oe ¥ % . Y * ata ee ty Ne ee eles ‘ : rege stat aN fa : a ’ Phra aeiny : . BES cet ete De ete ses a ; thats : . G ; z Ree Se en roe ena st Mey * aa & ' Wei ates Sea SOREN eer : ; : Ee They yas wy ental wet SEAL AST ara a Re Ayre a wise te Cea eres pares Bt sgseeu 2 ORG LeMay gag oo buns! rie iio heats oes Pewaye pn te 28 pany Fite aaa Cae as) ye ey! MUph ggy ie os Daa Baby Soe em reer sop : Bk PRE Gears 2 penis ih Save Haass Btesgs Ip thoy i324 HARVARD UNIVERSITY e Library of the Museum of Comparative Zoology At ay i ee aye ue ee Hy ts ne fi MUS. COMP, ZOOL LIBRARY @iamnn ate, wise! HARVARD UNIVERSIT Y THE CANADIAN FIELD-NATURALIST Volume 97 1983 THE OTTAWA FIELD-NATURALISTS’ CLUB OTTAWA CANADA MATA RAD: Aa T it i k f - | TeLte MUTA | te ' ‘ — é 6 4 <5 ¢ f of ub , in ) L. > ‘ Ne bey uh A ic ? ue | r . ‘ a — . i nl ; Tis j M y } a a a! The CANADIAN FIELD-NATURALIST Published by THE OTTAWA FIELD-NATURALISTS’ CLUB, Ottawa, Canada Volume 97, Number 1 January-March 1983 The Ottawa Field-Naturalists’ Club FOUNDED IN 1879 Patrons Their Excellencies the Governor General and Mrs. Edward Schreyer 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 information on these fields as widely as possible; to support and cooperate with organizations engaged in preserving, maintaining or restoring environments of high quality for living things. Honorary Members William J. Cody Clarence Frankton Thomas H. Manning Loris S. Russell Mary E. Stuart William G. Dore’ W. Ear! Godfrey George H. McGee Douglas B. O. Savile Sheila Thomson R. Yorke Edwards Louise de K. Lawrence Hugh M. Raup Pauline Snure 1983 Council President: D. F. Brunton W.R. Arthurs E. M. Dickson B. M. Marwood Vice-President: P. M. Catling R. E. Bedford S. Gawn E. G. Munroe Vice-President: J. K. Strang B. A. Campbell G. G. Gruchy K. W. Taylor Recording Secretary: E. F. Pope W. J. Cody G. M. Hamre R. Taylor Corresponding Secretary: W. K. Gummer F. R. Cook D. R. Laubitz P. S. Walker Treasurer: P. D. M. Ward S. J. Darbyshire P. M. D. Martin Those wishing to communicate with the Club should address correspondence to: The Ottawa Field-Naturalists’ Club, Box 3264, Postal Station C, Ottawa, Canada KIY 4J5. Forinformation on Club activities telephone (613) 722-3050. 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. Editor: Francis R. Cook, Herpetology Section, National Museum of Natural Sciences, National Museums of Canada, Ottawa, Ontario KIA 0M8 Copy Editors: Wanda J. Cook, Glen Keenleyside and Louis L’Arrivée Business Manager: Mr. W. J. Cody, Box 3264, Postal Station C, Ottawa, Ontario, Canada KIY 4J5 Book Review Editor: Dr. J. Wilson Eedy, R. R. 1, Moffat, Ontario LOP 1J0 Coordinator, The Biological Flora of Canada: Dr. George H. La Roi, Department of Botany, University of Alberta, Edmonton, Alberta T6G 2E9 Associate Editors: C. D. Bird A. J. Erskine D. E. McAllister Stephen M. Smith E. L. Bousfield Charles Jonkel W. O. Pruitt, Jr. C. G. Van Zyll de Jong Chairman, Publications Committee: R. E. Bedford All manuscripts intended for publication should be addressed to the Editor. Urgent telephone calls may be made to the Editor’s office (613-996-1755), or his home on evenings and weekends (613-269-3211), or to the Business Manager’s office (613-996-1665). Subscriptions and Membership Subscription rates for individuals are $13 per calendar year. Libraries and other institutions may subscribe at the rate of $25 per year (volume). The Ottawa Field-Naturalists’ Club annual membership fee of $13 includes a subscription to The Canadian Field- Naturalist. Subscriptions, applications for membership, notices of changes of address, and undeliverable copies should be mailed to: The Ottawa Field-Naturalists’ Club, Box 3264, Postal Station C, Ottawa, Canada KIY 4J5. Second Class Mail Registration No. 0527 — Return Postage Guaranteed. Back Numbers and Index Most back numbers of this journal and its predecessors, Transactions of The Ottawa Field- Naturalists’ Club, 1879-1886, and The Ottawa Naturalist, 1887-1919, and Transactions of The Ottawa Field- Naturalists’ Club and The Ottawa Naturalist — Index compiled by John M. Gillett, may be purchased from the Business Manager. Cover: A fungus fairy ring — Marasmius Ring |, June 1968. For further description see Figure 2 in Toohey pp. 9-15. The Canadian Field-Naturalist Volume 97, Number | January-March 1983 Nesting Biology of Solitary Wasps and Bees in the Eastern Townships Region, Quebec STEPHEN GODFREY! and DONALD F. J. HILTON? 'Redpath Museum, McGill University, Montreal, Quebec H3A 215 2Department of Biological Sciences, Bishop’s University, Lennoxville, Quebec JIM 1Z7 3Author for correspondence. Godfrey, Stephen, and Donald F. J. Hilton. 1983. Nesting biology of solitary wasps and bees in the Eastern Townships region, Quebec. Canadian Field-Naturalist 97(1): 1-8. Trap nests were placed at 23 sites within two locations in the Eastern Townships region, Quebec. Out of 48 trap nests occupied, 34 contained Ancistrocerus spp.(A. antilope or A. catskill), 8 contained Symmorphus spp. (S. albomarginatus or S. cristatus) (Vespidae: Eumeninae), 3 contained Auplopus caerulescens or Dipogon sayi(Pompilidae) and 3 contained Megachile inermis or M. mendica (Megachilidae). Information on nest architecture, cell size, sex ratios, prey, parasites and associates is provided for most species. Key Words: Solitary wasps, solitary bees, Ancistrocerus, Auplopus, Dipogon, Megachile, Symmorphus, trap nests. Many solitary wasps and bees construct brood cells in pre-existing natural cavities such as tunnels made in dead trees by wood-boring insects, or the hollow stems of pithy plants like elderberry (Sambucus spp.). Nest structure, nesting behavior, larval food, parasites and predators of these cavity-nesting hymenopterans can be investigated by placing artificial nest cavities at appropriate sites in suitable habitats. Krombein (1967) summarized work done up to that time and reported on his own extensive studies conducted over a 12-year period. In the U.S.A trap-nesting investiga- tions were carried out in Arizona, Florida, Missouri, New York, North Carolina, Tennessee, Wisconsin, and Washington, D.C. For Canada, Fye (1965a, b; 1972) described the biology of wasps and bees obtained in trap nests set out in northwestern Ontario, while Longair (1981) discussed sex ration variations in aculeate Hymenoptera similarly collected at Lake Opinicon, Ontario. To our knowledge, no trap-nesting studies have ever been done in Quebec and this work was under- taken to provide information on the nesting biology of those species of wasps and bees that use trap nests in this part of Canada. Materials and Methods Trap nests were constructed from 20 X 20 or 25 X 25 mm pine or spruce boards cut to 75 or 165 mm lengths. Each piece was cut in half lengthways, the two halves clamped together and a hole drilled longitudi- nally down the midline. Holes were drilled using a standard wood bit rather than the twist drills employed by Krombein (1967). Following drilling, the clamp was removed and the two halves were taped together with water-resistant masking tape. This method of trap construction allowed future nest con- tents to be periodically examined by removing the tape and gently separating the two halves. Five sizes of trap-nests were used: 20 X 20 X 75 mm, hole depth 70 mm, hole diameter 3.2 mm (4”); 20 X 20 X 165 mm , hole depth 150 mm, hole diameters either 4.8 mm (3,”) or 6.4 mm ('4”); and 25 X 25 X 165 mm, hole depth 150 mm, hole diameter either 12.7 mm('4”) or 16 mm (*%”). Beginning | June 1980 bundles of five (one of each hole diameter) or three (hole diameters of 4.8, 6.4 and 12.6 mm) trap nests were placed at II sites adjacent to Bishop’s Uni- versity (Figure 1) and 10 sites near Johnville (Figure 2). In addition, bundles were deposited 3.2 km N of Bishop’s University (site number 21) and 2.5 km NNW of Johnville (site number 23). The characteris- tics of each trap-nest site are given in Table 1. Trap nests were usually examined weekly until | October 1980 and any that had been filled were brought back to the laboratory and replaced by an empty trap nest of the same hole diameter. Vegetation in this part of Quebec is mixed decidu- ous forest of the Eastern Townships portion of the 2 THE CANADIAN FIELD-NATURALIST BISHOP'S CAMPUS SSSCSCSSSSCOSOS®S 2 GOLF COURSE Vol. 97 FIELDS FiGURE |. Trap-nest sites (numbers) near Bishop’s University (45°22’N, 71°51’W). A: refuse dump; B: Red Pine stand; C: mixed deciduous woods. Not to scale. Great Lakes — St. Lawrence Forest Region (Rowe 1972). Habitats similar to the boreal forest occur locally within the otherwise deciduous hardwood forests, particularly where impeded drainage has con- tributed towards the formation of Black Spruce- sphagnum (Picea mariana-Sphagnum spp.) bogs. The Johnville site (Figure 2) is one such location and its physiography and vegetation are described by Bowers (1966) and Palmer (1970). The vegetation of both localities has been much altered by agriculture, lum- bering, buildings, highways, and railways. Results and Discussion Cells were not constructed in trap nests with hole diameters of 3.2 or 16 mm. Diameters of 6.4 mm were most preferred as nesting cavities, followed by 4.8 and 12.7 mm (Table 2). These preferences probably reflect the insect’s size, type of cell constructed and kinds of materials employed. A. Eumeninae (Vespidae) 1. Ancistrocerus a. antilope Fourteen of 48 trap nests were occupied by this species (Table 3). Male cells were slightly (not signifi- cantly) smaller than female cells (Table 5) and consi- derably more female than male cells were constructed in each trap nest (4.8 mm, I0F:2M; 6.4 mm, 19F- 12M; 12.7 mm, 10F:2M). Observations on egg laying, cell construction, larval feeding and adult emergence agree with the detailed life-history information provided by Cooper (1953) and Krombein (1967). Prey consisted of larval Olethreutidae and Pyrali- dae (Lepidoptera). It is possible larvae from other families were used but we could not determine this since it was impracticable to remove all larvae and send them for identification. Furthermore, by the time some cells were opened the prey had been consumed and any remains were unrecognizable. Generally, prey larvae in each cell were similar (species? genus?) and mixtures of families were rare. Female cells contained seven-eight prey larvae while male cells had only four- five. This reflects the greater nutritional requirements of developing females and must mean the adult female recognizes, and differentially provisions, cells in which she had laid either female or male eggs. 1983 GODFREY AND HILTON: SOLITARY WASPS AND BEES 3 MIXED FOREST FIGURE2. Trap-nest sites (numbers) at a bog habitat near Johnville (45°20’N, 71°45’W). A: gravel and sand pit; B: water sheds in forest clearing: G: sphagnum moss and ericaceous shrubs. Not to scale. The ectoparasitic mite Kennethiella trisetosa (Saproglyphidae) occurred on wasps from 2/14 trap nests (site numbers 18 and 19). This mite’s life history is intimately connected with its host’s life cycle, and includes venereal transmission from male to female wasps (Cooper 1955). One trap nest (site number 18) also contained an Exeristes comstockii larva (Ichneumonidae) in the second cell of a three-celled nest. This wasp killed the larval A. a. antilope, consumed the provisions and, after reaching adulthood, chewed its way through the mud partitions to emerge. Macrocentrus nigridorsis, Meteorus sp. and Microtypus sp. (Braconidae) were each reared from single trap nests. These wasps are solitary endoparasitoids of Lepidoptera larvae and presumably were already present as eggs or young larvae in the prey used as cell provisions. Upon reach- ing adulthood, these braconids were unable to chew their way through the cell partitions and thus perished since A. a antilope was not ready to emerge. Two trap nests (site numbers 18 and 22) each con- tained one Eumea (= Aplomya) caesar (Tachinidae). Following emergence from their puparia, both these individuals died imprisoned in the host cell. One was in the innermost of three cells and the other was in the middle of three cells. E. caesar is transcontinental in distribution and has been reared from the larvae and pupae of various species of Arctiidae, Gracillariidae, Lymantriidae, Noctuidae, Olethreutidae, Pyralidae and Tortricidae (Arnaud 1978). It seems likely that E. caesar is a solitary endoparasitoid of certain Lepidop- tera larvae used as provisions by Ancistrocerus a. antilope. 2. Ancistrocerus c. catskill Twenty of 48 trap nests were occupied by this spe- cies (Table 4). Cell structure and egg placement appear to be similar to that described for A. a. anti- lope (Cooper 1953; Krombein 1967). However, female cells were significantly larger than male cells (Table 5) and twice as many female cells as male cells (29F: 14M) were constructed in trap nests with a boring diameter of 4.8 mm whereas the reverse occurred in trap nests with a 6.4 mm diameter (17F:32M). Male and female THE CANADIAN FIELD-NATURALIST Vol. 97 TABLE |. Characteristics of trap-nest sites, number of trap nests occupied, and number of provisioned cells constructed at each site. Insects utilizing trap nests Ancistrocerus Megachile Symmorphus Height anti- albomar- Site No. Location (cm) Aspect lope catskill Auplopus Dipogon inermis mendica_ ginatus cristatus 1 fence 89 open 0 0 0 0 0 0 0 0 2 building 203 open 0 1(10) 0 0 0 0 0 1(4) 3 fence 81 open 0 0 0 0 0 0 0 0 4 stump 38 open 0 1(6) 0 0 0 0 0 0 5 fence 109 open 0 2(7) 0 0 0 0 0 0 6 log 58 open 1(7)° 5(18) 0 0 0 0 1(6) 0 7 tree 51+150° shade 0 0 0 0 0 0 0 0 8 stump 102 shade 0 0 0 0 0 0 0 0 9 tree 132+406 open 0 1(6) 0 2(15) 0 0 0 0 10 shrub 51 open 0 0 0 0 0 0 0 0 11 tree 216 shade 0 0 0 0 0 0. 0 2(9) 12 log 5 open 0 2(4) 0 0 0 0 0 0 13 tree 153+330 open 0 0 0 0 0 0 0 0 14 log 5 open 1(4) 1(2) 0 0 2(7) 0 1(4) 0 15 tree 114 shade 0 1(11) 1(9) 0 0 0 0 0 16 building 140 shade 0 3(23) 0 0 0 1(6) 0 2(9) 17 tree 170 shade 0 0 0 0 0 0 0 0 18 tree 140 open 6(20) 0 0 0 0 0 0 0 19 tree 107 open 3(14) 3(16) 0 0 0 0 0 0 20 tree 198 shade 0 0 0 0 0 0 0 0 21 building 153 open 0 0 0 0 0 0 0 1(3) 22 tree 119 shade 3(12) 0 0 0 0 0 0 0 23 fence 61 open 0 0 0 0 0 0 0 0 Total 14(57) 20(103) 1(9) 2(15) 2(7) 1(6) 2(10) 6(25) Mean no. cells/trap nest 4.1 5.1 9 7.5 3.5 6 5 4.2 “Two sets of trap nests. ’Number of trap nests occupied (total number of provisioned cells). TABLE 2. Preferences for trap-nest hole diameters. Ancistrocerus Megachile Symmorphus Diameter (mm) antilope catskill Auplopus Dipogon inermis mendica albomarginatus cristatus 4.8 3/ 14° 9/20 0 1/2 0 0 0 1/6 (21.4%) (45%) (50%) (16.7%) 6.4 7/14 11/20 0 1/2 0 0 1/2 5/6 (50.0%) (55%) (50%) (50%) (83.3%) 12.7 4/14 0 1/1 0 2/2 1/1 1/2 0 (28.6%) (100%) (100%) (100%) (50%) “Number of trap nests of this hole diameter occupied/total number trap nests occupied by this species. 1983 GODFREY AND HILTON: SOLITARY WASPS AND BEES 5 TABLE 3. Nest characteristics of Ancistrocerus a. antilope. Measurements in mm. Trap number 49 38 21 23 DY 47 26 48 ! 8 9 32 17 18 Site number 6 14 18 18 18 18 18 18 19 19 19 22 22 22 Diameter GOA 647i GS 1G 4h Gt! GA) ay a A Ah eb ee Nad Empty space length —. 40 32 — — 12 11 — = = = Preliminary plug a +° + + — = ? + Number provisioned cells 7 4 3 4 4 3 4 2 5 4 5 4 3 5 Provisioned cell lengths” 12 16 14 26 22 19 2 12 15 20 21 18 10 15 10 14 15 23 15 16 11 11 15 17 21 15 10 14 11 19 15 19 22 13 13 15 17 20 13 9 13 13 19 15 19 14 13 17) 14 24 15 9 12 17 14 12 8 Provisioned cell partition thickness ! 2 1.5 2) 1 1.5 1.5 I 1.5 2 2 2 1.5 1S) Intercalary cell length 1] — — Vestibular cell length 31 _ 41 38 35 52 50 60 60 4] 26 47 86 34 Closing plug thickness 9 — 3 4 ? 6 3 i 2 4 5 CPS) 6 “Terminology used in Tables 3, 4 and 6 according to Krombein (1967). >Measurements start from the innermost cell. “Absent. “Present. cells of A. c. catskill were significantly smaller than those of A. a. antilope (Table 5). Prey provisions in all nests belonged to the Gele- chiidae, Olethreutidae and Tortricidae (Lepidoptera). One of the tortricids was identified as Clepsis clemensiana. The only parasite obtained was Chrysis coerulans (Chrysididae) from site number 5. Chrysidids are soli- tary ectoparasites of larval wasps and bees, first eating the host’s egg or larva and then the provisions. Krom- bein (1967) reared C. coerulans from A. c. catskill in New York as well as from other cavity-nesting eume- nine vespids in Florida, Maryland and New York. Apanteles sp., Dolichogenidea sp. and Meteorus sp. (Braconidae) were each reared from single trap nests of A. c. catskill. Some of the Apanteles sp. cocoons were hyperparasitized by Lysibia mandibula- ris (Ichneumonidae). All died imprisoned in the Ancistrocerus c. catskill cells. One trap nest from site number 22 had the contents of 2/4 cells destroyed by larval Megaselia sp. (Phori- dae), 7 females, and Oscinella sp. (Chloropidae), | specimen. The individuals of Megaselia sp. belong to the aletiae, iroquiana and sphinx group, “all of which are reported from dead grasshoppers, fungi, nests of yellowjackets and sphecid wasps” (B.V. Peterson [1980] personal communication). Krombein (1967) reared M. aletiae from Podium rupifes, Trypargilum collinum rubrocinctum and T. striatum (Sphecidae) from North Carolina, Maryland and New York, respectively and Pachyodynerus erynnis (Vespidae) from Florida. Liston (1979) found nearly 40 cocoons of Megaselia sp. in 1/16 Cimbex femoratus cocoons collected near East Lothian, U.K. Liston considers the presence of Megaselia to be fortuitous and probably the result of a female Megaselia mistaking a C. femo- ratus pre-pupa for a moribund larva and therefore ovipositing init. Further, he doesn’t think adult Meg- aselia mouthparts are adequate for chewing their way out of the C. femoratus cocoon. In opposition, how- ever, he cites R. H. L. Disney’s opinion (in a personal communication) that the phorids were obligate para- sitoids of the sawfly and would have emerged success- fully. Oscinella spp. are often plant pests (Cole 1969) but without knowing the species name for this speci- men it is not possible to infer what role it was playing in a cell of Ancistrocerus c. catskill. 3. Symmorphus spp. Two and six trap nests were occupied by S. al/bo- marginatus and S. c. cristatus, respectively (Table 6). These two wasps were similar to Ancistrocerus spp. in their biology and nest construction (Krombein 1967). However, in this part of Canada Symmorphus spp. appear late in the summer and are univoltine whereas Ancistrocerus spp. appear earlier and are bivoltine. This was the case in Fye’s (1965a) study. In addition, Symmorphus spp. are much more likely to construct ON THE CANADIAN FIELD-NATURALIST Vol. 97 TABLE 4. Nest characteristics of Ancistrocerus c. catskill. Measurements in mm. Trap number St 5) 0842716, 2529" 30" Bile 40) 4) 45.) 22570 Si, A Oe Vowel 25 Site number Diiserds eS D6" 6, OY 966 95 12D 14S 6S 16> S16: 19 eat9 19 Diameter 64 48 48 48 48 64 64 64 64 64 48 48 48 48 48 64 64 64 64 64 Empty space length 16 $15 27 5 20 — — 42 — ? 74 400 — ~— ~ ~ ~~ ~ — — Preliminary : plug GP heme eB SE ac RM Be yee NL Be TS ON ete Stee cy fit sk 2. 5 a nah rc ea + No. provisioned cells OEE AS 6 Sim ean FOP Seinen see al a6 1 3.02) ON a8 FS soe aes 6 Provisioned cell lengths I Uy RO ts Wa it Nas NA IAS TNs TO. I) 7 16 13 Lael Sel Deo ea eS = lS] OFS 98 era? 16 12 13 10 #410 211 #10 #19 15 I@ 3 Ie FY WW i esa Wate 14 12 10 10 211 8 20 14 il ly) 10 11 ls} 10 10 10 10 8 10 7 30 13 2 11 10 10 11 6 10 el ? 1] Oe Sa 13 12 7 10 8 10 7 7 10 8 10 7 10 10 9 10 10 Provisioned cell partition thickness ] ] | lee2eS) ] LO ESOS Ss? Sis 125 1 1 1 |i 1 AES lS Intercalary cell length: —_— — —- ? — 10 —~ ~ ~ ~ ~ ~ — — Vestibular cell length Di 2 22 SOM eS O24 1e 84 eh Sie 23 29S One Sie Sill 2eO2 aer/S 43 Closing plug thickness Tey cA vice: Gos De oa. Uae Spal rir Dian PAY ees) 3 Bat a Dee 3 “Measurements start from the innermost cell. Absent. “Present. TABLE 5. Size differences (in mm) between male (M) and female (F) cells of Ancistrocerus spp. Ancistrocerus antilope catskill M F M F antilope F : catskill F antilope M : catskill M Number of cells 12 33 38 48 — = Mean length 15.6 16.0 10.1 11.6 -- — Range 8-22 10-26 6-16 6-19 — — SD! 43 4.1 Dw, 2.8 ~ = Student’s t 0.3218 2.909 5.387 6.214 Significance level ns <0.01 <0.001 <0.001 1983 intercalary cells (Table 6) than are Ancistrocerus spp. (Tables 3, 4). Krombein (1967) also noted this pheno- menon. As is typical for Symmorphus, both species provis- ioned their cells with larvae of leaf beetles, Chrysom- ela sp. (Chrysomelidae). These are smaller than the lepidopteran larvae utilized by Ancistrocerus spp. and larger numbers are used to provision each cell (for S. c. cristatus; mean 19; range 18-26). One trap nest (site number 16) containing three cells of S. c. cristatus had one cell parasitized by Actia interrupta (Tachinidae) and another by Melittobia chalybii (Eulophidae). As was the case for Eumea caesar in the cells of Ancistrocerus a. antilope, this individual of Actia interrupta, following emergence from its puparium, died trapped in the Symmorphus c. cristatus cell. Actia interrupta is widespread in North America and has been reared from the larvae and pupae of a wide variety of Geometridae, Noto- dontidae, Olethreutidae and Tortricidae (Arnaud 1978). It apparently bears the same relationship to Symmorphus c. cristatus as Eumea caesar does to Ancistrocerus a. antilope (vide supra). Melittobia chalybii parasitizes the larvae of social and solitary wasps. Large numbers of individuals of both of the types produced by this dimorphic species were found and we captured 70 from the cell when it was opened but many more escaped. Details of its life cycle are provided by Clausen (1962), Evans and Eberhard (1970) and Spradbery (1973). Krombein (1967) consi- dered it to be “a very serious pest” of the various wasps he was investigating in his trap-nesting studies. GODFREY AND HILTON: SOLITARY WASPS AND BEES : 7 B. Pompilidae 1. Auplopus caerulescens subcorticalis Only one trap nest (diameter 12.7 mm) at site number |5 contained nine cells of this species. Cells are barrel-shaped structures (4-5 X 7 mm) made from fine-grained mud and lie loose within the boring hole in an end-to-top fashion. Cells were not opened for prey analysis but Krombein (1967) states that the ones he examined contained Clubiona spp. spiders (Clubi- onidae). In addition, Evans and Yoshimoto (1962) record the spiders Anyphaena pectora (Anyphaeni- dae), Phidippus audax (Salticidae) and Trachelas tranquillus (Clubionidae) as prey being carried by the adult wasp. However, they were never able to find any provisioned cells of Auplopus caerulescens subcorticalis. 2. Dipogon s. sayi Two trap nests, both from site number 9, contained seven and five cells of this species (Table 6). Nest and cell construction were similar to that reported by Evans and Yoshimoto (1962), Fye (1965a) and Krom- bein (1967). These trap nests were tied to the trunk ofa Red Pine (Pinus resinosa) on the edge of a P. resinosa stand. Female D. s. sayi apparently foraged for nest materials in the immediate vicinity since the cell parti- tions were made up of hundreds of tiny pieces of wood and bark of P. resinosa, seeds (unidentified) and bits of lichens (Cladonia sp. and Parmelia sulcata). Cells were provisioned with crab spiders (Thomisi- dae). This agrees with Evans and Yoshimoto (1962), Fye (1965a) and Krombein (1967) who say that Tho- TABLE 6. Nest characteristics of Dipogon s. sayi, Symmorphus albomarginatus and S. c. cristatus. Dipogon s. sayi Trap number 33 $2 Site number 9 9 Diameter 4.8 6.4 Empty space length a — Preliminary plug No. provisioned cells 7 5) Provisioned cell lengths* 9 6 13 13 13 5 9 8 7 3 4 6 Provisioned cell partition thickness eS 5 Intercalary cell length — — Vestibular cell length — 4 Closing plug thickness 23 15 “Measurements start from the innermost cell. b Absent. “Present. Symmorphus albomarginatus Symmorphus c. cristatus 50 46 36 54 53 37 56 19 6 14 2 1] 11 16 16 21 6.4 12.7 64 48 64 64 64 64 _ — — 14 = ue ats = + + + es 6 4 4 5 4 6 3 3 1] 11 12 8 10 10 13 9 12 10 9 10 1] 10 20 9 10 11 I] 1] 22. 12 15 10 lI 15 12 11 27 10 12 15 9 1] U5) ? l 1.5 l l 1.5 IS) | = — 3 a a lES l 20 4 50 65 46 52 33 36 34 _ 9 18 6 6 5 3 — 8 THE CANADIAN FIELD-NATURALIST misidae, especially Xysticus, are the typical prey utilized. C. Megachilidae 1. Megachile spp. M. inermis and M. mendica occupied two and one trap nests, respectively (diameters of 12.7 mm), at sites number 14 and 16(Table 1). Nest architecture of these leaf-cutter bees is similar to that described by Fye (1965b) and Krombein (1967). Seven cells of M. inermis had a mean length of 11 mm (range 10- 12 mm) and each was constructed from about 25 pie- ces of what were probably the leaves of Broadleaf Spirea (Spirea latifolia). Circular pieces (diameter approximately 10 mm) were used for the ends and oblong pieces (approximately |! X 20 mm) for the walls. Leaf fragments composing the cells of M. men- dica were packed in such a way as to make it extremely difficult to separate them for measurement and spe- cies determination. Acknowledgments We would like to thank the following individuals for kindly identifying specimens: L. LeSage (Coleop- tend) BeVeaeterson (Diptera): J .Re Barrons ide Goulet, M. Ivanochko, L. Masner and W. T. M. Mason (Hymenoptera); S. Allyson and W.C. McGuffin (Lepidoptera) all of the Biosystematics Research Institute, Agriculture Canada; and J. Car- penter (Eumeninae) of Cornell University. Literature Cited Arnaud, P. H., Jr. 1978. A host-parasite catalog of North American Tachinidae (Diptera). United States Depart- ment Agriculture; Science and Education Administration, Miscellaneous Publication No. 1319. Bowers, T. M. 1966. Bogs of the upper St. Francis River basin, Quebec. Phytosociological notes and field key for vascular plants. M.Sc. thesis, Bishop’s University. Len- noxville, Quebec. Clausen, C. P. 1962. Entomophagous insects. Hafner Pub- lishing Co., New York. 688 pp. Cole, F. R. 1969. The flies of western North America. Uni- versity California Press, Berkeley, 693 pp. Vol. 97 Cooper, K. W. 1953. Biology of eumenine wasps I. The ecology, predation, nesting and competition of Ancis- trocerus antilope (Panzer). Transactions American Ento- mological Society 79: 13-35. Cooper, K. W. 1955. Venereal transmission of mites by wasps, and some evolutionary problems arising from the remarkable association of Ensliniella trisetosa with the wasp Ancistrocerus antilope. Biology of eumenine wasps II. Transactions American Entomological Society 80: 119-174. Evans, H. E., and M. J. West Eberhard. 1970. The wasps. University Michigan Press, Ann Arbor. 165 pp. Evans, H. E., and C. M. Yoshimoto. 1962. The ecology and nesting behavior of the Pompilidae (Hymenoptera) of the northeastern United States. Miscellaneous Publica- tions Entomological Society America 3: 65—120. Fye,R. E. 1965a. The biology of the Vespidae, Pompilidae, and Sphecidae (Hymenoptera) from trap nests in north- western Ontario. Canadian Entomologist 97: 716-744. Fye, R. E. 1965b. Biology of Apoidea taken in trap nests in northwestern Ontario (Hymenoptera). Canadian Ento- mologist 97: 863-877. Fye, R. E. 1972. The effect of forest disturbances on popu- lations of wasps and bees in northwestern Ontario (Hyme- noptera: Aculeata). Canadian Entomologist 104: 1623-1633. Krombein, K. V. 1967. Trap-nesting wasps and bees: life histories, nests, and associates. Smithsonian Press, Washington, D.C. 570 pp. Liston, A. D. 1979. On Phoridae (Diptera) from sawfly cocoons (Hym.: Symphyta). Entomologist’s Record and Journal Variation 91: 303-305. Longair, R. W. 1981. Sex ratio variations in xylophilous aculeate Hymenoptera. Evolution 35: 597-600. Palmer, K.T. 1970. A vegetational analysis of five bogs of the central St. Francis River drainage basin, Quebec. M.Sc. thesis, Bishop’s University, Lennoxville, Quebec. Rowe, J.S. 1972. Forest regions of Canada. Canadian Forest Service Publication (1300): 172 pp. Spradbery, J. P. 1973. Wasps. An account of the biology and natural history of social and solitary wasps. University Washington Press, Seattle. 416 pp. Received 21 May 1981 Accepted 20 February 1982 Fungus Fairy Rings in Soil: Etiology and Chemical Ecology JOHN I. TOOHEY'! Queen’s University Biological Station, Elgin, Ontario 'Present address: Department of Medicine, University of California at Los Angeles, Los Angeles, California 90024 Toohey, John I. 1983. Fungus fairy rings in soil: etiology and chemical ecology. Canadian Field-Naturalist 97(1) 9-15. Thirty species of fungal sporophores were identified in association with fungus fairy rings on abandoned farm land ina 100 square km study area in the Rideau Lakes Region of Ontario. The rings were characterized with respect to their effects on herbacious vegetation and their rates of expansion. The vegetation was killed, stimulated, or unaffected by the various fungi. Stimulated vegetation appeared to be associated only with fungi which were growing saprophytically whereas killed or unaffected vegetation was associated with either saprophytic or mycorrhizal fungi. The rings reveal two interesting features of soil ecology. Firstly, they demonstrate the dramatic effects that soil fungi can have on herbacious vegetation. Secondly, they indicate that basidiomycete spores give rise to new mycelial colonies in soil at an extremely low rate. Possible mechanisms by which certain fungi affect plant growth are discussed. Key Words: Fairy rings, fungus rings, soil ecology. Ring-shaped patterns of altered vegetation in mea- dows and lawns have been recorded since antiquity. Early references to the phenomenon attributed it to supernatural causes, giving rise to terms suchas “fairy rings”, “hexenringe”, or “cercles magique”. Withering (1796) was the first to associate these rings with soil fungi and Marasmius oreades was the first fungus so associated. In 1917, Shantz and Piemeisel published a comprehensive treatise on the phenomenon, in which they reviewed the mythology associated with rings, the species reported to cause rings, and their own observations on rings in Colorado. They recognized three types of rings based on the effects produced on the herbacious vegetation: Type I, vegetation killed; Type ll, vegetation stimulated; and Type III, no effect on vegetation. They compiled a list of 56 species of fungi reported to cause rings up to that time. More recent reports bring to 127 the number of fungus species reported in association with rings. (A complete list of species, compiled in June, 1981, is available from the Depository of Unpublished Data, CISTI, National Research Council of Canada, Ottawa, Canada, KIA 0S2). In this report, observations on fungus rings in Eastern Ontario are reported. These observations are a sequel to an earlier study on the chemistry of Type | rings, then called “barren rings” (Toohey, et al. 1965a, 1965b). Methods The study was carried out in an area of about 100 square km in the vicinity of Queen’s University Bio- logical Station at Chaffey’s Lock, Ontario. The topo- graphy of the region is one of glaciated granite and limestone outcroppings interspersed with relatively flat areas of lacustrine-deposited clay soil. The latter florentinum, areas with sufficient natural drainage were cleared for farming about 160 years ago and most were aban- doned 30 to 60 years ago. Some fields have been unused and have regenerated a young forest; other fields have been used for grazing cattle and have remained as open fields. The herbacious plants in the fields were predominantly grasses, Phleum pratense, Danthonia spicata, Poa compressa, or Poa pratensis; or the sedge, Carex pennsylvanica. Other commonly occurring plants were Trifolium pratense, T. agra- rium, Chrysanthemum leucanthemum, Hieracium Ranunculus acris, Echium vulgare, Potentilla recta, and Taraxacum offinale. Each field was examined for rings by walking through it at different seasons. Rings were identified by altered vegetation and/ or by the presence of fungus sporophores. Wooden stakes were placed at the outer edge of the affected band of each ring. Each subse- quent year in September or October, new stakes were placed at the advancing edge of the band and mea- surements of yearly expansion were made at that time. The width of the affected band and the annual increase in radius are reported as averages for all rings of a given species over all the years of observation. The study was begun in 1961 but the rings have been under study for variable periods of time, since many rings have disappeared and newrings were found each year. Sporophores were collected when they first occurred, tentatively identified, preserved by drying, and for- warded to the Mycology Section, Biosystematics Research Institute, Department of Agriculture, Ottawa, for confirmation of identity. The specimens are in the permanent collection of the National Myco- logical Herbarium (DAOM). In addition to the systematic study described above, 10 THE CANADIAN FIELD-NATURALIST casual observations were made on an exceptional aggregation of rings located on Centre Island of the Toronto Island Parks, Toronto, Ontario. The lawn on which these rings occurred was created in 1960 by dredging sand from Lake Ontario to fill a marsh, followed by seeding with lawn grass seed. The classification system of Shantz and Piemeisel, based on effect on vegetation, was used with some modification. Each type was dichotomized on the basis of whether or not sporophores were observed on the rings, indicated in the terminology by placing a plus sign (+) or a minus sign (-—) after the numeral. Most of the rings of Types I and II were first recog- nized by the altered vegetation alone, subtype (—), and they were reclassified in subtype (+) in later years when they produced sporophores. A new type, Type 0, was created to accommodate rings in locations where her- bacious vegetation was sparse or absent, making it impossible to determine the effect of the fungus on vegetation. The following definitions apply to terms used in this report to describe patterns of fungus growth in soil. Band: a curvilinear area of affected soil or vegetation _ Marrow in one horizontal dimension and long in the other horizontal dimension. Ring: a regular curved band forming a complete turn of 360° and, therefore, enclosing an area of unaffected soil or vegetation. The term “ring” is frequently used in a generic sense to refer to any growth pattern of this type even though it is not geometrically complete. Arc: a segment of a ring. Results and Discussion Description of Fungus Rings 159 rings were studied and 30 species of sporo- phores were identified in association with the rings. Table | lists the species according to ring type and describes the characteristics of the rings produced by each species. The production of sporophores was very much dependent on weather conditions and in most years none were produced. On rings of Types I and II, the sporophores were produced at the outer edge of the altered vegetation. Sporophores of ring species were never observed in random distribution in the study area although sporophores of other species did occur randomly in the area. Type I Rings. On rings of Type I, the vegetation was suppressed. The degree of suppression was not the same forall species and this Type might be subdivided into two subtypes. In subtype a, the vegetation was completely killed producing a band | to 2 m wide devoid of plants. This effect was produced on rings of Type I+ by Clavaria vermicularis and Helvella con- nivens and was seen on many rings of Type I- (Figure 1). The appearance of these rings was not appreciably Vol. 97 affected by weather conditions. In subtype 5, the vege- tation was distinctly suppressed in density and vigour but not killed completely. The affected band was 30 to 60 cm wide and all plant species were equally affected. The degree of suppression was dependent on weather conditions, being most conspicuous in dry seasons and less marked in wet seasons. Rings of subtype b were produced by Amanita flavorubescens, Clavaria cinerea, Inocybe lacera, Russula decolorans, R. densi- folia, and R. aerugina. Most of the rings of Type! were located at the edges of fields within 30 m of the trees bordering the fields or, if they occurred in the centre of fields, there was at least one tree within 30 m. This was true for all of the rings of Amanita flavorubescens, Inocybe lacera, Russula aeruginea, R. decolorans, and R. densifolia, which is consistent with the reported mycrorrhizal nature of these species (Trappe 1962). However, the relationship of the rings to the trees was not clearly defined. Some rings of the Russula species occurred as arcs with the open portion of the arcs facing the line of trees, suggesting that they were related to the tree roots. Incontrast, other arcs and some complete rings, although near trees, did not show any geometrical relationship to the trees. A similar undefined relation- ship was reported by Hawksworth (1962) in rings of Polyporus confluens occurring under trees of the spe- cies Pinus contorta and Picea engelmanii. Type II Rings. On the bands of Type II rings, the vegetation was taller, denser, and a darker green than the adjacent vegetation (Figure 2). This effect was dependent on weather conditions: the stimulation was most pronounced in seasons of heavy rainfall and barely perceptible during periods of drought. In this study, six species of fungi were associated with Type II rings (Table 1). In addition to the rings described in the table, a large group of Type II rings was observed ona lawn on Centre Island, Toronto. These included 80 rings of Marasmius oreades, 25 rings of Lycoper- don curtisii, and 22 rings of Melanoleuca humile. They were all of diameters less than 4 m and were less than 21 years old since the lawn was created in 1960. Rings of Type II occurred in open fields, frequently several hundred m from the nearest tree, and there was no evidence that any of them were related to trees. Six of the eight species observed on stimulated rings are apparently non-mycorrhizal (Agaricus campestris, Calvatia fragilis, Clitocybe dealbata, C. subconnexa, Lycoperdon curtisii, and Melanoleuca humile). Marasmius oreades is said to be mycorrhizal on Pinus ponderosa but this tree does not occur in eastern Canada. Lycoperdon perlatum is mycorrhizal but many rings observed in this study were clearly unre- lated to trees. The observations of Shantz and Piemeisel on Type 1] FUNGUS FAIRY RINGS IN SOIL TOOHEY 1983 sI snuad JURUTUO pad ay} 10 JSaIBIU 94} A[UO “B19 [LIAAIS JO 919M $991} BY} “PouIW1a}ap ION, “QSRIDAB JY} MOTIG 10 JAOGK I%C] UL) SSa] SEM UONILIIPA SY Jy ‘paureu si saioads yuvuiwopaid 10 ysorvau ay} yuo ‘satoads 904) paxtu JO 19M Pasinos0 sBuls Q adA] YotyM Jopun s}sa10J Ys ‘poweu JOUIS ‘Pal 947 SULIAPI0G $991} YI JO WI YE ULYIM 19M SBULI UY] JO Sa1jUIO AY} LY) SaLIIPUL Pal} JO 99 Pq,,p ‘saplopsadord] vsoydosaisy ‘ajaohwiorpiseg ay} Aq paziisesed AjIAvay ataM sBurI YO UO saIpoq sulyInI4¢ ‘(z96] addeiy) feztyss004w aq 0} payiodai usaq sey satoods ay) yey) SaqeoIpul aureU satoads B Iaqye (A )z “s3UlI YIM UOTRIDOSse UI paylodai Ajsnotaaid jou saisads, ZZ O€ ZO wnajyd splay uodo Z umouyun quasqe paieynuins -Il O£-0¢ 0c1-0¢ Ga uinajYq — SPlal} JO sadpa ‘spjaly uado 0¢ umouyun JUSS, PETTY al 09 6-2 19} Jeo] paqna snIsaNG Jopun II (A) winuidl DULO;OY IAT 0€ Ol's 19})1] Jeo] SNqodIs snuig Japun G (A) |Sninjnupss snying ‘pu Z 19qiI] Jeol psouisad Snulg Japun I | DAaINJONUDAND DINSSNY ‘pu Or 19411] Jeo] WINADYIIDS 4aIp Japun I )Snajuvsis snyixpdosng] ‘pu ¢ 19] Je9] SISUapDUD) DENS], Japun I (A) |Snuisad sn14p1007 ZE Z1-8 JOVI] Jeo] paqnd snod4an@ Jopun g (W) \2/nssn4 snsoydossa FY pu I 19}1I] Jeo] pSOUlsad SNUuig Jopun I (W) |2ds149 DyjaajayH 61 GI Jaq] Je9] psouisad SNUIg Japun I ‘ds sniipUuijioy Ol 8 Jaq] Jed] WNADYIIDS AaIp Japun I (IW) Snidpgio snjjaapyiUuDdD a6 6-L 19}1I] Jeo] paqnd snz4anG Japun P ((W) Dlaposnu pjiuDU py 09 ~ Jaq] Jes] = DUDIIAaWD snulj Jopun I |SNULAPOYIUDX SNILDBY quasaid pourwsajapun O SZ ¢ piuoylung —-biny, eau “pial Jo adpa I (IW) Winasial DULOJOY IAT, snjndog 1e3u Ol GI‘ZI unalYd ‘splalj Jo sodpa Z (W) Sninjoaur snqjixod 0Z 9-¢ unajyd splaly uado € (W) ;Wnpunuiun win dydoaT SI €-Z uinajYd splay uedo 9 snaalu snsoydossd H] TZ o'r pluoyjung —- BjNJag 1evIU “pjalj Jo adpa Z ‘ds sniaDpuijd4oD quasaid yoajja ou +II1 GE 0Z 01-Z wnalYd spjaly uado IZ (I) Sapvaso sniluspiDpy 09 SP 8-¢ uinajYd spay uado € (W) wnjiojiad uopsadoIAT GL 06 G1-S DOg ‘xadDD Spjaly uado € |DxaUuUuoIgns aqdIoN)D 0 1 €-C uina|Yd spjary uado v ping|vep aqdooiy St (G3 8-S uinayd spjary uado 6 SILBDIf DIIDATDD 0@ O€ L-€ wungYyd splay uodo P SLUISAAUUIDI SNILIDB quasaid paiejnwns +I] snjndog 1e3U 0c 09 0C-C uinod ‘splay JO sadpa VC (W) \2yofisuep vjnssny DAAISCE. 1e9U (63 €C 07 SI uina|Yd ‘splaly Jo sadpo 6 (W) SIO LOSI | snjndog 1e3u 0Z CZ €-Z piuoyiung ‘splay Jo sad pa v ((W) \Daulsniap Djnssny 0€ Sv €l piluoyjUung —_—-vbjNlag 1eaU ‘platy Jo adpa I |DdaID] AQAIOUT 0€ 001 SI DOd puiydd] snyy Japun I |SUIAIUUOD DIJAAJAH CZ 0ZI L‘9 piuoyiung spaty uado Z [STUDINIIWAIA DIADAD]D L CZ €-Z xadDD 4 spjay uado v |DadaUl) DIADAD]D ~snz41an@ Or iS OI-L unajyd 1eauU ‘Spialj JO sad po g AW) |Suassagnioanlf DjlUDU py juasaid PITHY +] (44 /WI9) (wd) (Ww) I9A00 uo1}v907] SsUulI JO saisads sa1oydoiods UOT} RIITIA adAy snipes ul gpurq jo 19]9WIeIp punoi3 Joaquin uo 1994 asealoul UyJpIM quoting JUvUIWIOpald jenuue adVIDAV asvIDAV sdULI sn3unyj Jo uONKoIIsse]o pue uoNdiosaq “| I1aV 1 12 THE CANADIAN FIELD-NATURALIST ih ee eg Se 4 Bes) FiGurE |. A segment of a ring of Type la-, photographed June 1, 1964. The diameter of the ring was 9.1 m and the band of killed vegetation was 1.5 m wide. The grass in the normal area on the left was Phleum pratense. The upright stakes marked the inner and outer edges of the previous year’s growth. The soil in this zone was littered with dead stems of the grass. II rings were quite similar to those described above. They observed the following fungus species on Type II rings: Agaricus campestris, A. tabularis, Calvatia cya- thiformis, C. fragilis, C. polygonia, Catastoma sub- terraneum, Lycoperdon curtisii, L. perlatum, Marasmius oreades, and Melanoleuca melaleuca. The first seven of these are non-mycorrhizal and the last three are mycorrhizal. However, their study was car- ried out in agricultural land in Eastern Colorado and, although not specifically mentioned in their report, it is unlikely that any trees were present. Therefore, it appears that stimulated rings are associated only with fungi which are growing saprophytically. Type III and Type 0 Rings. Rings of Type III showed no effect on herbacious vegetation and were recognized only by the ring of sporophores (Figure 3). Five fungus species were associated with rings of Type III (Table 1). Eleven species produced rings of sporo- phores under a canopy of trees where herbacious vege- tation was sparse or absent and the ground cover consisted of leaf litter (Table 1). Since it was impossi- ble to determine the effect of these fungi on vegeta- tion, they are grouped separately in Type 0. With the exception of Hygrophorus niveus, all of the rings of Type II] and Type 0 occurred under trees or near trees and most of the associated fungi have been reported to be mycorrhizal. Origin of Fungus Rings Fungus rings occur classically in man-made mea- dows such as lawns, pastures, parks, or golf courses. The abandoned fields in this study are similar to these classical sites. In all of the sites where rings occur, the soil has been cultivated and then left undisturbed for many years. The cultivation, with its attendant disrup- tion, aeration, and drying, apparently alters the endo- genous microbial flora in such a way that newly- established colonies of certain fungi grow symmetrically, much like colonies grow on sterile arti- ficial medium. In this study, three rings of Type I were followed from an early stage. They first appeared as small complete circles of killed vegetation. As they expanded past a 2 m diameter, the centres regenerated TOOHEY: FUNGUS FAIRY RINGS IN SOIL FIGURE 2. A ring of Type II with sporophores of Marasmius oreades, photographed in June, 1968. The diameter of the ring was 2.2 m. The flowers were Chrysanthemum leucanthemum and Ranunculus acris. The band of vegetation inside the ring of mushrooms was dark green. new vegetation and the affected areas became bands. One of these rings eventually produced sporophores of Russula densifolia and two have remained as Type I-. A great variety of fungi are capable of producing this phenomenon. Although most of them are basidiomy- cetes, the ascomycetes are represented by the genera Helvella, Morchella, and Tuber. Many types of basidi- omycetes have been reported on rings including agar- ics, puffballs, boletes, hydnums, and coral fungi. Rings are produced by both saprophytic and mycorrhizal species. The diversity of types suggests that any fungus species could produce the ring phenomenon if given the right conditions. In fact, it is possible that some of the rings on which sporophores have never been observed are caused by imperfect fungi or other pro- tists which do not form visible sporophores. Moreover, it is likely that rings of Type III- exist (no effect on vegetation, no sporophores). Rings of this type would not be detectable by macroscopic methods. An interesting feature of fungus rings is the failure of spores produced on the rings to colonize adjacent soil. In suitable years many rings produce over 100 sporophores, each of which releases very large numbers of spores. If these spores were successful in giving rise to new mycelium in adjacent soil, even at a very low frequency, the species would become dif- fusely distributed. In fact, the distribution remains rigidly confined to the ring-shaped pattern for periods as long as several hundred years (see next section). In this study, satellite rings were not observed around established rings and rings of a given species did not regularly occur in clusters. The failure of the spores to establish new mycelium in adjacent soil may be attrib- utable to a characteristic which Garrett (1963) des- cribed as the “low inoculum potential of fungi”. According to this concept, a critical quantity of metabolically-active fungal protoplasm must be pres- ent fora fungus to invade and colonize new substrate. The ring phenomenon provides evidence that fungal spores achieve the critical energy threshold with extremely low frequency. Growth and Death of Rings The average annual increase in radius for rings of 14 THE CANADIAN FIELD-NATURALIST FIGURE 3. Aring of Type III with sporophores of Hygrophorus niveus, photographed in October, 1970. The ring was |.8 m in diameter. Although the lighting appears non-uniform, there was no alteration of the vegetation. each species is given in Table 1. The expansion rate varied from 7 cm per year for Clavaria cinerea to 75 cm per year for Clitocybe subconnexa. The expan- sion rate was characteristic for each species. A vari- ability of = 15% appeared to be related to the annual rainfall but exact correlations were not made. Based on the diameter of a ring and the expansion rate, the approximate age of the ring can be estimated. Calcu- lated in this way, the oldest ring in this study was a ring of Type I-, which was 75 years old when it disappeared. This calculation has been applied to rings in Colorado, where rings of Agaricus tabularis were determined to be 250 years old (Shantz and Piemeisel 1917) and to rings in England where a ring of Clitocybe geotropa was determined to be nearly 700 years old (Smith 1957). In this study, few rings survived past an age of 50 years, and during the study 75 rings disappeared for various reasons. Many rings were killed by grass fires; no ring has ever been observed to revive after being burned over. Other rings disappeared when fields were ploughed. Many rings occurred in fields where forests were regenerating and they disappeared as the ground cover changed from grasses to forest litter. In contrast, some species continued to flourish under a forest cover, most notably Tricholoma irinum and Hygrophorus russula (Table 1, Type 0). Many rings were disrupted when they encountered obstacles such as rocks, roadways, or trees, and frequently there was no obvious cause for the disruption. Segments of many rings disappeared without apparent cause, leav- ing residual arcs: 15 rings disappeared completely in this way. An interesting feature of fungus rings is the uni- directional outward growth. Reverse (inward) growth does not occur and sporophores are usually produced only on the outer edge of the band. In the absence of specific data, it is possible only to speculate on the reason for the unidirectional growth of the mycelium. The two most obvious explanations are that the myce- lium depletes essential nutrients in the soil or that autotoxic chemicals are released. In rings with altered vegetation, there is a return to normal vegetation on the trailing edge of the band as the ring expands. In rings with suppressed vegetation, the return to normal involves a specialized plant suc- cession which varies somewhat from site to site. A 1983 description of the succession which occurred in one location has been published (Toohey ef al. 1965a). In rings with stimulated vegetation, the return to normal is simply a decrease in the green coloration and vigour of the plants. Chemical Ecology of Fungus Rings The fungus ring phenomenon demonstrates the dramatic effects that soil fungi can have on herbace- ous vegetation. The chemical mechanisms of these effects are not completely understood at present. The mechanism by which plants are killed on rings of Type I has been partially elucidated. From the soil of these rings a phenazine carboxylic acid-producing strain of Pseudomonas aureofaciens has been isolated, while in the adjacent non-ring soil the same bacterium occurred as a non-phenzine-producing variant (Too- hey et al. 1965a). The phenazine carboxylic acids have herbicidal properties (Toohey et a/. 1965c). When the phenazine carboxylic acid-producing strain of the bacterium was maintained in vitro, it slowly reverted to the non-producing type (Toohey ef al. 1965a). These findings suggest that the bactertum, which is ubiquitous in soil, is induced to synthesize the herbici- dal compounds when it is growing in association with the fungal mycelium in the soil of Type I rings. The testing of this hypothesis in vitro has been impeded by the inability to obtain pure cultures of the appropriate fungi. Numerous attempts to obtain cultures from the spores or from sporophore tissue have been unsuccessful. The plants on Type II rings have the features of increased nitrogen availability. The concentration of ammonia and nitrate in the soil of these rings has been reported to be higher than that in adjacent soil (Bayliss-Elliot 1926; Smith 1957; Shantz and Piemei- sel 1917) and it has been speculated that the fungus mycelium accelerates the decomposition of organic matter in the soil, making the nitrogen available (loc. cit.). This hypothesis is supported by the finding dis- cussed above that Type II rings are associated with saprophytic fungi. However, an alternative hypothe- sis involving enhanced nitrogen fixation must be con- sidered. Although no fungus has been found to have the ability to fix atmospheric nitrogen, the mycelium of several species of fungi has been shown to enhance N, fixation by nitrogen-fixing bacteria when the two were grown in mixed culture (reviewed by Jensen and Holm 1975). This phenomenon has been well docu- mented in pure cultures under laboratory conditions, but its significance under natural conditions has not been tested. It is possible that the association of spe- cific fungi with nitrogen-fixing bacteria or blue-green algae in the soil of Type II rings results in enhanced nitrogen fixation. This hypothesis might be tested in TOOHEY: FUNGUS FAIRY RINGS IN SOIL 15 vitro if pure cultures of Type II ring fungi were avail- able. Attempts to obtain such cultures have been unsuccessful. Fungus fairy rings provide circumscribed examples of the pronounced effects which soil fungi can have on herbacious vegetation. It is likely that similar effects occur under less well-defined conditions resulting in generalized suppression or stimulation of plant growth over large areas. An understanding of the mechanisms by which specific fungi affect plant growth could have relevance to practical agronomy. Thus, the fungal flora of soils might be intentionally altered to increase or suppress the growth of herba- cious plants. Acknowledgments The author acknowledges the assistance of the fol- lowing in identifying fungi: the late J.W. Groves, the late Mary E. Elliott, D. Malloch, R.A. Shoemaker, M. Eversons, and S.A. Redhead of the Biosystematics Research Institute, Department of Agriculture, Ottawa, and R.H. Petersen of the Department of Botany, The University of Tennessee, Knoxville. Literature Cited Bayliss-Elliott, J. S. 1926. Concerning fairy rings in pas- tures. Annals of Applied Biology 13: 277-288. Garrett, S. D. 1963. Soil fungi and soil fertility. Pergamon Press, N.Y. , Hawksworth, F. G. 1962. Fairy rings associated with Poly- porus confluens. American Midland Naturalist 68: 495. Jensen, V., and E. Holm. 1975. Associative growth of nitrogen-fixing bacteria with other micro-organisms. Jn Nitrogen fixation by free-living micro-organisms. Edited by W. D. P. Stewart. Cambridge University Press, Cam- bridge. pp. 101-119. Shantz, H. L., and R. L. Piemeisel. 1917. Fungus rings in eastern Colorado and their effect on vegetation. Journal of Agricultural Research 11: 91-245. Smith, J. D. 1957. Fungi and turf diseases. Journal of the Sports Turf Research Institute 9: 324-352. Toohey, J.I., C.D. Nelson, and G. Krotkoy. 1965a- Barren ring, a description and study of causal relation- ships. Canadian Journal of Botany 43: 1043-1054. Toohey, J .I., C. D. Nelson, and G. Krotkov. 1965b. Isola- tion and identification of two phenazines froma strain of Pseudomonas aureofaciens. Canadian Journal of Botany 43: 1055-1062. Toohey, J. I., C. D. Nelson, and G. Krotkoy. 1965c. Toxic- ity of phenazine carboxylic acids to some bacteria, algae, higher plants, and animals. Canadian Journal of Botany 43: 1151-1155. Trappe, J.M. 1962. Fungus associates of ectotrophic mycorrhizae. The Botanical Review 28:538-606. Withering, W. 1796. An arrangement of British plants, edi- tion 3, volume 4. Printed by M. Swinney, London. Received 5 August 1981 Accepted 20 December 1982 Island Biogeography of Seed Plants in Lake Nipigon, Ontario KEVIN P. TIMONEY Department of Botany, University of Wisconsin, Madison, WI. 53706 Present Address: Department of Botany, University of Alberta, Edmonton, Alberta T6G 2E9 Timoney, Kevin P. 1983. Island biogeography of seed plants in Lake Nipigon, Ontario. Canadian Field-Naturalist 97(1): 16-25. Islands supporting boreal forest in Lake Nipigon, Ontario, were sampled for presence of angiosperms and gymnosperms by a timed random walk of constant duration. Sample islands were manifestly different only in size (1.5-13.5 ha) and distance to mainland (1.0-10.5 km). Island species richness was correlated positively with area and negatively with distance. A new isolation index is proposed and shown to be a reliable predictor of island species number. It is suggested that propagules of many plant species on Lake Nipigon decrease in density as | / distance? froma source, and that the total number of propagules produced by a source is directly proportional to the source area. The observed species-distance relation suggests that small distances may havea significant effect on island species number. Near islands are populated by equal numbers of animal and wind-water dispersed plant species. Distant islands are characterized by a predominance of wind-water dispersed plants. Key Words: Angiosperms, dispersal, distance, gymnosperms, island biogeography, isolation, Lake Nipigon, Ontario, species-area relation. At what distance does an island’s isolation begin to affect higher plant species richness, and how does area per se affect species richness? Do islands that differ in area or isolation also differ in their types of propagule dispersal? Much research has been done on oceanic and habi- tat islands, and most of that work has centered on the fauna. Continental lake islands have received little attention (c.f. McNeill and Cody 1978). Recent work indicates that higher plant species richness may be affected by isolation no greater than 1.5-10 km (Nip- van der Voortet al. 1979; Crowe 1979; Linhart 1980). The islands in Lake Nipigon, probably 8,500 to 9,500 years old (Bryson et al. 1969), are most likely near species equilibrium. The lake’s islands, of varying size and isolation, providea test of the effects of short distances, and area, on seed plant species richness. Islands were selected stringently for habitat homoge- neity, and recent disturbance by fire, humans, or windstorms resulted in rejection of an island. By sam- pling islands equal amounts of time, the likelihood of encountering more habitats on larger islands was min- imized. The more thorough sampling of small islands provided a rigorous test of the effects of island area on species richness. The purpose of this study was to find out if distance and area are significantly related to the species rich- ness of seed plants on boreal forest islands. Description of the Study Area Lake Lake Nipigon les about 60 km north of Lake Superior inthe Thunder Bay District of NW Ontario. The lake’s approximate geographic center 1s 88° 30’W, 49° 52’N. Nipigon is nearly elliptic in outline, measur- ing 100 km long by 55 km wide, or about 4300 km? (Figure |). Average lake level is 263 m ASL (Ontario Dept. of Lands and Forests 1965). The lake lies within the Precambrian Shield (Zoltai 1965). Most of the islands and mainland are underlain by Late to Middle Precambrian diabase and related 89°10 49° 25" FIGURE |. Lake Nipigon, sites appear as dots. NW Ontario, Canada. Sample 1983 TABLE |. Forest vegetation dominants of the sample sites. Dominants Sub-dominants Trees Abies balsamea Picea mariana Betula papyrifera Picea glauca Thuja occidentalis Populus tremuloides Populus balsamifera Shrubs and Saplings of A. Sorbus decora Small Trees balsamea and Cornus stolonifera B. payrifera Taxus canadensis Alnus rugosa Salix humilis Salix phylicifolia Rubus idaeus Alnus crispa Sambucus racemosa Ribes glandulosum Seedlings of A balsamea and B. papyrifera Linnaea borealis Trientalis borealis Cornus canadensis Moneses uniflora Aralia nudicaulis Mitella nuda Pyrola secunda Rubus pubescens Fragaria vesca Ground Layer mafic igneous rocks (Ontario Geological Survey 1980). Numerous other bedrock types occur, but no sample sites were located outside the mafic igneous zone. Island and mainland sites (Figure |) were chosen using the following criteria: homogeneous dominant tree vegetation of Abies balsamea (Balsam Fir) and Betula papyrifera (White Birch) with Picea glauca (White Spruce) and Picea mariana (Black Spruce) (Table 1; also see Cooper 1913); island size between |.5 and I5 ha; absence of swamps, meadows, heaths, jack pine forest, and disturbance due to fires, humans, and windstorms; average slope not exceed- ing 25%; maximum elevation not exceeding 30 m. Bedrock outcrops, depressions, and steep areas were avoided. Differences in topography on the sample islands were so small that microclimatic differences were imperceptible. Sample islands thus differed manifestly only in area and isolation. Silty to sandy till overlies the bedrock on the major- ity of mainland and island sites. Stratified and non- stratified lacustrine deposits of clay, silt, and sand are associated with the till (Ontario Dept. of Lands and Forests 1965). Till depth varies from zero on exposed bedrock to an average of 2.5 m (Zoltai 1965). The present Lake Nipigon shoreline and lower elevations, once covered by the waters of Glacial Lake Kelvin, are dominated by lacustrine deposits (D. A. Fawcett, pers. comm.). The soils of the area are broadly classed as humoferric podzols, with rockland, eutric brunis- ols, and gray luvisols in the rocky and stony phase (Agriculture Canada 1977). Thin soils over bedrock are common. Timber use capability, an index of plant growth potential, ranges from good to fair for the TIMONEY: ISLAND BIOGEOGRAPHY OF SEED PLANTS i7/ sample sites (Ontario Ministry of Natural Resources 1976). The Lake Nipigon basin is enclosed on the north, east, and west sides by the higher land of the Central Plateau (Rowe 1972). The slope of the land 1s gradual in the south, with some exceptions (e.g., Nipigon River, Pijitawabik Bay, Tchiatang Bluffs, South Bay, Cook Point). In general, the topography is rolling and the relief slight, rarely exceeding 40 m above lake level. Numerous rivers and creeks drain into the lake from the north, east, and west. At Pipestone Bay in the extreme SE the lake is drained by the Nipigon River which flows into Lake Superior near the town of Nipigon. Hundreds of islands dot the lake, the vegetated ones ranging in size froma few m? to about 10,000 ha(e.g., Kelvin Island). Islands are well distributed through- out the lake, though sparse near the eastern shore. The greatest distance between islands does not exceed 10 km. Typical islands appear in Figure 2. The larger Glacial Lake Kelvin came to occupy the present Lake Nipigon basin with the retreat of the Laurentide Ice Sheet about 9,000 years B.P. (Zoltai 1965; Bryson 1969). Afterwards, water levels fluctu- ated due to periodic readvances of glacial ice, vertical crustal uplifts, and erosion of lake outlets (Zoltai 1965). All sample islands are similar in elevation, however, and therefore emerged from the receding lake waters at nearly the same time. Thus, all sample islands are of similar age. Field Methods Mainland exposures and islands ranging in size from 1.5 to 13.5 ha were sampled for presence of angiosperms and gymnosperms. Presence was deter- mined during a timed random walk. Islands were divided into four quadrants which delimited NE, NW, SE, and SW exposures. Mainland sites were divided into two exposures and were sampled chiefly to determine the species present for island colonization. Fifteen minutes search time was allotted to each island quadrant and mainland aspect. Total search time on any island was one hour. Islands smaller than 1.5 ha were too small to allow one hour of random search time without sampling ground already covered. I set an upper limit of 15 ha; above 15 ha, much nonsearch time was spent travelling between quadrants. I wrote a description of each island and mainland site immediately after sampling. The description treated the following: slope, width and nature of shoreline, moisture conditions, dominant species in the overstory, shrub and ground layers, presence of blowdowns, clearings, or depressions, and nature of the forest floor (e.g., whether moss or leaf-covered, prominence of rocks and downed trees). Descriptions 18 THE CANADIAN FIELD-NATURALIST FIGURE 2. Typical islands in Lake Nipigon, June 1979. were later used in deciding whether a site conformed to the experimental habitat type. The actual time spent on any island was at least 4-5 hours. During random walks, | often stopped search tim- ing to allow travel time in difficult walking on strand, downed trees, or tangles, and to allow time to record species and notes and to identify and collect plants. When 15 minutes of sampling time expired, sampling ceased until I entered a new quadrant. The walking route was random. Sampling time in deep forest, forest edge, and strand was about proportional to the area of the habitat type in each quadrant. The use of quadrants ensured stratified coverage and provided data on possible exposure effects. Species tallies by quadrant were repetitive for any island indicating that most species present were discovered (e.g., 57% of an island’s species number was present in a single quad- rant; the ratio of species per quadrant: total island species was unrelated to area; the average number of new species between the third and fourth quadrants equaled 3.6). The sampling method was not meant to be an exhaustive search forall higher plants ina study area; some species undoubtedly were not tallied. At least 28 species were excluded from the study for one of three reasons. Some species, e.g., most Carex spp., were impossible to identify in vegetative condi- tion. A number of voucher specimens of rare species was misplaced by an unnamed agency; unidentified Species in the lost batch were excluded. Ephemeral Vol. 97 species, which either faded before or became visible after July 20, were excluded (e.g., Calypso bulbosa). | recognized one hundred species, 91 of which occurred in the samples. The limited number of species recorded for the sites can be attributed to: choice of only homogeneous Abies- Picea- Betula forest, inex- haustive search method, and exclusion of vegetative graminoids, lost voucher specimens, and ephemerals. Analytical Methods Numbers of species per island quadrant, island, mainland exposure, and mainland site were deter- mined from field data. Frequency occurrence of each species for islands, mainland sites, and overall, and dispersal mechanisms are given in Appendix I. The area of each island was determined by planime- ter from Canada Map Office maps (scale = 1:50000, published 1959, ’66, 67, ’69). Distance to the mainland was measured from an island’s nearest shore to the nearest mainland shore. Islands 10 km? or larger were observed to support species numbers essentially iden- tical to the mainland, and thus treated as mainland. Thornton (1967) pointed out that simple distance to nearest neighbour ignores the contribution of other islands. He proposed that the sum or the average of distances of each island to every other island in an archipelago would provide a better index of isolation (c.f. Power 1972). Power (1972) tested two isolation indices, the latter type taking into account that near 1983 islands are more likely to contribute propagules to a recipient than are distant islands. For specified maxi- - : & k mum distances he estimated isolation as |= 1 - ¥ — i= It tm where m is in miles, and 1s are islands or mainland points. Power follows Darlington (1938) in assuming that propagule density varies inversely as the distance from the source, not inversely as the square of the distance as in this study. His isolation index, moreover, does not take into account the area of each stepping stone. The effective isolation of an island can be viewed as its distance to the mainland minus the contribution of any stepping stones (1.e., islands closer than the main- land which may contribute propagules), the latter islands effectively shortening the distance to the main- land. The stepping stone factor, Sj, was devised to approximate the effective isolation. Calculation of S; involved two variables: the dis- tance in km from the recipient to each stepping stone island (D1), and the area in km? of each stepping stone (Ai) estimated by planimeter. S 22) The effec- tive isolation in km thus= I = W(1-K), where W is the distance from recipient island to mainland in km, and S;. K (the correction factor for the effect of stepping stones) ranges between 0 and |, the least isolated having a K value of | and therefore an effective isola- tion of zero, 1.e., located on the mainland. Four assumptions underlie the stepping stone fac- tor. The first and most questionable assumption 1s that propagule density decreases inversely as the square of the distance from the source (c.f. Johnson and Raven 1970). Such exponential dispersal may hold for water and air-borne propagules, but uniform dispersal may hold for animal borne propagules (MacArthur and Wilson 1967). Secondly, the number of potential propagules was assumed proportional to the stepping stone area. Thirdly, the maximum dis- tance for inclusion of stepping stones equaled the distance to the mainland, with islands 10 km? and larger considered as mainland. Finally, dispersal was assumed equiprobable in all directions. Regressions of island species number and inde- pendent variables were carried out. An alternate isola- S | . My AN: tion factor using > (thus Sj= » (S) in place of = i= | (Darlington 1938) was regressed upon island species number. The effect of exposure (slope aspect) upon species number was tested by ANOVA. Site summar- les appear in Table 2, and regression results in Table 3. Results Island species richness was correlated positively TIMONEY: ISLAND BIOGEOGRAPHY OF SEED PLANTS 19 with area (p < 0.01), log. area (p < 0.05), and average species richness/ quadrant (p < 0.01), and negatively correlated with distance (p< 0.05) and isolation (p < 0.05). The log. of island species richness was correlated positively with area (p<0.05) and species richness/ quadrant (p<0.01), and negatively corre- lated with distance (p<0.05) and isolation (p<0.05). Species richness/ quadrant was correlated positively with island area (p<0.05). Isolation, distance, and the dance ISolation index were all strongly correlated with each other (p<0.01), and all were unrelated to island area. Mainland and island exposures had no effect on species richness (ANOVA: mainland F = 1.78, dfl-= 2, df2 = 7; island F = 1.25, dfl = 3, df2 = 80). Species with animal borne propagules were numer- ous on near islands, whereas wind-water dispersed spe- cies outnumbered animal dispersed species on distant islands (distance p<0.01, isolation p<0.01, ahionce index p<0.05). Dispersal type was unrelated to island area. (Table 3). Of the 91 plant species which occurred in the sam- ples, 31 were dispersed by wind and/or water, 39 by birds and/or mammals; 18 species used both wind- water and bird-mammal dispersal. The dispersal mechanism of three species could not be determined with certainty. Twenty-one of the 91 plant species found at the sample sites were restricted to islands; 14 species were restricted to the mainland, and nine of these 14 species depended on bird-mammal dispersal. Discussion The species richness of seed plants in Lake Nipigon is positively correlated with island area. Area per se as expressed through higher immigration rates and lower extinction rates of large islands may help account for the increase in species diversity with area. Immigration rate is not independent of area since large islands provide a larger catchment surface for propagules than small islands (the “sampling effect”; Power 1972). The number of wind-borne propagules landing per unit area should be the same on large and small islands. This constant immigration per unit area implies that more propagules will immigrate to larger islands. Water borne immigration should be more a function of island perimeter than area. Areas being equal, a long thin island with low shoreline would receive more propagules than a circular island with a steep margin. Larger islands also might offer more of any given resource to animals bearing propagules. Enhancement of immigration to large islands may occur for propagules dispersed by the Woodland Caribou (Rangifer caribou sylvestris Richardson) during open water migration on Lake Nipigon. Cari- bou swim rapidly and well for distances of 5-6 km (Jackson 1961), and local fishermen have observed 20 THE CANADIAN FIELD-NATURALIST Vol. 97 TABLE 2. Name, area, loge area, total species, loge species, species/ quadrant, ratio of species/ quadrant tog total species, distance, isolation, isolation assuming |/ Distance dispersal, and ratio of wind and water dispersed to bird and mammal dispersed plant species of island and mainland sites. Map coordinates are available upon request. @ < McKee 13.2 Gypsy 1.5 Pipestone OF] June 1.5 Big Tchiatang 3.3) Storm 5.9 Lightning 2.8 Lone 4.4 Little Russell 5.0 Middle Tichnor ell South Mclvor HS) Middle Mclvor Bey W.S. Mclvor 1.5 Windy Mclvor 5.4 Cress 1.6 Cove Ded} East August BES East Remote 3.8 West Remote 6.0 West August 12.4 South August 4.1 Mean 5.6 Cook-Col. NE NW Frenchman’s Head NE NE Queen Anne Point SE NW Tchiatang SE NW Cloud SE Point SE MAINLAND Distance ONAN FOCTON—]NAWNANA — e i s ie eh $3 eb t Le 2 Seeman’ Bare 2.58 44 3.78 27 0.61 0.41 33 3.50 21 0.64 DED 5) 39 3.66 20 0.51 0.41 34 3.53 19 0.56 1.19 24 3.18 12 0.50 1.77 25 32D 15 0.60 1.03 3] 3.43 7 0.55 1.48 23 3.14 1] 0.48 1.61 38 3.64 18 0.47 1.96 38 3.64 7 0.45 0.41 32 3.47 20 0.63 2.46 Sy 3.95 29 0.56 0.41 24 3.18 13 0.54 1.69 29 3.37 16 0.55 0.47 31 3.43 21 0.68 0.99 36 3.58 22 0.61 2.60 35 3.56 23 0.66 1.34 28 3.33 17 0.61 1.79 22 3.09 12 0.55 D252) 35 3.56 17 0.49 1.41 33 3.50 21 0.64 1.72 33 3.50 19 0.57 44 36 34 38 33 30 40 37 14 45 36 29 43 29 35 42 31 Isolation = ro) 3.2 Isolation using |/D Dispersal 21 ISLAND BIOGEOGRAPHY OF SEED PLANTS TIMONEY oney [esiadsiq XT0+T1 *VS 0 = ‘q/| 3ulsn ‘uonryosy XI1O0+ITT xv0+90 **SC 0 **9L 0 rp UOTIEIOS] SO OXG OES O — oO sete Ur **9S 0 **«VL 0 «#66 0 = 20UrISIG saizadg [e101 L00 6£ 0 0c 0 610 Ez juvIpeNnd /saisads x00 +70 910 c0'0 ct 0- 6¢ 0- +0S 0 == juvIpend /sarsads x00-9'¢ x00-9¢€ XL81 +7 9P- saisads tC O 910" +60 0- +90 0- 900 **88 0 = 307 Xe1 +69¢ xXC1- OLE x90 +70 saisads tc 0- 610- +80 0- #bV 0- £00 +*88 0 = =a [RIO] xXtb +797 Roly €10 910 910 810 LE0O- Lo0 Or'0 +SP 0 < "307 xso+l sl xXO0+EE XOI+0LC 80°0 610 80°0 110 €10- #tV 0 PO) +#95 0 = = Baly cn aes z oo = 2 5 2 2 5% a = ye ss 2 s g Et —_ 10 feb) j=} Hn w 2 2 oe 5 ~E é E 3 3 A : = Pe ; 1983 ‘10°90 = d @ queoyiusis 44 $600 = d @ quroryiusis , 6p6'0 = 100=4 M1 ‘eer 0 = 60:0 =d @ 1 10119 JPG] ‘XQ4+k = A ‘SUONR[IIIOD JURIIJIUSIS 10} Paist] SUOIZeNba UOISsaIdII ‘Sa[qelIeA juapuadapul puv JUapuadap OJ sanjeA (I)UOTe[AIO)D “¢ A1aVL 22 THE CANADIAN FIELD-NATURALIST them swimming between the large Kelvin and Sha- kespeare Islands, an open water crossing of about 10 km. Caribou require plentiful amounts of lichens and browse for survival; thus, their crossings must be more frequent between large islands. The results, however, do not indicate a general enhancement of animal borne plant immigration to the larger islands. Island area on Lake Nipigon bears no relation to the type of dispersal used by the plant species. Rather, distance appears to govern dispersal type, with bird- or mammal-dispersed species com- prising a smaller percentage of the flora on distant islands than on near islands. The prevailing dispersal type of species on distant islands is wind-water. This is understandable in that animal (i.e., dispersal agent) species richness has been shown to decrease with isola- tion, whereas wind and water can provide transport subject only to the adaptations of the plant propagule and the decrease in propagule density from the source. Extinction rates are accepted as inverse to island size, but MacArthur and Wilson (1967) hypothesized that small islands may possess exceptionally high area-independent extinction rates. Whether Lake Nipigon’s small islands exhibit the special case of area-independent extinction can only be shown exper- imentally. Area-independent rates or not, small popu- lations may suffer high extinction rates (Pickett and Thompson 1978). Island populations of many species on Lake Nipigon often comprised only one to a few individuals, e.g., Calypso bulbosa, Clintonia borealis, Cornus canadensis, Pyrola secunda, P. virens, and Viola nephrophylla. The significant decrease in plant species richness in Lake Nipigon with increasing isolation and distance may result from lower immigration rates to the distant islands. For many species on Lake Nipigon, the small isolation distances must present little barrier to dis- persal. Seeds of Orchidaceae and Pyrolaceae are eas- ily blown hundreds of km (van der Pijl 1969). Yet many species with dust-like seeds are saprophytes (e.g., Corallorhiza trifida, Monotropa uniflora), mycophytes, or parasites, and the special substrates they require often limit their distribution. Other wind- borne seeds are less easily dispersed but also less exact- ing of habitat. Ldve (1963) lists average dispersal lim- its in km fora variety of species, some genera of which are present on Lake Nipigon: Fraxinus (0.03), Abies (0.09), Picea (0.3), Betula (1.6), Taraxacum (10) and Lycopodium (330). Some widespread species on Lake Nipigon can be dispersed by water in addition to their normal wind dispersal, e.g., Betula papyrifera, Salix spp., Populus tremuloides and P. balsamifera(USDA 1974). During the study I observed thousands of seeds several km from land, such as those of Paper Birch and Balsam Poplar afloat in water, and the comose seeds of Fireweed aloft in air. Vol. 97 Dispersal over ice from December through April must also contribute to immigration rates. For plants with winter-persistent fruits, such as birch and spruce, ice could provide a ready avenue for wind dispersal. Isolation distances for bird-dispersed propagules, such as Clintonia borealis berries dispersed by Ruffed Grouse (Ridley 1930), appear trivial on first inspec- tion. Mere flight of a bird between source and recip- ient islands, however, need not connote successful dispersal. The bird must first eat the fruit (or the fruit or seed become attached to the bird), the seeds must be viable, remain with the bird during the flight, and later be excreted on the recipient island in the proper habi- tat. Some herbivores, such as White-throated Spar- rows, were very active overwater fliers, but others, such as Ruffed Grouse, appeared sedentary. Success- ful bird dispersal, as with any dispersal, is only one factor determining a plant’s distribution; the propa- gule must establish itself and persist until other suc- cessful introductions. The above discussion emphasizes the roles of area per se and the sampling effect in determining island species number. Other explanations exist, such as habitat diversity. Though sample time was held con- stant over all island sizes, a greater variety of micro- habitats may have been encountered on large islands simply because large islands can contain more habi- tats than small islands. Although no between-island habitat differences were evident, the importance of habitat must not be overlooked. Koopman(1958) has shown that the critical factor in determining the bat species number of southern Caribbean islands is not area or isolation but vegetation. Rate of establishment is a function of the number of sites suitable for germi- nation (Harper 1965) and the rate of immigration. On many islands in Lake Nipigon, mesic herbs are res- tricted to forest edge. Establishment and persistence in the moss, acidic humus, and nutrient-poor soil may be enhanced in the brighter light of the forest edge. Competition from mosses may play a role in deter- mining seed plant species richness on islands in Lake Nipigon. Data and observations on the frequency of forest herbs, shrubs, and seedlings in the cushion moss mats (dominated by Dicranum fuscescens Turn., D. rugosum (Hoffm.) Brid., Hylocomium splendens (Hedw.) BSG, and Pleurozium schreberi (Brid.) Mitt.) of Nipigon’s islands indicate that common higher plants are able to grow in the moss mats. Whether occasional and rare species on Lake Nipigon owe their scarcity in part to moss competition can only be answered by more detailed study. Moss- seed plant competition on Lake Nipigon would not be surprising in that mosses and lichens inhibit estab- lishment of pine, spruce, shrubs and herbs elsewhere in the boreal forest (Braun-Blanquet 1932; Wilde and Krause 1960; Savile 1963; Walter 1973). 1983 Exposure did not affect species richness on either islands or mainlands, due perhaps to the low eleva- tions and gentle slopes of the sample sites. Conclusions The results indicate that seed plant species richness of Lake Nipigon islands is affected both by area and by isolation distances less than 10 km. The larger catchment surface of large islands may permit higher immigration rates than those of small islands. Lower species richness on distant islands may result from lower immigration rates than those of near islands. Island area in Lake Nipigon bears no relation to plant dispersal type. Wind-water dispersal is the primary means by which plants disperse to distant islands. Near islands are populated by equal numbers of animal and wind-water dispersed plant species. Acknowledgments lam indebted to Grant Cottam and George La Roi, who provided helpful criticism and encouragement, and to Rebecca Filer, whose assistance early in the fieldwork was invaluable. I thank John Shepherd, Chris Jones, Al Swain, Cynthia Williams, Barbara Reynolds, Don Waller, Ed Beals, Virginia Kline, Gary Breckon, and anonymous reviewers for their insight- ful comments. Fred Hermann kindly identified the mosses. Thanks to the Ontario Ministry of Natural Resources and the University of Wisconsin Herba- rium curators for their cooperation. The Davis Fund committee provided financial support for travel. Literature Cited Agriculture Canada. 1977. Soils of Canada. Supply and Services Canada, Ottawa. Braun-Blanquet, J. 1932. Plant Sociology. McGraw-Hill, INeYe Bryson, R. A., W. M. Wendland, J. O. Ives, and J.T. Andrews. 1969. Radiocarbon isochrones on the disinte- gration of the Laurentide ice sheet. Arctic and Alpine Research 1I(1): 1-14. Cooper, W.S. 1913. The Climax Forest of Isle Royale, Lake Superior, and its development. Botanical Gazette 55: 1-44, 115-140, 189-235. Cooper, W. S. 1922. The ecological life history of certain species of Ribes and its application to the control of the white pine blister rust. Ecology 3: 7-16. Crowe, T. M. 1979. Lots of weeds: insular phytogeography of vacant urban lots. Journal of Biogeography 6: 169-181. Darlington, P. J. Jr. 1938. The origin of the fauna of the Greater Antilles, with discussion of dispersal of animals over water and through air. Quarterly Review of Biology 13: 274-300. de Vlaming, V., and V. W. Proctor. 1968. Dispersal of aquatic organisms; viability of seeds recovered from the droppings of captive killdeer and ducks. American Jour- nal of Botany 55(1): 20-26. TIMONEY: ISLAND BIOGEOGRAPHY OF SEED PLANTS 23 Jackson, H. H. 1961. Mammals of Wisconsin. University of Wisconsin Press, Madison, WI. Linhart, Y. B. 1980. Local biogeography of plants on a Caribbean atoll. Journal of Biogeography 7: 159-171. Love, D. 1963. Dispersal and Survival of Plants. pp. 189-205, In: North Atlantic Biota and Their History. Edited by A. Love and D. Léve. MacMillan, N.Y. MacArthur, R. H. and E. O. Wilson. 1967. The Theory of Island Biogeography. Princeton University Press, Prin- ceton, N.J. McNeill, J., and W. J. Cody. 1978. Species-area relation- ships for vascular plants of some St. Lawrence River islands. Canadian Field-Naturalist 92(1): 10-18. Nip-van der Voort, J., R. Hengeveld, and J. Haeck. 1979. Immigration rates of plant species in three Dutch polders. Journal of Biogeography 6: 301-308. Ontario Department of Lands and Forests. 1965. Thunder Bay Surficial Geology. MS 265. Toronto, Ontario. Ontario Geological Survey. 1980. Geological Highway Map, Northern Ontario. Ontario Geological Survey, Map 2440. Ontario Ministry of Natural Resources. 1976. Ontario Land Inventory. Land Inventory Unit, Richmond Hill, Ontario. Pickett, S.T. A., and J. N. Thompson. 1978. Patch dynamics and the design of nature reserves. Biological Conservation 13(1): 27-38. Power, D. M. 1972. Numbers of bird species on the Cali- fornia islands. Evolution 26: 451-463. Praeger, R.L. 1911. Phanerogamia and Pteridophyta. Proceedings of the Royal Irish Academy. 31(1): part 10. Ridley, H. N. 1930. Dispersal of Plants Throughout the World. L. Reeve and Co., Ashford, Kent. Rowe, J. S. 1972. Forest Regions of Canada. Department of the Environment, Canadian Forestry Service Publica- tion 1300, Ottawa, Ontario. Savile, D. 1963. Factors limiting the advance of spruce at Great Whale River, Quebec. Canadian Field-Naturalist 77(2): 95-97. Scoggan, H.J. 1978. The Flora of Canada. National Museum of Natural Sciences, Publications in Botany, No. 7(1-4), Ottawa, Ontario. Thornton, I. B. 1967. The measurement of isolation on archipelagos, and its relation to insular faunal size and endemism. Evolution 21: 842-849. USDA Forest Service. 1974. Seeds of Woody Plants in the United States. Agriculture Handbook 450, USDA, Washington, D.C. van der Pijl, L. 1969. Principles of Dispersal in Higher Plants. Springer-Verlag, N.Y. Walter, H. 1973. Vegetation of the Earth. Springer-Verlag, N.Y. Wilde, A.,and H. H. Krause. 1960. Soil-forest types of the Yukonand Tanana Valley in sub-arctic Alaska. Journal of Soil Science |1: 266-274. Zoltai, S. C. 1965. Glacial features of the Quetico-Nipigon Area, Ontario. Canadian Journal of Earth Sciences 2: 247-269. Received 12 December 1980 Accepted 30 November 1982 24 THE CANADIAN FIELD-NATURALIST Vol. 97 APPENDIX |. Recognized species found at the sample sites, their site frequencies, and probable dispersal mechanisms. Nomenclature follows Scoggan 1978. Species Islands Abies balsamea 1.00 Acer spicatum 0.33 Achillea millefolium 0.38 Actaea rubra 0.05 Agrostis perennans 0.57 Alnus crispa 0.76 Alnus rugosa 0.43 Amelanchier sanguinea 0.29 Anaphalis margaritacea 0.05 Aralia nudicaulis 0.45 Arctostaphylos uva-ursi 0.05 Aster macrophyllus 0.00 Betula papyrifera 1.00 Bromus ciliatus 0.05 Calamagrostis canadensis 0.86 Cardamine pensylvanica 0.24 Carex disperma 0.43 Clintonia borealis 0.05 Coptis trifolia 0.14 Corallorhiza trifida 0.00 Cornus canadensis 0.57 Cornus stolonifera 0.95 Deschampsia caespitosa and D. flexuosa 0.19 Diervilla lonicera 0.00 Eleocharis compressa 0.00 Epilobium angustifolium 0.95 Fragaria vesca 0.43 Fragaria virginiana 0.00 Galium triflorum 0.14 Gaultheria hispidula 0.05 Goodyera repens 0.10 Hierachloe odorata 0.05 Hydrocotyle americana 0.00 Ledum groenlandicum 0.14 Linnaea borealis 0.38 Lonicera canadensis 0.00 Lonicera hirsuta 0.00 Maianthemum canadense 0.24 Mentha arvensis 0.48 Mertensia paniculata 0.05 Mitella nuda 0.29 Moneses uniflora 0.62 Monotropa uniflora 0.24 Oryzopsis asperifolia 0.00 Phalaris arundinacea 0.19 Physocarpus opulifolius 0.14 Picea glauca 0.95 Picea mariana 0.71 Pinus strobus 0.00 Poa glauca 0.62 Poa palustris 0.52 Polygonum sp. 0.43 Populus balsamifera 0.86 Mainlands 1.00 0.71 0.00 0.57 0.00 0.86 0.71 0.43 0.00 1.00 0.00 0.71 1.00 0.00 1.00 0.00 0.57 1.00 0.57 0.43 1.00 0.71 0.00 0.57 0.14 0.57 0.29 0.29 1.00 0.43 0.14 0.00 0.14 0.00 1.00 0.71 0.29 1.00 0.14 0.86 1.00 0.43 0.43 0.14 0.00 0.00 0.86 1.00 0.14 0.43 0.00 0.14 0.71 Overall 1.00 0.43 0.29 0.18 0.43 0.79 0.50 0.32 0.04 0.57 0.04 0.18 1.00 0.04 0.89 0.18 0.46 0.29 0.25 0.11 0.68 0.89 0.14 0.14 0.04 0.86 0.39 0.07 0.36 0.14 0.11 0.04 0.04 0.11 0.54 0.18 0.07 0.43 0.39 0.25 0.46 0.57 0.29 0.04 0.14 0.11 0.93 0.79 0.04 0.57 0.39 0.36 0.82 Probable Dispersal! wind*> Red Squirrel? birds?, water? wind, birds? wind, water, birds birds? birds?, wind?, water? wind, water? wind, water? birds wind? birds (e.g., Pine Grosbeak, Red-bellied woodpecker) birds (e.g., Blue Grouse), mammals wind?, birds? wind, water2>4, birds birds (e.g., Crow) birds?, wind?, water? water?, birds? birds?, water?, mammals? Ruffed Grouse gravity? wind3 birds (e.g., Pine Grosbeak, Crow) birds (e.g., Crow*, American Redstart) wind’ 2 birds (e.g., ducks and Killdeer®) wind? birds (e.g., Crow, Starling, Eastern Kingbird) birds (e.g., Crow) birds, mammals birds? wind wind?, water? birds? birds (e.g., Rock Ptarmigan), reindeer mammals (e.g., humans?, deer, hare), birds birds? birds (e.g., Grey Vireo, Red-eyed Vireo) birds (e.g., Magpie) water gravity? gravity? wind wind? birds? birds (e.g., Killdeer, Mallards®, Reed Bunting), water, wind wind?3 wind>>, rodents?, birds?, water2 wind>>, rodents?, birds?, water? wind>>, birds?, water? wind?, water? birds?, wind?, water? birds (e.g., Killdeer and ducks®) wind, water? Continued 1983 TIMONEY: ISLAND BIOGEOGRAPHY OF SEED PLANTS 25 APPENDIX |. Concluded. Species Islands Mainlands Overall Probable Dispersal! Populus tremuloides 0.81 0.86 0.82 wind, water4 Potentilla anserina 0.29 0.14 0.25 water, mammals, birds (e.g., domestic goose) Potentilla norvegica 0.86 0.29 0.71 birds (e.g., domestic chicken), water? Potentilla palustris 0.10 0.00 0.07 water?, birds? Primula mistassinica 0.48 0.00 0.36 wind? Prunus pensylvanica 0.14 0.43 0.21 birds (e.g., Crow, Robin‘4), mammals Pyrola secunda 0.67 0.71 0.68 wind Pyrola virens 0.43 0.29 ~ 0.39 wind Ranunculus pensylvanicus 0.29 0.86 0.43 birds?, water? Rhamnus alnifolia 0.00 0.14 0.04 birds? Ribes glandulosum 0.95 0.86 0.86 birds Ribes hudsonianum 0.19 0.00 0.14 birds Ribes lacustre 0.00 0.57 0.14 birds Ribes oxyacanthoides 0.43 0.29 0.39 birds (e.g., Catbird, towhee, Blue Jay, Cedar Waxwing, Robin, etc.)8 Ribes triste 0.00 0.14 0.04 birds, reindeer Rosa acicularis 0.38 1.00 0.54 birds (e.g., Prairie Chicken), mammals Rubus idaeus : 0.90 0.43 0.79 birds (e.g., Crow, Magpie) Rubus pubescens 0.33 0.86 0.46 birds Salix humilis 0.90 0.57 0.82 wind, water4 Salix phylicifolia 0.67 0.86 0.71 wind, water4 Sambucus racemosa 0.52 0.14 0.43 birds (e.g, Crow, Robin, Red-headed Woodpecker) Schizachne purpurascens 0.05 0.00 0.04 birds?, wind?, water? Sium suave 0.05 0.00 0.04 water Solidago graminifolia 0.67 0.00 0.50 wind?, water? Sorbus decora 0.95 1.00 0.96 birds, mammals Streptopus roseus 0.00 0.86 0.21 birds? Taraxacum sp. 0.19 0.00 0.14 wind, birds, water Taxus canadensis 0.90 0.43 0.79 birds Thuja occidentalis 1.00 0.86 0.96 wind?, birds? Trientalis borealis 0.81 1.00 0.86 mammals? Typha latifolia 0.10 0.14 0.11 wind’, water Urtica dioica 0.10 0.00 0.07 wind, birds (e.g., Magpie), mammals Vaccinium angustifolium 0.14 0.29 0.18 birds, humans? Vaccinium myrtilloides 0.10 0.43 0.18 birds, humans? Viburnum edule 0.29 1.00 0.46 birds Viola macloskeyi 0.19 0.14 0.18 birds?, ants? Viola nephrophylla 0.10 0.00 0.07 birds?, ants?, water? Viola renifolia 0.14 0.86 0.32 birds?, ants? 'Source is Ridely (1930) unless otherwise noted 2Personal observation 3van der Pijl (1969) 4USDA Forest Service (1974) 5Love (1963) 6de Vlaming and Proctor (1968) 7Praeger (1911) 8Cooper (1922) Some dispersal agents listed are not present at Lake Nipigon, but related species are, e.g., domestic goose — Canada Goose. “2” denotes that dispersal agent is documented for a plant species related to that at Lake Nipigon. Distribution sur les coniféres des lichens appartenant aux Genres Bryoria, Alectoria, Usnea et Ramalina sur la Cote Nord et la Céte Sud du Golfe Saint-Laurent M. LAFLAMME-LEVESQUE!, J. M. PERRON! et L. JOBIN? 'Département de biologie, Faculté des sciences et de génie, Université Laval, Québec GIK 7P4 2Laboratoire de recherches forestiéres des Laurentides, Environnement Canada, Ste-Foy, Québec G1V 4C7 Laflamme-Levesque, M., J. M. Perronet L. Jobin. 1983. Distribution sur les coniféres des lichens appartenant aux genres Bryoria, Alectoria, Usnea et Ramalina sur la Céte Nord et la Céte Sud du Golfe Saint-Laurent. Canadian Field- Naturalist 97(1): 26-32. Seventeen species of corticolous fruticose lichens have been identified in various conifer stands on the St-Lawrence north shore, some of the Mingan Islands, on Bonaventure Island, and in the Gaspé Peninsula. The genus Bryoria is represented by 9 species, the genus A/ectoria by 1, the genus Usnea by 4, and the genus Ramalina by 4 species. The data on the distribution of the lichens in relation to the substratum are compared with similar information obtained previously on Anticosti Island. Key Words: Québec, lichens, Bryoria, Alectoria, Usnea, Ramalina. Dix sept espéces de lichens fruticuleux corticoles ont été dénombrées dans différents peuplements conifériens de la Cote Nord du Saint-Laurent, de quelques iles en Minganie, de l’ile Bonaventure et de la Gaspésie. Le genre Bryoria est représenté par 9 espéces, le genre Alecroria par |, le genre Usnea par 4, et le genre Ramalina par 4 especes. Les données recueillies quant a la distribution des lichens sur le substrat sont comparées avec les informations obtenues précédemment a ile d’ Anticosti. Mots Clés: Québec, lichens, Bryoria, Alectoria, Usnea, Ramalina. Les connaissances sur la composition lichénique —_ permettra de comparer ces 2 milieux en regard du réle des régions insulaires de l’est du Québec se limitent qu’ils pourraient jouer lors de la ponte de linsecte. aux travaux de Lepage (1972), Grondin et Melangon (1978) aux iles de Mingan, de Reilly(1972) auxilesde Materiel et Méthodes la Madeleine et de Laflamme-Levesque et al. (1979) a Le territoire couvert par cette étude (Figure |) com- ‘ile d’Anticosti. De part et d’autre de ces iles, les — prend, sur la Céte Nord du Saint-Laurent, la région régions cotiéres bordant le fleuve Saint-Laurent n’ont —située entre Baie-Comeau et Havre-Saint-Pierre. été soumises a aucune étude quant a la végétation Parallélement sur la rive sud du fleuve, le territoire lichénique corticole. D’aprés ces auteurs, parmi les étudiéen Gaspésie s’étend de Sainte-Anne-des- Monts lichens fruticuleux qui colonisent l’écorce des coni- a Percé. De plus, quelques iles de la Minganie, la feres de ces iles, les genres Alectoria, Bryoria, Evernia, Grande Ile, l’ile Niapisca, Pile Eskimo, et Pile Bonav- Ramalina et Usnea sont représentés. Selon Jobin — enture en Gaspésie ont fait l’objet de notre recherche. (1973), certains d’entre eux constitueraient un site de Les données ont été recueillies au cours de l’été ponte pour Lambdina fiscellaria fiscellaria (Guén.). 1978, dans 45 places-échantillons sur la Cote Nord et Les observations qu’ila effectuées aucoursdesannées 40 en Gaspésie. Les anciens sites de pullulations de 1971-1972 a Vile d’Anticosti indiquent que les oeufs = Lambdina fiscellaria fiscellaria, la présence de sapi- sont déposés, dans 90% des cas, dans leslichens‘pend- _niére a maturité et la proximité de grandes riviéres ants’ des vieux coniféres. Une étude a démontré que __constituaient nos principaux criteres dans le choix ces lichens se composent principalement de9 espéces _d’une place-échantillon. Un relevé de la composition du genre Bryoria et d’une espéce du genre Alectoria. _ forestiére, de sa densité, de lage des arbres, un échan- Les données recueillies ont démontré qu'il existe une __ tillonnage des lichens fruticuleux corticoles ainsi relation entre l'intensité de recouvrement lichénique —_qu’une évaluation du recouvrement des coniferes par du substrat, l’'4ge des arbres et la densité forestiére ces lichens, constituaient les données accumulées a (Laflamme-Levesque et al. 1979). chaque place-échantillon. Les techniques utilisées Sachant que les relevés épidémiologiques de Lamb- __ pour recueillir ces informations et déterminer la den- dina fiscellaria fiscellaria sont concentrés dans des sité lichénique ont été les mémes que celles utilisées a régions insulaires et maritimes (Carroll 1956; de Gryse Yile d’Anticosti, lors d’une recherche antérieure et Sched! 1934), une meilleure connaissance de la (Laflamme-Levesque et al. 1979). composition lichénique corticole et de sa densité, sur L’identification des lichens a été faite d’apres l’ou- la Cote Nord du Saint-Laurent et en Gaspésie, nous —_ vrage de Brodo et Hawksworth (1977), pour les es- 26 1983 1St-Nicolas 2Franquelin 3.Godbout 4 Pointe-des-Monts 5. Havre-St-Pierre 6. Baie-St-Laurent 7Ste-Anne-des-Monts 8 Riv Darmouth 9.Parc Forillon 10.Gaspe 11 Perce 12 Parc de la Gaspesie M. LAFLAMME-LEVESQUE, PERRON ET JOBIN: DISTRIBUTION DES LICHENS 27 /le Nt ICOg ty FIGURE |. Localisation des régions étudiées sur la Céte Nord du Saint-Laurent et en Gaspésie. péces appartenant aux genres Alectoria et Bryoria. L’étude des Usnéacées a nécessité lutilisation du tra- vail de Dahl et Krog (1973). Enfin, pour les espéces du genre Ramalina, notre principale référence a été celle de Hale (1969). Les especes décelant des problémes d’ identification ont été vérifiées par I1.M. Brodo. Les spécimens sont déposés a lherbier Louis-Marie de l'Université Laval (QFA). Résultats et Discussion Composition et Distribution Lichénique Du point de vue composition lichénique, les coni- feres de la Cote Nord du Saint-Laurent et de la Gaspe- sie sont colonisés par les lichens fruticuleux apparte- nant principalement aux genres Bryoria, Alectoria, Usnea et Ramalina. Genre Bryoria Le genre Bryoria, subdivisé en sections selon les caracteres morphologiques, anatomiques et chi- miques des espéces (Brodo et Hawksworth 1977), est représenté par 9 especes sur le territoire étudié. I] s'agit, pour la section Bryoria, de Bryoria trichodes (Michx.) Brodo et D. Hawksw., Bryoria fuscescens (Gyeln.) Brodo et D. Hawksw., et Bryoria lanestris (Ach.) Brodo et D. Hawksw.; pour la section Divari- catae, de Bryoria simplicior (Vain.) Brodo et D. Hawksw., Bryoria furcellata (Fr.) Brodo et D. Hawksw. et Bryoria tenuis (Dahl) Brodo et D. Hawksw.; et pour la section Implexae, de Bryoria nadvornikiana (Gyeln.) Brodo et D. Hawksw., Bry- oria capillaris (Ach.) Brodo et D. Hawksw. et Bryoria salazinica Brodo et D. Hawksw. Dans la section Bryoria, les espéces les plus fréquemment rencontres sont B. trichodes et B. fus- cescens,; B. lanestris se retrouve plus rarement. B. trichodes est représenteé sur tout le territoire étudié par 2 sous-espéces, B. trichodes ssp. trichodes et ssp. americana (Mot.) Brodo et D. Hawksw. (Figure 2, A et B); cette derniére est la plus courante. B. trichodes occupe invariablement les 3 niveaux du substrat étudié, soit le sommet, la région médiane et la base des coniféres. I] colonise indifféremment Abies balsamea (L.) Mill., Picea glauca (Moench) Voss et Picea maria- na (Mill.) B.S.P. Les peuplements conifériens de la Céte Nord du Saint-Laurent et de la Gaspésie recélent labondance de B. fuscescens (Figure 2 C) de haut en bas du substrat. Enfin, B. /anestris, quoique faible- ment représenté, colonise les 2 rives du fleuve (Figure 2 D). Cette espéce se fixe sur le tiers supérieur des 28 THE CANADIAN FIELD-NATURALIST Vol. 97 FiGureE 2. Distribution des espéces de la section Bryoria du genre Bryoria sur la Cte Nord du Saint-Laurent et en Gaspésie: (A) Bryoria trichodes ssp. trichodes; (B) Bryoria trichodes ssp. americana; (C) Bryoria fuscescens; (D) Bryoria lanestris. arbres 2.5 fois plus souvent que sur le tiers inférieur. Cette observation concorde aussi avec le fait qu’il croit, sur 90% des arbres étudiés, en association avec B. simplicior, espece qui semble aussi avoir un grand besoin de lumiére (Brodo et Hawksworth 1977). Les lichens appartenant a la section des Divaricatae sont représentés par B. simplicior, B. furcellata et B. tenuis. Certaines conditions d’éclairement pourraient expliquer la présence conjointe de B. simplicior et B. furcellata, dans 85% des échantillons recueillis. Ces 2 espéces montrent une préférence pour de bonnes con- ditions de lumiere, colonisant ainsi le tiers supérieur des arbres, dans des proportions de 3:1 et 4:3 par rapport au tiers inférieur. Le matériel récolté a Vile d’ Anticosti (Laflamme-Levesque et al. 1979) démon- trait aussi cette particularité pour B. furcellata, mais de fagon plus évidente avec un rapport de 9:1. Leur présence est notée sur tout le territoire, a exception des iles de Mingan (Figure 3, A et B). Rarissime, B. tenuis n’a été identifié que sur le sapin a Havre-Saint- Pierre (Cé6te Nord) et au Parc Forillon (Gaspésie) (Figure 3 B). B. nadvornikiana, B. capillaris et B. salazinica illus- trent la section Implexae. Rare a Vile d’Anticosti (Laflamme-Levesque et al. 1979), tous les peuple- ments forestiers étudiés sur la Cote Nord du Saint- Laurent et en Gaspésie sont marqués par la présence de B. nadvornikiana (Figure 3 C). Sur le territoire étudié, il n’y a pas d’exigences particuli€res de cette espéce face aux conditions d’éclairement. Sur la Cote Nord et en Gaspésie, la distribution de B. capillaris (Figure 3 D) sur les coniféres confirme les résultats obtenus a ile d’Anticosti (Laflamme-Levesque et al. 1979) selon lesquels, il existe une préférence pour les régions les mieux éclairées du substrat. Seulement 2 relevés lichéniques effectués a l’ile d’ Anticosti permet- tent de signaler la présence de B. salazinica. Absente en Gaspésie, cette espéce n’est notée sur la Cote Nord du Saint-Laurent qu’a Godbout (Figure 3 C), sur la méme essence forestiére, le sapin. Genre Alectoria Une seule espéce représente le genre Alectoria, Alectoria sarmentosa (Ach.) Ach. ssp. sarmentosa. Fréquente sur la Céte Nord du Saint-Laurent, elle est plus rarement trouvée en Gaspésie (Figure 4). Con- 1983 M. LAFLAMME-LEVESQUE, PERRON ET JOBIN: DISTRIBUTION DES LICHENS 29 FIGURE 3. Distribution des espéces des sections Divaricatae et Implexae du genre Bryoria sur la Cote Nord du Saint-Laurent et en Gaspésie: (A) Bryoria simplicior; (B)(® ) Bryoria furcellata; (0) Bryoria tenuis; (C)(®) Bryoria nadvornikiana; (0) Bryoria salazinica; (D) Bryoria capillaris. trairement aux données recueillies a Vile d’Anticosti (Laflamme-Levesque et al. 1979), le mode de réparti- tion de cette espéce sur le substrat montre une préférence pour la partie ombrageée des arbres, se fi- xant ainsi sur le tiers inférieur des coniféres, dans des proportions de 2:1 par rapport au tiers supérieur. FIGURE 4. Distribution du genre A/ectoria sur la Céte Nord du Saint-Laurent et en Gaspésie: (®) Alectoria sar- mentosa ssp. sarmentosa, race KCt+; (0) Alectoria sarmentosa ssp. sarmentosa, race KC” Seules les iles de la Minganie, Vile Bonaventure et le Parc de la Gaspésie possédent en plus la race chimique KC- (Figure 4). Genre Usnea Les lichens fructiculeux corticoles appartenant au genre Usnea sont représentés sur le territoire étudié par 4 espeéces: Usnea filipendula Stirt., Usnea longis- sima Ach., Usnea sorediifera (Arn.) Lynge s. lat. et Usnea trichodea Ach. Les peuplements conifériens de la Cote Nord du Saint-Laurent et de la Gaspésie sont tous colonisés par U. filipendula (Figure 5A). Sur lécorce de quelques sapins seulement se sont fixés U. ongissima et U. sorediifera (Figure 5B). Ces especes ne montrent aucune exigence particuliére quant aux conditions d’éclairement, colonisant dans les mémes propor- tions, le sommet, la région médiane et !a base des arbres. Absente a lile d’Anticosti et en Gaspésie, U. trichodea n’a été identifié sur la Céte Nord qu’a St- Nicolas (Figure 5B), sur un sapin. Contrairement a Vile d’ Anticosti ol quelques spécimens de U. subflori- dana Stirt. avaient été récoltés, cette espece est man- quante sur tout le territoire étudie. 30 THE CANADIAN FIELD-NATURALIST Ficure 5. Distribution du genre Usnea sur la Cote Nord du Saint-Laurent et en Gaspésie: (A) Usnea filipendula; (B) (@ ) Usnea longissima; (° ) Usnea sorediifera, (&) Usnea trichodea. Genre Ramalina Dans les peuplements conifériens de la Cote Nord du Saint-Laurent et de la Gaspésie, 4 especes corti- coles représentent le genre Ramalina: Ramalina dilace- rata (Hoffm.) Hoffm. (Syn. R. minuscula (Nyl.) Nyl.), Ramalina roesleri (Hochst.) Nyl., Ramalina thrausta (Ach.) Nyl. et Ramalina farinacea (L.) Ach. L’abondance des espéces R. dilacerata et R. roesleri se vérifie sur tout le territoire (Figure 6, A et B). Les 2 espéces colonisent le sommet et la base des 3 essences forestiéres étudiées, le sapin, l’épinette noire et Pépinette blanche. Concordant avec les données recueillies a Pile d’ An- ticosti(Laflamme et a/., 1979), R. thrausta illustre une préférence marquée pour les milieux ombragés, se fixant sur le tiers inférieur des coniféres, dans des proportions de 2:1 par rapport au tiers supérieur. Rare sur la Céte Nord du Saint-Laurent, cette espece se retrouve fréquemment sur le sapin a l’extrémité de la péninsule gaspésienne (Figure 6C). Absente sur la Céte Nord, les informations recueil- lies sur R. farinacea en Gaspésie confirment les résul- tats obtenus a Vile d’Anticosti(Laflamme-Levesque et al. 1979) sur la préférence de cette espéce pour les Vol. 97 milieux sombres (Figure 6D). En effet, le tiers inférieur des coniféres est colonisé par R. farinacea 1.7 fois plus souvent que le tiers supérieur. Description du territoire La Céte Nord du Saint-Laurent et la Gaspésie con- stituent 2 territoires fortement différents quant a leur composition foresti¢re. Sur la Cote Nord, les peuple- ments forestiers qui recouvrent la région céti€re située entre Baie-Comeau a l’ouest et Havre-Saint-Pierre a lest, sont surtout composés de coniferes. Les pessi€res noires occupent une grande partie de la forét mais, a occasion, il y a présence de sapinieres qui sont alors situées dans la partie inférieure des bassins de riviére. Elles sont généralement trés anciennes, mais des per- turbations causées par l’exploitation forestiére, le feu ou les pullulations d’insectes en ont diminué l’impor- tance et elles sont souvent réduites a quelques arbres seulement. La région gaspésienne s’étendant de Ste-Anne-des- Monts a Percé est caractérisée, au point de vue fores- tier, par son manque d’homogénéeité. En effet, les nombreuses perturbations du milieu occasionnées a la suite des périodes successives d’exploitation fores- tiére, des pullulations d’insectes et de feux, ont trans- formé la forét en une véritable mosaique ou dominent tantot les feuillus, tantdt les coniféres. Les peuple- ments inventoriés en Gaspésie different nettement de ceux de la Céte Nord du Saint-Laurent par leur tres grande hétérogénéité et par le morcellement créé par les nombreuses perturbations. Ces régions cétiéres, quoique fort différentes actuel- lement de Tile d’Anticosti au point de vue forestier (Laflamme-Levesque et al. 1979) possédent une végétation lichénique corticole semblable. En effet, les mémes espéces du genre Bryoria se retrouvent sur les coniféres des 3 régions étudiées a l’exception de B. salazinica, absente en Gaspésie. B. simplicior rare a Vile d’Anticosti (Laflamme-Levesque et al., 1979) se retrouve trés fréquemment dans les milieux cOotiers inventoriés. La présence de la seule espéce illustrant le genre Alectoria, Alectoria sarmentosa ssp. sarmen- tosa est notée abondamment sur la Cote Nord du Saint-Laurent et a Tile d’Anticosti (Laflamme- Levesque et al. 1979); elle devient trés rare en Gaspé- sie. La race chimique KC-, notée pour la premiére fois a Vile d’Anticosti se retrouve également sur les iles de Mingan, l’ile Bonaventure et dans le Parc de la Gaspé- sie. Enfin, a l'exception de R. farinacea absente sur la Cote Nord, le genre Ramalina est représenté par les mémes espéces a l’ile d’ Anticosti (Laflamme-Levesque et al. 1979), sur la C6te Nord et en Gaspésie. Sur la Céte Nord du Saint-Laurent et en Gaspésie, le petit nombre de sapiniéres nous empéche d’établir une corrélation, comme ce fut le cas a l’ile d’Anticosti, entre la densité de recouvrement lichénique des 1983 M. LAFLAMME-LEVESQUE, PERRON ET JOBIN: DISTRIBUTION DES LICHENS 31 FIGURE 6. Distribution du genre Ramalina sur la Céte Nord du Saint-Laurent et en Gaspésie: (A) Ramalina dilacerata; (B) Ramalina roesleri; (C) Ramalina thrausta; (D) Ramalina farinacea. arbres, leur age et la densité forestiére. Toutefois, il est intéressant de noter que les arbres sur lesquels la den- sité lichénique dépasse 50% de recouvrement sont agés d’au moins 50 ans. Les sites dans lesquels les lichens recouvrent si intensément le substrat sont d’une part, des endroits ot une pullulation de L. fiscellaria fiscel- laria a déja été signalée (Frankelin, Godbout, Riv. Darmouth, Pointe-des-Monts, Riv. Mingan, Baie St- Laurent, ile Eskimo, ile Niapisca), et d’autre part de vieilles sapiniéres n’ayant pas subi de perturbations importantes (Havre-Saint-Pierre, Parc Forillon, Parc de la Gaspésie). Remerciements Nous sommes reconnaissants envers Ie Dr. I. M. Brodo du Musée des Sciences naturelles, Ottawa, pour son aide dans l’identification des lichens. L’as- sistance technique de M. Charles Coulombe fut gran- dement appréciée lors du travail sur le terrain. Cette recherche a pu étre réalisée grace aux octrois du Con- seil national de recherches du Canada, subvention A-5764. Références Brodo, I. M., and D. L. Hawksworth. 1977. Alectoria and allied genera in North America. Opera Botanica 42: 1-64. Carroll, W.J. 1956. History of the Hemlock looper, Lambdina fiscellaria fiscellaria (Guén.), (Lepidoptera: Geometridae) in Newfoundland, and notes on its biology. The Canadian Entomologist 88: 587-599. Dahl, E., and H. Krog. 1973. Macrolichens of Denmark, Finland, Norway and Sweden. Universitest Forlaget, Oslo, Bergen and Tromso. 185 pp. De Gryse, J. J.. and K. Sched]. 1934. An account of the eastern Hemlock looper Ellopia fiscellaria Gn., on the , hemlock, with notes on allied species. Scientific Agricul- ture 14: 524-539. Grondin, P., et M. Melangon. 1978. Etude phyto- écologique de la Grosse ile au Marteau et de Vile a Samuel, Archipel de Mingan, Québec. Thése de maitrise, Univer- sité Laval, Faculté Foresterie et Géodésie, Québec. 263 pp. Hale, M. E. Jr. 1969. How to know the lichens. Wm. C. Brown Co., Dubuque, Iowa. 226 pp. Jobin, L. 1973. L’arpenteuse de la pruche, insecte insulaire. Centre de recherches forestiéres des Laurentides, Envi- ronnement Canada, Québec, Milieu, No 7: 8-12. 32 THE CANADIAN FIELD-NATURALIST Laflamme-Levesque, M., J. M. Perron et L. Jobin. 1979. Etude des lichens appartenant aux genres Bryoria, Alecto- ria et Ramalina dans les foyers d’infestation de Lambdina fiscellaria fiscellaria (Guén.) a Vile d’Anticosti. Le Natura- liste Canadien 106: 505-510. Vol. 97 Lepage, E. 1972. Nouveau catalogue des Québec. Naturaliste Canadien 99: 533-550. Reilly, R. W. 1972. The macrolichens of the Magdalen Islands, Quebec, Canada. The Bryologist 75: 179-184. lichens du Recu le 10 novembre 1981] Accepte le 30 octobre 1982 Amphibians of the Province of Newfoundland JOHN E. MAUNDER Newfoundland Museum, Duckworth Street, St. John’s, Newfoundland AIC 1G9 Maunder, John E. 1983. Amphibians of the Province of Newfoundland. Canadian Field-Naturalist 97(1): 33-46. Eight species of amphibians have been confirmed for Newfoundland. The Striped Chorus Frog ( Pseudacris triseriata) and the Green Frog (Rana clamitans) are found on the island only. The Blue-spotted Salamander (Ambystoma laterale), the Two-lined Salamander (Eurycea bislineata) and the Mink Frog (Rana septentrionalis) are found in Labrador only. The American Toad (Bufo americanus), the Wood Frog (Rana sylvatica) and the Northern Leopard Frog (Rana pipiens) are found in both sections of the province. All six species confirmed for Labrador are native there. All five species found on the island have been introduced, but have spread, naturally and with human assistance, from their original sites of introduction. Two additional species, the Eastern Redback Salamander (Plethodon cinereus) and the Spring Peeper (Hy/a crucifer), have been reported from Labrador, but their status remains hypothetical. Key Words: frogs, toads, salamanders, Newfoundland, Labrador, introductions, distributions, range extensions, range limits, interspecific competition. This paper is a synthesis of what is known about the amphibians of the province, and covers past and pres- ent distributions, introductions, dispersal, and some breeding biology. Amphibians have been investigated only superfi- cially in the province of Newfoundland. Much of what is known can be found in reports dealing with broader geographic areas (Bleakney 1954, 1958; Harper 1956), or in the form of notes by transient field-workers and “travellers in the country” (Packard 1866, 1891; Maret 1867; Kindle 1924; Austin 1932; Ayre 1938; Cook and Folinsbee 1975; Cook and Preston 1979; de Graaf, Boles and Lovisek 1981). More specific papers have been written by Johansen (1926), Backus (1954), Cameron and Tomlinson (1962) and Buckle (1971). Sources of information, other than the literature, include personal interviews with field-workers, field work by the author to confirm some distribution records, and a unique school-participation project known locally in Newfoundland as “Project Frog”. Most of the new data concerning four amphibian species introduced to the island of Newfoundland between 1960 and 1966 (Buckle 1971) was gleaned from the excellent, continuing field observations of Mr. James Buckle, or obtained while on field excur- sions in his company. “Project Frog” was a cooperative effort involving the Natural History Section of the Newfoundland Museum at St. John’s and students from forty partici- pating junior high schools located throughout the island of Newfoundland and the mainland of Labra- dor. Following guidelines set out by the museum ina project handbook, and under the supervision of science teachers, students collected and documented specimens of frogs and toads. Information on the local histories of the various species was collected in some instances and, where appropriate, students 33 reported the apparent local absence of amphibians after searching the countryside and interviewing resi- dents. All specimens and data were deposited in the Newfoundland Museum. Data from the project added significantly to the knowledge of amphibian distribu- tion in the province. In the following species accounts, J. S. Bleakney’s Herpetofaunal Sections (Bleakney 1958) are exam- ined as they apply to the amphibian distribution on the island of Newfoundland (Figures 1, 3 and 5) and in Labrador (Figures 2 and 3). Blue-spotted Salamander Ambystoma laterale Only three specimens of this species have been col- lected from the province, all from Labrador (Figure 3). The first was a male taken by W. E. Beckel at Goose Bay on 25 July 1948 and reported by Bleakney (1954). Originally listed as A. jeffersonianum, the specimen was re-assigned to A. /aterale when Uzzell (1964) revised this species complex. The second spec- imen was a juvenile taken by John Porter approxi- mately 16 km south of Wabush on 22 September 1974 and reported by Cook and Folinsbee (1975). The third specimen was collected from a basement in Happy Valley, near Goose Bay, on 11 November 1981 (Frank Phillips, Provincial Wildlife Division, personal com- munication). There is also a recent and apparently reliable sight record of a Blue-spotted Salamander for Ugjoktok Bay, near Hopedale, by Bernard Chaulk of the Provincial Forestry Division (Frank Phillips, per- sonal communication). The Wabush specimen and the Ugjoktok Bay sighting are from climatic regions designated by Bleakney (1958) as Herpetofaunal Sec- tions (see caption of Figure |) 6. and 7 respectively (see Figure 2) indicating a more boreal distribution for this species than previously thought. 34 THE CANADIAN FIELD-NATURALIST Eastern Redback Salamander Plethodon cinereus Bleakney (1958, p. 15) reported that the occurrence of a small salamander in the Lake Melville area was mentioned on several occasions in 1955 to National Museums of Canada worker Mr. W. J. Smith, and suggested that from the descriptions this salamander “must be Plethodon c. cinereus”. Packard (1866, 1891) made several references to a salamander he variously identified “with some doubts” (1866, p. 272) as “Plethodon glutinosa” and “Plethodon glutinosus”, the Slimy Salamander. His observations were made just outside of what is now the province of Newfoundland, near the coast at Belles Amours, Quebec, approximately 22 km west of Blanc Sablon and the Labrador-Quebec boundary. Packard’s descriptions variously described “a little salamander ... of a slate color, with a paler light dorsal band” (1891, p. 112), and “a salamander of a dark slate color, with a paler dorsal stripe” (1866, p. 272; 1891, p. 406). The descriptions may indicate the Two-lined Salamander as suggested by Cook and Preston (1979), but also seem to fit the red-backed or striped phase of the Eastern Redback Salamander. However, Packard (1891, p. 112) stated that one of the salamanders “ran into the water, to [his] great disap- pointment, just eluding [his] grasp”. While fleeing to water is not unusual behavior for Two-lined Sala- manders, it would not be expected of Eastern Red- back Salamanders. An additional problem arises with the statement in Packard’s 1866 paper (p. 272) that the salamander was “about six inches in length”, possibly indicating the Blue-spotted Salamander, as suggested by Bleakney (1954). It is possible Packard encountered more than one species of salamander and unaccountably merged the descriptions together. If his “Plethodon glutinosus” identification actually referred to a species he thought to be the Slimy Salamander, the creature in question was probably the “similar” but shorter-tailed Blue- spotted Salamander. The “little salamander... witha paler light dorsal band” may well have been the Two- lined Salamander. Packard (1891, p. 106) also referred to “lizards” reported from “inland” that were “most probably Ple- thodon glutinosus’’, but it is unclear whether “inland” referred to Quebec or Labrador territory. Both the Blue-spotted Salamander and the Two-lined Sala- mander were first collected in south-central Labrador after Packard and Bleakney wrote their papers. The status of Eastern Redback Salamander in Lab- rador must be regarded as hypothetical at best. Two-lined Salamander Eurycea bislineata The Two-lined Salamander has been collected at only two provincial localities, both in Labrador (Fig- Vol. 97 ure 3). The first collection was of a recently trans- formed juvenile taken by John Preston approximately 16 km northeast of Labrador City in mid-August 1972 (Cook and Preston 1979). The second collection was of several larvae by Dirk de Graaf ina small tributary of the Cache River in central Labrador on 17 August 1979 (de Graaf, Boles and Lovisek 1981). Both collec- tions are from Bleakney’s Herpetofaunal Section 6. The boreal nature of these records may indicate a widespread distribution in southern Labrador. American Toad Bufo americanus This species is widely distributed in southern Lab- rador, and in the Corner Brook area on the island. In Labrador, American Toads occur to the north- ern limit of Bleakney’s Herpetofaunal Section 6 (Fig- ure 3). This distribution is supported by Packard (1891), Kindle (1924), Austin (1932), J. K. Doutt (unpublished summary of zoological field observa- tions entitled “Field Notes of the Carnegie Museum (Philadelphia) Number 19; Expedition to Hamilton River, Labrador to Sandgirt Lake, June 22 - Sep- tember I1, 1939”), Backus (1954), Bleakney (1954, 1958), Harper (1956), and my own records. On the island, toad distribution is centered in the Corner Brook area in the vicinity of James Buckle’s introductions (Figure 4). According to Buckle (per- sonal communication), and as present distribution shows, the toads have dispersed naturally from intro- duction and transplant sites at an average rate of 0.3 to 0.4 km per year, with a maximum of approximately | km per year. (Throughout this paper, Buckle’s fig- ures for distance dispersed per year by each species are estimated values based on over twenty years of regular field observations plus an examination of present distributions). In addition to natural dispersal, considerable human-assisted dispersal has taken place. The history of the American Toad on the island has been ascer- tained as follows. James Buckle, who brought the species there, made all his collections of young toads at Richmond Hill and Thornhill in the Toronto area. In 1960, approximately 100 newly-metamorphosed toads were released on the shore of a small pond near the Trans-Canada Highway just south of Corner Brook (U.T.M. 340193). (Introduction localities in this paper have been pinpointed to within 100 metres using Universal Transverse Mercator (U.T.M.) Mil- itary Grid References and 1:50 000 scale topographic maps issued by the Surveys and Mapping Branch, Department of Energy, Mines and Resources, Ottawa.) Road construction occurred at this intro- duction site soon afterwards, and toads were not seen there again. Buckle thereupon sent four new install- ments of young toads from Toronto to his father at 1983 MAUNDER: AMPHIBIANS OF NEWFOUNDLAND 35) FiGuRE 1. The island of Newfoundland showing Bleakney’s (Bleakney 1958) Herpetofaunal Sections (numbered, and bounded by heavy lines) and Environmental Index isopleths (designated by numbered “i’s”). Bleakney described a series of “herpetofaunal sections” (or regions) relating to amphibian and reptile distribution in eastern Canada. The boundaries of these sections were determined using a series of environmental temperature indices calculated by multiplying the length of the growing season by the mean July temperature at each of several locations. For simplicity, each number so calculated is divided by 1000. For Herpetofaunal Section 4, the environmental indices are || to 10; for Section 5 they are 10 to 8; for Section 6 they are 8 to 6; for Section 7 they are 6 to 5. Bleakney did not illustrate herpetofaunal sections for the island. They were plotted for this paper using Hare (1952). 36 THE CANADIAN FIELD-NATURALIST Corner Brook (60 in 1963, 45 in 1964, 41 in 1965, and 55 in 1966), who released the toads in his garden on Valley Road (U.T.M. 303223). Because this release site had poor breeding potential, 3 adult and 29 one- year-old toads were moved in 1967 to a site 2.4 km northeast of Steady Brook bridge on the south side of the highway (U.T.M. 412238). Since that time, much has happened. The known movement of toads from the Steady Brook bridge site is 4.5 km northeast to Little Rapids, 3.5 km southwest to the southwestern limit of habitation at Steady Brook village (including “Dogwood Station”), 0.5 km northwest across the Humber River (approximately 500 m wide at this point), and 2.5 km southeast of Steady Brook village up the gorge of Steady Brook stream (an elevation increase of approximately 250 m). The Steady Brook gorge record is based ona single sighting and may have been a toad moved by humans. Dispersal in the Steady Brook area appears to be considerably limited by the terrain, which includes the deep, narrow valley of the Humber River with hills rising steeply to 300 mand 500 mto the northwest and southeast respectively, and a very narrow river gorge downstream (west) from Steady Brook village. Dis- persal to the northwest, southwest and southeast is probably now stymied, but to the northeast it seems to be continuing. In 1970, approximately 150 young toads from Steady Brook were moved to Beaver Pond just northwest of the Trans-Canada Highway, approximately half-way along the southeast side of George’s Lake (U.T.M. 172015). By 1977, the toads had dispersed 3.5 km northeast and southwest, and 2 km southeast to Rocky Pond. By 1978, toads had appeared at Blue Ponds Provincial Park 4 km to the northeast. Also in 1970, approximately 50 young toads from Steady Brook were moved toaditch 1 km northeast of “Silver Birches” Salvation Army Camp near Pynn’s Brook (U.T.M. 592353). They have estab- lished a tremendously successful breeding population in the area near the camp entrance. In 1976, 100 tadpoles and six adults from Steady Brook were moved to the southwest end of Deer Lake (U.T.M. 500278). This population is now established and toads are breeding in the lake despite the constant presence of floating log booms destined for the paper mill at Corner Brook. In September 1977, 34 adult toads were “rescued” from an area of highway con- struction near George’s Lake and released at Picadilly on the Port-au-Port Peninsula. Unconfirmed reports from local residents in August 1980 indicated toads were still present there. In 1978, approximately 1000 individuals (mostly tadpoles) from the “Silver Birches” camp area near Pynn’s Brook were moved to roadside ditches on the Deer Lake Airport road. The Vol. 97 same year there were at least two transplants outside the Corner Brook area. Approximately 80 young toads from George’s Lake were moved to Boswarlos on the Port-au-Port Peninsula, and approximately 100 young toads from Steady Brook were moved to a small pond on the east side of the Trans-Canada Highway | km north of the entrance to Butterpot Provincial Park near St. John’s. Repeated searches of the Deer Lake Airport road and Butterpot Park areas (by Buckle and Maunder respectively) have shown no evidence of toad survival at either site. The fate of the Boswarlos toads is unknown, although there have been some positive reports from local residents. In 1980, 75 small toads were moved from Steady Brook toasmall farm owned bya Mr. Prowse and located at the bridge over the Humber River on the Northern Peninsula Highway at Deer Lake (U.T.M. 683490). Other locations where toads have been reported include Benoit’s Cove west of Corner Brook (one seen in 1978), Wild Cove Dump just northeast of Corner Brook (toads apparently put there, but not seen lately), and near the highway intersection at the extreme southeast corner of Humber Arm east of Corner Brook. Buckle reports that toads can now be found almost anywhere within the city limits of Corner Brook. An additional record comes from the Bonne Bay area approximately 70 km north of Corner Brook. Toads were heard trilling in an alder (Alnus sp.) swamp at the intersection of the Norris Point road and the Rocky Harbour cove road in Rocky Harbour in late May 1977. On 3 June of that year, a male was captured by Kevin Moore and Gre- gory Horne. The toad was photographed (photo- graphs on file at Gros Morne National Park), mea- sured and released. At the same time; two other toads were heard trilling at this location, and another was heard ina pond | km to the south. These individuals were transplanted to this location by humans, pre- sumably from toad populations in the Corner Brook area. It is certain that the considerable human-assisted dispersal of toads (and other amphibians) will con- tinue to occur on the island. Adults and children alike seem compelled to move them from place to place. Many pockets on rural school buses, for example, now carry toads, and the presence of a significant breeding population of toads at the “Silver Birches” summer camp (which attracts campers from across the province) has obvious consequences. There is even one known instance of children attempting (with uncertain success) to sell toads on the highway at Steady Brook. It seems that the American Toad is firmly estab- lished on the island. It can be expected to continue spreading at a steady rate on its own, and in a more 1983 MAUNDER: AMPHIBIANS OF NEWFOUNDLAND 37 FiGuRE 2. Labrador showing Bleakney’s (Bleakney 1958) Herpetofaunal Sections (numbered, and bounded by heavy lines) and Environmental Index isopleths (designated by numbered “i’s”). For explanation, see caption of Figure |. Replotted for this paper using more complete and up-to-date climatic data from Wilson (1971). 38 THE CANADIAN FIELD-NATURALIST erratic fashion over longer distances where humans are involved. If the species is unable to move into Bleakney’s Herpetofaunal Section 7, as appears to be the case, toads will probably never populate the northern reaches of the island’s Great Northern Peninsula (Figure 1). Toads in the Corner Brook area disperse through any type of vegetation, including thick spruce woods (Picea sp.). Dispersal along lakeshores and water- courses is at approximately the same rate as through woods except where strong river currents sweep indi- viduals along for a short distance. No investigations of toad diet have been carried out in the province, but it is widely reported that straw- berry beds (Fragaria sp.) are strikingly free of slugs in areas where toads are common. Dytiscid water beetles have been observed preying on toad tadpoles at “Silver Birches”. The degree of predation by Brook Trout (Sa/valinus fontinalis), the only major freshwater fish of small water bodies in the Corner Brook area, is unknown; but it is known that toad tadpoles prosper in the local trout streams. Buckle reports that he found about 20 breeding toads killed and disemboweled by Common Crows (Corvus brachyrhynchus) at the “Silver Birches” breeding area in 1981. Toads in the Corner Brook area breed ina variety of locations, including shallow ponds, roadside ditches, pools in gravel pits, mud flats, drained beaver ponds, lake shores, slow-moving brooks, and marshes. They are not particularly early breeders, laying after both Chorus Frogs and Wood Frogs. Calling began, typi- cally, at “Silver Birches” on 18 May 1980. Breeding at higher altitude locations southwest of Corner Brook, such as Blue Ponds (above 200 m), takes place as much as two to three weeks later because of cooler temperatures. Calling begins when the temperature rises to approximately 15°C. Buckle and others found that once calling begins it will continue for a time even if the temperature drops. However, little calling takes place at night in the Corner Brook area because spring nights are generally very cold, with temperatures often approaching the freezing point. Midday is acommon time to hear toads there. On the nearby mainland, Harper (1956) found American Toads breeding on 25 May 1953 at Sept Iles, Quebec and on 14 June 1953 at Caro] Lake, Labrador. The same author found these toads began calling at approximately 13°C, nearly the same temperature as on the island. Calling continues sporadically well into summer near Corner Brook. Newfoundland toads appear to breed three years after they hatch as tadpoles, according to Buckle’s observa- tions at introduction sites. Striped Chorus Frog Pseudacris triseriata This species is found only in the Corner Brook area Volgoy7, of the island (Figure 4). Specimens for introduction were collected by Buckle near Keel Street in the Toronto area. In 1963, 50 to 55 tadpoles were released at three locations in the vicinity of Steady Brook (14 to 19 at U.T.M. 395219, 18 at U.T.M. 402231, and 18 at U.T.M. 411238). No Chorus Frogs were located dur- ing subsequent midsummer visits to the area (Buckle 1971). However, in the early 1970’s, visits to Steady Brook in early spring (when the frogs were calling) turned up a surprisingly healthy population. In 1975, 75 tadpoles from Steady Brook were moved to a roadside pool at the west end of South Brook town (U.T.M. 536288). F. W. Schueler, A. Karstad and F. D. Ross collected two adults at this site on 3 and 6 May 1976. Chorus Frogs were still breeding in the pool in small numbers in 1981. How- ever, Buckle could not find any in the spring of 1982. No dispersal from the pool was noted. Also, in 1975, a few tadpoles from Steady Brook were moved to Mea- dows Brook at the northeast end of Pinchgut Lake south of Corner Brook (U.T.M. 295090). At the same time, approximately 60 more tadpoles from Steady Brook were literally “discarded” into a cold, fresh- water spring approximately 0.8 km southwest of the Meadows Brook site (U.T.M. 289087). The first of the two transplants at Pinchgut Lake included Wood Frogs, and was not successful. The second of the two transplants was of Chorus Frogs only. It was soon forgotten because no one thought Chorus Frogs would actually breed in such extremely cold water. In 1980, Buckle was surprised to learn that they did indeed breed there and were flourishing. Chorus Frog tadpoles were also found in pools of warmer water in the same vicinity. Of course, the tadpoles in the warmer pools developed much more quickly than those in the spring. However, both groups appeared equally successful. The population at Pinchgut Lake is presently the most vigorous Cho- rus Frog population on the island. In 1978, approximately 130 tadpoles from Steady Brook were moved to a site 2 km south-southwest of the Meadows Brook site at Pinchgut Lake (U.T.M. 289073). The success of this transplant has yet to be determined. Also in 1978, approximately 200 tadpoles from Steady Brook were moved to ditches along the Deer Lake Airport road. This transplant was not suc- cessful. In 1980, 20 to 30 tadpoles from South Brook were introduced to an open, wet area near the ferry terminal at St. Barbe, on the Great Northern Penin- sula. A surprising aspect of Chorus Frog existence in the Corner Brook area is the apparent ability of Wood Frogs to devastate Chorus Frog populations. The evidence is circumstantial, but may be worth consider- ing. The 1963 introduction of Chorus Frogs at Steady Brook is a case in point. On the west side of the brook, 1983 MAUNDER: AMPHIBIANS OF NEWFOUNDLAND 39 Two-lined alamander “i Salamander 2, American Toad Northern Leopard Frog FIGURE 3. Distribution of amphibian species in Labrador. Dark circles indicate sites where species have been collected or verified. 40 THE CANADIAN FIELD-NATURALIST Buckle released 14 to 19 Chorus Frog tadpoles. But, on the east side, along with 36 Chorus Frog tadpoles, he also released 50 to 55 Wood Frog tadpoles. The Chorus Frogs released on the west side of the brook gave rise to what was to become, by the late 1970's, a large and thriving population. But those released on the east side with the Wood Frogs were never seen again. Despite the initial success of the Chorus Frogs on the west side of the brook, their population began to decline rapidly assoon as Wood Frogs crossed into the brook (about 1978 or 1979). A striking example of this rapid decline occurred at a small pond below the Marble Mountain ski lift at Steady Brook. It was exclusively inhabited by Chorus Frogs until 1979. But searches in 1980 found none. Instead, the pond con- tained 52 Wood Frog egg masses. The situation was the same throughout the Steady Brook area. No Cho- rus Frogs have been recorded there since 1979. A second case in point is the failure of Chorus Frogs at the Meadows Brook site at Pinchgut Lake (where Wood Frogs occurred), and the success of Chorus Frogs at the “cold spring” site and vicinity at Pinchgut Lake (where Wood Frogs do not occur). These two cases cannot be satisfactorily explained by climatic factors, since the successes and failures occurred virtually side by side. Additionally, Steady Brook would seem an unlikely place to be affected since it has the most benign climate in the area. Habi- tat change, the likely cause of the gradual disappear- ance of Chorus Frogs at the established South Brook site, was not an apparent factor at Steady Brook, which has the most diverse Chorus Frog habitat of any introduction site. The natural dispersal of Chorus Frogs in New- foundland has been negligible. At Steady Brook, movement was less than a kilometre over the 16 years between 1963 and 1979, all this through continuously wet and open terrain. If Wood Frogs do, in fact, devastate Chorus Frog populations, it is likely that Chorus Frogs will eventually be extirpated on the island by advancing Wood Frogs (unless they are transported periodically to Wood Frog-free areas by humans). Chorus Frogs are early breeders. Vocalization has been recorded by Buckle as early as 29 April at South Brook. Calling begins when the temperature reaches 12° to 13°C, but continues until temperatures drop as low as 8° C. The species can usually be heard as late as the first week of June. Spring Peeper Hyla crucifer Bleakney (1954, p. 166) reported a description related to him by a “naturalist-inclined M.D. student” of “tiny tree frogs with suction disc finger tips and huge vocal sacs” ina ditch near Menihek Lake Camp Vol. 97 in Labrador. Bleakney thought the frog in question was a Spring Peeper. David Smith, of Hunter and Associates in St. John’s, reported (personal commun- ication) that he and fellow field-worker R.S. W. Bobbette heard what they believed was at least one Spring Peeper calling in the evening of 11 July 1980 at Thomas Brook in the Lower Churchill River Valley, Labrador. A specimen of this species was collected in 1974 by MacCulloch and Bider (1975) at a similar latitude, but far to the west, at Lac Nathalie, Quebec. The status of the Spring Peeper in Labrador must still be regarded as hypothetical. Green Frog Rana clamitans In the province of Newfoundland, this species is known to occur only on the island (Figure 5). Its early history here is rather hazy. Jukes (1842, p. 187) states that “Not a frog, nora toad, nora lizard, nor a snake has ever been seen in the country”. A quarter century later, Maret (1867) found “frogs” in ponds and lakes in the St. John’s area. In 1922, Johansen (1926) found Green Frogs to be fairly common around St. John’s, and noted that the frogs were “supposed to have been introduced (with hay ?) from Nova Scotia less than a century ago” and that “they [were] not found outside the Avalon Peninsula”. The actual mechanism and date of introduction will likely remain a mystery. The only previous effort to sort out the dispersal of the Green Frog in Newfoundland was made by Cameron and Tomlinson (1962). At the time of their study, the species occurred mainly over the central and eastern portions of the Avalon Peninsula, with scattered human-transplanted populations through- out the island. During the course of “Project Frog”, undertaken in 1979, a pattern of distribution emerged that loosely follows the well-established road and rail links on the island. Green Frogs are notably absent from the south coast of the island west of Baie d’Espoir (an area which still has no road or rail links except for a very recent connection at Burgeo), although there is a record of people keeping “frogs” as pets in Francois, and a record of an unsuccessful transplant to Grand Bruit. The species is also presently absent from the entire Great Northern Peninsula north of Sally’s Cove, an area which has received basic road links only in recent years. Some might take this as an indication of the human influence on frog dispersal in the pro- vince. However, while human-assisted dispersal appears to be very significant, climatic conditions probably play an equally important and coincidental role. Green Frogs do not appear to occur more boreally than the outer limits of Bleakney’s Herpeto- faunal Section 5 (Figures | and 5). If this statement is accurate, it is unlikely that Green Frogs will ever be 1983 MAUNDER: AMPHIBIANS OF NEWFOUNDLAND 4] American Toad Striped Chorus Frog Corner Brook Northern Leopard Frog Corner Brook 1979 = FIGURE 4. Dispersal of the recently introduced species on the island of Newfoundland. Dates indicate introductions or transplants to particular sites. Arrows indicate dispersal routes taken by transplanted stocks. Dark circles are introduction or transplant sites. Dark squares are sites to which introduced or transplanted populations have dispersed “on their own”. 42 THE CANADIAN FIELD-NATURALIST found in the Burgeo area on the south coast of the island, or on the Great Northern Peninsula. As in the past, humans will almost certainly con- tinue to be the main agents of Green Frog dispersal on the island. (According to Cameron and Tomlinson (1962), natural dispersal has averaged less than | km per year.) The Reverend Oliver Jackson transplanted Green Frogs to Bell Island (Wabana) “to show the children” (Ayre, 1938, p. 11). A population was trans- planted to the Eastport Peninsula by Boy Scouts (who did this sort of thing often, according to Cameron and Tomlinson). About 1940, there is a record of a transplant to the Burin Peninsula (Cameron and Tomlinson, 1962). I have another record of frogs being brought to Jones’ Pond near St. Lawrence “years ago”, which may be the same record. The origin of the Burin Peninsula transplants is unknown, but it is interesting to note that Green Frogs were introduced to the nearby French islands of Saint Pierre et Miquelon (specifically to Saint Pierre and the sub-island of Langlade) a few years before, in 1934 (Daniel Abraham, personal communication; Aubert de la Rue, 1937, p. 87). The species is now found through- out the French islands and evidence suggests that the present population there stems from the 1934 introduc- tion (Daniel Abraham, personal communication). A transplant to Springdale from the Burin Penin- sula some time later is also known. Bleakney (1954) reported Green Frogs from Terra Nova village. Green Frogs have been noted at Carmanville “for about fifteen years”. Sportsman and film-maker Mr. Lee Wulff, and family, brought Green Frogs to Winter- house Pond in Daniel’s Harbour (apparently an unsuc- cessful transplant) in 1955. At least two other unsuc- cessful transplants of Green Frogs to Daniel’s Harbour have been made by others in recent years. Cameron and Tomlinson (1962) wrote that Mr. Leslie M. Tuck had “recently” reported Green Frogs from Port-aux- Basques and Port Saunders (none are now known to occur at the second site). Green Frogs have been transplanted to Burgeo on several occasions, the last time being in 1976 when five tadpoles were released. Apparently, no frogs have survived at Burgeo. There are numerous and widespread reports of additional transplants of two or three frogs at a time. None of these enterprises have met with known success, but indicate the public interest in moving frogs. It seems certain that the Green Frog will continue to be successful on the island. It will likely continue to fill in the unoccupied areas within its spotty distribution. Overall, however, it may now be at the limits of its range in Newfoundland. Cameron and Tomlinson (1962) noted that Brook Trout feed on Green Frog tadpoles. However, both species regularly inhabit the same ponds. Vol. 97 No one has studied the breeding biology of Green Frogs in Newfoundland. Pope (1944) found that Green Frogs in the Chicago area began egg-laying when the temperature reached 24°C. On the island, egg masses were collected by Cameron near Holyrood on 19 June 1960. The same author saw tadpoles that had overwintered on 2 July 1960 about 3 km inland from Cape Broyle. Tomlinson collected a tailed tad- pole and an almost completely metamorphosed adult at Pouch Cove on 21 August 1961. Mink Frog Rana septentrionalis There are only four definite records for this species in the province, all from Labrador (Figure 3). Backus (1954) reported a specimen collected at Etagaulet Bay, Lake Melville, on 11 July 1950. Frank Phillips and Bruce Boles (personal communication) obtained a specimen (Newfoundland Museum catalogue number H-2) at Goose Cove on the Churchill River approxi- mately 3 km above the mouth of the Metchin River in 1977. C. Schryburt, J. Mayes, R. S. W. Bobbette and D. Smith (D. Smith, personal communication) sight- ed five adults in a small pool on the Goose River near Goose Bay on 23 July 1980. Two were collected, but the specimens have since been lost. Frank Phillips (personal communication) reported the species to be widespread in the Goose Bay area. In addition, eggs, tadpoles and adults of this species were collected (NMC 2267-2268, 2270-2276) in 1952 by Sherman Bleakney at Mile 134 of the Quebec-North Shore Railway, just 24 km south of the Labrador border. Packard (1866, 1891) reports a specimen from Okak that was identified as this species by E. D. Cope. Bleakney (1954) is cautious of this record, though ina later paper (Bleakney 1958) he states (p. 15) that “Recent collections have proved that Packard’s report is not at all unreasonable”. Hildebrand (1949), Wynne-Edwards (1952) and Backus (1954) all record this specimen as Wood Frog. Harper (1956) is skepti- cal of the record because of the extreme boreal nature (see Figures 2 and 3) of the Okak area. Indeed, the climate on the north coast of Labrador is far more severe than the climate in the Ungava Bay watershed to the west (Figure 2) where there is a Mink Frog record for Lac Aigneau (NMC 3209). In fact, the climate at Okak is more severe than it is even in the George River area on the east side of Ungava Bay where Wood Frogs reach the tree line and are the most boreal frogs otherwise reported on the peninsula (Bleakney 1954). An additional factor, with reference to the Okak record (which is from far beyond the tree line), is that the specimen was “presented. . . by one of the missionaries” (Packard 1866, p. 272), a statement which leaves in doubt the actual collection site. This whole controversy causes wonder about the 1983 MAUNDER: AMPHIBIANS OF NEWFOUNDLAND 43 Green Frog 0 100 ps km FIGURE 5. Distribution of the Green Frog (Rana clamitans) on the island of Newfoundland. Dark circles indicate sites where the species has been collected or verified in recent years. Open circles indicate sites where the species was sought but not found. Dates indicate known transplants, but all apparently stem from the “pre-1867” stock. Most of the information used to compile this map was contributed by “Project Frog”. 44 THE CANADIAN FIELD-NATURALIST other Mink Frog sightings reported by Packard (1891) for Stag Bay, Domino Harbour, Lewis Bay, and Henley Harbour to the south. Harper (1956) sug- gests that all of Packard’s records may have referred to Wood Frogs, a species completely overlooked by him. Of interest, however, is a convincing 1976 sight record of a Mink Frog at Hopedale (Figure 3) by Eric Loring (personal communication). Loring reported what “looked like a small Green Frog”. Bleakney (1958) recorded Mink Frogs breeding between 17 and 23 June 1952 at Mile 134 of the Quebec-North Shore Railway, just south of the Lab- rador border. Wood Frog Rana sylvatica This species occurs in Labrador, and as an intro- duced species on the island. In Labrador, the species occurs (Figure 3) at least in the western region and in the Lake Melville area (Backus 1954; Harper 1956; William Threlfall, per- sonal communication). As suggested in the account of Mink Frogs, the reports of that species by Packard (1866, 1891) for Okak and the coast south of Lake Melville may actually have been reports of Wood Frog. Austin (1932) states that a Mr. Hettasch told him that “frogs” could be found on the Labrador coast as far north as Webb Bay just north of Nain. At least the northernmost of these “frogs” were probably Wood Frogs. On the island the Wood Frog is found mainly near its introduction site in the Corner Brook area (Figure 4). The tadpoles used in the introduction were col- lected near Keel Street in the Toronto area. In 1963, 50-55 tadpoles were put in a roadside ditch 2.2 km northeast of Steady Brook bridge (U.T.M. 412238). Buckle found tadpoles numerous at the site in 1966 and 1967, and by 1968 found tadpoles as far as 0.8 km away. In 1967, 42 tadpoles from northeast of Steady Brook were put in a pool near the present site of the regional college in Corner Brook (U.T.M. 311210). By 1969, tadpoles were abundant at the site and, in 1976, adults were found there by F. W. Schueler, A. Kar- stad and F. D. Ross. Despite recent infilling of the pool, Wood Frogs were still present at the regional college in 1980. In 1975, over 200 tadpoles from Steady Brook were put at the northeast end of Pinch- gut Lake at Meadows Brook (U.T.M. 295090). Buckle found Wood Frogs there in 1979. In 1978, approxi- mately 1000 tadpoles from Steady Brook were moved to roadside ditches on the Deer Lake Airport road. In 1979, 100 tadpoles from Steady Brook were put 15 km into the country on the main woods road at Hawkes Bay on the Great Northern Peninsula. In 1980, two egg masses from Steady Brook were placed in a flooded fen on the Salmonier Line about 10 km north Vol. 97 of St. Catherine’s. The success of the last four trans- plants has not yet been determined. Wood Frogsare firmly established on the island, at least in the Corner Brook area. The species disperses through thick spruce woods like the American Toad at a rate that Buckle has determined to average approximately 0.4 km per year. Almost any pool will serve for breeding. Wood Frogs in the Corner Brook area lay eggs very early in the year, often in pools that skim over with ice within hours. Eggs have been seen as early as 29 April. Vocal- ization begins around this date and is usually finished by 21 May. At any given location the frogs can be heard for only a few days, after which time they become extremely hard to locate. Calling begins at approximately 7°C. As with other species, Wood Frogs will continue to call for a while after the temperature drops below this figure. Harper (1956) reports hearing Wood Frogs at temperatures as low as 1°C in Labrador. The same author noted the dates of first calling to be 25 May at Sept Iles, Quebec, and 5 June at Knob Lake, Labrador. At Fort Chimo, Quebec, Hildebrand (1949) reported egg masses at a stage where larvae would leave when the masses were shaken, on 21 June 1948. Of the amphibians recently introduced to the island, Buckle has found the Wood Frog to be the most successful. Evidence for this includes its steady dispersal, and the ease with which it inhabits even the most marginal of habitats. Northern Leopard Frog Rana pipiens This species is found in Labrador and on the island. In Labrador, it is found (Figure 3) in the southcentral region near Lake Melville (Kindle 1924; Backus 1954; William Threlfall, personal communication) and at Paradise River on the coast (Austin 1932). On the island, the species is found only near its introduction site in the Corner Brook area (Figure 4). The frogs used for introduction were collected near Malton Airport, Toronto. In 1966, 25 adults were released near the site of the present regional college in Corner Brook (U.T.M. 311210). Five more were released at the same time ina pond on the south side of the Trans-Canada Highway 6.5 km northeast of Steady Brook bridge (U.T.M. 447256) near Little Rapids. When Buckle wrote his 1971 paper, he was not aware that the introduction of this species had been successful. The introduction near Little Rapids seems, indeed, to have been a failure. The regional college site proved initially successful. In 1969, after plans to build the college near the frog pond were revealed, 60 tadpoles from the site were put in a small pond southeast of Corner Brook near the Trans-Canada Highway (U.T.M. 349187). The col- 1983 lege site was subsequently bulldozed, and there was no evidence of survival at the Trans-Canada Highway site. It was not until eight years later, in August 1977, that Joe Dunphy of Corner Brook caught an adult Leopard Frog at the Trans-Canada Highway pond. Since that time, the site has been seen to harbour a breeding population of Leopard Frogs, though there has been no dispersal at all to other locations. The threat of highway construction at this site arose in 1978, and 200 tadpoles were moved to Shoal Lake, almost 3 km from the highway south of Corner Brook (U.T.M. 303172). In 1979, two more transplants of tadpoles from the highway site were made to ponds south of Corner Brook. Approximately 500 tadpoles were put at Big Feeder Pond (U.T.M. 286164) and approximately 100 tadpoles were put at a small pond | km to the northeast (U.T.M. 290171). At the same time, approximately 500 tadpoles were put in an upstream section of Meadows Brook (U.T.M. 318117) which flows south into Pinchgut Lake. So far, no construction has taken place at the high- way site, and the population there is still flourishing. The fate of the 1979 transplants is not yet known. A search of the Shoal Lake site by Buckle in early June 1981 was unsuccessful, but tadpoles and year-old frogs were found there on August |, 1982. Buckle also found Leopard Frog tadpoles in Meadows Pond, approximately 5 km southeast of Shoal Lake, in 1982. The tadpoles appear to have reached Meadows Pond by descending Whaleback Brook from Shoal Lake. From Meadows Pond, the species 1s likely to make its way down Meadows Brook into Pinchgut Lake. Leopard Frogs in Newfoundland will move only short distances through swampy or damp terrain, and will not move at all through spruce woods, even to the next pond. Dispersal seems to require a continuous water connection. Buckle has observed Leopard Frog tadpoles being eaten by dragonfly (Odonata) larvae. They will live quite successfully with trout, and they seem to do best in clear pond water, rather than in muddy or boggy pools. Leopard Frogs on the island are late breeders. Call- ing begins only when the temperature reaches 18°C. Eggs were first seen on 20 May 1979 and 18 May 1980. The breeding period is very short and 1s easily missed. Buckle has seen Leopard Frog tadpoles swimming lazily beneath the ice as late as 4 November although no overwintered tadpoles have yet been observed. Acknowledgments The origin of this study was the school-museum experiment that spawned “Project Frog”. To the par- ticipating students and science teachers too numerous to list here, | wish to express my greatest appreciation for a job well done. I am deeply indebted to Jim Buckle of Corner Brook for sharing his field expe- MAUNDER: AMPHIBIANS OF NEWFOUNDLAND 45 rience and knowledge so enthusiastically. My appre- ciation also to Jim’s “family”, Terry and Thelma Dunphy, and Joe Dunphy for their helpfulness and hospitality. Thanks also to Francis Cook for his encouragement and helpful suggestions, and to Daniel Abraham, Edward Andrews, Bruce Boles, Todd Howell, Kevin Moore, Frank Phillips, Fred Schueler, David Smith, Rob Walker, and all the others who in some way helped me. Literature Cited Aubert de la Rue, E. 1937. Le territoire de Saint Pierre et Miquelon, étude de géographie physique et humaine. Journal de la Société des Américanistes (n.s.) 29: 239-372. Austin, Oliver L. 1932. The birds of Newfoundland Labra- dor. Nuttall Ornithological Club Memoir 7: 1-229. Ayre, Agnes M. 1938. Newfoundland Notes. The New- foundland Quarterly, Christmas Number: 10-12. Backus, Richard H. 1954. Notes on the frogs and toads of Labrador. Copeia 1954(3): 226-227. Bleakney, J.Sherman. 1954. Range extensions of amphibi- ans in eastern Canada. Canadian Field-Naturalist 68(4): 165-171. Bleakney, J. Sherman. 1958. A zoogeographical study of the amphibians and reptiles of eastern Canada. National Museum of Canada Bulletin 155: 1-119. Buckle, James. 1971. A recent introduction of frogs to Newfoundland. Canadian Field-Naturalist 85(1): 72-74. Cameron, Austin W. and A. J. Tomlinson. 1962. Dispersal of the introduced Green Frog in Newfoundland. Pp. 104-110 /n Contributions to Zoology 1960-1961. National Museum of Canada Bulletin 183. Cook, Francis R., and John Folinsbee. 1975. Second record of the Blue-spotted Salamander from Labrador. Canadian Field-Naturalist 89(3): 314-315. Cook, Francis R., and John Preston. 1979. Two-lined Salamander, Eurycea bislineata, in Labrador. Canadian Field-Naturalist 93(2): 178-179. de Graaf, Dirk, Bruce K. Boles, and James Lovisek. 1981. Two-lined Salamander, Eurycea bislineata (Amphibia: Caudata: Plethodontidae) in Labrador. Canadian Field- Naturalist 95(3): 366-367. Hare, F. Kenneth. 1952. The climate of the Island of New- foundland: a geographical analysis. Geographical Bulletin 2: 36-88. Harper, Francis. 1956. Amphibians and reptiles of the Ungava Peninsula. Proceedings of the Biological Society of Washington 69: 93-104. Hildebrand, Henry. 1949. Notes on Rana sylvatica in the Labrador Peninsula. Copeia 3: 168-172. Johansen, Frits. 1926. Occurrences of frogs on Anticosti Island and Newfoundland. Canadian Field-Naturalist 40(1): 16. Jukes, J. B. 1842. Excursions in and about Newfoundland during the years 1839 and 1840. Volume 2. John Murray, London, 354 pp. Kindle, E. M. 1924. Geography and geology of the Lake Melville district, Labrador Peninsula. Memoirs of the Geographical Survey of Canada 141: 1-105. 46 THE CANADIAN FIELD-NATURALIST MacCulloch, Ross D., and J. Roger Bider. 1975. New records of amphibians and garter snakes in the James Bay area of Quebec. Canadian Field-Naturalist 89(1): 80-82. Maret, E. 1867. Frogs on Newfoundland. Proceedings of the Nova Scotia Institute of Science I(3): 6. Packard, A.S., Jr. 1866. List of vertebrates observed at Okak, Labrador, by Rev. Samuel Weiz, with annotations by A.S. Packard Jr. Proceedings of the Boston Society 10: 264-277. Packard, A. S., Jr. 1891. The Labrador coast. New York, 513 pp. Pope, Clifford H. 1944. Amphibians and reptiles of the Chicago area. Chicago Natural History Museum, 275 pp. Vol. 97 Uzzell, T. M., Jr. 1964. Relationships of the diploid and triploid species of the Ambystoma jeffersonianum com- plex (Amphibia, Caudata). Copeia 1964(2): 257-300. Wilson, C. V. 1971. The climate of Quebec — part one, climatic atlas. Atmospheric Environment Service, Toronto. Wynne-Edwards, V. C. 1952. Freshwater vertebrates of the arctic and subarctic. Fisheries Research Board Bulletin 94: 1-28. Received 2 October 1981 Accepted 15 October 1982 Fish Predation and Other Distinctive Features in the Diet of Nogies Creek, Ontario, Largemouth Bass, Micropterus salmoides JAMES G. HAMILTON! and P. M. POWLES Biology Department, Trent University, Peterborough, Ontario K9J 7B8 '\Present address: Booth Aquatic Research Group Inc., 532 Queen Street East, Toronto, Ontario MSA 1 V2 Hamilton, James G., and P. M. Powles. 1983. Fish predation and other distinctive features in the diet of Nogies Creek, Ontario, Largemouth Bass, Micropterus salmoides. Canadian Field-Naturalist 97(1): 47-56. Fresh and back-calculated lengths and weights of fish prey pumped from 1252 Largemouth Bass (Micropterus salmoides) aged one to eight (95 to 500 mm TL) produced GM regressions of Y = .27x - 3.6(lengths) and Y = .03x — 3.3 (weights) in the Nogies Creek sanctuary, 1977. The fish prey averaged only 3% of the predator’s body weight, well below average available sizes, and below sizes reported elsewhere. Of the six main forage species, four showed a wide length range (15 to 125 mm TL) and significant positive correlation with their predator length. Contrarily, very narrow length ranges of 70 to 80 mm TL for Golden Shiner (Notemigonus crysoleucas), and 25 to 55 mm TL for basses, were consumed. Pumpkinseed (Lepomis gibbosus) was the most abundant prey, followed by Rock Bass (Ambloplites rupestris), perch (Perca flavescens), and several cyprinids. Next to the consumption of “small-packaged” fish prey, the most distinctive feature of Nogies Creek Largemouth Bass diet was the high predation on tadpoles. An inverse relationship was found between frog and fish diets; when frogs and tadpoles were important (up to 34% by weight, ages three to seven), fish were not. A small crayfish component (maximum 13%, at age six) and very small insect component (except for age one), characterised the Nogies Creek diet in 1977. Ages one and two took Ephemeroptera, Odonata, Pumpkinseed, and Yellow Perch, and their diet was distinct from older bass which were primarily fish/ frog feeders. Bass fed from mid-April to mid-October, ceasing at 10°C water temperature. Key Words: Largemouth Bass, Micropterus salmoides, fish predation, diet. The feeding of Largemouth Bass (Micropterus sal- Rock and Largemouth Bass (Crossman 1956, and moidas) in lentic habitats has been well described but Muir 1960). The sanctuary’s size and characteristics only three U.S. reports deal with lotic populations, have been described in Hamilton and Powles (1979). only two with fish predation, andtherearefewstudies Its eutropic waters support dense aquatic plant of the species in its northern range. In fact, only one _growthand algal blooms throughout the summer, and other published study (Keast 1970) examines Large- oxygen may occasionally become low at the bottom mouth Bass feeding in Canadian waters. (1 mg/L), but not limiting to fish life in the rest of the The food and growth of juvenile or young-of-year water column. (YOY) Largemouth in Nogies Creek has already been Trap net and seining studies were already available described (Hamilton and Powles 1979). The object of _ to evaluate the relative abundance of fish prey species the present study was to describe food of the adults, in the community. ages | to 8 and over, with emphasis on fish prey sizes ingested with increasing age. Lewis et al. (1974), des- Field Methods cribed sizes of fish prey ingested by Largemouth Bass Bass were captured live, by boat-mounted electro- in the field and Lawrence (1958) and Tarrant (1960) shocker, similar to that of Novotny and Priegel showed a positive relationship between sizes of forage (1974). Standard 6- and 8-foot trap nets were also fish and the bass predator. But Wright (1970) failedto employed during April 1977 to October 1976 and demonstrate this in the laboratory. Our studies were 1977. In the winter of 1977, gill nets were set under ice. designed to document predation on Nogies Creek for- Bass, hand-netted after shocking, were anaesthet- age fish species to allow a future comparison with _ ized in 1:10 000 parts MS,,, ;,, to facilitate the stom- Muskellunge, and to test the validity of the “optimal ach pumping procedure. After loss of equilibrium, foraging theory” within an open weeded lotic habitat | they were removed from the bath, weighed, measured for Largemouth Bass. and subjected to gastric lavage (Crossman and Hamil- ton 1978) and ascale removed for aging. The fish were Methods allowed to recover and then returned to the water. Nogies Creek is an Ontario provincial sanctuary in Partially digested fish species were identified by use which environmental parameters and the fish com- _ of a reference collection of skeletons from the locality. munity have been studied for some years. Specific past | A bone possessing unique characteristics (as the oper- research projects have focussed on Muskellunge, culum in Newsome 1977) was removed from partially 47 48 THE CANADIAN FIELD-NATURALIST digested prey, and compared to a collection of disar- ticulated skeletons from Nogies Creek. The prey was then identified to species. Using various regression of bones on total length, a back-calculated original length of fish prey was obtained as in Newsome and Gee (1978), and Pikhu and Pikhu (1970). Lengths of fish prey were plotted as original lengths (mm TL), and weights were computed from previous records to allow direct comparison with other studies. To determine changes in diet with size, each Large- mouth Bass was assigned to a length group corres- ponding to an age-class. Stomach samples were obtained from a total of 1252 Largemouth Bass, from one to eight years of age and over. The oldest fish captured was 12 years. A number of bass were captured more than once but all data were included in the analysis. Identification of prey was to species for all samples used in the parti- tioning analysis; otherwise, to Family or Order, in the case of partially digested organisms (except for fish). Largemouth and Smallmouth Basses were grouped together as one prey type “basses”. To compare the diet of Nogies Creek largemouth to diets in other communities, frequency and weight were chosen as acommon denominator. Our samples were broken down by frequencies, and lengths were con- verted to wet weights. Weights of invertebrates were taken from Cummins and Wuycheck (1971) using average weights. Thus, an Ephemeropteran weight of 0.009 g and a Cladoceran weight of 3.5 x 10° were used to estimate weights of such small invertebrates. The weights of frogs ingested by bass were calculated from a length/ weight series of formalin-preserved specimens, since all frogs were discarded from the stomach contents at the time of the gastric lavage. Results Fish predation Of the 20 fish species in Nogies Creek, Pumpkin- seed, Lepomis gibbosus, Rock Bass, Ambloplites rupestris and Yellow Perch, Perca flavescens were most abundant in 1977 along with Largemouth Bass (Table 1). Less numerous in the catches were Golden Shiners, Notemigonus crysoleucas, and Smallmouth Bass, Micropterus dolomieu. Occasionally, Carp and Yellow Bullheads, Cyprinus carpio and Ictaiurus natalis occurred, along with 3 species of Notropis, though not all these species were trapped in 1977. Four species of frogs are common and numerous, and the invertbrate fauna (typical for the region), are quite diverse (see Hamilton and Powles 1979). Largemouth Bass from 95 to 500 mm TL consumed fish prey ranging from 19to 170 mm TL (Figure 1). A GM regression analysis (Ricker 1973) between length of bass and length of fish prey yielded the equation Vol. 97 TABLE |. Relative numbers of the common fishes taken over 428 trap-net-days in Nogies Creek, Ontario in 1977. Pumpkinseed Lepomis gibbosus 6655 Rock Bass Ambloplites rupestris 3832 Yellow Perch Perca flavescens 2170 Largemouth Bass Micropterus salmoides 479 Muskellunge Esox masquinongy 387 Golden Shiner Notemigonus crysoleucus 321 Brown Bullhead Ictalurus nebulosus 75 Smallmouth Bass Micropterus dolomieu 7 Common Shiner Notropis cornutus 6 Yellow Bullhead Ictalurus natalis l Y = 0.27X — 3.6, where Y is the total length of the prey and X the total length of the Largemouth Bass, both in mm. The correlation coefficient (r = 0.67) was signifi- cant at the 0.01 leveland the 95% C.L. on V = + 0.021. For all fish prey, a linear regression characterized Nogies Creek Largemouth Bass. For Illinois, the shad prey relationship was curvilinear. Largemouth Bass, varying from 12 to 2100 g, showed the GM relationship between predator and weight of fish prey of Y = 0.03X — 3.3, where Y is wet weight of the prey and X, the weight of the Large- mouth Bass in grams (Figure 2). This relationship was statistically significant (p< 0.01, r= 0.62) with the 95% C.L. on V = + 0.002. Within the size range exam- ined, an average Largemouth Bass consumed a fish prey 3% of its own body weight in Nogies Creek. Again, the mixed prey line of Nogies Creek was linear relationship, while that for shad of Illinois (the only other comparable field data available) was curvili- near. Furthermore, the average weight of shad was for above the prey line for mixed prey of Nogies Creek. When Nogies fish prey consumption of Large- mouth Bass was broken down by length for each species (Figure 3), three clusters were apparent: 1) The length distribution of one group of prey species (Pumpkinseed, Rock Bass, Perch and Shiners) showed a uniform increase with predator size from 25 to 120 mm TL; 2) only intermediate sizes of Golden Shiner (70 to 90 mm) were consumed, and 3) the lower size-range of the Basses were ingested, from 20 to 60 mm TL. Poor positive correlations with length were thus obtained between both Golden Shiner (r = 0.42) and Basses (r = 0.28), as opposed to Pump- kinseed, Rock Bass, Perch and Shiners (r = 0.98, 0.87 and 0.73 respectively). The most numerous fish species in the community was Pumpkinseed and these headed the prey list. There were 174 Pumpkinseed ingested by 753 Large- mouth Bass over the 1977 feeding season (Table 2). Yellow Perch was the second most common fish prey species in spite of Rock Bass being more numerous in the waters (Table 1). 1983 Gizzard 150 Shad 100 FISH PREY TL mm 50 Illinois —_—___» 200 HAMILTON AND POWLES: DIET OF LARGEMOUTH BASS 49 500 300 BASS LENGTH TL mm 400 FiGureE 1. GM regression (Nogies Creek) and eye-fitted regression (Illinois) for fish prey and lengths of predator, the Largemouth Bass, Micropterus salmoides. (Gizzard Shad prey data from Lewis et al. 1974.) Predation on perch was characteristic of all sizes of Largemouth, but 2- and 3-year olds (128 to 260 mm TL) ate slightly more perch than the larger sizes. Rock Bass were not eaten in quantities relative to their apparent abundance, but ranked third in importance by frequency. There was a tendancy for Rock Bass to be taken by the larger Largemouth Bass (over 305 mm TL). Of the less common species of fish prey (Table 1), Golden Shiners were taken reqularly by all ages except l’s, somewhat more frequently than would be suggested by their relative abundance in the community. Other Food Organisms Next to the “smallness” in the size of the fish food, the most distinctive feature of the diet of the Nogies Creek Largemouth Bass was the high frog compo- nent, particularly tadpoles (Table 2). All but the young (2- and 3-year olds) fed heavily on frogs, but ages 3 to 7 ate the most. Ages four and five were highest, with 76 tadpoles in 175 stomachs. An interesting inverse relationship in the Large- mouth Bass diet existed between frogs (tadpoles) and fish in this community (Figure 4.). In the stomachs of younger fish, tadpoles were a low component, 6%. As tadpoles built up in the diet with increasing age, the fish component decreased until age 5. Past age 5 the importance of frogs diminished and the fish contin- gent again became high. The third principal component, insects, (particu- larly Ephemeroptera) were taken in large numbers by Largemouth Bass, but only by I- and 2-year old bass. Older M. salmoides of Nogies Creek did not utilize this resource, nor did they consume many crayfish (25 in 753 stomachs, all ages), a common bass food. Food Resource Overlap with Size The type of food items ingested by different age groups of Largemouth Bass in Nogies Creek were compared statistically (Table 3), and the degree of overlap in the diets was assessed by the similarity coefficient, C,, as by Kislalioglu and Gibson (1977) 50 PREY Golden Shiners Other Shiners Rock Basses @ Pumpkinseed 130 Perch 8 Basses x A Oo x TL mm 100 50 PREY LENGTH 100 BASS THE CANADIAN FIELD-NATURALIST 200 Vol. 97 300 400 LENGTH TL mm FiGURE 2. Relationships of individual prey species’ lengths to length of predator, the Large- mouth Bass, Micropterus salmoides in Nogies Creek, 1977. *Includes: largemouth and smallmouth. and others. This coefficient varies from zero, when diets are completely distinct, to one, when the diets are similar in proportions of the prey groups or species. Thus, any value greater than 0.60 is accepted as indi- cating an overlap in diet. Table 3 revealed that |- and 2-year olds, while feed- ing on similar food, ate substantially different propor- tions of prey from all of the other age groups. From age 3 onwards, however, Largemouth Bass fed on similar proportions of the 12 main food items. The younger Largemouth Bass were responsible for most of the predation on Ephemeroptera and Odonata, as well as Pumpkinseed and Yellow Perch(Table 2). One other major difference was that age 3 fish and older ate crayfish, while the l- and 2-year olds did not. In summary then, 0-age largemouth from Nogies Creek ate mainly small and varied invertebrates (Hamilton and Powles 1979); I- year olds ate tad- poles, small fish and invertebrates, and from age 3 onwards, the diets overlapped, being characterized by intake of six species of “small-sized” fish prey, a high proportion of frogs and tadpoles, followed by crayfish (of low importance). Composition Through the Feeding Season With regard to the length of the feeding season, some food was present in the guts from April to October (Table 4). Before the end of October (and before freeze-up) less than 50% of the stomachs con- tained food (52% empty in 1976, 70% empty in 1977). At mid-April, only four stomachs (20%) contained food, which was mostly vegetation. The length of the feeding season was thus estimated at about six months, mid-April through mid-October. The gill nets which we set under the ice in winter caught no Largemouth Bass, and so this conclusion is naturally 1983 HAMILTON AND POWLES: FISH PREY g 5 10 15 20 BASS WEIGHT 'oog FiGURE 3. Weight of shad (Illinois, Lewis et al. 1974) prey and mixed fish-prey GM linear regression for Nogies Creek Largemouth Bass, Micropterus salmoides, 1977. provisional. The other 16 empty stomachs (from April) were tightly contracted and contained heavy mucus secretions, suggesting that feeding had not occurred recently (Keast 1965). In May, feeding commenced on animal contingents, such as ephemer- opterans Pumpkinseed, Golden Shiner, Rock Bass and Yellow Perch. Tadpoles were prominent in the diet along with the much smaller odonatan larvae. By June, ephemeropterans had diminished to less than 2% of the total number of food items that month, whereas tadpoles constituted 24% of the food items. Bolstered by large broods of free-swimming 0-year- old Largemouth Bass, the incidence of mixed large and Smallmouth Bass in the diet rose to 15% in June. Unidentified fishes, Pumpkinseed, Rock Bass, Golden Shiner and crayfishes were also important items in June (21.6%). In July, tadpoles (17.5%) and Pumpkinseed (17.8%) were the main food items, although Yellow Perch, Golden Shiners and Rock Bass were all of some importance. Gut contents in August were similar to those in July, although Pump- kinseed were much more important (26.1%) than tad- poles (9.4%). Fragments of vegetation were somewhat common in the diet of bass in August, amounting to 5.1% of a all food items in that month. In September, Pumpkinseed and tadpoles were again the two most frequently ingested food items (21.2 and 14.3% respec- tively), although Rock Bass and plant fragments represented a combined total of 25% of all stomach items. In summary, the combined food frequencies (Table 4) over the study period indicated that ephemeropte- rans (all life stages, 18.8%) were the most frequent food items ingested by adult and juvenile bass in May. Tadpoles and older frogs (16.1%) represented the second most frequent food item, and these were eaten DIET OF LARGEMOUTH BASS >I in June, July, September and October, but particu- larly June and October ’76 and ’77. Pumpkinseed (13 to 29%) were eaten mainly from July onward. Discussion The diet of Largemouth Bass in Nogies Creek Sanc- tuary was characterized primarily by a high fish com- ponent made up of prey less than 3% of the predator’s weight. Unlike other areas, the young did not become steadily more piscivorous with increasing age. The l-year olds ate a large number of small frogs and tadpoles, and this frog consumption continued over the life span of the Largemouth Bass, declining only past age five. This was probably a function of the Nogies Creek community, where frogs were extremely numerous, as were their tadpoles. In no other study of Largemouth Bass feeding was the consumption of tadpoles so high, nor the predation on crayfish so low. (Lewis et al. 1974; Seaburg and Moyle 1964; Snow 1971; and others). The frog diet appeared to comple- ment a fish diet composed mainly of small-sized for- age fish. The only other largemouth study reporting lengths of fish prey was that of Lewis et al. (1974). The lengths of Gizzard Shad, the only fish prey measured, were considerably greater in relation to the predator, than the fish prey of Nogies Creek Largemouth Bass (Fig- ure 1). The fish prey lengths were almost double those of Nogies Creek at the 300 mm TL predator length. The regression line for Gizzard Shad was curvilinear, whereas for our fish species it was linear for fish prey, as a whole, and for 4 of the six major prey species. The Gizzard Shad prey data of Lewis et al. (1974) when plotted on the same graph in fact, fell far above the Nogies’ fish prey regression line, except beyond preda- tor weight 1500 g. At a Largemouth Bass weight of 1800 g, the two prey regressions for Nogies and Crab Orchard Lake intersected, suggesting that the larger Gizzard Shad in Illinois waters were becoming increasingly unavailable, unattainable or not pre- ferred. (Sevino and Stein 1982). The other quantitative fish predation study directly comparable to ours (Lewis et al. 1974) showed that as Largemouth Bass increased in size, the prey (shad) length increased proportionately. The prey weight, expressed as a percentage of the predator weight, however, gradually decreased from 10% at 270 g to 3% at 1500 g. In the study by Lewis et al. (1974) this proportional reduction was probably a function, not of the maximum size of the prey (48 cm), but possibly of increased swimming speed of the prey as the size increased, or change in habitat of the larger shad. Within the length range of 175 to 484 mm TL, the average shad consumption was 6.2% of predator weight (Lewis et al. 1974). THE CANADIAN FIELD-NATURALIST Vol. 97 52 ESL (SE) OF @ )/) EO (9 ) 80 (CE) oS (Gr). O@ (Comair, (LZ) OF€ (OL) €6 (Se). hs (SI) Ce (LLI) 6 €Z (101) El (ep )) 1G (OG) LG (QS) Pe (KE )) CY (98) II (6L ) SOI (PLI) 1°€Z (SzS) L'69 SON 6% [B10 001 (AY) Ol (i) =I @) = (GS ) OG (> On (C) O72 @) oO| (9) 09 (ZI) O7ZI (SI) Oey (SZ) 0SZ (1) O€I (9) 09 (Ga 0ie (6) 06 (E) OF (EE) ESS (O01) OO (9%) 0°92 (%6) 0°76 SON % INA0 2 Q WIL (BE (€ ) (0 ) (0 ) (8 ) (0 ) (p ) (Z ) (9 ) (1 ) (ZI) (€1) (Ge) (¢ ) (1) (9 ) Ge) (Z1) (Z ) (12) (pr) SON 8S £0! L072 7 9€ 6SL % (L) (9 ) (0 ) (0 ) (L ) (UY ) (G )) (p ) (6 ) (¢ ) (27) (L2) (11) (¢ ) (0 ) (01) (S) (ZI) (¢ ) (PI) (¢S) SON IT yg ( }) Se AGIE) (Oe) EO). (G-)) 60 &) cy i) ors @ ) Qh aa (Ge) Sy (¢) 861 (SE) €r~ (8¢) é6 (©) 7) me (ES) WG (é) Le (¢ ) 0) ee Co Sy (@) 971 (81) 96r (Eb) % SON (9) $8 901 CIC 9°0S % (S) 06 rau (i Ce ©) Fe O} = @) = QQ) I (SE) Le CG) CE Mh) 6 OO) =— (io BO @) CY <<) eo (G) GG MY "GO (OD) il ©) ow Ce | (G) SY WE) ihn Gor SGT QE) <6? (re) ire ©) 29 Te) gaI (G) Ce G08 ) ip (G) St OM) it GD) rin WG) ¢€e GM) oH (Cup aS UG ep CG) CE Wo) CSI (Go) PYG (ES) ONE (Oy) PIS (5) GSS SON % SON % (p) (¢) (jayxoRig ul adv) T] Wu azIs sseg (I (S€ (LE C81 -~- SS OO rOTéLr Yr rer Tr Yr ONTr Tr YN wa ) ) (Zr) SON (Z) WU O8E-19E WW YOE-LEE WW OFE-COE WW POE-[9Z WW Y9Z-POZ WW E0Z-8ZI Cl sud} POO} YIM sseg Jo ‘ON CC 20) oe uolyead9\, aS () SPUR IT Ons S9}VIQIIIAUT 19Y1Q == (0) = HSIAAVUO te (0) E®) $19aSUI 19yIO LC (On gee e]eUOPpO I'L (A) LX via}dorswaydq rol - G) Ses SLOASNI SO - ©) = aytuaant 79 yNpy GS Ci) 29 sojodpey OO py - 19 SOOUA biz (€) 002 Yst} PalfUusptu/) ie (0) = satoads ystj 124109 iG OQ. = JoUIYS ulyoyoRg by (@) = IaUIYS UapjoyH sseg yinouljeuls 8'€ (O)-- pure yynowasie7 OD LB) sseg YoY Gol () £9 Yolad MOTAA COC (7) eee 9G poosuryduing 082 (6) 0°09 HSIA Zo SON % SUID}I poo} (1) uuu /71-OS ‘dnoid (ade) azts Aq ‘1/6 | ‘4991 SAISON Ul ‘saploups snsaidosaipy ‘Sseg YNOWITIL'] JO SYIRUIO}s dy} UI PUNO] sUISTURTIO Jo sloquinu pure adejud0I9g “7 AIEVL 1983 100 - 80 ae O Lu > 60 _— ee AG U mee. 20 [os AGE HAMILTON AND POWLES: DIET OF LARGEMOUTH BASS 53 IN YEARS FiGuRE 4. Relative weights of fish, frogs, and crayfish in the diet of Nogies Creek Largemouth Bass, Micropterus salmoides, 1977, by size and age-group. In an experimental study, Werner (1977) reports that Largemouth Bass were able to consume fish 6.5% their own body weight, and obviously Lewis et al.’s (1974) samples verified this. Our Nogies Creek sam- pling suggested that the proportional weight of fish prey fell far below this level. Largemouth Bass of 1000 g took fish of only 2.5% of their own body weight, and bass of 2000 g consumed fish of 2.9% their body weight. Thus, it would appear that Nogies Creek bass did not feed at Werner’s (1974) “maximum effi- ciency” (optimal foraging theory) level for fish prey. Unfortunately, there are no other field data published on prey size consumption for Largemouth Bass. Snow (1971) found prey to be mainly Bluegill, Bullheads and perch, but the weights of the fish prey organisms were not separated by predator size group, and more than half of the total food was crayfish, both by weight and frequency of occurrence. In Nogies Creek, frogs and tadpoles represent a high calorie diet (5 kcal per gram) as opposed to crayfish, at 2 kcal per gram dry weight (Cummins and Wuycheck 1971). The population thus ate small packages of high energy food rather than consuming the more typical higher weight compo- nents offered by crayfish. Snow (1971), for example, found 54.6% by weight (50% by frequency) of crayfish in Wisconsin Largemouth Bass. Generally, crayfish are not abundant in areas with detritus bottoms (Ber- rill 1978) such as Nogies Creek. Abundance of frogs and scarcity of crayfish probably explain the relatively heavy utilization of frogs and fish in 1976. It is possible, though purely speculative, that the fast-growing muskellunge of the Nogies sanctuary, may “skim off’ the larger prey species (Hourston 1952) leaving the smaller prey for Largemouth Bass. The other possibility is that predator success is reduced in the heavy vegetation (Savine and Stein 1982) and the smaller prey are relatively more numer- ous than the larger (older) fish prey. The most efficient feeding would be achieved by ingesting larger forage as Largemouth Bass grow larger. Perhaps the tadpoles and frogs are relatively rich calorifically, and more easily digestible than cen- trarchids (no scales, thin skin), and being plentiful and slow, were easy to capture. This would have compen- sated for prey handling and smaller size of the food package in the energetics budget. Seasonal feeding trends were rather usual or typical for bass populations. For example, spring feeding of 54 THE CANADIAN FIELD-NATURALIST Vol. 97 TABLE 3. Comparisons of the |2 main food items! eaten by different age classes of Largemouth Bass in Nogies Creek. The figures in the table are values for the similarity coefficient (C) as determined by Kislalioglu and Gibson (1977) in which 0.60 denotes overlap. (N = 753) Age l 2 3 4 | — 0.91* 0.07 0.10 2 0.31 0.23 3 0.77* 4 5 6 7 S 8 5 6 7 8 & over 0.06 0.11 0.13 0.09 0.22 0.30 0.33 0.27 0.84* 0.89* 0.76* 0.64* 8.80* 0.72* 0.61* 0.47 0.86* 0.60* 0.52 0.82* 0.76* 0.90* *Denotes significant overlap in diets of the size classes. 'Pumpkinseed, Yellow Perch, Rock Bass, Bass (Largemouth and Smallmouth) Golden Shiner, Blackchin Shiner, tadpoles, frogs, ephemeropterans and odonatons (all stages), crayfish, and vegetation. ephemeropterans and canibalism on basses (Table 4) were probably no more than a reflection of biotic changes in the community and in water temperature. Heavy weed growth by July in Nogies Creek may have protected some prey and hampered the predator, which are less manoeuvreable in dense weeds than are cyprinids. More energy is also expended by bass in pursuit of prey in an area with dense cover than ina sparsely covered area (Glass 1971). These factors, less available large food, and high energy costs pursuing prey, could have reduced fish feeding efficiency of the Nogies Creek bass, explaining the heavy exploitation of slow tadpoles from June to September. Numbers, or frequency of prey items, while useful in qualitative comparisons characterizing a communi- ty’s food resource, often produce a false proportion or account in terms of bulk diet constituents. For exam- ple, although in May ephemeropterans ranked 18% by frequency, they amounted to only 1.6% by weight in the total diet for that month. In the case of fish prey, frogs and tadpoles, their importance was accentuated by documentation of weights. The fish component ranged from 91 to 66% by weight throughout the season, and frogs from 7.7 to 34% by weight, depending upon the sizes of the predator. The length and weight breakdown of fish prey in this study certainly makes us cautious in supporting the “optimal foraging” concept for Nogies Creek Lar- gemouth Bass. The community is of intermediate complexity, and productive, which should encourage predator stability (Glass 1971). Modified behaviour tactics are suggested on the part of basses and Golden Shiners, but we have no records on Muskellunge’s fish predation. This study demonstrates the need for more community-based and controlled feeding behavioural studies for prediction and interpretation of feeding data. Acknowledgments We would like to acknowledge with thanks, the cooperation and keen interest of Dr. E.J. Crossman, Director of Nogies Creek Station. Support was given by the Canadian National Sportsmen’s Show IR85 and N. R. C. grants A4672 to Dr. P. M. Powles and A1705 to Dr. E. J. Crossman. Dr. D. C. Lasenby read the manuscript and made helpful suggestions. Mr. Fergus McNeil’s assistance with the field work was invaluable and most appreciated. We are indebted to Miss D. E. Gray, who assisted in analysing and tabu- lating data. The Outboard Marine Corporation of Peterborough graciously consigned us a motor for the duration of this project, for which we are very thank- ful. We thank Trent University for its part during the completion of the senior author’s MSc thesis. Literature Cited Berrill, M. 1978. Distribution and ecology of crayfish in the Kawartha Lakes region of southern Ontario. Canadian Journal of Zoology 55(1): 166-177. Crossman, E. J. 1956. Growth, mortality and movements of a sanctuary population of maskinonge. Journal of the Fisheries Research Board of Canada 13(5): 599-612. Crossman, E. J., and J. G. Hamilton. 1978. An apparatus for sampling gut contents of large, living fishes. Environ- mental Biology of Fishes 3(3): 297-300. Cummins, K. W., and J. C. Wuycheck. 1971. Caloric equi- valents for investigations in ecological energetics. Interna- tional Association of Theoretical and Applied Limnology Communication 18. Glass, N. R. 1971. Computer analysis of predation energe- tics in the largemouth bass. /n Systems Analysis and Simu- lation in Ecology. Edited by B.C. Patten. Academic Press, N.Y. 607 pp. Hamilton, J. G., and P. M. Powles. 1979. Feeding habits and growth of young-of-the-year largemouth bass (Micropterus salmoides) near its northern limit, Nogies Creek, Ontario. Canadian Journal of Zoology 57(7): 1431-1437. 55 DIET OF LARGEMOUTH BASS HAMILTON AND POWLES 1983 Poulqwios YINoW]]eWIs pue YINoWwWAadIe], 660 Iv OS CSI 801 cL ££ 9] 4 syoeulojs Ajduia Jo ‘ON [RIO] CSCI 69 £0! P8t 06C 807 9¢1 07 (G4 poululexa Sseq YINOWITIR] “ON [PIO] ITI LI 9¢ OLT 897 8LC 68E 74 £C SUI9}T POO} “ON [BIOL (€Z ) 61 6¢ of Cl LC 61 (II re = I3Y1IO (9¢ ) EG Pv 6C CCl 1S 61 V0 ot 0001 ca UOT]EIATI A (Sr ) pe = rs cc CV CC 67 = = eByeUuUOPO ( 8°81 — — a — a Z'19 — — sigs I Letsdersbade (Sz ) 61 Te 81 AG 9 Os £0 5 = ystpAesD (Le ) IC ST 8 81 v0 (AG £0 ce OLD ssoly INpy (781) 6 el OLI € vl v6 CLI € VC L9 a PLI sajodpe (6€1) 90! 6S 6LI Cc Il £6 I~ 8C =a tv Yslf palfuapiu ys) (9¢ ) ag = a LY cv v0 VC cae — yslf 19q1O (ze ) LAG = 81 vl £9 CC 80 = = TouTysS urlyoyxoryg (IZ ) vs = 81 rs 06 OV ims aa 4 TOUIYS UIplory (Sz ) LS = VS 9 I'v 81 = = = xS9SSEG (06 ) 69 = CCl ps SL 6L 67 ar y9C ssegq Yooy (16 ) 69 6S ps 9SI 9 TT I] 97 = SAR YoI9q MOT[9A (161) 9'PI CEC CIC 197¢ SLI VO! L9 = sy paasuryduing (Aouanbayy [enioy) LL61 LL61 LL6I LL6I LL6I LL6I LL6I 9L6I Sw3a}I poo adejuadiod [v10 | pO das ‘ony A[ne oun Kew audy ‘pO OLIEIUO ‘YI91D saison ul sseg yinowasie7q apusanf pue ynpe 7¢Z] Aq patunsuod surat pooy Jo Aouanbay a8eyuaosad KPO; “p A1av | 56 THE CANADIAN FIELD-NATURALIST Keast, A. 1970. Food specialization and _ bioenergetic interrelations in the fish faunas of some small Ontario waterways, pp 377-411. /n Marine Food Chains. Edited by J. H. Steele. Oliver and Boyd, Edinburgh, England. Keast, A. 1977. Mechanisms expanding niche width and minimizing intraspecific competition in two centrarchid fishes. Evolutionary Biology 10: 333-395. Kislalioglu, M., and R. N. Gibson. 1977. The feeding rela- tionship of shallow water fishes in a Scottish sea loch. Journal of Fish Biology 11(3): 257-266. Lawrence, J. M. 1958. Mouth size of largemouth bass in relationship to size of forage-fishes, lowa State College Journal of Science 32: 212-213. Lewis, W. M., R. Heidinger, W. Kirk, W. Chapman, and D. Johnson. 1974. Food intake of the largemouth bass. Transactions of the American Fisheries Society 103(2): 277-280. Muir, B.S. 1960. Comparison of growth for native and hatchery-stocked populations of Esox masquinongy in Nogies Creek, Ontario. Journal of the Fisheries Research Board of Canada 17(6): 919-927. Newsome, G. E. (Buck). 1977. Use of opercular bones to identify and estimate lengths of prey consumed by pisci- vores. Canadian Journal of Zoology 55(4): 733-736. Newsome, G. E. (Buck), and J. H. Gee. 1978. Preference and selection of prey by creek chub (Semotilus atromacu- latus) inhabiting the Mink River, Manitoba. Canadian Journal of Zoology 56(11): 2486-2497. Novotny, D. W., and G.R. Friegel. 1974. Electrofishing boats: Improved designs and operational guidelines to increase the effectiveness of boom shockers. Department of Natural Resources, Wisconsin Technical Report Number 73. 48 pp. Pikhu, E. K.,and E. R. Pikhu. 1970. Reconstruction of the size of fishes swallowed by predators from fragments of their vertbral column. Journal of Ichthyology. (USSR) 10(5): 706-709. Vol. 97 Ricker, W. E. 1973. Linear regressions in fishery research. Journal of the Fisheries Research Board of Canada 30(3): 409-434. Savino, J. F., and R. A. Stein. 1982. Predator-Prey Inter- action between Largemouth Bass and Bluegills as Influ- enced by Simulated, Submersed Vegetation. Transactions of the American Fisheries Society 111(3): 255-265. Seaburg, K.G., and J.B. Moyle. 1964. Feeding habits, digestive rates, and growth of some Minnesota warmwater fishes. Transactions of the American Fisheries Society 93(3): 269-285. Snow, H.E. 1971. Harvest and feeding habits of large- mouth bass in Murphy Flowage, Wisconsin. Department of Natural Resources, Wisconsin, Technical Report Number 50. 25 pp. Tarrant, R. M., Jr. 1960. Choice between two sizes of for- age fish by largemouth bass under aquarium conditions. Progressive Fish Culturist 22: 83-84. Warden, R. L., and W. J. Lorio. 1975. Movements of lar- gemouth bass (Micropterus salmoides) in impounded waters as determined by underwater telemetry. Transac- tions of the American Fisheries Society 104(4): 696-702. Werner, E. E. 1974. The fish size, prey size, handling time relation in several sunfishes and some implications. Jour- nal of the Fisheries Research Board of Canada 31(9): 1531-1536. Werner, E. E. 1977. Species packing. and niche comple- mentarity in three sunfishes. American Naturalist 111: 553-578. Wright, L. D. 1970. Forage size preference of the large- mouth bass. Progressive Fish Culturist 32: 39-42. Received 6 May 1981 Accepted 30 October 1982 Effects of Moose, Alces alces, on Aquatic Vegetation in Sibley Provincial Park!, Ontario D. FRASER2 and H. HRISTIENKO? Wildlife Research Section, Ontario Ministry of Natural Resources, Box 50, Maple, Ontario LOJ 1E0 'Ontario Ministry of Natural Resources, Wildlife Research Section Contribution Number 81-15. 2Present Address: Animal Research Centre, Agriculture Canada, Ottawa, Ontario KIA 0C6, where requests for reprints should be sent. 3Present Address: Wildlife Branch, Manitoba Department of Natural Resources, Box 14, 1495 St. James St., Winnipeg, Manitoba R3H 0W9 Fraser, D.,and H. Hristienko. 1983. Effects of Moose, Alces alces, on aquatic vegetation in Sibley Provincial Park, Ontario. Canadian Field-Naturalist 97(1): 57-61. The effect of Moose (Alces alces) on aquatic vegetation was studied in Sibley Provincial Park, Ontario. Two small exclosures were built in preferred feeding lakes to protect vegetation from Moose. Both exclosures developed a dense growth of plants. In one, species palatable to Moose were much more abundant than in unprotected areas. The other lake supported very little vegetation outside the exclosure. Aquatic vegetation ina large preferred lake underwent a series of changes from the 1960’s to 1980. Nuphar variegatum and Potamogeton filiformis largely disappeared, leaving the lake sparsely vegetated in some years and dominated by annuals in-others. With recent reductions in Moose activity, Nuphar is becoming re-established. In the 23 lakes studied, Nuphar variegatum was absent or scarce in areas heavily used by Moose, but Potamogeton foliosus, an annual, was most common in such sites. Key Words: Moose, Alces alces, aquatic plants, feeding behaviour. With their remarkable appetite for twigs, Moose While studying aquatic feeding by Moose, we real- (Alces alces) can have a profound effect onthe forest ized that the animals had an important influence on communities that provide the bulk of their diet (e.g., | boththe abundance and species composition of aqua- Janke 1976). Much less is known, however, about the tic vegetation. Here we describe this influence by influence of Moose on their aquatic habitats. reporting (1) a study using exclosures in two ponds Isle Royale provides some of the best documenta- __ heavily used by Moose, (2) plant succession in a well tion. Evidently Water Lilies (Nymphaeaceae) were used lake, and (3) species composition of 23 lakes in plentiful on the island in the early 1900’s before Moose __ relation to Moose activity. became established. By 1930, when Moose were par- ticularly abundant, aquatic vegetation was greatly Study Area depleted: Water Lilies were “practically gone” and The study was done in Sibley Provincial Park, Onta- pondweeds (Zosteraceae) were scarce (Murie 1934). _—_ rio (48° 20’N, 88° 45’W), a 243-km? peninsula of prin- Aquatic vegetation had apparently recovered some- ___ cipally boreal forest on the north shore of Lake Super- what by 1960 (Krefting 1974), but was againdepleted _ ior. The Park is closed to hunting and trapping, and around 1970 (Jordan et al. 1973), corresponding to —_ supports conspicuous populations of Moose, White- fluctuations in Moose numbers described by Peterson Tailed Deer (Odocoileus virginianus), Black Bears (1977). (Ursus americanus), Beavers (Castor canadensis), and Similar evidence was provided by Wright (1956) in | some Timber Wolves (Canis /upus). The Park includes New Brunswick. He used aerial photographs to esti- 37 named lakes, plus numerous small, unnamed lakes mate the abundance of Water Lilies in areas where and ponds. Most of the water bodies are <50 ha in Moose were common, and concluded that the plants —_ surface area and have soft organic sediments; many are were depleted in lakes accessible to Moose. He also. <2 m deep. Rock outcropping in the peninsula con- recorded a depletion of Water Lilies around 1912 sists mainly of Precambrian sediments which impart a when Moose had been particularly abundant in the _ higher mineral content to the lake water than is com- province. mon in granitic areas of the Precambrian Shield. Total In an experimental study, Aho and Jordan (1979) alkalinity of the water is typically 15 to 150 ppm used exclosures to protect sections of ponds from CaCQ,, with pH from 6.5 to 8.5. Moose. At the end of the growing season, the standing crop was much greater in protected than unprotected Methods areas. The only apparent effect on plant species com- During June 1978, a sturdy exclosure was built in position was a relative increase in Potamogeton pusil- _ each of two small lakes, called Lakes 7B and 24A, that lus and Lemna minor in the protected parts of one _ were heavily used by Moose. (Locations on the UTM pond. Grid are 5369100 m N, 370500 m E for Lake 7B, and 37) 58 THE CANADIAN FIELD-NATURALIST 5358100 m N, 369800 m E for Lake 24A). Each exclo- sure covered 4.5 m2 of lake bottom in about 50 cm of water. They consisted of steel pipes driven into the lake sediment, with cross bars secured by scaffolding clamps. The structure was covered in welded wire screening with openings 3 x 3 cm. Exclosures were inspected several times in 1978 to 1980. Plants growing inside were harvested on 12 July 1979 and 29 July 1980 for Lakes 24A and 7B respec- tively. Lake bottom inside the exclosures was divided into 12 plots each 61 x 61 cm. The vegetation in each plot was uprooted by hand. It was then sorted by species; roots and rhizomes were removed from all species except Eleocharis acicularis (which has a very small root system); the number of stems was counted for all species except E. acicularis and the Characeae; and the wet weight of each species was recorded after the vegetation had drained in the air for 5 sec. A strip of five control plots, also 61 x 61 cm, was marked out on each side of the exclosures in water of similar depth and over a similar type of substrate extending 1-4 m from the exclosure. Plants in these plots were harvested and treated in the same way. Plants elsewhere in the lake were examined to ensure that control plots were typical of similar parts of the lake. Moose activity was studied from 1976 to 1980 in Joeboy Lake (UTM Grid 5370000 m N, 372500 m E), a known feeding area for Moose. Observations each year consisted of five to eight 3-h shifts between mid- June and mid-July, with half the shifts from 0600 to 0900 and half from 1800 to 2100. During shifts, an observer at a fixed shoreline location scanned the lake every 15 min and recorded the number of moose seen. Cobus (1972) also studied Moose behaviour in Joe- boy Lake in 1971. He recorded the number of Moose present in the lake in scans at 15-min intervals throughout daylight hours about 2 days per week. Because he presented his data by time of day for the early summer (his Fig. 3), it was possible to calculate the mean number of moose seen per scan from 0600 to 0900 and from 1800 to 2100. His results are based on about 12 mornings and 12 evenings from | June to 14 July. In 1975 to 1979, we made occasional notes on the vegetation in Joeboy Lake in spring and early summer, and did a quantitative survey during | or 2 days during late summer (26 July to 3 September). Each survey was based on 342 to 410 non-permanent plots at about 3 m intervals along four transect lines which were traversed by canoe. Approximately the same lines were used each year. In each 75 x 75 cm plot, the observer esti- mated the percentage of lake bottom covered by each plant species as if seen from above. In 1980, the four lines were traversed more quickly, and notes were made on plant abundance. Vol. 97 Our entire study included 24 lakes in Sibley Provin- cial Park. Detailed information on Moose activity, lake surface area, mean water depth, shoreline type, organic content of the sediment, water chemical com- position, and plant abundance will be given in a separate report. Briefly, preliminary study indicated that the relative amount of aquatic feeding in a lake could be assessed at the end of the feeding season from the amount of Moose tracks, trails, summer dung piles, and other signs around the lake perimeter. Therefore, Moose activity on the 24 lakes was rated subjectively from high (5) to none (0) based on these signs in August and early September, 1977. Records of moose sightings at the more accessible lakes gener- ally agreed with the subjective ratings. Abundance of different species of aquatic plants was also estimated in each lake in late summer of 1977. A canoe or boat was paddled near the shoreline and in several lines criss-crossing the lake. Areas with <2 m of water were divided into convenient sections. Abundance of each plant species was estimated in each section as the approximate percentage of lake bottom covered by the plants as if seen from above without interference of other vegetation. Estimates from the sections were then combined to give estimated abundance in shal- low water areas (<2 m) for each lake. To determine the effect of Moose activity on vegeta- tion, we did stepwise multiple regressions of the abun- dance of each plant species (expressed as percentage of all vegetation in the lake) on the Moose activity rating and other variables. Only plant species repres- ented in 10 or more lakes were included. Logarithmic transformations were used to reduce skew in the dis- tributions. One lake (Calcite Lake) was omitted because the indirect and direct measures of Moose activity did not agree. Because the many variables gave considerable opportunity for chance associa- tions, the 1% level of statistical significance was used. For all plants listed, voucher specimens were pressed and are located in the herbarium of the Uni- versity of Waterloo, Waterloo, Ontario. Plant names follow Fernald (1950). Results Exclosures Lake bottom inside the exclosures was trampled during construction in June 1978. In the Lake 24A exclosure, numerous small Potamogeton alpinus and P. epihydrus were growing by late summer of 1978, apparently from rhizomes which had grown into the exclosure from outside. When harvested in July 1979, the exclosure was crowded with plants. Expressed as weight per plot, the exclosure contained 10 to 20 times more Potamogeton alpinus, P. epihydrus, and Chara- ceae than control plots, but similar amounts of Eleo- charis acicularis (Table 1). Plots in the exclosure also 1983 TABLE 1. Mean + SD weight (g) of aquatic plants harvested from 12 plots in the exclosure and 10 unprotected control plots in Lake 24A. Species Exclosure Control P! Eleocharis acicularis 169+ 83 190+119 ns. Potamogeton alpinus 121+93 5+4 <0.002 Characeae 335) a2 2A) ae 3 <0.002 P. epihydrus Mae DY Sse 3 <0.05 P. foliosus 5+4 0+0 £<0.002 Najas flexilis Snes) 2, se 22 n.s. Myriophyllum verticillatum 2+4 0+0 n.s. Hippuris vulgaris 0+0 1+3 n.s. P. berchtoldi 0+0 1+1 n.s. 'Statistical significance of the difference by the Mann- Whitney U test, 2-tailed. contained more stems of P. alpinus and P. foliosus than plots outside (P< 0.002). P. alpinus and P. epihydrus from the exclosure were more mature and had greater weight per stem. Exclo- sure plants had generally grown to the water surface and had developed floating leaves and fruit. Those outside were shorter, with only submersed leaves. By July 1978, the 7B exclosure had developed a dense growth of Potamogeton foliosus. The rest of the lake was sparsely vegetated, mainly with small clumbs of the same plant. In 1979, the expected crop of P. foliosus failed to develop in Lake 7B and in several other lakes. The exclosure contained a sparse growth of Characeae, similar to the rest of the lake, plus a few stems of Sparganium chlorocarpum and Sagittaria latifolia. By July 1980, the last two species formed a dense growth in the exclosure, but the rest of the lake had almost no vegetation. Plant weights per plot aver- aged 1890+920g (mean+SD) for Sparganium chlorocarpum and 30 + 60 g for Sagittaria latifolia in the exclosure. There was almost no vegetation in the 10 control plots. Joeboy Lake There was a pronounced decline in Moose activity TABLE 2. Number of 3-h observation shifts on Joeboy Lake during mornings and evenings, first and last dates of shifts, and mean number of Moose seen during scans of the lake in six years. Year Shifts Dates Mean no. 1971! 