 1963. Factors limiting the advance of spruce at Great Whale River, Quebec. Canadian Field- Naturalist 77: 95-97. Savile, D. B. O., and J. A. Parmelee. 1964. Parasitic fungi of the Queen Elizabeth Islands. Canadian Journal of Botany 43: 699-722. Savile, D. B.O. 1964. Geographic variation and gene flow in Puccinia cruciferarum. Mycologia 56: 240-248. Savile, D. B. O., and D. R. Oliver. 1964. Bird and mam- mal observations at Hazen Camp, northern Ellesmere Island, in 1962. Canadian Field-Naturalist 78: 1-7. Savile, D. B. O. 1964. General ecology and vascular plants of the Hazen Camp Area. Arctic 17: 237-256. Savile, D. B. O. 1964. Puccinia podophylli by light mic- roscopy. Mycologia 56: 452-453. Savile, D. B. O. 1965. Puccinia karelica and species delim- itation in the Uredinales. Canadian Journal of Botany 43: 231-238. Savile, D. B.O. 1965. Spore discharge in basidiomycetes: a unified theory. Science 147: 165-166. GINNS AND DARBYSHIRE: TRIBUTE TO D. B. O. SAVILE 361 Savile, D. B.O. 1965. Some fungal parasites of Umbelli- ferae. Canadian Journal of Botany 43: 571-596. Savile, D. B. O. 1965. Comment in: Mason, W.R.M. Eco- logical Peculiarities of the Canadian North. The Arctic Circular 16: 18. Savile, D. B.O. 1965. A new rust on Erigeron. Myco- logia 57: 476-477. Savile, D. B. O. 1966. The rusts of Eriogonum, Chori- zanthe and Oxytheca. Canadian Journal of Botany 44: 1151-1170. Savile, D. B. O. 1966. Unity from diversity in biological research. Transactions of the Royal Society Canada Series 4, 4: 245-251. Savile, D. B. O. 1967. Evolution and relationships of North American Pedicularis rusts and their hosts. Canadian | Journal of Botany 45: 1093-1103. Savile, D. B.O. 1967. The sterigmata and basidiospores of Cronartium ribicola. Canadian Journal of Botany 45: 1454-1456. Savile, D. B. O. 1968. Species of Puccinia attacking Ver- oniceae (Scrophulariaceae). Canadian Journal of Botany 46: 631-642. Savile, D. B.O. 1968. Some fungal parasites of Scro- phulariaceae. Canadian Journal of Botany 46: 461-471. Savile, D. B.O. 1968. Parasitic relationships and disposi- tion of Filipendula. Brittonia 20: 230-231. Savile, D. B. O. 1968. The case against “Uredium”. My- cologia 60: 459-464. Savile, D. B. O. 1968. Botanical studies in arctic Canada. Greenhouse Garden-Grass 7(2): 1-3. Savile, D. B.O. 1968. The rusts of Cheloneae (Scro- phulariaceae): a study in the co-evolution of hosts and parasites. Nova Hedwigia 15: 369-392. Savile, D. B.O. 1968. Foreword. Jn J. A. Calder and R. L. Taylor, Flora of the Queen Charlotte Islands, Part 1. Queen’s Printer, Ottawa. Pages v—vil. Savile, D. B. O. 1968. The land plants of Hudson Bay. In Hudson Bay, land plants. Science, history and Hudson Bay. Vol. 1. Edited by J.S. Beals. Energy Mines and Resources, Ottawa. Pages 347-416. Savile, D. B. O. 1968. Possible inter-relationships among fungal groups. Jn The Fungi. Vol. 3. Edited by G. P. Ainsworth and A. S. Sussmann. Pages 649-675. Savile, D. B. O. 1969. Interrelationships of Ledum species and their rust parasites in Western Canada and Alaska. Canadian Journal of Botany 47: 1085-1100. Savile, D. B. O. 1969. Chives rust at Ottawa, Ontario. Canadian Plant Disease Survey 49(1): 29. Savile, D. B. O. 1969. Biology in the north. Science Af- fairs 3(4): 69-71. Savile, D. B.O. 1969. The meaning of “pleomorphism”. Mycologia 61: 1161-1162. Savile, D. B.O. 1970. Some Eurasian Puccinia species attacking Cardueae. Canadian Journal of Botany 48: 1553-1566. Savile, D. B. O. 1970. James Walton Groves 1906-1970. Proceedings of the Royal Society of Canada Series IV Vol. VIII. Savile, D. B. O. 1970. Autoecious Puccinia species attack- ing Cardueae in North America. Canadian Journal of Botany 48: 1567-1584. Savile, D. B. O. 1971. Microclimate and plant growth at Isachsen and Mould Bay. Arctic 24: 306-307. Savile, D. B. O. 1971. Methods and aims in the study of the rust fungi. Journal of Indian Botanical Society 50A: 41-51. 362 Savile, D. B. O. 1971. Co-ordinated studies of parasitic fungi and flowering plants. Le Naturaliste canadien 98: 535-552. Savile, D. B. O. 1971. Coevolution of the rust fungi and their hosts. The Quarterly Review of Biology 46: 211-218. Savile, D. B.O. 1971. Generic disposition and pycnium type in Uredinales. Mycologia 63: 1089-1091. Savile, D. B. O. 1971. Frank Lisle Drayton 1892-1970. Proceedings of the Royal Society of Canada Series IV Vol. 9: 49-50. Savile, D. B.O. 1972. Evidence of tree nestings by the marbled murrelet in the Queen Charlotte Islands. Can- adian Field-Naturalist 86: 389-390. Savile, D. B. O. 1972. Spring observations of Sabine’s gull. Canadian Field-Naturalist 86: 389. Savile, D. B. O. 1972. Some rusts of Scirpus and allied genera. Canadian Journal of Botany 50: 2579-2596. Savile, D. B. O. 1972. Arctic adaptations in plants. Can- adian Department of Agriculture Monograph No. 6. 81 Pp. Savile, D. B. O. 1973. A variety of Puccinia centaureae on Centaurea diffusa. Canadian Journal of Botany 51: 1077-1078. Savile, D. B.O. 1973. Vegetative distinctions in Canadian — species of Mitella and Tiarella. Canadian Field- Naturalist 87: 460-462. Savile, D. B. O. 1973. Aeciospore types in Puccinia and Uromyces attacking Cyperaceae, Juncaceae and Poa- ceae. Report of the Tottori Mycological Institute 10: 225-241. Savile, D. B. O. 1973. Rusts that pass import inspection. Canadian Plant Disease Survey 53: 105-106. Savile, D. B. O. 1973. Revisions of the microcyclic Puccinia species on Saxifragaceae. Canadian Journal of Botany 51: 2347-2370. Savile, D. B. O. 1973. Collection and care of botanical specimens. Reprinted with addenda. Canada Department of Agriculture Publication 1113. Ottawa. 128 pp. Savile, D. B. O. 1974. Dedication to Ibra L. Conners. Fungi Canadenses. July 1974. 2 pp. Savile, D. B.O. 1974. Puccinia urbanis. Fungi Canaden- ses No. 18. Savile, D. B.O. 1974. Puccinia parnassiae. Fungi Cana- denses No. 19. Savile, D. B. O., and J. A. Parmelee. 1974. Uromyces punctatus. Fungi Canadenses No. 24. Savile, D. B. O., and J. A. Parmelee. 1974. Uromyces phacae-frigidae. Fungi Canadenses No. 25. Savile, D. B. O., and J. A. Parmelee. 1974. Uromyces lapponicus. Fungi Canadenses No. 26 a,b. Savile, D. B. O. 1974. Phragmidium potentillae. Fungi Canadenses No. 41. Savile, D. B. O. 1974. Puccinia codyi. Fungi Canadenses No. 46. Savile, D. B. O. 1974. Puccinia holboellii. Fungi Cana- denses No. 47. Savile, D. B. O. 1974. Phragmidium fusiforme var. novi- boreale. Fungi Canadenses No. 54. Savile, D. B. O. 1974. Puccinia canadensis. Fungi Cana- denses No. 56. Savile, D. B. O. 1974. Orphanomyces, a new genus of cypericolous smuts. Canadian Journal of Botany 52: 341-343. Savile, D. B. O. 1974. Vegetation and animal life [Arctic THE CANADIAN FIELD-NATURALIST Volvils Islands.] In Encyclopaedia Brittanica, 15th Edition, Helen Hemingway Benton, U.S.A. pp. 1117-1118. Savile, D. B. O. 1974. Some new or poorly known rusts of Brassicaceae. Canadian Journal of Botany 52: 1501- 1507. Savile, D. B. O. 1975. Mundkurella mossii, a smut of Aralia nudicaulis. Mycologia 67: 273-279. Savile, D. B. O. 1975. Evolution and biogeography of Saxifragaceae with guidance from their rust parasites. Annals of the Missouri Botanical Garden 62: 354-361. Savile, D. B. O. 1975. Puccinia fergussonii. Fungi Cana- denses No. 64. Savile, D. B.O. 1975. Puccinia violae ssp. americana. Fungi Canadenses No. 75. Savile, D. B. O. 1975. Puccinia ornatula. Fungi Canaden- ses No. 77. Savile, D. B. O. 1975. Puccinia glacieri. Fungi Canaden- ses No. 78. Savile, D. B. O. 1975. Phragmidium arcticum. Fungi Can- adenses No. 79. Savile, D. B.O. 1975. Phragmidium occidentale. Fungi Canadenses No. 80. Kevan, P.G., Chaloner, W.G., and D. B. O. Savile. 1975. Interrelationships of early terrestrial arthropods and plants. Palaeontology 18: 391-417. Savile, D. B. O. and C.E. Savile. 1976. Sight record of Laughing Gulls (Larus atricilla) in Saskatchewan. Can- adian Field-Naturalist 90: 187. Savile, D. B. O. 1976. Twig abscission in maples (Section Rubra: Acer rubrum and A. saccharinum) as a defence reaction against water stress. Canadian Field-Naturalist 90: 184-185. Savile, D. B. O. 1976. Notes on some parasitic fungi from southern British Columbia, southwest Alberta, and adja- cent United States. Canadian Journal of Botany 54: 971-975. Savile, D. B. O. 1976. Phragmidium ivesiae and its allies in North America. Canadian Journal of Botany 54: 1690-1696. , Savile, D. B. O. 1976. Coniomitella (Saxifragaceae) and its rust. Canadian Journal of Botany 54: 1977-1978. Savile, D. B. O. 1976. Evolution of the rust fungi (Uredi- nales) as reflected by their ecologocal problems. Evolu- tionary Biology 9: 137-207. Savile, D. B. O., and L.K. Weresub. 1977. Mary Eliz- abeth Elliott (1923-1976). Mycologia 69: 460-462. Savile, D. B.O. 1977. Puccinia conglomerata. Fungi Canadenses No. 110. Savile, D. B. O. 1977 [1978] Leaf form and evolutionary patterns in Pedicularis. Proceedings of the Indian National Science Academy B. 43: 223-227. Savile, D. B. O. 1978. Paleoecology and convergent evo- lution in rust fungi (Uredinales). Bio-Systems 10: 31-36. Savile, D. B. O. 1978. Rust fungi in the British Columbia flora. Davidsonia 9: 1-5. Savile, D. B. O., and H.N. Hayhoe. 1978. The potential effect of drop size on efficiency of splash-cup and springboard dispersal devices. Canadian Journal of Botany 56: 127-128. Savile, D. B. O. 1979. The evolution of anamorphs in the Uredinales fungi, structure. Jn Whole Fungus Sex-Asex- Syn. Ottawa, National Museum of Natural Sciences and National Museums of Canada. 2: 547-554. Savile, D. B.O. 1979. Fungi as aids in higher plant classi- fication. Botanical Review 45: 377-503. 2001 Savile, D. B. O. 1979. Ring counts in Salix arctica from northern Ellesmere Island. Canadian Field-Naturalist 93: 81-82. Savile, D. B. O. 1979. Dispersal by falling water drops in Saxifragaceae. Davidsonia 10: 65-69. Savile, D. B. O. 1979. Fungi as aids to plant taxonomy: methodology and principles. Symbolae Botanicae Upsalienses 22(4): 135-145. Savile, D. B.O. 1979. Spring beauty associations. Trail & Landscape 13: 64-65: Savile, D. B. O. 1980. Ecology, convergent evolution, and classification in Uredinales fungi. Report of the Tottori Mycological Institute 18: 275-281. Savile, D. B. O. 1980. A naturalist’s approach to biology. Canadian Field-Naturalist 94: 105-109. Savile, D. B. O. 1981. A naturalist looks at arctic adapta- tions. Jn Evolution Today. Edited by G.G.E. Scudder and J. L. Reveal. Proceedings of the Second International Congress of Systematic and Evolutionary Biology. pp. 47-53. . Savile, D. B.O. 1981. Cueillette et montage de spécimens botaniques. Agriculture Canada Publication 1113F. Ottawa. 151 pp. Savile, D. B. O. 1981. Seed dispersal by falling water drops. Michigan Botanist 20: 43. Savile, D. B. O. 1981. The supposed rust of Echinochloa. Mycologia 73: 1007-1008. Parmelee, J. A., and D. B. O. Savile. 1981. Autoecious species of Puccinia on Cichorieae in North America. Canadian Journal of Botany 59: 1078-1101. Savile, D. B. O. 1982. Adaptations of fungi to arctic and subarctic conditions. Jn Arctic and Alpine Mycology. Edited by G. A. Laursen and J. F. Ammirati. University of Washington Press, Seattle. pp. 357-370. Savile, D. B. O., and Z. Urban. 1982. Evolution and ecolo- gy of Puccinia graminis. Preslia 54: 97-104. Savile, D. B. O. 1983. Puccinia striiformis. Fungi Cana- denses No. 250. Savile, D. B. O. 1984. Communication problems in inter- disciplinary research. Proceedings of the Indian Acad- emy of Science. Plant Sciences 93: 223-230. Savile, D. B.O. 1984. Randomness and change of func- tion in the evolution of structures. Systematic Botany 9: 494495. Savile, D. B. O. 1984. Taxonomy of the cereal rust fungi. 1: 79-112. In The Cereal Rusts. Edited by W. R. Bushnell and A. P. Roelfs. Academic Press, New York. 546 pp. Savile, D. B.O. 1985. Biology. In Canadian Encyclo- pedia. Hurtig, Edmonton. 1: 178-180. Baum, B.R., and D. B. O. Savile. 1985. Rusts (Uredi- nales) of Triticeae: evolution and extent of coevolution, a cladistic analysis. Botanical Journal of the Linnean Society 91: 367-394. Cody, W. J., Savile, D. B. O., and M. J. Sarazin. 1986. Systematics in Agriculture Canada at Ottawa 1886- 1986. Agriculture Canada Historical Series No. 28. 81 pp. [Version francaise: La recherche en systématique 4 Agriculture Canada, Ottawa 1886-1986. 83 pp.] Savile, D. B. O. 1987. Use of rust fungi (Uredinales) in determining ages and relationships in Poaceae. In Grass Systematics and Evolution. Edited by T.R. Soderstrom, K. W. Hilu, C.S. Campbell and M.E. Barkworth. Smithsonian Inst. Press, Washington. pp. 168-178. Savile, D. B.O. 1987. Biologie. In L’Encyclopédie du Canada. Stanké, Montréal. 1: 228-229. GINNS AND DARBYSHIRE: TRIBUTE TO D. B. O. SAVILE 363 Savile, D. B.O. 1987. Uromyces polygoni-avicularis. Fungi Canadenses No. 314. Savile, D. B.O. 1988. Terminology of spore states in Uredinales. Mycotaxon 33: 387-389. Savile, D.B.O. 1989. Harold Johnston Brodie, 1907-1989. Mycologia 81: 832-836. Savile, D. B. O. 1989. Raveneliaceae revisited. Canadian Journal of Botany 67: 2983-2994. Savile, D. B.O. 1989. A Tribute to Ibra L. Conners, 1894-1989. Canadian Field-Naturalist 103: 610-612. Savile, D. B.O. 1990. Relationships of Poaceae, Cyper- aceae, and Juncaceae reflected by their fungal parasites. Canadian Journal of Botany 68: 731-734. Savile, D. B. O. 1990. Coevolution of Uredinales and Ustilaginales with vascular plants. Report of the Tottori Mycological Institute 28: 15-24. Savile, D. B. O. 1993. Cladistic analysis of the rust fungi: a reappraisal. Nova Hedwigia 57: 269-277. Savile, D. B. O. 1993. Letter to the editor: Book Reviews Revisited. Inoculum 44: 12. Savile, D. B. O. 2001. Evolution of a naturalist. Canadian Field-Naturalist 115(2): 365-380. Reviews Savile, D. B. O. 1944. Review: Driver, E. C. 1942. Name that animal, a guide to the identification of the common land and fresh-water animals of the United States, with special reference to the area east of the Rockies. Smith College, Northampton, Mass. 527 pp. Canadian Field- Naturalist 58: 70. Savile, D. B. O. 1949. Review: Storer, J. H. 1948. The Flight of Birds. Analyzed through slow-motion photog- raphy. Cranbrook Institute of Science, Bloomfield Hills, Michigan. 94 pp. Canadian Field-Naturalist 63: 93-94. Savile, D. B. O. 1949. Review: Palmer, E. L. 1949. Field- book of Natural History. Whittlesey House, McGraw- Hill Book Co., Inc. New York and Toronto. 664 pp. Canadian Field-Naturalist 63: 166. Savile, D. B. O. 1951. Review: McCowan, D. 1951. Tide- water to Timberland. Macmillan Company of Canada Ltd. Toronto. 205 pp. Canadian Field-Naturalist 65: 126. Savile, D. B. O. 1951. Review: Bailey, A. M. 1951. Nature photography with miniture cameras. Museum Pictorial. No. 1. Denver Museum of Natural History. Denver, Colorado. 64 pp. Canadian Field-Naturalist 65: 188. Savile, D. B. O. 1954. Review: Gray, J. 1953. How Ani- mals Move. Cambridge University Press and Macmillan Co. of Canada. xii + 114 pp. Canadian Field-Naturalist 68: 142. Savile, D. B. O. 1955. Review: Dodd, E. (editor) 1955. Mark Trail’s Book of North American Mammals. McClelland and Stewart Ltd. Toronto. 242 pp. Canadian Field-Naturalist 69: 133. Savile, D. B. O. 1956. Review: Annual Report for 1955 of the Denver Museum of Natural History. Denver Colorado. 70 pp. Canadian Field-Naturalist 70: 186. Savile, D. B. O. 1956. Review: Hochbaum, A. 1956. Travels and Traditions of Waterfowl. The University of Minnesota Press, Minneapolis and Thomas Allen Ltd. Toronto. 301 pp. Canadian Field-Naturalist 70: 146-147. Savile, D. B. O. 1957. Review: Ford, A. (ed.) 1957. The Bird Biographies of John James Audubon. The Macmillan Co., New York and Brett-Macmillan Ltd., Toronto. 282 pp. Canadian Field-Naturalist 71: 160-161. Savile, D. B. O. 1957. Review: Mathews, G. V. T. 1955. 364 Bird Navigation. Cambridge University Press, Cam- bridge and Macmillan Co. of Canada, Toronto. 141 pp. (Cambridge Monographs in Experimental Biology, No. 3). Canadian Field-Naturalist 71: 202. Savile, D. B. O. 1957. Review: Snyder, L. L. 1957. Arctic Birds of Canada. Illus. By T. M. Shortt. University of Toronto Press, Toronto. 310 pp. Canadian Field-Natur- alist 71: 203. Savile, D. B. O. 1958. Review: Bailey, A. M. 1956. Birds of Midway and Laysan Islands. Denver Museum of Natural History, Denver, Colorado. 130 pp. Canadian Field- Naturalist 72: 183. Savile, D. B. O. 1958. Review: Elton, C. S. 1958. The Ecology of Invasions by Animals and Plants. Methuen, London. 181 pp. Canadian Field-Naturalist 72: 179-180. Savile, D. B.O. 1960. Review: Moore, W. C. 1959. Brit- ish Parasitic Fungi. Cambridge University Press, New York. xvi + 430 pp. Economic Botany 14: 338. Savile, D. B. O. 1961. Review: Mayfield, H. 1960. The Kirtland’s Warbler. Cranbrook Institute of Science. Bloomfield Hills, Michigan. 242 pp. Canadian Field- Naturalist 75: 46-47. Savile, D. B. O. 1961. Review: Pettingill, E. R. 1960. Penguin Summer. Potter, New York. 197 pp. Canadian Field-Naturalist 75: 104-105. Savile, D. B. O. 1961. Review: Greenewalt, C. H..1960. Hummingbirds. Doubleday, New York. 250 pp. Canadian Field-Naturalist 75: 105-106. Savile, D. B. O. 1961. Review: Tuck, L. M. 1960. The Murres. Department of Northern Affairs and National Resources, Canadian Wildlife Service, Ottawa. 260 pp. Canadian Field-Naturalist 75: 256. Savile, D. B. O. 1962. Review: Wiggens I. L. and J. H. Thomas. 1962. A Flora of the Alaskan Arctic Slope. Arctic Institute of North American Special Publication. No. 4, University of Toronto Press, Toronto. vii + 425 pp. Canadian Field-Naturalist 76: 70. Savile, D. B. O. 1962. Review: H. R. H. The Duke of Edin- burgh. 1962. Birds from Britannia. Longman’s Canada - Ltd., Toronto. 62 pp. Canadian Geographical Journal 65: 1X, XE Savile, D. B. O. 1964. Review: Arthur, J. C. 1962. Manual of the Rusts in United States and Canada. Supplement by George B. Cummins, Hafner Publishing Company, New York. 22 + 438 pp. Economic Botany 18: 286. Savile, D. B. O. 1964. Review: A and D. Love, eds. 1962. North Atlantic Biota and their History. A symposium held THE CANADIAN FIELD-NATURALIST Voliits at the University of Iceland, Reykjavik, July 1962 under the auspices of the University of Iceland and the Museum of Natural History. Sponsored by the NATO Advanced Institute Program. Macmillan, New York. 430 pp. Arctic 17: 138-141. Savile, D. B.O. 1965. Review: Gleason H.A. and A. Cronquist. 1964. The Natural Geography of Plants. Columbia University Press, New York, and Copp Clark Publishing Co., Toronto. 420 pp. Canadian Field-Natur- alist 79: 70-71. Savile, D. B. O. 1966. Review: Taylor, R. L. 1965. The Genus Lithophragma (Saxifragaceae) University of California Publications in Botany 37: 1-122, University of California Press, Berkeley and Los Angeles. Bulletin of the Torrey Botanical Club 93: 202-203. Savile, D. B. O. 1967. Review: Conners, I. L. 1967. An annotated index of plant diseases in Canada and fungi recorded on plants in Alaska, Canada and Greenland. Canada Department of Agriculture Publications, Queen’s Printer, Ottawa, Canada. 381 pp. Canadian Plant Disease Survey 47(2): 31. Savile, D. B. O. 1967. Review: Gilkey, H. M. and L.-R. J. Dennis. 1967. Handbook of Northwestern Plants. Oregon State University Bookstores Incorporated. Corvallis, Oregon. 505 pp. Canadian Field-Naturalist 81: 293-294. Savile, D. B. O. 1968. Review: Weber, W. A. 1967. Rocky Mountain Flora. University of Colorado Press, Bolder. viii + 437 pp. Canadian Field-Naturalist 82: 61. Savile, D. B. O. 1972. Review: Cummins, G. B. 1970. Rust Fungi of Cereals, Grasses and Bamboos. Springer- Verlag, New York. 570 pp. Economic Botany 26: 93-94. Savile, D. B. O. 1973. Review: Viereck, L.A. and E. L. Little. 1972. Alaska Trees and Shrubs. U. S. Department of Agriculture, Forest Service, Agriculture Handbook. Washington, D. C. 265 pp. Arctic 26: 172-173. Savile, D. B. O. 1973. Review: Ingold, C.T. 1971. Fungal Spores: Their Liberation and Dispersal. Clarendon Press, Oxford. Oxford University Press, New York. 302 pp. Mycologia 65: 259-263. Savile, D. B. O. 1974. Review : Irving L. 1972. Arctic Life of Birds and Mammals Including Man. Zoophysiology and Ecology. Springer-Verlog, New York, Heidelberg, Berlin. Vol. 2. 193 pp. Canadian Field-Naturalist 88: 121-122. Received 11 January 2001 Evolution of a Naturalist D. B. O. SAVILET Savile, D.B.O. 2001. Evolution of a naturalist. Canadian Field-Naturalist 115(2): 365-380. For the purpose of this discussion I use the term naturalist for trained biologists who have had ample field experience, at least in the early parts of their careers. Darwin and Wallace were the most impor- tant early exemplars of the discipline. Gerald Durrell calls himself an amateur naturalist; but he received substantial professional training in his youth and is really a professional, although he addresses himself to amateurs. The training is usually received, at least in part, in universities; but the committed naturalist pursues knowledge throughout his career. Darwin was trained in university by a reverend gentleman; but what could he have been taught beyond outlines of classification? He took many books with him on the Beagle and studied continuously after the voyage was completed. Beatrix Potter was brought up to be a “lady”, Victorian style, by parents who abhorred anything smacking of professionalism. She became a competent naturalist by her own endeavours. Among the outstanding biologists, of my acquaintance, Ernst Mayr and Nikolas Tinbergen are proud to be known as naturalists; and I am glad to follow humbly in their footsteps. There are laboratory workers who believe that they alone are biologists. They have my sympathy for their mental myopia. I suppose that all biologists recall one or more teachers who stimulated their interests and guided them into a career. I met with little such stimulus until I went to university, and consequently was slow in developing serious biological interests. 