24 I Jun-14 Jul 5.0 1976 5 21 Jun- | Jul 2.6 1977 8 21 Jun-14 Jul 0.4 1978 8 17 Jun-16 Jul 1.9 1979 7 25 Jun-17 Jul 0.7 1980 8 19 Jun-15 Jul 0.4 'Recalculated from Cobus (1972, Figure 3). FRASER AND HRISTIENKO: EFFECTS OF MOOSE ON VEGETATION 59 on Joeboy Lake between 1971 and the later years (Table 2). Cobus (1972) noted that Moose activity was highest in mid-June to mid-July. Therefore the inclu- sion of early June in his results, in contrast to our later results, could not account for the higher values in [o7ve Residents of the area report an abundant growth of yellow Water Lilies (Nuphar variegatum) in the lake in the 1960s. Subsequently the plant was almost elimi- nated. Cobus (1972) considered Potamogeton filifor- mis the most abundant and preferred plant in the lake in 1971. He presented a map showing “dense beds” of the plant covering about 30% of the lake (his Fig. 1), and noted that Moose concentrated on these beds. Cobus listed the larger pondweeds ( P. amplifolius and P. richardsonii) as second in importance to Moose in the lake, and Nuphar variegatum as third. Our 1975 survey showed only 6% of the lake’s surface to be covered in vegetation. The dense beds of P. filiformis previously mapped by Cobus were largely devoid of vegetation, and the species was judged to cover less than 0.1% of the lake. Most specimens of Nuphar were very small (leaf diameter about 3 cm) and may have been seedlings. Najas flexilis and Potamogeton foliosus, which grew as annuals, played a large but variable part in subsequent years. The 1975 survey estimated 6 to 7% plant cover. N. flexilis was the most common plant with 3% cover, followed by the large pondweeds at 2%. In 1977 and 1978, plant cover exploded to 42% and 55% respectively because of an abundant growth of P. foliosus and N. flexilis which together accounted for most of the vegetation. The larger pondweeds were next at 3%. P. filiformis, covering 1% of the lake, was much more common than in 1975 and 1976. In 1979 there was a poor crop of pondweeds and an almost total failure of P. foliosus in several lakes. Reasons for this are unknown, but the late spring may have contributed. Plant cover in Joeboy was 4%. However, there were many large specimens of Nuphar variegatum (leaf diameter of 20-30 cm). In 1980, large N. variegatum were judged to cover 1% of the lake, and Moose were seen feeding steadily on the species for the first time since our observations began in 1976. Based on the description by Cobus (1972), average water depth appeared not to have changed from 1971 to 1980. However, water depth declined from spring to late summer by 10-20 cm in most years, causing shallow shoreline areas to become exposed mud. Comparison of Lakes The abundance of Nuphar variegatum in the 23 lakes was closely related to Moose activity (Table 3). In the six lakes heavily used by Moose, N. variegatum comprised only 0.4+0.8% of the vegetation (mean + SD), compared to 7.1 + 6.2% in the other 17 60 THE CANADIAN FIELD-NATURALIST TABLE3. Moose activity rated from high (5) to none (0), and abundance of two plant species expressed as percent of all aquatic vegetation in the lake for 23 lakes. Moose Nuphar Potamogeton Lake activity variegatum(%) foliosus (%) 7B 4.6 0.0 82.0 24A 4.4 2.0 4.0 Joeboy 4.3 0.1 56.0 Gardner 4.0 0.1 30.0 15D 337/ 0.0 13.0 Talus 3a 0.01 0.0 Lizard DS) 19.0 0.03 Ferns 1.6 2.0 12.0 Norma 1.6 8.0 0.0 Pickerel eS) 7.0 2.0 7A he3} 17.0 2.0 ISA 3 0.7 3.0 Kay 1.2 5.0 0.0 Legend 1.0 0.6 0.0 Grassy 0.8 0.2 0.4 Holt 0.8 1.0 0.0 Sawbill 0.8 7.0 6.0 1A 0.6 3.0 2.0 Rita 0.6 13.0 0.01 13A 0.5 10.0 0.02 Helen 0.4 17.0 0.0 1B 0.3 5.0 0.1 Norwegian 0.3 5.0 9.0 lakes. Most specimens from heavily used lakes had small leaves (3-10 cm diameter), spindly petioles, and they rarely flowered. Conspicuous beds of N. variega- tum were confined to lakes with little Moose activity. The Moose activity rating was the only factor included in the stepwise multiple regression (r = —0.56, P<0.01). Potamogeton foliosus was most abundant in the heavily used lakes (Table 3). It comprised 31 = 32% of the vegetation in the six heavily used lakes (mean + SD), but only 2 + 4% in the other 17 lakes. Stepwise multiple regression included the Moose activity rating as the first factor (r = 0.61, P<0.01), and total phosphorus in the water as the second factor (r = 0.52, P< 0.01) related to abundance of the plant. Abundance of the other plant species was not signif- icantly related to Moose activity. Relationships between plant abundance and other lake variables will be reported separately. Discussion These studies suggest that Moose have a complex influence on aquatic vegetation. In some cases, presumably, the animals simply remove preferred food items by grazing. The dense growth of Potamogeton alpinus and P. epihydrus in one exclosure, and the initial growth of P. foliosus in Vole, the other, contrasted sharply with the sparse devel- opment of the same species in unprotected areas. Since these species are highly palatable to Moose (D. Fraser, unpublished data), their depletion outside the exclosure was likely due to grazing. Eleocharis acicu- laris is not a preferred food item, and was about equally abundant in protected and unprotected areas. Some species are probably depleted by Moose through mechanical disturbance rather than grazing. Emergent species such as Sparganium chlorocarpum and Sagittaria latifolia normally contain substantially less sodium than submersed or floating-leafed plants (Boyd 1978), and do not appear to be eaten a great deal by Moose. Suppression of these species in the unprotected areas of Lake 7B was probably caused by repeated disturbance when Moose sought preferred vegetation. In some areas, Moose activity may also influence plant growth by increasing the turbidity of the water (Aho and Jordan 1979), or by causing dis- turbed sediments to settle on the vegetation. Some plants appear to be particularly vulnerable to disturbance. Depletion of Nuphar variegatum by con- centrated aquatic feeding has been mentioned repeat- edly in the past and is fully supported here. Although not the most preferred species of aquatics (D. Fraser, unpublished data), Nuphar seem greatly affected by herbivory. For example, Seton (1953: 172) describes killing a bed of Water Lilies by repeated clipping of the leaves and petioles. Many other aquatic species, however, seem to survive substantial removal by her- bivores (Aho and Jordan 1979). In contrast to Nuphar, Potamogeton foliosus flour- ished in preferred feeding areas even though it is read- ily eaten by Moose (D. Fraser, unpublished data). Unlike most other pondweeds that usually developed from rhizomes or winter buds, P. foliosus commonly grew annually from fruit. This probably imparted a competitive advantage in areas where rhizomes are continually disturbed by Moose. Once the distur- bance is removed, as in the exclosure in Lake 7B, P. foliosus would presumably be choked out by perennials. In Joeboy Lake, the rapid changes of vegetation may have been influenced by the exceptionally shal- low water and soft sediment as well as the activities of Moose. However, the depletion of Nuphar variega- tum and the trend toward annuals ( Potamogeton foli- osus and Najas flexilis) are consistent with the appar- ent effects of Moose observed in other lakes. Moose seem particularly attracted by lakes with a mineral soil substrate such as the clay-bottomed Lake 7B and the silt-bottomed Lake 24A (cf. Fraser et al. 1980). Such sites can be puzzling when first encoun- tered. The intense Moose activity at Lakes 7B and 24A was obvious from tracks and trails, but 7B had very 1983 little vegetation, and 24A was dominated by Eleocha- ris acicularis which Moose rarely appear to eat. At first glance, one might suppose that these sites were unproductive aquatic habitat, and that Moose were attracted by something other than aquatic plants. The exclosures show, however, that these are fertile sites that would be crowded with aquatic plants but for the devastating effect of the Moose. We suggest that Moose continue to use these areas despite their depleted vegetation because of some superior quality of the plants imparted by the mineral soil bottom. Science belatedly came to recognize that the sediment is important in the nutrition of aquatic plants (Hut- chinson 1975: 276). Moose probably knew this long ago. Acknowledgments We are grateful to L. Walters and A. Hurly for assistance in the field, to Ministry of Natural Resour- ces staff at Sibley Provincial Park, Thunder Bay, and Maple for their kind cooperation, and to C. D. MaclInnes and J. D. Roseborough for support of the project. Dr. J. K. Morton kindly checked plant identi- fications. S. Strathearn, D. Baggley, and H. G. Lumsden provided valuable comments and assistance with the manuscript. Literature Cited Aho, R. W.,and P. A. Jordan. 1979. Production of aquatic macrophytes and its utilization by Moose on Isle Royale National Park. Pp. 341-348 Jn Proceedings of the First Conference on Scientific Research in National Parks. Edited by R. M. Linn. U.S. Department of the Interior, National Park Service Transactions and Proceedings Ser- ies No. 5. Boyd, C. E. 1978. Chemical composition of wetland plants. Pp. 155-167 In Freshwater Wetlands. Edited by R.E. Good, D. F. Whigham, and R. L. Simpson. Academic Press, New York. FRASER AND HRISTIENKO: EFFECTS OF MOOSE ON VEGETATION 61 Cobus, M. 1972. Moose as an aesthetic resource and their summer feeding behaviour. Proceedings of the North American Moose Conference and Workshop 8: 244-275. Fernald, M. L. 1950. Gray’s manual of botany, 8th edition. D. Van Nostrand Co., New York. Fraser, D., D. Arthur, J: K. Morton, and B. K. Thompson. 1980. Aquatic feeding by Moose Alces alces in a Can- adian lake. Holarctic Ecology 3: 218-223. Hutchison, G. E. 1975. A treatise on limnology, Vol. 3, John Wiley and Sons, New York. 660 pp. Janke, R. A. 1976. Moose-boreal forest ecology in Isle Royale National Park. Proceedings of the North Ameri- can Moose Conference and Workshop 12: 70-90. Jordan, P. A., D. B. Botkin, A. S. Dominski, H. S. Lowen- dorf, and G. E. Belovsky. 1973. Sodium as a critical nu- trient for the Moose of Isle Royale. Proceedings of the North American Moose Conference and Workshop 9: 13-42. Krefting, L. W. 1974. The ecology of the Isle Royale Moose with special reference to the habitat. University of Minne- sota Agriculture Experiment Station Technical Bulletin 297. 75 pp. Murie, A. 1934. The Moose of Isle Royale. University of Michigan Museum of Zoology Miscellaneous Publication 25. 44 pp. Peterson, R. O. 1977. Wolf ecology and prey relationships on Isle Royale. National Park Service Scientific Mono- graph Series 11. U.S. Government Printing Office, Washington, D.C. 210 pp. Seton, E. T. 1953. Lives of game animals, Vol. 3, Part 1. Charles T. Brantford Co., Boston. 412 pp. Wright, B.S. 1956. The Moose of New Brunswick. Nor- theastern Wildlife Station, Fredericton, N.B. 63 pp. Received 21 September 1981 Accepted 20 August 1982 Survival of Female Black Ducks, Anas rubripes, During the Breeding Season JAMES K. RINGELMAN!? and JERRY R. LONGCORE3 'School of Forest Resources, University of Maine, Orono, Maine 04469 2Present address: Colorado Division of Wildlife, 317 W. Prospect Rd., Fort Collins, Colorado 80526 3Maine Field Station, Patuxent Wildlife Research Center, U. S. Fish and Wildlife Service, University of Maine, Orono, Maine 04469 Ringelman, James K., and Jerry R. Longcore. 1983. Survival of female Black Ducks, Anas rubripes, during the breeding season. Canadian Field-Naturalist 97(1): 62-65. The Mayfield method was used to estimate the survival rate of 19 radio-marked, female Black Ducks (Anas rubripes) in southcentral Maine during 1977-80. An overall survival rate of 0.74 was estimated for the 121-day monitoring period that included the pre-laying and laying, incubation, brood rearing, and post-rearing stages. No differences in survival rates were detected among these stages. Two instrumented hens were killed by Red-shouldered Hawks (Buteo lineatus) and a third was killed by an unknown predator. We found no evidence that the attachment of radio transmitters affected hen survival. Key Words: Black Duck (Anas rubripes), biotelemetry, Maine, mortality, survival rate To understand changes in waterfowl populations it is important to know how annual mortality is parti- tioned between hunting and non-hunting causes, as well as among specific periods of the annual cycle (Anderson and Burnham 1976:42). It is particularly important to obtain such information for the Black Duck (Anas rubripes), a heavily-hunted species that has declined in numbers and for which seasonal mort- ality estimates are lacking. Because of low band re- covery rates during non-hunting periods and the diffi- culty in detecting deaths of wild, unmarked waterfowl, recent studies (Gilmer et al. 1974; Kirby and Cowardin in press) have used radio-equipped ducks to estimate survival during the breeding and post-breeding period. In the present study, we use biotelemetry to estimate survival rates of adult female Black Ducks in southcentral Maine breeding habitat. Study Area and Methods The 151 km? study area was located 30 km south- west of Bangor, Maine. The area supported 28-32 Black Duck pairs/ year on | 12 wetlands typical of the evergreen and deciduous scrub-shrub, deciduous forested, and emergent wetland types (Cowardinet al. 1979) found in the Northeast. Detailed descriptions of the study area are presented by Ringelman (1980). Nineteen female Black Ducks were captured during 1977-80 (five in April, 10in May, two in June, and two in July) with nest traps (Coulter 1958) and rocket nets (Wildlife Materials, Carbondale, Illinois: use of trade names does not imply U.S. Government endorsement of products). Birds were fitted with adjustable (Dwyer 1972) back-mounted radio packages (Cedar Creek Bioelectronics Laboratory) weighing 20-25 g with an effective range of 1.5-3.0 km. Transmitter life ranged 62 from 20 to 120 days (x = 63). Hens were located 3-4 times/ day from mobile tracking vehicles by conven- tional telemetry techniques. Locations of marked females were plotted on U.S.G.S. topographic maps (1:64 000) to a resolution of 1.0 ha. If a bird remained in an area for > 2 days and its radio signal indicated inactivity, we visited the site and determined whether the duck was dead or alive and healthy. Rustrack (model 288) recorders coupled to receivers (AVM Company model LA12) and 3-element yagi antennas allowed continuous monitoring of 16 females. Recorder stations detected attenuations in radio sig- nals that indicated duck activity. Visual observations of radio-equipped females were made once every three days, and individuals were occasionally flushed to reaffirm that transmitters were not inhibiting flight. Field necropsies of marked birds were conducted to determine cause of death and gen- eral condition. Survival rates were calculated by a method origi- nally developed for determining nesting success (May- field 1961, 1975) and recently applied in survival stu- dies of ducklings (Ringelman and Longcore 1982) and adult ducks (Kirby and Cowardin in press). Known days of survival represented days of “exposure”, and time of death was estimated as one-half the interval between the last date the bird was known to have been alive and the date that death was confirmed. The breeding season was partitioned into four stages: pre- laying and laying, incubation, brood-rearing, and post-rearing. Unsuccessful nesters were assigned to the post-rearing period following their final nesting attempt. Because hens were captured before or during incubation, birds were assigned to the brood-rearing or post-rearing stages based on behavioral data 1983 obtained by telemetry. Estimates of the duration of the prelaying-laying (16 days) and incubation (26 days) stages were used to back-date reproductive phe- nology and determine exposure for these early inter- vals. Comparisons of survival rates follow procedures described in Johnson (1979). Results and Discussion Three of 19 instrumented Black Ducks died during the monitoring period; one during pre-laying and lay- ing, two during post-rearing and none during incuba- tion and brood-rearing (Table 1). Differences in sur- vival rates could not be detected among stages (p > 0.10); thus the overall survival rate for the 121- day season was estimated as 0.74 (the probability of surviving | day, 0.99748, times itself 121 times). All dead birds had been killed by predators. Two ducks (one pre-laying, one post-breeding) were killed by Red-shouldered Hawks (Buteo lineatus), and a third female just completing the wing-molt was killed by an unknown predator. Another unmarked, incubating hen (not included in our estimate of survival) was killed by a Raccoon (Procyon lotor). All predator- killed ducks were in good body condition when reco- vered, exhibiting no lesions or other debilitating physi- cal effects attributable to the transmitter packages, and two of those birds whose esophagi were examined had been feeding immediately before death. Although it has been suggested that predation accounts for a small portion of nonhunting mortality in waterfowl (Stout and Cornwell 1976), it may havea disproportionately large effect on population levels by affecting nesting success (Duebbert and Lokemoen 1980) and sex ratios (Johnson and Sargeant 1977). Reduced survival rates of breeding and molting females, when examined in a computer simulation model, showed a potentially significant effect on the Black Duck population (Ringelman and Longcore 1980). Common predators of adult Black Ducks include large raptors (Mendall 1944, Wright 1954; Appendix II), Red Foxes (Vulpes fulva; Wright 1954: Appendix II), and Raccoons (Stotts 1959;168, Stotts and Davis 1960). Although Mendall (1944) reported RINGELMAN AND LONGCORE: SURVIVAL OF BLACK DUCKS 63 no waterfowl remains in 14 stomachs of Red- shouldered Hawks in Maine, this species obtains most of its prey from wetland habitats (Bednarz and Dins- more 1981). Thus Red-shouldered Hawk predation on Black Ducks in our study is not implausible. It is important to assess the effects of instrumenta- tion on survival because the validity of survival rate estimates derived from radio-equipped ducks is con- tingent upon marked ducks experiencing the same mortality factors as the wild, unmarked population. We flushed instrumented females on over 30 occa- sions, and none exhibited reduced flight ability. Some hens preened excessively around the transmitter dur- ing the week following instrumentation, but this behavior was not apparent after a one-week acclima- tion period. However, we could not determine whether the death of the hen during the pre-laying and laying period, which occurred during the first week of instrumentation, was related to behavioral changes associated with instrumentation. Three Black Ducks (1 male, 2 females) with radio packages still attached one year following instrumentation, were observed on the study area and all were paired with new mates and behaving normally. One of these females that we recaptured was of average weight and body condition, and showed no skin abrasion or feather wear beneath the harness or transmitter. Habitat preferences of marked and unmarked Black Ducks on our study area were similar (Ringelman 1980:22). Caution must be used when interpreting published accounts of the effects of transmitter packages on duck behavior and survival (Schladweiler and Tester 1972; Greenwood and Sargeant 1973; Gilmer et al. 1974; Kirby and Cowardin in press), because harness configuration, transmitter size and weight, tempera- ment of individual birds, and especially the care taken in installing the harness may influence behavior and survival. A review of previous studies, along with our data on instrumented Black Ducks, leads us to con- clude that the capture, handling, and instrumentation of free-living Anas species does not cause biased sur- vival estimates during the breeding and early post- breeding seasons. TABLE |. Survival rates of radio-marked, female Black Ducks during four stages of the breeding season. Period length Period N (days) Pre-laying and laying 18 16 Incubation 15 26 Brood-rearing 6 45° Post-rearing 13 34° Totals 121 95% Exposure Period Confidence (days) Deaths survival rate interval 278.5 I 0.94 (0.84-1.06 299.0 0 1.00 175.0 0 1.00 438.5 2 0.86 0.69-1.07 1191.0 3 “Length equals the average number of days that two radio-marked hens remained with broods (43-48 days). b : ° . F . Length equals the average number of days birds were monitored during this period. 64 THE CANADIAN FIELD-NATURALIST The breeding season survival rate (0.74) for Black Ducks does not differ (P> 0.50) from survival esti- mates obtained from radio-marked female Mallards (0.73) breeding in northcentral Minnesota (Kirby and Cowardin in press). These most recent biotelemetry data suggest that breeding and post-breeding season survival rates are lower than had been previously assumed (e.g. Cowardin and Johnson 1979, Ringel- man and Longcore 1980). Population management of waterfowl is accom- plished primarily through regulatory manipulation of the harvest. Yet, to recognize when and how much the harvest should be restricted or liberalized, managers should know the magnitude and timing of seasonal non-hunting mortality. For adult female Black Ducks in Maine the average annual survival rate is 0.51, slightly lower than the continental mean, 0.56 (W. Blandin, Office of Migratory Bird Management, per- sonal communication). Except for our estimate of Black Duck survival for the breeding season, other seasonal survival rates (post-breeding and molt, win- tering, and hunting periods) are unknown. If we assume, for purposes of illustration, that these rates approximate those of the closely related Mallard (A. platyrhynchos), then the product of the four seasonal survival rates would be breeding — Black Duck, 0.74 X post-breeding - Mallard, 0.947 (Kirby and Cowardin in press) X hunting — Mallards, 0.811 (Anderson 1975; Johnson and Sargeant 1977) X wintering — Mallard, 0.919 (Cowardin and Johnson 1979) = 0.520, a value close to the average annual survival rate for female Black Ducks in Maine. In actuality, Black Duck survival rates probably differ greatly from those of the Mallard at many times of the year. Only after additional research is conducted on season-specific survival can we hope to comprehend the dynamics of the Black Duck population. Acknowledgments This study was supported in part by contract #14- 16-0008-2125 from the Migratory Bird and Habitat Research Laboratory, U.S. Fishand Wildlife Service, and the University of Maine at Orono, School of Forest Resources. J. Boring, J. Connolly, J. Schoultz, M. Scott, J. Sease, and S. Staples helped with field work. We thank D. Gilmer, R. Kirby, J. Nichols and W. Blandin for providing helpful comments on early drafts of the manuscript. Two anonymous referees provided helpful revisions and one referee detected a critical arithmetic mistake, for which we are appreciative. Literature Cited Anderson, D. R. 1975. Population ecology of the mallard. Temporal and geographic estimates of survival, recovery, Voleo7 and harvest rates. U.S. Fishand Wildlife Service Resource Publication 125. 110 pp. Anderson, D. R., and K. P. Burnham. 1976. Population ecology of the Mallard. VI. The effect of exploitation on survival. U.S. Fish and Wildlife Service Resource Publica- tion 128. 66 pp. Bednarz, J. C., and J. J. Dinsmore. 1981. Status, habitat use, and management of Red-shouldered Hawks in lowa. Journal of Wildlife Management 45: 236-241. Coulter, M. W. 1958. A new waterfowl nest trap. Bird- banding 29: 236-241. Cowardin, L.M., V. Carter, F.C. Golet, and E. T. LaRoe. 1979. Classification of wetlands and deepwater habitats of the United States. U.S. Department of the Interior FWS/ OBS-79/31. 103 pp. Cowardin, L. M., and D. H. Johnson. 1979. Mathematics and Mallard management. Journal of Wildlife Manage- ment 43: 18-35. Duebbert, H. F., and J. T. Lokemoen. 1980. High duck nesting success in a predator-reduced environment. Jour- nal of Wildlife Management 44: 428-437. Dwyer, T. J. 1972. Anadjustable radio-package for ducks. Bird Banding 43: 282-284. Gilmer, D. S., I. J. Ball, L. M. Cowardin, and J. H. Riech- mann. 1974. Effects of radio packages on wild ducks. Journal of Wildlife Management 38: 243-252. Greenwood, R. J., and A. B. Sargeant. 1973. Influence of radio packs on captive Mallards and Blue-winged Teal. Journal of Wildlife Management 37: 3-9. Johnson, D. H. 1979. Estimating nest success: the Mayfield method and an alternative. Auk 96: 651-661. Johnson, D. H., and A. B. Sargeant. 1977. Impact of Red Fox predation on the sex ratio of prairie Mallards. U.S. Fish and Wildlife Service Wildlife Research Report 6. 56 Pp. Kirby, R. E.,and L. M. Cowardin. /n press. Seasonal survi- val of female Mallards from north-central Minnesota. U.S. Fish and Wildlife Service, Resource Publication. Mayfield, H. 1961. Nesting success calculated from expo- sure. Wilson Bulletin 73: 255-261. Mayfield, H. 1975. Suggestions for calculating nest success. Wilson Bulletin 87: 456-466. Mendall, H. L. 1944. Food of hawks and owls in Maine. Journal of Wildlife Management 8: 198-208. Ringelman, J. K. 1980. The breeding ecology of the Black Duck in south-central Maine. Ph.D. thesis. University of Maine, Orono. 89 pp. Ringelman, J. K., and J. R. Longcore. 1980. Computer simulation models as tools for identifying research needs: a Black Duck population model. Proceedings of the North- east Wildlife Conference 37: 182-193. Ringelman, J. K., and J. R. Longcore. 1982. Survival of juvenile Black Ducks during brood rearing. Journal of Wildlife Management 46: 622-628. Schladweiler, J. L., and J. R. Tester. 1972. Survival and behavior of hand-reared Mallards released in the wild. Journal of Wildlife Management 36: 1118-1127. Stotts, V. D. 1959. A study of the breeding ecology of Black Ducks on the upper eastern shore of the Chesapeake Bay in Maryland from 1953 to 1959. Pittman-Robertson Report W-30-R-7. 231 pp. 1983 RINGELMAN AND LONGCORE: SURVIVAL OF BLACK DUCKS 65 Wright, B.S. 1954. High tide and an east wind. Stackpole Company, Harrisburg, Pennsylvania, and the Wildlife Management Institute, Washington, D.C. 162 pp. Stotts, V. D., and D. E. Davis. 1960. The Black Duck in the Chesapeake Bay of Maryland: breeding behavior and biology. Chesapeake Science |: 127-154. Stout, I. J., and G. W. Cornwell. 1976. Nonhunting mor- tality of fledged North American waterfowl. Journal of Wildlife Management 40: 681-693. Received 7 October 1981 Accepted 19 October 1982 Winter Movements of Arctic Foxes, Alopex lagopus, in a Petroleum Development Area LESTER E. EBERHARDT,! ROBERT A. GARROTT,2 and WAYNE C. HANSON? !Applied Ecology Section, Pacific Northwest Laboratory, P.O. Box 999, Richland, WA 99352 2Group LS-6, Los Alamos National Laboratory, Los Alamos, NM 87545 3Dames and Moore, 800 Cordova Street, Suite 101, Anchorage, AK 99501 Eberhardt, Lester E., Robert A. Garrott, and Wayne C. Hanson. 1983. Winter movements of Arctic Foxes, Alopex lagopus, in a petroleum development area. Canadian Field-Naturalist 97(1): 66-70. A total of 35 Arctic Foxes (A/opex lagopus) were radio-tracked in the vicinity of the Prudhoe Bay oil field in northern Alaska from October 1979 through May 1980. Dispersal of Arctic Foxes from Prudhoe Bay appeared to take place in the fall and late winter—early spring periods. Most juveniles remained in their natal home ranges through at least January and relatively few individuals participated in fall dispersal. The late winter—-early spring dispersal involved a large proportion of marked animals and may be related to breeding activities. Arctic Fox use of developed sites and associated garbage and handout food sources reached a peak in December when large concentrations of foxes were common around dumps and other development-related facilities. Foxes appeared to use well-defined areas and did not move freely between developed sites within the Prudhoe Bay area. Key Words: Arctic Fox, Alopex lagopus, radio-telemetry, fall and winter movements. Little information is available on the fall, winter, bins, dumps, and incinerators, which attract Arctic and early spring movements and activities of individ- | Foxes, are associated with most development facilities ual Arctic Foxes (Alopex Jagopus). During this period and are located throughout the Prudhoe Bay area. Arctic Fox populations appear to be highly mobile | Hand-feeding foxes is unlawfully practiced by some with long-distance movements occurring (Eberhardt _ oil-field personnel. and Hanson 1978) and even mass migrations taking The study was conducted from October 1979 place in some regions (Wrigley and Hatch 1976). through May 1980 on approximately 240 km? of Recently, large areas of Arctic Fox habitat have extensively developed habitat. Because of the experienced intensive exploration for and develop- expense, only periodic visits were made to the study ment of petroleum resources. However, the impact of site; these included: 11 days in early October, 10 days exploration and development activities on Arctic in early December, 7 days in mid-January, 6 days in Foxes is poorly understood. This paper describes the mid-March, and | day in late May. Arctic Foxes were results of a study on the fall, winter, and early spring __live-trapped at personnel camps, dumps, and dens in movements of Arctic Foxes in the vicinity of the October, December, January, and March. In addi- Prudhoe Bay oil field in northern Alaska. The pur- _ tion, several adult and juvenile Arctic Foxes, which poses of this study are to document dispersal times of | were captured in June and July 1979 during a study of Arctic Foxes from Prudhoe Bay, the pattern of use of juvenile behavior (Fine 1980), were monitored during the Prudhoe Bay area by individual Arctic Foxes our investigation. Captured foxes were anesthetized during winter, and the impact of winter petroleum- with a combination of ketamine hydrochloride and development activities on this species. xylazine (Cornely 1979); identified as to sex; weighed; equipped with ear tags and streamers (Eberhardt and Study Area and Methods Hanson 1978); and, in most instances, fitted with The Prudhoe Bay oil field is situated on the north- _radio-collars. Foxes were classified as adults or juve- ern coast of Alaska at 70°20’N. The area is character- _ niles (<1 yr old) by annuli counts of the lower first ized by long, cold winters; a flat, treeless topography; premolar extracted during handling or, in the case of approximately 75 cm of snowfall (Selkregg n.d.), and previously marked animals, by birth dates or past the absence of direct sunlight from mid-November _ breeding performance. through mid-January. Petroleum-related develop- Radio-collars operated on individual frequencies ment activities at Prudhoe Bay include the construc- between 216.000 and 216.999 MHz, weighed approx- tion of an extensive network of gravel roads and col- imately 3% of fox body weight, had an estimated life lection pipelines; the drilling of wells; and the estab- _— of at least 5 mo, and a range of approximately 2.5 km. lishment and operation of personnel camps, airports, | Several individuals were refitted with new collars dur- and petroleum processing centers. Garbage-holding ing subsequent trapping periods and could be fol- 66 1983 lowed for periods greater than 5 mo. In October, December, January, and March radio-collared foxes were located by triangulation from a single vehicle equipped with a null-peak yagi antenna system. An aerial search of the Prudhoe Bay area for radio- collared foxes was conducted in May with a light aircraft equipped with two yagi antennas. During each field period emphasis was placed on locating as many of the radio-collared foxes as possible. The Prudhoe Bay area was searched several times during each ground-tracking period. Radio-collared foxes were presumed to have left the study area if not located during two successive field trips. It is possible that some of the foxes assumed to have left the area actually had nonfunctioning radios; however, pre- vious spring and summer experience with these radios (Eberhardt et a/. 1982) demonstrated a low failure rate. The low environmental temperatures character- istic of the Arctic during the winter did not appear to adversely affect transmitter function, presumably because of the high insulative value provided by the long, dense fur of Arctic Foxes and our use of lithium batteries for the transmitters. We realize that the lack of a radio signal provides only indirect evidence of the absence of a marked animal from the study area; however, the extreme expense of logistics, harsh weather conditions, and spe- cies behavioral characteristics [long movements in short periods of time (Eberhardt and Hanson 1978)] preclude intensive efforts at locating dispersing radio- collared foxes outside the immediate Prudhoe Bay area. The term dispersal, as used in this paper, indicates only a movement of marked foxes away from Prud- hoe Bay. It is not used in reference to any specific age group, nor does it imply that the dispersing individu- als were a resident of the Prudhoe Bay area for any specific time period. It is possible that some foxes, particularly in late winter and early spring, were trapped and marked as they were moving through the Prudhoe Bay area. Results and Discussion A total of 35 Arctic Foxes (14 males and 21 females) were equipped with radio-collars and 12 (5 males and 7 females) with ear tags only. Eleven of the collared animals (5 juveniles males, 2 juvenile females, 2 adult males, and 2 adult females) were initially captured and fitted with radios during the summer prior to our investigation (Fine 1980). All radios installed during the summer were capable of functioning through the first few months of our study. In addition, five of these summer-caught individuals were recaptured and fit- ted with new radios at least once during our investi- gation. Table | summarizes data on radio-collared foxes, the expected life of radio-collars relative to the study period, and the dates of relocations. EBERHARDT, GARROTT AND HANSON: MOVEMENTS OF ARCTIC FOXES 67 It appears we were able to relocate foxes present in the Prudhoe Bay area that had functioning radio- collars. In only three instances did we apparently miss a radio-collared fox (foxes 780, 826, and 417) during one visit and subsequently locate it in the Prudhoe Bay area during a later visit (Table 1). Dispersal Radio-collared foxes disappeared and presumably dispersed from the Prudhoe Bay area during two time periods. The first period (fall) occurred between late August and late September. The second dispersal (winter-spring), involving a larger proportion of marked animals, began gradually after mid- December, accelerated sharply after mid-January, and continued through March (Table 2). Indirect evidence of fall dispersal is based on the disappearance prior to October of two (adult male 879 and juvenile male 229) of the 11 foxes radio-collared during the summer (Table |). Radios on these foxes functioned consistently throughout July and August and should have been working during our October and December trips if the animals were present. Radios installed on the other nine individuals at approximately the same time functioned through at least mid-January. In addition, adult male 879 de- finitely did disperse from the Prudhoe Bay area at some time because it was trapped in April 1982 near Baker Lake, Northwest Territories approximately 2000 km east of Prudhoe Bay. Juvenile Arctic Foxes residing on or near our study area decreased their use of dens late in the summer (Eberhardt 1977; Fine 1980; Garrott 1980). Several other investigators have also noted a decreasing use of dens by juveniles in August and September and have related this to dispersal of young (Dementyeff 1958; Shibanoff 1958; Tchirkova 1958; Macpherson 1969). However, we found that most of the radio-collared juveniles did not actually disperse from their natal home range during this period. Out of seven juveniles collared during the summer, one male (229), noted previously, dispersed prior to October and a second (male 390) dispersed 8.4 km within the Prudhoe Bay area between mid-October and early-December. The other five juveniles, representing portions of two lit- ters, remained in the vicinity of their rearing dens and were frequently located together through mid- January when, in most cases, their radios were due to stop. A juvenile female (417), refitted with new radio- collars twice, was located repeatedly on her natal home range through the end of the study in May. Tchirkova (1958) related both the timing and dis- tance of fall dispersal by juvenile Arctic Foxes to the availability of local food sources. If dispersal timing is related to food, then the readily available and heavily utilized “unnatural” food such as garbage and hand- 68 THE CANADIAN FIELD-NATURALIST Vol. 97 TABLE |. Summary of Arctic Fox capture, radio-collar, and relocation data at Prudhoe Bay, Alaska during 1979-1980 Study Periods Marked Foxed Relocated‘ Fox Period of Estimated a Sy ee Ee ea No. Sex! Age? Radio Life} June July Aug. Oct. Dec. Jan. Mar. May 381 M A June-February = s +F = e 780 F A June-October * * + * * 879 M A July-December = Ad 894 F A July-December + a ad 390 M J July-February - « = 5 229 M J July-December ~ 687 F J July-February is - * ty - 554 M J July-March x i * ce # 417 F J July-May ce a a o a 407 M J July-December 3 + 2 2 826 M J July-December * + : a 430 F J October-February = x“ 615 F J October-February x 261 F J October-February 3 = * 055 F J October-February < ~ ee 850 F J October-February A iS 940 F U October-February 2 a 915 F A October-February = iM 529 F J October-February iS 3 347 F J December-April a 489 M J December-April * * 523 Je U December-April iA * 149 je J December-April * < 361 F A December-April 2 * 449 M A December-April o Se M U December-April 927 le A January-May hd ar 100 F J January-May e 972 E U January-May « ‘ bs 115 F U January-May * 863 M J January-May * 337 M A January-May : 042 M J January-May * 162 JE U March-July * 994 M U March-July ‘ 'F — female, M — male. 2A — adult, J — juvenile, U — unknown. 3Estimated 5 mo radio life. Some foxes were fitted with new radio-collars during the study. 4* _ fox located, + — fox not searched for. TABLE 2. Relocation rate of radio-collared Arctic Foxes at Prudhoe Bay, Alaska, 1979-1980. Study Periods October December January March May Number of foxes which should have had working radios! 10 18 20 15 10 Numbers of radio-collared foxed located? 6 16 7 I 2 Percentage of radio-collared foxed located 60 89 35 7 20 'Radio life estimated at 5 mo. 2Foxes located after their radios had exceeded expected life were excluded. 1983 outs (Eberhardt 1977; Fine 1980; Eberhardt et al. 1982) at Prudhoe Bay may have influenced the disper- sal times of resident juveniles. The fall dispersal of adult male 879 occurred subse- quent to his participation in the rearing of young. We have recorded one other instance in which a marked adult male dispersed, in this case 1246 km, sometime after raising young (unpublished data). Fall dispersal of adult male Red Foxes ( Vulpes vulpes) has also been recorded (Storm et al. 1976, p. 31). Both adult and juvenile Arctic Foxes dispersed from the Prudhoe Bay area between mid-December and March. Evidence of the winter-spring dispersal of foxes from Prudhoe Bay is based on a decrease in the proportion of radio-collared animals relocated (Table 2), a reduction in the amount of fox signs observed around petroleum development facilities, and decreased trapping success as winter progressed. The decline in the relocation rate of radio-collared foxes began after mid-December and continued through the remainder of the study (Table 2). During the latter part of the investigation trapping success decreased approximately 50% from 0.46 fox/trap-night (35 trap-nights) in October and 0.50 fox/trap-night (44 trap-nights) in December to 0.21 fox/trap-night (29 trap-nights) in January and 0.16 fox/trap-night (26 trap-nights) in March. The decrease in fox trapping success is probably a reflection of both fewer foxes and the fact that a proportion of the fox population residing in the vicinity of our heavily trapped areas was captured early in the study and became “trap-shy” during later trapping periods. However, we also repeatedly captured “trap-prone” animals, and these individuals disappeared as well in the latter part of the study. Urquhart (1973), in a study of Arctic Fox use of mobile seismic camps in Canada, also noted a gradual decrease of fox use through the winter-spring period. He attributed the decline in fox numbers to a high winter mortality of juveniles and an avoidance of humans by adult foxes during the breeding season in February and March. Juvenile mortality did not appear to bea significant factor in the decrease in fox use of the Prudhoe Bay area since none of 19 radio- collared juvenile foxes were known to have died while on the study area. However, fox mortality has been important in the past at Prudhoe Bay when incidence of rabies was high (unpublished data). We believe the disappearance of large numbers of foxes from the Prudhoe Bay area in late winter and early spring was related to the establishment of territories for breeding purposes. This species has been shown to be territorial during the spring and summer months (Eberhardt et al. 1982). Distances moved by some dispersing Arctic Foxes during the winter-spring period were extensive. Two EBERHARDT, GARROTT AND HANSON: MOVEMENTS OF ARCTIC FOXES 69 adult males moved 781 and 2000 km to the east during this study and three other males from earlier tagging efforts at or near Prudhoe Bay (Eberhardt and Han- son 1978) moved between 338 and 901 km during the winter-spring periods. Developed Site Use The amount of developed site-use by individual Arctic Foxes during this study was variable but greater than observed during spring and summer stu- dies (Eberhardt et al. 1982). Intensity of use varied from moderate in October, heavy in December, mod- erate in January, and very light in March. During December many foxes spent large amounts of their time on or near developed sites. One juvenile female (417) was located repeatedly over a 5-d period in December within the compound ofa large oil process- ing center. Another juvenile female (687), a sibling of fox 417, moved back and forth between this process- ing center and a personnel camp 2 km away during the same time period. Both of these foxes used a much larger area prior to December, most of which was undeveloped. Radio-collared foxes generally used well defined areas within the Prudhoe Bay study site. Only 2 of the 22 radioed foxes that were relocated at least once during subsequent trips to Prudhoe Bay made move- ments in the study area that exceeded normal home range movements as defined by summer radio- tracking data (Eberhardt et al. 1982). These animals, juvenile male 390 and juvenile female 615, moved 8.4 and 12.5 km, respectively. In addition a juvenile male, equipped with ear tags only, moved 14.3 km within the Prudhoe Bay study area. Territorial defense by Arctic Foxes appeared to be less strong during the fall and winter when compared to spring and summer periods (Eberhardt et al. 1982). Concentration of fox activities around developed sites was common in the fall and winter and up to 21 foxes were captured and marked within a 63-ha area. Dur- ing past winter studies in the Prudhoe Bay area up to 23 foxes were captured at a single dump in7 d(unpub- lished data). Other investigators have reported large winter concentrations of Arctic Foxes near food sour- ces (Chesemore 1968; Urquhart 1973). A number of captured foxes were in poor condition. It was assumed that these individuals were either dis- eased or recent arrivals to Prudhoe Bay. Known resi- dent foxes feeding on garbage were generally fat. Fall and winter movements of Arctic Foxes to Prudhoe Bay froma distance of up to 1000 km away have been reported (Wrigley and Hatch 1976; Eberhardt and Hanson 1978). The fall and winter use of the Prudhoe Bay area by relatively large numbers of foxes has potential nega- tive and positive aspects. The increase in fox-to-fox 70 THE CANADIAN FIELD-NATURALIST and fox-to-man contact is undesirable from the standpoint of disease transmission, particularly rabies, which is enzootic in the species (Crandell 1975). A number of Prudhoe Bay oil field personnel have been exposed to and treated for rabies during the fall and winter, especially in those years when the disease is epizootic. “Unnatural” food sources may contribute to the over-winter survival of some Arctic Foxes. Grace (1976) and Fuller and Keith (1980) have also speculated that dump-foraging by Gray Wolves (Canis lupus) may contribute to survival in that species. Acknowledgments This research was conducted under United States Department of Energy Contract DE-AC06-76RLO 1830. We thank the Prudhoe Bay Field operators, Sohio Alaska Petroleum Company and ARCO Oil and Gas for permission to use the study site; the Alaska Department of Fish and Game for permission to trap and tag foxes; H. Fine for capturing and marking animals during the summer; W.M. Tzil- kowski for sectioning Arctic Fox teeth; and J.W. Helmericks for aircraft support. Literature Cited Chesemore, D .L. 1968. Distribution and movements of White Foxes in northern and western Alaska. Canadian Journal of Zoology 46: 849-854. Cornely, J. E. 1979. Anesthesia of Coyotes with ketamine hydrochloride and xylazine. Journal of Wildlife Manage- ment 43: 577-579. Crandell, R. A. 1975. Arctic Fox rabies. Jn The natural history of rabies. Edited by G. M. Baer. Academic Press, Incorporated, New York, Volume II. Pp. 23-40. Dementyeff, N. I. 1958. Biology of the Arctic Fox in the Bolshezemelskaya tundra. /n Translations of Russian Game Reports. Volume 3. Department of Northern Affairs and National Resources, Ottawa, Canada. Pp. 166-181. Eberhardt, L. E., and W. C. Hanson. 1978. Long-distance movements of Arctic Foxes tagged in northern Alaska. Canadian Field-Naturalist 92: 386-389. Eberhardt, L. E., W. C. Hanson, J. L. Bengtson, R. A. Gar- rott, and E. E. Hanson. 1982. Arctic Fox home range characteristics in an oil-development area. Journal of Wildlife Management 46: 183-190. Vol. 97 Eberhardt, W. L. 1977. The biology of Arctic and Red Foxes on the North Slope. M.Sc. thesis, University of Alaska, Fairbanks. 125 pp. Fine, H. 1980. Ecology of Arctic Foxes at Prudhoe Bay, Alaska. M.Sc. thesis, University of Alaska, Fairbanks. 77 Pp. Fuller, T.K., and L.B. Keith. 1980. Wolf population dynamics and prey relationships in northeastern Alberta. Journal of Wildlife Management 44: 583-602. Garrott, R.A. 1980. Den characteristics, productivity, food habits, and behavior of Arctic Foxes in northern Alaska. M.Sc. thesis, The Pennsylvania State University, State College. 95 pp. Grace, E. S. 1976. Interactions between men and wolves at an arctic outpost on Ellesmere Island. Canadian Field- Naturalist 90: 149-156. Macpherson, A. H. 1969. The dynamics of Canadian Arc- tic Fox populations. Canadian Wildlife Service Report Series Number 8. 52 pp. Selkregg, L. L. No Date. Alaska regional profiles: Arctic region. State of Alaska, Division of Planning and Research, Juneau. 218 pp. Shibanoff, S. V. 1958. Dynamics of Arctic Fox numbers in relation to breeding, food and migration conditions. /n Translations of Russian Game Reports. Volume 3. Department of Northern Affairs and National Resources, Ottawa, Canada. Pp. 5-28. Storm, G. L., R. D. Andrews, R. L. Phillips, R. A. Bishop, D. B. Siniff, and J. R. Tester. 1976. Morphology, repro- duction, dispersal, and mortality of midwestern Red Fox populations. Wildlife Monographs 49: 1-82. Tchrikova, A. F. 1958. A preliminary method of forecast- ing changes in numbers of Arctic Foxes. /n Translation of Russian Game Reports. Volume 3. Department of North- ern Affairs and National Resources, Ottawa, Canada. Pp 29-49. Urquhart, D. R. 1973. Oil exploration and Banks Island wildlife: a guide for the preservation of Caribou, Muskox, and Arctic Fox populations on Banks Island, N.W.T. Game Management Division Report, Government of the Northwest Territories. 105 pp. Wrigley, R.E., and D.R.M. Hatch. 1976. Arctic Fox migrations in Manitoba. Arctic 29: 147-158. Received 9 December 1980 Accepted 28 September 1982 Habitat Selection and Food Habits of Marten, Martes americana, in the Northwest Territories RICHARD J. DOUGLASS!, LORNE G. FISHER2, and MARNIE MAIR3 '613 West College, Bozeman, Montana 59715 238 Belmead Gardens, Edmonton, Alberta, Canada T5N 1P6 39847 83rd Avenue, Edmonton, Alberta, Canada TSN 1P6 Douglass, Richard J.. Lorne G. Fisher, and Marnie Mair. 1983. Habitat selection and food habits of Marten, Martes americana, in the Northwest Territories. Canadian Field-Naturalist 97(1): 71-74. Habit data were gathered at Chick Lake and Moon Lake, Northwest Territories on permanently marked transects by recording tracks. Marten selected different habitat in different seasons and study sites. Scats collected on transects showed food habits to be similar to other areas of North America in that voles were the major food of Martens. However Marten in the Northwest Territories appeared to be selective in species consumed in contrast to more opportunistic feeding habits reported for other areas. Key Words: habitat selection, food habits, Marten, Martes americana, Northwest Territories. Little is known about winter habitat selection and food habits of Marten, Martes americana, in northern environments. Although Marten prefer mature stands of coniferous forest (Seton 1929; Marshall 1951; de Vos 1952; Lensink et al. 1953; Francis and Stephenson 1972) it is not clear whether they select specific habi- tats within forests. There have been no reports pub- lished on winter food habits of Marten in the North- west Territories. We had the opportunity to collect some data relative to their habitat selection and food habits in this area peripheral to a study designed for other purposes. Objectives in this study were to determine if Marten selected specific habitats within coniferous forests during periods of continuous snow cover and to determine the prey of Marten during the same period. We conducted this study in the northern boreal forest near Chick Lake and Moon Lake, Northwest Territories. Chick Lake (65°52’N, 128°07’W) and Moon Lake (65° 37’N, 127°30’W) are located east of the Mackenzie River between Norman Wells and Fort Good Hope. The study area is covered by mature stands of Black Spruce (Picea mariana) with some intermixed Larch (Larix /aricina) and Paper Birch (Betula papyrifera). Ground cover is predominantly lichen (Cladonia spp.), moss (Aulocomium sp.) with low densities of shrubs (Ledum groenlandicum, Vac- cinium vitis-idaea, V. uliginosum, and Salix spp.), sedges (Carex spp.) and grass (Calamagrostis sp.). Methods We constructed 44 km(23.4 at Chick Lake and 20.6 at Moon Lake) of permanently marked transects, examining them every day it did not snow during spring 1973 (15 March to | May), fall 1973 (20 October to | December), spring 1974 (25 February to 7 28 March) and fall 1974 (21 October to 13 November). As tracks were recorded they were obliterated to avoid duplication of counts. Marten scats were collected during both years and habitat data were collected during 1974. Habitat preference was examined by testing the following hypotheses: H, the habit at Marten track crossings is not differ- ent than the overall habitat in which the tracks occur. H, the habitat at Marten track crossings is different than the overall habitat in which the tracks occur. Habitat preference is inferred when the habitat at track crossings is found to be significantly different from that of the general habitat. Habitat variables were estimated at each Marten track crossing and at 0.2 km intervals along all transects. The distribution of classes of variables estimated at systematic points was considered to be representative of the overall habitat. Because we could be at the transects only during times of snow cover only variables measurable above the snow surface were recorded. These were tree dispersion (Cottam and Curtis 1956), tree height, and shrub density. We estimated tree dispersion and tree heights in meters (dispersion = the mean distance to the four nearest trees in quarters, and height = the mean height of the same four trees) and assigned them to seven dispersion and I! height classes. We esti- mated shrub density according to the following classi- fication: Class |, shrubs present; Class 2, low density; Class 3, moderate density; Class 4, high density. Information on the diet of Marten was obtained from the analysis of scats collected along transects. We identified prey remains to genera according to techniques developed by Mathiak (1938) and Day (1966), and with the aid of keys and a reference collec- U2 THE CANADIAN FIELD-NATURALIST tion. We identified species of rodents by comparing teeth contained in scats with those of specimens col- lected in the study area. We did not attempt to identify the species of birds, invertebrates or plant material. Results and Discussion Habitat Preference During the sampling sessions we estimated habitat variables where 2,214 Marten tracks crossed tran- sects. We compared the distribution of habitat varia- bles estimated at track crossings with the distribution Vol. 97 of the same variables estimated in the systematic sam- pling (Table 1). These data were analyzed by a chi- square one-sample test (Siegel 1956) to determine whether habitat classes at track crossings were dis- tributed randomly in respect to expected distribu- tions. Except for tree dispersion and shrub density at Moon Lake during spring, the distribution of classes of habitat variables at Marten track crossings differed significantly from the distribution in the overall habi- tat. However, inter-area correlation coefficients were insignificant for all variables. This indicates that the TABLE |. Distribution of classes of habitat variables at Marten track crossings compared to the distribution of the same classes estimated in systematic surveys of transects. Significant values indicate that the habitat where Marten tracks were encountered was different from the overall habitat. Chick Lake Moon Lake Spring Fall Spring Fall Obs." Exp.” Obs. Exp. Obs. Exp. Obs. Exp. Dispersion (m) Classes > 3.0 12 9.90 240 47.94 21 25.08 153 112.42 2.6- 3.0 10 26.40 iD, 127.84 50 47.88 102 214.62 Pls DS) 17 52.80 120 255.68 59 50.16 154 224.84 1.6- 2.0 49 26.40 152 127.84 50 57.00 245 255.50 te WS 36 23.10 128 111.86 23 25.08 204 112.42 0.6- 1.0 33 11.55 80 55.93 1] 15.96 133 71.54 <0.5 8 14.85 l 71.91 5 6.84 31 30.66 Total 165 799 228 1,022 > xX? (6df)104.45*** (6df)953.12*** (6df)8.75 (6df)223.90*** Height (m) Classes 0.0- 2.0 51 39.6 120 191.76 41 52.44 184 235.06 2.1- 4.0 57 70.95 239 343.57 116 93.48 388 419.02 4.1- 6.0 40 34.65 19] 167.79 55 66.12 358 296.38 6.1- 8.0 13 9.9 144 47.94 11 11.40 82 51.10 8.1-10.0 2 6.6 96 31.96 5 4.56 10 20.44 10.1-12.0 2 3.3 8 15.98 0 0 0 0 Total 165 799 228 1022 XX? (Sdf)11.54* (Sdf)386.29*** (3df)9.84* (3df)50.22*** Shrub Density class’ ] 30 135.3 320 655.18 194 202.92 715 909.58 D 81 18.15 376 87.89 30 20.52 266 91.98 3 36 9.9 96 47.94 4 4.56 31 20.44 4 18 1.65 I 7.99 0 0 10 0° Total 165 799 228 1,022 > X2 (3df)530.41*** (3df)1,170.21*** (2df)3.46 (2df)376.31*** “Distribution of classes of habitat variables at track crossings b 6 4 ec 5 Values based on relative availability of various classes based on transect results. “I=present 2=low cover 3=moderate cover 4=dense cover d 3 F A - 6 F . Shrubs existed in the area of this density class but none were encountered in the systematic sample. *<0.05 #*D<0.01 *€*D<0.001 1983 distribution of variables at track crossings was incon- sistent between areas during both spring and fall. In most cases the distribution of habitat variables at Marten track crossings was different from the overall habitat (i.e. H, was rejected, suggesting that Marten selected sub-habitats within the forest for travel. However, the inconsistencies 1n distributions between areas are perplexing. Results may have been incon- sistent as a result of: (1) interactions of the variables we examined with other environmental variables that we did not measure or could not incorporate in the analysis (e.g., snow depth, daily temperatures, changes in rodent densities and distribution); (2) vari- ation in preferences demonstrated by individual Marten. Douglass (1977) showed that Yellow- cheeked Voles (Microtus xanthognathus) underwent small population fluctuations (0.5/ha to 5/ha) at Chick Lake and that fluctuations were asynchronous. Also, Marten tracks at Chick Lake were found in disproportionately high numbers in areas of low tree dispersion (Table 1), and captures of both Northern Red-backed Voles (Clethrionomys rutilus) and Yellow-cheeked Voles were found to be negatively correlated with tree dispersion (Table 2). Marten may have selected certain areas within the forest on the basis of the abundance and availability of prey rather than solely on the basis of the habitat structure. This is consistent with the higher Marten activity in areas of high prey abundance described by More (1978). If Marten in the Northwest Territories do select areas on the basis of prey abundance and availability, the spa- tially nonsynchronous population fluctuations of Yellow-cheeked Voles may have accounted for some of the inconsistencies in the habitat selection. More research, possibly employing radio telemetry and multivariate analysis of habitats, is required to determine conclusively if Marten select habitats within the coniferous forest. Diet The frequency of occurrence of food items in scats does not necessarily accurately represent the impor- tance of food items and should be used cautiously TABLE 2. Correlation coefficients for the capture rate of three species of rodents in relation to tree dispersion at Chick Lake, Northwest Territories, during 1974." Captures/ 100 Trap Nights Species vs. Tree Dispersion Northern Red-backed Vole -0.7913* Meadow Vole —0.2012 Yellow-cheeked Vole —0.7717* “Data are based on 23,000 trap nights on two 250-trap live-trapping grids. *pn<0.05 DOUGLASS, FISHER AND MAIR: HABITAT AND FOOD OF MARTEN W3 (Day 1966). However, these data do provide some insight into the diet of Marten and form a basis for comparison with other studies. Table 3 summarizes the occurrence of food items in 172 scats collected at Chick Lake and Moon Lake. The diet of Marten at Chick Lake and Moon Lake was similar to that previously recorded in other stu- dies (Lensink et al., 1955, Weckwerth and Hawley TABLE 3. Percentage occurrence of food items in 172 scats collected during 1973 and 1974 and the relative abundance of 4234 small mammals live-trapped from 1973 through 1974 in the Northwest Territories. % Relative % Abundance Occurrence In Live Food Items In Scats Traps Microtus pennsylvanicus 12.8 20.4 M. xanthognathus 17.4 22.1 Total Microtus® 75.6 42.5 Clethrionomys rutilus 13.4 55.4 Snyaptomys borealis 2.3 0.6 Phenacomys intermedius 2.9 0.4 Total non-Microtus” 26.7 56.3 Sorex spp. 1.2 0.7 Tamiasciurus hudsonicus 0.0 0.4 Zapus hudsonius 0.0 0.02 Lepus americanus 0.0 0.0° Mustela erminea 1.2 0.1 Unidentified mammal DS) Bird 4.] Invertebrates 32.0 Plant: Seed 19.2 Moss 28.5 Wood 26.2 Grass NS Lichen 11.0 Leaf 4.7 Spruce Needles 16.9 Other 6.4 Total Vegetation 58.1 “This includes samples identified from hair. These could not be identified to species. *This includes C. rutilus, S. borealis, P. intermedius and all unidentified non-Microtus specimens. “From Douglass 1977. Numbers represent the total individ- uals captured at Chick Lake in 102,300 trap nights from 1973 through 1975. “Because Tamiasciurus hudsonicus appears to be fairly untrappable in small traps this number probably underes- timates their relative abundance. Tamiasciurus hudsonicus tracks in winter were generally more abundant than marten tracks. “Lepus americanus was too large for our traps but numbers of tracks during winter were very low (much lower than marten tracks). No Lepus americanus were observed during trapping sessions. 74 THE CANADIAN FIELD-NATURALIST 1962, Francis and Stephenson 1972, Soutiere 1979). Voles (Microtus spp.) occurred most frequently (75%) in scats while all other cricetid rodents occurred in only 26.7% of scats. Other items that occurred at relatively high frequencies were invertebrates (32%) and plant material (58.1%). We obtained a cursory examination of Marten prey selectivity by comparing the frequency of occurrence of small mammals in scats with their relative abun- dance in the environment. Small mammal abundance was determined during the summer months at Chick Lake (Douglass 1977, Table 2). During this time Microtus spp. populations were low (<10/ha) and Clethrionomys rutilus populations decreased from a high in 1973 of ~41/ha to a low of 4/ha in 1974 (Douglass, unpublished data). All species encoun- tered in scats were live-trapped in the forest. If these data collected during the summer accurately reflect abundance of small mammals in the study area during the winter, and if the prey analysis accurately reflects use of prey, then Marten were being very selective in the prey they consumed. Microtus spp. occurred in 75.6% of the scats but made up only 42.5% of the individuals captured in live traps during three summers (Douglass 1977). All other voles occurred in 26.7% of scats but comprised 56.3% of the individuals captured in live traps, suggesting that Marten at Chick Lake and Moon Lake preferred Microtus spp. over other voles. This contrasts with Lensink et al.’s (1955) results elsewhere that suggested Marten were oppor- tunistic in their feeding habits and consumed rodents in approximately the same ratios as they occurred in the environment. However, Weckwerth and Hawley (1962) found that although Marten feed on most rodents opportunistically, Microtus appeared in scats more frequently than in traps. Acknowledgments Data for this paper were gathered while field studies were being conducted for Canadian Arctic Gas Study, Ltd., by Renewable Resources Consulting Services, Ltd. We thank E. C. Murphy for critically reviewing Vol. 97 the manuscript and A. McNaughton and B. Wooley for assisting in the field work. Literature Cited Cottam, G., and J. T. Curtis. 1956. The use of distance measures in phytosociological sampling. Ecology 37: 451-460. Day, M.G. 1966. Identification of hair and feather remains in the gut and feces of stoats and weasels. Journal of Zoology 148: 201-217. De Vos, A. 1952. The ecology and management of fisher and marten in Ontario. Ontario Department of Lands and Forests. Technical Bulletin. 90 pp. Douglass, R. J. 1977. Population dynamics, home ranges and habitat associations of the yellow-cheeked vole (Microtus xanthognathus) in the Northwest Territories. Canadian Field-Naturalist 91: 237-247. Francis, G.R., and A.B. Stephenson. 1972. Marten ranges and food habits in Algonquin Provincial park, Ontario. Ontario Ministry of Natural Resources. Research Report 9. 53 pp. Lensink, C.J., R.O. Skoog and J.L. Buckley. 1955. Food habits of marten in interior Alaska and their significance. Journal of Wildlife Management 19: 363-368. Marshall, W. H. 1951. Pine marten as a forest product. Journal of Forestry 49(12): 899-905. Mathiak, H. A. 1938. A rapid method of cross- sectioning mammalian hairs. Journal of Wildlife Man- agement 2: 162-164. More, G. 1978. Ecological aspects of food selection in pine marten (Martes americana). M.Sc. thesis. University of Alberta, Edmonton. 94 pp. Seton, E. T. 1929. Lives of game animals. Doubleday Doran and Company, New York. 746 pp. Siegel, S. 1956. Nonparametric statistics for the behav- ioral sciences. McGraw-Hill, New York. 312 pp. Soutiere, E.C. 1979. Effects of timber harvesting on marten in Maine. Journal of Wildlife Management 43: 850-860. Weckwerth, R. P.and V. D. Hawley. 1962. Martenfood habits and population fluctuations in Montana. Journal of Wildlife Management 26: 55-74. Received 15 January 1981 Accepted 12 August 1982 Taxonomy of the Gaspé Shrew, Sorex gaspensis, and the Rock Shrew,? S. dispar THOMAS W. FRENCH!? and GORDON L. KIRKLAND, JR.2 IN. Y. Cooperative Wildlife Research Unit, Cornell University, Ithaca, New York 14853 2Vertebrate Museum, Shippensburg State College, Shippensburg, Pennsylvania 17257 3Nature Conservancey, 294 Washington Street, Boston, Massachusatts 02108 French, Thomas W.,and Gordon L. Kirkland, Jr. 1983. Taxonomy of the Gaspé Shrew, Sorex gaspensis, and the Rock Shrew S. dispar. Canadian Field-Naturalist 97(1): 75-78. Sixty-seven Gaspé Shrews, Sorex gaspensis, from Mount Carleton Provincial Park, one from Moose Mountain near Bath, New Brunswick, and a Rock Shrew, S. dispar, from Quaggy Joe Mountain, Aroostook Co., Maine, were compared to previously reported, but more geographically separated, specimens of S. dispar and S. gaspensis using a stepwise discriminate analysis. Non-overlapping scattergram clusters, indicating significant morphological differences, suggest that these two shrews should continue to be recognized as distinct species. Key Words: Gaspé Shrew, Sorex gaspensis, Rock Shrew, Sorex dispar, New Brunswick, taxonomy. The Gaspé Shrew, Sorex gaspensis, was described from the Gaspé Peninsula of eastern Quebec (Anthony and Goodwin 1924) and has since been captured on Cape Breton, Nova Scotia (Roscoe and Majka 1976) and Mt. Carleton, New Brunswick (Peterson and Symansky 1963). The closely related Rock Shrew, S. dispar, occurs from the higher eleva- tions of North Carolina and Tennessee northward along the Appalachians to Maine (see map in Kirk- land and Van Deusen 1979). A disjunct population of S. dispar has also been reported from southeast New Brunswick (see map in Kirkland 1981). These two closely related species are distinguished primarily on the basis of size, S. gaspensis being significantly smaller in virtually all characters (Kirkland 1981, Kirkland and Van Deusen 1979). The taxonomic status of S. gaspensis and S. dispar was studied by Kirkland and Van Deusen (1979) using a stepwise discriminant analysis of 18 morphological characters. Fourteen S. gaspensis from the Gaspé Peninsula and Cape Breton were available for com- parison to 108 S. dispar representing four geographi- cally distinct groups. These four groups were from North Carolina and Tennessee, Virginia and West Virginia, and mid-Atlantic states (New Jersey, Pen- nsylvania and New York), and New England (Ver- mont, New Hampshire, Massachusetts and Maine). The three Maine specimens examined were from Mt. Katahdin and South Branch Pond, Baxter State Park, Piscatoquis County, and Enchanted Pond, Somerset County. Specimens from the most northerly localities of §. dispar at Beaver Creek, Aroostook County, Maine (Godin 1977) and Riverside-Albert, Albert County, New Brunswick (Kirkland et al 1979) and the most southerly locality for S. gaspensis at Mt. Carleton, New Brunswick were not available for com- parison. The nearest specimen localities of these two Wd species in the Kirkland and Van Deusen (1979) analy- sis were thus approximately 300 km apart. Kirkland and Van Deusen (1979) detected a north- south cline in size for S. dispar progressing from larger in the south to smaller in the north. Sorex gaspensis was, however, much smaller than the northernmost population of S. dispar and they concluded that S. gaspensis continued to warrant species recognition, as it was distinctly differentiated from all populations of S. dispar in the stepwise discriminant analysis and because the magnitude of size decrease was substan- tially greater than the rate of clinal size decrease detected in S. dispar populations. This paper re-examines the taxonomic status of these two shrews in the light of new specimens acquired from localities in New Brunswick and Maine geographically intermediate to those represented in the previous study. Methods The new series of S. gaspensis was trapped by the senior author and two assistants, Beth Ann Sabo and Mark Dalton, on Sagamook Mountain, Mount Carleton Provincial Park and on Moose Mountain near Bath, New Brunswick. Mount Carleton, the highest point in New Brunswick, rises 820 meters above sea level. Trapping took place in various habi- tats throughout the park, although efforts were con- centrated on the north face of Sagamook Mountain (alt. 792 m), located 3.4 km north of Mount Carleton and part of the same ridge system. On Mount Saga- mook, seven traplines were set for varying lengths of time, between 16 June and 20 July, 1980. Snap traps were set for 10,602 trap nights and pitfalls for 985 trap nights. The more southerly location at Moose Moun- tain (alt. 381 m) was trapped for two nights between 21 and 23 July (122 snap trap nights). Also examined 76 THE CANADIAN FIELD-NATURALIST was the S. dispar reported from near Riverside- Albert, New Brunswick (Kirkland et al. 1979) and a new specimen (USNM 554224) from Quaggy Joe Mountain (345 m), Aroostook County, Maine. Body measurements were made in the field but all cranial measurements were made by the junior author in the same way as described by Kirkland and Van Deusen (1979). These measurements were then com- pared to the data sets for S. gaspensis and S. dispar from Kirkland and Van Deusen’s (1979) previous study using the same statistical procedures described in that study. Specimens Examined Specimens used in this study included 224 Sorex dispar and 23 S. gaspensis listed by Kirkland and Van Deusen (1979) and one S. dispar reported by Kirkland et al. (1979). The collection localities, numbers of specimens, and abbreviations for the collections hous- ing the previously unreported specimens used in this CANONICAL VARIABLE - 2 -6.0 Vol. 97 study are listed below. The following abbreviations are used: NBM (New Brunswick Museum, St. John), SSC (The Vertebrate Museum, Shippensburg State College, Pennsylvania), TWF (Thomas W. French —Private Collection), and USNM (National Museum of Natural History, Smithsonian Institution, Washington, D.C.). Sorex gaspensis (68) — NEW BRUNSWICK, Moose Mountain, near Bath, 1 (USNM 553302); Mt. Carleton Provincial Park, north slope of Mt. -Sagamook, 67 (USNM 553242-553301; NBM 1843- 1846; SSC 12018-12019; TWF 561). Sorex dispar (1) — MAINE, AROOSTOOK CoO., Quaggy Joe Mountain (USNM 554224). Study Area The forest type in Mt. Carleton Prov. park is prim- arily boreal with some areas transitional between boreal and mixed deciduous. The mature forests on Oo 6.0 CANONICAL VARIABLE -1 FiGuRE 1. Relationships of 122 Sorex dispar and 58 S. gaspensis in five groups as plotted by discriminant function analysis (BMD P7M). Groups are represented as follows: G = Sorex gaspensis, (Quebec, N.B., N.S.); N = S. dispar from New England (Maine, H.H., Vt., Mass.) and Albert Co., N.B.; M = S. dispar from Mid-Atlantic states (N.Y., N.J., Pa.); U = S. dispar from W. Va. and Va.; B= S. dispar from Tenn. and N.C. The triangle represents the S. gaspensis from Moose Mt., the star represents the S. dispar from Albert Co., N.B., and the black circle in N represents the S. dispar from Quaggy Joe Mt. Within G, specimens from Quebec are represented by Cs, those from Nova Scotia by open circles, and those from New Brunswick by closed circles. Ten New Brunswick specimens do not appear as separate circles because of overlap of coordinate values. The centroids of the five groups are represented by circles enclosing the letters of the groups. 1983 Mount Carleton and Sagamook Mountain are domi- nated by Black Spruce (Picea mariana), Yellow Birch (Betula lutea), Northern White Cedar (Thuja occiden- talis), and Balsam Fir (Abies balsamea). Quaking Aspen (Populus tremuloides) and Paper Birch (Betula papyrifera) are dominant in disturbed areas. The understory consists predominantly of Striped Maple (Acer pensylvanicum), Red Maple (Acer rubrum), and Hobblebush (Viburnum alnifolium). Wild Sarsa- parilla (Aralia nubieaulis), Oak Fern (Gymnocarpium dryopteris), Spinulose Woodfern (Dryopteris spinu- losa), and Bunchberry (Cornus canadensis) comprise the major portion of the herbaceous layer. The talus layer in the Mount Carleton/Sagamook Mountain area is quite substantial and makes trapping along talus slopes difficult. The talus penetrates to a depth of 1.5-2.0 mand consists of large rounded boulders rang- ing from .4 to 1.3 m in diameter. At Moose Mountain, the forest cover is a’more mixed deciduous, second-growth type dominated by Beech (Fagus grandifolia), Sugar Maple (Acer sac- charum), and Paper Birch. The understory consists of Hobblebush, Balsam Fir, and Striped Maple, while the herbaceous layer is dominated by Yellow Clinto- nia (Clintonia borealis), Red Trillium (Trillium erec- tum) and Canada Mayflower (Maianthemum cana- FRENCH AND KIRKLAND: GASPE SHREW AND ROCK SHREW V7 dense). At Moose Mountain, the talus layer consists of small, well consolidated rocks approximately 0.1 to 0.3 m across. Trapping was made easier by the fact that the talus layer averages only about 0.4 m deep. Results and Discussion Sixty-seven Sorex gaspensis were trapped on Sagamook Mountain between 290 and 488 mand one on Moose Mountain at about 335 m. Sixty-two of the Sagamook Mountain specimens were young of the year and five were adults. Three of the adults were males and both females were pregnant with six embryos each. The Moose Mountain specimen was a pregnant female with five embryos. These are the first embryo counts ever recorded from S. gaspensis (see Kirkland 1981). Fifty-seven of the Sagamook Mountain specimens had undamaged skulls and were used in the statistical analysis. The inclusion of these specimens in the step- wise discriminant analysis program yielded scatter- gram clusters virtually identical to those reported in the previous study with all of the Mount Sagamook specimens clustering with S. gaspensis, and this spe- cies being still well differentiated from all populations of S. dispar. The skull of the Moose Mountain specimen was TABLE 1. Measurements for five external and 16 cranial/ mandibular characters of Moose Mt., New Brunswick specimen (USNM 553302) and the ranges of measurements for these characters in 67 S. gaspensis from Mount Carleton Provincial Park, N.B. and 56 New England S. dispar. All linear measurements are in millimeters and weight is in grams. Character Moose Mt. Total length 115 Tail Length 55 Hind foot length 12.5 Body length 60 Weight 5.0 Greatest length — Condylobasal length — Interorbital breadth 2.95 Cranial breadth a= Molariform tooth row 3.65 Cheek tooth row 4.45 Total tooth row 6.75 Incisor width 1.15 Canine width IBS Molar width 3.6 Nasal length 6.1 Palatal length 6.6 Post-palatal length — Mandible length | 97 Mandible length I 10.35 Mandible height Sy *Moose Mt. specimen-exceeds range for character. **Moose Mt. specimen is less than range for character. Range Mt. Sagamook New England S. gaspensis S. dispar 110 =- 116 103 - 136.5 47 Saye 46° - *=6i 12 e M3 2 S 5) 51 - 62 48 3) 7) Do) 2s Dats AQ) oa Oe 162 = N73) N73) MNS.33 15.85 - 16.70 16.45 - 17.70 2.90 - 3.65 Sully 3h 6.50 - 8.15 Teor = 8.3 3.45 - 3.80 eis) = GE BD 4.05 - 4.75 4.1 - 4.9 6.55 - 7.00 6.8 - NA 1.00 - 1.15 leer 1.4 1.40 - 1.65 1.4 - 1.8 32,39) = 3.65 3.65 - aa yhcs 415 - 6.10 5.60 - q25 Os 9 Os 6.6 - 7.5 7.10 - lens) 7.30 - 8.35 D3 a, OY OWS = MNO. 9.80 - 10.55 10.10 - 11.35 2.85 - 5535 3.05 - 4.00 78 THE CANADIAN FIELD-NATURALIST damaged, so it could not be included in the initial analysis. Measurements of the greatest skull length, condylobasal length, cranial breadth, and post-palatal length of this specimen could not be obtained, so a second analysis using the 14 intact characters was made. In the new analysis, the Moose Mountain spec- imen clearly fell within the S. gaspensis cluster (Figure 1) but on the large end of the cluster. This individual is indeed a large S. gaspensis (Table 1) but only exceeds the extreme measurements of the 67 Sagamook Moun- tain specimens in tail length and equals the maximum measurements for incisor width and nasal length. Mea- surements of this specimen are less than the minimum measurements for 56 New England S. dispar (Kirkland et al. 1979) for five characters and equal to the min- imum for one. Interestingly, the weight of this preg- nant specimen exceeds the maximum for New England S. dispar but not for S. gaspensis. The Sorex dispar from Riverside-Albert, New Brunswick and the one from Quaggy Joe Mountain, Maine, clustered with New England S. dispar (Figure 1). In view of the fact that the localities of the new specimens from Moose Mountain and Mount Carleton are intermediate between the nearest locali- ties for S. gaspensis and S. dispar examined by Kirk- land and Van Deusen(1979), it is significant that these new specimens do not plot as intermediate specimens in the discriminant function analysis (Figure 1). Instead, they are interspersed throughout the S. gas- pensis polygon (G) among the previously examined specimens from Nova Scotia and Quebec. These results show that the morphological differences between S. dispar and S. gaspensis remain significant and essentially unchanged, even though the geogra- phic distance between specimens of S. dispar and S. gaspensis has been greatly reduced (to as close as 45 km in the case of the Moose Mountain and Quaggy Joe Mountain specimens). Therefore, we conclude that Sorex dispar and S. gaspensis are separate and distinct species and should continue to be classified as such. Vol. 97 Acknowledgments We express our appreciation to William Hooper, N. R. McKinley, and David Cartwright of the New Brunswick Department of Natural Resources, and to Leland McGaw and Robert Armstrong of the Department of Tourism, for their logistical assistance and for permission to trap within New Brunswick and specifically Mount Carleton Provincial Park. We are grateful to Blanchard MacDougall and the employees of Mount Carleton Provincial Park for their assist- ance. The field work for this project was conducted as a research internship of the Atlantic Center for the Environment, a division of the Quebec-Labrador Foundation, Ipswich, Massachusetts 01938. Literature Cited Anthony, H. E., and G. G. Goodwin. 1924. A new species of shrew from the Gaspé Peninsula. American Museum of Natural History Novitates 109: 1-2. Godin, A. J. 1977. Wild mammals of New England. John Hopkins University Press, Baltimore. 304 pp. Kirkland, G. L., Jr., and H. M. Van Deusen. 1979. The shrews of the Sorex dispar group: Sorex dispar Batchelder and Sorex gaspensis Anthony and Goodwin. American Museum of Natural History Novitates 2675: 1-21. Kirkland, G. L., Jr., D. F. Schmidt, and C. J. Kirkland. 1979. First record of the long-tailed shrew (Sorex dispar) in New Brunswick. Canadian Field-Naturalist 93: 195-198. Kirkland, G. L., Jr. 1981. Sorex dispar and Sorex gaspen- sis. Mammalian Species 155: 1-4. Peterson, R. S.,and A.Symansky. 1963. First record of the Gaspe Shrew from New Brunswick. Journal of Mammal- ogy 44: 278-279. Roscoe, B., and C. Majka. 1976. First records of the Rock Vole (Microtus chrotorrhinus) and the Gaspé Shrew (Sorex gaspensis) from Nova Scotia and a second record of the Thompson’s Pygmy Shrew (Microsorex thompsoni) from Cape Breton Island. Canadian Field-Naturalist 90: 497-498. Received 18 December 1981 Accepted 15 October 1982 Home Range Size, Movements and Habitat Use in Two Moose, Alces alces, Populations in Southeastern Alaska JOSEPH G. DOERR United States Department of Agriculture Forest Service, Petersburg, AK 99833 Doerr, Joseph G. 1983. Home range size, movements, and habitat use in two Moose, Alces alces, populations in southeastern Alaska. Canadian Field-Naturalist 97(1): 79-88. Fourteen radio-marked adult cow Moose (Alces alces andersoni) from two mainland populations in central southeastern Alaska were monitored for up to 29 months during 1978-81. Home range size during the study averaged 4030 ha (range 1280-7300 ha, n= 10), with three Moose exhibiting movements between separate seasonal home ranges. Riparian shrub and high-volume coniferous forests were preferred habitat during conditions of thick snow cover. River terrace Sitka Spruce (Picea sitchensis) stands were selected over mixed Western Hemlock (Tsuga heterophylla)-Sitka Spruce old-growth forests in both populations. Clearcuts under 30 years of age were utilized more heavily than unlogged old-growth forests. Key Words: Moose, Alces alces, southeastern Alaska, clearcuts, old-growth forests. Ecological studies of Moose in southeastern Alaska are lacking. Klein (1965b) and LeResche et al. (1974) discussed the spread of Moose to the mainland drain- ages of central southeastern Alaska in the last quarter of the 19th century, and Burris and McKnight (1973) described introductions of Moose in southeastern Alaska during the 1950’s and 1960’s. Kelsall and Telfer (1974) speculated on factors limiting Moose distribution in the moist temperate Western Hemlock-Sitka Spruce biome of western North America. This study reports on home range sizes, movements, and habitat use by telemetered cow Moose in two mainland populations of central south- eastern Alaska. Methods During March 1978, seven adult cow Moose in the Thomas Bay Area (Figure |) were immobilized froma Hiller 12J-3 helicopter utilizing a Cap-Char gun anda combination of etorphine (M-99, D-M Pharmaceuti- cals Inc., Rockville, MD) and xylazine hydrochloride (Rompun, Chemagro, Kansas City, MO). The Moose were fitted with radio transmitters (Wildlife Mate- rials, Inc., Carbondale, IL) and numbered collars. After handling, diprenorphine (M50-50, D-M Phar- maceuticals Inc., Rockville, MD) was administered as an antagonist. During April 1979, 10 adult cow Moose were captured in a similar manner along the Stikine River drainage, downstream from the Cana- dian border, and were fitted with radio transmitters (TELONICS, Mesa, AZ) and numbered collars. From April 1978 through September 1980, aerial searches were made approximately twice a month at Thomas Bay, utilizing a Cessna 180 with wing- mounted Yagi antennae. Similar searches were made along the Stikine River from May 1979 through 79 October 1981, excluding the months from October 1980 through January 1981. Relocations were recorded on U.S. Geological Survey maps (1:63360) and described with respect to elevation, slope (0-35%, 36-75%, greater 75%), and habitat type. With two exceptions, no aerial flights were made during the Moose hunting season (15 September-15 October on the Stikine River and 1-31 October at Thomas Bay) to avoid negative public reaction from hunters. Approx- imately 10 percent of additional relocation data was obtained from helicopter flights and ground surveys utilizing a radio receiver and hand-held antenna. Hunters and local workers provided a few sightings as well: the latter were used in assessing home range size but not habitat selection. Telemetered Moose were sighted during only 14 percent of 153 relocations at Thomas Bay and 26 percent of 241 relocations on the Stikine River from fixed-wing aircraft. This low observability was due to dense overstory cover. If visual contact was not made and if the suspected location overlapped on two or more habitat types, locations were recorded as “in- exact.” Telemetered Moose were visually located using a helicopter in September 1978 and December 1979, at Thomas Bay, and December 1979, on the Stikine River. Fall calf production of collared Moose was estimated from helicopter flights. Home range was determined from the area of the polygon created by connecting temporally close, out- side locations of each animal. Migratory Moose were defined as those having “two or more seascnally and spatially distinct ranges” (Mould 1976:6). Observed home range size was plotted as a function of the number of telemetry relocations for each Moose. Rate of increase in home range size averaged 169 ha/ relocation (range 29 — 374, n= 12) for the first 80 THE CANADIAN FIELD-NATURALIST BAIRD GLACIER Vol. 97 Bw -_- TO a 7 oo a na ( ‘Area Show in Fig.2 _ / ee eae | ee PATTERSON GLACIER? —— Z (a Kupreanof Island FIGURE |. Thomas Bay and Stikine River study area. ten relocations, 213 ha/relocation (range 36 — 626, n= 10) forthe next ten relocations, 100 ha/ relocation (range 0 — 346, n= 8) for the 21st through 30th reloca- tions, and 43 ha/ relocation (range 0 —- 95, n= 6) for the 31st through 40th-plus relocations. I feel that the increase in total home range size with increased number of relocations is more a result of gradual changes in range use over a period of years than inadequate telemetry sampling during specific time periods. Total home range data are only analyzed for Moose with at least 25 relocations, recognizing that this reflects a somewhat conservative statistic for Moose movements over a specific number of months. Percent area of different vegetation types within home ranges of Moose was calculated using a dot- grid. Seasonal habitat use at Thomas Bay was ana- lyzed by separating the data into four periods: spring (April-May), summer (June-August), late fall-early winter (November-December), and winter (January- March). Data from September and October were insufficient for this analysis. Age-specific use of second-growth stands was evaluated at Thomas Bay by dividing the second-growth stands into age classes spanning seven-year intervals. Chi-square analysis was used to determine if habitat preferences, seasonal habitat use, and age related use of second-growth Chen oh ( LE CONTE \ GLACIER 2c oor as tal a - \ ° Area Shown in Fig. 3 ye Scale ce ees me | (0) 8 12 16 20 Km. stands were evident in the data (Steel and Torrie 1960). The data were partitioned to avoid expected values less than five (Cochran 1954). Plant classification followed Hultén (1968). Study Area The Thomas Bay study area is located between the LeConte and Baird Glaciers and includes the drain- ages of the Muddy and Patterson Rivers (Figure 1). The Stikine River study area includes the lower 50 km of the river as well as adjacent smaller drainages. Glaciers, icefields, vertical rock cliffs, talus slopes, alpine tundra, krummbholz, estuarine meadows, mus- kegs, and floodplains are common in both areas. Coniferous forests are widely distributed from the beach to over 1000 m elevation. Three types of coni- ferous forests are recognized: old-growth, river ter- race, and muskeg-scrub. Old growth is composed of Western Hemlock (7Tsuga heterophylla) and Sitka Spruce (Picea sitchensis) with blueberry (Vaccinium spp.) as a major shrub understory species. These stands are characterized by large, live trees, large snags, high accumulation of woody litter, and an uneven forest age structure (Franklin et al. 1981), as well as substantial gross understory production and diversity (Alaback 1981). 1983 River terrace forests on the Stikine River were typi- fied by Spruce trees exceeding 76 cm dbh at variable spacing with a dense and varied shrub understory. Willows (Salix sitchensis, S. alaxensis, S. interior, S. Barclayi, S. commutata), Black Cottonwood (Popu- lus balsamifera trichocarpa), alder (Alnus crispa sinuata, A. oregona), Salmonberry (Rubus spectabi- lis), Stink Currant (Ribes bracteosum), Red-osier Dogwood (Cornus stolonifera), Devil’s Club (Echi- nopanax horridum), Highbush Cranberry (Viburnum edule), and Red-berried Elder (Sambucus racemosa) are common shrubs of the Stikine drainage with Wil- low, Black Cottonwood, and alder pioneering ripar- ian shrub communities. Black Cottonwood forests of variable ages exist along the Stikine drainage with Spruce invading the older stands. River terrace forests at Thomas Bay are typified by Sitka Spruce trees from 30 to 60 cm dbh and a varia- ble Vaccinium shrub understory. Alder, Devil’s Club, and Highbush Cranberry are often present in the younger river terrace forests, with Hemlock invading the more developed forests. Mixed stands of Sitka Willow (S. sitchensis), Black Cottonwood, and alder occur in appreciable quantities only in glacial river- wash soils and along logging roadsides. Alder thickets are common in slide areas and active floodplains. Muskeg-scrub forests are composed of stunted Sitka Spruce, Western Hemlock, Mountain Hemlock (T. Mertensiana), Lodgepole Pine (Pinus contorta), and Yellow Cedar (Chamaecyparis nootkatensis) interspersed with muskegs and ponds containing a variety of plants, including sedges (Cyperaceae), eri- cacious shrubs, and Bunchberry (Cornus canadensis). Riparian sedge (Carex spp.) muskegs within the floodplain of the Stikine River are bordered and often interspersed with stands of willows, alder, and Red- osier Dogwood. A soils-vegetation map, (F. R. Stephens, C. R. Gass, and R. F. Billings. 1968. Soils and site index in southeastern Alaska. Unpublished administrative study, USDA-Forest Service, Juneau, Alaska) was used to determine habitat availability at Thomas Bay. Site productivity of forest soils is measured by the total height that the average dominant and codomi- nant tree will attain at 100 years of age. Site indices of Sitka Spruce range from 80 to 150 on sites classified as commercial forest soils. The Alaskan portion of the Stikine River area has wilderness status and the habitat has not been appre- ciably altered by human activity. Conversely, Thomas Bay is in an area designated for intensive timber harv- est. From 1951 to 1976, over 2500 ha were clearcut in the drainages of the Patterson and Muddy Rivers. Logging activity ceased in 1976, except for minor salvage sales: Approximately 650 ha of 10-to-25-year- DOERR: HOME RANGE, MOVEMENTS AND HABITAT USE IN TWO MOOSE 81 old clearcuts were precommercially thinned from 1975 through 1980. Clearcuts under 30 years of age have higher forage production and diversity than surrounding old- growth forests (USDA-Forest Service unpublished data, Petersburg Ranger District). Bunchberry, cer- tain ferns, grasses, 5-leaf Bramble (Rubus pedatus), Salmonberry, alder, Trailing Black Currant (Ribes lacustre), and Red-berried Elder increase in early second-growth stages. Both study areas are affected by a maritime climate with the average annual precipitation ranging from 200 to 400 cm depending on elevation and distance from saltwater (U.S. Department of Agriculture 1979). Snowfall normally occurs from November through April and is likewise influenced by elevation and proximity to saltwater. Periodic rains commonly occur throughout the winter at Thomas Bay and the delta region of the Stikine River, drastically altering snowpack. Snow often persists into May in the eastern portion of the Stikine River study area with little melting occuring during January through March. This snow is typically hard packed by wind action, often easily supporting Moose on its crust. By con- trast, snow at Thomas Bay could seldom support Moose during this study. Fora more detailed descrip- tion of vegetation and climate see Taylor (1932), Klein (1965a), and Alaback (1981). I estimate densities of approximately 2.3 Moose/ km? in the best habitat of both study areas based on the number of marked to unmarked animals seen, pellet-group densities in trend transects, the number of animals sighted during optimum winter viewing conditions, and the level of hunter harvest. (USDA Forest Service unpublished data, Petersbury Ranger District). Results Home Range and Movements Mean total home range sizes (Table 1) were com- parable between both Thomas Bay and the Stikine River (4480 vs. 3840 ha, respectively, P > 0.50). Dis- tances of maximum separation between all individual relocations (Mould 1979) averaged 13.2 km, for Thomas Bay Moose and 15.1 km for Stikine River Moose, an insignificant difference (P > 0.50). Three of the Stikine Moose were migratory, with winter home ranges that were significantly larger than their summer home ranges (930 vs. 210 ha, P < 0.05). Observed movement patterns varied among indi- viduals in each study area (Figure 2). At Thomas Bay, Cow | was consistently located within 0.8 km of the Patterson River, except when she moved inland 3.2 kmtoa heavily logged area adjacent to the Muddy River during a period of severe snowfall in March 82 THE CANADIAN FIELD-NATURALIST Vol. 97 TABLE |. Home range sizes and straight-line distance of maximum separation between locations of telemetered cow Moose, Thomas Bay and Stikine River. Dates of Home Range Distance Max. Moose No. Location No. Locations Size (ha) Separation (km) Nomigratory: wl 3/78- 3/79 17 900 4.8 T2 3/78- 9/80 51 3530 11.4 T3! 3/78- 9/80 47 5140 18.4 T# 3/78- 9/80 53 4770 9.9 TS! 3/78- 1/80 18 1850 12.6 SS! 4/79-10/81 43 7240 18.4 S78! 4/79- 9/80 26 3530 2 S84! 4/79-10/81 43 3050 11.3 S85! 4/79- 6/80 27 7300 20.0 Migratory S79 5/79-12/79 9 190 — 1/80- 4/8/80 7 1200 — 4/28/80- 9/80 10 160 — 4/79- 4/81 29 2180 16.4 S81! 10/79- 5/21/80 14 1430 _ 5/28/80- 9/80 11 240 — 4/79- 4/81 34 2320 16.8 $83 5/79-12/79 9 250 — 1/80- 5/21/80 8 310 — 5/28/80- 9/80 10 230 — 2/81- 7/20/81 6 780 = 7/31/81-10/81 5 200 a 4/79-10/81 39 1280 11.8 x Thomas Bay — 3 Moose} 4480 (+ 843)4 13.2 (£ 4.5) X Stikine River — 7 Moose 3840 (+ 2450) 15.1 (43.7) 'Cows with fall calves *Summer-winter ranges overlapped in 1978-79, but not in 1979-80 3Total home ranges only analyzed for Moose with more than 25 relocations. 4(= STD) 1979. Radio contact was subsequently lost with this individual. Cow 2 spent most of her time along the lower 6.4 km of the Patterson River drainage. Nota- ble exceptions were from 26 April-27 July, 1978, 17 April-26 May 1979, and 12 June-10 July 1980, when she was found ca. 7.2 km farther inland along the Muddy River. Cow 5 was located by radio signal from April through September 1978, but only intermit- tently thereafter (visual sightings) due to radio failure. On 26 April and 28 June 1978, she was located on Ruth Island. All other locations were within 0.8 km of the Patterson River from its mouth to within 1.8 km of the glacier that is its source. Relocations of this Moose (n= 8) from April through September were within 2.8 km of saltwater, while nine relocations from November through March averaged 3.4 km dis- tance from saltwater (range 0.8-6.4 km). This suggests that Cow 5 tended to move farther inland during winter months. On the other hand, Cow 4 was located 5.4 to 8.8 km inland along the Muddy River during the periods 27 June-2 November 1978, 26 May-2 November 1979, and 21 May-13 September 1980, excluding two summer locations in the Patterson drainage in 1978, and within 3.8 km of the mouth of the Muddy River during the remainder of the study period. Cow 3 displayed the most wide-ranging movements of any telemetered Thomas Bay Moose, moving widely throughout the study area. From April through September in 1978 to 1980, she was generally located within 1.6 km of the Thomas Bay or Bock Bight shoreline, and swam at least twice to Ruth Island during the period late April to early June 1978 (Figure 2). From November 1978 to March 1979, Cow 3 inhabited a home range of 3030 ha that included portions of the lower Muddy and Patterson River drainages, as well as beach fringe forests and muskeg- scrub forests to 276 m elevation ca. 4.8 km south of the lower Muddy River. From November 1979 through March 1980, during a relatively moderate 1983 Patterson Glacier Fredrick Sound Clearcuts eee —_—_ ae | Moose Home Ranges FIGURE 2. Location of clearcuts and total home ranges of telemetered Thomas Bay Moose. winter, Cow 3 had an observed home range of only 250 ha confined to within 1.6 km of the southeastern shoreline of Thomas Bay. Four of the telemetered Stikine River Moose has seasonally overlapping home ranges located primarily within the Stikine River Valley and lower Andrew Creek drainage (Figure 3). Of the three migratory Moose, Cow 81 occupied a summer home range in 1979 and 1980 along a tributary of Thunder Creek, 6.2 km north of its winter range on the Stikine River. Cows 79 and 83 wintered in the Stikine River Valley and lower portions of Andrew Creek and had small summer home ranges along the upper north fork of Andrew Creek. Both Moose returned to their winter home ranges in December or early January, with observed minimum distances between seasonal ranges equalling 5.6 and 4.6 km for Cows 79 and 83, respec- tively. Aerial reconnaissance surveys flown during excellent tracking snow conditions on 25 January 1980,and 4 March 1981, revealed no Moose tracks or signin these summer home ranges and suggest that the drainage of the north fork of Andrew Creek beyond four km from its mouth is largely devoid of Moose during months of high snow accumulation. DOERR: HOME RANGE, MOVEMENTS AND HABITAT USE IN TWO MOOSE 83 Timing of movement to summer ranges varied among the individuals and the years. Cow 83 moved to her summer range between 21 and 28 May in 1980,and between 20 and 31 July in 1981. Cow 81 moved to her summer range between 4 and 12 June in 1979, and 21 and 28 May in 1980. Several of the telemetered Moose made movements that reduced the possibility of contact with hunters during the bull Moose hunting season. Cows 5 and 85 were located in the Canadian portion of the Stikine River 4.5 and 2.6 km above the border on the last day of the season in 1979. Cows 5 and 84 left the floodplain of the river, where hunters concentrate, and moved into mature forests north of the river during the early portion of the 1981 season. During the 1979 and 1980 hunting seasons, Cows 79 and 83 were on their summer ranges in the relatively inaccessible upper north fork of Andrew Creek. Habitat Use Thomas Bay Home ranges of telemetered Moose at Thomas Bay were largely restricted to relatively level, low elevation habitats containing a mixture of clearcuts, forests, and riparian habitats. Ninety-four percent of all relo- cations (N = 179) were at elevations under 77 m and 97 percent were on slopes from 0 to 35 percent. Only 11 relocations (6%) were more than 0.4 km from a clearcut or 0.8 km from the main course of the Muddy or Patterson River. Three of these relocations were in beach fringe old-growth during periods of high snow accumulation, while the remainder were in muskeg- scrub forests or low volume (site index less than 120) old-growth forests. Extensive tracts of old-growth above the river bottoms were largely unutilized by the collared Moose. Total home ranges of telemetered Moose over- lapped each other, and all home ranges were com- bined when testing for habitat preference and use. Relocations in glacial riverwash soils were signifi- cantly higher than expected, while relocations in muskeg-scrub forests and muskegs were significantly less than expected based on the proportion of these habitat categories in the combined home ranges of the telemetered Moose (P< 0.005). Use of Hemlock- Spruce forest soils was in proportion to their occur- rence in the area (Table 2). Fifteen to 20 percent of the Moose relocations were in glacial riverwash soils compared to ca. 70 percent in Hemlock-Spruce forest soils. Clearcuts from ages two to 30 and unlogged forests comprised 23 and 41 percent, respectively, of the combined home ranges of the Moose and accounted for 52 and 46 percent, respectively, of all exact reloca- tions in commercial forest soil types. Based on these 84 THE CANADIAN FIELD-NATURALIST Vol. 97 Travel Way FIGURE 3. Location of home ranges of telemetered Stikine River Moose. data, clearcuts were used significantly more than unlogged forests (P < 0.005). Differences in sightabil- ity between clearcuts and forests did not greatly affect the proportion of exact locations in clearcuts com- pared to forests as only seven telemetered Moose were sighted in clearcuts during fixed-wing flights. The proportion of relocations in river terrace soils was higher than expected, while the proportion of reloca- tions in low volume forest soils was lower than expected (Table 2, P< 0.005). Eighty-four percent of all exact relocations in river terrace soils were in unlogged stands compared to 25 percent of all relocations in old-growth soils with site indices from 120 to 150. The data suggest that unlogged river terrace forests were preferred over river terrace clearcuts (P < 0.005) and that clearcuts under 30 years of age on other forest soils were pre- ferred over high-volume old-growth forests (P < 0.005). Major differences in the proportion of relocations in glacial riverwash habitats, old-growth, clearcuts, and muskeg-scrub timber were observed with respect to season. Sample sizes during any period were small, however, and precluded a conclusive determination of seasonal use at Thomas Bay. We did detect a signifi- cantly (P < 0.025) higher percentage of relocations in muskegs, muskeg-scrub forests, and inexact locations associated with muskeg soils during January to March (34 %) and significantly lower use of the above habitats during November and December (3 %) com- pared to the entire year (16% overall). Likewise, the number of relocations in clearcuts compared to all exact relocations in forest soils was significantly (P < 0.025) lower in the months June to August (6 of 21) and higher in November and December (10 of 12) than the entire year (42 of 81 relocations). Heavy snowfall conditions (150+ cm in open areas) were present during only four weeks of the Thomas Bay study in February and March of 1979. Of eight relocations during that period, two were in open scrub 1983 DOERR: HOME RANGE, MOVEMENTS AND HABITAT USE IN TWO MOOSE 85 TABLE 2. Summary of locations of five telemetered Thomas Bay Moose, April 1978-September 1980 by soil-vegetation type. Soil-Vegetation Type No. Locations Riverwash (RW) 25 Hemlock-Spruce (F type) 108 Muskeg-Scrub (MF5) 8 Muskeg (M type) 2D RW-F Mixture? 12 M-F or MF-F Mixture? 18 Estuarine 0 Lakes, Ponds 0 173 F Types (62% of total) Uncut Forest 37 Clearcut (2 to 30 years) 42 Regrowth (31 to 70 years) D Uncertain? DY RW-F, M-F, and MF-F Mixture (17% total)? Uncut Forest* 21 Clearcut (3 to 29 years)° 3 Uncertain F Type (exact locations) River terrace soils High-volume old-growth soils” Low-volume commercial forest soils Percent Locations Percent Area* 14 5 62 64 5 16 13 7 oe 10 as = = 2 21 41 24 23 16 bu 12 = 2 = 3 Es 23 8 72 46 5 10 100 4Based on combined total home ranges of all Moose. Slopes greater than 75% accounted for 9% of the home ranges were excluded from calculations of area. > Approximate location overlapped on more than one type. °F type which could have contained the Moose. ‘Site index greater than 119. “Site index 80 to 119. riverwash habitat, two in young river terrace forests, and four in old-growth forests within 0.5 km of saltwater. Clearcuts of ages 2-8, 9-15, 16-22, and 23-29 com- prised 12, 28, 51, and 9 percent, respectively, of all clearcuts within the combined home ranges of all the telemetered Moose. Moose relocations (N = 40) in these respective age categories were 3, 23, 68, and 8 percent of relocations in clearcuts. The observed pro- portion in each age category 1s not significantly differ- ent than would be expected based on the proportion of these categories. Stikine River Habitat data were gathered on nine Stikine River Moose, with seven Moose contributing the bulk of the relocations. All Moose were closely associated with riparian habitat. Only two of 246 relocations were more than 0.4 km from the floodplain of the Stikine River or the stream courses of Andrew, Thunder, or North Arm Creeks. Only six locations (two percent) were in Tsuga- Picea old-growth forests with a Vacci- nium understory, one location was in a pond sur- rounded by Tsuga- Picea- Vaccinium forests, and eight inexact locations overlapped on such forests. Ninety- three percent of the relocations were in riparian shrub habitats, Spruce river terrace forests, Cottonwood forests, riparian sedge muskegs, or a mixture of these habitats. Two locations were on forested hillsides bordering the river with a dominant alder understory. The exact riparian plant community could not be identified in 55 percent of the relocations during April through November due to the complex mosaic of vegetative communities within the riparian zone, as well as the poor summer sighting conditions. During December through March, 43 percent of 57 reloca- tions were in Spruce river terrace forests, 24 percent in riparian shrub habitats, 14 percent in Black Cotton- wood forests, 3 percent in mixed Cottonwood-Spruce forests, 14 percent in inexact riparian habitat, and one location was in a sedge (Carex spp.) riparian muskeg. During December through March, 86 percent of the relocations were in the Stikine River Valley and 14 percent in adjoining drainages. During the rest of the 86 THE CANADIAN FIELD-NATURALIST year, 58 percent of the relocations were in the Stikine River Valley. These percentages are significantly dif- ferent (P < 0.005) and indicate that Moose tend to concentrate in the Stikine River floodplain during winter months. Cow 81 had asummer range between 150 and 300 m elevation, and Cows 79 and 83 had summer ranges located between 78 and 140 melevation. Ninety-seven percent of the other relocations were under 78 m ele- vation. Only two percent of all relocations were on slopes exceeding 35 percent. Discussion Moose are closely associated with riparian habitat in many portions of their North American range (Sumanik and Demarchi 1977; Mould 1979; Houston 1968: LeResche et al. 1974; Peek 1974; Berg and Phil- lips 1974). In forests with limited riparian habitat, Moose tend to select early successional stands created by fire or logging, especially during the fall and early winter and in the spring (Krefting 1974; Peek et al. 1976; Eastman 1974; Bergerud and Manuel 1968; Hamilton et al. 1980). Moose populations may increase several fold when early seral stands are created in mature or climax forests (Bergerud and Manuel 1968: Peek et al 1976; Cowan et al. 1950; Bishop and Rausch 1974; Aldous and Krefting 1946; Geist 1971). By contrast, Moose often prefer conifer forests during periods of heavy snowfall (Peek et al. 1976; Eastman 1974; Knowlton 1960; Stevens 1970; Van Ballenberghe and Peek 1971; Brassard et al. 1974; LeResche et al. 1974; Peterson 1977). In the Stikine River population, 94 percent of all relocations were in riparian habitat, including Spruce river terrace stands. At Thomas Bay, glacial riverwash soils, supporting quantities of Willows and Black Cot- tonwood, were preferred by Moose and accounted for 15 to 20 percent of all relocations. The preference for riparian communities is supported by observations of Klein (1965b) and LeResche et al. (1974) who noted a close association of Moose distributions in southeast- ern Alaska to mainland river drainages. In both populations, riparian communities with tall deciduous shrubs and Sitka Spruce river terrace forests were important habitats of Moose during con- ditions of thick snow cover. At Thomas Bay where these habitats are limited, old-growth Spruce- Hemlock forests were also important winter range. River terrace forests were selected over old-growth forests in both populations. At Thomas Bay, Moose selected soils with high Spruce productivity over poorer sites and favored clearcuts over unlogged old-growth stands. The pres- ence of large clearcut areas in early stages of succes- sions may be a factor accounting for the relatively Vol. 97 high Moose abundance observed during the study. Cowan et al. (1950) noted that the removal of old- growth Engelmann Spruce (Picea Engelmanni) forest by fires in central British Columbia was followed by an increase of Moose and that the highest densities of Moose occurred in young second-growth areas. Home range sizes and movements of Moose are highly variable among individuals and among various populations in North America (Ritchie 1978; Van Ballenberghe 1977; Phillips et al. 1973; Addison et al. 1980; Hauge and Keith 1981; LeResche 1974). Observed distances of movements tended to be shor- ter and home ranges smaller for southeastern Alaska Moose than those reported for Moose in Idaho (Rit- chie 1978), arctic and boreal Alaska (Mould 1979; Van Ballenberghe 1977; Taylor and Ballard 1979), and Alberta (Lynch 1976; Hauge and Keith 1981). Habitat management concerns vary between the two areas. The Stikine River population is highly dependent upon the vegetation produced within the Stikine floodplains, especially during winter months. Long-range effects of regulated water levels on the downstream riparian communities need to be deter- mined to evaluate the impact of proposed dam con- struction on the Moose population. The Thomas Bay Moose population extensively utilizes timber stands with site indices from 120 to 150 and clearcuts with high forage production. Alaback (1981) has shown that in unthinned clearcuts second- growth conifers dominate the site and deciduous browse and forb production is virtually eliminated 25 to 35 years after logging. Loss of early clearcuts, a preferred habitat, may have major impacts on the population. Long-range management of Moose at Thomas Bay should emphasize sustaining a mixture of high forage second-growth and forested winter hab- itat. The ability of precommercial thinning to prolong understory production and Moose utilization of second-growth stands needs to be studied. Acknowledgments This study was initiated and continually supported by E. L. Young, Jr., formally of the USDA-Forest Service, who now works for the Alaska Department of Fish and Game. Dr. A. W. Franzmann, and H. R. Merriam of the Alaska Department of Fish and Game and J. H-: Hughes and EB. L. Young, Jr. US DAq Forest Service, captured and marked the Moose. K. Becker, M. Madrid, and J. Brighenti, USDA-Forest Service, and V. Bier, Alaska Department of Fish and Game, assisted in relocating the Moose. Pilots of Alaska Island Air and Temsco Helicopters, Inc., assisted in the aerial flights. Figures were drawn by J. Hanks, USDA-Forest Service. Dr. V. Van Ballen- belghe, E. L. Young, Jr., and K. Becker reviewed an 1983 earlier draft of this manuscript. All help was greatly appreciated. Literature Cited Addison, R. B., J. C. Williamson, B. P. Saunders, and D. Fraser. 1980. Radio-tracking Moose in the boreal forest of northwestern Ontario. Canadian Field-Naturalist 94: 269-276. Alaback, P. B. 1981. Biomass and productivity of under- story vegetation in seral Sitka Spruce-Western Hemlock forests of southeast Alaska. Ph.D thesis, Oregon State University, Corvallis, 87 pp. Aldous, S. E., and L. W. Krefting. 1946. The present state of Moose on Isle Royale. Transactions of the North Amer- ican Wildlife Conference | 1: 296-308. Berg, W.E., and R.L. Phillips. 1974. Habitat use by Moose in northwestern Minnesota with reference to other heavily willowed areas. Naturaliste Canadien 101: 101-116. Bergurud, A. T., and F. Manuel. 1968. Moose damage to balsam fir-white birch forests in central Newfoundland. Journal of Wildlife Management 32: 729-746. Bishop, R. H., and R. A. Rausch. 1974. Moose popula- tion fluctuations in Alaska, 1950-1972. Naturaliste Cana- dien 101: 559-593. Brassard, J. M., E. Andy, M. Crete, and P. Grenier. 1974. Distribution and winter habitat of Moose in Quebec. Naturaliste Canadien 101: 67-80. Burris, O. E., and D. E. McKnight. 1973. Game trans- plants in Alaska. Alaska Department of Fish and Game, Game Technical Bulletin No. 4. 57 pp. Cochran, W. G. 1954. Some methods for strengthening the common X’° tests. Biometrics 10: 417-451. Cowan, I. McT., W.S. Hoar, and J. Hatter. 1950. The effect of forest succession upon the quantity and upon the nutritive values of woody plants used as food by Moose. Canadian Journal of Research, D, 28: 249-271. Eastman, D. S. 1974. Habitat use by Moose of burns, cut- overs, and forests in north-central British Columbia. Pro- ceedings of the North American Moose Conference Workshop 10: 238-256. Franklin, J. F., K. Cromack, Jr., W. Denison, A. McKee, C. Maser, J. Sedell, F. Swanson, and G. Juday. 1981. Eco- logical characteristics of old-growth Douglas-fir forests. USDA-Forest Service, Pacific Northwest Forest and Range Experiment Station, General Technical Report PNW-118. 48 pp. Geist, V. 1971. Mountain sheep-study in behavior and evo- lution. The University of Chicago Press, Chicago, Illinois. 383 pp. Hamilton, G. D., P. D. Drysdale, and D. L. Euler. 1980. Moose winter browsing patterns on clearcuttings in north- ern Ontario. Canadian Journal of Zoology 58: 1412-1416. Hauge, T. M., and L. B. Keith. 1981. Dynamics of Moose populations in northeastern Alberta. Journal of Wildlife Management 45: 573-597. Houston, D. B. 1968. The Shiras Moose in Jackson Hole, Wyoming. Technical Bulletin |. Grand Teton Natural His- tory Association. 110 pp. DOERR: HOME RANGE, MOVEMENTS AND HABITAT USE IN TWO MOOSE 87 Hultén, E. 1968. Flora of Alaska and neighboring Territo- ries. A manual of the vascular plants. Standford Univer- sity Press, Stanford, California. 1008 pp. Kelsall, J. P., and E.S. Telfer. 1974. Biogeography of Moose with particular reference to western North Amer- ica. Naturaliste Canadien 101: 117-130. Klein, D. R. 1965a. Ecology of deer range in Alaska. Eco- logical Monographs 35: 259-284. Klein, D. R. 1965b. Postglacial distribution patterns of mammals in the southern coastal regions of Alaska. Arctic 18: 7-20. Knowlton, F. F. 1960. Food habits, movements, and popu- lations of Moose in the Gravelly Mountains, Montana. Journal of Wildlife Management 24: 162-170. Krefting, L. W. 1974. Moose distribution and habitat selec- tion in north central North America. Naturaliste Canadien 101: 81-100. LeResche, R. E. 1974. Moose migrations in North Amer- ica. Naturaliste Canadien 101: 393-415. LeResche, R. E., R. H. Bishop, and J. W. Coady. 1974. Distribution and habitats of Moose in Alaska. Naturaliste Canadien 101: 143-173. Lynch, G. M. 1976. Some long-range movements of radio tagged Moose in Alberta. Proceedings of the North Amer- ican Moose Conference Workshop 12: 220-235. Mould, E. 1979. Seasonal movement related to habitat of Moose along the Colville River, Alaska. Murrelet 60: 6-11. Peek, J. M. 1974. Onthe nature of winter habitats of Shiras Moose. Naturaliste Canadien 101: 131-141. Peek, J. M., D. L. Urich, and R. J. Mackie. 1976. Moose habitat selection and relationships to forest management in northeastern Minnesota. Wildlife Monograghs Number 48. 65 pp. Peterson, R. O. 1977. Wolf ecology and prey relationships on Isle Royale. National Park Service Science Monograph Series, Number | 1. 210 pp. Phillips, R. L., W. E. Berg, and D. B. Siniff. 1973. Moose movement patterns and range use in northwestern Minne- sota. Journal of Wildlife Management 37: 266-278. Ritchie, B. W. 1978. Ecology of Moose in Fremont County, Idaho. Idaho Department of Fish and Game, Wildlife Bulletin Number 7. 33 pp. Steel, R.G.D., and J.H. Torrie. 1960. Principles and procedures of statistics. McGraw-Hill Book Company Inc., New York. 481 pp. Stevens, D. R. 1970. Winter ecology of Moose in the Gal- latin Mountains, Montana. Journal of Wildlife Manage- ment 34: 37-46. Sumanik, K. M., and D. Demarchi. 1977. Dispersion and relative abundance of Moose in northern British Colum- bia. Proceedings of the North American Moose Confer- ence Workshop 13: 252-257. Taylor, K. P., and W. B. Ballard. 1979. Moose movements and habitat use along the Susitna River near Devil’s Canyon. Proceedings of the North American Moose Con- ference Workshop 15: 169-186. Taylor, R. F. 1932. The successional trend and its relation to second-growth forests in southeastern Alaska. Ecology 13: 381-391. 88 THE CANADIAN FIELD-NATURALIST U.S. Department of Agriculture. 1979. Water Resources Atlas. Forest Service, Juneau, Alaska. Pages not numbered. Van Ballenberghe, V., and J. M. Peek. 1971. Radioteleme- try studies of Moose in northeastern Minnesota. Journal of Wildlife Management 35: 63-71. Vol. 97 Van Vallenberghe, V. 1977. Migratory behavior of Moose in southcentral Alaska. 13th International Congress of Game Biologists, Atlanta, Georgia. 102-109. Received 12 February 1982 Accepted 15 November 1982 Notes Northern Records of Risso’s Dolphin, Grampus griseus, in the Northeast Pacific HOWARD W. BRAHAM National Marine Mammal Laboratory, Northwest and Alaska Fisheries Center, National Marine Fisheries Service, Seattle, Washington 98115 Braham, Howard W. 1983. Northern records of Risso’s Dolphin, Grampus griseus, in the Northeast Pacific. Canadian Field-Naturalist 97(1) 89-90. On 12 March 1976 two Risso’s Dolphins were observed in the Gulf of Alaska 450 km north of their reported summer range. This is the most northern verified record for this temperate species and represents the earliest annual sighting in the North Pacific. Key Words: Grampus griseus, Risso’s Dolphin, Gulf of Alaska, Northeast Pacific, Vancouver Island, British Columbia. On 9 and 12 March 1976, while transiting from Seattle, Washington to Kodiak, Alaska aboard the NOAA ship SURVEYOR, I observed seven Risso’s Dolphins (Grampus griseus) in three separate groups. In addition, R. Mercer, Northwest and Alaska Fisher- ies Center, Seattle, reported to me a sighting he esti- mated to exceed 2,000 individuals made while he was aboard the F/V TRES CHER on 3 August 1981 off Washington state (Table 1). The sighting on 12 March 1976 constitutes the northernmost record of Risso’s Dolphin in the Pacific Ocean, and Mercer’s sighting of the largest concentration reported from the Northeast Pacific. Leatherwood et al. (1980) reviewed all information available on the species’ occurrence in the eastern North Pacific through 1978. They included as the northernmost record accounts by Guiquet and Pike (1965) of several sightings at 50°N, 145° W between July and October, 1958-1960. Reimchen (1980) extended the range of the species (Table 1). In the western North Pacific, Risso’s Dolphins have been reliably reported near the Kuril Islands, as far north as 51°N (Sleptsov 1961). G. griseus was also reported for the Aleutian Islands by Collins, Clark and Walker (1945) and Commander Islands at Lat. 54° N by Slept- stov (1952). Leatherwood et al. (1980) rejected as unsubstantiated reports by Collins et al. (1945) of Risso’s Dolphins around the Aleutian islands. Tomilin (1957) similarly regarded the reports in Slept- sov (1952) of Risso’s Dolphins in Commander Islands as unsupported. In addition, of the 13 species of dol- phins and porpoises reported by Collins et al. (1945) and Sleptsov (1952) as occurring in the southern Ber- ing Sea, eight are not documented in the literature (D. TABLE |. Recent sightings of Risso’s Dolphins in the Northeast Pacific Ocean. Date Time Location 12 March 1976 1040 55°49’N, 145°56’W 27 March 1978 1500 54° 11'N, 133°01’W 9 March 1976 1350 49°52'N, 128°37'W 9 March 1976 1338 49°50’N, 128° 30’W 3 August 1981 1100 47° 10’N, 125° 20’W 89 Number of Water individuals depth(m) Source 2 4000 Author’s observation, this paper 14 10-12 Reimchen (1980) 2 183 Author’s observation, this paper 3 183 Author’s observation, this paper 2000+ 290 R. Mercer, personal communication, February 1982 90 THE CANADIAN FIELD-NATURALIST Rice, National Marine Mammal Laboratory, per- sonal communication, May 1982), and therefore | believe them to be unsubstantiated. Several coastal sightings and strandings have occurred along the west and northwest coast of North America, but most were south of Canada and occurred in summer (Guiguet and Pike 1965; Stroud 1968: Hatler 1971). The spring records included in Table | are 3-4 months earlier in the year than pre- viously noted for the species in the northern portion of its range. These few sightings suggest that the north- ern limit probably does not exceed the southern por- tion of the Gulf of Alaska; the species constitutes a rare visitor to Alaskan waters as evidenced by the lack of strandings north of southern British Columbia. The August 1981 record (Table 1) is of special inter- est because of the size of the sighting. The 2000+ estimated Risso’s Dolphins were observed over an area 2-4 km long by 4 km wide. In association with these Risso’s Dolphins were approximately 1000 Right Whale Dolphins (Lisodelphus borealis), 500 White-sided Dolphins (Lagenorhynchus obliquidens) and 30 Dall’s Porpoise (Phocoenoides dalli). Acknowledgments This research was supported by the Bureau of Land Management under an interagency agreement with the National Oceanic and Atmospheric Administra- tion, Alaska Outer Continental Shelf Environmental Assessment Program, Juneau, Alaska project office. | also thank Dale Rice, David Rugh, and Robert Miller, National Marine Mammal Laboratory for their review, and Roger Mercer, Northwest and Alaska Fisheries Center, Seattle, Washington, for use of his unpublished sightings. Literature Cited Collins, H., A. Clark, and E. Walker. 1945. The Aleutian Vol. 97 Islands: Their People and Natural History. Smithsonian Institution, War Background Study 21: 54. Guiguet, C. J.,and G. C. Pike. 1965. First specimen record of the gray grampus or Risso’s dolphin, Grampus griseus (Cuvier), from British Columbia. Murrelet 46(1): 16. Hatler, D. G. 1971. A Canadian specimen of Risso’s dol- phin (Grampus griseus). Canadian Field-Naturalist 85(2): 188-189. Leatherwood, S., W. Perrin, V. Kirby, C. Hubbs, and M. Dahlheim. 1980. Distribution and movements of Risso’s dolphin, Grampus griseus, in the eastern North Pacific. Fisheries Bulletin 77(4): 951-963. Reimchen, T. 1980. Sightings of Risso’s dolphins (Gram- pus griseus) off Queen Charlotte Islands, British Colum- bia. Murrelet 61(1): 44-45. Rice, D. W. 1977. A list of the marine mammals of the world. United States Department of Commerce, National Oceanic and Administration Technical Report National Marine Fisheries Service SSRF-711. 15 pp. Sleptsov, M.M. 1952. Kitoobraznye dal’nevostochnykh morey (Cetaceans of far eastern seas). Izvestiya TINRO, 38. Vladivostok, USSR. Sleptsov, M. M. 1961. Usloviya sushchestrovaniya kitoob- raznykh v zohnakh smesheniya kholodaogo (Kurilo- Kamchatskogo) I Teplogo (Kurosio) Techemiy (Living conditions of cetaceans in zones where cold (Kurile- Kamchatka) and warm(Kuroshio) waters mix). TRUDY. Instituta Morfologii Univotn Ykh 34: 111-135. (Transla- tion, United States Naval Oceanographic Office, Washington, D.C., Translation number 481. 1970, 18 pp.) Stroud, S. K. 1968. Risso’s dolphin in Washington State. Journal of Mammology. 49(2): 347-348. Tomilin, A. G. 1957. Kitoobraznye (Cetacea). Vol. IX. Jn, Y. G. Heptner (eds), Zveri SSSR i Prilezhashchikh stran (Mammals of the USSR and adjacent countries). National Akademy, Moscow, 756 pp. (Translation, Israel Program of Scientific Translations, 1967. 717 pp., available through the National Technical Information Service, Springfield, Virginia.) Received 20 January 1982 Accepted 15 August 1982 1983 NOTES 91 The Swamp Saxifrage, Saxifraga pensylvanica, a Rare Plant in Canada, Newly Discovered in Saskatchewan VERNON L. HARMS Biology Department and W. P. Fraser Herbarium, University of Saskatchewan, Saskatoon, Saskatchewan S7N 0WO Harms, Vernon L. 1983. The Swamp Saxifrage, Saxifraga pensylvanica, a rare plant in Canada, newly discovered in Saskatchewan. Canadian Field-Naturalist 97(1): 91-93. Saxifraga pensylvanica, the Swamp Saxifrage, is reported as newly discovered in the Saskatchewan Pasquia Hills. The Canadian records for this rare species are reviewed. Key Words: Saxifraga pensylvanica, Swamp Saxifrage, Saskatchewan, rare plant. In June 1981, during a botanical foray in the Sas- katchewan Pasquia Hills, we made the rather startling discovery of Swamp Saxifrage, Saxifraga pensylvan- ica L. ssp. pensylvanica [about 1.2 km west of Km 21 (Mile 13) of the Fir River Road north of Veillardville (Sect. 2, Twp. 47N, Rge. 5N2M; 53°01’N, 102°38’W; elev. 1825’), very scarce on sedge hummocks in an open, marshy bog-fen; Harms et al. #29416 (SASK), see Figure 1]. Besides myself, the members of the field party included Patricia Sky, the Fraser Herbarium Technician, and local naturalists, Donald F. Hooper of Somme and Les Baker of Carragana, the latter actually being the first to sight this different- appearing plant. Some of the plant associates in the fen community were the codominant sedges: Eriopho- rum viridi-carinatum (Engelm.) Fern., E. gracile W. J. D. Koch, Carex prairea Dewey, C. tenuiflora Wahl., C. lasiocarpa Ehrh., C. magellanica Lam. var. irrigua (Wahl.) B.S.P., C. diandra Schrank, C. canes- cens Dewey, and C. interior Bailey; the dominant shrubs: Betula glandulifera (Regel) Butl., Salix can- dida Fluegge, Salix spp., and Chamaedaphne calycu- lata (L.) Moench; the more common forbs: Menyanthes trifoliata L., Potentilla palustris (L.) Scop., Galium trifidum L., Lysimachia thyrsiflora L., Equisetum fluviatile L., Triglochin maritima L., and Caltha palustris . Other Saskatchewan rare plants found at the boggy Black Spruce treed borders of the same fen were Pedicularis parviflora Sm., Cardamine pratensis L. var. palustris Wimn. & Grab., and Cypripedium calceolus L. var. pubescens (Willd.) Correll. The known range of Saxifraga pensylvanica, as extrapolated from Burns (1942) and a combination of current floras, has been from southeastern Manitoba, western and southern Ontario, and southern Maine, south to southeastern Minnesota, northern Missouri, Illinois, and western North Carolina. It is considered rare at least in most of the peripheral areas of its range. This species has been listed as rare in Canada (L. Kershaw, J. K. Morton, and J. Venn. 1976. Inventory of Rare and Endangered Vascular Plants of Ontario: Computer printout, Department of Biology, Univer- sity of Waterloo, Ontario; Argus and White 1977, White and Johnson 1980). Canadian records appear 7a29 FIGURE |. Saxifraga pensylvanica L. Collection from the Pasquia Hills, Saskatchewan. 92 THE CANADIAN FIELD-NATURALIST SCALE O 200 400 Miles O 200 600 Km Vol. 97 FiGuRE 2. Canadian Distribution of Saxifraga pensylvanica. quite limited for the plant, and it is reportedly scarce wherever found in this country. Macoun (1886) cited an early Ontario collection by David F. Day from “low places, near Fort Erie” (in Welland Co., 42°54’N, 78°56’W), but Looman(1973) indicated that a herbarium search by Dr. Bernard Boivin had failed to locate the Day specimens to substantiate that report. According to Looman, Boivin’s (1966) inclu- sion of the species for Ontario was based only on the more recent collections by Ward and C. E. Garton from the Rainy River area of far western Ontario. Scoggan (1978) further cites it, however, from Long Beach in Welland County, Ontario (42°52’N, 79°23’W), based on a report by Zenkert (1934). lam unaware of any other Ontario records for the Swamp Saxifrage. Looman (1973) recorded it from the Moose Lake area (49° 12’N, 95° 19’W), southeastern Manitoba, a location just across the Lake-of-the-Woods within 50 miles of the Rainy River station in Ontario. The recent White and Johnson (1966) listing of the Manitoba rare plants cites only this single locality for the species in Manitoba. It was omitted for any of the Prairie Provinces by Breitung (1957), Scoggan (1957), Budd and Best (1964), Boivin (1968), and even Scoggan (1978). Figure 2 shows the widely separated Canadian stations thus far recorded for the species. The present record from the Saskatchewan Pasquia Hills represents a surprising and phytogeographically quite interesting find, marking a 640 km (about 400 mile) northwestward extension of the species’ known range in Canada. The Saskatchewan Pasquia Hills, along with the Porcupine Hills and Duck Mountain upland, are parts of the Manitoba Escarpment, representing highlands that bordered the south shores of former Glacial Lake Agassiz. These hills are phyto- geographically interesting for including northwestern outlier stations (relict populations?) of a number of plant species belonging to the Eastern Deciduous Forest Element (e.g. Anemone nemorosa L. var. bifo- lia (Farw.) Boivin, Cypripedium arietinum R.Br., Mimulus ringens L., Polygala pauciflora Willd., Pru- nus pumila L., Trillium cernuum L.). The new record of Saxifraga pensylvanica, while more disjunctly iso- lated, and thus a more striking example, than others in the above group, seems phytogeographically classifi- able with them. It probably represents a persistent relict from the post-Pleistocene Climatic Optimum. At that time, associates of the present eastern decidu- ous forest appear to have extended their ranges con- 1983 siderably farther northwestward than at present, but have subsequently been largly eradicated by climatic deterioration. According to the hypotheses of Love (1959), and in apparent agreement with the more recent findings of Ritchie (1966, 1976) and other mod- ern workers, this maximum northwestward extension of the eastern forest element into Saskatchewan is probably referable to the earlier, more moist part of the “Hypsithermal Period”, dated at about 9000-6000 years B.P. Literature Cited Argus, G. W., and D. J. White. 1977. The Rare Vascular Plants of Ontario. Syllogeus (14) National Museums of Canada, Ottawa. Boivin, B. 1966. Enumeration des Plantes du Canada. II] -Herbidées, 1° partie: Digitatae: Dimerae, Liberae. Le Naturaliste Canadien 93: 583-646. Boivin, B. 1968. Flora of the Prairie Provinces. Part II - Digitatae, Dimerae, Liberae: Provancheria 3 (Reprinted from Phytologia 17). Breitung, A. J. 1957. Annotated catalogue of the vascular flora of Saskatchewan. The American Midland Naturalist 61: 510-512. Budd, A. C., and K. F. Best. 1964. Wild Plants of the Can- adian Prairies. Agriculture Canada, Ottawa. 519 pp. Burns, G. W. 1942. The taxonomy and cytology of Saxi- fraga pensylvanica L. and related forms. American Mid- land Naturalist 28: 127-160. Looman, J. 1973. Some new and interesting plant records for the Prairie Provinces. The Blue Jay 3(3): 176-179. NOTES 93 Looman, J.,and K. F. Best. 1979. Budd’s Flora of the Can- adian Prairie Provinces. Agriculture Canada publication no. 1662. Love, D. 1959. The postglacial development of the flora of Manitoba: a discussion. Canadian Journal of Botany 37: 547-585. Macoun, J. 1886. Catalogue of Canadian Plants. Part III - Apetalous Exogens: Additions and Corrections to Parts I-II, p. 523. Geological and Natural History Survey of Canada. 425 pp. Ritchie, J. C. 1966. Late-Pleistocene History of the Cana- dian Flora. Pp. 66-80. In The Evolution of Canada’s Flora. Edited by R. L. Taylor and R. A. Ludwig. Univer- sity of Toronto Press, Toronto. 137 pp. Ritchie, J. C. 1976. The late-Quaternary vegetational his- tory of the Western Interior of Canada. Canadian Journal of Botany 54: 1793-1818. Scoggan, H. J. 1957. Flora of Manitoba. National Muse- ums of Canada Bulletin 140: 619 pp. Scoggan, H. J. 1978. The Flora of Canada. Part 3 - Dicotyledoneae (Sauraceae to Violaceae). National Museums of Canada, Ottawa. White, D., and K. L. Johnson. 1980. The Rare Vascular Plants of Manitoba. Syllogeus (27) National Museums of Canada, Ottawa. Zenkert, C. A. 1934. The Flora of the Niagara Frontier Region: Ferns and Flowering Plants of Buffalo, N.Y., and Vicinity. Buffalo Society of Natural Sciences Bulletin 16: 1-328. Received 5 October 1981 Accepted 30 September 1982 Small Game Hunting Behaviour of Polar Bears, Ursus maritimus GARY D. MILLER! and DONALD R. WOOLDRIDGE? 'Biology Department, University of New Mexico, Albuquerque, New Mexico, 87131 ?Quintette Coal Ltd., 1508 102nd Ave. Dawson Creek, British Columbia VIG 2E2 Miller, Gary D. and Donald R. Wooldridge. 1983. Small game hunting behaviour of Polar Bears, Ursus maritimus. Canadian Field-Naturalist 97(1): 93-94. Anadult female Polar Bear ( Ursus maritimus) was observed catching a small rodent and a subadult Polar Bear was observed hunting and catching a Willow Ptarmigan (Lagopus lagopus). Key Words: Feeding behaviour, hunting behaviour, Polar Bear, Ursus maritimus. The major prey of Polar Bears (Ursus maritimus) throughout the Arctic is seals. Polar Bears are most successful hunting seals on sea ice (Stirling 1974), but can also hunt seals in ice-free water (Furnell and Oolooyuk 1980). Each year the bears along the west- ern coast of Hudson Bay are faced with three to four months of little or no ice cover (Stirling et al. 1977), and based on scat analysis, their diet shifts to include carrion, berries, grass, birds, and microtine rodents (Russell 1975). Polar Bears also encounter Porcupines (Jonkel 1968), but it is not known if they consume them. Polar Bears are known to take Canada Geese (Branta canadensis) on land, but most of their bird hunting is probably confined to open water (Russell 1975). During field observations of bear deterrent and 94 THE CANADIAN FIELD-NATURALIST detection systems near Churchill, Manitoba, we observed two incidences of Polar Bears preying upon small game. On 30 October 1978 we observed an adult female successfully hunt a small rodent. Then on 4 November 1978 we observed a subadult female stalk and take a Willow Ptarmigan (Lagopus lagopus). The adult female we were watching had a cub of the year with her. She was walking across an esker near a large willow thicket with her cub following 5m behind. There was approximately 10 cm of snow. As they were walking, the adult stopped suddenly and watched the ground at her feet with her ears directed toward the ground. When she stopped, her cub also stopped and remained about 4 m behind her, watch- ing. After watching and listening for 3 to 4 seconds, the mother raised up, lifted her front feet off the ground and then pounced down hard on the snow with her front feet together. She then bit the snow and came up with a small rodent (probably Microtus pen- nsylvanicus). She then resumed walking while chew- ing her catch and her cub followed immediately. The second incident of hunting we observed involved a subadult female stalking a Willow Ptarmi- gan. The bear approached a willow thicket from downwind and moved slowly through the brush. She occasionally disappeared from view, apparently crouching slightly. Suddenly she sprang up and for- ward, landing front feet first about | m ahead of her initial jumping point. The willow thicket was covered with a layer of snow approximately 40 cm thick and the attack generated a considerable spray of powder snow. The bear then walked directly out of the thicket carrying the ptarmigan in her mouth. She proceeded onto a nearby frozen lake and consumed the kill. During our research period (September to November 1978), large flocks of ptarmigan were seen briefly each day. Ptarmigan roost in the snow under willow cover and occasionally roost under the snow. Likewise, small rodents utilise tunnels under the snow once there is sufficient snow cover. Consequently, the bears probably locate such prey by smell and hearing rather than by sight. Polar Bears are not notable consumers of birds or Vol. 97 small mammals. Scat analyses have shown birds and small mammals among their food items (Russell 1975) but little is known about how the bears obtain such food. Our observations indicate that Polar Bears obtain small game opportunistically as in the case of the small rodent, but they also actively hunt small prey. The subadult bear was hunting through the wil- low thicket when she found and captured the ptarmi- gan. In Churchill, as in the rest of southern Hudson Bay, the ice-free period during the summer is rela- ‘tively long. There may be 3 to 4 months during which bears cannot hunt seals from the sea ice and must live ashore along the coast. A bear’s ability to find and capture small prey species during these ice-free months may be important for its survival during the ensuing winter. Acknowledgments We thank Scott Mair for his assistance during the project and C. Jonkel for his support. Gary Miller was funded by National Science Foundation Grant No. 7617644 to C. Jonkel. Don Wooldridge was funded by the Government of the Northwest Territories. Literature Cited Furnell, D. J. and D. Ooyooluk. 1980. Polar Bear preda- tion on Ringed Seals in ice-free water. Canadian Field- Naturalist 94(1): 88-89. Jonkel, C. J. 1968. A Polar Bear and Porcupine encounter. Canadian Field-Naturalist 82(3): 222. Russell, R. H. 1975. The food habits of Polar Bears of James Bay and southwest Hudson Bay in summer and autumn. Arctic 28: 117-129. Stirling, I. 1974. Midsummer observations on the behav- iour of wild Polar Bears (Ursus maritimus). Canadian Journal of Zoology 52(9): 1191-1198. Stirling, I., C. J. Jonkel, P. Smith, R. Robertson and D. Cross. 1977. The ecology of the Polar Bear (Ursus mari- timus) along the western coast of Hudson Bay. Occasional Paper No. 33, Canadian Wildlife Service Catalogue No. CW69-1 / 33. Received 26 March 1981 Accepted 19 August 1982 1983 NOTES 95 Birds First Described from Hudson Bay C. STUART HOUSTON 863 University Drive, Saskatoon, Saskatchewan S7N 0J8. Houston, C. Stuart. 1983. Birds first described from Hudson Bay. Canadian Field-Naturalist 97(1): 95-98. More than two centuries ago, Linnaeus, Forster and others first named 19 new bird species and 9 forms now recognized as subspecies from the vast territory then known as ‘Hudson’s Bay.’ The nearly-forgotten collectors were James Isham, Andrew Graham, Alexander Light, Humphrey Martenand Thomas Hutchins. Changes in type locality designation are recommended for three species and one subspecies. Key Words: Hudson Bay, history, collections, type localities, Marbled Godwit (Limosa fedoa), White Pelican (Pelecanus erythrorhynchos), Purple Martin (Progne subis), subspecies of Gyrfalcon (Falco rusticolus obsoletus). Between the late 1750s and 1770s, a handful of Hudson’s Bay Company fur traders collected bird specimens, 28 of which were given Latin names by Linnaeus and other early taxonomists. “These few people provided a remarkable concentration of early type specimens from one sparsely settled area. This article emphasizes the full extent of the type locality, ‘Hudson Bay’, and corrects errors and omis- sions in five earlier reviews of pre-1800 work. Stevens (1936) mistakenly credited the Golden Plover to Edwards and the Greater Yellowlegs and Tree Spar- row to Forster; Baillie (1946) told of eight species when in fact there were 13; McAtee (1950) covered Edwards very well, but the Hudson Bay provenance poorly; Allen (1951) omitted any mention of James Isham or Alexander Light; Snyder (1963) suggested an overly restrictive type locality of ‘northeastern Manitoba’ for 13 species. None of the five had access to two important but limited-edition books which contain Isham’s, Graham’s and Hutchins’ early bird observations from Hudson Bay (Rich 1949, Williams 1969). The general type locality of ‘Hudson Bay’ used in the Sth American Ornithologists’ Union Check- List (1957) is inadequate or misleading for several species, in addition to the geographic ambiguity inherent in the citation of a vast inland sea, up to 1600 km in length and up to 1000 km in width. Few ornithologists have appreciated that until 1870 the popular term ‘Hudson’s Bay (the possessive form here is no longer in official geographic use) designated an area of nearly 3.6 million km? extending west to the Rocky Moun- tains and draining into the bay (Rich 1967). In this area, Officially named ‘Rupert’s Land’ for 200 years, the people, as well as some of its birds and mammals, were sometimes called ‘Hudsonians’ (Hearne 1795). For example, when Joseph Sabine described the North American form of the Black-billed Magpie, now Pica pica hudsonia, from specimens collected 600 km inland at Cumberland House, he named it “Corvus Hudsonius, Hudson’s Bay Magpie” (Sabine 1823). Histories of the Hudson’s Bay Record Society similarly spoke of the 1714 negotiations “settling the boundary between Hudson Bay and Canada” (Davies and Johnson 1965). Linnaeus (1758) thus was 112 years ahead of his time when he used the term ‘Canada’ for the type locality of the Blue Goose, Surf Scoter, Spruce Grouse, Sharp-tailed Grouse, and one subspecies of Golden Eagle — all of which came from Hudson Bay. Canada was not an official name until 1791, and only after the purchase of the Rupert’s Land territory from the Hudson’s Bay Company in 1869-70 was it correct to use ‘Canada’ for the area which encompassed Hud- son Bay. Before Hearne established the first Hudson’s Bay Company (HBC) inland trading post at Cumberland House in 1774, Indians came many hundreds of miles to trade at the posts on Hudson Bay. Andrew Graham told how, before 1769, Indians were bringing him birds’ “skins stuffed and dried... from inland” ( Willi- ams 1969). Although an individual bird can wander far from its breeding territory, knowledge of typical habitat and present distribution offers a strong prob- ability that Isham’s Marbled Godwit was collected far inland, perhaps along the Saskatchewan River. For similar reasons, an inland derivation is almost as likely for the Purple Martin, White Pelican, and per- haps the Whooping Crane, although Samuel Hearne (1795) mentioned that occasionally (“not very often”) a few Whooping Cranes would visit the bay in spring. For at least the Marbled Godwit, Purple Martin, and White Pelican, ‘Hudson Bay Territory’ or Hudson Bay drainage basin’ would bea more accurate descrip- tion of the type locality than the present use of ‘Hud- son Bay’ or Snyder’s (1963) suggestion of ‘Northeast- ern Manitoba’. For the remaining species to be listed, ‘Hudson Bay’ is not a misleading designation, Thirteen of the new species for which the type local- ity of ‘Hudson Bay’ is given in the 5th AOU Check- 96 THE CANADIAN FIELD-NATURALIST Churchill York Severn Cumberland ® Vol. 97 HUDSON BAY 9 Albany: Moose Factory FiGureE |. Extent of Hudson’s Bay Territory, 1670-1870 (shaded area). Map redrawn from Encyclopedia Canadiana, edited by J. E. Robbins, 1957, with kind permission of the Grolier Society of Canada Limited. List (1957) were collected by James Isham and illus- trated in George Edwards’ A Natural History of Uncommon Birds, volume 3 (Edwards 1750). They were among the first birds to receive binomial Latin names (Linnaeus 1758) as follows: Ardea herodias (= Great Blue Heron), page 143 in Linnaeus (= blue morph of Snow Goose, now Chen caerules- cens), p. 124; (= Surf Scoter Melanitta perspicillata), p. 125; (= Spruce Grouse Dendra- gapus canadensis), p. 159; (= Sharp-tailed Grouse Tympanuchus phasianel- lus), p: 160; (= Whooping Crane Grus americana), p. 142; (= Sandhill Crane Grus canadensis), p. 141; (= Sora Porzana carolina), pp. 153-154; Anas caerulescens Anas perspicillata Tetrao canadensis Tetrao phasianellus Ardea americana Ardea canadensis Rallus carolinus (= Marbled Godwit Limosa fedoa), p. 146; (= Hudsonian Godwit Limosa haemastica), p. 147; (= Red Phalarope Phalaro- pus fulicaria), pp. 148-149; (= Northern Phalarope Phalaropus lobatus), Scolopax fedoa Scolopax haemastica Tringa Fulicaria Tringa lobata p. 148; Hirundo subis (= Purple Martin Progne subis), p. 192. Isham’s own observations, including mention of 39 species of birds, but only six of the above species, were not formally published until 1949 (Rich 1949). Following up on Isham’s collections, the Hudson’s Bay Company in 1770 requested that its employees’ make a concerted effort to send more specimens back to England (Williams 1969, 1978). In 1771, Andrew Graham sent 64 skins of 39 species from Severn, aug- mented by 17 skins of seven species from Humphrey Marten at Albany, 17 skins of eight species from Moses Norton at Churchill, and two skins from Fer- dinand Jacobs and surgeon Thomas Hutchins at York 1983 Factory (Williams 1969). From the eight boxes of specimens, Johann Reinhold Forster the next year described five valid new species, with exact type locali- ties (Forster 1772), the Eskimo Curlew from Albany, and the remaining four from Severn, as follows: (= Eskimo Curlew Nume- nius borealis), pp. 411 & 431-432 in Forster; (= Great Gray Owl), pp. 386 & 424-425; (= Boreal Chickadee), pp. 408 & 430-431; (= Blackpoll Warbler Den- droica striata), pp. 406 & 428-430; (= White-crowned Sparrow Zonotrichia leucophrys), pp. 403-404 & 426-428. John Latham (1785) first described the White Peli- can in his General Synopsis of Birds, volume 3, based on two specimens from ‘Hudson’s Bay’. The pelican bill sent from Cumberland House by William Tomi- son, the full specimen sent via York Factory in 1771 and the specimen sent from Albany by surgeon Edward Jarvis about the same time may have been involved (Williams 1969). J.F. Gmelin (1789), in the 13th edition of Systemae Naturae, after the death of Linnaeus, bestowed the binomial of Pelecanus erythrorhynchos. Scolopax borealis Strix nebulosa Parus hudsonicus Muscicapa striata Emberiza leucophrys Andrew Graham in the 1770s wrote notes on I11 species of birds, and one version of his manuscript accompanied the collections. Thomas Hutchins later copied Graham’s notes and added his own on another 11 species, and subsequently received inappropriate credit for Graham’s observations as well as his own. The matter of authorship was sorted out and the observations of both men were eventually published together (Williams 1969, 1978). Samuel Hearne (1795), probably a better naturalist than either Graham or Hutchins, provided the first recognizable description of what he called the Horned Wavey, but did not provide a Latin binomial for this small goose that was regular to the northwest of Chur- chill. Cassin (1861) named this goose for B.R. Ross, another HBC fur trader, some 66 years after Hearne’s description. Of the nine taxa now recognized as subspecies, each with the type locality of ‘Hudson Bay,’ two owe their provenance to another HBC employee, Alexander Light. His specimens, sent home to England even before those of Isham (Davies and Johnson 1965) were illustrated by Edwards (1743, 1747) in his first two volumes. The first was named by P.L.S. Muller (1776) in his supplement to Systemae Naturae, and the NOTES 97 second by Gmelin (1789) in his 13th edition of Syste- mae Naturae, as follows: Strix caparoch (= Northern Hawk Owl Sur- nia ulula caparoch), p. 69; (= Willow Ptarmigan Lagopus lagopus albus), Os FSW. Two other forms were named by Linnaeus from birds illustrated by Edwards in his first two volumes (1743, 1747). The first was an eagle shown with fea- thered tarsi but a white tail, brought alive to England by an unnamed “Gentleman employ’d in the Hudson’s-Bay Company’s Service” (Edwards 1743) and later designated on somewhat questionable grounds as the type for a subspecies: Falco canadensis (= Golden Eagle Aquila chrysaetos canadensis), p. 88. The second, definitely an Isham specimen, was the one Hudson Bay taxon first described by Linnaeus (1776) in his 12th edition: (= Northern Harrier Circus cyaneus hudsonius), p. 128. Lagopus albus Falco hudsonius When Latham (1790) began using Latin binomials for the first time in his Index Ornithologicus, he named froma Hudson Bay specimen the North Amer- ican form of the Whimbrel: Numenius hudsonicus (= Whimbrel Numenius phaeopus hudsonicus), Do VU Finally, Thomas Pennant (1785) in his first edition of Arctic Zoology described four more forms from Hudson Bay that were given the following Latin binomials by Gmelin (1789). (= Horned Grebe Podiceps auritus cornutus ), p. 591; (= Rock Ptarmigan Lago- Pus mutus rupestris), jo. Worle (= Great Horned Owl Bubo virginianus wapacuthu), pp. 291-292; (= Gyrfalcon Falco rustico- lus obsoletus), p. 268. Inthe example of F. r. obsoletus, the type locality of ‘Hudson Strait’ given in the Sth AOU Check- List is incorrect, and should be changed to ‘Hudson Bay’ as first indicated by Pennant (1785). The error evidently arose through an unfortunate literal translation of Gmelin (1789), who used ‘in freto Hudsonis’ in the alternative wider Latin sense of ‘the sea in general’ (Simpson 1963: 256) for the Northern Harrier, Great Gray Owl, and Gyrfalcon, all of which had been sent from the bay — not the strait. Colymbus cornutus Tetrao rupestris Strix wapacuthu Falco obsoletus 98 THE CANADIAN FIELD-NATURALIST I] hope the above material will correct some miscon- ceptions and also give belated recognition to the fur traders who, two centuries ago, made important bird specimen contributions from a remote region of what is now called Canada. Literature Cited Allen, E. G. 1951. The history of American ornithology before Audubon. Transactions of the American Philoso- phical Society 41: 387-591. American Ornithologists’ Union. 1957. Check-List of North American Birds, Sth edition. Lord Baltimore Press, Baltimore. 691 pp. Baillie, J. L. 1946. Naturalists on Hudson Bay. Beaver 277.2: 36-39. Cassin, J. 1861. Communication in reference to a new spe- cies of goose from Arctic America. Proceedings of the Academy of Natural Sciences, Philadelphia 13: 73. Davies, K. G.and A. M. Johnson. 1965. Letters from Hud- son Bay, 1703-40. Hudson’s Bay Record Society, London. Ixvi+ 455 pp. Edwards, G. 1743, 1747, 1750, 1751. A Natural History of Uncommon Birds (4 vols.). London. Forster, J. R. 1772. Anaccount of the birds sent from Hud- son’s Bay: with observations relative to their natural his- tory; and Latin descriptions of some of the most uncom- mon. Philosophical Transactions 62: 382-433. Gmelin, J. F. 1789. Systemae Naturae, 13th edition. J. B. Delamolliere, London. Hearne, S. 1795. A Journey from Prince of Wales Fort in Hudson’s Bay to the Northern Ocean. T. Cadell, London. xliv + 458 pp. Latham, J. 1781, 1785. A General Synopsis of Birds. Ben- jamin White, London. Latham, J. 1790. Index Ornithologicus. Volume 2. Leigh and Sotheby, London. Vol. 97 Linnaeus, Carolus (Linné, Karl von). 1758. Systemae Naturae, 10th edition. L. Salvi, Stockholm. 823 pp. Linnaeus, Carolus (Linné, Karl von). 1766. Systemae Naturae, 12th edition. L. Salvil, Stockholm. McAtee, W. L. 1950. The North American birds of George Edwards. Journal of the Society for the Bibliography of Natural History 2: 194-204. Muller, P. L.S. 1776. Supplement to Systemae Naturae. Pennant, T. 1785. Arctic -Zoology, volume 2. Henry Hughes, London. Rich, E. E., Editor. 1949. James Isham’s Observations on Hudson’s Bay, 1743. Champlain Society, Toronto. cv + 352 pp. Rich, E. E. 1967. The Fur Trade and the Northwest to 1857. McClelland & Stewart, Toronto. x + 336 pp. Robbins, J. E., Editor. 1957. Encyclopedia Canadiana, 10 vols. Grolier Society, Toronto. Sabine, J. 1823. Zoological Appendix (v), pp. 647-703, in: Franklin, J., Narrative of a Journey to the Shores of the Polar Sea in the Years 1819, 20, 21, and 22. John Murray, London. Simpson, D. P. 1963. Cassell’s New Latin Dictionary. Funk & Wagnalls, New York. Snyder, L. K. 1963. On the type locality of thirteen North American birds. Canadian Field-Naturalist 77: 128-129. Stevens, O. A. 1936. The first descriptions of North Ameri- can birds. Wilson Bulletin 48: 203-215. Williams, G., Editor. 1969. Andrew Graham’s Observa- tions on Hudson’s Bay, 1767-91. Introduction by R. Glover. Hudson’s Bay Record Society, London. Ixxu + 423 pp. Williams, G. 1978. Andrew Graham and Thomas Hut- chins. Collaboration and Plagiarism in | 8th-Century Nat- ural History. Beaver 308.4: 4-14. Received 8 February 1982 Accepted 21 July 1982 1983 NOTES 99 The Occurrence and Origin of Tiger Trout, Salmo trutta X Salvelinus fontinalis, in Ontario Streams LARRY D. WITZEL! College of Biological Sciences, University of Guelph, Guelph, Ontario NIG 2W1 ‘Present address: Ontario Ministry of Natural Resources, Nanticoke Fish Study, P.O. Box 429, Part Dover, Ontario NOA INO Witzel, Larry D. 1983. The occurrence and origin of Tiger Trout, Sa/mo trutta x Salvelinus fontinalis, in Ontario streams. Canadian Field-Naturalist 97(1): 99-102. Three natural hybrids of the Brook Charr (Sa/velinus fontinalis) x Brown Trout (Salmo trutta) are described; the first published report of wild Tiger Trout from streams in Ontario. Examination showed that body vermiculation and the distribution of vomerine teeth were the most variable characters among hybrids. It was postulated that hybrids were probably more a result of an excited rival male Brook Charr intruding upon the spawning of paired Brown Trout, rather than from the actual spawning of Brook Charr with Brown Trout or the chance fertilization of trout ova by the downstream drift of charr milt. Key Words: Brook Charr, Brown Trout, Tiger Trout, wild, intergeneric hybridization, fertilization, intraspecific spawning, hypothesis. Since the pioneer work of Day (1882) several authors have reported on artificial hybrids within the family Salmonidae. These studies have included the Tiger Trout, a hybrid of the Brook Charr, Sa/velinus fontinalis (Mitchill) and Brown Trout, Salmo trutta Linnaeus. Day’s (1882) investigations characterized the Tiger Trout’s sterility, distinct colouring, vermicu- lated body pattern, poor survival, and taxonomic characters as intermediate between those of the Brook Charr and Brown Trout. Subsequent studies by Alm (1955), Buss and Wright (1956, 1958), Susuki and Fukuda (1971, 1973), and Blanc and Chevassus (1979) have elaborated on these and other aspects of the Tiger Trout. Brown (1966) described three Tiger Trout collected from streams in Montana overa 1|5 year period. Allan (1977) described the first Tiger Trout reported in Alberta. Despite this hybrids apparent rare occur- rence in the wild, other feral specimens have undoubt- edly been sighted, but, they either were not reported or were not recognized as different when captured. Fowler (1944) identified two specimens of Tiger Trout from a tributary of the Delaware River in 1932 as dwarf Brown Trout, however, a photograph accom- panying Fowler’s report would seem to indicate these fish were Tiger Trout. This paper reports the first incidence of wild Tiger Trout from Ontario, a description of each specimen, and speculation on behavioural and biological events which could lead to the hybridization of Brook Charr with Brown Trout in nature. A search of the literature (on fish hybrids) compiled by Schwartz (1972) and Dangel et al. (1973) indicated that no one has observed Brook Charr spawning with Brown Trout, nor has there been an explanation of how these two species might hybridize in nature. Ontario Tiger Trout In Galt Creek near Aberfoyle, Ontario, I captured a Tiger Trout on 19 October 1978 while electrofishing for Brook Charr and Brown Trout. A written inquiry to all nine Ontario Districts of the Ministry of Natural Resources within the sympatric range of the charr and trout, and a search of records (by museum staff) at the Royal Ontario Museum (ROM) and the National Museums of Canada revealed that only two other wild caught specimens existed for Ontario (see Table 1) and that neither had been published. Both of these Tiger Trout were caught by anglers and donated to the ROM which lent them to the author for this study. A Tiger Trout was recorded in a 1975 Ministry of Natural Resources stream survey report of Gibsons Creek near Turkey Point. However, this specimen and others allegedly caught by anglers from Gibsons Creek have not been authenticated (apparently neither specimens nor photographs of them were kept). Characters of the Ontario Tiger Trout were mea- sured or counted according to the methods of Hubbs and Lagler (1964). Vertebral counts were made from radiograms. Ontario Tiger Trout demonstrated close similarity in the characters examined (Table |). Body vermiculations and the distribution of vomerine teeth showed the greatest variability among hybrids. These two characters were also among the most variable in 100 TABLE |. Catch and numerical information on wild Tiger Trout from Ontario. Specimens Date of 21 July 19 October 19 May capture 1969 1978 1979 Location of Potters Ck. Galt Ck. Bighead R. capture Norfolk Co. Wellington Co. Grey Co. ROM catalogue number 27155 — 35744 Weight (g) 111.4 32.0 96.4 Lengths (cm) standard 19.6 — 18.2 fork 13.9 — total 24.0 14.7 22.0 head 5.0 — od snout 3} — 1.4 orbit Ie — lee Counts branchiostegals 1] — 12 gill rakers 17 — 18 principal dorsal rays 11 — 1] pectoral rays 13 — 12 vertebrae 59 i 60 pored lateral line scales 111 — 114 teeth on head of vomerine 2 — 4 teeth on shaft of vomerine 4 — 2 Body markings along dorsum narrow small irregular small irregular circular shaped dark shaped spots bands spots along side circular © large irregular dark circular bands shaped dark bands aligned in spots becoming a row wide irregular bands in caudal half Age 2+ I+ 2+ natural hybrids reported by Brown (1966) and Allan (977): There is little doubt that the Galt Creek specimen is a Tiger Trout as the body markings of this particular cross are quite distinct from either parent. Although different from each other (Figure 1), the appearance _ of the three Ontario hybrids resembles that of cultured Tiger Trout (Buss and Wright 1958) and wild speci- mens captured outside of Ontario (Brown 1966, Allan IT). Possible Methods of Hybridization Ample opportunity for hybridization between the native Brook Charr and alien Brown Trout occurs where they co-exist in streams of southwestern Onta- THE CANADIAN FIELD-NATURALIST Vol. 97 rio. In southwestern Ontario, Brook Charr typically start to spawn one to two weeks before Brown Trout, but, the spawning activities of the charr generally overlap those of the trout by at least one week in most sympatric populations and up to three weeks in others (Witzel 1980). Interaction between charr and trout during reproduction is limited to a large extent by differences in preferred spawning sites (Reiser and Wesche 1977, Witzel 1980). Brook Charr, however, do occasionally spawn concurrently in the same areas as Brown Trout. This is more apt to occur where pre- ferred spawning sites are limited or where spawners are concentrated below stream barriers. In nature, cross-fertilization might occur by sperm drift in surface flow (Shuman 1950)-or in intragravel water flow (Stuart 1954). However, spermatozoa quickly lose motility when introduced to water (Gins- burg 1963) and are not viable after only a few seconds of exposure (Shuman 1950). Chances of fertilization are also lessened by the dilution of milt during drift. Similarly, unfertilized trout ova quickly harden in water and become impregnable after about three min- utes of exposure (Ginsburg 1963). Therefore, it is improbable that hybridization would occur from the dispersal of gametes by stream currents. Both reciprocal crosses of the Brook Charr with Brown Trout have survival rates lower than intraspe- cific crosses. Of the two interspecific crossings, hybrid survival is best (4 to 63%) from female Brown Trout with male Brook Charr (Alm 1955, Buss and Wright 1956, Blanc and Chevassus 1979). The reciprocal cross is much less viable, with 0 to 0.5% survival to the fry stage under experimental conditions (Alm 1955, Buss and Wright 1956). Brook Charr are not likely to actively spawn with Brown Trout in mixed pairs because of inherent dif- ferences in reproductive behaviour (such as courtship activities) which would likely disrupt intergeneric spawning. Although there is some evidence of inters- pecific intrageneric spawning in salmonids (Cordier- Goni 1939, Miller 1957), there is no direct evidence as judged from the literature cited by Schwartz (1972) and Dangel et al. (1973) that members of Sa/mo form spawning pairs with those of Salvelinus. Hybridization would seem most likely to occur when one or more rival male Brook Charr (usually the smaller ones) that are not attended by a female charr, invade the spawning of paired Brown Trout, releasing their milt coincidentally with the discharge of eggs from the trout. Attempts at hit-and-run fertilization by non-attending male Brook Charr with other spawning Brook Charr have been observed widely in nature (Hazzard 1932, Smith 1941, Bridges and Mullen 1972). Intraspecific intrusions by rival males during conspecific spawnings have been witnessed in other trout as well (Greeley 1932, Hobbs 1937, Jones 1983 J FIGURE |. Wild Ontario Tiger Trout captured from the Bighead River (upper specimen), Galt Creek (mid- dle), and Potters Creek (lower). and Ball 1954). Such behaviour in species of sunfishes (Lepomis: Centrarchidae) has been referred to as “cuckoldry” by Gross (1979). In Galt Creek, I have on at least one occasion observed several small male Brook Charr and Brown Trout without females, positioned peripheral to a rif- fle area (~ 6 m2) where paired trout and paired charr were spawning. It seems conceivable in this situation for an inexperienced male Brook Charr, excited by the spawning activities of other fish, to discharge its milt upon the ova of a paired female Brown Trout, unseen and coincidentally with the release of milt by the attending male trout. In this way, several Brown Trout ova are likely to be fertilized by a male Brook Charr, yet, except fora brief moment, the spawning of the two species remains intraspecific. Acknowledgments I wish to thank the many district staff of the Ontario Ministry of Natural Resources who provided essential information on Tiger Trout. Dr. E. J. Crossman, Curator of the Department of Ichthyology and Her- NOTES 101 petology, ROM, provided catch records and permit- ted examination of loaned museum specimens of wild Ontario Tiger Trout. lam grateful to L. Sztramko, S. Nepszy, and Dr. E. J. Crossman for their comments and criticisms of the earlier manuscripts. Barra Gots, Department of Zoology, University of Guelph, Guelph, Ontario was a co-discoverer of the Galt Creek Tiger Trout. This paper originated from studies on the reproductive biology of Brook Charr and Brown Trout which were supported by funds provided to Dr. H. R. MacCrimmon, Department of Zoology, Uni- versity of Guelph by the National Research Council of Canada and the Ontario Ministry of Natural Resources. Literature Cited Allan, J. H. 1977. First report of the tiger trout hybrid, Salmo trutta Linnaeus x Salvelinus fontinalis (Mitchill), in Alberta. Canadian Field-Naturalist 91: 85-86. Alm, G. 1955. Artificial hybridization between different species of the salmon family. Report, Institute Freshwater Research, Drottningholm 36: 13-56. Blanc, J. M., and B. Chevassus. 1979. Interspecific hybrid- ization of salmonid fish. I. Hatching and survival up to the 15th day after hatching in Fl generation hybrids. Aqua- culture 18: 21-34. Bridges, C. H., and J. W. Mullen. 1972. A compendium of the life history and ecology of the brook trout Sa/velinus fontinalis (Mitchill). Massachusetts Division of the Fish and Game Bulletin 23. 38p. Brown, C. J. D. 1966. Natural hybrids of Sa/mo trutta and Salvelinus fontinalis. Copeia 3: 600-601. Buss, K., and J. E. Wright, Jr. 1956. Results of species hybridization within the family Salmonidae. Progressive Fish-Culturist 18: 149-158. Buss, K., and J. E. Wright, Jr. 1958. Appearance and fertil- ity of trout hybrids. Transactions of the American Fisher- ies Society 87: 172-181. Cordier-Goni, P. 1939. Hypothése sur un hybride du sau- monet de la truite (Sa/mo salar L. — Salmo (Trutta) fario L.). Riviera Scientifique 26: 3-10. Dangel, J. R., P. T. Macy, and F. C. Withler. 1973. Anno- tated bibliography of interspecific hybridization of fishes of the subfamily Salmoninae. NOAA Technical Memo- randum, National Marine Fisheries Service NWFC-I. 48p. Day, F. 1882. On variation inform and hybridism in Salmo fontinalis. Journal Linnaen Zoological Society of London 17: 13-19. Fowler, H. W. 1944. The salmon-like fishes of Pennsylva- nia. Biennial Report Pennsylvania Board Fish Commis- sion for the period ending 31 May 1942, 55-63. Ginsburg, A. S. 1963. Sperm-egg association and its rela- tionship to the activation of the egg in salmonid fishes. Journal of Embryology and Experimental Morphology 11; 13-33. Greeley, J. R. 1932. The spawning habits of brook, brown and rainbow trout and the problem of egg predators. Transactions of the American Fisheries Society 62: 239-248. 102 Gross, M. R. 1979. Cuckoldry in sunfishes (Lepomis: Cen- trarchidae). Canadian Journal of Zoology 57: 1507-1509. Hazzard, A. S. 1932. Some phases of the life history of the eastern brook trout, Sa/velinus fontinalis (Mitchill). Tran- sactions of the American Fisheries Society 62: 344-350. Hobbs, D. F. 1937. Natural reproduction of Quinnat sal- mon, brown and rainbow trout in certain New Zealand waters. New Zealand Marine Department, Fisheries Bul- letin 6: 7-104. Hubbs, C. L., and K. F. Lagler. 1964. Fishes of the Great Lakes region. University Michigan Press, Ann Arbor, Michigan. 213p. Jones, J. W., and J. W. Ball. 1954. The spawning behav- iour of brown trout and salmon. Animal Behaviour 2: 103-114. Miller, R. B. 1957. Have the genetic patterns of fishes been altered by introductions or by selective fishing? Journal Fisheries Research Board of Canada 14: 797-806. Reiser, D. W., and T. A. Wesche. 1977. Determination of physical and hydraulic preferences of brown and brook trout in the selection of spawning locations. Completion report for project C-7002, Water Research and Technol- ogy. U.S. Department of the Interior, Water Resources Research Institute, University of Wyoming. 100p. Schwartz, F. J. 1972. World literature to fish hybrids with an analysis by family, species, and hybrid. Gulf Coast Research Laboratory Museum Publication No. 3. 328p. THE CANADIAN FIELD-NATURALIST Vol. 97 Shuman, R. F. 1950. On the effectiveness of spermatozoa of the pink salmon (Oncorhynchus gorbuscha) at varying distances from point of dispersal. Fisheries Bulletin 51: 359-363. Smith, O. R. 1941. The spawning habits of cutthroat and eastern brook trouts. Journal of Wildlife Management 5: 461-471. Stuart, T. A. 1954. Spawning sites of trout. Nature 173: 354. Susuki, R., and Y. Fukuda. 1971. Survival potential of F1 hybrids among salmonid fishes. Bulletin of the Freshwater Research Laboratory (Tokyo) 21: 69-83. Susuki, R., and Y. Fukuda. 1973. Appearance and numeri- cal characters of Fl hybrids among salmonid fishes. Bul- letin of the Freshwater Research Laboratory (Tokyo) 23: 5-32. Witzel, L. D. 1980. Relation of gravel size to spawning site selection and alevin production by Salvelinus fontinalis and Salmo trutta. M.Sc. thesis, University of Guelph, Guelph, Ontario. 203p. Received 7 April 1982 Accepted 31 August 1982 Two Sable Island Fungi, Peziza ammophila and Hygrocybe turunda, New to Nova Scotia S. A. REDHEAD and P. M. CATLING Biosystematics Research Institute, Research Branch, Agriculture Canada, C.E.F., Ottawa, Ontario KIA 0C6 Redhead, S. A.,and P. M. Catling. 1983. Two Sable Island Fungi, Pezizaammophila and Hygrocybe turunda, new to Nova Scotia, Canadian Field-Naturalist 97(1): 102-103. Peziza ammophila on dunes and Hygrocybe turunda on Sphagnum at Sable Island represent the first reports of these fungi from Nova Scotia and the first Canadian record of the Peziza. Key Words: Sable Island, Nova Scotia, fungi, first records, Peziza ammophila, Hygrocybe turunda. During a recent expedition to Sable Island (43° 56’ — 44°00’ N, 59°45 — 60°05’ W), N.S., to survey the vascular plant flora, the second author collected two fleshy fungi. Peziza ammophila Durieu et Montagne was found on the north shore, approximately 200 m east of West Point. It occurred in freshly deposited sand in an open dune sparsely covered with American Beach grass (Ammophila breviligulata Fernald). This is the first record of Peziza ammophila from Canada. Andersson (1950) gave a detailed account of this fun- gus with illustrations and mapped its distribution in the northern hemisphere. The first collections from North America (Florida) were described as Peziza funerata Cooke (1878). It has since been found on sand dunes in Michigan and in California (Seaver 1930, 1942). It occurs ina number of European coun- tries, the British Isles and Morocco(Andersson 1950), and has been reported from Australia (Seaver 1930), South Africa (Ramsbottom 1926) and Argentina (Singer 1968). In all cases it has been found on shifting sand containing little humic material. Hygrocybe turunda (Fr.) Karsten was collected on a thin Sphagnum cover over wet peaty sand with Large Bog Cranberry (Vaccinium macrocarpon Alt.), 1983 rushes (Juncus balticus Willd.) and Grass Pink Orchid (Calopogon tuberosus (L.) BSP) approximately 150 m N. of the weather station. It was also noted in similar associations at four other localities on the island. This species was not treated in Bird and Grund’s (1979) Nova Scotian monograph of Hygro- Phorus sensu lato. In the earlier North American monograph, Hesler and Smith (1963) illustrated and reported it from Idaho, Massachusetts, Michigan, Montana, Oregon and Washington. Hesler and Smith also recorded it (sub Hygrophorus turundus var. sphagnophilus (Peck) Hesler and Smith) from New- foundland. We confirmed this identification (see below). Collections from New Brunswick, northern Quebec and northern Ontario which extend the known range have also been seen. Hygrocybe turunda is characterized by its bright scarlet colours which can fade to yellow, fuscous scales of varying development on the pileus and large spores (see Hesler and Smith 1963) which distin- guishes it from the related Hygrocybe cantharella (Schw.) Murr. and H. miniata (Fr.) Kummer. Hygro- cybe cantharella occurs ina variety of habitats, almost always with mosses and sometimes on Sphagnum of a loose consistency, often in shady coniferous forests. Hygrocybe miniata is usually found in hardwood forests on well decayed mossy logs. Hygrocybe turunda occurs in open locations on sand or mosses such as Sphagnum over sandy substrates, but in New Brunswick it occurred in exposed sites in a raised bog on Sphagnum of a compacted form with robust capi- tulae. This exposed habitat is reflected by its overall distribution. Outside of North America it has been reported from Greenland, Iceland, the Faeroes, parts of northern Europe and Japan (see Hesler and Smith 1963). In more southerly latitudes it is restricted to the higher elevations or other exposed cool sites. Specimens examined: Peziza ammophila: N.S.: Sable Island, Aug. 25, 1981, P.M. Catling & W. Freedman (DAOM 180764). Hygrocybe turunda: Ont.: Algoma NOTES 103 Dist., L. Superior Prov. Park, Rabbit Blanket L., Sept. 14, 1972, D.W. Malloch 14.9.72.15 (TRTC). Que.: Reserve Chibougamau, 15.75 kmN. of Bochart on Hwy. 167, Aug. 29, 1976, J. Ginns & S. Redhead 2092 (DAOM 174756). N.B.: Kouchibouguac Natl. Park, Sept. 24, 1977, S.A. Redhead 2565 (DAOM G77). IWesss Sallie lsiernal, sue, ZS, WES LPS kyl. Catling & W. Freedman (DAOM 180763). Nfld: St. Anthony, Aug. 12, 1951, D.B.O. Savile 2856 & J. Vaillancourt (DAOM 28523). Acknowledgments We thank Dr. W. Freedman of Dalhousie Univer- sity for financial assistance enabling the second author to visit Sable Island. Dr. D. W. Malloch of the University of Toronto provided the loan of H. turunda from Ont. Literature Cited Andersson, O. 1950. Larger fungi on sandy grass heaths and sand dunes in Scandinavia. Botaniska Notiser, Sup- plement 2: 1-89. Bird, C. J., and D. W. Grund. 1979. Nova Scotian species of Hygrophorus. Proceedings of the Nova Scotian Insti- tute of Science 29(1): 1-131. Cooke, M. C. 1878. Ravenel’s American fungi. Grevillea 6: 129-146. Hesler, L. R.,and A. H.Smith. 1963. North American spe- cies of Hygrophorus. University of Tennessee Press, Knoxville. Ramsbottom, J. 1926. Chapter IX. Special groups of plants. A. Fungi. Pp. 152-173. /n Aims and methods in the study of Vegetation. Edited by A. G. Tansley and T. F. Chapp. British Empire Vegetation committee, London. Seaver, F. J. 1930. Photographs and descriptions of cup- fungi-XIII. Subhypogeous forms. Mycologia 22: 215-218. Seaver, F. J. 1942. The North American cup-fungi. (Oper- culates). Supplemented Edition. N.Y. Singer, R. 1968. Sand-dune inhabiting fungi of the south Atlantic coast from Uruguay to Bahia Blanca. Mycopa- thologia et Mycologia Applicata 34: 129-143. Received 21 October 1981 Accepted 12 August 1982 Note added in proof: Peziza ammophila has been reported and photographed in Japan on sand dunes. Otani, J. 1982. Some notes on two interesting fungi col- lected in Japan. Transactions of the Mycological Society of Japan 23: 379-384. 104 THE CANADIAN FIELD-NATURALIST Vol. 97 Observations on Primary Dispersal of White Spruce, Picea glauca, Seed JOHN C. ZASADA and DONALD LOVIG USDA Forest Service, Institute of Northern Forestry, Fairbanks, Alaska 99701 Zasada, John C., and Donald Lovig. 1983. Observations on primary dispersal of White Spruce, Picea glauca, seed. Canadian Field-Naturalist 97(1): 104-106. Primary dispersal of individual White Spruce (Picea glauca (Moench) Voss) seeds is described. Airborne time of seed accounted for 38% of the variation in the distance which filled seeds travelled from the parent tree. Observation of the flight paths of individual seeds indicated that seeds commonly attain heights greater than their cone of origin and that the distance of the landing point from the tree is not necessarily indicative of the distance traveled during primary dispersal. Key Words: primary seed dispersal, White Spruce, Picea glauca White Spruce (Picea glauca (Moench) Voss) depends on seed to naturally colonize disturbed sites. The dispersal of seed occurs in two phases (Watkinson 1978). During primary (phase 1) dispersal, the seed travels from the parent tree to a site on the ground. During secondary (phase II) dispersal, the seed is moved along the ground surface by one or more agents. In the case of White Spruce, the greatest movement usually occurs during primary dispersal except when secondary dispersal occurs over snow or via water on flood plains. Dobbs (1976) and Schlesinger (1970) concluded that the dispersal pattern for White Spruce follows the general pattern observed for winged seeds. That 1s, total seedfall decreases rapidly with distance from the seed source. Harper (1977) showed that seed dispersal fora number of species with different seed shapes and sizes generally can be described by the inverse square or cube laws. Van der Pijl (1972) discussed principles of dispersal in higher plants in detail and presented information on maximum dispersal distances for some tree species. The methods used to study tree seed dispersal usu- ally involve the placement of seed traps at regular intervals away from a seed source (Schlesinger 1970; Dobbs 1976). Schlesinger used this basic idea and released known quantities of seed from a standard height under ambient, but well-documented, weather conditions. Dobbs (1976) observed a combination of primary and secondary dispersal where over-snow movement probably contributed to the dispersal patt- erns reported. This study examined the airborne path and time of individual seeds naturally dispersed from White Spruce cones. Although these observations are not conclusive, they provide additional insight regarding primary dispersal. Methods An individual tree located about midway along the eastern edge of a forest clearing was selected for observation. This tree was 18.5 m tall with cones in the top 3.1 m of the tree. The clearing was approximately 100 m wide and 200 m long, with the long axis oriented north and south. Seeds could not be followed distances greater than 100 m. A cone was selected for observation, and the first seed that dropped from the cone was followed until it either landed or disappeared from sight. The time airborne was measured witha stop watch. The landing point was marked and its distance measured. The seeds were caught and cut open to determine if they were filled. A map was made of the airborne path of several seeds. The flights of 50 seeds were actually observed and measured. This represented about one-half of the observations started. Some were not followed to their landing site, or were lost because they traveled up and out of sight, if the observer fell or was otherwise distracted. Linear regression was used to examine the relationships between airborne time and distance landed from the tree. Observations were made during a 5-hour period (1100 hrs to 1800 hrs). The day was cloudless, with a slight but variable breeze during the observation period. Mean air temperature was 18°C. Results and Discussion About 38% (r2= 0.38) of the variation in the distance traveled by filled seeds from the tree could be attrib- uted to the time the seed was airborne. For empty seeds, the r? value was 61%. The shortest horizontal distance of a seed traveling from the tree was 2 m (Figure 1). Four percent of the seeds traveled more than 100 m. Mean distance travelled was 33 + 24 m. Neither the distance traveled nor time airborne of 16% of the seeds could be determined because they were carried up and out of sight shortly after release from the cone. The shortest time airborne was 7 s; the long- est was 125 s and this seed was still in flight 100 m 1983 NOTES 105 100 ° * * 80 D x Le e ® p=) ® i 60 * filled seed 8 x empty seed c 3 ‘ % seed airborne at 100 m. 0 ral 40 20 0 20 40 60 80 100 120 140 Time (seconds) FIGURE I. Relationship of airborne time and distance seed landed from tree for the primary phase of seed dispersal in White Spruce. ° . : Parent t from the tree (Figure 1). Average time airborne was feline 38 + 21 s. The average rate of fall was 0.6 = 0.4 m/s (range 0.2 to 2.4). Schlesinger (1970) reported a rate of fall of 0.5 + 0.1 m/s for White Spruce seed in still air. The distance of a seed’s landing from a tree did not necessarily indicate its actual airborne distance tra- velled. During flight an individual seed can follow a very circuitous path. The extent and nature of horiz- ontal and vertical movement is determined by the wind and air turbulence encountered. An example is one seed that travelled about 121 m yet landed only ’ Legend 13 m from the tree. During dispersal, this seed expe- a : Flight of seed 1 rienced two periods of upward and three of downward yy Cina ors ceeas s umeral indicates distance on that flight (Figure 2). A leg of flight; U or D indicates up or The height reached by seeds traveling up and out of fs See Clie securing sight shortly after leaving the cone could not be determined. These observations, however, indicate FIGURE 2. Flight path of two White Spruce seeds that the greatest height attained by a White Spruce during primary dispersal (seed | landed seed may exceed the height of the cone from which it 13 m from tree and was in flight 71 s; seed came. These seeds have the potential for longer dis- 11 > 100 m and > 125s). 106 persal then seeds which fall directly to the soil surface during primary dispersal. Literature Cited Dobbs, R. C. 1976. White spruce seed dispersal in central British Columbia. Forestry Chronicle 52: 225-228. Harper, J. L. 1977. Population biology of plants. Aca- demic Press. London. 892 pp. Schlesinger, R. C. 1970. Diffusion models applied to seed THE CANADIAN FIELD-NATURALIST Vol. 97 dispersal. Ph.D. Thesis. Syracuse University College of Forestry. Syracuse, N.Y. 103 pp. Watkinson, A. R. 1978. The demography of a sand dune annual: Vulpia fasciculata. 1] The dispersal of seeds. Jour- nal of Ecology 66: 483-498. Van der Pijl, L. 1972. Principles of dispersal in higher plants. Springer-Verlag. New York. 161 pp. Received 20 February 1981 Accepted 30 July 1982 A Golden-yellow Colored Sablefish, Anoplopoma fimbria, Caught off Quatsino Sound, British Columbia D. A. NAGTEGAAL Department of Fisheries and Oceans, Resource Services Branch, Pacific Biological Station, Nanaimo, B.C.V9R 5K6 Nagtegaal, D. A. 1983. A golden-yellow colored Sablefish. Anoplopoma fimbria, caught off Quatsino Sound, British Columbia. Canadian Field-Naturalist 97(1): 106. A Sablefish, Anoplopoma fimbria, golden yellow in color, was captured south of Kains Island by the M/V VIKING SUNRISE on 27 May 1980. Key Words: Sablefish, yellow. On 27 May 1980, a 58cm (FL) golden-yellow colored Sablefish, Anoplopoma fimbria, was cap- tured 22.5 km due south of Kains Island. It was caught in Blackcod trap gear by the M/V VIKING SUNRISE at a depth of 550-730 m. The whole body of the specimen was yellow including the peritoneum and the underside of the gill cover, although the color of the iris of the eye was a normal black (Figure 1). Adult Sablefish are usually slaty black or greenish gray on the dorsal surface and light gray on the ventral surface (Hart 1973). Juveniles are generally lighter on the dorsal surface and sometimes whitish in color on the ventral surface. This specimen could be labelled a partial albino since the iris of the eye was the normal color instead of the pink color commonly associated with true albinos. True albinism is an inherited condi- tion in which the organism lacks an enzyme that 1s involved in the production of the dark pigment melanin (Villee and Dethier 1971) and affects the entire body color. The specimen is deposited in the British Columbia Provincial Museum (BCPM 980-509). The morphometric and meristic characters of the yellow colored specimen do not deviate from the mea- surements of a normal Sablefish as described by Hart (1973). Specifically, the diagnostic dorsal fin and fin ray counts are 18 and 16, respectively. Examination of the specimen revealed that the peculiar pigmentation was not due to a pathological condition (personal communication, Dr. Ed Bilinski, Technology Services Branch, Department of Fisheries and Oceans). The skin appeared normal in areas lacking the black pig- ment and the flesh was not stained in spite of the presence of yellow pigmentation on the lining of the gill covers and on the skin. It is possible this abnormal pigmentation has a genetic origin. One other yellow Sablefish has been recorded, off California (Phillips 1952), although it had some vesti- gial smudges of dusty black on the body. Phillips also recorded some “calico” Sablefish that had a mottled appearance of light gray and slaty black. I thank Captain Inge Noringseth of the M/ V VIK- ING SUNRISE for recovering the specimen. I am also grateful to Nev Venables, of the Pacific Biological Station, for his helpful assistance. Literature Cited Hart, J.L. 1973. Pacific Fishes of Canada. Fisheries Research Board of Canada Bulletin (180): 740 pp. Phillips, J. B. 1952. Yellow Sablefish (Black cod) taken in Monterey Bay. California Fish and Game. 38(1): 437-438. Villee, C. A., and V. G. Dethier. 1971. Biological Princi- ples and Processes. W. B. Saunders Co., Philadelphia: 1009 pp. Received 20 February 1981 Accepted 18 February 1983 1983 NOTES 107 Distribution of the Badger, Taxidea taxus, in Southwestern Ontario WAYNE M. LINTACK! and DENNIS R. VOIGT? 'Ontario Ministry of Natural Resources, Huronia District Office, Midhurst, Ontario LOL 1X0 2Ontario Ministry of Natural Resources, Wildlife Branch, Research Section, P.O. Box 50, Maple, Ontario LOJ 1E0 Ontario Ministry of Natural Resources, Wildlife Branch Contribution No. 81-13. Lintack, Wayne M., and Dennis R. Voigt. 1983. Distribution of the Badger, Taxidea taxus, in southwestern Ontario. Canadian Field-Naturalist 97(1): 107-109. Thirty records of Badgers ( Taxidea taxus) were obtained from southwestern Ontario from 1972 through 1980. These data suggest Badgers are more common than reported earlier. Key Words: Badger, Taxidea taxus, distribution, range records, Ontario. Saunders (1932) reported on the presence of the Badger (7axidea taxus) in Ontario near Grand Bend, Lambton County, in the late 1890s, but later they were considered extinct in southern Ontario by Cross and Dymond (1929). Snyder (1935) recorded a single Badger near Port Dover, Norfolk County, on 5 November 1934 and in 1948 the Royal Ontario Museum(ROM) received a Badger from Woodgreen, Mosa Township, Middlesex County (ROM No. 19 630). Bartlett (1955) reported five Badgers in Oxford Township, Kent County, and two in Aldbo- rough Township, Elgin County, from June 1953 through May 1954. On 9 May 1961, a 7.7-kg male Badger was found killed on a road near Morpeth, Howard Township, Kent County (ROM No. 31 412). These records suggest the Badger is only rarely encountered in southern Ontario. However, J. D. McCabe (personal communication), a farmer and predator control agent for the Ontario Ministry of Natural Resources (OMNR), reported to us that he caught and subsequently killed or released 20-25 adult and immature Badgers in the Counties of Elgin, Mid- dlesex, and Lambton during the past 25 years. We investigated these and other records. The purpose of this paper is to report the distribution of Badgers in southern Ontario based on records obtained from 1972 through 1980. Methods Occurrence of Badgers was investigated during the collection of more than 14 000 specimens of terrestrial furbearers from more than 200 trappers and hunters during a rabies research program. Staff from the Mammalogy Department of the ROM and 12 OMNR administrative districts were contacted for additional records. Records of badgers were entire animals, skulls or pelts observed by OMNR staff; sightings substantiated by OMNR staff with physical evidence; specimens in the collection of the ROM; accounts by trappers known by us to be reliable. Results Thirty records of Badgers were obtained that included 19 Badgers caught in traps set for Red Fox (Vulpes vulpes) or Raccoon (Procyon lotor); 5 killed on roads; 2 shot; | killed by clubbing; 2 sightings (involving a single Badger and | family group); and | caught by OMNR staff (B. Ellah personal communi- cation). Ellah also observed diggings made by Badgers when he investigated the two sightings. One of these sightings was made by a farmer who reported shoot- ing at two adult and three young Badgers. Ellah found five separate trails of Badgers leaving the sighting- location through a hay field. Nine of 19 trapped Badgers were caught in No. 220 or 330 Conibear traps set for Raccoons on trails, or at the entrances of ground burrows, baited cubbies or drainage tiles. One Badger was snared and nine were caught in leg-hold traps set for Red Fox. The sex ratio of the 18 Badgers that were sexed was 50:50. Badgers were recorded in Waterloo and Haldimand-Norfolk Regional Munici- palities and the Counties of Kent, Lambton, Mid- dlesex, and Grey (Figure 1). Ontario specimens of Badgers Kent County 1. Harwich Township, 42°25’N, 82°05’W, 8/4/80, road kill. 2. Orford Township, 42°35’N, 81°50’W, 20/10/80, male, trapped. 3. Camden Township, 42°35’N, 82°05’W, 11/79, male, trapped. 4. Zone Township, 42°40’N, 81°55’W, 11/79, male, trapped. 5. Zone Township, 42°40’N, 81°55’W, 11/79, male, trapped. 6. Dover Township, 42°25’N, 82°20'W, 3/79, female, trapped. 7. Howard Township, 42° 30'N, 81°55’W, 9/9/78, killed by clubbing. 8. Orford Town- ship, 42°35’N, 81°50’W, 10-11/78, trapped. Middlesex County 9. Caradoc Township, 42°50’N, 81°30’W, 10/80, female, trapped. 10. Metcalfe Township, 42°50’N, 81°45’W. 10/80, road kill. 11. Caradoc Township, 108 THE CANADIAN FIELD-NATURALIST Vol. 97 Lake 3 Ontario Ay —_— MICHIGAN ¢ Recent record 4 Family group sighting K Kent County M Middlesex County HN Haldimand - Norfolk R.M. W Waterloo R.M. G Grey County FiGuRE 1. Records of the Badger in southwestern Ontario 1972-1980. Numbers correspond to specimens described in text. 1983 42°50’N, 81°30’W, 11/80, male, trapped. 12. Mosa Township, 42°45’N, 81°45’W, 11/80, female, trapped. 13. Ekfrid Township, 42°45’N, 81°40’W, 11/79, trapped. 14. Adelaide Township, 43°00’N, 81°40’W, 11/79, female, trapped. 15. Caradoc Township, 42°50’/N, 81°30’W, 7-8/78, shot. 16. Delaware Town- ship, -42°50’N, 81°20’W, 10/78, male, trapped. 17. Westminster Township, 42°55’N, 81° 15’W, 18/11/78, male, trapped. 18. Lobo Township, 43°00’N, 81°25’W, 10-11/77, female, trapped. 19. Lobo Town- ship, 43°00’N, 81°25’W, 10-11/77, female, trapped. 20. Delaware Township, 42°50’N, 81°20’W, 10/76, female, trapped. Grey County 21. Bentinck Township, 44° 15’N, 80°55’W, 11/80, male, trapped. Haldimana- Norfolk Regional Municipality 22. Town of Dunnville, 42°55’N, 79°40’W, 24/11/80, female, trapped. 23. Town of Simcoe, 42°50'N, 80° 15° W, 6/78, sighting of 5. 24. Delhi Township, 42° 50'N, 80° 25’W, 6-7/77, sighting of I. 25. Delhi Township, 42° 50’N, 80° 25’W, 6-7/77, road kill. 26. Delhi Township, 42°50'N, 80° 25’W, 6-7/77, caught and released. 27. Delhi Township, 42° 50’N, 80° 25’ W, 6-7/77, road kill. 28. Delhi Township, 42°50'N, 80° 25’ W, 6-7/72, road kill. Regional Municipality of Waterloo 29. City of Kitchener, 43°25’N, 80° 30’W, 26/8/79, male, shot. Lambton County 30. Euphemia Township, 42°45’N, 81°55’W, 20/11/79, female, trapped. Discussion Our data suggest Badgers are found in southwest- ern Ontario near the north shore of Lage Erie. Distri- bution was similar to that suggested by Peterson (1966) and Banfield (1974). Most records were in Kent or Middlesex County and Haldimand-Norfolk Regional Municipality in areas with sandy or sandy loam soils. Eight additional records in Kent and Middlesex Counties were reported by staff of the OMNR Chat- ham District from 1959 through 1978 (Kuja and McKay-Kuja, University of Toronto, unpublished data). The ROM has a record (ROM No. 27 559) of a Badger collected in 1956 from South Gower Town- ship, Grenville County, in eastern Ontario, but we were unable to obtain any recent evidence of Badgers in that area. The records of male Badgers from Grey County and Waterloo Regional Municipality were extra-limital for Ontario. These Badgers may have dispersed from more southern areas of Ontario. Messick and Hor- nocker (1981) reported that most young Badgers in NOTES 109 Idaho disperse during their first summer for distances of up to 110 km. In recent years, higher fur prices and a renewed interest in trapping has increased the harvest of Red Fox and Raccoon in southern Ontario which may have increased the number of incidental Badger catches. Only three of the 16 interviewed trappers had caught more than one Badger. Trappers that we inter- viewed recognized the rarity of a Badger catch and expressed reservations about killing captured Badgers. The pugnacious behaviour of Badgers (Ban- field 1974) makes the release of uninjured individuals a difficult task for unprepared trappers. However, Badger specimens were popular as collection items; only six of 18 marketable Badger pelts were offered for sale at fur auctions. The Badger is now classified as a furbearer in Onta- rio under the Ontario Game and Fish Act. The harvest season is from 15 October to 31 December and royal- ties are levied on all pelts which must also be sealed prior to sale. Our data suggest Badgers may be encountered in most counties in southwestern Onta- rio. Future changes in their distribution can be moni- tored from the harvest statistics of the OMNR. Acknowledgments The assistance of B. Ellah, A. Woodliff, E. Liver- more, D. Smith, 1. Watt and many staff of the OMNR and the Mammalogy Department of the Royal Onta- rio Museum is gratefully acknowledged. J. McCabe, W. Huber, K. Puyenbroek, B. Johnson, W. Reaume, C. Keech, J. Vloet, J. Saliniewicz, D. Butler, J. Van Every, D. Gailing, E. Bennet, S. Haggith, D. Wilson, G. Alderson, M. Myers, E. Vandenbroecke and D. Emke provided information on badgers. C. MacInnes reviewed the manuscript typed by E. Brolly. Literature Cited Banfield, A. W. F. 1974. The mammals of Canada. Univer- sity of Toronto Press, Toronto. 438 pp. Bartlett, C. O. 1955. Badgers in Kent and Elgin Counties, Ontario. Canadian Field-Naturalist 69:12-13. Cross, E. C., and J. R. Dymond. 1929. The mammals of Ontario. Royal Ontario Museum Zoology Handbook 1. University of Toronto Press, Toronto. 52 pp. Messick, J. P., and M. G. Hornocker. 1981. Ecology of the badger in southwestern Idaho. Wildlife Monograph 76:1-53. Peterson, R. L. 1966. The mammals of eastern Canada. Oxford University Press, Toronto. 438 pp. Saunders, W. E. 1932. Notes onthe Mammals of Ontario. Transaction Royal Canadian Institute XVIII, Part 2(40):286. Snyder, L. L. 1935. A badger specimen from Port Dover, Norfolk County, Ontario. Canadian Field-Naturalist 49:136-137. Received 17 September 1981 Accepted 3 August 1982 110 THE CANADIAN FIELD-NATURALIST Vol. 97 Responses of Two Groups of Mountain Goats, Oreamnos americanus, toa Wolf, Canis lupus CHRISTIAN A. SMITH Alaska Department of Fish and Game, Ketchikan, Alaska 99901 Smith, Christian A. 1983. Responses of two groups of Mountain Goats, Oreamnos americanus, to a Wolf, Canis lupus. Canadian Field-Naturalist 97(1): 110. The reaction of two small groups of Mountain Goats, Oreamnos americanus, to the presence of a Wolf, Canis lupus, illustrates the importance and effectiveness of rocky terrain to the goats, and different responses in different situations. Key Words: Mountain Goats, Oreamnos americanus, Wolf, Canis lupus, responses, habitat. The Rupicaprini, including Mountain Goats (Oreamnos americanus), are well adapted to the exploitation of steep, rugged terrain, and their anti- predator strategies rely on the distinct physical disad- vantages of most mammalian predators in such habi- tat (Schaller 1979). A recent observation of the interactions of two groups of goats and a Wolf (Canis lupus) illustrates the importance of escape terrain to this ungulate species. On 24 September 1981, two groups of Mountain Goats were being observed ona ridge in south coastal Alaska in the vicinity of Boca de Quadra, 55° 20’ N, 130° 30’ W. One group consisted of two adult females, two kids and one subadult (sex unknown), all of which were bedded in an area of broken rock interspersed with alpine vegetation on a 15-20° slope at 1100 m elevation 20 m below the ridge top. The second group consisted of a pair of adult females, each accompanied by a kid, and bedded on a smooth 30-35° slope of alpine vegetation 100 m below and 100-150 m south of the former group. About 50 m farther south was a steep rock outcrop, slope 45-50° approximately 30 m wide and 10-20 m high. At 0942 a single adult Wolf trotted over the crest of the ridge downwind and 50 m south of the first group. The instant the Wolf appeared, the upper group bolted north along the ridge, remaining below the ridge crest in rocky terrain. The Wolf initially chased the goats, but after 100-150 mit ceased the pursuit. All five goats continued to flee at a panicked gallop, angled up along the slope and traversed 800-900 m before they crossed over the ridge out of sight onto an extremely sheer rock face. In contrast to the first goats, the lower group rose quickly from their beds and walked at a deliberate but unhurried pace to the nearby rock outcrop. One pair climbed onto a narrow ledge approximately 5 m above the base and near the middle of the outcrop, making an approach by the Wolf difficult, if not impossible. The other pair moved to the base of the outcrop and backed up against a vertical rock face, thus preventing any flanking approach. When the Wolf turned from the aborted chase, it briefly observed the four lower goats as they gained their positions on the rock outcrop. It then loped to the south along the ridge crest, making no attempt to approach the lower goats. It continued along the ridge until it became aware of my presence, then descended the slope into the subalpine forest and disappeared. The different reactions of these two groups of goats may have been due to their relative positions when the’ Wolf appeared. The upper goats could have been effectively intercepted by the Wolf if they ran toward the rock outcrop. Their response was to flee to the nearest steep terrain, which was almost 900 m away. The lower group was able to reach adequate escape terrain before the Wolf could initiate an attack. From an energy standpoint, the response of the lower group was obviously advantageous. This may explain why McFetridge (1977) reported 95% of all observations of nursery groups of goats in Alberta as being within 412 m of escape terrain and why studies in Alaska indicate that distance to cliffs is the single most important factor determining habitat use by goats (Schoen and Kirchoff 1982; author’s unpubl. data). Literature Cited MeFetridge, R. J. 1977. Strategy of resource use by moun- tain goat nursery groups. /n Proceeding of the First Inter- national Mountain Goat Symposium. Edited by W. Samuel and W. G. MacGregor. British Columbia Fish and Wildlife Branch, Victoria, B.C. pp. 169-173. Schaller, G. B. 1979. Mountain Monarchs. University of Chicago Press, Chicago. 425 pp. Schoen, J. W., and M. D. Kirchoff. 1982. Habitat use by mountain goats in southeast Alaska. Alaska Department Fish and Game. Juneau. 67 pp. Received 28 February 1982 Accepted | September 1982 1983 NOTES 111 Occurrences of the Black-legged Kittiwake, Rissa tridactyla, in the Prairie Provinces and North-central United States. ALAN R. SMITH! and E. KUYT? 1Canadian Wildlife Service, 115 Perimeter Road, Saskatoon, Saskatchewan S7N 0X4. 2Canadian Wildlife Service, Room 1000, 9942-108th Street, Edmonton, Alberta TSK 2J5. Smith, AlanR.,and E. Kuyt. 1983. Occurrences of the Black-legged Kittiwake, Rissa tridactyla, in the Prairie Provinces and North-central United States. Canadian Field-Naturalist 97(1): 111-113. Fifteen records of Black-legged Kittiwakes (Rissa tridactyla) in the Canadian prairie provinces and four north-central American states are summarized. Ten records were made in late fall or early winter, and 12 were of young of the year. It is postulated that inexperienced birds are carried inland by low pressure systems moving out of the north Pacific. Key Words: Black-legged Kittiwake, Rissa tridactyla, extralimital records. The Black-legged Kittiwake (Rissa tridactyla) is one of the most maritime of gulls. Its breeding distri- bution is circumpolar and strictly coastal: from the shores of the Polar Basin south to Nova Scotia and France in the Atlantic and to the Kurile and Aleutian Islands in the Pacific Ocean. During the winter it is pelagic, occurring in offshore waters south of New Jersey and to northwest Africa in the Atlantic and to Japan and Baja California in the Pacific Ocean (God- frey 1966). Its occurrence 1000 km from the nearest ocean in the Province of: Alberta must therefore be considered unusual. On 23 December 1979, during the Christmas Bird Count, we, with Duane Sept, found a dead Black- legged Kittiwake on a dyke at the Sundance Cooling Pond south of Wabamun Lake, Alberta (cf. Smith and Prach 1980). The bird was found frozen in a squatting position, in feather-perfect condition — strong circumstantial evidence that it had died recently and non-violently. The bird appeared to be extremely emaciated, suggesting that it had died of starvation. The skin of the bird, an immature, is in the collection of the Department of Zoology, University of Alberta (Catalogue Number UAMZ 5415). Our recovery of the kittiwake marks the third known occurrence of this species in Alberta. Records in other prairie provinces and nearby states are also few: there are no records in Saskatchewan, one in Manitoba, one in Idaho, three in Montana, two in North Dakota and five in South Dakota (Table 1). These 15 records form a non-random pattern; all but three were in fall or winter (mainly November or December), and only one whose age was described was in adult plumage (unlike many gulls, Black-legged Kittiwakes have already assumed adult plumage in their second winter, Bent 1921). The Wabamun Lake specimen was sent to Henri Ouellet of the National Museum of Natural Sciences for a subspecific determination. Ouellet (personal communication) referred the specimen to pollicaris of the Pacific Ocean but added that that “is a very poorly marked subspecies which. . . could not withstand the test of modern revision. . . on the basis of the criteria used to characterize it”. It is most likely that only the Manitoba record derived from Atlantic populations. Weather systems which could have carried these birds inland tend to move from west to east. One might speculate that kittiwakes which migrate south from their breeding grounds in November (Dementiev and Gladkov 1969; Bent 1921) are displaced by low pressure systems moy- ing east out of the north Pacific. Some inexperienced immatures lose their orientation and move inland whereas the more experienced adults manage to remain on course. Birds arriving on the Great Plains would find little open water in November and even less in December, and they would be attracted to hydroelectric sites such as Garrison Dam or thermal electric sites such as Wabamun Lake, where open water is artificially maintained throughout the winter. Munyer (1965) advanced a similar theory to explain the occurrence of the Ancient Murrelet (Synthlibo- ramphus antiquus) in the interior of North America. He stated that “most of the records are directly or indirectly associated with weather disturbances over the Pacific coast”. He added that “about two-thirds of the records are dated November or late October, pre- sumably months of great migratory activity”. Con- nelly and Gates (1981), in summarizing the inland occurrences of the Black-legged Kittiwake in the northwestern United States, found that one or more low pressure systems had moved across the Pacific Northwest within two weeks prior to each inland kit- tiwake record. This explanation is consistent with the recovery of the Wabamun specimen and of two other records. On 10 December 1979 a low pressure area in northern British Columbia produced strong westerly winds 112 THE CANADIAN FIELD-NATURALIST Vol. 97 TABLE |. Black-legged Kittiwake records for the prairie provinces and adjacent north-central states. Location Date Number-Age Type of Record Source ALBERTA Calgary 13 Nov. 1976 | immature found dying Pinel and Riddell 1977 Beaverhill Lake 17 July 1977 1 adult sighting R. Ebel, pers. - comm. 1980 Wabamum Lake 23 Dec. 1979 1 immature found dead Harris 1980 MANITOBA Churchill 11 June 1980 | unknown sighting Gollop 1980 IDAHO Butte County 13 Feb. 1980 | immature found dead Connelly and - Gates 1981 MONTANA Ninepipe NWR 27 Apr. 1977 | immature collected Rogers 1977 Fort Peck Dam 23-26 Dec. 1978 | immature photographed Serr 1979 Broadview Marsh June 1980 1 unknown sighting Serr 1980b NORTH DAKOTA Garrison Dam 14 Dec. 1979 — | immature sighting Serr 1980a 5 Jan. 1980 McLean and Ward 7 Sept. — 1 immature sighting Serr 1981 Counties 28 Nov. 1980 SOUTH DAKOTA Big Ben Dam 3-28 Nov. 1967 1 immature photographed Harris 1967 Fort Randall Dam 24 Nov. 1968 | immature sighting Rose 1969 Lake Herman 5 Dec. 1969 | immature sighting Harris 1970 Sturgis 20 Dec. 1969 | immature found dying Schroeder 1970 Camp Crook 13 Dec. 1979 | immature sighting Serr 1980a along the Pacific Coast (Atmospheric Environment Service 1979). On 13 December an immature Black- legged Kittiwake was seen at Camp Crook, South Dakota, and the next day one was seen at the Garrison Dam in North Dakota (Serr 1980a). On 23 December the Wabamun specimen was found. It seems likely that these birds were part of the same movement. Acknowledgments We thank J. Alway, D. A. Boag, H. Ouellet, E. Serr and P. D. Skaar for their assistance in the preparation of this note. We also thank D. V. Weseloh and an anonymous referee for their comments on the manuscript. Literature Cited Atmospheric Environment Service. 1979. Canadian Weath- er Review 17: 1-7. Bent, A. C. 1921. Life Histories of North American Gulls and Terns. U.S. National Museum Bulletin 113. Washing- ton, D.C. 337 pp. Connelly, J. W., and R. J. Gates. 1981. First Record of a Black-legged Kittiwake in Idaho. Condor 83: 272-273. Dementiev, G. P., and N. A. Gladkov. 1969. Birds of the Soviet Union. Vol. 3. Isreal Program for Scientific Trans- lation, Jerusalem. 756 pp. Godfrey, W.E. 1966. The birds of Canada. National Museum of Canada. Bulletin 203. 428 pp. Gollop, J. B. 1980. Prairie Provinces Region. /n The Nest- ing Season June I-July 31, 1980. American Birds 34: 282-264. Harris, B. K. 1967. The Black-legged Kittiwake in South Dakota. South Dakota Bird Notes 19: 76-77. Harris, B. K. 1970. Third Record for the Black-legged Kit- tiwake for North Dakota. South Dakota Bird Notes 22: 56. Harris, W. C. 1980. Prairie Provinces Region. /n The Win- ter Season December |, 1979-February 29, 1980. Ameri- can Birds 34: 282-284. Munyer, E. A. 1965. Inland wanderings of the Ancient Murrelet. Wilson Bulletin 77: 235-242. Pinel, H. W., and J. R. Riddell. 1977. The first record of Black-legged Kittiwake for Alberta. Blue Jay 35: 89. Rogers, T.H. 1977. Northern Rocky Mountain - Intermountain Region. /Jn The Spring Migration April 1-May 31, 1977. American Birds 31: 1024-1027. Rose, B. J. 1969. Another Black-legged Kittiwake in South Dakota. South Dakota Bird Notes 21: 92. Schroeder, J. R. 1970. First Black-legged Kittiwake Spec- imen for South Dakota. South Dakota Bird Notes 22: 81. Serr, E. M. 1979. Northern Great Plains. /n The Winter Season December |, 1978-February 28, 1979. American Birds 33: 291-292. 1983 Serr, E. M. 1980a. Northern Great Plains. Jn The Winter Season December |, 1979-February 29, 1980. American Birds 34: 284-285. Serr, E. M. 1980b. Northern Great Plains. Jn The Nesting Season June I-July 31, 1980. American Birds 34: 906-908. Serr, E.M. 1981. Northern Great Plains. /n The Autumn NOTES 113 Migration August 1-November 30, 1980. American Birds 35: 196-198. Smith, A. R., and R. W. Prach. 1980. Wintering Water- birds of Wabamun Lake. Blue Jay 38: 171-176. Received 9 February 1982 Accepted 14 September 1982 Breeding Records of Northern Shoveler, Anas clypeata, along the Northern Coast of Ontario R. KENYON ROSS! and NORMAN R. NORTH? 'Canadian Wildlife Service, 1725 Woodward Drive, Ottawa, Ontario K1A 0E7 2Canadian Wildlife Service, 152 Newbold Court, London, Ontario N6E 1Z7 Ross, R. Kenyon, and Norman R. North. 1983. Breeding records of Northern Shoveler, Anas clypeata, along the northern coast of Ontario. Canadian Field-Naturalist 97(1): 113. Breeding of Northern Shoveler on the Hudson Bay coast of Ontario was confirmed by the observation of two unfledged broods near the mouth of the Shagamu River. Records of broods, fledging status unknown, along the James Bay coast suggest breeding at low densities along that shore. Key Words: Northern Shoveler, Anas clypeata, northern Ontario, breeding. The Northern Shoveler (Anas clypeata) has been known to occur occasionally along the Ontario coasts of James and Hudson Bays (Manning 1952; Godfrey 1966), but its status has never been clarified. The near- est recorded breeding sites are as distant as Churchill, Manitoba (Jehl and Smith 1970), and Lilabelle Lake near Cochrane (D.G. Dennis, CWS, personal com- munication). In this note, we describe breeding records of the Shoveler from the Hudson Bay and James Bay coasts of Ontario. On 8 August 1977, we encountered a flying female Northern Shoveler with three fully-feathered but flightless young on a pond near the mouth of the Shagamu River (55°51’N, 86°46’W). A male(420 g) and a female (410 g) were caught and photographed. Copies of the photos have been deposited with the National Museum of Natural Sciences in Ottawa. Later that day, a second family group (a female and four juveniles) was noted in the same vicinity. Consid- ering the birds’ age and occurrence near a river mouth, they could have hatched some distance upstream. There are no confirmed breeding records of the Northern Shoveler along the James Bay coast although there are several records of broods with unknown powers of flight. R.H. Smith, in an unpub- lished report to the Canadian Wildlife Service, noted a brood of at least three young, apparently unfledged, on the Opinnagau River (54°12’N, 82°25’W) on 24 July 1944 and another brood of nine “well grown young” near Hook Point (54°52’N, 82°13’W) on 15 August 1944. Also, R.K. Ross observed a female with five flying young at Big Piskwamish Point (51°42’N, 80° 35’W) on 18 August 1976. All these sightings indi- cate that the Shoveler breeds in small numbers along the James Bay shore. Literature Cited Jehl, J. R., Jr., and B. A. Smith. 1970. Birds of the Chur- chill region, Manitoba. Special Publication Number |, Manitoba Museum of Man and Nature, Winnipeg. 87 pp. Godfrey, W.E. 1966. The birds of Canada. National Museum of Canada, Bulletin 203. 428 pp. Manning, T.H. 1952. Birds of the west James Bay and southern Hudson Bay coasts. National Museum of Can- ada, Bulletin 125. 144 pp. Received 7 April 1982 Accepted 12 September 1982 114 THE CANADIAN FIELD-NATURALIST Vol. 97 Black Bear, Ursus americanus, Predation ona Mule Deer Fawn, Odocoileus hemionus. E. VERSPOOR Faculty of Environmental Design, University of Calgary, Calgary, Alberta T2N 1N4 Verspoor, E. 1983. Black Bear, Ursus americanus, predation on a Mule Deer fawn, Odocoileus hemionus. Canadian Field-Naturalist 97(1): 114. A Black Bear (Ursus americanus) attacked and killed a Mule Deer fawn (Odecoileus hemionus) in the Porcupine Hills region in southwestern Alberta, 28 May 1980. As Black Bear rarely attack ungulates, extreme spring hunger is the probable cause of the reported incident. Key Words: Black Bear, Ursus americanus, predation, feeding behavior, Mule Deer, Odocoileus hemionus. Most of the evidence of predation by Black Bears (Ursus americanus) on ungulates (Hatler 1972; Ban- field 1974; Franzmann et al. 1980) is inferential. The only observation known to me was reported by Beh- rend and Sage (1974) of a Black Bear attacking and killing two White-tailed Deer (Odecoileus virginia- nus) while they were in a holding pen. Bacon and Burghardt (1976) do not even discuss predatory behavior in their paper on the ingestive behavior of the Black Bear. An attack by a Black Bear ona Mule Deer (Odecoileus hemionus) and her fawn which resulted in the capture and killing of the fawn is doc- umented here. The observations were made just before noon on 28 May 1980 in the Willow Creek drainage in the Porcu- pine Hills of southwestern Alberta at approximately 50°12’N, 114°01’W. A female Mule Deer and her fawn were observed to cross the head of a side valley off Willow Creek and to enter the aspen woods on the slope opposite. At the time they were headed south about 4 km north and upwind from the author and a friend. Shortly thereafter, cattle grazing in the aspen woods on the slope directly opposite were seen to move down into the valley bottom, vocalizing contin- uously. About 10 minutes later a Black Bear was seen chasing a Mule Deer and her fawn (presumably the deer observed previously) in the area vacated by the cattle. It was possible to observe the last 80-120 m of the chase after the animals emerged from the trees into the open. It ended with the bear knocking the fawn over and grabbing it by the neck with its jaws. The bear repeatedly bit the fawn on the head, neck, and back without killing it, and after a few minutes dragged the fawn by the neck into the woods out of sight. While being mauled and for about 15 minutes after being dragged from view, the fawn continued to give out distress calls. During the chase the female deer remained with the fawn; while the fawn was being mauled, it paced back and forth about 20-25 m away. After the bear dragged the fawn into the woods it disappeared from view. The bear weighed an estimated 100 kg, and seemed thin and poorly nourished. It had noticeably darker paws and snout, and a light, almost grey coat. The fawn was small and therefore quite young, perhaps recently born. The observations were made at the time of year when Mule Deer give birth (Banfield 1974). The success of the attack was probably fortuitous. The situation favored the bear as the deer approached upwind from it in the restricted visibility of the woods. These factors probably allowed the bear to go unno- ticed. The bear may have been prompted to attack because of hunger. This would be consistent with its appearance and with the view that bears are generally undernourished in the spring until their more usual and abundant foods such as berries and _ insects become readily available (Banfield 1974). Literature Cited Bacon, E. S., and G. M. Burghardt. 1976. Ingestive behav- ior of the American Black Bear. /n Bears — their Biology and Management, 3rd International Conference on Bear Research and Management. Binghampton, New York and Moscow, USSR, June 1974. Banfield, A. W. F. 1974. The Mammals of Canada. Uni- versity of Toronto Press, Toronto. Behrends, D. F. and R. W. Sage. 1974. Unusual feeding behavior by Black Bears. Journal of Wildlife Manage- ment. 38(3): 570. Franzmann, A. W., C.C. Swartz, and R.O. Peterson. 1980. Moose calf mortality in summer on the Kenai Peninsula Alaska. Journal of Wildlife Management. 44(3): 764-768. Hatler, D. F. 1972. Food habits of Black Bears in interior Alaska. Canadian Field-Naturalist 86(1): 17-31. Received 21 May 1982 Accepted 24 March 1983 1983 Sphagnum at Prudhoe Bay, Alaska PETER D. SPATT NOTES NS Institute of Polar Studies and Department of Botany, Ohio State University, Columbus, Ohio 43210. Spatt, Peter D. 1983. Sphagnum at Prudhoe Bay, Alaska. Canadian Field-Naturalist 97(1): 115-116. Four populations of Sphagnum fimbriatum Wils. ex. Hook, co-occurring with S. girgensohnii Russ., represent the first records of Sphagnum from Prudhoe Bay, Alaska. The highly calcareous substrates here are thought to limit Sphagnum, an acidophilic moss, in this part of the arctic coastal plains. Key Words: Sphagnum, Arctic, Alaska, alkaline tundra. The coastal plain at Prudhoe Bay, Alaska, has been surveyed extensively for its bryophyte flora (Murray and Murray, 1974, 1975; Rastorfer et al. 1973; Steere 1976, 1978; Walker et al. 1980). The area is essentially flat and composed mostly of wet tundra (Walker et al. 1980). The influence of the Sagavanirktok River makes it quite uniformly calcareous (Steere 1976). No Sphagnum species have ever been reported from Prudhoe Bay (B. Murray personal communication; Steere 1976, 1978, personal communication) and this has been attributed to the calcareous nature of the substrate (Steere 1976). The occurrences of small, local deposits of acid peat, however, have been noted (Steere 1976). In August, 1981, Sphagnum was observed and col- lected in two localities along the western margin of Prudhoe Bay, in sites considered to be relatively acidic (Everett, personal communication; Walker et al. 1980). Collection Site I. 18 July 1981. A single, circular (70cm diameter) mat of Sphagnum fimbriatum (Spatt #450-453; allin ALA) raised 10-12 cm from the surrounding wet, low polygon tundra, was located approximately 0.5 km west of the West Dock Storage Pad and approximately 0.5 km south of the Beaufort Sea. Within and surrounding the mat were Eriopho- rum spp., Petasites frigidis (L) Fries, and Salix rotun- difolia Trautv. Collection Site II. 19 August 1981. Mixed popula- tions of S. fimbriatum and S. girgensohnii were observed approximately 2 km south of Site |. Popula- tion A (Spatt #376- S. girgensohnii, 377- S. fimbria- tum; ALA) was located along the margin of a pond on the side of a raised polygon. Other plants associated with this population were A/lectoria sp., Carex aquati- lis Wahl., Cassiope tetragona (L) D. Don., Cladonia sp., Dactylina arctica (Hook.) Nyl., Dryas integrifolia Vahl., Peltigera sp., Salix pulchra Cham., Salix retic- ulata L, and Thamnolia sp. Populations B (Spatt #382, 389- S. fimbriatum, 386- S. girgensohnii; ALA) and C(Spatt #390- S. fimbriatum; ALA) were located on a polygon ridge, with similar associated vascular and non-vascular plant species. Sphagnum fimbriatum and S. girgensohnii are closely related wide ranging, circumpolar species. In North America they occur from the arctic south to the northern part of the continental United States (Andrus 1974). Both species prefer minerotrophic habitats and are commonly found in mineral soils of bogs and fens (Andrus 1974). S. fimbriatum has been previously reported from Barrow, Meade River, Navagapak Point, and Putu Camp on the Alaskan arctic coastal plains, and S. girgensohnii was reported from Barrow and Okpilak Lake (Steere 1978). The collections at Prudhoe Bay extend their ranges along the northern Alaskan coast and enhance understand- ing of the soils and flora of the region. The importance of Sphagnum spp. in the low arctic, a region in which Prudhoe Bay 1s located, is greater than in other arctic regions, especially the high arctic (Vitt and Pakarinen 1977). Other populations of Sphagnum are likely to be found at Prudhoe Bay, especially in the area west of West Dock. This area has mildly to moderately acidic soils (Walker et al. 1980). Sphagnum is least likely to occur directly in the Sagavanirktok River delta or floodplain due to their highly alkaline substrates. Acknowledgments ] thank Barbara Murray, Jerry Snider, Roland Seymour, and Lee Klinger for their helpful comments, and Howard Crum for verifying the Sphagnum identi- fications. Literature Cited Andrus, R. E. 1974. The Sphagna of New York State. Ph.D. Dissertation, SUNY College of Environmental Science and Forestry, Syracuse. University Microfilms, Ann Arbor. 421 pp. Murray, B. M., and D. F. Murray. 1973. Checklists to he Flora of Alaskan U.S. I.B.P. Tundra Biome study sites. U.S. Tundra Biome Data Report 73-30. U.S. Interna- tional Biological Program, Washington. 104 pp. 116 Murray, B.M., and D.F. Murray. 1975. Provisional checklist to the vascular, bryophyte, and lichen flora of Prudhoe Bay, Alaska. Jn Ecological Investigations of the Tundra Biome in the Prudhoe Bay Region, Alaska. Edited by J. Brown. Biological Papers of the University of Alaska. Special Report no. 2. 215 pp. Rastorfer, J. R., H. J. Webster, and D. K. Smith. 1973. Floristic and ecologic studies of bryophytes of selected habitats at Prudhoe Bay, Alaska. Institute of Polar Stu- dies, Ohio State University, Report 49. 20 pp. Steere, W. C. 1976. Symposium II. Geography and ecology of bryophytes. Ecology, phytogeography, and floristics of arctic Alaskan bryophytes. Journal of the Hattori Botani- cal Laboratory 41: 47-72. THE CANADIAN FIELD-NATURALIST Vol. 97 Steere, W.C. 1978. The mosses of arctic Alaska. Bryo- phytorum Bibliotheca 14: 1-508. Vitt, D. H., and P. Pakarinen. 1977. The bryophyte vegeta- tion, production, and organic components of Truelove Lowland. Pp. 225-244 In Truelove Lowland, Devon Island, Canada: A High Arctic Ecosystem. Edited by L. C. Bliss. University of Alberta Press, Edmonton. 714 pp. Walker, D. A., K. R. Everett, P. J. Webber, and J. Brown. 1980. Geobotanical atlas of the Prudhoe Bay region, Alaska. CRREL Report 80-14. U.S. Army Corps of Engineers, Cold Regions Research and Engineering Laboratory. Hanover, N.H. 69 pp. Received | April 1982 Accepted 22 February 1983 Northern Range Extension of the Two-lined Salamander, Eurycea bislineata, in Ontario JAMES KAMSTRA R.R. #5, Oshawa, Ontario LIH 7K5 Kamstra, James. 1983. Northern range extension of the Two-lined Salamander, Eurycea b. bislineata, in Ontario. Canadian Field-Naturalist 97(1): 116. A Two-lined Salamander, Eurycea bislineata, was observed at the Onakawana River, 400 km north of the previous northernmost record in Ontario. Key Words: Two-lined Salamander, Eurycea bislineata, northern range, Ontario. On September 4, 1977, a single adult Two-lined Salamander (Eurycea bislineata) was found along the Onakawana River (50° 37’N, 81° 25’W) in Morrow Township of the Cochrane District, northern Ontario. This salamander was observed under unusual circum- stances; it emerged from under a large log against which a fire had been built. The log was situated ona gravel substrate about 0.5 meters from a moderately fast flowing river six meters wide. The surrounding forest consisted mainly of Balsam Fir (Abies balsa- mea) and Black Spruce (Picea mariana). The individual was about 8 cm in length, dull yellowish-brown dorsally with a broad dark lateral Stripe. Its belly was bright yellow. There was no pre- serving material available so the specimen was released after being photographed. Identification of the photo was confirmed by Dr. Francis R. Cook, Curator of Herpetology, National Museum of Natu- ral Sciences, Ottawa. This locality, only 100 km south of James Bay, is 400 km north of Lake Temagami, the closest record of E. bislineata cited by Logier and Toner (1961). The next nearest records are from Quebec (Cook and Pres- ton 1979): Lac Charon, 500 km ESE, and Rupert River, 600 km ENE of the Onakawana site. There is also a recent northern range extension into Labrador (Cook and Preston 1979.). The Labrador and Onakawana records of Two- lined Salamander extend the range delineated by Conant (1975) in both a northeast and northwest direction, suggesting a more widespread distribution in the boreal forest of eastern Canada. Literature Cited Conant, Roger. 1975. A Field Guide to Reptiles and Amphibians of Eastern and Central North America. Houghton Mifflin Co., Boston. 429 pp. Cook, Francis R., and John Preston. 1979. Two-lined sala- mander Eurycea bislineata in Labrador. Canadian Field- Naturalist 93(2): 178-179. Logier, KE. B. S., and G. C. Toner. 1961. Checklist of the amphibians and reptiles of Canada and Alaska. Royal Ontario Museum, Toronto. Life Sciences Division Con- tribution 53: 1-92. Received 3 December 1981 Accepted 21 September 1982 1983 NOTES i 7/ A Second Record of the Deer Mouse, Peromyscus maniculatus, from Newfoundland M. C. BATEMAN Canadian Wildlife Service, Sackville, New Brunswick E0A 3C0 Bateman, M. C. 1983. A second record of the Deer Mouse, Peromyscus maniculatus, from Newfoundland. Canadian Field-Naturalist 97(1): 117. Four Deer Mice, Peromyscus maniculatus, were collected from the Abitibi-Price Inc. Woods Camp, Southwest Brook, Newfoundland. The only previous record of the Deer Mouse on insular Newfoundland is 100 km southwest of this location. Key Words: Deer Mouse, Peromyscus maniculatus, Newfoundland, second record. Four Deer Mice (Peromyscus maniculatus) were collected at the Abitibi-Price Inc. Woods Camp, Southwest Brook, Newfoundland (48° 29’N, 58° 0l’W) in early December 1981. The only previous record of the Deer Mouse on the Island of Newfoundland is from Broom’s Brook in the Anguille Mountains (47° 56’N, 59° 10’W) (Gould and Pruitt 1969). Broom’s Brook is approximately 100 km southwest of the present capture site. A total of 5,900 trap nights (Museum Special and Victor snaptraps) from 29 August to 9 September 1980 and 31 August to 9 September 1981 between 2 and 27 km from the camp failed to catch any species other than Meadow Vole (Microtus pennsylvanicus), Masked Shrew (Sorex cinereus) and Eastern Chip- munk (Tamias striatus). Most trapping was done in mature Balsam Fir (Abies balsamea), Balsam Fir — White Birch (Betula papyrifera), and Black Spruce (Picea mariana) forest, although cutover (regenerat- ing Balsam Fir and White Birch) and softwood scrub were also sampled. The four specimens, trapped by company foremen after damaging supplies, were kept frozen until study skins were prepared in the Canadian Wildlife Service laboratory in Sackville, New Brunswick. Two speci- mens are in the Canadian Wildlife Service mammal collection at Sackville and two are at the National Museum of Natural Sciences, Ottawa (NMC 45720, 45721). Identifications were confirmed by Dr. van Zyll de Jong, Curator of Mammals, National Museum of Natural Sciences, Ottawa. Acknowledgments Specimens were collected during field work funded by Parks Canada, Atlantic Region. I thank Abitibi- Price Inc. for providing accommodation and freezer space, and G. Belyea for preparation of the study skins. Literature Cited Gould, W. P., and W. O. Pruitt, Jr. 1969. First Newfound- land record of Peromyscus. Canadian Journal of Zoology 47: 469. Received 22 March 1982 Accepted 22 July 1982 118 THE CANADIAN FIELD-NATURALIST Vol. 97 Dwarf Clearweed, Pilea pumila (Urticaceae) — New to Nova Scotia N. HILL and S. P. VANDER KLOET Department of Biology, Acadia University, Wolfville, N.S. BOP 1X0 Hill, N., and S. P. Vander Kloet. 1983. Dwarf Clearweed, Pilea pumila (Urticaceae) — new to Nova Scotia. Canadian Field-Naturalist 97(1): 118. We have documented the occurrence of Dwarf Clearweed, Pilea pumila (L.) Gray in the Cobequid Mountains of Nova Scotia. Key Words: Pilea pumila, distribution, Nova Scotia. On the 17 October 1980, seven fruiting specimens of Dwarf Clearweed, Pilea pumila (L.) Gray were disco- vered on a seepage slope in the rich maple-beech woods at West Branch, Pictou County, Nova Scotia (43°37’N, 63°04’W). Macoun (1884) listed several locations in New Brunswick and noted it was common in western Quebec and Ontario. Boivin in 1966 reported findings from Prince Edward Island. In their monographic treatment of the family Urticaceae for Canada, Bassett et al. (1974) do not record P. pumila for Nova Scotia. Our discovery, therefore, extends the range of P. pumila into Nova Scotia. Two plants were collected and have been deposited in the E. C. Smith Herbarium at Acadia University (ACAD). Pilea pumila is an annual and depends on seeds to survive the winter. We combed the area from May to October in the following year but found no plants. It appears the population failed to maintain itself but it may persist dormant for several years in the soil. Thompson and Grime (1979) predicted non- persistent seed banks would occur in these areas where plant growth was curtailed by low winter tempera- tures and where vegetational gaps stayed open until spring. Thirty 6 cm x 5 cmsoil cores were taken along the seepage slopes during July 1981, spread out in trays over sterilized sand, placed in the greenhouse and watered daily. After one year, no Pilea pumila, no Impatient biflora Walt. (Jewel Weed), nor any Cir- caea alpina L. (Enchanter’s Nightshade) had germi- nated although the latter annuals are common along these seepage slopes. The bulk of the seedlings in the trays were sedges (Carex spp) and rushes (Juncus spp), and on the basis of core size and number, it is estimated that Juncus spp had 9286 Seeds/m? and Carex spp. had 2833 seeds/m-?. Literature Cited Bassett, I.J.. C.W. Crompton, and D.W. Wood- land. 1974. The family Urticaceae in Canada. Canadian Journal of Botany 52: 503-516. Boivin, B. 1966. Enumeration des plantes du Canada (Suite). Le Naturaliste Canadien 93(4): 404. Macoun, J. 1884. Catalogue of Canadian Plants. Part I] — Gamopetale. Dawson Brothers, Montreal. 248 pp. Thompson, K., and J. P. Grime. 1979. Seasonal variation in the seed banks of herbaceous species in ten contrasting habitats. Journal-of Ecology 67(3): 893-922. Received 22 April 1982 Accepted 5 August 1982 News and Comment The American Association for The Advancement of Science: Arctic Division The 34th Alaska Science Conference, with the theme: Alaska/ Canada North: Neighbours in Science, will be held in Whitehorse, Yukon, September 28-October 1, 1983 First Announcement and Call for Papers In 1982 the AAAS — Alaska Division changed its name to AAAS -— Arctic Division in order to acknowledge the continual participation in the Alaska Science Conference of researchers from Yukon and Northwest Territories. The 1983 Conference will follow-up on this initiative by returning its activities to Whitehorse for the first time since 1968. The program focus is intended to be very general while still reflecting transboundary topics in Northern Science. Technical and poster sessions in the broad area of scientific and technical research concerning northern people, the northern lands and marine environment, and resources development and man- agement will be offered. Plenary Sessions will include one which highlights current research of special inter- est to the general public. Projected Special Events and Symposia: Caribou and Human Activity Yukon River Basin Studies Position Available October Migration Program Manager (Biologist). $11 000- 12 750. Responsible for all field operations: banding, maintenance, boats, surveys; data analysis. Qualifica- tions: banding eligibility, proven identification and banding competence, mechanical aptitude, supervi- sory ability, etc. Resume and three referees to Execu- tive Director, from whom fuller advertisement available. ARTHUR N. LANGFORD Executive Director Long Point Bird Observatory P.O. Box 160 Port Rowan, Ontario NOE 1MO Yukon Heritage and Museums Association 6th Annual Fall Conference Scientific contributions of early Yukon-Alaska Explorers September 30- October 2. Science and Frontier Hydrocarbon Exploration — The Beaufort Sea Experience Human Performance in the Cold Ecological Land Classification Wood as Fuel — Uses and Problems Call For Papers We plan to prepublish abstracts of the papers to be presented prior to the Conference. Authors wishing to present papers in the technical sessions should provide titles by May 15, 1983, and abstracts by July 15, 1983. Abstracts should not be longer than one double- spaced typewritten page with the title, author’s name, and affiliation at the top of the page. Those interested in the poster sessions should provide titles by July 15, 1983. All correspondence should be addressed to: Art Pearson Conference Chairman Box 4580 Whitehorse, Yukon YIA 2R8 (403) 667-4288 Notice of Publication The second part of the Atlas of the Rare Vascular Plants of Ontario, edited by G. W. Argus and D. J. White, will be published in July 1983 by the National Museum of Natural Sciences. It treats about 100 spe- cies in several important families including the ferns, Asteraceae, Ranunculaceae, and Scrophulariaceae. The Atlas is available free-of-charge from: The Rare and Endangered Plants Project Botany Division National Museum of Natural Sciences Ottawa, Ontario KIA 0M8 The recipients of Part | of the Atlas will automati- cally be sent this and subsequent parts, but if one does not arrive ina reasonable length of time, the Museum ‘should be contacted. G. W. ARGUS 119 Gamma-Ray Irradiation of a Boreal Forest Ecosystem: The Field Irradiator - Gamma (FIG) Facility and Research Programs JOHN E. GUTHRIE and JANET R. DUGLE Environmental Research Branch, Whiteshell Nuclear Research Establishment, Atomic Energy of Canada Limited, Pinawa, Manitoba ROE 1 LO Guthrie, John E., and Janet R. Dugle. 1983. Gamma-ray irradiation of a boreal forest ecosystem: the Field Irradiator— Gamma (FIG) facility and research programs. Canadian Field-Naturalist 97(1): 120-128. A long-term radiation ecology research project called Field Irradiator - Gamma (FIG) began at the Whiteshell Nuclear Research Establishment in 1968. The experimental area is in southeastern Manitoba and 1s located on the western edge of the Precambrian shield. The project studies the ecological effects continuous exposure to a gradient of gamma radiation has ona mixed boreal forest ecosystem. The gradient ranges from | to 460 000 times the natural background radiation level. This paper describes the forest, the gamma irradiator and its radiation field, and the research programs. A long-term radiation ecology project called FIG, an acronym for Field Irradiator - Gamma, began in 1968 at the Whiteshell Nuclear Research Establish- ment (WNRE). WNRE is located in southeastern Manitoba, about 115 km from Winnipeg. The FIG project studies the ecological effects of continuous long-term exposure to gamma radiation on a mixed boreal forest ecosystem. The investigations of the effects of exposure to lonizing radiation on individual organisms and popu- lations under laboratory conditions are numerous. The number of radiation ecology studies carried out in the field, however, is much smaller. Since the late fifties, several research projects have investigated the effects of ionizing radiation on plant and animal communities in natural ecosystems: an Oak-Pine forest (Woodwell 1963), a Mediterranean forest (Fabries et al. 1972), a desert (French et al. 1974.), a montane tropical rain forest (Odum and Pigeon 1970), a northern lake forest (Rudolph 1974; Zavitkovski 1977), anda pine plantation and old field (Platt 1963). One objective these projects share is to determine the amount natural background radiation must be increased to cause demonstrable ecological effects on free-living organisms, communities and ecosystems. The effects on boreal forests of continuous expo- sure to gamma radiation have not, to our knowledge, been previously investigated. The Boreal Forest is an important biome in Canada since it occupies about one-third of our land mass. The studies previously cited may be only indirectly applicable to Canada, because they involved different groups of plants and animals exposed to climatic conditions not typical of Canada. The first phase of the FIG project (pre-irradiation) started in 1968 and continued until 2 March 1973 when Phase II (irradiation) began. The recovery, or post-irradiation, phase will conclude the project. This paper describes the experimental forest, the gamma irradiator and its radiation field, and the various research programs. The FIG Experimental Area General The FIG area (Figure |) is on the property of the Whiteshell Nuclear Research Establishment (WNREB), at 50° 11’N, 96° 1’W and at 270 mabove sea level. Established as an ecological preserve in 1968, it is | kmin diameter, and surrounded by a 2.3 mchain- link fence and a 30m wide firebreak. The only entrance into the area 1s a gate adjacent to the irradia- tor control building. The area encompasses an ecotonal mixed boreal forest described as Lower English River type (Rowe 1972). Part of the area still shows the effects of logging carried out about 50 years ago, of forest fires as recent as 1961, and of more recent gravel-digging operations, which produced small marshes. The area also includes a Black Spruce-Sphagnum bog that drains from SS W to NNE. Beginning in 1968, extensive cataloging of the plants, small mammals and forest insects indigen- ous to the FIG area was carried out until the irradiator (Figure 2) was moved into the area and irradiation began in March of 1973. Climate and Meteorology The climate of FIG is that of the transition zone between temperate continental and boreal climates (Trewartha 1954.). The average annual precipitation in the area is 56 cm (38-50 cm as rain and about 130 cm as snow!'). The long-term average extreme maximum and minimum temperatures are shown in Figure 3 (Davis and Reimer 1980). Meteorological data collected in FIG include air temperature, dewpoint, humidity, wind speed, soil 'l cm snow ~ | mm rain 120 1983 GUTHRIE AND DUGLE: FIG FACILITY AND RESEARCH 121 FiGureE |. Field Irradiator - Gamma (FIG) area. This mixed boreal forest, 1000 m in diameter and surrounded by a firebreak, is located in southeastern Manitoba. Irradiation has continued at 19 hours per day since 1973 March. heat flux, net solar radiation, and precipitation (Rei- mer and Desmarais 1973). Flora and Fauna There are three general soil moisture regimes in the FIG area (Dugle 1972). The wert regime covers approximately 40% of the area of which Picea mari- ana (Black Spruce) — Larix laricina (Tamarack) — Sphagnum is the major association and Black Spruce the dominant tree. The very moist regime covers about 20% of the area. The major association in this area is Abies balsamea (Balsam Fir) — Picea mariana and P. glauca (White Spruce) — Ledum groenlandi- cum (Labrador Tea) with Balsam Fir as the dominant tree. The remaining 40% of FIG area is a moist regime supporting a mixed boreal forest association of Popu- lus tremuloides (Trembling Aspen), the dominant tree, and Betula papyrifera (Paper Birch), smaller stands of Pinus banksiana (Jack Pine) and Fraxinus nigra (Black Ash) and scattered individuals of Balsam Fir, White Spruce, Ulmus americana (American Elm), Populus balsamifera (Balsam Poplar) and Fraxinus pennsylvanica (Green Ash). The Woodridge series soil type predominates in the moist regime, with sphagnum moss and peat in the wet portion. The drier locations consist of lacustrine sand overlying lacus- trine clay, with the bog beginning at the eastern limit of the sand. A check-list of the flora of the WNRE area has been published (Dugle 1969a; Dugle and El-Lakany 1971) and voucher specimens are in the WNRE herbarium. An illustrated document on shrub flora, which also reports radiation effects, was com- piled (Dugle et al. 1979). The small mammal species studied in FIG (Iverson and Turner 1976) include mice, voles and shrews. Only those insect species living on the foliage and 122 THE CANADIAN FIELD-NATURALIST Vol. 97 FIGURE 2. FIG irradiation tower photographed in 1973, showing the 20 m tower, tower supports, beam stop, plantation, service road, and mixed boreal forest. branches of Balsam Fir, Jack Pine, Paper Birch and Trembling Aspen were studied in detail. Present were 518 species in 111 families. A new aphid species, Man- sonaphis pinawae, was first described from specimens found on Labrador Tea growing in FIG (Robinson 97/3): FIG Irradiator The irradiator, designed and built at WNRE, is located in the center of the FIG area. It consists of a 20 m tower mounted between two supports (Figure 2) containing equipment such as an air compressor anda radiation sensor. A lead shield (castle) for the source is attached at the lower end of the tower. The tower can be tipped between its supports to lower it for mainte- nance. A triple-encapsulated !3’Cs source provides the gamma radiation field. Its effective source strength was 370 TBgq. (10 000 Ci) when the irradiator was placed in service in March 1973. Compressed air lifts the source from its castle to either of two irradiator positions, panoramic or beam, where it is held by an electromagnet at 20 m or 2m heights above ground level. In the panoramic posi- tion, the '!3’Cs source is at the top of the tower and the entire FIG area is irradiated. In the beam position, a collimated beam of gamma radiation is directed along a trolley track into a beam stop (Figure 2). The beam position may be used for acute irradiation of flats of plants or cages of small animals placed on moveable trolleys on the track. The source is returned to its castle by switching off the electrical supply to the electromagnet at the control building, allowing the source to descend ona cushion of air. Inthe event ofa malfunction when the source is in the beam or pano- 1983 TEMPERATURE (°C) JAN. BEBE MAR. APR. MAY JUNE GUTHRIE AND DUGLE: FIG FACILITY AND RESEARCH 123 EXTREME ~~ _MINIMUM ~ JULY AUG. SEPT. OCT. NOV. DEC. JAN. MONTH FIGURE 3. Long-term average and extreme maximum and minimum temperatures at WNRE(Davisand Reimer 1980). ramic positions, the source automatically drops into its castle. Radiation Field and Dosimetry Lithium fluoride thermoluminescent dosimeters (LiF TLDs) measure radiation dose-rate at the exper- imental sites in FIG. Aluminum foil-wrapped packets of two LiF TLDs (Harshaw TLD-100) enclosed in plastic vials are located at ground level or at | or2 m heights above ground in all botany quadrats, insect- sampling sites and other areas of particular interest. The maximum dose rate at ground level is 65 mGy/h,? at a distance of about 22 m from the !3’Cs source (about 12-18 m from the.tower). The dose rate decreases with distance from the irradiator to 0.005 mGy/h at the perimeter fence. Figure 4 shows the individual dosimeter readings from several years and directions. Consequently, the variation 1s large. FIG Research Programs Botanical Research The main effort of the FIG project is on plant 21 m Gy/h = 100 mrad /h. ecology and biosystematics. Its objectives are to determine changes in plant associations taking place with time and at various dose rates, effects of radia- tion on plant species, especially mature trees and shrubs, combined effects of other stresses (cold, insects, drought) and gamma radiation, and the low- est dose rate at which any effect of gamma radiation can be detected on plant associations or species. During the pre-irradiation studies, check-lists of plants present in the area and ecological descriptions of the several plant associations were prepared (Dugle 1969a; 1969b; 1972; Dugle and El-Lakany 1971). Pub- lished radiation sensitivities of trees were reviewed (El-Lakany 1971) and those of species or genera found in the FIG area were documented (Dugle and EI- Lakany 1971). Several techniques were used to study the effects of radiation on individual plants and plant species, and on such successional changes as those affecting numbers of individuals and species diversity (Dugle and Thibault 1972; 1974). A summary discussion of these techniques follows: the plant associations in FIG are natural ones with the exception of a cultivated strip about 5 m wide and 100 m long bordering an access road. This road runs 124 100.0 60.25 36.21 21.88 13.18 7.94 4.22 2.24 1.19 0.63 0.34 0.18 0.094 0.050 0.026 0.014 0.0075 0.0040 DOSE RATE (mGy/h) (omereOntrco) sce 40nn On 6S DISTANCE FROM IRRADIATOR (m) THE CANADIAN FIELD-NATURALIST Vol. 97 100 126 158 FIGURE 4. Computer-generated plot of dose rates (mGy/h) at various distances from the !3’Cs source in FIG. The line is a second order polynomial fit. The change in slope at around 20-30 m from the source is due to self-shielding by the tower and source. from an old logging road to the irradiator located in the center of the FIG area. A number of trees and shrubs planted in the culti- vated strip were well established before the irradiation began. Regular survival counts, phenological obser- vations and measurements of growth were made. The radiation sensitivities of several taxa growing in the strip were determined. Eight representative associations of plants were selected for detailed study (Dugle 1961b; 1972). The methods used to analyze the 115 chosen quadrats were reported (Dugle and Thibault 1972; 1974). A statistical analysis of forest type by category (random, aggregate, or uniform) was completed at the end of the pre-irradiation phase from quadrat tree maps listing the species, condition, diameter breast height (dbh) and exact location. The irradiation phase studies give results relevant to the ecological signifi- cance of radiation effects on individual tree species. Changes in canopy cover of trees and shrubs are assessed in two ways — as part of the quadrat analysis studied annually, or as a study of marked individual trees or shrubs studied biweekly during the growing season and sampled annually (Dugle and Mayoh 1974). Some marked trees have been carefully mea- sured at least once a year — the “age, height, circum- ference” trees. Thus, the effects of radiation can be observed on mature trees subjected to chronic radia- tion. After the tree dies, its annual rings are measured, allowing comparison with the concurrent control trees. For details of these analyses see Dugle et al. (1974) and Dugle and Thibault (1974). Biosystematic studies provide baseline information on species, especially woody ones, indigenous to FIG. Much of the information is relevant to the radiation sensitivity of the species. Specific studies and the top- ics relevant to radiation effects are 1. Birch (Dugle 1966; 1969c; Dugle et al. 1979) — taxa present, chromosome information, morphol- ogy, hybrids, leaf chemistry. 2. Spruce (La Roi and Dugle 1968; Dugle and Bols 1971) — taxa present, morphology, chromosome information, hybrids, leaf chemistry. 3. Rose (Ziola and Dugle 1970; El-Lakany 1972a; 1972b) — taxa present, morphology, chromosome information, ploidy level, DNA content, hybrids. 4. Other shrubs (Dugle et al. 1979) — taxa present, identification, morphology, hybrids, ploidy levels. 5. Cattail (Dugle and Copps 1972) — taxa present, hybrids. 6. DNA(El-Lakany and Dugle 1972) — taxa present, DNA content of local plants. Permanent plot (and miniplot) locations were ran- domly selected in each quadrat during the pre- 1983 irradiation phase using a stratified nested sampling design. Twenty plots per quadrat were established for annual study during all three phases of the FIG pro- ject (Dugle and Thibault 1972; 1974). Information collected during the plot and miniplot analysis was used for ecological and species effects studies. Many of the species effects were discovered through canopy, plot and general phenology and morphology descrip- tions. Radiation effects such as somatic mutations, growth-rate increase or decrease, and survival have also been described (Dugle et al. 1979). Time and radiation dose rate are the controlled variables in studies of succession using the same plots since the start of the project. A cluster analysis (Parks 1970) with several variations was used for the ordina- tion of the associations and their changes during irra- diation. This type of analysis allows the use of pres- ence and absence observations, in addition to numerical data. The dimensions of the analysis are large, approximately 10 000 quadrats (or plots) vs. 230 variables. In addition, phenology and morphology were regu- larly checked during the growing season. Somatic mutations and voucher specimens of radiation- exposed plants are deposited in the WNRE Herbarium.3 Zoological Research Changes in the presence/absence and abundance of small-mammal and phytophagous-insect species in FIG have been studied since 1969. An annual census of the numbers of small-mammal species and individ- uals was made using depletion trapping at permanent 8 x 8 station grids over a 30-day period (Iverson and Turner 1973). Four of these grids are inside, and the other four (concurrent controls) are outside, the FIG- enclosing fence. Several plant associations and dose rates are represented in these grids. The presence/absence and abundance of herbivorous- and entomophagous-insect species inhabiting chronically irradiated stands of Aspen, Birch, Fir and Jack Pine were studied by randomly dividing the stands into five samples. Each sample consisted of five sub-samples each containing five randomly selected trees. Every year from 1969 to 1978 a cubic meter of branches and foliage of each of these trees was “beaten” during the first two weeks of July. The dislodged insects were caught on a cloth sheet, from which they were removed and preserved in vials. The insect samples were sent to Northern Forest Research Laboratory, Environment Canada, for identification. A pre-irradiation breeding-bird survey was com- 3WNRE Herbarium, Pinawa, Manitoba, ROE ILO, Curator Janet R. Dugle. GUTHRIE AND DUGLE: FIG FACILITY AND RESEARCH DS) pleted in 1972 (Seabloom 1975). A study of the breed- ing biology of House Wrens and Tree Swallows living in FIG was published (Zach and Mayoh 1982). Other Research Effects of loss of canopy on the microclimate in stands of Jack Pine and Balsam Fir were studied. In addition, there are the experimental programs carried out by members of university departments and federal government laboratories. The scope of these pro- grams is wide, ranging from bog ecology (e.g., Long- ton 1972) to snail biosystematics (Pip 1977). Discussion Studies were or are being made on a number of different ecosystems of the effects of exposure to gamma radiation. Examples of these are summarized in Table |: Oak — Pine forest (Whittaker and Wood- well 1968; Woodwell and Whittaker 1968), desert (Whicker and Fraley 1974), grassland (Fraley and Whicker 1973), tropical rain forest (Odum and Pigeon 1970), and ZEUS, a project studying the effects of chronic irradiation on small-mammal population dynamics (Turner and Iverson 1976). The radiation conditions and experimental protocol of these studies are compared in Table | (refer also to Hart 1980). Only four of the summarized studies are not located within the continental USA. Two of these, FIG and ZEUS, are in Canada, at WNRE. The duration of some of the studies (Table |) was as short as a few months. The lake forest biome studied in Wisconsin (Rudolph 1974; Zavitkovski 1977) most closely resembles that of the FIG project. However, it was only irradiated during one growing season and it differs in its experimental protocol. To our knowl- edge, the FIG project is the first long-term study of the effects of chronic irradiation on a boreal forest ecosystem. Most of the studies of radiation effects on FIG flora and fauna are unpublished, since effects on plants and animals, and at the community level of the ecological hierarchy, are still occurring. The ecological changes that occurred within 50 m of the irradiator, where the radiation field exceeds 100 000 times background, were expected. Many woody plants are dead, and a different assemblage of species now comprises the ground cover. As a consequence, there were localized changes in the microclimate. So far, no evidence has been obtained that indicates small mammals and phy- tophagous insects were directly affected. They have, however, reacted to the changes in availability of food plants and habitat caused by radiation. In general, the radiation sensitivities of the impor- tant FIG plant species are as anticipated. The effects of radiation on shrubs (Dugle et al. 1979) and of 18 months of irradiation on canopy cover (Dugle and Mayoh 1974) were reported. THE CANADIAN FIELD-NATURALIST Vol. 97 126 WD LT = DEL I (9) “4ead Yora ur | Jago190 Jo se payst] sinoY aaneiNUND (¢) “Your, [¢ 0} 19q0190 | SI polsad saqUIM tsaquiaidas O¢ 01 [Udy | si potiad Jawuings (p) aseyd uonripess ul duinunuos = 5 (¢) (,) Y4eW UONsanb e YM payeoipuls! Y10M payst|qnd wo.) paureigo jou pur paiajul UONPWIOJU] (Z) ‘Uostedwios Ajydus oy sun JP[IWIS 0} pa}iaAUOd Uaag sey UONBWIOJUT (]) :SNON (926] UOSIAAT PuR JauIN |) “1861 01 £96] Suoneindod S910} paxtw Aq papunosins ‘Play uoneIpes JOA Jo SaIpnys uONRIpRssi-ad w 0 ppeur) SMOPROU BY | [ENPIAIpUy Ud}NRLJ O1 pasn pjalys ssauyoiy)-dulAIe A *s[O1]U09 JUdLINDUOD = FuluoIssiwW0S ‘bal ppp 1861 ‘eqolUR (SNAZ) sss ‘AQURULIOP JOIULA\ SP 9AJ9S SMOPRaW JUddRIPy JOVRIPRLI] - 1861 SOc 9) JaQUIDAON | ‘PMPUIG Japus) JUaWUOIIAUY [POI1T0]00Z7 (9961 Sa105 *JBUONIAIIP-1uN SI pyaly :201n0S Pure YIWIODIP $8961 JAI) “ara Jad uoseas auo ‘suoseas ino] ayy 8961 w ¢"| ‘vwSsn uonejuryd aug JO yora duLINp suoneipesit [eNpIAIpuy ‘soipnys uoneipess-aid SyIBUWI syieuias syiewal 0} ‘baL OPE ~ “BuI[OIR YINOS PIP PIO “AIUBULIOP 19IUI AA, Pur sjoiuod JudIINndUOD 22S 23S 29S 6961 *SOuc he) 6961 “Udyly “SUONRIPRU] JOATY YBUURARS ‘payodas GS3IPNIS UONBIPBII-o1d GOOE &7 6008 | 609F | cL6l 219M P/Y p7 1B spoliad awos ‘1aAaMoy (A|uatinou09 OPO 07 OP9 € 099 € 1L61 ui | vSsn “P/Y OT 24 0} palunsse uoNeIpeut Aled palpnis pur 10yeIpeds Obl TI OPo’ 099 € 0L61 ‘DEL pZE ~ 6961 “Opesojoy (£261 JO¥IYAN Pur Aa[es4) “AQUBULIOP JOIUIA\ WO) JULISIP SBo1e [O1}U0) OPP S Ors I 009 ¢ 6961 *SDiti (ce) [udy ¢ “uunN ol TBIq! SSB1D)=110US) ‘y96I Ul SUOP S3IPNIS UOHRIPRUI-Ag Play voneipes “A[JUaLNUOD 6000 €2 UMOUY ION UMOUXION $6 wl vwsn (pL6l [P19 Youss4) UINELJ O} pasn pyarys ssauyoiy)-ulAieA paipnis puke 10}eIpeIt ‘baL OLLI (9.10 “BpPAIN qniys j1asaq “AQUPWOP JOWIUWINS WO} JUPISIP Seale [OIUOD UMOUY ION UMOUX ION P96] ‘Sour = “PLOL AUeW b961 “AIT[RA YIOY- wasaq aArlo; ‘$961 11961 SaIpNIs UONPIPRII-I1q “A[JUaLINdUOD A1DAOIIY — L961 W ¢Z (0L6] UoadIg puke WNpO) paipnis puke 10, RIpet A1DA0IIY — 9961 ‘bal OLE 6961 S961 ory Olend }SasO} ures [RIIdO1) QURIUOW Wd} JURISIP Svale [O1]U0) 877 O8S I 89 $961 “SDut1 tudy LZ Arenuep 6] “OPI9IA 1d jSS1ONPUON EIDE ROOM ROI ION 196] “Jaws ul 6000 €€1 z 3 O86l Ww ogg vsn (£961 []aMpoom) dUOP SIIPNys UONRIPRIII-I1qg — - - - ‘ba ZsE ~ 1961 ‘yIOK MIN JSa10J UIq — YRO *AQUPULIOP JO}UL, “sear [O1U0) — i0S9 € i059 € 1961 *‘SOre iD JaqUIadAON ZZ ‘uoidy UdARY YOOIg ‘sinoy 000 bl = syseuial we¢ 026] 9uNe Jo se ainsodxa aanejnuny 29S UMOUY ION UMOUXION OL6I ‘baL pr ~ 6961 a0uRly (ZL6] ‘1B 19 Sage) “AQURWIOP 19WIWING “paiiodal JON 000 6 UMOUY ION UMOUYION 696] *‘SOutt iO Aine |] _‘ayoRiepey }S010.4 URIUPIIAI Pay] “1L61 91 6961 “JOJRIPRII YIM Suajqoid jo asnesaq — SSIPNYS UONRIPRIUI-I1qg (LL61 uoseas Suimols jo wed Ajiva 3ulinp *A|WUaLINIUOD A1QA0D9Y _ - rl6l w¢ vwsn LYSMONALTZ “hL6] Ydjopny) P/Y OZ UY) Ss JO} pajeIpess jsa104 paipnis pue 10) eIpesil K19A0I2Y — - €L6l ‘bal OLE ZL61 “UISUOISIAY S210j 24P] “AOURWIOP JaUT AA, WO} JURISIP Seaie [OI]UOD LILT iPge “ELE Z ZL61 ‘SOucr 190190 9] ZL61 ABW € ‘astiduayuq }so104 uONPIpeY asiudiaquy : 9Sce = 1861 6LS CS 6Ps £ LOE E O861 tL SP cob te PSP t 6161 LLL 8t 9SP E ftp e 8261 ‘CL61 ©1 6961 888 It CBE E lect LL61 ‘sumoy SE] 9¢ = 1861 [dy | ~ SoIPN)}S UONPIPRI-I1q GL6 PC 80P & Sect 9L6l Jo se ainsodxa aanjejnuins “AWuaLinouod ZE0 81 bore GUS € SL6l W (QZ epeue) jsodOJ [RI10G 'P/Y 61 S! polsad uoNeipesa adeIaAy Palpnjs pue 10j;eIpes1| tr0 11 Ips t fob t PL6l ‘bEL OLE £L6l “eqo ue (DIq) ewwey “AOURWOP Jd}ULAA, WO] JURISIP Seale [OIIUOD 6£0 P 86S lpr e €L6l ‘SOc 5) youeyp Z “PMPUI JOJPIPCII] Plat] SyIPW9Y sjo1qUuOD (sinoy) (sanoy) (sunoy) Siva A punois papuq paweis uOol]R907] yoalo1d SAUSIRUDD esevw A uaununs 00ge NURI Goneipeas jo sa1eq ainsodxa uoneipes jo uoleing ern uoneipey “SoIPN}S UONRIPRIII Play JO uostIVdWOD “| ATAV_L 1983 GUTHRIE AND DUGLE: FIG The third phase of the FIG project will begin when the radiation field is “turned-off” for the last time. This phase, recovery or post-irradiation, will be as ecologically interesting as the pre-irradiation and irradiation phases. It will give us the opportunity to study the recovery processes in an ecosystem damaged by prolonged exposure to an environmental insult. Acknowledgments The field and laboratory work of our colleagues and by the more than 50 summer students who have partic- ipated in the FIG project to date is gratefully acknowledged. Literature Cited Davis, P. A., and A. Reimer. 1980. Diffusion microclima- tology of WNRE: 1964-1976. Atomic Energy of Canada Limited Report, AECL-6493. Dugle, J. R. 1966. A taxonomic study of western Canadian species in the genus Betu/a. Canadian Journal of Botany 44: 929-1007. Dugle, J. R. 1969a. Check list of the vascular plants of the Whiteshell area in Manitoba. Atomic Energy of Canada Limited Report, AECL-3421. Dugle, J.R. 1969b. Ecology of the Field Irradiator- Gamma area. Pre-irradiation studies. I]. Plant associa- tions. Atomic Energy of Canada Limited Report, AECL-3424. Dugle, J. R. 1969c. Some nomenclature problems in North American Betu/a. Canadian Field-Naturalist 83: 250-252. Dugle, J. R. 1972. Plant associations in the Whiteshell Nuclear Research Establishment controlled area. Atomic Energy of Canada Limited Report, AECL—3957. Dugle, J. R., and N. Bols. 1971. Variation in Picea glauca and P. mariana in Manitoba and adjacent areas. Atomic Energy of Canada Limited Report, AECL—3681. Dugle, J. R.,and T. P. Copps. 1972. Pollen characteristics of Manitoba cattails. Canadian Field-Naturalist 86: 33-40. Dugle, J. R., and M. H. El-Lakany. 1971. Check list revi- sions of the plants of the Whiteshell area, Manitoba, including asummary of their published radiation sensitivi- ties. Atomic Energy of Canada Limited Report, AECL-3678. Dugle, J. R.,and K. R. Mayoh. 1974. Boreal forest canopy cover changes after eighteen months of chronic gamma irradiation. Manitoba Entomologist 8: 70-79. Dugle, J. R., K. R. Mayoh, and P. J. Barclay. 1979. Sh- rubs of the Field Irradiator— Gamma area in eastern Manitoba: systematic biology and ecology. Atomic Energy of Canada Limited Report, AECL-6428. Dugle, J. R., and D.H. Thibault. 1972. Ecology of the Field Irradiator-Gamma area. Pre-irradiation studies. II. Botanical methods and vegetation sampling procedures. Atomic Energy of Canada Limited Report, AECL-4135. Dugle, J. R., and D.H. Thibault. 1974. Ecology of the Field Irradiator-Gamma area. III. Revisions to botanical methods and vegetation sampling procedures (AECL- 4135). Atomic Energy of Canada Limited Report, AECL-4668. FACILITY AND RESEARCH O77 Dugle, J. R., D. H. Thibault, and W. D. Fitzsimmons. 1974. Pre-irradiation study of tree size, age and growth in the Field Irradiator-Gamma area. Atomic Energy of Can- ada Limited Report, AECL—4758. El-Lakany, M. H. 1971. The effects of ionizing radiation on forest trees: a review. Atomic Energy of Canada Limited Report, AECL-3951. El-Lakany, M. H. 1972a. DNA content variation, ploidy levels, and relationships within the genus Rosa in Manit- oba. Atomic Energy of Canada Limited Report, AECL-3960. El-Lakany, M. H. 1972b. Quantitative variation in DNA as related to ploidy level and species in some wild roses. Canadian Journal of Genetics and Cytology 14: 347-351. El-Lakany, M. H., and J. R. Dugle. 1972. DNA content in relation to phylogeny of selected boreal forest plants. Evo- lution 26: 427-434. : Fabries, M., A. Grauby, and J. L. Trochain. 1972. Study of a Mediterranean type phytocenose subjected to chronic gamma radiation. Radiation Botany 12: 125-135. Fraley, L., Jr., and F. W. Whicker. 1973. Response of short-grass plains vegetation to gamma radiation. Radia- tion Botany 13: 331-353. French, N. R., B. G. Maza, H. O. Hill, A. P. Aschwanden, and H. W. Kaaz. 1974. A population study of irradiated desert rodents. Ecological Monographs 44: 45-72. Hart, D. 1980. A selected bibliography of radiation effects upon whole organisms, populations and ecosystems. Atomic Energy of Canada Limited Report, AECL-6798. Iverson, S. L., and B. N. Turner. 1973. Habitats of small mammals at Whiteshell Nuclear Research Establishment. Atomic Energy of Canada Limited Report, AECL-3956. Iverson, S.L., and B. N. Turner. 1976. Small mammal radioecology: natural reproductive patterns of seven spe- cies. Atomic Energy of Canada Limited Report, AECL-5393. La Roi, G. H., and J. R. Dugle. 1968. A systematic and genecological study of Picea glauca and P. engelmannii using paper chromatograms of needle extracts. Canadian Journal of Botany 46: 649-687. Longton, R. E. 1972. Growth and reproduction in north- ern and southern hemisphere populations of the peat- forming moss Polytrichum alpestre Hoppe with reference to the estimation of productivity. Proceedings of the 4th International Peat Congress Helsinki, 1: 259-275. McCormick, J. F., and F. B. Golley. 1966. Irradiation of natural vegetation. An experimental facility, procedures and dosimetry. Health Physics 12: 1467-1474. Miller, G. L. 1968. Influence of season on the radiation sensitivity of an old field community. Chapel Hill, North Carolina, University of North Carolina, 262 p. Thesis. Odum, H. T., and R. F. Pigeon, Editors. 1970. A tropical rain forest. A study of irradiation and ecology at El Verde, Puerto Rico. TID-24270, USAEC Division of Technical Information Extension, Oak Ridge, Tennessee. Parks, J. M. 1970. FortranlV program for Q-mode cluster analysis on distance function with printed dendrogram. Computer contribution 46. State Geological Survey, The University of Kansas, Lawrence. Pip, E. 1977. Frequent scalariformy in a population of the pond snail, Lymanaea stagnalis. The Nautilus 92: 52-54. 128 THE CANADIAN FIELD-NATURALIST Platt, R. B. 1963. Ecological effects of ionizing radiation on organisms, communities and ecosystems. /n Radio- ecology. Edited by V. Schultz and K. W. Klement, Jr. Reinhold, New York. pp. 243-256. Reimer, A., and R. Desmarais. 1973. Micrometeorological energy budget methods and apparent diffusivity for boreal forest and grass sites at Pinawa, Manitoba, Canada. Agri- culture Meteorology I 1: 419-436. Robinson, A. G. 1973. Two new Macrosiphini (Homo- ptera: Aphididae) from Manitoba. The Canadian Ento- mologist 105: 813-815. Rowe, J. S. 1972. Forest regions of Canada. Canadian Forestry Service, Environment Canada, Publication Number 1300. Rudolph, T. D., Editor. 1974. The Enterprise, Wisconsin, Radiation Forest. Pre-irradiation ecological studies. USAEC Report, T1D-26116-P1. Seabloom, R.W. 1975. Field Irradiator-Gamma: _ pre- irradiation occurrence of breeding birds in three boreal habitats. Atomic Energy of Canada Limited Report, AECL-S115. Trewartha, G. T. 1954. An introduction to climate. 4th Edition. McGraw-Hill Book Company, Toronto. Turner, B. N., and S. L. Iverson. 1976. Project ZEUS: a field irradiator for small-mammal population studies. Atomic Energy of Canada Limited Report, AECL-5524. Vol. 97 Whicker, F. W., and L. Fraley, Jr. 1974. Effects of ionizing radiation on terrestrial plant communities. Advances in Radiation Biology 4: 317-366. Whittaker, R. H., and G. M. Woodwell. 1968. Structure, production and diversity of the oak-pine forest at Brook- haven, New York. Advances in Ecological Research |: 155-174. Woodwell, G. M. 1963. Design of the Brookhaven experi- ment on the effects of ionizing radiation on a terrestrial ecosystem. Radiation Botany 3: 125-133. Woodwell, G. M., and R. H. Whittaker. 1968. Effects of chronic gamma irradiation on plant communities. Quar- terly Review of Biology 43: 42-55. Zach, R., and K. R. Mayoh. 1982. Breeding biology of tree swallows and house wrens in a gradient of gamma radia- tion. Ecology 63: 1720-1728. Zavitkovski, J.. Editor. 1977. The Enterprise, Wisconsin, Radiation Forest. Radioecological Studies. Technical Information Center Energy Research and Development Administration, T]1D-26113-P2. Ziola, B., and J. R. Dugle. 1970. A biosystematic study of Manitoba roses. Atomic Energy of Canada Limited Report, AECL-3468. Received | June 1981 Accepted 24 December 1982 Book Reviews ZOOLOGY Bears: Their Biology and Management Edited by Clifford J. Martinka and Katherine L. McArthur. 1980. Bear Biology Association Conference Series No. 3. Bear Biology Association, Boise, Idaho. 375 pp., illus. No price given. This volume comprises a selection of papers presented at the Fourth International Conference on Bear Research and Management held at Kalispell, Montana in 1977. In the past decade, considerable interest has developed in the biology of North American bears. This interest has been stimulated, sometimes with great emotional outcry, by such events as Grizzly Bear attacks on humans in Yellowstone and Glacier National Parks, a television special on the Polar Bear migra- tion through Churchill, Manitoba, and increased Polar Bear problems at sites of oil, gas, and mineral exploration in the Arctic. With increasing contacts between humans and bears, it has become clear that proper management requires a more complete knowledge of certain species; published proceed- ings such as this serve a useful role in delineating what is known and in stimulating and directing future research. The 60 papers are broadly divided into those dealing with Mammalian Population Genetics Edited by Michael H. Smith and James Joule. 1981. Univer- sity of Georgia Press, Athens, Georgia, 380 pp., U.S. $25.00. This book is a series of papers arising from a 1978 sympo- sium depicting recent developments in the field of mammal- ian population genetics. Persons interested in ecological genetics of wild mammal populations will find cautious per- usal (see below) of several papers informative (those by Schnell and Selander, Schwartz and Armitage, Massey and Joule, McClenaghan and Gaines). Broad subject areas include genetics and phylogeny, genetic variability and mor- phological variability, genetics and environmental character- istics, genetics on islands, and the Chitty hypothesis. The fact that field biologists have ignored genetics and geneticists have ignored field biology is evident throughout this book. Also of alarming evidence is the ignorance of proper statistical analyses by many of the geneticists. Many authors have performed parametric statistical tests (analysis of variance, correlation and/or regression analysis) on genetic proportions (polymorphism, heterozygosity, known to be non-parametric) without arcsine-transforming their data set. Many have performed correlation and regression analysis on proportions with the same denominator and made ungrounded conclusions on the results. Few authors properly analyzed their data and as a result many of the conclusions made in this text may be erroneous. Many references cited in the text are missing from the bibliography. Authors with the same last name and year of ecology (55%), management (37%), and anatomy and physi- ology (8%). By species, 22 deal with Black Bears, 17 with Grizzlies (another seven are on the Eurasian Brown Bear, the same species as the Grizzly), six with Polar Bears, and four with the Japanese Black Bear. Although variation occurs within this collection, the quality is generally high. There is an appropriate mixture of papers dealing with “hard data” and supported opinion, and between descriptive and analyti- cal presentations. Readers unfamiliar with Eurasian species will obtain a good impression of some of the serious man- agement problems involving bears in the Eastern Hemisphere. It is likely that bear biologists are already quite familiar with much of the information contained in these proceed- ings. Those less knowledgeable about bears, but interested in the broad fields of ecology and wildlife management (as this reviewer is), should find this volume interesting. DAVID A. LOVEJOY Westfield State College, Westfield, Massachusetts, 01086. publication are cited in the text many times without first initials. Citations by the same author in the same year many times have no “a” or “b” suffix in the text but a suffix appears in the Literature Cited section. New orcorroborative items of interest throughout the text include: the lower genetic variability found on islands when compared to the mainland for 96 populations of 20 species may be primarily caused by founder effects during initial colonization; heterozygosity is usually positively correlated to body weight within a species; marine mammals usually have very low levels of genetic variability; reduced gene-flow and high levels of inbreeding may be the cause of low genetic similarity among pocket gopher populations; there usually are conflicting relationships among habitat heterogeneity, morphological variability, and genetic variability; genetic distance is seldom related to suspected phylogenetic dis- tance; and gene frequencies and/or genetic variability, as determined by electrophoresis, were not related to age, sex, survival or reproductive success of marmots. This text will stimulate new questions on mammalian genetics, but if mammalian geneticists go on analyzing their data sets as many have here, few of these questions will receive respectable answers. RICHARD M. ZAMMUTO Kananaskis Centre for Environmental Research, University of Calgary, Seebe, Alberta TOL 1X0. 129 130 THE CANADIAN FIELD-NATURALIST Vol. 97 Summer Birds of the Northwest Angle Provincial Forest and Adjacent Southeastern Manitoba, Canada By Robert S. Ferguson. 1981. National Museum of Natural Sciences, Ottawa, Canada. Syllogeus No. 23, 23 pp. Free. This is a factual, concise report of an intensive study conducted near the extreme southeastern corner of Manit- oba between 25 May and 7 July 1978. In the course of collecting 191 specimens of 83 species for the Museum of Natural Sciences, Ferguson observed 155 species but men- tions firm breeding evidence for only 18 species. He docu- ments a modest eastward extension of range for the Marbled Godwit and Sprague’s Pipit, but fails to relate this to the clearing of woods for fields and pastures. Immature Broad- winged Hawks were flying by 17 June, an incredibly early date. Rowan’s nearby study in 1920, a 3-day visit by Soper in 1940, and a 2-day visit by Godfrey in 1951, are mentioned in the introduction, but their observations are not summarized Wildlife Population Ecology Edited by James S. Wakeley. 1982. Pennsylvania State Uni- versity, University Park, Pennsylvania. x + 385 pp. Cloth U.S. $18.50; paper U.S. $11.95. This book is a compilation of 31 reprints of classical scientific articles used by most researchers studying ecology of wildlife populations. Wakeley has simply photocopied (most in whole, some in part) the most important contribu- tions made in wildlife population ecology and republished them in this book. Eleven papers are reprinted from the Journal of Wildlife Management, three from Ecology, three from Evolution, and one or two from American Naturalist, Auk, Condor, Endocrinology, Ibis, Journal of Animal Ecol- ogy, Journal of Applied Ecology, Journal of Mammalogy, Journal of Zoology, Science and Wildlife Monographs. The photographic reproduction is excellent for most of these articles. A section entitled “Additional Readings” lists com- plementary articles to those appearing in the text. In my opinion, Wakeley has selected a nearly perfect Fish Gene Pools Edited by N. Ryman. 1981. Proceedings of a symposium, Stockholm, January, 1980. Ecological Bulletins 34. FRN, Stockholm. 111 pp., illus. SK90. Fish Gene Pools isa compilation of nine papers originally presented at an international symposium held at Stockholm, Sweden, in 1980. The central theme is the preservation of the genetic diversity of natural fish populations that are impacted by man’s activities. Emphasis is placed on intra- specific variation, rather than variation among species, and the conservation of stocks of fish, rather than species. One of the book’s objectives is to make fisheries scientists and man- in the species accounts, which deal only with Ferguson’s observations. Ferguson in fact lists 62 species not seen by Rowan, while Rowan listed 13 species not seen by Ferguson. The multi-authored Birder’s Guide to Southeastern Mani- toba, published by the Manitoba Naturalists Society in 1980, is not even cited. Such lists of the birds of a given area once formed a major part of ornithology journals. Sy//ogeus presents a wide var- iety of single topics of varying length in a useful and inexpen- sive format, in this instance an excellent vehicle for sharing the results of a museum expedition with the interested public. C. STUART HOUSTON 863 University Drive, Saskatoon, Saskatchewan S7N 0J8 assemblage of papers for his subject. Most of the papers are used in wildlife ecology and/or population courses at col- leges and universities throughout North America. Researchers in wildlife population ecology have probably read more than 75% of the papers and have probably been exposed to concepts of more than 95% of them. Il expect this book will receive wide use in wildlife ecology, especially with its attractive price. The book should be an excellent starting point for students and/or persons inter- ested in wildlife population ecology. It should also serve well as a useful treatise for professionals who cite these papers in their own publications. RICHARD M. ZAMMUTO Kananaskis Centre for Environmental Research, University of Calgary, Seebe, Alberta TOL 1X0 agers aware that the genetic variation within and between natural fish populations is a valuable natural resource that needs to be protected. The authors provide the reader with a series of recommendations for the management of natural fish populations. These recommendations advocate a man- agement strategy that minimizes further genetic erosion of natural fish populations. The book has the usual pluses and minuses of multi- authored publications. The individual presentations are quite varied in their scope and quality as well as their rele- vance to the main theme of the book. I particularly enjoyed the papers by Allendorf and Phelps; Ryman; and Stahl. The 1983 Allendorf and Phelps paper gives a very lucid discussion of the use of gel electrophoresis of enzymes as a tool for the study. of intraspecific genetic variation. Little knowledge of population genetics is required to follow the main arguments of these papers, since basic concepts in population genetics are reviewed for the reader. Each paper is intended to stand On its own, and can be read independently. However, this approach has resulted in considerable repetition, especially in the methodology sections of some of the papers. Also, the general concern of most of the authors about the conserva- tion of genetic fisheries resources is restated repeatedly, while not enough space is devoted to integrating the different points of view expressed by the population geneticist, ecolo- gist, and fisheries manager. The book is most effective in presenting some of the evi- dence that has recently come to light about the rich ecologi- cal and genetic diversity (mostly isozyme variation) within species. However, the taxonomic and geographic perspec- tives are not as broad as one might have expected from the title of the book. In fact, only three species of the same genus (Salmo) are discussed in any detail, and the geographic range is virtually restricted to Sweden and Norway, with one example from North America. Nevertheless, these examples serve to give the reader an appreciation of the ecological and genetic intraspecific variation found in natural fish popula- tions, and to make a strong case for the conservation of this variation. The Scandinavian perspective is particularly rele- An Atlas of the Birds of the Western Palaearctic By Colin Harrison. 1982. Princeton University Press, Prin- ceton, New Jersey. 322 pp., illus. U.S. $25.00. North American ornithologists who might expect the direct relevance of this atlas to be restricted to maps of species distribution shared by the Nearctic and Palaearctic may be missing an important reference source. Harrison’s inclusion in his family discussions of ecological equivalents in continents outside the Palaearctic provides a goldmine of potential ideas for further research. Although the book is based on the western Palaearctic, many of the maps include species of the much less-known eastern Palaearctic. Thus, this book also serves as a brief introduction to the birds of much of Asia. The book consists of a few introductory sections, followed by about 250 pages of maps and accompanying text, one page of further reading and indices to English and scientific names. The introductory sections are dominated by a discus- sion On interpreting the maps. This discussion includes prob- lems posed by the technical difficulties of representing the curved surface of the earth ona flat page, problems posed by behaviour of strays and irruptive species, and factors influencing the distribution of a species. The atlas itself is arranged by family, each being introduced by a general account of it in the western Palaeartic and in the world as a whole, with considerable emphasis on “replacement” species in other continents, and of related species in other parts of the Palaearctic Subregion. In families represented througha major portion of this subregion, a map is shown of the BOOK REVIEWS iil vant to Canada, since we too are the custodians of a rich variety of trout, charr, and salmon stocks. The authors recommend that genetic fisheries resources be identified. It is clear from the book how to proceed with regard to biochemical variation, but the approach to be taken to genetic differences in local adaptation is not given the same emphasis. In general, the recommendations state laudable goals. However, the methods for achieving those goals need to be developed. For example, one of the recom- mendations states that “fisheries should be managed so as not to harm the genetic characteristics of populations through selective harvesting”. The question remains, how this is to be accomplished? How does one balance the needs of the user with the concern of the conservationists? Every fisheries manager should be exposed to the ideas expressed in this volume. I recommend this book, as well, to those interested in fisheries population genetics, but lacking the time or the inclination to read journal articles. The authors successfully relate basic population genetic concepts to fisheries management problems and made a strong case for the conservation of genetic fisheries resources. PETER E. IHSSEN Ontario Ministry of Natural Resources, Research Section, Fisheries Branch, Box 50, Maple, Ontario LOJ 1E0 breeding distribution of all species in it. Family discussions are followed by individual species accounts for the western Palaearctic itself. Each species is alloted a map of permanent, breeding and/ or winter range, with arrows indicating direc- tion of migratory movements, and accompanied by a para- graph on geographic and altitudinal range, habitat, and other information related to distribution. In families with distinct groups (subfamilies, large genera, and other obvious divisions) these groups are usually given a general write-up, often accompanied by an additional map. A drawing of each Palaearctic species mentioned in the text appears at the bottom of the page. The map’ of the Yellow-billed and Common loons is Holarctic, but all other maps are restricted to the western Palaearctic or the Palaearctic as a whole. The maps, drawn by Crispin Fisher, have obviously been thoroughly researched, and contain few errors. The small donuts of seasonal colour in the midst of the permanent ranges of the Black-billed Magpie and Linnet appear to be slips of the paintbrush, and the dot of winter distribution shown in the Spitzbergen Islands for the tropical Chestnut-bellied Sand- grouse is an obvious error. The black-and-white drawings accompanying the text are generally pleasing, although a few (notably the larks, finches and sparrows) are distorted in shape, and the tarsi of the Arctic Tern are too long. On the whole, the text is thorough and well written. The introductory section on factors affecting distribution in the Palaearctic is an excellent precis of the principles of biogeo- graphy as applied to any large land mass. Although this 132 section contains numerous references to maps in the main part of the book, I found only one error in number — the Orange-tufted Sunbird mentioned on page 36 should have referred to map 583, not 584. I-found only one other such numbering error in the whole book (the Pied/ White Wagtail complex is mapped on map 422, not 393 as stated on page 208), and there are only about three other typographical errors in it. Other errors are equally scarce and minor. The statement that the Laurel Pigeon may be extinct on Gomera Island should have substituted the word extirpated, the word | “Plover” after Killdeer is redundant, and the Cedar Wax- wing does not breed only ineastern North America. I do not agree with all of Harrison’s statements on ecological equival- ents. If Blue-winged Teal indeed replace Gargeny on this continent, then this surely applies to the Cinnamon Teal also. Forster’s Terns do, indeed, breed in parts of the western U.S. not occupied by Common Terns, but the breeding ranges of the two species overlap extensively, and ecologically Fors- ter’s Tern is much more similar to the unrelated Swift Tern than to the Common. Arctic Animal Ecology By Hermann Remmert. 1980. Springer-Verlag, Berlin. 250 pp., illus. U.S. $24.80. This translation of a work by a German ecologist with fifteen years of experience on Spitsbergen promises to be a synthesis of information from all parts of the Arctic. It draws on an impressive range of sources, and uses section headings of good ecological generality. Unfortunately, though, some combination of author, translator, and publisher has made such a poor job of it that the result is an insult to the reader. Technical problems abound. By my tally, almost half of the citations in the text are not given in the list of references. Figures have unexplained symbols, missing units, and super- fluous data. Some do not show what they are supposed to, others seem to show nothing at all. A three-quarter page figure purporting to show the effect of caterpillar outbreaks on forests has tree-ring data but nothing on insects except for a line in the caption that says “minimum growth rates are usually connected with such outbreaks”. A discussion of a freshwater lake in Iceland fed by warm springs is illustrated by a feeding diagram for the Bering Sea. A table supposed to “reveal very clearly the typical pauperization of arctic regions with respect to large decomposers” has no data on body size, no data on ecological function, no comparative data from the south, and no groups finer than “Diptera lary.” or “Coleoptera ad.”. The style is emphatic throughout, with much use of words like always, impossible, obviously, completely, more often than not without justification in my opinion. Some extraor- dinary statements are made. Arctic diving birds are invaria- bly soaked to the skin and their “feathers and extremities freeze as soon as the animals emerge from the water”. Sim- ilarly, “poikilothermic organisms exhibit no active signs of life below freezing point (land and freshwater: 0°C, sea THE CANADIAN FIELD-NATURALIST Vol. 97 Such differences in interpretation and minor errors do not detract from the overall high quality of this book, and | found many intriguing differences between habits of species shared by our continents. For example, Horned and Eared grebes show much less overlap in breeding ranges in Europe than in Canada; Great Cormorants barely stray inland in Pri Edward Island, but breed on freshwater lakes in the Palaearctic; and Canada Geese introduced to Europe have / evolved a northward moult migration in the British Isles, but a southward post-breeding migration in Sweden! In short, this atlas should be standard fare on the shelves of European ornithologists. For North Americans, it offers far more than the peripheral distributional information that may be suggested by its title. MARTIN K. MCNICHOLL 320 Markham Street, Toronto, Ontario M6G 2K9. water: —1.7°C)”. By contrast, “the Arctic is a favourable environment for all warm-blooded organisms”. In the section “Population Cycles” we are given a figure showing the Canadian fur-trade statistics for Lynx and “mountain hare”, without acknowledgment of source and without critical comment, as evidence of a nine-year arctic cycle. There is no distinction made between cycle and simple variation. The summary figure for this discussion says nothing more than that populations in general can go up or down for many different reasons — it misses the point about cycles altogether. The section “Seasonal Migrations of Birds and Mammals” begins by saying that “the majority of birds breeding in the Arctic and a large proportion of the mam- mals leave the true Arctic in winter for other regions”. There are no references to mammals and no examples until several pages later when we are told that “most arctic mammals remain in the same geographical region winter and summer”. Elsewhere, the logic is of this standard: “If forest caribou and forest reindeer are in fact distinct species they must have evolved in different places and their similarities are thus analogies. Each of them must therefore be an individual species ...”. There is hardly a page in the book that does not yield other examples of these kinds. The considered opinion of an expe- rienced investigator can be a better integrator of disparate data than any other, and deserves much respect. However, in this case there are so many problems with logic and content and presentation that the result is most unsatisfactory. JOHN MIDDLETON Department of Botany, University of Dar es Salaam, P.O. Box 35060, Dar es Salaam, Tanzania 1983 BOOK REVIEWS 133 Handbook of Census Methods For Terrestrial Vertebrates Edited by David E. Davis. 1982. CRC Press, Boca Raton, Florida. 397 pp., U.S. $140.00. This guide should be useful to those wanting to estimate the abundance of amphibians, birds, mammals or reptiles in a habitat. It could be consulted as a starting-point reference (along with Schemnitz, S. D. 1980, Wildlife Management Techniques Manual) during planning, prior to a census of a species in a particular habitat. This guide is a series of more than 150 short articles written by 160 authors. It features articles on 3 amphibians, 43 birds, 60 mammals, 4 reptiles and 32 habitats (for combinations of species). Each article was written by a researcher (address provided for direct inquiries) who has personally tested and recently published (since 1974) a scientific article on the use of the censusing method he discusses. Included in the discus- sions are many basic facts on the species or habitat concerned. Each article may answer several questions: how many traps should be set?; how long should the traps be set?; what EDITOR’S NOTE The following two publications are only available in Chi- nese. The reviewers have combined linguistic and scientific backgrounds to provide a unique insight into the status of fisheries research in this vast and important geographic area. The Fishes of the Islands in the South China Sea By South China Sea Fisheries Institute, Xiamen Fisheries College, Institute of Oceanology, Institute of Zoology, East China Sea Fisheries Institute, South China Sea Insti- tute of Oceanology, Beijing Natural History Museum, and Fisheries Institute of Hainan Administrative Region. 1979. Science Press, Beijing, China. xxv + 613 pp. Cloth ¥11.80; paper ¥7.80 (approximately $8.00 and $5.50 Can- adian, respectively). China faces on three seas, the Yellow, East China and South China Seas. Because of their richness, the faunas ranging from boreal to tropical and comprising over 2 000 marine species, are of wide ichthyological interest. Although most of the larger ichthyological studies pub- lished in China in the last decade have been on freshwater fishes, several marine faunal works have been published in the 50’s and 60’s: Tchang Tchun-Line? al. 1955. Fishes of the Yellow Sea and Pohai, China. Science Press, Peking. 362 pp., 206 fig.; Institute of Zoology et al. 1962. Fishes of the South China Sea. Science Press, Peking, 1184 pp.; Cheng, C.T. et al. (eds) 1962. Economic fauna of China. Marine Fishes. Science Press, Peking. 174 pp. 32 pls. and Chu Yuan- Ting. 1963. Fishes of the East China Sea. Science Press, Peking. 642 pp., 442 fig. These works are scientific in style. A popular guide series has also been started: Institute of Ocea- nology and Shanghai Natural History Museum. 1975. areas should be checked?; how accurate will the estimate be?; and/or where is more information on the species or habitat? A comprehensive chapter contains general methods usually used during census work (basically a reprint of Chapter 14 in Schemnitz 1980). The index is very good — an important aspect of a hand- book of this kind. The text reads with facility, is informative, and has few typographical errors. Perhaps a problem with this book is its high price. Few researchers may want to obtain a personal copy whereas the guide will probably receive wide use as a reference. Persons interested in beginning a study on a terrestrial vertebrate or habitat (a few aquatic species and habitats are discussed) should find this guide a useful reference. RICHARD M. ZAMMUTO Kananaskis Centre for Environmental Research, University of Calgary, Seebe, Alberta TOL 1X0. Zhong Guo Hai Yang Yu Lei Yuan Se Tu Ji (The color pictorial guide to Chinese marine fishes). |. People’s Press, Shanghai. 230 col. pl. The Fishes of the Islands in the South China Sea covers the surrounding waters of the four groups of islands: Dongsha (Pratas Islands), Xisha and Zhongsha (Paracel Islands including Macclesfield Bank), and Nansha (reef islands around Investigator Shoal, roughly between 7° and 12°, west to Palawn of the Philippines and north to Borneo). As in all the other works mentioned, the text is naturally in Chinese, but scientific names in Latin are also included in the text, tables of contents, and captions. The book consists of the following sections: a detailed table of contents including a list of contributors (31) and their responsible taxa, and illustrators (16); scientific des- criptions including synonymies, keys to taxa, and other interesting information such as cooking recipes and medici- nal uses; literature cited (10 pages, somewhat incomplete); a Chinese index; an index to scientific names in Latin; and colour plates (34 species) and black and white plates (96 species). There are accounts for 521 species. Compare this with 386 species for the Pacific coast of Canada (Y. Jean, A. Peden, and D. E. McAllister. 1981. British Columbia Pro- vincial Museum Heritage Record Number 13). The format of the book is excellent, and some features such as synonymies should be adopted for Canadian fish books. The use of multi-character keys is commendable; drawings, paintings and their printing quality good. The drawings, prepared by 16 artists in seven institutions, are remarkably uniform in appearance. Important morphologi- cal characters in the species description are under-dotted. Four species of Chinese fishes were run through keys to correct identifications without any difficulty. Aside from a 134 handful of publications, all references predate 1975. Thus, references such as G. U. Lindberg and Z. V. Krasyukova, 1975, Fishes of the Sea of Japan, Zool. Inst. Acad. Naukk SSSR, were omitted. Some classifications are out of date, such as the use of Scopeliformes instead of Myctophiformes (J.S. Nelson’s 1975, Fishes of the world, is not mentioned). A preliminary report of the deep-sea fishes of the South China Sea by Q. Cheng and M. Tian(1981 Studia Marina Sinica 18: 235-275, 32 text-fig., | pl.) recently added 34 new records for China. Obviously The Fishes of the Islands in the South China Sea is an important contribution to science and will assist in managing fisheries and educating students. If translated, this and the previously mentioned faunal works would be of great service to western scientists. Canada, which is just develop- THE CANADIAN FIELD-NATURALIST Vol. 97 ing the field of aquaculture, could profit from China’s centur- les of experience. Readers may be interested to learn that a Chinese Society of Ichthyology was inaugurated 16 October 1979, which bodes well for the continued development of ichthyology in this ichthyologically rich country. A “Transactions of the Chinese Ichthyological Society” will be published by the Society. DON E. MCALLISTER and C. T. SHIH Vertebrate and Invertebrate Zoology Divisions, National Museum of Natural Sciences, National Museums of Canada Ottawa, Ontario KIA 0M8& A Study of the Lateral-line Canals System and that of Lorenzini Ampullae and Tubules of Elasmobranchiate Fishes of China By Yuanting T. Chu and Ching Wen Meng. 1979. Mono- graph of Fishes of China, No. 2. Shanghai Kexue Jishu Zhubanshe (Shanghai Science and Technology Press), Shanghai. 132 pp. 64 colour plates. This book, in Chinese, but with a seven page abstract in English, is devoted to the study of the structure and arran- gement of the lateral line system, the ampullae and tubules of Lorenzini, and Savi’s sacs, as well as the phylogeny of elas- mobranchs. The authors describe these epidermis-derived sensory systems in 73 species of sharks, rays, and skates found in the seas of China. They illustrate all but two in colour plates which show the lateral line of the head and often the body in blue, the ampullae and tubules of Lorenzini in red, and Savi’s sacs in green. Many of these have never been depicted before. Based on the arrangement of these structures, and referring to the classification system of Woodward (1889, 1891), Jordan (1923), Whiteley (1937), Bigelow and Schroeder (1948-1952), Berg (1955), Romer (1966), Rass and Lindberg (1971), and Compagna (1973), the authors develop a new classification system for Chinese elasmobranchs and present a key and a phylogenetic tree with a geological time scale. Several taxonomic changes are made in their new system. Galeorhinus japonicus (Muller et Henle) is removed from Carcharhinidae and placed in Tria- kidae and Triaenodon obescus (Ruppell) from Triakidae to Carcharhinidae; Rhinobatos granulatus (Cuvier) is assigned to the genus Scobatus and Dasyatis kuhlii ( Muller et Henle) to Urolophoides. A new family Cirrhoscylliudae is created to house Cirrhoscyllium expolitum Smith and Radcliffe, formerly of Orectolobidae. Clearly this study makes important contributions to the sensory systems and classification of the Elasmobranchii. It should be noted that this study appeared 16 years after the first of the series was published (Chu, Lo, and Wu, 1963: A study on the classification of sciaenoid fishes of China, with descriptions of new genera and species). Apparently the Chinese scientists have wasted no time in publishing studies that had been interrupted by the cultural revolution. Other works related to systematics known to us include Fauna Sinica, Economic Fauna of China, and monographs such as The Fishes of the Islands in the South China Sea. DON E. MCALLISTER and C. T. SHIH Vertebrate Zoology Division and Invertebrate Zoology Divi- sion, National Museum of Natural Sciences, National Museums of Canada, Ottawa, Ontario, Canada KIA 0M8 1983 BOTANY Shrubs of Ontario By James H. Soper and Margaret L. Heimburger. 1982. Royal Ontario Museum, Toronto. xxxi+ 495 pp., illus. $20.00. The new Shrubs of Ontario is far more than a revised edition of the familiar 100 Shrubs of Ontario, by the same authors. Whereas the usefulness of the earlier work was seriously limited by the omission of several major families, this new book provides thorough coverage of Ontario shrubs, ranging from the barely woody Chimaphila species to the usually arborescent Sassafras, and from Carolinian rarities restricted to the southernmost part of the province to arctic species occurring only along the shores of Hudson Bay. Naturalized species are also included. This book is intended primarily for the identification of shrubs by persons with limited experience in plant systemat- ics. However, its contents are sufficiently rich and diverse to make it a useful reference for professional plant taxonomists as well. Families are arranged in the familiar sequence of Engler & Prantl; genera and species within each family are arranged alphabetically. There are dichotomous keys to the genera and species. Evaluation of the keys will require their use by many persons, but they are simply and lucidly written and in general appear to be very good. They are based largely on vegetative characters, and do not require material, such as both flowers and fruit, unlikely to be present on a single specimen. The only fault I have detected is that Rhamnus cathartica keys out under “Leaves all opposite” rather than “Leaves mostly opposite but some subopposite.” The species descriptions are detailed, and are accompan- ied by succinct “field checks.” The illustrations, by Leslie A. Garay and Ronald A. With, are excellent. Locality dot maps indicate the distribution of the respective species in Ontario, and comments are numerous, on topics from traditional and modern utilization to variation in chromosome numbers. Die Kustenvegetation Ostkanadas By Dietert Thannheiser. 1981. Ferdinand Schoningh, Paderborn, Germany. 202 pp., illus. DM 41.50. This fine study dealing with the coastal vegetation of east- ern Canada, is based ona three-year survey along the Gulf of St. Lawrence. Preliminary chapters outline the physical geo- graphy of the region (including an abundance of hydrologi- cal and meteorological maps) and explain the sampling methodology. The main portion of the book is devoted to a survey of the vegetation of three major coastal communities: salt marsh, sand dune, and strand. The physiography, asso- ciations, distribution, and community dynamics of each are covered. Illustrative material includes distribution maps of more than 30 species, detailed vegetation maps for selected BOOK REVIEWS 135 Among the literature citations are some as recent as 1980. The nomenclature is consistently up-to-date. The authors have generally been conservative about accepting taxonomic departures from standard floras — poison ivy, for example, appears as Rhus rather than as Toxicodendron — but differ- ing treatments in recent works are-noted. The distributional data are likewise up-to-date, having been based on specimens examined at many Ontario herbaria, including recent accessions. The treatment of Salix, having been prepared under the guidance of George W. Argus, essentially provides a much- needed taxonomic revision of this genus for Ontario. Other acknowledgments attest to the wide range of expertise represented in both the taxonomic treatments and the distri- butional data. The treatment of Crataegus, however, is based largely on the drastic “lumping” of Gleason & Cronquist, with further restriction to selected species, although the recent monograph by Phipps & Muniyamma is cited. Proofreading has generally been good, but several lines on page 212 evidently were lost in the “stripping” process, with the result that some material on naturalized roses was omitted. The size of this book, 25 x 17 cm, reflects an excellent balance between legibility of text and clarity of figures, and cost and bulk. By current standards, the price is quite reasonable. In marked contrast to the several good wildflower guides and the plethora of tree-identification books, there have been no previous guidebooks really satisfactory for the identifica- tion of Ontario shrubs. I am pleased indeed that this need has been met by such a superb work as this. JAMES S. PRINGLE Royal Botanical Gardens, Hamilton, Ontario L8N 3H8 localities, and association diagrams. The study is by no means definitive, but it should serve as an excellent basis for subsequent ecological studies of the relevant communities. The author is particularly commended for having his collec- tions verified by a number of established taxonomists. The one real objection to the book is that it is in German, which may render it less accessible to the Canadian botanical com- munity than it deserves to be. A. E. STAHEVITCH Biosystematics Research Institute, Agriculture Canada, Central Experimental Farm, Ottawa, Ontario KIA 0C6 136 THE CANADIAN FIELD-NATURALIST Lichens: an illustrated guide By Frank Dobson. 1981. 2nd edition. Richmond, England (distributed by Mad River Press, Eureka, California). 320 pp., illus. + plates. U.S. $18.70. In Great Britain, and increasingly in Canada and the Uni- ted States, there is a large number of serious amateur natural- ists interested in mosses and lichens. This group is able and willing to use keys and fairly technical guide books to help them sort through the complexities of dealing with small organisms and their often microscopic characters. Yet, when allis said and done, there is nothing like a good illustration to provide the novice with corroboration of an identification. Dobson’s “Illustrated Guide” to British lichens is addressed to those amateurs willing to usea hand lens and microscope, and even try a few simple chemical tests, but who need good photographs for reliably pinning down a name. The book covers close to 450 frequently encountered lichens and some rarer species (out of approximately 1500 species) found in Great Britain. At least a third of the species covered do not occur in Canada, and less than half are common here; the book will therefore not be entirely reliable on this side of the Atlantic. Most species have been photo- graphed in black and white, and 50 are illustrated in colour. Identifications are made with the aid of keys, first to groups of genera, then to genera, and finally to species. I believe Dobson’s book will serve the diligent British amateur lichenologist fairly well, but not before he or she becomes fairly familiar with many lichen genera. The key to groups and genera I found to be quite complex and unrelia- ble. Actually, genera are not keyed out at all; they are instead listed together with superficially similar genera, with their distinguishing characteristics compared in a kind of table. This method will prove to be frustrating to those not able to eliminate a majority of the potential names through experience. The Illustrated Flora of Illinois: Flowering Plants, By Robert H. Mohlenbrock. 1982. Southern Illinois Univ. Press, Carbondale and Edwardsville. 234 pp., illus. U.S. $22.95. In my earlier review of two books of this series, ‘Willows to mustards” and “Magnolias to pitcher plants,” I presented a list of the nine titles published at that time (1982. Canadian Field-Naturalist 96: 235-6). To this list must now be added this tenth volume. “Basswoods to spurges” continues to fol- low a modified Thorne system of classification. Four orders are treated, namely, Malvales (families Tiliaceae, Sterculia- ceae and Malvaceae), Urticales (Ulmaceae, Moraceae and Urticaceae), Rhamnales (Rhamnaceae and Elaeagnaceae) and Euphorbiales (Thymelaeaceae and Euphorbiaceae). The cost is not given on the jacket, but | am informed by our library that it remains the same as the last volume. I pray that the escalating book prices are now stabilizing. This volume follows the pattern established in the preced- ing volumes. It is dedicated to the author's eldest son, Mark Vol. 97 The author provides short descriptions of each genus and species; in most cases, I found them to be accurate. Each genus description includes drawings of asci and spores which, unfortunately, are not very good. In some cases, (as with almost all ascus drawings, and the spores of such genera as Mycoblastus, Physcia, and Phaeographis), the drawings are misleading. The line drawings are rather poor asa whole, especially when juxtaposed with the excellent black-and- white photographs, an area in which the author obviously excels. The colour photographs are also good, but are often somewhat “distant” for really useful comparisons. It is a pity that some of the lichens picked for photography were atypi- cal (e.g., Cladonia coniocraea, Lecanora rupicola, and Lobaria pulmonaria). The writing is clear and concise, but is marred by a number of grammatical errors apparently missed in the preparation of the second edition. For those owning the first edition (published in 1979), I would advise against buying the newer version. Very few changes have been made except to the index, which is now much more accurate and complete. A list of recent nomenclatural synonyms is also given in the second edition, following the recent checklist of British lichens by D. L. Hawksworth, P. W. James, and B. J. Coppins pub- lished in volume 12 of the “Lichenologist” journal. In summary, I recommend the book to serious amateurs wanting a set of descriptions and good photographs of the common British lichens. Although the coverage of Canadian lichens is necessarily limited, Dobson’s volume 1s one of the few places one can find illustrations of crustose species, and for this reason it will be of some value to Canadian readers. IRWIN M. BRODO National Museum of Natural Sciences, National Museums of Canada, Ottawa KIA 0M8& Basswoods to Spurges William Mohlenbrock, who did all the illustrations. He has done a superb job. The outlines are sharp and clear, and stipple type shading is used for floral enlargements, fruits, seeds, etc., which accompany the overall illustration with magnifications listed in the caption. For each species syn- onyms are given and confusion in the application of names is explained and resolved. The maps use an arrangement and construction similar to the other volumes. Forty-two genera containing 103 species and 1I5 lesser taxa are dealt with in this volume. Some 27 of these species are either adventive from some other part of the United States or are introductions from somewhere else, mostly Europe. Most of these occur in 19 genera, of which nine consist only of weedy species. The usual map of Illinois counties, and an introduction containing an explanation of the modified Thorne classifica- tion, follows the foreword. Because it differs from the cus- tomary Engler and Prantl system, the complete classification is given with Mohlenbrock’s modifications indicated. The work is based on field studies, plus examinations of herba- 1983 rium material from 11 listed herbaria and some unlisted private collections. Maps were prepared by the author’s daughter, Ann, and the clerical work performed by his wife, Beverly, so the series is becoming a family affair. Following the species descriptions, there is a list of “Spe- cies Excluded,” mostly misidentifications, and then a numer- icalsummary of taxa treated, a glossary, literature cited, and finally an index. The key to the species of Ce/tis in Illinois deserves some attention. There are three species in Illinois, C. occidentalis, C. laevigata, and C. tenuifolia. Because the key employs leaf features, C. occidentalis comes out three times in the first four choices. C. /aevigata and C. tenuifolia come out twice each. There are seven couplets. Visually it looks very odd indeed. Then, there are additional keys to separate the three varieties of both C. occidentalis and C. laevigata, and to separate the two varieties of C. tenuifolia. If these species seemingly intergrade as the author contends, then perhaps it The Cacti of the United States and Canada By L. Benson. 1982. Stanford University Press, Stanford, California. 1044 pp., illus., plus distribution maps. U.S. $85.00. Although only four of some 152 cacti species which occur north of the Mexican border are found as native species in Canada, this book is of great interest. The Cactus family has attracted much attention because of its many remarkable growth forms. Some species are tree- like in aspect; numerous others are small and can readily be grown indoors in our northern regions. In more southern climates, fascinating displays can be found on rocky slopes in gardens, to say nothing of native habitats, where such species as the Organ Pipe Cactus, Barrel Cactus, Saguaros and Teddy Bear Chollas occur. This book, written by the foremost student of North American cacti, is more than just keys and descriptions of cacti. It contains a wealth of information in its two parts. The first (about one-quarter of the book) includes chapters on the structure, physiology and chemical characters of cacti; a discussion of the origin and relationships of taxa within the family Cactaceae; a discussion of the nature of species, varie- ties, and hybrids; the author’s policy in developing his classi- fication of the North American cacti; herbarium and field methodologies, including techniques of specimen gathering and notes required; geographic distributions and environ- ments in which cacti are found, including the relationship of floras and vegetation to world climate; the floras and floristic BOOK REVIEWS Sy would have been better to employ a comparative table, as done with Red and White Mulberry on page 88 and the hops on page 99. Incidentally, the same three species come out very easily in the key by Gleason & Cronquist (Manual of vascular plants of northeastern United States and adjacent Canada. Van Nostrand, 1963.) which uses leaf characters only. Beyond my adverse reaction to the Cel/tis key, I have nothing but praise for this volume. Undoubtedly, the author’s early decision to illustrate every species was an excellent one. Good illustrations of every species ensure the success of a flora; no matter how well a text is written, a picture is worth a thousand words. JOHN M. GILLETT Vascular Plant Section, National Museum of Natural Scien- ces, Ottawa, Ontario KIA 0M8 associations of North America; the uses of cacti; and the conservation of cacti. All are well illustrated. The second part is a detailed treatment of the 18 genera of cacti found in North America, with keys, detailed descrip- tions of the genera and species, ecology, distribution, special comments, line drawings, photographs, and distribution maps. The maps present some difficulty to one not too familiar with the geography of the United States because there are no place names and they depict only state and county borders. This is softened somewhat by the inclusion of a series of general reference maps near the end of the book which name all the counties in the various states where the family has its main range. A section on documentation gives bibliographic references, synonomy, and citation of types and selected specimens. Reference matter in the form of a glossary, references cited, the general reference maps, an index, and two fascicles of absolutely beautiful colour photographs complete the work. This is a magnificent book which should be on the shelf of anyone interested in cacti. In addition, the introductory chapters will serve any introductory course in botany as a most useful text. WILLIAM J. CODY Biosystematics Research Institute, Agriculture Canada, Central Experimental Farm, Ottawa, Ontario K1A 0C6 138 ENVIRONMENT The Nature of the Stratigraphical Record By Derek V. Ager. 1981. Second edition. Halsted (Wiley), Somerset, New Jersey. 122 pp., illus. U.S. $17.95. Stratigraphy, the branch of geology that treats the forma- tion, composition, sequence and correlation of stratified rocks, forms the basis of our understanding of the earth’s history. Ager believes that most stratigraphers have been immersed in the intricasies of the rock strata, and really have missed the geology for looking. He takes a more general overview by citing examples world-wide and drawing inter- continental comparisons of various geological formations. In doing so, he challenges the accuracy of some fundamental concepts that have shaped the science of stratigraphy from its beginnings. The book is divided into seven short chapters, each of which focuses on one stratigraphical principle, and ends with Ager’s rewritten principle. For example, every student of geology is familiar with the law of uniformitarianism (intro- duced in the mid !700’s) which dictates that “the present is the key to the past”. In light of the vast number of biotic extinctions evident in the fossil record, uniformitarianism is not always applicable, which prompts Ager to write; “Paleontologists cannot live by uniformitarianism alone”. In another chapter, Ager defines the “Principle of the Golden Spike” as a replacement for the International Code of Strati- graphic Nomenclature. Currently, geologists identify type sections, comparable to biological type specimens, for com- parison and correlation of stratigraphic units of similar age from other regions. However, Ager points out that contem- poraneous strata can differ due to local depositional or biotic factors, in turn superficially suggesting noncontempora- neity. Alternatively, he prefers to identify and correlate on the basis of marker boundaries (golden spikes) between rock units. THE CANADIAN FIELD-NATURALIST Vol. 97 His thoughts are summarized in the eighth and final chap- ter. The overriding message, which prompted Ager to write the book, is simply to draw attention to the inadequacy of traditional, and perhaps outdated, stratigraphic concepts. He goes almost so far as to imply that concepts have shaped current geological thought, rather than geology shaping the concepts. On the scale Ager has chosen to view the stratigra- phic record on, his point is well-taken. However I do not believe that geologists are nearly as blind as Ager may con- tend, since recent stratigraphy books do caution against short-comings of the concepts. Nonetheless, Ager’s goal has been achieved, as evidenced by reactions to his first edition which “ranged from near ecstasy to something approaching scorn”. The book is written in an intelligible, sound, and, as intended, relaxing style, though at times it is too digressive. His examples and arguments provide provocative, stimulat- ing, and interesting reading. Every geologist and hobbiest rock hound should havea copy on their bedside table. Unfor- tunately, I cannot recommend this book to non-specialists unfamiliar with the technical jargon, the geological time- scale, and the very stratigraphical principles Ager chooses to rewrite. Lastly, | must comment on the price. I find it excessive at $17.95 U.S. fora soft cover, mediocre paper stock, 122 pages, and 18 black and white plates. One’s book budget does not go far at these inflated prices. BARRY G. WARNER Department of Biological Sciences, Simon Fraser Univer- sity, Burnaby, British Columbia V5A 1S6 A Bibliography of the Natural History of Middlesex County, Ontario, to the Year 1980 with an Historical Introduction By William W. Judd. 1981. Phelps Publishing Company, 87 Bruce Street, London, Ontario. 157 pp. $12.00. This is a superb regional bibliography that covers the full breadth of natural history in Middlesex County. The com- piler, W. W. Judd, is also the author most often cited — 20 of the 157 pages list his publications. I have only two criticisms. There are 17 pages of items by “Anonymous”. Surely a little more digging in the archives of the organizations involved might have ascertained authorship of more of these items. Unfortunately, Judd fails to make clear to the casual reader that William E. Saunders was the son of William Saunders. Judd has researched a wide range of literature including some little-known and defunct journals in order to produce this useful and attractively printed list. C. STUART HOUSTON 863 University Drive, Saskatoon, Saskatchewan S7N 0J8 1983 NEW TITLES Zoology +A. B. A. checklist: birds of continental United States and Canada. 1982. By G. S. Keith, L. G. Batch, D. D. Gibson, R. G. McCaskie, C. S. Robbins, A. Small, P. W. Sykes, and J. A. Tucker. 2nd edition. American Birding Associa- tion, Austin, Texas. 80 pp. U.S. $9 plus U.S. $2.25 shipping. Advances in marine biology, volumes 19 and 20. 1982. Edited by J. H. S. Blaxter, F. Russell, and M. Yonge. Aca- demic Press, New York. 394 pp., and 448 pp. U.S. $16.50 and U.S. $65. Animal — sediment relationships. 1982. Edited by P. L. McCalland M. J. S. Trevesz. Plenum, New York. c260 pp. U.S. $42.50. Avian biology, volume 6. 1982. Edited By Donald S. Farner, James R. King, and Kenneth C. Parkes. Academic Press, New York. 512 pp. U.S. $65. +A bibliography of the sticklebacks (Gasterosteidae: Ostreichthyes). 1982. By Brian W. Coad. Syllogeus No. 35. National Museum of Natural Sciences, Ottawa. 142 pp. Free. The biology of Crustacea, Volumes I to 4. 1982. Editor-in- chief Dorothy E. Bliss. Volume |: systematics, the fossil record, and biogeography. Edited by Lawrence G. Abele; volume 2: embryology, morphology, and genetics. Edited by Lawrence G. Abele; volume 3: neurobiology structure and function. Edited by Harold L. Atwood and David C. Sandeman; volume 4: neural integration and behaviour. Edited by David C. Sandeman and Harold L. Atwood. Academic Press, New York. 336 pp., 432 pp., 512 pp., and 352 pp. U.S. $38.50, U.S. $42.50, U.S. $49.50, and U.S. $36. Biology of lampreys, volume 3. 1982. Edited by M. W. Hardisty and I. C. Potter. Academic Press, New York, 484 pp. U.S. $81. Biology of the Reptilia, volume 12: physiological ecology. 1982. Edited by Carl Gans and F. Harvey Pough. Academic Press, New York. 570 pp. U.S. $79.50. Bird migration in Africa: movements between six conti- nents, volumes 1 and 2. 1981. By Kai Curry-Lindahl. Aca- demic Press, New York. 444 pp., and 251 pp. U.S. $85 and U.S. $42.50. +The birds of Africa, volume 1. 1982. By Leslie H. Brown, Emil K. Urban, and Kenneth Newman. Academic Press, New York. 536 pp., illus. U.S. $99. *A bird-finding guide to Ontario. 1982. By Clive E. Good- win. University of Toronto Press, Toronto. 248 pp. Paper $12.50. British Anthozoa: keys and notes for the identification of the species. 1981. By R. L. Manuel. Academic Press, New York. 250 pp. U.S. $18. BOOK REVIEWS 139 Canadian Atlantic offshore fishery atlas. 1982. Edited by D. J. Scarratt. Revised edition. Canadian Special Publica- tion of Fisheries and Aquatic Sciences, 47. Supply and Services Canada, Hull. 101 pp. $8 in Canada; $9 elsewhere. Also available in French. *A celebration of birds: the life and art of Louis Agassiz Fuertes. 1982. By Robert McCracken Peck. Walker, New York. 178 pp., illus. U.S. $30. The Columbia River salmon and steelhead trout: their fight forsurvival. 1981. By A. Netboy. University of Washington Press, Seattle. 180 pp. U.S. $13.95. East African mammals: an atlas to evolution in Africa. 1982. By Jonathan Kingdon. Volume 3C: bovids, horned ungulates; volume 3D: bovids, horned ungulates (con- tinued). Academic Press, New York. 404 pp. and 358 pp. U.S. $92.50 each. Ecology of bats. 1982. Edited by Thomas H. Kunz. Ple- num, New York. c410 pp. U.S. $49.50. *The emergence of ornithology as a scientific discipline, 1760-1850. 1982. By Paul Lawrence Farber. Reidel (Dis- tributed by Kluwer, Boston). xxi + 191 pp., illus. U.S. $39.50. Environmental physiology of fishes. 1981. Edited by M. A. Al. Plenum Press, New York. x1+ 732 pp. U.S. $69.50. Feeding strategy. 1982. By Jennifer Owen. University of Chicago Press, Chicago. 158 pp., illus. U.S. $10.95. +Fifth annual (1982) spring migration report: Point Pelee National Park and vicinity. 1982. By Alan Wormington. Friends of Point Pelee and Parks Canada, Leamington. 21 pp., illus. No price given. +Fisheries and wildlife resources and the agricultural land base in Alberta. 1982. By William M. Glasgow. Alberta Energy and Natural Resources, and the Environmental Council of Alberta, Edmonton. 65 pp., illus. Free. Foraging behaviour: ecological, ethological, and psycho- logical approaches. 1981. Edited by A. C. Kamiland T. D. Sargent. Garland STPM Press, New York. 448 pp. U.S. $45. Great blue: the odyssey of a great blue heron. 1980. By M. R. Crowell. Time, New York. 143 pp. U.S. $10.95. *Gulls: a guide to identification. 1982. By Peter Grant. Buteo Books, Vermillion, South Dakota. 280 pp., illus. U.S. $32.50. Gulls: an ecological history. 1982. By Frank Graham, Jr. VanNostrand Reinhold, New York. vi+ 179 pp., illus. U.S. $8.95. 140 THE CANADIAN FIELD-NATURALIST Harper and Row’s complete field guide to North American wildlife-eastern edition. 1981. By H. H. Collins. Harper and Row, New York. xii+ 714 pp., illus. Cloth U.S. $17.50; paper U.S. $12.95. Hibernation and torpor in mammals and birds. 1982. By Charles P. Lyman, John S. Willis, Andre Malan, and Law- rence C. H. Wang. Academic Press, New York. 323 pp. U.S. $37.50. The integrated study of bird populations. 1981. Edited by H. Klompand J. W. Woldendorp. North-Holland Publish- ing, New York. 256 pp. U.S. $41.50. Insect clocks. 1982. By D. S. Saunders. 2nd edition. Per- gamon Press, Elmsford, New York. 420 pp., illus. Cloth U.S. $103.50; paper U.S. $55. Mammalia Africana. 1981. By Jonathan Kingdon. Aca- demic Press, New York. 64 pp., illus. U.S. $10.50. Mammal-like reptiles and the origin of mammals. 1982. By T.S. Kemp. Academic Press, New York. 378 pp. U.S. $44.50 *The mammals of Minnesota. 1982. By Evan B. Hazard. University of Minnesota Press, Minneapolis. x11+ 280 pp., illus. Cloth U.S. $39.50; paper U.S. $15.95. +Mammifeéres du Québec et de l’est du Canada, tomes | et 2. 1982. Par Jacques Prescott et Pierre Richard. Editions France-Amérique, Montréal. xii + 199 pp., illus., et xii + pp. 200-429. $11.95 et $11.95. +Man and wildlife in a shared environment. 1982. Edited by Roy Vontobel. Canadian Wildlife Service, Ottawa, 55 pp., illus. Free. + Migration, harvest, and population dynamics of mourning doves banded in the central management unit, 1967-77. 1982. By J.H. Dunks, R. E. Tomlinson, H. M. Reeves, D. D. Dolton, C. E. Braun, and T. P. Zapatka. Special Scientific Report — Wildlife No. 249. 128 pp., illus. Free. Parental care in mammals. 1981. Edited by D. J. Guber- nick and P. H. Klopfer. Plenum, New York. xx + 460 pp. U.S. $39.50. Perspectives in ethology, volume 5: ontogeny. 1982. Edited by P. P. G. Bateson and Peter H. Klopfer. Plenum, New York. 500 pp. U.S. $39.50. Plankton and productivity in the oceans, volume 2: zoo- plankton. 1982. By J. E.G. Raymont. 2nd edition. Per- gamon Press, Elmsford, New York. 700 pp., illus. Cloth U.S. $86.50; paper U.S. $21.90. Population biology of tropical insects. 1982. By Allen M. Young. Plenum, New York. 495 pp. U.S. $57.50. Principles of protozoology. 1983. By J. O. Corliss. Per- gamon Press, Elmsford, New York. c350 pp. Cloth U.S. $46; paper U.S. $23. Vol. 97 Problems in management of locally abundant wild mam- mals. 1982. Edited by Peter A. Jewell, Sidney Holt, and Donna Hart. Academic Press, New York. 374 pp. U.S. $23.50. +The rare breeding birds of Ontario. 1982. By Paul F. J. Eagles and John D. McCauley. Biology Series Number 24. University of Waterloo, Waterloo. 26 pp., illus. No price given. Harper and Row’s complete field guide to North American wildlife: western edition. 1981. By J. E. Ransom. Harper and Row, New York. xiit+ 810 pp. Cloth U.S: $17.50; paper U.S. $12.95. Rattlesnakes: their habits, life histories, and influence on mankind, abridged edition. 1982. By Laurence M. Klauber. University of California Press, Los Angeles. xxii+ 350 pp., illus. U.S. $19.95. Red deer: behavior and ecology of two sexes. 1982. By T. H. Clutton-Brack, F. E. Guinness, and S. D. Albon. University of Chicago Press, Chicago. c400 pp. Cloth U.S. $37.50; paper U.S. $12.95. Saving America’s birds. 1982. By Paula Hendrich. Lothrop, Lee and Shepard, New York. 160 pp., illus. U.S. $10.50. Selections from the distribution and abundance of ani- mals. 1982. By H. G. Andrewartha and L. C. Birch. Uni- versity of Chicago Press, Chicago. 288 pp., illus. Cloth U.S. $25; paper U.S. $7.95. Social insects, volumes 1 to 4. 1979, 1981, 1982, 1982. Edited by Henry R. Hermann. Academic Press, New York. 456 pp., 454 pp., 480 pp., and 416 pp. U.S. $49.50, U.S. $55, U.S. $58, and U.S. $52. Telemetric studies of vertebrates. 1982. Edited by C. L. Cheeseman and R.B. Mitson. Proceedings of asymposium, London, England, 21-22 November, 1981. Academic Press, New York. 388 pp. U.S. $54. Wildlife of the rivers. 1981. By W. H. Amos. Abrams, New York. 232 pp. U.S. $19.95. *Wild mammals of North America: biology, management, and economics. 1982. Edited by Joseph A. Chapman and George A. Feldhamer. Johns Hopkins University Press, Baltimore. xu + 1147 pp., illus. U.S. $50. A world guide to whales, dolphins, and porpoises. 1981. By D. S. Heintzelman. Winchester Press, Tulsa. 156 pp. U.S. $9.95. Botany Air pollution and forests: interactions between air contam- inants and forest ecosystems. 1981. By W.H. Smith. Springer-Verlag, New York. 379 pp. U.S. $29.80. 1983 tAgaves of continental North America. 1982. By Howard Scott Gentry. University of Arizona Press, Tucson. illus. U.S. $49.50. *Botany: plant biology and its relation to human affairs. 1982. By Jean H. Langenheim and Kenneth V. Thimann. Wiley, New York. xi+ 624 pp., illus. U.S. $27.95. Common plants of the mid-atlantic coast: a field guide. 1982. By Gene M. Silberhom. Johns Hopkins University Press, Baltimore. xii1 + 256 pp., illus. Cloth U.S. $17.50; paper U.S. $9.95. tDynamics of the shore vegetation of a north Swedish hydro-electric reservoir during a 5-year period. 1981. By Christer Nilsson. Acta Phytogeographica Suecica 69. Almgavist and Wiksell, Stockholm. 94 pp., illus. No price given. *A flora of Waterton Lakes National Park. 1982. By Job Kuyt. University of Alberta Press, Edmonton. xxiv + 684 pp., illus. $21. +Flowers of the wild: Ontario and the Great Lakes Region. 1982. By Zile Zichmanis and James Hodgins. Oxford Uni- versity Press, Don Mills, Ontario. xv + 272 pp., illus. $35. Handbook of seagrass biology: an ecosystem perspective. 1981. By R. C. Phillips. Garland STPM Press, New York. 450 pp. U.S. $45. {Plants of Essex County: a preliminary list. 1981. Compiled by Wilfred Botham. Essex Region Conservation Authority, Essex, Ontario. xiv + 223 pp. Limited edition. $5.95 plus $2.55 postage. Environment An annotated reader in environmental planning and management. 1982. By IT. O’Riordan and R. K. Turner. Pergamon Press, Elmsford, New York. 352 pp., illus. Cloth U.S. $52; paper U.S. $23. An arctic ecosystem: the coastal tundra at Barrow, Alaska. 1981. Edited by L. Bunnell. Dowden, Hutchinson and Ross, Stroudsburg, Pennsylvania. xxvill + 527 pp. U.S. $29.50. Biotic contents of Spooky Hollow Sanctuary and Short Hills Wilderness Area: nature reserves owned by Hamilton Naturalists’ Club. 1982. By Hamilton Naturalists’ Club, Hamilton, Ontario. $4 + postage. Camouflage and mimicry. 1982. By Denis Owen. Univer- sity of Chicago Press, Chicago. 158 pp., illus. Paper U.S. $10.95. Coastal area management and development. 1982. By the United Nations Department of International Economic and Social Affairs. Pergamon Press, Elmsford, New York. 196 pp., illus. U.S. $46. BOOK REVIEWS 141 The ecology of tomorrow’s world: industry’s environment. 1981. By J. Elkington. Halsted Press (Wiley), New York. xii + 312 pp. U.S. $29.95. Environmental biology for engineers: a guide to environ- mental assessment. 1981. By G. Camougis. McGraw-Hill, New York. 214 pp. U.S. $19.95. Environmental oceanography: an introduction to the behaviour of coastal waters. 1983. By T. Beer. Pergamon Press, Elmsford, New York. c280 pp., illus. Cloth U.S. $46; paper U.S. $16.10. The human impact: man’s role in environmental change. 1982. By Andrew Goudie. MIT, Cambridge, Massachu- setts. x+ 316 pp., illus. Cloth U.S. $22.50; paper U.S. $10. International organization and the conservation of nature. 1981. By R. Boardman. Indiana University Press, Bloom- ington. 224 pp. U.S. $22.50. Life in the sea. 1982. By Scientific American. Freeman, San Francisco. vill + 248 pp., illus. Cloth U.S. $24.95; paper U.S. $11.95. Making pollution prevention pay: ecology with economy as policy. 1983. Edited by D. Huisingh and V. Bailey. Per- gamon Press, Elmsford, New York. 165 pp., illus. U.S. $28.80. Oceanography and marine biology: an annual review, volume 20. 1982. By M. Barnes. Pergamon Press, Elms- ford, New York. 778 pp., illus. U.S. $95. + An overview of the environmental impacts of forestry, with particular reference to the Atlantic Provinces. 1982. By Bill Freedman. Institute for Resource and Environmental Sciences, Dalhousie University, Halifax. 219 pp. $10. + Private options: tools and concepts for land conservation. 1982. By the Montana Land Reliance and Land Trust Exchange. Island Press, San Rafael, California. 310 pp. UESHS25: Resource management and environmental uncertainty. 1981. Edited by M. H. Glantzand J. D. Thompson. Wiley- Interscience, Somerset, New Jersey. 419 pp. U.S. $42.50. Sexual strategy. 1982. By Tim Halliday. University of Chi- cago Press, Chicago. 158 pp., illus. U.S. $10.95. These are endangered. 1981. By C. L. Cadieux. Stone Wall Press, Washington. 288 pp. U.S. $15. + Worms eat my garbage. 1982. By Mary Appelhof. Flower Press, Kalamazoo. 110 pp., illus. U.S. $5.95 plus $1 shipping. Miscellaneous Charles Darwin: a commemoration 1882-1982. 1982. Edited by R. J. Berry. Reprinted from the Biological Jour- 142 THE CANADIAN FIELD-NATURALIST nal of the Linnean Society, volume 17, number |. Academic — Press, New York. 126 pp. U.S. $12.50. A day in the life of a marine biologist. 1982. By William Jaspersohn. Little, Brown, Boston. 96 pp., illus. U.S. $10.95. +K-TEC II: Cretaceous — Tertiary extinctions and extrater- restrial causes. 1982. Edited by D. A. Russell and G. Rice. Proceedings of a workshop, Ottawa, May, 1981. Syllogeus No. 39. National Museum of Natural Sciences, Ottawa. 151 pp. Free. *A manual of outdoor photography. By Michael Freeman. Zift Davis, New York. 224 pp., illus. U.S. $14.95. Natural resources information directory. 1982. By Alberta Energy and Natural Resources, Edmonton. $5. +Ocean yearbook 3. 1982. Edited by Elisabeth Mann Bor- gese and Norton Ginsburg. University of Chicago Press, Chicago. U.S. $49. Population systems: a general introduction. 1981. By A. A. Berryman. Plenum, New York. 238 pp. U.S. $16.95. + Public hearings on noise in Alberta: report and recommen- dations. 1982. By the Environmental Council of Alberta, Edmonton. 175 pp., illus. Free. +Spruce Woods Provincial Park. 1982. By Manitoba Department of Natural Resources, Winnipeg. 32 pp., illus. $2.50. +Station lists of marine biological expeditions of the National Museum of Natural Sciences in the North Ameri- can pacific coastal region, 1966 to 1980. 1982. By E. L. Bousfield and Norma E. Jarrett. Syllogeus No. 34. National Museum of Natural Sciences, Ottawa. 66 pp., illus. Free. Synopsis and classification of living organisms, volume 1 and 2. 1982. Edited by Sybil P. Parker. McGraw-Hill, New York. 2 volumes, illus. U.S. $149.50 set. W. E. Saunders, naturalist. 1949, 1981. Edited by R. J. Rutter. Reprint. Federation of Ontario Naturalists and University of Toronto Press, Toronto. 66 pp. $5. Vol. 97 Books for Young Naturalists Bears. 1982. By Susan Kuchalla. Troll, Mahwah, New Jer- sey. 28 pp., illus. Cloth U.S. $7.89; paper U.S. $1.95. Billions of bats. 1982. By Miriam Schlein. Lippincott, New York. 56 pp., illus. U.S. $9.50. Cactus: the all-American plant. 1982. By Anita Holmes. Four Winds, New York. viit 178 pp., illus. U.S. $14.95. Carnivorous plants. 1982. By Cynthia Overbeck. Lerner, Minneapolis. 48 pp., illus. U.S. $8.95. Diary of a rabbit. 1982. By Lilo Hess. Scribner’s, New York. 48 pp., illus. U.S. $10.95. How did we find out about life in the deep sea? 1982. By Isaac Asimov. Walker, New York. 61 pp., illus. U.S. $7.95. The Japanese crane: bird of happiness. 1981. By Dorothy Britton and Tsuneo Hayashida. Kodansha (Distributed by Harper and Row, New York.) 64 pp., illus. U.S. $15.50. Moose. 1981. By Jack Denton Scott. Putman’s, New York. 64 pp., illus. U.S. $9.95. Nature’s clean-up crew: the burying beetles. 1982. By Lous J. and Margery Milne. Dodd, Mead, New York. 62 pp., illus. U.S. $7.95. Some birds have funny names. 1981. By Diana Harding Cross. Crown, New York. 46 pp., illus. U.S. $7.95. Sun power: the story of solar energy. 1982. By Madeleine Yates. Abingdon, Nashville. 80 pp., illus. U.S. $8.95. Turtles. 1982. By Janet Craig. Troll, Mahwah, New Jersey. 30 pp., illus. Cloth U.S. $7.89; paper U.S. $1.95. Wonders of egrets, bitterns, and herons. 1982. By Wyatt Blassingame. Dodd, Mead, New York. 80 pp., illus. U.S. $7.95. * Assigned for review t Available for review TABLE OF CONTENTS (concluded) Distribution of the Bager, Taxidea taxus, in southwestern Ontario WAYNE M. LINTACK and DENNIS R. VOIGT Responses of two groups of Mountain Goats, Oreamnos americanus, to a Wolf, Canis lupus CHRISTIAN A. SMITH Occurrences of the Black-legged Kittiwake, Rissa tridactyla, in the Prairie Provinces and north-central United States ALAN R. SMITH and E. Kuyt Breeding records of Northern Shoveler, Anas clypeata, along the northern coast of Ontario R. KENYON ROss and NORMAN R. NORTH Black Bear, Ursus americanus, predation on a Mule Deer fawn, Odocoileus hemionus E. VERSPOOR Sphagnum at Prudhoe Bay, Alaska PETER D. SPATT Northern range extension of the Two-lined Salamander, Eurycea bislineata, in Ontario JAMES KAMSTRA A second record of the Deer Mouse, Peromyscus maniculatus, from Newfoundland M. C. BATEMAN Dwarf Clearweed, Pilea pumula (Urticaceae) — new to Nova Scotia N. HILL and S. P. VANDER KLOET News and Comment The American Association for the Advancement of Science: Arctic Division — Position Available October — Notice of Publication Gamma-Ray irradiation of a boreal forest ecosystem: The Field Irradiator - Gamma (FIG) facility and research programs JOHN E. GUTHRIE and JANET R. DUGLE Book Reviews Zoology: Bears: their biology and management — Mammalian population genetics — Summer birds of the Northwest Angle Provincial Forest and adjacent southeastern Manitoba, Canada — Wildlife population ecology — Fish gene pools — An atlas of the birds of the western Palaearctic — Arctic animal ecology — Handbook of census methods for terrestrial verte- brates — The fishes of the islands in the South China Sea — A study of the lateral-line canals system and that of Lorenzini ampullae and tubules of elasmobranchiate fishes of China Botany: Shrubs of Ontario — Die Kiistenvegetation Ostkanadas — Lichens: an illustrated guide —The illustrated flora of Illinois: Flowering plants, Basswoods to Spurges — The Cacti of the United States and Canada Environment: The nature of the stratigraphical record — A bibliography of the natural history of Middlesex County, Ontario, to the year 1980 with an historical introduction NEW TITLES Mailing date of previous issue, volume 96, number 4: 18 April 1983 107 110 111 113 114 115 116 117 118 119 120 129 THE CANADIAN FIELD-NATURALIST Volume 97, Number 1 Articles Nesting biology of solitary wasps and bees in the Eastern Townships region, Quebec STEPHEN GODFREY and DONALD F. J. HILTON Fungus fairy rings in soil: etiology and chemical ecology JOHN I. TOOHEY Island biogeography of seed plants in Lake Nipigon, Ontario KEVIN P. TIMONEY Distribution sur les coniféres des lichens appartenant aux genres Bryoria, Alectroia, Usnea et Ramalina sur la Cote Nord et la Cote Sud du Golfe Saint-Laurent M. LAFLAMME-LEVESQUE, J. M. PERRON et L. JOBIN Amphibians of the Province of Newfoundland JOHN E. MAUNDER Fish predation and other distinctive features in the diet of Nogies Creek, Ontario, Largemouth Bass, Micropterus salmoides JAMES G. HAMILTON and P. M. POWLES Effects of Moose, Alces alces, on aquatic vegetation in Sibley Provincial Park, Ontario D. FRASER and H. HRISTIENKO Survival of female Black Ducks, Anas rubripes, during the breeding season JAMES K. RINGELMAN and JERRY R. LONGCORE Winter movements of Arctic Foxes, Alopex lagopus, in a petroleum development area LESTER E. EBERHARDT, ROBERT A. GARROTT, and WAYNE C. HANSON Habitat selection and food habits of Marten, Martes americana, in the Northwest Territories RICHARD J. DOUGLASS, LORNE G. FISHER, and MARNIE MAIR Taxonomy of the Gaspé Shrew, Sorex gaspensis, and the Rock Shrew, S. dispar THOMAS W. FRENCH and GORDON L. KIRKLAND, JR. Home range size, movements and habitat use in two Moose, Alces alces, populations in southeastern Alaska JOSEPH G. DOERR Notes Northern records of Risso’s Dolphin, Grampus griseus, in the northeast Pacific HOWARD W. BRAHAM The Swamp Saxifrage, Saxifraga pensylvanica, a rare plant in Canada, newly discovered in Saskatchewan VERNON L. HARMS Small game hunting behaviour of Polar Bears, Ursus maritimus GARY D. MILLER and DONALD R. WOOLDRIDGE Birds first described from Hudson Bay C. STUART HOUSTON The occurrence and origin of Tiger Trout, Sa/mo trutta X Salvelinus fontinalis, in Ontario streams LARRY D. WITZEL Two Sable Island fungi, Peziza ammophila and Hygrocybe turunda, new to Nova Scotia S. A. REDHEAD and P. M. CATLING Observations on primary dispersal of White Spruce, Picea glauca, seed JOHN C. ZASADA and DONALD LOVIG A golden-yellow colored Sablefish, Anoploma fimbria, caught off Quatsino Sound, British Columbia D. A. NAGTEGAAL 1983 26 35 47 >] 62 66 71 75 79 89 g)! 93 95 99 102 104 106 concluded on inside back cover ISSN 0008-3550 The CANADIAN FIELD-NATURALIST Published by THE OTTAWA FIELD-NATURALISTS’ CLUB, Ottawa, Canada Volume 97, Number 2 April-June 1983 The Ottawa Field-Naturalists’ Club FOUNDED IN 1879 Patrons Their Excellencies the Governor General and Mrs. Edward Schreyer Tne 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 information on these fields as widely as possible; to support and cooperate with organizations engaged in preserving, maintaining or restoring environments of high quality for living things. Honorary Members William J. Cody Clarence Frankton Thomas H. Manning Loris S. Russell Mary E. Stuart William G. Dore W. Earl Godfrey George H. McGee Douglas B. O. Savile Sheila Thomson R. Yorke Edwards Louise de K. Lawrence Hugh M. Raup Pauline Snure 1983 Council President: D. F. Brunton W.R. Arthurs E. M. Dickson B. M. Marwood Vice-President: P. M. Catling R. E. Bedford S. Gawn E. G. Munroe Vice-President: J. K. Strang B. A. Campbell G. G. Gruchy K. W. Taylor Recording Secretary: E. F. Pope W. J. Cody G. M. Hamre R. Taylor Corresponding Secretary: W. K. Gummer Ee RaCook D. R. Laubitz P. S. Walker Treasurer: P. D. M. Ward S. J. Darbyshire P. M. D. Martin Those wishing to communicate with the Club should address correspondence to: The Ottawa Field-Naturalists’ Club, Box 3264, Postal Station C, Ottawa, Canada KIY 4J5. Forinformation on Club activities telephone (613) 722-3050. 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. Editor: Francis R. Cook, Herpetology Section, National Museum of Natural Sciences, National Museums of Canada, Ottawa, Ontario KIA 0M8 Copy Editor: Louis L’Arrivée; Assistant to the Editor: Barbara Stewart Business Manager: Mr. W. J. Cody, Box 3264, Postal Station C, Ottawa, Ontario, Canada KIY 4J5 Book Review Editor: Dr. J. Wilson Eedy, R. R. 1, Moffat, Ontario LOP 1J0 Coordinator, The Biological Flora of Canada: Dr. George H. La Roi, Department of Botany, University of Alberta, Edmonton, Alberta T6G 2E9 Associate Editors: Gy Ds Bird A. J. Erskine D. E. McAllister Stephen M. Smith E. L. Bousfield Charles Jonkel WO Pruitt Ir: C. G. Van Zyll de Jong Chairman, Publications Committee: R. E. Bedford All manuscripts intended for publication should be addressed to the Editor. Urgent telephone calls may be made to the Editor’s office (613- 996- 1755), or his home on evenings and weekends (613-269-3211), or to the Business Manager’s office (613-996-1665). Subscriptions and Membership Subscription rates for individuals are $17 per calendar year. Libraries and other institutions may subscribe at the rate of $30 per year (volume). The Ottawa Field-Naturalists’ Club annual membership fee of $17 includes a subscription to The Canadian Field- Naturalist. Subscriptions, applications for membership, notices of changes of address, and undeliverable copies should be mailed to: The Ottawa Field-Naturalists’ Club, Box 3264, Postal Station C, Ottawa, Canada K1Y 4J5. Second Class Mail Registration No. 0527 — Return Postage Guaranteed. Back Numbers and Index Most back numbers of this journal and its predecessors, Transactions of The Ottawa Field- Naturalists’ Club, 1879-1886. and The Ottawa Naturalist, 1887-1919, and Transactions of The Ottawa Field- Naturalists’ Club and The Ottawa Naturalist — Index compiled by John M. Gillett, may be purchased from the Business Manager. Cover: White Malaxis, Malaxis monophyllos var. diphyllos, a new orchid for Canada, photographed by William J. Beese, see pp. 215-216. The Canadian Field-Naturalist Volume 97, Number 2 April-June 1983 Movements of Collared Caribou, Rangifer tarandus, in Relation to Petroleum Development on the Arctic Slope of laste KENNETH R. WHITTEN and RAYMOND D. CAMERON Alaska Department of Fish and Game, 1300 College Road, Fairbanks, Alaska 99701 Whitten, Kenneth R., and Raymond D. Cameron. 1983. Movements of collared Caribou, Rangifer tarandus, in relation to petroleum development on the Arctic Slope of Alaska. Canadian Field-Naturalist 97(2): 143-146. Between April 1975 and May 1978, 160 Caribou (Rangifer tarandus granti) on the Central Arctic Slope of Alaska were marked with either visual numbered collars or radio collars. Through 1979, 92% of the Caribou with radio collars were relocated repeatedly within this region. Only 59% of the Caribou with visual poles were relocated, but comparable data on radio-collared Caribou suggest that most of the unobserved number-collared individuals remained in the study area. Resighting patterns corroborate the existence of a distinct Central Arctic Caribou Herd. A significantly higher proportion of collared bulls than of collared cows was observed from the road system associated with the Trans-Alaska Pipeline and Prudhoe Bay Industrial Area. Bulls were also resighted more frequently from the road system, and they crossed through the road corridor more often; in contrast, bull and cow resighting patterns in off-road areas were not significantly different. Thus, the cow/calf segment of the herd appeared to avoid disturbed areas more so than did bulls. The heavily developed Prudhoe Oilfield was an effective barrier to both bulls and cows. Key Words: Caribou, Rangifer tarandus, collared, movements, petroleum, development, disturbance. Construction of the Trans-Alaska Pipeline (TAP) and development of oil reserves near Prudhoe Bay focused considerable attention on the possible disrup- tion of Caribou movements and range use patterns. Results of systematic aerial surveys conducted since 1975 have shown that this region of the Arctic Slope is inhabited by a distinct Caribou subpopulation, the Central Arctic Herd (CAH) (Cameron and Whitten 1979, 1980). Seasonal movements of the CAH are principally north-south between winter range in the northern foothills of the Brooks Range and summer/ calving range on the coastal plain along the Arctic Ocean. In addition, east-west movements along the Arctic coast occur during midsummer. Surveys con- ducted along the Dalton Highway (previously known as the TAP haul road) through 1980 indicate that abnormally few cow-calf pairs occupy habitats within or near the Pipeline Corridor and Prudhoe Bay Indus- trial Area (PBA), particularly during summer (Cameron et al. 1979; Cameron and Whitten 1980, unpublished). In 1975 a complementary collaring program was undertaken to examine, in greater detail, the popula- tion characteristics of Caribou in the Central Arctic region and to assess any disturbance-related devia- tions in seasonal distribution and movements. This report deals with pertinent data gathered from col- lared Caribou between 1975 and 1979. Study Area and Methods The study area is on the Central Arctic Slope between the Canning and Itkillik rivers. The TAP Corridor is oriented north-south, roughly bisecting the study area, and the PBA lies near the Arctic coast at the origin of the TAP. Physiography and floristics of this region have been described by Spetzman (1959) and by Whitten and Cameron (1980). Between April 1975 and May 1978, 160 Caribou were equipped with either visual or radio-transmitter collars (Table 1). Caribou were located opportunisti- cally, generally within 20 km of the Corridor, and darted from a helicopter with Cap-Chur equipment. Anectine (Succinyl-choline choloride) was used to immobilize Caribou in 1975, and a combination of M99 (Etorphine hydrochloride) and Rompun (Diazi- none) was used from 1976 through 1978; in the latter case the antagonist M50-50 (Diprenorphine hydro- chloride) was administered to effect recovery. Visual collars consisted of red or blue background material with conspicuous yellow numbers on the top and sides. Radio-transmitter collars (Oceans Applied Research, San Diego, CA) used in 1975 were not 143 144 TABLE |. Schedule of collar placement on CAH Caribou. Date Radio Collars Visual Collars April 1975 SNe 10F, 1OM October 1975 4F 4F,3M April 1976 TF 44F,3M April-May 1977 12F 20F, 13M May 1978 8F 13F,4M Totals 36F 91F, 33M F = female, M = males numbered, but some were color coded. All other radio collars (AVM Instrument Co., Champaign, IL) were attached to standard visual collars. Caribou with visual collars were sighted incidental to routine ground surveys along the Dalton Highway (Cameron et al. 1979) and aerial surveys of the entire study area (Cameron and Whitten 1979), as well as during unscheduled trips and flights; no special effort was made to relocate any individual. Radio-collared Caribou were located on an opportunistic basis dur- ing most survey flights and were also specifically tracked from fixed-wing aircraft in November and February/ March, and at approximately 10-day inter- vals between April and October each year. Crossings of the TAP Corridor were occasionally witnessed but were more often inferred from consecutive resight- ings. Because sampling effort was presumed to be the same for all visual-collared individuals, any observed differences in resighting patterns between visual- collared bulls and cows and/or between road and off-road areas should reflect corresponding differen- ces in distribution. Chi-square contingency analysis was used to test differences in the proportions of bulls and cows resighted. Mean rates of resighting and Cor- ridor crossings were compared through Student’s t- test, with the activities of individual Caribou serving as the sample unit. Significance was evaluated at the 95% confidence level. Results and Discussion Herd Identity Transmitters used in 1975 functioned less than eight months, but all Caribou radio-collared that year were subsequently accounted for by known mortality, re- collaring, or visual sightings. All but three radio transmitters used after 1975 were still operating in 1979. Of the 36 Caribou equipped with radio transmit- ters, one emigrated from the study area, two were not relocated, and I! were never seen without the aid of radio-tracking equipment. Thus, only 22 (61%) of the radio-collared Caribou would have been resighted had they worn only visual collars, even though at least 33 (92%) were present in the study area. THE CANADIAN FIELD-NATURALIST Vol. 97 Seventy-three of the 124 Caribou with visual collars (59%), including 22 of 33 males and 51 of 91 females, were Observed at least once within the study area. These figures are similar to the proportion of radio- collared Caribou that would have been resighted without the aid of radio-tracking equipment. At least six different visual-collared female Caribou were later observed in either the Porcupine or Western Arctic Caribou Herds. Based on the radio collar findings, however, it is likely that most of the unobserved visual-collared Caribou remained in the Central Arc- tic area. High rates of resighting of both radio- and visual- collared Caribou within the study area provide evi- dence for the discreteness of the CAH. At least 59%, and probably more than 90%, of the Caribou marked in the Central Arctic area remained within that area over a 4-5 year period. Demographic data, including population size, recruitment, mortality, and seasonal movements also indicate that the Central Arctic Herd behaves as a discrete unit (Cameron and Whitten 1979; Whitten and Cameron 1983). The few cases of known emigration of CAH Caribou support the con- tention that, while Alaskan Caribou occur in separate herds, interchange is sufficiently frequent that all Caribou in the state constitute a single breeding popu- lation (Skoog 1968). To our knowledge, this 1s the first fully documented report of inter-herd movements in Alaska. Caribou Occupancy of the Tap Corridor and PBA Based on the tendency for female Caribou and their calves to avoid the PBA and TAP Corridor (Cameron et al. 1979, Cameron and Whitten 1980), one might expect differences in the resighting patterns for col- lared cows and bulls observed from the TAP haul road. In fact, the proportions of visual-collared bulls and cows resighted at least once during surveys con- ducted from the road were significantly different (61% of the males and 35% of the females), while the pro- portions resighted during aerial surveys away from the road (39% of the males and 51% of the females) did not differ significantly. The mean number of resightings per visual-collared bull (3.6) and cow (2.0) did not differ significantly. However, this comparison may be misleading, since cows apparently retained collars longer than did bulls. Most resightings of collared bulls occurred within the first calendar year, but cows were frequently seen two or three years after collaring (Table 2). Collars were attached loosely to bulls to accommodate neck swel- ling during rut and, consequently, may have slipped off during winter months when most adult bulls would have been antlerless. Overwinter loss of radio collars by bulls has been observed in other Alaskan Caribou 1983 TABLE 2. Relationship of last resighting to date of collaring for visual-collared Caribou. Last resighting (years past collaring date) Year Cohort Collared <1 | 2 3 4 Males 1975 2 4 3 0 0 1976 l 0 0 0 1977 9 | 0 1978 2 0 64%! 23% 14% 0% 0% Females 1975 2 | | 3 | 1976 5 6 7 5 1977 8 4 3 1978 4 2 37% 25% 2% 15% 2% ‘Percentage of last resightings occurring in year group. (J. Davis and P. Valkenburg, personal communica- tion). Incontrast, collars on cows were tightly secured and were less likely to be lost during the antlerless period each summer. Considering only Caribou that were resighted, and assuming collar retention only until last date seen, the resighting rates per unit time were significantly higher for bulls than for cows (X = 11.9 resightings/ year per individual bull vs. X = 4.1 resightings/ year per indi- vidual cow). When these adjusted resighting rates were subdivided into observations adjacent to and away from the road, the rate for bulls along the road was significantly higher than that for cows (X = 9.6 resightings/ year vs. X = 2.4 resightings/ year), while the rates away from the road did not differ signifi- cantly (X = 5.3 resightings/year vs. X = 3.0 resight- ings/ year). Clearly, a higher proportion of the col- lared bulls was resighted from the road system, and those resighted along the road were seen more often than were the collared cows. Away from the road system there were no significant differences in bull and cow resighting patterns. These data are consistent with previous reports of cow/calf avoidance of the Corridor (Cameronetal. 1979; Cameron and Whitten 1980). Crossings of the TAP Corridor Crossings of the TAP Corridor also reflect differen- tial use of the area by bulls and cows. A significantly higher proportion of visual-collared bulls than cows crossed the corridor (68% of all bulls resighted ys. 41% of all cows resighted). Among the visual-collared Caribou resighted, bulls also crossed more frequently (X = 6.3 crossings/ year) than did cows (X = 2.1 cross- ings/ year). However, recognizable bull groups were often observed on successive trips along the haul road, WHITTEN AND CAMERON: MOVEMENTS OF CARIBOU 145 while individual cow groups were rarely seen more than once. Repeated crossings by collared bulls in such “resident” groups inflate the mean crossing rate for bulls and may not be an accurate reflection of annual crossing activity (i.e., seasonal migrations across the Corridor). Nevertheless, these data indicate greater use of the area by bulls. Unfortunately, base- line data are insufficient to determine if recent cross- ing patterns differ from those before road and pipeline placement. A comparison of seasonal movements by radio-collared cows and bulls, disregarding observa- tions from the road, is required to determine if bulls actually do cross the Corridor more readily than cows. Caribou Movements in Relation to the PBA Midsummer movements of large post-calving aggregations of CAH Caribou have frequently included a gradual eastward drift along the Arctic coast during July, followed by a rapid westward movement and inland dispersal in early August (Roseneau et al. 1974; Roseneau and Stern 1974: Cameron and Whitten, unpublished observations). In the early 1970's such movements extended through the PBA(R. White, personal communication). Since 1975, however, no collared Caribou have been observed to move through the PBA during mid- summer. On several occasions, large post-calving groups including both radio- and visual-collared Caribou have approached the oilfield complex from both the east and west but have fragmented and dispersed; only individuals or sma!l groups (mostly adult males) actually entered the field. Conclusions Resightings of collared Caribou support the exist- ence of a distinct Central Arctic Herd. The data also substantiate previous evidence that female Caribou avoid areas of petroleum-related activity within their range; bulls appear to be far less sensitive. Although reduced occupancy of the TAP Corridor by cows and calves suggests a concomitant decrease in crossing success, further study is required to determine if the Pipeline and/or haul road constitute a serious impedi- ment to seasonal movements of Caribou. Nonethe- less, the Prudhoe Bay Oilfield does appear to disrupt midsummer movements of CAH Caribou. Acknowledgments The study was funded by Federal Aid in Wildlife Restoration Projects W-17 and W-21. Alyeska Pipe- line Service Company and Northwest-Alaska Pipeline Company also provided logistics and financial sup- port. W. T. Smith, D. D. Roby, and P. Valkenburg participated in the road and aerial surveys. We are grateful toS. J. Harbo, J. W. Coady, and H. V. Rey- nolds for their critical review of the manuscript. 146 THE CANADIAN FIELD-NATURALIST Literature Cited Cameron, R.D., and K.R. Whitten. 1979. Seasonal movements and sexual segregation of caribou determined by aerial survey. Journal of Wildlife Management 43(3): 626-633. Cameron, R. D., and K. R. Whitten. 1980. Influence of the Trans-Alaska Pipeline Corridor on the local distribution of caribou. /m Proceedings of the Second International Reindeer/Caribou Symposium, Roros, Norway, 1979. Edited by E. Reimers, E. Gaare, and S. Skejenneberg. Direktoratet for vilt og ferskvannsfisk, Trondheim. Cameron, R. D., K. R. Whitten, W. T. Smith, and D. D. Roby. 1979. Caribou distribution and group composi- tion associated with construction of the Trans-Alaska Pipeline. Canadian Field-Naturalist 93(2): 155-162. Roseneau, D. G., and P. M. Stern. 1974. Distribution and movements of the Porcupine caribou herd in northeastern Alaska, 1972. Arctic Gas Biological Report Series Volume 7. 209 pp. Roseneau, D. G., P. M. Stern, and C. Warbelow. 1974. Dis- tribution and movements of the Porcupine caribou herd in northeastern Alaska. Chapter IV, Pages 1-197 in Studies Vol. 97 of large mammal populations in northern Alaska, Yukon, and Northwest Territories, 1973. Edited by K.H. McCourt and L. P. Horstman. Arctic Gas Biological Report Series, Volume 22. 347 pp. Skoog, R. O. 1968. Ecology of caribou ( Rangifer tarandus granti) in Alaska. Ph.D. thesis, University of California, Berkeley. 699 pp. Spetzman, L. A. 1959. Vegetation of the Arctic Slope of Alaska. U.S. Geological Survey Professional Paper 302-B. 42 pp. Whitten, K. R., and R. D. Cameron. 1980. Nutrient dynamics of caribou forage on Alaska’s Arctic Slope. Jn Proceedings of the Second International Reindeer/ Cari- bou Symposium, Roros, Norway, 1979. Edited by E. Rei- mers, E. Gaare, and S. Skejenneberg. Direktoratet for vilt og ferskvannsfisk, Trondheim. Whitten, K.R., and R.D. Cameron. 1983. Population dynamics of the Central Arctic Herd, 1975-1981. Acta Zoologica Fennica 175: 159-161. Received 29 December 1981 Accepted 9 September 1982 An Estimate of the Black Scoter, Melanitta nigra, Population Moulting in James and Hudson Bays R. KENYON ROSS Canadian Wildlife Service, 1725 Woodward Drive, Ottawa, Ontario KIA 0E7 Ross, R. K. 1983. An estimate of the Black Scoter, Melanitta nigra, population moulting in James and Hudson bays. Canadian Field-Naturalist 97(2): 147-150. On 26 and 27 July 1977, an aerial survey was made of the flocks of moulting Black Scoter (Melanitta nigra) off the northern shore of Ontario. On that flight, 88 700 moulting male Black Scoters were counted by means of aerial photography; sucha total could represent up to 320 300 birds in eastern North America in the early fall. Only 7% of that number have been located during wintering ground surveys in the eastern United States. On regaining their powers of flight, most of the Black Scoter apparently moved from the Ontario shore to southeastern James Bay where they staged during August and September prior to the fall migration. Key Words: Black Scoter, Melanitta nigra, James Bay, Hudson Bay, moulting, staging, migration. Each summer moulting Black Scoters (Melanitta nigra) gather in large numbers at several locations off the James and Hudson Bay shoreline (Manning 1952; Manning and Macpherson 1952; Todd 1963; Bellrose 1978). A high proportion of those birds is found off the Ontario coast where their presence has long been known (Manning 1952), but no complete survey has previously been attempted. In late July 1977, the Canadian Wildlife Service undertook a photographic inventory of Black Scoter flocks along the Ontario shore to document flock size and distribution of this poorly-understood species. A population estimate has been generated from these results and this is compared to the relatively low counts made during surveys of the wintering areas along the eastern seaboard of the United States (Bellrose 1978). Methods The survey took place between | 1:00 and 16:00 EST on 26 and 27 July 1977 from a DeHavilland Otter aircraft flying at approximately 160 km per hour at sufficient height (150 m asl.) to avoid causing diving by the ducks; clouds were scattered and visibility was very good. As the flocks were easily visible on the water, a total census was attempted by flying a ziz-zag course along the coast. The outer limit of flock distri- bution was determined by ranging up to I5 km off- shore; no flock was detected more than approximately 10 km from the coast. Each flock was photographed from an almost verti- cal angle using a hand-held camera (Hasselblad 500 EL, 250 mm lens, Ektachrome 200 Professional film ASA 400). An assistant estimated the size of the few small flocks that could not be accommodated along the flight line. The resulting transparencies were pro- jected on gridded paper using a photographic enlarger 147 and all distinguishable birds were marked and counted. The timing of the survey was determined through ground observations of the Black Scoter flock at Long- ridge Point during July and August 1976. Numbers were seen to peak in late July and feeding activity tended to subside in midday thus reducing the proba- bility of not counting individuals because they were submerged. Some birds were undoubtedly under water during the survey but the numbers missed were likely very small (< 5%); survey counts should be considered minima. Results and Discussion The survey revealed 88 700 male Black Scoters in three locations along the Ontario coastline (Table 1, Figure 1). The birds around Longridge Point and Shell Brook were grouped ina few larger flocks (Table 1), mostly within 2 km of shore. Those along the north James Bay shore (192 km of coast) formed much smaller groups which were widely scattered as far as 10 km offshore. The presence of many moulting Black Scoters along the northern Ontario shoreline has long been known. H. G. Lumsden (in Bellrose 1978) estimated that 30 000 were present along the James Bay coast. Such results are similar to those reported here and imply that these are representative rather than chance con- centrations. Other flocks have been reported along the shore of Manitoba (Bellrose 1978) and Québec (Man- ning and Macpherson 1952; Todd 1963; S. G. Curtis, CWS, personal communication). The most systematic records are summarized in Table | and Figure |. S. G. Curtis’ observations were made incidental to a survey of eel grass (Zostera marina) beds and so represent minimum estimates. 148 THE CANADIAN FIELD-NATURALIST Vol. 97 QUEBEC ONTARIO PIAGOCHIOUI LOCATION OF MOULTING BLACK # SCOTERS FiGuRE |. Known distribution of moulting Black Scoters in James and Hudson Bay. TABLE |. Numbers of moulting Black Scoters along the shoreline of James and Hudson Bay (to the nearest hundred birds). Co-ordinates Total No. of | Mean Flock Survey Location N WwW Numbers Flocks Size Date Source Longridge Point 51°50’ 80°42’ 2 400 2 1 200 26 July 1977 This paper Ekwan Point- S326, 82205¢ 42 600 110 387 26 July 1977 This paper Hook Point SAMOS ONO Mouth of Shell SSeS. BSSt3? 43 700 12 3 642 27 July 1977 This paper Brook Mouth of Kettle River 56°99’ 89°22’ 2-3 000 N/A N/A 4 August 1961 Arthur and Vaught, in Bellrose 1978. Paint Hills to 53°00’ =78°48’ Mouth of Piagochioui 54°02’ 79°02’ 2 800 N/A N/A 25 July — S. G. Curtis CWS River to 6 August 1974 (personal communication) Cape Jones 42380 5) 1924 5% 16 700 N/A N/A 25 July — * % 6 August 1974 N/A — not available 1983 The size of the population of Black Scoters repre- sented by those post-breeding moulters may be esti- mated given the following assumptions on population dynamics and distribution: (a) a stable population; (b) 5 to 50% more moulters present along the Québec and Manitoba coasts (Table 1); (c) an overall sex ratio of 1.67 males per female (Bellrose 1978); (d) a fledging success range of 0.75 to 1.00 young per pair (cf. Hildén 1964; Bellrose 1978; Brown and Brown 1981) (e€) an overwintering mortality range for subadults of 65 to 70% (cf. Bellrose 1978); Using all possible combinations of the limits of the population parameters, eight different estimates of population size can be generated; these yield a mean value of 259 300 birds (range, 203 300-320 300) in the early fall. The seasonal distributions and movements of those Black Scoters are poorly understood. The breeding range was initially considered to be centred in Alaska (Kortright 1942); however, the breeding population of Black Scoters in Alaska has been estimated at 235 000, most of which apparently winter along the Aleutian Islands where approximately 250 000 adults and immatures have been noted (Bellrose 1978). Thus, it would appear that a different sub-population nests in the hinterland surrounding James Bay and Hudson Bay. Godfrey (1966) listed very few Canadian breed- ing records including several in southern Keewatin, Ungava, and Newfoundland. P. Lamothe, Québec Hydro (personal communication) observed breeding pairs in the Great Whale and Little Whale River bas- ins in northern Québec in 1976. No confirmed breed- ing records have come from the Ontario Hudson Bay Lowlands, although H. Lumsden, OMNR (personal communication) has occasionally seen possible Black Scoter broods there, and on 21 and 22 May 1982 I observed four pairs on muskeg ponds north of Missisa Lake (52°40’N, 85° 15’W). Also unknown is the distribution of the subadult cohort. In Alaska (Palmer 1976) and Denmark (Joensen 1973), the subadults commence moulting during June in locations later used by moulting breed- ers. Subadults, however, do not moult in James and Hudson Bay, probably because ice conditions are too severe in May. Instead, they apparently remain along the coastlines of the Atlantic provinces and Québec (Palmer 1976). The estimated small number of sub- adults (mean 19 300, range 12 500-27 900) could well be dispersed along such an extensive shore. After the Black Scoters on James and Hudson Bay complete their moult, the birds presumably move to wintering areas along the eastern seaboard of the Ross: BLACK SCOTER POPULATION MOULTING 149 United States. The chronology of the moulting flock at Longridge Point suggests, however, that this movement is not a direct one. In both 1976 and 1977, numbers declined during August (2500 — | August 1976, 800 — 28 August 1976, visual estimated; 2400 — 26 July 1977, 940 — 29 August 1977, photographic counts). The hiatus between their leaving the moulting grounds in August and arriving in more southern regions in late September and October (Quilliam 1973; Palmer 1976) suggests the existence of an inter- mediate staging area. The southeastern James Bay shoreline, especially around various of the outer islands, has consistently held high numbers of all scoter species, with counts apparently peaking during September. A scoter flock estimated at 100 000 birds, including many Black Scoters, was noted around the Cape Hope Islands (52°25’N, 78°45’W) on 12 September 1971 by G. Arsenault and D. I. Gillespie (A. Bourget, CWS, personal communication) and large flocks were observed by S. G. Curtis (personal communication) as far south as Charleton and Strut- ~ ton Islands (52°5’N, 79°10’W) on 10 October 1972. Such large concentrations would accommodate not only birds from Longridge Point but also the much larger moulting flocks further north. S. G. Curtis (personal communication) had records of only approximately 4000 scoters present around the islands of southeastern James Bay between mid-July and mid-August. Black Scoters again seem to disappear on reaching their wintering grounds. Mid-winter inventories along the Atlantic seaboard of the United States have accounted for an average of only 22 000 Black Scoters (Bellrose 1978), i.e. as little as 7% of the wintering population estimated here. Some birds may winter in Hudson Bay as do the local Common Eider race (Somateria mollissima sedentaria), and a few may remain on the Great Lakes, but it is unlikely that those areas account for many birds. Instead, as suggested by the Atlantic Flyway Council (1964), it is more likely that numbers along the eastern seaboard have been greatly under-estimated because the mid-winter sur- veys were not designed specifically for scoters and other ducks which raft far offshore. The location of the missing birds is problematic, although the large concentrations along coastal Georgia and South Carolina (Stott and Olson 1972) suggest a likely area. Significant numbers might also be found along the Gulf Coast of Texas where the Black Scoter has recently become more common (Palmer 1976). More study is needed to solve this problem. Acknowledgments I thank R.I.G. Morrison, N. R. North, K. B. Switzer, P.S. Taylor and D. A. Welsh who partici- 150 pated in the surveys and P. A. M. Angehrn and F. G. Brazeau who analyzed the photographs. A. Bourget and P. Lamothe generously shared unpublished information as did S. G. Curtis who also commented on the manuscript. Literature Cited Atlantic Waterfowl Council. 1964. The Atlantic flyway waterfowl management guide. 99 pp. Bellrose, F.C. 1978. Ducks, geese and swans of North America. 2nd Edition, revised. Stackpole Books, Harris- burg, Pennsylvania. 540 pp. Brown, P. W., and M. A. Brown. 1981. Nesting biology of the White-winged Scoter. Journal of Wildlife Manage- ment 45(1): 38-45. Godfrey, W.E. 1966. The Birds of Canada. National Museum of Canada Bulletin 203. 428 pp. Hildén, O. 1964. Ecology of duck population in the island group of Valassaaret, Gulf of Bothnia. Annales Zoologici Fennici 1(3): 1-279. Joensen, A. H. 1973. Moult migration and wing-feather moult of seaducks in Denmark. Danish Review of Game Biology 8(4): 1-42. THE CANADIAN FIELD-NATURALIST Vol. 97 Kortright, F. H. 1942. The ducks, geese and swans of North America. Stackpole Books, Harrisburg, Pennsyl- vania. 476 pp. Manning, T. H. 1952. Birds of the west James Bay and southern Hudson Bay coasts. National Museum of Can- ada Bulletin 125. 114 pp. Manning, T. H.,and A. H. Macpherson. 1952. Birds of the east James Bay coast between Long Point and Cape Jones. Canadian Field-Naturalist 66: 1-35. Palmer, R.S., Editor. 1976. Handbook of North Ameri- can Birds, Volume 3. Yale University Press, New Haven, Connecticutt. 560 pp. Quilliam, H.R. 1973. History of the birds of Kingston, Ontario. Second edition, revised. Kingston Field Natural- ists, Kingston, Ontario. 209 pp. Stott, R.S., and D. P. Olson. 1972. Differential vulnera- bility patterns among three species of sea duck. Journal of Wildlife Management 36: 775-783. Todd, W. E. C. 1963. Birds of the Labrador peninsula. University of Toronto Press, Toronto. 819 pp. Received 7 April 1982 (revised version 30 August 1982) Accepted 11 December 1982 Spatial Trends in Canadian Snowshoe Hare, Lepus americanus, Population Cycles CHARLES H. SMITH Department of Geography, University of Illinois, Urbana, Illinois 61801 Smith, Charles H. 1983. Spatial trends in Canadian Snowshow Hare, Lepus americanus, population cycles. Canadian Field-Naturalist 97(2): 151-160. Population levels of the Snowshoe Hare ( Lepus americanus) are well known to exhibit cyclic fluctuations. Little attention has been given to detailed analysis of the spatial trends associated with these, however, and in this work an argument and empirical foundation is laid for such studies. Data first reported as “The Canadian Snowshoe Rabbit Enquiry” in the 1930's and 1940’s are reformulated and statistically summarized and mapped. Immediate results indicate: |) the diffusion of the population level change “wave” across Canada is affected by topographic and ecological factors; and 2) a multi-nodal diffusion model is more appropriate than a single-nodal model in understanding the dynamics of the ten-year cycle system. Key Words: population cycles, Snowshoe Hare, Lepus americanus, ten-year cycle, spatial analysis, time series analysis. It has been well known since the late nineteenth century that a number of species of Canadian wildlife undergo regular nine- to ten-year period fluctuations in their population levels. What has not been so well appreciated, however, is the extent to which these fluctuations exhibit spatial regularities in addition to their temporal ones. In this work, data first published as “The Canadian Snowshoe Rabbit Enquiry” in a series of reports in the 1930’s and 1940’s (Elton 1933, 1934; Elton and Swynnerton 1935, 1936; Chitty and Elton 1937, 1938, 1939, 1940; Chitty and Chitty 1942; Chitty and Nicholson 1943; Chitty 1943, 1946, 1948, 1950) have been consolidated to a closer look at this aspect of the ten-year cycle condition. Before 1950, all ten-year cycle studies involved localized field work or descriptive analyses of primary and secondary data. A review of the extensive litera- ture surrounding the first approach is given by Mac- Lulich (1937) and Keith (1963). Those following the second route concentrated on collecting information that could be used to demonstrate firmly that the population cycles in fact existed. One preferred mode of inquiry was the questionnaire study; this allowed compilation of qualitative information on population level trends over time that could be compared withina locational framework. Examples from this period include MacLulich’s (1937) survey of Ontario snow- shoe hare population trends, Chitty and Chitty’s (1941) analysis of several arctic species, Elton and Nicholson’s (1942a) study of muskrats, and the aforementioned Canadian Snowshoe Rabbit Enquiry. Other investigations featured retrieval and interpretation of secondary sources of data — mostly fur trade statistics (Cross 1940; Elton and Nicholson 1942b; Butler 1951) — or attempts to promote causal hypotheses based on correlative relationships (Hunt- ington 1945; Rowan 1950). 15] Statistical analysis of the fur trade and other data began in the 1950’s after Palmgren (1949) and Cole (1951, 1954) expressed concern that more effort should be applied to distinguishing between statisti- cally significant cycles and random series. Moran (1952, 1953, 1954) led the way with examinations of game-bird data and Canada Lynx fur records. Butler (1953) criticized the remarks of Cole and presented evidence of both temporal and spatial population trends in a number of species. It was some twenty years before further statistical measures were applied. Bulmer (1974) published an extensive analysis of several sources of serial data fora number of organisms. In addition to demonstrating the existence of cycles in many of these series, he also arranged his results in such a form as to indicate relative lag periods among the cycles of the creatures involved. This information was later used to presenta model of cycle causation centering on the Snowshoe Hare (Bulmer 1975). Bulmer’s findings were followed up by Smith and Davis(1981), who applied univariate and bivariate spectral time series analysis to examine nineteenth and twentieth century fur data for the Canada Lynx. The results confirmed and extended Bulmer’s findings, and suggested an apparent long term shift in the location of the general nodal region representing highest negative phase lag of population cycling. For further information on the quantitative/ statistical measurement of population cycles and a review of the various causal theories that have been entertained to explain them, consult Finerty (1980). It is now established that; |) population level cycles of near ten years’ period exist in a number of Cana- dian wildlife populations, 2) there are phase lags in the cycles of these populations relative to one another, and 3) there are phase lags within species’ cycles asso- ciated with varying geographic location within the Sy THE CANADIAN FIELD-NATURALIST range of the organism, i.e., the cycles also exhibit the spatial characteristics of a recurring diffusion process. Regarding the last point, no one has moved beyond the identification of this fact to determine whether the patterns of population level change over space can be used themselves to expose the cause of the pheno- menon (though Fox, 1978, has presented an argument based on data aggregated at the provincial level that incidence of fire and population level change are correlated). It seems crucial at this stage in the investigation of ten-year cycles that increasing attention be placed on the system-wide spatial-temporal characteristics of the phenomenon. It is not enough, though increas- ingly the fashion (note the reviews in Finerty 1980), to generate and test organismal, population, or even community level models of cycle relationships and consider the matter solved, because such treatments are unextendable to the discussion of continental-- level patterns. This is a critical issue, because the most diagnostic structural aspect of the ten-year (and four- year) cycle condition is its (their) consistent and con- tinuous expression across vast spatial reaches. The field and modelling approaches used to study popula- tion irruptions (and even cyclic irruptions) are unable to penetrate this problem, whose organizational basis appears to be at some higher plane of departure. More than likely, the condition has devolved gradually over time in response to some accidental interplay of bio- geographical, and perhaps climatological/ geophysi- cal, forces of continental scale. The preceding comments hold whether one believes the cycles to be internally generated or a response to some external driving function. In either case, one may assume that the large scale spatial-temporal characteristics of the condition (that is, phase lag rela- tionships and the like) constitute information that can contribute to an understanding of the underlying cau- sal element. Such modelling would start with the pre- mise that observed cycle characteristics are a function of some causal factor whose effect varies over space (and time?) in some describable fashion. A sufficiently detailed model of the forces acting (as interpreted through the patterns of population change) might very well lead to direct identification of the factor itself, and to predictions at the organismal/ popula- tion level concerning the varying effect of the agent on these over space that could be directly tested. This seems an ultimately more persuasive attack than is now being pursued, since all studies to this date allud- ing to a search for causes can only be considered analyses of the effects of the condition until they are logically linked to the continental scale interactions holding the system together. Such modelling presupposes the availability of an appropriately detailed data base, and this is where the Vol. 97 Canadian Snowshoe Rabbit Enquiry can be useful. This seventeen year long study was the most ambi- tious of the questionnaire-based efforts. In it, hundreds of observers across Canada were asked to submit yearly their impressions of whether the local populations of Snowshow Hare had _ increased, decreased, or exhibited no trend when compared to levels of the year before. Each year these data were plotted on a map of Canada and subjectively dis- cussed. Inasmuch as this continental-level accumula- tion of information on spatial trends in population change is perhaps the most extensive and detailed of its kind in existence for any organism, it is somewhat curious that almost no use has been made of it. In the present work, it was used to generate a base for future spatial modelling approaches to the ten-year cycle problem; it is believed that such modelling will ulti- mately provide the framework within which a com- plete understanding of cycle causation can be constructed. Methods Consideration of how to deal with the seventeen years of data was constrained by several factors. First, most of the locations reporting only did so for por- tions of the entire seventeen year period. It was there- fore impossible to study the spatial relationships of a simple set of series without much — too much — loss of information. Second, the results of the first six years of the study were mapped as area data rather than point data (in the last eleven reports, it was assumed the point symbol mapped represented the center of the area reported on). This necessitated my employing some mode of representation not biased in favor of one or the other original modes. Third, the base map upon which data had been plotted for the years 1931-32 through 1935-36 was replaced with a different projection in 1936-37 that was retained the- reafter. The result was the necessity of devising a transformable sampling grid. Fourth, the absolute validity of each point datum was suspect on the basis of scale considerations: spatially and temporally local events might interfere with the assessment of greater scale trends. In addition to these constraints, there were the more general considerations of how to approach statistically the data of such a short series and how to represent the results graphically. A two-step procedure was applied. In the first step, Thiessen polygons were constructed around the point data of the later maps and between the limits of the area data of the earlier ones. This method made it possible to transform all the data into areal distribu- tions that in sum covered all of the study area. It was possible to determine whether the method treated both data representation forms equivalently, because luckily the results of one year of the original study 1983 were mapped in both forms (representational differ- ences proved trivial). Once the patterns of “increase,” “decrease,” and “no change” were established for the seventeen sets of data, a quadrat sampling grid was applied to each map. The size of each cell in the grid was determined under two constraints. The first involved the set of considerations associated with minimizing spatial sampling error, as outlined in Getis and Boots (1978). The second concerned the fact the original point and area data had been plotted ona row and column basis, with each “point” datum thereby representing an area of about 900 square miles. A grid cell length of some -multiple of thirty miles was thus highly desirable from the point of view of ease of construction and interpre- tation. A three row by three column cell size was statistically satisfactory and provided a desired level of resolution. Moreover, this degree of smoothing was subjectively deemed appropriate for reducing noise effects. The resulting grid network contained 370 cells to cover the exact portion of Canada involved. Within each cell, the proportions of area covered by “increase” (plus), “decrease” (minus), and “no change” (zero) were tallied and then added together to generate a value within the range of plus one to minus one. I his value was then assigned to a point represent- ing the center of the cell. Seventeen years of data were thus transformed to a 370 by 17 matrix. The first step in treating the 370 temporal series of seventeen values each was to eliminate those series for which too little original data contributed to the values obtained. Eighty of the series, most located along the study area periphery, were discarded on this basis. Values representing each element of the remaining series are in almost all cases derived from the effects of presence of one to several point data within the cell and/or all surrounding cells. For each of the seventeen years, mean proportions of change for the country as a whole were then com- puted from the remaining 290 values. The resulting series 1s clearly cyclic, correlating at r= 0.97+ witha simple cosine series (a period of about 8.9 years pro- vided the best fit). There is no wholly satisfactory solution to the prob- lem of determining phase relationships for cyclic ser- ies of 8.9 years period among series consisting of only seventeen values. Three ad hoc methods were investi- gated to derive such values here. | first compiled 290 series of running totals, plus one for the national mean series. All series were next subjected to a linear regres- sion operation through which gross serial trend was removed. Each detrended series could then be com- pared individually to the national mean. Method one consisted of directly comparing the relative position in each pair of series of all peaks, lows, and changes in residual sign, and dividing this integer total through SMITH: SNOWSHOE HARE POPULATION CYCLES 153 by the number of comparisons. This gives a reasona- bly unbiased value for phase lag, but only when the residual series compared are unambiguously cyclic, not all of which were. Method two consisted of directly correlating each pair of series and deriving the Pearson correlation coefficient, r, between each. Since the value r is distributed as a cosine series, phase lag in years could be read by taking the arcosine of r, divid- ing this by 360 (degrees), multiplying the results by the period of 8.9 years, and reinstating the sign. This approach is slightly biased at large lags by the slight asymmetry of these series, and very biased at small lags for its inability to distinguish between phase response and noise. Method three involved an analy- sis of covariance between each pair of series under the assumption that the variance and mean of all series were equal. This assumption was, in fact, violated, though the variance of most of the series was close enough to that of the national mean series that the slight discrepancy could be ignored. Those few that did vary greatly were eliminated from further processing. The third method described above produced results deemed the most acceptable overall among the three, and adopted to generate the phase lag values reported here. A final screening of the series emerging from this analysis followed, with a number more being elimi- nated for: 1) their failure to produce residuals that were significantly serially autocorrelated at the 0.025 level of significance (as indicated by the Durbin- Watson test statistic), or 2) their generation of a level of serial autocorrelation greater than that of a perfect cosine series of 8.9 years period (indicating the exist- ence of a trend, but one of longer period than the national mean to which comparisons were being made). Results The graphic results of the compilations may be viewed in Figures | through 4. Figure | is a year by year reformulation of the original data as described above. Each symbol represents one 90 miles squared area; all 370 of the original grid cells are represented except where lack of nearby original data precluded interpretation. The three symbols represent ranges of mean trend in each area: a “—” represents values of —1.0 through -0.4 (strong decline); a“o”, -0.3 through +(.3 (little or no trend); and a “mw”, +0.4 through +1.0 (strong increase). Figures 2, 3, and 4 are maps of summary statistics for the seventeen year series. Figure 2 is an isoplethic map of phase relationships, where 0.0 years corres- ponds to the national mean. Locations with negative values cycle “ahead” of the national mean, locations with positive values, behind. Areas that tested non- cyclic for the period of 8.9 years are shaded as such. THE CANADIAN FIELD-NATURALIST Vol. 97 154 ate Sy-PP6l | Roe yy-Erel SMITH: SNOWSHOE HARE POPULATION CYCLES 155 1983 Ww, 8r-Lrél “AQY puv uoIBVUR]Axd 10] 1X3} 99S “Bp 2OORIBIOB | & OBIE! OF ws, Ly-0v61 Lr6l YsnosY ZE-[ EG] “S[2Ag] UONR[Ndod s1eY BOYSMOUS UI spud [eIaUaH *| aUNdI4 156 THE CANADIAN FIELD-NATURALIST This map was constructed from 238 individual series lag values. Figure 3 is an isoplethic map of a surrogate used to help interpret relative amplitudes of the cycle in differ- ent areas. To derive this map, the proportions of the “no change” category noted at each grid cell location each year have been totalled for the seventeen year period. This value is an imperfect but useful surrogate for amplitude because it can be assumed that observers were likely to note “no change” more often for areas whose populations fluctuated less obviously. All 290 values were used to generate this map. Figure 4 is an isoplethic map of a surrogate statistic representing degree of non-random trend in the resid- uals of the detrended series. In most of the map, this can be considered to correspond to mean degree of cycle strength; in a few areas, however (notably the St. Lawrence Valley), trend was high despite lack of sig- Vol. 97 nificant 8.9 years periodicity. This situation arises because the surrogate used here is the Durbin- Watson statistic, which measures trend only (and not cyclicity per se). Apparently, some parts of the study area were either undergoing long term population irruptions (unlikely) or were simply cycling in periods greatly varying from the 8.9 year national mean. Areas of the map that tested “no trend” at the 0.025 level of signifi- cance are labelled such; of the remaining areas, lower values signify clearer trend. All 290 values were used to generate this map. Despite the many possible sources of error in deriv- ing these summary maps, it 1s felt they do a reasonably good job of representing the original data. A further validity check on the data of Figure 2 was carried out by averaging the values mapped on a province-by- province basis. The results are listed in Table 1. These results correlate very highly (r = 0.930) with Bulmer’s HUOS ON BAY FIGURE 2. Geographic variation in Snowshoe Hare cycle phase, 1931-32 through 1947-48. Stippled shading signifies areas for which data were unavailable; diagonal shading signifies areas that did not cycle at the system mean of about 8.9 years period. See text for further explanation. 1983 SMITH: SNOWSHOE HARE POPULATION CYCLES 157 HUOSON BAY FiGURE 3. Geographic variation in Snowshoe Hare cycle amplitude, 1931-32 through 1947-48. Stippled shading signifies areas for which data were unavailable. Higher values here represent lower amplitudes; see text for further explanation. (1974) provincial phase lag values for the Canada Lynx over the period 1919 to 1957, a good substantia- tion considering the two different types of data involved and the probability that Bulmer’s results are spatially biased by his use of a national norm weighted unevenly by the varying sales volumes of furs for each province. In addition, data of Figure 3 show a trend similar to that subjectively assessed by Adams (1959). Still, it should be emphasized that only limited confi- dence can be attached to the specifics of pattern pro- duced from such a short sampling period, regardless of the great cycle strength occurring in this case. Discussion and Summary Subjective descriptions of the trends visible in Fig- ure | accompany the original reports. A full analysis of these and the other data presented here will only be possible within the context of particular spatial- temporal modelling attempts; for the present, how- ever, several general trends can be pointed out. The low amplitude/degree of cycling in south central Canada has been recognized for some time; this is also true for southern British Columbia and parts of the Maritime Provinces. Here, however, Nova Scotia tests as strongly cyclic, and with a large negative phase lag (These patterns were also discovered in the Lynx data analysis in Smith and Davis 1981). A second trend, recognizable in all three summary maps, Is a tendency for patterns visible at the southern limits of range to be repeated at the northern limits; thus, phase lag tends to increase, amplitude to decrease, and strength of cycle to diminish geographically south- ward or northward away froma strip extending across central Canada (this impression has been confirmed in a trend surface analysis not presented here). A second set of interesting characteristics involves the rate of spread of the “wave” of population level change away from the main node in central Canada. 158 THE CANADIAN FIELD-NATURALIST Vol. 97 HUOS ON BAY FiGuke4. Geographic variation in non-randomness of Snowshoe Hare population trends, 1931-32 through 1947-48. Stippled shading signifies areas for which data were unavailable; diagonal shading signifies areas that did not pass the non-randomness test at the 0.025 level of significance (and are mapped as values of 1.0 or greater). See text for further explanation. TABLE |. Geographic variation in Snowshoe Hare cycle phase, 1931-32 through 1947-48, relative to the national mean, and aggregated at the provincial level. See text for further explanation. Province Lag (in years) Nova Scotia - 2.3 Manitoba 0.8 Saskatchewan -—0.7 Alberta -0.5 British Columbia =O Northwest Territories 0.0 Ontario + 0.4 Quebec + 0).5 Yukon Territory calles) First, it is apparent the manner of spread is not one of a simple node away from which diffuses a single wave of population change. Instead, minor nodes of change seem to appear in advance of the wave emanating from the main node, generating their own waves of advance. These eventually meet along “divides.” This aspect of system organization seems worth further consideration, perhaps from an_ epidemiological perspective. Second, it appears the diffusion of population level change is in some way slowed by physiographic and/or ecological barriers. Figure 2 clearly shows the effect of the Rocky Mountains and the Great Plains ona wave of change emanating from its Boreal Forest origin. The meaning of this relationship is certainly 1983 not obvious. One might argue that it is indicative of a change in community interaction rates associated with lower population densities in suboptimal habi- tats, but the question then becomes how this spatial- temporal translation is effected and why it doesn’t lead to an upset of standing phase relationships across the rest of the range. Moreover, it is difficult to defend central Alberta, for example, as being “suboptimal” habitat for the Snowshoe Hare. Nonetheless, this rela- tionship does cast doubt on the likelihood of the sim- ple external driving mechanism hypothesis of cycle causation, since it is difficult to consider this correla- tion between change in rate of diffusion and habitat change as a spurious relationship. In brief summary, this treatment has two general purposes. The first is to suggest the need for a view of ten-year cycle dynamics in which system-wide pat- terns are given primacy in the consideration of causal relationships. Failure to follow this route may result in much wasted effort as the effects of cycle operation are confused with its causes. The second is to produce from the best sources available a data base that can be used to support such spatial modelling attempts in whatever form they might take. Acknowledgments I would like to extend my thanks to Dr. Arthur Getis of the Department of Geography, University of Illinois at Champaign-Urbana, for offering helpful suggestions during the course of this work, and to Mr. Phillip Schneider, for assistance in the preparation of the maps. Literature Cited Adams, L. 1959. An analysis of a population of snowshoe hares in northwestern Montana. Ecological Monographs 29 (2): 141-170. Bulmer, M. G. 1974. A statistical analysis of the 10-year cycle in Canada. Journal of Animal Ecology 43 (3): 701-718. Bulmer, M. G. 1975. Phase relations in the ten-year cycle. Journal of Animal Ecology 44 (2): 609-621. Butler, L. 1951. Population cycles and color phase genetics of the colored fox in Quebec. Canadian Journal of Zool- ogy 29 (1): 24-41. Butler, L. 1953. The nature of cycles in populations of Can- adian mammals. Canadian Journal of Zoology 31 (3): 242-262. Chitty, D., and H. Chitty. 1941. The Canadian arctic wild life enquiry, 1939-40. Journal of Animal Ecology 10 (2): 184-203. Chitty, D., and H. Chitty. 1942. The snowshoe rabbit enquiry, 1939-40. Canadian Field-Naturalist 56 (2): 17-21. Chitty, D., and C. Elton. 1937. The snowshoe rabbit enquiry, 1935-36. Canadian Field-Naturalist 51 (5): 63-73. Chitty, D., and C. Elton. 1938. The snowshoe rabbit enquiry, 1936-37. Canadian Field-Naturalist 52 (5): 63-72. SMITH: SNOWSHOE HARE POPULATION CYCLES 159 Chitty, D., and C. Elton. 1939. The snowshoe rabbit enquiry, 1937-38. Canadian Field-Naturalist 53 (5): 63-70. Chitty, D., and C. Elton. 1940. The snowshoe rabbit enquiry, 1938-39. Canadian Field-Naturalist 54 (8): 117-124. Chitty, D., and M. Nicholson. 1943. The snowshoe rabbit enquiry, 1940-41, Canadian Field-Naturalist 57 (4,5): 64-68. Chitty, H. 1943. The snowshoe rabbit enquiry, 1941-42. Canadian Field-Naturalist 57 (7,8): 136-141. Chitty, H. 1946. The snowshoe rabbit enquiry, 1942-43. Canadian Field-Naturalist 60 (3): 67-70. Chitty, H. 1948. The snowshoe rabbit enquiry, 1943-46. Journal of Animal Ecology 17 (1): 39-44. Chitty, H. 1950. The snowshoe rabbit enquiry, 1946-48. Journal of Animal Ecology 19 (1): 15-20. Cole, L.C. 1951. Population cycles and random oscilla- tions. Journal of Wildlife Management 1I5 (3): 233-252. Cole, L. C. 1954. Some features of random population cycles. Journal of Wildlife Management 18 (1): 2-24. Cross, E. C. 1940. Periodic fluctuations in numbers of the red fox in Ontario. Journal of Mammalogy 21 (3): 294-306. Elton, C. 1933. The Canadian snowshoe rabbit enquiry, 1931-32. Canadian Field-Naturalist 47 (4): 63-69; 47 (5): 84-86. Elton, C. 1934. The Canadian snowshoe rabbit enquiry, 1932-33. Canadian Field-Naturalist 48 (5): 73-78. Elton, C., and M. Nicholson. 1942a. Fluctuations in numbers of the muskrat (Ondatra zibethica) in Canada. Journal of Animal Ecology I! (1): 96-126. Elton, C., and M. Nicholson. 1942b. The ten-year cycle in numbers of the lynx in Canada. Journal of Animal Ecol- ogy I 1 (2): 215-244. Elton, C., and G. Swynnerton. 1935. [he Canadian snow- shoe rabbit enquiry, 1933-34. Canadian Field-Naturalist 49 (5): 79-85. Elton C., and G. Swynnerton. 1936. The Canadian snow- shoe rabbit enquiry, 1934-35. Canadian Field-Naturalist 50 (5): 71-81. Finerty, J. P. 1980. The population ecology of cycles in small mammals: mathematical theory and biological fact. Yale University Press, New Haven. Fox, J. F. 1978. Forest fires and the snowshoe hare- Canada lynx cycle. Oecologia 31: 349-374. Getis, A., and B. Boots. 1978. Models of spatial processes. Cambridge University Press, Cambridge. Huntington, E. 1945. Mainsprings of civilization. Wiley, New York. Keith, L. B. 1963. Wildlife’s ten-year cycle. The University of Wisconsin Press, Madison, Wisconsin. MacLulich, D. A. 1937. Fluctuations inthe numbers of the varying hare (Lepus americanus). University of Toronto Studies, Biological Series No. 43. Moran, P. A. P. 1952. Thestatistical analysis of game-bird records. Journal of Animal Ecology 21 (1): 154-58. Moran, P. A. P. 1953. The statistical analysis of the Cana- dian lynx cycle. Australian Journal of Zoology | (2): 163-173; 1 (3): 291-298. Moran, P. A. P. 1954. [he logic of the mathematical theory of animal populations. Journal of Wildlife Man- agement 18 (1): 60-66. 160 THE CANADIAN FIELD-NATURALIST Vol. 97 Palmgren, P. 1949. Some remarks on the short-term fluc- Smith, C. H., and J. M. Davis. 1981. A spatial analysis of tuations in the numbers of northern birds and mammals. wildlife’s ten-year cycle. Journal of Biogeography 8 (1): Oikos | (1): 114-121. 27-35. Rowan, W. 1950. Canada’s premier problem of animal conservation: a question of cycles. New Biology 9: 38-57. Received 10 September 1981 Accepted 30 August 1982 Naturalization of the Rainbow Smelt, Osmerus mordax, in Lake Simcoe, Ontario HuUGH R. MacCRIMMON, ROBERT W. PUGSLEY! and BARRA L. GOTS Department of Zoology, University of Guelph, Guelph, Ontario NIG 2WI1 'Present Address: Lake Simcoe Fisheries Assessment Unit, Ontario Ministry of Natural Resources, Sutton West, Ontario LOE IRO MacCrimmon, Hugh R., Robert W. Pugsley, and Barra L. Gots. 1983. Naturalization of the Rainbow Smelt, Osmerus mordax, in Lake Simcoe, Ontario. Canadian Field-Naturalist 97 (2): 161-169. The Rainbow Smelt (Osmerus mordax), naturalized in Lake Simcoe from circum 1960, is now an important component in the sport fishery. Estimated annual winter angling harvest increased from three smelt in 1962 to a peak of 234 865 fish in 1973, but has since been lower. Spring spawning runs were first noted in 1965, with peak numbers being taken by sport fishermen from 1970 to 1974. The structure of the spawning population has shown a change in dominance from age group I to III in 1965, to age groups II tol V in 1972, and almost exclusively to age groups II to III in later years. Mean total length for fish in age groups I to III declined from 1965 to 1970, but since that time growth rate has remained relatively constant. Fecundity, significantly related to total length(P < 0.01), ranged from 4 000 to 58 100 eggs. Most spawning in each year normally took place at night in the gravel riffles of tributary streams (190 eggs cm “) at water temperatures of 7 to 11°C. Some spawning (1 — 2 eggs cm~) also occurred along the lake shore at water temperatures of 5 to 7°C. Growth of the smelt population has been concurrent with increased eutrophication of Lake Simcoe and a decline in Lake Whitefish (Coregonus clupeaformis) and Lake Trout (Salvelinus namaycush) fisheries. Key Words: angling harvests, population structure, growth rate, spawning, fecundity, naturalization, Rainbow Smelt, Osmerus mordax, Lake Simcoe. With the exception of a few eastern Ontario lakes where the species is considered to be indigeous (Dymond 1944), the occurrence of the Rainbow Smelt (Osmerus mordax) in the Great Lakes and a number of inland lakes inthe Great Lakes watershed westward to the Rainy River drainage is the result of naturaliza- tions occurring since 1923 (Scott 1963; Scott and Crossman 1973). Of these inland lakes now support- ing smelt populations, that in Lake Simcoe provides the most extensive fishery and is the best documented because of studies undertaken by the University of Guelph and the Ontario Ministry of Natural Resources. The origin of the Lake Simcoe smelt is unknown but its presence since circum 1960 is almost certainly the result of accidental or intentional introduction by man. Natural movements of smelt from Georgian Bay to Lake Simcoe through the Trent Canal system were blocked by an 18.7 m waterfall (power dam) at the Big Chute. Fish moving through the same canal network from Lake Ontario would have required passage through 41 locks and a series of shallow warmwater lakes. Prior to the first appearance of the smelt in Lake Simcoe, it is believed to have been absent from the entire Trent waterway between the Big Chute and Lake Ontario. The species was subsequently recorded in the Gull and Burnt River drainages in 1963 and 1965, respectively (White, personal communication). Smelt first appeared in the Lake Simcoe sport fishery in 1962. Winter creel records of the Ministry of Natural Resources have, since that time, provided a reliable picture of trends in the smelt population. The authors, in cooperation with the Lake Simcoe Fisher- ies Assessment Unit, have examined related biological phenomena occuring during the first 18 years of the smelt fishery. Description of Study Site Lake Simcoe, the sixth largest of Ontario inland lakes, lies 221.5 m above sea level (44° 23’N, 79° 18’W) with a surface area of 725 km?2, and mean and maxi- mum depths of 17.2 and 41.8 m respectively (Ander- son 1964). It forms a link in the 400 km Trent Canal system between Lake Ontario and Georgian Bay. The watershed is largely agricultural, but the shoreline is almost totally fringed by cottage and recreational development, plus several large towns. The eutrophi- fying effects of agricultural and domestic nutrient inputs are reflected by increasing algal blooms, local- ized hypolimnetic oxygen depletion, and a gradual shift in benthic communities (Veal and Clark 1970; Nicholls and MacCrimmon 1975). The lake supports a very substantial warmwater and coldwater sport fishery (MacCrimmon and Skobe 1970). Naturaliza- tion and growth of the smelt population has coincided with a decline in traditional Lake Trout (Sal/velinus namaycush) and Whitefish (Coregonus clupeaformis) fisheries (DesJardine 1977). 