1. Childhood and school Spending most of my first six years at Maseno, then a small, isolated settlhement on the equator at 1500 m in western Kenya (in those far off days still +Deceased. This personal account of his life was among D.B.O. Savile’s papers and well reflects the development of his contributions against the background of people and time he was influenced by. It is complementary to the text and bibliography presented by Ginns and Darbyshire as part of their tribute. It is (except for expanding some abbre- viations for clarity) unedited. If Dr. Savile was still living we would have been tempted to negotiate and perhaps soft- en some of his personal judgements of contemporaries, and eliminate some of the repetition toward the end. Readers should realize that many comments here may be more reflection of a particularly dry sense of humour than purely intemperate judgements. — F. R. Cook British East Africa), I had a promising start, as my mother was mildly interested in natural history. I never saw much of the larger animals, except zebras, giraffes, and ostriches, which we saw from the train. (In those days the train ran from Mombasa on the coast only to Kisumu on an arm of Lake Victoria). When my father went to Maseno, as a missionary and manager of a sisal plantation, the land was large- ly covered in tall grass, which had to be cut before he could run a survey of the land allotted for devel- opment. I suppose the later extensive plantings of maize and sisal may have discouraged a lot of herbi- vores. Certainly there were carnivores, which we sel- dom saw; but pet dogs and escaped tame rabbits had a brief existence. My father had to dispose of leop- ards occasionally, if they turned man-eater. I do remember the hyena. I trotted off one day to meet a visiting relative coming by the trail from Kisumu. Unfortunately, I picked the wrong fork where the trail divided. The guest had arrived before they missed me; and when an adult caught up with me I was standing still, face to face with a hyena. I recall that I was staring solemnly at it, and can only sup- pose that my steady gaze discomposed it. If I had known enough to be frightened things might have been different. But perhaps I just looked too skinny for it to bother with. We used to catch chameleons occasionally, trying to make them change colour to match any back- ground. They were also popular as flycatchers. Many of our observations were of insects. I recall mother showing us mantises that closely mimicked a curled up leaf. Very small children can focus down to a few centimeters. When I was perhaps four and my sister close to six, we spent an improving half hour or so watching termites repairing damage to their mud dwelling. Each animal placed its mud pellet in posi- tion and then rapped it quite hard with its mandibles. Mother was interested in the story, but, with her adult eyes, naturally could not confirm it. Whether this behaviour was already known I have no idea. Our attempt at similarly studying safari ants on the march was less successful. We knew enough not to touch the actual column, but thought we could watch from a step or so away. Naturally a scout brought in a company of soldiers to deal with us. My impres- sion was that they bit simultaneously all over me; but perhaps one started and the victim’s reaction set all the others biting. We ran screaming to the house, 365 366 and the household quickly peeled off our clothes. Removing the soldiers is not easy, as the mandibles lock into one’s flesh. Pulling often left the head behind and scissors were then used to separate the mandibles, which were pulled out individually. The lesson is that if you must watch the march, stand back about 2 meters — and still watch out for sol- diers climbing soft-footedly up you. Today, of course, I would use a pair of field glasses modified to focus down to under two meters (feasible with some center-focusing models). Returning to England to go to school I received absolutely no stimulus from any teacher in any field of natural history. In high school I did have science teachers, who dealt crudely with the merest fringes of physics and chemistry. One was a motor-cycle enthusiast, from whom I acquired some knowledge of internal combustion engines, to accompany what I learned of steam engines from a model engineering magazine. Throughout my years in England my elder brother and I tried to learn bird identification in sum- mer holidays spent in rural Devon; but, without even a pair of opera glasses between us, we did not achieve very much. 2. Introduction to Agriculture I moved to Canada in March 1928, to take a two- year course in agriculture at Macdonald College, with summers spent on farms. I travelled, courtesy of the Canadian Pacific Railway Company, who, with the Canadian government, offered assisted passages to potential farmers. The idea was that we would finally buy some of the land held by C.P.R. in Western Canada. It was clear from one summer on a farm that without capital that one could work a life time on farms without getting a down payment on one. However, off I went to a farm at Mystic in extreme southern Quebec, assigned to me by the fac- ulty member at Macdonald who coped with the diploma course students. The farmer, his wife, the hired girl and a permanent hired hand were all supremely ignorant of the world around them. The hired girl did warn me not to touch what she called “poison ivory’, but did not even explain that it was a plant. I got a few practical pieces of information from an Irish boy who had been working on a neigh- bouring farm for a year or two; he at least was vocal. As the weather warmed up I naturally saw a few animals: woodchucks, the more conspicuous birds and a few snakes. After I had seen my first garter snake, I casually asked if there were any poisonous snakes in Canada — a reasonable question from one reared in the tropics. The response was a mixture of hilarity and indignation. Of course, there were no poisonous snakes in Canada! How could I be so ignorant? Poisonous snakes were tropical! One day later in the summer the dog and I went along the lane to the back of the farm to fetch the cows. As we crossed the rocky, wooded ridge that THE CANADIAN FIELD-NATURALIST Vol. 115 divided the cultivated fields a coiled snake raised its head and buzzed at us. Not a very impressive sound, but I could see that the buzz was produced by the tip of the tail. Clearly a rattler! In the days of silent films I could have no idea what a rattler sounded like, but I had imagined something more startling. I grabbed the dog by his collar as he lunged at the snake; then I crippled the snake with a well-directed stone, finished it off with a stick, and hung the corpse on the fence. The rattle was perhaps a centimeter long but with several distinct segments. I was distinctly relieved to have held back the dog, because I would have been held responsible had he been bitten. As junior hired man I was responsible for everything that went wrong on the farm. I did not even mention the inci- dent. In my week or so at Macdonald I had met no biologists, and had not even heard of the National Museum. Consequently I left the snake on the fence and finally forgot about the episode. Many years later, after seeing an article on rattlesnakes, I recalled it and told the story to Clyde Patch. He confirmed that it had to be the Massasauga Rattlesnake, Sis- trurus catenatus, recorded sparingly in northern New York but not in Quebec*. I realize now that in prea- gricultural days it must have been more widespread, but clearing the land exterminated it except on rocky ridges. Returning to Mac, I entered the diploma course and, among others, I took a course on fungal dis- eases, put on by J. E. Machacek, who was then com- pleting his Ph.D. He was the first teacher who brought a botanical topic to life for me. Already con- sidering transferring to the B.S.A. course, I was per- suaded by his lectures to specialize in plant patholo- gy and mycology. 3. Training in Biology Transferring to the B.S.A. course was not very simple. Like other British students I had entered the diploma course with the understanding that I could take a transition year, followed by the standard third and fourth years. When the schedule was changed, making the first two years standard B.Sc. courses with no agriculture, the transition year had to be abandoned. Four of us protested vehemently and we ended up taking all the courses of the first two years that we could not talk our professors out of agreeing to. I had to skip first year chemistry and math. The chemistry did not matter, but the math showed me *More likely he meant a Timber Rattlesnake, Crotalus hor- ridus, which has occasionally been reported although never verified in Quebec, along the Quebec-New York border (see Claude Melanson 1961. Innconus et Mccones: amphib- ians and reptiles of Quebec. La Provancher Société, Quebec, Second edition). The Massassauga has never been reported from Québec — F.R. Cook 2001 how shamefully I had been ignored in my last year in school: I did not have the coordinate geometry that leads to the calculus. I survived but it was a gruelling undertaking. Dorothy Newton (later Swales) put on the freshman botany course. She was such a spirited lecturer that I would probably have majored in phanerogamic botany had there been such an option. In the summer of 1931 I worked for J. G. Coulson in the Plant Pathology Department, and this included the planting and care of field plots. When not needed by him, I helped Dorothy Newton in the then primi- tive herbarium, mounting specimens that had been 15 years or more in newsprint, and also finding my way about Gray’s Manual, edition 7, in hopefully identifying many specimens. In the third year I took elementary genetics, a technique course that covered cytological methods, and a comparative morphology course that included substantial cytology of the major plant groups. Thus I had a foundation for a field that I then had no idea of entering. I also had a half year of mycology and a full year of plant pathology. 4. Fire Blight and the Division of Botany In the spring of 1932 I applied for a position in the fire blight investigation at Abbotsford, Quebec, hav- ing heard that a second appointment was planned. As it happened they were considering appointing an entomology student, but no appointment was made that year. However, the Division of Botany at Ottawa now had my name. Thus, when the incum- bent was killed in a car accident, I was phoned by H.N. Racicot to get ready to leave, and he picked me up later in the day (a Saturday). So I was hastily indoctrinated into the experimental work and the operation of a weather station. We had to go to church next morning, to a memorial service for my predecessor. The plate came round, so all I could do was to put in the only money that I possessed, a 50- cent piece, which I had earned by inking diagrams for a graduate student’s thesis. So before Racicot returned to Ottawa I had to touch him for an advance. We were then in the depression and living on credit was normal. There was little spare time that summer, as the copying of the weather records was far behind schedule; and, because bacterial diseases were cov- ered in the final year, I had to read up all that I could find on fire blight. Fortunately the pollination of the trees in insect-proof cages had already been done. I was able to buy an ancient bicycle, which I chris- tened the Death Rattle: no bell was needed, for everyone could hear it coming. The chain kept com- ing off, owing to having been bent. It was not a stan- dard chain, but luckily I found its mate in Eaton’s catalogue and replaced it. I could then travel at 10 m.p.h. without mishap. Our plots were mainly in orchards 2 miles in one direction and 3 miles in the SAVILE: EVOLUTION OF A NATURALIST 367 other from the home base. On a state visit next year, Dr. Giissow disapproved of my bicycle (a much bet- ter one and fitted with a box for carrying Petri dish- es, etc.) and said I should buy a car to save the gov- ernment’s time. I replied that I should be glad to do so if given a salary that permitted it. I remained a cyclist. My pay worked out at about $90/mo., based on a 60-hour week. Fortunately, I paid no rent and did my own cooking; and thus I saved substantially toward the next winter’s costs. Being alone that first summer I put in a full 6-day week, well over 10 hours per day, and still had to tend the weather station on Sunday. However, browsing around on Sunday I used to collect a few plants, which I hopefully identified with an acquired copy of Gray’s Manual or Bailey’s Manual of Cultivated Plants. Thus I gradually learned many plants of what I later realized to be an attenuated Appalachian flora. (Some plants I did not see again until attending meetings in Massachusetts about 30 years later). What provoked my interest in learning to recognize plants? Surely Dorothy Newton- Swales’ stimulus was important; but I recall insisting that if I was to collect and study parasitic fungi I had to learn their host plants. Accordingly, I collected very few fungi during my first two or three summers at Abbotsford. As soon as Arthur’s Manual of the Rusts appeared in 1934 I bought a copy, and from then on the rusts were my main concern; but I con- tinued to collect other foliicolous fungi as time per- mitted. In the summer of 1933 I must have identified a good many plants, although, like most beginners, I was still avoiding the grasses and sedges. My final undergraduate year (1932-33) included systematic botany, under Dorothy Swales, in which I was ahead of the class, partly because I was familiar with the use of keys and partly because I could recognize many genera. I vaguely recollect that in the two exams I received the highest marks ever given for the course. In 1935 Fr. Marie-Victorin commended me for disentangling the dwarf Euphorbia spp., including E. glyptosperma, which was not in Flore Laurentienne. In the fall of 1933 I registered for my M.Sc. under Prof. J. G. Coulson. Because of my experience with fire blight, Coulson suggested that I study the limit- ing factors in some bacterial leaf spots, paralleling a recent study on some fungal leaf spots. Some of the observations seemed interesting, and Coulson sug- gested that I present them at the spring meeting of the Quebec Society for the Protection of Plants. I gave the talk and, at his request, turned the final manuscript over to Coulson who was to deliver it to the editor. In due course the proceedings appeared and my paper was not included. Surprise, surprise! Coulson had failed to transmit the manuscript. This was my first experience with a non-publisher and, 368 coupled with later experiences in Ottawa, it could have been disastrous. I wonder if supervisors (well, nominal supervisors anyway) appreciate how much their dilatoriness (or worse) may damage a student’s career. How can a student expect a fellowship if he has nothing to show that he can do research? I barely got my thesis turned in by the deadline for fall convocation. Each edition was pronounced per- fect by J.G.C.; then, after it was typed, he would want it all changed. I think there were five editions. Once I got the phone call on Friday even after my fellow-student had gone off for the weekend; and I could not move until Monday as someone had to maintain the weather station. I was to have gone to Ottawa for the winter but was fired by telegram at the last moment (H. T. Giissow having evidently cleaned out the money for one of his royal tours). After convocation I accordingly caught a boat to England for a family visit. Thus the winter was essentially wasted. The fire blight work continued in the spring of 1935 fairly satisfactorily, except that when I wrote . up a report on any aspect of the work it quietly dis- appeared. I continued with all of the work as origi- nally planned and new aspects as they occurred to me, most of which I cannot now recall. (Yes, despite statements to the contrary, ants were up the apple trees throughout the day and night in fine weather, dodging clearly exuding cankers, but surely carrying some bacteria. I checked one tree group hourly for 72 hours until I crossed the tent to turn off the alarm without waking!) Of course all the notes and drafts went home with Homer Racicot and were never heard of again. The fall of 1935 was an echo of 1934. I was packed and ready to go to Ottawa in the lab car when I was again fired by telegram (N.S.F.; perhaps anoth- er royal tour). Stopping at Macdonald I found that Harold Brodie had come on staff. I had already met him and was impressed especially by his productivi- ty — he had already published several papers. I found a small grant available to add to my $300 summer savings, and registered again in graduate school. I took advanced mycology from him and made a good start in rust cytology, using several dif- ferential staining techniques, some of my own devis- ing. Brodie and I were soon firm friends. On the first weekend I walked back to the college woods to look for rusts and to botanize generally; and promptly met the Brodies (Helen wearing a shocking old hat of Harold’s to keep the flies off) collecting agarics for class use. As naturalists we spoke the same lan- guage. Before winter’s end he realized that my inter- est in the rusts was serious and encouraged me to try to get to Ann Arbor and take my doctorate under E. B. Mains. In the spring of 1936 I returned to Abbotsford for the last season; really a waste, when I look back on THE CANADIAN FIELD-NATURALIST Vol. 115 it, as all my writeups kept disappearing, “taken home to read over the weekend.” A couple of years later when I read Don Marquis’ Archie and Mehitabel, I recognized Homer’s prototype: Freddie the Rat, another inhabitant of Don Marquis’ office, was a lit- erary critic. He would read one of Archie’s poems, sniff, and then eat it. Homer Racicot to the life! Soon after I reached Ottawa, Dr. Giissow, sick of the lack of results from the La Pocatiére Lab on what was then known as bacterial wilt and rot of potatoes, dumped the problem on Racicot and me, which in effect meant me. The “bacterial” in the common name was a lucky guess, for all that previous investi- gation had isolated were yellow saprophytic bacteria and soil yeasts. During the preceding year at Macdonald I had told J. C. Perrault that, if he would fix infected tissue, I would embed, section and stain it. That would have given the answer, for one of my stock methods for showing pathogens in plant tissue was Gram staining with a counterstain; but he declined the offer. When I received material for study I at once started pouring dilution plates, but simulta- neously fixed and embedded stem and tuber tissue material. As it happened I first got the pathogen from plates, slightly speeded up by colonies appearing near an actinomycete colony. Obviously it was a slow grower and everyone had thrown out plates after 5 days as sterile. Adding yeast extract to the medium shortened the incubation period. Critics belittled the finding as sheer luck; but a few days after the initial isolation I had stained sections showing the typical Gram positive short rods of what is now recognized as Corynebacterium sepedonicum. With Gram posi- tive plant pathogens so rare, its identification took a matter of minutes: bacterial ring rot, known for years in Europe, where small whole potatoes are planted, but causing little damage. In North America, where cut sets are planted (to reduce virus infection), it became extremely serious, for the cutting knife proved to be a more reliable means of inoculation than any that I could devise. I had to take a week’s leave and go to Macdonald, at my own expense, to confirm the identification, using differential media in the Bacteriology Department and controlled tempera- ture chambers in the Plant Pathology Department. (I never understood why such facilities were not avail- able in the Bacteriology Division at the C.E.F. How can one study bacteria without them?). I asked Racicot to check symptom development on my inoc- ulated plants in the greenhouse, but he never did so. Getting priority on this identification was critical, as pathologists in Maine were known to be working on the disease. Accordingly, I wrote up a short paper describing the disease, the pathogen, the main meth- ods of spread, and a Gram smear technique that I had devised for rapid confirmation of the disease. The paper was reviewed by I. L. Conners and F. L. Drayton and was ready for submission to Scientific 2001 Agriculture (long since defunct) when Homer roused from his winter dream world long enough to see what was happening. He immediately insisted that he had to be senior author. As he had done nothing all winter except for hindering my work, I said that he had no claim even to be junior author, all my work being done in spite of him. He whiffled off down- stairs to Dr. Giissow, who must have had a pretty good notion of the truth, which had been a main topic of conversation in the building for many weeks. He finally came back and said I could be senior author, which I had to agree. After I went to Ann Arbor he, with some additions by I. L. Conners, paraphrased the original paper for American Potato Journal, with me as second author. This may have been what persuaded him in later years that all the work had been his. 6. Ann Arbor Days This key paper fortunately appeared promptly and persuaded the Botany faculty at University of Michi- gan that I could do research. Consequently they yielded to Harold Brodie’s urging and offered me a small fellowship. I thus was able to spend two win- ters at Ann Arbor and secured my Ph.D. in the spring of 1939 under E. B. Mains. My thesis on nuclear structure and behaviour in rusts, successfully disen- tangled several gross misinterpretations that resulted largely from complete reliance on iron alum hema- toxylin staining, which is seriously non-selective. For example, strange figures in mycelium, purport- ing to be the telophase of one nucleus with a large nucleolus at each end, were shown by Feulgen stain- ing to be four nuclei following simultaneous division of a dikaryon. The “nucleoli” are strongly DNA pos- itive; and the “telophase” is two recently regenerated nuclei squeezing past each other to restore compati- ble pairing. World War II started just before my paper appeared in American Journal of Botany for October 1939. Such esoteric pursuits as rust cytology had to be set aside; J. H. Craigie seems to have for- gotten these and other findings when he optimistical- ly turned to rust cytology after his retirement in 1952. He failed to use modern methods and in 1959 published many of the same fictional interpretations commonly made prior to 1939. I must emphasize that rust cytology is not a topic that can be studied at odd times, especially late in life. I made good progress in the winter of 1935-36, and again in my time at Ann Arbor (1937-38 and 1938-39). Although I used a microscope at Ottawa in 1938 for many hours, mainly checking smears for ring rot bacteria, I received a shock on returning to Ann Arbor after four months absence from cytologi- cal study: All the objects in my rust sections seemed extremely small and I could not distinguish fine details. Checking with a stage micrometer assured me that the microscope was undamaged. The trouble had SAVILE: EVOLUTION OF A NATURALIST 369 to be in my eyes. To have your eyes fail you half way through a cytological thesis problem is a horrifying thought. I dared not tell anyone but persevered. Sometimes resolution of details was