161 162 THE CANADIAN FIELD-NATURALIST Assessment of overall smelt abundance in Lake Simcoe is based on winter angling harvests estimated by the Ontario Ministry of Natural Resources from creel census data collected each year between 1962 and 1980. Field studies by the authors and Ministry per- sonnel have been centred on a 54 km- section of the lake and the Sibbald Park Creek and Pefferlaw River tributaries (Figure 1). The lake area extends from Georgina Island to Jackson’s point and northward to the 22 m depth contour. The I1 km shoreline is typi- cally small stones (50%), sand (30%), and boulders (20%). Littoral substrates extend about 100 m off- shore, but beyond, the bottom becomes mud except for several sand and gravel shoals southwest of Geor- gina Island. Several small intermittent creeks discharge into the study area, the largest being Sibbald Park Creek which continues to flow until mid-summer. It drains about | km? of swamp but, at a distance of 300 m from the lake, becomes gravel and stone bottom and discharges through a sand beach to the lake. A per- manent stream, the Pefferlaw River, drains 120 km? of mixed agricultural and idle land with upriver movement of fish from Lake Simcoe limited by a Salvation 7 Army Ck. Area of more concentrated study FIGURE |. Lake Simcoe, showing area of more concentrated study. Vol. 97 milldam 3 km from the mouth. The Pefferlaw River enters the lake about 8.4 km east of Sibbald Park Creek. Mature smelt enter both streams in most years. Materials and Methods Visual observations on the presence and relative abundance of smelt were made in the onshore waters of Lake Simcoe during the spawning season in 1970 | and, thereafter, only in Sibbald Park Creek, the Pef- ferlaw River and along the adjacent shoreline. Detailed observation of spawning location, time, duration and success relative to environmental influ- ences has focused on the former tributary. The first adult smelt to enter each evening were caught by monofilament gill net (120 mm bar measure) set in a semi-circle around the creek mouth at a water depth of 0.75 to 1.5 min conjunction with a fine-meshed seine. Further nightly samples were taken at hourly intervals from the same location using a dip net. In addition, representative samples collected by the Ministry of Natural Resources during the spawning run at Sib- bald Park Creek in 1974 and 1976 to 1980 were incor- porated into the study. Distribution in the open waters of the lake was determined by means of mono- filament gill nets of various meshes. Biological features examined included total length (TL), age using scales taken one centimeter to the left of the insertion of the dorsal fin, gonadal condition, and fecundity using the dry weight method (Nagasaki 1958) checked by three actual counts. The relative size of annual spawning runs along the lake shore and up tributary streams within the study area was based on a qualitative visual judgment and the nightly success of sport fishermen. On this basis, the numbers of spawners each year could be assessed as negligible, low, average or high. During the 1972 spawning season, actual spawning success was evalu- ated for lower Sibbald Park Creek and nearby lake shoals by placing weighted sections (30 X 30 cm) of gray-brown canvas on the lake and stream bottom substrates in eight locations each night from 26 April to 9 May. The canvas was retrieved the following morning and the eggs counted. Egg densities in the main spawning area in the upper creek itself were determined directly from bottom samples taken on 10 and 17 May. Results The first three Rainbow Smelt known to be taken from Lake Simcoe were caught through the ice by anglers during the winter of 1962. These fish were an age 3 female (TL 204 mm) from Kempenfeldt Bay on 22 January, anage 3 female(TL 221 mm) from Cooks Bay on 10 February, and an age 2 female (TL 241 mm) from Kempenfeldt Bay on 17 February caught at 1983 depths of 31, 12 and 32 meters respectively (Scott, personal communication). No smelt captures are known for 1963, but catches of 208 and 250 fish were reported in 1964 and 1965 respectively. In the spring of 1965, spectacular aggregations of mature smelt appeared suddenly at various onshore and stream locations around the lake. The first of these runs was noticed in late April and early May along the south shore of the lake just to the west of Jackson’s Point (Holder, personal communication). The abundance of smelt, as revealed by winter angling catches increased dramatically in succeeding years (Table 1). By 1970, the winter harvest had reached an estimated 54 829 smelt with spring spawn- ing aggregations of mature fish evident on most sand and gravel beaches, and in most tributaries around the lake. A peak winter catch, estimated at 234 865 smelt, was reached in 1973. Both angler success (average: 0.42 fish per hour) and the contribution of smelt (41.9% by number) to the total winter harvest of all species also peaked in that year. Between 1973 and 1980, winter harvests have shown no trend with annual catches fluctuating between and estimated low of 40 567 fish in 1977 and a high of 145 636 fish in 1976. C.U.E. values follow a similar pattern(Table 1). After 1973, annual catches of smelt have made up less than 24% of the total winter harvest of all species by number. Beginning in 1970, spring onshore aggregations of adult smelt have provided substantial and continuing annual catches by night-time seine and dip-net MACCRIMMON, PUGSLEY AND GOTS: RAINBOW SMELT 163 fishermen. Actual harvests are unknown, but spot sampling of the 1974 spring fishery revealed average nightly catches of 70 smelt (1.9 kg) per fishermen at a C.U.E. of 43 smelt per hour (DesJardine and Law- rence 1975). Total catches, however, undoubtedly have varied greatly from year to year. Changes in structure of the adult smelt population between 1965 and 1980 as determined from an exami- nation of stream spawning aggregations have been as follows (Figure 2). The spawning population of 1965 was composed essentially of age I, II and III fish in about equal numbers with a few age IV (4.4% of the sample) present. Thereafter, mature age I fish have been absent from spawning aggregations sampled except for a comparatively few fish in 1972 (0.8%), 1977 (9.8%) and 1980 (2.2%). A single age V female smelt appeared among the 1966 spawners sampled. Although several age V fish are represented in most annual samplings, their contribution to the spawning population since 1965 is low (2.8%). During the same 18-year interval, each of the age II and age III classes have been the most abundant, the latter representing a strong age II year-class in the preceding year. Only in 1972 was there a strong representation of age IV fish (37.9%). The strongest single age-class represented in any of the spawning aggregations to date was age II in 1980 (94.5%). All five age groups were present only in the 1972 and 1977 samples. The oldest smelt yet recorded for Lake Simcoe have come from the winter sport fishery in 1975 when an age VI male and an age VII female were caught (Table 2). TABLE |. Estimated numerical angling harvest of fish from Lake Simcoe during the 75-day winter fishery, 1962 to 1980. Osmerus mordax % of Salvelinus Coregonus Coregonus Perca Angler Total Catch Year Number C.U.E. Catch namaycush clupeaformis artedii flavescens Hours (all species) 1962 3 — ~— 1963 0 -- 2 588 92 945 43 155 14 040 335 081 152 728 1964 208 — 0.1 2 128 84 389 46 710 21 279 321 936 154 714 1965 250 — 0.1 1 123 153 338 92 602 14 277 401 981 261 590 1966 1 088 0.5 1 920 159 425 47 872 16 384 390 208 226 689 1967 — — — -- — — _ — 1968 6 677 — DQ) 2 245 169 376 83 504 36576 499 360 298 378 1969 — = -- — — _ — _ — 1970 54 829 0.2 22.2 2 141 71 073 64 934 54 396 240 828 247 374 1971 161 330 0.3 34.2 945 43 209 99 039 167 274 471 797 471 797 1972 135 018 0.3 30.6 ls 72 268 61 045 171469 441 575 440 975 1973 234 865 0.4 41.9 745 43 400 65 908 PIS) SY) 560 253 560 233 1974 72 377 0.2 230 1119 20 481 44 584 175 907 440 187 314 468 1975 99 000 0.2 ed 1 200 11 000 152 000 295000 459000 559 000 1976 145 636 0.3 DB 4 889 6 982 84 512 370 757 490 656 615 296 1977 40 567 0.1 10.1 5 452 3 006 193 168 161 108 432 220 403 547 1978 127 873 0.2 16.8 6 604 4 345 370 711 252 231 633 270 762 329 1979 83 634 0.1 15.1 5 329 8 834 222 715 233 306 583 270 553 905 1980 64 211 0.2 19.9 3 297 9 894 70 967 173 708 398 788 322 077 164 THE CANADIAN FIELD-NATURALIST i ive v 1967-1969 no data collected Percentage o t HW Ul Age (yrs) FIGURE 2. Year-class representation in spawning aggrega- tions of smelt sampled in Lake Simcoe between 1965 and 1980. n refers to sample size. Diagonal lines fol- low the strong year-classes of 1968, 1972, and 1975. Vol. 97 The growth of Lake Simcoe smelt, using the more abundant age classes II and III for comparative pur- poses (Figure 3), was greatest in 1965 but had declined by 1970 toa rate which, despite annual variations, has not changed appreciably since that time. Examination of the three fish caught in 1962 also showed the rapid growth rate evident in the 1965 spawning population. No growth trend is apparent among the compara- tively few fish in the fishery greater than age III. Mean total lengths of mature fish at spawning time are recorded in Table 2. Smelt as large as 267 mm in total length and 105 g have been recorded in winter angler catches. Scale annuli are formed after the spawning time, with 60% being formed by mid-June and 99% by mid-August. Post-spawning mortalities of adult smelt were first evident in 1968 and became very widespread in 1969 (Skobe, personal communication). Since that time, substantial spring die-offs following spawning have been a regular annual occurrence around the lake. The most extensive of these is believed to be that occurring in April of 1980. The observed spawning sequence begins in late win- ter each year when adult smelt become aggregated around the periphery of the lake at depths of 3 to 10 meters. At that time, the lake is still ice-covered and under stable winter stratification with epilimnetic temperatures of 3 to 4°C. Subsequent movement of adult smelt to onshore spawning areas commences at water temperatures of 4 to 7°C, usually in the latter half of April with the gradual disappearance of sur- face ice. During abnormally warm weather in Febru- ary 1981, however, smelt gathered under the ice in the warmer effluents of the Sutton River and east shore (and probably other) creeks. In some years, pre- spawning inshore aggregations have been disrupted greatly by shifting ice and rough water. Following onshore movements of adult fish to depths of 1.5 to3 m, lake spawning commenced along sand beaches of water temperatures of 5 to 9°C fora period of up to 13 days. Spawning aggregations, how- ever were dispersed if wave action became pro- nounced, or if water temperatures were consistently above 11°C. The more intense spawning generally took place in tributary creeks at temperatures of 7 to 11°C. Spawning in Sibbald Park Creek failed to occur in 1975 when the water temperature quickly rose to 12°C without a subsequent decline. Spawning did not occur in other spawning creeks, presumably also because of high water temperatures, but smelt remained numerous in shoal areas of the lake at depths of | m (9°C) where spawning did occur. Subsequent to the break-up of spawning aggrega- tions, the adult smelt gradually move to deeper water and, by mid-summer, become more-or-less randomly 1983 MACCRIMMON, PUGSLEY AND GOTS: RAINBOW SMELT 165 TABLE 2. Age composition and mean total length (T.L.) of mature smelt in spring spawning aggregations in Lake Simcoe. ! 1965 1966 1970 Age Mean Mean Mean Seep (en) Ne 2D) EN ae penal pat Nowa 22) I 138 (29) 31.5 = = a -- — = II 192 (28) 30.4 173 (80) 42.3 148 (81) 62.3 Ill 198 (31) 33.7 174 (89) 48.1 169 (27) 20.8 IV 179 (4) 4.4 179 (15) 8.1 198 (13) 10.0 Vv — — — 206 (1) 0.5 223 (9) 6.9 177 (92) 100.0 175 (185) 100.0 163 (130) 100.0 1971 1972 1974 Age Mean Mean Mean Group’ 1T.L.(mm) N % T.L.(mm) N % T.L.(mm) N % I _ — — 110 (8) 0.8 -- —~ = II 161 (154) 31.2 162 (192) 20.3 159 (171) 85.0 Ill 169 (247) 50.1 172 (344) 36.3 171 (27) 35) IV 195 (74) 15.0 179 (359) 31) 207 (2) 1.0 Vv 230 (18) 3.7 189 (44) 4.7 208 (1) 0.5 173 (493) 100.0 173 (947) 100.0 161 (201) 100.0 1975 1976 1977 Age Mean Mean Mean Group’ T.L.(mm) N q T.L.(mm) N % T.L.(mm) N % I —_ = — _ — — 141 (4) 9.8 II 156 (55) 24.6 154 (39) 72.2 160 (31) 75.6 Ill 170 (149) 66.5 176 (15) 27.8 178 (4) 9.8 IV 188 (18) 8.0 = = — 186 (1) 2.4 Vv 212 (2) 0.9 — = = 196 (1) 2.4 168 (224) 100.0 160 (54) 100.0 161 (41) 100.0 1978 1979 1980 Age Mean Mean Mean SOT SPA a NT a ON i an Ta un) Nee EZ I oa — = — == —_ 143 (6) De II 152 (54) 37.3 154 (74) 74.0 152 (263) 94.6 Ill 164 (76) 52.4 176 (23) 23.0 167 (9) 32 IV 182 (9) 6.2 189 (3) 3.0 _ — Vv 196 (6) 4.1 — — _ — -— 162 (145) 100.0 160 (100) 100.0 152 (278) 100.0 11965, 1966 (Holder, personal communication); 1970-72 (Pugsley 1976); 1974 (DesJardine and Lawrence 1975); 1975 (Pugsley 1976; DesJardine 1979); 1976-1979 (DesJardine 1979); 1980 (Pugsley, personal communication). distributed on or near the lake bottom at depths between 18 and 40 m. However, they may rise on occasion to break through the warm surface waters over considerable mid-lake areas. While the smelt were usually taken by summer netting in waters of 6 to 9° C, they have been taken in temperatures ranging up to 17.5°C during their ephemeral migrations to sur- face waters. During the late autumn and winter, the fish seem to become more closely associated with the bottom while seeking out the warmest water available. Fecundity and egg deposition rates were examined for the Sibbald Park Creek population in 1972. Exam- ination of 51 mature female smelt showed the gonads to account for 21.5% of the total body weight. The two ovaries typically varied greatly in size with the larger ovary (12.2% of body weight) being positioned in the anterior ventral left portion of the body cavity, and the smaller ovary being located in the posterior right ventral part of the coelom. Total ripe egg counts ranged from 4 000 to 58 100 (x = 19 191) and showed 166 e—e 1965 wem-x 1970 [a Leemer 1975 E pA Oo—do 1980 £ Doses a ree 3 150 x7 i) = 120 0 Il Hl Age Group FiGureE 3. Mean total lengths of ages I] and II] Lake Simcoe smelt sampled from spring spawning aggregations in 1965, 1970, 1975 and 1980. a significant relationship (P < 0.01) with total length as expressed by the equation, Log E = 3.4048 + 3.4087 Log L (r = 0.89). There was no evidence of the parasite Glugea hertwigi (Delisle 1972). The density of eggs deposited on the onshore sand bottom adjacent to Sibbald Park Creek averaged only one to two eggs cm 2 in contrast to the creek bottom where numbers of eggs scattered over the humus and gravel were as high as 190 eggs cm *. Subsequent observations revealed that, one week after deposition, only those eggs adhering to protruding gravel or stones were alive. Survival rate of all eggs deposited on the creek bottom as estimated at 7.2%. After a similar time, none of the eggs deposited on the sand beach could be found. The survival of eggs dislodged by wave action is unknown. The summer dispersal of young-of-the-year smelt was followed in 1980, using a Reckahns mini-wing trawl. Substantial numbers were found near the bot- tom at depths of 10 to 14 mat various lake locations, with 15-minute hauls, on occasion, yielding more than 700 young-of-the-year smelt. Coincident with the naturalization of smelt in Lake Simcoe have been unique changes in the composition of the fishery which have been associated with the increasing eutrophication processes occurring in the lake (Nicholls and MacCrimmon 1975; DesJardine 1977). Principal changes in the composition of the winter angling catch occurring simultaneously with the rise of the smelt population have been a gradual THE CANADIAN FIELD-NATURALIST Vol. 97 decline in the abundance of Lake Whitefish (Corego- nus clupeaformis) and a decrease in the numbers of native Lake Trout (Sa/velinus namaycush), although the size of annual winter harvests of this latter species has been sustained by the release of hatchery-reared fish which now dominate the catch. On the other hand, catches of native Yellow Perch (Perca flaves- cens) increased progressively during the 1960’s and early 1970's. Also, large increases in winter harvests of Ciscoe (Coregonus artedii)' have occurred in the late 1970’s (Table 1). Discussion The sudden appearance of a large spawning popula- tion of smelt around much of the shoreline of Lake Simcoe in 1965, with nearly equal numbers of mature fish of age I, II and III, is clear evidence that fruitful spawnings occurred in 1962, 1963 and 1964. The breeding population must have been extremely smal! in these years in view of their negligible contribution to the winter sport fishery and the non-detection of spawning aggregations which have become so obvious since 1966, both onshore and in tributary creeks. The subsequent growth and stabilization of the smelt pop- ulation confirms both the suitability of Lake Simcoe for the species, and the plasticity of the smelt to utilize the diverse habitat offered (MacCrimmon and Skobe 1970; MacCrimmon and Pugsley 1979). The most notable biological changes in the smelt which have occurred since their naturalization in Lake Simcoe are, firstly, the pronounced change in age of maturity from I to II years which took place in the 1965 year class; secondly, a pronounced retardation in growth rate through ages II and III during the latter part of the 1960’s; and, thirdly, a gradual increase in the population to peak abundance in 1973. Time to maturity and growth rate are no doubt related phen- omena (Ivanova et al. 1969; Saunders and Powers 1970; Moulton 1974). The appearance of a few mature age I fish in the 1972, 1977, and 1980 spawning popu- lation represent a few precocious fish, which only in 1980 were above average size for their year-class. All of the above-noted three phenomena occurred within little more than a decade following the establishment of a self-sustaining population in the lake. Since the peak abundance of smelt in 1973, as reflected by winter angling success (C.U.E. of 0.4 and 41.9% of total harvest), winter harvests have fluctu- ated greatly (40 567 to 145 636 fish), but no popula- tion trend is apparent. In view of the close relationship among C.U.E. values (0.1 to 0.3), percentage of smelt in winter angling catches (10.0 to 23.7%) and total 'Likely all C. artedii, although a deepwater ciscoe (‘trout herring’) resembling C. hoyi was identified at one time (MacCrim- mon and Skobe 1970). 1983 annual harvests, it is reasonable to assume that the winter catch statistics now represent a fair measure of smelt abundance. During the late 1960's, however, actual smelt populations may have been higher than indicated from the winter creel data as the alien smelt was held in disdain by most traditional fishermen. By the early 1970’s the smelt had been accepted as an important component of the sport fishery and virtu- ally all captured fish retained regardless of size. While the quantificaton of annual spawning numbers has not been attempted, intensive (night- time) general observations have shown a remarkable variability in the extent and duration of both onshore and creek spawning runs which undoubtedly affect year-class strengths. The principal environmental fac- tor determining the onset and probably the duration of both onshore and creek spawning 1s undoubtedly water temperature (7 to 11°C for actual spawning), although other (undetermined) factors may regulate the absence, presence or abundance of spawning aggregations on a night-by-night basis even when water temperatures are satisfactory. While local flow conditions were observed to occasionally affect creek spawning, the adverse effects of wave action on the exposed sand beaches of the lake frequently affected not only spawning activity and duration, but the fate of the eggs deposited there. The effect on survival of eggs dislodged from shoal substrate by wave action is unknown, but the survival of ova in Sibbald Park Creek would seem to be limited to eggs adhering to well-oxygenated and unsilted substrate. The observed annual fluctuations in year-class strengths during the past decade have not been corre- lated with any observed environmental parameter, but the dominance of the 1968 year-class in the 1970, 1971, and 1972 spawning populations illustrates the impact of one highly-successful spawning on the fishery. Large variations in the size and timing of local annual spawning aggregations are known to occur, but there is no apparent correlation between subse- quent year-class strengths and what have been judged to be poor and good spawning years. A reasonable observation is that year-class strengths in Lake Sim- coe would seem to be determined by spawning success rather than the numbers of spawners. The two times of Smelt harvest from Lake Simcoe are the 75-day winter ice fishery and the spring onshore and creek sport netting operation during the spawning season. Because of the sporadic and unpre- dictable nature of the latter night-time fishery in and adjacent to many creeks around the lake, a meaning- ful census has proven to be impossible. Hence, the annual harvest of smelt from the lake is unknown. Using a mean weight of 23.5 g per fish for the esti- mated 234 865 smelt captured by winter angling dur- MACCRIMMON, PUGSLEY AND GOTS: RAINBOW SMELT 167 ing the peak year of 1973 (Pugsley 1976), the total harvest for the winter fishery totalled some 5 500 kg. If the assumption is made that spring sport netting yielded a like amount (probably a gross over- estimate), the annual harvest of smelt from Lake Sim- coe would have reached 10 000 kg or 0.14 kg ha!. Actual weights harvested annually would, of course, vary not only with the numbers harvested but also with year-class strengths represented in the fishery. Angling intensity during the winter ice fishery is alsoa function of weather conditions causing unsafe ice conditions, drifting snow, and other factors that can affect accessibility to the offshore fishing grounds spread over much of the 72 519 ha lake surface. Comparison of growth, maturity, fecundity and general biological characteristics of the Lake Simcoe smelt suggests no marked differences from other native and naturalized populations (Greene 1930; Ferguson 1965; Rupp 1968; Burbridge 1969; Lackey 1969). The principal environmental features that seem to have favored the establishment of the substantial naturalized smelt population in Lake Simcoe can be identified as, firstly, the presence of extensive creek and shoal spawning areas around the periphery of the lake; secondly, the extensive expanse of open-water with summer bottom water temperatures (<< 11°C) favourable for smelt feeding and growth; and, thirdly, an abundance and diversity of food available for year- round utilization (MacCrimmon and Pugsley 1979). The extent to which the accelerated process of eutrophication, which has been occurring in Lake Simcoe simultaneously with naturalization, has favoured establishment of the substantial smelt popu- lation is unknown. By 1970, changes in the benthic communities to favour more pollution-tolerant orga- nisms had been detected (Veal and Clark 1970), while critically low dissolved oxygen was an eminent con- cern in some hypolimnetic areas prior to the 1977 autumn turnover (DesJardine and Lawrence 1979). Whether-or-not pronounced changes in the native fish fauna (most obviously the Whitefish and Lake Trout) which have occurred coincident with naturali- zation of the smelt in Lake Simcoe can be attributed to the presence of the alien smelt, or to progressive eutrophication or overfishing, is a question of major management as well as biological interest. Although smelt have been cited as a probable cause of Whitefish declines in other waters (Cucin and Regier 1965), our present study has failed to identify any direct adverse effects of smelt on any of the coldwater fishes (White- fish, ciscoes, Lake Trout) which contribute to the winter or summer fisheries. Despite the seasonal presence of fish eggs (espe- cially smelt and sucker, Catostomus sp.) and small fish) especially young-of-the-year smelt and Emerald 168 THE CANADIAN FIELD-NATURALIST Shiner, Notropis atherinoides) in many of the smelt stomachs examined (MacCrimmon and Pugsley 1979) and the partly-piscivorous diet of smelt in most waters (Creaser 1925; Gordon 1961; Price 1963; Sel- geby et al. 1978), neither the eggs nor young of the Whitefish, ciscoes or Lake Trout have been identified in the stomachs of Lake Simcoe smelt. While the absence of these (and most other common fish species) in the stomachs may be a problem of inadequate sampling, this is not believed to be the case. Rather it is felt that because of the broadly omnivorous and rather selective diet of the smelt, it is more probable that the suppression of any sympatric species in the lake would be the result of competitive pressure on specific food items required at some critical stage in the early development of other fishes (MacCrimmon and Pugsley 1979). Adequate data on the food and feeding of Lake Simcoe fishes to support this conten- tion are simply not available. On the basis of present knowledge, the decline of native Lake Trout and Whitefish, and the recent co- dominance of Yellow Perch and ciscoes in the winter sport fishery, would seem to be more closely related to changing environmental parameters and fishery pat- terns than to the introduction of smelt to the lake. The Rainbow Smelt now occupies a unique and seemingly permanent ecological niche in the Lake Simcoe eco- system. While there is presently no evidence to suggest that the environmental requirements and_ bio- ecological features of the smelt are seriously in conflict with those of native species, their possible displace- ment of indigenous species is worthy of further investigation. Acknowledgments We are indebted to the staff of the Lake Simcoe Fisheries Assessment Unit, Ontario Ministry of Natu- ral Resources, and to Regional Biologist, R. L. Des- Jardine, for services provided and for making availa- ble unpublished data on the Lake Simcoe fishery. Financial assistance for the earlier part of the study was provided by the Canadian National Sportsmen’s Show and, more recently, by the National Research Council of Canada. Personal Communications Cited Holder, A. S. 1966. Fishery Biologist, Lake Simcoe Fisher- ies Assessment Unit, Ontario Ministry of Natural Resour- ces, Sutton West, Ontario (now Director, Fisheries Branch, Toronto). Pugsley, R.W. 1981. Fishery Biologist, Lake Simcoe Fisheries Assessment Unit, Ontario Ministry of Natural Resources, Sutton West, Ontario. Scott, W. B. 1975. Curator, Department of Ichthyology and Herpetology, Royal Ontario Museum, 100 Queen’s Park, Toronto, Ontario (now Huntsman Marine Labora- tory, St. Andrews, N.B.). Vol. 97 Skobe, E. 1970. Fishery Biologist, Lake Simcoe Fisheries Assessment Unit, Ontario Ministry of Natural Resources, Sutton West, Ontario (now University of Guelph, Guelph). White, D. D. 1975. District Manager, Minden District. Ontario Ministry of Natural Resources, Minden, Ontario. Literature Cited Anderson, D. V. 1964. The thermal regimes of Lake Sim- coe, Ontario. Ontario Department of Lands and Forests, Technical Series Research Report 57. 14 pp. Burbidge, R. G. 1969. Age, growth, length-weight relation- ship, sex ratio, and food habits of American smelt, Osme- rus mordax (Mitchill), from Gull Lake, Michigan. Tran- sactions of the American Fish Society 98: 631-640. Creaser, C. W. 1925. The establishment of the Atlantic smelt in upper waters of the Great Lakes. Papers of the Michigan Academy of Science, Arts and Letters 5: 405-424. Cucin, D., and H. A. Regier. 1965. Dynamics and exploita- tion of lake whitefish in Southern Georgian Bay. Journal of the Fisheries Research Board of Canada 23: 221-274. Delisle, C. E. 1972. Variations mensuelles de Glugea hert- wigi (Sporozoa: Microsporida) chez différents tissus et organes de l’éperlan adulte dulcicole et conséquences de cette infection sur une mortalité massive annuelle de ce poisson. Canadian Journal of Zoology 50: 1589-1600. DesJardine, R. L. 1977. Statement on Lake Simcoe fisher- ies problems. Ontario Ministry of Natural Resources, Lake Simcoe Fisheries Assessment Unit Report. 10 pp. DesJardine, R. L. 1979. A note on the spring spawning runs of smelt, Osmerus mordax, in Lake Simcoe, 1974- 1979. Ontario Ministry of Natural Resources, Lake Sim- coe Fisheries Assessment Unit Report 79-10. 9 pp. DesJardine, R. L., and J. N. Lawrence. 1975. The spring spawning run of rainbow smelt, Osmerus mordax (Mit- chill), of Lake Simcoe, 1974. Ontario Ministry of Natural Resources, Lake Simcoe Fisheries Assessment Unit Report. 38 pp. DesJardine, R.L., and J.N. Lawrence. 1979. The dis- solved oxygen and temperature regimes of Lake Simcoe during the winter and summer of 1978. Ontario Ministry of Natural Resources, Lake Simcoe Fisheries Assessment Unit Report 78-5. 70 pp. Dymond, J. R. 1944. Spread of the smelt (Osmerus mor- dax) inthe Canadian waters of the Great Lakes. Canadian Field-Naturalist 58(1): 12-14. Ferguson, R. G. 1965. Bathymetric distribution of Ameri- can smelt, Osmerus mordax, in Lake Erie. University of Michigan, Great Lakes Research Division Publication 13: 47-60. Gordon, W.G. 1961. Food of the American smelt in Saginaw Bay, Lake Huron. Transactions of the American Fisheries Society 90: 439-443. Greene, C. W. 1930. The smelts of Lake Champlain. Pp. 105-129 in A biological survey of the Champlain watershed. Supplemental to the Nineteenth annual report, New York State Conservation Department, 1929. 321 pp. Ivanova, M.N., I. Ye. Permitin, and S. N. Polovkova. 1969. Structure features and abundance of the population of landlocked smelt (snetok), Osmerus eperlanus eperla- 1983 nus morpha spirinchus Pallas, in the Rybinsk reservoir. Problems of Ichthyology 9(3): 325-331. Lackey, R. T. 1969. Seasonal depth distribution of land- locked Atlantic salmon, brook trout, landlocked alewives, and American smelt ina small lake. Journal of the Fisher- ies Research Board of Canada 27: 1656-1661. MacCrimmon, H. R., and R. W. Pugsley. 1979. Food and feeding of the rainbow smelt (Osmerus mordax) in Lake Simcoe, Ontario. Canadian Field-Naturalist 93(3): 266-271. MacCrimmon, H. R., and E. Skobe. 1970. The fisheries of Lake Simcoe. Ontario Department of Lands and Forests, Fish and Wildlife Branch. 140 pp. Moulton, L. L. 1974. Abundance, growth and spawning of the longfin smelt in Lake Washington. Transactions of the American Fisheries Society 103: 46-52. Nagasaki, P. 1958. Fecundity of Pacific herring. Journal of the Fisheries Research Board of Canada 15(3): 313-330. Nicholls, K. H., and H. R. MacCrimmon. 1975. Nutrient loading to Cook Bay of Lake Simcoe from the Holland River watershed. International Revue gesamt Hydrobiol- ogie 60(2): 159-193. Price, J. W. 1963. A study of the food habits of some Lake Erie fish. Bulletin of the Ohio Biological Survey (New Series) 11 (No. 1). 89 pp. Pugsley, R. W. 1976. Biology of the rainbow smelt, a new resident in Lake Simcoe. M.Sc. thesis, University of Guelph, Guelph, Ontario. MACCRIMMON, PUGSLEY AND GOTS: RAINBOW SMELT 169 Rupp, R.S. 1968. Life history and ecology of the smelt, Osmerus mordax, in inland waters of Maine. Maine Department of Inland Fisheries and Game, Fisheries Research Management Division, Final Report. 36 pp. Saunders, L. H., and G. Power. 1970. Age, growth, matur- ity and sex composition of American smelt, Osmerus mordax (Mitchill) from Matamek Lake, Quebec. Le Naturaliste canadien 97: 511-516. Scott, W. B. 1963. A revision of changes of the fish fauna of Ontario. Transactions of the Royal Canadian Institute 34(2): 111-125. Scott, W. B., and E. J. Crossman. 1973. Freshwater fishes of Canada, Bulletin of the Fisheries Research Board of Canada 184. 966 pp. Selgeby, J. H., W. R. MacCallum, and D. V. Swedberg. 1978. Predation by rainbow smelt (Osmerus mordax) on Lake Herring (Coregonus artedii) in western Lake Super- ior. Journal of the Fisheries Research Board of Canada 35: 1457-1463. Veal, D. M., and A. R. Clark. 1970. A preliminary report on water quality characteristics of Kempenfelt Bay and adjacent Lake Simcoe. Ontario Water Research Commis- sion (Ministry of the Environment). 17 pp. Received 23 March 1982 Accepted 22 July 1982 Range Extensions of Vascular Plants from the Northern Yukon Territory LES C. CWYNAR Department of Botany, University of Toronto, Toronto, Ontario MSS IA1 Cwynar, Les C. 1983. Range extensions of vascular plants from the northern Yukon Territory. Canadian Field-Naturalist 97(2): 170-176. Fifty-two vascular plant taxa are reported from the northern Yukon for either the first time or as range extensions. Four taxa are new to the flora of the Yukon: Carex amblyorhyncha, Ceratophyllum demersum (Hornwort), Armeria maritima ssp. arctica (Thrift), and Artemisia biennis (Biennial Wormwood). Forty-two are reported as the first records for the northern Yukon; 31 beyond Hultén’s (1968) predicted range limits and 1! within. Range extensions are also reported for 10 taxa previously collected within the northern Yukon. Key Words: range extensions, Carex amblyorhyncha, Ceratophyllum demersum, Armeria maritima ssp. arctica, Artemisia biennis, Yukon Territory. During. his studies of the Beringian flora, Eric Hultén visited the northern Yukon at a time when it was still poorly explored botanically. He later (Hultén 1968) noted that one of the most interesting areas remaining to be botanized in the Yukon was in the north around the upper Porcupine River because “the northern and southern extremities presumed for many plant ranges might merge there”. The flora of the northern Yukon has proved to be rich (Welsh and Rigby 1971; Weinet al. 1974; Nagy et al. 1979), partly because of the varied topography and geology (Bos- tock 1961, Welsh and Rigby 1971) and partly because most of the northern Yukon has escaped the repeated glaciations of the Quaternary Period (Hughes 1972). Although these recent collections have confirmed Hultén’s suspicion, they have not proved to be defini- tive. In the summers of 1975-1978 I made 1550 collec- tions of vascular plants from the northern Yukon, which I define as that area north of 65° latitude, corresponding with the important physiographic boundary of the Ogilvie Mountains. This paper des- cribes those collections that are either the first for a taxon in the northern Yukon or extend a taxon’s known range. Collection Sites The approximate location of collection sites is shown in Figure |. The precise location and general character of each site is briefly described below. 1) Beaufort Sea coast, 69°30’N, 140°40’W, 3m above sea level (a.s.1.), coastal sand dunes and beach. 2) Beaufort Sea coast, 69°30’N, 139°13’W, 3m a.s.l., coastal sand dunes and beach. 3) Firth River, 69°21’N, 139°30’W, 150 m a.s.l., a 4) 5) 6 — 7) 8 — 9 — 10 — 11 — 12) 13) 14) 170 rock knoll in mesic-xeric tundra in the vicinity of Richard S. McNeish’s archaeological excavations. Small un-named lake 2 km northeast of Trout Lake, 68°50’N, 138°44’W, 150 maz:s.I., tundra in north slope foothills of the British Mountains. Dog Creek at junction with Black Fox Creek, 68° 23’N, 138°46’W, 425 maz.s.]., tundra on lime- stone bedrock with well developed gallery forest of White Spruce along Black Fox Creek. Hanging Lake, 68° 23’N, 138°23’W, 460 mazss.l1., a lake perched in shale bedrock in tundra with a sparse vegetation cover within the drainage basin. Barn Mountains, 68°25’N, 139°52’W, 760 m a.s.l., bedrock of shales and siltstones with collec- tions from alpine meadows with lush growth, especially along streams. Bonnet Lake, 68°12’W, 139°52’W, 600 m a:s.1., snowflushes on soft shale tundra slope. McDougall Pass, Richardson Mountains, 67°41’N, 136°35’W, 300 m az.s.l., open White Spruce woods with some White Birch; thermo- karst lakes and ponds scattered throughout. Junction of Rat Indian Creek and Porcupine River, 67°34’N, 138°20’W, 300 m az.s.l., river- bank forests and thickets, river margins, and forested mountain slopes. Old Crow Flats, 67° 42’N, 139° 08’W, 260 ma.s.1., shallow thermokarst lakes, marshy infilled lakes, and wet meadows. Old Crow Flats, 67° 52’N, 139°27’W, 260 ma.s.1., wet heaths and margins of thermokarst lakes. Junction of Schaeffer Creek and Old Crow River, 672 S07Ne 139 25ile Wee2.6 0) imeialselemsSiiGc mam onl Schaeffer's cabin on riverbank, now somewhat overgrown but still a meadow. Second Caribou Lookout, 67°35’N, 139°31’W, 1983 CWYNAR: PLANTS FROM THE YUKON 171 136° OLD CROW FLATS NORTHWEST TERRITORIES 7. YUKON TERRITORY FIGURE |. Location of collection sites from the northern Yukon Territory. 172 280 m a.s.l., an outcrop of shales and siltstones with few spruce trees and a sparse cover of dry heaths. 5) ROldi Chie love Sa0NewI S937 aW 4 260omiaysales the site of intensive excavations by members of the Northern Yukon Research Programme, Univer- sity of Toronto. Terrace slopes support White Spruce woods with Salix glauca. The uplands beyond the terraces are wet with numerous bogs and muskeg. 16) Klo Kut, 67°34’N, 139°41’W, 270 ma:ss.l., site of the base camp for the Northern Yukon Research Programme. It lies on a meadow terrace between the Porcupine River and a small lake which abuts a limestone bluff. 17) Un-named thermokarst lake, 67° 34’N, 139° 45’W, 280 maz.s.l., marshy with scattered Black Spruce and some White Spruce. 18) Old Crow Village, 67°34’N, 139°49’W, 250 m a.s.l., collections made of weedy plants within the village. Big Bluff, 67°29°W,, 139°53°W, 275 m sas:lsa long silt bluff along the Porcupine River. Bluefish Basin, 67°22’N, 139°54’W, 280 mazss.l., a wet plain with many bogs and thermokarst lakes and scattered stands of spruce, mostly black. Bluefish Cave, 67°08’W, 140°47’W, 300 maz.s.l., limestone ridge with thick turf and scattered White Spruce. 22) Southern Richardson Mountains, 66°04’N, 135°50’W, collections made at various elevations from alpine tundra to low elevation closed White Spruce woods; the bedrock is limestone. 19 nm 20 ma” 21 — Vascular Plant Range Extensions Fifty-two taxa are reported with sites, habitats, and collection numbers. Nomenclature follows Hultén (1968), but much of the pioneering work in the Yukon and Mackenzie District by A. E. Porsild and W. J. Cody (e.g. 1968) uses a slightly different nomenclature so I have added in parenthesis the latters’s recent usage (Porsild and Cody 1980) where it differs. A set of voucher specimens has been deposited at the Cana- dian Department of Agriculture Herbarium(DAO) in Ottawa. Four taxa are new to the flora of the Yukon: Carex amblyorhyncha, Ceratophyllum demersum, Armeria maritima ssp. arctica, and Artemisia biennis. Forty- two are the first records for the northern Yukon: 31 beyond Hultén’s predicted range limits and 11 within. Range extensions are also reported for 10 taxa pre- viously collected within the northern Yukon. Lycopodiaceae Lycopodium annotinum ssp. annotinum. Site 9: in THE CANADIAN FIELD-NATURALIST Vol. 97 mixed While Spruce-White Birch woods, 30/. This is the first record for the northern Yukon, extending the range northward over 200 km beyond Hultén’s (1968) predicted limit. Potamogetonaceae Potamogeton filiformis. Site 9: in shallow water of sluggish stream, 296. Although well within Hultén’s predicted range, this collection is nevertheless a first for the northern Yukon. Juncaginaceae Triglochin palustris. Site 10. locally abundant on moist silty riverbank with Carex aquatilis ssp. aquati- lis and Eleocharis uniglumis, 920 and 926; site 16: seepage area along silty riverbank, /029. This species reaches interior Alaska and the Mackenzie River Delta but it has not been previously recorded in the northern Yukon where Hultén predicted it would not be found. It is an easily overlooked taxon frequently mixed with sedges and spike-rushes on damp silty riverbanks. Gramineae Agropyron subsecundum (A. trachycaulum s. lat.). Site 10: common along riverbank, 893A. This collec- tion extends the species’s range well north of Hultén’s predicted limit in the central Yukon. Beckmannia erucaeformis ssp. baicalensis (B. syzi- gachne). Site 10: rare, only 3 plants in seepage area along steep silty riverbank, 98/. The first record for the northern Yukon, this collection lies well beyond Hultén’s predicted limit in the central Yukon and west of collections from the Mackenzie Delta area. Puccinellia Nuttalliana (P. borealis). Site 19: exposed dry silt bluff, 8. An uncommon plant in Alaska- Yukon usually confined to alkaline soils in southwest- ern Yukon and southeastern Alaska. The first collec- tion from the northern Yukon and outside of Hultén’s predicted limit. Trisetum sibiricum ssp. sibiricum. Site 10: locally common in a Salix arbusculoides-Alnus incana thicket along the Porcupine River, 953 and 1/548. Huitén (1968) does not report this taxon from North America although he did earlier (Hultén 1941-1950). The confused status of this taxon in North America will be clarified in a forthcoming paper (Cwynar, in preparation). This is the first report for the northern Yukon. Cyperaceae Carex amblyorhyncha. Site 15: bog margin, 224; site 16: slumping winter road, 35/A. A little-collected 1983 subarctic plant, these are the first collections for the Yukon Territory and they fall within Hultén’s pre- dicted range. Carex capitata. Site 15: Sphagnum bog, 223: site 14: White Spruce woods, /546. Although within Hultén’s predicted range, these are the first collections from the northern Yukon, well beyond previous collections in the central Yukon. Carex disperma. Site 16: common along shaded, mossy intermittent stream, /027. This collection extends the range northward from collections south of the Ogilvie Mountains and westward from the Mac- kenzie Delta. Carex Franklinii. Site 22: abundant on calcareous, turfy, open slope with a few White Spruce, /288. This apparently rare plant of calcareous habitats has been collected only 3 times before inthe Yukon. Just within Hultén’s predicted range, this is the first collection from the northern Yukon. Carex rariflora. Site 16: slumping winter road, 346; site 12: abundant in boggy pools, sedge meadows, and shallow margins of thermokarst lakes, 760, 804, 8/8, and 8/9; site 5: abundant in sedge meadow, /056. When Porsild (1974) described material for a flora of the central Yukon, he noted that this arctic species was rare in the Yukon and suggested that it was probably restricted to alpine habitats. These are the first collec- tions from the northern Yukon where it is acommon plant of sedge meadows along the margins of thermo- karst lakes, especially in the Bluefish Basin and the southern half of the Old Crow Flats. Carex rostrata. Site 11: common in shallow water, 758: site 10: fringing lake margin, 9/0, and sparse on wet silty riverbank, 927. Porsild (1974) reported the first collections from the northern Yukon in the Ogil- vie Mountains. My collections extend the known range northward into the southern part of the Old Crow Flats where Hultén predicted it would be found. Eriophorum russeolum ssp. rufescens. Site 4: wet depression, /66; site 12: in Sphagnum at edge of lake, 779. This taxon was previously known from the Yukon at only one disjunct locality in the interior. These are the first records for the northern Yukon. Eleocharis uniglumis. Site 10: locally abundant on wet silty riverbank, 895; site 16: abundant on wet silty riverbank, /0/6. Previously collected infrequently in the southern Yukon, these are the first collections from the northern Yukon — several hundred kilome- CWYNAR: PLANTS FROM THE YUKON 173 tres north of previous collections and considerably beyond Hultén’s predicted range. Juncaceae Juncus alpinus s. lat. Site 10: common on seepage area of steep silty riverbank, 986. This collection extends the range northward from the central Yukon into the upper Porcupine River, outside of Hultén’s predicted limit. Luzula arcuata ssp. unalaschensis. Site 6: lakeshore, 1400. Nagy et al. (1979) reported this species from a site just north of this one but they did not identify their collection to subspecies. This is the first record for the northern Yukon and it is within Hultén’s predicted range. Luzula rufescens. Site 15: rare in rich fen, 206; site 16: common on margin of slumping winter road, 509, 5/0, 514. The first records for the northern Yukon, they lie beyond Hultén’s predicted limit. Cody and Porsild (1968) reported a collection from Canoe Lake in the northern Richardson Mountains, Northwest Territo- ries. The Canadian collections thus appear to be con- tiguous with those from the principal range in Alaska. Orchidaceae Amerorchis rotundifolia (Orchis rotundifolia). Site 22: rich White Spruce woods, /283. This is the first record for the northern Yukon, extending the range from west-central Yukon. Salicaceae Salix interior. Site 10: common along cobbly river- bank, 903; site 16: pioneering on riverbar, 566. The first collections from the northern Yukon, well north of previous collections from west-central Yukon and beyond Hultén’s predicted limit. Betulaceae Alnus incana ssp. tenuifolia. Site 10: forming thickets along riverbank, 946. The first collection from the northern Yukon and beyond Hultén’s predicted limit. This species was not found at several localities on the Old Crow River where it was looked for. Santalaceae Geocaulon lividum. Site 21: on limestone ridge with scattered White Spruce, //34. The first collection from the northern Yukon, it bridges the gap between previous collections from adjacent Alaska and Mac- kenzie District. Polygonaceae Polygonum amphibium ssp. laevimarginatum (P. 174 amphibium var. stipulaceum). Site 10: locally abund- ant on dry silty riverbank, 984. The first collection form the northern Yukon and beyond Hultén’s pre- dicted range, it bridges the gap between Alaskan and Mackenzie District collections. Polygonum caurianum. Site 10: common on cobble riverbank, 962: site 16: cobble riverbank, /023. These are the first records from the northern Yukon for this easily overlooked species. Previously known in the Yukon from only 3 collections in the interior and southwest. Chenopodiaceae Monolepis Nuttalliana. Site 16: cobble riverbank where winter road crosses, /022. Hultén puts the limit of this species at the Yukon border from the Alaskan side, witha locality marked on the Porcupine River at the Alaska-Yukon border. This collection is therefore either the first or second for the Yukon and extends the range beyond Hultén’s predicted limit. Portulacaceae Claytonia sarmentosa. Site 7: along stream through rich alpine meadow, /367. Well beyond Hultén’s pre- dicted limit, this is a significant range extension from previously reported localities in the southern quarter of the Yukon. Caryophyllaceae Arenaria longipedunculata (A. humifusa pro parte). Site 15: rare in wet crevices of limestone bluff, ///0. The first record for the northern Yukon, well beyond Hultén’s predicted limit. Melandrium Taylorae. Site 10: common on dry upper riverbank, 906 and 929; site 15: occasional along shaded stream, //03. This species has been recorded twice in the southwest corner of the northern Yukon as well as in the Mackenzie Delta area to the east. My collections lie beyond Hultén’s predicted limits, bridg- ing the gap between previously known collections and suggesting that the range is not as fragmented as shown by Hultén. Ceratophyllaceae Ceratophyllum demersum. Site 12: forming extensive mats in water 0.5 to 1.5 m deep around thermokarst lake, 761. This species is new to the flora of the Yukon Territory. It is disjunct from the main boreal forest range which reaches Great Slave Lake in the North- west Territories. Hultén shows several disjunct locali- ties in interior Alaska. Ranunculaceae Ranunculus confervoides (R. aquatilis var. eradica- THE CANADIAN FIELD-NATURALIST Vol. 97 tus). Site 16: common in shallow water of oxbow lake, 1092. This is the first collection from the northern Yukon although it lies within Hultén’s predicted limit. This collection together with Porsild’s (1974) indicate that the populations in the southwest Yukon are not disjunct from those in the north as shown on Hultén’s map. Ranunculus scleratus ssp. multifidus. Site 22: mud flat at lake margin, 307. Collected occasionally in the Yukon from south of the Ogilvie Mountains, this is the first collection in the north. Hultén shows an isolated collection from the Mackenzie Delta to the east. Thalictrum sparsiflorum. Site 16: a solitary plant ina stand of Balsam Poplar ona terrace of the Porcupine River, /0/4. Considerably north of Hultén’s predicted range, this is the second record for the northern Yukon, the other coming from the northern part of the Ogilvie Range. Fumariaceae Corydalis pauciflora. Site 22: common along stream draining late-melting snow patch, /267. Wein et al. (1974) extended the range from the British Mountains south into the central Richardson Mountains. My collection further extends the range to the southern Richardson Mountains towards collections from the Mackenzie Mountains. Cruciferae Eutrema Edwaradsii. Site 22: along seepage channel through shale fellfield, //9/. This collection lies out- side of Hultén’s limit and bridges the gap between localities in the northern Richardson Mountains and the Mackenzie Mountains. Smelowskia borealis. Site 22: common on limestone scree, 1/80. Although collected several times in the northern Yukon near the Alaska-Yukon border, this is the first collection from the Richardson Mountains, extending the range eastward, as well as northward from the Mackenzie Mountains. Saxifragaceae Saxifraga foliolosa var. foliolosa. Site 6: gravelly lakeshore, /430. Although within Hultén’s predicted range, this is only the third published collection from the Yukon and the first from the north. Rosaceae Geum macrophyllum ssp. perincisum. Site 18: weedy on riverbank, /549. This is the second collection for the northern Yukon and extends the range northward just beyond Hultén’s predicted limit. 1983 Potentilla pennsylvanica. Site 15: on south facing bluff dominated by Artemisia frigida, 182. This is the first collection for the northern Yukon and falls out- side of Hultén’s predicted limit. Sanguisorba officinalis. Site 16: common along river- bank, 75; site 10: abundant along riverbank, 970. These collections extend westward the range from the Alaska-Yukon border to beyond Hultén’s predicted range. Halagoraceae Mpyriophyllum spicatum (M. exalbescens). Site 12: commonly scattered in water 0.5 to 1.5 m deep in thermokarst lake, 755, 762, and 800. These are the first collections from the northern Yukon and they are within Hultén’s predicted range. Umbelliferae Cicuta mackenzieana. Site 20: forming outer fringe of Menyanthes mat in thermokarst lake, 649; site 17: scattered in marsh, 689. These collections are within Hultén’s predicted range and the first for the northern Yukon. Pyrolaceae Pyrola chlorantha. Site 22: locally abundant in White Spruce woods on limestone slope, /287. Previously collected in the Yukon from south of the Ogilvie Mountains, this is the first collection from the north- ern Yukon, north of Hultén’s predicted limit in the Yukon and west of collections from the Mackenzie District. Plumbaginaceae Armeria maritima ssp. arctica. Site 1: coastal sand dunes, 4//. A coastal plant that rarely occurs inland, this taxon is new to the flora of the Yukon Territory although its occurence along the north coast was expected. Boraginaceae Lappula occidentalis (L. Redowskii var. occidentalis). Site 14: abundant on south-facing silty bluff with Artemisia frigida, 707; site 10: dry, south-facing bluff, 980. These are the first collections from the northern Yukon, extending the range north from Hultén’s limit in the southern Ogilvie Mountains. Scrophulariaceae Veronica Wormskjoldii ssp. alterniflora. Site 22: occasional in lush growth of herbs and shrubs below late-melting snow patch, /269. This collection appears to be the first for the northern Yukon and extends the range north beyond Hultén’s predicted limit. CWYNAR: PLANTS FROM THE YUKON 175 Plantaginaceae Plantago major var. Pilgeri. Site 10: occasional along silty riverbank, 9/6 and 9/7. This is the first collection from the northern Yukon and only the second from the Yukon; the other is from near Dawson. Campanulaceae Campanula aurita. Site 22: in turf of open subalpine White Spruce stand on south-facing limestone ridge, 1169. Previously reported from the westernmost extent of the Ogilvie Mountains in the southwest corner of the northern Yukon and from the vicinity of Rampart House of the Alaska-Yukon border, this collection lies beyond Hultén’s predicted limit bridg- ing the gap between collections from the eastern Brooks Range and the northern Mackenzie Mountains. Compositae Antennaria rosea. Site 14; under Balsam Poplar along rocky Pleistocene beach line, /95. This is the first collection from the northern Yukon. It is beyond Hultén’s predicted limit and extends the range north- ward from the central Yukon. Artemisia biennis. Site 10: occasional along river- bank, 96/. This is the first collection for the Yukon Territory where Hultén did not predict its occurrence. It is probably an introduction arising from the boat traffic that historically entered the northern Yukon from the Mackenzie District via the Peel, Rat, Bell, and then Porcupine rivers. Artemisia furcata (A. hyperborea). Site 1: coastal dunes, 403. The first collection from the northern Yukon, extending the highly fragmented range northwestward from the Mackenzie District to the Yukon coastal plain. Erigeron elatus. Site 15: wet areas of 20-30 year-old forest fire, 363. The first record for the northern Yukon, it lies beyond Hultén’s predicted limit. Petasites palmatus. Site 22: locally common in rich spruce woods on side of moraine, /225. An uncom- mon plant in the Yukon, it has been collected only once previously in the north near the Old Crow River. This collection is the first for the Richardson Moun- tains and reduces the gap between the apparently disjunct populations in the northern Yukon and lower Mackenzie River area from the principal distribution farther south. Acknowledgments This research was made possible by the logistic 176 support offered by W. N. Irving and J. Cing-Mars of the Northern Yukon Research Programme. J. C. Rit- chie accompanied me on many of the collecting trips and gave me the opportunity to collect in the southern Richardson Mountains. I thank D. F. Murray of the University of Alaska Herbarium for verifying many of my collections and G. Argus of the Vascular Plant Herbarium, National Museums of Canada, for verify- ing my willow determinations. This is Contribution Number 47 to the Northern Yukon Research Programme. Literature Cited Bostock, H. S. 1961. Physiography and resources of the northern Yukon. Canadian Geographical Journal 63: 112-119. Cody, W. J.,and A. E. Porsild. 1968. Additions to the flora of Continental Northwest Territories, Canada. Canadian Field-Naturalist 82: 263-275. Hughes, O. L. 1972. Surficial geology of northern Yukon Territory and northwestern District of Mackenzie, Northwest Territories. Paper 69-36, Geological Survey of Canada, Department of Energy, Mines, and Resources, Ottawa, Ontario, Canada. Hultén, E. 1941-1950. Flora of Alaska and Yukon. Lunds University Arsskr. Lund. THE CANADIAN FIELD-NATURALIST Vol. 97 Hultén, E. 1968. Flora of Alaska and neighboring territo- ries. Stanford University Press, Stanford, California. 1008 Pp. Nagy, J. A., A. M. Pearson, B. C. Goski, and W. J. Cody. 1979. Range extensions of vascular plants in northern Yukon Territory and northwestern District of Mackenzie. Canadian Field-Naturalist 93: 259-265. Porsild, A. E. 1974. Materials fora flora of central Yukon Territory. National Museums of Canada, Publications in Botany, Number 4. 77 pp. Porsild, A. E., and W. J. Cody. 1968. Checklist of the vas- cular plants of Continental Northwest Territories, Can- ada. Plant Research Institute, Canada Department of Agriculture, Ottawa. 102 pp. Porsild, A. E., and W. J. Cody. 1980. Vascular plants of continental Northwest Territories, Canada. National Museums of Canada. 667 pp. Wein, R. W., L. R. Hettinger, A. J. Janz, and W. J. Cody. 1974. Vascular plant range extensions in the northern Yukon Territory and northwestern Mackenzie District, Canada. Canadian Field-Naturalist 88: 57-66. Welsh, S. L., and J. K. Rigby. 1971. Botanical and physio- graphic reconnaissance of northern Yukon. Brigham Young University Science Bulletin, Biological Series, volume XIV, Number 2. 64 pp. Received 3 February 1982 Accepted 14 October 1982 Habitat Segregation by Stream Darters (Pisces: Percidae) in the Thames River Watershed of Southwestern Ontario JOHN ENGLERT! and BENONI H. SEGHERS?% Department of Zoology, University of Western Ontario, London, Ontario N6A 5B7 'Present address: MacLaren Plansearch Corporation, Suite 401 — 750 West Pender St., Vancouver, B.C. V6C 217 2Present address: Kananaskis Centre for Environmental Research, University of Calgary, Seebe, Alberta, TOL 1X0 3Author to whom correspondence should be addressed Englert, John, and Benoni H. Seghers. 1983. Habitat segregation by stream darters (Pisces: Percidae) in the Thames River watershed of southwestern Ontario. Canadian Field-Naturalist 97(2): 177-180. Habitat segregation among five species of darters was studied in the Thames River watershed of southwestern Ontario during May-September in 1977 and 1978. Direct underwater observations and quantitative net sampling demonstrated that each species occupied a wide range of habitats; however, some degree of segregation was found. Rainbow (Etheostoma caeruleum) and Greenside (E. blennioides) darters predominated in riffles and Johnny (£. nigrum) and Blackside ( Percina maculata) darters were found mainly in pools or raceways. Fantail Darters (£\ flabellare) were common in both riffles and raceways. Key Words: darter, Etheostoma blennioides, Etheostoma caeruleum, Etheostoma flabellare, Etheostoma nigrum, Percina maculata, Percidae, habitat segregation, Thames River, Ontario. Six species of darters (Family Percidae, Subfamily Etheostomatinae) occur sympatrically in streams of the Thames River watershed in southwestern Ontario. Darters are small benthic fish in which the swim bladder is greatly reduced or absent. Habitat segrega- tion was studied for five species: the Blackside Darter (Percina maculata), Johnny Darter (Etheostoma nigrum), Rainbow Darter (E. caeruleum), Greenside Darter (E. blennioides) and Fantail Darter (E. flabel- lare). The sixth species found in the Thames River watershed is the Least Darter (E. microperca) but it was excluded from this study because its small size and cryptic behavior in dense vegetation made obser- vation difficult. Few studies of the ecology and behavior of darter populations in Canada have been made. Resource partitioning has been examined in Johnny and Black- side darters in Manitoba (Smart and Gee 1979) and in Rainbow, Fantail, Johnny and Least darters in south- ern Ontario (Paine et al. 1982). The Greenside and Rainbow darters are restricted in Canada to southern Ontario (Scott and Crossman 1973) and the Greenside Darter has been classified as rare by McAllister and Gruchy (1977). Study Areas and Methods Using a faceplate and snorkel, direct underwater observations of darters were made mostly in Medway Creek, approximately 12 km north of London, Onta- rio, between May and September in 1977 and 1978. Medway Creek is a 36-km tributary of the Thames River and is characterized by a low gradient and diverse bottom ranging in composition from silt to boulder. Darter habitats were categorized after Winn(1958a) as follows: a riffle is shallow, fast-moving water; a pool is deep, slow-moving water; a raceway is intermediate between riffle and a pool. For rocky bottoms, average stone diameter was estimated and substrate type was classified after Jenkins and Burkhead (1975): gravel, 0.3-8.0 cm.; rubble, 8.0-30 cm. Because it was easy to differentiate male and female Rainbow Darters in the field (males have brighter blue and orange coloration), each sex was treated as a distinct group. Sexual differences in habitat selection may also occur in the other species (Lachner et al. 1950; Winn 1958a, b) but our attempts to sex them in the field were unreliable. Quantitative data were collected by seining in a 360 m section of the Middle Thames River, approxi- mately 24 km east of London, Ontario. The river is generally shallow (maximum depth following spring run-off is less than 2 m) and slow-moving (Bietz 1980). The sampling periods were 21-30 June and 25 August — 16 September 1978. A seine net 6.1 m long and 6.4 mm in mesh diameter was used. Each seine haul consisted of an approximately 11 m long sweep upstream and towards shore; four sites were sampled (Table 1). Water velocity was measured with a flow meter 4 cm from the bottom. The most common ma- crophytes were Heteranthera dubia, Myriophyllum sp., Elodea canadensis, Potamogeton crispus and P. pectinatus (Bietz 1980). The downstream raceway was slower and deeper than the upstream raceway. The cove site was adjacent to a section of stream similar to the upstream raceway site, but displayed greater vari- ation in water velocity and depth. Lei 178 THE CANADIAN FIELD-NATURALIST Vol. 97 TABLE |. Habitat characteristics of darter collection sites inthe Middle Thames River, Ontario(7 July, 1978). Water velocity and depth data are based on a series of measurements (4, 4, % distance) across the stream. s = standard deviation; C = coefficient of variation. Downstream Raceway Water velocity x 0.029 (m/sec) S 0.022 C 75.9% Water depth x 34.6 (cm) s 9.5 Cc 27.5% Vegetation cover (%) 30 Substrate type silt, gravel Results Underwater observations in Medway Creek revealed that Greenside and Rainbow darters were usually the most common species in riffle areas. Greenside Darters generally occurred over rubble and Rainbow Darters over gravel bottoms. Female Rain- bow Darters were found in both raceways and riffles whereas male Rainbow Darters were common only in riffles. Fantail Darters were also found in riffles but unlike Greenside or Rainbow darters were more abundant in shallow water (i.e. << 35 cm). Fantail Darters were most common in riffles or raceways with a rubble bottom. Johnny and Blackside darters occurred mainly in pools or raceways but could also be observed in the relatively quiet waters along the edge of riffles or behind large rocks. Blackside Darters showed no clear preference for any particular sub- strate; however, Johnny Darters were most common over sand or silt bottoms. Habitat Upstream Riffle Raceway Cove 0.244 0.074 0.138 0.056 0.023 0.106 23.0% 31.1% 76.8% 21.8 21.4 24.9 3.0 4.7 9.0 13.8% 22.0% 36.1% 15 45 50 gravel rubble silt, rubble gravel Inthe Middle Thames River there were no apparent seasonal changes in habitat utilization from June to September and thus the quantitative data from the two sampling periods were combined. Because some collection sites were seined more frequently than others, the results were standardized by dividing the raw data by the number of seine hauls at each site (catch per unit effort, see Table 2). Seining was most difficult in the riffle site due to the high water velocity. The rubble bottom of the upstream raceway also reduced seining efficiency as did the macrophytes in the cove site. Consequently, more emphasis was placed on comparing darter abundance within than among the different collection sites. In the riffle site, Rainbow and Greenside darters were the most abundant species; surprisingly, female Rainbow Darters were significantly more common than males (x2 = 4.83, p < 0.05). They were also more common in the downstream raceway (x2= 11.04, TABLE 2. Number of darters captured inthe Middle Thames River, Ontario (June 21-30 and 25 August — 16 September 1978) expressed as the catch per unit effort. Numbers in parentheses are the percentage of the total catch of each species occurring at the different sites. Male and female Rainbow darters are tabulated separately; for comparisons among species these data should be summed. Catch per unit effort Number of Total Site seine hauls darters Blackside Johnny Rainbow(F) Rainbow(M) Greenside Fantail Downstream 23 419 4.9 3.5) 3.2 1.7 4.9 0.1 Raceway (49.0) (46.1) (32.7) (20.5) (31.2) (1.9) Upstream 20 223 0.2 1.0 2.1 37) 2.4 2.3 Raceway (2.0) (13.2) (21.4) (38.6): (15.3) (44.2) Riffle 12 101 0.6 2 2.0 0.9 2.8 0.9 (6.0) (15.8) (20.4) (10.8) (17.8) (17.3) Cove 15 279 4.3 1.9 2.5 2.5 5.6 1.9 (43.0) (25.0) (25.5) (30.1) (35.7) (36.5) Totals 70 1022 10.0 7.6 9.8 8.3 Sa7/ ayy 1983 p<0.01), but less common than males in the upstream raceway (x? = 4.57, p< 0.05). The greatest proportion of Blackside and Johnny darters occurred in the downstream raceway; Rainbow and Greenside darters were also common at this site. Most of the Fantail Darters were caught in the upstream raceway; however, Rainbow Darters were the most abundant species. The cove site contained a large proportion of the total catch of all five species. Presumably, this reflects in part the large variation in habitat character- istics at that site. Discussion Differences in behavior and morphology may have resulted in some variation in catchability among the different darter groups. Blackside Darters are usually less benthic than the other species (Englert 1979; Smart and Gee 1979) and therefore may be more vulnerable to seining. Another potential source of sampling bias is the antipredator behavior of the dif- ferent species, e.g. in response to a splashing stimulus Greenside Darters normally show a greater avoidance distance and Fantail Darters have a greater tendency to dive under cover than the other species (Englert 1979). Nevertheless, these data do provide a general description of the main habitat and relative abun- dance of each species in the upper Thames River watershed. The findings here generally support the existing literature on darter populations elsewhere. Winn (1958a) found Greenside, Rainbow and Fantail dar- ters in fast rubble or gravel riffles. Fahy (1954) and Miller (1968) observed Greenside Darters in algae- covered rocky riffles and raceways. Fantail Darters have been reported in fast riffles (Lake 1936) and rock riffles (Page and Schemske 1978). Winn (1958a) observed Johnny and Blackside darters in slower and deeper raceways; however, Smart and Gee (1979) found wide tolerances for water velocity and substrate type by both species. Lachner et al. (1950) captured Blackside Darters in riffles, but near the shoreline, where much vegetation was present and water velocity was reduced. In Ontario, Paine et al. (1982) reported that Rain- bow and Fantail darters were the dominant species in riffles and Johnny and Least darters were dominant in weed beds. Similarly, Grant (unpublished manu- script), seining in Medway Creek near our study site, found Johnny Darters in bare pools and vegetated stream edge habitats whereas Greenside, Rainbow and Fantail darters were captured primarily in riffles. It appears these five species of darters partition the available physical environment to some extent by segregating into different habitats. In the Thames River watershed, the critical riffle habitat often ENGLERT AND SEGHERS: HABITAT SEGREGATION BY DARTERS 179 appears limited, especially during periods of low water level in the summer. This may force a greater overlap in habitat among the different species than expected. During periods of high water in the spring, however, these species may segregate more strongly due to major differences in their reproductive behav- ior (Winn 1958a, b). Acknowledgments We thank Eva Heczko, Keith Bowser and Lillian Roth for assistance with the field work and James Grant for valuable discussion on darter behavior. This study was funded by N.S.E.R.C. of Canada Grant A9894 to B. H. Seghers. Literature Cited Bietz, B. 1980. The adaptive significance of territorial aggregations in longear sunfish (Lepomis megalotis pel- tastes Cope). Ph.D. thesis, University of Western Ontario, London, Ontario. 126 pp. Englert, J. 1979. A comparative study of the antipredator behaviour of stream darters (Percidae) in southwestern Ontario. M.Sc. thesis, University of Western Ontario, London, Ontario. 119 pp. Fahy, W. E. 1954. The life history of the northern green- side darter, Etheostoma blennioides blennioides Rafi- nesque. Journal of the Elisha Mitchell Scientific Society 70: 139-205. Jenkins, R. E., and N. M. Burkhead. 1975. Recent capture and analysis of the sharphead darter, Etheostoma acutic- eps, an endangered percid fish of the Upper Tennessee River drainage. Copeia 1975: 731-740. Lachner, E. A., E. F. Westlake, and P.S. Handwerk. 1950. Studies on the biology of some percid fishes from Western Pennsylvania. American Midland Naturalist 43: 92-111. Lake, C. T. 1936. The life history of the fan-tailed darter, Catonotus flabellaris flabellaris (Rafinesque). American Midland Naturalist 17: 816-830. McAllister, D. E., and C.G. Gruchy. 1977. Status and habitat of Canadian fishes in 1976. In Canada’s threatened species and habitats. Canadian Nature Federation, Ottawa, Ontario. pp. 151-157. Miller, R. V. 1968. Asystematic study of the greenside dar- ter, Etheostoma blennioides Rafinesque (Pisces: Perci- dae). Copeia 1968: 1-40. Page, L.M., and D. W. Schemske. 1978. The effect of interspecific competition on the distribution and size of darters of the subgenus Catonotus (Percidae: Etheos- toma). Copeia 1978: 406-412. Paine, M. D., J. J. Dodson, and G. Power. 1982. Habitat and food resource partitioning among four species of dar- ters (Percidae: Etheostoma) ina southern Ontario stream. Canadian Journal of Zoology 60: 1635-1641. Scott, W. B., and E. J. Crossman. 1973. Freshwater fishes of Canada. Fisheries Research Board of Canada Bulletin 184. 966 pp. Smart, H. J., and J. H. Gee. 1979. Coexistence and resource partitioning in two species of darters (Percidae), 180 THE CANADIAN FIELD-NATURALIST Vol. 97 Etheostoma nigrum and Percina maculata. Canadian Winn, H. E. 1958b. Observations on the reproductive hab- Journal of Zoology 57: 2061-2071. its of darters (Pisces — Percidae). American Midland Winn, H.E. 1958a. Comparative reproductive behavior Naturalist 59: 190-212. and ecology of fourteen species of darters (Pisces — Perci- dae). Ecological Monographs 28: 155-191. Received 27 January 1982 Accepted 20 October 1982 Aquatic Plants in Lake Superior Provincial Park in Relation to Water Chemistry! D. FRASER24 and J. K. MORTON? / ‘Ontario Ministry of Natural Resources, Wildlife Research Section, Contribution No. 82-01. 2Ontario Ministry of Natural Resources, Wildlife Research Section, Box 50, Maple, Ontario LOJ 1E0 3Department of Biology, University of Waterloo, Waterloo, Ontario N2L 3G] 4Present address: Animal Research Centre, Agriculture Canada, Ottawa, Ontario KIA 0C6 Fraser, D., and J. K. Morton. 1983. Aquatic plants in Lake Superior Provincial Park in relation to water chemistry. Canadian Field-Naturalist 97(2): 181-186. Species of submersed and floating-leafed aquatic plants and chemical composition of the water were studied in 14 lakes on the Canadian Shield in Lake Superior Provincial Park, Ontario. Most lakes were carbonate-poor and had typical softwater floras. Two lakes of higher alkalinity contained several hardwater species that were rare elsewhere in the park. Winter de-icing salt affected water chemistry in several roadside lakes, but had little apparent influence on plant species composition. Key Words: aquatic plants, Canadian Shield, water chemistry, salt. Despite considerable work in other environments, there has been little study of aquatic vegetation in relation to water chemistry on the Canadian Shield. Species lists for some southern Shield areas are given by Aiken and Gillett (1974), Crowder et al. (1977), and Miller and Dale(1979). Moyle’s (1945) study of aqua- tic vegetation in relation to water chemistry included some Shield areas in northeastern Minnesota. In the following study, we recorded species of aqua- tic plants and water chemical composition in 14 Shield lakes in Ontario. Because several of the lakes were near a major highway, the study also gives some information on vegetation in lakes contaminated with highway de-icing salt. Study Area The study was done in Lake Superior Provincial Park, a 1555-km2 area on the east shore of Lake Superior. The Park is dominated by boreal forest in the north and Great Lakes-St. Lawrence forest in the south(Rowe 1972). Outcropping rock consists mainly of igneous and metamorphic rocks of early Precam- brian age, as well as some more recent volcanic and sedimentary formations (Ontario Division of Mines 1971). The Park has a rugged terrain with elevations ranging from 200 to 600 m. Many of the lakes have rocky shorelines and mineral-poor water. An 85-km section of Trans-Canada Highway runs through the Park. Because of icy road conditions between November and April, rock salt (NaCl) is applied at a rate of 30 to 40 tonnes/km of highway each winter, as estimated by highway officials. Of the 14 lakes studied (Table 1), 11 were chosen because of easy road access, and because they gave a variety of sizes, depths, and shoreline types. Boulder Lake was included because previous study (Ontario Ministry of Natural Resources, unpublished data) indicated a relatively high total alkalinity. North and South Tugboat Lakes were included because their contact with sedimentary rock was expected to pro- duce relatively mineral-rich water. Methods Each lake was visited once between 19 July and 7 August 1980. The observer paddled slowly around the shoreline in a canoe, zig-zagging over all areas where vegetation could be seen. All submersed and floating- leafed plants were noted, and their abundance in the lake was scored subjectively from rare(1) to abundant (7). Voucher specimens were pressed and deposited in the herbaria of the University of Waterloo and Lake TABLE |. Geographic location of the 14 lakes, and minimum distance from the Trans-Canada Highway Distance from Lake Location highway (km) Doc Greig 47°29'N, 84°48’W <0.1 Rustle 47°32’N, 84°48’W 0.3 Sickle! 47°34.5’N, 84°49’'W <0.1 Crescent 4P1l7TN, 84°33'W 0.8 Collette 47°43’N, 84°49'W <0.1 Mom 47°33’N, 84°48’W <0.1 Kenney 47°17'N, 8 4°34’W <0.1 Dead Otter 47°38’N, 84°48’W <0.1 Moose 47°S50’N, 84°53’W <0.1 Rabbit Blanket 47°45'N, 84°50’W <0.1 Myin Pond! 47°41’N, 84°49’W <0.1 Boulder! 47°52'N, 84°52’W <0.1 South Tugboat! AOR IN, 7 t35)° Il sy WY 15 North Tugboat! 47°35.5'N, 85°2’W HI 'Names assigned, for the purpose of this study, to unnamed lakes. 181 182 THE CANADIAN FIELD-NATURALIST Superior Provincial Park. Some species in North Tugboat Lake were recorded merely as “Present” or “Absent” because of the limited time available at that location. A I-L mid-lake surface water sample was collected from each lake. Water was analyzed for specific con- ductance, total alkalinity (equal to bicarbonate alka- linity), and for other major cations and anions (Table 2) at the Ontario Ministry of the Environment labora- tory, Rexdale, Ontario, according to standard labora- tory techniques (Ontario Ministry of the Environment S75) Latin names of plants follow Fernald (1950) except where more recent work has indicated a need for revision. Common names have been included where available. For plants considered to be rare in the study area, known ranges were checked from Scoggan (1978) and by consulting specimens in the herbaria of the National Museum of Canada in Ottawa (CAN), of the Biosystematics Research Institute, Agriculture Canada in Ottawa (DAO), of the Department of Botany, University of Toronto (TRT), and of the Department of Biology, University of Waterloo (WAT). Results and Discussion Most of the lakes had “soft” water (TA < 40 ppm, as defined by Moyle(1945)) typical of granitic areas of the Canadian Shield (Table 2). North and South Tug- boat Lakes were slightly more alkaline, presumably because of carbonates in the sedimentary and volcanic bedrock in their area. In most lakes, Ca and bicarbo- nate (as reflected by TA) were the major ions, and their levels were closely correlated (Figure 1). How- ever, Moose Lake, Mijin Pond and Boulder Lake had very high levels of Na and Cl, presumably derived from salt applied to the highway in winter (Table 2). TABLE 2. Water chemistry of the 14 lakes, including total alkalinity (TA) in ppm of CaCO,, specific conductivity (SC) in pmhos/cm, and other cations and anions in ppm Lake Wes SC Ne 1K Ca Wie Gl SO, Doc Greig Am Py (0S) Oe DW OXS > Bx0) Rustle 33) WL) OP B45 OO — O69 75 Sickle i) Gl A) OS) SHS) O83) DA O.5) Crescent Sj 3 12 05° 28 O60 O2 7.0 Collette 1 Se CoO. S06 O45 15.5 80 Mom I ©) 33 OS) HS ell 4ay 9S Kenney lO - 333 O77 SA V3 7B 75 Dead Otter N70 35 Or <2 1.3 60° WS Moose MN Bil 4 O83) 1B N43 WB.) WO Rabbit Blanket” 22) 80) 92:9 07) 18:4" 40 47" 78.0 Miin Pond 2) 22) 2s OS 132 113d 280 loM Boulder AX) 2s) 2 OL 2D DAG ZOO 13.0 South Tugboat 48 100 2.6 0.3 11.2 3.2 0.5 6.0 North Tugboat 53°130 2.9 0.6 15.8 3.7. 0.5 11.0 Vol. 97 These lakes had higher levels of Ca than would be expected based on TA (Figure |). Extra Ca may have been introduced into these lakes by an ion exchange between soil and ground water caused by the high levels of dissolved NaCl. The four lakes located more than 0.1 km from the highway had very low levels of Cl compared with the other ten lakes. Much of the Cl in the study lakes is probably derived from the highway. Submersed and floating-leafed plants in the 14 lakes are listed in Table 3. Typical emergent species, not listed in Table 3, included Eleocharis palustris (Spike-rush), Equisetum fluviatile (Horse-tail), Hyp- ericum ellipticum (St. John’s-wort), Sium suave (Water-parsnip), Sparganium chlorocarpum (Bur- reed), Typha latifolia (Common Cat-tail), Dulichium arundinaceum (\Vhree-way Sedge), J/ris versicolor (Blue Flag), and Sagittaria latifolia (Arrowhead). Cardamine pensylvanica (Bittercress) was found growing partly submersed in a cold-water seepage area in Kenny Lake. Myriophyllum tenellum may be more common in the Park than Table 3 indicates: although flowering specimens were first noted in Doc Greig Lake near the end of the field work, small non-flowering stems seen in other lakes may have belonged to the same species. Potamogeton illinoiensis is somewhat outside its known range in Ontario. It was found only in Kenny Lake, which is immediately beside the highway, and could have been introduced by the considerable human activity in the area. Utricularia purpurea 1s also well north of its known range in Ontario. It was found only in Collette Lake, and was totally sub- merged in2to4 mof water. The unusually clear water of Collette Lake allowed the plant to be seen at these depths. It would easily be overlooked if it occurred at similar depths in other lakes in the area. Myriophyllum farwellii is listed as a rare vascular plant in Ontario by Argus and White (1977). How- ever, it was reasonably common in two lakes in our area, and also common in a northern Ontario lake about 120 km to the east (Fraser et al. 1980). Alkalinity of the water appeared to be the major factor influencing the plant species composition of the different lakes. The two most alkaline sites, North and South Tugboat Lakes, contained several species not found in the more softwater lakes, and not collected previously in the Park (Table 3). Of these, Potamo- geton pectinatus, P. foliosus, P. richardsonii and Elo- dea canadensis are listed as hardwater plants by Moyle (1945), and occur in other lakes of similar alkalinity in the Lake Superior area (unpublished data). Similarly, Swindale and Curtis (1957) include Potamogeton pectinatus, P. richardsonii and Myrio- phyllum exalbescens in a group of species found in mineral-rich water. 1983 Ca (ppm) 10 20 FRASER AND MORTON: AQUATIC PLANTS 183 30 40 50 TA (ppm) FiGuRE |. Relation between Ca and TA in the water of 14 lakes in Lake Superior Provincial Park. The regression line is based on II lakes, omitting (1) Moose Lake, (2), Mijin Pond, and (3) Boulder Lake, which had unusually high levels of Ca relative to TA. The anomaly is likely related to the high levels of Na and Cl, presumably derived from highway salt, in these three lakes. Lakes of lower alkalinity generally included typical softwater species such as Lobelia dortmanna, Erio- caulon septangulare, Scirpus subterminalis, and Isoetes echinospora (Vable 3). Potamogeton epihy-- drus was one of the most abundant pondweeds. I hese features are similar to the softwater Shield flora de- scribed in Minnesota by Moyle (1945), in softwater Ontario lakes by Miller and Dale (1979), and in the three most mineral-poor lakes studied by Crowder et al. (1977). Also, Keddy (1981) noted similar species in the very soft water of Axe Lake near Georgian Bay, Ontario, and Swindale and Curtis (1957) included Eriocaulon septangulare, Potamogeton epihydrus, and Myriophyllum tenellum in their group of plants with softwater affinities. Sparganium angustifolium, Nuphar variegatum and Nymphaea odorata were common in the softwater lakes studies here, but are also common in more alkaline waters in the Lake Superior area and elsewhere (unpublished data). Despite their high NaCl content and high specific conductivity, Moose Lake, Mijin Pond, and Boulder Lake retained a flora fairly typical of other lakes of similar alkalinity. However, Scirpus subterminalis, Isoetes echinospora and Lobelia dortmanna were absent from the three salt-rich lakes although com- mon in other lakes, including Rabbit Blanket Lake which had similar alkalinity. This may have been a chance effect, or it may be related to the salt content of these lakes. In southern Ontario, concentrations of highway salt appear to have caused major changes in roadside floras (Catling and McKay 1980). In lake water, however, highway salt forms more dilute solu- tions and presumably has less pronounced effects on the vegetation. Vol. 97 THE CANADIAN FIELD-NATURALIST mM] CO] +r] wuniyofisnsun wniupdsiods (IOMIOpprig) SLUDBINA DIADINIWIQ snapdyida ‘qd |SNUId]D “q (poam puod Wa}js-a1y\ ) sn3suojapid U0JasOUIDIOg (TOSI, 1938 A) mpjamanf{ unppAydoid PY 11ISUIGG Od UOJABOUDIOd i([IB1-S, 918) slipsjna siunddiyy (avs[e) avaovIRYyD SNIUILUDAS “J snyofyduin uojasoulnjod (SIIXA/f SDIDN ml wm] wo] wo] + Ww] A.) Qa} om (OFT, 1978 A\) suarsaqipxa un AYydolsd W |(PaaM191e AA ) SIsnapdDUD) DapoI] ((P9IaMpuod PRIY-PIy ) NUOSPADY IIA “q \snsouof ‘d |(paampuod O3kS) snjouijoad uUojasolwudjod SIdIN 20q ayisn y SSIS quao “S319 aMaIOD wo, Auusy 10 prod (J1OM1IVIS-19]P MM ) DUAIA IYIIIN JOD IsOoo| IyuR[g puog nqqey ull Jopjnog mxexe }e0q -3n] 9 -dn UNO Seto 184 AWUIPEY[L MO] JO sayR] Ul ATUreU WOO ay} Ie asoy) “AUER YstYy Jo sayey ut AJUreW puNo] a19M JST] ay) Jo do} ay) We SIULTY ‘PpayeUIsa aq JOU P[Nod sdURpUNG? JI paiou sem (q) 9 m) layers (Willson 1974) above these plots was estimated and averaged. I hese values were summed for each layer to produce a mea- surement of total vegetation volume. Canopy volume, the product of (0.10) (tree volume) (canopy height) was calculated. The occurrence of foliage in each corner of the column was also noted and totalled by layer for all plots, giving a maximum cover score of 40. Nest height, tree height, and nest tree dbh were recorded. Macrohabitat studies were conducted within circles of 3.14 ha (radius of 100 m) around the nesting site. The circle was divided into 10 X 10 m squares and cover-mapped to determine the area occupied by 200 1983 ARMSTRONG AND EULER: HABITAT OF BUTEO 201 —\s BURK'S “FALLS / \ ORILLIA \ HALIBURTON ONTARIO Ficure |. Central Ontario study area (bounded by dashed line) in relation to Lake Simcoe, showing major settlements and highways (route number circled). Inset shows position of area within Ontario. coniferous and deciduous forest, water, open area, and man-made clearings. The % slope, aspect, eleva- tion, and distances to the nearest lake, man-made clearing, and water body were recorded. The position of the nest tree on the slope in relation to the nearest body of water (lake position) and the local terrain (valley position) were recorded on a linear scale of | (lowland) to 6 (upland plateau). The type of water body closest to the nest was also recorded ona linear scale from | (seasonal stream) to 7 (lake). A Develop- ment Index (D1) for the nearest 100 m section of lake shoreline was measured, calculated as the greatest length of linear cleared shoreline. Data tor Red-shouldered and Broad-winged Hawk nests were analysed using Discriminant Function Analysis (DFA) (Dixon 1975) to select the habitat 202 THE CANADIAN FIELD-NATURALIST variables that most clearly differentiated the nesting sites of these raptors according to sample variance (James 1971; MacKenzie and Sealy 1981). The step- wise DFA selects, one at a time, those variables which best discriminate among the two groups of nests, and provides an F-Matrix to test significance. While the variables which provide maximum discrimination can be used to plot observations and group means accord- ing to two canonical variables, we were primarily interested in determining the habitat variables involved. Microhabitat and macrohabitat measure- ments were analysed separately to examine nest site selection and nesting habitat selection respectively. Data were re-entered if a nest was used in more than one year. We acknowledge the possible bias with small samples. Micro- and macrohabitat characteristics were sim- ilarly measured around known perching sites of the Broad-winged Hawk, and compared with nesting sites by DFA. All perching sites used more than once anda random selection of all remaining sites were studied. Field staff recorded observations of raptors oppor- tunistically during the breeding season, following a modification of Craighead and Craighead’s (1956) technique. Data recorded included general forest type and density, proximity to cottage development, and perching site. Habitat coverage was not uniform, but was dependent on other field activities. YOUNG NEAR NEST Voles7 Results Nesting Chronology An inventory of 101 potential nests resulted in the documentation of 27 active Broad-winged Hawk nests and nine Red-shouldered Hawk nests. Nest building was observed in early May for the Red-shouldered and from early May through mid-June for the Broad- winged Hawk (Figure 2). Most activity involved gar- nishing the nests with coniferous twigs. Many nests had been built in a previous year. As nests were only observed from the ground, it was not possible to ascer- tain the precise time of hatching. Nestlings of the Red-shouldered Hawk were observed as early as the third week of May, while young Broad-winged Hawks were observed in the second week of June. Fledging was observed in early July for the Red-shouldered Hawk and from mid- to late July for the Broad- winged Hawk. Back-dating suggests initial egg laying in late April for the Red-shouldered Hawk and mid-to late May for the Broad-winged Hawk (Matray 1974; Wiley 1975). Habitat Characteristics The DFA indicated that the nest microhabitats of these two raptors were separated by four habitat vari- ables: number of trees, ground cover, nest height, and percent coniferous trees (Table 1)(F = 7.31, df= 4, 31, P <0.05). All Red-shouldered Hawk nests and 88.9% of Broad-winged nests were classified correctly using -— RED-SHOULDERED HAWK -— —4 BROAD- WINGED HAWK OBSERVED FLEDGING NESTLINGS IN NEST APPARENT INCUBATION BEHAVIOR NEST ADORNMENT 7 142 28 5 2 iI) 2 APRIL MAY OF) 16) 23) S30) 07), Vaas 2 28) en 4tin ent een S ame JUNE JULY AUGUST FiGuRE2. Temporal trends in observed nesting activities of Broad-winged and Red-shouldered hawks; 1977 to 1979 data combined. 1983 ARMSTRONG AND EULER: HABITAT OF BUTEO 203 TABLE 1. Microhabitat characteristics (mean + Sb) of Broad-winged and Red-shouldered Hawk nests, and Broad-winged Hawk perching sites. Habitat Characteristic Red-shouldered Hawk Nest (9)! Canopy height (m) 24.44 3.5 No. shrub species AQ 28 O57) No. shrubs Wey 32 Ags) Coniferous shrubs (%) 8.0 + 12.2 No. tree species 3.0 a8 ie? No. trees 17.7+ 6.32 No. small trees (8-32 cm dbh) NAM se 75 No. large trees (32 cm dbh) Saf ae DO Coniferous trees (%) 10.7 + 13.6 Ground cover 25.7+ 9.52 Shrub cover 26.9 + 10.4 Tree cover 37.3 22 23 Ground volume (%) IO, [| se 92 Shrub volume (%) IS 28 ©, Tree volume (%) 296+ 7.0 Canopy volume? 72.4 + 20.4 Vegetation volume (%) 57/10 26 12 Nest tree height (m) 2B, \| 32 S27 Nest tree dbh (cm) 48.3+ 11.6 Nest height (m) 140+ 3.62 'Sample size. Broad-winged Hawk Nest (27) Broad-winged Hawk Perching Site (17) DN 22 2Qe7 114+ 8.9 3.3325 {40 i325 2,3 18.1 = 12.83 B32) 35 ISH) 16.3 = 24.3 11.44 17.3 48+ 1.5 A) ae 7,1 28.64 8.6 13.24 12.1 24.6 + 10.0 11.64 11.1 40+ 2.83 1S2 18 17.6 = 22.8 14.4 + 25.0 294+ 5.4 31.7 8.0 27.6+ 6.63 10.5 = 10.9 34.14 48 7.1 = 10:9 180+ 6.8 29.6+ 14.1 18.14 6.23 84+ 9.4 31.4 = 10.0 46+ 7.3 64.9 + 21.03 9.9 + 16.3 66.6 + 15.4 41.8 + 22.6 A\loll a8 BO a 44.2 + 16.2 = IES =e228 — ?Variables included in the discriminant function of Red-shouldered and Broad-winged Hawk nests. 3Variables included in the discriminant function of Broad-winged Hawk nesting and perching sites. 4(Tree volume x canopy height) + 10 these criteria. This separation characterized Red- shouldered Hawk nesting habitat as mature, open’ deciduous forest with low ground cover, and Broad- winged Hawk habitat as dense, deciduous-dominated, mixed forest with higher ground cover. Some Red- shouldered Hawk nests were also located in the latter type of habitat. Macrohabitat characteristics for nests of these rap- tors differed only for elevation (F = 7.93, df= 1, 35, P< 0.05) (Table 2). Separation was only fair: 86% and 57% of the Broad-winged and Red-shouldered Hawks respectively were correctly classified. Five microhabitat variables were included in the DFA between Broad-winged Hawk nests and perch- ing sites: number of shrubs, shrub cover and volume, number of large trees, and canopy volume (Table |) (F = 59.83, df = 5, 38, P< 0.05). All sites were cor- rectly classified using these variables. Three macro- habitat variables were useful in differentiating nesting and perching sites: area of man-made clearings, dis- tance to clearing, and slope(Table 2)(F = 21.6, df= 3, 41, P< 0.05). Most sites (88.9%) were classified cor- rectly, using these criteria. The Broad-winged Hawk nested in several species of tree including white birch, yellow birch (Betula alleghaniensis), and sugar maple (Table 3). The Red- shouldered Hawk nested in three species, primarily yellow birch and American beech. Nesting Success Of the 16 Broad-winged Hawk nests for which either nesting failure or the presence of young near fledging age were documented, 14 fledged young. Fledging success averaged |.5 young per nest and 1.7 young per successful nest. The highest number of young fledged per nest was 3. Five of six Red- shouldered nests with documented outcomes were successful. Fledging success was |.8 young per nest and 2.2 per successful nest. Highest number of young fledged per nest was also 3. Some nests of both species were not located until young had already hatched, possibly introducing some biases to the nesting suc- cess calculations (Mayfield 1961). Raptor Observations Totals of 577 Broad-winged and 64 Red-shouldered Hawk observations were made in both years from May through early September. A small number of Red-tailed Hawks (20) were also observed. Although there are obvious biases inherent in using casual observations, such observations yield general infor- 204 THE CANADIAN FIELD-NATURALIST Vol. 97 TABLE 2. Macrohabitat characteristics (mean + sp) of Broad-winged and Red-shouldered Hawk nests, and Broad-winged Hawk perching sites. Broad-winged Hawk Nest (27) Broad-winged Hawk Perching Site (17) Habitat Red-shouldered Characteristic Hawk Nest (9)! Lake position (1-6) 49+ 1.4 Valley position (1-6) 32pm leS Distance to lake (m) 223.9 + 146.8 Distance to clearing {m) Ps se NSP Distance to man-made clearing (m) 137.8 £ 158.3 Distance to water (m) 62.3+ 38.5 Type of water (1-7) Sof ae 19) Slope (%) DES) 32 ANS Area coniferous forest (100 m2) Hh ae) 2 Area deciduous forest (100 m2) 262.9+ 29.1 Area water (100 m2) [WES Tat 352. Area opening (100 m-) 40.4+ 28.5 Area man-made clearing (100 m2) 31.64 26.7 Shore Dev. Index 0.7+ 0.7 Sine of aspect 04+ 0.7 Cosine of aspect O.1+£ 0.7 Elevation (m) 295.5+ 60.92 ‘Sample size 42+ 1.6 3.5 S58 24) §.giae | es) Jae 20) 215.1 = 232.1 491.4 = 357.7 42.14 30.23 ae Qe 180.0 = 301.9 6.0 16.8 51.54 30.8 31.34 34.5 48+ 2.0 4:5 see? DORE 16893 144+ 11.1 28.94 51.7 16.2 26.2 231.4 71.8 161.0 60.2 Sp).0 28" S03) 58.8 + 54.8 52.54 40.6 133.24 46.3 2 Onli 255/23 CO) ae 37/3 OL ae = OL) = O1£ 0.8 0.3 0.8 Oas OF -0.2+ 0.6 350.44 48.8 362.54 24.4 ?Variables included in the discriminant function of Red-shouldered and Broad-winged Hawk nests. 3Variables included in the discriminant function of Broad-winged Hawk nesting and perching sites. mation on habitat use (Table 4). Most Broad-winged Hawks were located by sight and most Red- shouldered Hawks by sound. Significantly more Red- shouldered Hawks were found in dense forests than Broad-winged Hawks which were more common in forest areas with openings (x? = 66.6, P< 0.001). Greater proportions of Red-shouldered Hawks were also sighted in locations away from shores (x? = 21.4, P <0.001). Red-shouldered Hawks were observed significantly more often in uncottaged areas than were Broad-winged Hawks (x2= 14.7, P< 0.001). The most commonly observed Broad-winged Hawk perch- ing sites were power lines, while those of Red- shouldered Hawks were deciduous trees. Casual observations revealed that most territorial interac- tions occurred at forest canopy openings such as shorelines, powerlines and roads. Discussion Broad-winged Hawks were the most commonly observed raptors in the study area, outnumbering Red-tailed Hawks approximately 18 to 1. As bothare relatively conspicuous because of their perching and soaring habits, this figure probably reflects relative abundance. However, the secretive behaviour and preference for closed forest of the Red-shouldered Hawk reduces its visibility, so that it is likely more common than the ratio (7.2:1) of Broad-winged to Red-shouldered Hawk observations suggested. The Red-tailed Hawk is typically an open-country nesting bird (Gates 1972; McInvaille and Keith 1974), although it is adapted to a wide range of habitat conditions (Beebe 1974). Its nesting habitat is quite different from that of the other two species (Titus and Mosher 1981). Most observations of this raptor were made at perching sites along roadsides and in areas with open beaver meadows and swamps. Our limited behavioural observations suggest that Broad-winged Hawks react aggressively to the Red-tailed Hawk, as do Red-shouldered Hawks (Campbell 1975). Both Broad-winged and Red-shouldered Hawks ~ frequent deciduous, upland forests near riparian or lakeshore habitat. The Broad-winged Hawk often frequented partially open forests like those along roadways, while the Red-shouldered Hawk was almost totally restricted to closed forests. These two species had no significant differences in general nest- ing macrohabitat. Inclusion of elevation in the dis- criminant function seemed to be a result of sampling bias, as several Broad-winged Hawk nests and most 1983 TABLE 3. Tree species used for nesting by Red-shouldered and Broad-winged Hawks Number Nests Broad- Red- winged shouldered Species Hawk Hawk White birch 7 0 Yellow birch 5 4 Sugar maple 4 | American beech 0 4 Red oak 3 0 Large-toothed aspen (Populus grandidentata) 2 0 White pine | 0 Trembling aspen (Populus tremuloides) | 0 Eastern hemlock l 0 Basswood (Tilia americana) | 0 Other 2 0 TOTAL 27 9 TABLE4. Locations of observations of Broad-winged and Red-shouldered Hawks, 1978-1979 Number of Observations Broad- Red- winged shouldered Habitat Description Hawk Hawk Forest Type Coniferous 24 3 Coniferous dominated mixed 49 l Mixed 111 13 Deciduous dominated mixed 137 16 Deciduous 232 29 Forest Density Dense 162 43 Some openings 291 14 Many openings 81 3 Location! Lakeshore 96 4 Backshore 259 52 Over lake Si) l Road edge 289 9 Cottage Description Cottaged 47 | Near cottaged 153 7 Uncottaged 340 5 Perching Site Coniferous 13 6 Deciduous tree 70 2 Dead tree or hydro pole 37 2 Hydro wire 13 | Ground 4 | 1551 observations, some overlap. ARMSTRONG AND EULER: HABITAT OF BUTEO 205 Red-shouldered Hawk nests were in one area of low elevation. General habitat conditions for both species included location on a slope near a natural clearing, close to a water source, ina predominantly deciduous forest. This association with slope and water was not due to emphasis on field work conducted near lake- shores, as nests were located and studied away from shores. Broad-winged Hawk nests in the mid-western U.S. and nests of both species in the eastern U.S. show similar relationships with water and clearings (Keran 1978; Titus and Mosher 1981). Discriminant function analysis of nesting sites of four woodland raptor spe- cies in Maryland showed that Broad-winged and Red- shouldered Hawk nest sites could not be differentiated along the first discriminant function (weighted to dis- tance to water and nest height), but they were separ- ated along the second discriminant function (weighted to basal area, dbh of nest tree, and distance to nearest opening) (Titus and Mosher 1981). More specific habitat selection was involved in the precise nest location. Red-shouldered Hawk nests were usually found in areas with fewer but larger trees, and lower ground cover and conifer composition. Their nests were generally higher and nest height was important in the discriminant function. Red- shouldered Hawk habitat was thus typically mature, open deciduous forest, while Broad-winged Hawk habitat was younger, deciduous-dominated mixed forest. In southern Ontario, the Red-shouldered Hawk nests in similar habitats to those found in this study (Campbell 1975), although elsewhere they nest primarily in riparian areas (Stewart 1949; Portnoy and Dodge 1979; Galli et al. 1976). In this study, Red-shouldered Hawk habitat (nest sites and occu- pied breeding ranges) fell into two broad categories: lowland riparian and upland deciduous. The Red- shouldered Hawk nested in only three species of tree, principally beech and yellow birch. The three species used here also were the most commonly used nest trees in southwestern Quebec (Morris et al. 1982). Beech or other species of birch have been commonly used else- where as nest trees (Stewart 1949; Henny et al. 1973; Campbell 1975). The Broad-winged Hawk nested ina variety of trees, but predominantly white birch, yellow birch, and sugar maple. Early studies indicated that trees were used mainly in accordance with their abun- dance (Burns 1911). The overlapping of the nesting habitats of these two species would most likely occur in the moist slopes and valleys alongside waterbodies. The Red- shouldered Hawk appears to have the advantage in competition for nesting sites, because it begins nesting earlier (Burns 1911; Henny et al. 1973; Matray 1974; Portnoy and Dodge 1979). Red-shouldered Hawk nests were generally higher than those of Broad- winged Hawk nests, unlike the reverse situation in 206 western Maryland (Titus and Mosher 1981). Newton (1976) suggested that, where nesting sites are abun- dant, interspecific competition may only result in the displacement of one species to another site in the home range with no effect on breeding density. This would not likely be the case with these raptors,’ because of the aggressive interactions observed. Occupied territories appeared to be almost mutually exclusive. Both the Red-shouldered and Broad-winged Hawks require continuous forest areas. The Red- shouldered needs a minimum of 10 ha (Galli et al, 1976) and possibly as much as 2500 ha of generally forested habitat (Bednarz and Dinsmore 1981). It was rarely seen outside of dense forest cover, while the Broad-winged often was. The DFA of perching and nesting sites of the Broad-winged Hawk verified that open areas are important to that species. Nesting and perching sites were differentiated on the basis of clearing size and adjacency, and slope. Exposed sites such as dead trees, transmission line wires, and exposed deciduous branches are preferred perching sites. Road edges, beaver meadows and lakeshores were all frequented. Similar clearings are used in Minnesota and Wiscon- sin (Keran 1978). Although large-scale clearing of the forest would be detrimental, the existing degree of cottage develop- ment may not seriously disrupt Broad-winged Hawk breeding habitat. Territories, which appeared to be distinct and non-overlapping, were separated and defended along openings in the forest canopy, whether roadways, clearings, or shorelines. Most intraspecific interactions took place near such areas. The Broad-winged Hawk seems to defend exclusive home ranges as does the Red-shouldered Hawk (Craighead and Craighead 1956; Newton 1976), with the defended “activity centre” including a breeding, nesting and feeding territory (Keran 1978). If open areas are necessary to divide territories, the presence of lawns, roads, and hydro lines that occur with cot- tage development may slightly increase the potential number of Broad-winged Hawk territories in an area. Aside from major habitat alteration, one of the other major effects of cottaging on the Broad-winged Hawk may be human disturbance. This species is most conspicuous in late May, probably due to activity associated with nest establishment. As the first major influx of cottagers occurs at that time (Victoria Day weekend), nest disturbance or desertion could result. The site tenacity of most raptors is weakest at the time of territorial establishment which could make it one of the most critical periods for nest desertion (Fyfe and Olendorff 1976). Heavy recreational use in Wyoming has had significant negative impacts on nesting buteos (Craighead and Mindell 1981). THE CANADIAN FIELD-NATURALIST Vol. 97 The Red-shouldered Hawk did not respond posi- tively to cottage development, although it often nested close to clearings and, in some cases, roads. Propor- tionally more sightings of this raptor occurred in uncottaged areas. Cottage roads that encircle lakes 50 to 100 m from the shore could fracture and isolate potentially valuable riparian and lakeshore habitat. The importance of nesting sites near permanent or seasonal water has also been documented elsewhere (Wiley 1975). The nesting success rate of the Red- shouldered Hawk is directly proportional to the dis- tance between adjacent sites (Henny et al. 1973). This suggests that, if available breeding habitat were lost to cottage development, the response could be a compac- tion of the breeding population and a decreased reproductive success. A nesting population of the Red-shouldered Hawk had not been documented previously in the study area (Peck 1979). We documented 9 nesting attempts of this species although it is considered to be declining in central Canada and elsewhere in its range (Brown 1971; Fyfe 1976). It appeared to be widely distributed in the study area, as there are large areas of suitable habitat and Red-shouldered Hawks were observed in areas where nests were not located. Given its general population status and the fact that nest trees may be limiting factors in otherwise suitable habitats (Titus and Mosher 1981), the effects of cottaging on habitat quality must be carefully considered in the context of its overall status. Acknowledgments We would like to acknowledge the assistance of the 1978 and 1979 field staff. Particular thanks are extended to G. Racey, M. Heimberger, J. Hall- Armstrong, M. Sobchuk, and B. Ratcliff. Without their efforts, this study could not have been con- ducted. W. Richardson, B. Penak, S. G. Sealy and an anonymous reviewer kindly reviewed this manuscript. Literature Cited Bednarz, J. C., and J. J. Dinsmore. 1981. Status, habitat use and management of Red-shouldered Hawks in lowa. Journal of Wildlife Management 45: 236-241. Beebe, F. L. 1974. Field studies of the Falconiformes of British Columbia. British Columbia Provincial Museum, Occasional Paper no. 17. Brown, W. H. 1971. Winter population trends in the Red- shouldered Hawk. American Birds 25: 813-817. Burns, F. L. 1911. A monograph of the Broad-winged Hawk (Buteo platypterus). Wilson Bulletin 18: 143-320. Campbell, C. A. 1975. Ecology and reproduction of Red- shouldered Hawks in the Waterloo Region, Southern Ontario. Raptor Research 9: 12-17. Craighead, J. J., and F.C. Craighead, Jr. 1956. Hawks, owls and wildlife. Stackpole Co., Harrisburg, Pennsylvania. 1983 Craighead, F.C., Jr., and D. P. Mindell. 1981. Nesting raptors in western Wyoming, 1947 and 1975. Journal of Wildlife Management 45: 865-872. Dixon, W. J. 1975. BMDP Biomedical Computer Pro- grams. University of California Press, Los Angeles. Fyfe, R. W. 1976. Status of Canadian raptor populations. Canadian Field-Naturalist 90: 370-375. Fyfe, R. W., and R. R. Olendorff. 1976. Minimizing the dangers of nesting studies to raptors and other sensitive species. Canadian Wildlife Service Occasional Paper Number 23. Galli, A. E., C. F. Leck, and R. T. Forman. 1976. Avian distribution patterns in forest islands of different sizes in central New Jersey. Auk 93: 356-364. Gates, J. M. 1972. Red-tailed Hawk populations and ecol- ogy in east-central Wisconsin. Wilson Bulletin 84: 421-433. Godfrey, W.E. 1966. The birds of Canada. National Museums of Canada Bulletin No. 203. Henny, C. J., F. C. Schmid, E. M. Martin, and L. L. Hood. 1973. Territorial behavior, pesticides, and the population ecology of Red-shouldered Hawks in central Maryland, 1943-1971. Ecology 54: 545-554. James, F.C. 1971. Ordinations of habitat relationships among breeding birds. Wilson Bulletin 83: 215-236. Keran, D. 1978. Nest site selection by the Broad-winged Hawk in north central Minnesota and Wisconsin. Raptor Research 12: 15-20. MacKenzie, D. I. and S. G. Sealy. 1981. Nest site selection in Eastern and Western Kingbirds: a multivariate approach. Condor 83: 310-321. Matray, P. F. 1974. Broad-winged Hawk nesting and ecol- ogy. Auk 91: 307-324. Mayfield, H. 1961. Nesting success calculated from expo- sure. Wilson Bulletin 73: 255-261. MclInvaille, W. B., and L. B. Keith. 1974. Predator-prey relations and breeding biology of the Great Horned Owl and Red-tailed Hawk in central Alberta. Canadian Field- Naturalist 88: 1-20. ARMSTRONG AND EULER: HABITAT OF BUTEO 207 Morris, M. M. J., B. L. Penak, R. E. Lemon, and D. M. Bird. 1982. Characteristics of Red-shouldered Hawk, Buteo lineatus, nest sites in southwestern Québec. Cana- dian Field-Naturalist 96(2): 139-142. Newton, I. 1976. Population limitation in diurnal raptors. Canadian Field-Naturalist 90: 274-300. Peck, G. K. 1979. Ontario nest records scheme: Fifteenth report (1956-1978). Royal Ontario Museum, Toronto. Portnoy, J. W., and W. E. Dodge. 1979. Red-shouldered Hawk nesting ecology and behavior. Wilson Bulletin 91: 104-117. Postupalsky, S. 1977. A critical review of problems in cal- culating Osprey reproductive success. pp. !-I1 in J.C. Ogden, ed. Transactions of the North American Osprey research conference, U.S. Department of the Interior, National Parks Service Transactions Proceeding Series Number 2. Rowe, J.S. 1972. Forest regions of Canada. Canadian Forestry Service Publication No. 1300. Schnell, G. D. 1968. Differential habitat utilization of win- tering Rough-legged and Red-tailed Hawks. Condor 70: 373-377. Stewart, R. E. 1949. Ecology of a nesting Red-shouldered Hawk population. Wilson Bulletin 61: 26-35. Tate, J., Jr. 1981. The blue list for 1981. American Birds 35: 3-10. Titus, K., and J.A. Mosher. 1981. Nest-site habitat selected by woodland hawks in the central Appalachians. Auk 98: 270-281. Weller, M. W. 1964. Habitat utilization of two species of buteos wintering in central lowa. lowa Bird Life 34: 58-62. Wiley, J. W. 1975. The nesting and reproductive success of Red-tailed Hawks and Red-shouldered Hawks in Orange County, California, 1973. Condor 77: 133-139. Willson, M. F. 1974. Avian community organization and habitat structure. Ecology 55: 1017-1029. Received 5 January 1982 Accepted 24 September 1982 Dependence of Clark’s Nutcracker, Nucifraga columbiana, on Conifer Seeds during the Postfledging Period STEPHEN B. VANDER WALL! and HARRY E. HUTCHINS2 'Department of Biology, Utah State University, Logan, Utah 84322 ?Department of Forest Resources, Utah State University, Logan, Utah 84322 Vander Wall, Stephen B., and Harry E. Hutchins. 1983. Dependence of Clark’s Nutcracker, Nucifraga columbiana, on conifer seeds during postfledging period. Canadian Field-Naturalist 97 (2): 208-214. We studied the behavior of Clark’s Nutcracker ( Nucifraga columbiana) at two times during the postfledging period. First, we observed nutcrackers recover stored Whitebark Pine (Pinus albicaulis) seeds in subalpine areas at Grand Teton and Yellowstone National Parks, Wyoming, USA, during June and July of 1979-1981. Seeds from caches were the only food items that we saw adults feed to their young. Of 163 caches located by adult nutcrackers, 20% contained germinating seeds. Juvenile nutcrackers learned to find caches by locating germinating seeds, and so became independent of their parents about mid-July. Second, we observed the behavior of nutcrackers foraging on Pifion (Pinus monophylla) and Limber (Pinus flexilis) pine cones inthe Raft River Mountains, Utah, USA, from August-October in 1978 and 1980. Juveniles began foraging on cones in mid-August and first cached seeds in early to mid-September, both forms of behavior occurring 2-3 weeks later than in adults. Juveniles foraged significantly more slowly than adults in late August and early September, but after 15 September their foraging rates were similar. Juveniles were highly aggressive towards foraging adults prior to 7 September, but not thereafter. We conclude that Clark’s Nutcrackers depend heavily on conifer seeds throughout the postfledging period. Key Words: Clark’s Nutcracker, Nucifraga columbiana, conifer seeds, caching, seed-recovery, juvenile, behavior. Clark’s Nutcrackers (Nucifraga columbiana) depend extensively on conifer seeds during the winter and the breeding season. This dependence takes sev- eral forms. First, conifer seeds make up 70-100% of the diet in November to April (m= 281 stomachs; Giuntoli and Mewaldt 1978). Second, Clark’s Nut- cracker is one of the earliest breeding passerines in temperate regions, with egg laying occurring from late February to mid-April. Early breeding is thought to occur, however, only following large conifer seed crops (Vander Wall and Balda 1981). Third, nestling nutcrackers are fed conifer seeds almost exclusively (Johnson 1900; Bradbury 1917; Mewaldt 1956). The seeds for all these activities come from caches made during the fall, and nutcrackers must leave a region if seeds are unavailable for storage (Davis and Williams 1957, 1964; Vander Wall et al. 1981). Vander Wall and Balda (1981) argued that depend- ence on conifer seeds is an integrated component of a highly specialized seed-harvesting and storing syn- drome, and that level of dependence is a key factor in the evolution of specialized food-storage behavior in certain corvids. In this paper we examine dependence of Clark’s Nutcracker on conifer seeds during the postfledging period (from fledging to the post-juvenile molt), and discuss this dependence in the context of the model of seed-caching behavior developed by Vander Wall and Balda (1981). During this period nutcrackers engage in two important activities. First, nutcracker family groups congregate from May-July at communal cache areas where adults search for cached seeds and care for young.' Second, from August to November, juveniles learn to forage on conifer cones and store seeds for the winter. Tomback (1978) briefly discussed Clark’s Nutcracker postfledg- ing behavior, and Volker and Rudat (1978) described postfledging behavior of Eurasian Nutcrackers ( Nuci- fraga caryocatactes). Study Areas and Methods Cache-recovery behavior and cone-foraging behav- ior were studied in two separate areas where nut- crackers foraged for the seeds of different pines. (1) The recovery of cached Whitebark Pine ( Pinus albicaulis) seeds was observed at Surprise and Amphi- theater lakes (2960 m elev.) in Grand Teton National Park and Mt. Washburn (2700-2900 m elev.) in Yel- lowstone National Park, Wyoming. These areas are dominated by Whitebark Pine, but Subalpine Fir (Abies lasiocarpa), Engelmann Spruce ( Picea engel- mannii) and Lodgepole Pine (Pinus contorta) also occur. Quantitative data were gathered on 13 and 15 July 1979, 27 and 30 June, | July and 16-18 July 1980, and 17-19 June 1981. Adults and juveniles searching for cached White- bark Pine seeds were observed at close range (usually <20 m). All bill-probes into the substrate were assumed to be cache-recovery attempts. The number 'We operationally define “young” as nutcrackers dependent on their parents for food; the term “juvenile” refers to nutcrackers from independence to the completion of postjuvenile molt. 208 1983 of successful (one or more seeds recovered) and unsuccessful cache-recovery attempts was recorded for each foraging bird. The time to find a cache was measured with a stopwatch, and the distance traveled while searching was estimated in meters. Digging techniques used and numbers of seeds recovered were recorded. Following each successful cache recovery the excavation was inspected for unrecovered edible seeds, spoiled seeds, and germinating seedlings. The fate of recovered seeds was recorded as: |) shelled and eaten or placed in sublingual pouch at the cache site, 2) shelled and fed to young at the cache site, 3) trans- ported (in bill or pouch) intact away from the cache site, and 4) re-cached. The presence and behavior of attendant young were noted. (2) Cone-foraging adult and juvenile nutcrackers were observed at two sites in the Raft River Moun- tains of northwestern Utah from August to November . of 1978 and 1980: a Limber Pine ( Pinus flexilis) stand along a north-facing ridge (2525 m elev.) overlooking Clear Creek drainage (described in Lanner and Vander Wall 1980), and a pifion-juniper (Pinus monophylla-Juniperus osteosperma) woodland (1825-2450 m elevation) occupying a south-facing slope. ‘ Foraging rates of adults and juveniles were deter- mined by measuring with a stopwatch the time required to remove Pifion or Limber pine seeds from cones (search time), and to shell and ingest the edible contents (handling time). Mean search time of adults during an observation period was calculated as the unweighted mean of individual foraging observations in which three or more seeds were extracted. Foraging data were gathered at 5-7 day intervals throughout the fall to monitor changes in foraging rates as cones ripened, and to determine differences in adult and juvenile foraging rates as juveniles gained foraging experience. Qualitative differences in adult and juve- nile foraging behavior were noted. All agonistic encounters and the ages of participants were recorded. Results When the subalpine cache areas became accessible to us in early June, nutcracker family groups were already present. At this time most young were strong and skilful fliers, but they were still completely dependent on adults for food. Adult Cache- Recovery Behavior Adult nutcrackers frequently searched for caches in loose aggregations of 2-10 birds, but within these groups each individual foraged independently and, for the most part, ignored other birds. Two patterns of search were used to locate cached seeds. Birds fre- quently flew or hopped directly toa cache site, probed in the ground with their bills, and extracted seeds as if VANDER WALL AND HUTCHINS: CLARK’S NUTCRACKER 209 they knew the cache location. They also engaged in relatively prolonged searches which entailed slow deliberate inspection of the ground surface. Distance travelled during prolonged search ranged from 3-50 m (x +SD=12.0+9.8 m, n= 45) at a mean rate of 0.19 m/sec (SD = 0.09 m). Only 36% of prolonged searching bouts ended in cache discovery (n = 105). Nutcrackers retrieved seeds by either probing verti- cally with the bill (86.5% of observations, n = 148), flicking debris aside with lateral movements of the bill (8.1%), or a combination of the two (5.4%). Lateral bill movements were used only in areas overlain by forest litter. Probe holes averaged 1.2 X 2.1 cm in diameter (n= 23) with a mean depth of 2.5 cm (n = 14). Recovered caches ranged in size from one to 14 seeds (X + SD = 3.37 = 2.39 seeds, n = 155). Nut- crackers spent an average of 25.7 sec at cache sites (range = | — 224 sec, SD = 42.9 sec, n = 43). Time ata cache site was correlated with number of seeds retrieved (r = 0.48, P<0.05). The percent of probes resulting in cache recoveries was 31.5% in mid-June (1981), 33.3% in late June and early July (1980), and 82.4% in mid-July (1979). Overall success rate was 44.1%. These rates were calculated from data col- lected in different areas during different years, so they cannot be taken asa reliable trend in recovery success rate through the season. At least 32 of the 163 caches located by adults contained one or more germinating seeds. Nut- crackers located germinating seeds soon after the seed stem or hull broke through the soil surface. Nut- crackers detached the seed from its stem, shelled it, and ingested the remaining contents before excavat- ing the area around the stem to find other seeds. Most germinating seeds were located during prolonged search, and we suggest that birds employing this behavior are searching for germinating seeds. The proportion of recovered caches containing germinating seeds changed significantly during the summer (X2 = 32.6, P<0.001). Germinating seeds did not occur in caches recovered in mid-June (n = 41), but in late June and mid-July 61.1% (n= 36) and 43.5% (n = 23) of the caches, respectively, contained germinating seeds. Whitebark Pine seeds apparently began germinating between mid and late June during the years of this study. Bibikov (1948) reported that sprouting Cedar Pine (Pinus sibirica) seeds were fre- quently taken by Eurasian Nutcrackers during June and July, and Bossema (1968) found that many, if not all, acorns eaten by the jay (Garrulus glandarius) in June and July were taken from seedling oaks (Quercus Sp.). Nearly 55% of all recovered seeds were shelled at the cache site and either eaten or placed in the sublingual pouch (these actions were not always distinguishable). 210 In the latter case, seeds were then transported out of the cache area and presumably fed to young. Of the remaining seeds, 34.5% were placed in the sublingual pouch, transported short distances (usually <100 m), shelled, and either eaten or fed to young, 8.2% were opened at the cache site and fed directly to begging young, and 2.7% were immediately re-cached in new sites. Following cache recoveries we inspected 95 excava- tions for undiscovered seeds. Only five edible seeds remained. These seeds were from three caches and accounted for 1.6% of the total seeds cached. One of these seeds was frozen to the substrate. We suspect that the bird left seeds because of disturbances by the observers and begging young at the other two caches. Eight mouldy seeds were left at or near cache sites. Nutcrackers generally concentrated their search effort where snow had recently melted: 43.2% of all searches began within 1.0m of snow, and 69.2% began within 5.0 m of snow. On several occasions nutcrackers pecked through ice and snow at the edge of the snow pack to recover seeds. One nutcracker spent 75 sec to break and remove ice 1.0 cm thick to reach a cache situated 2.0 cm behind the edge of a snow bank. As nutcrackers tended to forage near the periphery of the melting snow pack, centers of cache- recovery activity shifted markedly during the course of the summer. Behavior of Young on the Cache Area While adult nutcrackers searched for cached seeds, groups of young usually waited in nearby trees. Young frequently gave kra-a-a calls similar to those given by adults but higher pitched and repeated more fre- quently (20-30 times/ min). When adults approached with small loads of shelled seeds in their pouches, the young flew or hopped toward the adults and began giving the hunger call (cf. Mewaldt 1956), a frantic near. \f the adults arrival was not observed by the young, the adult gave soft near or nert calls to which the young responded with hunger calls and rapid approach. As a young approached an adult its wings were held partially outstretched and quivered rapidly. The mouth was gaped wide and the head titled slightly upward. The adult then ejected seeds from its pouch with quick jerks of the head and, holding the seeds one at a time in the tip of the bill, inserted them deep into the young’s mouth. Each young typically received 3-7 whole seeds in quick succession. We never saw adults feed young anything other than Whitebark Pine seeds. At 24.2% ot adult cache recoveries, young hopped along the ground 1-5 m behind the adult, giving the high-pitched kra-a-a or hunger calls. Adults ignored these young, but when a cache was located the 1-4 young frequently crowded too close and were driven back with pecks to the face. Whole seeds were fed to THE CANADIAN FIELD-NATURALIST Vol. 97 the young individually as they were removed from the cache and shelled. On three occasions young were fed a mouthful of snow. Young nutcrackers from different broods roosted and moved about independently. Young from one brood were occasionally attracted to and approached begging birds ina nearby brood, but these young were ignored or driven away by adults. Young nutcrackers regularly “foraged” on their own between feedings. They probed into needle clus- ters, removed exfoliating bark, and manipulated a wide variety of inedible objects (e.g. twigs, pebbles, old cone cores). Young that followed foraging adults frequently examined excavations made by them, probed into the holes, and manipulated seed shells and mouldy seeds near the hole. During mid-July we saw young nutcrackers locate four caches; all contained germinating seeds. At this time many young appeared independent, and the vocalizations of most young were indistinguishable from those of adults. Young occasionally foraged near other young or adults but these foraging associations lacked the integrity of family groups. Two young observed for over 90 min were not visited or fed by adults. Although several feedings of young were observed, most of these were of a bird that stayed near and appeared dependent on an adult. Cone-foraging Behavior of Adults The behavior of adults foraging on cones 1s briefly summarized here for comparison with juvenile cone- foraging behavior. Prior to mid-August most foraging is for arthrop- ods and seeds cached the previous fall, but in late July and early August adult nutcrackers begin foraging on ripening conifer cones. At this time the seeds of pre- ferred pines (e.g. Pifion and Limber pines) are not fully developed, having dry weights only 5-35% and caloric values (per g ash-free dry weight) 60 to 90% those of mature seeds (Vander Wall unpublished data). By mid-August, seeds were 25-60% of their mature weight and 85-95% of their mature caloric value, and birds were frequently observed foraging in the tree-tops. A nutcracker grasps the base of a cone with its feet and hammers with its bill between the overlapping cones scales. The head is then twisted to the side and the bill, acting as a lever, pries the scales apart. After several such prying movements, a seed is exposed and extracted with the bill. Starting in the last few days of August, seeds may be deposited in the sublingual pouch to be stored later. Seed extraction rates changed predictably during the course of the harvest period. For Pifion and Limber pines, extraction rates in mid-August were | -2 seeds/ min. As cones dried and seeds ripened, the adult 1983 seed extraction rate steadily increased to a peak of 7-12 seeds/ min in late September, when most cones opened. The seed extraction rate decreased to <6 seeds/min from mid-October through November as the seed crop became depleted. The pattern is similar for Whitebark Pine (Hutchins and Lanner 1982). Development of Cone-foraging and Seed-handling behavior in Juveniles Juvenile nutcrackers were first seen foraging on Pifion Pine cones on 14 August in 1978 and on 19 August in 1980. On Whitebark Pine, juvenile foraging was first observed on 15 August in 1980. During the first few weeks of cone foraging, juvenile behavior differed both quantitatively and qualita- tively from that of adults. Until 15 September, juve- nile seed extraction rate was significantly slower than that of adults (t-test, P<0.05, Figure !) with juveniles frequently requiring 2-4 times longer to extract a seed. Juvenile nutcrackers hammered less forcefully than adults, did not concentrate hammering at a specific point, did not always direct hammering between the cone scales, and in general appeared less coordinated than adults. Juveniles failed to extract seeds in approximately half their foraging attempts. As juve- niles gained foraging experience and cones dried and began to open, the differences between adult and juvenile seed extraction rates and foraging behavior l2 Juv.-x Ad. Foraging Rate (sec/seed) 20 | VANDER WALL AND HUTCHINS: CLARK’S NUTCRACKER 211 disappeared. This occurred between 15 and 23 Sep- tember for birds foraging on Pifion Pine in 1980. Seed handling time also differed between adults and juveniles. Mean handling time for adults remained ‘fairly constant during the course of the harvest season: 9.6 sec/seed (SD = 4.5 sec, n = 813) for Limber Pine and 19.4 sec/seed (SD = 10.7 sec, n = 241) for Pinon Pine. As juveniles usually broke seeds during the extraction process, handling times by juveniles prior to 7 September were difficult to document. Adults seldom break seeds after late August. Limited data (n = 10) suggest that prior to 7 September juveniles required a mean of 20.5 sec(SD = 12.3 sec) more than adults to handle seeds. Long handling times for juve- niles prior to 7 September were due to excessive manipulation of seeds in the bill and difficulty in cracking the seed coat and in separating the contents from the seed coat. However, juveniles rapidly gained proficiency. After 12 September, they handled seeds at mean rates of 10.5 sec/seed (SD = 6.3, n = 30) for Limber Pine and 23.5 sec/seed (SD = 8.9, n = 6) for Pifion Pine; not significantly different from adults (Mann-Whitney U-test, P<0.05). Adult nutcrackers were first seen filling their sub- lingual pouches with Pifion seeds on 27 August 1978 and 2 September 1980, an indication that seed storage had begun. For Whitebark and Limber pines, nut- lO 20 SEP OCT Ficure |. Juvenile minus adult foraging rate from 26 August to 6 October on Pifion and Limber pines. Vertical line is the range, horizontal line is the mean, and vertical bar is + | SD. Number of foraging observations is given above vertical line. Two juvenile foraging observations on 14 August are represented in circles. 212 crackers first placed seeds in their pouches on 15 August in 1980 and 26 August in 1978, respectively. The first juvenile observed putting seeds into its sub- lingual pouch did so on7 September, but this behav- 10r was not commonly observed until 23 September. Although seed caching behavior in corvids appears to be innate (Ligon and Martin 1974), juvenile nut- crackers in the years of this study did not begin to store seeds until 2-3 weeks after adults had begun to do so. Juvenile Aggression We observed 145 aggressive interactions among nutcrackers foraging on Pifion Pine in 1980. In 106 of these, the ages of both participants were determined. Adults interacted aggressively 71 times, adults were aggressive toward juveniles 6 times, and juveniles were aggressive towards adults 29 times. To test whether the observed frequency of interactions between age classes was significantly different from random, we calculated expected frequencies of interactions using a binomial distribution. We assumed that the propor- tion of adults (p) in the population was 0.80. This value was selected for two reasons: |) it approximated the actual proportion of adults on our study site at the Aggressive Interactions /Obs. Hour 20 | AUG THE CANADIAN FIELD-NATURALIST Vol. 97 time the data were collected and 2) this value gives the highest probability of failing to reject the null hypothesis (1.e., aggressive interactions between age classes were random). Juveniles were aggressive toward adults more frequently than predicted by chance (X2 = 19.94, P<0.001). Juvenile aggression towards adults was very high on 19 and 26 August and rapidly decreased to zero by 7 September (Figure 2). Aggression among adults persisted throughout the harvest season at a low level and adult aggression toward juveniles was low and fluctuated erratically. In all cases (nm = 145), agonistic interactions began with the aggressor attempting to supplant a foraging bird. Most aggressive attempts (90.4%) were success- ful, resulting in the aggressor assuming the foraging station. Adults resisted juvenile attempts at displacing them only twice and juveniles failed to supplant adults only once (” = 29). Discussion Young nutcrackers are dependent on their parents for an unusually long time. The exact age of young observed in this study was not known, but assuming o—o Ad>Ad o----o Juv> Ad o—-0 Ad-» Juv 30 SEF FIGURE 2. Aggressive interactions per observation hour (adjusted to adult-juvenile ratio of 80:20, see text) between adult and juvenile nutcrackers foraging on Piffon Pine from 19 August-4 October 1980. Infrequent adult-adult aggression continued into November. 1983 eggs hatched in mid-April (e.g. Mewaldt 1956) nut- crackers achieving independence in mid-July were |3- 14 weeks old. Volker and Rudat (1978), working with individually marked Eurasian Nutcrackers, found that young separated from their parents at 15-17 weeks of age. These periods of dependence are much longer than those found in most other passerines (Skutch 1976). The extended period of parental care in nutcrackers is probably related to the unusual food supply on which the young depend. The seeds that make up the bulk of the diet of young nutcrackers are hidden in the soil, available only to the adults that recover them by remembering where they put them (Vander Wall 1982). Since young birds cannot efficiently forage for these seeds, they have no alternative but to rely on adults for food. As a consequence of this behavioral constraint, the development of the young’s foraging ability lags behind that of other behavior; e.g. by mid-June young nutcrackers are excellent fliers not easily distinguished from adults, though still depend- ent on them for food. When seeds begin to germinate, by late June in Whitebark Pine, they can be located and utilized by any foraging bird. It is at this time that young are first able to forage for seeds and make a significant contri- bution toward satisfying their energy demands. Young nutcrackers apparently acquire the ability to locate germinating seeds through observational learn- ing and socially facilitated behavior (Alcock 1969; Clayton 1978). Young, following their parents or other adults, may observe the adult pattern of search, see germinating seeds, and examine the excavation and seed shells at cache sites. Feedings, which fre- quently take place at the cache site, may reinforce the association between germinating seeds and a food source. [Alcock (1969, p. 320) suggested that “the interaction between observational learning and a bird’s reinforced socially facilitated experience could assist in the development and expression of a bird’s behavioral repertoire.”] Young nutcrackers achieve independence about the same time that they acquire the ability to find germinating seeds, processes that may be causally related. How long juveniles rely on germinating seeds is unknown, but since nutcrackers typically follow the receding snow pack to recover seeds, and since the snow pack may persist into August in alpine areas, it is probable that in many years germinating seeds can be utilized until mid- August, when foraging on new cones begins. Clark’s Nutcracker and several other seed-caching corvids are thought to be the primary dispersal agents for several species of wingless-seeded pines (Vander Wall and Balda 1977, 1981; Tomback 1978; Lanner and Vander Wall 1980; Lanner 1982; Hutchins and VANDER WALL AND HUTCHINS: CLARK’S NUTCRACKER 213 Lanner 1982). In an area less that 500 m2, we located 50 groups of one-year old seedlings, all, we suspect, germinated from nutcracker caches. While the exploi- tation of germinating seeds by nutcrackers may reduce the effectiveness of dispersal, this effect appears to be small, as the critical period during which germinating seeds may be destroyed appears to be very short, perhaps only 2-3 days following emer- gence. Unripe cones are difficult for nutcrackers to open, the task requiring considerable strength and skill. The development of juvenile cone-foraging skills, mea- sured by mean search time (Figure |), requires about four weeks to achieve the level of adults. Much of the juveniles’ improvement, however, is due to the drying and opening of cones (in Pifion and Limber pines), which greatly increases the foraging rate of all birds. Seed caching by juveniles begins 2-4 weeks later than in adults. This has important implications for juvenile winter survival because, relative to adults, the quan- tity of seeds stores by juveniles is likely to be small, and juveniles may not be able to store enough food to ensure survival through the winter in years of low to medium cone crops. Ashmole and Tovar (1968), Fogden (1972), and others have suggested that a long period of parental care is common in birds requiring a high degree of skill for foraging success. During this period of parental care the young develop foraging skills. Clark’s Nut- crackers, however, do not extend parental care into the fall seed harvest. Apparently, Clark’s Nutcracker has become so specialized at harvesting and storing conifer seeds that its entire annual cycle is dependent on its stored seed reserves (Vander Wall and Balda 1981). By the onset of the seed harvest adults must be free from their young to devote most of their time to harvesting and storing seeds. Consequently, juveniles must acquire cone-foraging skills on their own or through interactions with adults that are not necessar- ily their parents. Juveniles were aggressive toward foraging adults during the early portion of the seed harvest, a behav- ior which probably increased their access to seeds. Juvenile aggression may have been due to high hunger levels, a probable consequence of their slow, ineffi- cient foraging. Breaking into cones frequently took juveniles 2-5 min of constant hammering, an activity demanding much energy. Many foraging attempts ended in failure. It is possible that, while learning to forage on cones, juveniles were on the verge of starva- tion. Food-deprived birds may interact aggressively ata limited food source (Hinde 1952; Roth 1971), although others (e.g. Andrew 1956; Wiley and Hart- nett 1979) have argued that there may be no direct link 214 THE CANADIAN FIELD-NATURALIST between food deprivation and aggression. Aggres- sively displacing foraging adults may help juveniles maintain a positive energy balance. This supposition is supported by the observation that juvenile aggres- sion ceased by 7 September (Figure 2), about the time that juvenile foraging rates approached those of adults (Figure 1). Juvenile Steller’s and Pifion jays also typically dominate adults (Brown 1963; Balda and Balda 1978). This study extends the findings of Vander Wall and Balda (1981) by demonstrating that Clark’s Nut- cracker depends heavily on conifer seeds throughout the postfledging period. Conifer seeds, which are fed upon by young, juveniles and adults, appear to have had a pervasive role in shaping the natural history of Clark’s Nutcracker. This high level of granivory, exceeded by few other bird species, is made possible by the high nutritive value of conifer seeds (Botkin and Shires 1948; Lanner 1981) but has necessitated some ancillary adaptations of parental and foraging behavior to be successful. Acknowledgments We thank Peter Landres and Richard Mewaldt for their helpful comments and Marsha Sylvia for typing. This research was supported by grant DEB 78-02808 from the National Science Foundation to Ronald M. Lanner. Literature Cited Alcock, J. 1969. Observational learning in three species of birds. Ibis 111: 308-321. Andrew, R.J. 1956. Influence of hunger on aggressive behavior in certain buntings of the genus Emberiza. Phy- siological Zoology 30: 177-185. Ashmole, N. P., and S. H. Tovar. 1968. Prolonged paren- tal care in Royal Terns and other birds. Auk 85: 90-100. Balda, R. P., and J. H. Balda. 1978. The care of young Pifion Jays (Gymnorhinus cyanocephalus) and their inte- gration into the flock. Journal fiir Ornithologie 119: 146-171. Bibikov, R.I. 1948. On the ecology of the nutcracker. Trudy Pechorskogo-lIlychskogo Gosudarstvennogo Zapovendnika 4: 89-112. Bossema, I. 1968. Recovery of acorns in the European Jay (Garrulus g. glandarius .). Koninkliyke Nederlandse Akademie van Wetenschappen. Series C, 71: 1-5. Botkin, C. W., and L. B. Shires. 1948. The composition and value of Pinon nuts. New Mexico Experimental Sta- tion Bulletin 344: 3-14. Bradbury, W. C. 1917. Notes on the nesting habits of the Clark Nutcracker in Colorado. Condor 19: 149-155. Brown, J. L. 1963. Aggressiveness, dominance and social organization in the Steller’s Jay. Condor 65: 460-484. Clayton, D. A. 1978. Socially facilitated behavior. Quar- terly Review Biology 53: 373-392. Davis, J., and L. Williams. 1957. Irruptions of the Clark Nutcracker in California. Condor 59: 297-307. Vol. 97 Davis, J. and L. Williams. 1964. The 196] irruption of the Clark’s Nutcracker in California. Wilson Bulletin 76: 10- 18. Fogden, M.P.L. 1972. The seasonality and population dynamics of equatorial forest birds in Sarawak. Ibis 114: 307-343. Giuntoli, M., and L. R. Mewaldt. 1978. Stomach contents of Clark’s Nutcrackers collected in western Montana. Auk 95: 595-598. Hinde, R.A. 1952. Behaviour of the Great Tit (Parus major) and some other related species. Behaviour Sup- plement 2: 1-201 Hutchins, H. E.,and R. M. Lanner. 1982. The central role of Clark’s Nutcracker in the dispersal and establishment of whitebark pine. Oecologia 55: 192-201. Johnson, H. C. 1900. In the breeding home of Clark’s Nut- cracker. Condor 2: 49-52. Lanner, R. M., and S. B. Vander Wall. 1980. Dispersal of limber pine seed by Clark’s Nutcracker. Journal of Fore- stry 78: 637-639. Lanner, R. M. 1981. The pifion pine: a naturaland cultural history. Univ. of Nevada Press, Reno. 208 pp. Lanner, R. M. 1982. Avian seed dispersal as a factor in the ecology and evolution of limber and whitebark pines. 6th North American Forest Biology Workshop, Univ. of - Alberta, Edmonton, Alberta, August 11, 1980. pp. 16-48. Ligon, J. D., and D. J. Martin. 1974. Pifion seed assess- ment by the Pifton Jay, Gymnorhinus cyanocephalus. Animal Behaviour 22: 421-429. Mewaldt, R. R. 1956. Nesting behavior of the Clark Nut- cracker. Condor 58: 3-23. Roth, V. D. 1971. Unusual predatory activities of Mexican Jays and Brown-headed Cowbirds under conditions of deep snow in southeastern Arizona. Condor 73: 113. Skutch, A. F. 1976. Parent birds and their young. Univ. of Texas Press, Austin, Texas. 503 pp. Tomback, D. F. 1978. Foraging strategies of Clark’s Nut- cracker. Living Bird 16: 123-161. Vander Wall, S. B. 1982. Anexperimental analysis of cache recovery in Clark’s Nutcracker. Animal Behaviour 30: 84-94. Vander Wall, S. B., and R. P. Balda. 1977. Coadaptations of the Clark’s Nutcracker and the pinon pine for efficient seed harvest and dispersal. Ecological Monographs 47: 89-111. Vander Wall, S. B., and R. P. Balda. 1981. EcoCogy and evolution of food-storage behavior in conifer-seed- caching corvids. Zeitschrift ftir Tierpsychologie 56: 217-242. Vander Wall, S. B.,S. W. Hoffman, and W. K. Potts. 1981. Emigration of Clark’s Nutcracker. Condor 83: 162-170. Volker, V.,and W.Rudat. 1978. On the postfledging behav- ior of nutcrackers (Nucifraga caryocatactes L.). Zoolo- gische Jahrbucker fiir Systematik, Gkologie und geogra- phie dertiere. 105: 386-398. Wiley, R. H., and S. A. Hartnett. 1979. Effects of hunger. on aggression, approach, and avoidance in juncos (Junco hyemalis). Zeitschrift ftir Tierpsychologie 51: 77-83. Received 11 May 1982 Accepted 12 November 1982 ee Notes White Malaxis, Malaxis monophyllos var. diphyllos, an Addition to the Orchids of Canada from the Queen Charlotte Islands, British Columbia. WILLIAM J. BEESE MacMillan Bloedel Limited, 65 Front Street, Nanaimo, British Columbia V9R 5H9 Beese, William J. 1983. White Malaxis, Malaxis monophyllos, var. diphyllos, an addition to the orchids of Canada from the Queen Charlotte Islands, British Columbia. Canadian Field-Naturalist 97(2): 215-216. Two collections in 1981 and 1982 extend the range of Malaxis monophyllos var. diphyllos, approximately 2000 km southeast of those described from the Aleutian Islands, Alaska. Key Words: White Malaxis, Malaxis monophyllos var. diphyllos, Queen Charlotte Islands, range extension. During vegetation sampling for a forest habitat type classification in the Queen Charlotte Islands, | discovered a variety of White Malaxis or Adder’s- mouth, Malaxis monophyllos (L.) Schwartz var. diphyllos (Chamisso) Luer, previously described only from the Aleutian Islands in Alaska. This new record represents a disjunct population approximately 2000 km southeast of its known range (Figure 1). The first collection was made on September 11, 1980 ina topogenous bog at 390 m elevation on south- eastern Graham Island, 10 km north of Queen Char- lotte City, B.C. (53° 19’N, 132°07’W). The specimens had passed the flowering stage, so a positive identifi- cation was not possible. A second collection at the same locality on June 16, 1981 procured several flow- ering specimens. My identification of them as M. monophyllos was confirmed by Dr. Roy L. Taylor, Director of the Botanical Garden at the University of British Columbia, Vancouver, who determined them to be the variety diphyllos from Luer (1975). Two specimens have been deposited in the University of British Columbia herbarium. The species is new to the flora of the Queen Charlottes, and the variety is new to Canada. Collected specimens are 15-20 cm tall, arising from an ovoid corm I-1.5 cm in diameter, with two leaves of nearly equal size sheathing the stem. The flowering spike is a raceme of up to 60 small yellowish-green flowers with the lip uppermost, as in the typical variety. The presence of two robust leaves distinguishes this plant from the typical variety, which has a single leaf or occasionally two leaves, with the second being very inconspicuous. Hultén (1968) did not recognize the variety diphyllos in the Flora of Alaska, nor did Scog- gan (1978) in the Flora of Canada. A third variety, var. brachypoda, was collected previously in several locations in coastal B.C. (Szczawinski 1959). It differs ALEUTIAN : (ISLANDS QUEEN CHARLOTTE ISLANDS 0° 500 1000 KM Figure |. Previous known distribution of Malaxis mono- phyllos var. diphyllos is shaded with the new locality indicated with a solid circle on the map inset. PMV) 216 from the other varieties in that the flowers are inverted, so that the lip is lowermost. This variety 1s not recorded on the Queen Charlotte Islands; how- ever, a second species, Malaxis paludosa (L.) Sw., is found there (Calder and Taylor 1968). It is generally a much smaller plant with fleshy leaves and is not easily confused with M. monophyllos. Both species of Malaxis are rare in British Columbia or at least poorly collected because they are an inconspicuous element of the plant associations in which they occur. My collections were made on a site representing a transition between bog and closed coniferous forest. The open, scrubby tree canopy included Tsuga hete- rophylla, Tsuga mertensiana, Thuja plicata and Cha- maecyparis nootkatensis. Vaccinium ovalifolium, V. alaskaense and Menziesia ferruginea were the most common shrubs present. Malaxis occurred in a lush herbaceous patch in a small colony of scattered indi- viduals. Associated vascular plants included Vera- trum viride, Dodecatheon jeffreyi, Athyrium filix- femina, Monitia sibirica, Lysichiton americanum and Carex anthoxanthea. A dense carpet of bryophytes including Rhizomnium glabrescens, Leucolepis men- ziesii and Conocephalum conicum occupied most of the ground surface where Malaxis was present. Sphagnum spp. were notably not abundant. Plants THE CANADIAN FIELD-NATURALIST Vol. 97 were rooted in a surface soil horizon of well decom- posed, black, mucky humus overlying deep peat layers of moderately decomposed Sphagnum and Carex. ‘Observations of M. monophyllos in other localities in the Queen Charlottes have not as yet been made. I wish to thank Dr. R. L. Taylor and Dr. W. B. Schofield of the University of British Columbia for their assistance and review of the manuscript. Literature Cited Calder, J. A., and R. L. Taylor. 1968. Flora of the Queen Charlotte Islands. Part 1. Systematics of vascular plants. Research Branch, Agriculture Canada, Monograph Number 4, Part 1. 659 pp. Hultén, E. 1968. Flora of Alaska and neighboring territo- ries: a manual of vascular plants. Stanford University Press, Stanford, California. 1008 pp. Luer, C. A. 1975. The Native Orchids of the United States and Canada excluding Florida. The New York Botanical Garden, New York. 361 pp. Scoggan, H.J. 1978. The Flora of Canada. Part 2. National Museum of Canada, Publication in Botany 7(2). Szezawinski, A. F. 1959. The Orchids of British Columbia. British Columbia Provincial Museum Handbook No. 16, Third Printing 1975, Victoria, B.C. 124 pp. Received 18 June 1982 Accepted 15 August 1982 Extensive Overland Movement of Pintail, Anas acuta, Brood and Attempted Predation By Hawks DAVID C. DUNCAN Department of Zoology, University of Alberta, Edmonton, Alberta T6G 2E9 Duncan, David C. 1983. Extensive overland movement of Pintail, Anas acuta, brood and attempted predation by hawks. Canadian Field-Naturalist 97(2): 216-217. A Northern Pintail (Anas acuta) was observed moving her one-day-old brood approximately 5.2 km overland. The hen and brood survived predation attempts by a Northern Harrier (Circus cyaneus) and a Swainson’s Hawk (Buteo swainsoni). Key Words: Northern Pintail Anas acuta, brood, overland movement, predation, hawk, Alberta. Duck broods may undertake frequent and substan- tial overland movements (Evans et al. 1952; Ball 1973), with the hen apparently initiating such move- ments (Dzubin and Gollop 1972). Among North American ducks, Northern Pintails (Anas acuta) are reported to have one of the highest brood mobilities (Evans et al. 1952; Diem and Lu 1960). Sowls (1955) noted a Pintail hen moving her brood over 700 m the first day after hatch and Sayler (1962) recorded a Pintail brood travelling almost 5 km prior to fledging. Overland movements are thought to result in consid- erable duckling mortality (Sayler 1962; Dzubin and Gollop 1972; Ball 1973), often attributed to predation. The following observation involved the movement of a Pintail hen and her one-day-old brood over flat, grazed, prairie near Brooks, Alberta. They left Kinin- vie South Lake, the same waterbody from which Giroux (1980) observed broods of Canada Geese (Branta canadensis) moving overland. Early on 6 June 1981, a radio-marked Pintail hen took her newly-hatched brood of seven ducklings from their nest to Kininvie South Lake, a distance of 1983 350 m. At 11:40 the next day, the hen and brood were located and sighted about 500 m west of Kininvie South Lake. After this initial observation, | tracked the hen from a vehicle at a distance of 0.5-I km. About | h after the initial sighting | approached to within 50 m of the hen, causing her to fly from her brood and perform distraction displays. As | with- drew to about 500 m, the hen rejoined her ducklings and continued walking. At this time they were |.1 km northwest of Kininvie South Lake. Two and a half hours later, the hen and her brood crossed a road, enabling me to count seven ducklings. The hen and her brood had been travelling north- northwest, parallel to and about 500 m west of a trail and ditch which lead towards another waterbody, Kininvie Flat. Ten min after this, the hen was sighted heading directly for Kininvie Flat, having altered her course to northeast. As the hen approached Kininvie Flat, | moved to within about 300 m. At 16:10 a female Northern Harrier (Circus cya- neus) dove at the hen and brood. The hen stood in an alert posture and quacked as the hawk swooped to within0.5 m. The hawk landed about 10 maway from the hen. Both stood their positions for approximately 5 min after which the hen and her brood continued walking towards Kininvie Flat and the hawk flew off. About 15 min later a Swainson’s Hawk (Buteo swainsoni) dove at the hen and landed 10 m away. After a few minutes the hen again proceeded towards Kininvie Flat and the hawk flew away. Neither hawk made any further attack. At 17:00 the hen and brood reached the edge of Kininvie Flat and | observed that all seven ducklings were still present. The total distance walked was about 5.2 km while the shortest distance between the lakes is 4.3 km. Estimated time for the movement was 6 h, an average speed of 0.9 km/h. The hen generally traversed open prairie although for brief periods she walked in the taller cover of a dried lake bottom and along a buried pipeline. The hen and brood were very difficult to observe, particularly when the hen crouched low after being disturbed. On a number of occasions the hen was observed tilting her head to the side as if to look up. Overall, the movement appeared directed and the hen well-oriented. Kininvie South Lake, the waterbody vacated by the hen, is shallow and prone to dry up incertain years. In 1981, however, ample water persisted due to mid- summer water input, and many duck broods, includ- ing Pintails, fledged on the lake. Kininvie Flat is larger with a more stable water level and less emergent vege- tation. Hens generally move their broods to intended rearing areas when the ducklings are young (Ball 1973, Talent et al. 1982). The movement may have been in response to human disturbance as the hen was separ- NOTES ZAG], ated from her brood for some time the previous day when | attempted to count her ducklings. A number of Pintail nests in the Brooks area are located farther than | km from water (unpublished data). Thus, substantial overland movements by newly-hatched ducklings must be common. Although such overland travel entails an increased risk of preda- tion by terrestrial mammals, its survival value must generally be greater than the risk (Eriksson 1978). Overland movements may be less hazardous in sparsely-vegetated areas such as grazed, mixed prairie because of decreased probability of separation of ducklings from the hen. Four days after the movement described here, the hen was found alone on Kininvie Flat, her brood having apparently died in the interim. Acknowledgments I extend my thanks to D. A. Boag for his construc- tive comments and to the reviewers for their interested input. The observation was made while conducting research funded by a grant (A2010) from the Natural Sciences and Engineering Research Council of Can- ada to D. A. Boag. Logistic support was provided by the University of Alberta. Literature Cited Ball, I. J., Jr. 1973. Ecology of duck broods in a forested region of north-central Minnesota. Ph. D. thesis, Univer- sity of Minnesota, St. Paul. 67 pp. Diem, K.L., and K.H. Lu. 1960. Factors influencing waterfowl censuses in the parklands, Alberta, Canada. Journal of Wildlife Management 24: 113-133. Dzubin, A., and J. B. Gollop. 1972. Aspects of Mallard breeding ecology in Canadian parkland and grassland. Pp. 113-152 in Population ecology of migratory birds. United States Fish and Wildlife Service, Wildlife Research Report 2. Eriksson, M. O. 1978. Lake selection by Goldeneye duck- lings in relation to the abundance of food. Wildfowl 29: 81-85. Evans, C.D., A. S. Hawkins, and W.H. Marshall. 1952. Movements of waterfowl broods in Manitoba. Uni- ted States Fish and Wildlife Service, Special Scientific Report, Wildlife No. 16. 47 pp. Giroux, J.-F. 1980. Overland travel by Canada Goose broods. Canadian Field-Naturalist 94: 461-462. Sayler, J. W. 1962. Effects of drought and land use on prairie nesting ducks. Transactions of the 27th North American Wildlife and Natural Resources Conference: 69-79. Sowls, L. K. 1955. Prairie ducks; A study of their behav- iour, ecology and management. Wildlife Management Institute, Washington. 193 pp. Talent, L. G.,G. L. Krapu, and R. L. Jarvis. 1982. Habitat use by Mallard broods in south central North Dakota. Journal of Wildlife Management 46: 629-635. Received 27 April 1982 Accepted 11 December 1982 218 THE CANADIAN FIELD-NATURALIST Vol. 97 Predation by Fish and Common Mergansers on Darters (Pisces: Percidae) in the Thames River Watershed of Southwestern Ontario JOHN ENGLERT! and BENONI H. SEGHERS?? Department of Zoology, University of Western Ontario, London, Ontario N6A 5B7 \Present address: MacLaren Plansearch Corporation, Suite 401-750 West Pender St., Vancouver, B.C. V6C 2T7 2Present address: Kananaskis Centre for Environmental Research, University of Calgary, Seebe, Alberta, TOL 1X0 3Author to whom correspondence should be addressed Englert, John, and Benoni H. Seghers. 1983. Predation by fish and Common Mergansers on darters (Pisces: Percidae) in the Thames River watershed of southwestern Ontario. Canadian Field-Naturalist 97(2): 218-219. A survey of potential predators of darters in the |hames River watershed near London, Ontario revealed that darters were rarely consumed by other fish species; however, they did appear to be an important food resource for Common Mergansers (Mergus merganser) during their spring migration. Key Words: darter, Etheostoma blennioides, Etheostoma caeruleum, Etheostoma flabellare, Etheostoma nigrum, Percina maculata, Micropterus dolomieui, Ambloplites rupestris, Esox lucius, Semotilus atromaculatus, Mergus merganser, preda- tion, food habits, Thames River, Ontario. Five species of darters are common in the Thames River watershed of southwestern Ontario: the Black- side Darter (Percina maculata), Johnny Darter (Etheostoma nigrum), Rainbow Darter (E. caeru- leum), Greenside Darter (E. blennioides), and Fantail Darter (E. flabellare) (Englert and Seghers 1983). There is a paucity of information available on pre- dation on stream darters. Reports of predation on specific darter species in streams include Creek Chub (Semotilus atromaculatus) preying on Johnny Dar- ters (Moshenko and Gee 1973), Northern Pike (Esox lucius) on Blackside Darters (Hunt and Carbine 1950), Mottled Sculpins (Cottus bairdi) (Koster 1936) and Smallmouth Bass (Micropterus dolomieui) (Lachner 1950) on Fantail Darters, Spotted Bass (M. punctulatus) on Johnny Darters (Smith and Page 1969), Pickerel (Esox americanus) on Spottail Darters (E. squamiceps) (Page 1974), Cottus carolinae on Snail Darters (Percina tansi) (Starnes 1977), and juvenile Burbot (Lota Jota) on lowa (E. exile) and Johnny Darters (Hanson and Qadri 1980). Darter species are often lumped together as prey items, e.g. stream dwelling Smallmouth Bass were found to be feeding on unidentified darter species by Surber (1941), Tate (1949), and Swor and Bulow (1975) and Metzelaar (1928, 1929) reported predation on darters by Brook Trout (Salvelinus fontinalis), Brown Trout (Salmo trutta), and Rainbow Trout (S. gairdneri). A potential avian predator on stream darters 1s the Common Merganser ( Mergus merganser) which stops on the Thames River system during its spring migra- tion northwards. This species is known to search under stones for food (Lindroth and Bergstrom 1959) and was reported by Salyer and Lagler (1940) to feed on darters in Michigan streams. Study Area and Methods Potential fish predators from Medway Creek and the Thames River (middle and north branches) near London, Ontario were sampled in daylight hours dur- ing the summers of 1977 and 1978 and their stomach contents examined. Fish were captured mainly by angling but bag seining, electroshocking, and gill net- ting were also employed. A total of 17 Common Mergansers were shot on the Thames River (north branch) during April in 1977 and 1978 and examined for evidence of predation on darters. Results and Discussion Few darters were consumed by cther fish species and the only identifiable species was the Blackside Darter (Table |). Darters are apparently not an important summer food resource for Smallmouth Bass or Rock Bass (A mbloplites rupestris); their chief food was crayfish. Samples of Northern Pike and Creek Chub were too small to assess their importance as predators. Both species were often observed in the same habitat as Johnny and Blackside Darters (rela- tively deep and quiet water; see Englert and Seghers 1983). The low incidence of predation on darters by other fish species may reflect the rich repertoire of antipredator behavior of darters (Englert 1979). Var- ious escape tactics may render these species relatively immune to attack by larger fish. Also, we cannot rule out the possibility that darters are consumed more 1983 NOTES 219 TABLE 1. Evidence of predation on darters in Medway Creek and the Thames River near London, Ontario. Number Number Containing Predator Examined Fish Smallmouth Bass Til 13 Rock Bass 39) q Northern Pike 5 3 Creek Chub 17 0 Common Merganser 17 12 Number Number of Containing Darters and Darters Species l 1 unidentified 1 2 Blackside 1] 1 Blackside 0 eas 5 6 Greenside 5 Rainbow 3 Blackside Seen eeeeeeeeeeeeeeeeeeeeeeeeeSSSSSSSSS——————SEE— eee am—_0060 frequently in fall and winter when alternate inverte- brate prey become scarce. The analysis of the diet of the Common Mergansers reveals that these birds feed on darters during their stop in this area and that at least three species of darters, Greenside, Rainbow, and Blackside, are vulnerable. Interestingly, the first two species are pre- dominantly riffle species (Englert and Seghers 1983). Common Mergansers also fed on Yellow Perch ( Perca flavescens) and unidentified species of suckers and cyprinids. In our study area, darters undoubtedly are preyed upon by other bird species (heron, kingfisher), mam- mals, and perhaps reptiles. Our extensive field obser- vations would suggest that such occurrences are quite rare; however, this deserves further study. Acknowledgments We thank Eva Heczko and Ric Cole for help in collecting fish, and Dr. Dave Ankney and Geoff Bain for collecting waterfowl. This study was funded by N.S.E.R.C. of Canada grant A9894 to B. H. Seghers. Literature Cited Englert, J. 1979. A comparative study of the antipredator behaviour of stream darters (Percidae) in southwestern Ontario. M. Sc. thesis, University of Western Ontario, London, Ontario. 119 pp. Englert, J., and B. H. Seghers. 1983. Habitat segregation by stream darters (Pisces: Percidae) in the Thames River watershed of southwestern Ontario. Canadian Field- Naturalist 97(2): 177-180. Hanson, J. M., and S. U. Qadri. 1980. Morphology and diet of young-of-the-year Burbot, Lota /ota, inthe Ottawa River. Canadian Field-Naturalist 94(3): 311-314. Hunt, B. P.,and W. F. Carbine. 1950. Food of young pike, Esox lucius L. and associated fishes in Peterson’s Ditches, Houghton Lake, Michigan. Transactions of the American Fisheries Society 80: 67-83. Koster, W. J. 1936. The food of sculpins (Cottidae) in cen- tral New York. Transactions of the American Fisheries Society 66: 374-382. Lachner, E. A. 1950. Food, growthand habits of fingerling northern smallmouth bass, Micropterus dolomieu dolo- mieu Lacépéde, in trout waters of western New York. Journal of Wildlife Management 14: 50-56. Lindroth, A., and E. Bergstrom. 1959. Notes on the feeding technique of the goosander in streams. Report of the Insti- tute of Freshwater Research, Drottningholm 40: 165-175. Metzelaar, J. 1928. The food of the rainbow trout in Mich- igan. Transactions of the American Fisheries Society 58: 178-182. Metzelaar, J. 1929. The food of trout in Michigan. Tran- sactions of the American Fisheries Society 59: 146-152. Moshenko, R. W., and J. H. Gee. 1973. Diet, time and piace of spawning, and environments occupied by creek chub (Semotilus atromaculatus) in the Mink River, Manitoba. Journal of the Fisheries Research Board of Canada 30: 357-362. Page, L. M. 1974. The life history of the Spottail Darter, Etheostoma squamiceps, in Big Creek, Illinois, and Fergu- son Creek, Kentucky. Illinois Natural History Survey, Biological Notes No. 89. 20 pp. Salyer, J. C., and K. F. Lagler. 1940. The food and habits of American mergansers during winter in Michigan, con- sidered in relation to fish management. Journal of Wildlife Management 4: 186-219. Smith, P. W., and L. M. Page. 1969. The food of Spotted Bass in streams of the Wabash River drainage. Transac- tions of the American Fisheries Society 98: 647-651. Starnes, W. C. 1977. The ecology and life history of the endangered Snail Darter Percina (Imostoma) tansi. | en- nessee Wildlife Resources Agency Technical Report No. 77-52. 44 pp. Surber, E. W. 1941. A quantitative study of the food of the Smallmouth Black Bass, Micropterus dolomieu, in three eastern streams. Transactions of the American Fisheries Society 70: 311-334. Swor, C.I., and F. J. Bulow. 1975. Changes in the tood habits of various game fishes after stocking rainbow trout in the Cordell Hull section of the Cumberland River. Journal of the Tennessee Academy of Science 50: 12-15. Tate, W. H. 1949. Growthand food habit studies of Small- mouth Black Bass in some lowa streams. lowa State Col- lege Journal of Science 23: 343-354. Received 27 January 1982 Accepted 20 October 1982 220 THE CANADIAN FIELD-NATURALIST Vol. 97 Water Meal, Wolffia arrhiza (Lemnaceae) in Saskatchewan J. LOOMAN Research Station, Research Branch, Agriculture Canada, Swift Current, Saskatchewan. S9H 3X2 Looman, J. 1983. Water Meal, Wolffia arrhiza (Lemnaceae) in Saskatchewan. Canadian Field-Naturalist 97(2): 220-222. Wolffia arrhiza, an addition to the flora of Canada, was found in Saskatchewan in 1981 in a slough north of Pierceland. Further surveys in 1981 and 1982 revealed it in five additional localities. This is the first report of the species in North America. Key Words: Wolffia arrhiza, Lemnaceae, new records, distribution, ecology. Recently, Cody reported the occurrence of Wolffia columbiana Karsten in Manitoba (Cody 1980), where it was found in several ponds in Riding Mountain National Park. Cody’s report was the first record of the genus Wo/ffia in the Prairie Provinces. Previously Wolffia was known only from Southern Ontario and Quebec (Scoggan 1978). In August 1981, while collecting Lemna minor L. and a sample of water for analysis from a slough north of Pierceland, Saskatchewan, I found that a large amount of plant material had passed through the strainer used to separate Lemna from the water sample. Examina- tion of this plant material showed that it consisted of small, rootless fronds and might'be a species of Wolffia (Looman 22928, 6 August 1981, SCS). Microscopic examination of the fronds revealed they were round to ovate or globose in outline, 0.5 to 1.0 mm in diameter; the upper surface was slightly convex, the lower surface rather strongly convex, and without distinguishing features (Figure 1). The slightly flattened surface showed epidermal cells of 10 to 16 um, with very thick walls, averaging 4 um. The numerous chloroplasts were very small. Staining with iodine showed numerous stomata in young cells, but in old cells only a few stomata became visible, and sometimes none at all. The Canadian species of Wolffia were described by Dore (1957), the genus Wolffia by Daubs (1965) who includes the Eurasian W. arrhiza(L.) Wimm. Descrip- tions of this species were also available to me in Heu- kels and Wachter (1962), Tutin et al. (1980), and Ben- tham and Hooker (1947). Illustrations were provided in Heimans et al. (1965) and Fitch et al. (1949). Employing the various characters used in the above descriptions as well as those observed on my speci- mens, the following key could be constructed. 1. Thallus distinctly flattened above, with a single terminal papilla; punctate; epidermal cells 20-30 um W. punctata Thallus not distinctly flattened, rounded to globose O7 Ovatee eee 2 2. Upper surface with | to 3 papillae along the median line; epidermal cells 40 to 60 um, thin-walled W. columbiana Upper surface without papillae; epidermal cells 10 to 16 wm, thick-walled W. arrhiza eee ee ee eee ee ee ee eo This indicates that my specimens belong to W. arrhiza (L.) Wimm. (not W. arhiza Wimm. sensu Scoggan 1978). The thick cell walls are not mentioned in any of the descriptions, but Daubs (1965) mentions a distinct border around the exposed surface of W. arrhiza. Under high magnification, 100-X or more, the thick walls do indeed give the impression of a “border”. Comparison of my specimens with those of W. columbiana (Cody 24555) shows that the latter are, on an average, much larger, and have large, thin-walled epidermal cells. Cody (1982 personal communica- tion) notes: “There is no question but what your speci- mens are quite different from the Wolffia I collected at Riding Mountain National Park. Yours are much smaller than mine and tend to have flattened sides which makes them look almost cylindrical, and in TABLE |. Chemical analysis of water in which specimens of Wolffia arrhiza were collected. Coll. No. Ec N Na+ kK Ca + Mg Cl Abundance 22937 0.96 5.30 59.0 128.0 24.5 23036 0.63 1.12 34.3 81.6 41.5 23038 0.42 1.22 34.0 59.3 16.5 + 23037 0.38 4.72 46.5 41.5 30.2 chatete 22928 0.35 4.35 59.0 19.6 29.5 +++ 22936 0.33 2.80 29.0 44.7 34.5 + 1983 Ficure |. Wolffia arrhiza. 12x magnification addition, are covered with distinct fine markings.” The fine markings are caused by the thick cell walls, and are clearly visible under low magnification. The “flattened sides” are also caused by the thick cell walls, and under high magnification show up as the “border” used by Daubs (1965) as a distinguishing feature of W. arrhiza (Figure 2). Wolffia arrhiza has not previously been reported from North America. It is widespread in Europe, but lacking in the Scandinavian countries and the north- ern USSR, and is also known from Africa, Asia and Australia. Its occurrence at about 54°30’N is, there- fore, close to its northern limits in Europe. A survey in August 1981 of 25 bodies of water along a line from Fort Qu’Appelle in the south to Prince Albert inthe north revealed the presence of W. arrhiza in two other locations; a small lake 8 km northeast of Perigord (Looman 22937, 31 August 1981, SCS), and in a slough 10 km south of Prince Albert (Looman 22936, 31 August 1981, SCS). In both locations the surface of the water was covered witha mat of Lemna NOTES 22 FIGURE 2. Wolffia arrhiza, 24x magnification; showing the border effect caused by the thick cell walls minor, but in neither location was W. arrhiza as plen- tiful as in the Pierceland slough. A further survey was made in September 1982 along a line from Prince Albert to Pierceland, and then south to St. Walburg. In more than 50 sloughs and lakes samples of the Lemna vegetation were taken and strained, but only in three locations was Wolffia arrhiza present. The locations are: 20 km north of Debden (Looman 23036, 11 September 1982, SCS); 11 km north of Pierceland (Looman 23037, 12 Sep- tember 1982; SCS), and 5 km north of Red Cross (Looman 23038, 12 September 1982, SCS). Except for the second location north of Pierceland, where the species was very abundant, W. arrhiza occurred only in small numbers. Cody (1980) posed the question whether W. colum- biana is indigenous or whether it was introduced to Riding Mountain. He noted that the species occurred in six beaver ponds, but was absent from other, some- times adjacent, ponds. The same question arises about the occurrence of W. arrhiza. It occurred abundantly in two bodies of water, sparingly in four more, but was absent from numerous other sloughs and lakes in the same areas. Phytosociological studies of aquatic plant com- munities in Europe place Wolffia arrhiza in duckweed pips communities, Lemnetea, which occurin™. . . polluted waters containing some chlorine. Pollution can be caused by rotting foliage and may be anthropogenic.” (Westhoff et al. 1946). Knapp (1948) considers the communities indicative of “. . . water rich in nutrients _..,; Heukels and Wachter (1952) mention ™. . . high ” TAPMANURO PEM een. ct- Chemical analyses of the water in which | collected W. arrhiza shows these to be moderately to highly rich in nutrients, relatively rich in chlorine, and moderate to high in N-NO, and/or N-NHg. As evident in Table 1, a high nitrogen content does not guarantee abundance of Wolffia. It is quite possi- ble that the water of 22937 (Prince Albert) is some- what too high in nutrients; it is on the borderline of brackish. If this is so, it might well explain the appar- ent absence of Wolffia spp. in the prairie waters. Most of these, when high in N, are brackish or even saline. Although habitat conditions may explain the dis- tribution of W. arrhiza, and possibly that of W. columbiana, as presently known, the question of whether Wolffia is an indigene is not answered. How- ever, as noted by den Hartog and Segal (1964), most of the truly aquatic plants die rapidly when dried out. As Wolffia is possibly the most “aquatic” of the aquatic plants, transport by water fowl over more than short distances seem unlikely. This implies that both W. arrhiza and W. columbi- ana are indigenous in their respective areas, and in turn, poses another question: what caused the isola- THE CANADIAN FIELD-NATURALIST Vol. 97 tion of the two areas from the main distribution areas of the species? Literature Cited Bentham, G., and J. D. Hooker. 1947. Handbook of the British Flora. L. Reeve and Co. Ltd., Ashford, Kent. 606 Pp. Cody, W. J. 1980. Wolffia columbiana (Lemnaceae), Water-meal, new to Manitoba. Canadian Field-Naturalist 94: 193-194. : Daubs, E. H. 1965. A monograph of Lemnaceae. Illinois Biological Monographs 34: 1-118. Dore, W. G. 1957. Wolffia in Canada. Canadian Field- Naturalist 71: 10-14. Fitch, W. H., et al. 1949. Illustrations of the British Flora. L. Reeve & Co. Ltd., Ashford, Kent. 338 pp. Hartog, C. den, and S. Segal. 1964. A new classification of the water-plant communities. Acta Botanica Neerlandica 13: 367-393. Amsterdam. Heimans, E., H. W. Heinsius, and Jac P. Thysse. 1965. Geil- lustreerde Flora van Nederland. Versluys N. V., Amster- dam. 1182 pp. Heukels, H., and W. H. Wachter. 1952. Beknopte school- flora voor Nederland. Noordhoff N.V., Groningen. 414 pp- Scoggan, H.J. 1978. The flora of Canada. National Museum of Natural Sciences Publications in Botany, Number 7(2): 456-457. Tutin, T. G., etal. 1981. Flora Europaea, Vol. 5, Alismata- ceae to Orchidaceae, Cambridge University Press, Cam- bridge. xxxvi+ 452 pp. Received 21 January 1982 Accepted 10 January 1983. ‘Facing In’ is not General to all Gulls Nesting on Cliffs EDWARD H. BuRTT, JR., and WILLIAM CHOW Department of Zoology, Ohio Wesleyan University, Delaware, Ohio 43015 Burtt, Edward H., Jr., and William Chow. 1983. ‘Facing in’ is not general to all gulls nesting on cliffs. Canadian Field- Naturalist 97(2): 222-224. Herring Gulls (Larus argentatus) nesting on cliffs lack the cliff-facing behaviour found among incubating and brooding kittiwakes. The result strongly suggests that cliff-facing is another of the kittiwake’s behavioural specializations to its unusual and potentially dangerous nesting habitat. Key Words: Black-legged Kittiwake, Rissa tridactyla, Herring Gull, Larus argentatus, nesting behaviour, cliff-facing. Parental Black-legged Kittiwakes (Rissa tridactyla) show a number of behavioural adaptations that reduce the chance of the egg or chick falling from its nest (Cullen 1957). Among such adaptations 1s a pro- nounced tendency to face the rock wall during incuba- tion and a significantly greater tendency to face the rock wall during brooding (Burtt 1975; Hodges 1975). By facing the cliff, a brooding kittiwake places its body and especially its legs between the chicks and the abyss. The position restricts the chick’s movement. Is cliff-facing a behavioural product of the kittiwake’s evolutionary adaptation to its precipitous nest site or is cliff-facing a general reaction to vertical topo- graphy, a reaction found among all species of gulls? Herring Gull (Larus argentatus) chicks reared on cliff ledges behave like kittiwake chicks (Emlen 1963). 1983 -Cross-fostering experiments suggest that the altered behaviour is a product of experience at the cliff nest, but whether the chick’s behaviour is a product of habitat similarities or similarities in parental behav- iour is unclear. Hence, we address the question of the evolutionary origin of cliff-facing by studying the behaviour of Herring Gulls at nest sites comparable to those of kittiwakes. Kittiwakes typically nest on tiny ledges on sheer cliffs, but may occupy ledges up to 45 cm wide (Coul- son 1963). Herring Gulls commonly nest on flat ground, but a few occupy slopes and a very few occupy cliff ledges 40-50 cm wide (Emlen 1963). If cliff-facing were a generalized reaction to vertical topography, then cliff-nesting Herring Gulls should face the rock wall when incubating and brooding. If cliff-facing is a specialized evolutionary adaptation of the kittiwake, then Herring Gulls, few of which nest on cliffs, would be unlikely to face the rock wall during incubation or brooding. We report the orientation of incubating and brooding Herring Gulls nesting on cliffs and compare their orientation to that of kittiwakes and that of Herring Gulls nesting on sloping and flat ground. Between 6 and 26 July 1979, we observed unmarked Herring Gulls at individually identified nests on Kent Island, New Brunswick, Canada. Our study area included I! nests on cliffs, ground whose angle of inclination was 75-90° above the horizontal; 10 nests on slopes, angle of inclination 35-55° and 20 nests on flat ground, inclined 15° or less. Because so few nest sites were on slopes and cliffs, we were unable to statistically evaluate orientation of adults at individ- ual nests. However, orientation at no nest differed strikingly from the pattern for that topographical category, nor was any category dominated by data from one nest. Hence, we have combined orienta- tional data from all nest sites within a particular topo- graphical category. To maximize the sequential independence of obser- vations we recorded an adult’s orientation only oncea day. The effect of sun angle (Lustick et al. 1978) and wind (Gochfield 1978; D. E. Miller, personal com- munication) on orientation and posture of sitting gulls was minimized by collecting data only within 0.5 hr of solar noon (sun nearly overhead) on days with wind speeds less than 2.5 ms |. A gull that sat with breast feathers contacting the nest’s contents was “incubating” if the nest contained only eggs, or “brooding” if the nest contained at least one chick. With its long axis within 45° of perpendicu- lar to the cliff or upslope, a gull was “facing in” when its head was toward the cliff or upslope, and “facing away” when its tail was toward the cliff or upslope. If the long axis was between “facing in” and “facing away’, the gull was oriented “parallel”. NOTES 223 All nests on cliffs and slopes were along the island’s eastern shore, which runs approximately north-south. Hence, a gull that was “facing in” was also facing west. Nests on flat ground were also near the eastern shore. A gull nesting on flat ground was considered “facing in” if its longitudinal axis was within 45° of an east- west direction and the head west of the tail, “facing away” if the head was east of the tail, and “parallel” if its longitudinal axis was within 45° of a north-south direction and facing either north or south. Regardless of topography at the nest site, gulls oriented “parallel” and facing north were not differentiated from those oriented “parallel” and facing south. If incubating and brooding Herring Gulls orient randomly, half of the individuals will be “parallel”, one-quarter “facing in”, and one-quarter “facing away’. On flat ground both incubating and brooding adults oriented randomly (Table |) with no change in the orientation of adults before and after hatching (Chi-square test for independence, Table 1, x2 = 3.01, df = 2,0.25 > p > 0.1). On slopes and cliffs, incubat- ing gulls tended to “face away”, but brooding adults oriented randomly in both habitats (Table 1). The difference in orientation before and after hatching was not significant on slopes (Table 1, x2 = 0.49, df = 2, 0.9 >p>0.75) nor on cliffs (Table 1, x2= 0.99, df = 2, 0.75 > p> 0.5). Topography at the nest site had no significant effect (x? = 1.58, df= 2, 0.5 > p > 0.25) on the proportion of incubating Herring Gulls “facing in”, nor was topo- graphy a factor in the orientation pattern of brooding adults (vy? = 1.94, df = 2, 0.5 > p > 0.25). Unlike incubating and, especially, brooding Black- legged Kittiwakes, orientation of incubating and brooding Herring Gulls is essentially random, despite a slight tendency for incubating slope- and cliff- nesting gulls to “face away”. The smaller body size and more agile flight of kittiwakes enables them to nest on extremely narrow ledges. Differences in the width of nesting ledges may partially explain the orientional differences, Kittiwakes frequently cannot “face away’, because the tail cannot be accommodated by the narrow nesting ledge. Herring Gulls nesting on wider ledges are not subject to this spatial, morpho- logical constraint. However, room for the tail cannot explain the preference of adult kittiwakes for “facing in” over “parallel”, or the significant change in adult orientation at hatching. The data refute the hypothesis that cliff-facing is a generalized reaction of larids to vertical topography. Regardless of topography, Herring Gulls display essentially random orientation. Furthermore, the sig- nificant change in orientation of adult kittiwakes fol- lowing hatching of their eggs cannot be simply a reac- tion to topography. Cliff-facing appears to be another 224 THE CANADIAN FIELD-NATURALIST Vol. 97 TABLE |. Orientation of incubating and brooding Herring Gulls at topographically different nest sites. Topography of Nest Site Adult’s (number of nests) Activity “Facing in” Flat Incubating 37 (20) Brooding 20 Slope Incubating 38 (10) Brooding 18 Cliff Incubating 30 (11) Brooding 15 Orientation of Adult* 3 “Parallel” “Facing Away” (df = 2) 87 32 2.40 28 17 1.56 53 55 [S61 27 23 3.62 69 51 6.61*** 40 18 0.99 *If random, 25% of the gulls will be “facing in”, 50% “parallel”, and 25% “facing away”. **Significantly nonrandom, p < 0.01 ***Significantly nonrandom, 0.05 > p > 0.01 of the many behavioural and morphological speciali- zations (Cullen 1957; McLannahan 1973) of the kitti- wake to its unusual and potentially dangerous nest site. The data say nothing about the proximate con- trol of cliff-facing, but argue strongly for evolution by kittiwakes of a capacity to respond to the cliff wall and failure of such a capacity to evolve among species of gull whose nesting habitat, like the Herring Gull’s, is usually flat. A few species of larids, Glaucous Gull (Larus hyperboreus), Thayer’s Gull (L. thayeri), and Iceland Gull (L. glaucoides), frequently nest on cliffs where they show behaviour similar to that of kitti- wakes, but also nest on flat ground where their behav- iour resembles that of Herring Gulls (Smith 1966). These species may provide insight into the origin and function of cliff-facing. Acknowledgments Our study was supported by a grant from Sigma Phi Epsilon. Logistical support was provided by Bowdoin College. Charles E. Huntington offered his always cheerful advice. The manuscript benefited greatly from the thoughtful comments of A. John Gatz, Jr., Jack P. Hailman, Don E. Miller, and W. A. Monte- vecchi. Our study is contribution 48 from the Bow- doin Scientific Station. Literature Cited Burtt, E. H.,Jr. 1975. Cliff-facing interaction between par- ent and chick Kittiwakes Rissa tridactyla in Newfound- land. Ibis 117: 241-242. Coulson, J. C. 1963. Thestatus of the Kittiwake in the Brit- ish Isles. Bird Study 10: 147-179. Cullen, E. 1957. Adaptations in the-Kittiwake to cliff- nesting. Ibis 99: 275-302. Emlen, J. T., Jr. 1963. Determinants of cliff edge and escape responses in Herring Gull chicks in nature. Behav- iour 22: I-15. Gochfield, M. 1978. Incubation behavior in Common Terns; influence of wind speed and direction on orienta- tion of incubating adults. Animal Behavior 26: 848-851. Hodges, A. G. 1975. The orientation of adult Kittiwakes Rissa tridactyla at the nest site in Northumberland. Ibis 117: 235-240. Lustick, S. I., B. Battersby, and M. Kelty. 1978. Behavioral thermoregulation; orientation toward the sun in Herring Gulls. Science 200: 81-83. McLannahan, H. M. C. 1973. Some aspects of the onto- geny of cliff-nesting behaviour in the Kittiwake (Rissa tridactyla) and the Herring Gull (Larus argentatus). Behaviour 44: 36-88. Smith, N. G. 1966. Adaptations to cliff-nesting in some Arctic gulls (Larus). Ibis 108: 68-83. Received 13 July 1982 Accepted 13 December 1982 1983 NOTES 225 A River Otter, Lutra c. canadensis, of Record Size JAMES D. LAZELL, JR. The Conservation Agency, 8 Swinburne St., Jamestown, Rhode Island 02835 Lazell, James D., Jr. 1983. A River Otter, Lutra c. canadensis, of record size. Canadian Field-Naturalist 97(2): 225-226. A male otter, Lutra c. canadensis (Mammalia: Mustelidae), from Lynnfield, Essex Co., Massachusetts, measures 1321, 470, 129, 21, and is thus at least a percent larger than various published maxima for this form. The baculum is 102, the maxillary row 39.0 and the carnassial 12 by 10 mm. The specimen is Yale Peabody Museum (YPM) 4405. Key Words: River Otter Lutra c. canadensis, size, dentition, baculum. On 17 April 1977 I collected a road-killed, male River Otter on Route 128 just west of exit 31, just within the western limit of both the town of Lynnfield and Essex County, Massachusetts. The specimen was fresh and skinned immediately. The skull was badly broken and therefore the measurement I give (Table 1) for total length is probably unduly precise; the total length is about 132 cm or approximately 52 inches (4 feet 4 inches). The animal was not weighed but was judged at least 10 kg. The prepared skin is shown in Figure |. | follow Hoffmann (1976) in use of the generic name Lutra. TABLE !. Maximum sizes for Lutra c. canadensis in the literature and measurements of YPM 4405 in millimeters. Source Total Tail Hind Foot Hamilton (1943) 1100 400 100 Bailey (1946) 1045 330 105 Jackson (1961) 1220 475 133 Peterson (1966) 1300 500 140 Doutt et al. (1977) 1200 440 132 YPM 4405 1321 470 129 In popular literature some very large sizes are given for otters in general. For example, Trimm (1977) claims individuals “measuring up to 55 inches in length and weighing up to 30 pounds.” Although heis, by implication, writing about North American River Otters, he cites no source and does not imply that he means the eastern form, Lutra c. canadensis. Maximum size data from five standard sources for L. c. canadensis are given in Table |. Hall (1981) does not give data for subspecies; his measurements for the entire species include the large races of Canada and Alaska. Hall gives the maxima 1300, 507, 146; these measurements exceed YPM 4405 in tail and hindfoot, but the total length is still 1.5 percent below this Massachusetts individual. Hall and Kelson (1959, p. 948) give the maxillary tooth row of the Alaskan L. c. mira, a form distinguished by its large skull, as 38.8 mm. This is close to the 39.0 mm maxillary row of YPM 4405 (Figure 2). Friley (1949) gives the average length of the bacu- lum for oldest adults as 96.1 mm; his maximum is 106.4: the baculum of YPM 4405 is 102.1 mm (Figure 3). Stephenson (1977) gives a maximum of 106.9 mm. The present specimen is thus in excess of six years of age. vg a ae Se ve ce ac \© Of G2 Bz ce 92 s@ we? ca 2a i2 0€ Gi oi ci wl Siw cl mt wom 6 S78 8 re Ns as We 7 ee 2g el ae 2a! aa es ae a? a8! ee) 30) (31 aa 3a 4a | 38 a 4 a & ° c 2 ‘ FiGure |. Skin of the male river otter, YPM 4405, prepared to the measurements of the carcass. The scale is a yardstick. 226 NOILVYUOdHOD DIAILNZIOS FiGure2. Left maxillary tooth row of the river otter, YPM 4405. The first tooth on the maxilla is the canine: C. There are four premolars: P1-P4. There is but one upper molar present in the family Mustelidae: M. Godin (1977, p. 234) says males grow larger than females and may attain 130 cm. He gives tail and hind foot maxima as 50 cm and 1I5 cm, respectively. These measurements seem suspiciously round and even; | do not believe they were taken from an actual specimen. Similarly, Waters and Rivard (1962) say no more about the size of otters than they attain “three feet or more,” and have a maximum tail length of “fifteen inches” (p. 105). The photographs of YPM 4405 were made by Con- stance A. Rinaldo. The specimen was salvaged on Massachusetts permit 475, held by me. THE CANADIAN FIELD-NATURALIST W/ Vol. 97 SCIENTIFIC CORPORATION 60 ARSENAL ST., WATERTOW! 3 4) 5) sa. FiGureE 3. Baculum or os penis of the river otter YPM 4405 (A-89662 is a field number). Literature Cited Bailey, J. W. 1946. The mammals of Virginia. Williams, Richmond. xvi+ 416 pp. Doutt, J.K., C.A. Heppenstall, and J.E. Guilday. 1977. Mammals of Pennsylvania, Fourth Edition. Penn- sylvania Game Commission, Harrisburg. 283 pp. Godin, A. J. 1977. Wild mammals of New England. The Johns Hopkins Press, Baltimore. xii + 304 pp. Friley, C. E. 1949. Age determination, by use of the bacu- lum, inthe river otter, Lutra c. canadensis Schreber. Jour- nal of Mammalogy 30: 102-110. Hall, E. R. 1981. The mammals of North America. Second Edition, revised. John Wiley and Sons, Toronto. xv + 1181 pp. Hall, E.R., and K.R. Kelson. 1959. The mammals of North America. Ronald Press, New York. viii + 1083 pp. Hamilton, W. J. 1943. The mammals of eastern United States. Comstock Ithaca: 432. Hoffmann, R.S. 1976. Anecological and zoogeographical analysis of animal migration across the Bering Land Bridge during the Quaternary Period. Beringia in Cenoz- oic, Vladivostok pp. 354-367. Jackson, H. H. T. 1961. Mammals of Wisconsin. Univer- sity of Wisconsin Press, Madison. xii + 504 pp. Peterson, R. L. 1966. The mammals of eastern Canada. Oxford, Toronto. xxx + 465 pp. Stephenson, A. B. 1977. Age determination and morpho- logical variation in Ontario otters. Canadian Journal of Zoology 55(10): 1577-1583. Trimm, H.W. 1977. Otter. The Conservationist (New York) 31(6): 23-25. Waters, J.H., and C.J. Rivard. 1962. Terrestrial and marine mammals of Massachusetts and other New Eng- land states. Standard-Modern, Brockton, Mass. vi+ 151 Pp. Received 7 June 1982 Accepted 6 January 1983 News and Comment Notice of The Ottawa Field-Naturalists’ Club Annual Business Meeting The 105th Annual Business Meeting of the Ottawa Field-Naturalists’ Club will be held in the auditorium of the Victoria Memorial Museum Building, Metcalfe and MacLeod Streets, Ottawa on Tuesday, 10 January 1984 at 2000 h. FRANK POPE Recording Secretary The Alfred B. Kelly Memorial Fund of the Province of Quebec Society for the Protection of Birds, Inc. Annual Research Grants up to $1,000 will be avail- able for studies pertaining directly to Quebec orni- thology. Applications will be accepted from any inter- ested person regardless of place of residence. Applications must be postmarked by 15 February, 1984. Applicants will be notified of the committee’s decision by 31 March 1984. For application forms write to Marianne G. Ainley P.Q.S.P.B. Research Committee 4828 Wilson Avenue Montreal, Quebec Canada H3X 3P2 Announcing a Canadian Forestry and Wildlife Management Symposium A symposium on the integration of forestry and wildlife management in Canada will be held May 7, 8, 9 and 10, 1984 at the University of British Columbia near Vancouver. Knowledgeable speakers will address such topics as the legal basis for integrated resource management, the implementation process in various jurisdictions, the needs of resource users, and means of solving problems inhibiting integration. The program will be aimed at forest resource administra- tors and managers, and participation of field foresters and biologists is emphasized. Those wishing to receive a more detailed description of the symposium now and pre-registration materials in January 1984, con- tact T.C. Dauphine, Canadian Wildlife Service, 1725 Woodward Drive, Ottawa, Ontario, KIA OE7 (613-998-4693). Notice of Motion to Amend the Constitution of The Ottawa Field-Naturalists’ Club It is proposed that Article 1, now reading: “NAME AND STATUS This Club shall be known as The Ottawa Field- Naturalists’ Club. It is a non-profit organization incorporated under the laws of the province of Onta- rio (1884). All assets and other accretions...” be amended to read as follows: “NAME AND STATUS This Club shall be known as the Ottawa Field Natural- ists. It isa non-profit organization incorporated under the laws of the Province of Ontario (1884). ’ All assets and other accretions...” This is notice of a motion, proposed by Roger Tay- lor and seconded by Joyce Reddoch, to be presented at the 105th Annual Business Meeting. Explanatory Note: The purpose of this motion is to remove the word “The”, the hyphen, the apostrophe and the word “Club” from the organization’s name. Writers frequently forget these items and an editor’s job is appreciably simplified if they are removed from the official name. Eo ORE Recording Secretary 227 228 The Ottawa Field-Naturalists’ Club Awards This is a call to members for nominations for the following OFNC awards: 1. Honorary Member — in recognition of contri- bution in marked degree to the successful work- ing of the Club, or of outstanding contribution to Canadian natural history, generally over a period of time. 2. Member of the Year Award — for the member judged by Awards Committee to have contrib- uted the most to the Club in the previous year. 3. OF NC Service Award — in recognition of the member who has contributed significantly to the smooth running of the Club in that year. THE CANADIAN FIELD-NATURALIST Vol. 97 4. Conservation Award — in recognition of a member’s recent, outstanding contribution to the cause of conservation. 5. Anne Hanes Natural History Award — in recognition of an outstanding contribution by an amateur to our knowledge, understanding and appreciation of the natural history of the Ottawa Valley. Nominations for any of these awards should be sent to W.K. Gummer, Chairman, Awards Committee, 2230 Lawn Avenue, Ottawa, K2B 7B2 (Phone 596-1148) by 15 December, 1983. Call for nominations for the Council of the Ottawa Field-Naturalists’ Club A nominating committee has been chosen by the Council to nominate persons for election to offices and membership of the Council for the year 1984, as required by the Constitution. We would like to remind Club members that they also may nominate candidates as officers and other members of Council. Such nominations require the signatures of the nominator and seconder, and a statement of willingness to serve in the specified posi- tion by the nominee. Nominations should be sent to the Nominating Committee, The Ottawa Field- Naturalists’ Club, Post Office Box 3264, Postal Sta- tion C, Ottawa, Ontario KI Y 4J5, to arrive no later than 15 December 1983. The Committee will also consider any suggestions for nominees which members wish to submit to it by 15 December 1983. It would be helpful if some rele- vant background on the proposed nominees were provided along with the suggested names. ROGER TAYLOR Chairman, Nominating Committee Herpetology Position Available at the Royal Ontario Museum A curatorial position in herpetology is available in the Department of Ichthyology and Herpetology, Royal Ontario Museum, Toronto, Ontario. Appointment will be at senior Assistant Curator or Associate Curator level. Ph.D. graduate in herpetol- ogy with field, research, collection, and communica- tions skills and experience. This is a tenure stream, permanent position. Starting date — | July 1984. Closing date for receipt of applications and all documents — 15 March 1984. For further details, write Dr. E. J. Crossman, Chairman, Search Committee, Department of Ich- thyology and Herpetology, Royal Ontario Museum, 100 Queen’s Park, Toronto, Ontario, Canada MSS 2C6. 1983 Editor’s Report for 1982 In 1982, The Canadian Field- Naturalist 95(4) was mailed 7 April, 96(1) on 21 July, 96(2) on 30 Sep- tember, 96(3) on 14 December. The concluding issue of volume 96 was edited and completed to page proof (articles and notes) and galley (news and comments, tribute and book reviews) by the end of the year but was not published until 18 April. The number of manuscripts submitted in 1982 is about equal to each of the past two years (Table 1). The percentage of manuscripts accepted is up margin- ally, perhaps partly attributable to an increase in the quality of submissions as a result of the efforts of the previous editor, Lorraine Smith (1972-1981). Accep- tance totals for 1982 submissions are incomplete as many manuscripts are still with authors for revision. Acceptable revisions of a few pre-1982 submissions during the past year have modified totals given in the Editor's Report for 1981, The Canadian Field- Naturalist 96(2): 220-223, Table 4. The number of manuscripts published in The Can- adian Field- Naturalist volume 96 is given by field of study in Table 2. The number of pages of research (46 articles and 31 notes) was 384 (334 pages of articles and 50 of notes). News and Comment (including the Editor’s Report, the Annual Report of the Ottawa Field-Naturalists’ Club, and tributes to W.K. W. Baldwin and C.M. Sternberg) accounted for 37 pages, Book Reviews (including new titles) 90 pages additional notices (Instructions to contributors) for 3, and the annual Index 16 pages. The number of reviews and new titles published were: zoology 39 and 173, botany 14 and 64, environment 23 and 110, miscel- laneous 14 and 44, young naturalists 0 and 13. Further comment and comparisons between 1981 and 1982 from Wilson Eedy follow this report. In 1982, associate editors Charley Bird (botany), Ed Bousfield (invertebrates), Tony Erskine (ornithol- ogy), Charles Jonkel (predator-prey relationships), Bill Pruitt (mammalogy) and Steve Smith (entomol- TABLE !. Summary of manuscripts submitted and accepted by The Canadian Field- Naturalist Number of manuscripts Year Percent Submitted Submitted Accepted Accepted 1975 167 123 74 1976 147 93 63 IS 137 88 64 1978 149 93 62 1979 148 93) 63 1980 IS7/ 89 65 1981 136 9] 67 1982 138 54 NEWS AND COMMENT 229 ogy) were joined by Don McAllister (ichthyology) and Stan Van Zyll de Jong (mammalogy). Tony continued his role as super-associate by directly contacting refer- ees and returning manuscripts requiring revisions to authors for all bird manuscripts, though final accep- tance remains with the editor, as it does for all mate- rial. Tony has agreed to continue in this manner for one more year, through 1983. George La Roi con- tinued as co-ordinator for The Biological Flora of Canada series. No manuscripts were accepted for this in 1982, though a number are assigned and should soon be completed. E. Wilson Eedy continued as Book Review editor and Harvey Beck again compiled the index. The collective time and effort spent by all of these associates is staggering, vital to the production of the journal, and rewarded only by their dedication to the maintenance of its particular contribution to the dissemination of knowledge of natural history in Canada. Special thanks are also due to Wanda Cook and, at the end of the year, Glen Keenleyside, for long hours spent reading manuscripts and proof-reading galleys and page proof, and to Thérese Lapierre for typing. Bill Cody added a 34th consecutive year to his legend as Business Manager and was, as usual, a tower of strength and conscience behind the editor. Ron Bed- ford, chairman, and the publications committee, and Dan Brunton, president, and the council of the Ottawa Field-Naturalists’ Club, have provided encouragement and forums to discuss an evolving publications policy. During the past year the report of the ad hoc commit- tee on publications was thoroughly examined and the resulting guidelines and policy statement is published in The Canadian Field-Naturalist in this issue. The National Museum of Natural Sciences, National Museums of Canada, have continued to TaBLe 2. Number of manuscripts published in The Cana- dian Field- Naturalist 96(1982) by major field of study. Number of Manuscripts Subject Total (Articles + Notes) Research Mammals 29 (JI 3 113) Birds 27 (7 se 10) Amphibians and Reptiles (Zar 2) Fish 4 (CD 2 Invertebrates 3 ( jl se 2) Plants 10 (Sica) Total research Vi (46 + 31) Tribute 2 ( 2 ar (0) Grand total 79 (48 + 31) 230 provide space and support for the journal and I am particularly indebted for these to Henry Ouellet, Chief of the Vertebrate Zoology Division, and Chuck Gruchy, Acting Director. M.O.M. Printers in Ottawa again printed The Canadian Field-Naturalist, and individual acknowledgment is due Emil Holst and Ed Finnigan for their efforts. All research articles and notes, and all special arti- cles, submitted to The Canadian Field-Naturalist are reviewed not only by one or more associate editor(s) but also usually one, almost invariably two in orni- thology, external referee(s). The latter, by giving freely, promptly, and most often anonymously, of their time, are essential to the peer evaluation system on which the general quality of this journal, like any scientific publication, depends. The following list, compiled with the aid of A. J. Erskine for ornithology reviewers, includes those to whom we are indebted for comments on one or more papers during the 1982 calendar year: R. M. Alison, C. D. Ankney, G. Argus, M. Arm- bruster, P. J. Austin-Smith, P. W. Ball, A. W. F. Banfield, J. C. Barlow, J. Bartonek, T. S. Baskett, J. Bassett, IT. D. Beacham, J. Bedard, F. Bellrose, J. Bendell, J. R. Bider, D. M. Bird, J. S. Bleakney, H. Blockpoel, D. A. Boag, C. E. Bock, E. Broughton, R. G. B. Brown, D. F. Brunton, D. G. Busby, C. A. Campbell, R. W. Campbell, P. Catling, J. Christie, C. S. Churcher, W. J. Cody, B. Collins, F. Cooke, M. Crete, E. J. Crossman, A. Cyr, C. Dersle, J.-L. Des- Granges, G. Divoky, E. H. Dunn, A. Dzubin, R. Y. Edwards, D. Faber, E. B. Fenton, M. A. D. Fergu- son, J. C. Findlay, C. D. Fowle, B. Freedman, R. W. THE CANADIAN FIELD-NATURALIST Vol. 97 Fyfe, G. Geen, W. E. Godfrey, J. B. Gollop, S. W. Gorham, P. K. Gregory, C. R. Harington, B. Har- rington, G. E. Hartman, T. Hauge, C. J. Henny, P. Hicklin, D. J. T. Hussell, R. Ireland, R. D. James, J. Ro Jehly PR: Ac Keddy, L. Keith) PB: KelsalleGeale Kirkland, W. Klenner, R. W. Knapton, A. C. Kohler, E. Kott, C. J. Krebs, E. Kuyt, D. Laubitz, L. deK. Lawrence, S. Leatherwood, M. R. Lein, A. R. Lock, R. C. Long, H. G. Lumsden, C. D. MacInnis, W. J. Maher, D. Mallock, J. Maunder, R. A. McArthur, M. K. McNicholl, R. McNeil, I. A. McLaren, P. L. McLaren, L. R. Mewaldt, A. L. A. Middleton, J. S. Millar, F. L. Miller, E. Mitchell, W. A. Montevecchi, M. T. Myres, J.S. Nelson, R. W. Nelson, R. W. Nero, L. Oliphant, H. Ouellet, R. S. Palmer, W. B. Parama, D. B. Peakall, P. A. Pearce, M. R. Peterson, P. M. Peterson, R. L. Peterson, G. Power, P. M. Powles, W. B. Preston, J. Ranch, J. Reddock, T. E. Reimchen, M. E. Riske, C. S. Robbins, R. J. Robert- son, R. K. Ross, S. Rowe, J. P. Ryder, J.-P. Savard, D. B. O. Saville, S. G. Sealy, V. H. Schaefer, F. W. Schueler, G. Scotter, D. E. Sergeant, N. R. Seymour, M.W. Shoesmith, H. C. Smith, J. N. M. Smith, P. R. Stepney, K. W. Stewart, L. G. Sugden, P. S. Taylor, R. Taylor, C. F. Thompson, J. W. Thomson, B. Threlfall, C. E. Tull, N. A. M. Verbeek, K. Vermeer, D. Vitt, P. J. Weatherhead, R. Weeden, R. D. Weir, M. W. Weller, D. A. Welsh, D. V. Weseloh, D. D. Williams. FRANCIS R. COOK Editor Summary of Book Review Activities: Volume 96, 1982 The ever increasing number of titles in natural sciences has caused steady increases in the numbers of new titles listed and books reviewed over the years. Within five years a 40% increase in titles listed and almost 50% increase in books received have been recorded. Interestingly the number of reviews received and published has fluctuated from 50 to 86 through- out the period. This illustrates the two most signifi- cant difficulties for the Book-review Editor: identify- ing an appropriate, willing reviewer for each of so many books, and having reviews completed within a reasonable time period. In 1982 we have added a section for young natural- ists to our new titles. This results from a strong com- mitment to environmental education and the knowl- edge that the future of any organization relies on the commitment of young, new members. The following table summarizes book review activi- ties for 1981-82: Volume 96 Volume 95 1982 1981 Book Reviews Published 90 7\ Books Received 107 92 Book Reviews Completed 71 86 New Titles Listed 404 361 WILSON EEDY Book-review Editor A Publication Policy for The Ottawa Field-Naturalists’ Club The Ottawa Field-Naturalists’ Club (OFNC) has published The Canadian Field-Naturalist (and its predecessors) for over a century. In recent years in- volvement with publications has increased dramati- cally with the introduction of Trail & Landscape in 1967, The Shrike in 1976, and various Special Publi- cations from time to time. Until now no formal policy underpinning these publications has existed. The question of the need for such a policy was broached some time ago and, following numerous discussions, the Council of the OFNC approved in November 1978, the setting up of an Ad Hoc Committee on OFNC Publications to be charged with“. . . a broad mandate to examine any aspect of existing OFNC periodicals that it considers to be relevant. . .” together with some specific questions to be addressed. The committee, comprising Dr. R. A. Foxall (chair- man) OrekeRe Cook Dr I Me Brodos Mir. Ho 1. Dickson (secretary), was formed in November 1979 and submitted its report to Council in October 1980. This report was reviewed by the Publications Com- mittee in December 1980, following which both the report and the Publication Committee’s response to it were reviewed at a joint meeting of Council and the Ad Hoc Committee in January 1981. As a result of this meeting, the Executive Committee was asked to prepare a draft of a Publications Policy based upon the Ad Hoc Committee report and its subsequent reviews. This draft policy, submitted to Council in May 1982, was given intensive clause-by-clause study within the Publications Committee which, in turn, submitted a recommended revision to Council in November 1982. On December 8, 1982 Council approved (with some amendments) this revision which forthwith became the OFNC Publications Pol- icy. The text of this policy is printed here in its entirety. R. E. BEDFORD © Chairman, Publications Committee The Ottawa Field-Naturalists’ Club Publications Policy: February 1983 A — INTRODUCTION Although The Ottawa Field-Naturalists’ Club has been involved with publications almost since its incep- tion, it has never enunciated a formal publications policy governing their production. With the gradual increase in the number and diversity of these publica- tions, it has become desirable to formulate such a policy. An ad hoc committee was set up in 1978 to examine the status of the publications and to advise and recommend on their future courses. It is from the report of this committee, presented in 1980, that this Publications Policy has evolved. The Ottawa Field-Naturalists’ Club currently pub- lishes three periodicals and, from time to time, books or monographs as Special Publications. The three periodicals differ in purpose and scope. The Canadian Field-Naturalist is essentially a scientific journal devoted to the publication of refereed papers con- cerned with natural history that is relevant to Canada. It is the official organ of the Club* and appears quar- terly. Trail & Landscape, with five issues annually, is a less formal periodical geared to providing articles per- taining chiefly to the natural history of the Ottawa area and to providing information aimed at the local membership. The Shrike, a specialized bimonthly newsletter, reports to the bird-watching community *W herever the term “the Club” appears in this document, it is to mean The Ottawa Field-Naturalists’ Club. on bird sightings in the Ottawa District. Special Pub- lications embrace material better suited to separate publication. This document outlines the policy governing publi- cation of each of the above-mentioned items. B — THE CANADIAN FIELD-NATURALIST The Canadian Field- Naturalist, the official Club publication, is published quarterly and is available through separate subscription and to members of the Club. It is managed by annual budget. (1) — Objectives: (a) to be the official publication of The Ottawa Field- Naturalists’ Club, as described in the Constitution. (b) to support the Club’s objective “. . . to promote the appreciation, preservation and conservation of Canada’s natural heritage; to encourage inves- tigation and the publishing of the results of research in all fields of natural history and to diffuse information on these fields as widely as possible; to support and cooperate with organiza- tions engaged in preserving, maintaining or re- storing environments of high quality for living things”. (c) to publish refereed scientific papers by amateur and professional naturalists or field-biologists reporting observations and results of investiga- Dy) 232 tions in any field of natural history provided that they are original, significant, and relevant to Canada. (d) to publish news, comments, review articles, book reviews, and other such material that is in accord with objectives (a)-(c). (2) — Achievement of the Objectives: Useful indicators for the achievement of the objec- tives for The Canadian Field- Naturalist include: (a) unsolicited letters to the Editor or the Club. (b) feedback from the Council. (c) the degree to which Canada’s prominent field- naturalists support The Canadian Field- Naturalist by submitting papers. (d) submission of a sufficient number of high quality papers. (e) government responses to applications for funding. (f) trends in the number and geographic distribution of subscriptions. (g) citations in other publications. (h) adherence to a regular publication schedule. (3) — The Ottawa Field-Naturalists’ Club as Publisher: The Canadian Field- Naturalist is the only publica- tion of its kind in North America. It is recognized as being an asset to The Ottawa Field-Naturalists’ Club through the national and international acclaim it brings to the Club. The Club, therefore, has a large responsibility for The Canadian Field- Naturalist and recognizes the requirement to ensure its continued publication and high standards. The broadly-based interests of The Ottawa Field-Naturalists’ Club are seen as an important factor in maintaining the topical diversity and widely-based appeal that exists for The Canadian Field- Naturalist in Canada and beyond. The Ottawa Field-Naturalists’ Club will continue to publish The Canadian Field- Naturalist. (4) — Editorial Board: The Editor and Associate Editors should be: (a) scientists who are experienced in field-based nat- ural history investigation. (b) appointed for one year terms, renewable without limit. (5) — Support for Publication of Manuscripts: Independent authors are those who submit papers to The Canadian Field- Naturalist without the benefit of institutional support or research grant funding. The Ottawa Field-Naturalists’ Club recognizes the extra effort required by such contributors in the prepara- +Wherever the term “Council” appears in this document, it is to mean the Councikof The Ottawa Field-Naturalists’ Club. THE CANADIAN FIELD-NATURALIST Vol. 97 tion of their papers and that such self-motivated efforts have contributed significantly to the develop- ment of natural sciences knowledge in Canada. To assist such independent efforts, as part of The Ottawa Field-Naturalists’ Club’s commitment to the devel- opment of naturalists as well as natural history knowledge, a measure of logistical support will be provided by The Ottawa Field-Naturalists’ Club. For authors with minimal financial resources, limited journal funds are available to help offset publication charges. Requests for such financial assistance may be made to the Editor if the manuscript is accepted for publication. (6) — Special Status for Ottawa District Natural History Studies: The Ottawa District will not be afforded special status in evaluating submissions to The Canadian Field- Naturalist. Such papers must meet the same standards by which all other submissions are judged. (7) — Content and Readability: The content and readability of The Canadian Field- Naturalist are satisfactory at the present time. To satisfy its primary responsibility for reporting findings, The Canadian Field- Naturalist papers must continue to use the appropriate technical and scien- tific terminology. It is desirable however, that abstracts be written in fairly simple language (when this is possible without sacrificing accuracy) so that the widest possible readership is reached. (8) — The Canadian Field- Naturalist Reserve Fund: The Canadian Field-Naturalist Reserve Fund ensures that, in the face of an economic disaster within The Ottawa Field-Naturalists’ Club, the journal would have sufficient funds to publish one further volume and, thus, enough time to make other publish- ing arrangements. The Fund will not be permitted to decline below a minimum level established by the Finance Committee (and reviewed by that committee as required). C — TRAIL & LANDSCAPE Although the information presented in Trail & Landscape must be factually correct, its manner of presentation will usually be less rigorously structured and more discursive than for The Canadian Field- Naturalist. This reflects the more casual and less scientifically-oriented readership of Trail & Land- scape and this publication’s conscious effort to edu- cate interested lay persons concerning the natural his- tory of the Ottawa area. Trail & Landscape is published five times a year and is available to all local members, on demand to members outside the Ottawa district, and to institu- tions on subscription. It is managed by annual budget. 1983 (1) — Objectives: (a) to encourage the membership to make and to publish natural history observations that are related to the Ottawa District. (b) to promote and to report on Club activities and the participation of the membership in such activities. (c) to provide information and articles of interest regarding the natural history of the Ottawa area. (d) to present, analyze and comment on natural his- tory issues (such as conservation) of significance to the Ottawa area. (2) — Achievement of the Objectives: Useful indicators for the achievement of the objec- tives for Trail & Landscape include the following: (a) unsolicited comments by members on an issue- by-issue basis. (b) the effort required by editorial staff to obtain sufficient appropriate material. (c) the results from member surveys. (d) citations in other publications. (e) adherence to a regular publication schedule. (3) — Content and Readability: Trail & Landscape will continue to provide a bal- anced mixture of natural history articles oriented towards people, activities, and education. D — THESHRIKE The Shrike is a newsletter dedicated to the interests of bird-watchers which is available, by subscription only, to The Ottawa Field-Naturalists’ Club members and to the general public. It should be self-sustaining to the extent that the subscription fee should cover production, printing and mailing costs. (1) — Objectives: (a) to provide periodic reviews and analyses of recent bird sightings in the Ottawa District. (b) to serve as a depository for local bird observa- tional data, these data to be made available for scientific research purposes. (c) to offer birding advice and to provide notice of programs and activities of interest to local birders. (2) — Achievement of the Objectives: Useful indicators for the achievement of the objec- tives for The Shrike include the following: (a) the number of subscribers. (b) the number of contributors of observations and the extent to which their observations are repre- sentative of the Ottawa District throughout the year. (c) unsolicited comments by subscribers on an issue- by-issue basis. (d) the number of requests for access to the data base. PUBLICATION POLICY 233 (e) adherence to a regular publication schedule. Responsibility for overseeing the production of The Shrike rests with the Publications Committee. The Birds Committee will assist by providing guidance for the publication. (3) — The Shrike Data Base: The data deposited with The Shrike shall be made available to any person or group requesting them for scientific purposes. Within reason, in order to be use- ful, these data should be verified and accumulated ina form making them readily accessible to a researcher. Several programs have been written to provide output of these data in the form of potentially useful graphs and tables. Data entries are at the expense of The Ottawa Field-Naturalists’ Club. E — SPECIAL PUBLICATIONS PROGRAM The Club has three regular publications — The Canadian Field-Naturalist, Trail & Landscape, and The Shrike. \t should be recognized, however, that from time to time there may be material that would be better suited to separate publication. In considering a special publication the following guidelines should be used, with these general comments: (1) each case will be treated on its own merits. (2) authors are not to profit financially from the undertaking. (3) assistance to authors will be considered on an individual basis. (4) achievement of objectives will be assessed, as with other Club publications. (5) accountability to Council fora special publication will rest with the Publications Committee. Guidelines: (1) Council approval-in-principle to be sought, fol- lowing which, action is to be taken by the Publica- tions Committee. (2) The Publications Committee to assess proposal or manuscript and to recommend approval or rejec- tion to Council. If approval is recommended, to recommend also an Editor for the undertaking. (3) Council to approve or to reject the proposal. (4) Ifthe proposal is approved, Editor installs project committee, arranges contact with author(s), and appoints reviewers. (a) The Editorial Committee to include: — at least one specialist in an appropriate field of study — at least one Council member — the Editor — a person knowledgeable in publication marketing (not all members need be Club members) (b) The Editorial Committee to develop, first with the author and then with appropriate Club commit- 23 4 THE CANADIAN FIELD-NATURALIST Vol. 97 tees, a Publication Plan for Council approval. The Plan will include: assessment of the proposal’s significance with respect to the Club’s mandate — budget estimates and funding sources, including IF (1) possible cooperative publication efforts and avail- ability of grants contractual proposal for the author marketing assessment and plan production schedule The Publications Committee to review the Publi- cation Plan and, if it is acceptable, to pass it to Council for approval. Council approval of Publication Plan to lead to development of contract with author, printer, and illustrators. Contracts to be given final Council approval. ) Document to be published and distributed. Feedback on value of the publication to be sought, including volume of demand (vs. forecast of demand) and audience reaction. A report to Council to be made, evaluating the undertaking, by the Editor. — GENERAL ISSUES — The Ottawa Field-Naturalists’ Club Publica- tions Committee: (a) Duties: The duties of the Publications Committee are as follows: **(1) to supervise the policy, finances, and distribu- tion of the Club’s publications. **(11) to act in an advisory capacity to Council in all matters pertaining to the publications of the Club. **(i11) to recommend an Editor and a Business Man- ager for each publication, as required, for approval by the Council and to appoint Asso- ciate Editors. (iv) to ensure that The Ottawa Field-Naturalists’ Club publications are oriented to meeting the objectives of the Club by regularly reviewing the achievement indicators described for each. (v) to act as a liaison between Council and the Editors and to maintain a constructive dia- logue with the Editors on matters of policy, operations, and procedures. (vi) to resolve concerns of the Editors or members of the Club respecting the publications. (vil) to recommend an Editor for each Special Pub- **taken from The Ottawa Field-Naturalists’ Club By-laws. lication for approval by Council and to act as liason with him or her in the development of Special Publications in the manner prescribed in Section E. ; (vill) to ensure that the Editors of the various pub- lications meet occasionally with Council. (1x) to review the Publications Policy annually and to report on such review to Council. (b) Membership: The membership of the committee will consist of at least the following individuals: (i) the Editors of The Canadian Field- Naturalist, Trail & Landscape, and The Shrike. (11) the Business Managers of The Canadian Field- Naturalist and Trail & Landscape. (iii) an Ottawa Field-Naturalists’ Club Vice- president and three or more members in good standing of The Ottawa Field-Naturalists’ Club who are not directly associated with a publication. (iv) the Editor of each Special Publication as an ex officio member. (c) Chairman: The Chairman will be a member of Council and not directly associated with a publi- cation of The Ottawa Field-Naturalists’ Club. (2) — Editorials and Changes in Editorial Policy: In order to maintain a balance between editorial freedom and the policy - and direction-setting responsibility of The Ottawa Field-Naturalists’ Club, broadly accepted mechanisms for consultation and approval are required. The following mechanisms aim to maintain that balance: (a) policy change proposals will be presented to the Publications Committee by Editors; minor changes will be dealt with by the Committee and major policy changes will be referred to the Council. (b) each Editor will be invited to meet occasionally with Council to keep Council informed on the progress of the publications. (c) before the appointment of any new Editor, he or she will meet with the Executive and Publications Committees to discuss The Ottawa Field- Naturalists’ Club publications policy for that publication. (d) the authorship of all material must be specified. (e) Editors must send a copy of all editorials to the President prior to type-setting for his or her information. Book Reviews ZOOLOGY The Mammals of Minnesota By Evan B. Hazard. 1982. University of Minnesota Press, Minneapolis. xii + 280 pp., illus. Cloth U.S. $39.50; paper WESEEOIS9 5: As the first complete work dealing with the mam- mals of Minnesota in 30 years, this work is a signifi- cant contribution. Seventy-eight (plus three extir- pated) species are discussed. Except in the case of Peromyscus maniculatus bairdii and P. m. gracilis, subspecies are not treated in the text itself, but the subspecies found in the state are listed in an appendix. The large total number of species 1s a reflection of the variety of habitats in Minnesota; taiga, deciduous forest, pine forest, and prairie are the major types present. Species accounts comprise the bulk of the book. These include standard metric measurements, des- criptions of the whole animal and the skull, statements on range, habitat, natural history, and relationship to people, and a distribution map. A listing of pertinent references for each species, by author and year, is also included; the full citations are in the extensive 57 page bibliography which includes many general works as well as those specific to Minnesota. The bibliography includes citations through 1981. The distribution maps show county lines with the locality symbols centered on the appropriate townships from which specimens have been collected or recorded. This method of noting distribution is especially useful for Mammals of the National Parks By Richard G. Van Gelder. 1982. The Johns Hopkins Uni- versity Press, Baltimore, Maryland. xvi+ 310 pp., illus. Cloth U.S. $24.50; paper U.S. $8.95. This book introduces mammal-watchers to the commonly seen species of mammals in America’s national parks and monuments. It also discusses the parks themselves including suggestions on where to see particular species. Part One (162 pages) describes 48 parks and mon- uments, mostly in the western United States and Alaska. Maps are included for those parks providing good mammal-viewing. These maps are redrawn from National Park Service maps and show most main roads, visitor centers, and certain major features such as mountains or the various localities referred to in the the larger counties and for rare species. Overall, many areas of the state are poorly known. Hazard’s book clearly shows what is currently known and will likely stimulate further study. For many species, a black and white sketch of the animal and/or its skull are also included. A well-written and informative introductory sec- tion of eight pages describes mammalian characteris- tics, Minnesota habitats, and the use of scientific col- lections of mammals. The text also includes a useful glossary of anatomical and ecological terms and bracket keys to skulls and whole specimens (or skins). Overall, the keys are workable, but those unfamiliar with skull anatomy would benefit from a general fig- ure naming the various bones and foramina. Most of these terms are defined in the glossary or pictured within the key, but no comprehensive figure is included. As with most state or regional books, this work is intended for students, professionals, and the general reader. It should be of value to all these individuals as it is well-written, up-to-date, and provides easy access to the literature. Lastly, it isa pleasure to note that the book contains very few typographical errors. DAVID A. LOVEJOY Biology Department, Westfield State College, Westfield, Massachusetts 01086 accounts. Comments on the likelihood of seeing each species (e.g. readily observed, seldom seen) are included. Park accounts include a mammal checklist with notations indicating extirpated, introduced (non-native), and re-introduced species. Part Two is a 135 page treatment of the mammals themselves. Although all taxonomic groups are men- tioned, those species and species-groups most likely to be seen are emphasized. Therefore, shrews, moles, bats, mice, rats and pocket gophers are discussed only briefly. Wood rats (Neotoma) and the Brazilian Free- tailed Bat (whose evening flight from their nursery colony in Carlsbad Caverns National Park is an impressive sight) are the only mammals in the above groups which are discussed in depth. 235 236 An introductory section includes useful hints on observing and photographing mammals and general information on national parks and monuments. Designed for the amateur, the treatment is non- technical and common names are used throughout (an appendix lists scientific names). The book is well- written, easy to use, and nearly completely free of typographical errors. In addition to being a useful The Fisher: Life History, Ecology, and Behavior By Roger A. Powell. 1982. University of Minnesota Press, Minneapolis. xvi+ 217 pp., illus. U.S. $19.50. Like the wolverine, the fisher is perceived as a mys- terious, almost mythological animal, which lives a secretive existence in the forests of North America. Surprisingly, a considerable amount of research has been completed on this member of the mustelid fam- ily; however, few books deal with the fisher in a com- prehensive manner. Powell has successfully written a book which appeals to the general public, as well as the scientific community. This is a most difficult task, one which few authors successfully accomplish. Through 10 chapters Powell examines the ecology and biology of the fisher. He relies heavily on his own research and experiences with captive and free- ranging fishers. Powell complements the text with a fairly extensive literature review. The chapters dis- cuss: taxonomy; anatomy; life history (including a succinct review of the theories of delayed implan- tation); distribution and population density; general habits, home range and movements; food habits; hunting and killing behaviour; and the fisher’s rela- tionship to humans. Powell undertakes to integrate population, habitat and behaviour data throughout the book. This integration is limited only by virtue of the fact that his review of the fisher in the context ofa much larger ecological system is cursory. Many myths about the fisher have evolved through A Bird-finding Guide to Ontario By Clive E. Goodwin. 1982. University of Toronto Press, Toronto. 248 pp. $12.50. Authors who attempt regional bird-finding guides must necessarily face the question of what to include, and the more difficult question of what to leave out. For an area as vast as the whole of Ontario the pros- pect is daunting indeed. Clive Goodwin has in one small volume presented a list of the better known birding areas in Ontario along with detailed informa- tion on how to get there, when to go, and what one might reasonably expect to see. This book covers everything that would be expected THE CANADIAN FIELD-NATURALIST Vol. 97 guide for anyone interested in mammal-watching within the park system, this book is interesting and informative in its own right. | recommend it highly. DAVID A. LOVEJOY Biology Department, Westfield State College, Westfield, Massachusetts 01086 the ages. Powell works to dispel some and clarify others. This is evident in the chapter on the predator- prey relationship of the fisher and the porcupine. One myth is that fishers turn over porcupines and kill them by attacking the soft, unprotected ventral surface. Of six major predators, (ie. wolf, coyote, mountain lion, lynx, bobcat, and fisher) Powell reports that the fisher is the best adapted species to deal with porcu- pines. Unlike the other predators, the fisher is built low to the ground, which allows it to directly attack the face and head area of the porcupine. In addition, the fisher is powerful enough to inflict substantial wounds to the face and head, but is agile enough to avoid the porcupines tail during a confrontation. Once the porcupine ts killed or immobilized, the fisher then turns it over and feeds on the soft, unprotected ventral surface. The book 1s illustrated with black and white photo- graphs, and black line ink drawings and maps. A reference list is provided. I recommend this book to all who are interested in the biology and ecology of the fisher. It is an excellent introductory reference. PAUL A. GRAY Northwest Territories Wildlife Service, Yellowknife, Northwest Territories X1A 2L9 from such a guide, including brief descriptions of habitats, weather, a bird-life chronology, tips for vis- itors, and a checklist. Central to the book, and occup- ying 75% of the pages, are the lists of birding localities. To make this more manageable Goodwin has subdi- vided Ontario into six regions corresponding, more or less, to major ecological zones. The localities in each region are arranged alphabetically; the space devoted to each being roughly equivalent to its birding produc- tiveness. An outline map of each region is provided to show the relative position of the places mentioned. There are a few points readers should keep in mind , 1983 when using the book. There are a few typographical errors, which presents no serious problem. The com- plexities of some of the directions make it imperative that the appropriate road or street map be available. The confusion is compounded by some inaccurate street names: in the Ottawa account, on page 150, Moodie Drive is called Moodie Street, and nine (9) lines later Moodie Road. On the same page Carling Avenue is once referred to as Carling Street. Apart from the six regions, maps of an additional six specific locations are shown. Their choice seems random — why those six at the expense of all the other locations? Many more maps, perhaps at the expense of a few sewage lagoons, would be helpful and could have reduced the confusions resulting from the writ- ten directions. The birds associated with certain local- ities are sometimes inaccurate. In the Richmond sec- tion, page 151, the swamp (actually a fen), contrary to Goodwin’s statement contains few rails besides the Yellow Rail, but does have a large colony of Sedge Wrens (not mentioned). The Richmond fen, like sev- eral places listed, is also said to have “many other species” but there is no hint to what they might be. The Amphibians and Reptiles of Manitoba By William B. Preston. 1982. Manitoba Museum of Man and Nature, Winnipeg. 128 pp., illus. $9.95 plus $1 postage. Most peopleprobably do not associate the province of Manitoba with amphibians and reptiles. If they do, they probably think of the large communal hibernac- ula of the red-sided garter snake which have been widely publicized. The Manitoba herpetofauna, how- ever, 1S surprisingly large for the climate. Thirteen species of amphibians and eight of reptiles occur in the province, concentrated mainly in the south, although some species such as the Wood Frog range very far north. Two species, the Common Garter Snake and Northern Leopard Frog, are so abundant that they have been harvested commercially, necessitating regu- lations. This fauna is a mix of eastern species reaching their western range limits (e.g. Blue-spotted Sala- mander, Green Frog), western species at their eastern limits (e.g. Plains Spadefoot, Plains Hognose Snake), and a few essentially transcontinental species (e.g. Painted Turtle, Common Garter Snake). If nothing else, Preston points out to citizens of Manitoba and elsewhere the richness of this assemblage. As one who got his professional start in herpetology in Manitoba, I welcome this publication. Bill Preston had three objectives in writing this book: to provide easy identification of species in a regional guide format, to provide up-to-date informa- tion on amphibian and reptile biology, and to point BOOK REVIEWS 23 Almost anywhere will have “many other species”. In the Systematic Lists of the species, Goodwin’s warn- ing that the abundance designations could be mislead- ing to the unwary should be echoed. White Pelicans are listed as a rare summer resident in the north but only because of a breeding colony at Lake of the Woods. It would not be a “hoped for” species at Moosonee. What Goodwin has tried to do is compile in one volume a list of the places where the local experts do their birding. Overall he has produced a usable man- ual which would be most useful to an average birder planning a trip to, or within, Ontario. The binding appears strong and the book could probably tolerate the abuse of several camping trips. The price seems a bit high fora book with few maps and no pictures but may be well worth the cost in saved time and gas when hunting for a certain birding spot. BRUCE DILABIO 62 Grange Street, Ottawa, Ontario KIY 0N9 out areas where knowledge is lacking. He has suc- ceeded admirably in the first of these, and to a lesser degree in the other two. Preston’s book is written from his long experience as a naturalist and professional herpetologist. This is evident from the fact that most of the excellent black- and-white and colour photographs of both animals and habitats were taken by the author himself. Pres- ton also livens up the text in many places by recount- ing personal experiences. Finally, he departs from the usual format of popular field guides by referring to scientific literature in the text, using the literature citation style of scientific journals. This does not inter- fere at all with readability, yet allows one access to more in-depth treatment of a subject. This book is short but extensively illustrated. Its organization is straightforward and fairly conven- tional. Introductory sections briefly cover such topics as general features of biological nomenclature, basic biology of amphibians and reptiles, the Manitoba environment, and the distribution of amphibians and reptiles in Manitoba. The bulk of the book is devoted to individual accounts of all 21 species, each account including a photograph of the species, description of the animal and its geographic distribution, and cover- age of basic natural history (habitat, food habits, breeding, etc.). Maps showing locality records of each species are also included, but are all grouped together, somewhat inconveniently, at the back of the book. 238 Preston ends with a brief glossary, a list of all litera- ture cited in the text, and a list of other general books for the beginning herpetologist. He also includes a list of hypothetical species whose ranges approach Manitoba and might therefore be recorded there sometime in the future. I was also pleased to see a description of how to preserve specimens, a list of French common names, and separate keys for the identification of living and of preserved tadpoles. The major problem with this book ts that it is too short. This has two undesirable effects. First, the indi- vidual sections are so short as to make for very jumpy reading in places. Second, several important topics are given too brief a treatment or left out entirely. For example, the section on amphibian biology makes no mention of anuran vocalizations or paedogenesis (neoteny) in salamanders. Both are mentioned in indi- vidual species accounts but seem to me to be of sucha basic nature that they warrant more general treat- ment. Similarly, the discussion of ectothermy and thermoregulation omits consideration of costs, bene- fits, and adaptive significance of these phenomena; this is unfortunate, especially in view of the general misunderstanding of these concepts by lay people. Most noteworthy, however, is that there is no section dealing with factors limiting abundance and distribu- tion of north temperate amphibians and reptiles, and physiological and ecological responses made to these factors. Surely this is one of the most important areas of research that Canadian herpetology should address. 1 discovered few errors in reading this book. The only factual one concerns excretory products of amphibians and reptiles. According to Packard (1966, American Naturalist 100: 667-682), reptile embryos and adult amphibians both excrete mainly urea. Pres- ton, however, indicates that the former excrete uric THE CANADIAN FIELD-NATURALIST Vol. 97 acid (actually the main excretory product of adult squamates) and that the latter excrete ammonia (which is actually excreted only by aquatic amphibian larvae). A very minor error (possibly typographical) is the word “hemipene”, instead of hemipenis, among terms defined in the glossary. Preston also refers to the spadefoot as Scaphiopus and to Australian hylid frogs as Hyla, rather than Spea and Litoria respectively as is currently the fashion. However, this is really more a matter of systematic opinion than an error. My remaining quibbles with this book are mainly matters of style. Preston frequently switches from first to third person when referring to himself. Descrip- tions of full species ranges outside Manitoba are not consistently given in the individual species accounts; sometimes only the particular subspecies distribution is given. Standardized common names do not always correspond exactly to the subspecies names given. For instance, the section introducing colubrid snakes pairs standardized common names of species with scientific names of full species. I question the need for standard- ized common names, at least for subspecies, but if they are to be used, they should be used consistently. Overall, | think this book serves an important func- tion. It represents the first popular summary of mod- ern information on Manitoba’s amphibians and rep- tiles. If it spurs interest in these two groups and encourages people to be more concerned about con- serving them and adding to our knowledge of their natural history and distribution, it will play a valuable role. Bill Preston is to be congratulated for making this first step. PATRICK T. GREGORY Department of Biology, University of Victoria, Victoria, B.C. V8W 2Y2 Aquatic Insects and Oligochaetes of North and South Carolina Edited by A. R. Brigham, W. U. Brigham, and A. Gnilka. 1982. Midwest Aquatic Enterprises, Mahomet, Illinois. 837 pp., Illus. U.S. $39.50. Close on the heels of many a worthy predecessor comes this latest addition to a library of information on aquatic insects. With so many texts now available it is becoming increasingly difficult for students and professionals alike to choose the best and most appropriate one for their needs. In a somewhat nega- tive way this new book helps to narrow down the choice. Originally conceived as a guide for pollution biolo- gists, at the Duke Power Company, the book 1s still very much an in-house publication. This is primarily the result of two major restrictions that have appar- ently been designed into the book. First, as the title states, this is a study of aquatic insects of the Caroli- nas, and the relevance of much of the descriptive sections and most of the keys is restricted to this region. In fact much of the fauna of the northeastern U.S. and certainly most of eastern Canada lies beyond the scope of this work. The other major restriction arises as a result of the unidirectional orientation of the book to pollution biologists. One can only assume that pollution biolo- gists do not deal with anything other than the larval stages since, except for the Coleoptera and Hemi- ptera, only the larvae are considered. It is interesting 1983 to speculate what these biologists do with pupal stages encountered in their studies. The inclusion of the oligochaetes in this work, con- sidering the above orientation is easy to comprehend. However, by the same rationale one would also expect inclusion of other groups commonly encountered in the course of pollution studies, such as molluscs, crus- taceans, amphipods, and isopods. None of these appear and as a result the treatment of the groups is neither a complete account of the aquatic insects nora comprehensive treatment of organisms encountered in pollution studies. Rather, the book falls somewhere in between, fulfilling neither consideration completely. Within the area it does cover, the book treats its subject matter in considerable detail. Each chapter has a general description and diagnosis of the families in the order, as well as an extensive treatment of their life histories, though the latter may dwell at length upon only one, often well researched, genus. This is followed by a discussion of taxonomic problems within the family, including histories of synonymies and a discussion of the more useful taxonomic charac- ters. Some chapters, though by no means all, include detailed tables of ecological and habitat data, infor- mation that can often be invaluable when used in conjunction with the keys. A short checklist of species in the Carolinas pre- ceeds the main part of each chapter — the keys to genera, and, where applicable, to species. Unfortu- nately in compiling these keys no real consistency occurs from chapter to chapter. As a result, in some orders keys are provided to all the genera of eastern North America, while in others, only genera occurring Falcons of the World By T. J. Cade. 1982. Comstock/ Cornell University Press. Ithaca, New York. 192 pp., illus. U.S. $38.50. No one should be surprised that Tom Cade, a man witha lifelong fascination for birds of prey, especially falcons, has written a book on his favourite subject. However, as an apprentice in the study of falcon biol- ogy who can only benefit from Cade’s rich experien- ces, | am particularly pleased that he has produced more than just a coffee-table book. The book is divided into two sections. The first consists of ten chapters describing the special charac- teristics of the genus Falco, classification, distribution and migration, feeding adaptations, size and flying performance, hunting success, reversed size dimor- phism, social behaviour and reproduction, and finally, the falcon’s relationship with man. The second section provides us with descriptions of 39 species, perhaps only 38 if Cade’s conclusion about Falco BOOK REVIEWS 239 in the Carolinas are considered. Keys to species (not all genera are keyed to species level) are much more consistent and in virtually all cases are only for those found in the Carolinas. In most cases these are a synthesis of existing keys, often merely a contraction of more regional treatments to include only the local species. The only major addition is that taxonomic changes that have occurred since these keys were first published (in some cases nearly 40 years ago) have been incorporated, thus reducing confusion, particu- larly for those unfamiliar with the groups. The keys are on the whole well illustrated though occasionally those redrawn from other sources have suffered. Finally, a few mistakes, such as mislabelled illustra- tions are in evidence though these are no more fre- quent than in most books of this kind. Perhaps the most noteworthy point about the book, and that which distinguishes it from others of its kind, is in its design. Bound in an easily disassembled loose- leaf format, it provides for easy addition of future revisions and changes (purchased at extra cost) thus keeping the book up-to-date. For those working on the fauna of the Carolinas and their immediate surroundings, this book will no doubt serve as a more than adequate guide to the fauna. However, for those outside this area, and in particular in eastern Canada, the use of this book asa single source is not recommended. At most, it should be used only in addition to other sources, where appli- cable, to round out the information already available. REIN JAAGUMAGI Department of Entomology, Royal Ontario Museum, 100 Queen’s Park, Toronto, Ontario MSS 2C6 kreyenborgi is correct, and maybe even less if his desire for further taxonomic research on African spe- cies, in particular, is fulfilled. In fact, this latter point really reveals what I liked about the book. Using it as a forum to present ideas to stimulate new research, Cade has provided in his book a fresh approach to each chapter. For example, what is the real purpose of bony tubercles in the nares? Does the nasal secretion serve in some nutritional capacity? Does rangle aid in the removal of the koilin? I was especially interested in Cade’s explanation of the long-standing controversy, reversed size dimor- phism in birds of prey, which I believe can only stand as yet another untested hypothesis. The author seems convinced that the “big mother” hypothesis, 1.e. big mothers... big offspring... better survival. (K. Ralls. 1976. Quarterly Review of Biology 51: 245-276) is applicable to raptors. As far as I know, there is no 240 evidence for this as yet. In fact, a recent paper (D. M. Bird and P. C. Lagué. 1982. Canadian Journal of Zoology 60: 71-79) showed that physically smaller handreared American Kestrels (Falco sparverius) laid larger eggs than parent-raised birds. This may be an artifact of captivity, but there also is no evidence indicating that larger eggs produce larger fledglings more capable of surviving. Hence, I do not think we have seen the end of this great debate. I agree with Cade that “copulation in falcons has become ritualized into another form of courtship dis- play” and hasten to add more evidence. Spermatozoa can remain viable in the Kestrel oviduct up to 12 days (D. M. Bird and R. B. Buckland. 1977. Canadian Journal of Zoology 54: 1595-1597). Why should this be so if every copulation need result in sperm transfer? Another section I fancied was Cade’s argument for the existence of falconry. I have heard many argu- ments for and against this sport, some based on fact and others on pure sentiment, but I have never read a more eloquent and soundly based one as that put forth by Cade. This is required reading for anyone with strong feelings on the subject. It is understandably difficult to cover everything on a given topic, but I was surprised that little or nothing was mentioned of the falcon’s remarkable eyesight, nor the Kestrel’s astounding ability to fix its head in one position while its body undergoes various gyra- tions, allowing it to maintain a “fix” on prey while hovering or sitting on a moving perch. Also, finding falcon eggs to be as beautifully coloured as the birds themselves, I would have liked a word or two describ- ing them for each species (where known). My criticisms of the editing are relatively minor. In fact, throughout the entire book, I found no more than two dozen typographical, grammatical and spell- ing errors. The only annoying aspect of the writing style (which may not have been the author’s fault) was the inconsistency of the reference system. In some places, et al. was used and in others not. This even occurred on the same page, e.g. p. 120. Frequently, the BOTANY Survey of Canadian Herbaria By Bernard Boivin. 1980. Provancheria No. 10, Université Laval, Québec. Louis-Marie, Université Laval, Québec. Available from the author at Herbier Louis-Marie, Uni- versité Laval, Québec. 187 pp. $10. This book includes descriptions of 404 Canadian herbaria, both past and present, with 335 being insti- tutional, 69 private, and 248 still presumed active today. The author’s three-fold objectives for the book were: |) to provide information about the research THE CANADIAN FIELD-NATURALIST Vol. 97 periods at the ends of sentences preceded the reference which was not always in brackets. The range maps are extremely useful and a welcome addition. If Cade’s plea for further research on many species is heard, then these will undoubtedly change. For example, five years ago, I received a brood of orphaned American Kestrels from as far north as the east coast of James Bay (well above that shown on Cade’s map), and to top that, in the summer of 1982, I observed a male Kestrel perched on a hydro line beside the airport in Fort Chimo in Ungava Bay! As an aside, some readers may be confused by the lack of referencing for a few of the maps covering two species. Besides Cade’s authoritative writing, the book is richly endowed with 40 full-page colour paintings. I personally like R. David Digby’s style, which is to be expected because | also favoured D. M. Reid-Henry’s work (the former studied under the latter) above seven other artists in an earlier raptor book. Some readers might have liked to see both sexes and the immature plumages presented for each species (where war- ranted), but with the painstaking hard labour incor- porated into each painting, I think this is asking too much. Digby’s work is impressive and indicates he is no stranger to falcons. In summary, my criticisms of the book are largely miniscule when one considers the author is sharing with us more than 40 years of experience in observing both wild and captive falcons. As a “student of fal- cons”, I devoured the book and examined it with a fine-tooth comb. It has without doubt much to offer to those with an unquenchable curiosity for knowl- edge of these magnificent birds and to those with a refined appreciation for superb artwork. DAVID M. BIRD Macdonald Raptor Research Centre, Macdonald Campus of McGill University, Ste. Anne de Bellevue, Quebec H9X 1C0 capabilities of Canadian herbaria, 2) to suggest the herbaria that might contain specimens for particular research needs, and 3) to provide information on loca- tions of voucher specimens for previous studies. The nature and purposes of herbaria in general, previous herbarium surveys relevant to Canada, objectives and methodology of the present survey, and its format, are briefly described respectively in four short introduc- tory chapters. The actual descriptions of Canadian 1983 herbaria follow the introductory chapters and com- prise the main text. The book concludes with a brief chapter attempting to give some statistical summaries, an appendix reproducing some facsimiles of relevant historical documents, and a relatively extensive alphabetical index giving the herbarium acronyms and/or names, institutions, collectors, and plant taxa mentioned in the text. In the main text body, the included herbaria are alphabetically arranged and cross-referenced by their official and previously used acronyms, owner’s names (if the herbaria are private), and cities where located. The herbarium descriptions vary considerably from very brief (3-5 lines) to relatively extensive (‘4-2 pages). Information is usually given on date of origin (and sometimes also considerable history), size, area of concentration, main collectors, comments on exchanges and loans (but only in reference to Department of Agriculture, Ottawa (DAO) and the author), and some abbreviated references to informa- tion about the particular herbarium. This survey represents the results of a 30-year com- pilation of information about various Canadian her- baria by the author, a prodigious compiler, who obviously invested much time and effort on this task. Included is much useful and interesting historical information about various herbarium collections in this country. It has performed a valuable service in recovering and preserving such historical knowledge as part of our botanical heritage, aside from its stated primary objective of providing information on Cana- dian herbaria as resources for research. This publication is not, however, the even-handed and updated review of all present-day herbaria across the country that readers might expect. It was admit- tedly a personalized compilation and biased by the author’s own interests and his rather informal information-gathering methods, which consisted mainly of personal impressions and notes accumu- lated during his herbarium visits and studies spanning 30 years. The result, although highly informative and often fascinating in many details, is a distinctly uneven coverage of Canada’s herbaria, both in terms of the amount of detail included and the recency of much information. Some herbaria seem very well described The Rare Vascular Plants of the Yukon By G. W. Douglas, G. W. Argus, H. L. Dickson and D. F. Brunton. 1981. National Museum of Natural Sciences, Ottawa, Canada. Syllogeus No. 28.61 pp. (English), 64 pp. (French) + 35 pages of maps. Free. This is one of a series of publications dealing with the rare plants of the various Canadian provinces and BOOK REVIEWS 241 (e.g. CAN, DAO, MFJB, MT, MTMG, QFA, QK, TRT), but others are much less thoroughly reviewed. Sometimes the descriptions of particular herbaria appear fragmentary, subjective or uncertain, or scarcely if at all updated from information that must have been obtained 15, 20 or even 30 years ago (or often with only a size-amendation from /ndex Herba- riorum Edition 6, 1974). The lists of “main collectors” given with each herbarium description appear particu- larly non-uniform and in fact often seriously distorted with regard to inclusions vs. exclusions and to the recency of information. For some herbaria, the “main collectors” listed are strictly historical including none (or very few) more recent than 20-25 years ago, but for other institutions the lists may even include collectors up to the late 1970s. Obviously the author’s symbol- annotated alphabetical list of Canadian collectors in his extended concluding index can hardly be more consistent and complete than the individual lists from which it was compiled. One might have expected the author of a publica- tion, professing by its very title to be a national survey of herbaria, to have made a determined and syste- matic effort to update his information prior to publi- cation and to attempt giving a roughly equivalent treatment of all herbaria across the country. A form letter requesting needed information circulated to all curators (not just a selected few) might have avoided much of the excessive unevenness evident in this sur- vey. But as a curator himself, this reviewer can attest that no such communication concerning at least his herbarium was received during the last 14 years. This Survey of Canadian Herbaria represents an interesting and informative work of considerable his- torical value that is certainly recommended to readers. It should not, however, be uncritically accepted as an accurate, even-handed and up-to-date assessment of Canadian herbaria today or of the collectors asso- ciated with them. VERNON L. HARMS The W. P. Fraser Herbarium, University of Saskatchewan, Saskatoon, Saskatchewan S7N 0WO territories: #14(1977) Ontario; #17(1978) Alberta; #18(1978) Nova Scotia; #20(1979) Saskatchewan; and #27(1980) Manitoba. For their purpose, the authors consider a rare plant as “one that has a small population within the area under consideration. It may be restricted to a small 242 THE CANADIAN FIELD-NATURALIST geographical area, where it may be locally common, or it may occur in low numbers over a wide area”. In this volume, 313 species (about a quarter of the known flora) are considered as rare in the Yukon Territory. For each species the following information is presented: reference (herbarium location of speci- mens or bibliographic citation), distribution (both total and within the Territory), reference to distribu- tion maps, if any, habitat, status elsewhere in its range where it is rare, and sometimes additional comments, and a Yukon distribution map. Works such as this offer a challenge both to the amateur and the professional not only to find new sites and thus extend the known ranges of the species listed, but also to add new species to the list. In the process, some species may be found to be more fre- quent than hitherto supposed, and thus might have to Weeds By W.C. Muenscher, with foreword and appendices by P. A. Hyppio. 1980. Second edition. Comstock Publish- ing Associates, a division of Cornell University Press, Ithaca. xvii + 386 pp., illus. U.S. $29.50. The appearance of a new version of a classic text- book must always create ambivalent feelings. On the one hand there is the pleasure of seeing a familiar, well-proven work available once more, but on the other there is the question of whether it has been brought up-to-date adequately, and, perhaps more painfully, if indeed the old classic is capable of being updated or if it should have been allowed to die with its original author. Walter Leopold Conrad Muenscher, the original author of Weeds, was born in Germany in 1891, moved to the United States as a child, and taught at Cornell from 1916 to 1954. He died in 1963. The first edition of Weeds, a handbook of “the commonest weeds of the Northern United States and Canada”, appeared in 1935, long before the advent of modern herbicides. A second edition appeared in 1955. In preparing it, Muenscher faced the dilemma of how to handle the burgeoning field of chemical weed control. In retrospect, he was probably wise in deciding to omit all mention of it, save in his preface, and to keep the emphasis in the handbook on weed identification. Nevertheless this does lead to quaint paragraphs, red- olent of another era, that say, for example: “Control — Hand hoeing (5), hand pulling (4)”. The numbers refer to paragraphs in the |7-page chapter on “The Control of Weeds”, which deals almost exclusively with methods of mechanical control. On the positive side, this is an area which is all too easily overlooked to-day, and which should be of increasing importance Vol. 97 be dropped from future lists, and species new to the province or territory may be found. Froma conserva- tion point of view, works such as this help point out the areas which might be considered of special inter- est, and thus receive protected status. The work is a welcome and useful contribution to our knowledge of one of the most interesting regions of Canada. Additions, suggested deletions, comments and cor- rections are requested by The Rare and Endangered Plants Projects, Botany Division, National Museum of Natural Sciences, Ottawa, Ontario, KIA OM8. WILLIAM J. CODY Biosystematics Research Institute, Agriculture Canada, Central Experimental Farm, Ottawa, Ontario K1A 0C6 to home gardeners and certainly to those with a heightened awareness of the hazards to the environ- ment of the indiscriminate use of herbicides. The publication under review is a “Reissue” of this 1955 Second Edition with a new foreword and three new appendices by Peter Hyppio, Curator of the Her- barium and Extension Botanist at the L. H. Bailey Hortorium of Cornell University. These, respectively, explain the reissue, give a translation from the scien- tific names used by Muenscher to “current nomencla- ture”, list a “standardized common name” for each of the weeds mentioned in the book, and provide a 14 page bibliography of sources for the “current nomen- clature” and “common names” that Hyppio adopts. In addition the foreword indicates that typographical errors have been corrected. Is the reissue a success or should the book have been allowed to die? Paradoxically, | believe the answer to both questions is “no”. If the appendices are really needed, then surely the reader deserves to have this information in the text and not have to check every single scientific name to see if it is still in current usage. Likewise, if a standardized list of English-language names is really needed, as, for example, to meet the dictates of the Weed Science Society of America, then surely the user should have this name highlighted typographically in the text. In general, Hyppio has done a good job in his appendices, but he fails to draw the distinction, which would have been possible in Appendix I, between scientific name changes which represent corrections (e.g. Eragrostis minor for “E. poaeoides” or the author citation for Stellaria media as “(L.) Vill.” instead of “Cyrill”), and those which merely reflect a 1983 change in taxonomic judgement, which is not neces- sarily accepted by all botanists. An example of the latter is his adoption of the genus Toxicodendron and his recognition of the different races of poison ivy as distinct species even although Mulligan and Junkins (1978, Le Naturaliste canadian 105: 291-293) have provided evidence that specific rank is inappropriate. This leads to his use of 7. radicans and T. rydbergii where others would use Rhus radicans subsp. (or var.) radicans and R. radicans subsp. (or var.) rydbergii. Indeed Hyppio seems, in this case, to have confused two taxa that Muenscher correctly distinguished. Muenscher has “R. microcarpa Steud.”, “R. radicans L.”, and “R. radicans L. var. rydbergii Small”. Hyp- pio treats the first two as synonymous (T. radicans) and the last as 7. rydbergii. It is evident, however, from the distributions given by Muenscher coupled with the work of Gillis ( Rhodora 73: 72-159, 161-237, 370-443, 465-540, 1971) which included typification of the epithet radicans, that in current nomenclature these are, respectively, R. radicans L. subsp. radicans, R. radicans subsp. negundo (Greene) McNeill, and R radicans subsp. rydbergii (Small ex Rydb.) McNeill. In some other cases Hyppio has been much more conservative in maintaining Muenscher’s nomencla- ture than one would expect, as, for example, when he retains white cockle or white campion in the genus Lychnis (as L. alba) rather than placing it, as is usual to-day, in the genus Si/ene, beside its oft-confused look-alike, night-flowering catchfly (S. noctiflora). (In Silene, it has to be called S. /atifolia Poiret (= S. pratensis (Rafn) Godron & Gren.)). One evident fail- ure to correct Muenscher’s nomenclature is in the genus Euphorbia, where E. maculata L. applies to what Muenscher calls E. supina, whereas his E. macu- lata is correctly known as E. nutans Lag. In any revision, however, errors and omissions are inevitable, and are not the final arbiter of its worth- whileness. A wider issue is whether Muenscher’s tax- onomic treatment is not so dated as to make a mere nomenclatural appendix inadequate for present-day needs. In part, this is the case. For example, ENVIRONMENT Nature Conservation Day Compiled by I. J. Beechey and B. L. Raad. 1981. Proceed- ings of asymposium, 26 March 1980. Ontario Ministry of Natural Resources, Toronto. 86 pp. $2. This seminar featured speakers from provincial and other agencies concerned with protection of natural areas. Absence of participants from Parks Canada and the Canadian Wildlife Service (National Wildlife BOOK REVIEWS 243 Muenscher makes no reference to the widespread prairie and northern plains dock Rumex pseudona- tronatus Borbas. He includes Polygonum erectum, which appears to bea sporadic plant of native habitats to-day, yet omits the abundant, weedy P. achoreum Blake. The two species are well distinguished by Mit- chell and Dean (1978, N.Y. State Museum Bulletin 431: 38-42) and Muenscher clearly illustrates the former, even though it seems certain that in describing the habitat and distribution he is referring to the lat- ter. Likewise, although Muenscher lists only one dark-flowered dog-strangling vine or swallow-wort, which he calls Cynanchum nigrum (incidentally, another incorrect name that is over-looked), two read- ily confused species seem almost as widely distributed in the northeastern U.S. and adjacent Canada; in the segregate genus Vincetoxicum, these are correctly known as V. nigrum (L.) Moench and V. rossicum (Kleopov) Barbarich (cf. McNeill, 1981, Le Natura- liste canadien 108: 237-244.). Further examples can readily be found, particularly in long-confused genera such as Atriplex and Chenopodium. Yet, for all its defects, Muenscher’s work remains a very usable classic. It is one of the few weed manuals with identification keys; the coverage is good; the brief descriptive and distributional notes are helpful and the historic approach to weed dissemination, “ecol- ogy’, and control still has a relevance to our under- standing of weed biology. The book is not dead. This reissue, for all its faults, is probably worth its price, even in U.S. dollars. What would be even better, however, would be a third edition, or a new book unashamedly building on Muenscher, that would take account of recent advances in our understanding of the variation, taxonomy and distribution of the weeds of the Northern United States and Canada. Perhaps Dr. Hyppio will one day provide us with it. J. MCNEILL Department of Biology, University of Ottawa, Ottawa, Ontario KIN 6NS5 Sanctuaries) is notable. These agencies play a role in nature protection within Ontario, and could have added a national perspective to the discussions. The Minister’s message stated that the seminar was a success, with the objectives of increasing awareness, acquainting different agencies with mandates and programs, commemorating achievements, identifying 244 needs and problems, and exploring innovative approaches. However, the true success of the seminar should be best judged by problems solved and innova- tions adopted. With the proceedings published one year after the seminar, actions taken within that year could have been summarized. The Director of the Parks and Recreational Areas Branch presented a short historical introduction to provincial parks, and related legislation and policies. Unfortunately, the significance and substance of the acts and policies were not adequately described. For example, the Endangered Species Act of 1971 was mentioned, but its provisions for natural area protec- tion were not explained. The Manager of the Planning Section of the same branch attempted to explain the inventory of natural areas. The reader was thrown into a sea of agency jargon which is poorly defined. The reader encounters terms such as “landscape representation”, “ecosystem representation’, “representation principle”, “life science framework”, “earth science features”, and others. The concepts of biological versus physical features in this chapter were confusing. Besides biological fea- tures, the “life science framework” included substrate, moisture, microclimate and landform patterns. The relationships of various types of reserves, parks, zones and classes was also confusing for those not imprinted on the appropriate jargon. By contrast, the late Mr. Coffin, Executive Director of the Niagara Escarpment Commission clearly defined the jargon used by the commission. The reader must not assume that similar terms used by different agencies have similar definitions. Mr. Coffin tackled the problem of similarities to Ministry of Nat- ural Resources jargon directly. Mr. Coffin clearly defined the commission’s princi- ples, objective, activities and mechanisms for natural area protection. The General Manager of the Federa- tion of Ontario Naturalists, Director of Conservation Authorities, and Projects Director of the Nature Con- servancy of Canada also presented clear, concise de- scriptions of their agencies’ principles, objectives, programs and organization. The water management THE CANADIAN FIELD-NATURALIST Vol. 97 mandate of the Conservation Authorities was logi- cally connected with their activities in natural area preservation. The discussion of joint projects by the Authorities, the Conservancy and the Federation were especially pertinent to the seminar’s objectives. Six representatives of the Nature Conservancy of Canada (other than the Projects Director) also made presentations. The information in these presentations was insufficient and fragmentary. Ihe viewpoints were largely redundant with repeated calls for improved inter-agency coordination and cooperation. The New England Field Director of the Nature Conservancy provided a concise, factual and clear explanation of the perceived need for nature protec- tion inthe United States, government’s response, and the Nature Conservancy’s objectives, organization and programs. The appendix of background informa- tion on the natural area inventory program was very useful. Other chapters would have benefitted from such appendices. The summary of the panel discussion reiterates the problems identified earlier, and provides some solu- tions. The Ministry of Natural Resources appeared reliant on one provincial plan, blueprint or strategy to solve the problems of inter-agency coordination and cooperation. However, regional, provincial and fed- eral governments, Federation of Ontario Naturalists and the Nature Conservancy of Canada answer to different constituencies. Alignment behind one pro- vincial strategy may be difficult. A forum for open discussion between agencies, as recommended by Dr. Fowle of the Nature Conservancy of Canada, could be critical to solve identified problems. This compendium of viewpoints should be read by naturalists, biologists, and system planners who want a concise overview of natural area protection in Onta- rio. Students will find it a good starting point fora more thorough study of this subject. MICHAEL A. D. FERGUSON Department of Renewable Resources, Government of the Northwest Territories, Frobisher Bay, Northwest Territories X0A OHO The Mitigation Symposium: A National Workshop on Mitigating Losses of Fish and Wildlife Habitats Co-ordinated by G. A. Swanson. 1980. General Technical Report RM-65. U.S. Department of Agriculture, Fort Collins, Colorado. 685 pp. Mitigation is not a new concept. In fact, it was conceived in the U.S. Congress and developed under the auspices of the Fish and Wildlife Coordination Act in 1934 (see L. M. Krulitz’s paper entitled Federal Legal Background for Mitigation). However, it has evolved as a legitimate science/ art only in the last two decades. And although such rapid evolution should be applauded, it has not emerged without problems, which range from variability in biological techniques 1983 and definitions to legal and political issues. Therefore, it was with pleasure that I read this volume, a fine attempt to deal with many of the outstanding ques- tions surrounding mitigation. The objectives of this symposium were: (1) “To review the magnitude, and the seriousness of the losses to fish and wildlife habitat as a result of changing land and water use with par- ticular emphasis on federal development projects. (2) To review the extent to which these habitat losses are being and have been mitigated, and (3) To develop strategies, and practical recom- mendations, for minimizing fish and wildlife habitat losses and achieving more effective mitigation.” To meet these objectives, 124 papers and poster session presentations were prepared and published in the proceedings. The volume is divided into numerous sections which deal with: the concept of mitigation; coastal zone wetlands; inland wetlands; economic considerations; mining, oil and gas; water develop- ments; planning, evaluation, inventory of habitat and impacts; surveys of mitigation problems and impacts; power projects; terrestrial management techniques; aquatic management techniques; legal and political considerations; transportation systems; state perspec- tives; and poster session papers. The vast majority of the papers in the proceedings are case history analyses of specific land and water management projects. Before readers review the case histories, I strongly recommend that the 16 papers comprising the first section be read. It is important that the reader grasp the concept of mitigation. L. R. Jahn’s paper is par- ticularly valuable because it contains a succinct sum- mary of the conference proceedings, and a definition BOOK REVIEWS 245 of mitigation. Jahn reports that mitigation can be defined to include: (1) “Avoiding the adverse impact altogether by not taking a certain action or parts of an action (preventative). (2) Minimizing impacts by limiting the degree or magnitude of the action and its implementation. (3) Rectifying the impact by repairing, rehabilitat- ing or restoring the affected environment. (4) Reducing or eliminating the impact over time by preservation and maintenance operations during the life of the action. (5) Compensating for the impact by replacing or providing substitute resources or environments.” The disappointing aspect of this volume is that only two papers were presented by Canadian participants. B. Stubbs and B. Markam discuss wildlife mitigative measures for oil and gas activity in Alberta, and K. G. Peterson reviews benefit-cost analysis techniques as a basis for compensation and mitigating decisions for hydroelectric projects in British Columbia. Given the vast array of development projects of all sizes now underway or planned for in Canada, it would be extremely useful to conduct a similar symposium foc- using on Canadian issues. The text is supplemented with black and white pho- tographs, maps, and black ink sketches. Most of the technical papers contain a literature cited section. I recommend this volume to all professionals con- cerned with resource management and interested in mitigation. PAUL A. GRAY Northwest Territories Wildlife Service, Yellowknife, Northwest Territories X1A 2L9 Natural Heritage: Classification Inventory, and Information By A. E. Radford, D. Kay, S. Otte, L. Otte, J. R. Massey, P. D. Whitson, and contributors. University of North Carolina Press, Chapel Hill. 485 pp. U.S. $25.00. The need for a standard system to inventory areas of natural significance is well recognized. Many types of inventories: have been proposed, the IBP system being most notable. Radford and his colleagues have developed a system from their North Carolina expe- riences and present their methodology and its funda- mentals in the present volume. As noted by the authors, the primary purpose of the book is for “.. diversity inventory use in natural areas by conser- vationists, heritage workers, and students of flora and fauna and for habitat analysis of special sites by advanced students of systematics, ecology, and spe- cies biology. A system of classification for ecological diversity and inventory procedures was developed as a basis for effective and efficient use of resources and field time. The classification system was designed as a usable and useful tool, of practical and theoretical Keleviance meet: The system described is as detailed as any produced for ecosystem inventory. Seven components form the basis of the inventory: Climate, Soils, Geology, Hydrology, Topography, Physiography, and Biology. Each component has seven major subcomponents: System, Subsystem, Class, Subclass, Generitype, Type, and Population type (Site type). Learning any new system entails trial, error, study and comparison. The system proposed here has 246 reduced much of the frustration by defining the terms used and by having a very tight methodological framework which must be followed. The end product of the inventory is a Natural Area Diversity Sum- mary, which contains the Community Diversity Summaries for each of the seven components. These summaries enable different areas to be compared. The methodology presented certainly does not pro- fess to be perfect. Certain problems do exist, such as the detailed level of investigation required to perform these inventories and the need for repeated surveys at different times of the year. The botanical aspects of a natural area are heavily emphasized, due to the area of specialization of the originators. Initially the book implies that anyone with a bit of interest or training could undertake an inventory. However, that is definitely not the case as stated later THE CANADIAN FIELD-NATURALIST Vol. 97 in the book (p. 95): “basic inventory should be made by professionally trained naturalists...” Why wait so long to indicate this major point? The book proposes to aid management of natural areas. However, it only succeeds in providing a great wealth of information on how to conduct baseline surveys. It is simply a handbook for inventories. A great deal of further research would be necessary in order to effect proper management decisions. The book is large, has good print size and is easy to read, free of typographical errors. The only printing problem is that italicized Latin names are difficult to distinguish. ROBERT WOODS Canadian Forestry Service, Environment Canada, Ottawa, Ontario KIA 1G5 Estimation of Density from Line Transect Sampling of Biological Populations By Kenneth P. Burnham, David R. Anderson, and Jeffrey L. Laake. 1980. Wildlife Monographs 72. The Wildlife Society, Washington. 202 pp. U.S. $4.00. In order to make inferences about the abundances of populations, field researchers often use line tran- sects, recording the distances and angles at which individuals are sighted. The five parts of this valuable volume synthesize available knowledge about this technique. The first part discusses basic concepts and quantitative background. Following a reader’s guide which anticipates a diverse audience ranging from biological technicians to statisticians, methodological problems associated with experimental design and data collection are considered. An outline of the mathematical and statistical background underlying the methods leads to the second part, dealing with robust estimation methods. Topics covered include a variety of estimators and their assumptions, and tests for goodness of fit, useful in the different types of observations, together with a number of exemplar data sets. The third part examines the statistical esti- mation theory for these robust methods, including MISCELLANEOUS maximum likelihood, generalized linear models, rela- tive efficiencies of estimators, and results from compu- ter simulations. The final two parts include comments on other estimators and appendices of supplementary information and examples. Different aspects of the book will appeal to different users. The authors have prepared their material with this in mind by attending to both biological and statistical issues. The emphasis on methods robust to the failure of underlying assumptions makes good sense. The extensive exam- ples, illustrations, and literature citations are all wel- come features. Overall, this work succeeds in achiev- ing the stated objective of providing “a rigorous, comprehensive, and practical reference on line tran- sects”. As such, it isa worthwhile addition to a mono- graph series with a tradition of authoritative and yet inexpensive volumes. PATRICK COLGAN Department of Biology, Queen’s University, Kingston, Ontario K7L 3N6 A Celebration of Birds — The Life and Art of Louis Agassiz Fuertes By Robert McCracken Peck. 1982. Walker and Co., New York. xi+ 178 pp., illus. U.S. $30.00. In the autumn of 1982 the Academy of Natural Sciences, Philadelphia presented a travelling exhibi- tion of the art of Louis Agassiz Fuertes (1874-1927), showing many of his works for the first time. A Cele- bration of Birds, written by the Academy’s Vice- President, was planned to accompany this exhibition, which will be shown in numerous cities across North America during 1983-84. Peck, who is both an art historian and a naturalist presents a well researched, readable, fascinating account, focusing in three separ- 1983 ate sections on Fuertes: the man, the artist, and the naturalist. Fuertes, named after the Swiss-born Harvard zool- ogist Louis Agassiz, was the son of a civil-engineering professor at Cornell University. He was born in Ithaca, and lived all his life there. A childhood interest in animals led to his painting them in his early teens. Fuertes studied and collected birds, but never killed them just for the sake of it. His passion for birds was so overwhelming that by the time he became a student at Cornell University in 1893, it was evident that sub- jects such as philosophy and mathematics just could not compete with vertebrate zoology. Whilst still a university student Fuertes met two people who greatly influenced his life. A fortuitous coincidence led to his meeting Elliott Coues, the uncle of one of his friends from the university “Glee Club” and one of the foremost American ornithologists of the time. This chance encounter led to an exhibition of his work at the annual meeting of the American Orni- thologists’ Union (A.O.U.) in 1895. The following year Fuertes attended an A.O.U. meeting for the first time, and met many of the leading ornithologists of the late 19th century. At the same A.O.U. meeting Fuertes met the pain- ter Abbott Thayer, who proved to be another major influence on his life and art. Thayer, who was known for his controversial theories of protective coloration in animals suggested that Fuertes join him the follow- ing summer for a training session. The two became friends and Fuertes subsequently spent many happy days with Thayer and his family. Fuertes was a good field ornithologist and took advantage of all opportunities to escape from his stu- dio and spend time out of doors, near his home in Ithaca, with Thayer in New Hampshire, and to partic- ipate in expeditions to various parts of the world. His first extended journey was to Alaska in 1899 as a member of the now famous Harriman Alaska expedi- tion. Others quickly followed: Texas in 1901, the Bahamas in 1902, the western states in 1903, the Mag- dalen Islands, Quebec, in 1909, the Yucatan in 1910, and Colombia in 1911 and 1913. During World War One, money was scarce for expeditions and natural history work in general. This was the only period of his life when Fuertes was not BOOK REVIEWS 247 inundated with commissions for bird paintings and illustrations. The need to feed his growing family, plus a request from the National Geographic Society, induced Fuertes to branch out and begin to paint mammals. While his series of mammal plates for National Geographic Magazine were successful, Fuertes’ first love remained bird art and he returned to this as soon as possible. With the end of the war commissions for bird art increased again, and soon Fuertes was busier than ever. In the early 1920s Fuertes was widely sought after as lecturer by various natural history and conservation societies. In 1922 his association with Cornell was formalized and he became “resident lecturer” at his Alma Mater. His demanding schedule prevented him venturing far afield for a while and he had to decline the offer to accompany an expedition of the American Museum of Natural History to Peru in 1923. Three years later the Field Museum (Chicago) mounted a large expedition to Abessynia. This was an opportun- ity Fuertes was not willing to forego. The eight month trip was to be his last expedition. In August 1927 Fuertes was killed in an automobile accident near his home. A Celebration of Birds is also a celebration of the genius of Fuertes. Practically every page has either a black-and-white or a colour illustration. These dem- onstrate Fuertes’ progress as artist-naturalist. Most are so breathtakingly beautiful that they tend to divert the reader’s attention from Peck’s interesting, infor- mative text. Peck made skillful use of primary mate- rial, consisting of the letters and journals of the artist. The book contains a useful chronology, and a section on the major collections of Fuertes’ work, in addition to a list of the illustrations in the book. The “Notes” and “Selected Bibliography” are well arranged and useful, not only for references, but also for further research. This is a superbly illustrated, well written, reasonably priced book with one major fault. It 1s too short! MARIANNE GOSZTONYI AINLEY History and Philosophy of Science Program, McGill Uni- versity, Montréal, Québec H3A 1G5 Aleksander Tamsalu, 1891-1960: a Botanist in Exile By John B. Lord. 1980. Technical Bulletin No. 11. Royal Botanical Garden, Hamilton. 127 pp., illus. $3.50 plus $1 postage. Aleksander Tamsalu was an Estonian botanist and agronomist who is still recognized in his native land as an able and productive student of plant communities. His professional career, as described by John Lord in this brief and poignant but thorough and balanced study, was in part frustrated by his inability to recog- nize or properly manage some of the opportunities 248 THE CANADIAN FIELD-NATURALIST which came to him and also by events outside his control which forced the abandonment of some major projects. Son of a successful farmer with some English ances- try (the family surname was originally Tomson), Tamsalu early evinced great interest in natural his- tory. His father refused to encourage the lad’s aca- demic bent, so Tamsalu was channelled into voca- tional pursuits, becoming for a time an elementary school teacher. Deciding in his early twenties on a career in plant science, and barred from a conven- tional university career because of his lack of a matur- ity certificate, he worked his way through the Institute of Agriculture in St. Petersburg. He completed both the high school and college level programs in his nine years there, finishing with the equivalent of a Master’s degree in 1921. This accomplishment was achieved despite the loss of his eldest son and the trauma of the Russian Revo- lution, during which he rose froma minor clerkship in the city’s Central Railroad Office to the post of Rail- road Commissar for the Petrograd Region. His suc- cess was all the more remarkable in view of his loath- ing of Bolshevism, which at one point led to his arrest, brief detention, and a spirited discussion of political philosophy with Lenin himself. Declining a teaching post with the Institute, Tamsalu returned to Estonia, which had secured its independence in 1918. In his haste to leave, he neglected to retrieve his hard-won maturity certificate from the files of the Institute, an oversight that was to hamper him for years. During the 1920’s and early 1930’s, Tamsalu was briefly assistant director of an agricultural station and a teacher of agronomy at an agricultural school. Per- sonality clashes with colleagues were partly to blame for his loss of both positions. His unorthodox research techniques, which stressed field work with little time given to the literature and minimal publish- ing, won him little support. Lord points out that Tam- salu was a mere amateur in purely botanical research, his formal preparation having been in field crop production. By 1931, at the age of 40, Tamsalu was seemingly at a dead end in his chosen field. Virtually no one would read his research papers, and he was compelled to fall back on schoolteaching to keep his family fed. He then met Theodor Lippmaa, a near contemporary, who was professor of botany at the University of Tartu and a pioneering phytosociologist. Tamsalu was unable to begin doctoral work under Lippmaa because he could not secure a copy of his maturity certificate. Not until 1940 would this matter be resolved. He did, however, begin doing field work under Lippmaa’s close supervi- sion and guidance. He contributed with others to a vegetation map of Estonia and became Lippmaa’s most trusted and productive associate. Vol. 97 In the late 1930’s, fortune seemed to shine on Tam- salu. He was invited to teach in Finland at twice his Estonian salary, but elected to stay with Lippmaa. In the spring of 1940, Lippmaa elected to move on to other projects, and Tamsalu was placed in charge of the vegetation survey project. His appointment was contingent upon his securing his doctorate, but the way finally seemed clear, since the requisite creden- tials had finally arrived from Leningrad. Then the roof fell in. The Russians seized Estonia in the summer of 1940, and Tamsalu’s hard-won post was swept away. Tamsula, his wife and daughter were deported to a German labor camp, where his wife died. Tamsalu and his daughter ultimately reached the American zone of occupation at war’s end. He and his daughter lived and worked in a succession of Dis- placed Persons camps until she was able to emigrate to Canada in 1948. Tamsalu himself arrived in Connecti- cut in 1949. His years there were not happy. He had several jobs as a manual laborer and spent most of his free time carrying out field research projects which he had set for himself. He hoped to impress some prospective employer with his talents as a botanical researcher. He was never at home with English, however, and most of those asked to review his manuscripts admired his determination and the scope of his work while finding fault with the manner in which he carried it out. Virtually no one had heard of Lippmaa, and Tamsalu could produce no credentials. Tamsalu’s research methods were unfamiliar or, at best, dated to Ameri- can colleagues, and he often seemed to be unaware of the relevant literature in his field. His foreign national- ity precluded his being considered for any posts with the federal government, something Tamsalu could not understand. Growing desperate, and aware that his advancing age might mitigate against securing a scien- tific post in America, Tamsalu accepted the only offer he did have, that of skilled laborer at the Royal Botan- ical Gardens in Hamilton, Ontario, with the under- standing that full-time work as a researcher would be his if budgetary considerations made it possible. After a bit more than a year at the RBG, Tamsalu’s demon- strated abilities as researcher and collector finally won him a full-time post as an ecological plant researcher. He had undertaken herculean tasks for a man of his age, however, and the cumulative strains upon his reserves of strength and health were too much. A variety of ailments rapidly took their toll. Tamsalu was naturally disappointed that none of his American or Canadian manuscripts were ever published. As Lord points out, Tamsalu was never a theoreti- cian of botany, but rather an indefatigable describer of plant communities. He faithfully followed the Lippmaa method until the day he died, failing to appreciate the fact that his mentor’s theories had, 1983 BOOK REVIEWS 249 while “greeted with polite curiosity ... certainly never gained widespread attention in North America.” Yet Tamsalu did leave behind a legacy of accomplishment. The RBG Herbarium contains nearly 10000 speci- mens collected by Tamsalu, who “remains by far its most prolific contributor.” When in 1965 a text onthe distribution of vegetation in Estonia was finally pub- lished, Tamsalu was acknowledged as the chief con- NEW TITLES Zoology *The amphibians and reptiles of Manitoba. 1982. By Wil- liam B. Preston. Manitoba Museum of Man and Nature, Winnipeg. 128 pp., illus. $9.95 plus $1 shipping. The animals’ who’s who. 1982. By Ruthven Tremain. Scribner, New York. xvi+ 336 pp., illus. U.S. $17.95. *The barn owl. 1982. By D. S. Bunn, A. B. Warburton, and R. D. S. Wilson. Buteo Books, Vermillion, South Dakota. 280 pp., illus. U.S. $32.50. Birds of prey. 1982. By Emma Ford. Batsford, Newton Centre, Massachusetts. 64 pp., illus. U.S. $8.95. {Breeding birds of Ontario, nidiology and distribution, volume 1: non passerines. 1983. By George K. Peck and Ross D. James. Royal Ontario Museum, Toronto. xii + 321 pp., illus. $25. +Crabs and their relatives of British Columbia. 1982. By Josephine F. L. Hart. Handbook 40. British Columbia Pro- vincial Museum, Victoria. 267 pp., illus. $5. *Eastern chipmunks: secrets of their solitary lives. 1982. By L. Wishner. Smithsonian Institute Press, Washington. 144 pp., illus. U.S. $17.50. The ecology of whales and dolphins. 1982. By David E. Gaskin. Heinemann, Exeter, New Hampshire. x11 + 460 pp., illus. U.S. $45. Elephants. 1982. By S.K. Eltringham. Blandford, New York. x + 262 pp., illus. U.S. $15.99. +The Empidoidea (Diptera) of Fennoscandia and Denmark, II: general part, the families Hybotidae, Atelestidae, and Microphoridae. 1983. By M. Chvala. Fauna Entomologica Scandinavica Volume 12. Scandinavian Science Press, Klampenborg, Denmark. 279 pp., illus. 200 Dkr. Ethology: its nature and relations with other sciences. 1982. By Robert A. Hinde. Oxford University Press, New York. 320 pp., illus. U.S. $19.95. Fishes of the Minnesota region. 1982. By Gary L. Phillips, William D. Schmid, and James C. Underhill. University of tributor of vegetation maps to the project. Lord’s text concludes with a detailed explication of the Lippmaa method, extracted from Tamsalu’s letters and unpub- lished papers. KEIR B. STERLING Pace University, Pleasantville, New York, USA 10570 Minnesota Press, Minneapolis. x + 248 pp., illus. Cloth U.S. $25: paper U.S. $12.95. +Good-bye bugs! A practical guide to coping with insects in the great outdoors. 1983. By Allen West and Bev Small- man. Grosvenor House Press, Toronto. 144 pp., illus. $8.95. The great whale book. 1981. By John Kelly, Scott Mercer, and Steve Wolf. Acropolis, Washington. 116 pp., illus. U.S. $7.95. Hibernation and torpor in mammals and birds. 1982. By Charles P. Lyman, John S. Willis, Andrés Malan, and Law- rence C. H. Wang. Academic Press, New York. xii + 320 pp., illus. U.S. $37.50. IUCN amphibian — reptile red data book, part 1 (Testu- dines, Crocodylia, Rhynchocephalia). 1982. By the Inter- national Union for Conservation of Nature and Natural Resources. Unipub, New York. 480 pp. U.S. $20. IUCN mammal red data book, part 1 (new world). 1982. By the International Union for Conservation of Nature and Natural Resources. Unipub, New York. 560 pp. U.S. $20. Jellyfish and other sea creatures. 1982. By Oxford Scientific Films. Putman’s, New York. 32 pp., illus. U.S. $8.95. tJust bats. 1983. By M. Brock Fenton. University of Toronto Press, Toronto. x + 165 pp., illus. Cloth $25; paper $9.95. Lords of the Arctic: a journey among the polar bears. 1982. By Richard C. Davids. Macmillan, New York. xx + 140 pp., illus. U.S. $29.95. +Make prayers to the raven: a Koyukon view of the northern forest. 1983. By Richard K. Nelson. University of Chicago Press, Chicago. xvi+ 292 pp., illus. U.S. $25. Mammals of the southwest. 1982. By E. Lendell Cockrum. University of Arizona Press, Tucson. 176 pp., illus. Cloth U.S. $11.95; Paper U.S. $5.95. *Marine birds and mammals of Puget Sound. 1982. By A. Angell and K. C. Balcomb, III. University of Washington Press, Seattle. xiii + 145 pp., illus. U.S. $14.50. 250 Marine fish larvae: morphology, ecology, and relation to fisheries. 1982. Edited by Reuben Lasker. Washington Sea Grant Publications, Seattle. 132 pp., illus. U.S. $8.50. Mountain in the clouds: a search for the wild salmon. 1982. By Bruce Brown. Simon and Schuster, New York. 239 pp. WESSS12295: +Nature’s night life. 1982. By Robert Burton. Blandford Press (Canadian distributor Oak Tree Press, Toronto). 160 pp., illus. $22.95. +Les noms frangais des oiseaux d’Amérique du Nord. 1983. Par Henri Ouellet et Michel Gosselin. Syllogeus No. 43. National Museum of Natural Sciences, Ottawa. 36 pp. Free. North America tortoises: conservation and ecology. 1982. Edited by R. Bruce Bury. Fish and Wildlife Service, Washington. vilit+ 126 pp., illus. *The northern Yellowstone elk: ecology and management. 1982. By Douglas B. Houston. Macmillan, New York. xix + 474 pp., illus. U.S. $48. +Once ariver: bird life and habitat changes on the middle Gila. 1983. By Amadeo M. Rea. University of Arizona Press, Tucson. Illus. U.S. $24.50. Report of IUCN workshop on marine mammals/fishery interactions. 1982. Edited by the International Union for Conservation of Nature and Natural Resources. Unipub, New York. 68 pp. U.S. $7. *Reptiles of North America: a guide to field identification. 1982. By Hobart M. Smith and Edmund D. Brodie. Golden Press (Canadian distributor Whitman Golden, Cambridge, Ontario). 240 pp., illus. Cloth U.S. $9.95; paper U.S. $6.95. +Saskatchewan cougar: elusive cat. 1982. By Tom White. Special Publication Number 14. Saskatchewan Natural His- tory Society, Regina. 80 pp., illus. Seals and man: a study of interactions. 1982. By W. Nigel Bonner. Washington Sea Grant Publications, Seattle. 170 pp., illus. U.S. $9.95. +The semiaquatic bugs (Hemiptera, Gerromorpha): phylo- geny, adaptations, biogeography, and classification. 1982. By N. Moller Andersen. Scandinavian Science Press, Klam- penborg, Denmark. 455 pp., illus. U.S. $38. The shelduck: a study in behavioural ecology. 1982. BylI. J. Patterson. Cambridge University Press, New York. x + 276 pp., illus. U.S. $49.50. +Studies on amphipod crustaceans of the northeastern Pacific region. 1982. 1, Family Ampeliscidae, genus Ampelisca. By J. J. Dickinson. 2, Family Ampithoidae. By K. E. Conlan and E. L. Bousfield. 3, Family Aoridae. By K. E. Conlanand E. L. Bousfield. 4, Family Lysianassidae, genus Hippo- medon. Publications in Biological Oceanography, No. 10. THE CANADIAN FIELD-NATURALIST Vol. 97 National Museum of Natural Sciences, Ottawa. 128 pp., illus. Free. Study and management of large mammals. 1982. By Thane Riney. Wiley-Interscience, New York. x + 552 pp., illus. WES2 959,95: +The systematics and distributional ecology of the super fam- ily Ampeliscoidea (Amphipoda: Gammaridea) in the north- eastern Pacific Region, II: the genera Byblis and Haploops. 1983. By John J. Dickinson. Publications in Natural Science 1. National Museum of Natural Sciences, Ottawa. 38 pp., illus. Free. +Waterfowl ona pacific estuary. 1982. By Barry Leach. Spe- cial Publication No. 5. British Columbia Provincial Museum, Victoria. 211! pp., illus. $7. +The wolf in the southwest: the making of an endangered species. 1983. Edited by David E. Brown. University of Arizona Press, Tucson. Cloth U.S. $19.95; paper U.S. $9.95. Woodpeckers of the world. 1982. By Lester L. Short. Del- aware Museum of Natural History, Greenville. xvill + 676 pp., illus. U.S. $99.95. Botany Agaves of continental North America. 1982. By Howard Scott Gentry. University of Arizona Press, Tucson. xiv + 670 pp., illus. U.S. $49.50. +Alaska wild berry guide and cookbook. 1983. Edited by Alaska Magazine. Alaska Northwest Publishing, Anchor- age. 200 pp., illus. U.S. $16.95 plus $1 postage. Botanical exploration of southern Africa: an illustrated his- tory of early botanical literature. 1981. By Mary Gunn and L. E. Codd. Balkema, Cape Town. xvi+ 400 pp., illus. U.S. $56. Edible horticultural crops: a compendium of information on fruit, vegetable, spice, and nut species. 1982. By C. Hackett and J. Carolane. Academic Press, New York. 720 pp. U.S. $59.50. A field guide to mushrooms and their relatives. 1982. By Booth Courtenay and Harold H. Burdsall. Van Nostrand Reinhold, New York. 144 pp., illus. U.S. $18.95. Fluoride emissions: their monitoring and effects on vegeta- tion and ecosystems. 1982. Edited by Frank Murray. Aca- demic Press, New York. 250 pp. U.S. $29.50. Green planet: the story of plant life on earth. 1982. Edited by David M. Moore. Cambridge University Press, New York. 288 pp., illus. U.S. $27.50. The Larousse guide to house plants. 1982. By Alan Titch- marsh. Larousse, New York. 287 pp., illus. U.S. $9.95. 1983 Mires: swamp, bog, fen, and moor: part A — analytical studies and part B — descriptive studies. 1983. Edited by A. J. P. Gore. Elsevier Scientific, New York. xii + 438 pp. and xii + 452 pp. U.S. $161.75 each. *Moss flora of the maritime provinces. 1982. By Robert R. Ireland. National museum of Natural Sciences Publications in Botany No. 13. McClelland and Stewart, Toronto. 738 pp., illus. $20. Photosynthesis: volume 1, energy conversion by plants and bacteria; and volume 2, development, carbon metabolism, and plant productivity. 1982, 1983. Edited by Govindjee. Academic Press, New York. 832 pp., and 592 pp. U.S. $79 and U.S. $59 or set U.S. $116. Plants and people: vegetation change in North America. 1982. By Thomas R. Vale. Association of American Geo- graphers, Washington. vill + 88 pp., illus. U.S. $5. Prairie grasses: identified and described by vegetative cha- racters. 1983. By Jan Looman. Supply and Services Canada, Ottawa. illus. $9.95 in Canada; $11.95 elsewhere. +Some common plants of the sub-boreal spruce zone. 1982. By J. Pojar, R. Love, D. Meidinger, and R. Scagel. Land Management Handbook No. 6. British Columbia Ministry of Forests, Victoria. 102 pp., illus. Limited copies available from author. Free. VNR color dictionary of mushrooms. 1982. Edited by Colin Dickinson and John Lucas. Van Nostrand Reinhold, New York. 160 pp., illus. U.S. $12.95. Environment Acid rain: a plague upon the waters. 1982. By Robert Ost- mann, Jr. Dillon, Minneapolis. 208 pp., illus. U.S. $12.95. Acid rain: the devastating impact on North America. 1982. By Ross Howard and Michael Perley. McGraw-Hill, New York. 206 pp., illus. U.S. $6.95. Air pollution: assessment methodology and modeling. 1982. Plenum, New York. xvii + 330 pp., illus. U.S. $39.50. *Biogeographical dictionary of Rocky Mountain naturalists: a guide to the writings and collections of botanists, zoolo- gists, artists, and photographers, 1682-1932. 1981. By Joseph and Nesta Dunn Ewan. Bohn, Scheltema, and Hol- kema (North American distributor Kluwer, Hingham, Mas- sachusetts). U.S. $37.50. Carbon dioxide: emissions and effects. 1982. By Irene M. Smith. [IEA Coal Research, London, England. 132 pp., illus. £20. The coralreef. 1981. By Alan Emery. Canadian Broadcast- ing Corporation, Toronto. 112 pp., illus. $14.95. Crucible for conservation: the creation of Grand Teton National Park. 1982. By Robert W. Righter. Colorado BooK REVIEWS 251 Associated University Press, Boulder. ix + 192 pp., illus. Cloth U.S. $12.50; paper U.S. $5.95. Dammed Indians: the Pick — Sloan Plan and the Missouri River Sioux, 1944-1980. 1982. By Michael L. Lawson. Uni- versity of Oklahoma Press, Norman. xxvi+ 261! pp., illus. U.S. $19.95. Ecology of northern lowland bogs and conifer forests. 1982. By James A. Larsen. Academic Press, New York. 320 pp. U.S. $34. +The economy of nature: a textbook in basic ecology. 1983. By Robert E. Ricklefs. 2nd edition. Chiron Press, Concord, Massachusetts. 510 pp., illus. U.S. $24.95. Environmental planning guidelines for strategies and plans. 1982. By the International Union for Conservation of Nature and Natural Resources. Unipub, New York. 88 pp. (English/ French/ Spanish). U.S. $8. Estuaries and enclosed seas. 1983. Edited by B. H. Ket- chum. Elsevier Scientific, New York. xi1+500 pp. U.S. $170.25. Evolution of the flora and fauna of arid Australia. 1982. Edited by W. R. Barker and P. J. M. Greenslade. Papers from a symposium, Adelaide, 1980. Peacock, Frewvill, South Australia. viii + 392 pp., illus. A $32.95. Hazardous waste in America. 1982. By Samuel S. Epstein, Lester O. Brown, and Carl Pope. Sierra Club, San Francisco. xi + 593 pp. U.S. $27.50. Impact of marine pollution onsociety. 1982. Edited by Vir- ginia K. Tippie and Dana R. Kester. Praeger, New York. vit 314 pp., illus. U.S. $27.95. The importance and values of wild plants and animals in Africa. 1983. By J. B. Sale. Unipub, New York. 44 pp., illus. U.S. $10. Managing the ocean resources of the United States. 1982. By D. P. Finn. Springer-Verlag, New York. 193 pp., illus. U.S. $16. The marine flora and fauna of Hong Kong and southern China. 1982. Edited by Brian Morton and C. K. Tseng. Proceedings of a workshop, Hong Kong, April, 1980. Hong K ong University Press, Hong Kong. xvi + 934 pp., illus. U.S. $228.50. The world’s greatest natural areas: an indicative inventory of natural sites of world heritage quality. 1982. By the Interna- tional Union for Conservation of Nature and Natural Resources. Unipub, New York. 77 pp., illus. + maps. U.S. $10. Miscellaneous Abusing science: the case against creationism. 1982. By Philip Kitcher. MIT Press, Cambridge, Massachusetts. eae ALB} foyoy, WES a adlle). Day Astronomy handbook. 1982. By James Muirden. Arco, New York. 189 pp., illus. U.S. $8.95. The Cambridge photographic atlas of the planets. 1982. By Geoffrey Briggs and Fredric Taylor. Cambridge University Press, New York. 255 pp., illus. U.S. $24.95. *Canadian Arctic recollections: Baffin Island 1923-1931. 1981. By J. D. Soper. Mawdsley Memoir 4. University of Saskatchewan, Saskatoon. 141 pp. + map. $20. Constituents and properties of soils. 1982. By M. Bonneau and B. Souchier. Translation editor V. C. Farmer. Academic Press, New York. 520 pp. U.S. $67. Darwinism defended: a guide to the evolution controversies. 1982. By Michael Ruse. Addison-Wesley, Reading, Massa- chusetts. xvill + 356 pp., illus. U.S. $12.50. Evolution now: a century after Darwin. 1982. Edited by John Maynard Smith. Freeman, San Francisco. 239 pp., illus. Cloth U.S. $19.95; paper U.S. $9.95. Evolution without evidence: Charles Darwin and the origin of species. 1982. By Barry G. Gale. University of New Mex- ico Press, Albuquerque. ix + 238 pp. U.S. $21.95. *Mountain islands and desert seas: a natural history of the U.S.-Mexican borderlands. 1982. By F. R. Gehlback. Texas A+M University Press, College Station. xvi + 298 pp., illus. U.S. $19.95. The nature and origin of the biological world. 1982. By E. J. Ambrose. Hérwood, Chichester, England and Halsted (Wiley), New York. 190 pp., illus. U.S. $49.95. Physiological bases of phytoplankton ecology. 1981. Edited by Trevor Platt. Canadian Bulletin of Fisheries and Aquatic Sciences 210. Supply and Services Canada, Hull. x + 346 pp., illus. $17.95 in Canada; $21.95 elsewhere. The study of biology. 1982. By Jeffrey J. W. Baker and Garland E. Allen. 4th edition. Addison-Wesley, Reading, Massachusetts. xvi+ 1022 pp., illus. U.S. $26.95. Temperate deserts and semi-deserts. 1983. Edited by Neil E. West. Elsevier Scientific, New York. xii+ 522 pp. U.S. $170.25. Tropical rain forest ecosystems, part A: structure and func- tion. 1983. Edited by F. B. Golley, H. Lieth, and M. J. A. Werger. Elsevier Scientific, New York. xii + 382 pp. U.S. SMS: Ecology of coastal waters: a systems approach. 1982. By R.K. Mann. University of California Press, Berkeley. x + 322 pp., illus. Cloth U.S. $36; paper U.S. $18. Water: the life sustaining resource. 1982. By Robert Gardner. Messner, New York. 223 pp., illus. U.S. $9.29. THE CANADIAN FIELD-NATURALIST Vol. 97 The wilderness war: the struggle to preserve America’s wild- lands. 1982. By Edward B. Weinstock. Messner, New York. 191 pp., illus. U.S. $9.29. *Joseph Le Conte: gentle prophet of evolution. 1982. By Lester D. Stephens. Louisiana State University, Baton Rouge. xix + 340 pp. U.S. $22.50. +Journey Skagway- Whitehorse: a travellers’ guide to the land and its people. undated (1982). By Susan Ellenton and the Yukon Conservation Society. Yukon Conservation Society, Whitehorse. 18 pp., illus. $5. *Lucy Audubon. 1982. By Carolyn Delatte. Louisiana State University Press, Baton Rouge. xili+ 148 pp., illus. U.S. $15.95. Oceanography: the present and the future. 1983. Edited by P. G. Brewer. Springer-Verlag, New York. 392 pp., illus. U.S. $39.80. +Recherches sur la baie @’ Hudson et la baie James/scientific studies on Hudson and James Bays. 1982. General Editor, I. P. Martini. Proceeding of a symposium at the University of Guelph, 28-30 April, 1981. Volume 109 No. 3 and 4, Le Naturaliste Canadien. Les Presses de l'Université Laval, Québec. 1019 pp + indices, illus. $28. +Recreational land use: perspectives on its evolution in Canada. 1982. Edited by G. Wall and J. Marsh. Carleton University Press (distributed by Oxford University Press, Don Mills). 436 pp., illus. $14.95. Skyguide: a field guide for amateur astronomers. 1982. By Mark R. Chartrand. Golden Press, New York. 280 pp., illus. Cloth U.S. $9.95; paper U.S. $6.95. +Survey of invertebrate zoologists in Canada — 1982. 1983. By Chang-tai Shih and Diana R. Laubitz. Syllogeus No. 42. National Museum of Natural Sciences, Ottawa. 93 pp. Free. Books for Young Naturalists The adventures of Charles Darwin. 1982. By Peter Ward. Cambridge University Press, New York. 96 pp., illus. U.S. $7.95. Biological clocks. 1982. By Sarah R. Riedman. Crowell, New York. 120 pp., illus. U.S. $10.50. Cactus. 1982. By Cynthis Overbeck. Lerner, Minneapolis. 48 pp., illus. U.S. $8.95. Dinosaurs. 1982. By David Lambert. Watts, New York. 32 pp., illus. U.S. $7.90. Elephants. 1982. By Elsa Posell. Childrens Press, Chicago. 45 pp., illus. U.S. $6.95. Inside an egg. 1982. By Sylvia A. Johnson. Lerner, Min- neapolis. 48 pp., illus. U.S. $7.95. 1983 Living fossils. 1982. By Howard E. Smith, Jr. Dodd, Mead, New York. 79 pp., illus. U.S. $8.95. Mushrooms. 1982. By Sylvia A. Johnson. Lerner, Minnea- polis. 48 pp., illus. U.S. $8.95. Oceans. 1982. By Katharine Jones Carter. Childrens Press, Chicago. 48 pp., illus. U.S. $6.95. Science looks at mysterious monsters. 1982. By I homas G. Aylesworth. Messner, New York. 126 pp., illus. U.S. $9.29. The smallest dinosaurs. 1982. By Seymour Simon. Crown, New York. 47 pp., illus. U.S. $8.95. Snowy owls. 1982. By Patricia Hunt. Dodd, Mead, New York. 62 pp., illus. U.S. $7.95. BOOK REVIEWS 253 Swallowtail butterflies. 1982. By Jane Dallinger and Cyn- thia Overbeck. Lerner, Minneapolis. 48 pp., illus. U.S. $8.95. Vampire bats. 1982. By Laurence Pringle. Morrow, New York. 62 pp., illus. U.S. $7.50. The way of the dolphin. 1981. By Michael Fox. Acropolis, Washington. 96 pp., illus. U.S. $8.95. What makes it rain? The story ofaraindrop. 1982. By Keith Brandt. Troll, Mahwah, New Jersey. 32 pp., illus. Cloth U.S. $7.89; paper U.S. $1.95. *assigned for review tavailable for review The Ottawa Field-Naturalists’ Club Special Publications 1. Autobiography of John Macoun A reprint of the 1922 edition of the fascinating life story of one of Canada’s outstanding early naturalists, with a new introduction by Richard Glover and bibliographical essay, footnotes, and index by William A. Waiser, plus three maps of John Macoun’s western travels. Individuals Libraries $12.50 plus $2 postage and handling $15.00 plus $2 postage and handling 2. Transactions of The Ottawa Field-Naturalists’ Club and The Ottawa Naturalist — Index. Compiled by John M. Gillett A complete author, title, and subject index to the predecessors of The Canadian Field- Naturalist, the first thirty-nine volumes of the publications of The Ottawa Field-Naturalists’ Club. $25 plus $2 postage and handling Centennial Bird Record Songs of the Seasons More than fifty eastern North American birds and amphibians are presented in full stereo- phonic sound as recorded in the wild by wildlife recording expert F. Montgomery Brigham. $9.11 (postage and handling included but Ontario residents must add 7% sales tax Please send orders to: The Ottawa Field-Naturalists’ Club Box 3264 Postal Station C Ottawa, Ontario, Canada K1Y 4J5 - Instructions to Contributors Content The Canadian Field-Naturalist is a medium for the publi- cation of scientific papers by amateur and professional natur- alists or field-biologists reporting observations and results of investigations in any field of natural history provided that they are original, significant, and relevant to Canada. All readers and other potential contributors are invited to sub- mit for consideration their manuscripts meeting these crite- ria. For further information consult: A Publication Policy for the Ottawa Field-Naturalists’ Club, 1983. The Canadian Field- Naturalist 97(2): 231-234. Manuscripts Please submit, in either English or French, three complete manuscripts written in the journal style. The research reported should be original. It is recommended that authors ask qualified persons to appraise the paper before it is sub- mitted. Also authors are expected to have complied with all pertinent legislation regarding the study, disturbance, or collection of animals, plants or minerals. The place where voucher specimens have been deposited, and their catalogue numbers, should be given. Type the manuscript on standard-size paper, if possible use paper with numbered lines, double-space throughout, leave generous margins to allow for copy marking, and number each page. For Articles and Notes provide a biblio- graphic strip, an abstract and a list of key words. Generally words should not be abbreviated but use SI symbols for units of measure. Underline only words meant to appear in italics. The names of authors of scientific names should be omitted except in taxonomic manuscripts or other papers involving nomenclatural problems. Authors are encouraged to use “proper” common names (with initial letters capitalized) as long as each species is identified by its scientific name once. The names of journals in the Literature Cited should be written out in full. Unpublished reports should not be cited here but placed in the text. Next list the captions for figures (numbered in arabic numerals and typed together on a separate page) and present the tables (each titled, numbered consecutively in arabic numerals, and placed on a separate page). Mark in the margin of the text the places for the figures and tables. Extensive tabular or other supplementary material not essential to the text, typed neatly and headed by the title of the paper and the author’s name and address, should be submitted in duplicate on letter-size paper for the Editor to place in the Depository of Unpublished Data, CISTI, National Research Council of Canada, Ottawa, Canada K1A 082. A notation in the published text should state that the material is available, at a nominal charge, from the Depository. The Council of Biology Editors Style Manual, 4th edition (1978) available from the American Institute of Biological Sciences, is recommended as a guide to contributors. Web- ster’s New International Dictionary and le Grand Larousse Encyclopédique are the authorities for spelling. Illustrations— Photographs should have a glossy finish and show sharp contrasts. Photographic reproduction of line drawings, no larger than a standard page, are preferable to large originals. Prepare line drawings with India ink on good quality paper and letter (don’t type) descriptive matter. Write author’s name, title of paper, and figure number on the lower left corner or on the back of each illustration. Reviewing Policy Manuscripts submitted to The Canadian Field- Naturalist are normally sent for evaluation to an Associate Editor (who reviews it himself or asks another qualified person to do so), and at least one other reviewer, whois a specialist in the field, chosen by the Editor. Authors are encouraged to suggest names of suitable referees. Reviewers are asked to give a general appraisal of the manuscript followed by specific comments and constructive recommendations. Almost all manuscripts accepted for publication have undergone revi- sion — sometimes extensive revision and reappraisal. The Editor makes the final decision on whether a manuscript is acceptable for publication, and in so doing aims to maintain the scientific quality and overall high standards of the journal. Special Charges Authors must share in the cost of publication by paying $60 for each page in excess of six journal pages, plus $6 for each illustration (any size up toa full page), and up to $60 per page for tables (depending on size). Reproduction of color photos is extremely expensive; price quotations may be obtained from the Business Manager. When galley proofs are sent to authors, the journal will solicit on a voluntary basis acommitment, especially if grant or institutional funds are available, to pay $60 per page for all published pages. Authors may also be charged for their changes in proofs. Limited journal funds are available to help offset publica- tion charges to authors with minimal financial resources. Requests for financial assistance should be made to the Editor when the manuscript is accepted. Reprints An order form for the purchase of reprints will accompany the galley proofs sent to the authors. 254 TABLE OF CONTENTS (concluded) News and Comment Notice of the Ottawa Field-Naturalists’ Club Annual Business Meeting — The Alfred B. Kelly Memorial Fund of the Province of Quebec Society for the Protection of Birds, Inc. — Announcing a Canadian Forestry and Wildlife Management Symposium — Notice to Amend the Constitution of the Ottawa Field-Naturalists’ Club — The Ottawa Field-Naturalists’ Club Awards — Call for nominations for the Council of the Ottawa Field-Naturalists’ Club — Herpetology Position Available at the Royal Ontario Museum — Editor’s Report for 1982 — Summary of Book Review Activities: Volume 96, 1982 A Publication Policy for The Ottawa Field-Naturalists’ Club Book Reviews Zoology: The mammals of Minnesota — Mammals of the National Parks — The Fisher: life history, ecology, and behavior — A bird-finding guide to Ontario — The amphibians and reptiles of Manitoba — Aquatic insects and oligochaetes of North and South Carolina — Falcons of the world Botany: Survey of Canadian herbaria — The rare vascular plants of the Yukon — Weeds Environment: Nature conservation day — The Mitigation Symposium: a national workshop on mitigating losses of fish and wildlife habitats — Natural heritage: classification inventory, and information — Estimation of density from line transect sampling of biological populations Miscellaneous: A celebration of birds: the life and art of Louis Agassiz Fuertes — Aleksander Tamsalu, 1891-1960: a botanist in exile New Titles Instructions to Contributors Mailing date of previous issue: volume 97, number |: 16 September 1983. 227 23 235 240 243 246 249 254 THE CANADIAN FIELD-NATURALIST Volume 97, Number 2 1983 Articles Movements of collared Caribou, Rangifer tarandus, in relation to petroleum development on the Arctic Slope of Alaska KENNETH R. WHITTEN and RAYMOND D. CAMERON 143 An estimate of the Black Scoter, Melanitta nigra, population moulting in James and Hudson bays R. KENYON ROSS 147 Spatial trends in Canadian Snowshoe Hare, Lepus americanus, population cycles CHARLES H. SMITH SII Naturalization of the Rainbow Smelt, Osmerus mordax, in Lake Simcoe, Ontario HUGH R. MacCRIMMON, ROBERT W. PUGSLEY, and BARRA L. GOTS 161 Range extensions of vascular plants from the northern Yukon Territory LES C. CWYNAR 170 Habitat segregation by stream darters (Pisces: Percidae) in the Thames River watershed of southwestern Ontario JOHN ENGLERT and BENONI H. SEGHERS a Aquatic plants in Lake Superior Provincial Park in relation to water chemistry D. FRASER and J. K. MORTON 181 Avian breeding and occurrence notes from the Sutton Ridges area of northeastern Ontario R. D. JAMES, J. A. Dick, S. V. NASH, M. K. PECK, and B. E. TOMLINSON 187 A patterned fen on the north shore of Lake Superior, Minnesota PAUL H. GLASER 194 Habitat usage of two woodland Buteo species in Central Ontario EDWARD ARMSTRONG and DAVID EULER 200 Dependence of Clark’s Nutcracker, Nucifraga columbiana, on conifer seeds during the postfledging period STEPHEN B. VANDER WALL and HARRY E. HUTCHINS 208 Notes White Malaxis, Malaxis monophyllos var. diphyllos, an addition to the orchids of Canada from the Queen Charlotte Islands, British Columbia WILLIAM J. BEESE Dis Extensive overland movement of Pintail, Anas acuta, brood and attempted predation by hawks DAVID C. DUNCAN 216 Predation by fish and Common Mergansers on darters (Pisces: Percidae) in the Thames River watershed of southwestern Ontario JOHN ENGLER and BENONI H. SEGHERS 218 Water Meal, Wolffia arrhiza, (Lemnaceae) in Saskatchewan J. LOOMAN 220 ‘Facing in’ is not general to all gulls nesting on cliffs EDWARD H. Burtt, JR and WILLIAM CHOW 22D A River Otter, Lutra c. canadensis, of record size JAMES D. LAZELL, JR. 225 concluded on inside back cover ISSN 0008-3550 The CANADIAN «= nv FIELD-NATURALIS®. Published by THE OTTAWA FIELD-NATURALISTS’ CLUB, Ottawa, Canada July-September 1983 Volume 97, Number 3 The Ottawa Field-Naturalists’ Club FOUNDED IN 1879 Patrons Their Excellencies the Governor General and Mrs. Edward Schreyer 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 information on these fields as widely as possible; to support and cooperate with organizations engaged in preserving, maintaining or restoring environments of high quality for living things. Honorary Members William J. Cody Clarence Frankton Thomas H. Manning Loris S. Russell Mary E. Stuart William G. Dore W. Earl Godfrey George H. McGee Douglas B. O. Savile Sheila Thomson R. Yorke Edwards Louise de K. Lawrence Hugh M. Raup Pauline Snure 1983 Council President: D. F. Brunton W.R. Arthurs E. M. Dickson B. M. Marwood Vice-President: P. M. Catling R. E. Bedford S. Gawn E. G. Munroe Vice-President: J. K. Strang B. A. Campbell G. G. Gruchy K. W. Taylor Recording Secretary: E. F. Pope W. J. Cody G. M. Hamre R. Taylor Corresponding Secretary: W. K. Gummer F. R. Cook D. R. Laubitz P. S. Walker Treasurer: P. D. M. Ward S. J. Darbyshire P. M. D. Martin Those wishing to communicate with the Club should address correspondence to: The Ottawa Field-Naturalists’ Club, Box 3264, Postal Station C, Ottawa,Canada KIY 4J5. Forinformation on Club activities telephone (613) 722-3050. 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. Editor: Francis R. Cook, Herpetology Section, National Museum of Natural Sciences, National Museums of Canada, Ottawa, Ontario KIA 0M8 Copy Editor: Louis L’Arrivée; Assistant to the Editor: Barbara Stewart Business Manager: Mr. W. J. Cody, Box 3264, Postal Station C, Ottawa, Ontario, Canada KIY 4J5 Book Review Editor: Dr. J. Wilson Eedy, R. R. 1, Moffat, Ontario LOP 1J0 Coordinator, The Biological Flora of Canada: Dr. George H. La Roi, Department of Botany, University of Alberta, Edmonton, Alberta T6G 2E9 Associate Editors: C. D. Bird A. J. Erskine D. E. McAllister Stephen M. Smith E. L. Bousfield Charles Jonkel W. O. Pruitt, Jr. C. G. Van Zyll de Jong Chairman, Publications Committee: R. E. Bedford All manuscripts intended for publication should be addressed to the Editor. Urgent telephone calls may be made to the Editor’s office (613-996-1 755), or his home on evenings and weekends (613-269-3211), or to the Business Manager's office (613-996-1665). Subscriptions and Membership Subscription rates for individuals are $17 per calendar year. Libraries and other institutions may subscribe at the rate of $30 per year (volume). The Ottawa Field-Naturalists’ Club annual membership fee of $17 includes a subscription to The Canadian Field- Naturalist. Subscriptions, applications for membership, notices of changes of address, and undeliverable copies should be mailed to: The Ottawa Field-Naturalists’ Club, Box 3264, Postal Station C, Ottawa, Canada KIY 4J5. Second Class Mail Registration No. 0527 — Return Postage Guaranteed. Back Numbers and Index Most back numbers of this journal and its predecessors, Transactions of The Ottawa Field- Naturalists’ Club, 1879-1886, and The Ottawa Naturalist, 1887-1919, and Transactions of The Ottawa Field-Naturalists’ Club and The Ottawa Naturalist — Index compiled by John M. Gillett, may be purchased from the Business Manager. Cover: Lynx, Lynx canadensis, photo courtesy of Ludwig N. Carbyn, Canadian Wildlife Service, Edmonton, Alberta, see article by Carbyn and Patriquin pp. 262-267. The Canadian Field-Naturalist Volume 97, Number 3 July-Septethber1983 js 6 1094 Banff National Park, Alberta JOSEPH S. NELSON Department of Zoology, The University of Alberta, Edmonton, Alberta T6G 2E9 Nelson, Joseph S. 1983. The tropical fish fauna in Cave and Basin Hotsprings drainage, Banff National Park, Alberta. Canadian Field-Naturalist 97(3): 255-261. The Cave and Basin Hotsprings drainage in Banff National Park has been the site of introduction of various tropical aquarium fishes. A collection made in May 1981 indicated that there has been a change in the species present since 1968. A review of past collections and a description of the present tropical fish fauna, Mosquitofish (Gambusia affinis), Sailfin Molly (Poecilia latipinna), and Jewelfish (Hemichromis bimaculatus), is presented. The first two poecilliid species are well established and occur in places in dense numbers; the last cichlid species is relatively rare (at least in May). The highest temperature recorded where fish were found was 30° C in Basin Spring. Mosquitofish occurred at this site. Two native species found with the tropical fishes were Longnose Dace (Rhinichthys cataractae), most commonin the outlet of one small cave at 24° C, and Brook Stickleback (Culaea inconstans) in quiet shoreline waters with aquatic plants in temperatures up to 21°C. Key Words: Mosquitofish, Gambusia affinis, Sailfin Molly, Poecilia latipinna, Jewelfish, Hemichromis bimaculatus, Longnose Dace, Rhinichthys cataractae, Brook Stickleback, Culaea inconstans, tropical fishes, Cave and Basin Hotsprings, Banff National Park. The Cave and Basin Hotsprings site in Banff National Park is historically important as part of the small area acquired in 1885 which became Canada’s first national park. Before man’s activities in the area the native fish fauna consisted of Longnose Dace, Rhinichthys cataractae Valenciennes, and perhaps Brook Stickleback, Culaea inconstans (Kirtland). Individuals of several other species native to the Bow River may have utilized the outlet area. Several spe- cies of tropical fishes have been reported from the drainage. One, the Mosquitofish, Gambusia affinis (Baird and Girard), was established by the govern- ment for mosquito control in 1924 while the others were introduced by local aquarists, at least in part as an unauthorized experiment and to serve as breeding stock for their aquaria. It is a fauna which although artificial and incongruous with a national park is established and unique in Canada. This paper reports the current status of the tropical fish fauna in the Cave and Basin drainage and, because of the uniqueness of the fauna in Canada, describes the collecting sites. Unfortunately, in the absence of detailed biological studies prior to the introductions, it will probably be impossible to know the full impact of the introduc- tions upon the system, either in introducing other exotic organisms or in causing the extirpation of indi- genous organisms. The hotsprings may contain or may have contained an endemic subspecies of Long- nose Dace and is one of the sites for a snail of very restricted distribution, the Banff Springs Physa, Physa johnsoni (Clark 1973, 1981). Methods Intensive sampling was conducted 16 and 17 May 1981, using fine mesh dipnets of various sizes. General observations were made on 17 May 1980, 18 May 1981, and 5 August 1982. Twelve collections, each consisting of at least several dipnet attempts, were preserved; many other collections were made and the contents examined for new species and then released. Care was taken to try not to permanently change the existing populations through over-collecting in the springs or habitat disturbance. The 12 preserved col- lections were made at eight general sites (Figure 1). All specimens are catalogued in the University of Alberta Museum of Zoology (UAMZ 4613-4615; 4631-4640). In fin-ray counts, the last ray is defined as consisting of two ray elements, following Hubbs and Lagler (1958); the last element is unbranched while the preceding one is branched in the three tropical fishes (in the Mosquitofish, the bases of the last two ele- ments, comprising the last ray, are distinctly separ- 255 256 THE CANADIAN FIELD-NATURALIST Bathhouse FIGURE |. Diagrammatic map showing collecting sites 1-7 (eighth at western outlet and not shown) in Cave and Basin Hotspring drainage. Water level and spring courses are made from field notes of May 1981, and are intended only to show the relative position of various features. The origin of the springs is not shown. Slight change in water level would modify the shoreline contour. Inset map shows the general loca- tion of the drainage. B.S. = Basin Spring (E.F. = East Fork; W.F. = West Fork); C.S. = Cave Spring; V1- V3 = First to Third Vermilion Lake. ated). Scale counts were taken from behind the top of | the gill cover insertion to the end of the hypural plate (the few scales on the caudal fin were excluded). Elevations in the description of the study area were taken from Site Plan, NWB 71/ R42, of the former Department of Indian Affairs and Northern Devel- opment (scale | inch = 50 feet). Temperature was measured with a hand-held laboratory thermometer and temperature and conductivity with a YSI model 33 S-C-T meter. Study Area The Cave and Basin Hotsprings drainage, consis- ting of hotsprings and a shallow pond (or marsh) draining into the Bow River, is situated about 1-1/2 km southwest of the Banff townsite, immediately south of the Bow River and at the base of Sulphur Mountain. The pond, situated at about 51° 10°20"N, 115°35°35”W, and an elevation of about 1382 m, lies within Pleistocene and Recent deposits of the Bow Valley while the springs arise from the western portion of the Sulphur Mountain thrust fault bordered by Upper Devonian formation to the west and undiffer- entiated rock to the east (Price and Mountjoy 1972). Three main thermal springs exist, Pool Springs to the southeast, Basin Spring to the southwest, and Cave Spring in the middle (only the latter two, which Vol. 97 were investigated, are shown in Figure 1). Pool Springs, described by van Everdingen (1972: 25-26), arise higher on the mountain than the other two (behind Cave Spring on the east side of the Bath- house), consist of two relatively small sulphurous springs, and enter the pond at the east end. Cave Spring arises behind the eastern portion of the Bath- house, south of the present parking area. These two springs are piped beneath the parking area (which lies at an elevation of 1400 m in this area), an expansion of the Sundance Road, emerge on the north side (Basin Spring runs a short distance through the conduit), cross a combination footpath and horse trail at an elevation of about 1390 m, continue down a steep hillside, and enter the pond through various channels (several, which are not shown in Figure |, are under- ground and enter the pond at or near pond level through small caves). Changes, some man-made, have occurred in the drainage system (e.g., differences occur in the stream course depicted in maps prepared by the former Department of Indian and Northern Affairs in 1971, W. D. Reynolds in 1976, and myself in 1981). Recent channels are sometimes apparent by a white precipitate (e.g., dotted line in Figure 1). The shoreline adjacent to the pond area generally has a soft bottom. There are two outlets to the pond which cross a path beneath small bridges and enter the Bow River. The eastern outlet has a small amount of surface flow through a beaver dam; the western outlet, located about halfway along the northern shoreline, 1s blocked off by earthfill with the pond side being almost | m higher than the river level at the time of the study (an outlet a short distance upstream on the Bow River drains a marshy area adjacent to the Cave and Basin Pond). Van Everdingen (1972: 21-25) gives a physico- chemical description of the springs. He notes that the flow in the Basin Spring is less than in the Cave Spring but its temperature and total dissolved solids is higher (a maximum of 34.5 vs. 32.8°C and 1828 vs. 1040 ppm). His data suggest that the coolest period is about May when discharge is relatively high. Pritchard (1971) noted one of the springs to vary from 30°C to 26° C during its course but to vary only 1° C (26 to 27) at one site during a period when air temperature var- ied from —20°C to 32°C. History of Fish Collections Eigenmann (1895) made the first known fish collec- tions in the area. In 1892 he collected five species from the Bow River at the Banff townsite and in Echo Creek (= Vermilion Creek, the outlet of Vermilion lakes on the other side of the Bow River from the Cave and Basin area) (Figure 1). He referred to a tributary of the Bow from the hot sulphur springs and reported that Longnose Dace, given as Rhinichthys dulcis 1983 (Girard), were very abundant (and common in the adjacent Bow River). Nichols (1916) described a new subspecies of Longnose Dace, Rhinichthys cataractae smithi, on the basis of five specimens collected in 1915 from the Cave and Basin Hotsprings having, by his counts, a relatively low number of lateral line scales. Mosquitofish, Gambusia affinis, were introduced into the Cave and Basin Hotsprings outflow in 1924 for mosquito control and Mail (1954) reported them to be thriving. Aquarists began introducing tropical fishes into the drainage at least by the autumn of 1960 when several pairs of all-black (permablack) Sailfin Molly, Poecilia latipinna (Lesueur), were the first, as far as is known, to be added (C. Yarmoloy, personal communication); this is the only known introduction of this species. There is some feeling that earlier introductions of fishes may have been made by local aquarists but this is unsubstantiated. The first documentation of the unauthorized introduction of aquarium fishes into Cave and Basin Hotsprings was made by McAllister (1969), who, on the basis of collections made in June and August of 1968, reported the presence of four species of exotic fishes in addition to the Mosquito- fish, namely: Guppy, Poecilia reticulata Peters, the all-black variety of Sailfin Molly, Poecilia latipinna (Lesueur), Green Swordtail, Xiphophorus helleri Heckel, and Convict Cichlid, Cichlasoma nigrofascia- tum (Gitinther). The swordtails introduced were of the typical red aquarium variety but they apparently reverted to the wild green color in the pond and were always rare (C. Yarmoloy, personal communication). Of about 600 individuals collected for McAllister’s (1969) study, the Guppy was about four times more numerous than Mosquitofish which in turn were about nine times more numerous than either the molly or swordtail. Only four Convict Cichlids were obtained and he noted reports of Angelfish, presuma- bly Pterophyllum scalare (Lichtenstein), being pres- ent. As far as is known, the Angelfish along with the three-spot gourami variety of the Blue Gourami, Tri- chogaster trichopterus (Pallas), and Siamese Fighting Fish, Betta splendens Regan, were introduced into waters east of Cave Spring (Pool Springs?); the Gou- rami did breed but died out when the flow was turned off (C. Yarmoloy, personal communication). No native fishes were obtained in the 1968 collections. J. C. Ward (1974. The fishes and their distribution in the mountain national parks of Canada. Manu- script Report prepared for Parks Canada by the Can- adian Wildlife Service, Calgary. 41 pp. + appendices) made two collections from the Cave and Basin waters after 1968 but failed to find any Green Swordtails (he recorded Longnose Dace and Brook Sticklebacks). In 1976, W. D. Reynolds (1976. A report on Cave and NELSON: TROPICAL FISH FAUNA IN BANFF SSI Basin interpretive research. Manuscript Report to Director, Western Region Parks Canada. 53 pp.) noted the continued presence of Mosquitofish and Black Sailfin Molly, the absence of Green Swordtail and Convict ( = Zebra) Cichlid (a temporary diversion of the hotspring outflow was one factor he noted as being responsible for their disappearance), and the appearance of Jewelfish (Hemichromis bimaculatus Gill). The Jewelfish is thought to have been estab- lished from the introduction of a single breeding pair (C. Yarmoloy, personal communication). Reynolds found the Mosquitofish to occur generally through- out the spring-influenced areas while the Molly was restricted and the Jewelfish very limited. Based on recent communication with W. D. Reynolds and C. Yarmoloy, it seems fairly certain that the report of the Guppy still being present in Reynolds’ report is in error; unfortunately, no specimens of what were thought to be Guppies at the time were saved. Rey- nolds also found Brook Sticklebacks to be numerous in the shallow regions of the pond but to be absent where the exotic fishes occurred. He did not obtain any Longnose Dace but noted C. Yarmoloy’s observa- tions of their presence at the base of the Cave and Basin outflow from about June to August (the pre- sumed spawning time) and at the northern part of the pond, near the outlet, at other times. In addition to my study, fish from the Cave and Basin drainage were also collected in 1981 by J. Lanteigne, biologist on contract to the Committee on the Status of Endan- gered Wildlife in Canada (COSEWIC), assisted by A. Westhaver and F. McAllister, in mid-September ina study of the Longnose Dace. In addition to Longnose Dace, they obtained Mosquitofish and Sailfin Molly in the warm-water area (Mosquitofish outnumbered the molly about 3 to 1). Brook Trout, Sa/velinus fon- tinalis (Mitchill), were present in the pond near the beaver dam of the eastern outlet (J. Lanteigne, per- sonal communication). Results Collecting Sites and the Fishes The collecting sites of this study are described below. The number in parentheses after a species name indicates the number of specimens preserved in dipnet sampling (not just obtained or seen). Inevita- bly, a collecting bias exists with dipnetting in deter- mining the ratio of one species to another. The black Sailfin Molly, while much more conspicuous than the pale-colored Mosquitofish, was faster in swimming away and thus more unlikely to be caught in open water; extra effort was made to ensure collecting some mollies, when present, when only Mosquitofish might otherwise be caught. Although I lack data on the degree of this bias from differential escapes (which 258 could be obtained from large but possibly destructive seine hauls), it was apparent in a few areas that the molly outnumbered Mosquitofish yet the catch data suggested the opposite. It did seem apparent, how- ever, that Mosquitofish were more abundant and widespread than the molly in the pond and springs. Water temperatures are given for the time of mak- ing the collections. The air temperature was relatively cold, generally 5° C; a light snow and rain had fallen the night before commencing the study. Temperatures from some adjacent waters not influenced by thermal springs were 5-1/2°C in the adjacent Bow River, 7'4° Cina recently flooded beaver pond slightly over | km west of Cave and Basin, and 5-1/2°C ina nearby pond possibly at the head of the small creek entering the southwestern part of the Cave and Basin Pond. Conductivity readings were 1880 »mhos/cm (aver- age) in Basin Spring (collecting sites | and 2), 1100 in Cave Spring and at collecting site 6 (both values are very similar to that found by van Everdingen, 1972), and 1550 in the main part of Cave and Basin Hot- springs pond adjacent to site 7. In contrast, the Bow River was 120 and the above mentioned ponds to the west were 115 and 190, respectively. Tropical aquatic plants (of various ones, at least a Tape Grass, Vallisneria spirilis, and a Bushy Pond- weed, Najas microdon, appear to have been intro- duced with the tropical fishes, (C. Yarmoloy, personal communication) and some native plants were green and abundant in some areas influenced by the thermal springs. Vegetation elsewhere was brown. SITE 1. West Fork of Basin Spring and adjacent bay. This branch of Basin Spring is the larger of the two branches and enters a shallow, marshy, soft- bottomed bay through a relatively straight channel with swift current. Mosquitofish occurred a short dis- tance up the channel in the 30° C water. Mollies were abundant in much of the bay, even where the depth was only 2-5 cm. A sharp temperature change occurred proceeding from the deepish area along the north shoreline (18°C) to as little as 1 m away in the bay (27°C). Both areas had Mosquitofish and Sailfin Molly, but shallow backwater areas at 15°C did not. Mosquitofish (30 2 , 3, 37 young) and Sailfin Molly (62 ,2o%, 26 young) were in the bay; two dead Jewel- fish (1) were seen. On 5 August 1982, some 15 dead adult Jewelfish and numerous dead young were in very shallow and heavily vegetated water in the bay; most of the adults had red breeding coloration. SITE 2. Middle East Fork Basin Spring. This site is located between the base of the hill on the upstream end where water cascades over impassable chutes and a small falls on the downstream end formed by sticks and probably also impassable. Above the path the temperature was 33° C in Basin Spring, but was 30° C THE CANADIAN FIELD-NATURALIST Vol. 97 at the start of this section where Mosquitofish occurred and 27° Cat the end. Four very small springs with clear water (in contrast to the slightly milky- colored water of Basin Spring) arise from the hillside below a windfall and enter this section (shown in circled inset of Figure 1); the upper three are shorter than the lowest one and have their sources within about 1.5 m of each other and are about 2 min length. The second lowest spring, with a 10 cm wide slit-like opening (just large enough to get an aquarium net partially inside), was 22°C and rich in algae; young Mosquitofish (10) and a few adults were dipnetted from within the opening and a red-colored fish, possi- bly a Jewelfish in breeding color, was seen entering it. The uppermost of the four springs was 23°C and also contained small Mosquitofish. Many adult Mosquito- fish (22 and 2c) and several Jewelfish (1) in non- breeding color were present in the main spring chan- nel. Aquatic vegetation covers the four small springs and much of the main channel in the summer. SITE 3. Lower East Fork Basin Spring. This sec- tion of the spring has a wide channel and is between the small falls noted above and the entrance of the spring into the marshy bay of the pond opposite the western end of the island (a very small drop in water level of the pond would join this island to the shore and drain the bay). Temperature varied from 27 to 26° C. Based on visual observation, the Mosquitofish (259 ,9o) greatly outnumbered the molly (19,1), but the latter were present in good numbers. SITE 4. Shoreline between Basin and Cave springs. This site was along the marshy shoreline, below a small cliff and adjacent to small caves. Brook Stickle- backs (2) were common by the bank in the aquatic plants where the water was 19°C and Mosquitofish were abundant nearby. No Sailfin Mollies were observed but in one small cave to the east where the temperature was 20-1/2°C there were a few mollies (10) along with numerous Mosquitofish (25 9 , 5c). SITE 5. Cave east of Cave Spring. Cave Spring was 26° C at the path and entered the pond in numerous trickles (22°C) radiating out in a conical pattern. No fish were observed in this Spring but a few Mosquito- fish were in a small pool adjacent to the path and Mosquitofish, Sailfin Molly, and one Jewelfish were observed off the fan-like mouth. The cave (no higher than 30 cm) at the collection site immediately to the east of Cave Spring opened at about pond level in rock below a knoll. Its floor was heavily covered with empty snail shells. The current from the cave was strong and had an estimated discharge of about twice that of Basin Spring. The temperature was 22-1/2 -24° Cin the mouth; a few meters offshore in the pond it was 17°C. In collections made on 16 and 17 May the Mosquitofish (53 9 , 6c’) were more numerous than 1983 Sailfin Molly (59, 4c). Of the various sites, Long- nose Dace (6) were commonest here and swam in and out of the cave along with the tropical fish. The six Longnose Dace were obtained by specifically netting individuals sighted; they are relatively rare compared to the other fish and no young were observed. Most of the Longnose Dace were orange-colored in part of the snout and mouth area and were assumed to be males in breeding condition. SITE 6. Spring about 3 m east of Site 5. This 20- 1/2°C spring has a slow current emerging from the hillside. Only Mosquitofish (7 9 , 3c’) were captured or seen. SITE 7. Bay between island and west shore. This relatively deep bay was 21° Cand had a small 8° C creek (probably temporary) entering from the west. A slight current exists at the northern constriction with the island. The greatest diversity of fish was obtained in this bay. Collections made on 16 and 17 May yielded five species, namely Mosquitofish (189, 7¢%, 8 young), Sailfin Molly (1%, 16 young), Brook Stickleback (13), Jewelfish (1), and Longnose Dace (1). The temperature along the west shoreline of the pond decreased from 21°C at Site 7 to 13-1/2°C a short distance away, opposite a coniferous forest, to 11°C past a sedge marsh near the northern shoreline. No fish were seen along this shoreline past the island. SITE 8. Western outlet channel. There was no sur- face drainage through this blocked off outlet and the dam was about 15 cm higher than the pond level. Only Brook Sticklebacks (6) were dipnetted on the pond side of the elongated and deep channel. Description of the Fishes MOSQUITOFISH: The meristic characters of 205 specimens over 1.0 cm SL are as follows: Anal fin rays GS) 0) 2) 3) Sl Bw 33 2 al ® 7 Me 10 92 2 ie 38 The first and usually the second rays of the dorsal fin are unbranched. The first two (or rarely three) rays in the anal fin are relatively short and the first three rays are unbranched. The third to fifth rays of the male are elongated to form the gonopodium. In females the origin of the dorsal fin is about even with, or slightly in front of, the insertion (posteriormost part of base) of the anal fin while in males the anal and pelvic fins are farther forward than in females (the insertion of the anal fin is well ahead of the origin of the dorsal fin). Males are smaller and more slender than females. The largest females were 4.4 cm SL (3) with most females being between 2.8 and 3.8 cm (all size ranges Lateral scale rows Dorsal fin rays 5 @ 7 § NELSON: TROPICAL FISH FAUNA IN BANFF 29) were present). The largest male was 2.8 cm with most males being 1.8-2.3 cm. The smallest Mosquitofish captured was about 7 mm SL. The largest number of eggs was 88 in a 4.4 cm female; females between 3.0 and 4.0 cm usually had 15-55 fertilized eggs. Most had all their embryos at about the same level of develop- ment (e.g., either younger than the eyed stage, eyed, or with embryos well developed) but many had various stages present. SAILFIN MOLLY: The meristic characters of 42 indi- viduals are as follows: Dorsal fin Anal fin Lateral scale rays rays TOWS 12 13 14 8 9 25 26 Dy 28 3 16 23 6 36 4 9 26 3 The origin of the dorsal fin in the female Sailfin Molly is about even with the pelvic fin insertion and far forward of the anal fin origin. In males the pelvics are slightly farther forward than in females and the modified anal fin is well within the pelvic fins and its origin is about even with the dorsal fin origin. Many of the scales in the dorsal and ventral regions have a microscopic depression near the focus (this is the pit organ or neuromast; the Mosquitofish lack this char- acter although a weak trace of it was found in one individual). The largest female Molly was 5.0 cm SL and the largest male 3.3 cm. Males over 2.9 cm had developed the enlarged dorsal fin (the sailfin). Snelson (1982) notes that males with the large sail-like dorsal fin are typically 5.0-5.5 cm SL, unlike small mature males which are phenotypically similar to females except for the gonopodium. The smallest Molly captured was 8 mm. The largest number of eggs, 90, occurred in the 5.0 cm female. Of the relatively few mature females examined, most had 15-60 embryos which were at various stages of development. A wide diversity of color patterns was present in the Mollies. The jet black (commonest), checkered black, marbled black, and golden (very rare) strains were present. JEWELFISH. The four Jewelfish examined (4.6-6.6 cm SL) each had 14 spines and I! soft rays in the dorsal fin, three spines and nine soft rays in the anal fin, and 24 or 25 lateral scale rows on the body. BROOK STICKLEBACK. The number of dorsal, anal, and pectoral rays of the 21 individuals (2.7-4.6cm SL) are within known values for this species from other localities (Nelson, 1969). Thirteen individuals have five dorsal spines and eight have six spines; all have both pelvic spines. LONGNOSE DACE. The eight specimens were 3.35- 5.4 cm SL. Longnose Dace at Cave and Basin Hot- spring are of special interest because of the possibility 260 of an endemic subspecies being or having been pres- ent. Itis unknown, at present, whether or not existing specimens are descendant from the population which existed before man-made changes to the system or when Nichols (1916) did his study. The relationship between Bow River and Hotspring Dace is unknown but there is a belief that movement between the two areas does take place (C. Yarmoloy, personal com- munication). If such an exchange does occur the ques- tion remains as to whether or not the original popula- tion made sucha migration or whether it died out and was replaced by Bow River stock. It is notable that the Dace were most concentrated in current froma cave at 24°C (D.E. McAllister, personal communication, states that collections in his care were taken in temperatures as high as 26°C). This is an unusually high temperature for Longnose Dace and well above, to my knowledge, that which Bow drainage dace ever otherwise encounter. C. B. Renaud, University of Ottawa, is studying the taxonomic status of the Cave and Basin Dace at the National Museum of Natural Sciences, Ottawa. Discussion The tropical fish fauna in Cave and Basin Hot- spring drainage presently consists of the livebearing and salt-tolerant Mosquitofish and Sailfin Molly which are well established. Adult Jewelfish, which are aggressive, territorial egg-layers exhibiting strong parental care, are relatively rare and very secretive. Although no young were obtained in May they are very common under the parental care of the adult in the summer months (C. Yarmoloy, personal commun- ication). Apparently there is a heavy mortality during the winter but it would appear that a viable popula- tion has been established. The Guppy, Green Sword- tail, and Convict Cichlid, present in 1968, appear to be no longer present. Mosquitofish, native to the Gulf of Mexico (inland to Illinois) from northern Mexico to Florida and north on the Atlantic coast to New Jersey, have been introduced to many places in the western United States (Lee et al. 1980) and throughout the world (e.g., Africa, Europe, Australia, and New Zealand) for mosquito control (Welcomme 1981). It is the most widespread and abundant tropical fish in Cave and Basin drainage and has been taken nearby inthe warm spring area of Third Vermilion Lake (Bow Valley Naturalists 1978). This species has been observed under the ice in the Cave and Basin pond whereas the Sailfin Molly has not (C. Yarmoloy, personal communication). Two subspecies of Mosquitofish are generally rec- ognized (e.g., Hubbs 1961): G. a. holbrooki (Girard), the Atlantic coastal form (New Jersey to Florida) with THE CANADIAN FIELD-NATURALIST Vol. 97 more black speckles, especially in males (black males have been obtained in aquaria), 7 dorsal fin rays, and 10 anal fin rays, and G. a. affinis (Baird and Girard), the inland form (primarily Mississippi drainage to the Rio Grande) witha paler body, 6 dorsal fin rays, and 9 anal fin rays. The specimens taken from Cave and Basin had a pale body with a few black speckles usu- ally on the caudal fin (and on other parts) in both males and females. In fin ray counts they also resem- ble G. a. affinis. Mail (1954) stated that the Cave and Basin Mosquitofish were imported from California and Moyles (1976) noted that the subspecies intro- duced into California seems to be mostly G. a. affinis. The Sailfin Molly, exhibiting many variations in color patterns which are bred for by aquarists, prefers warmer waters than the mosquitofish and is native to Gulf of Mexico drainage from Yucatan Peninsula in Mexico to Florida and north on the Atlantic coast to North Carolina. It has been introduced in three southwestern states of the United States (Lee et al. 1980) and in several other parts of the world either for mosquito control or through releases from aquaria (Welcomme 1981). It seems probable in the Cave and Basin Hotsprings Pond that the checkered black, marbled black, and golden color strains have reverted from the permablack form, the only one known to have been introduced. The Jewelfish is native to western, central, and northeastern Africa. The only record given for intro- duction by Welcomme (1981) is for Florida where it occurs in canals in Dade County (Lee et al. 1980). The Cave and Basin record is the only other known occur- rence in the wild in North America. The Guppy, introduced in the 1960’s and perhaps also in the early 1970’s, was never very common and never fully established; it was probably gone by 1976. The relatively high numbers compared to Mosquito- fish noted in McAllister (1969) may be the result of collecting bias. Although the habitat may not be suit- able for the Guppy it is also possible that the aggres- sive (and more cold-tolerant) Mosquitofish was involved in preventing their continued existence. Myers (1965) notes cases of reduction and extinction of several fish species in various parts of the world where Mosquitofish had been introduced (and calls attention to the existence of other fish species equally capable of mosquito control which are not as destruc- tive to other fishes). Lee et al. (1980) reported that Mosquitofish caused the extirpation of many rare, localized populations of native fishes in the south- western United States. In contrast, the Sailfin Molly survives, perhaps, by being a faster swimmer than the Mosquitofish and thereby better able to avoid aggres- sive advances of the latter. As far as known, with the exception of Mosquito- 1983 fish possibly being in Third Vermilion Lake, tropical aquarium fishes are presently established in Banff National Park only in Cave and Basin drainage. There appears to be no difference in the meristic characters examined with those known in other popu- lations. Given the successful introduction of three tropical fish species so far north it would be of interest to determine the breeding pattern during the year, under the markedly variable photoperiod existing at this latitude. Acknowledgments I am grateful to Parks Canada for allowing me to collect in the Cave and Basin Hotsprings drainage. Their wardens and interpretive staff were helpful in all phases of the study. Thanks are due to D. E. McAllister (Curator of Fishes at the National Museum of Natural Sciences, Ottawa), A. A. Peters (Project Director, Cave and Basin Centennial Centre, Parks Canada), P. Wiebe (Wildlife and Fisheries Officer, Parks Canada), W. D. Reynolds (former contractor to Parks Canada for Cave and Basin resource inventory, presently Pro- gram Co-ordinator of John Janzen Nature Centre, Edmonton), C. Yarmoloy (former biology teacher in Banff and tropical fish hobbyist), and M. A. Harris (Zoology Department, The University of Alberta) for providing valuable criticisms of the manuscript. Iam especially grateful to Cecil Yarmoloy for freely shar- ing his intimate knowledge of the hotspring fishes with me. I have also benefited from discussions with J. Lanteigne (Department of Fisheries and Oceans, Ottawa). My four children, Brenda, Janice, Mark, and Karen assisted in making the collections. T. Smith provided technical assistance in the laboratory. This study was financed by Natural Sciences and Engineering Research Council grant No. A5457. Literature Cited Bow Valley Naturalists. 1978. Vermilion Lakes Banff National Park, an introductory study. Bow Valley Natu- ralists, Box 1693, Banff, Alberta TOL OCO. 68 pp. Clark, A. H. 1973. The freshwater molluscs of the Cana- dian Interior Basin. Malacologia 13(1-2): 1-509. NELSON: TROPICAL FISH FAUNA IN BANFF 261 Clark, A. H. 1981. The freshwater molluscs of Canada. National Museum of Natural Sciences, Ottawa. 446 pp. Eigenmann, C. H. 1895. Results of explorations in western Canada and the northwestern United States. Bulletin of the United States Fish Commission 14 (for 1894): 101-140. van Everdingen, R. O. 1972. Thermal and mineral springs in the southern Rocky Mountains of Canada. Water Management Service, Environment Canada. 151 pp. Hubbs, C. L. 1961. Isolating mechanisms in the speciation of fishes. Pp. 5-23 in Vertebrate speciation. Edited by W. F. Blair. University of Texas Press, Austin. Hubbs, C. L., and K. F. Lagler. 1958. Fishes of the Great Lakes Region. Cranbrook Institute of Science Bulletin No. 26: 1-213. Lee, D. S., C. R. Gilbert, C. H. Hocutt, R. E. Jenkins, D. E. McAllister, and J. R. Stauffer, Jr. 1980. Atlas of North American freshwater fishes. Publication 1980-12, North Carolina Biological Survey. North Carolina State Museum of Natural History. 854 pp. Mail, G. A. 1954. The mosquito fish Gambusia affinis (Baird and Girard) in Alberta. Mosquito News 14(3): 120. McAllister, D. E. 1969. Introduction of tropical fishes into a hotspring near Banff, Alberta. Canadian Field- Naturalist 83(1): 31-35. Moyles, P. B. 1976. Inland fishes of California. University of California Press, Berkeley. 405 pp. Myers, G.S. 1965. Gambusia, the fish destroyer. Tropical Fish Hobbyist. January, 1965: 31-32, 53-54. Nelson, J. S. 1969. Geographic variation in the brook stic- kleback, Culaea inconstans, and notes on nomenclature and distribution. Journal of the Fisheries Research Board of Canada 26: 2431-2447. Nichols, J. T. 1916. On a new race of minnow from the Rocky Mountains Park. Bulletin of the American Museum of Natural History 35(8): 69. Price, R. A.,and E. W. Mountjoy. 1972. Map sheet 1294A, Banff (east half). Department of Energy, Mines and Resources, Ottawa. Pritchard, G. 1971. Argia vivida Hagen (Odonata: Coena- grionidae) in hot pools at Banff. Canadian Field- Naturalist 85(2): 186-188. Snelson, F. F., Jr. 1982. Indeterminate growth in males of the sailfin molly, Poecilia latipinna. Copeia 1982(2): 296-304. Welcomme, R.L., Compiler. 1981. Register of interna- tional transfers of inland fish species. FAO Fisheries Technical Paper No. 213, Rome. 120 pp. Received 28 March 1982 Accepted 20 August 1982 Observations on Home Range Sizes, Movements and Social Organization of Lynx, Lynx canadensis, in Riding Mountain National Park, Manitoba L. N. CARBYN and D. PATRIQUIN Canadian Wildlife Service, 9942 - 108 Street, Edmonton, Alberta T5K 2J5 Carbyn, L. N., and D. Patriquin. 1983. Observations on home range sizes, movements and social organization of Lynx, Lynx canadensis, in Riding Mountain National Park, Manitoba. Canadian Field-Naturalist 97(3): 262-267. Ofa total of five Lynx (Lynx canadensis) captured, three were radio-collared and their movements studied. All radio-collared Lynx were later killed by trappers adjacent to the park. Home ranges for two females averaged 156 km? and that of a single male was 221 km2. These figures are higher than reported in the literature. Two females travelled with their kittens together for a short interval. Riding Mountain National Park does not appear to be large enough to sustain a viable Lynx population over time, and reinvasion of Lynx from outside areas may be necessary to maintain populations. Key Words: Lynx canadensis, Lynx, home range sizes, movements, Riding Mountain National Park. Although widely distributed in Canada and impor- tant as a commercial fur animal, the social behaviour of Lynx has not been extensively studied. Several studies have centred on food habits (Van Zyill de Jong 1966; Nellis and Keith 1968; Bergerud 1971; Nellis et al. 1972; Brand et al. 1976; Parker 1980). Data on home ranges and movements have been obtained from trapping and recaptures (Saunders 1963), snowtracking (Brand et al. 1973) and radio- tracking (Berry 1973; Koehler et al. 1979; Mech 1980; Parker, in preparation). Mech (1980) compared the methods and concluded that radio-tracking over- comes difficulties inherent in the other methods. This paper describes the movements of Lynx cap- tured and radio-collared in Riding Mountain National Park. Data were collected secondarily to other large carnivore studies on wolves and coyotes. Description of Study Area Riding Mountain National Park, 2944 km? in size, is located in southwestern Manitoba (Figure 1), approximately 200 km northwest of Winnipeg. The Park is surrounded by agricultural lands and as such forms an isolated wilderness system. Bailey (1967) described 12 main cover types within four vegetational associations. These associations are grassland areas, interspersed through the western por- tion of the park, two subcomponents of the boreal forest and northern elements of deciduous forests (Rowe 1972). Predominant vegetation are aspen (Populus tremuloides) and mixed forest stands. Relief, drainage and fire history govern plant cover in localized areas. Within historical times, haying, cattle grazing and logging contributed to human-induced changes, however such activities have ceased in recent years. Interspersed throughout the park are numerous ponds, lakes and a few streams. In the northern and eastern portions of the park the plains are deeply incised by streams flowing across an escarpment along the eastern portion of the park. Westward the terrain blends into the prairies and forms a plateau with a series of gently undulating hills. Methods Lynx were captured in No. 4 or No. 14 leg hold traps incidental to capture programs for Wolf (Canis lupus) and Coyote (Canis latrans). Trapped Lynx were anesthetized with equal porportions of phencyc- lidine hydrochloride and promazine hydrochloride at dosages of | mg of each drug per kg of Lynx weight. Captured Lynx were measured, weighed, ear-tagged, radio-collared and then released. Radio fix locations were monitored with a Maul MS aircraft and plotted on topographical maps. Home range sizes and distan- ces between fixes were calculated using a Hewlitt Packard 9830 desk computer. Fixes were sequentially numbered to detect whether spread of locations reflected movements within a home range or dispersal to new areas. Efforts were made at each relocation to see the animals and record group sizes of females with kittens. All radio fixes of the females were from the winter (November- April) period and only 4 of the 49 fixes of the male were from the August-October period, the remainder covered the winter period. Results Five Lynx were captured and three of these were radio-collared. Two animals from different family groups and a lone animal were radio-collared (Table 1). Radio-marked Lynx were studied over periods ranging from 86 to 228 days, and in one case recovery 262 1983 CARBYN AND PATRIQUIN: LYNX IN RIDING MOUNTAIN PARK 263 e Whiripooi L. Riding Mountain y National Park (Geese *=——="“McCreary Lynx (ML) SARY 1 WO) ZIYON/ 13 Oo Capture location [a] Kill site *—. Muskrat Lake” Lynx (MY) 15/08/78 to 29/11/80 PLM UT WS 8/02/78 4 Capture location °o Capure location Kill site ‘O) Kill site FIGURE |. Home range delineations of radio-collared Lynx studied in Riding Mountain National Park, Manitoba. ——- “Rolling River”Lynx (RR) Vol. 97 THE CANADIAN FIELD-NATURALIST 264 ‘({ 91n31,4 99S) AJUO sjurod apisul apnjout 0} paisn[py_, 916] Jaquiaidas ZZ (pose][od Jou 3nq) poinjdeo 3s1147 *poinjdeos sem dnoiz ay} ul xuAT Pole][Od ay} UDYM 8/6] AleNIgd7 B [MUN 1943930} pa[aaes} XUAT 9ATj JO dnoid ay |, ‘Ud}ITy YIM I[Npe puodas e Aq 8/6] Alenues 6 UO poulol sem dnoisd ATW SITY] | ‘08-1 1-6¢ poddex yewiuy ‘6L-€0-1E poddo}s auole dad SuOWUOW —« TZ pe =u CC +UASE B8te+wisr 240491 C8 +L 6P 78L-8-S I SI surjoaray, = yyNpy ae W ‘uMouyUN y[npe puosss pue sud}}1y poinjdes 3914} JO uayM 218} °8/-70-8 sud} y podden (811) Om YIM IW qyewof LI e] =u 61 +4UATe OF +WMICr LOFVKC €9 + 9P CC [AANEG 601 suppor, ynpy yew, ‘uMoUyUN uaiy | Jo poeinjdes 218} -8/-1-L7Z uayM poddeyn sudqqy u9}}1] au0 OM} YIM AW pue gewot =e 8 =u 66 FUAShp cSezrUyaoe L047? BL 4+ 9S vl LL-I1-6 Qe] Buypoaery, ynpy ajeuraj ae 7 921 ‘as7x ‘as7x ‘asFx ‘ds7x SOXIJ a1ep (3%) snjes adv Xa osurl gjduies s[eA1aqUI skep ¢ potiod JO‘ON oimdeg WWsIaAK, [RID0S oweH Aep ¢ [210.L S[BAIOJU] [B10 (wy) (skep) saxlj UIaMJaq SsadURISIG Aquanba14 3ul0jU0jy yleg [euoneN uleyuNopy BUIPIy Ul poie[[oo-orpes xUAT Jo sodues JWOY pu ‘s}UdIAOW ‘snjRIs [BIOs ‘Aouonbaly SULIOWUOW “[ ATAVL 1983 of a Lynx occurred 836 days after capture (Table 1). An adult female (ML) weighing 10.9 kg was cap- tured on 2 November 1977. There was no evidence that the female was still nursing her young. This animal travelled with two kittens, and based on 14 radio fixes covered an area of 138 km? (Figure 1). These three animals were joined by two others (non- collared) which from aerial observations appeared to be a female with kitten. On five different occasions, spanning a period of 30 days (9 January to 8 Febru- ary) these five animals were seen travelling together. The radio-collared female was killed by a trapper outside the park on 8 February 1978. Another adult female (MY) was captured 9 November 1977 and weighed 13.6 kg. A kitten, travel- ling with her and weighing 8.2 kg was captured at the same time. Home range, based on 22 fixes, covered by these two Lynx was 177 km? in size (Figure 1). Based on inside points only (see exclusion of shaded area) the home range was 118 km2. A trapper killed the animal 27 January 1978. An adult male (RR) weighing 15 kg was captured and radio-collared 15 August 1978. This animal had been previously trapped and ear tagged on 22 Sep- tember 1976, but was not radio-collared at that time. This Lynx was killed by a trapper adjacent to the park 29 November, 1980. Home range (15 August 1978 to 29 November 1980) of this male was 221 km? (Figure 1). Locations of radio fixes were plotted (Figure 1) and the time intervals listed (Table 1). Mean distance between fixes for adult females were 5.6 km and 4.7 km respectively and 4.8 km for the male. Since time intervals between monitoring dates were often widely spaced apart the distances for winter months were recalculated to include fixes which were obtained within three day intervals. Average distances between fixes in three day periods were 4.8 km, 3.1 km and 3.8 km for the two females with kittens and for the male respectively. Actual distances travelled probably were considerably longer. Discussion The status of Lynx populations in the past within Riding Mountain National Park has been uncertain. The species was not listed by Green (1932). Soper (1953) believed that only a few individuals migrate from time to time into the park. Soper did not see much evidence of the species while conducting field work in the 1940’s. Other previous records indicate the often sporadic abundance of the species. During the duration of this study Lynx were reported only for the eastern portion of the park. Intensive winter field work was carried out in the western portion on wolf/ungulate studies and on Showshoe Hare (Lepus americanus) (Leonard 1979, CARBYN AND PATRIQUIN: LYNX IN RIDING MOUNTAIN PARK 265 Poll 1981, Parks Canada and C.W.S. unpublished reports). Observers failed to see any sign of Lynx at that time. Since then wardens (A. Cochrane, personal communication) have seen a few Lynx in the winter 1980/81 in the western portion of the park. Concen- tration of Lynx in localized pockets were also found in Newfoundland (Bergerud 1971), Alberta (Brand et al. 1976) and Minnesota (Mech 1980). One trapper around the park reported trapping 16 Lynx in a re- © stricted area (about 3 km?) overa short period of time (January, early February) in 1982 (P. Paquet, per- sonal communication). From observations elsewhere in Manitoba it appears that the provincial Lynx har- vest peaked in 1978/79 and since then numbers have declined (R. Stardom, personal communication). Harvest statistics have revealed that the peaks begin in the northwestern portion of the province and move progressively outward from that focus. The peak in Lynx pelt production in the western part of the pro- vince, including the RMNP area, occur 4-S years after the harvest has peaked in the Pukatawagun/ Nelson House area. The province-wide peak in pelt produc- tion appears to occur when Lynx numbers are high over the greatest area. This study was conducted at a time when Snowshoe Hare numbers were at or near peak levels in the eastern portion of the province and Lynx in the park could have dispersed from more northerly areas. The Snowshoe Hare population in the park previously peaked in 1970/71 and was rapidly increasing by 1978 (Poll 1981, unpublished CWS report). In the early to mid 1970’s Lynx were seen on several occasions by L. Carbyn and G. Trot- tier in the central portion of the park. Warden obser- vations for that period suggest only sporadic distribu- tion of Lynx (Kingsley and Stelfox 1978, CWS unpublished report). Fur returns for the province of Manitoba indicated record high Lynx catches for 1971 (R. Stardom, personal communication). South of Manitoba Lynx began appearing in Minnesota in 1972 (Mech 1980) which probably dispersed from the north. The general absence of large numbers of Lynx in the early 1970’s and increase at a time when the Snowshoe Hare cycle was increasing could explain why the home range sizes reported here are larger than any reported to date in the literature. Previously the largest home range sizes for female Lynx ranged from 51-122 km? (Mech 1980). Female Lynx home ranges in Riding Mountain were 156 km? and unlike Minnesota Lynx, the females in this study were travelling with kittens. These figures are based on comparable methods in home range delineation as described by Mech (1980). If only the inside points of the home range perimeter (see Figure 1) are used in the aerial calculations then average female home range sizes would be 128 km?. 266 Calculation of territory size and home ranges for car- nivores requires more detailed attention. In Wolf stu- dies within the Park (Carbyn 1980, unpublished CWS report) a large number of fixes in relatively small territories were plotted at five fix intervals to deter- mine asymptotes at which upper limits of territories were defined. This approach probably is less applica- ble to social units or individuals with large home ranges. However, since only small numbers of radio fixes were available for Lynx in this study, it is likely that the larger territory sizes, using outside points only, are more representative of true home range sizes. The home range of the single male (221 km?) was similar to the range reported for Minnesota. Exten- sive radio-tracking in saturated Lynx populations in Nova Scotia resulted in much smaller territory sizes (G. Parker, personal communication). Mech (1980) suggested that spatial organization of Lynx in Minnesota was similar to that of Mountain Lions, Felis concolor, (Seidensticker et al. 1973) but differed from Bobcat (Felis rufus) (Bailey 1974). That is, female Lynx ranges tend to overlap both spatially and temporally, whereas male ranges to not overlap with other males or with females. In this study both females with kittens were caught at the same locations and ranges overlapped. The range of RR male Lynx overlapped with the territory of ML female, but this could have been as a result of the loss of ML female from her range, since monitoring of the male’s move- ments occurred after the female’s death. Our data therefore are inconclusive as to whether male/female ranges were separated, although results do imply this to be the case. Radio-tracking of Lynx indicated that the park population is vulnerable to outside non-park oriented activities. All five captured Lynx, including the three radio-collared animals, were killed by trappers. High human-induced Lynx mortality was also reported in the radio-collared Lynx in Minnesota (Mech 1980). We believe that in years of low Lynx numbers and high trapping pressures it is possible that the entire population in the park could become extirpated and numbers may then have to be replenished from movements by Lynx from other areas. Under present conditions, therefore, it does appear that the park may not be large enough to sustain a viable Lynx popula- tion over time, when a combination of factors such as high trapping pressures around the periphery of the park and low hare densities coincide and persist fora number of years. Further research would be required for clarification. Degree of dependence of kittens on their mother is important in setting harvest regulations. In this study kittens travelled with adult females well into midwin- ter. Judging from teat size it is improbable that young THE CANADIAN FIELD-NATURALIST Vol. 97 were still nursing. It was not possible to determine when kittens begin to establish their own home ranges or become separated from the parent. Trappers in northern Alberta recognize the vulnerability of family groups to trapping and take advantage of it by setting several traps in specific areas. Cases have been reported where only adult females were caught and kittens froze to death at the side of the dead parent (P. Galbraith, personal communication). Parker (1980) stated “hunting success must be an important learned behaviour by kittens during their first year of life. The extent of mortality among kittens, whose mother is trapped early in winter is completely unknown.” Our observations agree that this could be an important factor in management of the species. The fact that we observed two adult females travelling with their kit- tens, would suggest that under some circumstances the loss of one female may lead to the adoption of the surviving kittens by other adults in the group. Because of the economic importance of this species, we urge further research on its social behaviour for better understanding of life history as a basis for effective management. Acknowledgments The study was funded by Parks Canada. We grate- fully acknowledge the cooperation of G. Rochester, former Superintendent and his staff. Warden Holz- miller assisted in the capture of Lynx. Numerous wardens, from time to time assisted in the radio- tracking of Wolves, Coyotes and Lynx. Trappers C. Levandoski and A. Townley provided us with trap- ping information. R. Stardom, Manitoba Depart- ment of Natural Resources read the manuscript and allowed us to use unpublished information. Pilots from Dauphin Air provided us with efficient services. F. Anderka, Canadian Wildlife Service, Ottawa con- structed the radio collars. R. Leonard and E. Telfer reviewed the manuscript. Literature Cited Bailey, R.H. 1968. Notes on the vegetation in Riding Mountain National Park, Manitoba. Department of Forest and Rural Development, Forest Management Institute, Ottawa. 80 pp. Bailey, T. N. 1974. Social organization ina bobcat popula- tion. Journal of Wildlife Management 38: 435-446. Bergerud, A. T. 1971. The population dynamics of New- foundland caribou. Wildlife Monograph 25: I-55. Berry, P. M. 1973. Ecology and status of the lynx in inte- rior Alaska. In The World’s cats. Edited by R.L. Eaton. World Wildlife Safari, Winston, Oregon 1: 349 pp. Brand, C. G., L. B. Keith, and C. A. Fischer. 1976. Lynx responses to changing snowshoe hare densities in central Alberta. Journal of Wildlife Management 40: 416-428. 1983 Green, H. U. 1932. Mammals of the Riding Mountain National Park, Manitoba: a compilation of field notes and observations. Canadian Field-Naturalist 46: 149-152. Koehler, G. M., M. G. Hornocker, and H.S. Hash. 1979. Lynx movements and habitat use in Montana. Canadian Field-Naturalist 93: 441-442. Mech, L. D. 1980. Age, sex, reproduction, and spatial organization of Lynxes colonizing Northeastern Minne- sota. Journal of Mammalogy 61: 261-267. Nellis, C. H.,S. P. Wetmore, and L. B. Keith. 1972. Lynx- prey interactions in Central Alberta. Journal of Wildlife Management 36: 320-329. Parker, G. 1980. Winter habitat use and hunting activities of lynx (Lynx canadensis) on Cape Breton Island, Nova Scotia. Pp. 221-248 in Worldwide Furbearer Conference Proceedings. Volume |. Edited by J. A. Chapman and D. Pursley. 652 pp. Rowe, J.S. 1959. Forest Regions of Canada. Canada Department of Northern Affairs and Natural Resources, Forest Bulletin 123. 71 pp. CARBYN AND PATRIQUIN: LYNX IN RIDING MOUNTAIN PARK 267 Saunders, G. K. 1963. Movements and activities of lynx in Newfoundland. Journal of Wildlife Management 27: 390-400. Seidensticker, J. C., M. G. Hornocker, M. V. Wiles, and J.P. Messick. 1973. Mountain lion social organization in the Idaho primitive area. Wildlife Monograph 35. 60 pp. Soper, J.D. 1953. The mammals of Riding Mountain National Park. Canadian Wildlife Service Wildlife Man- agement Bulletin Series |. Manitoba, Canada. No. 7. 34 pp. Van Zyll de Jong, S. 1966. Food habits of lynx in Alberta and the Mackenzie District, N.W.T. Canadian Field- Naturalist 80: 18-23. Received 15 June 1982 Accepted 15 February 1983 An Ecological Study of Cypripedium passerinum Rich. (Sparrow’s Egg Lady-slipper, Orchidaceae) on the North Shore of Lake Superior C. J. KEDDY!, P. A. KEDDY?, and R. J. PLANCK?3 1644 Chapel St., Ottawa, Ontario KIN 7Z9 2Department of Biology, University of Ottawa, Ottawa, Ontario 3Limnoterra, 275 Lancaster St. W., Kitchener, Ontario N2H 4V2 KIN 6N5 Keddy, C. J., P. A. Keddy, and R. J. Planck. 1983. An ecological study of Cypripedium passerinum Rich. (Sparrow’s Egg Lady-slipper, Orchidaceae) on the north shore of Lake Superior. Canadian Field-Naturalist 97(3): 268-274. An ecological study of Cypripedium passerinum Rich. was conducted in the summer of 1979 on the north shore of Lake Superior near the mouth of the Pic River. Here C. passerinum occurs in sand dune complexes in a wide variety of habitats ranging from open grass-covered dunes to stabilized dunes with Picea glauca forest. Details of morphology, phenology, growth, reproduction and habitat are presented for the 52 groups of plants (899 plants) found. Cypripedium passerinum begins to flower in mid-June and immature capsules can be found in early July. Flower mortality was caused by insect damage (1%) and an unknown agent (12%). High capsule set in undamaged flowers (99%) suggested autogamy was occurring and experimental pollination studies supported this. Seed dispersal is not synchronized in any manner — both dehisced and undehisced capsules of the previous year were found in July. Cypripedium passerinum may take approximately 15 years to flower from seed. Four species from three insect orders (Hymenoptera, Diptera, Lepidoptera) fed on C. passerinum. Examination of 42 groups of flowering plants showed that only five contained seedlings and that within-group recruitment was largely the result of vegetative reproduction alone. Seedlings were found only in early successional habitats within the dune complex. The habitat occupied by C. passerinum is a changing environmental mosaic, and this has important implications for species management. Key Words: Cypripedium passerinum, rare species, autecology, Lake Superior, sand dunes, orchids, lady-slipper, reproduction. Cypripedium passerinum Rich. (Franklin’s or Sparrow’s Egg Lady-slipper) is one of the few members of the orchid family that grows within the Arctic Circle. Its North American distribution is con- tinuous from Alaska to James Bay, south to Lake Winnipeg, southwestern Saskatchewan, northwestern Montana and eastern British Columbia (Luer 1975; J. L. Riley, personal communication). The two most disjunct pockets are found at the mouth of the Pic River (48°36’N, 86° 18’W) in Ontario, 400 km south of the more or less continuous range in the Hudson Bay lowlands, and in the Mingan Islands (50°15’N, 63°30’W) near the north bank of the St. Lawrence River, isolated from the main range by 900 km. Figure | shows the Ontario range of C. passerinum. The disjunct station at the mouth of the Pic River, on the north shore of Lake Superior, is the most south- erly location for this species in Ontario. In this area, the species was first reported north of the Pic River by Macoun (1888) in the 1800’s and was later relocated in 1964 by Soper (1965). In 1976 it was also found south of the Pic River, in Pukaskwa National Park, by a team of botanists from the University of Toronto (P. M. Catling, personal communication). Because of its disjunct populations mentioned above, C. passerinum was included on the list of rare plants of Ontario (Argus and White 1977) and on the list of rare species in the Canadian flora (Kershaw and Morton 1976). It was subsequently excluded from the atlas of the rare vascular plants of Ontario (Argus and White 1982) because the disjunction Is not as great as originally thought, as shown by botanical exploration in the intervening region; but this does not diminish the significance of the population at the Pic River (G. W. Argus, personal communication). As with most rare species, little was known about the ecology of C. passerinum. Therefore, Parks Canada had insuf- ficient information on which to base management decisions regarding this species in Pukaskwa National Park. The principal object of this study was to obtain basic ecological data on the reproductive biology and habitat of this rare species. Methods A ‘group’ refers to a cluster of one or more plants spatially distinct from other such clusters by at least 1 m. Three clusters, although less than | m from one another, were recognised as groups because of obvious habitat differences. Since the distribution of C. passeri- num near the Pic River was poorly known, the search for groups continued throughout the study. Therefore, for each characteristic studied, sample size was deter- 268 1983 FiGurRe |. Distribution of C. passerinum in Ontario based on herbarium specimens (circles courtesy G. W. Argus, Botany Division, National Museum of Canada, and triangles generously provided by J. L. Riley). mined by the number of plants or groups found at the time the characteristic was observed or measured. Field work was carried out 6 June to 19 July, 1979. To obtain information on growth in previous years, rhizomes were examined on 26 flowering or large vegetative plants distributed among 22 groups. Surfi- cial debris and sand were gently brushed aside to reveal the shallowly-buried rhizomes. Distances between stalk scars (Figure 2), depth, and direction of growth were noted. Mean annual growth increments were calculated for 5-year intervals up to 1975-1979. Floral development was observed for flowering shoots in 34 groups. The relative contribution of sex- ual and asexual reproduction to group expansion and maintenance was first determined for 20 flowering groups. Seedlings were identified by a gradually taper- ing rhizome less than | cm long. In groups where seedlings were not obvious, the rhizome of the young- est plant was examined by carefully loosening the soil around it. Based on experience gained from these 20 groups, the presence/absence of seedlings in the remaining 22 flowering groups was determined with- out disturbing the soil, through observations of size, morphology and distance to other plants. Several procedures were carried out in an effort to determine the insect pollinators of C. passerinum. The KEDDY, KEDDY AND PLANCK: SPARROW’S EGG LADY-SLIPPER 269 exit holes at the base of the column in 10 flowers were plugged with cotton in order to trap pollinators in the lip. The lips were examined several times a day for a number of days after the exits were plugged. Secondly, insects were collected from nearby plants to be examined for the presence of C. passerinum pollen, but these were lost in transit. Thirdly, six hours (dur- ing both daylight and dusk) of four days during the peak of flowering were spent observing the flowers for insect activity. The success of self-pollination in producing swollen Ovaries was investigated. Fine mesh bags were tied over 12 unopened flowers as tightly as possible with- out constricting the stalk. After the flowers opened, ten were self-pollinated and two were left untouched as control plants. The mesh bags were immediately replaced following manipulation. Insect larvae found on C. passerinum were reared on leaves from nearby plant species to obtain adults for identification. The presence/absence of vascular plants, mosses and lichens within a 30 cm-diameter ring around C. passerinum plants was recorded for 40 groups. Common names for vascular plants were obtained from Fernald (1950) and Gleason (1968). Neen \ =p Stalk Scar we A Rhizome | E( Vel KE" FP YA [} 12, NS fst is Loe (S GS oaP- y Bud ~ ve u a ’ Roots FiGurE 2. Morphology of a plant of C. passerinum from the mouth of the Pic River. 270 Results Distribution Fifty-two groups (899 plants) were found at the Pic River, 22 to the south and 33 to the north. Groups varied in size from one to 137 plants and, typically, were composed of both clones and separate individu- als. More detailed information on distribution is recorded in Keddy (1979. An Ecological Study of Cypripedium passerinum and Listera borealis. Report to Parks Canada, Ontario Region, Cornwall, Ontario. 126 pp.). Although other sandy shorelines in Pukaskwa National Park were explored during the period of peak flowering, no new locations were found. Morphology C. passerinum is a perennial species with a horizon- tal rhizome which grows a new segment each year and produces a leafy aerial shoot at its tip. In the specimen examined (Figure 2), two roots were produced by each annual segment. Buds on the rhizome develop into lateral branches. After many years and much branch- ing, a rhizome network is formed and many appar- ently separate plants may be interconnected beneath the sand surface forming a large clone. Rhizomes generally grow parallel to the soil sur- face. The older portion of the rhizome tends to occur slightly deeper than the most recent portion, likely due to sand and litter accumulating above it over the years. At those sites where sand was obviously accumulating (sand rained down during field work), the rhizomes were growing at a 45° to 60° angle from the horizontal plane, rather than parallel to it. Rhi- zomes were buried deepest in moss-covered areas, where they occurred at the interface between a dark organic layer and 5to9cm of partially decayed moss. Rhizomes were shallowly buried (2.5-4 cm) where the surface was sand or litter. Plants can be approximately aged by counting the stalk scars (one scar per year); however, after several years, the oldest part of the rhizome rots and disinte- grates. Thus, ages determined by rhizome examina- tion may underestimate the actual age of old plants. The oldest rhizome visible was 23 years old, indicating that the group was of an even greater age. Rhizomes older than 17 years were found in six of the 22 groups examined. Mean annual rhizome increment (calcu- lated for 5-year intervals, up to 1975-1979) ranged from 3 mm to 18 mm (avg. 10 mm). Floral Development and Capsule Production The following scale was used to quantify floral phenology: 1. flower not yet visible 2. flower visible but unopened THE CANADIAN FIELD-NATURALIST Vol. 97 3. lateral petals open 4. dorsal sepal raised, exposing lip opening 5. flower wilted On 5 June, all plants resembled green pencils up to 5 cm high with two or three leaves visible. Stem elon- gation and leaf production gradually occurred, and flower buds (stage 2) were first observed 14 June. Individual flowering shoots ranged in height from 9 to 40 cm(n = 200). Flowers opened and were receptive to pollinators (stage 4) about a week later. There is considerable variation in phenology; some flowers were not fully open by 11 July. Wilting (stage 5) occurred between two and eight (usually four or five) days after flower opening. A few days later, the ovary was obviously swollen but by 19 July none had reached its full, mature size. Figure 3 shows floral development for flowering plants in two groups selected to illustrate the amount of variation observed. The most rapid development occurred on an open, grass-covered dune; this group 1s not shown in Figure 3 since it was found on 30 June with wilted and dried flowers, and the date these plants reached stage 5 could not accurately be deter- mined. Rapid development also occurred in a warm, sheltered area behind a high dune (A in Figure 3). The slowest rate of development was observed in dense shade beneath a low Picea glauca (White Spruce) canopy (B in Figure 3). Dates for seed maturation on hh ine) MEAN FLORAL DEVELOPMENT STAGE — ol i 13, 29 2b ass 6 iO «(14 JULY JUNE FiGuRE 3. Mean floral development stages (with 95% Cl) for two representative C. passerinum groups: A = shel- tered leeward slope of dune, B = level site with dense P. glauca canopy (stage |- flower not visible, 2-flower visible but unopened, 3-lateral petals open, 4-lip open, 5-flower wilted). 1983 remain to be determined. The time of capsule opening and seed dispersal varies. Typically, plants produce one flower or capsule per stalk. Only two double-flowered plants were found. A total of 256 flowers were produced by the 29 groups observed. Flowers were naturally damaged in two ways. The ovaries of 1% (3/230) were partially eaten by caterpillars. The flower and/or the peduncle of 12% (27/230) turned brown and wilted as a result of an unknown agent. Excluding those plants which were either manipulated for experimental purposes (26) or naturally damaged (30), 200 remained. Of these, 198 (99%) set capsules. Pollination The flowers of C. passerinum are white and have a light, sweet fragrance emitted by the lateral petals. It appears that the purple-dotted and yellow-tipped column serves to guide insects into the lip. Lines of purple dots inside the lip then probably guide them past the stigma, the anther and to the exit holes on either side of the column base. The size of the exit hole, 2.5 x 3.0 mm in C. passerinum, restricts the size of the pollinator. No insects were observed visiting C. passerinum at the peak of flowering and no insects were trapped in flowers whose exits were plugged. Most Cypripedium species are particularly attractive to bees (Stoutamire 1967), but only two bee species were found in the vicinity of C. passerinum (on Rosa acicularis, Rose). One was clearly too large to fit into the lip. The second, slightly smaller, was released into the lip, but got trapped and was much too large to use the exit holes. Swollen ovaries were produced on both self- pollinated and control plants. The capsules were still immature when the study ended; thus, seed produc- tion from these experiments could not be determined. Seed Production and Dispersal Large numbers of small, light-weight, wind- dispersed seeds are produced in each capsule. Capsule opening and seed dispersal is not synchronized among plants in groups of C. passerinum and seed dispersal occurs over a period of time. Capsules produced in the previous year remain attached to the plant through the winter. All stages of seed dispersal, from com- pletely vacant capsules to capsules not yet open and full of seeds, were observed in capsules that had over- wintered. Capsules produced during this study were not mature by 19 July. Seedlings Anexamination of stalk scars on the rhizome of the smallest plant at one site showed that it had been KEDDY, KEDDY AND PLANCK: SPARROW’S EGG LADY-SLIPPER Dial growing for at least 15 years and yet the plant had probably not reached flowering age (the stalk scar diameters were much smaller than those of nearby flowering individuals). Seedlings had recently established in only five of the 52 groups found, although flowering plants were pres- ent in 42. Thus, within-group recruitment was largely the result of vegetative reproduction alone. Herbivore Damage Only insect herbivores were observed feeding on C. passerinum and eight plants (<1%) were damaged. Two species of moth (Apantesis celia, Archips sp.), one species of sawfly (Nematus sp.) and one species of leaf miner (Diptera, section Acalyptratae) were found. Both A. celia and Archips sp. fed on C. passer- inum leaves. Larvae of Archips sp. also destroyed two flowers by chewing through the ovary. Despite an intensive search, these larvae were not found on any other species in the vicinity of C. passerinum plants; however, they were successfully reared on Linnaea borealis (Twin-flower), a nearby plant, indicating that they were not species-specific feeders. The sawfly larva chewed holes in the leaves and ate half the ovary of one plant. The leaf miners were found two to a leaf between the epidermal layers. Habitat Across Canada, C. passerinum is generally a plant of stream banks and lakeshores — habitats which are unstable but always present (Luer 1975). Both north and south of the Pic River, C. passerinum occurs in dune complexes formed from lactustrine sands depos- ited by the Pic River during glacial times (Zoltai 1967). The dunes north of the Pic River form the largest shoreline dune complex on the Canadian shore of Lake Superior. It extends for 1200 mina band 300 m wide with dunes reaching heights of up to 20 m above the mean water level of Lake Superior. Part of this complex is stabilized by a forest of Picea glauca with some Thuja occidentalis (White Cedar). Other areas are stabilized by a variety of species including Ammo- phila breviligulata (Beach Grass), Elymus mollis (Wild Rye), Equisetum hyemale (Scouring Rush), Populus balsamifera (Balsam Poplar), Cornus stolo- nifera (Red Osier Dogwood), R. acicularis, Juniperus horizontalis (Creeping Cedar), J. communis (Juniper) and Arctostaphylos uva-ursi (Bearberry). In a few areas, unstable sand is advancing upon the forest behind the dunes. Cypripedium passerinum was found ina variety of areas ranging from dunes stabil- ized by A. breviligulata to the shady, sheltered forest behind the dunes. South of the Pic River, in Pukaskwa National Park, DAD: the sand dune complex is smaller ina real extent and has less topographic variation. Most of the dunes are forested by P. glauca with some Larix laricina (Larch) and T. occidentalis. Between the lake and the forest, and in open pockets within the forested dunes, A. uva-ursi, J. communis, J. horizontalis, P. balsamifera, Shepherdia canadensis (Buffalo Berry), E. hyemale and Prunus pumila (Sand Cherry) stabilize the sand. Other typical sand dune species such as Arabis lyrata (Rock Cress), Festuca saximontana (Rocky Moun- tain Fescue) and Anemone multifida (Anemone) are also present. Beneath the trees, moss carpets (largely Pleurozium schreberi) are commonly formed where the terrain is flat. Cypripedium passerinum was found in a variety of areas, ranging from open sand between trees to mossy P. glauca forest. Table | shows that C. passerinum is most frequently associated with L. borealis, Carex concinna (sedge), Pyrola secunda (One-sided Pyrola) and Pleurozium schreberi, and only two lichens. Mature plant vigour, an indicator of habitat suita- bility, varied among groups. Plants in most groups had green leaves, robust rhizomes and vertical stalks. Four groups, in open dry areas and densely shaded areas (the two successional extremes), were less vigorous. The five groups with seedlings occurred in a variety of habitats, but a narrower range of habitats than is occupied by mature plants. One seedling site occurred TABLE |. Vascular plants, mosses and lichens associated with C. passerinum (n= 40). Species Frequency (%) Vascular Plants Linnaea borealis (Twin-flower) 58 Carex concinna (Sedge) 38 Pyrola secunda (One-sided Pyrola) 32 Equisetum hyemale (Scouring Rush) 30 Juniperus communis (Juniper) 30 Fragaria virginiana (Strawberry) 28 Cornus canadensis (Bunchberry) 25 Arctostaphylos uva-ursi (Bearberry) 18 Mitella nuda (Mitrewort) 18 Goodyera repens (Rattlesnake Plantain) 15 Rosa acicularis (Rose) 15 Solidago hispida (Goldenrod) 15 Mosses Pleurozium schreberi 60 Abietinella abietina 21 Drepanocladus uncinatus 14 Plagiomnium cuspidatum 14 Lichens Cladonia rangiferina 8 Cetraria islandica 3 THE CANADIAN FIELD-NATURALIST Vol. 97 under a P. glauca canopy with the ground covered by a thick layer of needles, a second occurred on a dune crest stabilized by A. uva-ursi and J. communis witha partial P. glauca canopy, a third under a canopy of J. communis with needle litter, a fourth on a dune with A. breviligulata and the last ina mixture of sand and litter in a clearing in the transition from forest to open dunes. Thus, seedings were found in early successional habitats in the dune complex. None were found at sites with moss cover. Variation in the suitability of sites occupied by seed- lings was also evident. Very dry weather in July resulted in browning and shrivelling of all seedlings at the site on the open grass-covered dunes and many of those on the dune crest established by A. wva-ursi and J. communis. Although the above-ground portion of the plants died, the rhizome and roots may have sur- vived. Seedlings remained green where some canopy protection was available. Discussion Although the flowers of C. passerinum are appar- ently designed to attract insect pollinators (Stouta- mire 1967), the high percentage of flowers setting capsules (99%, 198/200 flowers not attacked by insects or the unknown agent) suggested that self- pollination might be occurring. This was supported by the results of the pollination experiments which showed that C. passerinum is self-compatible. A detailed investigation of flower morphology (Catling 1980) showed that, in contrast to other Cypripedium species (e.g. C. reginae, Showy Lady-slipper), the anthers of C. passerinum are in contact with the mar- gin of the stigma allowing self-pollination. Self-pollination is rare in orchids (Sanford 1974). In species such as Epipactis helleborine (Helleborine) where 100% capsule set occurs regularly, it is sus- pected that self-pollination occurs. In other species, such as Platanthera blephariglottis (White Fringed Orchid), per cent capsule set may be commonly low (A. A. Reznicek, personal communication). Several years of observations on C. reginae indicate that typi- cally less than 5% of the flowers set capsules (G. Harvais, personal communication). Curtis (1954) determined per cent capsule set for several Cypripe- dium species over twenty years with the following results: C. calceolus var. parvi-florum (Small Yellow Lady-slipper), 49%; C. calceolus var. pubescens (Large Yellow Lady-slipper, three habitats), 40-70%; C. reginae, 23%; C. candidum (White Lady-slipper), 22%. In comparison to these figures, per cent capsule set recorded for C. passerinum is very high (86%, 198/230, taking into account those flowers damaged naturally). According to Pijl and Dodson (1966), the known 1983 KEDDY, KEDDY AND PLANCK pollinators of other Cypripedium species are bees. They list the following examples: C. arietinum (Ram’s Head Lady-slipper)—Megachile sp., C. calceolus (Yel- low Lady-slipper)—Andrena sp., C. calceolus var. par- viflorum—Ceratina sp., Zaodontomerus sp.. Based on the characteristics of typical bee flowers that they give, C. passerinum may also be visited by bees although none were observed during this study. Pollinators, if any, would be of a size to escape via the flower exit holes (2.5 X 3.0 mm). Flower mortality caused by both insect damage (1%) and the unknown agent (12%) is insignificant in determining the success of sexual reproduction. While sexual reproduction is required for group establish- ment, its importance in group maintenance is small as shown by the small percentage of flowering groups with seedlings (12%). The number of years required from germination to flowering presumably varies with environmental con- ditions. The period required for other Cypripedium species has been recorded by Curtis (1943): C. acaule (Stemless Lady-slipper), 10-12 yr; C. reginae, 14-16 yr; C. calceolus var. pubescens, 9 yr. It appears that C. Passerinum may require at least 15 years to flower. This is consistent with the observations of other spe- cies. While stalk production and the resulting stalk scars on the rhizome can be used as age indicators for older plants, it has been suggested that seedlings could be aged by counting the constrictions on the roots (Curtis 1943). In C. passerinum however, the roots were uniform, showing no constrictions; thus this method cannot be used. Nor can vegetative plants be aged by counting the number of leaves produced. Except for minor insect damage, herbivores appeared to have little effect on C. passerinum. Her- bivores known to feed upon other Cypripedium spe- cies include deer (Stuckey 1967), mice, snails and slugs (Reynolds 1976). The dunes at the mouth of the Pic River provide a wide range of stages in sand dune succession, ranging from open, unvegetated sand through dunes with A. breviligulata to dunes densely forested with P. glauca and L. Jaricina with a thick moss carpet. All seedlings occurred at sites representing early stages of succes- sion, with needle litter or sandy surfaces. Despite abundant seed production at moss-covered sites, no seedlings were observed there, suggesting that condi- tions at these sites may no longer be suitable for seed germination. (Curtis (1943) found that seedlings of C. reginae which germinated in Sphagnum moss turned brown and died before they reached two years of age. He suggested that the acidity of the moss was too great for development and observed that germination was best on the peat surface below the moss.) Grubb (1977) discussed the importance of the : SPARROW’S EGG LADY-SLIPPER DAS, regeneration niche in explaining the distribution and abundance of plants, emphasizing the rarity of ideal conditions for recruitment. Orchids generally have rather specific seed germination requirements (Arditti 1967; Harvais 1973; Stoutamire 1974; Ernst et al. 1970). Given that C. passerinum sets abundant, easily dispersed seed, but that only 12% of the present flow- ering groups contain seedlings, it seems that C. passer- inum recruitment may be largely limited by the rarity of suitable conditions for seedling establishment. Rar- ity in time must be considered as well as rarity in space. Some sites within the dunes may well be more suitable than others, but such spatial variation may be inconsequential relative to the temporal variation in rainfall, temperature and soil characteristics. Thus the distribution and population size of C. passerinum might be limited more by the rarity in time than the rarity in space of suitable conditions for establishment. Morrison and Yarranton (1974) have emphasized the vegetational heterogeneity found during primary sand dune succession. Our observations suggest the following general relationship between colony devel- opment and succession. As succession proceeds, habi- tat quality of a site for seedling establishment and growth decreases while habitat quality for mature plants increases, up to a point, and then decreases. The relationship between succession and plant distri- bution on the dune complex is such that only a portion of the complex at any one time may bea habitat for C. Passerinum and, of this portion, only a small fraction may be suitable for seedling establishment. To better understand the present distribution, research on (1) seedling habitat requirements, (2) the fungal sym- biont(s), and (3) survivorship in different habitats is necessary. In protecting species that occur in an environmen- tal mosaic, such as C. passerinum, it is not sufficient to merely protect those sites which are presently suitable for the species. The area must be large enough to provide similar sites as the environment changes over time with climatic fluctuations or vege- tation succession. That is, the protected area must be large enough to accommodate changes in both space and time. The larger the area available to the species, the greater the probability that C. passerinum will survive at the Pic River. Acknowledgments We thank H. Goulet, H. J. Teskey, D. Lafontaine and S. Allyson of Agriculture Canada (Ottawa) for identifying the insect herbivores, and R. R. Ireland and P. Y. Wong of the National Museum for identify- ing the associated mosses and lichens. P. M. Catling provided useful discussions prior to commencing the 274 study, and J.L. Riley kindly allowed us to use unpub- lished distribution data. Comments onthe manuscript by P. M. Catling and J. D. Ambrose were most helpful. The co-operation and assistance of the staff of Pukaskwa National Park was much appreciated. This work was supported by a contract from Parks Canada. Literature Cited Arditti, J. 1967. Factors affecting the germination of orchid seeds. Botanical Review 33: 1-97. Argus, G. W., and D. J. White. 1977. The rare vascular plants of Ontario. National Museums of Canada, Syllo- geus No. 14. Argus, G. W., and D. J. White. Editors. 1982. Atlas of the rare vascular plants of Ontario. Part I. National Museum of Natural Sciences, Ottawa, Ontario. Catling, P.M. 1980. Autogamy in northeastern North American orchids. Botanical Society of America Miscel- laneous Publication 158: 20 (Abstract). Curtis, J. T. 1943. Germination and seedling development in five species of Cypripedium. American Journal of Botany 30: 199-206. Curtis, J. T. 1954. Annual fluctuation in rate of flower pro- duction by native Cypripediums during two decades. Bul- letin of the Torrey Botanical Club 81: 340-352. Ernst, R., J. Arditti, and P. L. Healey. 1970. The nutrition of orchid seedlings. American Orchid Society Bulletin 39: 599-605, 691-700. Fernald, M. L. 1950. Gray’s Manual of Botany. 8th Edi- tion. Van Nostrand Co., New York, New York. Gleason, H. A. 1952. Illustrated Flora of the Northeastern United States and Adjacent Canada. Hafner Publishing Co. Inc., New York, New York. Grubb, P. L. 1977. The maintenance of species-richness in plant communities: the importance of the regeneration niche. Biological Reviews of the Cambridge Philosophical Society 52: 107-145. Harvais, G. 1973. Growth requirements and development of Cypripedium reginae in anexic culture. Canadian Journal of Botany 51: 327-332. THE CANADIAN FIELD-NATURALIST Vol. 97 Kershaw, L.,andJ. K. Morton. 1976. Rareand potentially endangered species in the Canadian flora —a preliminary list of vascular plants. Bulletin of the Canadian Botanical Association 9: 26-30. Luer, C. A. 1975. The Native Orchids of the U.S. and Canada, Excluding Florida. The New York Botanical Garden, New York, N.Y. Macoun, J. 1888. Catalogue of Canadian Plants, Part IV. — Endogens. Dawson Bros., Montreal, Quebec. Morrison, R. G., and G. A. Yarranton. 1974. Vegetational heterogeneity during a primary sand dune succession. Canadian Journal of Botany 52: 397-410. Pijl, L. van der, and C. H. Dodson. 1966. Orchid Flowers, Their Pollination and Evolution. University of Miami Press, Coral Gables, Florida. Reynolds, C. F. 1976. Growing “Lady-slippers” in a wild garden. American Orchid Society Bulletin 45: 480-483. Sanford, W. W. 1974. Theecology of orchids. Pp. 1-100in The Orchids, Scientific Studies. Edited by C. L. Withner. J. Wiley and Sons, New York, N.Y. Soper, J. H. 1965. Rediscovery of Cypripedium passeri- num in the Lake Superior region. Michigan Botanist 4: 91-92. Stoutamire, W.P. 1967. Flower biology of the lady’s- slippers (Orchidaceae: Cypripedium). Michigan Botanist 6: 159-175. Stoutamire, W. P. 1974. Terrestrial orchid seedlings. Pp. 101-128 in The Orchids, Scientific Studies. Edited by C. L. Withner. J. Wiley and Sons, New York, N.Y. Stuckey, I. H. 1967. Environmental factors and the growth of native orchids. American Journal of Botany 54: 232-241. Zoltai, S.C. 1967. Glacial features of the north-central Lake Superior region, Ontario. Canadian Journal of Earth Sciences 4: 515-528. Received | April 1982 Accepted 10 December 1982 Comparison of Great Blue Heron, Ardea herodias, Reproduction at Boot Island and Other Nova Scotia Colonies TERRY E. QUINNEY Department of Biology, Acadia University, Wolfville, Nova Scotia BOP 1X0 Present address: Department of Zoology, University of Western Ontario, London, Ontario N6A 5B7 Quinney, Terry. E. 1983. Comparison of Great Blue Heron, Ardea herodias, reproduction at Boot Island and other Nova Scotia colonies. Canadian Field-Naturalist 97(3): 275-278. Nesting behaviour and reproductive success of Great Blue Herons (Ardea herodias) on Boot Island, Nova Scotia, are reported for 1977 and 1978. Forty-two pairs bred in 1977 and 26 pairs in 1978. Mean clutch sizes were 4.6 and 5.0 and averages of 2.6 and 3.1 young were fledged per pair in 1977 and 1978, respectively. Additional colonies censused showed that fledging success in Nova Scotia was among the highest reported. Great Blue Herons were sensitive to human disturbance, particularly early in their nesting cycle. However, marine islands in Nova Scotia that usually are free from human disturbance and less susceptible to predation but are close to adequate feeding areas, provide superior heron breeding habitat. Key Words: Great Blue Heron, Ardea herodias, breeding, clutch size, fledging success, human disturbance, predation. Great Blue Heron fledging success reported for Tobacco Island, Nova Scotia (McAloney 1973) is among the highest reported (Vermeer 1969; Pratt, 1970, 1972, 1974; Henny and Bethers 1971; Henny 1972; Werschkul et al. 1977; English 1978; Des- Granges et al. 1979; DesGranges and Laporte 1979, 1981). Approximately 70% of the known Great Blue Heron colonies in Nova Scotia, including Tobacco Island are on marine islands (A.D. Smith, Canadian Wildlife Service, personal communication). Is Tobacco Island representative of Great Blue Heron colonies which are free from human disturbance and close to adequate foraging areas? To answer this ques- tion, I chose to study marine island colonies that had similar (Bon Portage Island) and contrasting envir- onments (Boot and Pinnacle Islands) to that of Tobacco Island. Because only McAloney (1973) and Pratt (1974) attempted to census all nests at various times in a given colony, information is still lacking about the influence upon numbers of young fledged of such variables as clutch size, hatching success, preda- tion, annual fluctuation in numbers breeding, size, and location of colonies. I provide data on these vari- ables, determine the number of young fledged in sev- eral Nova Scotia Great Blue Heron colonies and com- pare these results with data from other locations. Study Area Boot (45°08’N,64° 16’W) and Pinnacle (45°23’N, 64°07’W) Islands are located in the Minas Basin of the Bay of Fundy. Waters surrounding these islands are extremely turbid owing to the nature of sediments and strong tidal action. In contrast are Bon Portage (43° 28’N,65°45’W) and Tobacco (45°01’N,62°55’W) Islands that lie about 3-4 km off the coast in the Atlantic Ocean. Waters around these islands are rela- tively clear as there is little suspended sediment. Also, spring tides rise as much as 15 m around Boot and Pinnacle Islands but only 4.5 m above the low water mark around Bon Portage and Tobacco Islands. This is particularly important for the Minas Basin herons. For example, numerous tidal pools and streams are exposed at low water in three estuaries near Boot Island and are used for foraging (Quinney and Smith 1980). Great Blue Herons nested 4-12 m above ground in the tops of White Spruce, Picea glauca, or Balsam Fir, Abies balsamea, trees on all of the islands. All of these islands have been occupied by breeding herons for at least 10 years and all were uninhabited by people except for lighthouse keepers on Bon Portage Island. Methods To determine the fate of as many eggs and nestlings as possible, nesting Great Blue Herons were observed almost daily on Boot Island in 1977 after egg-laying began. Two hundred and thirty-five hours were spent observing unobstructed nests from blinds that had been set up on the ground prior to nest construction. Nests were also examined by climbing the trees or using an extension ladder; these visits were restricted to a maximum of 2 hours per day. In 1978, I examined nests to determine the onset of laying but postponed more intensive study until most laying was completed and incubation was well underway. I also measured some nestlings every 2-3 days in both years to follow their growth (Quinney 1982). The Bon Portage Island heronry was censused twice during the 1977 breeding season (31 May and 6 July) and at two week intervals in 1978. An observation blind, 10 m above ground, was erected in the centre of the Bon Portage Island pie) 276 colony in October 1977, after the departure of the herons, for observations during the following breed- ing season. The Pinnacle Island heronry was censused on 29 June 1978, approximately one week before the first nestlings were expected to fledge. Chi-square analysis and Mann-Whitney U-tests (Sokal and Rohlf 1969) were used to determine statistical significance. Results Unless otherwise stated, results pertain to Boot Island. First eggs were laid on 13 April 1977 and 20 April 1978. Forty-two breeding pairs initiated clutches in 1977; 20 clutches were lost owing to Com- mon Crow, Corvus brachyrhynchos, predation, and 14 of these pairs laid replacement clutches in new nests. This resulted in 36 nests from which at least one nestling hatched. One hundred and sixty-seven eggs were laid in the 36 successful nests and 76 were laid in the 20 nests that failed to produce any young. I deter- mined the outcome of all but 18 of these 243 eggs. In 1978, 26 breeding pairs produced 26 clutches and 25 pairs produced at least one nestling. The outcome of all but 13 of the 131 eggs laid was known. Hatching dates were known or extrapolated for 70% and 92% of all the nestlings in 1977 and 1978, respectively. Forty- three and 35% of all nestlings hatched during 25-28 May in 1977 and 1978, respectively. However, despite this relative synchrony of the peaks in hatching between years, hatching was later in 1978 (X2 = 40.6, p < 0.001; Table 1). Intervals from first to last hatch- ing of eggs ina givenclutch were 4-8 days in five nests from 1977 and 1978. Maximum ages at which eight nestlings first flew from the colony were 48, 51, 52, 54(2), 57(2), and 59 days. Newly-fledged juveniles joined the adults on their foraging grounds by the last week in July. Further details of nesting phenology and behavioural development of nestlings are presented elsewhere (Quinney 1979). Twenty to 27% of the eggs failed to hatch in Boot Island nests where at least one nestling hatched. How- ever, most of these failures were not due to predation, and nestling survivorship to fledging was 80-86% (Table 2). There were no differences (p > 0.05) between years in the number per nest, of eggs that hatched, that failed to hatch, or were lost to predators, or were missing, or in the numbers of nestlings that fledged or died. Clutch sizes and fledging success of TABLE |. Nestling hatch dates on Boot Island. May June 9-16 17-24 25-1 2-9 1977 11 34 47 Dy) 1978 0 14 44 27 THE CANADIAN FIELD-NATURALIST Voleo7 TABLE 2. History of Boot Island nests where Great Blue Heron nestlings hatched. Percentage of total eggs laid in parentheses. 