slightly improved and I began to suspect that the problem was physiological or psychological. After a week there was a definite improvement; but it was nearly three weeks before resolution was fully back to normal. This experience should warn anyone against trying to do anything as visually demanding as rust cytology at odd times. Only recently, when I read Evelyn Fox Keller’s A Feeling for the Organism (W.H. Freeman, New York and San Francisco, 1983), did I realize that I was not alone in this grim experience. In 1944 Barbara McClintock was invited to visit Stanford to help with the cytological aspects of George Beadle’s work on Neurospora. She was delayed in completing arrange- ments before she could pack her microscope and go by train across the continent; so I judge that she must have been away from microscopic work for several weeks. As the story is told, she was also worried that she might not be able to contribute to the project. She set up her microscope and for three days she could see nothing. Realizing (very wisely, I think) that she had to do something positive, she set out for a walk and finally rested on a shaded bench on the Stanford campus. Then suddenly she was sure that everything was going to be all right, and, indeed, everything was all right; and the project was very successful. Her recovery was rapid, perhaps partly reflecting a rela- tively short absence from her studies. It must also be noted that Dr. McClintock was 42 at the time of her experience, solidly established in cytogenetics and probably well prepared to cope with a psychological problem; whereas I, aged 29 and not established in research, was poorly prepared. Whatever the funda- mental cause of such a problem, the experience fur- ther emphasizes that fungal cytology is not something to be taken up at odd moments, and especially late in life when sustained microscopic study is a severe strain. In 1949, I had a happier experience of visual adap- tation. Various people had insisted that the Chimney Swift, Chaetura pelagica, beats its wings alternately, despite this being a palpable impossibility because the body would simply oscillate about its long axis with the wings producing little lift or thrust. How- ever, I rigged up a shutter stroboscope in front of one objective of my binoculars and could often briefly stop the wings of a bird flying toward or away from me. The wings were, of course, always both up or both down. The interesting point is that, after doing these observations daily for some two weeks, I could fully resolve the wing motion without the strobo- scope (Auk 67: 499-504, 1950). Visual acuity obvi- ously is not wholly explained by geometric optics. Ann Arbor was a stimulating place to work at in 370 those days. The Botany Department was large and the staff had very varied interests and experience. Unfortunately I could take few courses in the time available. I did take a cytology course (with unfortu- nately nothing new in it); and I took Professor Ehlers’ course in agrostology, finally reducing my ignorance of grasses and sedges (but unfortunately this was 20 years before modern concepts of the grass subfami- lies began to appear). I should have profited from another year, with time to learn more about bryophytes, marine algae and geology. However, my contacts were many and stimulating. I retain a mental picture of Carl Larue (experimental morphologist) and myself staring in at some shrub with leaf galls that looked like spinulose gooseberries. We specula- ted as to whether the resemblance was merely ran- dom. In those days, before the acceptance of trans- posable genes and DNA transfer, the idea naturally did not get off the ground. 7. War-time Turmoil Returning to Ottawa after commencement Connie and I were married in late July, which was some compensation for having to work nominally under - Homer Racicot. That fall I helped inspectors in their field inspection of potatoes. Usually ring rot, our main problem other than viruses, was clearly dis- tinct; but in one area abnormal weather caused it and black leg to look nearly identical, and Gram smears had to be used freely. The outbreak of war put pure science out of the picture. I have a blurred recollec- tion of committees for boosting food production and of posters that presumably appeared in village post offices where they may or may not have been read. All the junior staff had to help in the famous national registration. The one memorable question was “Can you milk a cow?” Farm boys looking for a lucrative job in a munitions plant said “No”. City boys hoping for a safe rural job often said “Yes”. Presumably some beneficial outbreeding in the pop- ulation resulted. The enormous mass of forms, which could never have been indexed by available means, was stored in an Ottawa building that collapsed under the weight. There are less tedious ways of destroying a building. I recall two technical surveys by which people with assorted talents could be reached. The military people, in search of tropic-proofing talent later in the war, could have found me by looking under meteo- rology, fungi and optical instruments. But they did not know of the registers, and found our Division by blind luck (good luck as it turned out). During the early part of the war I examined count- less decaying fruits and vegetables; but my only research paper was a short one on malformation of potato starch grains due to viral infection reducing the elasticity of the leucoplasts. That one seems to have slipped through while Homer nodded THE CANADIAN FIELD-NATURALIST Vol. 115 (American Journal of Botany 29: 286-287, 1942). H.H. Bartlett enthusiastically proclaimed that this would become known as the Savile phenomenon; but the war was on and nobody even noticed it! Foolishly I joined the R.C.A.F., trying to make some use of my appreciable and varied technical knowledge. Too late I appreciated a warning to the effect that as a civilian scientist you can talk to peo- ple of any rank, and are often listened to, (as indeed I found much later as a tropic-proofing adviser); but as a junior officer you cannot advise or help anyone of a rank above you. The next 18 months were continuous frustration to anyone with a conscience, leavened only by an improved understanding of machine-shop operations and of low-speed aerodynamics; not that I expected to make use of the latter, but I was still interested in learning. I did also develop a simple device to stop static charge from bunching up paper in the met. office teletype. This delighted the C.N. teletype technician, who presumably got credit for introducing it at other stations in his area. My com- manding officer, Elmer F. (no, not Elmer Fudd, who was funnier and probably smarter) took a rabid dis- like of me, mainly because when he asked my opin- ion on some problem involving several variables, I remarked that an intelligent answer required consid- erable thought. He blew right up with “That’s the trouble with you civilians; you always want to stop and think. In military action a snap decision is all you want.” This was accompanied by a snap of his fin- gers, one thing that he was good at. Because I was still hoping for a transfer to Operational Research I had to clamp on my tongue, instead of replying that that seemed to explain why we lose the first three years of a war, until some intelligent amateurs get into the higher ranks. Much later I found that I had been asked for by Operational Research but Elmer had turned down my transfer. Too late he found out that it was I (the brainless thinker) who had solved the teletype problem. (There being one in the main office he had heard about.) Suddenly, with my minor gadget that any schoolboy with an inventive turn of mind might have dreamt up, I was an engineering genius to his simple mind. He was plainly embar- rassed because he had pressed for my discharge. Even the most honorable discharge is distressing dur- ing war, and the fact that Elmer had guessed wrongly about me was not particularly heartening. Most guessers expect about a fifty percent score; but Elmer’s score was surely under ten percent; he was an uncanny wrong guesser. However, I may have been more use back in Agriculture than in Operational Research as it turned out. I returned to the Division as assistant to Ibra Conners, who, as curator of the mycological herbari- um, compiler of the plant disease survey reports, and registrar of commercial fungicides, certainly needed some help. Things as once went more smoothly. 2001 Lyle Drayton, having had to deal considerably with Racicot in my absence, had found that my com- plaints about his obstructionism were more than jus- tified. As the odd jobs section, Conners and I received not only fungal disorders, but mite or insect injuries, physiological disorders, algae, cyanobacteria, etc. Thus when an army officer, worried about fungal damage to optical glass, came to the Division he was brought up to us. This really was an odd job, as I soon found out. I took over the project as I had some familiarity with optical instruments and their servic- ing, and knew a good deal about meteorology. (I used to coach the met. observers at Centralia on cloud recognition and even cloud transformation.) Thus instrument damage and its prevention became my problem for the last two years of W.W.II, during which time I devoted ca. 15 hours per 60-hour week to it. (This included time spent at home making tools for stripping or adjusting instruments.) Salaries were frozen during the war; however, people with extra work loads could receive a $300 p.a. war duties sup- plement; but no-one in our Division did. (After tak- ing such an emotional beating in W.W.I, Dr. Giissow probably did not dare to apply for any.) Thus I ended the war at a salary designated for a new graduate with no graduate training or experience. It soon became clear that identification of the fungi in instruments was academic: Many common and widespread species were involved. (The situation was probably somewhat different for fabrics and insulating materials, handled by G. A. Ledingham in N.R.C.) My approach was to improve sealing and thus keep molds, mites and moisture out of the instruments. As Mole said to Ratty, after he cut his shin, my attitude was “Never mind what done it.” Two models of binoculars being built by Research Enterprises Ltd. (R.E.L.), a crown company, were substantially improved by minor modifications, aimed at preventing moisture penetration through pressure changes induced by fluctuating temperature. A more significant change, which would have cured the problem by matched contour milling of the cover plates and mating ends of the body castings, was vio- lently rejected by the plant superintendent. He was quite content to go on filling the gaps with luting compound, which cracks on shrinking just as it does when you caulk your window frames. Yet even in 1943 contour milling equipment and techniques were available. A rifle scope was a trickier problem. A computer (They were human in those far-off days) decided to recompute the optics, to eliminate most of the residual aberrations; so I had time for action before the instrument went back into production. I completely redesigned the sighting head, to guaran- tee a positive seal. Six instruments were hand made for trials. The one that I saw was certainly a beauty. However, the war was coming to an end and C. D. SAVILE: EVOLUTION OF A NATURALIST 371 Howe, who had organized the company, cancelled all contracts and the company instantly disintegrated. The army group with which I had worked also disap- peared through transfers and demobilization. I even- tually turned my notes and drawings for the sighting head over to the newly founded Defence Research Board as the logical safe depository. All my tropic- proofing reports had security ratings; and so I had no acknowledgeable publications for the work. Although my time in W.W. II was not entirely unproductive, most of it did not promote my devel- opment as a naturalist. My varied activities did at least deter me from becoming an extreme specialist. A working knowledge of low-speed aerodynamics, acquired during the war, led eventually to a detailed understanding of the adaptive characteristics of bird wings (Evolution 12: 212-224, 1957). But even by 1948 I was thinking increasingly in terms of evolu- tionary adaptations in birds and, certainly soon after- ward, of those of plants and fungi. 8. Biological Exploration of Canada After the war I was able to spend slightly more time on mycological studies; but, with Lyle Drayton serving full time as Associate Dominion Botanist, I was still stuck with all the work on diseases of orna- mentals and plant quarantine interceptions until a position in this field was filled. At this point, credit must be given to K. W. Neatby, who, as Director of Science Services, strongly encour- aged his staff to explore the country. Tragically he died just when the results of his stimulus were starting to show. With the onset of the Northern Insect Survey, with which I served three seasons as a botanist, describing biting fly habitats, I inevitably learned something of the ecology of the plants and their para- sites. For example, in 1949, my first full season in the field, I found that at a tree-line site (Great Whale River or Poste de la Baleine) obligately heteroecious rusts occurred only if the alternate hosts were virtually in contact. Thus my studies of rusts and other para- sites were increasingly in terms of hosts and host environment. | think that, by the time I joined J. A. Calder for the start of the British Columbia floristic survey in 1953, I qualified as a professional naturalist although I obviously had a lot to learn — especially as neither of us had set foot in the Cordillera. Working with Jim Calder was an education in itself. Not only was he a meticulous collector, whose specimens were clean, well-folded and well-dried; but he put more ecological data on his labels than was usual at that time. I like to think that I contribut- ed my bit in 1953 when we invaded completely new country without an appropriate flora for reference. I took along an aircraft altimeter, which allowed us to record altitudes accurately. Equally important in mountain country, as we soon came to realize, is the direction of slope: the difference between north and a2 south exposure may be equivalent to several degrees of latitude. During the previous winter I found that our British Columbia. collections of many plants were devoid of fruit. In case we should not be able to collect fruiting specimens later, we accordingly included old fruiting stems of flowering perennials. Calder’s primary training was in geology; and from him I learnt more of physiography and glacial geolo- gy than I had worked out for myself in the Hudson Bay region and northernmost Newfoundland. In British Columbia we thought from the start in terms of biogeography and late glacial history. This approach also influenced my mycological work; one example being studies in the co-evolution of Saxifragaceae and their rusts, which developed over many years, terminating in a summary in Annals of the Missouri Botanical Garden 62(2): 354-361, LOWS. I worked with Calder in 1953, 1954 and 1957, the last season being spent in the Queen Charlotte Islands. Botanically the Q.C.I. were exciting, demonstrating the occurrence of a low-level refu- gium on the west coast contiguous to the deep water . of the Pacific Ocean. Unhappily I caught a cold on the ship from Vancouver and, in the dank hotel at Queen Charlotte City, never completely threw it off. I was a drag on the party climbing the small, but steep and slippery mountains; and decided that I was too old for such work. I accordingly was glad to volunteer, on Harold Senn’s behalf, to join a party from McGill in 1958 on Somerset Island. I also had a brief arctic field trip before the 1959 Montreal Botanical Congress. In 1960 I worked out of Isachsen on Ellef Ringnes Island, with the Polar Continental Shelf Project. My experience in British Columbia partly prepared me for interpreting what I saw on Somerset and Ellef Ringnes islands. The latter is in the heart of the sup- posedly ice-free northwestern Queen Elizabeth Islands, pronounced, by those who had never studied them, to be the centre for all Canadian arctic endemic plants. What struck me, as the season developed, was that the flora round Isachsen is very small, the indi- vidual plants are very small and sparsely distributed, and, except for Puccinellia angustata, which is wide- spread across the northern islands and Greenland, the endemics are absent. I was able to work out during the summer, from local observations and the known distribution of all the arctic vascular plants, that, although the northwestern islands had intermittently been covered by snow and ice, even since the end of the Hypsithermal Interval, the ice had not been heavy enough to flow appreciably under its own weight. Thus it had wiped out the plants but had preserved rather than erasing small relief features. The fully documented story was presented in Canadian Journal of Botany 39: 909-942, 1961. I had clear pri- ority in this crucial discovery, as was fully recog- THE CANADIAN FIELD-NATURALIST Vol. 115 nized by arctic workers at Ottawa at that time. (The story was plagiarized a few years later, independently by two other investigators, both of whom had my paper!) It was certainly a breakthrough in our under- standing of arctic biology and late glacial to post- glacial history, completely reorganizing several con- cepts. The extensive collecting by our botanical and mycological staff saw the introduction of various new techniques, including the evolution of the field press, and caused me to write Collection and Care of Botanical Specimens Canada Department of Agri- culture, 1962, 1973, which went through two English and one French edition, not to mention a pirated Russian edition. 9. Some Fruits of Field Work In 1962 I spent the summer at Hazen Camp in northern Ellesmere Island, collecting plants and fungi and supplying the ecological background for studies by the entomologists. Fortunately, I was once more in good condition, for in that dry Shangri-la I had to do a good deal of mountain climbing to reach some species. For a high-arctic site Hazen Camp was very rewarding botanically, mycologically and ornithologically — and indeed entomologically as my colleagues demonstrated. With complementary studies by J. A. Parmelee on Axel Heiberg Island in 1961, and at various DEW Line sites in 1963 and 1967, our picture of geographic patterns of vascular plants and parasitic fungi in the Canadian Arctic was reasonably complete. Thus, I had ample data for bio- geographic and evolutionary studies of both plants and rusts. For example, I was able to demonstrate clinal variation in Puccinia cruciferarum, a micro- cyclic rust without pycnia (Mycologia 56: 240-248, 1964) exploding the dogma that without pycnia there could be no genetic recombination. The conspicuous distinctions of Puccinia poae-nemoralis ssp. hyp- arctica, in spore size, pigmentation, wall thickness, wall sculpturing and host specialization, have all accumulated since the onset of the Wisconsin glacia- tion, supplying the strongest support for parasexual recombination in the rusts (Arctic Adaptations in Plants, page 61, 1972). The incomplete segregation of Poa hartzii from Poa glauca evidently occurred in the same period. _ My booklet Arctic Adaptations in Plants (Canada Department of Agriculture Monograph 6, 1972) brought together most of my own and other peoples’ observations on plants and fungi. The first printing of 2000 was exhausted in two years. A second print- ing of 3000 was exhausted by 1974; and a third printing of 3000 is still in occasionally demand. Clearly it filled a need. About half the content was either original or from my own publications. It cer- tainly represented a large return on a few seasons of field study. 