1977 1978 Nests 36 25 Eggs laid 167 126 Eggs hatched 134 (80) 93 (74) Eggs predated 7 (4) 2 (2) Eggs disappeared 13 (8) 10 (8) Other egg losses! 13 (8) 21 (17) Nestlings fledged 107 (64) 80 (63) Nestlings dead 27 (16) 13 (10) 'Eggs addled, damaged by adults, broken by handlers, and nestlings dead in pipped eggs. Boot Island Great Blue Herons are compared with other Nova Scotia marine island colonies in Table 3. Some of the 1977 Boot Island clutches were from replacement clutches and this probably was the major factor contributing to the lower clutch size (p < 0.01) in 1977 compared with 1978. Replacement clutches are often smaller than initial ones (Lack 1954: 32). Clutch sizes were also lower on Tobacco Island in 1971 (McAloney 1973) than on Boot Island in either 1977 or 1978 (p< 0.01). However, 22% of the nests that later had nestlings on Tobacco Island had clutches of three eggs. On Boot Island, no incubated nests contained a clutch of less than four eggs. Colony sizes, clutch sizes and fledging success of Great Blue Herons in Nova Scotia are compared with other local- ities in Table 4. Fledging success in Nova Scotia has higher than averages reported elsewhere. Discussion Fledging success of Boot Island Great Blue Herons was among the highest reported. Furthermore, with the exception of Bon Portage Island in 1978 (when only 10 pairs bred), the numbers of young fledged from the four Nova Scotia colonies mentioned was very similar even in different years. Tobacco Island thus appears to be representative of comparably-sized marine island colonies. On Boot Island in 1977, my presence in the colony during egg laying facilitated crow predation of heron eggs. If disturbed before the clutch was complete or serious incubation had begun, the herons took flight from their nests sooner and remained absent for longer periods than during incu- bation (see also Milstein et al. 1970). In 1978, I made brief visits to the Boot Island colony to determine the number of breeding pairs and the onset of laying but daily nest examinations were postponed until most laying was completed. In my opinion, this approach was responsible for the much lower incidence of pre- dation during egg laying in 1978. Tremblay and Elli- 1983 QUINNEY: GREAT BLUE HERON REPRODUCTION ANT) TABLE 3. Clutch sizes and fledging success of Great Blue Herons in Nova Scotia. Mean clutch size Location Year 1+SD Tobacco Island? 1971 4.2+0.9 Pinnacle Island 1978 — Bon Portage Island 1977 4.8 + 0.6 1978 47+ 0.8 Boot Island 1977 4.6+ 0.6 1978 5.0 + 0.5 No. fledged / No. fledged Breeding breeding successful pairs pair nest! 42 2.6 341 4] Da} 2.8 50 3.0 Bw 10 1.0 1.7 42 2.63 3.0 26 3.1 3.3 'Nests where at least one young fledged. 2McAloney 1973. 3Assumes renesting by 14 pairs after Common Crow predation. son (1979) showed that visits to Black-crowned Night Heron Nycticorax nycticorax colonies just before and during laying provoked abandonment of newly con- structed nests and either predation of eggs or aban- donment of eggs followed by predation. Crows were the principal egg predators. The disturbances I created during egg laying in 1977 on Boot Island may also have contributed to the decline in the number of breeders during the following year. Tremblay and Ellison (1979) also showed a higher percentage of empty nests in frequently visited colonies. The pres- ence of the aerial observation blind in the Bon Portage colony in 1978 may have contributed to the decreased number of breeders compared with the previous year. The poor success of herons on Bon Portage Island in 1978, when most egg losses were caused by predation, may have been related to the small number of breed- ing pairs. DesGranges (1979) suggested that one of the advantages of colonial nesting in Great Blue Herons was that larger colonies offered more efficient protec- TABLE 4, Great Blue Heron reproduction in North America. Mean no. Latitude No. breeding Location (CN) colonies pairs/colony California 38 I 52 Oregon 44 26 — 43-45 5 93 44-45 7 35) 45 | 27 Nova Scotia 43-45 4 35 Quebec 45-49 22 29 Alberta 49-56 27 21 'Nests where at least one young fledged. *Based on 32 nests in a colony of 55 nests. 3Based on | colony only. tion of the clutch because the proportion of peripheral nesters in the colony decreases with increased size of the colony. The apparent relationship between human distur- bance and numbers breeding is not a simple one. Numbers in the Boot Island colony decreased from 50 to 26 pairs between 1970 and 1976 when the colony was censused only once annually, late in the nestling period, to determine numbers of breeders (C. K. Coldwell, Acadia University, personal communica- tion). Inthe colony of Audubon Canyon Ranch, Cali- fornia, there were decreases of 30% between 1968 and 1971 in numbers breeding (Pratt 1974), even though Pratt observed nests from an observation point 90- 185 m distant. The colony was never entered and thus effects of disturbance by the observer appeared neglig- ible. Mallette (1972; cited in Pratt 1974), froma state- wide survey of heronries conducted from 1969-1971, showed that of 109 heronries active in 1971, only 32 had been active in all three years. Lack (1954) com- Mean no. fledged / Mean clutch breeding successful size pair nest! Source 3.4 1.6 2.) Pratt (1974) 4.2 2.0 2.62 Henny & Bethers (1971) — — 2.4 Werschkul et al. (1977) = 2.0 2.4 English (1978) 4.5 1.9-2.3 2.3-2.7 Blus et al. (1980) 4.7 DES 258) McAloney (1973) Quinney (this study) 4.5 Il) 23) DesGranges and Laporte (1981) 5.0 2.2-2.3 2.63 Vermeer (1969) 278 mented that switching between colonies of some pairs of Grey Herons Ardea cinerea occurs, and mortality on the wintering grounds and during migration could further decrease breeding populations. Assuming that Great Blue Herons begin breeding as 2-year-olds, Henny (1972) estimated that 1.9 young must be fledged per breeding pair to maintain stable popula- tions in the northern United States and southeastern Canada. The Nova Scotia colonies mentioned in this paper fledged an average of 2.5 young per pair. How- ever, the number of breeding pairs on Boot and Bon Portage Islands fluctuated considerably from year to year. Although the effects of human disturbance should not be understated, these fluctuations are probably the result of a number of factors. I urge heron observers to exercise caution, particularly early in the nesting cycle of these birds. Nevertheless, the high fledging success suggests that marine islands in Nova Scotia, that are largely free from human distur- bance and thereby less susceptible to predation but are close to adequate foraging areas, provide superior heron habitat. Acknowledgments I thank P. C. Smith, Kees Vermeer, C. D. Ankney, S. G. Sealy, and C. J. Henny for valuable editorial assistance. I appreciate the efforts of the following in the field: S. K. Mainguy, B. N. Miller, P. C. Smith, J.S. Boates, C. K. Coldwell, R. D. Elliot, P. W. Hicklin, G. R. Milton, R. J. Sowerby, D. T. Lickley, G. L. Hansen, J. M. Porter, D.C. Boersma, and K. R. S. Quinney. This study was financially sup- ported by the Canadian Wildlife Service, National Sciences and Engineering Research Council, and Federal Summer Job Corps Program. Literature Cited Blus, L. J., C. J. Penny, and T. E. Kaiser. 1980. Population ecology of breeding Great Blue Herons in the Columbia Basin, Oregon and Washington. Murrelet 61: 63-71. DesGranges, J.-L. 1979. Adaptive value of social behav- iour in the Great Blue Heron Ardea herodias. Proceedings of the 1978 Conference of the Colonial Waterbird Group 2: 192-201. DesGranges, J.-L., P. Laporte, and G. Chapdelaine 1979. First tour of inspection of Quebec heronries, 1977. Canadian Wildlife Service Progress Note No. 93. DesGranges, J.-L., and P. Laporte. 1979. Second tour of inspection of Quebec heronries, 1978. Canadian Wildlife Service Progress Note No. 105. THE CANADIAN FIELD-NATURALIST Vol. 97 DesGranges, J.-L., and P. Laporte. 1981. Third tour of inspection of Quebec heronries, 1979. Canadian Wildlife Service Progress Note No. 123. English, S. M. 1978. Distribution and ecology of Great Blue Heron colonies on the Willamette River, Oregon. National Audubon Society Research Report 7: 235-244. Henny, C. J.,and M. R. Bethers. 1971. Population ecology of the Great Blue Heron with special reference to western Oregon. Canadian Field-Naturalist 85: 205-209. Henny, C. J. 1972. Analysis of the population dynamics of selected avian species. U.S. Fish and Wildlife Service Wildlife Research Report 1. Lack, D. 1954. The stability of the heron population. Brit- ish Birds 47: 111-121. McAloney, K. 1973. The breeding biology of the Great Blue Heron on Tobacco Island, Nova Scotia. Canadian Field-Naturalist 87: 137-140. Milstein, P. le S., 1. Prestt,and A. A. Bell. 1970. The breed- ing cycle of the Grey Heron. Ardea 58: 171-257. Palmer, R. S. 1962. Handbook of North American birds. Volume 1. Yale University Press, New Haven. 567 pp. Pratt, H. M. 1970. Breeding biology of Great Blue Herons and Common Egrets in central California. Condor 72: 407-416. Pratt, H. M. 1972. Nesting success of Common Egrets and Great Blue Herons in the San Francisco Bay region. Con- dor 74: 447-453. Pratt, H. M. 1974. Breeding of Great Blue Herons and Great Egrets at Audubon Canyon Ranch, California, 1972-1973. Western Birds 5: 127-136. Quinney, T. E. 1979. Reproductive success, growth of nes- tlings and foraging behaviour of the Great Blue Heron. M.Sc. Thesis, Acadia University, Wolfville, Nova Scotia. 143 pp. Quinney, T. E.,and P. C.Smith. 1980. Comparative forag- ing behaviour and efficiency of adult and juvenile Great Blue Herons. Canadian Journal of Zoology 58: 1168-1173. Quinney, T. E. 1982. Growth, diet and mortality of nestling Great Blue Herons. Wilson Bulletin 94: 571-577. Sokal, R.R., and F. J. Rohlf. 1969. Biometry. W.H. Freeman and Company, San Francisco. 776 pp. Tremblay, J., and L. M. Ellison. 1979. Effects of human disturbance on breeding of Black-crowned Night Herons. Auk 96: 364-369. Vermeer, K. 1969. Great Blue Heron colonies in Alberta. Canadian Field-Naturalist 83: 237-243. Werschkul, K., E. McMahon, M. Leitschuh, E. English, C. Skibinski, and G. Williamson. 1977. Observations on the reproductive ecology of the Great Blue Heron (Ardea herodias) in western Oregon. Murrelet 58: 7-12. Received 2 November 1982 Accepted 2 May 1983 Vascular Plants of the Hayes Sound Region, Ellesmere Island, Northwest Territories! J. BRIDGLAND?2 and J. M. GILLETT 'Contribution No. 16 from the Cape Herschel Project, Geological Survey of Canada. 2Department of Biology, Memorial University, St. John’s, Newfoundland A1B 3X9 3Botany Division, National Museum of Natural Sciences, Ottawa, Ontario KIA 0M8 Bridgland, J., and J. M. Gillett. 1983. Vascular plants of the Hayes Sound region, Ellesmere Island, Northwest Territories. Canadian Field-Naturalist 97(3): 279-292. Recent collections of 109 species bring the total known vascular flora of the Hayes Sound Region to 117 species. Of 22 species new to the region, Carex amblyorhyncha and C. scirpoidea are reported for the first time from Ellesmere Island. Analysis of factors controlling the distribution of plants in the region indicates that the current climate is the principal factor contributing to the diversity of the flora. Key Words: Arctic, vascular plants, floristics, continentality, Hayes Sound, Ellesmere Island. The Hayes Sound region (79°N, 77° W) on east central Ellesmere Island, has been of interest to botan- ists since the first collection of plants were made there over 120 years ago. Few records, however, have been published for the region since the early 1900’s. We report here the floristic results of three independent botanical studies made in the Hayes Sound region during the summers of 1979-1980. J. B. was based at Cape Herschel, from 20 June to 29 July 1979, and again from 19 May to 16 June 1980. J. M. G. accom- panied by M. J. Shchepanek, visited Alexandra Fiord from 12 to 24 July 1979, in order to conduct floristic studies in the region. In 1980 a group from the Univer- sity of Toronto and Dalhousie University initiated a several-year ecosystem study at Alexandra Fiord. Their floristic results (Hill and Ball. 1980. Vascular plants at Alexandra Fiord. Pp. 105-108 in Ecology of a high arctic lowland oasis, Alexandra Fiord, (78° 53’N, 75° 65’W), Ellesmere Island, N.W.T., Can- ada. 1980 Progress Report. Edited by J. Svoboda and B. Freedman. Departments of Botany, University of Toronto and Dalhousie University. 109 pp.), are included here. Physiography The Hayes Sound Region (Figure 1) occupies approximately 15000 km? of central Ellesmere Island. Physiographically, the region comprises a ser- ies of plateaux, dissected by deep glacial fiords and outlet glaciers of the central Ellesmere Island ice cap. The ice cap rises to altitudes in excess of 2 000 m and occupies nearly two-thirds (9500 km?) of the lands- cape. The region lies at the northern tip of the Pre- cambrian Shield. The crystalline basement rocks, including gneiss, granulite, amphibolite and granite, are unconformably overlain by Proterozoic and Paleozoic sedimentary rocks — sandstone, shale, dolomite, limestone, conglomerate and breccia. The basement rocks are exposed at the head of Sverdrup Pass, and along the east coast, south of Knud and Bache Peninsulas (Christie 1962, 1967, 1972; Frisch et al. 1978). In addition to being glaciated by the ice cap to the west, areas on the east coast have also been inundated by glacier ice flowing from the north. This flow deposited tills, including calcareous Paleozoic rocks and marine sediments on top of the plateaux — at 285m on Cape Herschel and at 550m on Pim Island (Christie 1962; Blake 1977). Studies of raised coastal features and lake sediments (Blake 1978, 1981) indicate that terrestrial habitats were available for plant colonization by approximately 9000 years B.P. History of Exploration Central Ellesmere Island is not now inhabited, but excavations in the Bache Peninsula region have shown it was occupied by Inuit of the Arctic Small Tool tradition from more than 4000 years B.P. until approximately A.D. 1000, and subsequently by early Thule-culture Inuit of western arctic origin until 1700 A.D. (Schlederman 1978; Schlederman and McCul- lough 1980). While Smith Sound, named by Baffin and Bylot in 1616, was probably visited by the Euro- pean whalers who hunted the North Water of Baffin Bay from the mid-1700’s to the mid-1800’s, records of its navigation date from 1852 when Inglefield sailed briefly into Kane Basin searching for Franklin(Taylor 1964). The first white man to report visiting Ellesmere Island was Dr. I. I. Hayes, surgeon on Kane’s expedi- tion of 1853-55, who sledged across to Cape Fraser on the northwest shore of Kane Basin from that expedi- tion’s winter quarters in Greenland. He discovered Hayes “Sound” between Knud and Thorvald Peninsu- 279 280 las and believed it ran across what is now known as Ellesmere Island, separating Ellesmereland, to the south, from Grinnell Land, to the north. While the existence of the “sound” was disproved by both Peary and Sverdrup, the name Hayes Sound persisted in botanical literature until the late 1940’s to describe the southwest coast of Kane Basin and we see no reason to discard it. Botanical exploration on the east coast of Ellesmere Island dates from 1861 when Hayes (1867) noted Salix arctica, Saxifraga oppositifolia and Festuca ovina at Cape Fraser. At Cape Isabella, he somewhat raptu- rously reported grass, moss, poppies, and buttercups, and collected “a nosegay of bright flowers, which are so pleasing an association that they will not find place in the ‘botanical collections’, but rather in another collection, -mementos” (Hayes 1867, p. 423). These in °o THE CANADIAN FIELD-NATURALIST 78° Vol. 97 fact appear to be the only collections he brought back from Ellesmere. Durand (1863[4]) published records of nine species collected by Hayes at Cape Isabella and Gale Point. H. C. Hart, naturalist on the British expedition of 1875-76 led by G.S. Nares, made a major contribution to botanical knowledge of the Hayes Sound area (Hart 1880). He collected 31 species from Cape Sabine at the east end of Pim Island and 56 species from four localities on the south shore of Hayes Sound. Hart found a great difference between the floras of the south shore of Hayes Sound and Grinnell Land, and judged that Hayes Sound was an important barrier to plant migration. The Greely expedition wintered on the north side of Pim Island during its tragic retreat from Fort Conger in 1883, but its members were so close to death from fe} 80° f dE 82 ee 79° 20' 3 NPI Baffin Bay 50 km FicureE |. Collecting localities: AF - Alexandra Fiord Post area, 78° 53’N,75° 50’W; AFW - West end of Alexandra Fiord, 78° 49’N,76° 25’W: BI - Baird Inlet, 78° 29’N,76° 31’W; BF - head of Beitstad Fiord, 79°04’N, 78°50’W and 79°03’N, 78° 53’W: BP - Bache Peninsula (former R.C.M.P. post), 79° 06’N, 75° 45’ W; CH - Cape Herschel, 78° 37’N, 74° 42’°W; CR - Cape Rutherford, 78° 50’N, 74° 55’W; EG -Ekblaw Glacier, 78° 29’N,76° 46’ W; FB - Flagler Bay (estuary at end of Sverdrup Pass), 79° 10’N,76° 10’ W; IB - Irene Bay (Thumb Mountain area), 79° 02’N,81° 30’W; KP - Knud Peninsula, 79° ()2’N, 76° 10’W; LG -Leffert Glacier, 78° 42’N,74° 47’ W; NPI - North side of Pim Island, 78° 41’N, 74° 25’W; PI - Pim Island, 78° 41’N, 74° 25’ W; SG - Stygge Glacier nunatak, 78° 44’N,78° 29’ W; SP - Sverdrup Pass, 79° 1 1’N,79° 25’W and 79° 09'N,79° 25’W. Stippled area ice-covered. 1983 starvation that their only records of plants there were saxifrage and rock tripe used, in desperation, as food (Greely 1886). In 1898-99 the Second Norwegian Arctic Expedi- tion in the Fram, under the command of Otto Sver- drup, wintered at Fram Harbour at the west end of Pim Island. H. G. Simmons, botanist on the expedi- tion, made substantial collections of localities between Pim Island and the present site of the Alex- andra Fiord police post, and between the mouth of Flagler Bay at the head of Beitstad Fiord. In total, he reported 80 species for the region. Simmons’ work was a landmark in the botany of Arctic North America. He reviewed all previous collections from Ellesmere Island in light of his own, so that when he published his Flora of Ellesmereland (Simmons 1906), it was better known floristically than any other island of the Canadian Arctic Archipelago. His analysis of plant distributions throughout the archipelago (Simmons 1913) concluded that, while the majority of plants in the flora immigrated into the archipelago from the mainland after the last glacial maximum, areas of Ellesmere Island appeared to have harboured plants during that period. Despite numerous visitors to the region, botanical records since Simmons are sparse. A. P. Low (1906) reported a few species at Cape Sabine collected in 1903. W. E. Ekblaw, botanist on D. B. MacMillan’s 1915-17 Crocker Land Expedition, confined his col- lecting to northwest Greenland (Ekblaw 1918). M. O. Malte visited the Bache Peninsula R.C.M.P. post in 1927 while preparing, with C. H. Ostenfeld, a flora of the Canadian Arctic. The deaths of both men pre- vented completion of this work and only a few records were published posthumously (Malte 1934). Expeditions visiting the area in the 1930’s, during which plants were apparently collected (Taylor 1964), included the Oxford University Ellesmereland Expe- dition in 1934-35, the expeditions of D. Haig-Thomas in 1937-38 and of J. M. Wordie in 1937, and the Danish Thule and Ellesmereland Expedition of 1939- 40. Polunin’s (1940) summary of botanical explora- tion on Ellesmere mentions only the last of these, during which no vascular plants were collected. Haig- Thomas (1940) stated only that a large number of vascular plants were deposited “at the Natural History Museum”, presumably in London. Porsild (1955, 1964) made no reference to any collectors working on the east side of Ellesmere Island. Climate The climate of eastern Ellesmere Island is domi- nated by a semi-permanent cold circumpolar vortex situated over the Queen Elizabeth Islands. The cold arctic air mass is disrupted in the Hayes Sound Region BRIDGLAND AND GILLETT: VASCULAR PLANTS OF ELLESMERE ISLAND 281 principally by incursions of warmer, moister maritime-arctic air masses associated with cyclonic systems originating in the Labrador Sea moving north through Baffin Bay (Maxwell 1980). Members of Sverdrup’s expedition collected cop- ious weather data while overwintering in Rice Strait between September 1898 and July 1899. The mean temperature for that period was —18.9°C. The mean monthly temperature rose above freezing in June (1.67°C) and July (2.86°C) (Mohn 1907). The R.C.M.P. post at Bache Peninsula provided the first year-round records of weather for the region. From 1930 to 1933 the mean annual temperature was ~15.6° C (Department of Transport 1944). June, July, and August were the only months with mean tempera- tures above freezing, with temperatures of 2.7°, 5.0°, and 3.3° C(36.8° ,41.4°, and 37.9° F) respectively. The mean annual precipitation during this period was 6.05 cm (2.38 in) rainin July and August and 71.4 cm (28.1 in) snow falling mainly in September and November. Weather data for Cape Herschel are available for the period from June 1973 to December 1974, when it was a Station for climatological studies of the North Water Polynya in northern Baffin Bay (F. Muller, H. Blatter, R. Braithwaite, H. Ito, A. Ohmura, K. Schroff, and A. Zust. 1975. Report on North Water Project activities, | October 1974 to 30 September 1975. Swiss Federal Institute of Technology, Zurich and McGill University, Montreal; F. Muller, W. Bachmann, P. Berger, H. Blatter, R. Braithwaite, J. Crawford, C. Dutter, H. Ito, S. Ito, G. Kappenberger, H. Muller, A. Ohmura, G. Schriber, K. Schroff, H. Siegenthaler, A. Zust, and J. Weiss. 1976. Report on North Water Project activities, | October 1975 to 30 September 1976. Swiss Federal Institute of Technol- ogy, Zurich, and McGill University, Montreal.) The mean temperature during this period was —14.0°C. While mid-winter mean monthly temperatures were found to be 5° to —8° warmer than those at Resolute, reflecting the warming influences of Baffin Bay and the North Water Polynya, the summer mean tempera- tures were not appreciably different from those at Resolute or at other arctic stations. Precipitation from June 1973 to July 1974 was 265 mm, 100 mm greater than the annual precipitation at Resolute. Aviation weather reports were made concurrently at Alexandra Fiord and Cape Herschel between 11 June and 24 July 1979. During this period, Alexandra Fiord had a mean temperature of 4.0° C, with a mean daily range of 5.9°C. Cape Herschel had a mean temperature of 2.2°C, with a mean daily range of 4.2°C. Figure 2 shows the mean daily temperatures for both stations. At the same time, Cape Herschel received I12 cm of fresh snow and had traces of precipi- FIGURE 2. Mean daily temperature for Alexandra Fiord (solid line) and Cape Herschel (broken line) from 11 June to 24 July 1979. tation on seven different occasions, while Alexandra Fiord had traces of precipitation on only two occa- sions. Cape Herschel was much windier than Alexan- dra Fiord, having an average wind speed of 18.0 km/h as opposed to 2.4 km/h for the latter station. Both stations had similar amounts of total cloud cover, at an average 6/10 of the sky in 87 observations at each station. The average low cloud cover at Cape Herschel was 4.7/10, while at Alexandra Fiord it was 4.0/ 10. This mostly reflects the higher incidence of coastal fog at Cape Herschel, which had fog at the station ten times while Alexandra Fiord had fog only once. Annotated Checklist For each of the species reported here, the collector is indicated either by name or by our own collection numbers. Collections with numbers between 18050 and 18425 were made by Gillett and Shchepanek. Their first set of voucher specimens is deposited at the National Museum of Canada, Ottawa (CAN); dupli- cate sets will be distributed. Bridgland’s 1979 collec- tions have numbers between 400 and 750; his 1980 collections are prefixed by 80-. The first set of these are deposited at the Biosystematics Research Insti- tute, Ottawa (DAO). Two collections made in 1980, by Svend Funder, of the Geologisk Museum, Copen- hagen, while attached to the Geological Survey of Canada field party, are deposited with Bridgland’s collection at DAO. Critical specimens of species reported by Hill and Ball (1980) were verified by Gil- lett, and are deposited at the University of Toronto Erindale Campus (TRTE). Thane Anderson’s collec- tions from Beitstad Fiord were verified by Bridgland, and remain in Anderson’s personal collection at the Geological Survey of Canada. Collections made in 1981 by W. Blake, Jr., were identified by Bridgland, THE CANADIAN FIELD-NATURALIST Vol. 97 and are deposited at DAO. All Drabae were deter- mined or verified by G.A. Mulligan. Simmons’ (1906) records of species not collected by ourselves have been included in parentheses, where they have been verified or otherwise unconditionally accepted by both Polunin (1940) and Porsild (1964). First records for our region are preceded by an asterisk (*). Species reported for our region by Sim- mons or Polunin but not mapped as such by Porsild (1964) are preceded by a plus mark (+). With the exception of the genus Draba, where we have followed the taxonomy of Mulligan (1970, 1971, 1974a, 1974b, 1976), and in the Caryophyllaceae, we have followed the classification and nomenclature of Porsild (1964). Cystopteris fragilis (L.) Bernh. AF: 1/8/11, Hill and Ball (1980); BP: 18348A; CH: 682. Rare. Appar- ently restricted to xeric igneous rocks in our area, much as reported by Barrett and Teeri (1973) from Devon Island. Dryopteris fragrans (L.) Schott. AF: Hill and Ball (1980). Rare. Also reported by Simmons (1906), this acidophile is at the northern limit of its range in our region. (Woodsia alpina (Bolt.) S.F. Gray) Simmons’ (1906) report from Cape Viele in the vicinity of Alexandra Fiord remains the only record from Ellesmere Island (Polunin 1940; Porsild 1964). Barrett and Teeri (1973) reported it from Devon Island so it may still be found in our area. Woodsia glabella R. Br. AF: 18109, Hill and Ball (1980); AFW: 18373; CH: Blake 1981; CR: 18411. Rare, in rock crevices and gravel on cliffs and out- crops, also on dry tundra. Equisetum arvense L. AF: 18424, Hilland Ball (1980); CH: 537, 586. Occasional. In wet sedge-moss mea- dows and along streams. Equisetum variegatum Schleich. AF: 18257, 18420; CH: 589; SP: 18069. Occasional. Generally, dwarfed plants were found in wet moss and in shal- low water and once on the gravel slope of an esker. Lycopodium selago L. AF: 18141, Hill and Ball (1980); EG: Blake 1981. Occasional on dry Cassiope- Vaccinium heath and in sedge meadows. Alopecurus alpinus L. AF: 18253, Hill and Ball (1980); CH: 653, 674; FB: 18088; KP: 18224; SP: 18065. Occasional. Scattered on wet mineral soil and moss, and dominant only once where it grew with Poa glauca below a bird cliff. Arctagrostis latifolia (R. Br.) Griseb. AF: 18249, 18324, 18416, 18423, Hill and Ball (1980); CH: 590, 671. Occasional, in moss and Sa/ix hummocks in wet sedge moss meadows and at pond margins. *Calamagrostis purpurascens R. Br. SG: Blake 1981. Rare. 1983 Colpodium vahlianum (Liebm.) Nevski AF: Hill and Ball (1980); CH: 475, 503. Rare, on moist mineral soil and submerged at edge of tundra ponds. (Deschampsia pumila (Trin.) Ostf.) Simmons’ collec- tion of Aira flexuosa from Fram Harbour was re- classified by Polunin (1940) as this species. A sterile plant (480), which had the habit of the species as illustrated in Porsild (1964, figure Sf), was found growing in cracks ona salt-spray swept rocky island off Cape Herschel. *Dupontia fisheri R. Br. AFW: 18369. Rare. Sandy border of a glacial stream. *Festuca baffinensis Polunin CH: 544. South-facing raised sandy beach. +Festuca brachyphylla Schultes AF: 18414, Hill and Ball (1980); CH: 531, 763; LG: Blake 1981. Occa- sional on dry unstable sand and by ponds on raised gravel terraces. + Hierochloe alpina (Sw.) R. & S. AF: 18145, Hill and Ball (1980). Rare. On rocky, sandy moraine. Phippsia algida (Sol.) R. Br. AF: Hill and Ball (1980); LG: Blake 1981; PI: 748. Rare. At Pim Island it was found uncharacteristically on dry clay barrens on the plateau, although the site may have late snow cover. *Pleuropogon sabinei R. Br. AF: Hill and Ball (1980); CH: 430, 654. Rare. At Cape Herschel this grew as scattered single plants on wet mineral soil in seep- age and streams. - Poa abbreviata R. Br. AF: 18054, 18158, Hilland Ball (1980); BP: 18347A; CH: 456, 507, 568, 577, 631, 785; IB: 18310; KP: 18228; SP: 18083. Common on rock outcrops and gravel; occasional in moss by pools. Poa alpigena (Fr.) Lindm. var. colpodea (Fr.) Schol. AF: 18149; CH: 672. Rare in wet moss meadows. Poa arctica R. Br. AF: 18114, 18248A, 18254, 18329, 18363, Hill and Ball (1980) as ssp. caespitans (Simm.) Nannf.; CH: 435, 505, 564, 627, 634; EG: Blake 1981; LG: Blake 1981. Occasional on gravel and mineral soil in seeps and streams. Poa glauca M. Vahl AF: 18256, 18419, Hill and Ball (1980); BF: 409; CH: 80-176 B. Occasional on sandy and gravelly slopes. At Cape Herschel it was found on frost hummocks with Alopecurus on ledges and screes below a bird cliff. + Poa hartzii Gand. AF: 18106A; SG: Blake 1981. The Alexandra Fiord collection comes froma wet pond margin. +Puccinellia angustata (R. Br.) Rand & Redf. AF: 18167, Hill and Ball (1980); FB: 78099; IB: 18311; LG: Blake 1981; SP: 18195. Occasional on unstable gravel slopes and beach mud. + Puccinellia phryganodes (Trin.) Scribn. & Merr. AF: 18383A, Hill and Ball (1980); CH: 650; IB: 1831/2; BRIDGLAND AND GILLETT: VASCULAR PLANTS OF ELLESMERE ISLAND 283 KP: 18216. Occasional by muddy estuaries, sea beaches, and brackish pools. *Trisetum spicatum (L.) Richt. AF: Hill and Ball (1980); CH: Blake 1981. Rare. *Carex amblyorhyncha Krecz. AF: Hill and Ball (1980). Rare. This is the first record of this species occurring on Ellesmere Island. Porsild (1964, map 81) shows one record for the Queen Elizabeth Islands on Axel Heiberg Island. Barrett and Teeri (1973) found it on Devon Island. Carex atrofusca Schk. AF: 18271A; IB: 18293; KP: 18231; SP: 18200. Occasional. On wet gravel, muddy polygons and in moss. *Carex bigelowii Torr. AF: Hill and Ball (1980); AFW: 18378; SP: 18067. Rare. On sandy stream banks and wet tundra. These collections represent the third report of this species from Ellesmere Island. Porsild (1964, map 84) shows it at Eureka, and Brassard and Beschel (1968) report it at Tanquary Fiord. Otherwise it is known in the Queen Elizabeth Islands only from Dundas Harbour (Polunin 1940). *Carex capillaris L. AF: 18252, 18263, Hill and Ball (1980); AFW: 18372; BP: 18346A; EG: 724; IB: 18280. Occasional. Chiefly on dry tundra, gravel slopes, and sandy ledges of cliffs. At Bache Penin- sula it was found on a wet mossy pond margin. (Carex glacialis Mack.) Reported by Simmons (1906) from Feilden’s collections in Hayes Sound as C. pedata. The same specimen was assigned to C. gla- cialis by Polunin (1940) after the two species were separated by Mackenzie. Carex maritima Gunn. AF: 18176, Hill and Ball (1980); BP: /8344A. Rarer than expected. The Alexandra Fiord collection was found on a dry raised beach terrace. Bocher (1954) found it to be continental in Southwest Greenland, occurring only in the interiors of the deepest fiords. Carex membranacea Hook. AF: 18245, 18361, 18362, 18418, Hill and Ball (1980); CR: 18407; IB: 18278. Common. Wet meadows, margins of ponds and from frost cracks on gravel beaches. Carex misandra R. Br. AF: 18170, 18264, 18362A, 18412, Hill and Ball (1980); AFW: /8376; BP: 18359 CH: 45954853; 539; 947, 991 NOU, 630 Gs 703; NPI: 18402, 18403; SP: 18068. Common, locally abundant. The most abundant Carex, occupying a variety of habitats from dry to very wet tundra. Carex nardina Fr. var. atriceps Kuk. AF: 1/81/57, 18171, 18270, Hill and Ball (1980); BF: Anderson 1980; CH: 603; EG: 720; IB: 18294; NPI: 1/8399; KP: 18227, 18242; SP: 18085. Common. Almost always present as scattered tufts on arid till, moraines, and gravel and rock slopes. 284 +Carex rupestris All. AF: 18/59, Hill and Ball (1980); IB: 18302; NPI: 18401; SP: 18084. Occasional, locally abundant. At Alexandra Fiord it forms extensive mats on arid raised beach terraces along with Dryas, Salix, and Saxifraga oppositifolia. *Carex scirpoidea Michx. AF: 18105, 18260, 18261, 18271, Hill and Ball (1980). At pond margins on rocky wet tundra. This is the first record for Elles- mere Island. The only other record for the Queen Elizabeth Islands is Polunin’s (1940) report from Dundas Harbour on Devon Island. +Carex stans Dre}. AF: 18415, Hilland Ball (1980); BF: 418; BP: 18342A, 18349; CH: 487, 587; IB: 18276, 18279; NPI: 18405; SG: Blake 1981; SP: 18203. Common on wet mossy meadows and at pond mar- gins, occasionally in drier habitats. Carex ursina Dewey AF: Hill and Ball (1980). This replaces their record of Carex cf. glareosa Wahlenb. (Dr. J. Svoboda, personal communication). Sim- mons (1906) reported it from Cocked Hat Island which is north of Pim Island. + Eriophorum angustifolium Honck. AF: 18151; AFW: 18368; CH: 493; SP: 18189. Rare. On mossy pond margins at Cape Herschel and mixed with £. scheuchzeri in a wet meadow in the Sverdrup Pass. Eriophorum scheuchzeri Hoppe AF: 18246, Hill and Ball (1980); AFW: 18366; BI: 694; CH: 670; IB: 18275. Common in wet places, almost always in water. Eriophorum triste (Th. Fr.) Hadac & Love AF: 1/8151, Hill and Ball (1980) as E. angustifolium ssp. triste (Th. Fr.) Hult.; AFW: 18368; CH: 451, 474, 485, 510; 1B: 18277; KP: 18218; SP: 18070. Common by streams and on wet tundra; not found in water like the two preceding species. Kobresia myosuroides (Vill.) Fiori & Paol. AF: 18172, Hill and Ball (1980). Occasional in tufts on raised beach ridges. *Kobresia_ simpliciuscula (Wahlenb.) Mack. AF: 18417, Hill and Ball (1980). Rare. By ponds on raised beach terraces. Porsild (1964, map 68) gives only two other localities for this species on Elles- mere Island. *Juncus albescens (Lge.) Fern. AF: Hill and Ball (1980). Rare. Juncus biglumis L. AF: Hill and Ball (1980); AFW: 18370; CH: 501, 534, 621, 680; FB: 18100; SP: 18073, 18198. Occasional to common on wet gravel and moss. Luzula confusa Lindebl. AF: 18149, 18156, 18332, 18413, Hill and Ball (1980); BI: 697; CH: 460, 567, 576, 606, 690, 767; EG: 737. Common to ubiquitous on moss or damp ground, generally not far from water. : THE CANADIAN FIELD-NATURALIST Vol. 97 Luzula nivalis (Laest.) Beurl. AF: 18250, Hilland Ball (1980) as L. arctica Blytt; BP: 18359; CH: 425, 469, 499, 500, 539, 562, 574, 614, 623, 780; CR: 18410. Common to ubiquitous on moss and muck on seepy slopes and near late snowpatches. + Salix arctica Pall. AF: 18133, 18135, 18136, 18137, Hill and Ball (1980); BF: 472; BI: 697; BP: 18351, 18352; CH: 450, 490, 559, 765; EG: 727, 756; FB: 18086; 1B: 18272, 18273, 18283; KP: 18210, 18229; NPI: 18397, 18398; SG: Blake 1981; SP: 18060, 18061. Ubiquitous. Most prevalent in dry to mesic sites where it was protected by drifted snow in winter. Oxyria digyna (L.) Hill AF: 18/04, Hill and Ball (1980); BF: Anderson 1980; CH: 431, 454, 488, 513, 608; IB: 18304. Occasional on pond margins, mossy seeps, heaths and meadows. Polygonum viviparum L. AF: 18259, Hill and Ball (1980); AFW: 18377; CH: 502, 518, 540, 597; EG: 708; NPI: 18392; SP: 18204. Common. Usually at margins of ponds and creeks but also in soil- covered hollows of outcrops and cliffs where it gen- erally made better growth. *Arenaria humifusa Wahlenb. AF: Hill and Ball (1980). Rare. Otherwise known only from Harbour Fiord on Ellesmere (Simmons 1906) and otherwise only from Resolute in all of the Queen Elizabeth Islands (Porsild 1964, map 147). Cerastium alpinum L. AF: 18365, Hill and Ball (1980); BF: 417; CH: 465, 543, 569, 601, 610, 762; FB: 18098. Occasional as scattered individuals on dry tundra, on sand and gravelly till; rare on moss or at pond margins. ssp. Janatum (Cham.) Aschers. & Graebn. AF: 1/8337. On patches of soil in rocky tundra. Cerastium arcticum Lge. CH: 593, 740. Rare. In wet moss of meadows and seepage slopes. Cerastium regelii Ostenf. CH: Blake 1981. Rare. Reported from Pim Island and from the vicinity of Cape Rutherford by Simmons (1906) as C. alpinum forma pulvinatum. Polunin (1940) assigning Sim- mons’ plant to this species erroneously describes it as common everywhere on Ellesmere Island, which is how Simmons (1906) described the typical form of C. alpinum. Porsild (1964, map 141) accepted Simmons’ record. +Minuartia rubella (Wahlenb.) Hiern. (=Arenaria rubella in Porsild (1964)). AF: 18268, 18315, Hill and Ball (1980); BF: 408; BP: 18354; CH: 453, 515, 584, 683, 743; FB: 18096; NPI: 18388; SP: 18066, 18079. Common to ubiquitous in almost all xeric habitats, especially on raised beach terraces. (Sagina intermedia Fenzl) Cocked Hat Island, which is just north of Pim Island, was the only locality in 1983 all of Ellesmere where Simmons (1906, p. 119) found this species. The record is accepted by Polunin (1940, p. 207) who found the species com- mon as far north as Craig Harbour, and by Porsild (1964, map 145). Silene acaulis L. var. exscapa (All.) DC. AF: 18147, Hill and Ball (1980); AFW: 18367; CH: 604; LG: Blake 1981; NPI: 18396; SG: Blake 1981. Occa- sional, never abundant; on well-drained till and sand slopes, usually with some seepage or by streams. Simmons’ (1906) observation that the spe- cies favours south-facing slopes in our area was found to be true. Silene involucrata (C. & S.) Bocq. ssp. involucrata (= Melandrium affine in Porsild (1964)). AF: 18107, 18364, Hilland Ball (1980); KP: 7/8275. Rare. Rock crevices, gravel and mucky soil. Silene uralensis (Rupr.) Bocq. ssp. apetala (L.) Bocq. (=Melandrium apetalum ssp. articum in Porsild (1964)). AF: Hill and Ball (1980); BP: 78348; CH: 444,521,581, 626; FB: 18089; IB: 18296: SG: Blake 1981; SP: 78197. Occasional to common, occurring usually in moss on seepy slopes or by streams and ponds. +Stellaria humifusa Rottb. AF: 18382, Hill and Ball (1980); KP: 78223. Rare, on muddy sea beaches with Puccinellia phryganodes. Stellaria longipes Goldies.1. AF: 18057, 18132, 18169, 18316; BP: 18358; CH: 426, 457, 498, 565, 622, 784; FB: 18097, 18101; IB: 18308; KP: 18240; SG: Blake 1981; SP: 18066A. Occasional and rarely in flower. While determination to the collective species was generally easy, the lack of flowers usually made further identification impossible. S. /Jongipes was found ina variety of habitats from moss in seeps, on cliffs and by creeks where it grew as scattered stems, to mucks, small patches of soilamong rocks and on coarse gravel where it grew in scattered tufts. The Stygge Glacier specimen was S. crassipes Hult. S. laeta Richards. was found at Beitstad Fiord (406) growing onsteep unstable sand ofa raised beach. S. monantha Hult. was found in similar habitats at both Beitstad Fiord (4/5) and Cape Herschel (535), and was reported from Alexandra Fiord by Hilland Ball (1980). S. crassipes Hult. and S. laxmannii Fisch. were reported by Bocher (1951b) who rede- termined Simmons’ material. Ranunculus hyperboreus Rottb. AF: Hill and Ball (1980); BP: 18339; CH: 482. Rare. In brackish pool and in crevices on a small! rocky salt-spray swept island. (Ranunculus nivalis L.) Although he thought it was probably overlooked, because of its similarity to R. sulphureus, Simmons (1906, p. 110) listed this spe- cies aS uncommon but definitely present in the Hayes Sound area. BRIDGLAND AND GILLETT: VASCULAR PLANTS OF ELLESMERE ISLAND 285 (Ranunculus pygmaeus Wahlenb.) Simmons’ (1906) report of this species from Fram Harbour was accepted by Polunin (1940) and Porsild (1964, map 172). Simmons found it did not flower before autumn. Ranunculus sulphureus Sol. CH: 443, 476, 572, 648, 782, 786. Common at Cape Herschel on wet seepy slopes in moss, and on mineral soil, and in water at the margins of ponds. Papaver radicatum Rottb. AF: 18056, 18317, Hilland Ball (1980) as P. lapponicum (Tolm.) Nordh. ssp. porsildii Knaben; CH: 428, 436, 455, 462, 495, 511, 560; FB: 18087; IB: 18307; LG: Blake 1981; SP: 18077, 18194. Common on sand and gravel; occa- sional or sparse on moss and wet tundra. (Arabis arenicola (Richards.) Gel.) Reported by Simmons (1906) from Beitstad Fiord, and accepted by Polunin (1940) and Porsild (1964, map 197). Braya purpurascens (R. Br.) Bunge AF: /8051A, 18063, 18168, Hill and Ball (1980); CH: 773; FB: 18092; IB: 18290, 18286, 18309; KP: 18211, 18220C; SP: 18063, 18206. Common. Mainly in open gra- velly places and on frost polygons. Cardamine bellidifolia L. AF: 18333, Hill and Ball (1980); CH: 423, 522; NPI: 18387. Rare. Single plants in crevices on granite outcrops and on patches of soil among rocks. Cardamine pratensis L. var. angustifolia Hook. AF: Hill and Ball (1980); CH: 643, 655, 662. Rare. At Cape Herschel this was always sterile. Its best growth was in a wet moss meadow dominated by the moss genera Calliergon and Meesia. Scattered dwarf individuals were found in nearby ponds and brackish areas of streams. Cochlearia officinalis L. AF: 18174, 18383, Hill and Ball (1980) as C. fenestrata R. Br.; CH: 481, 651; FB: 18094; KP: 18212, 18213. Occasional on sea beaches and a small salt-spray swept island. *Drabaadamsii Ledeb. AF: 18180B, 18181. Rare. This circumpolar high-arctic species collected here ona beach in moss, is known from the western islands of the Queen Elizabeth group and from Pearyland (Mulligan 1974a). Draba alpina L. AF: 18106, 18381, Hill and Ball (1980) as D. alpina and D. adamsii; CH: 508, 551A, 637; FB: 18090; NPI: 18385; PI: 750; SP: 18064. Common. The second most common Draba in the region, occurring as stunted individuals on wet soil and moss, by streams and ponds, and on wet tundra polygons. +Draba cinerea Adams AF: 18155, 18422A, Hill and Ball (1980); NPI: /8400A. Rare. This was found on rocks and sandy moraines. These collections and those of Barrett and Teeri (1973) complete the east- ern range of this species in the Arctic Archipelago, which was mapped by Mulligan (1971, figure 6). 286 Several collections of Simmons’ were assigned to this species by Polunin (1940, p. 245). Draba corymbosa R. Br. ex D.C. ©D. bellii in Porsild (1964)). AF: 18173, 18180A, 18331, 18381, Hilland Ball (1980); BF: 4/0; BP: 18341; CH: 557, 744; IB: 18284, 18295; KP: 18221B, 18225; PI: 749; SP: 18064. Ubiquitous. The most common Draba of the area, always present in the mesic to xeric microhab- itats of the cobble barrens of the raised beach terraces. +Draba fladnizensis Wulf. AF: Hill and Ball (1980); EG: 717, 722A. Rare. At the Ekblaw Glacier station this grew in moss on a south-facing gneissic cliff dominated by Vaccinium and Empetrum. Although collections at DAO and CAN indicate its northern limit in the vicinity of Coutts Inlet on northern Baffin Island (Mulligan 1974b, figure 11), Simmons (1906) reported it as rather common in the Hayes Sound region, having collected it at seven localities between what is now Alexandra Fiord and Pim Island. * Draba glabella Pursh AF: Hill and Ball (1980). Rare. This report is the second for Ellesmere Island. Por- sild (1964, map 193) shows it present on Axel Hei- berg Island. Mulligan (1970, fig. 1) shows its nor- thernmost Canadian locations as Bylot Island and the Borden Peninsula. We have not seen Brassard’s collection (Brassard and Beschel 1968) from Tan- quary Fiord which is at Queen’s University (QK). Draba lactea Adams AF: 18164, Hill and Ball (1980); BF: 414; BP: 18345; CH: 556, 607B; CR: 18408; EG: 722B; KP: 18211, 18220A; NPI: 18400A; SP: 18081. Occasional. On a variety of substrates from dry sand to rocks and moss in wet meadows. + Draba nivalis Liljebl. AF: 18110, 18133 B, 18165, Hill and Ball (1980); CH: 684A; LG: Blake 1981. Occa- sional. Not shown for our region by Porsild (1964, map 188) or Mulligan (1974b, figure 16), but Sim- mons (1906) reported it as locally abundant in the Hayes Sound region. It was found in wet places, from mossy creek banks to ledges on outcrops. Draba oblongata R. Br. (includes D. groenlandica E. Ekm. in Porsild (1964) according to Mulligan (1974a)). AF: 18055, 18133A, 18175, 18180A, 18380, Hill and Ball (1980); BF: 407; BP: 18343; KP: 18220B, 18221A; SG: Blake 1981; SP: 18080, 18205. Occasional. On sand and gravelly raised beach terraces. + Draba subcapitata Simm. CH: 440; FB: 18095. Rare. By maps in Porsild (1964, map 190) and Mulligan (1974b, fig. 19), this is the first record for east- central Ellesmere Island. Simmons (1906) reported fragmentary fruiting specimens which seemed to represent this species from Cape Rutherford and Pim Island. Our specimens were found on wet sandy gravel. THE CANADIAN FIELD-NATURALIST Vol. 97 Erysimum pallasii (Pursh) Fern. AF: Hill and Ball (1980); BF: Anderson 1980. Rare. Simmons (1906) reported it from the same two localities. Eutrema edwardsii R. Br. AF: Hill and Ball (1980); CH: 5/2, 554. Rare. At Cape Herschel this was found on wet sandy gravel of a raised beach. * Halimolobos mollis (Hook.) Rollins AF: 18258, Hill and Ball (1980). Rare. Ona gravel slope. This is the second record for the Queen Elizabeth Islands, the other being at Lake Hazen (Powell 1961). It is known also from Etah in Greenland (Bocher et al. 1968; Porsild 1964, map 198). +Lesquerella arctica (Wormsk.) Wats. AF: 18053, Hill and Ball (1980); BF: Anderson 1980; IB: 18287; SG: Blake 1981. Rare, on gravel. Saxifraga caespitosa L. s.1. AF: 18166, Hill and Ball (1980); NPI: 78390. ssp. uniflora (R. Br.) Porsild CH: 441, 570, 582, 585. Occasional, in wet moss on gravel terraces, rock outcrops and seepy slopes. Found once in a sedge-moss meadow. Saxifraga cernua L. AF: 18243, 18379, Hill and Ball (1980); CH: 442, 514, 583, 640, 661, 675, 688; FB: 18093; KP: 18217; LG: Blake 1981; PI: 751A. Common, on wet mineral soil, in creeks and mossy seeps. Always dwarfed. The Cape Herschel material always had bulbils in place of the apical flower. Saxifraga flagellaris Willd. ssp. platysepala (Trautv.) Porsild CH: 427, 434, 449, 458, 550, 605. Occasional at Cape Herschel. Mostly on wet mineral soil, but also in damp to wet moss on cliffs and by streams. Saxifraga foliolosa R. Br. AF: Hill and Ball (1980); BI: 696; CR: 18406. Rare in moss along streams and pond margins. * Saxifraga hirculus L. var. propinqua (R. Br.) Simm. CH: 80-406A; SP: 18192. Rare in mesic to wet sedge meadows. Saxifraga oppositifolia L. AF: 18134, Hill and Ball (1980); BF: 402; CH: 463, 479, 489, 685, 766, 771; FB: 18091: IB: 18297; KP: 18234; LG: Blake 1981; NPI: 18386; SG: Blake 1981; SP: 18074. Ubiqui- tous. Inalmost every plant community but the very wettest. It made its best growth on raised gravel terraces where it was dominant. A white-flowered plant was found at Cape Herschel (766). Saxifraga rivularis L. AF: 18160, 18163, Hilland Ball (1980); NPI: 1/8391. Rare in wet places among coarse boulders and rocks. + Saxifraga tenuis Sm. AF: 18108, 18326, Hill and Ball (1980); CH: 439, 445, 532, 573, 625, 742, 769, 778; EG: 710; NPI: 18389. Common, in all the habitats listed for it by Porsild (1964), and in an Empetrum- heath at the Ekblaw Glacier site. Saxifraga tricuspidata Rottb. AF: 18244, 18269, 18421, Hill and Ball (1980); BF: Anderson 1980; CH: 571, 600, 686; LG: Blake 1981. Occasional, locally abundant, on dry gravel and sandy cliff 1983 ledges. It dominated the south-facing slope of a raised delta not far from the 1979 Pim Island site. Dryas integrifolia M. Vahl AF: 18052A, 18162, Hill and Ball (1980); BF: 405, 416; CH: 492, 519, 602, 755, 761; EG: 734; 1B: 18282; SG: Blake 1981; SP: 18076. Ubiquitous. Abundant and dominant with Saxifraga oppositifolia on dry tundra barrens where it tends to favour frost cracks, Dryas is found in practically all plant habitats in this region. It often forms hummocks in mesic graminoid meadows. Potentilla hyparctica Malte AF: 18138, 18318, Hill and Ball (1980); CH: 424, 509, 561, 635; EG: 718; NPI: 78384. Common on turfy hollows and ledges of outcrops and cliffs. * Potentilla nivea L. ssp. chamissonis (Hult.) Hiit. AF: Hill and Ball (1980) as P. hookeriana Lehm. ssp. chamissonis (Hult.) Hult.; BP: 18340; IB: 18303. Rare on dry shingle slopes and a muddy gravel beach. * Potentilla pulchella Pursh BF: 417. Rare on dry stable silty gravel of a raised beach in barrens dominated by Salix. Potentilla rubricaulis Lehm. SG: Blake 1981. Sim- mons (1906) excluded P. nivea from the Hayes Sound region after careful study by him showed that all of his collections of the latter were in fact this species. Reported by Malte (1934) from Bache Peninsula as P. pedersenii, and listed for our area by both Polunin (1940) and Porsild (1964). * Potentilla vahliana Lehm. AF: Muc (1980. Vascular plant communities. Pp. 12-28 in Ecology of a high arctic lowland oasis, Alexandra Fiord (78° 53’N, 75° 65’W) Ellesmere Island, N.W.T., Canada. 1980 Progress report. Edited by J. Svoboda and B. Freedman. Departments of Botany, University of Toronto and Dalhousie University.); CH: 80-405; SP: 18082, 18197. Rare. The collection from Cape Herschel came from a south-facing Carex stans meadow. Empetrum nigrum L. AF: Hill and Ball (1980) as ssp. hermaphroditum (Lge.) Boch.; EG: 729, 752. Rare. This acidophile is at the north end of its range in our area. At the Ekblaw Glacier site it was codominant with Vaccinium on a south-facing gneissic cliff. *Epilobium arcticum Samuelss. BF: 4/3. Rare. One specimen, left from the previous season, was found on dry silty gravel of a south-facing raised beach. This is the third record of this species, both for Ellesmere Island and for the Queen Elizabeth Islands. Otherwise known only from Lake Hazen and from Van Hauen Pass (Brassard and Longton 1970). Epilobium latifolium L. AF: 18144, 18425, Hill and Ball (1980); BF: Anderson 1980; CH: 517, 523. BRIDGLAND AND GILLETT: VASCULAR PLANTS OF ELLESMERE ISLAND 287 Rare, locally abundant on gravelly sand by streams. Cassiope tetragona (L.) D. Don. AF: 18140, Hill and Ball (1980); BF: 407; CH: 520, 693, 759, 775; EG: 754. Occasional, locally dominant. Possibly under- collected in parts of our area. Dominant on slopes covered by snow in winter but snowfree in summer. Isolated plants occur in hollows on stony barrens and gravelly moraines. +Pyrola grandiflora Rad. AF: Hill and Ball (1980). Also reported by Simmons (1906). Vaccinium uliginosum L. var. alpinum Big. AF: 18113, 18314, Hill and Ball (1980) as var. micro- phyllum Lge.; BF: Anderson 1980; EG: 754. Occa- sional. Locally dominant with Cassiope or with Cassiope and Empetrum on acidic cliffs and out- crops; also common on soil among boulders and snowpatches. (Pedicularis arctica R. Br.) Though missing from our area in Porsild (1964, map 295), we accept Sim- mons’ record on the basis of his picture (Simmons 1906, pl. 2). Pedicularis capitata Adams AF: 18102, 18262, Hill and Ball (1980); BF: 404; IB: 18288. Rare. On moist to wet sand and rocky slopes. Pedicularis hirsuta L. AF: 18103, 18146, 18248, Hill and Ball (1980): CH: 496, 516, 781; KP: 18233; SP: 18062, 18190. Occasional to common. The most common Pedicularis in the region, occurring mainly in wet meadows, in mossy hummocks by creeks and on mossy seepage slopes. +Pedicularis lanata Cham. & Schlecht. IB: /8281/. Rare. At Irene Bay this species was found widely scattered on dry tundra. Reported by Malte (1934) from Bache Peninsula. * Pedicularis sudetica Willd. IB: 18274. Rare. Beside a wet muddy stream in a meadow. Campanula uniflora L. AF: Hill and Ball (1980); CH: Funder 1980; LG: Blake 1981. Rare. On silt ona south-facing slope at Cape Herschel. Antennaria ekmaniana Pors. EG:698, 758. Rare. This species, a calciphobe near the northern limit of its range here according to Porsild (1964), was found once on dry sandy ledges of a south-facing gneissic cliff. *Arnica alpina (L.) Olin ssp. angustifolia (Vahl) Maguire LG: Blake 1981. Rare. Erigeron compositus Pursh IB: 18306; SG: Blake 1981; SP: 18078. Rare. On a muddy gravel beach and on a gravel kame. +Taraxacum phymatocarpum J. Vahl AF: Hill and Ball (1980); CH: 525; IB: 18305; SP: 18196. Occa- sional, on gravel slopes. Excluded Species The following species are indicated on maps in Por- 288 sild (1964) as occurring in our area. We question the origins of the reports or differ in our taxonomic con- cepts and exclude them for the present. Deschampsia brevifolia R. Br. Simmons’ Poa evagens from Fram Fiord on the south coast was redeter- mined as this species by Polunin (1940). The origin of Porsild’s dot in our area is unknown. Carex glareosa Wahlenb. var. amphigena Fern. Sim- mons (1906) reported the species as rare from the Hayes Sound region. While Polunin (1940, p. 116) accepted the record, leaving the variety in question, he cast doubt on it on the basis of the habitat reported by Simmons. Porsild (1964, map 82) appears to have accepted the record and to have assigned it to this variety. This northernmost record should perhaps be excluded pending verification. Armeria maritima (Mill.) Willd. Simmons (1906) reported this from Fram Fiord on the south coast. Porsild (1964, map 279) put it at Fram Harbour. Antennaria compacta Malte. The only previous record of an Antennaria being collected in our region is a single collection from the mouth of Flagler Bay made by Simmons (1906, 1913) which was assigned by him to A. alpina L. No other collec- tion is mentioned by either Malte (1934) or Polunin (1940) who placed Simmons’ collection tentatively in A. labradorica Nutt. Revising the genus for northwest Canada, Porsild (1950) made no mention of the plant collection by Simmons, but excluded all of Polunin’s (1940) records of A. compacta from the eastern Arctic. In 1955, however, Porsild (p. 182) equated Simmons’ 4. alpina “not L.” (sic) from Melville Island to A. compacta Malte and listed (p. 50) A. ekmaniana as present on Ellesmere Island. Both A. compacta and A. ekmaniana are mapped for our region in the 1957 edition of Porsild’s Illus- trated Flora of the Arctic Archipelago but neither that publication nor Porsild (1955) gives any record of any botanical collection on the east coast of Ellesmere Island. Taraxacum hyparcticum Dahlst. and T. pumilum Dahlst. We prefer to follow the taxonomic concepts of Simmons (1906, p. 20) and Polunin(1940, p. 368) and combine these species of Dahlstedt’s with T. phymatocarpum. Porsild (1964) recognized both species and mapped both for our area, even though Dahlstedt’s redeterminations of Simmons’ material showed the latter species to be found only in Grant Land and on the South Coast. Discussion The known vascular flora of east-central Ellesmere Island now numbers 117 species of which 109 were collected in the summers of 1979-81. Of these recently collected species, forty-two are not shown on maps in THE CANADIAN FIELD-NATURALIST Vol. 97 Porsild (1964) as occurring in our area. Of this forty- two, twenty have been reported for the area by either Simmons (1906) or Polunin (1940). Factors affecting the floristic diversity of the Hayes Sound Region include the edaphic and climatic hete- rogeneity of the region and the vegetational history of adjacent regions. Simmons (1906) recognized that the presence, in the Hayes Sound Region, of both acidic and calcareous rocks increased the number of species found there. In fact, however, the flora includes only five species which could be considered acidophiles. Dryopteris fragrans, Hierochloe alpina and Empetrum nigrum do not occur north of Hayes Sound on Ellesmere Island. Lycopodium selago and Vaccinium uligino- sum cannot be obligate acidophiles since they occur in northern Ellesmere (Brassard and Longton 1970). Thirty-five of our species are listed by Porsild (1964) as calciphiles, of which Poa abbreviata, Carex rupes- tris, Draba subcapitata, Erysimum pallasii, Saxifraga caespitosa and Potentilla vahliana are said to be pro- nouncedly calciphilic. The presence of littoral habitats in the region is responsible for the occurrence of halo- phytic species such as Dupontia fisheri, Carex mari- tima, C. ursina, Stellaria humifusa, Puccinellia phryganodes, and Cochlearia officinalis. The role of regional climate in determining the dis- tribution of plants in the Hayes Sound region can be approached in two related but slightly different ways: in terms of continentality and in terms of aggregate summer warmth. The climate of the interior part of the region is more continental than that of the coast. This is reflected in Maxwell’s (1981) placement of his border between the Nares Strait and northern Baffin Bay climatic subre- gions on the Johan Peninsula. The increases of mean temperature and mean daily temperature range over the 40 km between Cape Herschel and Alexandra Fiord are similar to those found over similar distances from the coast in Alaska (Haugen and Brown 1980). Bocher (1954) classified the plants of southwest Greenland in terms of continentality. None of his oceanic species occurs in our area, but many of his continental ecogeographic and climatic indicator spe- cies do, as do many of the species which he found were not restricted to either oceanic or continental regions. Grouping our stations into coastal, intermediate and interior regions, and expressing the number of contin- ental species as a fraction of the total species for each region, we found a general increase in the continental- ity of the flora proceding westward from the coast (Table 1). The Baird Inlet stations and Irene Bay have a high proportion of continental species, which is not altogether unexpected since Baird Inlet was covered with ice to a distance of 30 km to the east of the fiord 1983 TABLE |. Distribution of continental species in the Hayes Sound flora. Total Continental Region Species Species Coastal Region (CH, CR, PI, NPI) 72 30 (42%) Intermediate Region (AF, AFW, BP, KP) 91 42 (46%) Interior Region (BF, FB, SP) 53 26 (49%) Irene Bay (IB) 28 16 (57%) Baird Inlet (BI, EG) 20 13 (65%) head when it was visited in late July, and since Irene Bay is situated in the most continental part of the island. However, because both areas were visited only briefly, they may be undercollected for more common species. Since the Hayes Sound area is very continental by comparison with southwest Greenland, it is not sur- prising that Bocher’s classification does not strictly apply there. Two of his continental ecogeographic species, Ranunculus sulphureus and Saxifraga flagel- laris were found, by us, only on the coast. But several of his continental species, including Calamagrostis purpurascens, Pedicularis lanata, Erigeron composi- tus, Lesquerella arctica, Potentilla nivea, and Pedicu- laris capitata, appear, by their distribution in our area, to be restricted, even in the high Arctic, to areas of relatively continental climate. Following the same criteria, Epilobium arcticum and Pedicularis sudetica, which do not occur in southwest Greenland, could be placed in the same category. Carex maritima, which Bocher describes as a continental littoral species, was not found in the coastal region. Seven of our species were described by Brassard and Beschel (1968) as more frequent in oceanic parts of the high Arctic. Of these, Cerastium regelii has been found only on the coast; Phippsia algida, Pleuro- pogon sabinei, and Potentilla hyparctica were found only in coastal and intermediate regions, while Silene acaulis and Pedicularis hirsuta ranged from the coast to the interior and Silene involucrata tended towards the interior. A large component of a continental climate is temperature. Young (1971) divided the Arctic into four floristic zones based on a correlation of the northern distributional limits of large groups of spe- cies with isopleths of aggregate summer warmth. While he places our region in his third zone, a= 12-20, values ofa calculated from the available weather data put the region in his second zone, a = 6-12, with Cape Herschel (a = 5.4 in 1973, a = estimated 6.3 in 1979) bordering on the first zone and Alexandra Fiord (a = BRIDGLAND AND GILLETT: VASCULAR PLANTS OF ELLESMERE ISLAND 289 estimated 11.9 in 1979) and Bache Peninsula (a= 11.0 in 1930-33) bordering on the third zone. Plants with limits in Young’s third zone found at Alexandra Fiord but not at Cape Herschel include Dryopteris fragrans, Kobresia simpliciuscula, Carex scirpoidea, C. capillaris, C. bigelowii, Draba glabella, D. cinerea, Empetrum nigrum, and Pyrola grandi- flora. Plants with northern limits in Young’s second zone which were found at Alexandra Fiord but not at Cape Herschel include Lycopodium selago, Hierachloe alpina, Carex rupestris, Kobresia myosuroides, Silene involucrata, Saxifraga foliolosa and _ Pedicularis capitata. Since the late 1800’s various portions of Greenland and the Canadian Arctic Archipelago have been pro- posed as botanical refugia during the last glaciation (cf. Bocher 1951la; Love 1962; Savile 1961; Brassard 1971). Detailed discussion of this topic is beyond the scope of this paper. Simmons (1913), however, did propose that the Hayes Sound Region harboured plant life during the last glacial maximum and this matter will be addressed here. Simmons based his conclusion on the scattered dis- tribution of species on Ellesmere Island and the absence of 29 species of plants for several degrees of latitude to the south. Of these, 17 are now known to be more widespread on Ellesmere and have been found on Devon Island or other intervening areas (Polunin 1940; Barrett and Teeri 1973; Bliss 1977) and can no longer be considered disjunct on Ellesmere Island. In our area this group is represented by Arnica alpina, Hierchloe alpina, Dupontia fisheri, Kobresia simpli- ciuscula, Carex ursina, Equisetum variegatum, Ranunculus pygmaeus, Sagina intermedia, Empe- trum nigrum and Pedicularis arctica. Taraxacum phymatocarpum is not isolated on the south coast of Ellesmere Island, and had Simmons not acceded to the arbitrary redeterminations of Dahlstedt, he would not have considered it as such. Of the remaining twelve species, six occur in our area. These include Arenaria humifusa, Dryopteris fragrans, and Carex glacialis. All of these are wide- spread on the west coast of Greenland, with centres of occurrence in continental West Greenland (Bocher et al. 1968). Their occurrence in the Hayes Sound Region is easily interpreted as an extension of the Greenlandic range. Arabis arenicola, Pyrola grandi- flora and Erigeron compositus are somewhat more problematic and are dealt with below. Of the species we found new to the flora of the Hayes Sound Region, Festuca baffinensis, Trisetum spicatum, Carex scirpoidea, Draba glabella, Saxi- fraga hirculus var. propinqua, Potentilla pulchella, Potentilla vahliana, Pedicularis lanata, and Pedicula- 290 ris sudetica are widespread in the Arctic Archipelago, and probably migrated into our area from the west. Carex bigelowii extends to our area from either West Greenland or Baffin Island. Nine species which occur in the Hayes Sound area could be considered as truly disjunct in the Queen Elizabeth Islands and/ or northwest Greenland (NWn of Bocher et al. 1968). These are Carex amblyorhyn- cha, C. capillaris, Potentilla nivea ssp. chamissonis, Epilobium arcticum, Draba cinerea, Arabis arenicola, Pyrola grandiflora, Halimolobos mollis, and Eri- geron compositus. Carex capillaris, Draba cinerea, and Epilobium arcticum are listed by Young (1971) as being limited to his second most southerly floristic zone (Zone 3). Carex amblyorhyncha, Arabis areni- cola, Pyrola grandiflora, and Halimolobos mollis are also found here at the northernmost end of their ranges. It would seem more likely that these species have migrated northward during a postglacial warm period (cf. Koerner and Fisher 1981) and remain as relict at the present time, than that they survived the last glaciation at Hayes Sound. Bocher (1951a), observing the weediness of Halimo- lobos in western Greenland and its association with native hunting camps there, has reasoned that the species may be considered a good candidate for post- glacial migration with the Inuit. The two species which show sizable disjunctions between the high and low Arctic both have the capac- ity for long distance dispersal. The seed of Erigeron compositus is well adapted to wind transport and Potentilla nivea ssp. chamissonis belongs to a genus in which the tap root is commonly collected by western Arctic Inuit for medicinal purposes (Young and Hall 1969). With 117 species, the flora of the Hayes Sound region is moderately rich. Counting all the infraspe- cies of Stellaria longipes, Brassard and Beschel (1968) reported 143 species from northern Ellesmere Island. Using the same taxonomy, Muc and Bliss (1977) revised the number of species at Truelove Lowland to 93, and of Devon Island to 117. The greatest floristic diversity in the Hayes Sound Region is found in the vicinity of Alexandra Fiord. While it is possible that some plant species may have survived the last glacia- tion on nunataks in that portion of the region, the diversity at Alexandra Fiord is more than adequately accounted for by the present-day climatic ecotone found there, with minor contributions from saline and granitic substrates. Acknowledgments We thank Dr. Henri Ouellet of National Museums of Canada and Dr. Peter Schlederman of the Arctic Institute of North America for arranging accommo- THE CANADIAN FIELD-NATURALIST Vol. 97 dation at Alexandra Fiord for Gillett and Shche- panek. Dr. Weston Blake, Jr. of the Geological Sur- vey of Canada was very kind to allow Bridgland to pursue his botanical interests while at Cape Herschel, and offered much useful criticism on various parts of the manuscript. Logistic support for both field parties was provided by the Polar Continental Shelf Project, Energy, Mines, and Resources. Dr. Josef Svoboda of the University of Toronto graciously allowed us to include in our list species collected by his party at Alexandra Fiord in 1980. Dr. Thane Anderson of the Geological Survey of Canada and Dr. Svend Funder of the Greenland Geological Survey also kindly allowed us the use of their collec- tions. Dr. Bea Alt of the Polar Continental Shelf Project and Dr. J. B. Maxwell of the Atmospheric Environment Service permitted us to publish some of the weather data. We thank Mr. G. A. Mulligan for verifying or rede- termining all of our collections of Draba. Bridgland thanks Drs. G. R. Brassard and P. J. Scott for help with some of the determinations. Literature Cited Barrett, P. E., and J. A. Teeri. 1973. Vascular plants of the Truelove Inlet Region, Devon Island. Arctic 26: 58-67. Blake, W., Jr. 1977. Glacial sculpture along the east-central coast of Ellesmere Island, Arctic Archipelago. Jn Current Research, Part C, Geological Survey of Canada, Paper 77-1C: 107-115. Blake, W., Jr. 1978. Coring of Holocene pond sediments at Cape Herschel, Ellesmere Island, Arctic Archipelago. Jn Current Research, Part C, Geological Survey of Canada, Paper 78-1C: 119-122. Blake, W., Dr. 1981. Lake sediment coring along Smith Sound, Ellesmere Island and Greenland. /n Current Research, Part A, Geological Survey of Canada, Paper 81-1A: 191-200. Bliss, L. C. 1977. Vascular plants of Truelove Lowland and adjacent areas including their relative importance. Appendix 6in Truelove Lowland, Devon Island, Canada: A high arctic ecosystem. Edited by L. C. Bliss. University of Alberta Press, Edmonton. 714 pp. Bocher, T. W. 195la. Distributions of plants in the circum- polar area in relation to ecological and historical factors. Journal of Ecology 39(2): 376-395. Bocher, T. W. 1951b. Studies on the distribution of the units within the collective species of Ste/laria longipes. Botanisk Tidsskrift 48(4): 401-421. Bocher, T. W. 1954. Oceanic and continental vegetational complexes in southwest Greenland. Meddelelser om Grgn- land 148(1): 1-336. Bocher, T. W., K. Holmen, and K. Jakobsen. 1968. The flora of Greenland. P. Haase & Son, Copenhagen. 312 pp. Brassard, G. R. 1971. The mosses of northern Ellesmere Island, Arctic Canada. I. Ecology and phytogeography, with an analysis for the Queen Elizabeth Islands. The Bryologist 74(3): 233-281. 1983 Brassard, G. R., and R.E. Beschel. 1968. The vascular flora of Tanquary Fiord, northern Ellesmere Island, N.W.T. Canadian Field-Naturalist 82: 103-113. Brassard, G. R., and R. E. Longton. 1970. The flora and vegetation of Van Hauen Pass, northwestern Ellesmere Island. Canadian Field-Naturalist 84: 357-364. Christie, R. L. 1962. Geology, Alexandra Fiord, Ellesmere Island, District of Franklin (map with marginal notes). Geological Survey of Canada, Map 9-1962. Christie, R. L. 1967. Bache Peninsula, Ellesmere Island, Arctic Archipelago. Geological Survey of Canada, Memoir 347. Christie, R. L. 1972. Central Stable Region. Pp. 40-87 in Guidebook to Excursion A66, The Canadian Arctic Islands and the Mackenzie Region. Edited by D. J. Glass, XXIV International Geological Congress, Montreal. Department of Transport. 1944. Meteorology of the Cana- dian Arctic. Meteorological Division. Toronto. 85 pp. Durand, E. 1863(4). Enumeration of the arctic plants col- lected by Dr. I. I. Hayes, between parallels 78th and 82nd, during the months of July, August and beginning of Sep- tember 1861. Proceedings of the Academy of Natural Science of Philadelphia 1863 (1864): 93-96. Ekblaw, W. E. 1918. The vegetation about Borup Lodge. Pp. 397-402 in Four years in the white north. By D. B. MacMillan. Harper, New York. Frisch, T.. W. C. Morgan, and G. R. Dunning. 1978. Reconnaissance geology of the Precambrian Shield on Ellesmere and Coburg Islands, Canadian Arctic Archipe- lago. Jn Current Research, Part A, Geological Survey of Canada, Paper 78-1A: 135-138. Greely, A. W. 1886. Three years of arctic service. Scribners, New York. Haig-Thomas, D. 1940. Expedition to Ellesmere Island, 1937-38. Geographical Journal 45: 265-277. Hart, H. C. 1880. On the botany of the British Polar Expe- dition of 1875-76. Journal of Botany New Series 9: 52-56, 70-79, 111-115, 141-145, 177-182, 204-208, 235-242, 303-306. Haugen, R. K., and J. Brown. 1980. Coastal-inland distri- butions of summer air temperature and precipitation in northern Alaska. Arctic and Alpine Research 12(4): 403-412. Hayes, I. I. 1867. The open polarsea. Hurd and Houghton, New York. Koerner, R. M., and D. A. Fisher. 1981. Studying climatic change from Canadian High Arctic ice cores. Pp. 195-218 in Climatic change in Canada 2. Edited by C. R. Harring- ton. Syllogeus No. 33. National Museums of Canada. Love, D. 1962. Plants and Pleistocene. Problems of the Pleistocene and arctic 2(2): 17-39. McGill University, Montreal. Low, A. P. 1906. Report on the Dominion Government Expedition to Hudson Bay and the Arctic Islands on board the D.G.S. Neptune. Government Printing Bureau, Ottawa. Malte, M. O. 1934. Critical notes on plants of Arctic Amer- ica. Rhodora 36: 172-193. Maxwell, J. B. 1980. The climate of the Canadian Arctic Islands and adjacent waters. Volume |. Climatological Studies 30. Atmospheric Environment Service, Toronto. 531 pp. BRIDGLAND AND GILLETT: VASCULAR PLANTS OF ELLESMERE ISLAND 291 Maxwell, J. B. 1981. Climatic regions of the Canadian Arc- tic Islands. Arctic 34(3): 225-240. Mohn, H. 1907. Meteorology. Report of the Second Nor- wegian Arctic Expedition in the “Fram” 1898-1902, No. 4. Videnskabs-selskabet, Kristiania. 399 pp. Muc, M.,andL. C. Bliss. 1977. Plant communities of True- love Lowland. Pp. 143-154 in Truelove Lowland, Devon Island, Canada: A high arctic ecosystem. Edited by L. C. Bliss, University of Alberta Press, Edmonton. 714 pp. Mulligan, G. A. 1970. Cytotaxonomic studies of Draba glabella and its close allies in Canada and Alaska. Cana- dian Journal of Botany 48(7): 1431-1437. Mulligan, G. A. 1971. Cytotaxonomic studies of the closely allied Draba cana, D. cinerea, and D. groenlandica in Canada and Alaska. Canadian Journal of Botany 49(1): 89-93. Mulligan, G. A. 1974a. Confusion in the names of three Draba species of the arctic: D. adamsii, D. oblongata, and D. corymbosa. Canadian Journal of Botany 52(4): 791-793. Mulligan, G. A. 1974b. Cytotaxonomic studies of Draba nivalis and its close allies in Canada and Alaska. Canadian Journal of Botany 52(8): 1793-1801. Mulligan, G. A. 1976. The genus Draba in Canada and Alaska: key and summary. Canadian Journal of Botany 54(12): 1386-1393. Polunin, N. 1940. Botany of the Canadian Eastern Arctic. Part 1. Pteridophyta and spermatophyta. National Museum of Canada Bulletin No. 92. Ottawa. 408 pp. Porsild, A. E. 1950. The genus Antennaria in northwestern Canada. Canadian Field-Naturalist 64(1): 1-25. Porsild, A. E. 1955. The vascular plants of the western Canadian Arctic Archipelago. National Museum of Can- ada Bulletin 135. Ottawa. 226 pp. Porsild, A. E. 1964. Illustrated flora of the Canadian Arctic Archipelago (1957). Revised edition. National Museums of Canada Bulletin No. 146. Ottawa. 218 pp. Powell, J. M. 1961. The vegetation and micro-climate of the Lake Hazen Area, northern Ellesmere Island, N.W.T. Defence Research Board, Department of National Defence. Canada. 112 pp. Savile, D. B. O. 1961. The botany of the northwestern Queen Elizabeth Islands. Canadian Journal of Botany 39: 909-942. Schlederman, P. 1978. Preliminary results of archeological investigations in the Bache Peninsula Region, Ellesmere Island, N.W.T. Arctic 31(4): 459-474. Schlederman, P., and K. McCullough. 1980. Western ele- ments in the early Thule culture of the eastern High Arctic. Arctic 33(4): 833-841. Simmons, H. G. 1906. The vascular plants in the flora of Ellesmereland. Report of the Second Norwegian Arctic Expedition in the “Fram” 1898-1902, No. 2. Videnskabs- selskabet, Kristiania. 197 pp. Simmons, H. G. 1913. A survey of the phytogeography of the Arctic American Archipelago with some notes about its exploration. Lunds Universitets Arsskrift. N.F. Afd. 2, Bd 9, Nr 19, Lund. Taylor, A. 1964. Geographical discovery and exploration in the Queen Elizabeth Islands. Canada, Department of Mines and Technical Surveys, Geographical Branch, Memoir 3, Ottawa. 292 THE CANADIAN FIELD-NATURALIST Vol. 97 Young, S.B. 1971. The vascular flora of St. Lawrence ethnobotany of the St. Lawrence Island eskimo. Anthro- Island with special reference to floristic zonation in the pological Papers of the University of Alaska 14(2): 43-53. arctic regions. Contributions from the Gray Herbarium, No. 201, Harvard University. 115 pp. Received 20 July 1982 Young, S. B., and E. S. Hall, Jr. 1969. Contributions to the Accepted 10 February 1983 Seasonal Feeding of Carp, Cyprinus carpio, in the Bay of Quinte Watershed, Ontario PERCE M. POWLES,! HUGH R. MACCRIMMON,? and DONALD A. MACRAE? \Department of Biology, Trent University, Peterborough, Ontario K9J 7B8 2Department of Zoology, University of Guelph, Guelph, Ontario NIG 2W1 380 Jeffcoat Drive, Toronto, Ontario Powles, Perce M., Hugh R. MacCrimmon, and Donald A. Macrae. 1983. Seasonal feeding of Carp, Cyprinus carpio, in the Bay of Quinte watershed, Ontario. Canadian Field-Naturalist. 97(3): 293-298. Observations from three areas and three seasons in 1977-78 suggest that Carp resident in the waters of the Bay of Quinte watershed are omnivorous, but primarily ingest invertebrate organisms. Feeding is curtailed but does not cease during the autumn and winter months. Plant materials make up asmall part of digestive tract contents in the fall, changing to seeds in the winter. The presence of considerable debris during summer and fall suggests an active scouring of the substrate during feeding. A low incidence of plant materials in the digestive tract contents at any time of the year indicates a low nutritional dependence on aquatic plants. This suggests that, at present population densities, Carp are not a threat (in terms of direct ingestion) to indigenous aquatic plants though they may still act the role of a control species for many native macrophytes. Key Words: Carp, Cyprinus carpio, feeding, macrophytes, benthos, invertebrates, seasonal, polyphagous, Wild Rice Carp, Cyprinus carpio, became naturalized in the Bay of Quinte watershed early in the present century likely as a result of a combination of easterly move- ments of the species through the Trent Canal system from Lake Simcoe, and invasions from plantings made in New York State where Carp had been intro- duced into the Lake Ontario watershed as early as 1890 (Smith 1892). Presence of the Carp was first reported from the Bay of Quinte in 1907 and from most of the Kawartha Lakes by 1915 (McCrimmon 1968: MacCrimmon and Skobe 1970). There is a longstanding concept, and some evi- dence, that the feeding habits of alien Carp may be detrimental to pristine aquatic ecosystems, especially indigenous vegetation and dependent fish and wildlife species (Cahn 1919; Gerking 1950). Early Canadian evidence of their deleterious effects on aquatic resour- ces is limited historically to a relationship observed between Carp overabundance and the destruction of rooted aquatics, in particular, Wild Rice (Zizania aquatica), Wild Celery (Valisneria americana), and Water Milfoil (Myriophyllum excalbens), in marsh habitats in the lower Great Lakes basin (McCrimmon 1968; MacCrimmon and Skobe 1970). Other negative relationships between the Carp and the environment are speculative and, in fact, the presence of this species may prove to be beneficial to eutrophifying waters (Uhler 1944; McCrimmon 1968; and others). Despite concern over potential negative effects of Carp populations on the aquatic resources of waters such as those in the Bay of Quinte watershed, no comprehensive investigation of the feeding and diet of the species has been undertaken in Canadian waters. This paper, therefore, records just such information on the Carp by reporting on the gastrointestinal con- tents of the species taken from three Bay of Quinte watershed locations over a 12-month period. Materials and Methods Carp were collected in the Bay of Quinte watershed from the Indian River, Bay of Quinte itself, and Stur- geon Lake (Figure 1). A detailed description of the watershed is provided by Johnson and Owen (1971). BAY OF QUINTE WATERSHED : () 25 50 N aa {| SCALE (mi) t ) D> \ QQ p \ A . Z [f i SAF // - “Take Ontario FIGURE |. Bay of Quinte Watershed showing the Kawartha Lakes District and sampling areas: A, Indian River; B, Bay of Quinte; C, Sturgeon Lake. 293 294 THE CANADIAN FIELD-NATURALIST Vol. 97 Collections from the Indian River were made dur- ing the spring and summer of 1977 by seine and fyke e e 2 net. The Carp were transported to the Trent Univer- a ilsese o & a! sity laboratory for routine examination. Fish were a a Ra N va) = sorted into two size groups and the entire gastro- 2 |ls reine intestinal tract preserved in 10% formalin for later = les study. Samples also were taken from Carp caught = commercially by gill nets during the fall of 1977 from S a the Bay of Quinte, and during the winter from Stur- eS 28 = 5 So N \o wt geon Lake, in 1978. g/£aS a 6 S The volumes of digestive tract contents up to 10 ml 5 3 3 were measured in graduated cylinders to the nearest 5 0.1 ml. If the total volume was greater than 10 ml, a 4 ee mixed 10% subsample was taken from larger samples. ro Ss Food items were recorded by frequency of occurrence 5 Ee |-2 ~so eo = and relative volume (Ricker 1968). Because of the = = 5) macerated nature of much of the material in the diges- ew ls = tive tracts of most fish, some animal life was classified = E only to order or group, such as plants, (green frag- OY |S |e ments and seeds), debris (sediments and dead plant oie is N 00 oo N material), or unidentified invertebrates. Statistical 3 2 x differences between- and within-group were examined © s using a “student t-test” (Steel and Torrie 1960). S E im = = iss} wa ° Results Sher z| a8 x a Indian River Carp specimens collected during the S - spring and summer months contained food in 92% of © the digestive tracts with a mean volume of 12.9 ml/kg eS SEEN ja aq rei body weight (Table 1). Separation of the fish into two E me! =F & eA size groups showed a higher percentage of animal a= = Sis a % = organisms (91%) in the diet of the younger fish. The 8 es 2) el dominant invertebrates in relative importance (Table s 2) in both size groups were Diptera. Plant remains = @ occurred in 39% of all digestive tracts of the smaller o =, E re S S = Carp (although only 2% by volume) and included e 5 | OE S 0 Ceratophyllum demersum (Coontail), Zizania aqua- 8 ee 5 TaN “) tica (Wild Rice), Valisneria americana (Wild Celery), es Elodea canadensis [= Anacharis canadensis] > Rt (Waterweed), Najas flecilis, (Bushy Pondweed), Zz a> oO = Potamogeton crispus (Curly Pondweed), Potamo- = 2 3 ye iS e S geton pectinatus (Sago Pondweed) and Potamogeton S ye =s o eS richardsonii (Redhead Grass). Seeds from Z. aqua- 2 = tica, N. flexilis, P. crispus, and P. pectinatus were = found in 25% of the stomachs. The above plant species 3 z=) E occurred in the digestive tracts of the larger Carp with es SORE eres 20 S the addition of Nymphaea odorata (White Waterlily) S Eo = = and Myriophyllum excalbens (Water Milfoil). Seeds = Za of Z. aquatica, N. flexilis, P. crispus and P. pectinatus = occurred in the stomachs of 25% of the smaller and 2 5 é 27% of the larger Carp. &§ S55 = 7s The spring and summer diet of Indian River Carp a as = 5 5 5 = was predominantly animal material (91 and 82% by 5 eleece es S oy volume, Table 2). Debris was the second most impor- a § 5 Ss = ss = & tant food category. Statistical analysis (P = 0.05) eS 5 S5 S a Se revealed no significant difference in food item prefer- 1983 ence by prey group with sex or size at this location. Adult Bay of Quinte Carp specimens collected dur- ing the autumn months contained food in 86% of the digestive tracts with a mean volume of 5.6 ml/kg fish (Table 1). Diptera dominated the invertebrate forms (Table 2) with their larvae and pupae making up 38% by volumetric proportion. Debris was the second highest level of relative importance at 18% by volume. The plant component was only 7% by volume (Table 2), principally green fragments and consisting of the same species identified in the Indian River samples. The autumn diet of the Bay of Quinte Carp was domi- nated by animal material with debris and plant items making up 18 and 8% respectively (Table 1). There were no significant differences between the presence of major food groups and the sex or two size categories. Mature Sturgeon Lake Carp specimens collected by large mesh (over 18 cm) gill nets during the winter months contained food in 74% of the digestive tracts with a mean volume of 1.4 ml/kg body weight (Table 1). Diptera made up 43% and Cladocera 29% of their diet by volume (Table 2). The plant component, which occurred in 32% of the digestive tracts, made up only two percent of the diet by volume and consisted almost entirely of seeds of Zizania aquatica, Najas flexilis, Potamogeton crispus, and Potamogeton pec- tinatus. The winter diet of Sturgeon Lake Carp was found to consist almost entirely of animal items (98% by volume, Table 1). There was no significant differ- ence between food items ingested and either size or sex (t-test, P = 0.01). Discussion Analysis of the Carp digestive tract contents between sites revealed no significant difference (P = 0.05) between categories or volumes of ingested groups. Thus comparison of the data by water body or season was simplified. This finding was rather surpris- ing, since some authors have clearly related Carp size to food preference (Bailey and Harrison 1945; Ali- kunhi 1958). It is possible that clear food/fish size relationships were inconsistent because sampling sites were different over the seasons. It was necessary, how- ever, to use Carp specimens when available from sev- eral waters, to round out data in the annual feeding cycle. From this study there is no evidence that availa- ble food items varied appreciably within the three waters that were essentially warmwater and eutrophic (Johnson and Owen 1971; R.W. Lewies 1976. The fisheries of the Kawartha Lakes. Jn: Kawartha Lakes Water Management Study/ Water Quality Assess- ment (1972-1976). Ontario Ministry of the Environ- ment and Ministry of Natural Resources, MS pp. 83-106), but future studies should focus on one site TABLE 2. Incidence of occurrence of mean volumes of materials in digestive tracts of Carp from the Bay of Quinte Watershed, by area and principal food grouping. Indian River Spring/Summer, 1977 Bay of Quinte Autumn, 1977 Sturgeon Lake Autumn/Winter, 1978 Large Fish Large Fish Large Fish Small Fish 625 mm 600 mm 257 mm 256 mm i) Volume % Occurrence we Food Volume Occurrence Occurrence Volume Occurrence Volume Category 77 66 73 29 69 65 26 20 63 Cladocera 12.5 43 63 Copepoda 9.5 38 12.5 32.5 4] 69 81 Trichoptera Diptera 79 71 43 49 28 86 32 8.5 6.5 10.5 72 Mollusca Other POWLES, MACCRIMMON AND MACRAE: SEASONAL FEEDING OF CARP Se) D* 15 25 24 39 88 Invertebrates Plant Material Debris 84 67 71 18 Ds 10 81 Number of 295 128 120 48 56 Stomachs *Seeds only 296 THE CANADIAN Over a season, to confirm our preliminary observations. The omnivorous nature of the digestive tracts’ con- tents is typical of Carp generally (McCrimmon 1968; Scott and Crossman 1973; McAllister and Coad 1974), and suggests that most littoral plant items are likely to be ingested either by choice or incidentally. The presence of substantial amounts of debris (2 to 8% by volume) in fish from all waters confirms that con- siderable sucking and digging activity by the Carp may occur while feeding (Vass 1957; Moen 1953). However, the items ingested by each Carp population were principally invertebrates (74 to 97% by volume) although a review of the literature (Table 3) shows that the relative amount of animal and plant materials in Carp from various waters is extremely variable (19 to 98% by volume). Only during the winter period of ice cover, when macrophyte production would be minimal, did plant material not make up an appreciable percentage (7 to 18%) of the volume of ingested items by the Carp. Seeds were found in 32% of the Carp in winter, and would represent high caloric-value food. Another noteworthy observation was that Carp did, in fact, feed under the ice in winter. This finding is new, and has not been reported for the Carp in European waters. Opinions differ on the dietary importance of vegetable matter, and there is some question whether feeding on aquatic plants if obligatory, as asserted by Gaevskaya (1969). While some authors (Cole 1905; Tracey 1910) reported direct grazing, other authors (Eder and Carlson 1977) concluded that the presence of plant materials in the digestive tract are a conse- quence of intensive searching for invertebrates. Oth- ers (Summerfelt et al. 1971) argue that there may be no FIELD-NATURALIST Vol. 97 nutritional reward because of the inability of the spe- cies to digest cellulose. On the other hand, Carp are known to feed strictly on vegetation in rice fields (Gaevskaya 1966); and Gunnetal. (1976) have clearly demonstrated that Bullhead (/ctalurus nebulosus) can assimilate algae. If we add to this the findings of Kevern (1966), who showed that yearling carp ingest and assimilate both detritus and algae, we must also support the argument on the side of herbivory, with the addition of seeds to the winter diet. With more eutrophic conditions in recent years in the Quinte region, both Carp and plant production have increased noticeably (J. Christie, personal com- munication). One impact on aquatic vegetation would seem to be uprooting of macrophytes by scouring activity, and there is no doubt (McCrimmon 1968) that high Carp populations can cause marked ecologi- cal change by uprooting vast areas of more susceptible rooted aquatics, especially Wild Rice, Wild Celery, and Water Milfoil. Accompanying such activity, the turbidity may reduce photosynthesis. However, popu- lation levels of Carp in the Bay of Quinte drainage would seem to be substantially below levels which cause vegetative devastation, perhaps because a signif- icant commercial fishery is maintained, thus keeping the Carp populations within reasonable size. To maintain primarily an animal diet, as judged by volume (which may or may not be a legitimate method), the Bay of Quinte Carp ingested a wide variety of invertebrate organisms, with Diptera, Cladocera, Trichoptera and Copepoda the dominant categories in Carp from the three sites (Table 2). The considerable variations in dominant invertebrate organisms ingested by Carp (Moen 1953; Nakamura 1955; Sigler 1958; Rehder 1959; Walberg and Nelson TABLE 3. Food of Carp as reported by various authors, expressed in % of total volume of digestive tract contents, showing the relative importance of plant and animal contingents. Reference Region/ Habitat Size % Animal % Plant Birznek (1962) in Gaevskaya 1966 Russian Rice Fields All Mainly Plants and seeds Ewers and Boesel 1935 Buckeye Lake, Ohio Young 88 12 Egereva (1958) in Gaevskaya 1966 Volga River All 22 78 Harrison 1950 lowa Lakes Adult 63 32 Moen 1953 lowa Lakes Adult 75 25 Pearse 1918 Wisconsin Lakes Adult 90 6 Rehder 1959 Des Moines River Iowa Adult 23 a, Sibley 1929 Lake Erie Young 62 14 Sigler 1958 Utah Lakes Adult 23 Vd Struthers 1929 Erie-Niagara System Adult 44 56 Struthers 1930 St. Lawrence Watershed, New York Adult 50 50 Present Study Bay of Quinte Watershed Young 91 8 Adult 74-98 1-8 1983 1966; Summerfelt et al. 1970; Eder and Carlson 1977) would seem to be explained largely by the local abun- dance and relative seasonal availability of aquatic invertebrates. A greater prevalence of chironomids in Bay of Quinte Carp than occurred in the two Kawar- tha waters, for example, may be related to the high abundance of such benthic organisms in the Quinte region (Johnson and Brinkhurst 1971). On the other hand, a high chironomid component could reflect a seasonal or site-specific bias. While fish eggs, including those of the Carp them- selves, have been recorded as infrequent constituents of Carp stomachs elsewhere (Cole 1905; Leach 1919), predations by Quinte Carp on fish eggs or fish was also of negligible consequence in this study. Conver- sely, Carp are in permanent co-existence with various piscivorous fishes, including esocid and centrarchid species in the Bay of Quinte watershed (MacKay 1963) and juvenile Carp are known to contribute to the diet of these species (McCrimmon 1968). Carp may, there- fore, bean amenity to local sport fisheries at this time. Seasonal variability in feeding activity, or intensity of feeding (Table 1), is apparent from mean volumes of digestive tract contents in the Carp stomachs. The most active feeding seems to occur in the spring and summer months (12.9 ml/kg fish body weight) when the water is warmest, and the least intensive during the cold winter months (1.4 ml/kg fish body weight), when metabolic rates are the highest and lowest, respectively. Similarly, the numbers of Carp contain- ing food are greatest in the spring and summer, and least in the winter. Also, the decreased abundance of debris and plant materials in their diet during the winter suggests a lesser foraging among vegetation. Macrophyte production at the time of sampling under a thick ice-cover, would have been at its lowest annual level. There is no evidence from the present study that Carp in the Bay of Quinte watershed, at present popu- lation levels, are deleterious to either indigenous aquatic vegetation or preferred fish species. Any sub- stantial increase in the Carp populations in these waters may, however, lead up to possibly severe environmental impact, such as has occurred histori- cally in other Canadian waters when Carp become too numerous (McCrimmon 1968). Acknowledgments We would like to thank Mrs. Barra Gots for her assistance in collating the material for analysis, and in preparing this manuscript. Our thanks to Dr. Lasenby for reading the manuscript in its original form, and for helpful suggestions. We are grateful to the Fishery Co-operative, Picton, for use of their facilities, and to Mr. Glen van Cott, who supplied us with samples POWLES, MACCRIMMON AND MACRAE: SEASONAL FEEDING OF CARP 297 from his winter commercial fishery for Carp in the Kawartha Lakes. Literature Cited Alikunhi, K. F. 1958. Observations on feeding habits of young Carp fry. Indiana Journal of Fisheries 5(1): 95-106. Bailey, R. M., and J. M. Harrison, Jr. 1945. The fishes of Clear Lake, lowa. Iowa State Journal of Science 20(1): 57-77. Cahn, A. R. 1919. The effect of Carp on a small lake: The Carp as a dominant. Ecology 10(2): 271-275. Cole, L. J. 1905. The German Carp in the United States. Reports of the U.S. Bureau of Fisheries 1904: 525-641. Eder, S.,and C. A. Carlson. 1977. Food habits of Carp and White Suckers in the South Platte and St. Vrain Rivers and Goosequill Pond, Weld County, Colorado. Transac- tions of the American Fisheries Society 106(4): 339-346. Ewers, L. A., and M. W. Boesel. 1935. The food of some Buckeye Lake fishes. Transactions of the American Fisheries Society 65: 57-70. Gaevskaya, N.S. 1966. The role of higher aquatic plants in the nutrition of animals of fresh-water basins, Vol. II. Edited by K.H. Mann; Translated by D.G. Maitland Muller. 1969. National Lending Library for Science and Technology, Boston Spa, Yorkshire, England. Gerking, S. D. 1950. A Carp removal experiment at Oliver Lake, Indiana. Investigations of Indiana Lakes and Streams 3(10): 373-388. Gunn, J. M., S. U. Quadri, and D. C. Mortimer. 1977. Fil- amentous Algae as a Food Source for the Brown Bullhead (Ictalurus nebulosus). Journal of the Fisheries Research Board of Canada 34(3): 396-401. Harrison, H. H. 1950. The foods used by some common fish. Proceedings of the lowa Conservation Commission’s Biology Seminar, July 1950: 31-84. Johnson, M. G., and R. O. Brinkhurst. 1971. Associations and species diversity in benthic macro-invertebrates of Bay of Quinte and Lake Ontario. Journal of the Fisheries Research Board of Canada 28: 1683-1697. Johnson, M. G., and G. E. Owen. 1971. Nutrients and nu- trient budgets in the Bay of Quinte, Lake Ontario. Journal of the Water Pollution Control Federation 43(6): 836. Kevern, N. R. 1966. Feeding Rate of Carp Estimated by a Radioisotopic Method. Transactions of the American Fisheries Society 95(3): 363-371. Leach, G. C. 1919. The artificial propagation of Carp. Uni- ted States Bureau of Fisheries Economic Circular No. 39: 1-19. MacCrimmon, H. R., and E. Skobe. 1970. The fisheries of Lake Simcoe. Ontario Department of Lands and Forests, Toronto. 140 pp. MacKay, H.H. 1963. Fishes of Ontario. Ontario Depart- ment of Lands and Forests, Fish and Wildlife Branch. 300 Pp. McAllister, D. E., and B. W. Coad. 1974. Fishes of Cana- da’s national capital region. Fisheries Research Board, Miscellaneous Special Publications 24. 204 pp. McCrimmon, H.R. 1968. Carp in Canada. Research Board of Canada Bulletin 165. 93 pp. Moen, T. 1953. Food habits of the Carp in northwest lowa lakes. lowa Academy of Science Proceedings 60: 655-686. Fisheries 298 Nakamura, N. 1955. Comparison of the Carp and the Cru- cian Carp from the piscicultural point of view. Bulletin of the Japanese Society for the Science of Fisheries 21(2): 77-81. Pearse, A. S. 1918. The food of the shore fishes of certain Wisconsin Lakes. Bulletin of the United States Bureau of Fisheries 35: 245-292. Rehder, D. D. 1959. Some aspects of the life history of the Carp, Cyprinus carpio in the Des Moines River, Boone County. lowa State Journal of Science 34: 11-26. Ricker, W. E. 1968. Methods for assessment of fish pro- duction in fresh waters. IBP Handbook Number 3. Black- well Scientific Publications. SBN 632 04660 0. Scott, W. B., and E. J. Crossman. 1973. Freshwater fishes of Canada. Bulletin of the Fisheries Research Board of Canada 184. 966 pp. Sibley, C. K. 1929. The food of certain fishes of the Lake Erie drainage basin. New York Conservation Department Supplements to the 18th Annual Report: 180-188. Sigler, W. F. 1958. The ecology and use of Carp in Utah. Logan Agricultural Experimental Station, Utah Univer- sity Bulletin 405. 63 pp. Smith, H. M. 1892. Report on the investigation of the fisheries of Lake Ontario. Bulletin of the United States Fisheries Commission, 1890, Volume 10: 177-215. Steel, R.G.D., and J. H. Torrie. 1960. Principles and procedures of Statistics. McGraw-Hill Book Company, Incorporated, New York. 481 pp. Struthers, P.H. 1929. A biological survey of the Erie- Niagara system. New York Conservation Department Supplements to the 18th Annual Report, Survey Number 3: 308-319. THE CANADIAN FIELD-NATURALIST Vol. 97 Struthers, P. H. 1930. A biological survey of the St. Law- rence watershed. New York Conservation Department Supplements to the 20th Annual Report, Survey Number 5: 217-229. Summerfelt, R. C., P. E. Mauck, and G. Mensinger. 1970. Food habits of the Carp Cyprinus carpio L. in five Oklahoma reservoirs. Proceedings of the Annual Confer- ence, Southeast Association of the Fish and Game Com- mission 24: 352-377. Tracey, H. C. 1910. Annotated list of the fishes known to inhabit the waters of Rhode Island. Pp. 354-360 in 40th Annual Report of the Commission of Inland Waters for Rhode Island. Uhler, F. M. 1944. Control of undesirable plants in water- fowl habitats. Transactions of the 9th North American Wildlife Conference. p. 302 Vass, K. F. 1957. Studies on food and feeding habits of the common Carp in Indonesian ponds (1) Changes in the diet during growth. Indo-Pacific Fisheries Council, 7th Ses- sion, Bondurg, Indonesia. 25 pp. Walberg, C. H., and W. R. Nelson. 1966. Carp, river carp- sucker, smallmouth buffalo, and bigmouth buffalo in Lewis and Clark Lake, Missouri River. United States Fish and Wildlife Service Research Report, Number 69: 1-30. Received 21 July 1982 Accepted 4 July 1983 Characteristics of Gray Wolf, Canis lupus, Den and Rendezvous Sites in Southcentral Alaska WARREN B. BALLARD and JAMES R. DAU Alaska Department of Fish and Game, P.O. Box 47, Glennallen, Alaska 99588 Ballard, Warren B., and James R. Dau. 1983. Characteristics of Gray Wolf, Canis lupus, den and rendezvous sites in southcentral Alaska. Canadian Field-Naturalist 97(3): 299-302. Physiography, internal structure, vegetation and spatial relationships of Gray Wolf (Canis /upus) den and rendezvous sites in southcentral Alaska were studied from 1976 to 1981. Eighteen denand six rendezvous sites had an average of 4.5 holes per site. No significant differences existed between the average number of holes between den and rendezvous sites. Mean dimensions of tunnel entrances and chambers are reported. No correlation existed between numbers of wolves which had used a site and numbers of holes. Den and rendezvous sites were usually located on knolls or hillsides with sandy, frost-free soil and mixed, semi-open stands of spruce (Picea spp.), Quaking Aspen (Populus tremuloides), and willow (Salix spp.). Wolves generally selected sites with south and/or east exposures. Mean elevation for all sites was 777 m while mean distance to water was 257 m. Average distance between contiguous, concurrent den sites was 45.3 km. Key Words: Canis lupus, Gray Wolf, den site, rendezvous site. Human encroachment and the development of nat- ural resources often have deleterious effects on wild- life. Minimizing these effects requires a sound under- standing of a species’ ecological requirements. While studying predator-prey relationships in southcentral Alaska, we investigated the denning ecology of Gray Wolves. This paper describes the external and internal characteristics of wolf den and rendezvous sites exam- ined, 1976 to 1981. Materials and Methods Den and rendezvous sites were located from fixed- wing aircraft while tracking radio-collared wolves from 1976 to 1981 in the Nelchina and upper Susitna River Basins of southcentral Alaska. The study area has been described elsewhere (Rausch 1967; Skoog 1968; Bishop and Rausch 1974; Taylor and Ballard 1979: Ballard et al. 1981, 1982). The location of each site was recorded on topographic maps (United States Geological Survey, scale 1:63 360). Ground examina- tions occurred after wolves had vacated the sites, usu- ally during August or September. Transportation to sites was usually by float-equipped fixed-wing aircraft. Den sites were distinguished from rendezvous sites by their temporal pattern of use and the presence and size of pups. Dimensions of tunnel structures were determined by entering tunnels with a flashlight and crawling as far as possible. Likewise, the disturbance to sites was minimal as no tunnels were excavated. All measurements were taken with steel measuring tapes calibrated in centimeters. Distances to water were estimated during ground and aerial observations. Dis- tances between concurrent main den sites for contigu- ous packs were determined by plotting den locations on topographic maps (U.S.G.S., scale 1:250 000) and measuring straight-line distances. Contiguous packs were identified by mapping home range territories from visual observations using the minimum home range method (Mohr 1947). When a pack simultane- ously reared two litters at different den sites, the den used by the established alpha female was denoted the main den and this was used for distance computations. Percent cover of over/understory vegetation was subjectively estimated during ground observations. The nomenclature of vascular plants follows Hulten (1968). Soil types were classified by particle texture and color, and elevation was determined from topo- graphic maps (U.S.G:S., scale 1:63 360), while the aspect of holes was determined with a compass. External openings of tunnels (“holes”) were strati- fied into two categories based upon size. The size index of each opening was computed using height and width measurements as if each hole was rectangular. There was a discontinuity in the area indices of holes around 1290 cm?. Holes with an area index greater than this were considered large. The terms “large” and “small” were used to describe each category of holes in lieu of “adult” and “pup” to preclude any erroneous inference, that certain age classes of wolves used holes of a certain size. This method does not exclude Red Fox (Vulpes vulpes) holes apparently unmodified by wolves at den or rendezvous sites that appeared to have been used by pups. However, fox-sized holes at sites that did not obviously show any indication of use by wolves were not included in these analyses. Shal- low holes less than | m deep were also excluded. Sta- EY) 300 tistical comparisons were either by F-test, f-test, linear regression, or Chi-square (Johnson 1980). Results Den and rendezvous site use by wolves were found to be fairly traditional. Of the 23 sites examined, at least six have been used three seasons since 1975; six have been used twice. The average elevation for 22 sites was 777 m= 110m ranging from 610-1097 m. Elevation of the study area ranged from 610 - 1829 m. Seventeen den sites contained an average of 2.5 + 1.2 large holes and 2.1 = 1.8 small holes. Six rendezvous sites contained an average of 2.5 + 1.0 large holes and 2.8 + 2.4 small holes. There was no significant difference (t = -0.74, pooled data, df = 21, P > 0.05) in the total number of holes between den and rendezvous sites. No significant correlation was found between the numbers of pup or adult wolves and the numbers of large or small holes per year of use (r= -0.17 and -0.35, respectively, P > 0.05). Mean height and width of 50 large holes was 42 + 9 cm by 51 = 13 cm, while 37 small holes averaged 23 + 6 cm by 28 = 7 cm. Tunnels usually descended at a steep (20-45° ) slope for approximately | m and then either ascended at a gentle (< 10°) slope or were level. A detailed analysis of tunnel lengths is not included in this discussion because tunnels often became too small to allow observers to reach or estimate their endpoints; how- ever, tunnel lengths generally ranged from |.5-5.5+ m. Small tunnels, 23 - 30 cm in diameter, often radiated from chambers. These tunnels appeared too small for use by adult wolves and could represent residual holes of foxes or be the excavations of wolf pups. Except for the area immediately adjacent to hole entrances, all tunnels were clear of vegetative debris and scats; none were observed lined with fur as described by Haber (1968). All den and rendezvous sites contained at least | chamber. Mean dimensions for 18 chambers were 45 + 14 cm high by 120 + 40 cm wide by 102 + 31 cm deep. Three den sites and one rendezvous site con- tained more than one chamber; however, not all tun- nels or tunnel complexes possessed a chamber. One tunnel complex which contairied a large chamber also contained two small chambers with dimensions of approximately 48 by 46 by 61 cm. Some chamber floors consisted of hard-packed sand that was unlike the powdery sand found on floors of other parts of the tunnels. Since the attendant female and pups proba- bly spent most of their time in the chamber, this was probably an incidental result. The distances from tun- nel openings to chambers averaged roughly 2 to 3 m. Generally, holes of both den and rendezvous sites were oriented in a southerly and/ or easterly direction THE CANADIAN FIELD-NATURALIST Vol. 97 (X2 = 25.0, n= 100, P< 0.01). The average distance to water for den and rendezvous sites was 257 + 263 m (n= 16), and there was no significant difference (t = 0.75, pooled data, P > 0.05) in mean distance to water between den and rendezvous sites. This distance was exaggerated because ephemeral water sources present during spring and early summer were dry by August or September and could not be measured. All sites were located in stable, sandy, frost-free soils which, in some cases, contained varying quantities of clay and gravel. Most den and rendezvous sites were located on elevated knolls with homogenous or mixed stands of spruce (Picea mariana and P. glauca), Quaking Aspen and willow with occasional Paper Birch ( Betula papy- rifera). Tree canopies were semi-open or interspersed with small glades providing places for sunning and vantage points. Rose (Rosa acicularis), decumbent willow (Salix spp.), Dwarf and Shrub Birch (Betula nana and B. glandulosa), grasses (Elymus innovatus, Poa spp., Calamagrostis spp.), Lowbush Cranberry (Vaccinium vitis-ideae), Labrador Tea (Ledum palus- tre), and Fireweed (Epilobium angustifolium) com- monly occurred in the undercover. Den sites were usually located roughly within the center of the observed pack territories but a few sites were located relatively close to territorial boundaries (Ballard et al. 1982). The mean distance between 19 pairs of concurrent natal dens of contiguous packs was 45 + 14 km; distances ranged from 27 to 67 km. Two wolf packs simultaneously raised two litters of pups at separate den sites. Distances between main and secondary dens within a pack were approximately 8 and 18 km, much less than the distance separating unrelated packs. There was no significant difference (F = 0.14, P > 0.05) in mean distance separating con- tiguous den sites among years. Discussion We did not observe the typical characteristics of den and rendezvous sites such as hollow logs, rock caves or dug-out Beaver (Castor canadensis) lodges des- cribed in other regions (Joslin 1966; Mech 1970; Peterson 1974, 1977; Allen 1979). Trees in this region usually do not attain a diameter adequate to house wolves. However, Beaver lodges were numerous and rock caves were also available in the study area. We did not observe pit dens which have been described for wolves from the Brooks Range of Alaska (Stephenson 1974). Most den and rendezvous sites in this area consisted of burrows dug either in elevated knolls or on the sides of hills. Holes were frequently located beneath spruce or aspen trees. Wolves often enlarge the tunnels of foxes and other small mammals (Murie 1944; Mech 1970; Stephenson 1983 1974; Haber 1977). The presence of Porcupine (Ere- thizon dorsatum) scats and quills suggests this species also uses tunnels while wolves are absent. Although we made no effort to quantify the abun- dance of potential wolf den sites, our casual observa- tions suggest that such sites are numerous in south- central Alaska. Therefore, we do not envision any shortage in wolf den sites due to human encroachment in the immediate future so long as Red Fox densities remain similar. Several den and rendezvous sites had beds located within 200 m of the main holes. These beds, roughly 60 cm in diameter, were located within sandy, litter- free areas that were | to3 mindiameter. Four of nine beds overlooked tunnel entrances. Five others were located on ridges out of sight of the holes and appeared to have been selected for their view and exposure to direct sunlight. Although one bed was beneath a large White Spruce, other beds did not indicate selection for areas beneath evergreens as reported by Joslin (1966). The low correlation between the number of holes and the number of wolves was not surprising. Mech (1970) reported that alpha females do most of the digging at den sites. Therefore, the number of holes used at a den site may be more attributable to the number of Red Fox holes present when the site was discovered by wolves than a function of pack size. Attendant yearlings and pups (6 - 8 weeks old) some- times enlarge existing holes and may dig new holes for play (C. Gardner and J. Westlund, personal commun- ication). Since wolves traditionally re-used den and rendezvous sites, on two occasions interchangeably, hole size and the number of holes could be influenced by the number of years a site was used. Soil density, rocks and root development undoubtedly affect tun- neling by wolves as well. Although pups may account for most of the digging at rendezvous sites (Peterson 1974), these other factors apparently overwhelm any correlation between the number of pups and holes at rendezvous sites (r = -0.42, n= 6). Mean dimensions reported herein for large holes fall within the 35 - 63 cm range reported by Mech (1970) for diameter of tunnel openings. Mean dimen- sions of small holes were slightly less, however. Mean dimensions for chamber size were also slightly less than those reported elsewhere (Criddle 1947; Joslin 1966; Stephenson 1974; Haber 1977), but this could be due to differences in the methods of measurement. The number of published, quantitative descriptions of chamber size is inadequate for meaningful comparisons. Haber (1968) provides the only description of a “nest” chamber lined with hair. Hair shed by adult wolves can adhere to the sides of tunnel structures but BALLARD AND DAU: GRAY WOLF IN SOUTHCENTRAL ALASKA 301 probably does not account for Haber’s observations. Our observations for the earliest date of denning, 13 April (Ballard et al. 1981), and dates observed for vernal molt (mid-March through April for captive wolves maintained in outside facilities at Fairbanks, Alaska: D. Hartbauer, personal communication) indicate little overlap between periods of denning and shedding. The average distance of 45 km between contiguous natal dens is comparable with the 40 km computed by Stephenson and Johnson (1973) for the northcentral Brooks Range. Knowledge concerning the average distance between natal dens in conjunction with knowledge of the types of sites used by wolves could allow managers to arrive at a crude approximation of number of denning wolf packs in an area. Also, searches of potential den site areas could aid in locat- ing denning wolves for study, as occurred during this project in 1978. Acknowledgments K. Schneider, S. Peterson; S. D. Miller and K. Wiley, all Alaska Department of Fish and Game (ADF&G) employees, reviewed early drafts of the manuscript. L. Metz, C. Gardner, S. Eide, D. Corne- lius, and J. Westlund, all ADF&G employees, assisted with collection of data. S. M. Miller (ADF&G) advised on statistical procedures; S. Brophy identified grasses. This study was funded in part by Alaska Federal Aid in Wildlife Restoration Project W-17-R and by the Alaska Power Authority. Literature Cited Allen, D. C. 1979. Wolves of Minong. Houghton Mifflin Co., Boston, Massachusetts. 499 pp. *Ballard, W.B., R.O. Stephenson, and T. H. Spraker. 1981. Nelchina Basin wolf studies. Alaska Department of Fish and Game. Pittman-Robertson Project Final Report, W-17-8, W-17-9, W-17-10, and W-17-11. 201 pp. *Ballard, W. B., C. L. Gardner, J. H. Westlund, and J. R. Dau. 1982. Susitna Hydroelectric Project Phase I Final Report — Big Game Studies — Volume V — Wolf. Alaska Department Fish & Game, Anchorage. 220 pp. Bishop, R. H. and R. A. Rausch. 1974. Moose population fluctuations in Alaska, 1950 - 1972. Le Naturaliste cana- dien 101: 559-593. Criddle, S. 1947. Timber wolf den and pups. Canadian Field Naturalist 61: 115. Haber, G. C. 1968. The social structure and behaviour of an Alaskan wolf population. MA thesis, Northern Michi- gan University, Marquette, Michigan. 198 pp. Haber, G. C. 1977. Socio-ecological dynamics of wolves and prey ina subarctic ecosystem. Ph.D. thesis, University of British Columbia, Vancouver, B.C. 786 pp. Hulten, E. 1968. Flora of Alaska and Neighboring Territo- ries. Stanford University Press, Stanford, CA. 1008 pp. 302 Johnson, R. 1980. Elementary Statistics, 3rd ed. Duxbury Press, Belmont, CA. 607 pp. Joslin, P. W. B. 1966. Summer activities of two timber wolf (Canis lupus) packs in Algonquin Park. M.Sc. thesis, Uni- versity of Toronto, Toronto, Ontario. 96 pp. Mech, L.D. 1970. The Wolf. Natural History Press, Garden City, New York. 384 pp. Mohr, C.O. 1947. Table of equivalent populations of North America small mammals. American Midland Natu- ralist 37: 223-249. Murie, A. 1944. The Wolves of Mt. McKinley. National Park Service Fauna Serial No. 5. United States Govern- ment Printing Office, Washington, D.C. 238 pp. Peterson, R. O. 1974. Wolf ecology and prey relationships on Isle Royal. Ph.D. thesis, Purdue University, Lafayette, Indiana. 368 pp. Peterson, R. O. 1977. Wolf Ecology and Prey Relation- ships on Isle Royal. National Park Service Monograph Serial No. 11. United States Government Printing Office, Washington, D.C. 210 pp. Rausch, R. A. 1967. Some aspects of the population ecol- ogy of wolves, Alaska. American Zoologist 7: 253-265. THE CANADIAN FIELD-NATURALIST Vol. 97 Skoog, R. O. 1968. Ecology of caribou (Rangifer tarandus granti) in Alaska. Ph.D. thesis, University of California, Berkeley, California. 699 pp. *Stephenson, R. O. 1974. Characteristics of wolf den sites. Alaska Department of Fish and Game. Pittman- Robertson Project Final Report, W-17-2, W-17-3, W-17- 4, W-17-5 and W-17-6. 27 pp. *Stephenson, R.O., and L. Johnson. 1973. Wolf report. Alaska Department Fish and Game. Pittman-Robertson Project Report, W-17-4. Job No. 14.3R, 14.4R, 14.5R, 14.6R and 14.7R. 52 pp. Taylor, K. P., and W. B. Ballard. 1979. Moose movements and habitat use along the upper Susitna River near Devil’s Canyon. Proceedings of the North American Moose Con- ference and Workshop 15: 169-186. *Available from Alaska Department Fish and Game, P.O. Box 3-2000, Juneau, Alaska 99802 Received 12 May 1982 Accepted 15 June 1983 Denning and Foraging Habits of Red Foxes, Vulpes vulpes, and Their Interaction with Coyotes, Canis latrans, in Central Alberta, 1972-1981 DICK DEKKER 3819-112A Street, Edmonton, Alberta T6J 1K4 Dekker, Dick. 1983. Denning and foraging habits of Red Foxes, Vulpes vulpes, and their interaction with Coyotes, Canis latrans, in Central Alberta, 1972-1981. Canadian Field-Naturalist 97(3): 303-306. Red Fox (Vulpes vulpes) dens near Edmonton, Alberta, were often dug under barns and uninhabited farm buildings. Ten litters of pups numbered from 3 to 7, and averaged 5. Two adults brought food to each litter, but at one den there were three foraging adults. Active play between two adults and pups was seen at one site. A dead pup was carried away from the den by an adult. A male fox led three pups away to another den during mid-day, while one or two pups remained at the first site where they were later seen with the vixen. At one den all prey seen brought to the young were small rodents. Hunting methods of adult foxes catching voles are described for summer and winter. Other prey was seldom taken. Coyotes (Canis latrans) chased foxes on nine occasions. Interspecific strife with Coyotes is believed to be the reason foxes occur close to human habitation in central Alberta. Key Words: Red Fox, Vulpes vulpes, Alberta, foraging, denning, interaction with Coyotes, Canis latrans. The Red Fox’s (Vulpes vulpes) distribution in Alberta covers all but the treeless prairies of the south (Soper 1964). Banfield (1974) considered the species very rare on the southern plains. I reported Red Foxes virtually absent in central Alberta until the late 1960's when it became locally common in agricultural regions south and east of Edmonton (Dekker 1973). This paper gives information on den sites, litter sizes, parental care, foraging habits and interactions with Coyotes (Canis latrans). Study Area and Methods From 1972 to 1981, I used three study areas of about 7 km? each, but local extirpation of foxes forced me to abandon two areas in 1975. The first area was situated 6 km east of the town of Tofield; the second within the eastern outskirts of the city of Edmonton; and the third 4 km west of Leduc. Areas 1 and 2 each included one occupied fox den until 1974; the third area contained two dens with pups in 1975, but only one thereafter. All three areas are dominated by gently sloping agricultural fields, intersected by country roads ona l-mile grid pattern. The Edmonton and Leduc areas contain a narrow creek ravine with mixed woods, mainly Trembling Aspen (Populus tremuloides) and willow (Salix spp.). The Tofield study area straddles the provincial Highway No. 14 and the C.N.R. tracks, and it contains several wet depressions with marshy vegetation. In general each quarter section of land (64 ha) has a cluster of farm buildings, usually flanked by a windbreak of White Spruce (Picea glauca) ora mixed woodlot. To study the foxes’ foraging habits during winter and to locate denning sites, I followed fox tracks in snow on a total of 124 days (466 h). Dens were checked for signs of use in early June. Several occu- pied dens were reported to me by local residents. I observed occupied sites for varying lengths of time on 70 days in June for a total of 215 h, usually between 0600 and 1000 or between 1900 and 2200. Most watch- ing was done from a car parked on a country road 100-500 m from the den, and I used 10 binoculars and a 20x telescope. I did not seek data on specific subjects, and neither watched systematically nor did any experimentation. Observations that seemed noteworthy were written down upon arriving home. Results and Discussion Pelage Variation All Red Foxes observed during this study were of the red phase. None resembled the so-called cross and silver phases, which together can constitute 22-61% of Canadian populations (Banfield 1974). The foxes under study probably belonged to the subspecies V. v. regalis, the Northern Great Plains Red Fox, the larg- est and palest of the western Red Foxes (Soper 1964). Toward late spring, some foxes had patches of bleached pale-yellow winter fur on their backs and necks, contrasting with the brown summer fur of the flanks and legs. There was a noticeable variation in the extent of black on lower legs, and of white on the tip of the tail. Six of the pups had one or two white front feet or white toes. At those den sites where I had a good view of both adults together, I noted a difference in size of the two 303 304 foxes. I assumed that the larger animal was the male, and the smaller the female or vixen. In June, the male appeared to be relatively thicker-furred with a longer and fuller tail. His pelage was generally paler, proba- bly because he spent more time above ground than the vixen. Dens During winter, I frequently flushed foxes from their daytime resting places in the open or in a woodlot. However, some foxes apparently spent the day or part thereof underground in dens or borrows, in dry bea- verlodges, or under man-made structures such as barns, haystacks, uninhabited farm houses, or grain elevators. In the Leduc study area, all of six fox families denned under isolated field barns or deserted farm houses. In the Edmonton area, two litters alternately used two dens under buildings and one ground den. In the Tofield area, two fox families used three ground dens in knolls on grain fields or pasture land. Farmers often ploughed or cultivated over ground dens, evenif they were occupied. Inhabited dens were re-opened by the foxes. Ground dens had three to six entrances. Dens under buildings had access holes on two or more sides of the structure. I did not visit the study areas during Apriland May. The first date I located occupied dens was | June. The last date I observed adults with pups at a den was 24 June. During the three weeks or so of my observa- tions, six fox families changed den sites once and one family moved twice. On 19 June 1981, I saw a depar- ture, which was probably prompted by my only attempt to photograph the foxes. My car had been parked about 60 m from the den during the morning. At midday, an hour or so after I had left the field and was watching from the road, the larger of the pair of foxes, probably the male, approached the denning barn and left shortly with three pups, trotting over the fields in a single file. One pup, which happened to be on a side of the barn where it could not have seen the departure of its siblings, stayed behind. Presently it was joined by the vixen, which habitually spent the day under the barn. Next morning, she was there with two pups, but the site was deserted the following day. The foxes had moved toa barn ona farm yard, 0.7 km away. Pups and parental care The number of pups in 10 litters that I observed in June averaged 5 and ranged from 3 to 7. Banfield (1974) gives average litter sizes as 5.1. In 1976, I reported a mortality rate of 46% for seven litters dur- ing their first three months of life in Central Alberta (Dekker 1976). As main causes of death I listed shoot- ing, traffic and dogs. Direct persecution by humans THE CANADIAN FIELD-NATURALIST Vol. 97 was probably the reason foxes disappeared from two of my study areas in 1974. On 21 June 1974, at 0630, I observed a pair of foxes transport a dead pup away from the den. The smaller of the pair, probably the vixen, twice lifted the carcass in her mouth but dropped it again. She left and pres- ently returned with the male, who carried the dead pup 220 m while the vixen followed. Twice the male put the carcass down briefly and finally abandoned it 150 m from the spot where I was sitting on the edge of a woodlot. The carcass was autopsied by J. D. Henry. Cause of death was gun shot wounds. The dead pup’s weight was 2.7 kg. By the middle of June, the pups may spend time above ground at any hour of the day, especially in early morning and in the evening when they are most active and involve each other in play. At two dens I saw the vixen join in play with the pups by pushing and pursuing them. At another den both adult foxes played with the pups. The male was the most active. He began by “bowling over” a pup, and then he ran around the denning barn repeatedly at great speed, ambushing the pursuing pups around the corner and knocking them over. On one occasion, the vixen joined in the chase, but on others she sat on her haunches near the barn. The male jumped over her during his race around the barn and repeated his leap over the sitting vixen three times. Playful interactions between vixen and pups have been described by other field observers (e.g., Tembrock 1958; Allison 1971) but, to my knowledge, active play involving both adults and pups has not been reported before. At all occupied dens Isaw two different adult foxes bring food to the pups, but at one den there were three adults, probably one male and two vixens, which I saw together on several occasions. One vixen could be distinguished by a thin tail. This fox remained some- what distant from the other two, which frequently “nose-touched” each other but did not exchange such greetings with the thin-tailed animal. All three foxes brought food to the six pups, which appeared to be of the same size and age. Dens with two vixens and two litters of pups of different sizes and age were reported by Sheldon (1950) from New York State, and by Niewold (1976) from the Netherlands. Murie (1961) saw three adults at a den in Alaska. In experimental fox groups, studied in enclosures, Macdonald (1979) observed two non-breeding vixens bring food to the pups of the breeding vixen within their group, but he saw only circumstantial evidence that barren vixens did so in similar groups of wild foxes in Britain. Food and hunting habits I was usually unable to identify prey items brought to the pups by the adult foxes because of the distance from which I was watching. However, at one Edmon- 1983 ton area den I had exceptional opportunities for close observation. All of 18 food deliveries consisted of small rodents, probably Meadow Voles (Microtus pennsylvanicus) which abounded in surrounding fields. Twelve loads contained 3-7 rodents each. On 20 June 1974, from 0630 to 0800, the foxes delivered five loads, totalling about 24 voles. There appeared to be an ample supply of food. At times, the vixen collected the carcasses of rodents scattered about the den area, and carried them inside the den under the barn. Sca- venging birds such as Magpies (Pica pica) regularly took dead rodents from the immediate vicinity of the den. I frequently observed adult foxes hunting and saw about 200 prey caught. A few of these were Pocket Gophers (Thomomys talpoides) and frogs, but the vast majority were voles and mice. During June of 1974, the three foxes at the Edmonton area den cap- tured voles with apparent ease ona fallow field, where I observed them on 15 dates. On 20 June, I saw the male fox capture 14 rodents in 25 minutes. He ate four, cached five, and carried three to the den, while he added two along the way. The hunting technique used by foxed to capture small rodents is the pounce, as reported by many observers (e.g. Smith 1944). Henry (1976, 1980), who described the Red Fox’s hunting methods in detail, measured a lunge on level ground at 4.5m and a downhill one at 7.5 m. Ona fallow field in the Edmon- ton area, foxes seldom pounced and caught most prey with a quick stabbing thrust of the front feet, while the animal’s hind legs remained on the ground. If the fox failed to pin the prey down under its feet, it began to dig with quickly alternating front feet. Several times, while I was watching through the scope, I saw a vole emerge from the soil and run. The fox seized it in its mouth in the next instant. The foxes killed voles witha few chewing bites, and then either ate the prey, cached it, or carried it to the den. In the evening and early morning, I frequently saw a resting fox rise and begin hunting. It ate the first 2-6 voles it caught, and then dropped several more onto the ground. Eventually, after 3-6 voles had been left on the field, the fox collected the carcasses and headed towards the den. Sometimes, it deposited its prey load and hunted addi- tional prey, or dug up a cache. Newly acquired voles were either added to the load, or cached. Several times, a Magpie followed a hunting fox, but the bird was chased away by the fox if it had left prey on the ground. During winter, I frequently observed foxes foraging on snow-covered fields or grassland. They hunted small rodents by stalking. After a pause to pinpoint the prey, either by smell or hearing, or both, the fox pounced and attempted to trap it under its feet. In DEKKER: RED FOXES AND COYOTES IN CENTRAL ALBERTA 305 shallow loose snow, these pounces covered distances of up to 4m. On thick and crusted snow, the foxes often walked slowly to and fro, head low as if smelling or listening, and pounced straight up and down, land- ing front feet first in an attempt at breaking the crust. If successful, they often dug extensively. Probably some foxes pounced in this vertical manner to try and collapse snow tunnels used by rodents. The fox secured such prey by digging and thrusting its snout into the snow. However, their slow search and vertical pounce method apparently is most often used to uncover cached food items and rodent nests with young. Some Foxes lunging in this vertical way left craters of up to 0.6 m deep, reaching down to the soil. At the bottom of these craters or onthe snow nearby, I found torn grass-made nests, probably of voles. In several instances, I observed foxes chew on small prey items that I believed to be baby voles. It is possible that foxes locate vole nests with young by sound. On 30 November 1970, shrill squeaks guided me to a nest with six baby voles under thick matted vegetation. On 15 February 1974, I found where a fox had dug numerous craters in 30-80 cm of wind-packed snow covering unharvested swathed grain. Pheasant (Pha- sianus colchicus) tracks led from one crater to the next. Apparently the bird had been feeding on grain exposed by the digging fox. Foxes caught some prey opportunistically. Tracks revealed that a fox had jumped out of its bed to seizea vole running on top of the snow. It was cached nearby. Such passive methods of hunting may be a deliberate tactic. I often found fox beds on the edge of ditches crossed by numerous tracks of small rodents, or in between the burrow mounds of Ground Squirrels (Spermophilus richardsonii) after they had emerged from hibernation in late winter or early spring. During winter, I seldom found evidence that the foxes hunted prey other than mice and voles. On two occasions, a fox had stalked a covey of Gray Par- tridges (Perdix perdix). One covey, huddled together in a communal bed, had escaped. The other covey, which had been feeding or sleeping in tunnels under crusted snow, had one bird captured, partly eaten and cached. Snowshoe Hares (Lepus americanus), occur- ring in several woodlots in all study areas, were sel- dom hunted. When snow cover was thick and soft in the woods, the foxes did not enter. Tracks indicated that one hare had been chased unsuccessfully for about 400 m on an open field. A Red Squirrel (7ami- asciurus hudsonicus), travelling over open fields from one woodlot to another, had been intercepted by a fox, which I saw carrying and caching the prey. During June of 1975, I frequently observed foxes hunt in a pasture where Pocket Gopher mounds were numerous. The foxes stalked slowly, apparently sens- 306 ing the gopher’s underground movements, and caught them ina swift dash. They also seized small prey items which I believed to be frogs. I rarely observed the foxes stalk and rush birds, although feathers of several species of ducks and small passerines were much in evidence around den sites. On 18 and 19 June 1975, I observed two foxes scavenge from a recently dead carcass of a domestic calf. Large chunks of meat and internal organs were carried to the den, about 0.5 km away. The calf had also been fed upon by one or more Coyotes and domestic dogs, which probably opened the carcass. Interaction with Coyotes On 18 June 1975, I saw a Coyote crouch and wait for an approaching fox which dodged the Coyote’s rush and escaped. On eight occasions, I observed foxes follow and bark at Coyotes that were within 0.8 km from a fox den with pups. The Coyotes repeatedly chased the foxes, which dodged and fled until the Coyote gave up. Murie (1961) saw foxes follow and bark at Wolves (Canis /upus) near fox dens in Alaska. On 16 June 1974, I saw a fox approach and bark at two Coyotes, each of which briefly chased the fox. Eventually, the Coyotes entered a woodlot in which the young foxes were hidden. As the fox came near, one Coyote rushed out of the trees and chased the fox along the woodlot, until the second Coyote emerged and took over. Eventually, this Coyote gave up too, but the fox didn’t return. The total distance of the chase was about | km. On 21 June 1977, I was observ- ing four fox pups sleeping near a field den, when suddenly a Coyote ran towards them. Some distance away, the vixen barked a warning and the pups bolted into the den in the nick of time. Coyotes may be a serious danger to Red Foxes, especially the young, and foxes avoid Coyotes by spatial segregation. In central Alberta, Red Foxes occur mainly near human habitation and roads. I have found them denning in the immediate vicinity of farms with large dogs, which appeared to keep Coyotes away. Rural residents of the Tofield region told me of three cases where foxes hid their pups under buildings on their yard. Conversely, I have never seen foxes and only once found their tracks on the more remote pasture lands around Beaverhill Lake near Tofield where I have spent about 70 days each year since 1972 and where I saw one or more Coyotes on 120 days. THE CANADIAN FIELD-NATURALIST Vol. 97 Red Foxes have only quite recently established themselves on the farmlands of central Alberta (Dekker 1973), perhaps by taking advantage of a depressed Coyote population decimated during the late 1960’s by snowmobile hunters. Foxes will proba- bly continue to exist on the Alberta farmlands, but only locally in areas where Coyotes are scarce or absent. Acknowledgments I thank J. D. Henry for critically reading an earlier version of the manuscript, for making many helpful comments and for providing relevant literature. J. Gunson read the final version. Literature Cited Allison, L. M. 1971. Activity and behaviour of Red Foxes in central Alaska. M.Sc. thesis, University of Toronto. 92 Pp. Banfield, A. W. F. 1974. The mammals of Canada. Univer- sity of Toronto Press and National Museum of Natural Sciences. 438 pp. Dekker, Dick. 1973. Red Foxes make a comeback after 30 years. Blue Jay 31: 43-44. Dekker, Dick. 1975. Red Foxes incentral Alberta. Edmon- ton Naturalist 3: 7-8. Dekker, Dick. 1976. Mortality rates of Red Fox pups, and causes of death of adult foxes in central Alberta. Alberta Naturalist 6: 65-67. Henry, J. D. 1976. Adaptive strategies in the behaviour of the Red Fox. Ph.D. thesis, University of Calgary. 259 pp. Henry, J. D. 1980. Fox hunting. Natural History 89: 60-69. Macdonald, D. W. 1979. “Helpers” in fox society. Nature 282: 69-71. Murie, A. 1961. A Naturalist in Alaska. Doubleday and Co. Inc. New York. Niewold, F. J. J. 1976. Aspecten van het sociale leven van de vos. Natura 73: 234-241. Sheldon, W. G. 1950. Denning habits and home ranges of Red Foxes in New York State. Journal of Wildlife Man- agement 14: 33-42. Smith, W. P. 1944. Red Fox’s method of hunting field mice. Journal of Mammalogy 25: 90-91. Soper, J. Dewey. 1964. The mammals of Alberta. The Queen’s printer, Edmonton. 402 pp. Tembrock, G. 1958. Spielverhalten beim Rotfuchses. Zoo- logische Beitrage. 3: 423-496. Received 14 June 1982 Accepted 27 June 1983 Dynamics of the Winter Distribution of Rosy Finches, Leucosticte arctoa, in Montana PAUL HENDRICKS! and JON E. SWENSON?2 'Department of Zoology, University of Montana, Missoula, Montana 59812 2Montana Department of Fish, Wildlife and Parks, 1001 Ridgeway Drive, Livingston, Montana 59047 Hendricks, Paul, and Jon E. Swenson. 1983. Dynamics of the winter distribution of Rosy Finches, Leucosticte arctoa, in Montana. Canadian Field-Naturalist 97(3): 307-310. During the winter of 1977-78 Rosy Finches (Leucosticte arctoa ssp.) were reported across the eastern prairies of Montana, whereas in 1978-79 they were limited to grasslands adjacent to the western mountains. Winter snowfall and days of snow cover on the prairies were similar during both winters. January — July precipitation in 1977 was 23% below normal, whereas it was 50% above normal in 1978. Variability in the winter distribution of Rosy Finches appeared to bea function of early spring and summer precipitation on the prairies (more rain produced more seeds and taller plants) and snow depth relative to plant height (which affected food availability). When food was scarce the finches moved farther onto the prairies. Key Words: Rosy Finch, Leucosticte arctoa ssp. winter distribution, Montana. Rosy Finches (Leucosticte arctoa tephrocotis and L. arctoa atrata) breed in alpine areas of western North America, and overwinter primarily on lowland grasslands (Bent 1968). On the wintering areas these finches forage predominantly on seeds of grasses, mustards, Russian thistle (Sa/sola kali), composites, etc. (Cameron 1907, Bent 1968). During the winters of 1977-78 and 1978-79 we noted differences in Rosy Finch distribution in Montana. In this paper we corre- late these differences with food availability and cli- matic conditions. Methods Distributional records were obtained from Christ- mas Bird Counts, published observations in American Birds, a two-year wildlife inventory on an area in extreme eastern Montana (W. Matthews, Bureau of Land Management, personal communication), and field observations in December through March dur- ing 1977-78 and 1978-79. The field observations were made opportunistically, but the same general areas were covered each winter. Climatological data were compared for Billings, Glasgow, Great Falls, Lewis- town, and Miles City in central and eastern Montana. Estimates of yields of dryland cereals, dryland hay, and livestock numbers were obtained from the Mon- tana Department of Agriculture (1978, 1980). Results and Discussion During the winter of 1977-78 Rosy Finches were observed throughout most of Montana east of the Continental Divide, based on all sources of data (Table 1, Figure 1). In contrast, during the winter of 1978-79 Rosy Finches were not reported ranging onto the eastern prairies, but were restricted to grasslands near the western mountains (Table |, Figure 2). The findings of Turner and Taylor (1980) were consistent with our observations. Of the seven observations they reported of Rosy Finches east of their normal winter range in the Northern Great Plains during these two winters, six occurred in 1977-78 and only one in 1978- 79. Turner and Taylor (1980) reported two sightings from the 1978-79 winter, but one of these, in Bismark, North Dakota, was actually observed during the 1977- 78 winter (Johnson and Randall 1980). A similar dis- tributional variation may have been noted by Cameron (1907: 403), who reported an absence of Rosy Finches in eastern Montana during the winter of 1905-06 “in great contrast to previous years”. The observed difference in winter distribution of Rosy Finches may be related to a difference in food availability during the two winters. Total snowfall and days of snow cover were similar during both winters (157.2 cm snowfall and 120.8 days with> 2.5 cm snow cover in 1977-78, compared with 147.9 cm and 136.6 days in 1978-79). Although seed production data on wild food crops utilized by wintering Rosy Finches were not available, a positive correlation between summer precipitation and grass seed production has been reported (Ellison and Woolfolk 1937, Brown et al. 1979). Climatological data show that precipitation during January through July 1977 (i.e. preceding and during the 1977 growing season) was 23% below nor- mal (xX = 191.5 mm for the five weather stations). Dur- ing the same period in 1978, precipitation was 50% above normal (x= 372.1 mm). Yields of dryland cereal crops responded accordingly, with less seed production during the drought year of 1977 (25.4 bushels/ acre for wheat and barley) than the wet year of 1978 (32.8 bushels/acre), a difference of 29.1% 307 308 THE CANADIAN FIELD-NATURALIST Vol. 97 TABLE 1. Numbers of Rosy Finches Reported on Christmas Bird Counts in Montana, 1977-78 and 1978-79. 1977-78 1978-79 L. a. L. a. Ih Ge ona: Region® and Count Location tephrocotis atrata tephrocotis: atrata Western Big Fork (BF)? 0 0 0 Bozeman (BZ) 0 0 0 0 Ennis (E) 417 0 ] 0 Glacier Park (G) 0 0 0 0 Missoula (M) 26 0 0 0 Ninepipe (N) NCS NC 0 0 Stevensville (S) 0 0 NC NC Troy (T) 0 0 0 0 Whitehall (W) 0 0 NC NC Central Billings (B) 202 0 0 0 Chester (C) 301 0 75 0 Helena (H) 0 75 12 0 Highwood (HW) 0 0 0 0 Lewistown (L) 129 0 0 0 Park County (P) 22 250 103 Townsend (TN) NC NC 0 0 Yellowstone Park (Y) 9 20 150 40 Eastern Fort Peck (F) 0 0 0 0 Miles City (MC) 0 0 0 0 and 7 EE Regions as defined by Montana Department of Agriculture (1978) and shown in Figures | and 2. >T ocations shown on Figures | and 2. “No count. (Montana Department of Agriculture 1978, 1980). Stem height data were also unavailable, but our impression was that prairie grasses were much taller in 1978 than in 1977. Long-time area ranchers also reported that the growth of prairie grasses was excep- tional in 1978. Ellison and Woolfolk (1937) noted marked differences in plant height in southeastern Montana related to dramatic variations in summer precipitation, and Newbauer et al. (1980) have shown a positive correlation between forage yield and grow- ing season precipitation. Yields of dryland hay (mostly native grasses) increased 16.9% from 1977 (1.31 tons/acre) to 1978 (1.49 tons/acre), perhaps indicating a change in stem height. In addition, the 10.5% decrease in cattle numbers from 1977 (2,318,000) to 1978 (2,074,000) in central and eastern Montana resulted in reduced grazing pressure when seed and forage production were higher. Therefore, more food was probably available to Rosy Finches during the winter of 1978-79 than in 1977-78, even though snow cover was similar for the two winters. Based on observations of population fluctuations at a roost in Utah, King and Wales (1964) concluded that snow cover, which impedes feeding by Rosy Finches, was more significant than ambient temperature in limiting winter flock distribution and in causing tem- porary displacements of populations. We agree that snow cover is probably important in affecting food availability and thereby in influencing winter distribu- tions of ground foraging granivores. However, Dun- ning and Brown (1982) showed that density of winter- ing sparrows in southeastern Arizona can be correlated with summer precipitation (reflected in the resulting seed crop size). Bock and Lepthien (1976) suggested that seed crop size is the primary factor behind “eruptive” southward migrations of boreal seed-eating birds. The winter-to-winter variations in Rosy Finch distributions observed during this study may be analogous in some aspects to the pattern of winter movements found in granivorous species which exhibit eruptive invasions, in that these movements are a response to local shortages of winter food resources. When food availability was more restricted the finches moved farther onto the prairies. Food availability is determined by seed production and stem height from the growing season prior to winter, and by snow cover during winter. The combined effects of these variables probably are important in regulating Rosy Finch winter distribution in Montana. 1983 HENDRICKS AND SWENSON: ROSY FINCHES IN MONTANA 309 | ijt ills Mh Lh ; a : ill AN c AW })) WAT | MH i} \} \) | Hi HH c MN a Ho ===. neve i ¥, ay | Ht : al | | | A Ki ) : ie se af a 9 FiGcureE |. Winter Rosy Finch distributionin Montana, 1977-78. (Circles and letters refer to Audubon Christmas Counts, see Table |, the hexagon to atwo-year study by Matthews (personal communication), and triangles to field observations. Open circles and hexagon indicate no Rosy Finches were sighted; filled circles, hexagon and triangles indicate the presence of Gray-crowned Rosy Finches (Leucosticte arctoa tephrocotis); boxed open circles indicate the presence of Black Rosy Finches (L. a. atrata); boxed filled circles and triangles indicate the presence of both subspecies. Shaded areas indicate mountains. The state is divided into western, central, and eastern Montana, corresponding with the regions used by the Montana Dept. of Agriculture. The dashed line is the Continental Divide.) . . \ = | - ‘ly Ne Sin Wim “S: | \ im! i VA : “\ (4 4 re | | } FIGURE 2. Winter Rosy Finch distribution in Montana, 1978-79. (See Figure | for explanation of symbols.) 310 Acknowledgments We thank Richard E. Johnson and Richard Hutto for their valuable criticism of an earlier draft of the manuscript. We also thank Richard E. Johnson and William Matthews for making their observations available to us. Literature Cited Bent, A. C. 1968. Life histories of North American cardi- nals, buntings, towhees, finches, sparrows and allies. Uni- ted States National Museum Bulletin 273, Part 1. Bock, C. E.,and L. W. Lepthien. 1976. Synchronous erup- tions of boreal seed-eating birds. American Naturalist 110: 559-571. Brown, J. H., O. J. Reichman, and D. W. Davidson. 1979. Granivory in desert ecosystems. Annual Review of Ecology and Systematics 10: 201-227. Cameron, E.S. 1907. The birds of Custer and Dawson Counties, Montana. Auk 24: 389-406. Dunning, J. B., and J. H. Brown. 1982. Summer rainfall and winter sparrow densities: a test of the food limitation hypothesis. Auk 99: 123-129. THE CANADIAN FIELD-NATURALIST Vol. 97 Ellison, L., and E. J. Woolfolk. 1937. Effects of drought on vegetation near Miles City, Montana. Ecology 18: 329-336. Johnson, K.J., and R.N. Randall. 1980. Gray-crowned Rosy Finch at Bismark, North Dakota. Prairie Naturalist 12: 24. King, J. R., and E. E. Wales, Jr. 1964. Observations on migration, ecology and population flux of wintering Rosy Finches. Condor 66: 24-31. Montana Department of Agriculture. 1978. Montana agri- cultural statistics. Volume 19. County Statistics 1976 and 1977. Montana Department of Agriculture. 1980. Montana agri- cultural statistics. Volume 20. County Statistics 1978 and 1979. Newbauer, J. J., III, L. M. White, R. M. Moy, and D. A. Perry. 1980. Effects of increased rainfall on native forage production in eastern Montana. Journal of Range Man- agement 33: 246-250. Turner, B. N., and P. Taylor. 1980. The Gray-crowned Rosy Finch in Manitoba and vacinity. Blue Jay 38: 238-245. Received 25 May 1982 Accepted 17 March 1983 Variation in Red Fox, Vulpes vulpes, Summer Diets in Northwest British Columbia and Southwest Yukon DONALD M. JONES and JOHN B. THEBERGE Department of Biology, University of Waterloo, Waterloo, Ontario N2L 3G] Jones, Donald M., and John B. Theberge. 1983. Variation in Red Fox, Vulpes vulpes, summer diets in northwest British Columbia and southwest Yukon. Canadian Field-Naturalist 97(3): 311-314. Diets of Red Fox (Vulpes vulpes) were examined in alpine, subalpine, and boreal forest habitats on and near the Chilkat Pass, British Columbia. Above the boreal forest habitat Arctic Ground Squirrel (Spermophilus parryii) and voles and mice predominated, especially in whelp scats from dens, but also in adult scats collected elsewhere. Non-den scats contained significantly fewer ground squirrel and more voles and mice remains. Scats at subalpine dens were significantly different from those at alpine dens, the latter containing fewer ground squirrel items. Scats collected from the boreal forest zone were dominated by Snowshoe Hare (Lepus americanus), and a greater diversity of species. Significant difference in results were found using “equal weight” and “random sampling” methods of analysis. These data show the considerable dietary plasticity and opportunism of Red Fox predation in this heterogeneous northern environment. Key Words: Canidae, Red Fox, Vulpes vulpes, food habits, predation, scat analysis. Dietary studies of the Red Fox (Vulpes vulpes) in North America indicate that their major prey is small SN rodents (eg. Microtus spp.) and lagomorphs (eg. Sy/- STUDY AREA vilagus spp.) (Errington 1935; Scott 1943; Scott and 4 Klimstra 1955; Stanley 1963; Pils and Martin 1978). Ry : Scat - Collecting Transect Although the range of the Red Fox extends to the \ oO & Den Site Canadian Arctic Archipelago[70° N] (Banfield 1974), no dietary studies have been conducted north of 55° N except those of Murie (1944). The prey in northern environments is characterized by wide spacial, sea- sonal, and annual differences in abundance. In this paper, we compare the diets of the Red Fox in boreal forest, subalpine, and alpine habitats of northwest British Columbia, and southwest Yukon Territory. We also compare two different methods of scat analysis: the traditional “equal weight” method, used by the authors mentioned above, and the “ran- dom selection” method, used less frequently but with increasing commonness in canid food habits studies (Wyman et al. 1975; Johnson and Hansen 1977a, 1977b; Green and Flinders 1981). Provincial Boundary International Boundary Study Area The study area was divided into the Haines Road and Alaska Highway sections. The Haines Road Sec- tion was the intensive study area, located along the Haines Road from km 91 to km 144 (Figure 1), in the Chilkat Pass. It is above tree-line in the St. Elias Mountains, at elevations exceeding 900 m. The ALASKA Alaska Highway Section is located along the Alaska ‘Highway from Haines Junction, km 1622, northward to km 1823, and is within the Shakwak Trench where 59° 30’ N FiGuRE |. Location of the study area showing five den sites the lowest elevation is approximately 600 m. and eight scat-collecting transects. The elevations of Three general vegetation zones occur in the study the den sites are as follows: den A 914 m; den B914 m; area. The boreal forest zone, with White Spruce den C 1005 m; den D 945 m; den E 1219 m. Slt St (Picea glauca) predominant (Rowe 1972). Stands of Trembling Aspen (Populus tremuloides) and Balsam Poplar (Populus balsamifera) occur as the result of fire disturbance and extend to approximately 900 m. Dense, subalpine shrub vegetation occurs between approximately 900 and 1000 m, characterized by Dwarf Birch (Betula glandulosa) and willow (Salix spp.) 1.0-1.5 m high. Scattered pockets of White Spruce occur infrequently. Higher elevations support alpine tundra lichens (predominantly Cladonia alpes- tris), Crowberry (Empetrum nigrum), prostrate wil- lows, dwarf birch and ericaceous shrubs. Repeated glaciation has resulted in a variety of surficial deposits and soils in both sections (Oswald and Genyk 1977). The mean May-September and January temperatures are 9° C and -21° C respectively (data from Haines Junction, Oswald and Senyk 1977). Methods Scats were collected from May to September 1978 every two weeks at five fox dens (two in the subalpine and three in the alpine) and along eight transects that were established on infrequently used roads and clear- ings along an abandoned pipeline (Figure 1). Except for one alpine den, all scats from the Alaska Highway Section were collected below the tree-line. To avoid confusion with Coyote (Canis latrans) and Wolf (Canis lupus) scats, only scats collected at fox dens or having a maximum diameter less than 19 mm were considered to be fox (Jones 1980). Known fox scats collected at dens ranged up to 23 mm, so choosing 19 mm as a maximum diameter reduced possibilities of overlap with coyote scats. The degree of weathering and the date of the previous collection on the transect were used to identify “summer” scats— mid-April to the end of September, which was the active season of the Arctic Ground Squirrel (Spermophilus parryii). Each scat was sterilized, washed and dried. For the “random selection” method of compilation, four sub- samples (usually hair) were randomly selected for identification from each scat. After these four sub- samples were identified every different food item in each scat was identified (similar to traditional metho- dology) in order that we could compare our “random selection” compilation method with the traditional “equal weight” compilation method. Hairs were identified by comparison with a refer- ence collection of photomicrographs using the hair impression technique (Williamson 1951). Mice and voles were grouped. Non-hair items (bones, seeds) were identified by comparison with a reference collection. Data for both the random selection and equal weight compilation methods are expressed as percent frequence of occurrence per item, based on the total THE CANADIAN FIELD-NATURALIST Vol. 97 number of food items. Two-way contingency tables were used to test the null hypothesis that there was no difference between a given pair of diets at the 5% level of significance. Statistical tests were performed only on data pairs consisting of 10 or more scat samples. Contingency tables included only pairs of food items for which one or both items in the pair had a greater than 5% frequency of occurrence. Results There were highly significant differences (p < 0.001) between the diet compiled by the random selection method and the diet compiled by the “equal weight” method (Table 1). The percent composition of avian, plant and miscellaneous items in the diet was greater with the “equal weight” method while the per- cent of 12 of 14 mammalian food items was less than the respective percentages compiled by the “random selection” method. For reasons to be explained in the Discussion, the remaining results were compiled by the random selection method. The main prey in den site scats were Arctic Ground Squirrels, voles, and mice. Collectively, they accounted for 84 and 79% of the identified items at subalpine and alpine dens respectively (Table 1). Scats collected at dens were assumed to be from whelps, because adults rarely defecate at dens (Scott 1943; Jones 1980). All other mammalian prey items, com- bined, contributed less than 5% of the food items at den sites. Fox fur in scats was not considered a food item residue. Vegetation, found in all three alpine den site scats, contributed 6.7%, whether deliberately or accidentally ingested, and avian prey contributed less than 3%. Summer diets based on the subalpine den site scats, collectively, were significantly different from alpine den site scats (p< 0.001) (Table 1). Scats collected at subalpine dens contained voles and mice less often, and ground squirrel remains more frequently, than did scats collected at alpine dens. Significant differen- ces in diets were also found among every pair of individual den sites between and within habitat types, due primarily to the relative contributions of Arctic Ground Squirrels and small mammals (Jones 1980). Significant differences (p< 0.001) in summer diet were found between den and summer, non-den loca- tions above the tree-line on the intensive study area on the Haines Road (Table 1). Scats collected away from dens were assumed to be adult droppings. The diet at summer, non-den locations was dominated by small mammals (30%), vegetation (23%), ground squirrels (22%), and Beaver (Castor canadensis) (18%). In con- trast, ground squirrels and small mammals made up 57% and 21%, respectively, of the diet at the dens. 1983 JONES AND THEBERGE: RED FOX IN BRITISH COLUMBIA AND YUKON S13 TABLE |. Differences in Red Fox diets between the random selection compilation method and the equal weight compilation method among various habitats, and den versus non-den sites in northwestern British Columbia and southwest Yukon.* Non-den location Total Total Den Site Above tree-line Below tree-line Random Equal Select Weight Subalpine Alpine Total® Summer Total Summer Total Method Method Food item (176)° (1056) (1028) (60) (244) (140) (612) (856) (542) Mammals Voles and Mice 19.34 31.4 20.5 30.0 39.8 20.0 19.9 25.6 15.1 Arctic Ground Squirrel 64.2 47.4 57.1 21.7 35.3 32.1 NTS) 22.6 13.8 Snowshoe Hare _- 0.8 0.5 — 2.9 35.9 39.5 29.1 3.3} Muskrat 2.3 0.5 0.9 -- 3.3 2.9 3.4 3.4 1.7 Red Fox 2.3 6.8 7.0 — — = 1.1 0.9 0.4 Porcupine — 0.4 _ = _ — 1.5 1.1 Io Shrew 1.1 0.8 1.0 6.7 4.1 — — 1.2 0.7 Dall Sheep or Mountain Goat — 0.2 0.2 — — = 2.0 1.4 0.6 Unknown ungulate — 0.6 — — — — 0.3 0.2 0.2 Unknown mammal _ 0.5 0.5 oo — 5.0 2.3 1.6 0.9 Moose — — _— —- 1.2 1.4 47 357 2.6 Beaver -- — — 18.3 4.5 2.9 1.3 Dp 0.9 Red Squirrel a — — — — — 2.6 1.9 0.7 Black Bear -- a _ — — — 0.7 0.5 0.2 Birds Ptarmigan — 0.4 0.3 i - a _ I. Unknown avian 0.4 2.1 2.5) 1.6 — — — ll Unknown Tetraonidae — — _- — — -- 0.5 0.4 0. Unknown Anseriformes — = — —- _ — — - 0.2 Plants General vegetation — 6.7 6.8 23.3 7.4 -- 2.6 4.0 Dil) Empetrum nigrum seeds 1.1 0.5 0.7 — — — — — 4.1 Vaccinium spp. seeds 1.1 — 0.2 a — _ — — — Unknown seeds 1.1 0.1 0.3 — — os — — 4.2 Arctostaphylos uva-ursi seeds — — _ _ — — — 3.5) Eleagnus commutata — — — — — — _ — 0.6 Miscellaneous 3.4 0.9 1.6 — — — a — 2.6 “Differences in diets between the “random selection” compilation method and the “equal weight” compilation method, in summer, and between non-den locations above and below tree line were each tested by 2-way contingency tables and are significant p < 0.001). ’Total number of food items. “The total excludes one alpine den location outside the intensive study area on the Haines Road, to facilitate comparison with scats collected at non-den locations above tree line (data column 4) which represented the same general area. ‘Percent frequency of occurrence based on total number of food items. Throughout the year, the diet of adult foxes at non-den locations below the tree-line in the boreal forest was significantly different (p< 0.001) from the diet of foxes residing above the tree-line (Table 1). Snowshoe Hare (Lepus americanus) was the most common, year-round item in the scats collected in the boreal forest, comprising 36-40% of the food items. In contrast, hares accounted for only 3% of the total diet above the tree-line. The above tree-line scats con- tained twice as many voles, mice, and ground squirrels as did the scats collected from the boreal forest habitats. 314 Discussion The striking contrast between data compiled by the “random selection” and “equal weight” methods is worthy of note. The traditional “equal weight” method overemphasizes the importance of fringe die- tary items suchas general vegetation. We question the ecological validity of this method where hypotheti- cally two food items representing 0.1% and 99.9% of identifiable items ina scat are treated equally. Data in Table 1, last column, allow comparison with other studies, but for the reason advanced, the following discussion is based upon the random selection method. Dietary data show the habitat-specific, opportunis- tic nature of Red Fox predation in this heterogeneous, mountain environment. While fox whelps during the denning season on the subalpine and alpine tundra are largely dependent on Arctic Ground Squirrels and small mammals, significant differences in vulpine diet among all den sites were caused by variation in the relative contribution of these prey. The percentage of ground squirrels exceeded that of small mammals in the diets at three of the four dens that were located less than 200 m from the Haines Road. Here, ground squirrels appeared to be very abundant, possibly due to soil disturbance associated with the construction of both the Haines Road and the pipeline (now abandoned). Significant differences in summer diet were found between whelp and adult scats collected above the tree-line in the Haines Road Section of the study area. Adults used more small mammals and fewer ground squirrels than did the whelps. These differences in diets suggest an efficient, parental strategy to bring fewer, relatively large items back to the den to feed their young, as suggested by Murie (1944) in Mount McKinley National Park, Alaska. In contrast to the diet of the tundra-dwelling fox, the vulpine diet of the boreal forest habitats was, like the prey-base, more diversified, and included Snow- shoe Hares, as a major prey item, as well as five other minor food items not found in scats collected above the tree-line. This diet approximates the diverse vul- pine diets reported for rural agricultural habitats south of 55° N (Errington 1935; Scott 1943; Scott and Klimstra 1955; Stanley 1963; Pils and Martin 1978). The dietary plasticity described in this paper is evi- dence of the opportunistic character of Red Fox pre- dation, which has enabled the species to live in a heterogeneous, northern environment. Acknowledgments We are grateful to A. M. Jones for her capable assistance in both the field and laboratory. We thank THE CANADIAN FIELD-NATURALIST Vol. 97 the following people for their assistance: G. W. Ben- nett, D. Campbell, R. Fuch, M. Hoefs, R. D. James, R. L. Peterson, and S. M. Smith. This paper is part of the M.Sc. Research of D. M. J., financed in part by grantstoJ. B. T. fromthe Natural Sciences and Engi- neering Research Council of Canada and the Univer- sity of Waterloo, anda NSERCC post graduate schol- arship to D. M. J. Literature Cited Banfield, A. W. F. 1974. The mammals of Canada. Univer- sity of Toronto Press, Toronto, Ontario. 438 pp. Errington, P.L. 1935. Food habits of mid-west foxes. Journal of Mammalogy 16: 192-200. Green, J. S., and J. T. Flinders. 1981. Diets of sympatric red foxes and coyotes in southeastern Idaho. Great Basin Naturalist 41(2): 251-254. Johnson, M. K.,andR. M. Hansen. 1977a. Comparison of point frame and hand separation of coyote scats. Journal of Wildlife Management 41(2): 319-320. Johnson, M.K., and R.M. Hansen. 1977b. Foods of coyotes in the lower Grand Canyon, Arizona. Arizona Academy of Science 12: 81-83. Jones, D. M. 1980. The home range, habitat use and diet of the red fox (Vulpes vulpes) in the heterogeneous environ- ments of northwest British Columbia and southwest Yukon. M.Sc. thesis, University of Waterloo, Ontario. 140 pp. Murie, A. 1944. The wolves of Mount McKinley. United States Department of Interior National Park Service, Fauna Series 5. 238 pp. Oswald, E. T.,and J. F.Senyk. 1977. Ecoregions of Yukon Territory. Canadian Forest Service, Pacific Forest Research Centre, Victoria, British Columbia. 115 pp. Pils, C. M., and M. A. Martin. 1978. Population dynam- ics, predator-prey relationships and management of the red fox in Wisconsin. Department of Natural Resources Technical Bulletin 105. 56 pp. Rowe, J. S. 1972. Forest regions of Canada. Department of Environment, Canadian Forest Service Publication No. 1300. 172 pp. Scott, T. G. 1943. Some food coactions of the northern plains red fox. Ecological Monograph 13: 427-479. Scott, T. G., and W. D. Klimstra. 1955. Red foxes and a declining prey population. South Illinois University Monograph Series, No. 123 pp. Stanley, W. C. 1963. Habits of the red fox in northeastern Kansas. University of Kansas Museum of Natural History Miscellaneous Publication No. 34. 31 pp. Williamson, V.H.H. 1951. Determination of hairs by impressions. Journal of Mammalogy 32: 80-84. Wyman, P., D.N. Meinzer, and J.T. Flinders. 1975. Foodniche of coyotes in the Rolling Plains of Texas. Journal of Range Management 28(1): 22-27. Received 5 March 1982 Accepted 4 January 1983 Ospreys, Pandion haliaetus, Relocate Nests from Power Poles to Substitute Sites PETER J. AUSTIN-S MITH! and GLEN RHODENIZER2 'Nova Scotia Department of Lands and Forests, Kentville, Nova Scotia B4N 3X3 2Nova Scotia Department of Lands and Forests, Bridgewater, Nova Scotia B4V 1V8 Austin-Smith, Peter J.. and Glen Rhodenizer. 1983. Ospreys, Pandion haliaetus, relocate nests from power poles to substitute sites. Canadian Field-Naturalist 97(3): 315-319. Eight substitute Osprey ( Pandion haliaetus) nest platforms were raised in autumn 1979 next to eight nest sites on power poles in Lunenburg County, Nova Scotia. Six platforms were occupied in 1980 and 1981 and five in 1982 by Ospreys, which produced a total of 27 young. Nest success at the platform sites was not significantly different from that at natural sites. Osprey use of nest sites possibly is related to visibility of neighbouring nests as well as of foraging waters. Key Words: Osprey, Pandion haliaetus, nest relocation, Nova Scotia. Electric power poles and towers are used frequently as nest supports by Ospreys (Pandion haliaetus) throughout much of their North American range. Reports of such nest sites have come from several states (Dunstan 1968; Melquist and Johnson 1975; Reese 1977; Henny et al. 1978) as well as from every Canadian province except Quebec and Saskatchewan (Stocek 1981). In Nova Scotia, as many as 40 pairs of Osprey have been reported nesting on power line support structures (Stocek 1981). The birds nest most commonly on the double wooden crossarms from which the power lines are suspended, and less so on crossbraces between double T poles and onthe top most horizontal bracing of steel power line towers. Power line maintenance personnel are concerned about the potential hazards caused by the presence of these nests. Recently, Olendorff et al. (1981), in a comprehen- sive review of the problems caused by Ospreys and other raptors on power lines, listed various measures for alleviating such problems. I hese measures include placing alternate nesting platforms for Ospreys on poles along the right-of-way or relocating nests to nearby trees. Also, triangles of wood or other mate- rials are placed on the top of power line poles to prevent the birds from rebuilding on them. Another apparently successful approach to the problem is to install nesting platforms at safe locations on existing or new power lines. This paper presents the results of a project to encourage Ospreys to relocate from power line pole nests to platforms situated on a power line right-of- way. The objectives were to: (1) monitor the use of alternate nest platforms by Ospreys nesting on power line poles; (2) compare platform nest success with that at natural nest sites and at previous power line nests. Ss) Study Area A group of eight pairs of Ospreys nesting on the Indian Path power line in Lunenburg County was chosen for the nest relocation project (Figure |). This 69 kV line extends from Lunenburg to the Riverport substation over gently rolling terrain. The forests in the area were burned and cut, with the last extensive harvest operation completed in 1972. They consist of immature even-aged spruce and fir ( Picea-A bies) with numerous short snags but few older and taller trees except scattered White Pines (Pinus strobus). The shallow waters of Upper and Lower South Coves and the Lower Le Have River, including Parks Brook inlet, provide foraging areas for Ospreys. The power line was constructed in late 1970, and according to local residents the first Osprey nest appeared the following spring at site 4 (Figure 1). Table | shows nest site development and annual occu- pancy on the power poles from 1971 through 1979. Materials and Methods Eight nest platforms were constructed by cutting off the flat ends of discarded wooden cable spools. The platforms, each 100 cm in diameter, were placed on 12.2 m poles so that the pole top protruded through the platform centre. Each platform was fastened securely to a pole using two pieces of lumber bolted to the pole and to the underside of the platform. Several nails were driven part way into each platform top to provide additional nest anchoring points. All wood was treated chemically to retard rot. Nest platforms were raised at the edge of the right- of-way close to power line nest sites in mid-November 1979, and the old nests were removed from the power poles. Materials from destroyed nests were arranged on the platforms to simulate crude nests (Postupalsky and Stackpole 1974). 316 THE CANADIAN FIELD-NATURALIST TABLE |. Nest site development and annual occupancy on the Indian Path power line previous to relocation project.! Nest site Year l 2 3 4 5 6 7 8 1971 X 1972 xX X 1973 xX xX 1974 xX x xX 1975 x xX x 1976 x xX x xX 1977 xX XxX xX xX xX 1978 x x x > x x x 1979 xX x x x x x x x '1971-72 data based on reports from local residents. Periodic ground observations to monitor Osprey activity were conducted at the power line and plat- form sites beginning | April and ending in late Sep- tember each year. Helicopter surveys were made on 6 May and 2 July 1980, on 13 July 1981 and 17 May 1982. In 1978, aerial surveys of the power line nests had been conducted on 12 May and 19 July. Eggs and nestlings were recorded in power line and platform nests and also in tree nests on the mainland and islands in Mahone Bay, Lunenburg County. Results and Discussion Platform nest use In 1980, Ospreys were observed first in the Indian Path area on4 April. By 2 May six platform nests were occupied, one power line nest at site 4 was active, and Vol. 97 site 6 was unoccupied. An aerial survey confirmed these observations and indicated the birds were incu- bating eggs. A final aerial survey recorded ten young Ospreys in the Indian Path platform nests. In late August, one pair of Ospreys built a rudimentary nest on a power pole east of site 3, hereafter recorded as 3A. By 20 September, all Ospreys were gone from the Indian Path area. In 1981, the first Ospreys were observed in the Indian Path area on 4 April. On 14 May, nest plat- forms at sites |, 2, 3, 5 and 7 were occupied and there were two occupied power pole nests, 3A and 4. At site 8, where five Ospreys had been recorded on 13 April, there were no birds after 26 May. An aerial survey later recorded ten young birds in platform nests, two young in the power line nest at site 4 and two eggs at site 3A. All Ospreys had left the Indian Path area by 22 September. In 1982 Ospreys again appeared in early April and on 17 May there were five occupied nest platforms, two occupied power pole sites, and nesting material on power poles between sites 3A and 4. Ground observations determined that 12 young birds were fledged from the nests (Table 2). A pair of Ospreys constructed a partial nest on a power pole between sites 6 and 7. Ospreys were last observed in the power line area on 26 September. Reproductive Success The mean clutch size of 38 natural (tree) nests for 1978 and 1982 was 2.55 + 0.72, and of 70 eggs counted in 28 nests surveyed later for young, 74% hatched. Clutch size distribution was 2 nests with | egg, 16 with TABLE 2. Reproductive success of Ospreys at natural nest sites and on Indian Path power poles and platforms. ! Occupied Year nests 1978 Natural sites 34 Power poles 6 1980 Natural sites 19 Power poles | Platforms 6 1981 Natural sites 20 Power poles 2 Platforms 6 1982 Natural sites 4 Power poles 2 Platforms 5) Total 1978, 1980-82 Natural sites 77 Power poles 1] Platforms 17 Productive No. Young/ occupied nests young nest 25 51 1.50 5 9 1.50 13 26 1.37 0 0 0 5 10 1.67 14 33 1.65 | 2 1.00 5 10 1.67 4 10 2.50 2 5 2.50 4 7 1.40 56 120 1.56 (1.12?) 8 16 1.45 (1.13) 14 27 1.59 (0.87) 'Nest terminology is as follows: Occupied — nest attended by one or more birds Productive — nest with well-developed nestlings ?Standard deviation 1983 AUSTIN-SMITH AND RHODENIZER: OSPREYS RELOCATE NESTS ST \ First WA South’, S\ D LUNENBURG BAY ‘/CORKUM | 2 ZISLAND | \\ Qah0D > i Ij Indian ee LI -¢ 6 \ \ — ’ ff SS N Power Line +---* y Riverport SS Vd Nest Platform @ Ss Ficure |. Locations of Osprey nest sites on the Indian Path power line, Lunenburg County. 318 two eggs, 17 with three eggs and 3 with four eggs. Complete egg counts were unavailable in 1980 and 1981. Breeding success rates of Ospreys at natural sites and on man-made structures in Lunenburg County are compared in Table 2. Reproductive success is measured as young per occupied nest (Postupalsky 1974). The four-year average productivity at natural sites and on the platforms was 1.56 and 1.59 young respectively. Power pole nests produced 1.45 young over the same period. There were no significant differ- ences in reproductive success between the natural, power pole or platform nest sites (X*=0.178, p > 0.05). These recruitment rates are above the 0.95 to 1.30 range required to maintain a stable Osprey population (Henny and Wight 1969). Prevost et al. (1978) calculated an average productivity for a two- year period of |.22 and 1.08 young on power poles and at natural nest sites, respectively, in Antigonish County, Nova Scotia. Nest site characteristics Ospreys occupied nests on six of the eight substitute platforms in 1980 and 1981 and on five in 1982. This degree of acceptance was not expected because four occupied nest platforms were at lower elevations than the previous power pole nests. All of those four nests produced young birds. Table 3 presents the character- istics of platform and power pole nests in relation to occupancy by Ospreys. As the power line poles were not modified to pre- vent the birds from rebuilding on them (in 1980 only one pair renested ona pole, at site 4), the occupation of the lower nest sites suggested that Ospreys were attracted to the nesting material. Nest height, how- ever, often is an important factor in nest site location. Although Ospreys nest on elevated structures which THE CANADIAN FIELD-NATURALIST Vol. 97 normally provide good visibility and protection from terrestrial predators, it is height relative to the sur- rounding terrain which seems most important. In some areas, Ospreys will nest on low man-made struc- tures (hunting blinds, channel markers) over water (Reese 1977). Newton (1979), discussing raptor nest dispersion patterns, noted that where foraging grounds are widespread, Ospreys nest in loose colonies. This nest- ing pattern enhances detection of unpredictable and patchy foods, giving rise to communal feeding areas. Earlier Prevost et al. (1978) referred to social foraging by Ospreys and suggested that nesting in colonies may augment this behaviour. According to Bayer (1982), colonial nesting birds learn of food sources by cueing to flight behaviours of colony members (social facilita- tion). Our observations of groups of six and seven Ospreys fishing over sections of Upper South Cove and two freshwater lakes may support the communal feed- ing concept. Hovering and other flight behaviours of Ospreys when they detect prey takes place well above the water surface. On occasion, such activities might be visible to birds at nest sites. For birds in the Indian Path colony, views of flight paths to and from nests and views over the foraging grounds may be as impor- tant as direct views of the water surface. It is tempting to suggest that the crude nests, combined with unim- peded views of adjacent nests and views in the direction of the primary foraging waters, stimulated Osprey reproductive behaviour at the four lower sites. The platform at site 4 remained unoccupied both years by Ospreys, which renested on the adjacent but higher power line pole. This high nest site was the first to be established on the line and has been occupied by Ospreys each year. It apparently provides optimum nesting conditions. The situation at site 6 is unclear TABLE 3. Characteristics and status of Osprey platform (P) and power line (L) nest sites, Lunenburg County, Nova Scotia. Obstructions to horizontal! Visible! Cre age Nearest ; d Nest Power pole Nest platform water Svusilbuliigy wudnt foraging LOganjorss) wes, KeceiGin site height (m) height (m) (degrees) 15.2 m 30.5 m waters (m) 1980 1981 1982 l 10.4 10.4 158 0 0 98 P P P 2 15.8 10.7 D2 0 0 152 P P P 3 15.8 11.0 55 0 0 400 P P P 3a? S72 no platform 2 0 0 550 — L L 4 16.8 10.7 48 0 0 396 L L L 5) 15.8 11.0 16 0 0 550 P P P 6 15.8 10.7 16 Xx x 550 _ — = 7 11.0 11.0 159 0 0 400 P P 8 12.8 11.0 92 0 xX 350 P P — 'As viewed from platform nest rim 2Nesting material on power line pole in late August 1980 1983 because Ospreys nested on the power line pole in 1979 yet the site remained unoccupied in succeeding years of this study. Site features are similar to those at occupied site 5 (Table 3) except for a stand of tall conifers within 8 m of the platform which blocks a portion of the horizon in the direction of Upper South Cove when viewed from the platform. The trees also top the power line pole by almost 3 m but are about 16 m from the poles. The platform at site 8, situated 30 m from deciduous trees which thinly obscure a view of the horizon to the Upper South Cove side of the power line corridor, was occupied in 1980 after gravel removal operations ceased nearby. Although five birds were noted there early in the 1981 breeding season, none were present after mid-June that year nor throughout the 1982 breeding season. In summary, substitute nest platforms were effec- tive in reducing the number of Ospreys occupying power pole nests while maintaining the productivity and size of the local breeding population of these birds. Continued use of the Indian Path platforms may depend upon the presence of platform nest mate- rials each spring, adequate visibility over the forest canopy to the foraging waters, and no substantial increase in human disturbance. Acknowledgments We thank the following for their assistance; Gerald Dickie and personnel of the South Shore Subdivision, Nova Scotia Department of Lands and Forests; Clyde Brown, John Gow and Keith Wentzell, Nova Scotia Power Corporation; Sheldon Mossman, River-port Sea Products, and Donald Richardson, National Sea Products. AUSTIN-SMITH AND RHODENIZER: OSPREYS RELOCATE NESTS 319 Literature Cited Bayer, R. D. 1982. How important are bird colonies as information centers? Auk 99: 31-40. Dunstan, T. C. 1968. Breeding success of Osprey in Minne- sota from 1963 to 1968. Loon 40: 109-112. Henny, C.J., and W.M. Wight. 1969. An endangered Osprey population: estimates of mortality and produc- tion. Auk 86(2): 188-198. Henny, C.J., D. J. Dunaway, R. D. Mallette, and J. R. Koplin. 1978. Osprey distribution, abundance, and sta- tus in western North America: |. The northern California population. Northwest Science 52(3): 261-271. Melquist, W. E., and D. R. Johnson. 1975. Osprey popula- tion in northern Idaho and northeastern Washington — 1972. Raptor Research Report Number 3: 121-123. Newton, I. 1979. Population ecology of raptors. Buteo Books, Vermillion, South Dakota. Olendorff, R. R., R.N. Lehman, and A.D. Miller. 1981. Suggested practices for raptor protection on power lines: the state of the art in 1981. Raptor Research Report Number 4. I 11 pp. Postupalsky, S. 1974. Raptor reproductive success: some problems with methods, criteria and terminology. Raptor Research Report Number 2: 21-31. Postupalsky,S.,andS. M.Stackpole. 1974. Artificial nest- ing platform for Ospreys in Michigan. Raptor Research Report Number 2: 105-112. Prevost, Y. A., R. P. Bancroft, and N. R. Seymour. 1978. Status of the Osprey in Antigonish County, Nova Scotia. Canadian Field-Naturalist 92: 294-297. Reese, J. G. 1977. Reproductive success of Ospreys 1n cen- tral Chesapeake Bay. Auk 94: 202-221. Stocek, R. 1981. Bird related problems on electric power systems in Canada. Research Report Canadian Electrical Association, Montreal, Quebec. 119 pp. Received 24 June 1982 Accepted 3 December 1982 Notes Distribution and Ectoparasites of Little Brown Bats, Myotis lucifugus, on Prince Edward Island GWILYM S. JONES! and HOWARD H. THOMAS!” 'Department of Biology, Northeastern University, Boston, Massachusetts 02115 2Present Address: Department of Biology, Fitchburg State College, Fitchburg, Massachusetts 01420 Jones, Gwilym S., and Howard H. Thomas. 1983. Distribution and ectoparasites of Little Brown Bats, Myotis lucifugus, on Prince Edward Island. Canadian Field-Naturalist 97(3): 320-321. Little Brown Bats (Myotis lucifugus) have been collected at four localities on Prince Edward Island. The following ectoparasites were collected from Cherry Valley: Pygmephorus mahunkai, Spinturnix americanus, Macronyssus crosbyi, and Myodopsylla insignis. The specimen of Pygmephorus mahunkai constitutes the first record of a pygmephorid mite froma bat; the other species are provincial records. Key Words: Little Brown Bat, Myotis /ucifugus, Prince Edward Island, ectoparasites, mites, flea, new records. Although seven species of bats occur in New Brunswick and five in Nova Scotia, only the Little Brown Bat (Myotis lucifugus) has been reported from Prince Edward Island. Three were collected at Mt. Hebert, Queen’s County and were deposited in the U. S. National Museum of Natural History (Cameron 1952). Five, previously unreported, fluid preserved specimens were collected on 15 June 1938 at Ellerslie, Prince County and were deposited in the Royal Onta- rio Museum. Another unreported series of 22 Little Brown Bats was collected on 19 June 1958 in the attic of Green Gables House, Cavendish, Queens County; they are on deposit inthe Acadia University Museum. Finally, eleven specimens were collected in an attic in Cherry Valley, Queen’s County on 12 August 1980; they are preserved as skins, skulls, and skeletons in the Northeastern University Vertebrate Collection. Ectoparasites were collected from the Cherry Val- ley bats by searching through the fur under a dissect- ing microscope and then washing each specimen with detergent, filtering the solution through a Buchner funnel and recovering the ectoparasites from the filter paper. Two of the seven males and all four females harbored at least one ectoparasite. Four species of ectoparasites were found — the mites Spinturnix americanus (n = 1), Macronyssus crosbyi (n = 3, ex. 2 bats), Pygmephorus mahunkai (n = 1), and the flea Myodopsylla insignis (n = 12, ex. 6 bats). Representative specimens are preserved in the Northeastern University collection. The mites were found only on female bats. Both Spinturnix america- nus and Macronyssus crosbyi have been previously reported from the Little Brown Bat and both infest a wide variety of bat species (Whitaker and Wilson 1974). Macronyssus crosbyi has been reported from Myotis sp. in Nova Scotia (Wright 1979), but this is the first record of Spinturnix americanus in south- eastern Canada. The specimen of Pygmephorus mahunkai is the first record of a pygmephorid mite from any bat species (Smiley and Whitaker 1979); P. mahunkai has previously been recorded from Scalo- pus aquaticus and Rattus norvegicus from Indiana (Smiley and Whitaker 1979) and Zapus hudsonius and Napaeozapus insignis from Prince Edward Island (Jones and Thomas 1982). Little is known about host distribution of phoretimorphic pygmephorids. All parasitized bats harbored at least one flea (range |-4). Myodopsylla insignis is a frequent ectoparasite of Myotis lucifugus (e.g. Whitaker 1973). Spinturnix americanus, Macronyssus crosbyi, and Myodopsylla insignis are new provincial records. Acknowledgments We thank Nelson G. Hurry, and Alfred P. Godfrey who collected the specimens at Cherry Valley, Judith L. Eger, Royal Ontario Museum, for loaning the spec- imens from Ellerslie, Robert L. Smiley, U. S. Dept. of Agriculture for the specific identification of Pygme- phorus, and John O. Whitaker, Jr., Indiana State University, for confirming the other parasite identifi- cations, and Thomas H. Kunz, Boston University and Diana B. Jones for help with various aspects of the preparation of this report. This project was funded in part by Grant RR 07143, U. S. Department of Health and Human Services. 320 1983 Literature Cited Cameron, A. W. 1952. Notes on a small collection of mammals from Prince Edward Island. Bulletin of the National Museum of Canada 126: 185-187. Jones, G. S., and H. H. Thomas. 1982. Mites, ticks, and fleas of the mice Zapus hudsonius and Napaeozapus insig- nis from the Maritime provinces and Gaspé Penninsula, Quebec, Canada. The Canadian Entomologist 114: 1031-1035. Smiley, R. L., and J. O. Whitaker, Jr. 1979. Mites of the genus Pygmephorus (Acari, Pygmephoridae) on small mammals in North America. Acta Zoologica Acad. Sci. Hungarica 25: 383-408. NOTES 374 Whitaker, J. O., Jr. 1973. External parasites of bats of Indiana. Journal of Parasitology 59: 1148-1150. Whitaker, J. O., Jr.. and N. Wilson. 1974. Host and distri- bution lists of mites (Acari), parasitic and phoretic, in the hair of wild mammals of North America, north of Mexico. American Midland Naturalist 91: 1-67. Wright, B. 1979. Mites, ticks, fleas, and lice in the Nova Scotia Museum and Acadia University Museum Collec- tion. Proceedings of the Nova Scotian Institute of Science 29: 185-196. Received 20 August 1981 Accepted 15 June. 1983 Heterospecific Vocal Mimicry by Six Oscines DOUGLAS B. McNAIR! and RICHARD A. FORSTER? 'Department of Zoology, Clemson University, Clemson, South Carolina 29631 2Natural History Department, Massachusetts Audubon Society, Lincoln, Massachusetts 01773 MeNair, Douglas B., and Richard A. Forster. 1983. Heterospecific vocal mimicry by six oscines. Canadian Field-Naturalist 97(3): 321-322. Heterospecific vocal mimicry in nature, especially by vireos, is described. Key Words: Blue Jay, Cyanocitta cristata, vireos, Vireo spp., Rose-breasted Grosbeak, Pheucticus ludovicianus, vocal mimicry. Heterospecific vocal mimicry in nature may occur in many avian species (Armstrong 1973). We present evidence of vocal mimicry in nature for several avian species, especially vireos, for which we believe such mimicry is under-recorded, overlooked, or rare. Terres’ (1980) descriptions of primary song, calls, or other vocal sounds are used. Our combined field ex pe- rience is about 40 years, and we believe our aural knowledge of bird sound is adequate to judge that the examples recorded are true mimicry and not normal variation of songs or single notes. The Blue Jay, Cyanocitta cristata, may readily imitate other sounds, both avian and non-avian (Bent 1946). Blue Jays are renowned for mimicry of the call of the Red-shouldered Hawk, Buteo lineatus. The latter species has been extirpated from many former breeding sites in eastern Massachusetts during the last 15-20 years and mimicry of their “kee-yoo, kee-yoo’ call has often been replaced by mimicry of the Broad- winged Hawk, B. platypterus, call ‘pweeee’; the latter breed at many sites in eastern Massachusetts. Jays still frequently imitate Red-shoulders at Boxford, Massa- chusetts, where the latter still nest. The same Blue Jay may imitate calls of both hawks; on 30 October 1976 in Wellesley, Massachusetts, a lone Blue Jay perched on top of a pine, scanned the sky for 10 sec, then uttered a high pitched imitation of a Red-shoulder, followed 5 sec later by a high-pitched call of a Broad- wing. Blue Jays may also imitate the ‘kree-e-e-e’ call of the Red-tailed Hawk, B. jamaicensis, though they do so less frequently. Imitations of the Cooper’s Hawk, Accipiter cooperii, call ‘cuck-cuck-cuck’ in the south- east United States have been heard about 20 times. Our observations are similar to those of Norris (1957). Mimicry of all four hawks has been heard at all sea- sons, whether or not the hawk imitated may have been present. Most North American vireos seldom mimic other avian sounds, though White-eyed Vireo, Vireo gri- seus, mimicry may be more frequent than recorded. We have many records, fromat least 30 individuals, of vocal mimicry for this species from April through August in southern New England and southern Uni- ted States. Examples are the ‘scrip’ note of Red- cockaded Woodpecker, Picoides borealis, and the ‘chuck’ note of Gray Catbird, Dumetella carolinensis, and songs of Bobwhite, Colinus virginianus, Yellow- billed Cuckoo, Coccyzus americanus, and Carolina Wren, Thryothorus ludovicianus. Usually only parts of a song were mimicked, rarely the complete song. Bye THE CANADIAN FIELD-NATURALIST Most imitated sounds were interspersed with the primary song of the White-eyed Vireo, though some mimicked sounds were given alone. White-eyed Vireo vocal imitations in southwest Virginia were predomi- nantly non-territorial sounds (Adkisson and Conner 1978). A Yellow-throated Vireo, V. flavifrons, singing on territory for at least a week in deciduous canopy at Acoaxet, Massachusetts, had snatches of the primary song of the White-eyed Vireo interspersed in its prim- ary song. A Solitary Vireo, V. solitarius, in spruce woods on 10 August 1977 in Windsor, Massachusetts mimicked the ‘pse-ek’ call of the Yellow-bellied Flycatcher Empidonax flaviventris for 15 min. We have heard Red-eyed Vireos, V. o/ivaceus, mim- icking the ‘que-bec’ and ‘hick-three-beers’ calls of the Least, E. minimus, and Olive-sided, Nuttallornis borealis, flycatchers in New England and Canada. Red-eyed Vireos incorporate these songs into their own primary song and this mimicry is fairly frequent. A red-eyed Vireo was also heard mimicking a Yellow- bellied Flycatcher call in a deciduous and evergreen grove 24 km north of Winnipeg, Manitoba. Finally, an adult Rose-breasted Grosbeak, Pheuc- ticus ludovicianus, heard and seen singing for 15 min in deciduous-coniferous forest on 27 June 1977 at Savoy, Massachusetts, mimicked the songs of several species. Interspersed, at about 10 sec intervals in its primary song, were imitations of Red-eyed Vireo primary song and its ‘tschay’ alarm note. Gray Cat- bird ‘mew’ call and a scold note, two variations of Northern Cardinal, Cardinalis cardinalis, primary song, and Rufous-sided Towhee, Pipilo erythroph- thalmus, ‘teeeeeeee’ song and ‘che-wink’ call. All imitated species were locally common to abundant. We believe mimicry in nature by the Blue Jay, White-eyed Vireo, and possibly Red-eyed Vireo, has been often overlooked. Mimicry in nature for the Vol. 97 other species is probably quite rare, though the proc- livity of mimicry by vireos suggests it too may be overlooked. Mimicry has been reported before for all of these species (Bent 1950, 1968; Benton 1952; Borror 1961; James 1976). The majority of mimicking species mentioned in this note used elements of primary song or other avian vocalizations in their own primary song or advertising calls. Acknowledgments We thank M. Gochfeld and A. L. A. Middleton for helpful comments on the manuscript. Literature Cited Adkisson, C.S., and R.N. Conner. 1978. Interspecific vocal imitation in White-eyed Vireos. Auk 95: 602-606. Armstrong, E. A. 1973. A study of bird song. Dover, New York, New York. 347 pp. Bent, A. C. 1946. Life histories of North American jays, crows, and titmice. United States National Museum Bul- letin 191. Bent, A. C. 1950. Life histories of North American wag- tails, shrikes, vireos, and their allies. United States National Museum Bulletin 197. Bent, A.C. (and collaborators). 1968. Life histories of North American cardinals, grosbeaks, buntings, towhees, finches, sparrows, and allies. United States National Museum Bulletin 237. Benton, A. H. 1952. Song mimicry of Red-eyed Vireo. Kingbird 2: 343. Borror, D. J. 1961. Intraspecific variation in passerine bird song. Wilson Bulletin 73: 57-78. James, R. D. 1976. Unusual songs with comments on song learning among vireos. Canadian Journal of Zoology 54: 1223-1226. Norris, R. A. 1957. The Blue Jay as an imitator of Hawks. Oriole 22: 19-20. Terres, J. K. 1980. The Audubon Society Encyclopedia of North American Birds. Alfred A. Knopf, New York. Received 10 December 1981 Accepted 25 February 1982 1983 NOTES 323 Photoperiod as an Environmental Cue for Hibernation in Juvenile Richardson’s Ground Squirrels, Spermophilus richardsoni TERESA M. DOLMAN Department of Biological Sciences, the University of Lethbridge, Lethbridge, Alberta T1K 3M4 Dolman, Teresa M. 1983. Photoperiod as an environmental cue for hibernation in juvenile Richardson’s Ground Squirrels, Spermophilus richardsoni. Canadian Field-Naturalist 97(3): 323-325. Richardson’s Ground Squirrels raised in the laboratory under a simulated natural photoperiod entered their first period of torpor more readily than did squirrels raised under a constant photoperiod of LD 12:12. There were no significant differences in body weights between the two groups of animals. Although squirrels of the natural photoperiod tended to be fatter than those of the LD 12:12, the differences were not significant. Key Words: Richardson’s Ground Squirrel, Spermophilus richardsoni, hibernation, photoperiod. Species of hibernating mammals vary in their dependence on environmental cues, such as tempera- ture or photoperiod, to prepare for and enter the hibernation state (Mrosovsky 1978). For Richard- son’s Ground Squirrel (Spermophilus richardsoni) the role which photoperiod plays is unclear. Harding (1980) found that reduction of metabolic rate, food consumption and activity, cessation of growth, moult- ing, building of deep nests and docility all occurred prior to photoperiod or temperature manipulations; she thus concluded that photoperiod is relatively unimportant. Scott and Fisher (1970) found that annual rhythms of hibernation persisted under con- stant photoperiod of LD 12:12 at 0°, 6° or 18°C, whereas no hibernation was observed in squirrels maintained under LD 12:12 at 0°C (Pengelley and Fisher 1961) or under LD 12:12 at 15°C or room temperature (Hudson and Deavers 1976). Demeneix and Henderson (1978) reported that 75% of their squirrels entered hibernation when exposed to a decreasing photoperiod, light intensity and tempera- ture regime (final conditions: LD 2:22, 0.2-1 lux, 6° C), and Abbotts and Wang (1980) induced hibernation by holding squirrels in total darkness at 5°C. Although these studies imply that photoperiod per se is not an important cue for entry into hibernation, they all used field-caught squirrels, which may have been entrained to the natural photoperiod before their exposure to artificial photoperiods. This study reports on an experiment designed to determine if photoperiod has an effect on the tendency of laboratory-born, young of the year Richardson’s Ground Squirrels to prepare for their first hibernation, as assessed by fat status (the long period of winter dormancy of many hibernators is fueled only by their “on-board” fat stores) and to enter their first hibernation state or torpor, as assessed by actual drop in body temperatures. Methods In mid April 1981 four pregnant Richardson’s Ground Squirrels were trapped in the vicinity of Red Deer, Alberta and transferred to the animal holding facilities of the Department of Biological Sciences, University of Lethbridge at Lethbridge, Alberta. Each female and her subsequent litter were housed ina large plastic cage (40 X 50 X 20 cm) and provided with nest- ing material, water and food ad /ibitum. Within three days of birth each litter was culled to five pups in order to minimize possible effects produced by differential growth rates of litters of different sizes. Before the pups’ eyes opened, two litters were transferred to a room with constant photoperiod of LD 12:12. The other two litters were exposed to a photoperiod reset about every two weeks to approximate the natural photoperiod of Lethbridge; it began with LD 15:9 in May, lengthened to 16:8 in June and shortened to 13:11 in September. Both rooms were maintained at 20 = 5° C. At six to seven weeks of age (in early to mid June) all animals were transferred to individual hang- ing wire mesh cages (18 X 24 X 18 cm). During the second week of September 1981 all squirrels were placed individually in plastic cages with sawdust, cotton and paper towels for nesting material, but no food or water. The cages were then placed inan environmental chamber set at 4 + 2°C, in darkness, for 24h. On their removal next day, it was noted which squirrels appeared torpid and their rectal temperatures were measured. All animals were returned to their hanging cages, at 20°C. One week later all were sacrificed and weighed (minus stomach contents) to the nearest g. The paired abdominal white fat deposits and the paired axillary brown fat deposits were dissected out and weighed to the nearest mg. The weight of single fat deposits is a reliable indicator of general body fatness (Morton and Tung 1971). Brown 324 fat deposits represent a special class of adipose tissue used to generate heat when a hibernator returns to normal body temperature from a period of torpor (Joel 1965). Results All animals during their 24h exposure to low temperature constructed substantial nests of shredded paper mixed with cotton. The animals which had not gone into torpor were immediately active and vocal when removed from their cages. The others remained immobile, were in a hibernating posture (see figures in Pengelley and Fisher 1961), and had low respiration rates. Only one individual, a male, of the LD 12:12 photoperiod group became torpid (body tempera- ture = 16°C), while six squirrels of the natural photo- period group did so (3 males, with mean + S.D. body temperature of 8.2 + 1.4° Cand 3 females, with mean body temperature of 9.7 £1.0°C). These squirrels shortly began shivering when transferred to 20° Cand aroused from their torpor. Table | indicates that fat deposit weights and body weights were higher in the natural photoperiod animals than those of LD 12:12, but the differences were not significant (p > 0.05). The LD 12:12 male that became torpid was not signif- icantly heavier, nor did he have significantly more white or brown fat than the other males of his group. Likewise the males and females of the natural photo- period group that became torpid were not signifi- cantly heavier, nor did they have significantly more white or brown fat than the other males and females, respectively, of that group. Discussion Despite the relatively small sample sizes in this experiment, it does appear that Richardson’s Ground Squirrels raised under a simulated natural photo- period are more prone to enter hibernation than those raised under non-varying photoperiod. Wang’s (1978) laboratory and field studies of animals with implanted transmitters for monitoring body temperature have TABLE |. Influence of photoperiod on weights of white and brown fat deposits in young squirrels. Variability expressed as + one standard deviation of the means. Body weights at termination of experiment. Mean Wt. Mean Wt. of White of Brown Mean’ Fat Deposit Fat Deposit Photo- Numbers Body Wt. (mg/g body (mg/g body period Sex (g) weight) weight) ILD WIA S males 4 ae 74s BOM) ae AO. 7S as St 5 females 350+52 104.8433.4 70+1.8 Natural 5 males 424+76 106.7+£17.6 7.91.1 5 females 353423 131.6+£19.8 7.52+0.8 THE CANADIAN FIELD-NATURALIST Vol. 97 shown that with maximum rates of body cooling, 19 to 26 h are required for an animal to drop its body temperature from a normal 37°C to 9°C (the body temperature reached by animals in this study under natural photoperiod). At maximum, then, five hours elapsed between the time the animals were first placed in the cold and the onset of torpor, and some of this time was spent in first constructing nests. These ani- mals therefore were predisposed to enter torpor, and required only the stimulus of cold temperature and/ or food withdrawal to make them do so. This does not imply that the LD 12:12 squirrels were incapable of hibernating. In fact, one of them had entered torpor and perhaps more would have if allowed more time. They were simply not “as prepared” or “as inclined” to do so as the natural photoperiod animals were. One obvious way in which many squirrels prepare for hibernation is to accumulate body fat. If fat depo- sition in Golden-mantled Squirrels is prevented, hiber- nation is delayed (Phillips 1979). Phillips (1980) also suggested that photoperiod may indirectly influence the date of onset of the initial hibernation period by controlling fat storage, and he found that laboratory- born Citellus lateralis raised under LD 6:18 acquired more body lipid than those raised under LD 18:6. However, there was no significant difference in fat status of the two photoperiod groups in my experi- ment and thus fat status as such may not account for the proclivity of the natural photoperiod group to enter torpor. One must be cautious in applying the results of this experiment to animals in the wild. Laboratory raised squirrels are generally better fed and consequently heavier than wild squirrels (Harding and Rauch 1981) and a photoperiod effect on fattening might show up only under natural conditions. As well, wild squirrels are not out of their burrows from dawn to dusk and therefore the effective photoperiod may be considera- bly different from the natural photoperiod by virtue of the animals’ own activity patterns. This experiment suggests that young squirrels are capable of using an environmental cue, photoperiod, to time or to trigger entrance into hibernation, but it does not reveal the importance of photoperiod relative to other possible environmental cues, nor does it indicate which aspect of the photoperiod is important (the increasing day- length component, the decreasing daylength compo- nent, both, or change itself, regardless of direction). The answers to these and many other questions must await further laboratory experiments and _ field observations. Acknowledgments I thank Cliff and Mary Soper for allowing me access to their land and I appreciate very much the help of Bruce McMullin in caring for the animals. 