2001 Successful evolutionary studies must usually have an interdisciplinary aspect. When H. J. Brodie (Cana- dian Journal of Botany 29: 224-234, 1951) clarified the splash-cup dispersal mechanism he spoke of rain- drops as the active agents; and probably no-one seri- ously questioned his terminology. However, even then, through my work on fire blight and my reading in meteorology, I knew that rain rarely falls vertically and that the drops are small; but I did not pursue the question. In the summer of 1958 I found Chrysosplenium rosendahlii Packer (misnamed by me in Canadian Journal of Botany 37: 999, 1959, as C. iowense) growing freely in moist grass and sedge meadows in southern Somerset Island Although rain is not rare in that region, it is usually fine and slanted by wind. Could the large drops falling scarcely half a metre from grass or sedge panicles possibly operate the splash-cups? It seemed unlikely; but I was uneasi- ly aware that drops from a canopy are large, for their sound on an umbrella is much louder than that of raindrops in the open. I searched the meteorological literature for help in vain. (Meteorologists are more concerned with terminal velocities of drops than with their initial acceleration). In 1972, as the publisher’s reader of the first draft of The Bird’s Nest Fungi, 1 came on Brodie’s report of the dry plain around Lima, Peru, where only fog, and no true rain occurs. The fog supports low tussocks of woody vegetation. There, to his astonishment, he found well-developed specimens of the bird’s nest fungus Cyathus olla on debris under many of the larger and denser tussocks. I was not astonished, but I was exasperated: This was my Somerset Island puzzle in repetition. Again I failed to get any answers to my questions. Years later, when I was again searching in the Agrometeorology library, the director, Dr. W. Baier, asked what my problem was. When I explained he turned me over to Dr. Henry Hayhoe, a mathematician. With published information on terminal velocities of falling drops in a vertical wind tunnel, Dr. Hayhoe was able to derive an equation to give the velocity of a drop of given size at a given distance of fall (Savile and Hayhoe, Canadian Journal of Botany 56: 127-128, 1978). Later calculations showed that a 4.5 mm drop (proba- bly close to the minimum size shed from the canopy) after 0.25 m of fall has a much higher relative momentum than a 2.5 mm drop at terminal velocity (Savile, Davidsonia 10(4): 65-69, 1979). It must be noted that natural raindrops rarely exceed 2.2 mm diameter, for larger drops break up in a long fall. Had I had mathematical expertise available, this problem might have been solved in 1958. Not surprisingly, recognition of the mechanism of the splash-cup quickly led to further observations. Brodie demonstrated the springboard dispersal system in Salvia, Ocimum and Kalanchoe (Canadian Journal of Botany 33: 156-167, 1955). Independently I recog- nized it in Tiarella trifoliata in the spring of 1953, SAVILE: EVOLUTION OF A NATURALIST 373 simply because Calder and I took our first look at the coastal rain forest of British Columbia on a drizzly day. I recorded it casually in a preliminary study of the rusts of Saxifragaceae (Canadian Journal of Botany 32: 400-425, 1954) without christening the mechanism. The misconception by Brodie, following Buller’s lead, that these devices are being powered by rain- drops became general and even resulted in a comic strip sequence showing Cyathus cups (quite well drawn) with peridioles being dispersed, quite impos- sibly, by strongly slanted raindrops. Until the deriva- tion of Dr. Hayhoe’s equation no-one could be con- vinced that drops from a low canopy could function adequately. Even Brodie, who had a very receptive mind, could not be convinced until he saw the fig- ures in Canadian Journal of Botany (1978) and Davidsonia (1979). At last he saw how Cyathus olla survived in the Peruvian desert (where, I am sure, some rodent emerges in the cool dusk and effects more distant transfer). Would I have stumbled on the springboard action of the Tiarella capsule if I had seen it only on a fine day? Probably not as promptly, but plainly I was thinking about Saxifragaceae in terms of dispersal and was puzzled by the strange capsule of Tiarella. I could scarcely have missed it next year in the humid forest east of Prince Rupert where both T. trifoliata and T. unifoliata flourish. Looking back after some 35 years, my activities in British Columbia indicate that I was thinking as a nat- uralist, considering plants and fungi in terms of Pleistocene glaciation, biogeography, means of dis- persal, etc., although I had a lot to learn. Well, the active naturalist never stops learning. For example, I was trying to shake off the appallingly artificial lump- ing of the rusts of Saxifragaceae. But I was still treat- ing at varietal level rusts that, with more abundant material, later proved to be good species that do not intergrade on contact. Thus it is was not until 1973 (Canadian Journal of Botany 51: 2347-2370) that I could present a treatment of all these rusts that still seems essentially satisfactory. My revision was still fully morphological but recognized small, but consis- tent, differences that most authors had ignored. From work on bacterial diseases and rust cytology I was used to oil-immersion microscopy, which was gener- ally avoided by mycologists because cedar oil had to be wiped off the immersion lens promptly. I pio- neered using tri-immersion in examining rusts, and other fungi about the end of World War II, using medicinal paraffin until non-drying immersion oils were available. I found I could measure spores more accurately and faster than with the 4 mm objective. With these characters the rusts of Mitella diphylla and Tiarella cordifolia (Puccinia heucherae var. minor and var. heucherae respectively) were usually separa- ble; but there were a few misfits. Having a suspicious 374 nature, I checked the foliage of all material in DAO, and found that the petiole hairs were reliably distinct, although a few non-fruiting specimens had been mis- named (Canadian Field-Naturalist 87: 460-462, 1973). Returning to the mycological specimens all the misfits vanished. These two plants often grow togeth- er in the eastern hardwoods, and occasionally the helpful collector had included an inflorescence — but of the wrong plant. Thus recognition of the small but reliable differences between these two plants led to a reliable way to distinguish vegetative material of Mitella and Tiarella. 10. Coevolution Although the term coevolution apparently did not come into use until 1964 (Ehrlich and Raven, Evolu- tion 18: 586-608), the phenomenon had been consid- ered for more than a decade previously. But its early history is obscure, for it is often uncertain whether there have been reciprocal genetic changes that are the sign of true coevolution or only a more or less random coexistence. However, when one can trace parallel advancing lineages in both partners I think the occurrence of coevolution must be accepted. (See - K. A. Pirozynski and D. L. Hawksworth, Chapter 1 in Pirozynski and Hawksworth, editors, Coevolution of Fungi with Plants and Animals. Academic Press, London, 1988). Because I have been cited as a pio- neer in coevolutionary studies, I must try to review my role in this field. My early observations were mainly aimed at using rust or smut data pragmatical- ly to improve our understanding of plant relation- ships. In the first summary of my ideas (Science 120: 583-585, 1954), I presented the rather miscellaneous conclusions of a few earlier studies. It was shown that the same rust (not merely a morphologically similar one) attacks Acorus calamus and Sparganium eurycarpum, and that the two plants, despite superfi- cial distinctions, have several anatomical similari- ties. Following the mycological conclusions (Parmelee and Savile, Mycologia 46: 823-836, 1954) Acorus has been variously disposed. Accord- ing to Cronquist (Evolution and Classification of Flowering Plants, 2nd Edition, New York Botanical Garden, Bronx, New York, 1988) it differs from Araceae in so many ways that it has been put into a separate family Acoraceae. However, it will proba- bly prove to be very close to Sparganium eury- carpum. In this example the rust is useful to the systematist, but there is no very clear sign of coevo- lution. On the other hand, it was shown that the rusts of Poaceae, Cyperaceae and Juncaceae are, on aver- age, much more primitive than the rusts of Liliales. Some rust clans can be traced through a group of grass or sedge rusts into greatly advanced autoecious liliicolous species. Here there is evidence of abun- dant coevolution, not all evident in 1954 but increas- ingly so as time went by. THE CANADIAN FIELD-NATURALIST Volkais By about 1960, although most of my papers dealt with detailed taxonomy of various groups of fungi, I was increasingly concerned with that of the host plants; and minor emendations of the latter were occa- sionally indicated, including: disposition of Allium (Mycologia 53: 31-52, 1961; Nature 196: 792, 1962); Eriogonium (Canadian Journal of Botany 44: 1151-1170, 1966); Filipendula (Brittonia 20: 230-231); Veroniceae (Canadian Journal of Botany 47: 1085-1100, 1969); Ledum (Canadian Journal of Botany 47: 1085-1100, 1969); Scirpus etc. (Canadian Journal of Botany 50: 2579-2596, 1972); Saxifra- gaceae (Canadian Journal of Botany 51: 2347-2370, 1973; Annals of the Missouri Botanical Garden. 62: 354-361, 1975); Brassicaceae (Canadian Journal of Botany 52: 1501-1507, 1974); Pedicularis (Proceed- ings of the Indian Natural Sciences Academy 438(6): 223-227, 1977). Several families, including Poaceae and Cyperaceae-Juncaceae, were discussed in Botanical Review 45(4): 377-503, 1979. Poaceae was further discussed in Chapter 16 of Grass Systematics and Evolution, edited by T.R. Soderstrom et al., Smithsonian Inst. Press, Washington, 1987. Interrela- tionships of Poaceae, Cyperaceae and Juncaceae, with confirmatory evidence of Juncaceae evolving from Cyperaceae, are discussed in Canadian Journal of Botany (68: 731-734). 11. Why are Scientific Advances so Slow? Fully professional biologists seem to have appeared ca. 1870 in Germany, where the universi- ties were supported by local princes and not con- trolled by religious sects; and science consequently flourished. In England, on the contrary, Oxford and Cambridge were rigidly controlled by the anglican church until 1871; until then admission was only to anglicans, and all senior staff positions were held by anglican divines. A vow of celibacy for students and teachers was abandoned only in 1878. Even after 1871 individual colleges could, and most did, refuse admission to jews, catholics and protestant dis- senters. Thus presbyterians stayed in Scotland where science and engineering flourished; but science stag- nated in England, where only brilliant individuals such as Darwin could advance without aid from the universities. According to C. D. Darlington (The Evolution of Man and Society, London, George Allen and Unwin, 1969) the same attitude was still common well into the twentieth century; but new universities, and new colleges, free from the reli- gious system, finally allowed great advances. Independent achievements were ill-received by self-satisfied members of the establishment. When Beatrix Potter, a remarkable self-trained naturalist, discovered the lichen symbiosis she was ridiculed by the learned men of Kew: Not only was she a woman but she was not a university graduate; ergo she could not discover anything. Luckily for generations of 2001 children, she went on to fame and fortune with her animal stories; and the lichen symbiosis was soon rediscovered in Germany. Professional biology in the United States seems to have developed gradually, perhaps from 1870 to 1890. In Canada, with its small and scattered popula- tion, professional training seems to have developed mainly between 1900 and 1910, with delays certain- ly due in large part to the stagnation of science in England. But in most countries the numbers of pro- fessional biologists seem to have been few until about 1920; and progress was understandably slow. However, even in recent years we do not see one breakthrough leading directly to the next. Why is this so? Are biologists as a whole too slow in their thinking to take faster steps? I do not think so. Consider Stephen Hawking, the theoretical physicist believed by many of his colleagues to have perhaps the finest intellect of the twentieth century. In dis- cussing Hawking’s work, John Boslough (1985. Stephen Hawking’s Universe. Win. Morris, New York) shows that, despite his many contributions, his ideas do not come in a steady stream. After a long period a new idea is born (it may be in the middle of the night) for no obvious reason; then he and his col- leagues again pursue the origin of the universe a stage further. Boslough quotes Hawking as saying: “T think we’ll come to the unifying theory within the next two decades, probably in a series of small steps. But you know, once we find it, it will rather taken the fun out of theoretical physics.” Fortunately biolo- gy is so much more complex that we shall surely not lose the fun of it for a long time; but still work hope- fully toward that end. Perhaps we lesser mortals, such as botanists and mycologists, need not be too ashamed of our halting progress; but surely we should try to understand its causes, and then perhaps we may find some reme- dies. When a theoretical physicist gets a new idea it presumably is born from a particular association of prior ideas. It is like rotating a kaleidoscope to get a new pattern. When a field biologist (i.e. a true natu- ralist) gets a new idea, or a new understanding, it is often derived from a newly observed structure or mechanism associated with an old idea, or a familiar structure associated with a new idea. When I saw Tiarella capsules hit by drops from the canopy I knew at once what was happening because the action of falling drops on splash cups was already in my mind. Would I have understood what was happening if I had seen the plants without having been involved with splash-cups? Well, certainly not immediately; but obviously the mechanism had to be seen in the field to be understood. No one, including myself, who had studied herbarium specimens of Tiarella seems to have suspected a function for those “ridicu- lous” capsules. A character can be “ridiculous” only when we do not understand it. The poet disparages SAVILE: EVOLUTION OF A NATURALIST 15 the ragged fingers of a crow; but the crow is large enough to need the extra lift from its wing-tip slots for economical travel even in level flight. Recognizing the value of field study to clarify a function, let us look at the lamentably slow growth in our understanding of the rust fungi, which was cer- tainly not entirely due to a paucity of students. The history, up to 1928, is told in some detail by J.C. Arthur et al. (The Plant Rusts, Uredinales, John Wiley, New York, 1929), with pertinent references. Anton de Bary, an intellectual giant, understood the morphology and parasitism of the rusts as early as 1853; but, despite insistence by farmers of the abun- dance of wheat stem rust near barberries, he did not countenance the identity of the aecia on barberry with the uredinia and telia on wheat for ten years: The con- cept of heteroecism was simply too preposterous for an educated man to accept. (Here is a parallel of Beatrix Potter and the lichens: We must have open eyes and minds). He finally proved the connection by reciprocal cultures in 1864 and 1865. He then also connected up some other heteroecious species. De Bary’s unwillingness to recognize heteroecism in the rusts is curious, for he already knew of the phenom- enon in some animal parasites. Although the pycnia were suspected of being spermogonia by Tulasne as early as 1851, there seems to have been little or no progress in understanding them for many years, despite various cytological studies. They are, in fact, a combination of spermogonia, receptive hyphae and a nectary, and are a unique structure whose significance is overlooked by a few workers who call the whole organ a spermogonium. Because the spores did not germinate by germ tube, they were held to be func- tionless. Surely by 1900 or thereabouts someone should have suspected that a functionless organ would not persist for millions of years in genus after genus. Finally J.H. Craigie (Nature 120: 116-117, and 765-767, 1927) demonstrated that insects feeding on the pycnial nectar transferred pycniospores to pycnia of the opposite mating strain, resulting in the forma- tion of dikaryotic aecia. Why did this step have to take three-quarters of a century? Pycnia are prominent in many rusts. I recall in my student days, a few years after Craigie’s work, watching assorted insects feed- ing on the conspicuous pycnia of a Gymnosporangium on Crataegus leaves. Had no naturalists seen insects feeding on pycnial nectar during all that time? How was the understanding finally achieved? Craigie’s work was done at the then newly founded Rust Research Laboratory at Winnipeg. Craigie must certainly have been influenced by A. H.R. Buller, professor of botany at the University of Manitoba, who, with his intense sense of curiosity, was always interested in the work of other investigators. Staff at the Rust Lab were inclined to give full credit to Craigie, with the inference that Buller was merely in the way. However, Harold Brodie, who was Buller’s 376 student at the time, told me many years later that Buller saw flies visiting the nectar on the inoculated barberry plants in the greenhouse, guessed what was happening, and told Craigie to grow seedlings in insect-proof cages and do controlled nectar transfers. After the successful experiments had been completed he urged Craigie to write it up for Nature. According to Brodie, who was working in Buller’s office at the moment, the first draft of the note was so unsatisfac- tory that Buller practically dictated the final version. They have all gone now, Brodie being the last, and we shall never know the truth in every detail. As one who several times saw Buller in action and who later worked under Craigie, I can well believe that in his enthusiasm Buller may have told Craigie what to do (even if Craigie was already doing it). Harold Brodie was meticulous in giving credit, and I therefore believe his interpretation of the scene in Buller’s office. However, Craigie was a canny Scot, not to be rushed in any action. He was not a ready writer (or speaker), but would keep polishing a statement and changing adjectives until he was finally satisfied. I recall a memorandum that went between him and me for a week before he released it (little changed from my original, but certainly no worse). Therefore, I would not expect the draft that Buller saw to have been wholly satisfactory: It may have been only a first draft, which Craigie would have polished with- out help (eventually). Buller, in contrast, seems to have been a very ready writer who never had to revise extensively. Buller had tremendous enthusi- asm and considerable imagination; but he seems to have been inclined to dictate experiments to others in preference to doing them himself. (Years later he visualized the splash dispersal in Cyathus and conned Brodie into finding specimens and running the tests that provided the splash-cup mechanism). Next A. M. Brown (Nature 130: 177, 1932) demon- strated that uredinial cultures of the autoecious Puccinia helianthi could dikaryotize isolated pycnial infections by nuclear transfer following hyphal fusion. In 1939 (American Journal of Botany. 26: 585-609). I demonstrated cytologically the fusion of pycniospores to receptive hyphae, including a nucleus in transit and the clear circles left on the wall of the receptive hyphae after the pycniospores fell away. Using advanced staining procedures, including the Feulgen method, I was able to interpret realistically some of the improbable figures that had added mystery to rust cytology (see section 6). Incidentally, I showed that the so-called Blackman and Christman fusions in the aecial fundament have no taxonomic distinction; both types and intermediates may occur in a single aecium, and they represent the shortest available route for an introduced nucleus initiating a dikaryon. In the meantime, it was widely assumed that rusts without pycnia were evolutionary dead ends, inca- pable of recombination. One of the conspicuous THE CANADIAN FIELD-NATURALIST Vol. 115 results of our extensive post-war field program, with its good geographic coverage, was the finding that such rusts are usually morphologically uniform over a widely occupied area; but that geographically iso- lated populations tended to differ from one another although each was homogeneous. It was clear that genetic recombination was operating in such popula- tions: they were not fragmenting like a fully apomic- tic dandelion. Then I demonstrated a morphological cline between two subspecies of Puccinia crucifer- arum, and it became clear that any two adjacent dikaryotic hyphae that are genetically distinct must trade nuclei when they meet (Mycologia 56: 240-248, 1964). Finally, the high-arctic Puccinia poae-nemoralis ssp. hyparctica, isolated by the onset of the Wisconsin glaciation, was seen to differ very uniformly from the parental subspecies in the size, pigmentation, wall thickness and sculpturing of its urediniospores and in its host range (Arctic Adap- tations in Plants. Canada Department of Agriculture Monograph number 6, Ottawa, 1972). Because P. poae-nemoralis never produces telia in the arctic, meiosis cannot be involved; and the accumulation of all the mutations adapting ssp. hyparctica to an extremely arid climate can only be accounted for by parasexual recombination, following nuclear transfer. The jumping of rusts to new hosts also involves their genetic make-up. In 1954, collecting in the mossy coastal forest of southwestern British Colum- bia, Calder and I repeatedly found Pyrola spp. and Goodyera spp. growing in close association. Pyrola often bore uredinia of Pucciniastrum pyrolae; and Goodyera occasionally bore pustules of Uredo goody- erae, which is excluded from Pucciniastrum because it lacks telia, but which has urediniospores only slight- ly distinct from those of P. pyrolae..Except for the scarcely separable U. ishikariensis in Japan this is the only such rust on a monocotyledon. Clearly it arose by a jump from contiguous Pucciniastrum pyrolae. Although I saw, in the humid coast forest, what had happened, I did not see its full significance and thought it a rare freak. Twelve years later, when revis- ing some rusts of Scrophulariaceae, I recognized another unmistakable jump, by Puccinia palmeri on Penstemon to Pedicularis, with the evolution of Puc- cinia rufescens. Both rusts have the relatively uncom- mon O, I, II life cycle with repeating aecia, and sev- eral conspicuous morphological resemblances; but P. rufescens has clearly rugose teliospores, in contrast to the smooth or faintly roughened spores of P. palmeri, and is certainly the derived species. With field experi- ence in the Cordillera I saw what had happened. Penstemon and the cordilleran Pedicularis have strong geographic and ecological overlap; and, although Penstemon is clearly a modern genus, the cordilleran Pedicularis are even more modern, having evidently originated from a late Tertiary asiatic immi- grant and radiated mainly in the Pleistocene when cli- 2001 matic fluctuations tended to fragment populations and stimulate speciation. The conditions that promote a successful jump evidently include strong ecogeo- graphic overlap, a young and genetically diverse parental rust (with a large gene pool), and a young and genetically diverse potential host (younger than the parental host). With these conditions the chance of compatible genomes meeting is maximized. I now realize that jumps have occurred abundantly in the evolution of the rusts. At last it was seen why Dietel’s early observation that rusts and hosts reflect each other’s ages of origin is valid. Why was it not general- ly understood for 67 years? Well, it should have been understood after a mere 63 years, but ignorance inter- vened. A fully documented account of the mechanism and supporting taxonomy was submitted in 1966 to Canadian Journal of Botany. Both editor and review- er accepted the taxonomy but insisted that the discus- sion be purged (editing by axe!). Eventually it was enthusiastically accepted for Nova Hedwigia (24: 369-392, 1968 [1969]). This journal deals with all cryptogams but not spermatophytes, so the paper was not seen by many phanerogamists. The jump mecha- nism was re-explained in a symposium paper in Quarterly Review of Biology (46: 211-218, 1971) where everyone saw it. (I stopped counting reprint requests at about 800). When the mechanism was finally explained it was through my field observations combined with detailed microscopic work. This major step in understanding rust evolution was possible because I was an experienced field naturalist, which is surely how most evolutionary advances are achieved. 12. Progress can be speeded The answers to problems relating to rust biology, and many similar topics, lie in the need for copious field work, but field work with the eyes and mind wide open. Also the collector should either be involved with the laboratory studies or be in close contact with that person. In this way cause and effect are most easily related. In my nine full seasons of field work between 1949 and 1962, I doubled as botanist and mycologist both by inclination and by necessity; and the value of this broad approach was soon very clear. At three arctic sites where my stay covered the nesting period (Chesterfield, Isachsen and Hazen Camp), I was able to run a breeding bird census, which gives an approximate measure of biomass productivity. A botanist patrolling an area devoid of tall plants can locate nearly all nests; but such counts are impossi- ble in wooded country. If we go into the field with the responsibility to collect and study ecologically all plants and fungi, and develop an intense interest in them, we can scarcely avoid discovering new information about many of them. Apart from the per- sonal satisfaction that our observations give us, this maximum yield of information justifies the substan- SAVILE: EVOLUTION OF A NATURALIST S77 tial cost of keeping a party in the field. Once we have fair coverage of an area, specialized collecting, to complete information in a particular field of study, is justified; but to go into an inadequately studied region with the blinkers on, neglecting all but one small group of organisms, may be worse than use- less, making it difficult to promote an adequate attack on that region later. The wealth of new biological information secured in the years 1949 to 1962 (after which funding for biological exploration was drastically reduced) must greatly exceed that of any previous period in Canada. Personal field and microscopic study explained how a rust jumped from Penstemon to Pedicularis (see section 11). Because he had described so many rusts J.C. Arthur received many specimens for study from other collectors; and his field observations became increasingly limited to the vicinity of Purdue University. Studies by others at arctic or alpine tree- line have shown the limitations of heteroecious rusts. When Holway sent Arthur the type collection of Puc- cinia praegracilis on Agrostis thurberiana, collected near treeline, he sent with it a note stating that it “grew adjacent to the Habenaria aecidium, and no where else.” Arthur named the rust but later buried it in P. coronata because he did not appreciate Hol- way’s warning; but Holway was absolutely right, as I have shown repeatedly. Near treeline these exclusive associations are more reliable than artificial rust cul- tures in which contaminations do occur. If puzzles are to be solved promptly the collector must include full information and the identifier must not ignore it. Most of my discoveries in various organisms result- ed from field observations made with an open mind conditioned by previous experiences. Field work and microscopic study demonstrated gene flow and geo- graphic races in Puccinia cruciferarum, which lacks pycnia. I discovered P. poeae-sudeticae ssp. hyparcti- ca in arid Hazen Valley because, from experience in arid southern British Columbia, I knew enough to look in the axils of the grass leaves where dew lingers; and its distinctness confirmed the occurrence of para- sexual recombination in a rust lacking teliospores (see section 9). Inspired by H. J. Brodie’s seminal paper on splash- caps (Canadian Journal of Botany 29: 224-234, 1951), I demonstrated splash-cups in Chrysosplenium and Mitella (Science 117: 250-251, 1953), but was puzzled by the strange capsules of Tiarella. Soon after my note appeared, as I stood in the coast forest of southwest British Columbia in a drizzle, I saw cap- sules of Tiarella trifoliata flicker as drops from the canopy hit them, and the operation of this elegant springboard was revealed (see section 9: Fruits of Field Work). Five years later, seeing Chrysosplenium rosendahlii spreading freely in marshes on Somerset Island, with only nodding grasses and sedges as a 378 canopy, I was puzzled at the thought of drops falling only half a meter being effective dispersal agents. Many years later H. N. Hayhoe derived an equation demonstrating the effectiveness of large drops falling very short distances (information important to plant pathologists) and our understanding of splash-cups and springboards was finally nearly complete (section 9). Among observations on plants apparently originat- ing with me: Alopecurus alpinus and Papaver radica- tum (s. lat.) have such low temperature tolerances that even in the high arctic their growth is limited mainly by aridity. Field observations disproved claims that Poa glauca is fully apomictic, when it was found freely crossing and backcrossing with Poa hartzii at Hazen Camp, and also hybridizing elsewhere with a member of the Poa arctica complex. Other strongly self-fertile plants were shown to outcross on occa- sion; e.g. hybrids between Stellaria edwardsii and S. laeta were found at Hazen Camp. I showed that Saxifraga oppositifolia was fully self-fertile at Isachsen, two distinct biotypes being present but no intermediates; sustained observations showed that, about two days after the stigmas became receptive, elongation and curvature of the filaments brought the anthers approximately into contact with the stigmas. (The only abundant potential pollinators were chi- ronomid midges, which prefer white flowers and seemed to feed mostly at Stellaria and Cerastium.) In contrast, at Hazen Camp, which is much warmer in summer than Isachsen, two bumble-bees (Bombus polaris and B. hyperboreus) were present, and it was difficult to find any two plants of Saxifraga oppositi- folia that were convincingly identical. At least three biotypes seem to have been present originally and the bees had mixed them freely. Later P.G. Kevan (In- sect pollination of high-arctic flowers. Journal of Ecology 60: 831-847, 1972) showed that S. oppositi- folia at Hazen Camp is about 90% self-sterile. Detailed observations of plant growth at Isachsen, physical conditions, and the total ranges of plants allowed me to interpret the late glacial and post- glacial history of the northwestern Queen Elizabeth Islands. The Islands were lightly snow- and ice- covered; in the postglacial hypsithermal interval they were well vegetated; but with increasing cold many plants were eliminated, leaving broken distribution patterns. The islands were not, as once suggested, a glacial refugium (see section 8). Thus the glacial his- tory of the region was completely reinterpreted. My observations of the Peary Caribou (Rangifer tarandus pearyi on Ellef Ringnes Island explained why, in defiance of Bergmann’s rule, this is the smallest, rather than the largest, race. Two distantly separated family groups (buck, doe and one fawn) each seemed, from their tracks, to patrol some 700-800 km/?, clipping off plant tops as they moved. As these regions usually have less than 1% of the ground covered by very small plants, continuous THE CANADIAN FIELD-NATURALIST Voli 15 movement of the animals is necessary for their sur- vival; and survival is best assured by small body size, which allows reproduction with minimal food intake. The Red-throated Loon (Gavia stellata) is the small- est member of its genus and also has the most northerly limits. Here small body size functions by allowing take-off and landing in relatively small ponds that become ice-free promptly. Large lakes, such as Lake Hazen, are used for fishing, but ice- shove keeps the shores bare of vegetation and impos- sible for nesting (Savile. A naturalist looks at arctic adaptations. Jn Evolution Today, pages 47-53, G.G.E. Scudder and J. L. Reveal Editors. Procedings of the 2nd International Congress of Systematic and Evolutionary Biology, Hunt Institute, Pittsburgh). I spent nine more or less full seasons (1'/)-4 months) in the field between 1949 and 1962, with shorter periods in two other years. This abundant exposure to the living world obviously contributed substantially to most of my success in interpreting biological (and occasionally physical) phenomena. It should be equally obvious that exposure to the biota cannot guarantee success. One must keep the eyes and the mind wide open. The unperceptive collector can bale botanical year after year and discover noth- ing — in fact not even see that unsolved problems exist. To break new ground the field biologist must either have an innate sense of curiosity or have been stimulated by associates (I was twice blest), prefer- ably both, for it generally pays to look at several aspects of a problem. The six blind men could not adequately describe the elephant because none would walk round it. Unusual characters do not evolve randomly; nor do they exist simply to aid taxonomists. If we can recognize their functions we may get clues to the evolution and paleoecology of the organisms. Some- times the function is obscure except under particular conditions. The team approach may then speed up recognition; but all team members must obviously be alert to the problem. Seed dispersal is a critical prob- lem in most flowering plants. Years of observation in Saxifragaceae (senso stricto), a clearly natural group, revealed a surprising variety of dispersal methods, both local and long-range (D. B.O. Savile. Evolution and biogeography of Saxifragaceae with guidance from their rust parasites. Annals of the Missouri Botanical Garden 62: 354—361, 1975). It is worth noting that all these mechanisms have also evolved in other plant groups, reminding us that in important problems there may be repetition of one solution as well as multiple solutions. In the large and ecologically diverse genus Carex the fruits may be scattered by wind, lodge in the fur or feathers of passing animals, float across water by virtue of blad- dery perigynia, or even be ingested by birds as in C. aurea with fleshy, sweet and bright yellow perigynia (Savile, Botanical Review 45: 488, 1979). The Asi- 2001 atic C. baccans Hara similarly has fleshy but red- purple perigynia. As in flowering plants, dispersal in fungi is criti- cally important, involving many modifications. A long-cycled rust initially has two dispersing spore states, aeciospores and urediniospores, which might seem ample; yet in the three advanced rust families, with pedicellate teliospores, diasporic teliospores have evolved in at least 17 genera by at least 9 meth- ods. Spore release is usually accompanied by newly adaptive changes: nearly uniform rather than apical wall thickening, broadened and shortened spores, sculpturing of spore walls, and a tendency for germ pores to drift from the apical or septal position. In the big genus Puccinia (and the related Uromyces) I was able to recognize the attainment of deciduous teliospores in more than 30 lineages by six methods of release (Savile. Evolution of the rust fungi (Ure- dinales) as reflected by their ecological problems. Evolutionary Biology 9: 137-207, 1976). Diasporic teliospores must be particularly important in short- cycled rusts; but they have also evolved in 47 long- cycled Puccinia or Uromyces in North America, and they must therefore be strongly adaptive even when aecia and uredinia are present. J.C. Arthur (Manual of the Rusts in United States and Canada. Purdue Research Foundation, Lafayette, Indiana, 1934) divided Puccinia into two sections: Eupuccinia with firm and Bullaria with fragile pedicels. Even in 1954 (D. B.O. Savile. Cellular mechanics, taxonomy and evolution in the Uredinales and Ustilaginales. Mycologia 46: 736-761) the functions of the correlat- ed changes in diasporic teliospores were clear and I identified several lineages of Puccinia and Uromyces that occur in both of Arthur’s sections. His sections are thus completely unnatural and taxonomically mis- leading. Indeed some species have incipiently dias- poric teliospores and defy disposition to section. Arthur apparently thought more in terms of a conve- nient pigeon-holing system, comparable to that of Engler and Prantl for the flowering plants, than of a classification reflecting active evolution in the rusts. 13. Truth and Beauty in Biological Research What drives the research biologist? Scientists expect to make an adequate living from their work; but if money were their main objective they would be in other professions. Recognition by one’s peers, or rarely by the general public, may be satisfactory, but is a result of rather than an impulse to research. Surely we are driven largely by the excitement of discovery and the ultimate delight of establishing a truth built up from a series of observations. Basically we seek to establish facts and, from them, to estab- lish the truth of a system. But truth, for a scientist, goes beyond the bald statement of truth vs. falsity sought in a court of law. Establishing a scientific truth is a matter of great satisfaction, the true picture becoming a thing of beauty. The search for, and SAVILE: EVOLUTION OF A NATURALIST 8 demonstration of, truth must surely be the driving force behind all dedicated scientists, whether or not they equate truth with beauty; but the imaginative worker must generally recognize beauty. Although Keats’ statement — “Beauty is truth, truth beauty” — is surely the most famous equation of truth and beauty in English, Chandresekar’s essay (S. Chandresekhar. Ph. 4, Beauty and the quest for beauty in science, in Truth and Beauty: Aesthetics and Motivations in Science. University of Chicago Press, Chicago and London, 1987) shows us that many people down the centuries have discussed the same theme. The concept of beauty in truth is surely an impor- tant stimulus to the research biologist. Much of his time in the field or back in the lab is taken up by rou- tine collecting, processing or identification of speci- mens, leavened by unexpected observations or the finding of unexpected species. But once in a while he discovers the function of a character or finds the specimen that explains earlier observations. How wonderful life suddenly is! Such revelations of beau- ty do not occur very often (just as well, for they can be pretty intoxicating), but they break the routine. The discoveries need not be major, but they clarify a problem, as two examples show. (1) On a drizzly day in southwestern British Columbia. I saw a Tiarella capsule flicker under a falling drop; and suddenly its strange form was revealed as a function- al and beautifully engineered springboard (Section 9); and it was clear that Mitella and Tiarella diverged in humid forests from a heucheroid ances- tor, each evolving an effective but separate splash dispersal mechanism. (2) When Calder and I were revising the subspecies of Saxifraga punctata, we found plants near Prince Rupert that possessed char- acters of two mainland subspecies but also some of an unknown plant. Suspecting these characters to be from a Queen Charlotte Islands race, we advanced our visit to these Islands. I can think of few more rewarding moments than when, after struggling up Tan Mountain, we first saw S. punctata ssp. carlot- tae looking exactly as we anticipated. It has been suggested that its aesthetic value may justify a theory that is shown to disregard facts; but surely an honest scientist cannot support such an idea. If a theory is false it is so defaced that, to me, it immediately loses any beauty. However, it is sadly true that some workers find it difficult or impossible to discard an old belief despite accumulating evi- dence against it. Although it was plain even in 1954, partly from parasite data, that Liliales are younger than Poaceae and Cyperaceae, a few authors have even recently indicated them to be older (e.g. A. L. Takhtajan. Outline of the classification of flowering plants, oragnoliophyta. Botanical Review 46: 225-359, 1980; R. M. T. Dahlgren and H. T. Clifford. The Monocotyledons: A Comparative Study. Aca- demic Press, London, 1982). 380 Biology differs from the physical sciences in sev- eral respects. In physical sciences simple laws are applicable because the systems are simple. In biolo- gy, as Ernst Mayr has shown more than once, the systems are infinitely complex: All the molecules of a chemical compound in the universe are identical; but in sexually reproducing organisms no two indi- viduals, except identical twins, are ever the same. The physical scientist, often after years of labour, produces laws or principles that are applicable because the systems are simple. Chandresekhar indi- cates that most physical scientists make their major contributions early in life; and he contrasts them with great poets, writers and musicians, who often produce very fine works late in life. It is notable that many biologists show a pattern similar to that of workers in the arts. Charles Darwin published The Origin of Species (his first major work) at age 50; and his other volumes appeared during the next 24 years. Ernst Mayr, born in 1904 and arguably the greatest evolutionary biologist of this century, has been publishing on evolutionary topics continuously since 1940, with important books appearing in 1942, 1963, 1970, 1976, 1982 and 1988. Many other biolo- gists of my acquaintance have continued in produc- tive research to formal retirement and beyond, as long as their health permitted. A biologist’s main contributions may all be made relatively late in life, as was true of Darwin, simply because accumulating countless small observations and fitting them into a useful and informative struc- ture inevitably takes a long time. This tendency reflects the complexity of biological systems. The young field biologist collects specimens for taxonom- ic study, and records their geographic range, general ecology, habitat (with orientation and altitude where pertinent), and associations. Eventually all these bits of information will lead to a broader understanding of the biology of the organisms. Thus we may often pass most of our careers before we are prepared to present a complete picture of a topic. But, because we are dealing with living and genetically variable organisms, the picture is seldom really complete, and in later years we, or others, will add to and modify it. There can be no mathematical proof that the picture is correct, for the pieces of the puzzle may change in shape and numbers at any time. What we aim at is the simplest explanation that conforms with all the data. If the data are abundant the most parsimonious solu- tion is nearly always correct; but we are dealing with probability rather than mathematical proof. As time passes we generally discover additional pieces to fit into the puzzle; and the probability of a correct solu- tion finally becomes enormous and can be accepted. About 1973 I started to assemble observations, which I had been accumulating for about 30 years, into a paper on evolution and ecology of the rusts (D. B.O. Savile. Evolution of the rust fungi, Ure- dinales, as reflected by their ecological problems. THE CANADIAN FIELD-NATURALIST Vol. 115 Evolutionary Biology 9: 137-207, 1976). When I finished it late in 1975, I hoped that the story was reasonably complete. Scanning my annotated copy I see considerable additional information from more recent publications, but nothing to change the main conclusions. Thus the timing was evidently appropri- ate. If I had attempted it many years earlier it would have been very inadequate; and if I had delayed it for more than 10 years I probably could not have achieved it, because of other involvements. When Arthur Cronquist suggested, in 1977, that I write an article for the Botanical Review on fungi as aids in plant classification, I had no illusions about a complete treatment, for much information is buried in papers whose titles and abstracts contain no appropri- ate key words. However, by postponing the attempt until after a related symposium at Uppsala in 1978, I was able to present considerable information in this field. I also made a few discoveries; one that surprised and excited me was that three rust lineages that arose in lower Cyperaceae reached their greatest morpho- logical advancement not only in Carex (as expected) but also in Juncus or Luzula. As the lineages are based on several correlated characters, and the trends in morphological advancement are widely recognized in other rusts, the age and relationship of Juncaceae, as an offshoot of Cyperaceae, seemed incontrovertible. As Juncaceae have been classically accepted as ances- tral to Cyperaceae, the indication of their advanced position is understandably unpopular, and was dispar- aged by two recent authors who, however, attempted no other explanation of the data. The multiple mor- phological characters of each rust lineage make it extremely unlikely that random similarity would occur even once, the probability being surely at least 100:1 against duplication. That it should occur three times is really unthinkable. Elementary probability tells us that in repeated similarities the odds are multi- plied. Thus the odds against random duplication in all the lineages are at least 1 000 000:1. However, in a recent restudy of these families (D.B.O. Savile. Relationships of Poaceae, Cyperaceae and Juncaceae reflected by their fungal parasites. Canadian Journal of Botany 68: 731-734.) I found that a smut, Entor- rhiza caricicola, occurs on Eleocharis gracilis, Carex spp. and Juncus spp., further indicating and advanced position for Juncus and relationship with Carex. The paper dealing with this and related smuts (J. M. Fineran. A taxonomic revision of the genus Entor- rhiza C. Weber. Nova Hedwigia 30: 1-68, 1978, received in Ottawa 9 April 1979) appeared too late to be incorporated in my Botanical Review paper. However, the host range of E. caricicola causes the odds against random similarity in the parasites Junca- ceae and Carex to be perhaps 100 000 000:1. It would probably be hard to find a biological conclusion closer to mathematical proof than that! Received 11 January 2001 Book Reviews ZOOLOGY A Field Guide to the Birds of China By John MacKinnon and Karen Phillipps. 