1983 Literature Cited Abbotts, B., and L. C. H. Wang. 1980. Seasonal thermo- genic capacity ina hibernator, Spermophilus richardsonii. Journal of Comparative Physiology 140: 235-240. Demeneix, B. A., and N. E. Henderson. 1978. Serum 14 and 13 in active and torpid ground squirrels, Spermophi- lus richardsoni. General and Comparative Endocrinology 35: 77-85. Harding, D. H. 1980. Changes in body weight and activity of juvenile Richardson’s ground squirrels, Spermophilus richardsoni, under controlled laboratory conditions. M.Sc. thesis, University of Manitoba, Manitoba. Harding, D.H., and J.C. Rauch. 1981. Comparison of body weights between captive and free-ranging juvenile Richardson’s ground squirrels (Spermophilus richard- soni). Canadian Journal of Zoology 59: 1916-1920. Hudson, J. W.,and D. R. Deavers. 1976. Thyroid function and basal metabolism in the ground squirrels Ammos- permophilus leucurus and Spermophilus spp. Physiologi- cal Zoology 49: 425-444. Joel, C. D. 1965. The physiological role of brown adipose tissue. Pp. 59-85 in Handbook of physiology. Adipose tissue. Edited by A. E. RenoldandG. F. Cahill, Jr. Amer- ican Physiological Society, Washington, D.C. Section 5. Morton, M. L., and H. L. Tung. 1971. The relationship of total body lipid to fat depot weight and body weight in the Belding ground squirrel. Journal of Mammalogy 52(4): 839-842. NOTES 325 Mrosoysky, N. 1978. Circannual cycles in hibernators. Pp. 21-65 in Strategies in cold: Natural torpidity and thermo- genesis. Edited by L.C. H. Wang and J. W. Hudson. Academic Press, New York. Pengelley, E.T., and K.C. Fisher. 1961. Rhythmical arousal from hibernation in the Golden-mantled Ground Squirrel, Citellus lateralis tescorum. Canadian Journal of Zoology 39: 105-120. Phillips, J. A. 1979. Indirect body fat manipulation and its effect on hibernation cycles in Cite//us lateralis. Canadian Journal of Zoology 57: 976-978. Phillips, J. A. 1980. Effect of photoperiod on fat accumula- tion in juvenile Citellus lateralis born in captivity. Cana- dian Journal of Zoology 58: 2150-2152. Scott, G. W., and K. C. Fisher. 1970. The lengths of hiber- nation cycles in mammalian hibernators living under con- trolled conditions. Proceedings of the Pennsylvania Academy of Science 44: 180-183. Wang, L. C. H. 1978. Energetic and field aspects of mam- malian torpor: the Richardson’s ground squirrel. Pp. 109-145 in Strategies in cold: Natural torpidity and ther- mogenesis. Edited by L. C. H. Wang and J.W. Hudson. Academic Press, New York. Received 28 April 1982 Accepted 15 May 1983 A Survey and Census of the Endangered Furbish Lousewort, Pedicularis furbishiae, in New Brunswick ROBIN T. DAY Department of Biology, University of Ottawa, Ottawa, Ontario KIN 6N5 Day, Robin T. 1983. A survey and census of the endangered Furbish Lousewort, Pedicularis furbishiae, in New Brunswick. Canadian Field-Naturalist 97(3): 325-327. Known Canadian locations of the endangered plant Furbish Lousewort ( Pedicularis furbishiae) were explored and a census indicated there were 546+ Canadian plants contributing to a world total of approximately 5546. This is about 146 more Canadian plants than previously reported. Plants were most common on moist, disturbed, morning-shaded riverbank habitat. Insects associated with this plant were identified and both American and Canadian sites were mapped. Key Words: Pedicularis furbishiae, Furbish Lousewort, census, habitat, endangered species, New Brunswick. Furbish Lousewort, Pedicularis furbishiae S. Wats., occurs predominantly on moist, disturbed banks of the St. John River through its northern drainage of the State of Maine and Province of New Brunswick. This species is a perennial herb that produ- ces 3-6 leaves by the first week in June. Leaves are often tinged with an anthocyanin-red at this time. Racemes are produced on most large plants in early June and these begin to flower around 10 July. Seed is dispersed by wind or water during autumn and winter and some seed may be retained in the capsules late into the following summer in sites undisturbed by river flood. In the early growth stages Furbish Louseworts must make haustorial root contact with a host plant or die (Macior 1980). By carefully unearthing mature plants Macior (1978) found no root contact with sur- rounding plants. The obligatory parasitism must therefore be considered a temporary juvenile characteristic. I have observed that if the microclimate remains favourable and there are no major disturbances for three or more years, flowering stalks usually will be 326 produced. If, however, there is too much shade the plants remain in a vegetative state. Habitat Description and 1981 Census During the summer of 1981 a survey and census of Canadian Furbish Lousewort populations was com- pleted. (Day, R. T. 1981. New Brunswick Ecological Reserves 1981 Field Work and Plant Collecting. Internal Report, Department of Natural Resources, Lands Branch, Ecological Reserves Program. Cen- tennial Building, Fredericton, New Brunswick. 18 pp). The location of populations is given in Figures | and 2. Site | was discovered by Mr. H. Hinds in 1977 near a railway embankment (46°44’N, 67°43’W), at the confluence of the Saint John and Aroostook Rivers. This habitat was atypical for this species. Here the plants were found along 25 meters of a steep bank beside a railway track. Brush cutting by a railway crew was evident from the remaining tree stumps. The rem- oval of shrub and tree growth seemed to benefit the herbaceous layer by reducing deep shade. My 1981 census gave a total of 80 plants (all age classes included). Stirrett (1980) reported “about 33 plants” from a 1978 count. Site 2 is along the east bank of the Saint John River immediately south of the Little River delta (46°52’N, NEW BRUNSWICK FIGURE |. Distribution of Pedicularis furbishiae. Sites | to 3 in Canada (Day 1981), 3 to 10 in the United States (Dyer 1981): | = confluence of Saint John and Aroostook Rivers, 2 = near confluence of Saint John and Little River, 3 = spans the Maine - New Brunswick border near Hamlin and Grand Falls, 4 = Van Buren, 5 = Fort Kent, 6 = St. Francis, 7 = Allagash, 8 to 10 = Townships T14R13, TISR13 and T16R12 on the Saint John River. THE CANADIAN FIELD-NATURALIST Volgo7 Grand Falls 67°40'w /)) ® J 14) — “” xe) ® — er our ~ Falls Fort Fairfield FIGURE 2. Locations of Furbish Lousewort within New Brunswick, Canada: anenlargement of part of Figure 1, showing sites I, 2, and 3. 67°41’W), near North Tilley. Here the Furbish Louseworts occur in a narrow linear zone along the Saint John riverbank. The plants are found only along the upper riverbank between the edge of the stable forested slope and the lower riverbank where seasonal flooding and ice-push cause disturbance to soil and plants. Plants usually grow in river-deposited calcare- ous silt. Stirrett and Tribe found “about 70+ plants of Pedicularis furbishiae scattered along 0.8 miles of riverbank” (Stirrett, G.M. 1977. Report on Investiga- tions of the Flora of Northern Maine and Northern New Brunswick with Particular Reference to Pedicu- laris furbishiae and other Rare Plants. Report on Contract No. DACW 33-77-M-0885. U.S. Depart- ment Army, Corps of Engineers. Waltham, Massa- chusetts. 61 pp.). In 1979 a total of “115 or 69 mature flowering plants and about 46 young plants” were counted (Stirrett 1980). Prolonged searching during my 1981 census led to the discovery of 212 plants along 1.5 km of the riverbank south of the Little River delta. Site 3 is at the Maine-New Brunswick border (47° 04’N, 67°47’W), at the base of a very steep slope. Plants were usually growing within one meter of the water’s edge on nearly vertical carpets of moist Feather Moss, Pleurozium schereberi, at the base of 1983 the stable forested slope. My 1981 census produced a total of 102 plants. In 1977, Stirrett and Tribe made a more extensive survey of this site along 1.2 miles of the riverbank and they found “about 254+ plants”: 154 flowering plants and 100 young non-flowering plants. (Stirrett 1977, 1980). There has been little disturbance to these populations since 1977, and therefore 254+ plants is probably the best estimate of current numbers. All three sites were at the base of fairly steep slopes where seepage water kept the soil near saturation. Many of the plants of Site 3 rooted in moss, however, exhibited moisture stress on 17 July as indicated by their flaccid leaves. The three sites were shaded by the trees and steep banks until approximately 1130-1200 h. In 1980, a survey of U.S. populations was com- pleted by Richard Dyer in which only flowering stems were counted. From approximately 2055 flowering plants counted (R. W. Dyer, personal communica- tion) he makes a rough total population estimate of 5000 American plants (Dyer 1981). This estimate assumes a 1:1 relationship between flowering and non-flowering plants, and an additional 20 percent factor for colonies that may not have been observed. My 1981 census has increased the Canadian total by 146 plants from approximately 402 (Stirrett 1980) to 546+. The Furbish Lousewort 1981 total world popu- lation is therefore (254+) + (80) + (212) = 546+ Canadian plants and approximately 5000 American plants = 5 546+. The results of the 1982 New Brunswick census were: site 1 = 125, site2 = 213,site3 = 117, a total of 455 plants (Don Brown, personal communication, 4 Au- gust 1982. Wildlife, Department of Natural Resour- ces, New Brunswick). There appears to be a consider- able increase in numbers at the railway embankment Site 1, from 80 in 1981 to 125 in 1982. Site Disturbance Because of the water seepage and steepness of the banks at Sites 2 and 3, the soil with its vegetation cover periodically slips downslope onto an unstable part of the riverbank where the plants are destroyed by flood and wave action. On more stable rocky and sunnier parts of the riverbank competing vegetation grows more vigorously over time, thus, the Furbish Louse- worts are suppressed by the heavy shade that devel- ops. Site | is temporarily maintained as a good Fur- bish Lousewort habitat because of brush cutting by railway crews. On this seepage slope, soil slippage was not evident. In summary, the Furbish Lousewort usually inhab- its unstable, morning-shaded, seepage slopes where the competitive effects of associated plants are NOTES 327 reduced. Because of the unstable nature of the sites (river erosion, steep slipping banks) new Furbish Lousewort habitat is being created while old patches are being destroyed. Thus the Furbish Lousewort is a typical “fugitive species” (Grime 1979) occupying temporary habitats which are periodically destroyed. At Site 2a Groundhog (Marmota monax (L.)) had clipped a number of Furbish Lousewort flower stalks from several plants. The following herbivorous insects were present at all sites: two species of Spittlebug (Cercopidae), Aphrophora gelida (Wlk.) and nymphs, probably Neophilaenus lineatus (L.), were often observed to stunt and deform the flowering stems (see Macior 1978) and lepidopterous larvae fed on the leaves. These larvae died in rearing trials because they had been parasitized by Ichneumon Wasps (Macrocentrus sp.). Forest Tent Caterpillars (Malacosoma disstria Hbn.) were frequently collected on Furbish Louseworts during their population boom in the summer of 1981 but were never observed to cause damage to the leaves. Acknowledgments I wish to thank Dr. Stirett, Fred Tribe, and Erwin Landauer for helpful discussion, and Mel Fitton, my helpful supervisor, with the Lands Branch, Ecological Reserves Program, Department of Natural Resour- ces. Richard Dyer supplied the American distribution data. Insect identifications were kindly completed by W.R. M. Mason, S. Allyson, and K. G. A. Hamilton of the Biosystematics Research Institute, Ottawa, Canada. Richard Staniforth of the University of Win- nipeg kindly reviewed a draft of this note and Michele Quaglia helped with drafting. This research was funded by the New Brunswick government. Literature Cited Dyer, R. W. 1981. Furbish Lousewort Recovery Plan. Technical Draft. Region 5 of the U.S. Fish and Wildlife Service, U.S. Department of the Interior. Newton Corner, Massachusetts 02158. 13 pp. Grime, J. P. 1979. Plant Strategies and Vegetation Pro- cesses. John Wiley and Sons, Toronto. 222 pp. Macior, L. W. 1978. The Pollination Ecology and Endemic Adaptation of Pedicularis furbishiae S. Wats. Bulletin of the Torrey Botanical Club 105 (4): 268-277. Macior, L. W. 1980. Population Ecology of the Furbish Lousewort, Pedicularis furbishiae S. Wats. Rhodora 82 (829): 105-111. Stirrett, G. M. 1980. The Status of Furbish’s Lousewort, Pedicularis furbishiae S. Wats. in Canada and the United States. The Committee on the Status of Endangered Wild- life in Canada, Canadian Wildlife Service, Environment Canada, Ottawa, Ontario. 75 pp. Received 17 March 1982 Accepted 20 August 1982 328 THE CANADIAN FIELD-NATURALIST Vol. 97 Nodding Thistle, Carduus nutans: an Addition to the Vascular Flora of Alberta BERYL HALLWORTH! and MICHAEL MYCHAJLUK? ‘Herbarium, Department of Biology, University of Calgary, Calgary, Alberta) T2N IN4 22616 11th Avenue S.E., Calgary, Alberta 1T2A 0E3 Hallworth, Beryl, and Michael Mychajluk. 1983. Nodding Thistle, Carduus nutans: an addition to the vascular flora of Alberta. Canadian Field-Naturalist 97(3): 328. The first record of the Nodding Thistle, Carduus nutans, in the province of Alberta is reported from the Calgary area. It is readily distinguished from true Thistles, Cirsium spp., and its dispersal and habitat characteristics may facilitate its spread in Alberta. Key Words: Nodding Thistle, Carduus nutans, Alberta, habitat, characteristics. The Nodding Thistle, Carduus nutans L. is a Eura- sian immigrant to Canada that is a serious weed of waste places, pastures and cultivated fields. Prior to this report it was known “in every province except Prince Edward Island and Alberta” (Mulligan 1976); it is here reported from Alberta. In February 1976 it was collected by Beryl Hall- worth and Desmond Allen, north of Calgary, near Beddington (now included in the city); in July of the same year a large stand was located by Cathy Osborne northwest of Calgary; Michael Mychajluk later located it in two places in the southeastern part of the city, and in 1981 it was found by Dr. L. V. Hills to be growing one-quarter mile west of Calgary. Specimens of these plants have been deposited at ALTA, CAN, and UAC. The specimen in CAN has broad heads (ca. 5 cm broad heads), distinctly reflexed outer involucral bracts, and glabrous leaves, which place it in subspe- cies leiophyllus (Petrovic) Stoj. & Stef., as treated by Moore and Frankton (1974). The Nodding Thistle, also called the Musk Thistle, flowers from July to September. It is a striking plant, standing 3-25 dm tall, and bearing many large reddish-purple, thistle-like heads (2.5-5 cm broad). The heads are borne on long, erect peduncles that nod at the ends. Only tubular disk-florets are present, surrounded by large, broad involucral bracts, each terminated by a spine. These bracts occur in several series and the outer bracts are reflexed. The spiny thistle-like leaves are up to 40 cm long, and clasp spiny stems. The fruit is a cypsela, (an inferior achene), with a white “parachute” composed of simple pappus hairs. The pappus is an effective aid to wind dispersal. Carduus nutans may be distinguished from the true thistles, Cirsium spp., by the reflexed involucral bracts, long peduncles, and simple, not feathery, pap- pus hairs. The habitat of this plant is “dry sandy and clay soils in grasslands, roadside, pastures, margins of culti- vated fields, and waste lands” (Moore and Frankton 1974). It “often forms very dense stands on rocky, hilly soil where there is little competition” (Frankton and Mulligan 1971). B. Hallworth and D. Allen found the Nodding T his- tle specimens associated with gravel pits. L. Hills found them ina similar habitat. The plants found by C. Osborne were in a roadside ditch, in disturbed sandy prairie soil, while those found by M. Mychajluk were growing in rocky soil. These habitats conform to that given by Frankton and Mulligan. It is probable that, because of its efficient mode of dispersal and its ability to occupy habitats that may be unsuitable for other species, it will soon be found all across southern Alberta. Literature Cited Frankton, C.,and G. A. Mulligan. 1971. Weeds of Canada. Canada Department of Agriculture. Publication 948: 184-185. Moore, R.J., and C. Frankton. 1974. The thistles of Canada. Canada Department of Agriculture, Monograph (10): 58-61. Mulligan, G. A. 1976. Common weeds of Canada. McLel- land and Stewart in association with Information Canada and the Department of Agriculture. 102 pp. Received 25 May 1982 Accepted 15 December 1982 1983 NOTES 329 Ambivalence of the Concealing Pose of Owls S. BONDRUP-NIELSEN Department of Zoology, University of Alberta, Edmonton, Alberta 16G 2E9 Bondrup-Nielsen, S. 1983. Ambivalence of the concealing pose of owls. Canadian Field-Natrualist 97(3): 329-330. It is argued that the “concealing pose” of owls is ambivalent and that applying this functional name is inappropriate. Rather, Ligon’s (1968: 63) term “erect posture” should be used with modifications. Key Words: concealing, owls, behaviour. Hiding, protective (Bent 1938), erect (Ligon 1968) and concealing (Catling 1972) poses, all suggested to have a concealing function, have been attributed to various species of owls: e.g. Screech Owl (Otus asio), Long-eared Owl (Asio otus) (Bent 1938), Elf Owl (Micrathene whitney) (Ligon 1968), Great Gray Owl (Strix nebulosa) (H6glund and Lansgren 1968), Saw- whet (Aegolius acadicus) and Boreal Owls (Aegolius funereus) (Catling 1972). All poses are similar, but there are some important differences among them that suggest to me that all cannot serve the same function. Bent (1938: 255), quoting Durfee, described the hid- ing pose of two Screech Owls as follows: “The attitude was long drawn out, . . .the wings and feathers held as close to the body as possible . . . (giving) the appear- ance of two long stubs, the top of the head being nearly square across. The eyes were slanted slits, and while the head was directly toward me, the body was swung sideways so as to keep the wing in front as a shield: in other words, they were looking over their shoulder” As Durfee walked around the owls, one suddenly swung halfway around so that it was looking over the other shoulder. Bent (1938: 256) further quoted Shelley’s description for the same species: “As we came into sight, very slowly the bird attained the protective pose of a dead stub.... the eyes half closed.”. On the other hand, Catling (1972) described what he called the concealing pose of Saw-whet and Boreal Owls, both of which stood erect on their legs, with the plumage of the breast and upper back appressed, and thus had a narrow oblong appearance. The wing nearest the intruder was directed toward him and raised to the level of the bill. The frontal crown feathers immediately above each eye were raised while those in the center of the crown remained flattened. The feathers between and somewhat above the eyes were fanned out, which resulted in expansion of the white areas between the eyes. The eyes were fully open. What Catling (1972) failed to mention is that the above pose is adopted abruptly rather than slowly (personal observation). In fact, the abrupt movement made by Saw-whet Owls as they assumed the “concealing pose” often led me to detect otherwise well-concealed individuals in their roost. The sudden flash of fully open eyes with white between them was revealing rather than concealing. The Elf Owl has an erect posture, which has been described as concealing (Ligon 1968), and which is similar to the concealing posture of the Saw-whet Owl. Again the eyes are fully open and there is con- spicuous erection of the white feather patches, one above the eyes and one below the facial discs. Brewster (1883:28), who was quoting Stephens, wrote of the Elf Owl: “I could just see its eyes over the wing, and had it kept them shut I might have overlooked it, as they first attracted my attention.”. Putman (1958) described the concealing pose of a Screech Owl in the presence of House Sparrows (Passer domesticus). The owl adopted the pose abruptly, but otherwise its pose appeared identical to that of the Screech Owl described by Bent (1938). Putman could not see whether the eyes were open or closed. Indeed, if owl eyes are releasers for mobbing by passerines (Curio 1975), it would be surprising if the owl’s eyes were open. In the Long-eared (Bent 1938) and Great Gray Owls (Héglund and Lansgren 1968), the eyes are fully open during the so-called “concealing pose.” Long-eared Owls will even close and open their eyes abruptly several times in succession during the pose (Kay McKeever, personal communication). Common features of all these poses are |) upright posture, 2) the wing nearest the intruder is raised to the bill, 3) appression of the body plumage, and 4) expansion of the lateral crown feathers. However, three features of the pose do not support the hypothe- sis that the function of the pose is to conceal the owl: Open eyes, increased exposure of white feathers around the eyes, and the abrupt manner in which the pose is adopted. For an animal to achieve concealment, it should avoid abrupt movement, because predators are gener- ally extremely alert to movement (Alcock 1975). Although Catling (1972) maintained that the white on the wing and between the eyes of the Saw-whet Owl served as disruptive camouflage, cases of disruptive 330 camouflage usually involve bold lines and patches that alter the outline of an animal (Alcock 1975). In most vertebrates, eyes are conspicuous and are impor- tant signalling devices. Eye spots are used by many animals, especially Lepidoptera and some Amphibi- ans, to divert attack by predators to less vital areas, or to surprise predators (Alcock 1975). Many animals also make their eyes less obvious by having various markings through the eye, e.g. tropical reef fishes. The posture of the Screech Owl, which is adopted slowly and in which the eyes are mere slits, does unquestionably conceal the bird. However, this was not achieved by the other owl species described, espe- cially the Saw-whet and Elf Owls in which the pose was adopted abruptly, with fully open eyes anda flash of white feathers catching the attention of at least human intruders. It appears, therefore, that two different postures are involved in the so-called “concealing pose” of owls, and to give them the same or similar functional names is premature. Until this posture is more thoroughly studied, I believe it is more appropriate to name the poses by descriptive terms; I propose, therefore, that Ligon’s (1968: 63) term “erect posture” be used with the addition of “abrupt” and “gradual” to distinguish between the posture that is adopted abruptly with open eyes and white markings and the posture adopted slowly with eyes reduced to mere slits. THE CANADIAN FIELD-NATURALIST Vol. 97 Acknowledgments I would like to thank L. Maslen for her help during preparation of this work, and W. A. Fuller, V. Lewin, J.O. Murie and F. Zwickel for criticizing earlier drafts on this note. Literature Cited Alcock, J. 1975. Animal behavior. An evolutionary approach. Sunderland Assoc. Inc., Massachusetts. Brewster, W. 1883. Ona collection of birds lately made by Mr. F. Stephens in Arizona. Auk 8: 21-36. Bent, A. C. 1938. Life histories of North American birds of prey. Part 2. Smithsonian Institution, United States National Museum. Bulletin 170. Catling, P.M. 1972. A behavioral attitude of Saw-whet and Boreal Owls. Auk 89: 194-196. Curio, E. 1975. The functional organization of anti- predator behavior in the Pied Flycatcher: a study of avian visual perception. Animal Behavior 23: 1-115. Héglund, N. H., and E. Lansgren. 1968. The Great Gray Owl and its prey in Sweden. Viltrevy 5: 361-416. Ligon, J. D. 1968. The biology of the Elf Owl, Micrathene whitneyi. Miscellaneous Publications. Museum of Zool- ogy, Univ. of Michigan, No. 136. Putman, Wm. L. 1958. Use of concealing posture by a Screech Owl. Auk 75: 477-478. Received 15 December 1982 Accepted 6 March 1983 Interspecific Food Piracy of Rough-legged Hawk, Buteo lagopus, on Red-tailed Hawk, B. jamaicensis, and Scavenging by Rough-legged Hawk GEOFFREY G. HOGAN Department of Biology, University of Prince Edward Island, Charlottetown, Prince Edward Island CIA 4P3 Hogan, Geoffrey G. 1983. Interspecific food piracy of Rough-legged Hawk, Buteo lagopus, on Red-tailed Hawk, B. jamaicensis, and scavenging by Rough-legged Hawk. Canadian Field-Naturalist 97(3): 330-331. A Rough-legged Hawk, Buteo lagopus, attacked a Red-tailed Hawk, B. jamaicensis, feeding upona Common Crow, Corvus brachyrynchos, and usurped the prey. Twenty minutes later a second Rough-legged Hawk scavenged the remains. A heavy snow crust may have caused a reduction in the availability of normal prey items for these raptors, thus prompting other means of obtaining food. Key Words: Red-tailed Hawk, Buteo jamaicensis, Rough-legged Hawk, B. lagopus, food piracy, scavenging, Prince Edward Island. Red-tailed Hawks, Buteo jamaicensis, and Rough- legged Hawks, B. lagopus, rely heavily upon small mammals, especially Meadow Voles, Microtus penn- sylvanicus, as winter prey, yet reduce competition on their sympatric winter range through differences in habitat utilization (Schnell 1968; Baker and Brooks 1981). These workers did not mention direct contron- tations between the two species. Bildstein (1979), however, recorded wintering Red-tailed Hawks and Northern Harriers, Circus cyaneus, as victims of piracy in 20% of encounters (103 interspecific) among four raptor species studied in south-central Ohio but 1983 did not say what species was the aggressor. Red-tailed Hawks are known to be occasional scavengers (Tufts 1973) although this behavior appears to be less com- mon among Rough-legged Hawks (Smith 1975). Severe winter weather may disrupt normal feeding methods and Klein and Mason (1981) reported that several raptors, including Rough-legs, shifted from food territories to patrolling highways for dead or living prey after a storm. At 11:55 on6 February 1982, several colleagues and I] observed an adult Red-tailed Hawk feeding upon a Common Crow, Corvus brachyrhynchos, carcass on the edge of the frozen North River, Prince Edward Island (46°17’N, 63°10’W). After about 5 min a Rough-legged Hawk (dark phase) flew in and attacked the Red-tail. The two birds fought vigor- ously onthe ice for 10-15 sec, with talons outstretched and wings spread, after which the Red-tailed Hawk retreated and the Rough-legged Hawk usurped the prey. Over the next 5 min, the Red-tailed Hawk remained 3-4 m away on the ice nearby while the Rough-legged Hawk mantled the prey but did not feed. At 12:15 the Red-tail flew 15-18 m to a White Spruce, Picea glauca, whence it watched the Rough- leg. The latter then commenced plucking and feeding upon the crow until 12:30, when it flew off suddenly, carrying part of the crow in its talons. At this time a second Rough-legged Hawk (light phase) flew in and began feeding upon the portion remaining. Almost immediately the Red-tailed Hawk flew off, alternately soaring and flapping over adjacent fields. The second Rough-leg fed until 12:55, during which time several crows investigated but did not harass the hawk. One crow on the ice approached within 2-3 m of the hawk and fed briefly on the remains when the hawk flew away. During the observation period viewing condi- tions were good despite a southwest wind that carried blowing snow and, later, freezing rain. The air temperature was around —5°C. Upon investigation we found only the outer right wing and scattered feathers, but no blood on the freshly fallen snow. As we did not see the Red-tailed Hawk capture the crow, we do not know whether the NOTES 331 hawk had killed it or was scavenging. Rough-legged and Red-tailed Hawks are the only buteos found on Prince Edward Island during the winter, and Rough- legs are generally more numerous. Small woodlots interspersed with farmland over much of the island provide suitable winter habitat for both species. The weather for the week previous to 6 February 1982 alternated between mild and freezing tempera- tures, thus creating a hard crust over the otherwise deep snow. Klein and Mason (1981) suggested that behavioral changes in hunting techniques by raptors following severe winter weather are examples of adapt- ability. The unusual behavior reported here also sug- gests that the formation of a hard snow crust promp- ted alternative feeding methods. Acknowledgments I wish to thank Martin K. MecNicholl, lan Mac- Quarrie, Daryl Guignion and two anonymous referees for reviewing the manuscript. Literature Cited Baker, J. A.,and R. J. Brooks. 1981. Distribution patterns of raptors in relation to density of meadow voles. Condor 83: 42-47. Bildstein, K. L. 1979. Behavioral ecology of Red-tailed Hawks (Buteo jamaicensis), Rough-legged Hawks (B. lagopus), Northern Harriers (Circus cyaneus), American Kestrels (Falco sparverius) and other raptorial birds win- tering in south-central Ohio. Raptor Research 13: 29-30 [abstract]. Klein, R. J., and D. R. Mason. 1981. Change in raptor hunting behavior following heavy snowlall. Raptor Research 15: 121. Schnell, G. D. 1968. Differential habitat utilization of win- tering Rough-legged and Red-tailed Hawks. Condor 70: 373-377. Smith, T. G. 1975. Rough-legged Hawks, Buteo lagopus (Pontoppidan) as carrion feeders in the Arctic. Canadian Field-Naturalist 89: 190. Tufts, R. W. 1973. The birds of Nova Scotia. Nova Scotia Museum, Halifax. Second Edition. 532 pp. Received 29 July 1982 Accepted 3 December 1982 B82 THE CANADIAN FIELD-NATURALIST Vol. 97 Ivory Gulls, Pagophila eburnea, Nesting on the Brodeur Peninsula, Baffin Island, N.W.T. AUSTIN REED and PIERRE DUPUIS Canadian Wildlife Service, 1141 Route de l’Eglise, P. O. Box 10100, Ste-Foy, Québec GI V 4H5 Reed, Austin, and Pierre Dupuis. 1983. Ivory Gulls, Pagophila eburnea, nesting on the Brodeur Peninsula, Baffin Island, N.W.T. Canadian Field-Naturalist 97(3): 332. Two small colonies of Ivory Gulls, Pagophila eburnea, were found on the northwest portion of the Brodeur Peninsula, Baffin Island, in August 1981. This is the first confirmed observation of nesting on the Brodeur Peninsula, an area long believed to harbour breeding Ivory Gulls. Key Words: Ivory Gull, Pagophila eburnea, breeding, distribution, Baffin Island. The present knowledge of breeding distribution of the Ivory Gull has been summarized by Blomqvist and Elander (1981). Based on confirmed records, the Can- adian breeding range included Prince Patrick Island and the Polynia Islands (Sites abandoned since their discovery by McClintock in the 1800’s: MacDonald and Macpherson 1962; MacDonald 1976), a floating ice island off Amund Ringnes Island (probably also abandoned: MacDonald 1976), reefs off Meighen Island (MacDonald 1976), Seymour Island (MacDo- nald 1976) and on eastern Ellesmere Island (Frisch and Morgan 1979; Witts and Morrison 1980). The Brodeur Peninsula on Baffin Island had often been identified as a probable breeding location (MacDo- nald and Macpherson !962; Renaud et al. 1979), but confirmation was lacking until now. On8 August 1980, while flying across the interior of the Brodeur Peninsula from Nanisivik to Resolute Bay in a fixed-wing aircraft, we saw two groups of small, whitish gulls rise from the ground and circle ina manner typical of gulls disturbed from a breeding colony. The altitude and speed of the aircraft pre- vented us from confirming our suspicion that they were breeding Ivory Gulls. On 10 and I! August 1981, we crossed the same peninsula by helicopter, during goose surveys. Two groups of gulls were located on I] August in similar habitat to those seen the previous year. On approach- ing with the helicopter we confidently identified the adult birds as Ivory Gulls (30 at one site, 18 at the other) and confirmed their breeding status by observ- ing several flightless young about 4 to 4% grown. We immediately retreated from the colonies without land- ing in order to minimize disturbance. The colonies were on the interior plateau about 15 kminland from Prince Regent Inlet and near 73°N (a detailed description of their location is on file with the Canadian Wildlife Service). Both colonies, like the suspected ones of the previous year, were on gently sloping mounds or ridges. The sites were character- ized by greyish boulders contrasting with the sur- rounding terrain, which was more smoothly textured and of a beige colour. Other Ivory Gulls were observed in the area during the same period in 1981, either as singles or small groups flying over the peninsula (total 61 birds) or as feeding groups on coastal waters (6 birds near Arctic Bay and 52 on Prince Regent Inlet). Our observations, confirming those of past generations of local Inuit, suggest other small colonies scattered over the Bro- deur Peninsula. A complete inventory would require an intensive aerial search. Acknowledgments We benefited from logistic support by the Polar Continental Shelf Project and from discussions with S.D. MacDonald (National Museum of Natural Sciences) and D. N. Nettleship (Canadian Wildlife Service). Literature Cited Blomqvist, S.,and M. Elander. 1981. Sabine’s Gull (Xema sabini), Ross’s Gull (Rhodostethia rosea) and |vory Gull (Pagophila eburnea) in the arctic: Gulls in the arctic — a review. Arctic 34: 122-132. Frisch, T.,and W. C. Morgan. 1979. Ivory Gull colonies in southeastern Ellesmere Island, Arctic Canada. Canadian Field-Naturalist 93: 173-174. MacDonald, S. D. 1976. Phantoms of the polar pack ice. Audubon 78: 2-19. MacDonald, S. D.and A. H. Macpherson. 1962. Breeding places of the Ivory Gull in Arctic Canada. National Museum of Canada Bulletin, 183: 111-117. Renaud, W.E., S.R. Johnson, and P.D. Hollingdale. 1979. Breeding birds of Arctic Bay, Baffin Island, N.W.T., with notes on the biogeographic significance of the avifauna. Arctic 32: 122-134. Witts, B., and R.I.G. Morrison. 1980. Joint services expedition to Princess Marie Bay, Ellesmere Island, 1980: Preliminary report. Wader Study Group Bulletin 30: 34-35. Received 14 June 1982 Accepted 31 October 1982 1983 NOTES 333 Long Distance Movement by Gray Wolves, Canis lupus WARREN B. BALLARD!, RICK FARNELL?, and ROBERT O. STEPHENSON? 'Alaska Department of Fish and Game, P.O. Box 47, Glennallen, Alaska 99588 ?Yukon Game Branch, P.O. Box 2703, Whitehorse, Yukon Territory Yl A 2C6 3Alaska Department of Fish and Game, 1300 College Road, Fairbanks, Alaska 99701 Ballard, Warren B., Rick Farnell, and Robert O. Stephenson. 1983. Long distance movement by Gray Wolves, Canis lupus. Canadian Field-Naturalist 97(3): 333. Between 10 April 1978 and 10 January 1979 at least two and probably four Gray Wolves (Canis /upus) from the same pack moved 732km from the Nelchina Basin of southcentral Alaska to the Brooks Mountain Range of northeastern Alaska. Key Words: Canis lupus, Coleen River, Nelchina Basin, movements, dispersal. The longest movement of a Gray Wolf (Canis lupus) reported in the literature is670 km(Van Camp and Gluckie 1979). Recently, however, the record movement for a single Wolf was extended to 917 km (S. Fritts, personal communication). This report doc- uments a minimum movement of 732 km by at least two and probably four Gray Wolves from the same pack. On 10 February 1978, a 42-kg male Gray Wolf, judged by tooth wear to be a least two years old, was captured (Ballard et al. 1982) and radio-collared for movements studies in the vicinity of St. Anne’s Creek (61°47’N,146°0’W) in southcentral Alaska (Ballard et al. 1981). The marked wolf was observed with the pack on six occasions during 10 February — 10 April 1978. On7 and 10 April 1978, two additional wolves in the pack were radio-collared. After 10 April contact was lost with the male captured on 10 February and with a young adult male captured on 10 April. During subsequent observations of the pack only eight of the original 12 wolves were present. Harvest records did not indicate any of the wolves had been killed by hunters or trappers. The wolf captured on 10 April was known to have dispersed at least as far as Tanada Lake by August 1978, 142 km northeast of its capture location. Con- tact was lost after this observation until 10 January 1979 when biologists heard two radio signals with frequencies and pulse rates identical to those of the two missing radio-collared St. Anne’s wolves in the Brooks Mountain Range of northeastern Alaska (67° 41’N, 143°03’W). On2 March 1979, the radio-collared adult was observed with three Gray Wolves which probably included the radio-collared yearling about 64 km north of the 10 January location. The wolves were last observed there on 28 March 1979. Other researchers in the region were not missing radio- collared wolves with either of the two identified fre- quencies. Therefore, we believe the wolves originated from the Nelchina Basin in southcentral Alaska, a minimum straight-line movement of 732 km. Since wolves commonly travel along waterways and through river valleys, the movement undoubtedly was considerably greater than 732 km. Because both radio-collared wolves disappeared from the St. Anne’s area at the same time the pack’s numbers declined by 4, and since four wolves were observed, all 4 may have been from the Nelchina Basin. The cause of this long movement by at least two and probably four wolves is unknown but may have been related to the low availability of prey inthe St. Anne’s area (Ballard et al. 1981). Other radio-collared wolves in this study area have dispersed either singly or in pairs from 22 to 193 km, often into vacant habitat. The observations reported here are a record dispersal for a group of wolves and are significant because they illustrate the ability of wolves to traverse long distan- ces and repopulate areas with suitable habitat and prey. Acknowledgments We thank K. Schneider and K. Wiley, both of the Alaska Department of Fish and Game for reviewing the manuscript. The study was funded in part by the Alaska Federal Aid in Wildlife Restoration Project W-1I7-R. Literature Cited *Ballard, W.B., R.O. Stephenson, and T.H. Spraker. 1981. Nelchina Basin Wolf Studies. Alaska Department of Fish and Game. P-R Final Report, W-17-8, W-17-9, W-17-10, and W-17-11. 201 pp. Ballard, W.B., A. W. Franzmann, and C.L. Gardner. 1982. Comparison and assessment of drugs used to immobilize Alaskan gray wolves (Canis lupus) and wolve- rines (Gu/o gulo) trom a helicopter. Journal of Wildlife Diseases. /n press. Van Camp, J., and R. Gluckie. 1979. A record long- distance move by a wolf (Canis lupus). Journal of Mam- malogy 60: 236. * Available from Alaska Dept. Fish and Game, P.O. Box 3-2000, Juneau, Alaska 99802 Received 17 May 1982 Accepted 4 November 1982 334 THE CANADIAN FIELD-NATURALIST Volsoy7, Reduction of the Pelvic Skeleton in the Threespine Stickleback, Gasterosteus aculeatus, in Two Lakes of Québec THOMAS A. EDGE and BRIAN W. COAD Ichthyology Section, National Museum of Natural Sciences, Ottawa, Ontario KIA 0M8 Edge, Thomas A., and Brian W. Coad. 1983. Reduction of the pelvic skeleton in the Threespine Stickleback, Gasterosteus aculeatus, in two lakes of Québec. Canadian Field-Naturalist 97(3): 334-336. Populations of Gasterosteus aculeatus having reduced pelvic skeletons are reported for the first time in Eastern Canada from the Saint John River drainage in Québec. Examination of 120 specimens from 12 collections revealed pelvic skeletal variation from total absence to a full complement of incompletely developed pelvic elements. Bilateral asymmetry in pelvic structure was common. Brook Trout, Sa/velinus fontinalis, were recorded from both lakes and were known to eat G. aculeatus. Key Words: Threespine Stickleback, Gasterosteus aculeatus, pelvic skeleton, Eastern Canada. The Threespine Stickleback, Gasterosteus aculea- tus, shows considerable morphological and meristic variation throughout its circumboreal distribution. Populations of G. aculeatus having reduced pelvic skeletons are known from the Queen Charlotte Islands, B.C. (Moodie and Reimchen 1973), Texada Island, B.C. (Bell 1974; Larson 1976) and the Outer Hebrides, Scotland (Campbell 1979). Similar pelvic reductions are known for other stickleback species, Pungitius pungitius (Nelson 197la; Coad 1973), Culaea inconstans (Nelson 1969) and from Pliocene fossil records of Gasterosteus doryssus (Bell 1974). The purpose of this study is to report, for the first time in Eastern Canada, the presence of two popula- tions of G. aculeatus with reduced pelvic skeletons. Lake surveys were conducted by le Ministere du Loi- sir, de la Chasse et de la Péche in Temiscouata County, Québec during the summers of 1980 and 1981. Surveys of Lac Croche (47°47'50’N, 68°47’/18”W) at 259m altitude and Lac Rond (47° 44’42’”N, 68°44’01”W) at 229 m altitude within the Saint John River drainage produced 12 collections with 120 specimens of Gasterosteus aculeatus, all with reduced pelvic skeletons. The morphological varia- tion and frequency of pelvic skeletal reduction exhi- bited are shown in Figure | and Table | respectively. Pelvic skeletal elements were described according to Nelson (1971b) and the figure was drawn from six specimens stained in alizarin. None of the fish examined had a completely devel- oped pelvic skeleton and bilateral asymmetry in pelvic structure was found in 58% of the fish examined. All specimens had a normal complement of three dorsal spines and lateral plate development was consonant with that reported elsewhere for freshwater popula- tions (Hagen and Moodie 1982). In specimens longer than 34 mm both the completely-plated morph and the partially-plated morph were found, with the latter predominating (> 90%). Keel development in the partially-plated morph was very weak. Counts of lat- eral plates for specimens longer than 34 mm were 10-21 (mean 13.8, standard deviation 2.5) for the partially-plated morph based on 31 specimens. Dorsal spine length in standard length for 30 specimens of the partially-plated morph was 10.8-15.3 (mean 13.2, standard deviation 1.2). Limnological data available from surveys showed Lac Rond to have a maximum depth of 25 m, a sur- face area of 150.3 ha, and a conductivity of 100 umhos. Lac Croche has a maximum depth of 15 m, a surface area of 80.4 ha, a conductivity of 80 umhos, and a pH of 7. Allsticklebacks were caught in depths of less than | m by electro-fishing or by seining over rocks, gravel and sand, or among vegeta- tion and dead trees. Salvelinus fontinalis has been recorded from both lakes since 1973 but evidence of its presence or absence previous to this date is lacking (C. Banville, personal communication, 1981). The cyprinid Pimephales promelas has been recorded from Lac Rond but not Lac Croche. Stomach content analysis of S. fontinalis from Lac Croche revealed the presence of G. aculeatus in 25 of 75 trout examined. The co-existence of S. fontinalis with G. aculeatus having pelvic reductions in Lac Croche and Lac Rond shows that no simple correlation of pelvic skeletal reduction with the absence of predatory fish exists, as is also emphasized by Reimchen (1980) and indicated in Campbell (1979). G. aculeatus populations with reduced pelvic skeletons reported by Reimchen (1980; 1982 ms: “Status report on endemic stickleback (Gas- terosteus aculeatus) on the Queen Charlotte Islands, British Columbia”, submitted to the Committee on the Status of Endangered Wildlife in Canada, 14 pp., 4 figs.) in the Queen Charlotte Islands were from small, shallow, bog lakes with a low pH in contrast to the lakes reported here. Environmental factors may have B35) NOTES 1983 XY Py te ~ CUES. oO. 2.50 9 4.50 8.50 4.25 Species 5 6 3 4 4.50 376 adjacent portions of southern Bylot Island (73°N, 78° W), which has been visited by ornithologists in 12 years over six decades. Here the vegetation in places was much more luxuriant than anywhere at Polar Bear Pass, and birds were more than twice as abund- ant (Van Tyne and Drury 1959). Some species com- mon at Polar Bear Pass such as the Red Knot and Sanderling rarely or never nested at Pond Inlet, and several species including the Golden Plover (Pluvialis dominica), Horned Lark (Eremophila alpestris), and Water Pipit (Anthus spinoletta) nested regularly here although absent from Bathurst Island. Notable addi- tions at Pond Inlet were small numbers of Eurasian species, the Ringed Plover (Charadrius hiaticula) and Wheatear (Oenanthe oenanthe) (Renaud et al. 1981). Of the arctic censuses available to me, the most similar in methods and duration at high latitudes were conducted on two tracts of coastal plain at Barrow, Alaska (71° 18’N, 156° 42’W). Here intensive censuses were carried out for six consecutive years (Myers et al. 1981). The mean numbers of pairs were 135 and 133/km2, about 10 times the density of the wetlands of Polar Bear Pass. In these two locations separated by about 1800 km, the species composition was almost totally different, and catastrophic reductions in nest- ing characterized as “nonbreeding years” do not occur at Barrow (J. P. Myers, in litterature, 1982). Acknowledgments For help in this study I am indebted to S. D. Mac- Donald and the Museum of Natural Science, National Museums of Canada, for the hospitality of the High Arctic Research Station on Bathurst Island and to the Polar Continental Shelf Project for providing trans- portation and other services in the arctic. Associates engaged in their own studies who provided help on these censuses were as follows: D. A. Gill, D. R. Gray, D. F. Parmelee, Charles Roots, Tom Sadler, Philip Taylor, and particularly John Geale and Pierre Lamothe, who supplied data for years when I was not present. J. P. Myers read the manuscript and gave valued suggestions. THE CANADIAN FIELD-NATURALIST Vol. 97 Literature Cited Anonymous. 1981. Polar Bear Pass: an ecological check- list. Northern Perspectives 9(6). Beschel, R.E. 1969. Floristic relation of the Nearctic islands. Botanicheskii Zhurnal 54: 872-891, English sum- mary p. 891. Dement’ev, G. P., N. A. Gladkov, and E. P. Spangen- berg. 1969. Birds of the Soviet Union, Volume 3. Israel Program for Scientific Translations. Freedman, B., and J.Svoboda. 1982. Populations of breed- ing birds at Alexandra Fjord, Ellesmere Island, Northwest Territories, compared with other Arctic localities. Cana- dian Field-Naturalist 96: 56-60. Godfrey, W.E. 1966. The birds of Canada. National Museum of Canada Bulletin no. 203. MacDonald, S.D. 1981. Scientific progress: terrestrial biology, an overview. Pp. 171-186 in Proceedings of the 23rd Symposium, A Century of Canada’s Arctic Islands 1880-1980, Royal Society of Canada at Yellowknife, N.W.T. Mayfield, H. F. 1978. Undependable breeding conditions in the Red Phalarope. Auk 95: 590-592. Myers, J. P., T. A. Sordahl, B. J. McCaffery, and F. A. Pitelka. 1981. Wet coastal plain tundra. American Birds 35: 95-96. Nettleship, D. N., and P. A. Smith. 1975. Ecological sites in northern Canada. Canadian Commission for IBP. Con- servation Terrestrial - Panel 9. Ottawa. Pattie, D. L. 1977. Population levels and bioenergetics of arctic birds on Truelove Lowland, Devon Island, Canada: a high arctic ecosystem. University of Alberta Press, Edmonton. Renaud, W.E., W.G. Johnston, and K.J. Fin- ley. 1981. The avifauna of the Pond Inlet region, N.W.T. American Birds 35: 119-129. Savile, D. B. O. 196la. The botany of the northwestern Queen Elizabeth Islands. Canadian Journal of Botany 39: 909-942. Savile, D. B. O. 1961b. Bird and mammal observations on Ellef Ringnes Island in 1960. National Museum of Canada Natural History Papers 9: 1-6. Savile, D. B. O., and D. R. Oliver. 1965. Bird and mammal observations at Hazen Camp, northern Ellesmere Island, in 1962. Canadian Field-Naturalist 78: 1-7. Van Tyne, J., and W.H. Drury, Jr. 1959. The birds of southern Bylot Island, 1954. Occasional Paper, Museum of Zoology, University of Michigan, no. 615. 37 pp. Received 23 February 1982 Accepted 22 August 1982 The Ecology of the Muskrat, Ondatra zibethicus, at Luther Marsh, Ontario GILBERT PROULX and FREDERICK F. GILBERT! Department of Zoology, University of Guelph, Guelph, Ontario NIG 2W1 \Present address: Wildlife Biology, Washington State University, Pullman, Washington 99164-4220 Proulx, Gilbert, and Frederick F. Gilbert. 1983. The ecology of the Muskrat, Ondatra zibethicus, at Luther Marsh, Ontario. Canadian Field-Naturalist 97(4): 377-390. The ecology of Muskrat (Ondatra zibethicus) populations inhabiting areas of different floristic composition and subjected to seasonal changes in water level was studied at Luther Marsh, Ontario. House building sites consisted of open water areas and heavy emergent vegetation stands, particularly cattail (7ypha sp.) stands, with > 15 cm water. The average home range hada ratio of vegetation:open water areas of 1:1. With a decline in water level, Muskrats extended their home range significantly (p < 0.05) from 484 (+ 238.4) m? in June-July to 1112 (+ 842.7) m2 in August-September and increased significantly (p < 0.05) the average number of houses/ family from 1.5 (+ 0.7) in early summer to 3.9 (+ 1.7) in late summer. Cattail was the most important foodstuff but food habits depended upon the movements of the animals and the diversity of the flora. Adult and juvenile populations usually had an even sex ratio in summer. Most females had two litters with an average of 6.3 embryos/litter. In cattail-rich areas, females produced | to 4 more young/ litter than females of other habitat types and there were more animals/ha of home range ( > 100 animals) than in any other habitat type (< 80 animals). The survival rate of juveniles was estimated at 66.4% in summer and 31.8% in winter. The maximum life span of a Luther Marsh Muskrat would be five years. Key Words: Muskrat, Ondatra zibethicus, Luther Marsh, home range, populations. Muskrats (Ondatra zibethicus) inhabit a wide range of community types but generally prefer lentic water containing vegetation (Perry 1982). The effect of water depth on habitat selection by Muskrat has been noticed by Sather (1958) and Danell (1978) and the direct effect of fluctuating water levels on muskrat populations has been considered greater than the indi- rect effect through altering the composition of the vegetation habitat (Bellrose and Brown 1941). On the other hand, McDonnell and Gilbert (1981) suggested stability in habitat usage by Muskrats even when water levels declined substantially. An obvious lack of information still exists concerning the relationship between muskrat populations and their environment and the effects of environmental conditions on musk- rat reproduction (Errington 1937; Bellrose and Low 1943; Arata 1959) and food habits (Errington 1941) are poorly known. The present study concerned the ecology of Musk- rat populations inhabiting different habitats of a marsh subjected to seasonal changes in water level. The primary objective was to determine and quantify the effects of vegetative composition and water level fluctuations on the selection and utilization of marsh habitat by muskrats. Muskrat population productiv- ity and changes over the year, were also investigated. Study Area The Luther Marsh Management Area is located 65 km north of Guelph, Ontario. The reservoir aver- ages about 120 cm in depth with annual water level fluctuations of 50-70 cm. The climate is humid con- tinental with mean maximum and minimum tempera- tures for July of 23.6°C and 12.1°C, and for January, -5.6°C and -13.5°C, respectively. Mean annual pre- cipitation is 89.8 cm (Heidorn 1975). Marsh areas have predominantly submerged and emergent vegeta- tion including cattail (Typha sp.), smartweed (Polyg- onum sp.), pondweeds (Potamogeton sp.), sedges (Carex sp.) and bulrush (Scirpus sp.) (Schiefele 1973). Peat and muck are the two major soil types of the Management Area (Canada Department of Agricul- ture 1962, 1963). In 1979, five study areas representative of Luther Marsh habitats were delineated: East Bay (11.2 ha), Western Shore (0.5 ha), Teal Bay (0.8 ha), Creek (0.2 ha), Pond 1 (0.4 ha) and Pond 2 (0.3 ha) (Figure 1). In 1980, a sixth study area, Upper Shore of Teal Bay (0.3 ha), was studied (Figure 1). Methods In May 1979 and 1980, the vegetative composition of each study area and the degree of open water areas were determined. When possible, the vegetation of a study area was subdivided into belts, each of which consisted of a distinct cover type (Takos 1947). A transect was randomly located through each study area. Measurements of the width and length of each vegetation belt and its position relative to free water areas were made at 10 m intervals along the transect. Allinformation was transferred onto maps at an orig- inal scale of 1:1000. Vegetation and free water areas ST) 378 THE CANADIAN FIELD-NATURALIST LUTHER MARSH FIGURE |. Luther Marsh showing location of the study areas. were determined with the aid of a conventional pla- nimeter. The reservoir water level at the dam and the amount of precipitation were recorded by Grand River Con- servation Authority staff. Daily water depths in the total area of each study area, except East Bay, were recorded at 0, 5, 10, and 15m on each side of the transect in May and early June. In East Bay, water levels were recorded every month at Muskrat houses. In late summer, three transects were established between the shoreline and the back of East Bay and water depths were recorded every 2m along each transect. These values, along with the reservoir water level at the dam, permitted determination of monthly water level fluctuations and delineation of regions, within the study area, with distinct water depths. The vegetation was analysed with the aid of per- manent 0.5 m x 1.0 m plots placed at random in each of the vegetation belts. In Western Shore and Teal Bay, the number of plots ina belt was proportional to the area covered by that belt. Therefore, 44 plots were inventoried in Western Shore (horsetail belt, 25; cat- tail belt, 12; sedge belt, 7) and 38 in Teal Bay (cattail belt, 25; spike rush belt, 13). The vegetation of the other habitats was treated as one belt and 25 plots were placed randomly in each habitat. Due to the size of East Bay, 100 quadrats were inventoried. The number of individual shoots of each taxon was counted in each plot. Two inventories were carried out in summer 1979; the first in late May-early June and the second in early July. In 1980, similar inventories were conducted with a third one in August. Consider- ing that some plants grow slower than others and that Vol. 97 their representation was low in the May sample, only the July results were used to compare the floristic composition of the habitats. A similarity index between 1979 and 1980 floristic compositions of each habitat was calculated according to the formula Se e100 A+B where S is the similarity index in percent, A is the number of taxa in sample A, Bis the number of taxa in sample B, and C is the number of taxa common to both samples (Odum 1971). For each plot, the relative density of each plant was calculated and habitats were compared to each other by a stepwise discriminant analysis (F-to-enter = 4.0, F-to-remove = 3.9; Klecka 1975). This analysis is a multivariate statistical analy- sis technique that allows one to describe, differentiate and classify elements characterized by a set of “p” variables. In this analysis, according to the linear combination of “p” variables that best separated and characterized the sample plots of one study area from that of other habitats being compared, a sample plot was Classified in its original study area or in another study area to which it was floristically more similar. The variables were the relative densities in a sample plot of: cattail, bulrush, spike rush (Eleocharis sp.), bur reed (Sparganium sp.), Graminae, horsetail (Equise- tum sp.), arrowhead (Sagittaria sp.) and ‘others’. All taxa, except those included in ‘others’, were used by Muskrats of all study areas in their building and feed- ing activities. ‘Others’ included taxa which were found in 1979 or 1980, but not in both years. A forage ratio was used to relate the quantities of the various taxa found at the feeding platforms to the proportions available in the study area (Takos 1947). A ratio of 1.0 implied that the plant was found on the feeding platforms in frequencies proportional to its occurrence in the habitat. When the ratio was < 1.0, the plant was used less than would be predicted if selection was random. A ratio > 1.0 implied preferen- tial selection of the taxon by Muskrats. Only plat- forms of recent origin (characterized by fresh rem- nants) were tabulated. Surveys of feeding platforms lasted approximately one month and were related to the period during which a vegetation analysis was done. Study areas were surveyed for Muskrat houses dur- ing the last week of May 1979 and the second week of May 1980. Thereafter, weekly checks for houses were made during the live-trapping period. Inhabited houses were recognized by fresh signs of Muskrat activities. In 1979, the vegetation surrounding the houses was recorded in late May-early June, as the vegetation inventory was being conducted. In early June 1980, the vegetation surrounding all Muskrat 1983 houses was recorded. Each house was located by tri- angulation and plotted on the habitat map. Muskrat were trapped with National live traps (17.8 x 17.8 x 50.8 cm; Tomahawk Livetrap Co.) from 25 May to 23 September 1979 and from 16 May to 25 September 1980. Two traps were set in close proximity to each active house, usually at feeding stations, defecation sites or along runways. Those sites permitted optimum sampling (Aldous 1946; Erickson 1963; Vincent 1972). Traps were maintained at each site throughout the trapping season and baited with carrots (Erickson 1963; MacArthur 1978). Adult muskrats were immobilized with an intra- muscular injection of 10 mg Ketamine (Ketaset™-BTI Products Inc.). Animals were sexed by manipulation of the urinary sheath (Sather 1958) and exposure of the penis (Baumgartner and Bellrose 1943). A num- bered No. | monel tag was put in one ear and a numbered rabbit tag in the other (National Band and Tag Co.). Adults and juveniles were released after their tag number and capture location were recorded. Because grid trapping is inadequate for the study of Muskrat movements (Proulx 1981), related methods for the determination of the home range size (Sander- son 1966) were rejected. The home range size of a family corresponded to the immediate site of the dwell- ings of a family plus the surrounding area delineated by the lines interconnecting the outermost capture sites, defecation points, feeding platforms and trails (connected to the water pool of a dwelling). However, since areas used by Muskrats wandering far from their dwelling might have been missed, either because some feeding platforms were not found or Muskrat signs were absent, the home range sizes determined in this study corresponded to minimum areas, intensively used by muskrats. : In order to determine natural mortality, all study areas were free of any trapping pressure during fall 1979 — spring 1980. However, carcasses of muskrats harvested in adjacent areas by a resident trapper were obtained in fall 1978 and 1979. Mean litter size and number of litters per breeding female were estimated from placental scars and used to provide an estimate of summer juvenile mortality. The summer mortality rate, along with live-trapping information, permitted an estimate of fall Muskrat populations, and that, together with knowledge of the breeding stock the following year, permitted an estimate of the winter mortality rate. A life table, based on summer and winter mortality rates, was developed according to Caughley (1977). Results Habitat Characteristics The reservoir water level was approximately 30 cm PROULX AND GILBERT: MUSKRAT AT LUTHER MARSH 379 higher in 1980 than in 1979. The amount of precipita- tion from | May to 30 September was 33.9 cmin 1979 and 53.4 cm in 1980. Comparison of the floristic composition of the hab- itats by discriminant analysis indicated significant dif- ferences (p < 0.05) among marsh habitats. East Bay plots were characterized by high relative densities of cattail (> 46%) and Graminae (> 46%) shoots, West- ern Shore plots, by their relative density of horsetail shoots (> 28%) and Teal Bay plots, by their relative density of spike rush (> 41%) and arrowhead (> 11%) shoots. None of these plants was characteristic of Creek plots which consisted predominantly of sedge shoots (> 69%). Upper Shore of Teal Bay was similar to Teal Bay in its floristic composition. According to the linear combination of all these variables, > 70% of the plots of each study area remained in their original study area in which they were inventoried. This analy- sis indicates that each study area had a distinct floris- tic composition. On a scale of ascending importance of cattails, Western Shore was the lowest, followed by Creek, Teal Bay, Upper Shore of Teal Bay and East Bay. Indices of similarity > 70% indicated that the taxa composition of each habitat did not change markedly from 1979 to 1980 (Table 1), although spe- cies which grow better on dry or wet land than in flooded areas were more frequent in 1979. The difference in vegetation between ponds and marsh habitats is attributed to a lack of emergent aquatic vegetation in the former (Table 1). A Grami- nae belt surrounded both Ponds | and 2 and indices of similarity > 80% indicated that their respective taxa composition did not change from 1979 to 1980. In East Bay, open water accounted for 18% of total habitat area in June-July 1979, and during the entire summer 1980. In August and September 1979, 15% of the habitat was covered by free water areas. Four water regions were delineated in East Bay (Figure 2a). In 1979, only region IV was covered by more than 15 cm of water from May to September. In 1980, regions II, III], and IV were covered by more than 15 cm of water during most of the summer months. In Western Shore, five water regions were delineated in relation to the vegetation belts (Figure 4a). In 1979, only regions IV and V had > 15 cm of water from May to Sep- tember. In 1980, all regions had > 15 cm of water during all summer months. In 1979, 56% of Teal Bay was covered by > 15 cm of water from June to Sep- tember. In 1980, the entire habitat was covered by that much water from June to September. The water con- ditions of Creek varied with the amount of precipita- tion but there was > 15cm of water throughout summers 1979 and 1980. Some shores of the ponds were more affected than others by a drop in water level during summer 1979. In 1980, however, all shores Vol. 97 THE CANADIAN FIELD-NATURALIST CC 9L cl IZ ty SBE LI OV 0 =r OV = Ga tv tv 08 6L 08 08 6L 08 6L 08 6L 08 6L 08 6L 08 6L DIOADI snonvqg ‘ds winisd DIIDAD DID ‘ds uinuuayiuvstay)D ‘ds winyofiay ‘ds po1uosa asuaddD uUnjasinby ‘ds spidaposp ‘ds piupjyjainog ‘ds suatjoduy ‘ds osnjunjd ‘ds piyopuiisaT] ‘ds sndooaT ‘ds piupsvay ‘ds osppijos ‘ds xauvd ‘ds vydt] ADUIUDAL ‘ds wnupnjos ‘ds winiypy apupianyf{ unjasinby ‘ds pyiuap ‘ds winiupsavds ‘ds snjnounuoy ‘ds wnuosdjog snpypa sndig ‘ds p1inj 130g ‘ds stupyr0ajq JADNS WINS ‘ds WinoDxXDAD I ‘ds winiqgojidy 08 6261 Wed ees | ers | Wd Wd 2104S ysny ayids pened aspas pened [le1as10 H Z puod |] puog saddyq YIIIO Avg [ea], IIOYS UIIISI A Avg yseq 380 JUdSII BXBL YUM ILUQEH YOe| Ul $}O[q JO JUIdIIg uOXxe | ‘O861 Puke 6L6| Ane Ul ‘OLIeIUG ‘Ysie] J9YyIN] 78 s}eIQey jeIysn UDAIS JO UOIISOdWOD ISO, “| JIV | 1983 were well supplied with water. No information was recorded relative to water levels in Upper Shore of Teal Bay. Muskrat Installations and Home Ranges East Bay — In 1979, the vegetation stands sur- rounding 11 of 24 houses built in spring were carefully recorded. The majority (82%) of houses were sur- rounded by mixed Reed Canary Grass (Phalaris arundinacea)-cattail stands. In 1980, there were only 14 houses active in May and 67% of them were sur- rounded by the same mixed stands. The frequency of those stands was not significantly different from that observed in the habitat in 1979 (x2 = 0.297, p > 0.05) and 1980 (x2 = 0.601, p > 0.05). In 1979, 10 of the 24 installations were used intensively and maintained by Muskrats during June and July. These were the main dwellings where the young-of-the-year were born and weaned. The other 14 houses were used in early summer by bachelors and in July by first litter juve- niles investigating the surroundings of their main dwelling. In 1980, 10 of the 14 houses were main dwellings and the four others were used as shelters by adults of some families or first litter juveniles moving around the main dwelling. In July, another main dwel- ling, a burrow, was found. The average distance between two main dwellings in 1979 was 87.6 (+ 51.4) m and was not significantly different from that in 1980 (87.2 + 39.3 m; t = 0.023, p > 0.05). In 1979, two of the main dwellings were built in region II, two in region III and six in region IV. Region I was a dry-out in May and no main dwellings were built there. The reachable free water areas covered 19 280 m2 and 19% of the covered area was in region II, 31% in region III and 50% in region IV (Table 2). The hypothesis that muskrats utilize open water areas in exact proportion to their relative importance in each region of the study area was tested by the chi-square test. No significant difference existed between the observed occurrence of main dwellings in each region and the one expected according to the proportion of free water areas in each region of the study area PROULX AND GILBERT: MUSKRAT AT LUTHER MARSH 381 (x2 = 0.406, p > 0.05). In 1980, the proportion of main dwellings / region was also similar to the proportion of open water areas in each region (x2 = 0.119, p > 0.05; Table 2). In early summer of both years, Muskrats were concentrating their activities by their main dwel- lings (Figure 2a, 3a). The average area of 1979 June- July home ranges (484 + 238.4 m2) was not different from that of 1980 (302 + 202.3 m2; t= 1.664, p > 0.05). No home range in early summer 1979 or 1980 had < 25% of either open water or vegetation and the average home range had a ratio of vegetation:open water areas of 1:1 (Table 3). The average number of houses used by a family in June-July was 1.5(= 0.7) in 1979 and 1.2(+ 0.6) in 1980 (t = 1.867, p > 0.05). Building activity occurred in late summer 1979 and there were 44 active houses in August-September. The observed occurrence of main dwellings per region was significantly different from the one expected accord- ing to the proportion of open water areas in each region of the study area (x?= 11.097, p< 0.05; Table 2). Then, animals concentrated their building activity towards the edge of the lake. They were “over-using” region IV, the region where free water areas were surrounded by vegetation stands with > 15 cm of water. Only eight new houses were built in August- September 1980 and a total of 23 installations were used by Muskrats in late summer. The proportion of dwellings in each region was not significantly different from the proportion of open water areas found in each region (x2 = 0.790, p > 0.05; Table 2). The average distance between the main dwellings and the new houses was 39.6 (+ 33.1) min 1979 and 33.4(+ 19.7) in 1980 (t = 1.460, p > 0.05). In 1979, the late summer home ranges (Figure 2b) were significantly larger than early summer ones (t’ = 2.268, p < 0.05) and averaged 1112 (+ 842.7) m2. In 1980, the average late summer home range was 470 (+ 182.7) m?(Figure 3b) and was not significantly larger than that of June-July (t = 1.664, p > 0.05). It was, however, significantly smaller than August-September 1979 (t’ = 2.361, p< 0.05). There was no summer home range in 1979 or 1980 TABLE 2. Distribution of Muskrat dwellings and relative importance of open water areas (%) in East Bay in summers 1979 and 1980. June-July August-September 1979 1980 1979 1980 Open Water Open Water Open Water Open Water Water Dwellings Areas Dwellings Areas Dwellings Areas Dwellings Areas Regions % % % % I 0 0 l 5 0 0 | 5 II 2 19 2 18 2 8 4 18 Ill 2 3] 3 29 6 35 5 29 IV 6 50 5 48 36 57 13 48 382 THE CANADIAN FIELD-NATURALIST Vol. 97 1,2,... family number boundaries of home ranges ---- boundaries of water regions 1.1... water regions 6 main dwelling e house built inspring OG house built in august-Saptember — free water FiGuRE 2. Home ranges of families inhabiting East Bay in summer 1979: a) June-July home ranges; b) August-September home ranges. 1,2,...family number —— boundaries of home ranges ----boundaries of water regions LIL... water regions 4& main dwelling @ house built in spring Ohouse built in august-september “— free water FIGURE 3. Home ranges of families inhabiting East Bay in summer 1980: a) June-July home ranges; b) August-September home ranges. 1983 PROULX AND GILBERT: MUSKRAT AT LUTHER MARSH 383 TABLE 3. Average composition of early and late summer Muskrat home ranges in East Bay in 1979 and 1980. August-September June-July 1979 Average Average Area (m2) SED: Area (m2) Vegetation Stands 213.1 121.8 154.3 Open Water Areas 270.9 185.0 148.4 1980 1979 1980 Average Average SD Area (m2) S.D. Area (m2) S.D. 119.3 599.4 493.3 244.6 113.0 104.1 512.9 368.3 225) 100.0 with < 25% of either open water or vegetation. In August-September of both years, water covered 50% of the area of the average home range (Table 3). There were 3.9 (+ 1.7) houses/ family in late summer 1979. This average was significantly higher than in early summer 1979 (t’ = 4.128, p< 0.005). In late summer 1980, there were 2.2 (+ 0.7) houses/ family. This aver- age was significantly higher than in early summer 1980 (t = 3.382, p< 0.005) but significantly lower than in late summer 1979 (t’ = 2.943, p< 0.005). Western Shore — Three of four installations built in spring 1979 were main dwellings: one house in the cattail stands, and two burrows in the horsetail stands. At the end of July, the surroundings of the burrows were dried out and animals relocated in houses built approximately 40 m from shore, in water > 70 cm deep. During summer 1979, two captures occurred in the sedge belt, nine in the cattail belt and 16 in the horsetail belt. The observed occurrence of captures in each vegetation belt was not different from the expected occurrence of captures for each vegetation belt (x2 = 0.415, p > 0.05) and Muskrat appeared to use the belts at random. The summer home ranges averaged 2283 (+ 816) m2 (Figure 4b). In spring 1980, four of five installations were main dwellings: three houses in the cattail belt and one burrow in the horsetail belt. New building occurred in August and a total of nine installations were used by Muskrats: six in the cattail belt and three in the horse- tail belt. The number of installations in each vegeta- tion belt was not distributed proportionately to occur- rence of vegetation belts (x2 = 5.577, p < 0.05) which implies that the cattail belt was “over-used.” However, no significant difference existed between the observed occurrence of installations in each water region and the one expected according to the proportion of water regions in the study area (x2= 1.153, p > 0.05). Two of 17 captures prior to August occurred in the horsetail belt and 15 inthe cattail belt. The observed occurrence of captures in each vegetation belt was different from the expected occurrence of captures in each belt and Muskrat captures were not distributed proportion- ately to occurrence of vegetation belts (x? = 30.547, p < 0.005). In August and September, 31 of 46 cap- tures occurred in the cattail belt, 13 in the horsetail belt, and two in the sedge belt. Again, Muskrat cap- tures were not distributed proportionately to occur- rence of vegetation belts (x2 = 34.203, p< 0.005) and Muskrat were concentrating more their activities in the cattail belt than in the others. Trapping results and field observations occurred in the vegetative portion of the habitat and the proportion of free water areas used by muskrats in their movements from one site to another is unknown. However, using an estimate of 50% water and 50% vegetation based on East Bay data, home range averaged 1682 (+ 783) m? (Figure Ac). Teal Bay, Creek and Upper Shore of Teal Bay: In 1979, Teal Bay and Creek were used by only one family which established its house in the cattail stands of the creek. The same house was used in 1980 by one family and another main dwelling was found in Teal Bay, inaclump of willows (Salix sp.) at the junction of the cattail and spike rush belts. In Upper Shore of Teal Bay, three main dwellings were found among cattail stands. One of them was found destroyed in July and Muskrats built a new one among cattail and bulrush stands. Captures and feeding platforms were res- tricted to the immediate site of the houses and the area covered by the movements of the animals was not determined in any of these habitats. Ponds: In early summer 1979, seven burrows were found in Pond | but only two were still active in August. All others were abandoned as soon as their entrances became exposed by the drop in water level. In August, two more burrows were built in the shore of the island. Muskrats were using trails up to 10 m long leading from their burrows to feeding grounds. The home range was estimated at 3900 m?. It had 2100 m2 (54%) of vegetation stands and 1800 m2 (46%) open water. In 1979, in Pond 2, two burrows, located on the only shore well supplied with water, were used during allsummer months. Muskrats were using trails up to 8 m long leading from their burrows to feeding grounds. Their home range was estimated at 3200 m2. Vegeta- tion stands covered 1700 m2 (53%) and open water, 1500 m2 (47%). In 1980, Ponds | and 2 were used by the same family. Burrows that had been used from May to 384 THE CANADIAN FIELD-NATURALIST a Cc : a” F ° & amily a a A -t a if peer ares) | IP, eS MS. eta ae. ail a/Bt Ea ap ah a iFamily _-Ep ! 3 ai! 3 E Ht ' | A 1 ; Sod) a psy 6 =I = a | 'Famil | Vora main dwelling built in spring house built in spring house built in August-September - - Summer home ranges 2 horsetail belt W water region (H cattail belt — free water (5a) belt [trees and bushes A @ o Vol. 97 FIGURE 4. Vegetation belts and water regions (a) and location of muskrat installations and home ranges of families inhabiting Western Shore in summers 1979(b) and 1980(c). 1983 PROULX AND GILBERT: MUSKRAT AT LUTHER MARSH 385 September 1979 were used again. The home range was estimated at 7100 m2. Vegetation stands covered E a Sloe OS 5 69 6 3800 m2 (54%), and open water, 3300 m2 (46%). S RR ERI meses i) Food Habits < 3BzO asad = Sufficient information to study Muskrat food hab- 30 hae its was obtained from East Bay and Western Shore. In 3 =O laa +o = East Bay, cattail and Reed Canary Grass (Graminae) 2 had the highest frequency indices and were also 2 2 obs Sms Se important Muskrat food items (Table 4). Cattail was 5 ac E always abundant on the feeding platforms but its for- 2 Sesolkcea == age ratio decreased from May to August (Table 4). 5 cca In Western Shore, in 1979, all species, except bul- 5 SOa( 5 85 a= rush, were in proportions lower than their availability ® sn (Table 4). In 1980, Muskrat food habits were also AS Sr Eee oe Be diversified but the frequency of cattail on feeding ms ors S platforms at different periods of the year was always = SZ8O|AALT TA & very high and the species was used in proportions = SAE greater than its availability (Table 4). S a= |S25SS 5 Muskrat Populations = Spt l|Oonann In 1979 and 1980, most females in each habitat had = ove ra Se aac Se two litters. In 1979, the number of first litter juveniles 2 a aS Oils OS captured per female averaged 5.7(+ 2.3) in East Bay 2 a os (10 families) and 5.0(+ 2.0) in Western Shore (three =| Deas ied | eR CERIN & ee families). No significant difference existed between 5 q these two values (t= 0.473, p > 0.05). It appeared that = el Noe 4 (os females which had a second litter in East Bay (six < AER a rah SN Xe families) produced more young than in Western Shore S Bo olz 3aa a (three families) and the averaged number of second a ee = oad ae litter juveniles captured per family was 3.8(+ 2.5) and z TASzlSs BSS 9 1.3(4 1.5) respectively (t = 1.564, p > 0.05). In 1980, iS an average of 6.6 (+ 2.4) first litter young were cap- = ROR ANG ate ns Die tured per family in East Bay (11 families), 5.7(+ 2.9) e Lees). SS in Western Shore (four families) and 5.7 (+ 0.6) in ‘e s 2 € o/= ss Upper Shore of Teal Bay (three families). These values 2g 20 = were not significantly (p > 0.05) different from each ee “2 Sauls 2a other. The average number of second litter juveniles 5 captured per family in East Bay (eight families) and g naela Rance Western Shore (three families) was 5.4(+ 2.5) and 5.3 zy o PN BES. Oc os (+ 4.0) respectively (t = 0.049, p > 0.05). Few second = 5 Ae Sl A2AQa juveniles were captured in Upper Shore of Teal Bay ep a because of a lack of favorable trap sites in late summer a SAS "SP 2an 1980 and the trapping results relative to this age were & not considered for that habitat. & an 2t|rnata Juveniles represented between 66 and 92% of all & gis Beil mose trapped populations (Table 5). When all muskrat i So s olS7an populations of all habitats were grouped together, 5 2 Se 7 juveniles comprised 74.4% of the trapped population & De Py Gil Ral ensice) les in 1979 and 85.5% in 1980. The sex ratio of most ac population segments was even. When all populations 25 nice 3 were grouped together, adult sex ratios favored males = iS Bais = Wee (x? = 5.333, p< 0.05) in 1979 but were even in 1980. se AS BS SS Bl as First litter sex ratios were even in 1979 (x2 = 2.539, u 2 S ve nS 3 S Sih S p > 0.05) but departed significantly froma 1:1 ratioin S eS aS 5 § & s= s = 1980(x2= 6.931, p < 0.05). Second litter juveniles had be NWHONHAMHON 386 THE CANADIAN FIELD-NATURALIST V ol97. TABLE 5. Muskrat captures in Luther Marsh study areas in summers 1979 and 1980. Upper Shore Age and Teal Bay _— Teal of Teal Sex Classes East Bay Western Shore & Creek Bay Creek Bay Pond | Pond 2 1979 1980 1979 1980 1979 1980 1980 1980 1979 1980 1979 Adults 36 23 6 6 2 I l 3 D 2 2 Male £25 12 4 4 — — I ! I ! Female 1] 11 2} D I | l 2 ] ] 1 First Litter 57 73 15 22 3 9) 3 18 6 9 8 Male 88 *49 12) 7 2 5 2 10 2 7 3 Female 24 24 3 13 | 4 l 6 4 2 5 Unknown — — — 2 — — — 2 Second Litter 22. 38 4 15 l 3 _ 5 4 4 Male 12 23 3 9 | 2 _ 3 — 2 I Female 10 15 l 6 — | — 2 — 2 3 Third Litter — 9 7 — a — — Male oe 7 — — a o _ — — = Female — 2 — — - — — *Sex ratio significantly different (P<0.05) from 1:1. an even sex ratio in 1979 (x2 = 0.322, p > 0.05) and 1980 (x2 = 2.615, p > 0.05). Estimates of Population Size and Density In fall 1978, 15 female adults were autopsied and 186 placental scars were counted (12.4 scars per female). In fall 1979, 424 scars were counted for 33 breeding females (12.8 scars per female). Since most of the females of the studied habitats had two litters, these results indicate that female Muskrats produced an average of 6.3 embryos per litter, This average was applied to the live-trapping results to estimate popula- tion size in each habitat. The population of East Bay was the largest, followed by the populations of West- ern Shore and Upper Shore of Teai Bay, Teal Bay and Creek and Ponds (Table 6). These population estimates were used in a determi- nation of maximum relative densities which would have occurred if all animals were alive (Table 6), The relative densities were calculated with respect to the habitable portion of the habitats, according to water levels and open water areas. With the exception of Pond 2, there were between 19.3 and 22.4 animals/ha in 1979 and between 18.1 and 22.6 animals/ha in 1980. Population densities can also be estimated by the areas covered by the home range (Table 6). East Bay populations had markedly more animals/ha of home range ( > 100 animals) than in any other populations (< 80 animals), Yearly Muskrat Population Changes From spring 1979 to spring 1980, growth rates of 1.1., 1.3, 2.0, and 1.0 were obtained for breeding muskrat populations of East Bay, Western Shore, Teal Bay and Creek, and Pond | respectively. The TABLE 6. Estimates of Muskrat population size and density in Luther Marsh study areas in summers 1979 and 1980, Maximum Population Muskrat/ha Muskrat/ha of Study Area Estimate of Habitat Home Range 1979 East Bay 121*(137) 21.2 109 Western Shore 44 19,3 65 Teal Bay + Creek 15 22.4 — Pond | 8 20.5 20 Pond 2 15 46.9 47 1980 East Bay 148 DBE5 287 Western Shore 52 22.6 Uy Teal Bay 15 18.1 = Creek 8 — == Pond | 15 21.1 21 Upper Shore of Teal Bay 44 — — *Value obtained when one considers that thae adult popula- tion was made up of only 10 breeding pairs, The value within brackets includes adult male bachelors. 1983 Muskrat population of Pond 2 was completely exter- minated from 1979 to 1980. Considering that all popu- lations had a large number of juveniles in 1979, a considerable reduction of the densities occurred over winter 1979-80. In fall 1978, 159 carcasses were analysed and there were 8.6 juveniles per female adult. Considering that there were 12.4 placental scars per female, the juvenile mortality rate from birth to the fall trapping season was estimated at 30.6%. In fall 1979, there were 8.1 young and 12.8 placental scars per adult female ina sample of 338 carcasses, which give an estimate of mortality between birth and the fall trapping season of 36.7%. The juvenile mortality over the two summers averaged 33.6%. The reduction of Muskrat populations over winter was calculated according to the maximal summer population estimates (Table 6), the sex ratio of each population segment and the average summer survival rate of the juveniles. A 10% adult mortality during summer (Errington, personal communication in Olsen 1959), an equal mortality for males and females and an immigration rate equalled to the emigration rate were also assumed. In East Bay, the surplus of male adults in summer 1979 consisted of bachelors which left the study area over summer. Thus, only 10 males were considered as part of the summer adult population. Also, the winter natural mortality rate in this habitat was calculated according to a potential of 12 breeding pairs in East Bay in spring 1980 (one female died in early summer from Tyzzer’s disease). The winter reduction ranged from 60 to 75% in East Bay, Western Shore, Teal Bay and Creek, and Pond | (Table 7). The average was 68.2(+ 6.4)%. Reduction of male and female populations averaged 65.3(+ 13.1)% and 68.3(+ 6.5)% respectively. Male populations were not reduced significantly more than female populations (t = 0.407, p > 0.05). Life Table If the summer mortality rates of adults and juve- niles are fixed at constant rates of 10 and 33.6% respectively, and the winter kill is 68.2%, a hypotheti- TABLE 7. Winter reduction (%) of fall muskrat populations in Luther Marsh study areas in 1979-80. % Reduction Male Female Total Habitat Population Population Population East Bay 71.4 71.4 71.4 Western Shore 66.7 80.0 75.0 Teal Bay + Creek 60.0 60.0 60.0 Pond | 50.0 75.0 66.7 PROULX AND GILBERT: MUSKRAT AT LUTHER MARSH 387 TABLE 8. Hypothetical muskrat life table for a cohort of 1000 juveniles at Luther Marsh. Age Mortality Survival (months) Survival Mortality Rate Rate x Ix dx qx px 5 1000 336 0.336 0.664 5 664 453 0.681 0.319 12 211 21 0.099 0.901 17 190 130 0.684 0.316 24 60 6 0.100 0.900 29 54 37) 0.685 0.315 36 17 2 0.118 0.882 4] 15 1] 0.733 0.267 48 5 ] 0.200 0.800 33} 4 3 0.750 0.250 60 I _ — - 65 — — — -- cal life table can be built for a cohort of 1000 juveniles (Table 8). More than half the animals would die dur- ing the fall and winter months following their birth and only 21.1% would be present during the first breeding season. Maximum longevity would be 60 months. A survivorship curve for such a population would be characterized by a high mortality during the summer months of the first year and during each winter season (Figure 5). Discussion Muskrat houses were found in areas affording two essential constituents: 1) heavy building material and 2) enough water to cover house channels and provide Muskrats with access to their house beneath the water’s surface. According to water conditions which prevailed in regions where intensive building occurred and those where abandonment of houses was observed, a minimum of 15 cm of water was necessary to accomodate lodge-building Muskrats. This esti- mate is in agreement with the findings of Bellrose and Brown (1941). Abandonment of houses was observed with a decrease in water level (Coulter 1948; West- worth 1974; Danell 1978). Therefore, water plays the role of a dispersion factor. However, the selection of cattail stands by Muskrats in Western Shore 1980 suggests that, when water levels are > 15 cm deep everywhere within a habitat, vegetation stands become a dispersion factor. Therefore, the suitability of a site for Muskrat building activities depends on the simultaneous presence of heavy emergent vegetation and water. Although previous studies pointed out that muskrats intensively use burrows in summer (Dil- worth 1966; Philips 1979), the present study indicates that Muskrats used such structures as an alternative to houses only when heavy emergent vegetation was absent and/or water depths were < 15 cm. 388 1000 100 Number of Animals ° 0 15 30 45 60 Age (months) FIGURE 5. Hypothetical survivorship curve of a cohort of 1000 juveniles at Luther Marsh. Until now, no one has ever quantified the propor- tions of a home range in water and vegetation stands and thus determined the composition of suitable Muskrat habitat. In this study, the average home range had a ratio of vegetation stands:open water areas of |:1 and it is suggested that vegetation stands cannot be dissociated from water areas when one considers the composition of a suitable habitat. According to previous studies (Sather 1958; Erickson 1963; Neal 1968), summer home ranges can be enclosed by a circle 45 to 60 m in diameter. In this study, such a circle would have overestimated home range sizes by including areas unused by Muskrats, such as vegetation stands isolated from water areas and thus inaccessible. Furthermore, in this study, late summer home ranges often consisted of disconnected areas, each occupied by some family members concen- trating their activities around the installations that they built. The minimum home range sizes determined in this study would appear to be realistic in that they measured areas intensively used by Muskrats and not areas traversed by animals in the re-establishment of their areas of daily activity. Home range expansions during the summer could be due to juveniles which were forced to leave the breeding lodge before the birth of the next litter (Errington 1961). In this study, home range expan- sions occurred in late July. At this time of the year, second litter juveniles started to investigate the imme- diate surroundings of the main dwelling and first litter juveniles re-established themselves in new homes. However, with a decrease in the water level in late summer 1979, several vegetation stands became inac- cessible. An expansion of the home ranges could have THE CANADIAN FIELD-NATURALIST Violen served to re-establish a balance of water to vegetation and to avoid frictions usually observed in degraded environments (Errington 1951, 1954; Neal 1968). Muskrat food habits were related to three factors acting simultaneously: 1) diversity of the flora; 2) movements of the animals; and 3) preference of the animals. When possible, muskrats concentrated their activities near their installations. Also, food habits reflected the floristic composition of the surrounding stands. This explains the high occurrence of cattail on the feeding platforms. When the surroundings had a diversified floristic composition, as in Western Shore, there was a larger number of plants with a high per- centage of occurrence on feeding platforms. When vegetation and water did not make up, together, the surroundings of Muskrat installations, Muskrat food habits became more diversified. Captures in Western Shore, 1979, showed that feeding activities there occurred far from the installations and animals used the different vegetation stands as they encountered them. Finally, a variation in the forage ratio of the plants, such as cattail, indicated that there was some seasonal preference in plant food. This was also noticed by Butler (1940) and Takos (1947). In 1979 and 1980, most females had two litters in all study areas. This agrees with other studies carried out in temperate Canada and the adjacent United States, e.g. McCann (1944) in Minnesota, Gashwiler (1950) in northern Maine, Fuller (1951) inthe Athabasca-Peace Delta, MacLeod and Bondar (1952) in southern Manitoba and Stewart and Bider (1974) in southern Quebec. Bellrose and Low (1943) and Arata (1959) have suggested that litter size may vary with environ- mental conditions. At Luther Marsh, environmental conditions might have had an effect on the number of animals born alive or surviving during the few hours or days following their birth. Trapping results indi- cated that juveniles made up a larger portion of the population in 1980 than in 1979. In Western Shore, although no significant difference was observed between the average number of juveniles captured per family in 1979 and 1980 (because of small sample size), the fact remains that 1980 families produced four more second litter juveniles than the 1979 families. Considering that 1980 families were located in an environment more favorable with respect to vegeta- tion type and water conditions, a higher productivity of juveniles would have been expected as an outcome. Also, the difference between captures of East Bay and Western Shore in 1979, although not found statisti- cally significant, could be meaningful on a long term basis, if the same environmental situation was repeated over many years in each habitat. In 1980, the similarity between captures of juveniles of East Bay, Western Shore, and Upper Shore of Teal Bay appears to have been a result of families of all habitats having 1983 been subject to similar environmental conditions. During summers 1979 and 1980, adult populations usually had an even sex ratio. This was also found by Aldous (1947), Coulter (1948), and Erickson (1963). An even sex ratio and the fact that only one adult male and one adult female were captured per main dwelling supports the idea of a monogamous breeding system (Stevens 1955; Sather 1958; Mathiak 1966). The gen- eral trend of the juvenile populations was also towards an even sex ratio. That result agrees with the findings of McDonnell (1979) and with the theory of Fisher (1930) who concluded that, at equilibrium, an optimal organism should allocate half its reproductive effort to progeny of each sex. Very few publications concern Muskrat densities in summer. An average of 21 animals/ha was obtained in all habitats in 1979 and 1980. This is markedly smaller than the 123 and 55 Muskrats/ha reported by Lynch et al. (1944) and Vincent and Quéré (1972) respectively for more moderate climes, but is similar to 21.7 Musk- rats/ha estimated by McDonnell (1979) for Luther Marsh. Previous studies correlated population densi- ties to the type of vegetation present in a habitat (Bellrose and Brown 1941; Errington 1941, 1948; Alexander 1955). A vegetation-Muskrat density rela- tionship becomes apparent when one considers the density of animals/ha of home range. In this respect, there were markedly more animals in East Bay per unit area than in any other habitat. This observation agrees with Smith and Jordan (1976) who associated the largest concentrations of Muskrats/ha with the habitat richest in cattails. The summer survival rate was very high at Luther Marsh and was greater than the values reported by Baumgartner and Bellrose (1943), McCann (1944), Alexander (1955), Dorney and Rusch (1953) and Olsen (1959). On the other hand, the winter reductions estimated at Luther Marsh were very high, compared to 38 and 44% reported by Errington (1939) and Ste- wart and Bider (1974) respectively. Relatively low summer population densities could have favored a high summer survival rate of the juveniles and concur- rently, allowed too high a concentration for the harsh winter environment at Luther Marsh. Intraspecific strife, starvation and disease might have occurred in animal groups subjected to restricted winter feeding grounds (Errington 1954, 1961). According to the life table, 99.5% of a cohort of 1000 juveniles would not reach four years of age. Errington (1961) estimated the Muskrat life span to be three to four years but he had no numerical values to substantiate his conclusions. The shape of the survi- vorship curve implies a constant rate of mortality, independent of age. Although many Muskrats die in their first year of age, the reduction did not produce a concave survivorship curve as hypothesized by Giles PROULX AND GILBERT: MUSKRAT AT LUTHER MARSH 389 (1978). Probably no real population has a constant survival rate throughout its whole life span and the sigmoid curve reported in studies of other rodents (Caughley 1966; Barkalow etal. 1970) might be appli- cable also to Muskrat populations. Animals in their prime might be more capable of coping with the environmental conditions than are older ones, and the slope of the corresponding portion of the survivorship curve might be less steep in the real world. Data collected ina minimum five-year study would provide a more reliable survivorship curve. Acknowledgments We would like to thank the Ministry of Natural Resources and Grand River Conservation Authority for allowing us to study at Luther Marsh and for providing living accomodations. We are grateful to Nancy Wilson for assistance in data collection and to Dr. Charles Robbins, Washington State University, for review of the original manuscript. Funding was provided by Natural Sciences and Engineering Research Council of Canada Grant A 9546 to F. F. Gilbert. Literature Cited Aldous, S. E. 1946. Live trapping and tagging muskrats. Journal of Wildlife Management 10: 42-44. Aldous, S. E. 1947. Muskrat trapping in Sand Lake National Wildlife Refuge, South Dakota. Journal of Wild- life Management | 1: 77-90. Alexander, N. M. 1955. Variations in winter muskrat habi- tats and harvest. American Midland Naturalist 53: 61-70. Arata, A.A. 1959. Ecology of muskrats in strip-mine ponds in southern Illinois. Journal of Wildlife Manage- ment 23: 177-186. Barkalow, F.S., Jr., R. B. Hamilton, and R. F. Soots, Jr. 1970. The vital statistics of an exploited gray squirrel population. Journal of Wildlife Management 34: 489-500. Baumgartner, L. L., and F. C. Bellrose, Jr. 1943. Determina- tion of sex and age in muskrats. Journal of Wildlife Man- agement 7: 77-81. Bellrose, F. C., and L. G. Brown. 1941. The effect of fluc- tuating water levels on the muskrat population of the Illinois River Valley. Journal of Wildlife Management 5: 206-212. Bellrose, F. C.,andJ. B. Low. 1943. The influence of flood and low water on the survival of muskrats. Journal of Wildlife Management 24: 173-188. Butler, L. 1940. A quantitative study of muskrat food. Canadian Field-Naturalist 54: 37-40. Canada Department of Agriculture. 1962, 1963. Soil maps of Dufferin and Wellington counties. Soil Research Institute. Caughley, G. 1966. Mortality patterns in mammals. Ecol- ogy 47: 906-918. Caughley, G. 1977. Analysis of vertebrate populations. John Wiley and Sons, New York. 234 pp. Coulter, M. W. 1948. A study of movements and habitat preferences of muskrats in Maine. M.Sc. thesis, University of Maine, Orono, Maine. 114 pp. 390 Danell, K. 1978. Intra- and interannual changes in habitat selection by the muskrat. Journal of Wildlife Management 42: 540-549. Dilworth, R. L. 1966. The life history and ecology of the muskrat under severe water level fluctuations. M.Sc. the- sis, University of British Columbia, Vancouver. 125 pp. Dorney, R. S.,and A. J. Rusch. 1953. Muskrat growth and litter production. Wisconsin Conservation Department Technical Bulletin 8: 3-32. Erickson, H. R. 1963. Reproduction, growth, and move- ment of muskrats inhabiting small water areas in New York State. New York Fishand Game Journal 10: 90-117. Errington, P. L. 1937. The breeding season of the muskrat in northwest Iowa. Journal of Mammalogy 18: 333-337. Errington, P. L. 1939. Observations of young muskrats in Iowa. Journal of Mammalogy 20: 465-478. Errington, P. L. 1941. Versatility in feeding and population maintenance of the muskrat. Journal of Wildlife Man- agement 5: 68-89. Errington, P. L. 1948. Environmental control for increas- ing muskrat population. Transactions of the North Amer- ican Wildlife Conference 13: 596-608. Errington, P. L. 1951. Concerning fluctuations in popula- tions of the prolific and widely distributed muskrat. Amer- ican Naturalist 85: 273-292. Errington, P. L. 1954. The special responsiveness of minks to epizootics in muskrat populations. Ecological Mono- graph 24: 377-393. Errington, P. L. 1961. Muskrats and marsh management. Wildlife Management Institute Publication, University of Nebraska Press, Lincoln. 183 pp. Fisher, R. A. 1930. The genetical theory of natural selec- tion. Clarendon Press, Oxford. 272 pp. Fuller, W. A. 1951. Measurements and weights of northern muskrats. Journal of Mammalogy 32: 360-362. Gashwiler, J. S. 1950. Sex ratios and age classes of Maine muskrats. Journal of Wildlife Management 14: 383-389. Giles, R. H., Jr. 1978. Wildlife Management. W. H. Free- man and Co., San Francisco. 416 pp. Heidorn, K. C. 1975. Climatology of the region in the vicin- ity of Luther Marsh. A Quality and Meteorology Sector, Air Research Branch, Ontario Ministry of Environment (Loose-leaf publication). Klecka, W. R. 1975. Discriminant analysis. Pp. 434-467 in Statistical Package for the Social Sciences. Second edi- tion. McGraw-Hill, Inc., Toronto. Lynch, J. J., T. O'Neil, and D. W. Lay. 1947. Management significance of damage by geese and muskrats to Gulf Coast Marshes. Journal of Wildlife Management 11: 50-76. MacArthur, R. A. 1978. Winter movements and home range of the muskrat. Canadian Field-Naturalist 92: 345-349. MacLeod, J. A., and G. F. Bondar. 1952. Studies on the biology of the muskrat in Manitoba. Part I: Oestrus cycle and breeding season. Canadian Journal of Zoology 30: 242-253. Mathiak, H.R. 1966. Muskrat populations studies at Horicon Marsh. Wisconsin Conservation Department Technical Bulletin No. 36. 56 pp. McCann, L. J. 1944. Notes on growth and sex and age ratios and suggested management of Minnesota muskrats. Journal of Mammalogy 25: 59-63. THE CANADIAN FIELD-NATURALIST Vol. 97 McDonnell, J. A. 1979. Behavioral responses of muskrats (Ondatra zibethicus) to water level fluctuations at Luther Marsh, Ontario. M.Sc. thesis, University of Guelph, Guelph. 69 pp. McDonnell, J. A., and F. F. Gilbert. 1981. The responses of muskrats (Ondatra zibethicus) to water level fluctua- tions at Luther Marsh, Ontario. Pp. 1027-1040 in World- wide Furbearer Conference Proceedings. Edited by J. A. Chapman and D. Pursley. R. R. Donnelly and Sons Co., Falls Church, Virginia. : Neal, T. J. 1968. A comparison of two muskrat popula- tions. Iowa State Journal of Science 43: 193-210. Odum, E. P. 1971. Fundamentals of ecology. Third edi- tion. W. B. Saunders, Philadelphia. 574 pp. Olsen, P. F. 1959. Muskrat breeding biology at Delta, Manitoba. Journal of Wildlife Management 23: 40-53. Perry, H. R., Jr. 1982. Muskrats (Ondatra zibethicus and Neofiber alleni). Pp. 282-325 in Wild Mammals of North America. Edited by J. A. Chapman and G. A. Feld- hammer. The Johns Hopkins University Press. Philips, D. W. 1979. Muskrat population dynamics on a controlled wetland in southern Saskatchewan. M.Sc. the- sis, University of Regina, Regina, Saskatchewan. 118 pp. Proulx, G. 1981. Relationship between muskrat popula- tions, vegetation and water level fluctuations and man- agement considerations at Luther Marsh, Ontario. Ph.D. thesis, University of Guelph, Guelph, Ontario. 239 pp. Sanderson, G. C. 1966. The study of mammal movements — areview. Journal of Wildlife Management 30: 215-235. Sather, J. H. 1958. Biology of the Great Plains muskrat in Nebraska. Wildlife Monograph No. 2. 35 pp. Scheifele, G. 1973. Vegetation at Luther Marsh Wildlife Management Area. Ontario Ministry of Natural Resour- ces, Cambridge District, Cambridge. 58 pp. Smith, H.R., and P. A. Jordan. 1976. The effect of increased trapping pressure on the age structure and stabil- ity of an estuarine muskrat population. Northeast Fish and Wildlife Conference 33: 119-124. Stevens, W.E. 1955. Adjustments of the northeastern muskrat (Ondatra zibethicus spatulatus) to a northern environment. Ph.D. thesis, University of British Colum- bia, Vancouver. 189 pp. Stewart, R. W., and J. R. Bider. 1974. Reproduction and survival of ditch-dwelling muskrats in southern Quebec. Canadian Field-Naturalist 88: 428-436. Takos, M. J. 1947. A semi-quantitative study of muskrat food habits. Journal of Wildlife Management | 1: 331-339. Vincent, J. P. 1972. Etude d’une population de rats musqués (Ondatra zibethica) par marquage et recapture. Mamma- lia 34: 8-21. Vincent, J. P.,and J. P. Quéré. 1972. Quelques données sur la reproduction et sur la dynamique des populations du rat musqué Ondatra zibethicus L. dans le Nord de la France. Annales Zoologie-Ecologie Animale 4: 395-415. Westworth, D. A. 1974. Ecology of the muskrat (Ondatra zibethicus) on the Peace-Athabasca Delta in Wood Buf- falo National Park. M.Sc. thesis, University of Alberta, Edmonton. 147 pp. Received 26 October 1982 Accepted 28 March 1983 Some Effects of Winter Shelter Conditions on White-tailed Deer, Odocoileus virginianus, Fawns F. F. GILBERT! and M. C. BATEMAN? School of Forest Resources, University of Maine, Orono, Maine 04473 \Wildlife Biology, Washington State University, Pullman, Washington 99164-4220 2Canadian Wildlife Service, Sackville, New Brunswick EOA 3C0 Gilbert, F. F:, and M. C. Bateman. 1983. Some effects of winter shelter conditions on White-tailed Deer, Odocoileus virginianus, fawns. Canadian Field-Naturalist 97(4): 391—400. The behavior of 11 White-tailed Deer (Odocoileus virginianus) fawns was observed from December 1970 through March 1971. The animals were maintained in individual pens under cover conditions that were naturally wooded, clear-cut, or clear-cut with an artificial windbreak. Commercial feed was supplied ad libitum and consumption measured weekly. The responses of the animals to winter climatic conditions were monitored. The deer maintained in the clear-cut pens with an artificial windbreak lost a greater percentage of their body weight, had a lower kidney fat index, and ate less relative to body weight than the deer in the other pens. Fewer opportunities for socialization by these deer was considered partially responsible for these results. Solar radiation accounted for about 28 percent of the variation in food consumption. Food intake was maximum at a mean daily ambient temperature of -15°C, minimum at -4°C and decreased below -15°C. Greater feeding and other activity occurred when skies were overcast, when the barometer was falling and when vapor pressure humidity was increasing. The location of bed sites was affected primarily by the location of deer in adjacent pens. Sites that were potentially solar exposed were selected more often on clear than overcast days. The deer in the clear-cut pens maintained trails around the periphery of their enclosures and used those portions of the trails along fences where other deer were present more than those portions where no deer were adjacent. The animals seldom moved off their trail networks when snow thickness exceeded 30 cm. Energy costs of maintenance under these free-ranging conditions were higher than those predicted from laboratory studies by a factor of 2. Key Words: White-tailed Deer, Odocoileus virginianus, winter, shelter, energy requirements. Winter weather conditions at the northern extremi- ties of the White-tailed Deer’s (Odocoileus virginia- nus) range have long been considered a major limiting factor for this species. Low temperatures, high winds and deep snow cover place energy demands on the animal that it partially meets by lowering its basal metabolic rate (Silver et al. 1969). Previous studies have indicated that shelter from unfavorable weather is a primary reason for use of “varding” areas during the winter (Cook and Hamil- ton 1942; Webb 1948; Severinghaus 1953). Verme (1965) found that areas with abundant browse but little shelter were not used as wintering areas. Ozoga (1968) demonstrated that yard areas had smaller fluc- tuations in temperature and lower wind velocities than other sites, which would mean energy require- ments might be lessened in such locations due to more favorable climatic and snow thickness conditions. Robinson (1959) concluded that even in sparse cover deer could find adequate shelter from adverse weather. Although the general characteristics of deer winter- ing areas have been described by many authors (e.g. Severinghaus and Cheatum 1956; Gill 1966) and it is accepted that protection from severe weather is an important factor in determining yard location, there have been few studies which show how the lack of cover may affect wintering deer. Tarchinski (1969) attempted to show the effects of temperature and wind velocity on the winter nutritional requirements of White-tailed Deer fawns but found only that variation in coat insulation was an important factor. Robinson (1959) showed that penned deer on a restricted diet bedded in areas with the most favorable microclimate, but Westover (1971) found no difference between the cover conditions at bed sites and those at random points in Michigan deer yards. Hugie (1973), studying a northern Maine deer population, found that the animals completely abandoned a winter shelter area in favor of an active cutting operation nearby in more open cover conditions. It appears that White-tailed Deer employ a number of behavioral and physiological strategies designed to ameliorate or minimize the effects of severe winter weather conditions. The purpose of this study was to relate differences in the behavior and condition of White-tailed Deer penned under three different cover situations to environmental factors. Methods Description of Study Area The study was conducted at the big game enclosures 39] 592 located in the D. B. Demeritt Forest at Orono, Maine. This area is included in Fobes’ (1946) Central and Southwestern Interior Climatic Division. The annual mean snowfall for this division is 193.0 cm and the mean January temperature is between -7.8° and =H (Ce One deer was confined in each of eleven 0.10 ha pens that provided 3-4 replicates of each experimental cover condition (Figure 1). The experimental condi- tions were: (1) natural tree cover (N), (2) absolute clear-cut (A), and (3) clear-cut with vertical wall windbreaks (C). Each pen was gridded into 12 separ- ate areas by red wooden posts. Each deer was desig- nated the letter-number sequence of the pen it occupied. The A and C pens were cleared of all woody growth the summer of 1969. All edible brush and branches were removed from the N pens toa height of 1.8 m. In order to reduce the shelter effects of trees adjacent to, but outside, the experimental pens, 9 and 15 m wide strips were clear-cut at the east and west ends, respec- tively, of the A and C pens during the summer of 1970. Although the difference in elevation was small, all pens sloped west. The overstory density for each area within each pen was measured witha spherical densiometer (Lemmon 1956) in January, 1971. The average value of the over- “clearcut I a 3 1cm=8.5m + red post e) pyrheliograph a hygrothermograph Scale: THE CANADIAN FIELD-NATURALIST Vol. 97 story density measured when facing north, west, south, and east in the center of each area of each pen was determined. Average N, A, and C values (in %) were 72, 6, and 8, respectively. The total basal surface area in the N pens ranged from 4.01 to 1.34. In area 9 of each C pen (Figure 1), a vertical-wall windbreak consisting of three sides radiating from a central post was constructed of rough lumber. Each side was 2.4 m long and 2.1 m high. Description and Treatment of Experimental Animals Six male and 5 female fawns were used in the study which began during December 1970. Three males (N4, A4, C4) and 3 females (N2, A2, C2) were received 2 October 1970 from the game farm of the Maine Department of Inland Fisheries and Game located at Gray. Two males (NI and Al) and one female (N3) were part of the Orono deer herd and had been raised by their does in the experimental pens. The other fawns (A3 and Cl) were captured on Swan Island Game Management Area at Richmond, Maine, 7 August and 27 August 1970, respectively. Deer N1, Al, Cl, N3 and A3, are designated Orono-raised; deer received from the game farm are referred to as game farm-raised. All experimental animals were allowed complete freedom of the 1.2 ha experimental area from the time they were received (or from the first of 0 observation tower ——P Z clearcut anemometer feeder *% F )— __ windbreak p precipitation gauge FIGURE |. The experimental pen lay-out showing positions of wind-breaks, climatological instruments, feeders, grid markers and grid number. The clear cut areas adjacent to the pens are also shown. 1983 September when the Orono does were separated from their fawns) until placed in the experimental pens. Description of Behavioral Observations The basic observation period was a 10-min interval during which the behavior of each deer was recorded sequentially on a data sheet. These 10-min intervals were repeated continuously for about 3 h. The loca- tion of each deer within its pen and the animal’s behav- 10r were noted (Bateman 1972). The behavior of deer in each replicate was recorded as quickly as possible. Generally, 5-6 min separated the observations of the first and last replicates within one 10-min period. The deer were observed during all hours of the day so that the total of the observations gave information on a complete solar day. Data recorded at the beginning of each 10-min observation period included: the local time of day (E.S.T.), the time before or after sunset (BSS, ASS) or sunrise (BSR, ASR) corrected to local incident time, cloud conditions, wind speed and direction, and precipitation. Measurement of Feed Consumption The deer were allowed to eat New Hampshire deer pellets (Thompson et al. 1973) ad libitum from indi- vidual feeders. Each feeder was located within 3.0 m of the red marker post nearest a corner of the pen so that the deer had to move away from the fence area to eat. Each week the amount of feed required to fill each feeder to a permanent line was weighed to the nearest 0.2 kg. This was assumed to be the weight of feed eaten by the deer during the previous week. The accuracy of this method was determined by refilling feeders from which various amounts of feed had been previously removed and weighed. From a series of 20 of these trials it was estimated that the weight of feed con- sumed weekly could be measured with 95 percent confidence within + 1.6 kg. Two control feeders which could not be opened by the deer were set up to determine feed loss to small mammals and birds. One control was placed in H4 while the other was moved from A4 to C4 after one month. Although Red Squirrels (Tamiasciurus hud- sonicus) and small birds were seen leaving both the control and experimental deer feeders, the small amount taken could not be measured by the tech- niques of this study. No spillage by the deer was observed. The weekly food consumption was divided by 7 and expressed as kg/day, or, using the average of the December-January and March body weights, expressed as kg/day/100 kg body weight. Collection of Meteorological Data Relative humidity, temperature, solar radiation, and barometric pressure were recorded using Casella GILBERT AND BATEMAN: WHITE-TAILED DEER FAWNS 398, 931 hygrothermographs, Belfort 53850 pyrhelio- graphs, and a Short and Mason A2868 barograph. The positions of these instruments during the study are shown in Figure |. Average daily maximum and minimum tempera- tures, average daily percent humidity, vapor-pressure humidity, and mean temperature were calculated for each week. The temperature and humidity measured by each instrument were checked periodically with a standard thermometer and Bendix Model 566 psych- rometer. The amount of precipitation that fell during each storm was measured ina U.S. Weather Bureau type precipitation gauge at the location shown in Fig- wre I ; Solar radiation, in langleys per day, was calculated from the average of three measurements of the area under the daily curves on the pyrheliograph charts. Totalizing anemometers (five Belfort 5-349, three Casella w1200/1) (Figure 1), recorded weekly total wind mileage. The average snow thickness of each cover condition was determined from 10 permanent snow stakes installed 15.2 m apart ina line through the center of each cover type. In addition, a meter stick was used to measure the snow accumulation in each area of each pen. These snow thicknesses were mea- sured at approximately the same place each time. Deer sinking depths were determined by measuring the depth of deer tracks when deer moved off the trails, or estimated with a snow compaction gauge (described by Verme, 1968) when deer used trails exclusively. All snow measurements and deer sinking depths were recorded after precipitation (several hours to two days), or at least once every week. Trails and bed sites were mapped when snow thicknesses were measured. If the deer had travelled the same path three or more times, it was marked as a trail. The location of tracks was also mapped. Weighing and Necropsy Procedures All deer except CI were weighed to the nearest 0.4 kg December 18 or 19. Cl was weighed February 2. The animals were killed during the period 30 March -11 April, weighed and autopsied. A kidney fat index was calculated for each animal, using the method of Anderson and Medin(1965). The wet weights of the kidneys were recorded and the amount of body fat described. Three femur bone mar- row samples from each deer were frozen for later determination of the fat content by a modification of the Association of Official Agricultural Chemists’ (1965) ether extract method. This analysis was per- formed on two of the samples from each deer. When there was a large difference between samples from the same deer, a third sample was analysed and the two closest values averaged. 394 Statistical Analysis of Data Analyses of variance were run on data grouped to test for significance between cover conditions, between individual deer, between sexes, and, in some cases, between the Orono and game farm-raised animals. Significance, or nonsignificance was deter- mined by the Student-Newman-Keuls’ (SNK) test (Steel and Torrie, 1960) at the 5 percent level unless otherwise stated. If the F ratios from analyses of var- iance were not significant, nonsignificance was assumed and SNK differences were not calculated. In most cases the December-January data were ana- lysed together because a relatively small number of observations were made in December. All measure- ments except organ weights were made in English system units and converted to metric units after analysis. Results Condition Indices The C deer lost a greater percentage of their December body weight (9.6 percent) than the A (6.2 percent) or N (6.1 percent) deer even though their December body weights were 3.0 kg lighter than the N deer (Table 1). The game farm deer lost two percent less weight than the larger Orono-raised deer. A negative r value (-0.36) between weight gain and December body weight of all deer was not significant. The A and C deer had much lower kidney fat indices (39.6 and 32.7 percent, respectively) than the N deer (53.9 percent). They also averaged lower bone marrow fat percentages (70.4 and 75.6 compared to 86.0 percent). Females lost more weight on the average than males (P < 0.10) but both kidney fat indices and bone mar- row fat percentages were higher in the females. The game farm-raised fawns had a much lower (29.7 per- cent) kidney fat index than the Orono fawns (58.8 percent), but the bone marrow fat percent was higher in the game farm deer (80.2 vs. 74.1 percent). The differences in condition indices (kidney fat THE CANADIAN FIELD-NATURALIST Vol. 97 indices, bone marrow fat percentages, weight gain in pounds, or weight gain expressed as percent of December body weight) between cover types, sexes, and origins were not significant. Food Consumption The differences in average food consumption of deer by cover condition were not significant, but when the food consumption of game farm-raised deer was analysed on an individual deer basis, N and A deer ate significantly more than C deer (Table 2). Food con- sumption of Orono-raised fawns was not tested separ- ately because of the small sample size. When data from all deer were analysed, no signifi- cant differences in food consumption by deer in the different cover conditions were evident with the exception of the 22 January to 26 March period when N and A deer ate significantly more than C deer. Orono-raised deer ate significantly less (P < 0.01) per unit body weight than game farm-raised deer. However, the difference in daily food intake was not significant. The correlation between average weekly feed con- simption and weight gain of each deer over the study period approached significance (r = 0.50, significant r = 0.58). Combined data from all deer showed a significant negative correlation (P < 0.01) between food consumption by deer and total weekly solar radiation in langleys (r = -0.33). Food consumption by the N deer was less correlated with total solar radiation (r = -0.54) than was food consumption of either the A (r = -0.62) or C(r = -0.71) deer. General Activity Most activity occurred during the hour after sunrise and the hour before sunset in all cover types in all months. Activity peaks occurred more frequently in March (at two to three hour intervals) than in Febru- ary or January (at three to four hour intervals). Deer activity during periods of falling snow was not significantly different from activity at all times. Activ- ity increased significantly (P < 0.01) as temperature increased. The percent of observations of deer lying TABLE 1. Mean December and March body weights of experimental deer. Body weight (kg) Weight gain Percent Deer Group December March (kg) change N 41 38 -2.7 -6.1 A 38 34 -2.3 -6.2 C 38 34 -3.5 -9.6 Male 40 38 -2.3 -4.9 Female 36 B8 -3.2 -9.1 Orono 41 37/ -3.2 -7.9 Game farm 37 34 -2.3 -5.9 1983 GILBERT AND BATEMAN: WHITE-TAILED DEER FAWNS 395 TABLE 2. Average daily weight of feed consumed by deer for each month based on weekly totals (kg/day/ 100kg). Game Farm Orono Date N A C N A C Farm Orono Dec. 4.1 4.0 3.9 3.1 3.6 3.6 4.0 3,5) Jan. Del 3.3 3.0 2.4 2.8 2.9 3.1 2.6 Feb. 2.9 Dell Dell 2.4 2.3 2.3 2.9 2.4 Mar. 32 Del Dep 8) 2.1 1.6 De 2.0 down decreased froma mean of 95.6 when air temper- ature was -18° to -34°C to 77.5 when the ambient temperature was -5.6° to -12°C, and to 26.6 when the temperature rose to between -5.6° and 10°C. The difference between activity in the temperature range -12° to -18°C and in the temperature range -5.6° to -12°C was significant. When the deer’s activity at each temperature range was analysed for each cover type separately, activity at the highest temperature range (-5.6° to 10°C) was in each case significantly greater (P < 0.01) than activ- ity at any of the lower temperatures. Activity of deer was compared in three ranges of humidity expressed as vapor pressure in (mm) of mer- cury: | = 0.000 to 0.050 mm; 2 = 0.051 to 0.200 mm; 3 = 0.201 to 0.300 mm. Activity increased but not always significantly for each range of greater humid- ity. The percent observations lying in humidity range 1 (57.2) were significantly greater than for humidity range 3 (44.5). These differences were not evident when each cover condition was analysed separately. Deer activity was greater under overcast than under clear skies, and greater when barometric pressure was falling than when rising or steady (Table 3). Activity when winds were moderate averaged 2.9 percent less than when there was no wind. Differences due to barometric pressure and cloud cover were significant when all cover types were grouped (Table 3). Feeding Activity The effect of weather conditions on feeding activity was similar to weather effects on general activity. The average number of deer eating each 10 min was greater under overcast (0.4) than under clear (0.3) skies, and greater when barometric pressure was falling (0.4) than when it was rising (0.3) or steady (0.3). Both of these differences were significant in the open but not in the N pens. The average number of deer eating during each 10 min observation period was the same (0.4) when wind speed was moderate, low or calm. There was no linear relationship between food con- sumption by deer and temperature. A third degree polynomial regression using average daily tempera- ture as the independent variable and food consump- tion as the dependent variable had an R? value of 0.48 and the F value was significant (y = 2.7820 + 0.2340x — 0.0199x? + 0.0004x3). A stepwise multiple regression analysis using food consumption as the dependent variable and total weekly solar radiation (langleys) (X1), average daily temperature (°F) (X2), and average daily wind mileage (X3) as independent variables resulted in the equation y = 3.4109 — 0.006X1 + 0.0107X2 -— 0.0073X3 Although average daily total wind mileage had a higher linear correlation (r = -0.33) with food con- sumption than did average daily temperature (r = -0.05), a greater amount of variation in food con- sumption was due to temperature as indicated by the results of a multiple linear regression. The only signif- icant reduction in variation was due to solar radiation which accounted for approximately 28 percent of the variation in food consumption. Since the R2 value was only 0.32, the equation was not further refined. Trail Length The C deer maintained an average trail length (116 TABLE 3. Inactivity expressed as percent of observations that deer were lying down during selected weather conditions. (Significant differences indicated by lines). Gover Cloud cover Wind Speeds Barometric pressure type clear overcast none light moderate falling rising steady N 65.5 59.3 58.9 56.2 59.3 59.6 62.5 66.3 A 63.4 58.9 57.4 55.4 62.9 54.4 62.3 62.4 € 64.1 hate Moe 53.0 60.0 48.2 58.3 57.7 All 64.0 S51 57.8 54.9 60.7 54.1 59.6 61.2 **P < 0.01 396 THE CANADIAN FIELD-NATURALIST Vol. 97 m) shorter than A (125 m) or N (132 m) deer. The average total trail length was shorter in pens of Orono-raised deer (119 m) than in pens of game farm deer (131 m). None of these differences were significant. The length of N3 trails (98 m) averaged significantly less than those of N4 (149 m), NI (143 m), A2 (140 m) or N2 (137 m). A simple correlation between trail length and deer sinking depth was significant (P < 0.01) (r = 0.34). The deer seldom moved off trails when sinking depth was greater than 30 cm. Selection of Bed Sites The most frequently used bed sites were in corners of pens close to bed sites of other deer. Ninety percent of all bed sites were within 4.5 m of bed sites of one or more other deer. All but four (16 percent) of the most frequently used bed sites were within a 4.5 m radius of a pen corner. Selection of solar-exposed sites by the A and C deer increased from January through March. During Jan- uary, 67 percent of the daytime changes in bed site selection of these deer was to more solar exposed sites when skies were less than one-half clouded. The Feb- ruary and March percentages were 70 and 80 percent, respectively. The N deer showed fewer changes in bed site preferences due to cloud cover, and a smaller percentage of these changes was to solar exposed sites. Bed site selection appeared affected by temperature only during December-January. During this period six of the deer selected sites of greater overstory den- sity when temperatures were less than -23.3°C (-10° F). Deer were recorded in some seldom-used bed sites when temperatures were between -5.6° and -17.7°C only when the sites were solar exposed. The selection of a bed site was dependent on location and temperature of the bed site (Table 4). Some changes in bed site preference were evident during periods of snowfall. Deer Al, Cl, C2, and C4 selected bed sites in the open more frequently during precipitation. However N4 was observed lying in the most sheltered area of its pen 37.1 percent of the observations during snowfall compared to 16.4 per- cent of the observations when there was no precipita- tion. i Most association with other deer occurred when the animals were lying down. Differences in the amount of association between deer in the three cover condi- tions were not significant during December-January, or March, but N and C deer were both significantly less “social” than A during February. The C deer were significantly less “social” than A deer during the ASR quarter of the day during both February (49.2 percent of the observations by no other deer compared to 27.1 percent) and March (38.1 percent compared to 35.5 percent). During the BSS quarter of the day, N deer were observed within 4.5 m of one or more other deer significantly less than A deer in February (54.1 percent compared to 74.4 percent) and in March (57.3 compared to 65.4 percent). The greatest differences in deer association between cover types were recorded in February. Discussion Percent bone marrow fat content, kidney fat index, and weight loss (expressed as weight gain) were used as condition indices. Since kidney fat is usually metabolized before bone marrow fat, it was probably the better indicator of deer condition in this study. Since all deer were in relatively good condition, the variation in bone marrow fat percent probably only indicated the naturally occurring variation in fawns. Robinson (1959) found indications that smaller deer grew more than larger deer during the winter. His TABLE 4. Frequency of utilization of each bedsite at each relative temperature for each month and the result of Chi-square tests. (Parentheses indicate number of times each bedsite showed each relative temperature). Relative Bedsite Bedsite Bedsite Month temperature* area | area 3 area 9 Jan. l 5(34) 2( 5) 5(16) 2 -( 6) 3(23) 6(24) 3 2( 6) 1(18) 3( 6) x2 = 9.39 sig. x? = 9.49 Feb. | 8(30) -( 4) 8(21) 2 6(13) -(16) 7(21) 3 2(2) 5(25) I( 3) x? = 21.12 sig. x?(p S 0.01) = 13.13 Mar. | 2( 8) -( 2) 6(24) 2 11(15) 2(12) 3( 5) 3 4( 7) 1(16) 1) *1 = coldest, 3 = warmest x? = 10.87 sig. x? = 9.49 1983 study showed that deer that lost the most weight grew the least during the winter. Skeletal growth was not measured in this study but there are some indications that the smaller deer lost less weight. The game farm- raised deer were smaller, lost less weight and had less body fat reserves, as measured by kidney fat indices, than Orono deer. This perhaps suggests more true growth occurred in the game farm deer. This is'also indicated by a lack of significance in correlation between kidney fat indices and weight loss or initial body weights. If all the deer had not grown or had grown the same amount, the amount of body fat present should have been related to the body weight lost. If, however, the smaller deer grew more, and at the same time metabolized more of their fat reserves, there would be little relation between remaining body fat and weight loss. Thompson et al. (1973) suggested tissue energy and nitrogen retention were better indi- ces of growth in fawns than body weight change. However, they too concluded that growth continued during the winter period. Food Consumption A significant difference in food consumption rela- tive to body weight between Orono and game farm fawns made it necessary to consider these two groups separately. The study showed that as much food per kg body weight is required to maintain a small fawn as a large one under similar circumstances. The difference between average food consumption of deer in different cover conditions was more pro- nounced in the latter part of the winter. This may, in part, have been due to increased solar radiation, since solar radiation caused a large part of the variation in feed consumption. Because the C pens received more solar radiation than N or A, those deer would there- fore have been more subject to changes caused by differences in solar energy levels which might explain their lower food consumption rates. Northern White-tailed Deer voluntarily restrict their food intake with a resulting decrease in body weight during the winter months (McEwenet al. 1957; Silver et al. 1969; Ozoga and Verme 1970). A sudden increase in food consumption during the weeks pre- ceeding 11 and 18 December corresponded to an abrupt drop in average daily temperature from 0.3°C to -13.3°C. Since this was the first cold weather of the season, it is quite possible the deer required two weeks to become acclimated to these severe temperatures. Colder average daily temperatures later in the winter did not have a similar effect on food consumption. Temperature appeared to have a polyphasic effect on food consumption. A low feed intake when average daily temperatures dropped to less than approxi- mately -15°C was probably due to conservation of energy by inactivity. Some homeothermic animals GILBERT AND BATEMAN: WHITE-TAILED DEER FAWNS 397 such as deer are known to conserve heat during cold weather by remaining inactive (Moen 1976). Maxi- mum food consumption when daily temperatures averaged from -15° to -12°C may indicate a tempera- ture range where energy requirements are high due to heat loss and relatively uncurtailed activity. A decreasing food consumption when average air temperatures were -12°C to -4°C probably reflected decreased thermoregulatory requirements. Silver et al. (1969) recorded a continuous drop in fasting meta- bolic rate during the fall until November-December. Food consumption in this study did not reacha stable low until the end of January. Silver (1971) recorded a continuous drop in food consumption during the win- ter and Ozoga and Verme (1970) found that fawns increased their food consumption at the end of March. Food consumption per se is not a very precise mea- sure of energy metabolism but by comparing the results to laboratory studies of the maintenance energy requirements for deer fawns (Thompson et al. 1973; Baker et al. 1979), some predictive values for free-ranging animals can be obtained. The fawns in our study ate about the same dry weight of feed/kg body weight as the Colorado Mule Deer (Odocoileus hemionus) fawns (Baker et al. 1979), but 40-50% more than the New Hampshire White-tailed Deer fawns fed the same diet as ours (Thompsonet al. 1973). Baker et al. (1979) found the winter maintenance requirements for the Mule Deer fawns held outside under ambient conditions to be 158 kcal ME/kg W® while Thomp- son et al. (1973) computed 121 kcal ME/kg W®”? in January and 108 kcal ME/kg W®? in March. Our comparative data (using ME values for the New Hampshire deer ration reported by Thompson et al. (1973) would be 277 and 225 kcal ME/kg W°.? for January and March, respectively, anincrease > 100% over the predicted basal rates. When it is considered that our fawns had ready access to feed ad libitum and thus would have little energy expenditure for foraging it is possible to imagine energy requirements in the wild for White-tailed Deer fawns as much as three times those obtained in the laboratory by Thompson et al. (1973). Bed Site Selection The average bed site ina Maine deeryard has been described by Gill (1966) as being about 25 cm deep and about 1-2 m from a 20 cm dbh spruce in a spruce-fir stand with 41 to 70 percent overstory density. Day’s (1963) measurements supported these findings. Day (1963), Gill (1966), and Robinson (1959) have pres- ented evidence that suggests deer select bed sites that provide a more moderate microclimate than the sur- rounding area. However, Westover (1971) has shown that in a northern Michigan yard, areas of bed sites 398 did not differ from randomly selected areas in the yard. Location of bed sites within the pens appeared to be determined at least partially by the opportunity for association with other deer. Multiple bed sites are common in the wild; Day (1963) reported finding bed sites in groups of one to three and Westover (1971) calculated the average number of bed sites ina group to be three. The importance of such socialization may be greater in fawns than in adults, and is probably emphasized by the confining conditions of a pen study. Even so, the presence of other animals may be more important psychologically to deer survival than is realized. Despite the difficulties in distinguishing between bed site selections caused by shelter and those caused by social factors, some indications of variation in utilization caused by weather were present. Socializa- tion appeared less important in bed sites during adverse weather (storms, etc.), and during the night. Less grouping during the night might indicate a greater tendency to select bed sites which give shelter from the environment. The density of trees around a bed site has an impor- tant effect on snow thickness, wind speed, and temperature. Snow thickness under pen conditions proved to be relatively unimportant in bed site selec- tion. Day (1963) suggests that deer in the wild main- tain fewer bed sites in more sheltered areas when conditions are severe. Under conditions of this study wind speeds decreased with increasing distance into the N pens, increased from west to east, and were often least along the north edge of the C pens. Wind speeds in N pens also appeared affected by density of stems. Wind speed did not have any effect on bed site selec- tion. However, the influence of wind in bed site selec- tion could have been masked by other environmental factors. Temperature was probably affected by overstory density and stem density as well. The importance of nocturnal radiant energy has been hypothesized for White-tailed Deer by Moen (1968a). The net long wave radiation value is affected by the density of the canopy cover (Lull and Relgner 1967). Moen (1968b) found that the infrared radiation from a tree affected radiometers at ground level to a distance of about one-half the height of the trees. The environment of the clear-cut pens would be affected, at most, one-half way across each pen. The A deer consistently selected bed sites in the radiant environment of the N pen trees. The C deer showed a tendency, at night when radiant energy from the trees would be most valuable, to select bed sites in the open. Under cold windy conditions, deer Cl and C2 often selected more sheltered bed sites. When temperatures were as low as -26°C and winds THE CANADIAN FIELD-NATURALIST Vol. 97 were light, the C deer preferred the open sites. The A deer also were recorded lying in the open at low temperatures. Similar action has been reported for wild deer by Kramer (1971) in Alberta and Moen (1968c) in western Minnesota. Moen calculated that a 40 kg deer on full feed would reach a negative energy balance when standing under the conditions of his study at approximately 10 km/h wind speed and temperatures of -25°C. Since lying deer conserve heat energy, the A and C deer in this study probably were not experiencing thermal stress. Trails Most trails were close to the fences. Trails beside fences with deer on the opposite side were used more frequently than trails on the periphery of the pen complex. This again implies that the presence of other deer may be important in determining behavioral responses. The N deer maintained a trail through the center of their pens, where snow depth was often less, instead of along the outside edges. Deer in N pens made more tracks off trails than A or C deer, probably because the average sinking depth was less in the N pens. A positive correlation between trail length and deer sink- ing depth indicates that the track-trail relationship proposed by Severinghaus (1953) may be a good mea- sure of mobility. Under pen conditions increased trail length resulted from confinement to trails. Confine- ment to trails occurred when deer sinking depth reached about 30 cm. Total snow thickness was 46 to 51 cm at this time. Telfer (1970) concluded a snow thickness of 50 cm caused confinement. General Activity The observed greater activity at sunset than at sun- rise was similar to that recorded in summer by Tibbs (1967). Since all night observations were made when moonlight was bright enough to allow observations, and Anderson (1959) found nocturnal activity to be greatest on bright nights, the night activity recorded in this study possibly represented a maximum value. Colder temperatures and a less favorable radiant energy environment at night may result in a reduction of nocturnal activity. Daily activity patterns were similar to those des- cribed for unconfined deer by Marchington (1968) and Tibbs (1967). Inactivity at low temperatures is a well known behavior pattern which conserves heat. Activity of the experimental deer was significantly greater in each of four progressively warmer temperature ranges. Beh- rend (1966) found activity levels during January and February correlated with maximum daily tempera- tures. Halloran (1943) reported decreased road counts in Texas during cold weather. 1983 Precipitation caused no evident effects on activity, although deer were observed on several occasions to get up and move about as snow began to fall. Behrend (1966) noticed increased activity before and during the first stages of a storm, but prolonged storms caused inactivity. Allen (1970) also recorded reduced activity during heavy snowfall. Marchington (1968) observed increasing activity prior to a front of bad weather and Anderson (1959), found Columbian Black-tailed Deer (Odocoileus hemionus columbianus) to be more active on cloudy and stormy nights in summer. In this study, greater activity when skies were overcast, when barometric pressure was falling, and with increasing humidity vapor pressure indicates greater activity preceding a storm. A relative increase in activity before storms may be an inherent behavior trait rather than a directed movement to shelter. Increased feed- ing before stormy periods has survival value if inactiv- ity, and consequently less feeding, occurs during storms. No reduction in activity due to snowfall was shown in this study, but it may occur in lengthy storms as previously reported. Under natural conditions more energy would probably be required during and after storms in order to break trails and move greater distances from shelter to feeding areas. Feeding Activity The effects of weather on feeding activity were sim- _ ilar to the effects on general activity. However, unlike general activity, the effects of the direction of baro- metric pressure and of cloud cover were significant for A and C deer tested separately. Possibly the deer in the clear-cut pens were more influenced by changing weather conditions. Since a greater feeding activity when barometric pressure was falling was recorded under clear as well as overcast skies, the effect appears to be a result of falling pressure rather than a com- bined effect of pressure and cloud cover. Conclusions In summary, it was shown that if fawns have access to an adequate food supply, they can withstand severe winter weather even in the absence of shelter. While adequate shelter may reduce the maintenance requirements of deer so that winter survival is more likely, in this study shelter did not cause a reduction in the amount of food consumed. This indicates that increasing attention should be given to the quality of food available on winter range in relation to over winter carrying potential and to physical condition of fawns in the autumn prior to the demands of winter (Verme and Ozoga 1980). Energy requirements for maintenance of free-ranging deer during winter may be 2-3X above basal metabolic values determined from laboratory studies. Temperature and solar radiation have value as pre- GILBERT AND BATEMAN: WHITE-TAILED DEER FAWNS 399 dictors of winter energy requirements of deer on northern ranges. Although food consumption is reduced in cold weather, the energy drain may be sufficient to necessitate ingestion of more food once temperatures moderate. The presence of other deer may prove very impor- tant for the winter survival of White-tailed Deer fawns. Although other deer may be an advantage for fawns for purely mechanical reasons (e.g. breaking trail), the effects on behavior and physiology may also be important. Acknowledgment The research was financially supported by Federal Aid to Wildlife Restoration, Maine, Pittman- Robertson Project W67-R. Literature Cited Allen T. J. 1970. Telemetry studies of deer movements and habitat utilization at Acadia National Park. M.Sc. thesis, University of Maine, Orono, Maine. 153 pp. Anderson, A. E., and D. E. Medin. 1965. Two condition indices of the Cache La Poudre Mule Deer herd and their application to management. Colorado Division of Game, Fish and Parks. Game Information Leaflet No. 23. Anderson, C. F. 1959. Nocturnal activities of the Colum- bian Black-tailed Deer, (Odocoileus hemionus columbia- nus, Richardson), affecting spotlight census results in the Oregon Coast Range. M.Sc. thesis, Oregon State College, Corvallis, Oregon. 86 pp. Association of Official Agricultural Chemists. 1965. Official methods of analysis of the Association of Official Agricul- tural Chemists. Tenth edition. Washington, D. C. 957 pp. Baker, D. L., D. E. Johnson, L. H. Carpenter, O. C. Wal- Imo, and R. B. Gill. 1979. Energy requirements of Mule Deer fawns in winter. Journal of Wildlife Management 43: 162-169. Bateman, M. C. 1972. Winter Shelter: some effects on the behavior and physiology of penned White-tailed Deer. M.Sc. thesis, University of Maine, Orono, Maine. 108 pp. Behrend, D. F. 1966. Behavior of White-tailed Deer in an Adirondack forest. Ph.D. thesis, Syracuse University, Syracuse, New York. 206 pp. Cook, D. B., and W. J. Hamilton, Jr. 1942. Winter habits of White-tailed Deer in central New York. Journal of Wildlife Management 6: 287-291. Day, B. W., Jr. 1963. Winter behavior of White-tailed Deer in north-central Maine. M.Sc. thesis, University of Maine, Orono, Maine. 151 pp. Fobes, C. B. 1946. Climatic divisions of Maine. Maine Technical Experimental Station Bulletin No. 40. Orono, Maine. 44 pp. Gill, J. D. 1966. What's a deer-yard in Maine? The North- ern Logger, February: 14-15, 28. , Halloran, A. A. 1943. Management of deer and cattle on the Aransas National Wildlife Refuge, Texas. Journal of Wildlife Management. 7: 203-216. Hugie, R. D. 1973. A winter study of deer mobility in west- central Maine. M.Sc. thesis, University of Maine, Orono, Maine. 78 pp. 400 Kramer, A. 1971. Notes on the winter ecology of the Mule and White-tailed Deer in the Cypress Hills, Alberta. Can- adian Field-Naturalist 84: 141-145. Lemmon, P. E. 1956. A spherical densiometer for estimat- ing forest overstory density. Forestry Science 2: 314-320. Lull, H. E., and I. C. Relgner. 1967. Radiation measure- ments by various instruments in the open and in the forest. United States Forest Service, Research Paper NE-84. 21 Pp. Marchington, R. L. 1968. Telemetric study of White-tailed Deer movement — ecology and ethology in the southeast. Ph.D. thesis, Auburn University, Auburn, Alabama. 139 Pp. McEwen, L.C., C.E. French, N. D. Magruder, R. W. Swift, and R. H. Ingram. 1957. Nutrient requirements of the White-tailed Deer. Transactions of the North Ameri- can Wildlife and Natural Resources Conference 22: 119-132. Moen, A.N. 1968a. Surface temperature and radiant heat loss from White-tailed Deer. Journal of Wildlife Man- agement 32: 338-344. Moen, A.N. 1968b. Thermal energy exchange of a birch tree and a spruce tree at night. Ecology 49: 145-147. Moen, A. N. 1968c. Energy exchange of White-tailed Deer, western Minnesota. Ecology 49: 666-682. Moen, A. N. 1976. Energy conservation by White-tailed Deer in the winter. Ecology 56: 192-198. Ozoga, J. J. 1968. Variations in microclimate in a conifer swamp deeryard in northern Michigan. Journal of Wild- life Management 32: 574-585. Ozoga, J. J., and L. J. Verme. 1970. Winter feeding patt- erns of penned White-tailed Deer. Journal of Wildlife Management 34: 431-439. Robinson, W. L. 1959. A study of winter shelter require- ments of penned White-tailed deer. M.Sc. Thesis, Univer- sity of Maine, Orono, Maine. 161 pp. Severinghaus, C. W. 1953. Springtime in New York — another angle. New York State Conservationist 7: 2-4. Severinghaus, C. W., and E. L. Cheatum. 1956. Life and times of the White-tailed Deer. Pp. 57-186 /n The Deer of THE CANADIAN FIELD-NATURALIST Vol. 97 North America: their history and management. Edited by W. P. Taylor. Stackpole Co., Harrisburg, Pennsylvania. 668 pp. Silver, H., N. F. Colovos, J. B. Holter, and H. H. Hayes. 1969. Fasting metabolism of White-tailed Deer. Journal of Wildlife Management 28: 445-448. Steele, R.G. D., and J. H. Torrie. 1960. Principles and procedures of statistics. McGraw-Hill Book Co. Inc., New York, New York. 481 pp. Tarchinski, R. 1969. The effects of temperature, humidity and wind velocity on the nutrition requirements of White- tailed Deer fawns. M.Sc. Thesis, Michigan Technological University, Houghton, Michigan. 51 pp. Telfer, E.S. 1970. Winter habitat selection by Moose and White-tailed Deer. Journal of Wildlife Management 34: 553-559. Thompson, C. B., J. B. Holter, H. H. Hayes, H. Silver, and W. E. Urban, Jr. 1973. Nutrition of White-tailed Deer. I. Energy requirements of fawns. Journal of Wildlife Man- agement 37: 301-311. Tibbs, A. E. 1967. Summer behavior of White-tailed Deer and the effects of weather. M.Sc. thesis, Pennsylvania State University, University Park, Pennsylvania. 93 pp. Verme, L. J. 1965. Swamp conifer deeryards in northern Michigan: Their ecology and management. Journal of Forestry 63: 523-529. Verme, L. J., and J. J. Ozoga. 1980. Influence of protein- energy intake on deer fawns in autumn. Journal of Wild- life Management 44: 305-314. Webb, W.N. 1948. Environmental analysis of winter deer range. Transactions of the North American Wildlife Con- ference 13: 442-450. Westover, A. J. 1971. The use of hemlock-hardwood win- ter yards by White-tailed Deer in northern Michigan. Occasional Papers of Huron Mountain Wildlife Founda- tion. 59 pp. Received 10 June 1982 Accepted 21 July 1983 Breeding Ecology of the Horned Grebe, Podiceps auritus, in Southwestern Manitoba ROBERT S. FERGUSON! and SPENCER G. SEALY Department of Zoology, University of Manitoba, Winnipeg, Manitoba R3T 2N2 1Present address: Wildlife Management Division, Department of Renewable Resources, Government of the Northwest Territories, Yellowknife, N.W.T. X1A 2L9 Ferguson, Robert S., and Spencer G. Sealy. 1983. Breeding ecology of the Horned Grebe, Podiceps auritus, in southwestern Manitoba. Canadian Field-Naturalist 97(4): 401-408. Nesting Horned Grebes, Podiceps auritus, studied in 1974 and 1975 in pothole marsh habitats near Minnedosa, Maniteba, selected permanent, open ponds with peripheral stands of emergent vegetation. This reflected their dependence on open water feeding areas and emergent nesting cover, and possibly alleviated competition with Pied-billed Grebes, Podilymbus podiceps. Annual variation in initiation of first clutches was related to spring weather conditions. Clutch size varied from 3 to 8 eggs and declined significantly as the nesting season progressed. Nest losses were high, with predators accounting for at least 54% of the total egg loss. Following nesting failure, pairs made up to three renesting attempts. Asynchronous hatching resulted in considerable age and size differences between first and last hatched chicks of a brood. Brood division, parental aggression, and selective feeding of smaller chicks appeared to be adaptations for reducing the competitive disparities among chicks that resulted from asynchronous hatching. The reproductive biology of Horned Grebes in southwestern Manitoba suggests ecological parallels with grebes of the genera Rollandia and Tachybaptus which are also adapted to nesting in unstable marsh habitats. Key Words: Horned Grebes, Podiceps auritus, breeding ecology, pothole marshes, Manitoba. Grebes and other waterbirds nesting in pothole marshes of the Canadian prairies and parklands are subjected to rapidly changing environmental condi- tions. The availability and quality of nesting cover in the form of emergent vegetation vary widely from year to year in response to fluctuating water levels (Weller and Spatcher 1965; Kiel et al. 1972). As a result of their anatomical and behavioural adaptations to aquatic habitats, grebes appear less tolerant of envir- onmental change than relatively non-specialized spe- cies. Some marsh-nesting icterids, for example, exploit alternative habitats when marshes are unfa- vourable for nesting (Robertson 1972; Voigts 1973). Despite its specific nesting requirements, the Horned Grebe (Podiceps auritus) is a prominent species in the avian community of pothole marshes in southwestern Manitoba. In 1974 we initiated a breeding study of the Horned Grebe near Minnedosa, Manitoba, in order to determine its ecological adaptations for nesting in unstable marsh habitats. A holarctic species, the Horned Grebe breeds throughout much of northern Europe, northern Asia, and northwestern North America (Fjeldsa 1973a). No detailed nesting studies of any North American popu- lations (P. a. cornutus) have been published. The reports of Bent (1919), Dubois (1919), Munro (1941), and Palmer (1962) were based on casual observations of a few nesting pairs from widely scattered localities and were generally incomplete. Storer (1969) des- cribed and analyzed the behavioural repertoire prior to egg-laying, and Faaborg (1976) and Sugden (1977) studied aspects of habitat selection in North Dakota and Saskatchewan, respectively. The breeding habits of P. a. auritus and P. a. arcticus are better known (Clase et al. 1960; Onno 1960; Hogstrom 1970; Fjeldsa ISVS, EC; Gl); Study Area and Methods Horned Grebes were studied in an area of pothole marshes (75 km2) south of Minnedosa, Manitoba (50° 15’N; 99°50’W). That area lies within the aspen parkland and has been described in detail in numerous papers. Descriptions of topography, hydrology, soils, vegetation, and climate of the Minnedosa pothole district are provided by Ehrlich et al. (1957), Bird (1961), Kiel et al. (1972), and Adams and Gentle (1978). Pothole marshes were classified into four types on the basis of water permanency and the distribution of wetland vegetation according to Millar’s (1976) classification. The four wetland types are Wet Mea- dow, Shallow Marsh, Emergent Deep Marsh, and Open Water Marsh. Field studies were conducted from 27 April through 27 August 1974, and from 17 April through 30 Sep- tember 1975. Each nest found was visited at least once every three days. Eggs were measured (width and length) to nearest 0.1 mm, and numbered inconspicu- ously with a waterproof marker in the order of laying. Behavioural aspects of territoriality, incubation, par- ental care, and chick development were studied froma 40] 402 blind or parked car. During incubation, adult birds were captured with a gill net set in the water between the nest site and open water (Ferguson 1980). Forty- three adults and seven young were captured, sexed (adults only), measured, and banded. In May and June, adults were sexed on the basis of sexual differ- ences in onset of the prebasic molt, length of the nuptial plumes, and colouration of the neck and flanks (see Storer 1969; Fjeldsa 1973a, d). In 1975, 12 females and 9 males were individually colour-marked with feather dye appled to the yellow head plumes and white secondaries. The head marking lasted until the plumes were shed during the prebasic molt, usually in early July. Pothole sizes were determined by planimetry from aerial photographs supplied by the National Air Photo Library, Department of Energy, Mines and Resources, Ottawa. Meteorological data for Minne- dosa were obtained from the Atmospheric Environ- ment Service, Winnipeg. Means are expressed as + | standard deviation. Results Territorial Establishment and Pothole Selection Horned Grebes were first observed on the study area on 29 April 1974 and 30 April 1975; the peak influx occurred during the first week of May. Estab- lishment of territories commenced as early as 7 May and continued throughout the month of May. Nesting Horned Grebe pairs showed a distinct preference for Open Water Marshes (Table 1). Potholes used for nesting varied in size from 0.1 to 8.4 ha (mean 1.2 + 1.3 ha; n = 65). Only three potholes supported more than one nesting pair of Horned Grebes. Two potholes of 1.1 haeach supported two pairs and one of 2.6 ha supported four pairs. Each pair defended a Type A territory (Nice 1943) which consisted of an area of open water and associated shoreline. In the absence of conspecifics, a nesting pair used all areas of a pothole for feeding and brood-rearing. THE CANADIAN FIELD-NATURALIST Vol. 97 Nest-Site Selection All nests located on the study area were anchored to emergent vegetation (Table 2). Nests built before 25 May were usually anchored to residual vegetation from the previous growing season. Unless flattened in winter by snow, residual vegetation such as bulrush and cattail provided sufficient nesting cover early in the season before new growth had appeared above the water surface. Water depth, measured at 119 nests at the time of clutch initiation, averaged 39.2 + 11.7 cm. Only 2 nests were located in water < 20 cm deep. A minimum water depth of 20 cm seems necessary to allow the grebes easy underwater access to and froma nest site. Nests located in shallower water may become stranded on dry land as water levels in many potholes fell by as much as 15 to 20 cm during a month of hot, dry weather. Egg- Laying Eggs were laid over a period of 53 days in 1974 (22 May to 13 July) and 57 days in 1975 (17 May to 12 July). Initiation of laying of first clutches (n = 70) ranged from 17 May to II June (Figure 1). In 1974 (n = 34) and 1975 (n = 36), 50% of first clutches had been initiated by | June and 25 May, respectively. Within a clutch, the interval between laying of con- secutive eggs varied from | to 3 days, but 3-day inter- vals were relatively uncommon. Total egg-laying periods for the three most frequent clutch sizes were 7-8 days for clutches of five eggs, 8-9 days for clutches of six, and 9-11 days for clutches of seven. Following nesting failure, most pairs renested at least once. Of 40 pairs whose first nests were des- troyed, 68% renested on the same territory. Although many birds were not individually marked, we are con- fident that these renests involved the same pairs. The promptness with which lost clutches were replaced and the distribution of nesting pairs (one per pond) support this assumption. Only 6% of 73 renests were located on previously unsuccessful sites. After TABLE |. Differential habitat selection by nesting Horned Grebes. Relative Availability of Potholes Used by Nesting Horned Grebes Pothole Type Pothole Types (%) Shallow Marsh 17 Emergent Deep Marsh 28 Open Water Marsh 60 Totals 100 x2) = 2357: \p < 0.005. Expected Frequency Observed Frequency 11 | 15 6 39 58 65 65 Note: Relative availability of pothole types was determined by classifying all wetlands within a central area (8.4 km?) of the study block. Wet Meadow wetlands were excluded from the analysis owing to their ephemeral water supply and unsuitability as nesting habitat for grebes. 1983 TABLE 2. Frequency distribution of Horned Grebe nests according to the dominant species of emergent cover. Number of Species Nests (%) Whitetop (Scolochloa festucacea) 41 (32) Bulrushes (Scirpus spp.) 39 (30) Cattails (Typha spp.) 29 (22) Water Parsnip (Sium suave) 8 (6) Goosefoots (Chenopodium spp.) 355 (2) Willows (Salix spp.) 2D (2) Sedges (Carex spp.) 1 (1) Water Crowfoots (Ranunculus spp.) 1 (1) Unidentified 5 (4) Total 129 (100) repeated failures, six pairs renested twice and another siX pairs renested three times, all within their original territories. Elapsed time between nesting failure and clutch initiation varied from < 2 to 9 days. When nesting failure occurred during laying, 22 of 28 replacement clutches were initiated within four days. For nests that failed after laying was completed, replacement clutches were rarely initiated within four days (1 of 8 nests). Egg-laying in the remaining seven nests began 5-9 days following nesting failure. 20 | 15 n ® <= 12) = oO o 10 o 2 E =] z 5 } June FIGURE 1. FERGUSON AND SEALY: BREEDING ECOLOGY OF THE HORNED GREBE 403 Two pairs laid a second clutch after their first clutch had hatched (Figure 1). In 1974, a pair initiated a second clutch 10 days after the last egg of the previous clutch had hatched. The same nesting platform was used for the second clutch. In 1975, a pair initiated a second clutch 16 days after the previous clutch had hatched. That pair, in which the female was individu- ally marked, built a new nesting platform 3.5 m from the first nest. Chicks from the first broods were still being attended by their parents when the second clutches were initiated. Mean dimensions of 591 eggs were 44.3 mm (range = 39.0-48.4) x 30.2 mm (range = 27.8-33.0). Clutches varied from 3 to 8 eggs (mean 5.9 + 1.2 eggs; n= 79). Clutches of 5, 6, and 7 eggs were most com- mon, comprising 81% of all nests examined. Variation in clutch size in relation to date of laying is shown in Figure 2. The regression lines did not differ signifi- cantly between years (F = 0.14; 1, 75df; NS; F = 2.20; 1, 76df; NS), thus the data were pooled. A significant correlation (r = -0.70) between clutch size and laying date was obtained (t = 8.60; P< 0.001). Incubation Incubation began as soonas the first egg of a clutch was laid. The incubation period for 17 marked eggs First nests (n=70) al Renests (n= 71!) BY Second nests (n=2) | i 15 15 July Seasonal pattern of egg-laying as revealed by dates of laying of the first egg in each clutch. (Each bar represents a 4-day interval.) 404 size Clutch 20 1 10 May Clutch THE CANADIAN FIELD-NATURALIST Vol. 97 e First nests o Renests x Second nests 20 1 10 July June initiation FIGURE 2. Variation in clutch size in relation to date of laying. from 10 clutches was 23-24 days, but in 2 nests hatch- ing did not begin until the first eggs had been incu- bated for 28 and 34 days. Attentive periods of males and females from 8 pairs ranged from 17 to 102 min (mean 45.3 + 20.2 min; n= 33). During egg-laying incubation was shared almost equally (55% by females, 45% by males), but after egg-laying females spent considerably more time on the nest (63%) than did males (37%). Hatching During both years, first eggs in 47 nests hatched between 10 June and 2 August (Figure 3). Clutches hatching after 5 July were either replacement or second clutches. Generally, the interval between hatching of consecutive eggs reflected the laying inter- val; most chicks hatched at 1-2 day intervals. Hatch- ing periods for the three most frequent clutches were 7-8 days for clutches of five eggs, 8-9 days for clutches of six, and 9-10 days for clutches of seven. Hatching success was low in both years (30.3% of 637 eggs). Predators destroyed 38% of eggs laid, waves generated by high winds washed out 12%, and adults deserted 2%. Eggs that disappeared from nests (18%) were not attributed to a specific cause, but predators were likely responsible for most of those losses as well. The appearance of eggshell remains (see Rearden 1951) indicated that Raccoons (Procyon lotor) were responsible for most of the predation. Most destroyed eggs were characterized by an opening in the large end of the shell which suggested that the end had been bitten off. The Raccoon’s activity as a predator on over-water nests in the Minnedosa district 1s well doc- umented (Kiel et al. 1972; Cowan 1973). Canvasbacks (Aythya valisineria) nesting on the study area in 1974 and 1975 also suffered high egg losses to Raccoons (J. Serie, personal communication). Potential avian pre- dators of eggs included American Crow (Corvus bra- chyrhynchos), Black-billed Magpie (Pica pica) and American Coot (Fulica americana) (see Burger 1973). Chick Development and Parental Care Shortly after hatching, chicks climbed onto the incubating parent’s back where they were brooded among the feathers between the adult’s slightly raised wings. During their first three days, chicks were brooded almost constantly although they spent brief periods in the water to drink, defecate, or regurgitate a 1983 w (eb) at ey oO ~ = (@) — [o) = cab) QO £ =) FZ 15 1 June FERGUSON AND SEALY: BREEDING ECOLOGY OF THE HORNED GREBE 405 July August FIGURE 3. Distribution of hatching of first eggs in Horned Grebe clutches. (Each bar represents a 4-day interval.) pellet. Pellet-casting was observed six times by young ranging in age from 3 days to 7 weeks. Chicks readily swallowed feathers offered by adults within 24 hours after hatching, an indication of the important role of feathers in the grebes’ digestive tract (Storer 1969). At 4-5 days of age chicks spent longer periods in the water, and by 8-9 days they no longer were brooded during the day. This corresponds to the age at which preening and waterproofing had developed in a cap- tive chick (McAllister 1963). At 12-13 days, chicks began to capture their own prey. At 19-21 days, most chicks were independent of parental care although the younger brood members frequently were fed until they were 21-24 days old. Seventy-five percent of 51 known-aged young fledged between 45 and 50 days of age. The youngest chicks observed in sustained flight above the water were 41-42 days old. In at least 11 of 16 broods, the older chicks departed from their natal pond before the younger siblings. In 1974 and 1975, fledging dates ranged from | August to 15 September. From an average brood size of 4.3 chicks at hatching (n = 38 broods), 64% survived to fledging. Chick mortality was greatest in the first few days after hatching; only 72% survived the first 10 days. None of the 3 chicks that hatched from second clutches survived to fledg- ing. Causes of post-hatching mortality were generally unknown because few dead chicks were found. Chick survival to fledging was independent of initial brood size (Table 3). Participation of adults in parental care was influ- enced by brood size. In six of eight families with < 4 chicks, one parent abandoned the territory within the week following the brooding period, leaving its mate to care for their young. In four families where the adults were individually marked, the female remained with the young in two cases, the male in the other two cases. In six of seven families with four or more chicks, the young were divided between the parents. Brood division also occurred in one family of three chicks and one family of two chicks. Brood division is similar to that described for the Great Crested Grebe, Podi- ceps cristatus (Simmons 1974). On the basis of size differences among chicks, they did not appear to be divided between the parents according to age. Brood division did not always involve the entire brood. In families of five or more young, the age difference between first and last hatched chicks was often as great as 8 or 9 days and the oldest one or two chicks were nearly independent when the brooding period TABLE 3. Chick Survival to fledging in relation to initial brood size. Initial Number of Brood Size Broods 2-3 6 4-5 17 6-7 10 Totals 33 Number of Young Proportion Surviving Fledged/ Hatched* to Fledging 10/14 0.71 48/78 (0.62 44/63 0.70 102/155 0).66** *y2= 0.44; p>0.05. **Proportion for all broods combined. 406 ended. In those families, the younger chicks were divided between the parents, while the older chicks foraged alone or with other independent siblings. During the brooding period, older chicks had a competitive advantage over their younger siblings because of their larger size and greater mobility. Chicks that were being warmed on a parent’s back simply waited for the second parent to provide them with food. Older chicks that did not require brooding developed a conspicuous begging display and usually rushed toward the feeding parent. That display was similar to one described for the Great Crested Grebe (Simmons 1968) and involved loud begging calls and splashing movements of the feet and legs. On many occasions, a parent refused to feed an older brood member that was exhibiting high-intensity begging displays but offered the food item to a younger, brooded chick. When older chicks (16-18 days of age) approached their parents for food, they were fre- quently chased away in a threatening manner. Such aggression was most pronounced when a parent was attempting to feed a smaller chick. Parental aggres- sion was observed only when a family, or family unit in the case of divided broods, comprised two or more chicks. Discussion Pothole marshes in the aspen parkland are highly productive (Barica 1974; Driver and Peden 1977) and perhaps support the highest breeding densities of the Horned Grebe in North America (Sugden 1977). At Minnedosa, nesting pairs showed a distinct preference for permanent, open ponds with peripheral stands of emergent vegetation. Selection of that habitat reflects the grebes’ specialized foraging techniques and their dependence on open water feeding areas (Wetmore 1924). Only Emergent Deep Marshes and Open Water Marshes retain water long enough to permit a pair of Horned Grebes to rear a brood to fledging. Emergent Deep Marshes, with good interspersion of open water and emergent vegetation, offer a wide choice of pro- tected nesting sites, but competition with Pied-billed Grebes (Podilymbus podiceps) may restrict the Horned Grebe’s use of that habitat type. Only three of 65 marshes that supported nesting Horned Grebes were used by Pied-billed Grebes. Similar nesting dis- tributions were obtained for Pied-billed Grebes in Minnedosa (Sealy 1978) and for Horned and Pied- billed Grebes on small ponds in North Dakota (Faa- borg 1976). Faaborg (1976) noted that Pied-billed Grebes selected more heavily vegetated ponds than Horned Grebes, and suggested that segregation by habitat allowed those ecologically similar species to co-exist in the glaciated prairie region of North America. THE CANADIAN FIELD-NATURALIST Vol. 97 Horned Grebes depended on emergent vegetation for protection and support of over-water nesting plat- forms. A nest site also must meet the minimum depth requirement of 20cm. During dry years when receding water levels reduce the availability of emergent nest- ing cover, grebes select exposed sites far from shore- line vegetation and nest losses to wave action may be high (see also Glover 1953; Nuechterlein 1975; Faa- borg 1976). Water levels at Minnedosa were high in 1974 and 1975, and peripheral stands of tall emergents provided an abundance of nesting sites well protected from wind and waves; Horned Grebes sustained low egg losses to wave action. An important determinant of nesting success is the water level through its control- ling influence on the distribution of emergent vegeta- tion. Even in years when habitat conditions are good, egg losses to predators can be high. In unpredictable habitats where the probability of producing young is uncertain, species which maxim- ize their reproductive potential are favoured (Cody 1971). At Minnedosa, Horned Grebes appeared to nest as early as weather conditions permitted. Earlier clutch initiation in 1975 than in 1974 was attributed to weather differences during the pre-laying period (1-20 May). Precipitation (5.5 cm, in the form of rain and snow) fell on 15 days during that period in 1974, but on only 6 days in 1975 (1.3 cm, rain only). Mean maximum daily temperatures during the same period were 8.9° Cand 16.8°C in 1974 and 1975, respectively. Pairs that nest early potentially have more time for the laying of replacement clutches. After repeated fail- ures, some pairs made as many as four nesting attempts. Within wide limits, any increase in the number of nesting attempts by a pair augments its chances of producing young. In North Dakota, the first territorial pair of Horned Grebes to occupy a large pond with a limited number of nesting sites usually selected the most sheltered site and nested successfully, whereas late-arriving pairs were forced to select more exposed sites and sustained higher nest losses to wind and wave action (Faaborg 1976). We were unable to test this hypothesis at Minnedosa because so few ponds supported more than one nest- ing pair. Two or more nesting pairs occurred only on ponds ofarea > | ha, as found also by Fjeldsa (1973) for pairs nesting on eutrophic ponds in Iceland. A third advantage of early nesting is that early clutches are potentially more productive than later ones. Clutch size decreased as the nesting season pro- gressed and was much reduced after 15 June. The reproductive potential is also enhanced by brood div- ision and parental aggression. Many aspects of paren- tal care appear to be adaptations for reducing the competitive disparities among chicks that result from asynchronous hatching. During the brooding period, 1983 parents ensured that older chicks did not receive all the food by selectively feeding the younger chicks. When the older chicks were nearly self-sufficient, par- ents directed their attention to the younger brood members and fed them until they were fully indepen- dent. That prolonged period of parental care for younger chicks may compensate for any competitive disadvantages that they experienced earlier in the brood-rearing period. Indirect evidence that competi- tive disparities were effectively reduced was provided by examining chick survival in relation to initial brood size. Age differences and presumably competi- tive differences among chicks are more pronounced as brood size increases. At Minnedosa, chick survival to fledging was independent of initial brood size. The family structure of Minnedosa Horned Grebes sug- gests ecological parallels with grebes of the genera Rollandia and Tachybaptus which are adapted to nesting in temporarily flooded habitats by raising as many young as possible in a brief period (J. Fjeldsa, personal communication). Climatic variability, fluc- tuating water levels, and frequent short-term changes in the availability of emergent nesting cover attest to the unpredictable nature of prairie pothole marshes as breeding habitats for aquatic birds (Eisenlohr 1969; Kiel et al. 1972). In contrast, Icelandic Horned Grebes are adapted to using more predictable permanent waters such as Lake Myvatn, and appear to maintain an optimal population level by moderate reproductive efforts through density-dependent mechanisms (Fjeldsa 1973c). Acknowledgments We are indebted to the landowners of the Minne- dosa community for their co-operation and permis- sion to conduct field studies on their properties. Jon Fjeldsa and Robert W. Storer kindly reviewed an earlier draft of the manuscript. Field work was funded by Natural Sciences and Engineering Research Coun- cil of Canada, Northern Studies Committee of the Department of Indian and Northern Affairs, Gulf Oil Canada Limited, Manitoba Department of Renewa- ble Resources and Transportation Services, and National Audubon Society. Literature Cited Adams, G. D., and G. C. Gentle. 1978. Spatial changes in waterfowl habitat, 1964-74, on two land types in the Manitoba Newdale Plain. Canadian Wildlife Service Occasional Paper Number 38. Barica, J. 1974. Extreme fluctuations in water quality of eutrophic fish kill lakes: Effect of sediment mixing. Water Research 8: 881-888. Bent, A. C. 1919. Life histories of North American diving birds. United States National Museum Bulletin Number 107. Dover reprint (1963). FERGUSON AND SEALY: BREEDING ECOLOGY OF THE HORNED GREBE 407 Bird, R. D. 1961. Ecology of the aspen parkland of western Canada in relation to land use. Canada Department of Agriculture Contribution Number 27. Burger, J. 1973. Competition between American Coots and Franklin’s Gulls for nest sites and egg predation by the coots. Wilson Bulletin 84: 449-451. Clase, H. J., F. Cooke, T. A. Hill, and W. J. Roff. 1960. A survey of the Slavonian Grebe at Myvatn, Iceland. Bird Study 7: 76-81. Cody, M. L. 1971. Ecological aspects of reproduction. Pp. 461-512 in Avian Biology, Volume |. Edited by D.S. Farner and J. R. King. Academic Press, New York. Cowan, W. F. 1973. Ecology and life history of the Rac- coon (Procyon lotor hirtus) in the northern part of its range. Ph.D. thesis, University of North Dakota, Grand Forks. Driver, E. A., and D. G. Peden. 1977. The chemistry of surface water in prairie ponds. Hydrobiologia 53: 33-48. DuBois, A. 1919. An experience with Horned Grebes. Auk 36: 170-180. Ehrlich, W. A., L. E. Pratt,and E. A. Poyser. 1957. Report of reconnaissance soil survey of Rossburn and Virden map sheet areas. Manitoba Soil Survey Report Number 6. Eisenlohr, W. S., Jr. 1969. Hydrology of small water areas inthe prairie pothole region. Pp. 35-39 in Canadian Wild- life Service Report Series Number 6. Faaborg, J. 1976. Habitat selection and territorial behavior of the small grebes of North Dakota. Wilson Bulletin 88: 390-399. Ferguson, R. S. 1980. A technique for live-trapping nesting Horned Grebes. Journal of Field Ornithology 51: 179-180. Fjeldsa, J. 1973a. Distribution and geographic variation of the Horned Grebe Podiceps auritus (Linnaeus 1758). Ornis Scandinavica 4: 55-86. Fjeldsa, J. 1973b. Feeding and habitat selection of the Horned Grebe, Podiceps auritus (Aves), in the breeding season. Videnskabelige Meddelelser fra Dansk Naturhis- torisk Forening 136: 57-95. Fjeldsa, J. 1973c. Territory and the regulation of popula- tion density and recruitment in the Horned Grebe Podi- ceps auritus arcticus (Boje 1822). Videnskabelige Medde- lelser fra Dansk Naturhistorisk Forening 136: 117-189. Fjeldsa, J. 1973d. Antagonistic and heterosexual behav- lour of the Horned Grebe, Podiceps auritus. Sterna 12: 161-217. Glover, F. A. 1953. Nesting ecology of the Pied-billed Grebe in northeastern Iowa. Wilson Bulletin 65: 32-39. Hogstrom, S. 1970. Svarthakedoppingen Podiceps auritus pa Gotland. Var Fagelvarld 29: 60-66. Kiel, W. H., Jr.. A. S. Hawkins, and N. G. Perret. 1972. Waterfowl habitat trends in the aspen parkland of Manit- oba. Canadian Wildlife Service Report Series Number 18. McAllister, N. M. 1963. Ontogeny of behaviour in five spe- cies of grebes. Ph.D. thesis, University of British Colum- bia, Vancouver. Millar, J. B. 1976. Wetland classification in western Can- ada. Canadian Wildlife Service Report Series Number 37. Munro, J. A. 1941. The grebes: Studies of waterfowl in British Columbia. British Columbia Provincial Museum Occasional Papers Number 3. 408 Nice, M. M. 1943. Studies in the life history of the Song Sparrow. Volume 2. Transactions of the Linnaean Society of New York Number 6. Neuchterlein, G. L. 1975. Nesting ecology of Western Grebes on the Delta Marsh, Manitoba. M.Sc. thesis, Colorado State University, Fort Collins. Onno, S. 1960. Zur okologie der Lappentaucher (Podiceps cristatus, grisegema and auritus) in Estland. Pp. 577-582 in Proceedings of the 12th International Ornithological Congress. Palmer, R. S. Editor. 1962. Handbook of North American Birds. Volume |. Yale University Press, New Haven. Rearden, J. D. 1951. Identification of waterfowl nest pre- dators. Journal of Wildlife Management 15: 386-395. Robertson, R. J. 1972. Optimal niche space of the Red- winged Blackbird (Agelaius phoeniceus). 1. Nesting suc- cess in marsh and upland habitat. Canadian Journal of Zoology 50: 247-263. Sealy, S. G. 1978. Clutch size and nest placement of the Pied-billed Grebe in Manitoba. Wilson Bulletin 90: 301-302. Simmons, K. E. L. 1968. Some observations on families of Great-crested Grebes. Bristol Ornithology 1: 21-26. THE CANADIAN FIELD-NATURALIST Vol. 97 Simmons, K. E. L. 1974. Adaptations in the reproductive biology of the Great-crested Grebe. British Birds 67: 413-437. Storer, R. W. 1969. The behavior of the Horned Grebe in spring. Condor 71: 180-205. Sugden, L. G. 1977. Horned Grebe breeding habitat in Saskatchewan parklands. Canadian Field-Naturalist 91: 372-376. Voigts, D. K. 1973. Food niche overlap of two lowa marsh icterids. Condor 75: 392-399. Weller, M. W., and C. E.Spatcher. 1965. Role of habitat in the distribution and abundance of marsh birds. lowa State University of Science and Technology Agricultural and Home Economics Experiment Station Special Report Number 43. Wetmore, A. 1924. Food and economic relations of North American grebes. United States Department of Agricul- ture Department Bulletin Number 1196. Received 19 April 1982 Accepted 8 July 1983 Variability in Size and Age at Sexual Maturity of Witch Flounder, Glyptocephalus cynoglossus, in the Canadian Maritimes Region of the Northwest Atlantic Ocean TERRY D. BEACHAM Department of Fisheries and Oceans, Fisheries Research Branch, Pacific Biological Station, Nanaimo, British Columbia VOR S5K6 Beacham, Terry D. 1983. Variability in size and age at sexual maturity of Witch Flounder, G/yptocephalus cynoglossus, in the Canadian Maritimes region of the northwest Atlantic Ocean. Canadian Field-Naturalist 97(4): 409-422. The variability in length and age at which 50% of individuals were mature was investigated for Witch Flounder (G/yptocepha- lus cynoglossus) in the Northwest Atlantic Ocean from 1959 to 1979. Median length and age at maturity generally declined through time, and males matured at smaller sizes and younger ages than did females. During 1975-1979, median length at maturity for males ranged from 29.2 cm (NAFO Division 4W) to 33.0 cm (Division 4T), while median age at maturity ranged from 5.1 yr (Division 4X) to 9.2 yr (Subdivision 4Vn). Median length (and age) at maturity for females ranged from 33.0 cm (8.8 yr) in Subdivision 4Vn to 34.3 cm (7.2 yr) in Division 4X. Witch Flounder in warmer waters matured at younger ages than did those in colder waters. Biological and fishery mechanisms causing these changes are examined. Key Words: age, length, maturity, Northwest Atlantic, Witch Flounder, G/yptocephalus cynoglossus. Canadian groundfish surveys have indicated that Witch Flounder (Glyptocephalus cynoglossus) 1s widely distributed in Northwest Atlantic Fisheries Organization (NAFO) Subarea 4 (Figure 1) (Scott 1976). Witch Flounder is more abundant along the edge of the Laurentian Channel in Subdivision 4Vn, between Banquereau and Sable Island Banks, and in deeper areas of the Bay of Fundy than in other areas of NAFO Subarea 4 (Halliday 1973). Limited seasonal movements of witch may occur between the southern Gulf of St. Lawrence and the Scotian Shelf (Powles and Kohler 1970). Catches of Witch Flounder in the commercial fishery have been largest in Division 4V and 4W, the fishery mainly conducted by otter trawlers. Spawning occurs from late spring to summer (Leim and Scott 1966), and thus summer groundfish surveys would be timed appropriately in order to assess the reproductive state of an individual. Vertebrate population dynamics are determined by the combined effects of reproduction, growth, disper- sal, and mortality. The age at which individuals attain sexual maturity has a large impact upon the potential for population growth rates (Cole 1954; Stearns 1976). Growth rates of individuals and reproductive potential of a population are linked by the size and age at which individuals attain sexual maturity, and there- fore these parameters are of importance in an investi- gation of population dynamics. Long term population monitoring of Witch Flounder in several areas of NAFO Subarea 4 presented an opportunity for an investigation of variability in median size and age at sexual maturity of Witch Flounder among areas. The major purpose of this paper is to present historical changes in median size and age at maturity and to attempt to account for biological variability among witch in different areas. Materials and Methods The data analyzed in the present paper were derived from Canadian groundfish surveys of the M/ V HA- RENGUS, E. E. PRINCE, and A. T. CAMERON from 1959-1979. Annual values for median (50% mature point) length and age at maturity were calcu- lated where possible. Groundfish surveys were usually conducted in September of each year in the southern Gulf of St. Lawrence (Division 4T). Witch Flounder in the other areas surveyed were sampled during July surveys of the A. T. CAMERON. All autumn surveys in a particular time interval were grouped for the analysis, as were all summer surveys on the Scotian Shelf. Maturity ogives, which are percentage mature in size Or age Class versus size or age, were also calcu- lated if possible for four periods: 1959-1964, 1965- 1969, 1970-1974, and 1975-1979. Details of the sur- veys, including vessels and gear used and areas surveyed, were outlined by Halliday and Koeller (1981). During the surveys, total length of Witch Flounder was recorded to the nearest cm. The stages of maturity outlined by Powles (1965) for American Plaice (Hippoglossoides platessoides) were used in assessing maturity of Witch Flounder. Witch Flounder were aged by otoliths according to the method used for Cod (Gadus morhua) by Kohler (1964). The validity of using otoliths for age determi- nation in Witch Flounder was established by Powles and Kennedy (1967). The transition from the immature to the mature condition in fish usually occurs over a range of length 409 410 2 7 Georges / r7 Bank ,/ lA THE CANADIAN FIELD-NATURALIST Vol. 97 Na 2 Banquereau /} alt ei ye Par Emerald Rank FIGURE |. Northwest Atlantic Fisheries Organization Divisions in Subarea 4. and age in the form of a sigmoid curve. The median length at sexual maturity was calculated by probit analysis (Leslie et al. 1945) from the percentage of mature individuals (gonads in ripening, ripe, spawn- ing, spent, and recovering condition) in 2-cm length intervals. Median age at maturity was calculated by probit analysis by grouping the data in I-yr intervals. Results Division 4T There were not sufficient data to compute annual variability in median size and age at sexual maturity, so multi-year groupings were used. Maturity ogives based on length indicated that median length at maturity for male Witch Flounder was relatively con- stant during 1959-1979 at 33 to 34 cm (Figure 2). However, median length at maturity for females did decline from 42 cm during 1959-1964 to 33 cm during 1975-1979 (t = 8.83, P < 0.01). During 1959-1964, 70% of females 45 cm in length were mature, but during 1975-1979, 100% of 45-cm females were mature. Maturity ogives based on age indicated that median age at maturity for males was relatively con- stant at 8.0 yr, but median age for females declined from 12.6 yr during 1959-1964 to 8.8 yr during 1975- 1979 (t = 3.55, P < 0.05) (Table 1). Subdivision 4Vn As was the case for witch in Division 4T, annual variability in median size and age at maturity could not be investigated. However, median length at maturity for males declined from 37 cm during 1959- 1964 to 33 cm during 1975-1979 (t = 3.30, P < 0.05) (Figure 3). Median length at maturity for females declined from 44 to 33 cm during the same period (t = 5.11, P < 0.05). A larger percentage of witch was mature at smaller sizes, with no females 35 cm in length mature during 1959-1964, but with over 60% of 35 cm females mature during 1975-1979 (Figure 3). 1983 BEACHAM: VARIABILITY OF WITCH FLOUNDER 411 Male Female 1959 - 64 mei e27 1959-64 mel72Z 80 80 60 60 40 40 20 20 : 20 30 40 50 : 20 30 40 50 1970 - 74 Nek SS Si OnewG nebaq fed) 5 80 80 fe) > 60 60 ie 40 40 (eb) vs 20 20 (ab) Oo 0 0 20 30 40 50 20 30 40 50 ——— 100 80 60 GO 20 0 20 30 40 50 20 30 40 50 Length (cm) Length (cm) FIGURE 2. Maturity ogives of Witch Flounder derived from Canadian groundfish surveys in Division 4T, 1959-79. 412 THE CANADIAN FIELD-NATURALIST Vol. 97 TABLE |. Percentages of sexually mature Witch Flounder by age and sex caught during Canadian groundfish surveys in the southern Gulf of St. Lawrence, 1959-1979. Sample sizes in parenthesis after individual ages and 95% confidence limits after Aso (yr). Age (yr.) 1959-1964 1970-1974 1975-1979 Male 6 0.0(4) —(0) 0.0(1) 7 40.0(10) 100.0(1) 50(2) 8 50.0(12) 100.0(1) 66.7(3) 9 77.8(9) 100.0(4) 80.0(5) 10 83.3(12) 85.7(7) 64.7(17) 1] 50.0(4) 100.0(4) 88.5(26) 12 50.0(2) 100.0(11) 100.0(19) 13 100.0(2) 100.0(2) 96.3(27) 14 100.0(1) 100.0(10) 100.0(30) Aso 7.97(6.94-9.14) — 8.04(6.75-9.58) Female 6 0.0(2) — (0) 40.0 (5) 7 6.3(16) 0.0(2) 33.3(3) 8 6.3(16) 0.0(4) 25.0(4) 9 21.4(14) 33.3(9) 57.1(7) 10 26.7(15) 0.0(7) 16.7(6) 11 30.0(10) 42.9(7) 75.0(16) 12 0.0(2) 36.4(11) 73.9(23) 13 66.7(6) 42.9(7) 100.0(25) 14 100.0(1) 100.0(4) 92.9(28) 15 100.0(4) 100.0(8) 89.2(37) 16 100.0(3) 100.0(3) 100.0(40) Aso 12.64(10.20-15.68) — 8.77(7.43-10.36) Maturity ogives based on age indicated a non- significant increase in median age at maturity for males from 7.8 yr during 1959-1964 to 9.2 yr during 1975-1979(t = 1.53,P > 0.05), although less than 10 fish per age were sampled during 1975-1979 (Table 2). Median age at maturity for females declined from 10.5 yr during 1959-1964 to 8.8 yr during 1975-1979, although the decrease was not significant (t = 0.91, IP > 0.05). Subdivision 4Vs Annual variability in median size at maturity was investigated for witch in Subdivision 4Vs. Median length at maturity generally declined during 1964- 1978, with males maturing at smaller lengths than females (Figure 4). With data grouped multi- annually, maturity ogives based on length indicated that median size at maturity declined from 36 to 31 cm for males (t = 3.77, P < 0.05) and 43 to 34 cm for females (t = 15.53, P < 0.01) during 1959-1964 to 1975-1979 (Figure 5). As with witch in Division 4T and Subdivision 4Vn, smaller witch became mature through time as the 0% and 100% mature points shifted towards smaller fish. For males ages 4-7 and females ages 4-8, there was a general increase in per- centage mature at age with time (Table 3). Median age at maturity also declined during 1959-1979, but more rapidly than median length at maturity. Median age at maturity for males declined from 6.9 yr during 1959- 64 to 6.1 yr during 1975-1979 (t = 1.40, P > 0.005), while median age for females declined from 9.2 to 7.6 yr during the same interval (t = 3.63, P < 0.05). Division 4W Trends in median size and age at maturity for witch in Division 4W were similar to those of witch in the other areas investigated. There was an increasing per- centage mature at shorter lengths through time, so that maturity ogives shifted towards smaller fish (Fig- ure 6). Median length at maturity for males declined from 36 cm during 1959-1964 to 29 cm during 1975- 1979 (t = 12.72, P < 0.01), while those for females declined from 42 to 33 cm, respectively (t = 3.85, P < 0.05). Median ages at maturity for males also declined from 7.7 yr during 1959-1964 to 5.8 yr during ISTSA1I9 TON (GE =94-8 he E005) 5 wihiles those maton females declined from 9.2 to 7.1 yr, respectively (t = 6.13, P < 0.05) (Table 4). The proportion of 1983 BEACHAM: VARIABILITY OF WITCH FLOUNDER 413 Male Female 100 ISS) =z me&20 1959-64 Me B56 80 80 60 60 40 40 20 20 : 20 30 40 50) : 20 st 40 50 100 1970 Ws 100 1970 =e % A= S10 n=514 2.2 810 80 =| jo) i) 5.60 60 6 40 Cc (ab) DS 2 20 (ab) S 0 AW 30 40 50 20 30 40 50 20 310 40 50 20 30 40 50 Length (cm) Length (cm) FIGURE 3. Maturity ogives of Witch Flounder derived from Canadian groundfish surveys in Subdivision 4Vn, 1959-79. 414 THE CANADIAN FIELD-NATURALIST Vol. 97 TABLE 2. Percentages of sexually mature Witch Flounder by age and sex caught during Canadian groundfish surveys in Subdivision 4Vn, 1959-1979. Sample sizes are in parenthesis after individual ages and 95% confidence limits after Asp (yr). Age (yr.) 1959-1964 1970-1974 1975-1979 5 0.0(1) Male 0.0(4) 0.0(5) 6 50.0(2) 0.0(3) 0.0(0) 7 50.0(16) 36.4(11) 50.0(6) 8 40.5(42) 33.3(12) 40.0(5) 9 70.7(58) 64.0(25) 12.5(8) 10 81.0(63) 54.2(24) 75.0(8) MI 87.5(32) 94.1(17) 75.0(8) 12 97.4(39) 75.0(12) 60.0(5) 13 93.1(29) 75.0(4) 66.7(3) 14 100.0(13) 100.0(4) 100.0(4) Aso 7.83(7.29-8.40) 8.56(7.76-9.45) 9.21(7.93-10.69) Female 5 0.0(3) —(0) 0.0(0) 6 20.0(5) 50.0(2) 66.7(3) 7 6.7(15) 44.4(9) 0.0(1) 8 14.3(42) 41.2(17) 25.0(8) 9 20.8(53) 42.9(14) 33.3(6) 10 40.0(65) 60.0(25) 66.7(18) i 65.7(35) 42.3(26) 57.1(7) 12 75.0(28) 47.1(17) 90.0(10) 13 85.7(35) 71.4(14) 44.4(9) 14 92.3(26) 86.7(15) 66.7(9) 15 86.4(22) 90.0(10) 100.0(6) 16 100.0(16) 100.0(6) 100.0(3) 17 93.3(15) 100.0(1) 100.0(4) 18 100.0(5) 100.0(1) 100.0(1) ie 10.45(10.05-10.85) 9.46(8.04-11.11) 8.80(6.23-12.43) individuals mature at each age generally increased through time for males ages 4-8 and for females ages 5-9, and thus median age at maturity declined. Division 4X Maturity ogives based on length indicated that there was a decline in median size at maturity for witch in Division 4X, as in the other areas. Median length at maturity for males declined from 34 cm from 1959- 1964 to 30 cm during 1975-1979 (t = 2.01,P < 0.05), while those for females declined from 44 to 34 cm, respectively (t = 4.21, P < 0.05) (Figure 7). During 1959-1964, 60% of females 45 cm in length were mature, but during 1975-1979, 100% of 45-cm females were mature. The number of fish sampled at each age was generally small during 1959-1964 and 1975-1979 (Table 5), but median age at maturity for males declined from 7.7 yr during 1959-1964 to 5.1 yr during 1975-1979 (t = 3.01, P < 0.05). Median age at maturity for females declined from 10.2 to 7.2 yr during the same period (t = 4.01, P < 0.05). Comparisons among areas Several consistent trends were apparent when all areas were considered in the multi-year grouping of the groundfish surveys. Median lengths at sexual maturity generally declined through time (Table 6), as did median age at maturity. Males matured at younger ages and smaller sizes than did females. Median lengths at maturity were relatively constant among areas within an interval. Witch Flounder in the more northerly areas (Division 4T and Subdivision 4Vn) had older median ages at maturity than did those on the southern Scotian Shelf (Divisions 4W and 4X). Bottom water temperatures derived from all sets on the July groundfish surveys on the Scotian Shelf dur- ing 1975-1979 were 3.80°C in Subdivision 4Vn, 3.99°C in Subdivision 4Vs, 6.48°C in Division 4W, and 7.19°C in Division 4X. There was a trend for higher median age at maturity for Witch Flounder caught in colder waters. Growth rates of Witch Flounder have been correlated positively with water temperature (Beacham 1982), so there was a trend for Witch Flounder that had slower growth rates to mature at older ages, but at similar lengths, than did faster-growing witch. Discussion Little research has been conducted on variability in size and age at maturity of Witch Flounder in the Northwest Atlantic. Bowering (1976) reported that 1983 _ BEACHAM: VARIABILITY OF WITCH FLOUNDER 415 49 £0 > ec =) (e) = oe) & g 35 aes D Cc ra) al (< e) U0 ca) > 30 up) 1964 1966 1968 1970 ISI 2 1974 (SIV IS73 Year FIGURE4. Median length at sexual maturity for Witch Flounder caught in Canadian groundfish surveys in Subdivision 4Vs, 1964-78. Vertical bars indicate 95% confidence limits. 416 THE CANADIAN FIELD-NATURALIST Vol. 97 TABLE 3. Percentages of sexually mature Witch Flounder by age and sex caught during Canadian groundfish surveys in Subdivision 4Vs, 1959-1979. Sample sizes are in parenthesis after individual ages and 95% confidence limits after Asp (yr). Age (yr.) 1959-1964 1965-1969 1970-1974 1975-1979 Male 4 0.0(2) 0.0(6) 0.0(12) 4.2(24) 5 0.0(3) 0.0(1) 50.0(10) 38.2(34) 6 —(0) 22.2(9) 43.8(16) 59.1(22) 7 —(0) 43.8(16) 50.0(22) 62.5(8) 8 66.7(21) 66.7(9) 66.7(36) 66.7(12) 9 88.9(18) 69.6(23) 88.9(18) 66.7(15) 10 85.7(14) 79.5(39) 82.4(17) 80.0(20) 1] 83.3(6) 100.0(25) 100.0(2) 100.0(13) 12 83.3(6) 100.0(12) 100.0(7) 71.4(7) 13 100.0(1) 100.0(4) 100.0(3) 100.0(7) Aso 6.87(5.56-8.48) 7.50(6.85-8.22) 6.53(5.92-7.20) 6.07(5.34-6.91) Female 4 0.0(5) 0.0(4) 0.0(2) 0.0(26) 5 0.0(1) 0.0(4) 8.3(12) 15.2(33) 6 0.0(2) 9.1(11) 50.0(4) 68.2(22) 7 0.0(5) 0.0(7) 30.0(20) 54.6(11) 8 0.0(5) 16.7(12) 18.4(38) 30.0(10) 9 72.7(11) 23.1(13) 42.9(28) 64.3(14) 10 66.7(21) 35.3(17) 72.2(18) 42.9(14) 11 55.6(9) 72.4(29) 100.0(9) 80.0(20) 12 100.0(12) 92.0(50) 90.0(20) 94.1(17) 13 100.0(8) 96.8(31) 100.0(10) 87.5(8) 14 100.0(6) 95.8(24) 100.0(6) 100.0(7) 15 100.0(4) 100.0(8) 100.0(6) 100.0(6) Aso 9.16(8.39-9.98) 9.76(9.27-10.28) 8.88(8.29-9.51) 7.55(6.80-8.39) for Witch Flounder caught between 1958-1974, median age at maturity for males ranged between 4.2 yr in Divisions 2J-3K L (Southern Labrador and Nor- theast Newfoundland Shelf) to 6.2 yr in the northern Gulf of St. Lawrence (Division 4RS). Age at 50% maturity for females ranged from 8.4 yr on the Grand Bank (Division 3NO) to 10.2 yr on St. Pierre Bank (Subdivision 3Ps). Length at 50% maturity for males ranged from 25-30 cm, while that for females ranged from 40-50 cm. Bowering (1976) concluded that faster-growing witch matured at an earlier age, but all matured at approximately the same size, and the results of the present study support that conclusion. Powles and Kohler (1970) reported that for Witch Flounder caught on the Scotian Shelf in 1965, median length at maturity was 37 cm for males and 44 cm for females, with median ages of 7 and 10 yr, respectively, similar to the results in the present study. For Witch Flounder in most of the areas investi- gated in the present study, median length and age at sexual maturity generally declined through time. To the extent that size and age at maturity has a genetic component (Alm 1959), there has been selection for fish that mature at smaller sizes and younger ages. Witch Flounder have been exploited commercially for many years, and selection would favour those geno- types that reproduce at lengths not fully recruited to the commercial fishery, whereas a genotype that matures at a larger size might be removed before spawning. This mechanism would account for the decreasing abundance of larger, immature fish through time. In a study of variability in median size and age at sexual maturity for Haddock (Melanogrammus aeglefinus), Templeman and Bishop (1979) attributed a decline in median length at maturity to decreasing growth rates, and a decline in median age at maturity to increasing growth rates. However, Templeman et al. (1978) reported that median age at maturity of Grand Bank Haddock declined during 1953-1956 as successful year-classes appeared and the growth rate decreased. Thus median age at maturity declined in periods of increasing and decreasing growth rates, and trends through time may not be simply related to changes in growth rates. In North Sea Sole (Solea solea), de Veen (1976) reported that median length at maturity increased while median age at maturity remained relatively constant during a 10-year period, 1983 BEACHAM: VARIABILITY OF WITCH FLOUNDER Male Female 10 1959-64 100 1959 -64 As iBS n=119 80 80 60 60 40 £0 20 20 0 0 20 30 40 50 20 30 40 50 60 100 NGS — 619 1005, * 185-9 y Ase 232 n =153 80 80 60 60 eee EO 40 = 2 20 io) =S 0 0 20 30 40 50 20 30 40 50 60 1970 -74 n=485 ISIO=7E 100 n= 354 Percent 20 30 40 50 20 30 40 50 60 1975 -79 1975-79 n=220 n=265 80 80 60 60 40 40 20 20 ; 20 30 40 50 2 20 30 40 a0 60 Length (cm) Length (c m) FIGURE 5. Maturity ogives for Witch Flounder caught in Canadian groundfish surveys in Subdivision 4Vs, 1959-79. 417 418 THE CANADIAN FIELD-NATURALIST Volnoy TABLE 4. Percentages of sexually mature Witch Flounder by age and sex caught during Canadian groundfish surveys in Division 4W, 1959-1979. Sample sizes are in parenthesis after individual ages and 95% confidence limits after Aso (yr). Age (yr.) 1959-1964 1965-1969 1970-1974 1975-1979 Male 3 0.0(3) 0.0(1) 0.0(3) 0.0(6) 4 0.0(8) 0.0(16) 0.0(7) 20.0(5) 5 10.0(10) 12.9(31) 33.3(3) 8.3(12) 6 28.1(32) 10.5(19) 25.0(4) 72.7(22) 7 26.3(57) 54.6(22) 83.3(12) 81.3(16) 8 50.8(61) 72.0(25) 82.4(17) 84.6(13) 9 78.8(33) 92.9(28) 100.0(20) 82.4(17) 10 93.3(15) 82.6(23) 100.0(18) 79.0(19) 11 100.0(9) 100.0(14) 100.0(4) 94.7(19) 12 100.0(4) 100.0(8) 100.0(9) 100.0(14) Aso 7.68(7.33-8.04) 7.01(6.61-7.43) 6.16(5.52-6.86) 5.75(5.09-6.50) Female 5 0.0(17) 0.0(25) 0.0(2) 0.0(7) 6 4.3(47) 5.0(20) 25.0(4) 33.3(12) 7 9.2(65) 0.0(24) 33.3(6) 40.0(5) 8 16.5(85) 5.6(18) 38.5(13) 45.5(11) 9 56.0(75) 13.8(29) 69.2(13) 100.0(6) 10 65.2(46) 48.2(27) 70.0(10) 88.2(17) 11 80.7(31) 68.8(32) 100.0(16) 93.3(15) 12 100.0(21) 87.0(23) 93.8(16) 95.2(21) 13 94.1(17) 100.0(13) 100.0(14) 100.0(19) 14 100.0(5) 100.0(8) 90.0(10) 100.0(13) Aso 9.17(8.88-9.48) 10.24(9.89-10.59) 8.16(7.34-9.06) 7.12(6.42-7.90) with the increase in length at maturity being accounted for by an increase in growth rates. Many groundfish stocks in the Northwest Atlantic endured high rates of exploitation in the 1960s and early 1970s, and consequently declined in stock bio- mass. If growth rates of Witch Flounder are inversely correlated with stock biomass, then an increase in growth rate could account for a decline in age at maturity, but it is difficult to account for a decline in size at maturity. Catch quotas were introduced in the mid-1970s and thus stock biomass may increase in the future. Further research is necessary in order to dis- tinguish between selection for early maturing individ- uals producing a decline in size and thus age at matur- ity or changes in environmental parameters such as stock biomass or temperature accounting for the decline in size and age at maturity. If changes in size and age at maturity are largely dependent upon stock biomass, then these parameters should increase if stock biomass increases under a management regime of reduced exploitation. During 1975-1977, median age at sexual maturity varied from 8 to 9 years for males and 9 years for females in the northern Division 4T and Subdivision 4Vn populations to 5 years for males and 7 years for females in the southern Division 4X population. Mean growth rates in the populations generally declined after most of the individuals became sexually mature (Beacham 1982). The present study indicated that Witch Flounder inhabiting regions of warmer water temperatures matured at younger ages than did witch in regions of colder water temperatures. This general trend was noted by Gunter (1950) who stated that fish inhabiting regions of higher water temperature grew faster initially, attained sexual maturity earlier, and were of smaller final size than the same species in regions of lower water temperature. However, Fleming (1960) found that Cod in the Labrador region of Newfound- land attained sexual maturity at younger ages, but grew more slowly than Cod in stocks further south. Fleming attributed this result to Cod in poorer envir- onments maturing earlier than those in more favoura- ble environments. The results of the present study indicate that individuals with higher growth rates attained sexual maturity earlier than those with slower growth rates. Alm (1959) noted that fish witha high growth rate attained sexual maturity at an earlier age than did slower-growing fish, and the results of the present study on regional variability in age at maturity support that conclusion. 