2000. Oxford University Press, New York, Don Mills, Ontario 586 pp., illus. Ornithologically isolated for decades, China now welcomes a growing trickle of birdwatchers. With a species list of over 1300, 100 plus endemics and some of the worlds’ most impressive birds, including 62 species of pheasant and nine cranes, it’s not hard to see why birders flock to China. But China has long suffered from the lack of a good field guide. This situation has now been rectified by the publica- tion of this niche-filling book. The Birds of China follows a standard format, with a short introductory section to familiarize the reader with the area covered (which includes Taiwan and Hong Kong), the history of ornithology in China, and key conservation challenges, including a depressingly long list of endangered species. For the non-sinologists, the political map inside the back- cover is critical to understanding species distribu- tions. A particularly useful feature is a listing of birdwatching and conservation organizations. The 128 colour plates are grouped together at the front of the book, followed by the species accounts. Each species account includes a brief description, and notes on voice, range, distribution and status, and habits. A Chinese version is planned and the current edition features Chinese names. The text is terse and to the point. The species accounts will generally be sufficient to identify most birds seen, including difficult groups such as snipe, but suffer from an inconsistent use of vague terms such as the “smallish” Common Tern at 35 cm and the “medium-sized” Black-napped Tern at 33 cm. Some of the more complex species will remain a mystery if this book is your only reference, for example, immature argentatus-type gulls are declared unidentifiable. Some of the descriptions are overly simplistic; one would be foolhardy to identify a female rufous-tailed Robin on the basis of its rufous tail alone as several similar species share this feature. That said, there is plenty of great informa- tion in this book that is bound to come in handy some day: how many people know that Asiatic House Martin has black wing linings that differenti- ate it from its European counterpart? What about the plates? As is often the case in guides covering mega-diversity regions, the quality is variable. Some of the passerine plates are very good; the robins, buntings, pipits, and larks are excellent, a fact which should make sorting out the little brown jobs that much easier. The quality of the non-passerines plates is generally not as good; the gulls are remarkably poor. Nonetheless, many of the paintings show important fine details such as the dif- ference in primary projection between Pacific and American Golden Plover. While a great attempt has been made to illustrate major forms and plumages there are omissions, for example, basic and alternate Teminck’s Stints are illustrated but not the immature plumage. Some paintings, and in particular the large water- fowl, have suffered through excessive shrinking caused in scaling them down to fit the book’s format. Generally the colour separation is good, although in the review copy the green tones on one of the leaf warbler plates (plate 99) is much too strong. The arrangement of species on the plates is sensible and the paintings are numbered, making it easy to figure out which species is which. The facing page to the plates feature useful and easy to read colour range maps, however, this arrangement comes at the expense of the helpful identification pointers that are a feature of other major field guides. Some odd birds are left out the plates, including Ibisbill which receives only a line drawing, surely one of the coun- try’s most sought-after birds. One of the strongest features of the book is the fact that it actually is a field guide. While too hefty to fit in most people’s pockets, at 20 x 14cm, and 4cm deep it can easily be tossed into a backback, and it benefits from easy to read type and a generally reader friendly format. The binding and covers are rugged and will probably stand up to usual field use. Some of the printing was blurred in my review copy. Unfortunately, this book suffers from serious proof-reading problems; it is littered with typograph- ical errors, some of which will cause confusion. For example, the phalarope range maps are mislabelled, the Pied and Collared falconet are reversed, and the text incorrectly indicates that there is a painting of Southern Grey Shrike. Additionally, some of the range maps are suspect, for example, Crested Shelduck is shown with a far more extensive range in northeasern China than the handful of records and questions over its continued survival would warrant: a question mark in the general area of the historical sightings would have been preferable. These mis- 381 382 takes are frequent enough that some second guessing is required; hopefully they will be addressed in a future addition. The taxonomy follows Sibley and Munroe’s 1990 Distribution and Taxonomy of Birds of the World. While it is a safe bet that the taxonomy of the relative- ly poorly known Chinese avifauna will be in a state of flux for generations to come, the insertion into this guide of a polemic on the merits of the phylogenic ver- sus the biological species concept is a bit misplaced; perhaps better left for the more arcane journals. This book will inevitably be compared with recently released volumes covering other mega- diversity regions such as India and Kenya, and will suffer in the comparison. Part of this is due to the authors’ commitment to producing a book that is tru- ely a field guide: in scaling back the text the authors have lost the sort of detail that some readers will be looking for. Nonethless, the bottom line is that the authors have produced an invaluable book that fills a major gap. This is not a state-of-the-art book, but it is an important birding and conservation tool; this is THE CANADIAN FIELD-NATURALIST Vol. 115 the first time some species are illustrated in a popu- lar field guide. Indeed, here for the first time I have seen an illustration of one species on my life list, Chestnut Bulbul. If you are planning to birdwatch in China, this book is a must. If you dream of finding asiatic vagrants on your next trip to Alaska, bring this book along. If you are a avid birding book collecter who hasn’t been able to sleep nights worrying about the lack of guides for the oriental region, you should buy this book. References Zimmerman, Dale A., Donald A. Turner, and David J. Pearson. 1997. Birds of Kenya and Northern Tanzania, Princeton University Press, New Jersey. Grimmet, Richard, Carol Inskipp, Tim Inskipp. 1999. A Guide to the Birds of India, Pakistan, Nepal, Bangladesh, Bhutan, Sri Lanka and the Maldives, Princeton University Press, Princeton, New Jersey. MARK GAWN 1354 Viking Drive, Ottawa, Ontario K1V 7J6 Canada The California Condor: A Saga of Natural History and Conservation By Noel Snyder and Helen Snyder. 2000. Academic Press, San Diego. 410 pp., 124 colour plates, 4 maps, 8 graphs, 23 tables. U.S. $29.95 This large, beautifully-illustrated, fact-filled book introduces us to heroic people and intersperses tri- umph with tragedy. It is the well-researched history of a remarkable bird and a superb conservation story — of a large bird and all the people who studied it. Only when a material object like a British Guiana penny black postage stamp, or a bird such as the California Condor, becomes extremely rare, are we willing to spend large sums of money. In the case of the California Condor, the efforts were perilously close to being, at first, too little. More recently, the expenditure of more than a million dollars a year has been almost too late. The California Condor’s wingspan of nearly 3 m is the largest of any North American bird species, and allows it to soar almost continuously. Archaeo- logical evidence includes bones of 63 condors at one human site in Oregon, well north of its 20-century range. Numerous aboriginal tribes employed condors in ritual sacrifice, perhaps one per village per year in the San Diego region. Others were shot for their quills. The condor declined further, beginning in the 1870s, when the use of strychnine and later cyanide of potassium were used to kill mammalian predators. Collectors also took a heavy toll, including at least 20 birds and 7 eggs taken in 1897-98 alone. Killing or collecting condors or their eggs was made illegal in California in 1905. William L. Finley made the first serious study of condors, found his first nest in 1906, and took 250 photographs. Finley considered that condors were extremely tolerant of humans. Carl Koford began his condor studies in 1939, identified their main nesting and roosting areas and estimated that by 1950 only 60 condors remained; in retrospect, the Snyders esti- mate that there were then about 150. Koford showed that young fledged at about 145 days of age, and then were dependent on parent feeding for another 6 months. He claimed that condors were incredibly shy and sensitive, completely intolerant of the presence of humans, and recommended that a large sanctuary be established to protect them. A full-time condor warden, John Borneman, was hired by the National Audubon Society (NAS) in 1965. The condor received protection under the Endangered Species Acts in 1966 and 1973. Fred Sibley began four years of condor studies for the U.S. Fish and Wildlife Service (USF&WS) in 1966; with his “superhuman” endurance, Sibley roped down every likely-looking cliff to find both active and many former nest sites. The 1966 condor count showed at least 51 still alive, but detractors claimed, without foundation, that Sibley’s close approaches to nests had caused the species to cease reproduction. In 1968, Sanford Wilbur replaced Sibley and began putting out carcasses as supplementary food for condors. Wilbur led a Recovery Team which pro- 2001 duced a Recovery Plan in 1974 and a Contingency Plan in 1976. Wilbur also supported the appointment of a joint panel of the American Ornithologists’ Union and the NAS, which reported in 1978, and was endorsed by the Fish and Wildlife Service. The NAS successfully lobbied Congress to expand the condor conservation effort in 1979. Noel Snyder of USF&WS and John Ogden of NAS developed a new intensive field effort in 1980. Meanwhile, Friends of the Earth opposed any “hands-on” activities and branded radiotelemetry as “mutilative biology.” The unfortunate death of a struggling condor chick strengthened their argument. In 1981 an ingenious flight photograph file was developed for each condor, using missing feathers to identify each individual. The wild condor population numbered 21 in 1982, 14 adults and 7 immatures. It then dropped precipitously to 19 in 1983, 15 in 1984, and only 9 in 1985. In October 1982 the first free- flying condor was caught safely in a cannon net, blood samples taken and a radio-transmitter applied; eight other wild condors received radios, seven of them in 1984. Snyder found surprises. Older condor nestlings were fed every 10 hours on average, though 13 times in 605 days of observations the chick went two days without being fed. An individual adult might incu- bate the single egg for up to 10 days before accept- ing relief from the mate. Pairs changed nest sites each year, moving an average of 3.8 km to the new nest. Individual condors ranged so widely that they were familiar with the entire range of the population; an individual might fly up to 225 km in one day. Snyder quickly realized that the great wariness of the Species is a myth, perpetuated in part to support the sanctuary ideal. In spite of occasional predation of the single con- dor egg by Common Ravens, and one egg that rolled out of the nest because of its sloping floor, productiv- ity during the 1980s study (7 or 8 nests successful in 17 analyzable nesting attempts) was satisfactory. Snyder found the decline in numbers, contrary to general belief, was fuelled by excessive mortality, about 25% per year for adults and immatures alike, mainly from lead poisoning. The radios did not con- tribute to mortality; indeed, only 3 of 9 radioed birds died whereas 12 of 15 non-radioed birds disappeared. Genetic studies of blood samples indicated low genetic diversity in the small remnant population. An observation in 1982 that, after loss of an initial egg, a condor laid a second after 40 days, allowed a bold experiment of replacement clutching. Begin- ning in 1983, each wild pair had their first egg taken from the nest and incubated artificially, thus devel- oping a captive population in two zoos, for eventual breeding purposes. The condor was then allowed to raise its second young in the wild. Between 1983 and 1986, 13 of 16 eggs taken in this way produced sur- viving chicks. Counting those eggs hatched in incu- BOOK REVIEWS 383 bators and those in the wild, six young were raised in 1983 and seven in 1984. Controversy about taking adults into captivity swirled between the Recovery team and the USF&WS; the latter organization “vacillated between moments of true wisdom and courage and unfortu- nate lapses in resolve and insight.” However, with the catastrophic mortality of six condors during the win- ter of 1984-85, and lead poisoning of another in December 1985, the official Recovery Plan suddenly became irrelevant. All wild birds were trapped and taken into captivity. By then, the stress of seven years of political duels had taken its toll and Noel Snyder had left the Service, for a “much less confrontational existence.” By 1987, in the Los Angeles Zoo and the San Diego Wild Animal Park, there were 14 females and 13 males in captivity, most fortunately an even sex ratio. In captivity, the level of productivity “far exceeded even the most optimistic projections.” The population grew rapidly to reach 150 by 1998. Fertility improved with each bird’s experience. By 1999, all 27 birds had been involved in production of at least one fertile egg! The release of condors back into the wild, based on success with Andean Condors, has not yet been an unqualified success. The release territory still contains carcasses of animals killed with lead ammunition. Puppet-reared nestlings cannot perfectly mimic adult condors in behaviour. When released into the wild, these birds approach humans without fear, enter towns, perch on buildings and poles, and trash property rang- ing from satellite dishes and roof shingles to screen doors, a habit unknown to former wild condors. Clearly the parent-reared birds have done best after release: 16 of 21 are still alive in the wild, while only 33 of 67 puppet-reared birds are still alive. Hunting with lead bullets is still allowed in condor release areas, and three released condors have developed lead poisoning, one requiring recap- ture and chelation therapy to clear its body of much of the lead. Snyder presents a forceful case that only the new TTB (TinTungstenBismuth) ammunition should be allowed in condor territory and that the maladapted condors in the wild should be recaptured and used again as breeding stock. Sadly, only 7 of 18 young condors raised in captivity in 1999 were parent- reared. Snyder recommends that only nestlings reared by live adults should be released in the years to come. The situation deteriorated further after the book went to press. At the AOU meeting in St. John’s, Newfoundland, in August 2000, Noel Snyder told me that the mortality rate of released birds is exces- sive, with 20 birds lost in the past year, almost equal to the number (22) produced in captivity in the same period. To date, five released condors are known to have died of lead poisoning while another 16 acutely poisoned with lead have only survived because of 384 emergency chelation therapy. The cause of death for more than half the birds lost is unknown, since many were never recovered. Noel and Helen Snyder call a spade a spade, pro- viding one perspective on a complex situation. They recount the deep rift between two camps, the tradi- tional “hands-off” conservation philosophy adopted by the followers of Carl Koford, and the activists who favoured captive breeding. I have one com- plaint; the heavy paper that allows near-perfect pho- tographic reproduction makes this book too heavy for comfortable handling while reading in bed! THE CANADIAN FIELD-NATURALIST Vol. 115 At this incredibly low price, you can buy one copy for yourself and another for a friend concerned about endangered species. This book is a landmark study that must be consulted by those involved in any recovery program for any endangered bird species, in the hope that mistakes made with the condor need not be repeated. C. STUART HOUSTON 863 University Drive, Saskatoon, Saskatchewan S7N 0J8 Canada Mammals of North America: Temperate and Arctic Regions By Adrian Forsyth. 1999. Firefly Books, Willowdale, Ontario. 352 pp., illus. $40. This is a book any naturalist would be proud to have on his or her bookshelf. Highly readable, visu- ally stimulating with outstanding photographs grac- ing nearly every page, and filled with fascinating information, it’s a resource I know [ll consult again and again. The author, Adrian Forsyth, an Ottawa native, well-known natural science writer, and Senior Biodiversity Scientist based at the Smithsonian’s Museum of Natural History in Washington, has cre- ated a unique North American mammal resource. It’s partly a field guide offering the following informa- tion on each mammal species: names, meaning of the scientific name, general physical description, colour, total length, weight, gestation period, total number of offspring, age of maturity, longevity, diet, habitat, and even dental formula. Unlike field guides, this book does not provide a visual representation for each species. Yet a number of mammals — the more widespread and/or often better-known species such as the masked shrew, little brown bat, snowshoe hare, groundhog, deer mouse, red fox, black bear, harbour seal, mink, bobcat, beluga whale, elk, and many more — are given a more in-depth profile. These profiles offer deeper insight into the animal, from distinctive features and behaviour, to its eco- logical role and threats to its survival. Readers will meet a social shrew, a mouse that howls like a wolf, a seal that naps on the ocean floor, a skunk that climbs trees, and a whale that sports the world’s largest tooth. Readers will learn how male deer mice go about providing parental care to their offspring, why racoons wet their food, how the Arctic fox keeps its feet from freezing, why white- tailed deer flash the white of their tails, and how the matriarchal structure of the orca whale works. Scattered among the species descriptions and por- traits are other brief essays on topics as diverse as mammal infanticide, migration, rumination, scientif- ic nomenclature, co-operative hunting, mammal milk, learning at play, bioconcentration of pollutants, and the advantages of eating meat. I was surprised at the cost of growing antlers, the success of the bat (Chiroptera) family, and the energy requirements of a harem bull elephant seal. I was impressed by the polar bear’s hunting strategies, the sperm whale’s diving capacity, the sensitivity of the kangaroo rat’s hearing. I was distressed by the references to species decline, habitat loss, and the general insensitivity of our Own species. But above all, I was fascinated, from cover to cover. As Forsyth writes in his introduction: “This book is a small attempt to interest readers in the wild mammals that live around them. It is generally true that the more one knows about something, the more one wants to know. And to conserve another organ- ism, even to want to conserve another organism, knowledge seems necessary. This is my agenda.” The agenda worked with me. I learned a lot from this book — particularly how complex the world of mammals is, and how much [ don’t know. So I’m intending to find out more. And I learned, I think, that Forsyth is a modest individual, because this book is anything but a “small attempt.” It’s a big book, an enormous accomplishment, and a signifi- cant achievement. R. SANDER-REGIER RR5 Shawville, Quebec JOX 2Y0O Canada 2001 Snipes of the Western Palearctic By Richard Rouxel. 2000. Eveil Nature, Saint Yrieux sur Charente, France. 304 pp., illus. English or French 2I15FF. This book is a detailed scientific review of the published information on the Common Snipe (Gallinago gallinago), Jack Snipe (Lymnocryptes minimus), and Great Snipe (Gallinago media). It does not include Pintail Snipe (Gallinago stenura), which live at the western edge of the Western Palearctic, between the Pechora and Chornaya River basins across to the Urals (Russia). Nor does it include a discussion of Common Snipe in North and South America, although many North American texts are cited (particularly Tuck’s CWS monograph on snipe). In addition, it does not clarify the status of an African subspecies of Common Snipe mentioned in the EBCC Atlas of European Birds (often consid- ered full specie — G. nigripennis). After an introduction to the taxonomy of Charadriiformes through to the Gallinagininae, the author begins on the species accounts. About half the book is devoted to the Common Snipe and the bal- ance split between the other two species. Each account contains a very detailed description of these birds. This is literally a feather by feather description of the important feather tracts that might help the researcher determine age and sex. These descriptions also include the molt sequences. Other biometrics such as measurements and weight contribute to information that could be used on captured birds. Most of this material would not be useful for field observations. The section on voice, however, is more helpful for field identification. The section on breeding is heavily weighted on distribution and numbers. For Europe the breeding population is estimated at 800 000 to 1 000 000. Adding in the Russian estimate of 1 to 10 million makes the range of the estimate a little absurd. Nesting, egg-laying and nurturing chicks is also described in detail. The movement of birds both diurnally and on migration leads to population estimates and a solid discussion on the validity of the data. This is then linked to mortality, particularly hunting. From even a cursory glance it is obvious that northern Europe is the snipe “factory” for hunters located in southern Europe. This pattern is followed for all three species of snipe. There is less information on Jack Snipe, pre- sumably because of its secretive nature. There is somewhat more on the Great Snipe, which at least can be counted during its lek displays. As the Common BOOK REVIEWS 385 Snipe is the most abundant, widespread, and studied, it accounts for the largest portion of the book. The book is illustrated with coloured photographs of the birds and their habitat. While the quality of the photographs is good the printing is less so. This is surprising as the paper quality and text printing is very good. This book is most difficult to read. It was original- ly written in French. The problems originate from several factors arising from the translation. I am well aware of the difficulties of changing languages. I have access to a very skilled group of translators who, not only are bilingual, but also have extensive experience in the technical and idiomatic language of my domain. They make sure the sense of my text means the same in either language. With this book I think the principal problem is the translation is too literal. This made the sentences convoluted and ram- bling. In some cases I could see the French version behind the English words and I could better compre- hend the sense. Often, though, I got lost and had to back track. This was particularly true of passages involving lots of data and multiple theories on inter- pretation. I have never seen so many colons in English text before. Their use extended the run-on nature of the sentences. There were a number of misused words. The prime example is the plural of snipe is snipe and not snipes. Another example is changing the French migrateur to migrator (which is not in my Oxford English Dictionary) instead of the correct word, migrant. Snipe do not browse or pasture, but feed. Pointer dogs “point” or indicate the location of birds; they do not block. The translator mixes quills up with feathers. The North American Marsh Hawk (Circus cyaneus) is confused with the European Marsh Harrier (Circus aeruginosus) when quoting from Tuck. There are errors of sense too. For example, “must be” is used in places where “ought to be” would be more appropri- ate. The author or translator uses some correct but obscure words like “haulm” (stems and stalks of plants). As a final comment I noted several typograph- ic errors. While none of these items prevent the reader from understanding the text, they do make that task very difficult. So this book is a very detailed compila- tion of the current knowledge on the three snipe species but ... A mon avis ¢a serait mieux d’acheter ce livre dans la langue originale. Roy JOHN 2193 Emard Crescent, Beacon Hill North, Gloucester, Ontario K1J 6K5, Canada 386 The Nature of Spiders, Consummate Killers By Adrienne Mason. 1999. Greystone Books, Douglas & McIntyre, Vancouver. 128 pp., illus. $34.95 This book is an excellent introduction to the world of spiders. It offers clearly written information, detailed drawings to support certain material, and lots of superb photos. I would venture to say that nearly half the book consists of photographs. With so much of the book devoted to photogra- phy, it still manages to convey an awful lot of infor- mation about spiders, including a summary of the major spider groupings (in the first chapter and in a 4-page appendix), descriptions of spider body parts and functions, a glimpse into various courtship and reproduction rituals, and a look at life after hatching, followed by a profile of different types of spider colonies. The book also provides an overview of diverse hunting techniques — including spitting, fishing, lassoing, stealing, and trapping through a wide range of webs — plus a look at creative strate- gies for spider self-preservation. It addresses the relationship between humans and spiders, from the spider’s role in folklore and mythology, to spider conservation and threats to survival. The author — trained as a biologist, now working as a freelance writer and naturalist — imparts to her topic a sense of awe that comes through in the first paragraph of the book: “Wherever you are, there is a spider within a meter of you ... We share our planet with close to 38,000 known species of spiders, and their diversity is remarkable. Whereas approximately 4,000 mammals roam the earth, there are close to 4.4000 species of jumping spiders alone.” In the pages of the book, readers meet some fasci- nating members of that 38,000- strong arachnid clan, including the world’s largest spider, the goliath bird- ENVIRONMENT THE CANADIAN FIELD-NATURALIST Vol. 115 eater (Theraphosa blondi) tarantula, whose legs could span a small pizza; the European cob-web weaver (Theridion sisyphium), who cares for her young by feeding them regurgitated food; fishing spiders (Dolomedes) who appear to use the ends of their legs as lures; jumping spiders (Salticidae) who let out silk to act as a bungee cord while they capture insects in flight; and the wheel spider (Carparachne aureoflava) who flips sideways and cartwheels down sand dunes to escape predators. The author’s fascination for her subject continues throughout the book, coming out as profound respect in the last paragraph, where she puts the spider’s future on this planet into perspective: “Long after we have cut our last tree, drained our last wetland, or perhaps even populated ourselves out of a home, somewhere, I have no doubt, spiders will be spin- ning. Their diversity, industry, and ski!l seems with- out parallel; after 400 million years, they have secured their place as one of nature’s most success- ful terrestrial animals.” Comforting thoughts. This book would make a wonderful addition to any naturalist’s library, if only for the photographs. The information covered will satisfy the curiosity of anyone wanting to know more about spiders, and will whet the appetite of budding arachnophiles to know more. The only thing I don’t like about this book is the subtitle, Consummate Killers. It sends an unneces- sary message of ferocity that will not help the spider overcome its unfortunate reputation. R. SANDER-REGIER RRS5 Shawville, Quebec, JOX 2Y0, Canada Encounters With Nature: Essays by Paul! Shepard Edited by Florence R. Shepard. 1999. Island Press, Washington. 200 pp. $38.50. Until his death in 1996, Paul Shepard, Professor Emeritus of Human Ecology at Pitzer College and Claremont Graduate School in California, wrote twelve books and many essays for publication. His work spanned five decades and his thoughts on human ecology and the relation of humans to the natural world had global perspective. This book compiled by his wife Florence contains essays and book parts written throughout his career with some previously unpublished notes and parts of manuscripts. Z I had never heard of Paul Shepard; but I found the material quite readable, entertaining, and thought provoking. His thoughts are critical of the human condition when referring to the natural world, which we share with animals, critical of human perceptions of animals and the world of nature. We have shaped our world in myth, art, children’s stories, and popu- lar culture but have not always represented other val- ues besides our own interests when seeing a beauti- ful picture, a productive land, or cuddly creatures. Part one of the book deals with animals, hunting, and human judgements. The section on bears is par- ticularly interesting as it examines current ideas and images, the myths of indigenous peoples, and chil- dren’s toys. Another chapter examines the writings of Dr. Albert Schweitzer, the physician, theologian and missionary in the jungles of Africa. Schweitzer’s Reverence for Life is critiqued as a European barn- yard ethic, protecting those animals which are to be husbanded for consumption and profit, while 2001 destroying predators and pests. Shepard contrasts the perspective of a naturalist with the attitudes of agri- culture and profit-motivated animal husbandry. The second part of the book is a series of essays on the importance of place. Shepard is outraged at how the modern world has taken industrial values and pro- moted the place of industry as pleasing and attractive. As the director of Green Oaks, the Knox College (Illinois) biological field station where he spent much time replanting trees, indigenous aquatic flora and fauna, and restoring fields to tallgrass prairie, and as conservation chairman for the National Council of State Garden Clubs he had much opportunity to see the changes which industry made on landscapes. In Shepard’s discussion industry includes manufactur- ing, mineral extraction, agriculture, forestry, and pas- turing. His essays contain experience from around the globe and insight into the movement of people, their domesticated animals, extraction industries, and build- ing strategies over the millennia. We need generalists like Paul Shepard with educa- tion in many disciplines to look at the big picture and research the interconnectedness of our world. I find River in a Dry Land: A Prairie Passage By Trevor Herriot. 2000. Stoddart Publishing, Toronto. 356 pp., 2 maps. Canadian $34.95. Trevor Herriot, a naturalist, writer, and artist, spent part of his early childhood in and near the east- ern end of the Qu’Appelle Valley, Saskatchewan. The valley is deeply rooted in his soul. This, his first book, is a tribute to the valley, its first peoples, its flora and fauna, and much more. Herriot takes us on an unusually personal journey, an exploratory tour with no holds barred. He mixes with admirable skill his observations on geography, geol- ogy, history, archaeology, anthropology, agrology, sociology, philosophy, economics, botany, mammal- ogy, and, his great love, ornithology. And through- out and above all, he demonstrates superb ecological insight and wisdom remarkable for a man so young. Herriot sees the world through the eyes of an artist, offering vivid word-pictures, writing with nos- talgia but also with sensitivity and integrity. His lyric, sometimes poetic prose ranks him as the suc- cessor to Wallace Stegner. His book should place Tantallon and the valley on the map, equivalent to Stegner’s success in Wolf Willow that placed the Cypress Hills and Eastend in public view. The dust-cover blurb is “right on” when it says “Herriot blends personal memoir and natural history and family legend with social commentary. ... gener- ations whisper to generations, in the voices of early explorers and utopian colonists, archaeologists and Cree elders, homesteaders and geologists. Their BOOK REVIEWS 387 that his research informed me on several different levels and his insight validated my experience. Industrial sites are not pleasing to the eye. Hunters are not always evil. Pastoral landscapes are only pleasing in art when your life experience of agricul- ture has to do with pasturing grazing animals. Overgrazing by sheep and goats is not a biblical value even though it seems to have been the normal practice (and still is) in the Middle East. Our society has forgotten or lost our reverence of searching and finding game, indigenous plants, nourishment, and shelter given to us by the earth and its Creator. Paul Shepard calls us back to that reverence in his cyni- cism of the ways we delude ourselves and in his out- rage at our wrong directions. I enjoyed the book, was able to identify with human ecology as he presented it and have discussed ideas from it with many friends. I recommend it to you. JIM O’ NEILL St. Mark’s College, 5935 Iona Drive, Vancouver V6T 1J7, Canada words arise always at the side of pathways both ancient and modern; the footpaths of buffalo hunters, the cart trails of displaced Métis and land-hungry immigrants, the spoor of coyote and bear, the rail lines abandoned by the very economy that built them, and, of course, the wandering, eloquent lines of the river itself.” Herriot’s broad sweep covers the entire Qu’ Appelle watershed, from Canada’s first wildlife preserve at the north end of Last Mountain Lakesouth to the former brick plant at Claybank, and from the west end adjacent to the South Saskatche- wan River elbow, to the east end of the valley where it joins the Assiniboine River near the Saskatchewan - Manitoba boundary. Because he tells the truth, some of what he says may cause resentment among those who worship the almighty dollar and confuse prosperity with progress. The first peoples of the valley and their relatives the Métis, with an ethic of local subsistence, were semi- nomadic. In contrast, Caucasian settlers, fixed to a half-section or section of land, were dependent upon distant economies and after a century or less left their farms, as Herriot says, “in the next movement to prosperity, better land, better jobs, a better El Dorado.” We are “exporting the life of the soil that fed the buffalo and prairie peoples in a sustainable cycle over thousands of years. ... We have too few farmers using too much machinery to feed too many people too cheaply too far away.” He catalogues 388 much evidence of the white man’s short-sightedness and stupidity. He vacillates between extreme pessi- mism and cautious optimism, and offers only a faint ray of hope for the future. On the credit side, as Caucasian farmers increase their individual holdings and their numbers become fewer, there has been an increase in elk, bears, ravens, and eagles. We learn about many people, including: the explor- ers, David Thompson, John Palliser, Henry Youle Hind; the poet, Pauline Johnson; early naturalists, John Macoun and R.D. Symons; the lifetime chroni- cler of valley bird life, Manley Callin; and Professor Zenon Pohorecky, who tried to preserve Mistaseni, the giant 400-ton buffalo rock, sacred to the Cree. We are also told about the short-lived Hamona colony, which failed because not everyone contributed equally to the farm labours and because the promised railroad failed to materialize; breadroot, wild onions, wild turnips, seneca root, the sources of natural food and medicine; the two species of towhee and two species of wood pewee, whose ranges meet in the valley; the MISCELLANEOUS THE CANADIAN FIELD-NATURALIST Voids endangered Burrowing Owls and Piping Plovers, the declining Upland Sandpiper; Swainson’s Hawks and Richardson’s Ground Squirrels. Recurring themes include the Bank and Barn Swallows and Say’s Phoebe. A review such as this can only hint at the superior quality of writing, especially Herriot’s ability to meld and synthesize a wide range of topics. This gifted wordsmith will provide many unexpected pleasures and provoke thought for any reader. A best-seller, appealing to a wide audience, it made the short list for the Governor-General’s award for non-fiction. Reprinting was necessary less than three months after its release, in spite of a substantial first printing run. This book is the perfect gift for any naturalist and for anyone who wishes to learn more about Saskatchewan. C. STUART HOUSTON 863 University Drive, Saskatoon, Saskatchewan S7N 0J8, Canada Yellowstone and the Great West: Journals, Letters, and Images from the 1871 Hayden Expedition Edited by Marlene Deahl Merrill. 2000. University of Nebraska Press: 315 pp. U.S. $29.95, £19.95. The US government’s surveys of the American West rank among the great scientific endeavours of the nineteenth century. In assessing the American frontier’s potential for commercial development, these surveys were a form of applied science with a government subsidy. Their achievements in geology, palaeontology, cartography, and other fields put them among the most influential scientific institu- tions of the period. Of all these scientific forays to the West, the 1871 expedition by geologist Ferdinand Hayden to the Yellowstone country has left perhaps the most visi- ble legacy. Hayden’s reports of this and later expedi- tions were mass-produced and widely read, and did much to bring Yellowstone scenery into middle-class drawing-rooms. Hayden was accompanied by a large retinue, which included William Henry Jackson, the photographer who first captured much of the Yellowstone country on plates, and Thomas Moran, the artist whose paintings of Yellowstone’s won- drous nature transfixed the public. Together, the geologist, photographer, and artist did much to con- vince U.S. legislators that Yellowstone should be made a national park - the first in the United States - and, subsequently, to persuade Americans to visit it. Yet, for all the importance of Hayden’s first expe- dition, we know little about how it operated, or what the party actually did from day to day. Oddly, nei- ther Hayden, Jackson nor Moran kept a diary that survived. It is as if the early nineteenth-century explorers Meriwether Lewis and William Clark had ventured on their mission along the Missouri River to the Pacific and then lost their notes. Marlene Deahl Merrill’s edited version of the journals of two other members of the team, geologist George Allen and mineralogist Albert Peale, thus fills an important gap, and constitutes the first daily account of the his- toric 1871 Yellowstone Survey. These journals are, in fact, the only diaries of the expedition known to exist. Integrated here with selections from Allen’s field notebook and letters that Peale wrote to news- papers during the expedition, they form-a valuable addition to our knowledge of the survey. This is a good thing, for previous accounts (most of them written by Hayden and Jackson years later) con- tain substantial errors. Jackson, for example, recalls in his autobiography that the Hayden party in 1871 was the first group of white men to visit Mammoth Hot Springs, one of the park’s central features. How strik- ing, then, to read here that when the party arrived they were met by two white settlers who had claimed the springs and were bent on turning them into a resort for health-seekers - such as the syphilitics who were also there to greet the expedition. 2001 Hayden himself, for a host of reasons, discounted Indian claims to the Yellowstone region, and con- tributed to the myth of Yellowstone as an “uninhab- ited wilderness”. In his journal, Peale reports sight- ings of Indians, and writes of the precautions the party took against Indian attack. (The worry about Indian attack was probably needless, but an idea of what the Indians in question might have looked like is given by a photograph of a family of Bannock Indians taken by Jackson shortly after he left Yellowstone.) Here, too, are glimpses of the scientific expedi- tion’s little-known support staff, people who are invisible in the official reports, including guides, horse wranglers, a driver nicknamed “Dummy” and a Mexican-American hunter known only as “Jos”. This is more than a compilation of journals, how- ever. Interspersed between Allen’s sentimental maundering and Peal’s scientific enthusiasm are a number of letters from Hayden to Spencer Baird, the assistant secretary of the Smithsonian Institution, reporting on the expedition’s progress. Pen-and-ink panoramas of the Yellowstone country, executed by the highly skilled (and self-taught) expedition topog- rapher Henry Elliott grace the headings of each of BOOK REVIEWS 389 the 10 chapters, and the work throughout is illustrat- ed with well-chosen photographs by Jackson. Merrill’s editing is superb. In 47 pages of end- notes, she integrates the journals with historical and scientific scholarship about Yellowstone and the American West. The achievements of the expedition become as salient as its curious errors, which include the mismeasuring of various peaks and geological features. The glossary of scientific terms is most helpful to the novice, and the four appendices (which include capsule biographies of expedition members) help flesh out the piece. Those wanting to know about Yellowstone at the time it became a national park have long relied on Hayden’s official reports. Now that body of work has an important supple- ment. Nobody who seeks a deeper understanding of Yellowstone’s natural systems as they were in 1871, or how nineteenth-century science was yoked to westward expansion, should miss this remarkable piece of editing and scholarly reconstruction. C. STUART HOUSTON 863 University Drive, Saskatoon, Saskatchewan S7N 0J8, Canada News and Comment The 123rd Annual Business Meeting of The Ottawa Field-Naturalists’ Club: 8 January 2002 The 123rd Annual Business Meeting of The Ottawa Field-Naturalists’ Club will be held in the auditorium of Victoria Memorial Museum (Canadian Museum of Nature), McLeod and Metcalfe streets, Ottawa, on Tuesday 8 January 2002 at 7:30 p.m. (19:30 h). The Council for 2002 will be elected at this meeting and a brief review of the activities during 2001 will be given, as well as a statement of the Club’s finances. KEN ALLISON Recording Secretary Call for Nominations: The Ottawa Field-Naturalist’s Club 2002 Council Candidates for Council may be nominated by any mem- ber of The Ottawa Field-Naturalist’s Club. Nominations require the signature of the nominator and a statement of willingness to serve in the position for which nominated by the nominee. Some relevant background information on the nominee should also be provided. COLIN GASKELL Chair, Nominating Committee Call for Nominations: The Ottawa Field-Naturalists’ Club 2001 Awards Nominations are requested from members of The Ottawa Field-Naturalists’ Club for the following: Honorary Membership, Member of the Year, George McGee Service Award, Conservation Award: Member, Conservation Award: Non-member, and the Anne Hanes Natural History Award. Descriptions of these awards have appeared in The Canadian Field-Naturalist 113(4): 689. With the exception of nominations for Honorary Member and Conservation Award: Non-member, all nominees must be members of The Ottawa Field-Naturalist Club in good standing. IRVIN BRODO Chair, Awards Committee Point Pelee Natural History News 1(1) Spring 2001 The first issue of a natural history newsletter for Point Pelee, Ontario, edited by Alan Wormington (e-mail: worm- ington@juno.com) contains features by Wormington, Alfred H. Rider, Henrietta T. O’Neil and Sarah Rupert: In the Beginning — The Point Pelee Natural Heritage Research Committee — The 2000 Invasion of Zebra Swallowtails at Point Pelee — Pawpaw at Point Pelee — Noteworthy Bird Records: August 2000 to February 2001 — Early Migrants at Point Pelee: Spring 2001 — Point Pelee Butterflies: Annual Summary for 2000 — Eurasian Teal: A new species for Point Pelee? — Point Pelee Alberta Wildlife Status Reports: (26 to 31) The Fisheries and Wildlife Management Division of the Alberta Natural Resource Status and Assessment Branch, Alberta Environmental Protection, has released new Wildlife Status Reports. The Series Editor is Isabelle M. G. Michaud, the Senior Editor is David R. C. Prescott, and the illustrations are by Brian Huffman. For a listing earlier numbers in the series, see The Canadian Field-Naturalist 112(1): 169 for 1- 11; 113(2): 311 for 12-17; 113(4): 686 for 18-21; and 114(1): 151 for 22-25. Recent reports issued in 2000 and 2001 are: 26. Status of the Trumpeter Swan (Cygnus buccinator) in Alberta, by Richard D. Lauzon. 17 pages. 27. Status of the Pygmy Whitefish (Prosopium coulteri) in Alberta, by William C. Mackay. 16 pages. 28. Status of the Short-eared Owl (Asio flammeus) in Alberta, by Kort M. Clayton. 15 pages. 29. Status of the Willow Flycatcher (Empidonax traillii) in Christmas Bird Count: December 18, 2000 — Recent Additions to Point Pelee’s “Checklist of Birds’, and sec- tions: In the Field, News and Announcements, Upcoming events and outings [April-May-June]. Subscription rates are Canada: CAN $15 (one year) or $30 (two years); International: US $15 (one year) or $30 (two years). Send payment (and e-mail address, optional) to The Friends of Point Pelee, 1118 Point Pelee Drive, Leamington, Ontario N8H 3V4. Issues will be mailed in March, June, September, and December, and back issues will be available for $15 per Volume/ $5 per issue (postage paid). Alberta, by Bryan Kulba and W. Bruce McGillivray. 15 pages. 30. Status of the Woodland Caribou (Rangifer tarandus caribou) in Alberta, by Elston Dzus. 47 pages. 31. Status of the Western Spiderwort (Tradescantia occi- dentalis) in Alberta, by Bonnie Smith. 12 pages. For copies contact the Information Centre - Publications, Alberta Environmental Protection, Natural Resources Service, Main Floor, Great West Life Building, 9920 - 108 Street, Edmonton, Alberta TSK 2M4, Canada (telephone: (780) 422- 2079), or Information Service, Alberta Environmental Protection, #100, 3115 - 12 Street NE, Calgary, Alberta T2E 7J2, Canada (telephone: (403) 297-3362); or Sherry Feser, 7th Floor, O. S. Longman Building, 6909 - 116 Street, Edmonton, Alberta T6H 4P2: telephone (780) 427-1248; fax (780) 422- 9685; e-mail sherry.feser@ gov.ab.ca 390 2001 The Boreal Dip Net 5(2) The March 2001 issue of the Newsletter of the Canadian Amphibian and Conservation Network/ Reseau Canadien de Conservation des Amphibiens and Reptiles (http://eqb- dge.cciw.ca/partners/carcnet/intro.html). Contents: Notes from the editor — Ontario Herp Summary: an update — Depositing your data — New Books to get — Some other meeting announcements — Terrestrial salamanders need woody debris — P.E.I. Annual Meeting: an announcement NEws AND COMMENT 39] — An ambystomatid coping at northern extremes — Sharp- tailed Snake study — Excerpts from Herp News — Membership/Donation forms. For membership and other information on the CARCN/RCCAR contact Bruce Pauli, Canadian Wildlife Service, National Wildlife Research Centre, 100 Gamelin Boulevard, Hull, Quebec, Canada K1A OH3. Froglog: Newsletter of the Declining Amphibian Populations Task Force (43, 44) Number 43, February 2001, contains: Surveillance of Amphibian Populations for Infectious Diseases (Rick Speare, Diana Mendez, Lee Berger, and Alex Hyatt) — Poorly Known Endemic Amphibians of Northeast India (Md. Firoz Ahmed) — Induction of Cell Death by Synthetic Pyrethroid Insecticide Cypermethrin in the Developing Brain of Physalaemus biligonigerus tadpoles from Argentina (M. F. Izaguirre, R.C. Lajmanovich, P. M. Peltzer, A. Peralta Soler and V.H. Casco — CARCNET/ RRCR Update (Christine Bishop) — Froglog Shorts — Publications of Interest. Number 44, April 2001, contains: Ultraviolet (UV) radi- ation and the decline of amphibian populations in Central Spain (Miguel Lizana and Adolfo Marco — The Crested Newt (T7riturus cristatus): Distribution, Biology, Ecology and Protection: 11-12 November 2000, Rostock, Germany (Richard A. Griffiths) — Monitoring Protocols Working Group Report (Richard A. Griffiths) — Amphibian harvest- ing in Romonia (Nemes Szilard and Kovasznay Csengele — Froglog Shorts — Publications of Interest. Froglog is the bi-monthly newsletter of the Declining Amphibian Populations Task Force of the World Conservation Union (IUCN)/Species Survival Commis- sion (SSC) and is supported by the Open University, The World Congress of Herpetology, the Smithsonian Insti- tution, and Harvard University. The newsletter is edited by John W. Wilkinson, Department of Biological Sciences, The Open University, Walton Hall, Milton Keynes, MK7 6AA, United Kingdom; e-mail: daptf@open-ac.uk. Funding for Froglog is underwritten by the Detroit Zoological Institute, P.O. Box 39, Michigan 48068-0039, USA. Froglog can be accessed at http:/www2.open.ac.uk/biology/froglog/ Frog Log 1(1): A newsletter for Canadian “Frogwatch” The inaugral (Spring 2001) issue of the Canadian newsletter Frog Log (not to be confused with the long- standing international newletter Froglog, see above) is a new project of the Canadian Nature Federation designed to keep participants in the (Canadian) “Frogwatch” inititive informed and up-to-date on program developments. Frogwatch is a joint program of Environment Canada’s Ecological Monitoring and Assessment Network (EMAN) and the Canadian Nature Federation. This issue of the newsletter features: How to get involved, Frog-watching Marine Turtle Newsletter (92) The April 2001 issue, 36 pages, contains: EDITORIAL: Making the books balance and a look at the future — ARTICLES: Update the Kemp’s Ridley Turtle Nesting in Mexico — Evaluation of the Black Turtle Project in Michoacan Mexico — Nest morphology in the Leatherback Turtle — Using GIS for Sea Turtle research at the Fog Bay rookery in northern Australia — Eastern Mediterranean ‘holiday hotspots’ versus sea turtle ‘nesting hotspots’ — Notes: From hook to hook: The odyssey of a Logerhead Sea Turtle in the Mediterranean — MEETING REPORTS — ANNOUNCEMENTS — News & LEGAL BRIEFS — RECENT PUBLICATIONS. tips, Frog-Faqs, What’s in a Name?, Frog Facts & Folklore. Coming this fall will be another Nature Watch program: “Wormwatch” (see fall, Nature Canada for details. For more information on Frogwatch, check EMAN site at http://www.cnf.ca/frog/. To obtain printed copies of the national Frogwatch identifiacation poster (in English or French) and the Frogwatch newsletter contact: Canadian Nature Federation, | Nicholas Street, Suite 606, Ottawa, Ontario K1N 7B7 Canada; e-mail: frogwatch @cnf.ca The Marine Turtle Newsletter is edited by Brendan J. Godley and Annette C. Broderick, Marine Turtle Research Group, School of Biological Sciences, University of Wales, Swansea, Singleton Park, Swansea SA2 8PP Wales, United Kingdom; e-mail MTN @swan.ac.uk; Fax +44 1792 295447. Subscriptions to the MTN and donations towards the production of MTN and its Spanish edition NTM [Noticiero de Tortugas Marinas] should be sent to Marine Turtle Newsletter c/o Chelonian Research Foundation, 168 Goodrich Street, Lunenburg, Massachusetts 01462 USA; e- mail RhodinCRF@aol.com; fax + 1 978 582 6279. MTN website is: . Editor’s Report for Volume 114 (2000) Mailing dates for issues in volume 114 were: (1) 7 March 2000, (2) 7 September 2000, (3) 27 December 2000, (4) 4 May 2001. Totals for circulation to mem- bers of the Ottawa Field-Naturalists’ Club and indi- vidual and institutional subscribers to The Canadian Field-Naturalist in 2000 together with those of 1999 are given in Table 1. The number of articles and notes in volume 114 is summarized in Table 2 by topic; totals for Book Reviews and New Titles are given in Table 3, and the distribution of content by page totals per issue in Table 4. An important aid to publication was created by Council decision to make 80% of the annual interest from the capital of the Thomas Manning bequest available to The Canadian Filed- Naturalist to offset the publication cost of northern papers where other funds (authors and institutional contributions) are insufficient to cover the page charges. This was made retractive to 1999, and in 2000 Council approved fund expenditures of totalling $1430: in 113(3): 375-385, $835 of the cost for the paper “Important bird and mammal records in the Thelon River Valley, Northwest Territiories: Range expansion and possible caunses”; in 113(4): 641-645 all of the $475 cost of publication for “Maritime Quillwort, Jsoetes maritima (Isoetaceae) in the Yukon Territory”; and 663-664 all of the $120 cost of “Frogs consumed by Whimbrels, Numenius phaeopus, on breeding grounds at Churchill, Manitoba” St. Joseph M.O.M. Printers, 300 Parkdale Avenue, Ottawa, set and printed the journal and special thanks are due Emile Holst, and to the Pre-press sec- tion particularly Yolande and Cecilia, and all the the MOM staff whose efforts make each issue possible. Wanda J. Cook proof-read the galleys and Bill Cody as Business Manager handled all reprint requests and bills and oversaw and proofed the compilation of the Index prepared by Leslie Durocher. Wilson Eedy continued as Book-Review Editor. Manuscripts (excluding book reviews, notices, club or journal reports) submitted to The Canadian Field-Naturalist totalled 154 in 2000, a substantial increase from recent years, 30 more than the previ- ous year and 41 more than the year before, and not topped since 1975 (when 167 manuscripts were sub- mitted). Unfortunatly this influx contributed to a slowing in the processing of papers, and substantial delays in their return. We expect to be caught up before the end of 2001. It has also meant an internal reconsideration by the editor of existing acceptance criteria for papers, both by content, region, and abili- ty to share costs of publication, but no change in our long-standing policies and broad focus are contem- plated at the present time. I am most grateful for the consideration, general good-will and patience extended by the many supportive authors for whom publication in The Canadian Field-Naturalist is important. The following returned reviews in 2000: Associate Editors: Birds: A. J. Erskine, Canadian Wildlife Service, Sackville, New Brunswick (40); Earl Godfrey, Nepean, Ontario; Botany: Charles D. Bird, Erskine, Alberta (18); Paul M. Catling, Agriculture and Agri-food Canada, Ottawa (14); Entomology: R. Anderson Canadian Museum of Nature (2); Fish and Marine Mammals: Robert R. Campbell, Ottawa, Ontario (7); Fish: Brian W. Coad, Canadian Museum of Nature, Cttawa, Ontario (3); Mammalogy: W. O. Pruitt, Jr., University of Manitoba, Winnipeg (40); Warren B. Ballard, Texas Tech University, Lubbock (37). Additional reviewers: L. Adams, U.S. Geological Survey, Ancorage, Alaska; S. Aiken, Canadian Museum of Nature, Ottawa; R. Anderson, Seattle Aquar- ium, Washington; C. D. Ankney, University of Western Ontario; C. Braun, Colorado State University, Fort Collins; J. Barr, University of Guelph, Ontario; E. Bayne, Canadian Wildlife Service, Saskatoon; G. V. Byrd, U. S. Fish and Wildlife Service, Adak, Alaska; J. R. Bider, MacDonald College of McGill University and Ecomuseum, Montreal (2); R. Bird, British Antarctic Survey, Cambridge, England; TABLE 1. The 2000 circulation of The Canadian Field-Naturalist (1999 in parenthesis). Membership totals from Annual Report of the Ottawa Field-Naturalists’ Club, January 2001; subscription totals compiled by W. J. Cody. Forty percent of membership dues and 100% of subscriptions go to publication of The Canadian Field-Naturalist. Members vote on Club } affairs, subscribers and institutions do not. Canada Memberships Family & individual 910 O21) Subscriptions Individuals 184 (184) Institutions 178_ (181) Totals 362 (365) TOTALS 1272 (1286) USA Other ~ Totals 32 (30) 6 (6) 948 (957) 61 (62) 6 (7) 251 (253) DS (277) 38 (39) 491 (Gen 316 (339) 44 (46) 742 (138)y 348 (369) 50 (52) 1690 (1707) Note: 22 countries are included under “Other” (outside Canada and United States): Austria, Belgium, Brazil, Denmark (2), | United Kingdom (9: including 1 to Scotland), Finland (2), France (3: including | to St. Pierre & Miquelon), Germany Iceland, Ireland, Japan, Mexico, Netherlands (3), New Zealand, Norway (4), Poland, Russia, South Africa, Spain (3), Sweden (2), Switzerland (2), Trinidad and Tobago. 392 2001 TABLE 2. Number of articles and notes published in The Canadian Field-Naturalist Volume 114 (2000) by major field of study. Subject Articles Notes Total Mammals 21 19 40 Birds 16 i fe) Amphibians + reptiles - 1 5 Fish 5 1 6 Invertebrates iL 1 8 Plants 11 pi 13 Tributes 3 0 3) Totals 67 ot 98 D. A. Boag, Brendwood Bay, British Columbia; J. Bogart, University of Guelph, Ontario; S. Bondrup-Neilson, Acadia University, Wolfville, Nova Scotia; W. S. Boyd, Canadian Wildlife Service, Delta, British Columbia; M. Boulet, MacMaster University, Hamilton, Ontario; I. Brodo, Canadian Museum of Nature, Ottawa; R. J. Brooks, Unviversity of Guelph, Ontario (2); R. T. Brooks, U.S. Department of Agriculture, Amherst, Massachusetts; M. Cadman, Canadian Wildlife Service, Guelph; L. Carbyn, Canadian Wildlife Service, Edmonton; M. Crete, Ministre de 1’Environment et de la Faune, Quebec; W. J. Cody, Agriculture and Agi-food Canada, Ottawa; J. Connelly, Idaho Department of Fish and Game, Pocatello; D. Christie, Albert, New Brunswick; A. Crowder, Queen’s University, Kingston; A. W. Diamond, University of New Brunswick, Fredericton (2); D. Dunbar, British Columbian Ministry of Environment, Surrey; J. Duncan, Ministry of Natural Resources, Winnipeg, Manitoba; E. H. Dunn, Canadian Wildlife Service, Hull; D.S. Erskine, Willow- dale, Ontario; A. O. Fortin, Montreal, Quebec; B. Theresa Fowler, Canadian Wildlife Service, Hull, Quebec; B. Freedman, Dalhousie Universitey, Halifax, Nova Scotia (2); T. D. Galloway, Department of Entomology, Uni- versity of Manitoba, Winnipeg; A. J. Gaston, Canadian Wildlife Service, Hull, Quebec; J. Gillen, Nova Scotia Museum, Halifax (2); P. Goosen, Canadian Wildlife Ser- vice, Edmonton (2); S. A. Graham, Kent State University, Ohio; P. J. Gregory, University of Victoria, British Columbia (2); W. Grimm, Eastern Ontario Biodiversity Museum, Kemptville, Ontario; Eric Haber, National Botanical Services, Ottawa (2); R. L. Haedrich, Memorial University of Newfoundland, St. Johns; P. Hamilton, Canadian Museum of Nature, Ottawa; F. Harrington, Mount St. Vincent University, Halifax, Nova Scotia (2); J. TABLE 3. Number of reviews and new titles published in Book Review section of The Canadian Field-Naturalist Volume 114 by topic. Reviews New Titles Zoology 24 84 Botany 4 22 Environment 16 58 Miscellaneous 4 24 Young Naturalists 1 66 Totals 49 254 EDITOR’S REPORT FOR 114 393 TABLE 4. Number of pages per section published in The Canadian Field-Naturalist Volume 114 (2000) by issue. (1) (2) (3) (4) Total Articles 141 124 2 160; “134° 560 Notes 8 26 18 16 68 News and Comment 4 6 4 3 A Tributes 1] 0 0 op gg Li Annual Meeting 0 0 7 0 7 Book Reviews* 21 ‘if 13 F150 Index 0 0 0 34 8634 Advice to Contributors 1 l 1 0 3) Totals 186 164 204 204 758 tIncludes a 21 page review article on status of amphibians and reptiles in Canda which appears after News and Comment. *Total pages for book review section include both reviews and new titles listings. D. Henry, National Parks Service, Haines Junction, Yukon; D. Hik, University of Alberta, Edmonton; G. Holord, Canadian Wildife Service, Edmonton; C. S. Houston, Saskatoon, Saskatchewan; R. James, Sutherland, Ontario (4); H. M. Jahns, Heirich Heine Universital, Dusseldorf, Germany; M. Jean, Centre St. Laurent, Montreal; W. Klenner, Kamloops, British Columbia; B. Kessell, University of Alaska Museum, Fairbanks; R. Lein, University of Calgary, Alberta; L. Licht, York University, Toronto, Ontario (2); J. Lien, Memorial University, Newfoundland; J. Mather, University of Lethbridge, Alberta; D. E. McAllister, Canadian Museum of Nature, Ottawa; R. McCulloch, Royal Ontario Musuem, Toronto (2); B. McCune, Oregon State Museum, Corvallis; B. McGillivray, Alberta Provincial Museum, Edmonton (2); D. McNicol, Candian Wildlife Service, Ottawa; D. J. Mech, North Central Forest Research Station, U. S. Geological Survey, St. Paul, Minnesota (3); R. W. Nero, Manitoba Natrual Resources, Winnipeg (2); J. D. McPhail, University of British Columbia, Vancouver; P. H. Martin, Canadian Wildife Service, Burnaby, British Columbia; J. Nelson, Unviersity of Alberta, Edmonton; D. J. Nettleship, Tantullon, Nova Scotia; M. J. Oldham, Ontario Natural Resources, Perterbough; M. Petrie, Ducks Unlimited, Memphis, Tennessee; S. Petrie, Long Point Waterfowl and Wetlands Research Fund, Port Rowan, Ontario; P. J. Pietz, Northern Prairie Wildlife Research Centre, Jamestown, North Dakota; C. Renaud, Canadian Museum of Nature, Ottawa; A. A. Reznicek, University of Michigan, Ann Arbor; J. D. Rising, University of Toronto, Ontario; G. R. Robertson, Canadian Wildlife Service, Delta, British Columbia (2); L. Rogers, Northwoods Research Centre, Ely, Minnessota; J. Romo, University of Saskatchewan, Saskatoon; D. Schecl, Alaska Pacific University, Anchorage; M. Schroeder, Washington Department of Fish and Game, Bridgeport; F. W. Schueler, Eastern Ontario Biodiversity Museum, Kemptville, Ontario (2); B. Scott, Kingston, Ontario (2); G. Scudder, University of British Columbia, Vancouver; S. C. Sealey, University of Manitoba, Winnipeg; K. W. Stewart, University of Manitoba, Winnipeg (2); I. Thompson, Canadian Forest Service, Saulte Ste. Marie, Ontario (8); R. Titman, MacDonald College of McGill University, Montreal; D. H. Vitt, University of Alberta, Edmonton; R. Weeber, Bird 394 Studies Canada, Port Rowan, Ontario; T. I. Wellicome, Canadian Wildlife Service, Edmonton. Once again, I am also indebted to Eleanor Zurbrigg, President of the Ottawa Field-Naturalists’ Club and the Club Council for continuing support of the journal; Chairman Ron Bedford and the Publi- cations Committee of the OFNC for editorial encour- agement and support, to the Canadian Museum of Nature for access to its library and the facilities at the Natural Heritage Building, 1740 Pink Road, THE CANADIAN FIELD-NATURALIST Vol. 115 Aylmer, Quebec, and to Joyce for everything else. Special acknowledgment and best wishes are due to Dr. Warren B. Ballard, who steped down at end of the year 2000 as an associate editor for mammalogy to assume editorial duties with another journal. His advice and encouragement, like that of all the associ- ate editors, has meant a great deal to me. FRANCIS R. COOK Editor TABLE OF CONTENTS (concluded) jotes ‘hree new taxa and a summary of the Mustard Family, Brassicaceae (Crucierae), in Canada and Alaska GERALD A. MULLIGAN rillium ovatum Pursh variety hibbersonii (Taylor et Szczawinski) - Douglas et Pojar, variety nova GEROGE W. DOUGLAS and JIM POJAR illing of a Bison, Bison bison, by a Wolf, Canis lupus, and four Coyotes, Canis latrans, in Yellowstone National Park DouUGLAS W. SMITH, KERRY M. Murpuy, and STAN MONGER hort-eared Owl, Asio flammeus, attack on a Burrowing Owl, Athene cunicularia, in Suffield National Wildlife area, Alberta SARAH D. CARNEGIE, ERIN J. URTON, and DAviID L. GUMMER \rboreal courtship behaviour by Eastern Garter Snakes,Thamnophis sirtalis sirtalis, in September in Bruce County, Ontario DavipD A. GALBRAITH creases and expansion of the New Brunswick breeding population of Black-legged Kittiwakes, Rissa tridactyla F, PATRICK KEHOE and ANTONY W. DIAMOND . Yellow Wood Lily, Lilium philadelphicum, from Nantucket Island, Massachusetts, with notes on its occurrence in New England DENVER W. HOLT and WESLEY N. TIFFNEY 1itial movement of juvenile Piping Plovers, Charadrius melodus, from natal sites in | northwestern North Dakota JEFFREY M. KNETTER, ROBERT K. MurRPHY, and R. Scott LUTZ vidence of an indirect dispersal pathway for Spotted Knapweed, Centaurea maculosa, seeds via Deer Mice, Peromyscus maniculatus, and Great Horned Owls, Bubo virginianus DEAN E. PEARSON and YVETTE K. ORTEGA ‘ooperative foraging by Steller Sea Lions, Eumetopias jubatus Scott M. GENDE, JAMIE N. WOMBLE, MARY F. WILLSON, and BRIAN H. MARSTON ‘ributes and Biography . tribute to Douglas Barton Osborne Savile, 1909-2000 J. GINNS and STEPHEN DARBYSHIRE Vvolution of a naturalist D. B. O. SAVILE 300k Reviews oology: A Field Guide to the Birds of China — The California Condor: A Saga of Natural History and Conservation — Mammals of North America: Temperature and Arctic Regions — Snipes of the Western Palearctic — The Nature of Spiders, Consummate Killers Nvironment: Encounters with Nature: Essays by Paul Shepard — River in a Dry Land: A Prairie Passage fiscellaneous: Yellowstone and the Great West: Journals, Letters, and Images from the 1871 Hayden Expedition Jews and Comment he 123rd Annual Business Meeting of The Ottawa Field-Naturalists’ Club: 8 January 2002 — Call for Nominations: The Ottawa Field-Naturalist’s Club 2002 Council — Call for Nominations: The Ottawa Field-Naturalists’ Club 2001 Awards — Point Pelee Natural History News 1(1) Spring 2001 _— Alberta Wildlife Status Reports: (26 to 31) — The Boreal Dip Net 5(2) — Froglog: Newsletter _ of the Declining Amphibian Populations Task Force (43, 44) — Frog Log 1(1): A newsletter for _ Canadian “Frogwatch” — Marine Turtle Newsletter (92) ‘tors Report for Volume 114(2000) FRANCIS R. COOK - uling date of the previous issue 115(1): 10 July 2001 341 343 343 345 347 349 | a a2 354 S55 357 365 381 386 388 390 392 THE CANADIAN FIELD-NATURALIST Volume 115, Number 2 Articles Tiger Salamander, Ambystoma tigrinum, movements and mortality on the : Trans-Canada highway in southwestern Alberta) ANTHONY P. CLEVENGER, MIKE MclIvor, BRYAN CHRUSZCZ, and KARI GUNSON Possible microclimate benefits of roost site selection in the Red Bat, Lasiurus borealis, in mixed mesophytic forests of Kentucky JEFFREY T. HUTCHINSON and MICHAEL J. LACKI Ectoparasites in lekking Sharp-tailed Grouse, Tympanuchus phasianellus LEONARD J. S. Tsusi, JIM D. KARAGATZIDES, and GERARDO DEILUIS Vascular plants of a successional alvar burn 100 days after a severe fire and their mechanisms of re-establishment PAUL M. CATLING, ADRIANNE SINCLAIR, and DON CUDDY New records for land snails from the mountains of northwestern British Columbia ROBERT G. FORSYTH The establishment and proliferation of the rare exotic plant, Lythrum hyssopifolia (Hyssop-leaved Loosestrife), at a pond in Guelph, Ontario Marc T. J. JOHNSON and CARL J. ROTHFELS Records of Canada Lynx, Lynx canadensis, in the Upper Peninsula of Michigan, 1940-1997 DEAN E. BEYER, JR., BRIAN J. ROELL, JAMES H. HAMMILL, and RICHARD D. EARLE Diet of the Prairie Rattlesnake, Crotalus viridis viridis, in southeastern Alberta MARGARET M. A. HILL, G. LAWRENCE POWELL, and ANTHONY P. RUSSELL Effects of enclosed large ungulates on small mammals at Land Between the Lakes, Kentucky CLARE C. WEIKERT, JOSEPH C. WHITTAKER, and GEORGE A. FELDHAMER Thermal habitat use and evidence of seasonal migration by Rocky Mountain Tailed Frogs, Ascaphus montanus, in Montana, USA SUSAN B. ADAMS and CHRISTOPHER A. FRISSELL Prey and reproduction in a metapopulation decline among Swainson’s Hawks, Buteo swainsoni JOSEPH K. SCHMUTZ, C. STUART HOUSTON, and SAMUEL J. BARRY Use of host-mimicking trap catches to determine which parasitic flies attack Reindeer, Rangifer tarandus, under different climatic conditions JOHN R. ANDERSON, ARNE C. NILSSEN, and WILLY HEMMINGSEN Germination potential, updated population surveys, and floral, seed and seedling morphology of of Symphyotrichum laurentianum, the Gulf of St. Lawrence Aster, in the Prince Edward Island National Park SARAH E. STEWART and CHRISTIAN R. LACROIX Mule, Odocoileus hemionus, and White-tailed, O. virginianus, deer in the Yukon MANEFRED HOEFS New records of vascular plants in the Yukon Territory III WILLIAM J. CoDy, CATHERINE E. KENNEDY, and BRUCE BENNETT Observations of change in the cover of Polargrass, Arctagrostis latifolia, and Arctic Lupine, Lupinus acticus, in upland tundra on Hershel Island, Yukon Territory C. E. KENNEDY, C. A. S. SMITH, and D. COOLEY Survey of freshwater mussels in the Petitcodiac River drainage, New Brunswick JOHN MARK HANSON and ANDREA LOCKE 200: Ay, 234 24 247) 25° DS 274% 287 296 307 32.5 329 (continued on inside back cover) ISSN 0008-3550 0 3 9088 01226 6 SMITHSONIAN INSTITUTION LIBRARIES | ANSEL OLe Sg aay SANT So, 11 nO 3594 “eH Sedan) Fo Wages ae Leconte A ag Ciwsce eat eke sea,