1983 BEACHAM: VARIABILITY OF WITCH FLOUNDER 419 Male Female 100, 1959 -64 100 1959 -64 ne S06 AN =eO 7 80 80 60 60 40 40 20 Soca 20 20 30 40 50 20 30 40 50 100, 1965-69 : 100 1965-69 ne 256 n= 318 804 80 60 60 404 40 = 20 20 jo) = 0 0 pe 20 30 40 50 20 30 40 50 5 100 IG7O = 7 100 1970 = 7A ba in eke mn = S64 w 80 80 (QL 60 60 40 40 20 20 @) 6) 20 30 40 50 20 30 40 50 IS7a—79 1975 -79 nse 2is7 N= e277. 80 60 40 20 0 210 30 40 50 20 30 40 50 Length (cm) Length (cm) FIGURE 6. Maturity ogives for Witch Flounder caught in Canadian groundfish surveys in Division 4W, 1959-79. 420 THE CANADIAN FIELD-NATURALIST Vol. 97 TABLE 5. Percentages of sexually mature Witch Flounder by age and sex caught during Canadian groundfish surveys in Division 4X, 1959-1979. Sample sizes are in parenthesis for individual ages and 95% confidence limits after Asp (yr). Age (yr.) 10 1959-1964 —(0) 0.0(1) WW) 0.0(4) 0.0(3) 71.4(7) 71.4(7) 87.5(16) 100.0(4) 85.7(7) 100.0(1) 7.74(6.51-9.20) 0.0(0) =) 0.0(7) 0.0(5) 14.3(7) 22.2(9) 33.3(9) 62.5(8) 90.0(10) 100.0(5) 100.0(1) .18(9.54-10.86) Male Female 1970-1974 0.0(3) 33.3(3) 0.0(4) 100.0(2) 45.5(11) 80.0(15) 94.4(18) 85.7(14) 91.7(12) 100.0(3) 80.0(5) 5.96(4.63-7.67) 0.0(1) 0.0(4) 0.0(6) 20.0(10) 27.8(18) 60.0(20) 87.1(31) 85.7(14) 100.0(22) 100.0(19) 92.3(13) 8.52(8.03-9.03) 1975-1979 100.0(1) 33.3(9) 66.7(12) 88.9(9) 83.3(6) 83.3(6) 100.0(10) 91.7(12) 100.0(8) 100.0(15) 5.09(4.05-6.39) 0.0(2) 20.0(5) 25.0(4) 25.0(8) 71.4(7) 77.8(9) 100.0(6) 100.0(9) 92.9(14) 94.4(18) 100.0(14) 7.21(6.41-8.10) TABLE 6. Median length at sexual maturity and 95% confidence limits of Witch Flounder in Divisions 4T, 4V, 4W, and 4X determined from Canadian groundfish surveys from 1959-1964, 1965-1969, 1970-1974, and 1975-1979. Males Division 1959-1964 1965-1969 1970-1974 4T 34.22 — 32.97 (32.66-35.85) (32.04-33.92) 4Vn 36.92 — 33.14 (36.00-37.86) (32.28-34.03) 4Vs 36.40 35.32 : 32.42 (34.58-38.32) (33.92-36.78) (31.66-33.19) 4W 35.71 32.45 30.96 (34.76-36.68) (31.35-33.60) (29.89-32.07) 4X 33.76 — 33.00 (31.37-36.32) (31.30-34.79) Females Division 1959-1964 1965-1969 1970-1974 4T 42.45 — 39.26 (40.92-44.03) (38.50-40.03) 4Vn 43.90 a 37.69 (43.09-44.73) (36.75-38.66) 4Vs 43.37 42.76 37.64 (42.16-44.55) (41.56-44.00) (36.88-38.42) 4W 42.37 43.56 36.44 (41.69-43.06) (42.75-44.37) (35.52-37.88) 4X 43.92 — 39.41 (40.63-47.46) (38.04-40.83) 1975-1979 32.95 (31.62-34.34) 32.88 (30.58-35.36) 30.77 (29.77-31.81) 29.20 (27.82-30.65) 30.13 (27.90-32.54) 1975-1979 33.36 (31.39-35.45) 32.96 (28.50-38.13) 33.54 (32.28-34.86) 33.10 (31.51-34.77) 34.29 (32.51-36.17) 1983 BEACHAM: VARIABILITY OF WITCH FLOUNDER 421 Male Female 100 1959 =64 i) 2 SS} 1959 - 64 mn so& 80 80 60 60 40 40 20 20 0 20 30 40 50 20 30 40 50 100 1970-74 100 1970 -74 n=194 = Soe = 80 80 jo) PA 60 60 me 40 40 oO Se 20 20 (eb) aa 0 0 20 30 40 50 20 30 40 50 20 30 40 50 20 30 40 50 Length (cm) Leme@in (te ma) FIGURE 7. Maturity ogives for Witch Flounder caught in Canadian groundfish surveys in Division 4X, 1959-79. 422 Acknowledgments I am indebted to those people who have crewed the research vessels and sampled and aged the Witch Flounder. Dr Charles J. Krebs of the University of British Columbia wrote the computer program for probit analysis. Two referees provided helpful com- ments on the manuscript. Albert Vromans ably assisted in much of the analysis. The manuscript was prepared with the assistance of the staff at the Pacific Biological Station. Literature Cited Alm, G. 1959. Connection between maturity, size and age in fishes. Report of the Institute for Freshwater Research Drottningholm 40. 145 pp. Beacham, T. D. 1982. Growth and Canadian exploitation of witch flounder (G/yptocephalus cynoglossus) in the Maritimes area. of the Northwest Atlantic Ocean. Cana- dian Technical Report of Fisheries and Aquatic Sciences 1112. 30 pp. Bowering, W. R. 1976. Distribution, age and growth, and sexual maturity of witch flounder (G/yptocephalus cyno- glossus) in Newfoundland waters. Journal of the Fisheries Research Board of Canada 33: 1574-1584. Cole, L. C. 1954. The population consequences of life his- tory phenomena. Quarterly Review of Biology 29: 103-137. Fleming, A. M. 1960. Age, growth, and sexual maturity of cod (Gadus morhua L.) in the Newfoundland area, 1947- 50. Journal of the Fisheries Research Board of Canada 17: 775-809. Gunter, G. 1950. Correlation between temperature of water and size of marine fishes on the Atlantic and Gulf coasts of the United States. Copeia 1950(4): 298-304. Halliday, R. G. 1973. The flatfish fisheries of the Scotian Shelf. International Commission on Atlantic Fisheries Redbook, Part III: 79-99. Halliday, R. G.,and P. A. Koeller. 1981. A history of Can- adian groundfish trawling surveys and data usage in ICNAF Divisions 4TVWX. Pp. 27-41 in Bottom trawl surveys. Edited by W. G. Doubleday and D. Rivard. Can- adian Special Publications of Fisheries and Aquatic Sciences 58. 273 pp. THE CANADIAN FIELD-NATURALIST Vol. 97 Kohler, A. C. 1964. Variations in the growth of Atlantic cod (Gadus morhua L.) Journal of the Fisheries Research Branch of Canada 21: 57-100. Leim, A. H., and W. B. Scott. 1966. Fishes of the Atlantic coast of Canada. Fisheries Research Board of Canada Bulletin 155. 485 pp. Leslie, P. H., J.S. Perry, and J.S. Watson. 1945. The determination of the median body weight at which female rats reach maturity. Proceedings of the Zoological Society of London 115: 473-488. Powles, P. M. 1965. Life history and ecology of American plaice (Hippoglossoides platessoides F.) in the Magdalen Shallows. Journal of the Fisheries Research Board of Canada 22: 565-598. Powles, P. M., and V.S. Kennedy. 1967. Age determina- tion of Nova Scotian greysole, G/yptocephalus cynoglos- sus L., from otoliths. International Commission on Atlan- tic Fisheries Resources Bulletin 4: 91-100. Powles, P. M., and A.C. Kohler. 1970. Depth distribu- tions of various stages of witch flounder (G/yptocephalus cyvnoglossus) off Nova Scotia and in the Gulf of St. Law- rence. Journal of Fisheries Research Board of Canada 27: 2053-2062. Scott, J.S. 1976. Summer distribution of groundfish on the Scotian Shelf, 1970-74. Fisheries and Marine Services Research Division Technical Report 635. 51 pp. Stearns, S.C. 1976. Life-history tactics: a review of the ideas. Quarterly Review of Biology 51: 3-47. Templeman, W., V. M. Hodder, and R. Wells. 1978. Sex- ual maturity and spawning in haddock, Melanogrammus aeglefinus, of the Southern Grand Bank. International Commission on Northwest Atlantic Fisheries Resources Bulletin 13: 53-65. Templeman, W., and C. A. Bishop. 1979. Sexual maturity and spawning in haddock, Melanogrammus aeglefinus, of St. Pierre Bank. International Commission on Northwest Atlantic Fisheries Resources Bulletin 14: 77-83. de Veen, J. F. 1976. Onchanges in some biological parame- ters in the North Sea sole (Solea solea L.). Journal du Conseil. Conseil International pour l’Exploration de la Mer 37: 60-90. Received 16 July 1982 Accepted 21 July 1983 Food Habits of Young Anadromous Alewives, Alosa pseudoharengus, in Lake Ainslie, Nova Scotia ROBERT S. GREGORY, GAIL S. BROWN, and GRAHAM R. DABORN Department of Biology, Acadia University, Wolfville, Nova Scotia BOP 1X0 Gregory, Robert S., Gail S. Brown, and Graham R. Daborn. 1983. Food habits of young anadromous Alewives, A/osa pseudoharengus, in Lake Ainslie, Nova Scotia. Canadian Field-Naturalist 97(4): 423-426. Gut contents of 60 larval and 170 juvenile Alosa pseudoharengus, in Lake Ainslie, Nova Scotia, collected in 1979, were examined concurrently with zooplankton samples. Until August, the larvae fed primarily on the dominant zooplankters Bosmina spp., Daphnia retrocurva and Diaptomus minutus. In August, however, the young Alewives switched to feeding on adult Diptera and Ephemeroptera when zooplankton abundance was low. A substantial drop in average gut biomass was noted at this time. The contribution of these two orders of insects to the dry weight biomass of the gut was as high as 82.6% on 23 August. Key Words: Alosa pseudoharengus, anadromous population, zooplankton, Margaree River system, opportunistic feeding. The feeding habits of landlocked Alewives (A/osa pseudoharengus) and Blueback Herring (A. aestiva- lis) have been well studied. In general they are consi- dered to be largely or solely planktivorous in habit, and through size-selective feeding, to exert major influence on the specific composition of the zooplank- ton (e.g. Brooks and Dodson 1965; Brooks 1968; Hut- chinson 1971; Wells 1970; Warshaw 1972; Vigerstad and Cobb 1978). Some studies have shown, however, that Alewives become more omnivorous with increas- ing size (Norden 1968; Morsell and Norden 1968; Watt and Duerden 1974; Kohler and Ney 1980). In contrast, relatively little is known about the feed- ing habits of anadromous Alewives, or of their impact on lake communities during their limited freshwater phases. Less is understood about the effects of feeding by young-of-the-year of these populations. Vigerstad and Cobb (1978) interpreted changes in zooplankton composition of a small New England reservoir as resulting from anadromous Alewife predation. Adult anadromous Alewives generally return to the estuary or the sea during mid or late summer, and juveniles follow in late summer or fall. The impact is thus seasonally highly variable. B. Jessop (personal communication) has estimated that 3 X 10° adult Alewives successfully evaded the approximately 50 commercial fishing operations on the Margaree River and entered Lake Ainslie in May and June, 1979. This study was conducted to examine the food habits of larval and juvenile Alewives in Lake Ainslie in relation to seasonal changes in zooplankton density and composition. Study Area Lake Ainslie is a large shallow freshwater lake in Cape Breton Island, Nova Scotia, Canada (46° 08’N, 61° 11’W). The lake is 19 km long and 7 km wide, with a total surface area of 57.4 sq km. Mean depth is 5.75 m, reaching a maximum depth of 18 m. Lake Ainslie is drained by the Southwest Margaree River, which flows north-westerly to the Gulf of St. Lawrence. The shore line consists of gravel to boulder sized stones interspersed with the occasional sand beach. The lake bottom is mostly sand, silt, or a sand-silt mixture. The lake’s sparse macrophytic vegetation is found mostly in the northwestern portion of the lake and consists mainly of bulrushes (Scirpus sp.) cattail (Typha sp.) and water lilies (Nuphar sp. and Nym- phaea sp). The lake 1s oligotrophic: oxygen levels in the water column normally exceed 90% saturation and the lake remains unstratified during the summer because of wind mixing. Other fish occurring in the lake include the American Smelt (Osmerus mordax), White Perch (Morone americana), Brook Trout (Sa/velinus fonti- nalis), Atlantic Salmon (Sa/mo salar), White Sucker (Catostomus commersoni), Killifish (Fundulus dia- phanus), American Eel (Anguilla rostrata), Three- spined Stickleback (Gasterosteus aculeatus), Nine- spined Stickleback (Pungitius pungitius) and Creek Chub (Semotilus atromaculatus). Methods Samples of larval fish were obtained between 20 May 20 and 28 July 1979 using a 1050 um, 0.5 m diameter plankton net. Juveniles were collected between 12 June and 23 August with a 15.2 X 2.4m beach seine, the bag of which consisted of 2.4 meters square of 0.6 cm delta nylon knotless mesh. Fish from 13 stations were used for gut analysis; specimens were immediately preserved in 95% ethanol. Vertical zoo- plankton hauls of the full water column were taken concurrently in close proximity to these sites using a No. 20 Wisconsin net. Lengths of larvae (L) were measured to + 0.01 mm by eyepiece micrometer in an Olympus microscope, and juveniles to the nearest 0.05 mm using Vis calip- 423 424 ers. Juveniles were designated as those specimens of 19.1 mm L or greater (Jones et al. 1978). Sixty larvae and | 70 juveniles were examined. Spec- imens were selected to represent sample sites, dates, and sizes as completely as possible. Larval fish were cleared with 3% KOH and glycerine to make gut material visible through the digestive tract, which is relatively undifferentiated until 20 mm L. Total gut analysis was necessary up to 23.0 mm FL, after which only stomach contents were analyzed. Dry weight biomass of the gut contents was esti- mated from mean dry weight of the predominant food organisms. These were vacuum dried for 24 hours ina Fisher Isotemp vacuum oven at 70°C. Dry weight was determined with a Cahn electrobalance. Six groups were determined for each food organism, when possi- ble, the number of organisms per group varying with their availability. Cladocerans and copepods were obtained from the plankton samples. Dipteran and ephemeropteran insects, ostracods, and amphipods were obtained from juvenile stomachs; in order to avoid discrepancies in mass due to digestion, these organisms were only taken from the anteriormost por- tion of the stomach. Only intact specimens were used. The dry weight of rotifers was not estimated. We could not find intact specimens, and their size ren- dered them relatively insignificant to the diet. Results Changes in abundance and species composition of zooplankton (Figure 1!) exhibited distinct seasonal trends. Total abundance reached a peak in June, declined in July and remained very low in August. By the end of September, there was a substantial increase in zooplankton. The major species of zooplankton were Diaptomus minutus, Cyclops sp., Daphnia retrocurva, Bosmina spp., Asplanchna priodonta, Keratella cochlearis, and Polyarthra vulgaris. Diaptomus, Daphnia, and Bos- mina were dominant during the spring maximum, whereas Cyc/ops was most numerous in early July and late September. During August, when zooplankton density was low, the predatory rotifer Asplanchna priodonta was the most abundant organism. By the end of September, Bosmina and Cyclops were domi- nant at all stations. The species assemblage and changes in seasonal abundance are considered fairly typical of temperate lakes. All sampling stations exhibited similar trends in changes of abundance of zooplankton except for Loch Ban, which was identified as the major area of spawning, hatching, and early development for Ale- wives. We presume that the low springtime abundance of zooplankton in Loch Ban is the result of feeding pressure exerted by Alewife larvae. THE CANADIAN FIELD-NATURALIST Vol. 97 The cladocerans Bosmina spp. and Daphnia retro- curva and the copepod Diaptomus minutus were the primary food items of young Alewives until early August (Figure 2). The three species represent by far the largest proportion of the zooplankton in the water column (Figure |). Except for 23 May, rotifers were present in the diet from 20 May to 27 June in reason- able numbers. Calculations of feeding indices (Berg 1979) suggest that rotifers and copepod nauplii are strongly selected at lower lengths of larvae and juve- niles, but negatively selected by larger fish. During August, when rotifers represented the largest propor- tion of the plankton, they were not consumed. The weight of food in the gut showed a significant drop early in August (Figure 3) followed by a sharp rise at the end of the month. The decline in the amount of food in the gut was apparent on three successive days. Our data clearly indicate that anadromous alewives are opportunistic feeders, rather than determined planktivores. With increase in body size during devel- opment the variety of food types increases, generally in accordance with availability. Organisms utilized included both planktonic and benthic species, and the presence of adult insects indicates a propensity for surface feeding also. 60 e@— DIAPTOMUS COPEPODA O——9 CYCLOPS One DAPHNIA 50 CLADOCERA ®---®@ BOSMINA w- -@ ASPLANCHNA 4 KERATELLA a POLYARTHRA ROTIFERA 40 Obes. A --1--©6 30 ORGANISMS / LITER 20 AUGUST “SEPTEMBER MAY JUNE JULY FIGURE |. Changes in mean abundance of zooplankton at 13 stations in Lake Ainslie, Nova Scotia, 1979. 1983 100 90 80 icp) 2 70 < @—* COPEPODA = +0 DAPHNIA 60 CLADOCERA u. @---@ BOSMINA i xem DIPTERA AND ze oo EPHEMEROPTERA O ul a 40 30 20 10 MAY GREGORY, BROWN AND DABORN: FOOD HABITS OF ALEWIVES 425 AUGUST FIGURE 2. Changes in weight percent of food ingested by young-of-the-year Alewives (A/osa pseudoharengus) in Lake Ainslie, IO), Discussion Brooks and Dodson’s (1965) theory that the Ale- wife is an obligate planktivore selecting larger prey sizes as length increases, was generally true until August. At this time, large numbers of adult chiro- nomids and mayflies were incorporated into the diet. The period of heavy reliance on insects corresponds with the period of lowest zooplankton density. Sim- ilarly, Pritchard (1929), Odell (1934), Norden (1968), Hutchinson (1971), Watt and Duerden (1974), and Kohler and Ney (1980) supply data suggesting that young Alewives become more omnivorous with increasing size. Larvae in May were taking | or 2 food types whereas juveniles in August were taking 14 to 18. Brooks and Dodson (1965) have suggested selective predation as a cause of cyclomorphosis, the cyclical seasonal change in shape exhibited by some Daphnia populations; Daphnia retrocurva inhabiting lakes with landlocked populations of Alewives is often strongly cyclomorphic. Despite intense predation on Daphnia retrocurva by young Lake Ainslie Alewives until August (when it disappeared from plankton samples), it underwent neither changes in body size or form. The low zooplankton densities exhibited during August would undoubtedly have an effect ona plank- tivorous fish. It has been suggested that low food resources could be a cause of downstream migration of Alewive juveniles (Richkus 1975; Vigerstad and Cobb 1978). We have no evidence of early August migration from Lake Ainslie in 1979, although there seems to have been a general movement of juveniles toward the outlet of the lake at this time (O’Neill 1980). Low plankton densities during August may normally cause downstream migrations, but the switch in feeding behaviour from planktonic to sur- face feeding on insects, could have enabled some juve- niles to remain in the lake. Although the extensive changes in weight of gut contents are not entirely explained, they may demonstrate that drastic switches in feeding behaviour can be accomplished after a period of adjustment. The Alewife is clearly capable of drastically altering zooplankton densities, and perhaps species composi- tion. In Lake Ainslie, a lake of apparently modest productivity, the production of an estimated 3 X 10! young Alewives from anescapement of 3 X 10° adults into the lake, is of major significance. Much more needs to be learned regarding the impact of anadrom- ous species like the Alewife on the trophic structure of lakes used for spawning. 426 1.8 AVERAGE GUT BIOMASS (mg) Or ON ne Fo ee (o>) (ee) (o) nh - or) ©) is 0.2 MAY JUNE JULY AUGUST FIGURE 3. Changes in total food ingested by young-of-the- year Alewives (A/osa pseudoharengus) in Lake Ains- lie, 1979. Acknowledgments This study was financially supported by an Envir- onment Canada contract (Number OSC78-00029) and NSERC grant A9697 to Graham Daborn. Weare deeply indebted to John T. O’Neill and H. Carolyn Peach of Acadia University for their unfailing aid in data collection and for their advice. Thanks are extended to Brian Jessop (Fisheries and Oceans) for his support and provision of equipment during this project. Literature Cited Berg, J. 1979. Discussions of methods of investigating the food of fishes, with reference to a preliminary study of the prey of Gobiusculus flavescens (Gobiidae). Marine Biol- ogy 50: 263-273. Borror, D. J., and R. E. White. 1970. A field guide to the insects. Houghton Mifflin Company, Boston. Brooks, J. L., and S.I. Dodson. 1965. The effect of a marine planktivore on lake plankton illustrates theory of size, competition, and predation. Science 150: 28-35. Brooks, J. L., and S. I. Dodson. 1968. The effects of prey size selection by lake planktivores. Systematic Zoology 17: 273-291. Eddy, S.,and A. C. Hodson. 1961. Taxonomic keys to the common animals of the north central states. Burgess Pub- lishing Company, Minneapolis. THE CANADIAN FIELD-NATURALIST Vol. 97 Hutchinson, B. P. 1971. The effect of fish predation on the zooplankton of ten Adirondack lakes, with particular ref- erence to the Alewife, A/osa pseudoharengus. Transac- tions of the American Fisheries Society 100: 325-335. Jones, P. W., F. D. Martin, and J. D. Hardy, Jr. 1978. De- velopment of fishes of the Mid-Atlantic Bight. Volume 1. Acipenseridae through Ictaluridae. Fish and Wildlife Ser- vice, U.S. Department of the Interior. 395 pp. Kissil, G. W. 1974. Spawning of the anadromous Alewife, Alosa pseudoharengus, in Bride Lake, Connecticut. Tran- sactions of the American Fisheries Society 103: 312-317. Kohler, C. C., and J. J. Ney. 1980. Piscivority in a land- locked Alewife (Alosa pseudoharengus) population. Can- adian Journal of Fisheries and Aquatic Sciences 37: 1314-1317. Morsell, J. W.,and G.R. Norden. 1968. Food habits of the alewife, A/osa pseudoharengus (Wilson) in Lake Michi- gan. Pp. 96-102 in Proceedings of the 11th Conference on Great Lakes Research. Norden, C. R. 1968. Morphology and food habits of larval Alewife in Lake Michigan. Pp. 103-110 im Proceedings of the 11th Conference on Great Lakes Research. Odell, T. T. 1934. The life history and ecological relation- ships of the alewife in Seneca Lake, New York. Transac- tions of the American Fisheries Society 64: 118-126. O'Neill, J. T. 1980. Aspects of the life histories of anad- romous Alewife, A/osa pseudoharengus (Wilson), and the Blueback herring, A. aestivalis (Mitchell), in the Southw- est Margaree River and Lake Ainslie, Nova Scotia, 1978- 1979. M.Sc. thesis, Acadia University, Wolfville, Nova Scotia. Pritchard, A. L. The Alewife (Pomolobus pseudoharengus) in Lake Ontario. University of Toronto Studies, Publica- tion of the Ontario Fish Research Laboratory 39: 39-54. Richkus, W. A. 1975. Migratory behaviour and growth of juvenile anadromous Alewives, A/osa pseudoharengus, in a Rhode Island drainage. Transactions of the American Fisheries Society 104: 483-493. Warshaw, S. J. 1972. Effects of Alewives (A/osa pseudoha- rengus) on the zooplankton of Lake Wonoskopomuc, Connecticut. Limnology and Oceanography 17: 816-825. Watt, W. D. and F. C. Duerden. 1974. Aquatic ecology of the Saint John River - Volume 2, Report No. 15G, The Saint John River Basin Board..112 pp. Wells, L. 1970. Effects of Alewife predation on zooplank- ton populations in Lake Michigan. Limnology and Ocea- nography 15: 556-565. Vigerstad, T. J.,and J.S. Cobb. 1978. Effects of predation by sea-run juvenile Alewives (A/osa pseudoharengus) on the zooplankton community at Hamilton Reservoir, Rhode Island. Estuaries 1: 36-45. Received 16 February 1982 Accepted 10 August 1982 Coastwide Distribution and Ocean Migration Patterns of Stream- and Ocean-Type Chinook Salmon, Oncorhynchus tshawytscha M. C. HEALEY Department of Fisheries and Oceans, Fisheries Research Branch, Pacific Biological Station, Nanaimo, British Columbia VOR 5K6 Healey, M. C. 1983. Coastwide distribution and ocean migration patterns of stream- and ocean-type Chinook Salmon, Oncorhynchus tshawytscha. Canadian Field-Naturalist 97(4): 427-433. Chinook Salmon, Oncorhynchus tshawytscha, hich spend a year or more in fresh water before migrating to sea (stream- type) have been considered to be an environmentally produced variant of those migrating to sea in their first year of life (ocean-type). The available evidence, however, shows that these two life-history types differ in the coastal distribution of their spawning populations, in their ocean distribution as immatures, and in the seasonal timing of their spawning migrations. These differences in distribution and behaviour cannot be accounted for by the fresh-water environmental conditions that might influence fresh-water residence. Rather, it appears that stream- and ocean-type Chinook Salmon are distinct races. Key Words: Chinook Salmon, Oncorhynchus tshawytscha, racial identification, coastal and ocean distribution, migration. Gilbert (1913) showed that Chinook Salmon (Oncorhynchus tshawytscha) may migrate seaward during the spring and summer of their first year of life, or after one or two years in fresh water. He termed those chinook which migrated to sea during their first year sea (ocean-) type and those which reared for one or two years in fresh water stream-type. Rich (1925) noted that in the spawning run of Chinook Salmon to the Columbia River, which extends throughout the spring, summer and autumn, stream-type chinook predominated early in the run and ocean-type chi- nook late in the run. Rich (1925) also noted that the most upstream spawning populations in the Colum- bia tended to be dominated by stream-type fish. These same features distinguish stream- and ocean-type chi- nook in the Fraser River (Ball and Godfrey 1968a; P. F. Starr, C. L. Cross, and F. J. Fraser. 1981. Chinook and coho of the Fraser River: summary 1951-1978. Canada Fisheries and Marine Service, unpublished ms. 152 pp.), the Nanaimo River (Hea- ley, unpublished data), and probably other rivers hav- ing both life history types. On the basis of samples from Monterey Bay, the Columbia River, the Strait of Georgia, and the Yukon River, Rich (1925) observed that the proportion of stream-type chinook in spawn- ing runs was positively correlated with both latitude and altitude of the spawning beds. Rich (1925) con- cluded that the tendency to remain a year in fresh water was determined by the harshness of the fresh water rearing environment, with more chinook remaining a year in fresh water in harsher environments. Considerable additional information now exists on the occurrence of stream- and ocean-type chinook in spawning rivers along the coast of North America, 427 and on the distribution of the two types at sea. These data suggest that, rather than phenotypic variants determined by environmental conditions, stream- and ocean-type chinook may be separate races. This paper presents the evidence in support of this alternate hypothesis. The Coastwide Distribution of Spawning Stream- and Ocean-type Chinook To describe the coastwide occurrence of spawning stream- and ocean-type chinook, precise estimates of the contribution of each type to unexploited spawning runs are desirable. Such estimates do not exist for several reasons. Firstly, sampling of spawning runs has generally been motivated by the presence of an intensive chinook fishery. The two life-history types, because of their different run timings, may have been subject to different exploitation rates. They may also differ in their stock and recruitment parameters. Samples taken decades after the start of commercial exploitation may, therefore, misrepresent the contri- bution of the two life-history types to the unexploited population. The estimates reported below are the ear- liest available for each river system, and should, there- fore, be the least altered by exploitation. The esti- mates are, however, mainly from samples taken many years after the start of commercial fisheries. Secondly, some of the estimates are based on a single sampling. The contribution of stream- and ocean-type chinook to a spawning run can vary considerably from year to year (Snyder 1931; P. F. Starr, C. L. Cross and F. J. Fraser 1981 unpublished ms). A single sample may not represent the average contribution of the two life history types to the population. Finally, stream-type chinook in spawning runs are distinguished by the 428 presence of a fresh water annulus on their scales. The identification of a fresh water annulus is not always a straightforward matter (Koo and Isarankura 1967; Tutty and Yole 1978), therefore some fish may be misclassified. In spite of these difficulties, it is unlikely that the patterns observed in the data are a result of sampling problems, and some important general con- clusions are possible. Most of the data on spawning run composition are from rivers along the British Columbia coast, although data are available from four Alaskan rivers, two Washington-Oregon rivers and two California rivers (Table 1). Stream-type chinook comprise 0-100% of spawning runs to these rivers. In Alaska, including rivers crossing the panhandle, chinook populations are virtually 100% stream-type (Table 1) and some fish spend two years in fresh water before migrating to sea. In British Columbia, stream- type chinook comprise 0-57% of spawning runs. Low THE CANADIAN FIELD-NATURALIST Vol. 97 percentages of stream-type chinook are found in both northern and southern British Columbia rivers, although three populations with the highest recorded percentage of stream-type chinook (Nass, Skeena and Yakoun rivers) are on the northern British Columbia coast (Table 1). In the Washington, Oregon and Cali- fornia rivers the overall percentages of stream-type chinook is similar to British Columbia and ranges 11-22% among rivers (Table 1). While these data may be interpreted to represent a latitudinal cline in occur- rence of stream-type chinook, there are many excep- tions to the cline. Of particular importance to the theme of this paper are the rather abrupt change from predominantly ocean-type chinook to virtually 100% stream-type about 55-56° North latitude and the occurrence of high percentages of stream-type chi- nook in large rivers even to the southern limit of the chinook’s range. Compare, for example, the Stikine River at 56°40’ N which has 100% stream-type chi- TABLE I. Occurrence of stream-type Chinook Salmon in spawning runs to rivers along the west coast of North America. The rivers are ordered in descending latitude. River Approximate Information Percent System(s) N. Latitude Source Stream-Type Alaska Yukon 60° 30’ Gilbert 1913 100 Cook Inlet 61° 30 Yancey and Thorsteinson 1963 97-99 Taku 58° 30’ Mehan and Siniff 1962, Kissner 1973 100 Stikine 56° 40’ Kissner 1973 100 British Columbia Nass ajo Ay Godfrey 1968 42 Skeena 54° 20° Godfrey 1968 48 Kitimat 54° 0! Canada, Fisheries and Marine Service (Unpublished Data) 12 Yakoun 53° 20° Canada, Fisheries and Marine Service (Unpublished Data) 57 Bella Coola D2 ay Canada, Fisheries and Marine Service (Unpublished Data) 14 Wannock (Rivers Inlet) 51° 40’ Schutz 1975 3 Quinsam (Campbell) 50° 0’ Canada, Fisheries and Marine Service (Unpublished Data) i Big Qualicum 49ers Canada, Fisheries and Marine Service (Unpublished Data) 0 Fraser 49° 20 Godfrey 1968 34 Nanaimo 49° 10’ Healey Unpublished Data 5 Natinat 48° 50’ Healey Unpublished Data I Chemanus 48° 50’ Canada, Fisheries and Marine Service (Unpublished Data) 0 Cowichan 48° 50’ Canada, Fisheries and Marine Service (Unpublished Data) 10 Washington/ Oregon Columbia 46° 10’ Rich 1925 22. Sixes 42° 50’ Reimers 1971 12 California Klamath 41° 30’ Snyder 1931 14 (Monterey Bay) 36° 40’ Snyder 1931, Rich 1925 11 1983 nook with the adjacent Nass River at 55° 20’ N which has only 42% stream-type chinook. In fact, all rivers south of the Stikine are dominated by ocean-type chinook except for the Yakoun, while all those north of the Nass are virtually 100% stream-type. South of the Stikine, most larger rivers (Nass, Skeena, Fraser, Columbia, Klamath) have a substantial percentage of stream-type chinook while the smaller rivers tend to be primarily ocean-type. Rather than a simple latitudinal cline, these data suggest a discontinuity in the coastal distribution of the two life-history types at the latitude of the Nass and Stikine Rivers. North of the Nass only stream- type chinook occur. South of the Stikine ocean-type fish predominate with stream-type being most suc- cessful in the headwater tributaries of the larger rivers (Table 2). Temperature and Fresh-water Residence Rich (1925) inferred that northern rivers and head- water tributaries have a more severe environment than southern rivers and lowland tributaries, resulting in slower growth of chinook fry and longer stream residence. The only indicator of environmental sever- ity available for the rivers considered in this paper is temperatures collected at streamflow guaging sites (Environment Canada 1977). These data provide an indication of the length of the growing season and the temperature regime during the growing season, two important factors affecting whether juvenile chinook can reach smolting size during their first summer. The temperature in rivers having 100% stream-type chinook (Yukon, Taku, Stikine) is not uniformly lower than in rivers having a high proportion of ocean-type chinook [Nass, Skeena, Yakoun, Kitimat, Bella Coola] (Table 3). Average annual temperatures inthe Yukon, Taku and Stikine are only slightly lower than in the Nass, Skeena and Kitimat. Temperatures HEALEY: MIGRATION PATTERNS OF CHINOOK SALMON 429 in the Yukon and Stikine during the summer growing season (June-August) are warmer than in the Nass, Skeena, Kitimat or Bella Coola (Table 3). The Yakoun, which has a high proportion of stream-type chinook, is much warmer throughout the year than either the Kitimat or Bella Coola, which have low proportions of stream-type chinook. The temperature data available for these rivers may not be directly representative of the temperatures in the microhabi- tats where juvenile chinook are found, so a detailed statistical analysis of these data is not warranted. Nor is temperature necessarily the best environmental sev- erity indicator to correlate with the abundance of stream-type chinook. Nevertheless, the absence of a consistent relationship between river temperature, a factor commonly associated with fish growth and production, and the relative abundance of stream- type chinook, weakens the argument that length of stream residence is mainly environmentally determined. Ocean Distribution of Stream- and Ocean-type Chinook Salmon Information on the ocean distribution and migra- tion patterns of stream- and ocean-type chinook comes from a variety of sources. Healey (1980a,b), Hartt (1980), and Argue (1970) present information on the occurrence of first ocean year chinook in coas- tal waters. Samples from the commercial troll fishery, which operates outside the surf line, provide informa- tion on the occurrence of older chinook within about 80 km of the coast (Milne 1964; Ball and Godfrey 1968b, 1969, 1970; Wright et al. 1972; Parker and Kirkness 1956). River mouth gillnet fisheries which Operate in the estuaries of the major rivers provide information on the occurrence of maturing chinook near the river mouths (Godfrey 1968; P. F. Starr, TABLE 2. Summary of characteristics distinguishing stream- and ocean-type Chinook Salmon. Characteristic Distribution of spawning popula- tions. Timing of spawning runs. Ocean distribution and migration. Stream-Type 100% of runs to rivers. north of the Nass. In the Nass and rivers to the south dominant in headwater spawning populations. Dominant in early runs. Smolts move offshore during first summer at sea. Dominant in offshore waters throughout marine life. Ocean-Type Dominant in the Nass and all rivers to the south. Dominant in downstream spawning populations. Dominant in late runs. Smolts remain in inside sheltered waters during first summer at sea. Dominant in onshore waters throughout marine life. 430 THE CANADIAN FIELD-NATURALIST Vol. 97 TABLE3. Monthly and annual average temperature (° C) of selected Chinook Salmon spawning rivers. Source, Environment Canada 1977. River Bella Month Yukon Taku Stikine Nass Skeena Yakoun Kitimat Coola January 0.0 0.1 0.0 0.3 0.2 1.5 0.4 0.5 February 0.0 0.0 0.0 0.2 0.3 15) 1.7 1.1 March 0.0 0.3 0.0 1.4 0.6 2.6 1.5 2.8 April 0.1 2.8 0.1 49 4.0 5,22 3.7) 7.4 May 4.8 S72 5.8 WD 6.0 7.8 5.3 8.8 June 9.1 8.7 8.6 V2 7.6 13.2 6.3 9.6 July 13.8 11.4 12.8 9.0 12.1 14.3 10.3 9.0 August 12.9 9.5 12.4 10.4 12.1 17.2 10.0 10.0 September 8.6 8.9 93 9.3 10.0 WA Q.7/ 7.4 October 3.3 7 2.8 4.0 6.0 10.0 6.6 7.0 November BS 1.8 0.0 2.6 1.6 See 1.5 8,5) December 0.0 —- 0.5 1.0 0.7 3.0 1.8 2.0 ALL 4.7 4.4 4.4 4.8 5, 7.8 4.9 5.8 Percent Stream-type Chi- nook Salmon 100 100 100 42 48 S7/ 12 14 C. L. Cross, and F. J. Fraser 1981 unpublished ms). Information on the high seas distribution of chinook comes from both published and unpublished data of the International North Pacific Fisheries Commission (Manzer et al. 1965; Vic Aro, personal communica- tion, 1980). While none of these data are complete or convincing alone, together they provide a reasonable picture of the occurrence and migratory behaviour of the two life history types in the ocean. The timing of seaward migration of stream- and ocean-type chinook juveniles is similar. Stream-type smolts generally migrate in April and May, slightly later than recently emerged ocean-type fry migrants but before the downstream movement of ocean-type smolts (Major and Meghell 1969; Healey 1980a and unpublished data). First ocean year stream-type chi- nook are present in the Strait of Georgia, a major juvenile salmon nursery area in southern British Columbia, for only about two months following downstream migration. Ocean-type chinook, how- ever, remain abundant in the Strait of Georgia throughout their first and second years of ocean life, and contribute heavily to the local sport and commer- cial fisheries of this region (Healey 1980a; Argue and Marshall 1976). In Juan de Fuca Strait, which con- nects the Strait of Georgia with the open Pacific, Argue (1970) observed a peak in the catch of first ocean year chinook during September, most of which were stream-type. Thus, the disappearance of stream- type juveniles from the Strait of Georgia is followed by an increase in their abundance further seaward in Juan de Fuca Strait, suggesting an outmigration of stream-type juveniles from the Strait of Georgia in late summer. Most of the ocean-type juveniles appear to remain behind in the Strait of Georgia. Hartt (1980) sampled by purse seine throughout the eastern north Pacific, and captured first ocean year chinook both nearshore and offshore. Greatest catches were in July, and in all months most fish were caught within 40 km of the shore. Of 253 fish whose ages were determined, 245 were stream-type (Colin Harris, personal communication 1979). Fifty of the fish had two fresh water annuli and these were a greater proportion of the catch north of 55° N than south (31.5% compared with 11.4%). Most of the fish with two fresh water annuli were probably from rivers in Alaska, where fresh-water residence times of more than a year have been observed. The other stream- type fish were presumably from populations all along the coast. These observations indicate that many, if not most, ocean-type chinook spend their first ocean year in sheltered inside waters, or very close to shore. Stream- type chinook, on the other hand, begin to move off- shore in July or August and during their first ocean year are most abundant along the open coast outside the surf line (Table 2). The contribution of the two life-history types to catches on the high seas, in the commercial troll fishery, and in the river mouth gillnet fisheries shows that differences in ocean distribution also exist among fish at sea for more than one year. Manzer et al. (1965) reported that a sample of 847 chinook, mostly in their second and third ocean years, captured east of 170° E in 1960 by the Japanese mothership fishery, were all stream-type. About 10% of the fish had two fresh 1983 water annuli. This sample might, however, have included a considerable number of Asian chinook. Eighty chinook in their second to fifth ocean years captured on the high seas north of 45° N and east of 170° W during 1961-1967 by Canadian research ves- sels were 65% stream-type (Vic Aro, personal com- munication 1980). These fish were almost certainly from North American populations. Since stream-type chinook constitute about 25% of all chinook in spawn- ing populations from the Columbia River to southeast Alaska, the percentage of stream type chinook in the high seas catches is greater than expected. The high proportion of stream-type chinook in the high seas catch could be due to most samples coming from north in the Gulf of Alaska coupled with the possibility that ocean-type chinook have a southerly distribution in keeping with the southerly distribution of their spawning populations. If this were the case, a higher proportion of ocean-type chinook should be found in the most southerly areas sampled onthe high seas. The total catch by Canadian research vessels was equally distributed north and south of 49° N and equal numbers of stream- and ocean-type chinook were captured north and south of 49° N. A high pro- portion of stream-type fish occur in offshore waters regardless of latitude (Table 2). Closer to the coast the situation is reversed. The proportion of stream-type chinook in the British Columbia commercial troll fishery ranged from 3.4% in the Strait of Georgia to 20.6% in the Queen Char- lotte Islands area during the 1960’s (Table 4). Off Washington and southeast Alaska, troll catches of stream-type chinook were similar, at 15 and 15.7% HEALEY: MIGRATION PATTERNS OF CHINOOK SALMON 431 respectively. The proportion of stream-type chinook in all troll fishery samples since 1940 has been less than 21% (usually below 15%) in all regions of the coast. This is low compared with the average proportion of 25% stream-type fish in the spawning populations. Stream-type chinook undertake their spawning migration earlier on average than ocean-type chi- nook. One possible explanation for the low propor- tion of stream-type chinook in the coastal troll fishery is the early departure of maturing stream-type fish from the fishing grounds. However, stream-type chi- nook are a low proportion of both the immature and maturing fractions of the troll catch (Ball and Godfrey 1968a, b, 1969, 1970). Early departure of maturing stream-type chinook from the fishing grounds is not, therefore, a sufficient explanation of the low percen- tage of stream-type fish in the troll catch. Before 1940 the percentage of stream-type chinook in the troll fisheries along the British Columbia coast was significantly higher, ranging from 17.5-28.0% (t= 3.10, P< 0.05) (Table 4). Several factors may explain this apparent decline in the contribution of stream-type chinook to the troll catch. Stream-type fish may have constituted a higher percentage of all chinook populations in the past. There is evidence that stream-type chinook have declined in abundance in the Columbia, Sacramento, and other major rivers due to human activities such as dam building and hydraulic mining. Stream-type fish might also be less able to withstand intensive fishing pressure than ocean-type and may have declined relative to them in recent years due to increased fishing pressure. Finally, prior to 1940, trollers may have fished closer to river TABLE 4. Percent of stream-type Chinook Salmon in the coastal troll catch and the rivermouth gillnet catch during several decades Data Decade Fishery Source 1911-20 1921-30 1941-50 1951-60 1961-70 1971-80 Troll S.E. Alaska I IS. 7/ Northern B.C. 2 23.0 5.6 20.6 Central B.C. D 9:3 Vancouver Island 2 20.0 12.4 10.9 3.9 Georgia Strait DZ. 28.0 We 6.5 3.4 Washington State 3 IS)s33 Gillnet Fraser River 4 34.4 42.8 12.8 Skeena River 4 48.1 28.1 Nass River 4 46.2 Data Source: |. Parker and Kirkness 1956. 2. Milne 1964; Ball and Godfrey 1968b, 1969, 1970. 3. Wright et al. 1972. oe . Godfrey 1968; Ball and Godfrey 1968a; P. F. Starr, C. L. Cross and F. J. Fraser 1981: unpublished ms; Ginetz 1976. 432 mouths were maturing stream-type fish are relatively abundant (see below). Unfortunately, no suitable river mouth or high seas catch data are available for these early years to determine whether stream-type chinook were more abundant in all regions prior to 1940. In river mouth gillnet fisheries the proportion of stream-type chinook is high (Table 4). Although the gillnet fisheries occur in rivers known to have a high percentage of stream-type fish, their abundance in the net catches shows that these fish are highly available in the river mouths even though they were not further offshore in the troll fishing areas. These data suggest a predominantly coastal distri- bution of ocean-type chinook and an offshore distri- bution of stream-type chinook throughout their lives (Table 2). The low proportion of stream-type chinook in the troll fisheries further suggests that maturing stream-type fish complete their onshore migration quickly and are available for only a short time to the coastal troll fisheries. Following their rapid onshore migration, maturing stream-type chinook delay inthe river estuaries for some time prior to upstream migra- tion, at which time they are available to the river mouth gillnet fisheries. Discussion The data available show consistent differences between stream- and ocean-type chinook in adult run timing, in the distribution of their spawning popula- tions both along the coast and within river systems, and in their ocean distribution and migration behav- iour (Table 2). These differences cannot easily be explained on environmental grounds. Water tempera- ture, as an example of the kind of environmental factor which might affect fresh-water residence, does not account for the absence of ocean-type chinook north of the Nass River or for the high proportion of stream-type chinook in the Yakoun River. Also, fresh-water environmental conditions that might result in a longer fresh-water residence seem unlikely to produce different oceanic distributions and adult run timings. In view of these difficulties, it seems unlikely that stream-type chinook are a phenotypic variant produced by fresh-water rearing conditions. Instead, it seems likely that stream- and ocean-type chinook represent genetically unique races as reflected by their different behaviour patterns. A number of studies point to genetic segregation between chinook spawning populations. Although I know of no experiments dealing specifically with stream- and ocean-type chinook, Ricker (1972) cites studies on the Columbia and Sacramento Rivers in which spring run chinook (presumably mostly stream- type) were transplanted to tributaries having natural THE CANADIAN FIELD-NATURALIST Vol. 97 runs of fall chinook (presumably mostly ocean-type) and vice versa. Transplanted spring chinook consist- ently returned in the spring and fall chinook in the fall regardless of the characteristics of the stream to which they were transplanted. Adult run timing, therefore, appears genetically determined. More recently, Kris- tiansson and McIntyre (1976) found that the fre- quency of certain polymorphic enzymes in spring run Columbia River chinook was significantly different from fall run chinook. These results provide further support for genetic separation of the two life-history types. Both Ricker (1972) and Kristiansson and McIn- tyre (1976), however, also present evidence for signifi- cant genetic differences among fall run populations of chinook. Simply to demonstrate that stream- and ocean-type chinook differ genetically does not consti- tute sufficient evidence for racial separation, although this is a necessary condition. Further investigation is required to determine, unequivocally, that these life history types do constitute separate races. Should racial classification of these life-history types prove correct, the question of their origin and maintenance arises. Various possibilities suggest themselves. The two life-history types may have arisen through allopatric evolution during the last glaciation with their present distributions representing post- glacial colonization of suitable habitats, or the life- history types may have arisen through a process of disruptive selection in the larger river systems. Under this scenario, spring and summer adult migration would be favoured in upriver areas because the rivers are easier to ascend to the headwaters during spring and summer high water levels, and longer fresh-water residence would be favoured because larger smolts are better able to undertake the longer downstream migration. Maintenance of the separate life-history patterns could result from behavioural or chemical isolating mechanisms during spawning. Physical sep- aration of the life-history types at spawning 1s not universally the case, so that maintenance of the life- history patterns cannot result simply from allopatry and precise homing. As a final alternative, the two life-history patterns may exist in a balanced genetic polymorphism. Such a polymorphism would provide additional survival opportunities to a population liv- ing in an uncertain environment. A thorough consid- eration of these, and perhaps other, alternatives must, however, await a more formal analysis of the genetic relationships between stream- and ocean-type chinook. Acknowledgments I am indebted to several people who provided me with unpublished information on various aspects of our chinook populations: in particular, Vic Aro, 1983 Colin Harris, and Alan Hartt. Dr. Brian Riddell and Dr. Kees Groot criticized a draft of the manuscript. Literature Cited Argue, A. W. 1970. A study of factors affecting exploita- tion of Pacific salmon in the Canadian gauntlet fishery of Juan de Fuca Strait. Canada, Department of Fisheries and Forestry. Technical Report 1970-11. 259 pp. Argue, A. W., and D. E. Marshall. 1976. Size and age of chinook and coho salmon for subdivisions of the Strait of Georgia troll fishery, 1966. Environment Canada, Fisher- ies and Marine Service. Technical Report PAC/T-76-18. 175 pp. Ball, E. A. R.,andH. Godfrey. 1968a. Lengths and ages of chinook salmon taken in the British Columbia troll fishery and the Fraser River gillnet fishery in 1966. Fisheries Research Board of Canada. MS Report 954. 143 pp. Ball, E. A. R.,andH. Godfrey. 1968b. Lengths and ages of chinook salmon taken in the British Columbia troll fishery in 1967. Fisheries Research Board Canada MS Report 998. 79 pp. Ball, E. A. R., and H. Godfrey. 1969. Lengths and ages of chinook salmon taken in the British Columbia troll fishery in 1968. Fisheries Research Board of Canada MS Report 1073. 41 pp. Ball, E. A. R., and H. Godfrey. 1970. Lengths and ages of chinook salmon taken in the British Columbia troll fishery in 1969. Fisheries Research Board of Canada MS Report 1121. 191 pp. Environment Canada. 1977. Water Temperatures: British Columbia and Yukon Territory. Environment Canada, Inland Waters Directorate, Water Survey of Canada, Vancouver. Volumes 1-4. 1874 pp. Gilbert, C. H. 1913. Age at maturity of the Pacific coast salmon of the genus Oncorhynchus. Bulletin of the U.S. Bureau of Fisheries 32: 1-22. Godfrey, H. 1968. Ages and physical characteristics of maturing chinook salmon of the Nass, Skeena and Fraser Rivers in 1964, 1965 and 1966. Fisheries Research Board of Canada MS Report 967. 38 pp. Hartt, A.C. 1980. Juvenile salmonids in the oceanic ecosystem- the critical first summer. Pp. 25-58 in Sal- monid ecosystems of the north Pacific. Edited by W. J. McNeil and D. C. Himsworth. Oregon State University Press, Corvalis, Oregon. Healey, M.C. 1980a. Utilization of the Nanaimo River estuary by juvenile chinook salmon, Oncorhynchus tsha- wytscha. Fishery Bulletin 77: 653-668. Healey, M. C. 1980b. The ecology of juvenile salmon in Georgia Strait, British Columbia. Pp. 203-230 in Sal- monid ecosystems of the north Pacific. Edited by W. J. McNeil and D. C. Himsworth. Oregon State University Press, Corvalis, Oregon. Kissner, P. D. 1973. Annual progress report for a study of chinook salmon in Southeast Alaska. Alaska Department of Fish and Game, Division of Sport Fisheries, Study AFS-41-1. 24 pp. Koo, T.S. Y., and A. Isarankura. 1968. Objective studies of scales of Columbia River chinook salmon, Oncorhyn- HEALEY: MIGRATION PATTERNS OF CHINOOK SALMON 433 chus tshawytscha (Walbaum). Fisheries Bulletin 66: 165-180. Kristiansson, A. C., and J.D. McIntyre. 1976. Genetic variation in chinook salmon (Oncorhynchus tshawytscha) from the Columbia River and three Oregon coastal rivers. Transactions of the American Fisheries Society. 105: 620-623. Major, R. L.,andJ. L. Meghell. 1969. Egg-to-migrant sur- vival of spring chinook salmon (Oncorhynchus tshawyt- scha) in the Yakima River, Washington. Fishery Bulletin 67: 347-359. Manzer, J. I., T. Ishida, A. E. Peterson, and M. G. Hana- van. 1965. Salmon of the north Pacific Part V. Offshore distribution of salmon. International North Pacific Fisheries Commission Bulletin 15. 452 pp. Meehan, W.R., and D.B. Siniff. 1962. A study of the downstream migrations of anadromous fishes in the Taku River, Alaska. Transactions of the American Fisheries Society 99: 399-407. Milne, D. J. 1964. Sizes and ages of chinook (Oncorhyn- chus tshawytscha) and coho (O. kisutch) salmon in the British Columbia troll fisheries (1952-1959) and the Fraser River gill-net fishery (1956-59). Fisheries Research Board of Canada MS Report 776. 36 pp. Parker, R. R., and W. Kirkness. 1956. King salmonand the ocean troll fishery of Southeastern Alaska. Alaska Department of Fish and Game, Juneau. Research Report 1. 64 pp. Reimers, P. E. 1971. The length of residence of juvenile fall chinook salmon in Sixes River, Oregon. Ph.D thesis, Oregon State University 99 pp. Rich, W. H. 1925. Growth and degree of maturity of chi- nook salmon in the ocean. Bulletin of the U.S. Bureau of Fisheries 41: 15-90. Ricker, W. E. 1972. Heredity and environmental factors affecting certain salmonid populations. Pp. 19-160 in The stock concept in Pacific salmon. Edited by R. Simon and P. Larkin. H. R. McMillan Lectures in Fisheries. Univer- sity of British Columbia, Vancouver, British Columbia. Schutz, D. C. 1975. Rivers Inlet chinook sport fishery, 1971- 1974. Environment Canada, Fisheries and Marine Service Technical Report PAC/T-75-9. 24 pp. Snyder, J.O. 1931. Salmon of the Klamath River, Califor- nia. California Division of Fish and Game, Fishery Bul- letin 34. 130 pp. Tutty, B. D., and F. Y. E. Yole. 1978. Overwintering chi- nook salmon in the upper Fraser River system. Fisheries and Environment Canada, Fisheries and Marine Service MS Report 1460. 23 pp. Wright, S., R. Kolb and R. Brix. 1972. Size and age charac- teristics of chinook salmon taken by Washington’s com- mercial troll and ocean sport fisheries, 1963-1969. Pacific Marine Fisheries Commission Bulletin 8: 38-48. Yancey, R. M., and F. V. Thorsteinson. 1963. The king sal- mon of Cook Inlet, Alaska. U.S. Fish and Wildlife Service Special Scientific Report on Fisheries 440. 18 pp. Received 15 June 1982 Accepted 25 September 1982 Wintering Biology of Mourning Doves, Zenaida macroura, in Ontario EDWARD R. ARMSTRONG! and DAVID L. G. NOAKES Department of Zoology, University of Guelph, Guelph, Ontario NIG 2W1 'Present Address: Ministry of Natural Resources, P.O. Box 730, Cochrane, Ontario POL 1CO Armstrong, Edward R., and David L.G. Noakes 1983. Wintering biology of Mourning Doves, Zenaida macroura, in Ontario. Canadian Field-Naturalist 97(4): 434-438. Mourning Dove, Zenaida macroura, populations in Ontario contain both migratory and non-migratory birds. Wintering doves often suffer damage to their feet, and moulting of immature and second-year doves declines or ceases during winter. Doves are overwintering in Ontario in increasing numbers, probably as a result of increasing availability of food supplies. Greatest early winter densities are found along northern Lake Erie and southern Lake Huron. Key Words: Mourning Dove, Zenaida macroura, wintering, moult, Ontario The Mourning Dove (Zenaida macroura) was orig- inally migratory in Ontario and other northern areas -of its range (Cole 1933; Saunders and Dale 1939). A changing food supply has been suggested as responsi- ble for recent increases in overwintering of doves in northern areas (Hennessy and Van Camp _ 1963; Richards 1966; Alison 1976; Freedman and Riley 1980). During winter, doves in agricultural areas of Ontario feed almost exclusively on waste or stored corn (Zea mays) (Armstrong and Noakes 1981). There is no Open hunting season on Mourning Doves in Ontario. As part of an ecological study of Mourning Doves in southern Ontario (Armstrong 1977), we investigated the relatively recent advent of overwinter- ing in the Ontario Mourning Dove population. We were specifically interested in assessing recent trends in the number of doves overwintering, and relating this to possible food sources. We also investigated the implications of overwintering to moulting activity and frost damage to the feet. Study Area and Methods The major study area was Norfolk County on Lake Erie, with some work in Wellington and Huron coun- ties to the north and northwest respectively. Three- quarters of the 100 000 ha of farm land in Norfolk County is in agricultural crops, mainly grain corn, winter wheat ( 7riticum aestivum) and tobacco ( Nico- tiana spp.) (Ontario Ministry of Agriculture and Food 1975). There are about 20 000 ha of woodland. The median period of snow cover is from 22 November to 30 March (Brown et al. 1974). Lowest mean monthly temperatures occur in January or February, being -5°C in 1975 and -9°C in 1976 (Environment Canada data, Simcoe, Ontario). During the winters of 1974-75 and 1975-76, feeding and roosting doves were collected for this and related studies. All doves were examined for frost damage to their feet, ranging from loss of one claw to complete loss of several digits. Doves with at least one buffy-tipped covert were classed as immature (before | January) or second-year birds (after | January) (Reeves et al. 1968). Doves lacking buffy-tipped coverts were similarly classified if the ninth and tenth primaries had smooth edges and buff-coloured fringes (Wight et al. 1967). Hatching dates were determined by moulting rates as reported for doves in Illinois (Hanson and Kossack 1963); that study was geographically closest to and therefore likely to be comparable to that of our study area. Ifa dove was not actively moulting its date of hatching was estimated by comparing the last moulted primary with known-age birds. The occurrence of suspended moult was also recorded, 1.e. when one primary had been completely replaced while the next had not yet dropped (Mead and Watmough 1976). Fourteen immature doves, captured live in August 1975, were maintained in an outdoor aviary with shel- ter at the University of Guelph during the following winter. Although maintained principally for other studies, they were examined at intervals for damage to the feet. Christmas Bird Count data from 1970-73 were con- sulted to determine the early winter distribution of doves in southern Ontario and adjacent states by plot- ting the average number of doves seen per 10 party hours (Bystrak 1971). Christmas Bird Count data were also compiled for 1950 through 1974 for St. Thomas, Elgin County, to assess increases in winter- ing doves on the north shore of Lake Erie (Audubon Field Notes 1951-70; American Birds 1971-75; e.g. see Auckland 1975). Data on the annual area planted to grain corn in Norfolk County (e.g. Ontario Ministry of Agriculture and Food 1975) were also compiled for the same period and compared with the wintering dove count by correlation analysis. Tests of association between occurrences of both suspended moult and damaged feet and time of year was done by chi-square tests. Expected frequencies 434 1983 were calculated from the assumption that the row and column classifications were not associated. Results There were significant seasonal differences in the proportion of adult doves (which would have included second-year doves that had completed moult) with damaged feet, with the highest incidence during January-March (Figure 1; x2= 10.66, p<0.05). There was no damage to the feet of immature doves in summer or fall, but four of seven known second-year birds collected during February and March had such injuries. None of the second-year birds in the aviary had this damage in early January, but by March five of seven doves did. During summer most immature doves were moult- ing One to three primaries simultaneously; during fall and winter only one primary was replaced at any one 100 x eZ 23 9 a 2 w ‘80 (&} < = < a = 7) (e) i ~ 60 z Ww < 30 a 23 fo) a re fo) z ° Ee 20 (oa [e) a (e) x oO 0 Jan.-March April-June July-Sept. Oct.-Dec. SEASON FiGure |. Trimonthly changes in the proportion of adult Mourning Doves with apparent frost damage to the digits. ARMSTRONG AND NOAKES: WINTERING BIOLOGY OF MOURNING DOVES 435 TABLE 1. Occurrence of suspended primary feather moult in immature* and second-year® doves. No. with No. with Moulting Suspended Primaries Moult? Period n (%) (%) June-September 10! 99 (98.0) 2 (2.0) October-November 24 18 (75.0) 6 (25.0) December-February 12 2 ((N®.7) 10 (83.3) March D 2 (100.0) 0 (0.0) “Includes all doves that had initiated primary feather moult, including live-captured birds on date of capture. ’Chi-square analysis of occurrence of suspended moult between June-September and October-February, x2 = 42.3, d.f. = 1, p< 0.005. time, and many doves had completely replaced one primary while the next one had yet to drop (Table 1). A chi-square test showed a significant association between this suspended moult and season (June - September versus October - February; p < 0.005). The hatching period determined by backdating from immature doves collected prior to | October was April through August, with a peak during May and June similar to nesting study results (Armstrong 1977). However, when data after | October were included there was a less-pronounced hatching peak in late June, and 24% of all hatching was estimated to have taken place between September and November. The greatest early winter concentration of doves in Ontario was along the northern shore of Lake Erie (including Norfolk County), continuous with similar populations in northern Ohio and southern Michigan (Figure 2). The Mourning Dove population on the north shore of Lake Erie has increased dramatically since the early 1950's, significantly correlated with increases in the area planted to grain corn (Figure 3; r= 0.55, p< 0.01). Discussion Frost damage to the feet is a common injury to doves wintering in northern areas (Thompson 1950; Hanson and Kossack 1963), reported for example in Ontario (Alison 1976), Ohio (Hennessy and Van Camp 1963) and Michigan (Nickell 1964). The highest proportion of such injuries occurred during winter, with a reduced proportion in spring as unaffected migrants joined the resident population. With refine- ment and larger sample sizes, the ratio of the propor- tion of summer to winter doves with frost damage may provide an estimate of the proportion of the local population that overwinters. Major assumptions include no regeneration of claws, and that doves with frost damage had wintered in the area. The initial estimate (33% overwintering) is comparable to the 436 THE CANADIAN FIELD-NATURALIST Vol. 97 0 DOVES < 1 DOVES 1-5 DOVES 5-20 DOVES 20 DOVES FIGURE 2. The early winter distribution of Mourning Doves in southern Ontario and neighbouring states. estimate of 32% resulting from banding studies in Ohio (Ruble and Urban 1977). Mourning Doves apparently lack adaptations to the cold possessed by Wood Pigeons (Columba palumbus), which have foot papillae that are thicker and more pointed in winter than in summer to reduce heat transfer and substrate contact (Lennerstedt 1975). The tarso-metatarsal temperature of Mourning Doves falls rapidly with cold exposure, and the toes are still exposed when perching (Bartholomew and Dawson 1954). The extreme loss of several toes apparently had little effect on the doves, although perching behaviour would be affected. The high proportion of immature doves with sus- pended moult indicated a cessation of moulting dur- ing the winter. This conclusion was confirmed by the discrepancy in nesting periods and peaks established by backdating all immature doves examined, and by backdating doves examined only before | October. The latter procedure corroborated concurrent nesting studies (Armstrong 1977), while results of the former method indicated an improbable and unsubstantiated nesting chronology with a delayed midsummer peak and a late fall extension. Similar delays in moult occurred in immature doves wintering in Wisconsin and Oklahoma (Thompson 1950; Ault et al. 1976). Suspended moult may occur in Mourning Doves as a result of severe environmental conditions, although we did not demonstrate a cause and effect relation- ship. Suspended moult would be advantageous for late-hatched doves, which are most likely to overwin- ter locally (Hanson and Kossack 1963; Hennessy and Van Camp 1963), and which would otherwise be com- pleting moult during late fall and winter. Metabolic rates of immature doves are higher than those of adults even without the additional cost of moult (Ivacic and Labisky 1973). Mourning Doves feed heavily upon waste or stored corn throughout the winter (Armstrong and Noakes 1981). The physiological consequences of this reliance on corn are uncertain, but immature wild doves feed- ing on corn during nonwinter months grew faster, weighed more, and had more body fat than those feeding on other foods (Hanson and Kossack 1957). The linear relationship between winter counts of doves and the amount of farmland planted to corn suggests that the increasing acreage in grain corn may be partially responsible for the increasing population of overwintering doves in southern Ontario. Other factors such as bird feeders probably are involved, as they have increased wintering populations of other birds (Bock and Lepthien 1976; Middleton 1977). The phenomenon of increasing Mourning Dove popula- tions is consistent with Cohen’s (1967) model of sea- sonal migratory behaviour, which predicts that the overwintering portion of a population will increase with the food supply and survival of the overwintering proportion. 1983 e—— —-e Corn AREA PLANTED TO CORN (x 1000 hectares) 1950-54 1956-59 YEAR ARMSTRONG AND NOAKES: WINTERING BIOLOGY OF MOURNING DOVES e———-e Doves 1960-64 437 DOVES SEEN PER TEN PARTY-HOURS 1965-69 1970-74 _ Figure 3. Recent trends in the area planted to grain corn in Norfolk County, and in the number of Mourning Doves seen per ten party-hours near St. Thomas, Elgin County. A successfully overwintering Mourning Dove pop- ulation could contribute substantially to local recruitment. Wintering doves avoid the hazards of migration and hunting, and are available to take advantage of favourable early spring conditions that may increase the length of the breeding season. Acknowledgments J. Baker, R. Cawthorn, R. Craig, D. Dennis, A. Garbutt, W. Mitchell, and R. Rolefson assisted in collecting specimens; D. Joyner, L. March, and A. L. A. Middleton provided advice; the Canadian Wildlife Service provided funding; several Ontario Ministry of Natural Resources’ staff provided advice; numerous landowners allowed access to their prop- erty for field work; and contributors to Christmas Bird Counts provided valuable information. The assistance of all those mentioned is acknowledged with thanks. Literature Cited Alison, R. M. 1976. Mourning Doves wintering in Ontario. Canadian Field-Naturalist 90: 174-176. Armstrong, E. R. 1977. Reproductive biology and ecology of the Mourning Dove in southern Ontario. M.Sc. thesis, University of Guelph, Guelph, Canada. 140 pp. Armstrong, E. R.,and D. L. G. Noakes. 1981. Food habits of Mourning Doves (Zenaida macroura) in southern Ontario. Journal of Wildlife Management 45: 222-227. Auckland, L. (compiler). 1975. St. Thomas, Ontario. American Birds 29: 197. Ault, J. W., III, V. J. Heller, J. C. Lewis, and J. A. Morri- son. 1976. Delayed molt of primary feathers of Mourning Doves during winter. Journal of Wildlife Management 40: 184-187. Bartholomew, G. A., Jr., and W. R. Dawson. 1954. Body temperature and water requirements in the Mourning Dove. Ecology 35: 181-187. Bock, C. E., and L. W. Lepthien. 1976. Changing winter distribution and abundance of the Blue Jay. American Midland Naturalist 96: 232-235. 438 Brown, D.M., G.A. McKay, and L.J. Chapman. 1974. The climate of southern Ontario. Environment Canada, Atmospheric Environment, Climatological Stu- dies no. 5. 50 pp. Bystrak, D. 1971. Howto prepare a winter range map from Christmas Bird Count data. American Birds 25: 952-955. Cohen, D. 1967. Optimization of seasonal migratory behavior. American Naturalist 101: 5-17. Cole, L. J. 1933. The relation of light periodicity to the reproductive cycle, migration, and distribution of the Mourning Dove (Zenaidura macroura carolinensis). Auk 50: 284-296. Freedman, B., and J. L. Riley. 1980. Population trends of various species of birds wintering in southern Ontario. Ontario Field Biologist 34: 49-68. Hanson, H. C., and C. W. Kossack. 1957. Weight and body fat relationships of Mourning Doves in Illinois. Journal of Wildlife Management 21: 169-181. Hanson, H. C., and C. W. Kossack. 1963. The Mourning Dove in Illinois. Illinois Department of Conservation, Technical Bulletin 2. 133 pp. Hennessy, T.E., and L. Van Camp. 1963. Wintering Mourning Doves in northern Ohio. Journal of Wildlife Management 27: 367-373. Ivacic, D.L., and R.F. Labisky. 1973. Metabolic responses of Mourning Doves to short-term food and temperature stresses in winter. Wilson Bulletin 85: 182-196. Lennerstedt, I. 1975. Seasonal variation in foot papillae of Wood Pigeon, Pheasant and House Sparrow. Compara- tive Biochemistry and Physiology 51 A: 511-520. Mead, C. J.,and B. R. Watmough. 1976. Suspended moult of trans-Saharan migrants in Iberia. Bird Study 23: 187-196. THE CANADIAN FIELD-NATURALIST Vol. 97 Middleton, A. L. A. 1977. Increase in overwintering by the American Goldfinch, Carduelis tristis, in Ontario. Cana- dian Field-Naturalist 91: 165-172. Nickell, W. P. 1964. The effects of probable frostbite on the feet of Mourning Doves wintering in southern Michigan. Wilson Bulletin 76: 94-96. Ontario Ministry of Agriculture and Food. 1975. Agricul- tural Statistics for Ontario, 1974. Ontario Ministry of Agriculture and Food, Publication 20. 52 pp. Reeves, H. M., A. D. Geis, and F. C. Kniffen. 1968. Mourn- ing Dove capture and banding. United States Fish and Wildlife Service Special Scientific Report, Wildlife 117. 63 Pp. Richards, A. P. 1966. Some aspects of the ecology of the Mourning Dove in Amherst and Holyoke, Massachusetts. M.Sc. thesis, University of Massachusetts, Amherst, Mas- sachusetts, U.S.A. 99 pp. Ruble, P.,and D. Urban. 1977. Seasonal movements of the Ohio Mourning Dove. Ohio Department of Natural Resources, Ohio Fish and Wildlife Report 6. 14 pp. Saunders, W. E., and E. M. S. Dale. 1939. History and list of birds of Middlesex County, Ontario. Transactions of the Royal Canadian Institute 19: 161-251. Thompson, D. R. 1950. Foot-freezing and arrestment of post-juvenal molt in the Mourning Dove. Wilson Bulletin 62: 212-213. Wight, H. M., L. H. Blankenship, and R. E. Tomlinson. 1967. Aging Mourning Doves by outer primary wear. Journal of Wildlife Management 31: 832-835. Received 3 March 1982 Accepted 7 May 1983 Adaptation of Atlantic Salmon, Salmo salar, to a Restricted Freshwater Environment J.-P. CUERRIER P.O. Box 7196, Ottawa (Vanier), Ontario KIL 8E3 Cuerrier, J.-P. 1983. Adaptation of Atlantic Salmon, Sa/mo salar, to a restricted freshwater environment. Canadian Field-Naturalist 97(4): 439-442. In an experimental stocking of a restricted freshwater environment, Papin Lake, Pontiac County, Quebec, Atlantic Salmon fingerling (25-35 mm long) growth was remarkable because of the introduction and adaptation of smelts utilized by the landlocked salmon as forage food. After three summers, some salmon had reached 2.0-2.84 kg. In Papin Lake, salmon reached first sexual maturity after three summer growing seasons, age 2+. No natural propagation was successful due to the absence of an inflowing stream. The last catch of a landlocked salmon in Papin Lake was made at the end of June 1981. The specimen was 8 years old and had been stocked in 1973. Key Words: Atlantic salmon, Sa/mo salar, landlocked salmon, growth, sexual maturity, smelt. Des fretins de saumon atlantique mesurant 25-35 mm ont été ensemencés dans un milieu dulcicole, le lac Papin, comté de Pontiac, Québec. La croissance a été remarquable grace a l’introduction et adaptation d’éperlans utilisés par le saumon dulcicole comme nourriture. Aprés trois étés, des saumons avaient atteint 2.0 a 2.84 kg. Au lac Papin, le saumon a atteint sa premiere maturation sexuelle aprés trois étés de croissance, soit al’age de 2+ ans. Aucune propagation naturelle n’a été réussie a cause de l’'absence d’une riviére de charge. La derniére prise d’un saumon dulcicole au lac Papina été effectuée a la fin de juin 1981. Ce spécimen était 4gé de 8 ans, ensemencé en 1973. Mots clés: Saumon atlantique, Sa/mo salar, saumon dulcicole, croissance, maturité sexuelle, éperlan. Landlocked salmon, also called freshwater salmon and ouananiche, occur in several lakes of North American states: Maine (Havey and Warner 1970), New Hampshire, Vermont (Lake Champlain) and New York (Anonymous 1977, 1981); in Trout Lake, near North Bay, in several other Ontario lakes (Scott and Crossman 1973; MacCrimmon and Gots 1979); in Memphremagog, Mégantic, Musquaro (Bryant 1964), Tremblant (Blais et Legendre 1976; Legendre 1980), Sorcier, St. John (Saguenay) lakes; in several lakes along the North Shore of the St. Lawrence River in Québec; and in several lakes in Newfoundland (including Labrador). Some occurrences result from such natural pheno- mena as the ouananiche and sebago; others are man- made introductions such as the landlocked salmon. For a detailed account on the nomenclature of anadromous (sea) salmon, freshwater salmon, land- locked salmon, lake salmon, ouananiche and sebago, and for a detailed list on the distribution of salmon in freshwater habitats, readers are referred to Mac- Crimmon and Gots (1979). The adaptation of landlocked salmon to a fresh- water environment offers an interesting challenge for biologists both as a management tool in the use of forage minnow and sucker populations for a fast growing fish and in providing angling opportunities (Legendre 1967). Authorities in the United States have comprehensive projects underway on landlocked 439 salmon (Anonymous 1967 and 1981; Pfeiffer 1979; Ingham 1981). This paper reports on experimental stockings of Atlantic salmon fingerlings, which started in 1961, in Papin Lake, Township of Auray, County of Pontiac, Québec (76°06’N, 46°09’W). Stockings On the basis of a policy of diversification in the availability of game fish in the former Pontiac Game Club waters, Papin Lake, with an area of nearly 160 ha, was selected for an experimental adaptation of Atlantic salmon because it seemed to offer suitable ecological conditions (oligotrophic) and because it was uninhabited by other species of game fish. The lake has a maximum depth of 33 meters. Data on the vertical distribution of temperature and dissolved oxygen on 23 August 1967 were as follows: Depth Temperature Dissolved Oxygen (Meters) (KES) (mg/ 1) 0 20.0 6.1 18.9 8.0 6.7 14.5 7.0 Hill 9.1 8.4 2 6.7 8.5 18.3 6.1 7.0 Records on stockings carried out between 1961 and 1973 in Papin Lake are presented in Table 1. 440 THE CANADIAN FIELD-NATURALIST Vol. 97 TABLE |. Stocking records of Atlantic Salmon, Sa/mo salar, and Rainbow Smelt, Osmerus spectrum, in Papin Lake, Quebec Date Species Number Size Origin of Eggs 1961, June Atlantic Salmon 4 000 35 mm Tadoussac Hatchery 1965, May Smelt Millions Eyed Eggs Meach Lake, Gatineau Park 1966, May Smelt Millions Eyed Eggs Meach Lake, Gatineau Park 1966, May Atlantic Salmon 10 000 25-35 mm Gaspé Hatchery 1971, June Atlantic Salmon 4 000 35 mm Tadoussac Hatchery 1973, May Atlantic Salmon 10 000 25-35 mm Tadoussac Hatchery On 24 June 1961, four thousand salmon fingerlings (about 25-35 mm in length) were planted in Papin Lake. They originated from eyed eggs obtained from the Tadoussac Hatchery, Tadoussac, Québec, where Atlantic salmon entering the Saguenay River are col- lected and stripped for fish culture purposes. Eyed eggs were hatched and fry raised at the Ripon Hatchery, Québec. The fingerlings were fed with dry feed for a month and a half prior to stocking. During the spring of 1965, several million eyed smelt eggs (free of G/ugea), taken froma creek flowing into Meach Lake, Gatineau Park, Québec, were released in shallow waters along a sandy beach at the south end of Papin Lake. The release of smelt eyed eggs was repeated in the spring of 1966. The objective was to establish a population of smelt that would provide a suitable forage food for salmon, a policy for years adopted in the states of Maine and New Hamp- shire and in Québec (Legendre 1967a). It was feared that the minnows (Cyprinidae) and suckers (Catos- tomidae) would drop in abundance and would not sufficiently sustain heavy predation by salmon. In fact, large schools of minnows can no longer be observed in shallow waters. A smelt population is now well established in Papin Lake. A statement by Cooper (1940, p. 103) is quoted here: “These rather limited data on food habits of the salmon substan- tiated the already established and well known fact that the smelt is the chief food of landlocked salmon in Maine lakes”. DeRoche (1976, p. 21) wrote similar comments in his study of Sebago Lake, Maine. The smelt population in Meach Lake originated from Utopia Lake, New Brunswick. This pigmy smelt is now classified as Osmerus spectrum Cope, 1870 (Lanteigne and McAllister 1982). Also during the spring of 1966, ten thousand Atlan- tic salmon fingerlings originating from eggs obtained from the Gaspé Hatchery, Gaspé, Québec, were released in Papin Lake. In order to avoid complications in the identification between salmon which might originate from natural propagation and those originating from ulterior hatchery sources, no further plantings of salmon were carried out until the spring of 1971, when four thou- sand fingerlings were released. During the spring of 1973, ten thousand salmon fingerlings were again released. In both cases, the eggs were obtained from the Tadoussac Hatchery. Results Salmon caught by angling and by gill netting are presented in Table 2. Because angling activities during the early sixties were very limited and unsuccessful, sampling with gill nets was carried out on 2 October 1964, and resulted in the catch of five salmon (speci- mens | to 5 inclusive in Table 2). The first salmon caught by an angling Club member was reported on 19 May 1965. During 1966, several salmon, then five years old, were reported: a) 5.0 kg: caught by angling; b) 8.2 kg: found dead along the shore by a Club member; and c) 7.5 and 7.3 kg: both caught with gill nets. Although the salmon caught by gill netting between 1964 and 1968 inclusive were mature and in spawning conditions (see sex and stage of maturity in Table 2), there was no evidence by the fall of 1968 that the landlocked salmon in Papin Lake spawned success- fully. Observations were carried out between 25 October and 5 November 1968, to examine the stage of salmon gonads and the possible spawning areas. Data on the four specimens netted appear in Table 2 (speci- mens no. | 1-14 inclusive). Stomach contents of gill net catches revealed the presence of smelt ranging in length from 10.1 to 22.9 cm; the largest smelt specimen, being two years old, originated from the 1966 egg planting. Because salmon were still ripe, an intensive search was made of the shallow waters with a spotlight powered by a portable generator in an effort to locate salmon spawning areas. The entire shoreline, to a depth of 4m was scanned between 2100 and 0100 h. Minnows, smelts and darters (Percidae) were observed, but no indication of spawning activities by salmon was obtained. The inlet creek with several small channels had a very low flow; several dams were maintained by beavers. 1983 CUERRIER: ADAPTATION OF ATLANTIC SALMON 44] TABLE 2. Data on Landlocked Atlantic Salmon, Salmo salar, caught in Papin Lake, Québec Length Weight Specimen (Total) in Number Date cm kg ] 2 October 1964 72.4 4.3 2 2 October 1964 58.4 1.5 3 2 October 1964 69.8 3377 4 2 October 1964 67.3 3.2 5 2 October 1964 65.4 33,1 6 19 May 1965 TES 4.2 7 1966 2 5.0 8 16 July 1966 91.4 8.2 9 27 August 1966 83.8 Vo 10 27 August 1966 86.4 Ted} 1] 27 October 1968 65.0 2.8 12 27 October 1968 64.5 Del 13 27 October 1968 59.4 2.0 14 3 November 1968 58.4 2.4 15 2 July 1974 48.9 1.4 16 2 July 1974 27.3 0.3 17 18 May 1975 Soll 2.0 18 23 July 1977 76.2 37) 19 13 May 1978 66.0 ? 20 13. May 1978 ? 6.1 21 13 May 1978 63.5 ? 22 14 May 1978 78.7 4.9 23 14 May 1978 64.8 3.4 24 14 May 1978 76.2 4.6 25 14 May 1978 63.5 ? 26 16 May 1978 71.1 4.5 27 21 June 1981 78.7 6.8 Stage of Age in Catch Sex Maturity Years Method M Immature Bits Netting F Ripe 3+ Netting M Ripe Shr Netting M Ripe Bits Netting M Ripe 3+ Netting M y 4 Angling ? ? 5 Angling ? ? ? Found dead on shore M Ripe $+ Netting F Ripe Sale Netting M Ripe Zar Netting M Ripe Dar Netting M Ripe 2+ Netting F Ripe Dap Netting M 2 3+ Angling ? ? I+ Angling ? 2 D, Angling M 2 4+ Angling 2 2 5 Angling ? Q 5 Angling 2 ? v Angling M Adult 5 Angling F Adult 5 Angling M* Adult 5 Angling ? ? 5 Angling Fe Adult 5) Angling ? ? 8 Angling *Scales with spawning mark erosion in 1975, when this male salmon was 2+. **Specimen with 10 ova of previous (1977) spawning still in body cavity; also 36 smelts were counted in the stomach. After repeated examination of the scales, it was concluded that the four salmon caught in 1968 were 2+ years old, had completed three summer’s growing sea- sons, and, therefore, were all from the 1966 stocking. The length and weight of the four specimens caught by netting in 1968 are remarkable, comparable to those reached by Atlantic salmon after two years at sea that return to spawn as grilse in the Alma River, Fundy National Park, N.B. (Dadswell 1968). The landlocked salmon specimens caught in Papin Lake revealed that the spring fingerlings achieved excellent growth following planting. After three summers, some had reached between 2.0 kg and 2.84 kg. At five years, the weight of 7.3 kg had been obtained. The 7.3 kg female caught on 27 August 1966 (Table 2) had 11 200 ova attached to the ovarian tissues. This 7.3 kg female was to spawn in the fall. The 7.5 kg male caught on the same date was also sexually mature. The live weight of the 8.2 kg salmon found dead along the shore with the abdominal wall and the body cavity partly rotted and removed by scavenger animals might have been close to 9.1 kg. During our contact with angling activities in Papin Lake, 27 landlocked salmon were reported and/or examined. The last specimen was caught on 21 June 1981. This salmon, 6.8 kg in weight, was 8 years old and from the 1973 stocking. Conclusions Atlantic salmon resulting from hatchery plantings grew exceptionally well in Papin Lake and thrived on introduced smelts. The smelts planted at the eyed-egg stage maintained themselves in large quantities through natural propagation. The growth rate of landlocked salmon varied widely; the five-year-old salmon caught by angling or by gill netting ranged from 3.4 to 7.48 kg. Salmon reached sexual maturity when 2+ years old, at the end of the third summer growing season, and had weight increment rates similar to sea-run grilse. However, the landlocked salmon in Papin Lake, which does not have a large inlet stream, did not spawn successfully. Females got rid of their ova but no progeny were caught. 442 Until 1981, we concluded that landlocked salmon in Papin Lake did not live beyond their sixth summer’s growth. However, one salmon caught in Papin Lake, in 1981, was eight years old. The introduced salmon are believed to have disappeared from Papin Lake. The observations on smelt in the lake and in the stomachs of salmon demonstrated that the habitat is suitable for this species of fish, and that salmon made good use of this forage food. The addition of smelt as forage indicates it resulted in furthering the growth of salmon. The size of the two-year-old salmon caught in 1968 compared advantageously with the three-year- old salmon caught in 1964. Acknowledgments These investigations (including purchase of finge- rlings and aircraft transportation) and the publication were funded by the former Pontiac Game Club and the Canadian Sport Fishing Institute. The assistance of Dr. M. J. Dadswell, then graduate student, and Mr. G. Laver, technician, who at one time or another participated in the field work, is greatly appreciated. Vianney Legendre (retired director of the Research Division of the Service de la Faune for the province of Québec) provided valuable references and authorita- tive criticism with respect to the manuscript. Many of the scales samples have been examined and their age determination confirmed by Mr. Imre Babos, techni- cian with the Québec Salmon Research Centre. Their cooperation is greatly appreciated. Literature Cited Anonymous. 1977. A strategic plan for development of salmonid fisheries in Lake Champlain. New York State Department of Environmental Conservation FW-P 201(2/82). 20 pp. Anonymous. 1981. Fisheries programming in New York State Division of Fish and Wildlife, Department of Envir- onmental Conservation. 100 pp. Blais, J. P., and V. Legendre. 1976. La ouananiche Salmo salar du lac Tremblant, Québec. Ministére du Tourisme, Chasse et Péche, Québec. 102 pp. THE CANADIAN FIELD-NATURALIST Vol. 97 Bryant, N. 1964. Land of the landlocked. Field & Stream June: 63, 102-104. Cooper, G. P. 1940. A biological survey of the Rangeley - Lakes, with special reference to the trout and salmon. Fish Survey Report No. 3, Maine Department of Inland Fisheries and Game. 182 pp. Dadswell, M. J. 1968. Observations on Atlantic Salmon in the Upper Salmon (Alma) River, Fundy National Park. DeRoche, S. E. 1976. The Sebago Lake study. Maine Department of Inland Fisheries and Wiidlife, Fisheries Research Bulletin No. 9. 56 pp. Havey, K. A., and K. Warner. 1970. The landlocked sal- mon (Sa/mo salar), its life history and management in Maine. Maine Department of Inland Fisheries and Game. 129 pp. Ingham, W. 1981. Strategic plan for the management of landlocked salmon in New Hampshire. Part |. Fish and Game Commission, State of New Hampshire: 60 pp. Lanteigne, J.,. and D.E. McAllister. 1982. The pygmy smelt Osmerus spectrum Cope, 1870, a forgotten sibling species of eastern North American fish. Manuscript sub- mitted for publication. National Museum of Natural Resources. National Museums of Canada. 46 pp. Legendre, V. 1967. Eperlan: introductions pour ouana- niches. Service de la faune, Québec. Pp. 16, 22-24, 27 in Bulletin 11. Legendre, V., J.-R. Mongeau, J. Leclerc, et J. Brisebois. 1980. Les salmonidés des eaux de la Plaine de Montréal. 1. Historique, 1534-1977. Service de Paménagement et de Yexploitation de la faune, Québec. Rapport technique no. 06-27. 280 pp. MacCrimmon, H. R., and B. L. Gots. 1979. World distri- bution of Atlantic salmon, Sa/mo salar. Journal of the Fisheries Research Board of Canada 36(4): 422-457. Pfeiffer, M.H. 1979. A comprehensive plan for fish resource management within the Adirondack zone. New York State Department of Environmental Conservation. 207 pp. Scott, W. B., and E. S. Crossman. 1973. Freshwater fishes of Canada. Fisheries Research Board of Canada. Bulletin No. 184. 966 pp. Received 18 November 1982 Accepted 10 March 1983 Elk, Cervus elaphus, Foraging Related to Forest Management and Succession in Idaho LARRY L. IRWIN! and JAMES M. PEEK2 'Department of Zoology and Physiology, The University of Wyoming, Laramie, Wyoming 82071 2College of Forestry, Wildlife and Range Sciences, The University of Idaho, Moscow, Idaho Irwin, Larry L., and James M. Peek. 1983. Elk, Cervus elaphus, foraging related to forest management and succession in Idaho. Canadian Field-Naturalist 97(4): 443-447. Diets of Elk (Cervus elaphus) were correlated with availability of forage, indicating Elk were not selective of individual plant species. In spring, Elk preferred grasses and sedges in seral brushfields, forbs in clearcuts, and evergreen shrubs in timber stands. Insummer, they preferred shrubs in seral brushfields and in timber stands, but fed more often in clear-cuts on forbs. In fall, Elk fed upon low-growing, evergreen shrubs in dense timber stands. In winter, they preferred shrubs in seral brushfields. Early successional stages provide large amounts of forage and were preferred feeding areas in all seasons except fall and early winter. Elk are generalist feeders which maximize energy intake through mechanisms of habitat selection rather than food selection. Key Words: Elk, Cervus elaphus, forage selection, forest management, plant phenology, successional stage. Food habits studies of Elk (Cervus elaphus) indi- cate that diets vary according to season and dominant vegetation (Kufeld 1973). However, relationships of forage selection to forage availability and to succes- sional stages within an area are not well understood. This information is needed to better incorporate Elk habitat management with the intensive timber man- agement typical of northern Rocky Mountain forests. This paper reports a three-year study of Elk foraging patterns in relation to forage availability and forest succession in northern Idaho. It was undertaken to clarify relationships between foraging and cover characteristics. Study Area The study area was along the Coeur d’Alene River within the Idaho Panhandle National Forest. Precipi- tation averages about 120 cm annually, and snow fre- quently accumulates in depths greater than 60 cm. Climax vegetation associations, or habitat types (Daubenmire and Daubenmire 1968), include primar- ily Western Hemlock (Tsuga heterophylla)/ Myrtle Pachistima (Pachistima myrsinites) and Grand Fir (Abies grandis)/ Myrtle Pachistima. The driest slopes at the lowest elevations are occupied by Douglas-fir (Pseudotsuga menziesii)/ Mallow Ninebark (Physo- carpus malvaceus), while climax vegetation at the highest elevations includes Subalpine Fir (A bies /asi- ocarpa) or Mountain Hemlock ( Tsuga mertensiana), with understories of Rusty Menziesia ( Menziesia fer- ruginia) or Common Beargrass (Xerophyllum tenax). Several successional stages are included within these habitat types. About half of the study area ts covered by old growth Western Hemlock, Grand Fir, White Pine (Pinus monticola), and Western Larch 443 (Larix occidentalis). These forests have few natural openings and average 86% canopy closure, as mea- sured by a spherical densiometer (Lemmon 1957). The generally sparse understory includes Rocky Moun- tain Maple (Acer glabrum) and/or Pacific Yew (Taxus brevifolia). The remainder of the area has either been clearcut during the past 15 years or is in other successional stages following wildfires in 1910, 1919, and 1931. Some areas have burned 3 times. The clear-cuts are comprised chiefly of forbs and sedges, including Fire- weed Willow-herb ( Epilobium angustifolium) and Elk Sedge (Carex geyeri). Mesic slopes within the wildfire- burned areas are covered by mature timber stands or by dense thickets of pole-sized timber. These stands average 82% canopy closure. Drier slopes are covered by scattered pole timber and seral brushfields mainly composed of Scouler’s Willow (Salix scouleriana) and/or Snowbrush Ceanothus (Ceanothus velutinus). Breaks along the Coeur d’Alene River support a grass- shrub association which includes Elk Sedge, Blue- bunch Wheatgrass (Agropyron spicatum), Saskatoon Serviceberry (Amelanchier alnifolia), and Mallow Ninebark. Methods The entire area was sampled once per week via an extensive network of logging roads and trails, which were generally equally distributed throughout all vegetative types. In winter only the winter range, composed of shrubs and pole-timber stands, was sampled. Feeding site examinations (Cole 1956, Knowlton 1960) occurred where Elk were observed feeding or where fresh Elk tracks were visible. If there was evidence that deer (Odocoileus hemionus, O. vir- 444 ginianus) recently fed in the feeding site, no data were recorded. An instance of use was considered to be that portion of a plant removed during | bite. If Elk fed in two distinct, adjacent cover types, two feeding sites were recorded. Data were summarized by average aggregate percent use (Martin et al. 1946). Availability of plant species within each feeding site was determined by estimating relative percent cover in 10% classes within a circular 100-m2 plot. All species present were assigned a relative percentage of the total plant cover. Field workers gained proficiency by measuring cover within practice plots. Because of sim- ilarities in understory, data were combined for mature, old-growth, and pole-timber stands. Shrub- fields and grass-shrub types were also treated as a single type. Clear-cuts comprised a third successional stage sampled. Individual species eaten in each succes- sional stage were designated as preferred (use > availa- bility) using the electivity index of Alcoze and Zim- mermann (1973). Seasonal groupings, based on plant phenology, Elk distribution, and shifts in habitat use (Irwin and Peek 1983), were: 21 April-15 June (spring), 16 June - 8 September (summer), 9 Sep- tember — 20 December (autumn), and 21 Decem- ber — 20 April (winter). The hypothesis that major foods were selected in relation to availability among all feeding sites in each season was tested by calculating correlations between use and availability. A positive correlation coefficient was used to indicate that Elk were not selective, 1.e., most species were used in proportion to availability over all feeding sites, collectively (Nudds 1980). Pref- erence refers to comparisons within feeding sites ina single successional stage in each season, and selection refers to comparisons of use and availability among all feeding sites in each season. To be selective, Elk would prefer or avoid certain species whenever they were available. Results Spring Grass-shrub and Seral Brushfields. — In spring 3.903 bites were counted in 31 feeding sites. Most feeding activity was observed in grass-shrub types and seral brushfields (16 sites), and clear-cuts (11 sites). When Elk fed in grass-shrub types and seral brush- fields during early spring, Bluebunch Wheatgrass, Elk Sedge, and groundsels (Senecio spp.) accounted for about 62% of the diet (Table 1). Although 33 of 58 species available were eaten, only groundsels, Elk Sedge and penstemon (Penstemon spp.) were pre- ferred, according to the electivity indices. Mallow Ninebark and Big Whortleberry (Vaccinium mem- branaceum) were abundant but not eaten. Clear-cuts. — As snow receded, Elk used clear-cuts THE CANADIAN FIELD-NATURALIST Vol. 97 at higher elevations, and fed upon 34 of 66 available species, including Fireweed Willow-herb, Elk Sedge, and Siberian Minerslettuce (Montia sibirica). These species were abundant, and 60% of the diet in clear- cuts was comprised of succulent forbs. However, use was similar to availability and no preference was indi- cated. Green Mountain Ash (Sorbus scopulina), Mountain Bluebell (Mertensia ciliata), and currants (Ribes spp.) were preferred, and their combined use was about 25% of the diet in clear-cuts. Timber Stands. — Myrtle Pachistima and Gold- thread (Coptis occidentalis), the only species fre- quently used in the four feeding sites examined in timber stands during spring, were eaten prior to green- up. Both are evergreen plants. All Areas. — Graminoids, shrubs, and forbs consti- tuted 30, 38, and 42% of the spring diet, respectively. Diets varied among successional stages in relation to availability, as graminoids and shrubs, shrubs and forbs, and shrubs were most heavily used in shrub types, clear-cuts, and timbered stands, respectively. A positive correlation (r = 0.60, P < 0.025) between use and availability of all species among all successional stages indicated Elk were not selective, because they used the most abundant forage species in spring. Summer Grass-shrub and Seral Brushfields. — Eighty-five summer feeding sites were examined, 17 of which occurred in grass-shrub types and brushfields. Forty- six of 68 available species were used, but Elk fed mainly upon leaves and flowers of shrubs. Saskatoon Serviceberry, Snowbrush Ceanothus, and Scouler Willow comprised 50% of the diet chosen in these successional types (Table |). Mallow Ninebark, Birch- leaf Spiraea (Spiraea betulifolia), and Common Snowberry (Symphoricarpos albus) were common but not eaten. Clear-cuts. — We examined 58 feeding sites in clear-cuts and counted | 1 653 bites. Ninety-seven spe- cies were available and Elk fed upon 74 species, mainly forbs. However, only Fireweed Willow-herb and Utah Honeysuckle (Lonicera utahensis) were pre- ferred. Elk made heavy use of Western Thimbleberry (Rubus parviflorus), Fairybells (Disporum orega- num), and young shoots of Myrtle Pachistima in some clear-cuts during flowering. These species received progressively less use as they became less succulent. Timber Stands. — After vegetation in clear-cuts matured in late summer, Elk were observed feeding more often in the nearly closed timber stands, where succulent forage was still available. We examined 10 feeding sites and found use on 51 of 76 available species. Elk preferred leaves and shoots of Rocky Mountain Maple and Utah Honeysuckle. Although 1983 IRWIN AND PEEK: ELK FORAGING IN IDAHO 445 TABLE |. Major species of plants available to and eaten by Elk in three successional stages in the Idaho Panhandle National Forest, 1975-772 Spring Summer Fall Winter Grass- Clear- Grass- Clear- Grass- Clear- Grass- Clear- shrub cut Timber shrub cut Timber shrub cut Timber shrub cut’ Timber At Acer glabrum ah& ma 02 20.0) ac : — eee a 3 3.1 0.9 13.3 4.5 2.0 6.1 4.5 Alnus sinuata 353 6.4 Amelanchier alnifolia OW AS MO MG an BUG ee, eae 1.2 18.1 0.5 0.8 4.5 0.8 18.1 4.5 Berberis aquifolium a 2 00 Oe 2 2.5 2.8 Bel 4.5 DP Ceanothus velutinus a) Us) AU Ue OU OD 95. 4.4 0.6 18.4 0.4 0.9 9.0 12.8 4 00. O03 6G" 17.8" 0.0 Lonicera utahensis SE ety EI, ike SUED SNE Meese 2.4 0.3 S22 10.0 5.4 i Pachistima myrsinites Ue CO OR eS OM I 2 Mell 1.0 35.6 0.5 4.1 2 9.0 ALS) OD 1.8 50.0 10.0° Ribes sp. : 7.4 Rubus parviflorus Oo 22 Bool tee 00 0 2 0D 4.0 DP 6.5 4.6 4.5 352 7.9 5.0 Mp f 0.3 2.1 8.3 1.9 0.6 0.0 — Ags 0.0 Salix scouleriana RS fe ice aig Mya el baer weer peel pees 0.3 D3} 12 1.9 0.7 37) 19.9 4.1 + Sorbus scopulina a 3 2! Je 4.5 3.4 2.0 I 2 Spiraea betulifolia Oo Ow a ac 0 0D 6.9 13.2 il. 0.9 4.5 5.0 y ee 22.4 gropyron spicatum ears 19.5 Anaphalis margaritacea OL ae) OS 0.1 32 0.3 Arnica latifolia g 0 OY OL We. 2 1.5 1.2 | 8.6 é 4.9 0.0 0.0 Bromus vulgaris AGES leis nee 4.5 4.6 z : Me TOS “OI 1.7 Ie arex geyeri eae Ua ce ei) Ree ee psy 16.1 17.0 0.1 2.1 4.5 Coptis occidentalis oe no) 0) nehe oe 14.5 5. I [Qe e520) 30.2 ; 4.4 0.0 Disporum oreganum SREB oy ee 4.2 3.1 + Epilobium O22 WES BE Be) SAY (98) angustifolium Oe 2254 0.4 23.2 4.2 S15) 5.4 a Mertensia ciliata no:8) 4.6 Montia sibirica 14.2 23 13.4 2.0 AO OV Penstemon sp. continued 446 THE CANADIAN FIELD-NATURALIST Vol. 97 TABLE | (concluded) Spring Summer Fall Winter Grass- Clear- Grass- Clear- Grass- Clear- Grass- Clear- shrub cut Timber shrub cut Timber shrub cut Timber shrub - cut’ Timber Polemonium ives pulcherrimum 6.4 ee 18.37 enecio IES Trillium ovatum Sey 2.7 _0.9 2.9 3.5 0.9 reer Dri 19.0 57) O.8 42OQ27 35.3 a7 0.6 34.8 8.4 4.4 1.1 t ——_ ns ——— —— ———— —————— ——_—— —_— Ee ——— —— Saad OS 62 69 ©S 66 D2 SO M7 OS ae 32.4 No. of sites examined 16 11 17 58 10 | 2 9 15 0 l No. of bites tallied 1611 1853 439 2767 #11653 1 143 74 408 1143 2375 0 87 No. of species eaten 33 34 8 46 74 51 3 6 D2 15 0 3 No. of species available 58 66 17 68 97 76 31 52 41 31 0 17 4 Average aggregate percent use/ average relative cover in |00 m2? circular plots. A plus sign indicates preference in relation to availability, using the electivity index of Alcoze and Zimmerman (1973). Includes species comprising less than 2% of the aggregate. Elk preferred these shrubs, numerous forbs were eaten. Goldthread, American Trailplant (Adenoca- lulon bicolor), and Big Whortleberry were abundant but not eaten. All Areas. — In summer, shrubs and forbs each constituted about 50% of the diet. Shrubs were most important when Elk fed in timber stands or seral brushfields, but forbs were most important when Elk fed in clear-cuts. As in spring, most species were eaten in proportion to availability; thus, Elk were not selec- tive (r = 0.78, P < 0.001). Twenty-three of 31 preval- ent species were used in proportion to availability, and 7 were avoided. Only Utah Honeysuckle appeared to be preferred wherever it occurred, i.e., it was selected. Fall Timber Stands. — In autumn, Elk were observed less often in open areas, where we found only 3 feeding sites. We examined 9 sites in timber stands where forbs and shrubs constituted 42 and 57% of the diet, respectively. Myrtle Pachistima, Goldthread, West- ern Thimbleberry, and Fireweed Willow-herb com- prised 84% of the diet, but only Myrtle Pachistima was preferred. Only 24 species were used in fall, but the correlations between use and availability (r = 0.94, P < 0.001) were particularly high, indicating Elk were not selective. Winter Seral Brushfields. — The winter range occurred within a 1931 buygn. Half the area was covered by grass-shrub types and seral brushfields, and no clear- cuts were present. Fifteen of 16 feeding sites examined were in seral brushfields, and only 16 species were used. Elk fed mainly on Scouler Willow, but also used Snowbrush Ceanothus, Saskatoon Serviceberry, and Rocky Mountain Maple. Scouler Willow and Rocky Mountain Maple were preferred. Elk pawed to Myrtle Pachistima and Elk Sedge when snow was not deep, but shrubs constituted 98% of the winter diet (Table 1). Correlations of use vs. availability of plant species indicated foraging was related to availability (r = 0.52, P< 0.05). Discussion Elk appear to be food generalists. While several species were preferred in certain successional stages, Elk were generally not selective, in comparisons among all successional stages. Thus, diets varied among the successional forest stages. Plant phenology also appeared to play a role in forage use, suggesting nutritional quality may influence forage selection. Elk preferred some species before or during flowering, but later shifted to other species in immature stages in other foraging areas. This information suggests that more emphasis should be placed in management and research on data describing mechanisms of habitat selection, including home range selection and food patch selection. Because Elk appear to be generally non-selective, managers should worry less about species composi- tion than the successional stages where feeding occurs. Black et al. (1976) defined distinct foraging and cover areas for Elk in eastern Oregon. However, forage areas are not distinct from optimal cover areas in 1983 northern Idaho. Late summer and fall feeding occurred primarily in nearly-closed timber stands, also suitable for thermal and security cover. Clear- cuts and grass-shrub types apparently become less useful at that time due to phenological changes. Condition of foraging areas used in late fall and early spring may be important in maintaining Elk herds in deep-snow areas. Importance of conditions prior to winter in maintaining winter ungulate popu- lations has been noted by Julander et al. (1961), Verme (1965), Klein (1965), and Mautz (1978). While forage quality is important, use of early successional stages with highest quantities of forage in late summer and early fall also may increase survival of animals through winter when forage sources have naturally deteriorated in both availability and quality. Managers should strive to enhance amounts and distribution of early successional stages which pro- duce large quantities of forage. In our observations, early greening areas near winter ranges appear to be in relatively short supply, and should be protected as much as possible from development. Some of the drier timber areas adjacent to riparian zones and near known winter ranges could be clear-cut and burned to provide early-growing grasses, sedges, and forbs. Forage-producing clear-cuts or seral brushfields occur on many summer ranges, but we observed those on north or east slopes are used most. These develop rapidly into dense sapling and pole stands, which are important for Elk in autumn. Such thickets might be lightly thinned to encourage production of palatable understory plants such as Myrtle Pachistima, which will increase following thinning within the Western Hemlock type (Irwin and Peek 1979). Seral shrub- fields currently used as winter range are very impor- tant and should be renovated as indicated by Leege (1969). Acknowledgments This study was supported by MclIntire-Stennis Pro- ject MS-22, College of Forestry, Wildlife and Range Sciences, University of Idaho, and the Intermountain Forest and Range Experiment Station, U.S. Forest Service. Contribution No. 233, Forestry, Wildlife and Range Experiment Station, University of Idaho. IRWIN AND PEEK: ELK FORAGING IN IDAHO 447 Literature Cited Alcoze, T., and E. Zimmermann. 1973. Food habits and dietary overlap of two heteromyid rodents from the mes- quite plains of Texas. Journal of Mammalogy 54(4): 900-908. Black, H., R.J. Scherzinger, and J. W. Thomas. 1976. Relationships of Rocky Mountain elk and Rocky Moun- tain mule deer habitat to timber management in the Blue Mountains of Oregon and Washington. Pages 11-31 in Proceedings Elk-logging-roads Symposium, Edited by S. R. Hieb. University of Idaho, Moscow. Cole, G. F. 1956. The pronghorn antelope. Its range use and food habits in central Montana, with special reference to alfalfa. Montana Agricultural Extension Station T ech- nical Bulletin 516. 63 pp. Daubenmire, R. F., and J. B. Daubenmire. 1968. Forest vegetation of eastern Washington and northern Idaho. Washington State University Agricultural Experiment Station Technical Bulletin 60. 104 pp. Irwin, L. L., and J. M. Peek. 1979. Shrub production and biomass trends following 5 logging treatments within the cedar-hemlock zone of northern Idaho. Forest Science 25(3): 415-426. Irwin, L. L. and J. M. Peek. 1983. Elk habitat use relative to forest succession in Idaho. Journal of Wildlife Man- agement 47(3): 664-672. Julander, O., W. L. Robinette, and D. A. Jones. 1961. Rela- tion of summer range conditions to mule deer productiv- ity. Journal of Wildlife Management 25(1): 54-60. Klein, D. R. 1965. Ecology of deer range in Alaska. Ecolog- ical Monographs 35: 259-284. Knowlton, F. 1960. Food habits, movements, and popula- tions of moose in the Gravelly Mountains, Montana. Journal of Wildlife Management 24(2): 162-170. Kufeld, R. C. 1973. Foods eaten by the Rocky Mountain elk. Journal of Range Management 26(2): 106-112. Leege, T. A. 1969. Burning seral brush ranges for big game in northern Idaho. Transactions of the North American Wildlife and Natural Resources Conference 34: 429-435. Lemmon, P.E. 1957. A new instrument for measuring forest overstory density. Journal of Forestry 55(9): 667-669. Martin, A. C.,R. Gensch, and C. Brown. 1946. Alternative methods in upland game bird food analyses. Journal of Wildlife Management 10(1): 8-12. Mautz, W. W. 1978. Sledding on a bushy hillside: the fat cycle in deer. Wildlife Society Bulletin 6(2): 88-90. Nudds, T. D. 1980. Forage “preference”: theoretical con- siderations of diet selection by deer. Journal of Wildlife Management 44(3): 735-740. Verme, L. J. 1965. Reproductive studies on penned white- tailed deer. Journal of Wildlife Management 29(1): 74-79. Received 22 September 1982 Accepted 15 July 1983 Notes Caspian Terns, Sterna caspia, Breeding in Labrador A. R. LOCK Canadian Wildlife Service, Bedford Institute of Oceanography, P.O. Box 1006, Dartmouth, Nova Scotia B2Y 4A2 Lock, A. R. 1983. Caspian Terns, Sterna caspia, breeding in Labrador. Canadian Field-Naturalist 97(4): 448. In 1979 Caspian Terns (Sterna caspia) were found breeding ina colony of Ring-billed Gulls (Larus delawarensis) on an island at the eastern end of Lake Melville, Labrador. Although there are no previous records of this species breeding in Labrador, it is suggested that this record represents the first documentation rather than a range expansion. Key Words: Caspian Tern, Sterna caspia, Labrador, breeding range. On 7 July 1979 four adult Caspian Terns (Sterna caspia) were observed over a Ring-billed Gull (Larus delawarensis) colony on Gull Island (54°00’N, 58°43’W), a treeless grassy island about 40 m long situated at the eastern end of Lake Melville, Labra- dor. A search revealed a large, downy Caspian Tern chick approximately one week of age concealed in long grass about 6 m above water level. The adults hovered over the area in which the chick was con- cealed, screeching continuously. Nest sites were not positively identified nor did a thorough search dis- cover further chicks. On the same island 322 pairs of Ring-billed Gulls, two pairs of Great Black-backed Gulls (Larus marinus) and two pairs of Arctic Terns (Sterna paradisaea) were also found breeding. This observation is the northernmost breeding record of Caspian Terns in eastern Canada. Caspian Terns have a wide but disjunct breeding distribution (Voous 1960) with the majority of the Canadian population breeding in central Manitoba and the Great Lakes. In eastern Canada they are known to breed only in Quebec and insular New- foundland (Godfrey 1966). Audubon (1897) noted them as breeding on the north shore of the Gulf of St. Lawrence in 1833 though he misidentified them as “Cayenne Terns’. Their numbers have dwindled since that time and only four birds were counted on the 1977 census of the bird sanctuaries on the Gulf North Shore (Chapdelaine 1980). Caspian Terns are also suspected to breed in the Magdalen Islands but not in the Mari- time Provinces. Their status on the island of New- foundland is less certain: there are several past records of breeding (M. Martin 1978. Status Report on Cas- pian Terns; Committee on the Status of Endangered Wildlife in Canada) but the number breeding there at present is unknown. There are no previous Labrador breeding records. Austin (1932) did not list the Caspian Tern as a species observed on the Newfoundland Labrador and Todd (1963) listed observations only in James Bay and on the Gulf North Shore. The only previous record of their being seen anywhere in Labrador is that of Grayce (1947), a July sighting of an individual at Antill Cove (52°13’N, 55°40’W) in southern Labra- dor, about 300 km SE of Gull Island. Todd (op. cit., p. 769) dismissed this record as a doubtful identification but the present record of breeding in Lake Melville lends credence to Grayce’s report. There is no reason to believe that Caspian Terns have extended their range northward in recent decades: they often breed in close association with Ring-billed Gulls, and that spe- cies was reported breeding in Lake Melville by Macoun (1900). Literature Cited Audubon, M. R. 1897. Audubon and his journals with zo- ological and other notes by Elliot Coues. Two volumes. New York. 537 pp. Austin, O. L., Jr. 1932. The birds of Newfoundland Labra- dor. Memoirs of the Nuttall Ornithological Club No. 7. 229 pp. Chapdelaine, G. 1978. Onziéme inventaire et analyse des fluctuations des populations d’oiseaux marins dans les refuges de la Cote Nord du Golfe Saint-Laurent. Cana- dian Field-Naturalist 94: 34-42. Godfrey, W.E. 1966. The Birds of Canada. National Museum of Canada Bulletin 203. 428 pp. Grayce, R. L. 1947. Birds of the MacMillan Labrador Expedition, 1946. Auk 64: 275-280. Macoun, J. 1900. Catalogue of Canadian Birds. Part 1. Geological Survey of Canada, Ottawa. 218 pp. Todd, W. E. C. 1963. Birds of the Labrador Peninsula and Adjacent Areas. University of Toronto Press, Toronto. 819 pp. Voous, K. H. 1960. Atlas of European Birds. Nelson, Lon- don. 284 pp. Received 15 April 1982 Accepted 20 October 1982 448 1983 NOTES 449 Range Extension of the Sage Artemisia rupestris ssp. woodii in Southwestern Yukon M. HOEFs, D. RUSSELL, and BETH EREAUX Yukon Wildlife Division, P.O. Box 2703, Whitehorse, Yukon YIA 2C6 Hoefs, M., D. Russell, and Beth Ereaux. 1983. Range extension of the sage Artemisia rupestris ssp. woodii in southwestern Yukon. Canadian Field-Naturalist 97(4): 449-450. A new location is given for Artemisia rupestris ssp. woodii, some 130 km from its nearest reported locality in Kluane National Park. Key words: Artemisia rupestris ssp. woodii, flora of Yukon. The distribution of flora and fauna elements of the Yukon is incompletely known, and frequent docu- mentations of range extensions can therefore be expected. The following observation, however, war- rants official recording because of the very restricted North American occurrence of the species in question. During a brief aerial assessment of Dall Sheep win- \\ resco ie Ciara eulloa t ter ranges in the southern Yukon, when short landings were made to compile species lists and physical parameters of representative sites, the occurrence of the sage Artemisia rupestris ssp. woodii was observed on the east side of Rose Lake. This location is shown on the accompanying map. The coordinates are approximately 135°55’W and 60°21’N. A number of S.W. YUKON TERRITORY €§ Locations oF ARTEMISIA RUPESTRIS ordenskiold =k Cag Ss : 5 * mS YW, ~ J wes Junction a (| \ >) Veg *\ WHITEHORSE \\MacRae ie t 2 or —<————> (Turki, YZ Lake 139° FIGURE 1. Map of the southwestern portions of the Yukon Territory showing the locations for Artemisia rupestris which are discussed in the text. 450 vigorous clumps of plants in bloom were located on 17 July 1982. The site was the southwest facing slope of Mt. Primrose at an elevation of about 1300 m and a slope gradient of 45°. This species was associated with Oxytropis viscida, Artemisia frigida, Saxifraga tri- cuspidata and Calamagrostis purpurascens in a typi- cally dry, exposed “steppe association”, as described by Hoefs et al. (1975) for Sheep Mountain, Kluane National Park, where Artemisia rupestris ssp. woodii constitutes a co-dominant or associate in a number of grassland communities. Specimens from this new site were preserved and are kept in the collection of the Yukon Wildlife Division. Neilson (1968) was the first to report this species for North America. He collected it on Sheep Mountain, Kluane National Park. Douglas et al. (1981) docu- mented additional locations in Kluane National Park, and the senior author also observed it at Cultus Bay, east of Kluane Lake. These previously known loca- tions of this species are shown on the accompanying map. THE CANADIAN FIELD-NATURALIST Vol. 97 The range of Artemisia rupestris ssp. rupestris 1s in the Lena River drainage of eastern Siberia, a distance of some 5200 km to the west. The newly reported location may also be disjunct, since it has not been reported for intermediate location, despite considera- ble vegetation work in the area during the past decade. Literature Cited Douglas, G. W., H. L. Argus, H. L. Dickson, and D. F. Brunton. 1981. The rare vascular plants of the Yukon. Syllogeus 28: 1-60. Hoefs, M., 1. McT. Cowan, and V. J. Krajina. 1975. Phyto- sociological analysis and synthesis of Sheep Mountain. southwest Yukon Territory, Canada. Syesis 8(Supplement 1): 125-228. Neilson, J. A. 1968. New and important additions to the flora of the southwestern Yukon Territory, Canada. Can- adian Field Naturalist 82(2): 114-119. Received 18 August 1982 Accepted 21 July 1983 Nesting of the Thick-billed Murre, Uria lomvia, in British Columbia ANNE VALLEE! and RICHARD J. CANNINGS? ‘Department of Animal Science, University of British Columbia, Vancouver, B. C. V6T 1W5. Deceased July 1982. *Cowan Vertebrate Museum, Department of Zoology, University of British Columbia, Vancouver, B. C. V6T 2A9. Vallee, Anne, and Richard J. Cannings. 1983. Nesting of the Thick-billed Murre, Uria Jomvia, in British Columbia. Canadian Field-Naturalist 97(4): 450-451. In 1981, about 19 pairs of Thick-billed Murres (Uria omvia) breeding on Triangle Island represented the first breeding record for this species in British Columbia. Triangle Island is now the southernmost known breeding site of Thick-billed Murres in the eastern Pacific. Key Words: Thick-billed Murre, Uria Jomvia, British Columbia, breeding, range extension. The Thick-billed Murre (Uria Jomvia) is an arctic and subarctic alcid, breeding in the North Pacific, North Atlantic, and Arctic Oceans (Tuck 1961). In the Atlantic it is known to breed as far south as Cape St. Mary’s, Nfld. (46°48’N, 54° 12’W) (Brownet al. 1975). In southeast Alaska, Thick-billed Murres have been reported nesting only as far south as St. Lazaria Is. (56°59’N, 135°43’W) near Sitka, where a colony of 2 000 birds was discovered in 1981 (A. L. Sowls, USFWS, personal communication). They have never been found breeding on the Pacific coast of Canada. There are two breeding season (mid-June to late August: Birkhead and Nettleship 1981) records of Thick-billed Murres in British Columbia: S. G. Sealy (U. Man., personal communication) saw four in breeding (alternate) plumage along the east side of Langara Island, Queen Charlotte Islands, on 20 June 1970, and George Sirk photographed a bird in breed- ing plumage on the eastern side of the Houston- Stewart Channel at the south end of the Queen Char- lotte Islands on 16 July 1980 (British Columbia Provincial Museum PDF#664). In this note, we report about 19 pairs of Thick-billed Murres nesting on Tri- angle Island, British Columbia. Triangle Island (50°52’N, 129°05’W) is a small tree- less island at the northern tip of Vancouver Island, 46 km west of Cape Scott. It harbours the largest breed- ing seabird colony of any site in British Columbia, including Cassin’s Auklets, Ptychoramphus aleuticus (360 000 pairs), Tufted Puffins, Fratercula cirrhata (25 000 pairs), Rhinoceros Auklets, Cerorhinca mon- ocerata (15 000 pairs), and Common Murres, Uria 1983 aalge (3 000 pairs) (Vermeer 1979, Vermeer et al. 1976, 1979). The topography, climate, and habitats of Tri- angle Island were described by Carl et al. (1951) and Vermeer et al. (1979). Thick-billed Murres were first seen on the island on 5 July 1980 when AV photographed two in the Com- mon Murre colony on the southwestern corner of the island. No Thick-billed Murres were seen again that year, but no extensive searches were made for them. On 19 June 1981, AV saw five Thick-billed Murres on a different ledge in the Common Murre colony, and on 18 and 21 August RJC photographed two birds with eggs. On 27 August, AV and Robin Cohen photographed seven adults with chicks. Copies of these slides are on file at the British Columbia Provin- cial Museum (PDF#666). A search of the Common Murre colony yielded 19 pairs of Thick-billed Murres on 27 August, but several suitable but inaccessible areas were not censused, and the lower cliffs visible from above were only scanned through a 20X tele- scope from distances of about 200 m. Thick-billed Murres were still present on 29 August, but AV could not find any on 5 September, although Common Murres were still numerous. We assume, therefore, that the chicks seen on 29 August had fledged before 5 September. In 1982, AV and Moira Lemon counted 68 Thick- billed Murres on ledges in the Common Mutrre col- ony, and Michael Rodway saw two Thick-billed and 20 Common Murres on cliffs on the east coast of the island. These sightings apparently indicate a recent arrival of the Thick-billed Murre on Triangle Island. The murre colony there was censused thoroughly in 1978 (R. W. Campbell, B. C. Provincial Museum, personal communication), but a small number of Thick-billed Murres nesting in inaccessible areas easily could have been overlooked. To the north, Thick-billed Murres were not breeding on St. Lazaria Island in 1912 (Wil- lett 1912; Sowls et al. 1978); thus the colony of 2 000 birds there in 1981 probably indicates a considerable population movement during the intervening years. Until that colony was discovered, the southernmost known breeding site in the eastern Pacific was on Middleton Island, Alaska (59°26’N, 146°21’W), 625 km northwest of St. Lazaria Island (Sowls et al. 1978). Triangle Island is almost 800 km southeast of St. Lazaria. Thick-billed Murres, therefore, seem to be extending their breeding range southeastward in the NOTES 451 eastern Pacific, as are the Northern Fulmar (Ful/ma- rus glacialis) (K. Vermeer, Canadian Wildlife Service, personal communication), and the Horned Puffin, Fratercula corniculata (Campbell et al. 1979). These two species have also been sighted on Triangle Island in the last few years and may breed there in very small numbers (personal observation, and K. Vermeer, per- sonal communication). Acknowledgments We would like to thank S. G. Sealy, M. Peterson, and N. A.M. Verbeek for critically reading this manuscript. Field work in 1982 was funded by the Canadian Wildlife Service and the British Columbia Provincial Museum. Literature Cited Birkhead, T.R., and D.N. Nettleship. 1981. Census methods for murres, Uria species: a unified approach. Canadian Wildlife Service, Occasional Papers No. 43. Brown, R. G. B., D. N. Nettleship, P. Germain, C. E. Tull, and T. Davis. 1975. Atlas of eastern Canadian seabirds. Canadian Wildlife Service, Ottawa. 220 pp. Campbell, R. W., H. R. Carter,andS. G. Sealy. 1979. Nest- ing of Horned Puffins in British Columbia. Canadian Field-Naturalist 93: 84-86. Carl, G. C., C. J. Guiguet, and G. A. Hardy. 1951. Biology of the Scott Island Group, British Columbia. Report of the Provincial Museum of Natural History and Anthropol- ogy, 1950: B21-B63. Sowls, A. L., S. A. Hatch, and C. J. Lensink. 1978. Cata- log of Alaskan Seabird Colonies. United States Depart- ment of the Interior, Fish and Wildlife Service FWS/OBS -78/78. Tuck, L. M. 1961. The Murres. Canadian Wildlife Service Monograph No. 1. Vermeer, K. 1979. Nesting requirements, food, and breed- ing distribution of Rhinoceros Auklets, Cerorhinca mon- ocerata, and Tufted Puffins, Lunda cirrhata. Ardea 67: 101-110. Vermeer, K., K. R. Summers, and D. S. Bingham. 1976. Birds observed at Triangle Island, British Columbia, 1974 and 1975. Murrelet 57: 35-42. Vermeer, K., R. A. Vermeer, K. R. Summers, and R. R. Billings. 1979. Numbers.and habitat selection of Cassin’s Auklets breeding on Triangle Island, British Columbia. Auk 96: 143-151. Willett, G. 1912. Report of G. Willett, agent and warden stationed on St. Lazaria Bird Reservation, Alaska. Bird- Lore 14: 419-426. Received 21 May 1982 Accepted 12 February 1983 452 THE CANADIAN FIELD-NATURALIST Vol. 97 Ring-billed Gull, Larus delawarensis, Predation on Bat, Myotis, Injured by an American Kestrel, Falco sparverius GEOFFREY L. HOLROYD! and ELISABETH G. BEAUBIEN2 1Canadian Wildlife Service, 1000, 9942-108 Street, Edmonton, Alberta TSK 2J5 2Parks Canada, Box 2440, Banff, Alberta TOL 0CO Holroyd, Geoffrey L., and Elisabeth G. Beaubien. 1983. Ring-billed Gull, Larus delawarensis, predation on bat, Myotis, injured by an American Kestrel, Falco sparverius. Canadian Field-Naturalist 97(4): 452. An American Kestrel (Falco sparverius) unsuccessfully attacked a small bat which was then killed and carried off by a Ring-billed Gull (Larus delawarensis). Key Words: American Kestrel, Falco sparverius, Ring-billed Gull, Larus delawarensis, bat, Myotis, predation. On i3 September 1981 at 1100 as we canoed across Lac des Arcs, Exshaw, Alberta (51°03’N, 115° 10’W), we were overtaken by an American Kestrel (Falco sparverius) pursuing a small bat. The kestrel made numerous swoops from above and behind the bat but the bat managed to evade the falcon fora minute after we first saw them. The head movements of the bat indicated that it was visually avoiding the kestrel rather than relying on echolocation. At this point an adult Ring-billed Gull (Larus dela- warensis) joined the attack on the bat. A few seconds after the gull arrived the kestrel gripped the bat with its claws. The kestrel uttered a short sharp cry, (possi- bly the bat bit it) and released the bat within 5 seconds. The gull made two swoops at the bat, hit it with its bill, and then the bat fell to the water. The gull picked the bat off the water and flew away with it in its bill, only to be chased by six other adult Ring-billed Gulls which had been standing ona sandbar.about one-half kilometer away. A chase ensued but the original gull still had the bat when we lost sight of them about 5 minutes later. The attack occurred over water. The shore behind the bat and falcon when we first saw them is vegetated with shrubs and aspen (Populus tremuloides). The town of Exshaw is within 200 m of the shore. We do not know where the bat came from or why it was flying during daylight. The weather was sunny, calm, 15°C, with light cirrus cloud cover. The above observations were made with the aid of 8 x 35 binoculars in good light. The total observations took less than 5 minutes to the death of the bat and covered about 300 m. The relative size indicated that the victim was likely a Little Brown Bat (Myotis lucifugus), the commonest bat in nearby Banff National Park (unpublished data). Although kestrels are known to eat bats (Stoner 1939; Baker 1962), this kestrel appeared to have con- siderable difficulty catching the bat, likely because kestrels catch most prey including birds on the ground (Jenkins 1970; Page and Whitacre 1975). Presumably, the arrival of the gull distracted the bat and the injur- ies inflicted by the kestrel facilitated the capture by the gull. The ultimate success of the Ring-billed Gull is an example of opportunistic feeding by this species. Acknowledgments We thank L. Oliphant for constructive comments on an earlier draft. Literature Cited Baker, J. K. 1962. The manner and efficiency of raptor depredations on bats. Condor 64: 500-504. Jenkins, R. E. 1970. Food habits of wintering Sparrow Hawks in Costa Rica. Wilson Bulletin 82: 97-98. Page, G., and D. F. Whitacre. 1975. Raptor predation on wintering shorebirds. Condor 77: 73-83. Stoner, D. 1939. Eastern Sparrow Hawk feeding on big brown bat. Auk 56: 474. Received 22 September 1982 Accepted 12 February 1983 1983 NOTES 453 High Nesting Density of Ducks on an Island in Saskatchewan PHILIP M. BROWNE!, DAVID A. DUFFUS?2, and RONALD W. BOYCHUK? 1Ducks Unlimited Canada, Box 4465, Regina, Saskatchewan S4P 3W7 2Department of Geography, University of Regina, Regina, Saskatchewan S4S 0A1 3Ducks Unlimited Canada, 1290 South Mackenzie Avenue, Williams Lake, B.C. V2G 3Y1 Browne, Philip M., David A. Duffus, and Ronald W. Boychuk. 1983. High nesting density of ducks on an island in Saskatchewan. Canadian Field-Naturalist 97(4): 453-454. Forty-seven duck nests, primarily Mallards (Anas platyrhynchos), were found on a small natural island of 0.4 ha near Meacham, Saskatchewan, in 1979. The secure nature of the site owing to dense vegetative cover and the island’s distance from shore were felt to be responsible for this high nesting density. Key Words: Ducks, dense nesting, natural island, Saskatchewan On 24 May 1979, while assessing potential of water- fowl nesting habitat near Meacham, Saskatchewan, we discovered a remarkably high density of nesting ducks on a natural island 0.4 ha in size. Three species of ducks: Mallard (Anas platyrhynchos), Northern Pintail (A. acuta) and Gadwall (A. strepera) nested there as well as some Canada Geese (Branta canadensis). The island was one of three searched for waterfowl nests in two adjacent sloughs in the area. On the island with high numbers of nesting ducks (‘A’) we found 47 nests, under dense shrub cover. Thirty-four active nests were of Mallard, with four abandoned, three hatched and three predated also of this species. One Pintail and two Gadwall nests were also present. Nest- ing density on island ‘A’ was 0.012 duck nests/sq. m. Spatial measurements were kept to a minimum to reduce disturbance; many of the nests were less than | m apart. Distance of duck nests from the water varied from 2 m to 28 m (the maximum possible). Common Crows (Corvus brachyrhynchos) from other islands in the immediate area may have been responsible for the predation noted. A second island (‘B’) in the same wetland, approxi- mately 150 m from island ‘A’, was also searched. Three nests, two of Pintail and one of Gadwall, were discovered. Initiation dates and clutch sizes were not determined for these nests. On the last island searched (‘C’), in another nearby wetland, no nests were discovered. Clutch initiation dates for nests on island ‘A’ were estimated using the method devised by Westerskov (1950) and back dating on the assumption that one egg was laid each day, with incubation starting on the day the last egg was laid. More than half of the clutches were started between 13 and 27 April(n = 22), witha second wave between 30 April and 11 May (n= 13) and two clutches 17-20 May, all dates plus or minus three days. The mean clutch size overall was 9.32 (n = 37), with the first wave averaging 10.8 eggs and the second 8.9 eggs. Lower mean clutch size of nests initiated between 30 April and 11 May suggests that these may have been re-nesting attempts. However, this was impossible to confirm as individual identities of nesting hens had not been previously established. All three islands were located in moderately saline sloughs approximately 0.75 m deep. Both wetlands had soft flocculent/ organic substrates. Island ‘A’ was 170 m from shore and oval in shape: 57 m wide and 92 m long. Island ‘B’ was approximately 80 m from shore, crescentic in shape and roughly three times the size of island ‘A’. Island ‘C’ was less than 30 m from shore, 6.5 ha in size and rectangular in shape. This latter island is depicted on mapsheet 73A/4(NTS 1:50 000 series, 1974) as being connected to the mainland. Inundation by high spring water levels of the isthmus between island ‘C’ and the mainland would appear to be temporary, as the depth of water was only approx- imately 0.4 m. Islands ‘A’ and ‘B’ were more perman- ent, although these shallow ponds probably dry out in some years. The vegetation on island ‘A’ differed markedly from that of island ‘B’. Shrub cover on island ‘A’ consisted primarily of round-leaved hawthorn (Crataegus rotundifolia) in dense stands approximately 3 m high, interspersed with lesser amounts of saskatoon (A me- lanchier alnifolia) \-2 m high. Low shrubs included mainly gooseberry ( Ribes spp.), rose (Rosa spp.) and snowberry (Symphoricarpos spp.). Grasses and Cow Parsnip (Heracleum lanatum) were encountered throughout, with the latter being a major component of the ground cover. Island ‘B’ had few shrubs and a small number of trees (Populus spp.) which were 4-5 m high. The ground cover was predominantly grass. Island ‘C’ had vegetation similar to island ‘A’, but hawthorn stands appeared to be more dense at site‘C’. Nesting ducks may have been attracted to island ‘A’ by its secure nature owing to the dense vegetative 454 cover and the island’s distance from shore. Island *B’ may not have provided adequate cover to attract high numbers of waterfowl to nest. Although the cover at island ‘C’ was similar to that at island ‘A’, easy access for terrestrial predators to island ‘C’ may have pre- vented ducks from nesting successfully there. Acknowledgments Thanks are extended to G. Bain, R. Seguin and THE CANADIAN FIELD-NATURALIST Vol. 97 J. B. Gollop for critical review of the manuscript. Literature Cited Westerskov, Kaj. 1950. Methods for determining the age of game bird eggs. Journal of Wildlife Management 14: 55-66. Received 24 March 1982 Accepted 12 February 1983 Recoveries of Saskatchewan-Banded White-winged Scoters, Melanitta fusca C. STUART HOUSTON! and PATRICK W. BROWN? 1863 University Drive, Saskatoon, Saskatchewan S7N O0J8 2Gaylord Memorial Laboratory, School of Forestry, Fisheries, and Wildlife, University of Missouri, Puxico, Missouri 63960 3Present address: School of Forest Resources, University of Maine, Orono, Maine 04473 Houston, C. Stuart, and Patrick W. Brown. 1983. Recoveries of Saskatchewan-banded White-winged Scoters, Me/anitta fusca. Canadian Field-Naturalist 97(4): 454-455. Of 23 White-winged Scoters, Melanitta fusca deglandi, banded in Saskatchewan, 12 were recovered in the west, 8 east and | south, and 2 were recovered locally. Only three recoveries occurred in the first six months. One bird represents a new longevity record for the species, at least 15'4 years of age at death. Key Words: White-winged Scoter, Melanitta fusca deglandi, migration, longevity. In this note we report the recovery distribution and longevity records of 23 White-winged Scoters, Mela- nitta fusca deglandi, banded in central Saskatchewan in July and August. In addition to the 11 recoveries of adult females we banded on their nests, we use recov- eries of 9, 2, and | local or juvenile scoters that were banded by D. C. Lowery and C. E. Graham, by W. Fuchs, and by A. J. Matheson, respectively. These scoters take divergent routes from Saskat- chewan to reach their wintering grounds on distant oceans (Figure |). Of the 23 known recoveries to date, 11 have been from the Pacific Ocean, 4 from the Atlantic Ocean and | from the Gulf of Mexico, while | and 4 others were evidently on their way to the Pacific and Atlantic Oceans, respectively. Only two were rec- overed in Saskatchewan. Ten birds were found dead, 10 had been shot by hunters, and single individuals were caught in a net, caught in a trap, and sighted by binoculars. Only 3 recoveries were direct within the first calendar year and another 6 were between 6 and 12 months after banding. Three were recovered in the second year (1 and 2 years after banding), then 2, 2, 0, 3, 1,0, 1,0, 0, 1,0 and 1 inconsecutive years thereafter. The last, an adult female with band no. 637-97350, was found freshly dead on 22 January 1981 at Comox, British Columbia (May Carragher, personal com- munication); it was banded at Redberry Lake, Sas- katchewan, on 16 July 1967. Since scoters do not nest until at least two years of age (Brown and Houston 1982), 637-97350 must have been at least 15!4 years of age, 134 years after her first and only capture on her nest. This surpasses the previous longevity record of an adult female scoter recaptured on her nest at Red- berry Lake |! years after banding, when at least 13 years of age (Brown and Houston 1982). These 23 recoveries offer further evidence that White-winged Scoters have a longer survival and lower annual mor- tality rate than most other ducks. Bellrose (1976), in briefly summarizing the 12 early recoveries from Saskatchewan-banded scoters, and from those banded in Alberta and the Northwest Ter- ritories (NWT), speculated: “Band recoveries suggest that the farther north and east the whitewings breed, the more likely they are to migrate to the Atlantic Coast, and, conversely, the farther south and west they breed, the more likely that the migration is to the Pacific Coast.” All four recoveries of scoters banded near Yellowknife, NWT, were from the Atlantic Coast. Our subsequent results are also consistent with 1983 NOTES 455 FIGURE |. Places of recovery of White-winged Scoters banded during summer in central Saskatchewan. Squares denote recoveries in the same year as banding; triangles are recoveries within the first six months of the succeeding year; and circles are recoveries more than one year after banding. a wide northwest-southeast line passing through cen- tral Saskatchewan, acting as a “migration divide,” a bit east of the line postulated for Buffleheads banded in Alberta (Erskine 1972). Acknowledgments We thank A. J. Macaulay of Ducks Unlimited for permission to use the A. J. Matheson record, and Duane C. Lowery and Wilhelm Fuchs for permission to give details of their banding results. The field assis- tance of Mary Ann Brown, K. Rautenstrauch, C. Thomas and D. White is acknowledged. Funds for part of this study were provided by the Wildlife Man- agement Institute and the Office of Migratory Bird Management, U.S. Fish and Wildlife Service (con- tract HUSDI 14-16-009-77-930), Gaylord Memorial Laboratory, and Missouri Agricultural Experimental Station (Project 170 & 183). Literature Cited Bellrose, F. C. 1976. Ducks, geese and swans of North America. Stackpole, Harrisburg. 544 pp. Brown, P. W., and C. S. Houston. 1982. Longevity and age of maturity of White-winged Scoters. Journal of Field Ornithology 53: 53-54. Erskine, A. J. 1972. Buffleheads. Canadian Wildlife Ser- vice Monograph Series 4. 240 pp. Received 21 July 1982 Accepted 2 November 1982 456 THE CANADIAN FIELD-NATURALIST Vol. 97 Albino Eastern Garter Snakes, Thamnophis sirtalis sirtalis, from Ontario WAYNE F. WELLER 9 Mississauga Road North, Mississauga, Ontario L4H 2H5 Weller, Wayne F. 1983. Albino Eastern Garter Snakes, Thamnophis sirtalis sirtalis, from Ontario. Canadian Field-Naturalist 97(4): 456. Two albinos were present ina litter of 14 Eastern Garter Snakes (Thamnophis s. sirtalis) born in captivity to a female of normal pigmentation taken in 1970 at Moira Lake, Hastings County, Ontario. Although reports of albinistic newborn and adult Eastern Garter Snakes are not uncommon, this represents the first documented record for Ontario. Key Words: Albinism, Eastern Garter Snake, Thamnophis s. sirtalis, Ontario. Hensley (1959) presented records of albino or par- tially albino Eastern Garter Snakes, Thamnophis s. sirtalis, from Michigan and Indiana eastward through Ohio, Pennsylvania, and New York to Connecticut. Subsequently, Dyrkacz (1981) reported albinos from West Virginia and Maryland, and Bartlett (1981) from Massachusetts. Two previous Ontario reports are vague. Hensley (1959: 155) noted that an albino specimen was “kept alive for some time in the Zoology Department, Uni- versity of Toronto”, but no locality was recorded. No records of albinistic garter snakes from Ontario are documented in the collections or files of the Royal Ontario Museum, Toronto (J. Lovisek, personal communication, May 1982) or the National Museum of Natural Sciences, Ottawa (F. R. Cook, personal communication, May 1982). An albino garter snake was reportedly collected at Fort Erie, Ontario, some- time in 1963-66 and given to the Buffalo Zoo but no other details are available (B. Froom, personal com- munication, May 1982). The following record is therefore noteworthy in giving the first documented evidence of albinism in Ontario garter snakes. A female of normal pigmentation for Thamnophis s. sirtalis, taken at Moira Lake (44°28’N, 77°28’W, near the town of Madoc, Hastings County), gave birth to 14 young on 10 August 1970. Of these, two were albino. No measurements were recorded at birth. Both albinos were active from birth and fed initially - on earthworms, Redback Salamanders (Plethodon cinereus), juvenile American Toads (Bufo america- nus), and, in later months, on young-of-the-year Leo- pard Frogs (Rana pipiens). One albino lived 18 months, the other, 27 months. Colour slides taken of both individuals, (W. W. photo numbers 10-14; 147- 154; 970-975) and one of these was earlier published by Froom (1972: 8e) who omitted any details of the specimen’s origin. The eyes of both specimens were bright pink. The background colour was chalk-white, but the striped pattern was apparent as the dorsal stripe and the two laterals were bright yellow. The two blotches on the parietal head scales were also bright yellow. Regrettably no measurements or weight were recorded for these specimens and neither was pre- served. One other albino garter snake was seen in the Moira Lake area in 1970. The same week that the normal pigmented adult female was collected, a cot- tage resident on Moira Lake reported a “2-foot long white snake with yellow stripes and pink eyes”. It had been seen basking on rocks at the shoreline of Moira Lake approximately 0.4 km from where the female which gave birth to the albinos was taken. Acknowledgments The assistance of F. R. Cook, J. Lovisek, and B. Froom in obtaining information on albino Ontario garter snakes is appreciated. Literature Cited Bartlett, R.D. 1981. An albino eastern garter snake, Thamnophis sirtalis, from western Massachusetts. Herpe- tological Review 12(2): 52. Dyrkacz, S. 1981. Recent instances of albinism in North American amphibians and reptiles. Society for the Study of Amphibians and Reptiles, Herpetological Circular No. 11. 31 pp. : Froom, B. 1972. The Snakes of Canada. McClelland and Steward Limited, Toronto/ Montreal. 128 pp. Hensley, M. 1959. Albinism in North American amphibi- ans and reptiles. Publications of the Museum, Michigan State University 1(4): 133-159. Received 8 September 1982 Accepted 10 July 1983 1983 NOTES 457 Wolverine, Gulo gulo, in Lake St. John area, Quebec JACQUES PRESCOTT Jardin zoologique de Québec, Ministére du Loisir, de la Chasse et de la Péche, 8191, avenue du Zoo, Charlesbourg, Québec G1G 4G4 Prescott, Jacques. 1983. Wolverine, Gu/o gulo, in Lake St. John area, Quebec. Canadian Field-Naturalist 97(4): 457-458. What could be the first sighting of a live Wolverine (Gu/o gulo) in the Province of Quebec since one in Gatineau Park in 1972, was made in June 1982 at the Petite Manouane River, 240 miles north of Chicoutimi. Key Words: Wolverine, Gulo gulo, Quebec. A lone Wolverine was seen by Michel Gravel and his wife on 27 June 1982, on the banks of Petite Manouane River (50°00’N, 70°55’W), a few kilome- tres north of Lake Duhamel, and about 240 km north of Chicoutimi. Mr. Gravel was cruising mid-stream up the Petite Manouane ina small 10 HP motor boat when at 1230 hr he noticed the presence of a bear-like animal on the west bank of the river. In that area, the river is 70 to 76 m wide. It flows at an altitude of 274 m and is borded to the west by a 250 m-high forested escarp- ment and to the east by a forested valley. The animal followed the boat at a good pace for 0.4 to 0.8 km(4 to 4 mile) on the river bank and stopped each time the boat slowed down. It was observed for about 10 minutes at a distance no greater than 25 to 45 m. The weather was sunny and clear with light wind, air temperature about 20°C. Mr. Gravel described the animal as being about 1.2 m long (4 feet) and 45 cm tall (1.5 feet), face and body dark colored, slightly paler on the sides. Its tail was longer than a bear’s and its front legs shorter than its hind legs. The sighting was long enough for Mr. Gravel to make sure the animal was neither a bear, nor a canid, Fisher, or Otter. Mr. Gravel has 20 years experience as a hunter and trapper, and has hunted in different locations in New Brunswick, Quebec, and Maine. This was his third hunting and fishing trip to the Lake Duhamelarea. He has killed nine bears over the years and trapped many different species of mammals. It is improbable that he could have confused the Wolve- rine with a bear or any other fur bearer. He reported his observation to the Jardin zoolo- gique de Quebec on 5 July 1982, and was interviewed by the author two days later. Confronted with color pictures of different mammals he unhesitantly recog- nized one of a Wolverine, as identical to the animal he saw. The Lake Duhamel area is inhabited by Moose, Black Bear and Red Fox but no Wolf sign or tracks have been reported there for a long time. Mr. Gravel noticed no carrion or carcasses, and no crows or rav- ens were seen near the sighting area. However, a large logging campsite 1s located about 25 km to the west. Beaver and Showshoe Hare were trapped in the area last winter. Wolverine are known to visit trap lines and investigate cabins (Banfield 1974). The presence of these in the area may have attracted the animal. Wolverine is considered extremely rare in Quebec (Harper 1961; Peterson 1966; Van Zyll de Jong 1975). From 1940-41 to 1981-82, only 72 Wolverine pelts have been registered in the Quebec Division of fur bearer production pelt statistics (Table 1). It is believed that most, if not all, of these pelts have been TABLE 1. Wolverine pelts registered on the Québec fur market as being harvested in Québec in the seasons 1940- 1941 to 1981-1982. Years Pelts 1940-41 1941-42 1942-43 1943-44 1944-45 1946 to 1962-63 1963-64 1964-65 1965-66 1966-67 1967-68 1968-69 1969-70 1970-71 1971-72 1972-73 1973-74 1974-75 1975-76 1976-77 1977-78 1978-79 1979-80 1980-81 1981-82 SSS Awan Rowuk OwornrsaosroowexrD 458 harvested outside Quebec, especially those since 1972, as no trapper has directly reported the capture of any Wolverine for a long time (M. Beaudet, personal communication). Due to the rarity of the species, Wolverine hunting and trapping has been forbidden in Quebec since 1981. This observation confirms the actual presence of the Wolverine in Quebec and provides new evidence that the species still occurs farther south of the normal range (north of the 52nd parallel) given by Peterson (1966) and Van Zyll de Jong (1975). Acknowledgments I am indebted to Marcel Beaudet, Head of the Division of Fur-bearer Production, Ministére du Loi- sir, de la Chasse et de la Péche, for providing me the statistics on Wolverine harvesting in Quebec. THE CANADIAN FIELD-NATURALIST Vol. 97 Literature Cited Banfield, A. W. F. 1974. Mammals of Canada. University of Toronto Press, Toronto. 406 pp. Harper, F. 1961. Land and fresh-water mammals of the Ungava Peninsula. University of Kansas, Miscellaneous Publications 27. 178 pp. Miller, F. L. 1972. Wolverine in Gatineau Park, Quebec. Canadian Field-Naturalist 86(4): 390. Peterson, R. L. 1966. The mammals of eastern Canada. Oxford University Press, Toronto. 465 pp. Van Zyll de Jong, C. G. 1975. The distribution and abun- dance of the wolverine (Gu/o gulo) in Canada. Canadian Field-Naturalist 89(4): 431-437. Received 23 August 1982 Accepted 4 October 1982 Presumed Breeding Record of Brewster’s x Brewster’s Warbler, Vermivora chrysoptera x pinus, in Ontario A. GEOFFREY CARPENTIER 42 Wallis Drive, Peterborough, Ontario, K9J 6B7 Carpentier, A. Geoffrey. 1983. Presumed breeding record of Brewster’s x Brewster’s Warbler, Vermivora chrysoptera x pinus, in Ontario. Canadian Field-Naturalist 97(4): 458-459. On 6 July 1982, a Brewster’s Warbler was observed feeding a fledgling in Belmont Township, Peterborough County, Ontario. The two birds were attended by a male Brewster’s Warbler. No nest was found, nor were any other fledglings. A careful search of the area revealed no Golden-winged or Blue-winged Warblers. This was a presumed successful Brewster’s x Brewster’s Warbler breeding record. Key Words: hybrid, warbler, Brewster’s Warbler Vermivora chrysoptera x pinus, Vermivora leucobronchialis, Ontario. The Golden-winged Warbler (Vermivora chrysop- tera) and Blue-winged Warbler (V. pinus) commonly hybridize where their ranges overlap, producing two hybrid forms known as Brewster’s and Lawrence’s Warblers. The former is by far the more common of the two. Both hybrids are fertile, readily leading to backcrosses with one of the parental types. Hybrid- hybrid matings are apparently very rare. K. C. Parkes (1951) reported finding only two published records of Brewster’s x Brewster’s matings — one in 1922 by Sutton in Pennsylvania and one in 1927 by Hicks in Ohio. On 6 July 1982, at the south end of Hubbel Road, Belmont Township, Peterborough County, Ontario, the author with Clayton Vardy identified a Brewster’s Warbler. The bird was skulking in a hawthorn (Cra- taegus sp.) tangle, and constantly scolded the observers. It apparently was a female, very similar to one pictured in Godfrey’s Birds of Canada (1966) except that no yellow was evident on the breast, and the back was greenish. The bird was very drab witha gray-white breast, yellow crown, black eye-stripe and pale yellow wing bars. It represented the first known record of this hybrid for Peterborough County (D. Sadler, personal communication 1982). Shortly thereafter, the bird was observed carrying food. A search for evidence of breeding revealed a recently fledged bird being fed by the adult. The fled- gling was generally drab gray and white with few plumage contrasts. A small amount of yellow was seen on the wings. The breast was white, while the back and crown were grayish, with no striking patterns on the head, back or face. The young bird skulked during the entire observation period. A third bird appeared shortly thereafter and started scolding us. It was a brightly plumaged Brewster’s 1983 Warbler, presumably a male. It resembled the female seen earlier except that it was much more brightly and boldly marked. The back was gray, the breast was immaculate white with no yellow wash, and the black eye-stripe was wider than that pictured by either God- frey or Parkes. This male constantly scolded us as the female and fledgling skulked away. The male followed the other two birds as they moved away through the underbrush after we ceased our observations. The entire observa- tion period lasted approximately 20 min. No other Brewster’s, Golden-winged, or Blue- winged Warblers were found in that area on that date or subsequently. The birds reported here were not seen on later trips to the area. We believe that this sighting represented a success- ful Brewster’s x Brewster’s Warbler nesting. In 1982, three Brewster’s Warblers were reported during the spring migration period in Ontario (Weir 1982a) anda Brewster’s x Golden-winged Warbler successful nest- NOTES 459 ing was reported from Long Point, Norfolk County, Ontario (Weir 1982b). Acknowledgments The author is grateful to Ron Weir, Doug Sadler and Doug McRae for comments on earlier drafts of this paper and for historical data on Brewster’s Warblers. Literature Cited Godfrey, E. 1966. Birds of Canada. National Museum of Canada Bulletin 203. Parkes, K. C. 1951. The Genetics of the Golden-winged x Blue-winged Warbler Complex. Wilson Bulletin 63: 5-15. Weir, R.D. 1982a. The Changing Seasons — Ontario Region. American Birds 36: 846-849. Weir, R. D. 1982b. The Changing Seasons — Ontario Region. American Birds 36: 970-974. Received 14 January 1983 Accepted 16 March 1983 Two Recent Ice Entrapments of Narwhals, Monodon monoceros, in Arctic Canada D. E. SERGEANT! and G. A. WILLIAMS2 ‘Canada, Department of Fisheries and Oceans, Arctic Biological Station, 555 St. Pierre Blvd., Ste. Anne de Bellevue, Quebec H9X 3R4 2Arctic Bay, Northwest Territories, XOA 0A0 Sergeant, D. E., and G. A. Williams. 1983. Two recent ice entrapments of Narwhals, Monodon monoceros, in Arctic Canada. Canadian Field-Naturalist 97(4): 459-460. A group of about 20 Narwhals stranded at the head of Dundee Bight, Bathurst Island (75°05’N, 100° 15’W) some time prior to July 1979. The locality is close to a published sighting of 10 or more Narwhals ina polynya in September 1976. A second group of at least 108 animals stranded in Agu Bay, Gulf of Boothia (70° 18’N, 86° 30’W) in early October 1979. Both strandings were mainly of females and young. Possible reasons for this are discussed. Key Words: Cetacea, Narwhals, arctic Canada, ice entrapment Ice entrapments (savssats) of both White Whales (Delphinapterus leucas) and Narwhals (Monodon monoceros) are well known in both West Greenland (Porsild 1918, 1922) and arctic Canada. Freeman (1968) described an entrapment of 150-200 White Whales in winter 1966/67 in Jones Sound, and Free- man (1973) one of at least three White Whales in winter 1969/70 off SE Ellesmere Island. Hill (1967) described an entrapment of White Whales in winter 1966/67 in the Eskimo Lakes, Beaufort Sea. However we know of no detailed accounts of entrapment of Narwhals in the Canadian arctic. Two such events, observed in 1979, one current and one from an unknown, earlier date, are described here. Inearly August 1979, Mr. Don Kushner, a geologist with Chevron-Standard Ltd., sent Mr. K. A. Hay photographs of about 20 dead Narwhals discovered by him in late July 1979, at Dundee Bight, Bathurst Island (76°05’N, 100°15’W, Figure 1). The carcasses were lying on top of the ice close to land, and were old. A few were partly eaten leaving the rib cage exposed. Most of the animals could be identified as tuskless animals (females or young males) with a few calves. It is not clear how the animals reached the ice surface, but the agency of man can be discounted at this remote site. Freeman (1973) reported the removal from the water of three entrapped White Whales by a Polar Bear, Ursus maritimus. The whales were then 460 partly eaten by bears and scavenged by Ravens, Cor- vus corax, and Glaucous Gulls, Larus hyperboreus. The region is far northwest of the common occur- rence of Narwhals in Lancaster Sound and Barrow Strait in summer. However, Roe and Stephen (1977) reported seeing a group of at least 10 mostly tusked Narwhals in a large polynya at 77°20’N, 103°30’W (Figure 1) on 6 September 1976. Ice charts of this period show open water in Queens, Maury and Wel- lington channels extending towards the region of the polynya. Since summer 1978 provided severe ice con- ditions in Lancaster Sound restricting the westward travel of Narwhals, it seems likely that the ice entrap- ment can be dated to 1977 or an earlier year, presuma- bly in autumn. A savssat was discovered at Agu Bay, NW Baffin Island (70°18’N, 86°30’W) on or before 9 October 1979. G. A. Williams arrived at the site on 14 October FIGURE 1. Map showing place names and sites of strandings (+) and extralimital occurrence (®) of narwhals de- scribed in text. THE CANADIAN FIELD-NATURALIST Vol. 97 after arrangements had been made for Igloolik hun- ters, as well as those at Agu Bay, to hunt the animals. The animals were then 10-12 km from open water in Fury and Hecla Strait. Ice charts for the period show open water southward into Foxe Basin, from which direction the animals must have come; there was solid ice northward in the Gulf of Boothia. The hunting was organised to avoid wounding and waste, and 108 animals were removed, with an addi- tional four lost. Of 89 animals killed on 15, 16 and 17 October, the composition by age and sex was: adult females 34; juvenile females 22; adult males 1; juvenile males 8; calves of the year of both sexes 24. There were no stomach contents except mud in a few animals. This entrapment, and the earlier one, were therefore very deficient in adult male (long-tusked) animals. Several factors could account for this. Adult males, being more loosely attached to the herds, could break away and escape more readily. Also it is observed that adult male Narwhals in summer tend to inhabit deeper water than the females and calves, and so could be entrapped less often. Acknowledgments Mr. K. A. Hay, Canada Department of Fisheries and Oceans, St. John’s, Newfoundland, most kindly supplied details of the northern entrapment. Literature Cited Freeman, M. M. R. 1968. Winter observations on beluga (Delphinapterus leucas) in Jones Sound, N.W.T. Cana- dian Field-Naturalist 82: 276-286. Freeman, M. M. R. 1973. Polar bear predation on beluga in the Canadian Arctic. Arctic 26: 163-164. Hill, R. M. 1967. Observations on beluga whales trapped by ice in Eskimo Lakes, winter, 1966/67. Inuvik Research Laboratory Report, Inuvik, N.W.T. 13 pp. Porsild, M. P. 1918. On “Savssats:” A crowding of arctic animals at holes in the sea ice. Geographical Review VI(3): 215-228. ° Porsild, M. P. 1922. Scattered observations on narwhals. Journal of Mammalogy 3: 8-13. Roe, N. A.,and W. J. Stephen. 1977. Narwhals(Monodon monoceros) observed near King Christian Island, Northwest Territories. Canadian Field-Naturalist 91: 309-310. Received 29 June 1982 Accepted 20 July 1983 News and Comment Second Announcement: The XIX International Ornithological Congress The XIX International Ornithological Congress will take place in Ottawa, Canada, from 22 to 29 June 1986. Prof Dr. Klaus Immelmann (West Germany) is President and Dr. Henri Ouellet (Canada) is Secretary General. The programme is being planned by an international Scientific Programme Committee chaired by Professor J. Bruce Falls (Canada). The programme will include plenary lectures, symposia, contributed papers (spoken and posters), and films. There will be a mid-congress free day. Pre- and post- Alfred B. Kelly Ornithological Scholarships Two Alfred B. Kelly Ornithological Scholarships valued at $500 each are available annually from the Province of Quebec Society for the Protection of Birds, Inc., to post-graduate students working on Quebec birds, regardless of place of residence. Selec- tions will be based on the merit of both the application and his/her project, as well as need. Holders of major scholarships and bursaries do not qualify. For appli- cation forms, please write to: congress excursions and workshops are planned in various interesting ornithological regions of Canada. Information and requests for application forms should be addressed to: Dr. Henri Ouellet Secretary General XIX Congressus Internationalis Ornithologicus National Museum of Natural Sciences Ottawa, Ontario, Canada KIA 0M8 Dr. David M. Bird PQSPB Education Committee P.O. Box 210 Macdonald College of McGill University 21 111 Lakeshore Road Ste-Anne de Bellevue, Quebec H9X 1C0 Canada The deadline for completed applications is 31 May 1984 and applicants will be notified by 30 June 1984. Raptor Collisions with Utility Lines: A Call for Information The United States Bureau of Land Management, Sacramento, in cooperation with the Pacific Gas and Electric Company, is assembling all available pub- lished and unpublished information concerning colli- sions of raptors with power lines and other utility lines. Actual case histories — no matter how circum- stantial or fragmentary — are needed. Please acknowledge that you have such information by writ- ing to Dr. Richard R. (Butch) Olendorff, U.S. Bureau of Land Management, 2800 Cottage Way, Sacra- mento, California 95825 U.S.A. (Phone (916) 484- 4541). A form of which to record your information will then be sent by return mail. Honorary Membership and 1983 Ottawa Field-Naturalists’ Club Awards Four Ottawa Field-Naturalists’ Club awards were presented at the 1983 Soirée. Certificates were pres- ented to the winners by Dan Brunton, President, and once again our thanks to Anne Gruchy for the cali- graphy of the certificates. The Anne Hanes Natural History Award was not presented this year. The award citations are reproduced here. They have been given earlier in Trail & Landscape 17(4): 198-200. Honorary Member: Hue N. MacKenzie A member of The Ottawa Field-Naturalists’ Club 461 for over twenty years, Hue has a long history of Club participation. He was a member of the Council between 1964 and 1980, was Vice-president in 1966 and was President from 1967 to 1969. He has been a member of various committees and chairman of sev- eral. His most recent chairmanship was that of the Centennial Steering Committee. Hue was instrumental in developing what is now the very active Conservation Committee. He also served as Club representative to the Federation of Ontario Naturalists for many years. He played a key role as co-ordinator and participant in the Club’s Native 462 Orchid Location Survey. Hue has contributed to both Trail & Landscape and The Canadian Field- Naturalist, has made presentations in the monthly meeting series and has led field trips on behalf of the Club. Indeed, it is difficult to point to a major Club activity over the past two decades that has not profited from the “MacKenzie Touch”. In the mid-1970s Hue carried out a study of Club Operations which led to a report to the Council entitled “Club Policies and Management Practices”. This report led, in turn, to increased efficiency in Operations and a strengthening of Committees and better definition of their roles. In 1974 Hue moved the resolution that led to the Centennial Planning Group, and subsequently, to the Centennial Steering Committee, which he chaired until its dissolution in late 1982. The enormous effort that Hue put into the preparation, planning and exe- cution of Centennial Year projects is largely responsi- ble for so many of these being so successful. In his efforts on behalf of The Ottawa Field- Naturalists’ Club, we have seen Hue’s dedication to the Club and his achievement as a naturalist. We are happy indeed to have Hue MacKenzie join our select group of Honorary Members. Member of the Year: Roger Taylor The Member of the Year Award is given to the member judged to have contributed the most to The Ottawa Field-Naturalists’ Club during the previous year. For the Club’s second such award, the Awards Committee was unanimous in the selection of Roger Taylor. Roger has a well-deserved reputation for stepping into the breach, to prevent an ongoing program from faltering, or to take a new initiative. In essence he is a catalyst. The Member of the Year Award is in recognition of this type of activity, specifically the reorganization of The Shrike including the creation of a computer- based file of bird records; the lead role as a member, and later chairman, of the Conservation Committee in bringing Federal and Provincial Government atten- tion to the potential damage that raccoon dogs could do to the Canadian environment; his co-ordination of letter writing campaigns on behalf of Ontario Parks and wetlands; and his co-ordination of the fight against Amendment 24 of the Regional Municipality of Ottawa-Carleton. In addition to the foregoing, Roger is Chairman of the Conservation Committee and the Nominating Committee, a member of the Federation of Ontario Naturalists Board of Directors, liaison for the Mem- bership Committee with the National Research THE CANADIAN FIELD-NATURALIST Vol. 97 Council Computation Centre, and is active in birding and other Club outings. Service Award: William H. Knight Behind every great man, they say, there is a great woman. Behind every great naturalists club there are those toilers who do all the important things that need to be done to keep things moving, and that are so often taken for granted in the smooth operation of the club. It is our pleasure this year to select from our band of background workers for the 1982 Service Award, Bill Knight. Over the years Bill has been a tireless worker. He is the backbone and mainstay of the Education and Publicity Committee and often is the one to ensure that the Club meets its commitments for displays and publications. Bill prints and distributes The Shrike, and, as well, he prints notices, tickets, placemats (such as for the Centennial), certificates and awards. He is familiar with the situation in which fast action is required because everything was left a little bit late, or because we all know Bill will get it done. Bill has produced the OFNC Award certificates, so it is particularly fitting — and a real pleasure — to present him with one of these and to add to it our most sincere thanks for all his efforts on behalf of the Club. Conservation Award: H. Loney Dickson Loney Dickson was a Vice-President of the Club for a year and a half prior to his departure for Alberta in 1981. He was an active member, and later chairman, of the Conservation Committee, and by virtue of his leadership the committee broadened its outlook to. cover the range of issues it now handles. His voice of reason has been heeded in times of controversy, and today the Conservation Committee is widely recog- nized as the most important conservation voice in eastern Ontario. Loney co-authored (with Stephen Darbyshire) a report on significant natural areas for the National Capital Commission. It has been important in estab- lishing protection for a number of valuable natural areas in Ottawa-Carleton. Loney played a prominent role in the re- establishment of special interest groups within the Club. The success of this move his been quite evident. The Ottawa Field-Naturalists’ Club’s 1982 Conser- vation Award is made to Loney Dickson as our recog- nition of this young man’s energy and efforts to teach us all the value of natural areas and their contents. BILL GUMMER and members of the Ottawa Field-Naturalists’ Club Awards Committee Book Reviews ZOOLOGY Fisheries and Wildlife Resources and the Agricultural Land Base in Alberta By William M. Glasgow. 1982. Fish and Wildlife Division, Alberta Energy and Natural Resources and Environment Council of Alberta, Edmonton. 65 pp. Free. The Environment Council of Alberta is presently con- sidering threats to the Province’s Agricultural land base. This report was prepared through the Alberta Fish and Wildlife Division as a background document on fish and wildlife resources in agricultural areas. The stated purposes are to document the general capabilities, the historical changes, and the uses made of the fish and wildlife resources as well as the land use conflicts caused by coexistence of agricultural and fish - and wildlife government programs. Suggestions are made for future programs. The abstract clearly states that agricultural land supports important fish and wildlife populations which are used by Albertans. Agricultural practices modify these resources. Cooperation and wise land use will be keys to fish and wildlife as well as agricultu- ral productivity. The report states that management techniques will be discussed. However, it expands on most of these points, except management techniques. This is a result of the lack of distinction between agricultural and non-agricultural land, the general nature of the concepts considered, and the lack of attention to the costs of government programs. Rarely is a distinction made between the fish and wildlife resource and its use in agricultural versus non-agricultural Alberta. Furthermore there are important differences among types of agricultural areas which are not addressed. It is very difficult to suggest techniques, except in the most general terms, if you don’t know whether you are planning for a trout fishery or duck hunting on a body of water. Superficial treatment of important concerns is fre- quent. For example the section on “Management The Park Buffalo By Sheilagh C. Olgivie. 1979. National and Provincial Parks Association of Canada, Calgary. 68 pp. illus. $4.95 plus postage. Seeing Bison ina National Park is a memorable expe- rience for many people. This book details the history of these Bison from the days of overhunting, to the origin of captive herds, their subsequent growth and spread in Canada’s National Parks. It willadd a great Problems Common to Fishand Wildlife and Agricul- ture” discusses planning land use on private land, hunter access to private land, and wildlife damage to agricultural products. However, it ignores major problems such as level of water table, salinization, and soil erosion. Curiously, the financial benefits of fishing and hunting, the money from fishermen and hunters ear- marked for enhancement programs, and the cost of wildlife damage to crops and livestock are always explicitly stated but the costs of government enhancement programs are never stated. Whether stocking sloughs with Rainbow Trout or offering incentives to raise pheasants or controlling magpies are sensible programs depends ona comparison of the costs with the results. How much does each trout or pheasant cost to produce? How much does it cost to killa magpie? Who benefits? One recommendation of this report is increased funding for wildlife manage- ment. This cannot be seriously considered without an assessment of the value of existing programs, future needs, and efficiency of use of present funding. Glasgow states that “If goals for fish and wildlife management on agricultural land were more clearly defined, it is likely that Alberta Agriculture would be willing to discuss these goals along with their own and explore cooperative land management and assistance programs.” This statement, like much of the report, is good common sense and common knowledge. How- ever, this report does not clearly or explicitly define management goals or suggest a starting point for their definition. DAVID McCORQUODALE Department of Zoology, University of Alberta, Edmonton, Alberta T6G 2E9 deal of understanding and appreciation to seeing these Bison. Much interesting information is presented ina very appealing fashion that is suitable for anyone high school age or older. Tales of colourful people (such as Samuel Walking Coyote, a Pend d’Oreille with a Flat- head wife and a more than passing interest in a Black- foot maiden, who used Bison calves — ancestors of 463 464 today’s park Bison — as an unsuccessful peace offer- ing to his alienated Flathead relatives) maintain inter- est throughout as information is easily assimilated. Concentration on the actions of specific people and specific events is a major reason why this book is successful. An optimistic thread runs throughout, lar- gely because of the success of Bison conservation. Despite this, unmitigated disasters, such as the intro- duction of diseased Plains Bison into the only known herd of Wood Bison in 1925, are not glossed over. The contrast of successes and failures allows reflection and consideration of how to best carry out a conservation program. Occasionally, wording is sloppy or melodramati- cism takes over, but generally the attention to detail and enthusiasm wins out. My only serious complaint is about the suggestion that some cattalo should have been released into northern Alberta instead of being slaughtered. After the Wood Buffalo National Park THE CANADIAN FIELD-NATURALIST Vol. 97 experience and the capability to introduce pure Bison, this should never have been considered. The book is a sturdy paperback, relatively free of typographical errors, and reasonably priced. The illustrations, line drawings, historical photographs, and reproductions of C. M. Russell paintings, add a great deal to the value and appeal of the book. The reference list of technical and popular accounts of bison and conservation will allow the curious reader to pursue their interests at their own level. Anyone whois interested in the origins of the Bison they have seen in Canada’s National Parks will find this book worth reading. DAVID McCORQUODALE Department of Zoology, University of Alberta, Edmonton, Alberta T6G 2E9 Eastern Chipmunks: Secrets of their Solitary Lives By Lawrence Wisher. 1982. Smithsonian Institution Press, Washington. 144 pp., illus. U.S. $17.50. The author, a professor of chemistry at Mary Washington College in Virginia, has written a detailed study of the eastern race of chipmunks (Tamias stria- tus lysteri) inhabiting Eastern Canada and the Nor- theastern United States. The 108 particular individu- als of which he reports occupied 14 acres of his land near Fredericksburg, and were observed over a period of six years. His account is “structured as a chronological narra- tive into which are woven the details and conclusions concerning the chipmunk’s life history.” The last of the nine chapters in the book summarizes the essential biological and ethological data concerning the East- ern Chipmunk. Appendices provide biographical sketches of five of the individuals whose lives were most closely researched, a catalogue of chipmunk behaviors, the taxonomic history and description of the species, and the reproductive events in the lives of four female chipmunks. The bulk of the book, after providing a family tree of North American squirrels and a brief assessment of chipmunks in nature, proceeds to a detailed discus- sion of the life and times of some of the individuals Wishner observed from 1973 through 1980. The 101 photographs (20 in color), maps, diagrams and draw- ings do much to enhance the reader’s understanding of the text account. One delightful example of the color camera’s ability to freeze thousandths of a second in time 1s a picture showing a male cardinal and a chipmunk seemingly dining peacefully together on sunflower seeds on a tree stump. In fact, as the author indicates, the cardi- nal hadn’t as yet divined the fact that it had company when the photo was taken, and “Within a fraction of a second both had fled the scene.” Tamias, the author reminds us, “is evolutionarily and behaviorally one of the most primitive of genera of living squirrels.” They are solitary animals, “geneti- cally programmed” to store food. One animal, it was estimated, collected some 928 acorns over her eight hours of harvesting activity ina single day. This would tend, as Wishner emphasizes, to encourage their soli- tary behavior. While chipmunks can distinguish “between perishable and non-perishable food for storage purposes,” they “cannot seem to stay out of traps,” even those which lack bait. The evidence for their intelligence would therefore seem to be mixed at best. This is a most delightful and informative volume, and one worth careful reading as well as reference. KEIR B. STERLING Department of History of Science, University of Wisconsin, Madison, Wisconsin 53706 1983 BOTANY BOOK REVIEWS 465 The Arctic-Alpine Element of the Vascular Flora at Lake Superior By David R. Given and James H. Soper. 1981. Publications in Botany, No. 10. National Museum of Natural Sciences, Ottawa. 70 pp., illus. Free. Ever since the time of Louis Agassiz, Lake Superior has been famous for the occurence of a disjunct arctic- alpine element on or near its shores. Although there has been much collecting of these plants and a number of articles on small areas or specific species, there has never been a general overview of the topic. Here we have one — a thorough and well illustrated treatment. Prefaced by a brief introduction to the literature and historical background, a major part of this account is a thorough discussion of the arctic-alpine floristic element. Forty-eight species were selected as representative of this element. A few species that were admittedly more wide ranging, such as Primula mis- tassinica, were included “because although they do not occur north of the Low Arctic, they are intimately associated with typically arctic-alpine species throughout much of their range. . .” No author would come up with the same selection of representative species, but certainly, enough species, including all the major ones, were selected to give a clear picture of the phenomenon. The 48 species are treated in more detail in an annotated list appended after the bibliography. Note that Carex heleonastes, rejected on p. 6 as being unsubstantiated in the region, has recently been found in the Lake Superior region of Michigan (Michigan Botanist 21: 169. 1982). Future field work will un- doubtably turn up additional species. A discussion of the vegetation in which these species occur forms an important part of this account. Frag- mentary shoreline and cliff and talus vegetation, which is the habitat of these species, is very difficult to classify. In fact, the first sentence introducing the vegetation section states that “Although discrete vege- tation units have been described, most are probably part of a continuum.” This states the case well, but then it is somewhat surprising to see the description of vegetation “units” go on for 13 full pages. One sus- pects that if more sites had been examined, there would have been still more “discrete units”. The distributions in the Lake Superior region (North America in the case of Cerastium alpinum and Castilleja septentrionalis) are mapped, very com- pletely, for 14 species. The distributions of all 48 spe- cies are grouped into six patterns preliminary to the discussion of the microclimatological effects of Lake Superior on the distributions. Perhaps the most inter- esting part of this account is the discussion of the origins of the arctic-alpine element. Palynological, geological and climatological evidence is synthesized into a plausible account tracing the migration of the arctic-alpine element from the edge of the glacial front northward and describing how they may have later become “stranded” in the Lake Superior region. If this account has any significant weakness, it is that there is too much concentration on Lake Super- ior, with its obvious climatic impact. The conclusions draw little from the occurence of arctic-alpine species (including a few, like Rhododendron lapponicum and Saxifraga aizoides that do not now occur near Lake Superior) in areas to the south such as the gorges in Algonquin Park, the driftless zone of Wisconsin, and even western New York. There is work to be done yet before these occurrences can be reconciled with those of Lake Superior to produce a comprehensive hypothesis for the disjunct ocurrence of northern spe- cies south of their continuous ranges. Also, little men- tion is made that other disjunct elements, notably western species and northern boreal forest species, which also are a prominant feature of the flora of the Lake Superior region. These might well provide clues to the survival of the arctic-alpine element. These are minor quibbles. Everyone interested in the phytogeography floristics and vegetation of the Great Lakes region should get a copy of this account. A. A. REZNICEK University of Michigan Herbarium, North University Bldg., Ann Arbor, Michigan 48109 466 ENVIRONMENT THE CANADIAN FIELD-NATURALIST Vol. 97 Mountain Islands and Desert Seas: A Natural History of the U.S.-Mexican Borderlands By Frederick R. Gehlbach. 1981. Texas A and M University Press, College Station. xi + 298 pp. U.S. $19.95. This book is the result of more than a quarter century of travel and study of the “Borderlands” from the mouth of the Rio Grande to California. The author, a professor of biology and environmental studies at Baylord University, skillfully weaves the natural and unnatural (i.e., man-made) histories of this region into an interesting, informative narrative. Gehlbach’s approach to his subject is a combination of the scient- ist, naturalist, and philosopher. While pondering the inevitable changes wrought by humans, his observa- tions are not only put in context of other studies done ona particular area since 1950, but are also compared to the observations of early naturalists visiting this region. Two of the 19th century naturalists were employed by the first U.S.-Mexican Boundary Sur- vey, 1849-1855. They were John Bartlett, U.S. Boun- dary Commissioner (1850-53), and his successor Wil- liam Emory (1854-55), whose itinerary is followed by Gehlbach. The author uses their personal discoveries, as recorded in their diaries, for comparative purposes. He also utilizes the writings of Edgar Mearns, army surgeon, ornithologist, naturalist, who was employed by the second U.S.-Mexican Boundary Survey of 1891-96 for a two year period (1894-96); and those of Vernon Bailey of the U.S. Biological Survey, who studied the border region early in the 20th century. Contrasting the natural history of the varied terrain MISCELLANEOUS Casey A. Wood (1856-1942): A Bio-Bibliography Compiled by Effie C. Astbury. 1981. Occasional Papers. Graduate School of Library Science. McGill University. Montreal. 66 + vi pp. $5.00 (paper). A quick perusal of the Bibliography of Canadian bib- liographies assures one that bio-bibliographies, pro- ducts of various library schools in Canada, are not on the endangered species list. Astbury’s work on the existing data on ophthalmologist, naturalist, orni- thologist, bibliographer and world traveller Casey Albert Wood is of particular interest to North Ameri- can naturalists. Wood was born at Wellington, Onta- rio, of American parents, and received all his educa- tion in Canada. He graduated from the Medical Faculty of the University of Bishop’s College, Mont- real and Lennoxville, in 1877. He practiced general of the “Borderlands” as seen by a modern biologist in the second half of the 20th century with the observa- tions of earlier naturalists enables the author to dem- onstrate both the natural changes in the mountains, canyons and deserts in this long study area and the destruction of the natural areas wrought by the “arch predator”, man. Environmental changes inevitably followed habitat destruction, some of them having undesirable, far reaching effects. Gehlbach uses side-by-side photographs, mostly in colour, to illustrate the natural and unnatural histo- ries of certain areas. Other photographs depict natural communities, plants, birds, fish, amphibians, and rep- tiles found in the “Borderlands”. There are three use- ful maps of the eastern, middle and western sections of the boundary. The numerous reproductions of Bart- lett’s and Emory’s original illustrations of both land- scape and plant and animal life, which are included in each chapter, add considerably to the interest of this work. A word of warning! This book is not easy reading. Obviously the work of a university professor, every chapter is well-documented with a profusion of foot- notes. These references, however, add considerably to the usefulness of the work. It is recommended for serious naturalists and conservationists. MARIANNE GOSZTONYI AINLEY 4828 Wilson Avenue, Montreal, Quebec H3X 3P2 medicine and surgery in Montreal for a while and was also professor of chemistry at Bishop’s. In 1886, Wood married Emma Shearer of Montreal and decided to leave for the U.S. to specialize in eye and ear diseases. It is not clear from this bio-bibliography when Wood developed an interest in nature. Astbury cites his first “ornithological” paper, one published on the eye and eyesight of birds, in the journal Ophthalmol- ogy in 1907. This was obviously a precursor for his beautifully illustrated, much acclaimed book, the Fundus Oculi of Birds, which was published in 1917, the year Wood joined the American Ornithologists’ Union. It was in 1917 that Wood retired from private practice and moved to California, and as far as one 1983 can gather developed a greater interest in ornithology and the bibliography of ornithology. In 1919 he founded and endowed the Emma Shearer Wood Library of Ornithology at McGill University. The following year he persuaded Robert R. Blacker, retired industrialist from Pasadena, to establish and endow a library of zoology at McGill, to be known as the Blacker Library of Zoology. From the early 1920s until his death in 1942 Wood, accompanied by his wife, and their pet John III, a “double yellow-headed parrot from British Hondu- ras” (which incidentally appears on many of the book- plates of the library), travelled around the world col- lecting treasures. These included manuscripts, books, and paintings on birds. His interest in ornithology earned Wood a number of honorary positions, such as honorary curator, collaborator, or lecturer at various North American institutions. In the mid-1920s Wood persuaded a number of well-known naturalists to deposit their correspondance in the archives of the Emma Shearer Wood Library of Ornithology. It was about this time that he began working on a catalogue of all zoological works in the McGill Libraries, which Dammed Indians: The Pick-Sloan Plan and the By Michael Lawson. 1982. University of Oklahoma Press, Norman. xxvi + 261 pp., illus. U.S. $19.95. The author, an historian with the Office of Rights Protection in the Bureau of Indian Affairs, is con- cerned here with the effects upon the Sioux Indians of U.S. Army Corps of Engineers dams constructed on the Missouri River in North and South Dakota from 1946 into the 1960's. The Pick-Sloan Plan, devised and approved during the later stages of the Second World War, came about as the result of a political marriage between the two federal agencies most concerned with water engineer- ing, the Engineers and the Bureau of Reclamation. These two agencies had responded to demands for better management of the Missouri River and certain of its tributaries following disastrous floods in the spring of 1943 and 1944. Though Col. Lewis A. Pick, then Divisional Engineer for the Corps in Omaha, and William Glenn Sloan, assistant director of the Bureau of Reclamation’s Billings, Montana office, had come to cordially loathe one another when debate over the merits of their respective river plans became bogged down in Congress, they disliked the prospect of another Tennessee Valley Authority (TVA)-like pub- lic corporation in the area even more. The result was a quick coordination of plans in 1944 aptly termed by one critic “a shameless, loveless shotgun wedding.” BOOK REVIEWS 467 has constituted one of the important zoological refer- ence works ever since. Whilst Astbury is restricted to a brief chronology of Wood’s life, there is enough material there to whet someone’s appetite to write a proper biography. The remaining sections of this publication consist of a chronological listing of Wood’s publications, and a list of sources used, both biographical and bibliogra- phical. These are full of useful information on the productive Dr. Wood. The bibliography does not merely list his works, but also includes reviews, where appropriate, of Wood’s writings. This section would have been even more useful to the naturalist if Wood’s publications were separated and grouped under sub- ject headings, such as ornithology, ophthalomology, etc., instead of chronologically. This may not be the way things are done, however, in library schools. Ms. Astbury is to be recommended for a thorough job. Its unpreposessing appearance notwithstanding, this is a useful and inexpensive publication. MARIANNE GOSZTONYI AINLEY 4828 Wilson Avenue, Montreal, Quebec H3X 3P2 Missouri River Sioux, 1944-1980 While the Engineers had stressed flood control and navigation and the Bureau irrigation and power development in their respective plans, Pick-Sloan simply incorporated most of the merits, as well as the defects of the original proposals. Some 107 dams were called for, and many details were left to be worked out later. Despite the admission by its proponents that there might not be enough water in the river “to pro- vide for both irrigation and navigation,” the plan was approved by Congress in 1944. Little consideration appears to have been given to the rights of Indians living in the river basin. In building their dams, the Corps of Engineers des- troyed much of the prime Sioux Indian lands and dislocated hundreds of families. The physical and psy- chological stresses which resulted were enormous, and seem not to have been anticipated by any of the government agencies involved. The problem was compounded by the fact that no central authority coordinated all aspects of the work and by the fact that the Army was preoccupied with engineering con- siderations. The Bureau of Indian Affairs was unfor- tunately hobbled by budget cuts and by threats to eliminate it altogether from the federal establishment. Later efforts to revise and rework Pick-Sloan lar- gely failed despite the wide publicity given the failures of the plan’s administration. Public cries for more 468 flood control in the wake of more disastrous floods and able public relations by the Corps kept everything essentially on track. The Indians, therefore, concentrated on seeking cash settlements for their lands and additional funds for relocation and rehabilitation. These proved inadequate by most objective standards and did not compensate the Indians for the continuing damage done to their remaining property by various man- made projects. Nor have the prospects for electric power and irrigation panned out. The amount of traffic on the Missouri River has not been adequate “to justify the huge expense in keeping the river open.” Congress rejected proposals ten years ago for $60 billion in additional funding to keep the river dredged and open. Recreational use of the man- made reservoirs on the main stem of the river has been made difficult because, in some cases, entire forests had been left standing in areas that were later flooded, resulting in man-made lakes of limited use for boating due to navigational obstructions. While the fishing in these reservoirs is generally excellent, the federal agencies having stocked them with game species, the boom in tourism and recreation hasn’t benefitted the Indians, who suffer in some parts THE CANADIAN FIELD-NATURALIST Vol. 97 of the region from a “strong anti-Indian economic bias.” Indians have not traditionally been concerned with fishing, boating, and swimming. Hunting, once their favorite sport, has been largely devastated by the destruction of much of the habitat for the local game animals by dammed-up waters. The author concludes that Pick-Sloan “would cer- tainly not have passed muster” in the 1970’s because of better federal standards for the protection of the environment and the enhanced ecology mindedness of the general public. The Indians now have a good deal more political sophistication than they had forty years ago. Yet various federal agencies continue to propose land use plans which violate basic Indian treaty rights. The total picture is one which demonstrates “the sham of native rights and hypocrisy of federal Indian poli- cies in the twentieth century.” This is a sobering and provocative volume whose lessons should be read and digested. KEIR B. STERLING Department of History of Science, University of Wisconsin, Madison, Wisconsin 53706 Biogeographical Dictionary of Rocky Mountain Naturalists: A Guide to the Writings and Collections of Botanists, Zoologists, Artists, and Photographers, 1682-1932 By Joseph and Nesta Dunn Ewan. 1981. Bohn, Scheltema, and Holkema/ Junk (North American distributor Kluwer, Boston). xvi + 253 pp. U.S. $37.50. This excellent reference, originally published in 1950 under the title of Rocky Mountain Naturalists, has been expanded, revised, and updated, though the ter- minal date (for coverage) of 1932 has been retained. The introductory essays in the earlier work have been omitted here, and the biographical and bibliographi- cal material has been expanded by about one third. Smaller type has been used in this edition, but the volume is well printed and bound. In any work of this nature, there are bound to be errors, some trivial, as for example “Westpoint” for “West Point” in the sketch for Amiel Weeks Whipple (p. 237). Others, however, are a bit more serious. The sketch for Louis Agassiz, for example, makes no men- tion of his founding of the Museum of Comparative Zoology at Harvard. The sketch for Joel Asaph Allen neglects his role in the founding of the American Ornithologists’ Union (AOU) in 1883, and his presid- ency of that organization, but does give credit for being a charter member and later president to C.F. Batchelder, not as central a figure. No mention is made of the AOU or their presidencies of it in the sketches of William Brewster or Elliott Coues either, though both of them and Joel Asaph Allen were important figures in the founding of this influential group. The sketches for John and Joseph Le Conte are unfortunately not entirely accurate (pp. 130-131). John Le Conte was acting president of the University of California, rather than his brother Joseph. Nor was John Le Conte the first president, but rather the uni- versity’s third chief executive. The reference to Frank M. Chapmanasa “lifetime associate” of the American Museum of Natural History is a bit misleading; Chapman was an associate curator and later a long- time curator and chairman of the Department of Ornithology. Othniel C. Marsh was a professor of paleontology at Yale from 1866, not 1886, though he received no salary until 1896. Edward Drinker Cope, his great rival, is recorded here as having made his first trip to Europe in 1873-1874, whereas it actually was a decade earlier. Despite these caveats, however, this volume is an essential one which ought to be on the shelves of every individual concerned with the Rocky Mountain 1983 region and its history. The authors have defined their geographical boundaries very broadly and point out that it is “not physiographically restricted.” Natural- ists, historians of science, biologists, taxonomists, wildlife management specialists, students of Western life and culture, and many others will find much information here not easily accessible in several Lucy Audubon: A Biography By Carolyn De Latte. 1982. Louisiana State University Press, Baton Rouge. xiii + 148 pp. U.S. $15.95. John James Audubon (1785-1851) was perhaps in many respects the quintessential North American artist-naturalist of the 19th century. His life and his portraits of North American birds and mammals have received repeated attention over the years from var- ious authors. Library shelves groan with various edi- tions of his works and extracts from them, to say nothing of the serried ranks of biographies which have been written about him. Very little attention, how- ever, has been given to the indomitable lady who married him, shared his triumphs and trials for nearly 43 years, lost two daughters in infancy, raised two sons — themselves artists of no mean ability — and outlived all of them. The English-born Lucy Green Bakewell enjoyed a life of genteel affluence for much of her life from the time of her birth in 1787 to her marriage in 1808. Her father suffered business reverses in the United States following his arrival in Pennsylvania in 1802, and Lucy’s mother died several years later. Her father remarried, but it soon became apparent that the step- mother would increasingly try to improve her own position at the expense of that of her stepchildren. De Latte does not essentially revise the standard accounts of John James Audubon’s frequent lack of consideration for his long-suffering family, his decid- edly mixed record in matters of business, his pro- tracted absences in the field or in Europe, nor the fabrications about his origins and training which have sometimes misled students of his career over the years. She does shed considerable light on part of Lucy’s life, especially the period from 1807 until 1830, where the sources are fullest. We learn, for example, that “Lucy never lost the knack of making people, particu- larly other females, feel somehow inferior and always at a disadvantage.” Since she had to accept employ- ment as a teacher of young children during much of her early married life, her path cannot have been a very easy one, though De Latte records that most of her employers were very happy with her work. Lucy’s early circumstances cannot have prepared BOOK REVIEWS 469 shelves full of other references. It deserves wide circu- lation despite its comparatively high price. KEIR B. STERLING Department of History of Science, University of Wisconsin, Madison, Wisconsin 53706 her for the “turbulent marriage, grim poverty, and life on the rowdy American frontier.” On one occasion, Lucy and John James decided to visit her parents in Pennsylvania. This entailed a horseback trip of some 800 miles in the dead of winter with a two year old son who rode with his father most of the way and seems to have thrived on the experience. The trip was com- pleted in 24 days, including a four day break with another relative. Lucy, whose “experiences with stagecoaches had been most unpleasant,” seems not to have objected to this arrangement, for she was an_ accomplished horsewoman. We are told that Lucy had a lively sense of humor, without which “she could not long have endured a man like John James.” Fortunately, too, her “excel- lent education” prepared her to bea full partner to her husband in his various ventures, and he relied upon her judgement. She always felt that he could find recognition for his artistic and ornithological talents in her native country, and her convictions ultimately bore fruit. The strains of separation on this occasion, however (three years and eight months) brought serious misunderstandings in its train, and Lucy feared that her husband had placed his bird projects ahead of her. Once convinced that this was not the case, she labored long and hard to make the Birds of America a success. This was an unusual arrangement for that period, since through force of circumstance and “by mutual consent, Lucy became family pro- vider [for some time] and guardian of the Audubon coffers.” Lucy’s life “recedes into near obscurity” following Audubon’s death in 1851, though we are told she had 15 grandchildren, some of whom helped lighten her last years. The oft-told tale of her having been forced to sell her husband’s original bird paintings to the New York Historical Society for a tiny fraction of their value is not related here, though it would have been of interest to readers unfamiliar with it; indeed, the last 43 years of her life are compressed into but 14 pages. This gives the reader a strangely skewed picture of “Madame Audubon,” and does some injustice to her accomplishments in this long period. This is not 470 entirely the fault of De Latte, though one wishes she had made more generous use of the available secon- dary sources for this period. One charming article by George Bird Grinnell, for example, later founder of the first Audubon Society and long time editor-publisher of Forest and Stream, was published in the Auk about 70 years ago. This told of his impressions of life in the Audubon Park neigh- borhood, where he lived as a child. Grinnell related something of Lucy’s talents as a teacher and his delight when she left him a picture he was fond of which had hung in her house for many years. He also tells us something of Audubon’s son John. Regretta- bly, this item is not mentioned in De Latte’s notes, nor NEW TITLES Zoology Antarctic wildlife. 1982. By Brian Sage. Facts on File, New York. 160 pp., illus. U.S. $22.95. The canine clan: a new look at man’s best friend. 1983. By John C. McLoughlin. Viking, New York. XIV + 162 pp., ills. U.S. $15.75. +Ce qu'il advient des truites ensemencées dans les eaux de la plaine de Montréal, d’aprés les expériences d’étiquetage et de recapture, de 1977 a 1981 inclusivement. 1982. Par Jean- René Mongeau et Jocelyne Brisebois. Ministére du Loisir, de la Chasse et de la Péche, Québec. 47 pp.., illus. Effects of highways on wildlife populations and habitats: phase I: selection and evaluation of procedures. 1982. By Lowell Adams and Aelred Geis. United States National Technical Information Service, Springfield, Virginia. 171 pp. U.S. $16.50. +A field guide to birds of the U.S.S.R. 1983. By V. E. Flint, R. L. Boehme, Y. V. Kostin, and A. A. Kuznetsov. Prin- ceton University Press, Princeton. c420 pp., illus. U.S. $65.00. +Five new world primates: a study in comparative ecology. 1983. By John Terborgh. Princeton University Press, Prin- ceton. c312 pp., illus. Cloth U.S. $40.00; paper U.S. $13.50. The great extinction: the solution to one of the great myster- ies of science, the disappearance of the dinosaurs. 1983. By Michael Allaby and James Lovelock. Doubleday, New York. x + 182 pp. U.S. $13.95. Herbivorous insects: host-seeking behavior and mecha- nisms. 1983. Edited by Sami Ahmad. Academic Press, New York. c590 pp. no price given. The king fisher. 1982. By David Boag. Blandford, New York. vii + 120 pp., illus. U.S. $14.39. THE CANADIAN FIELD-NATURALIST Vol. 97 in her bibliography, though she has used other Auk articles among her sources. Withall, it is refreshing to see the distaff side of this family given some well merited, if long overdue atten- tion. Lucy Audubon had a good measure of courage and devotion and deserves greater credit than most writers have previously given her for her husband’s successes. KEIR B. STERLING Department of History of Science, University of Wisconsin, Madison, Wisconsin 53706 Mammalian biology in South America. 1982. Edited by Michael A. Mares and Hugh H. Genoways. Papers froma symposium, Linesville, Pennsylvania, May, 1981. Univer- sity of Pittsburgh Pymatuning Laboratory of Ecology, Linesville. x1i + 540 pp., illus. U.S. $30.00. *Mammals of the northern great plains. 1983. By J. Knox Jones, jr., David M. Armstrong, Robert S. Hoffmann, and Clyde Jones. University of Nebraska Press, Lincoln. xl + 379 pp., illus. U.S. $32.50. Mosquito. 1982. By Oxford Scientific films. Putnam’s, New York. 32 pp., illus. U.S. $8.95. Parental behaviour of rodents. 1983. Edited by R.W. Elwood. Wiley-Interscience, New York. x + 296 pp., illus. U.S. $49.95. The sharks of North American Waters. 1983. By Jose I. Castro. Texas A&M University Press, College Station. xil + 180 pp., illus. + plates. Cloth U.S. $19.50; paper U.S. $9.95. ; *Vanishing fishes of North America. 1983. By Dana Ono, James D. Williams, and Anne Wagner. Stone Wall Press, Washington, xiii + 257 pp., illus. U.S. $27.50 (Cdn. $34.95). Waterfowl of the Chesapeake Bay. 1982. By B. Meanley. Tidewater Publications, Centreville, Maryland. xiv + 210 pp. U.S. $19.95. Elk of North America: ecology and management. 1982. Edited by J.W. Thomas and D.E. Toweill. Stackpole, Har- risburg, Pennsylvania. 720 pp. U.S. $39.95. Wolves of the world: perspectives of behavior, ecology, and conservation. 1982. Edited by Fred H. Harrington and Paul C. Paquet. Noyes, Park Ridge, New Jersey. 474 pp. U.S. $48.00. 1983 Botany Atlas of airborne fungal spores in Europe. 1983. Edited by S. T. Nilsson. Springer-Verlag, New York. 139 pp., illus. + plates. U.S. $50.00. Biology and ecology of weeds. 1982. Edited by W. Holzner and M. Numata. Junk, The Hague. 464 pp. Dfl. 265. Contributions to the ecology of halophytes. 1982. Edited by D. Navin Sen and K. Singh Rajpurohit. Task for Vegeta- tion Science 2. Junk, The Hague. 272 pp. Dfl. 165. The genus A triples (Chenopodiaceae) in Canada. 1983. By J. J. Bassett, C. W. Crompton, J. McNeill, and P. M. Taschereau. Monograph No. 31. Agriculture Canada, Ottawa. 72 pp. Grass genera of western Canadian cattle rangelands. 1983. By Susan G. Aiken and S. J. Darbyshire. Agriculture Can- ada Monograph No. 29. Supply and services Canada, Ottawa. 173 pp. $9.00 in Canada; $10.80 elsewhere. Man’s impact on vegetation. 1983. Edited by W. Holzner, M. J. A. Werger, and J. Ikusima. Junk, The Hague. xiv + 370 p. Dfl. 225. {Mate choice in plants. 1983. By Mary F. Wilson and Nancy Burley. Princeton University Press, Princeton. c244 pp., illus. Cloth U.S. $35; paper U.S. $12.50. Metals and micronutrients: uptake and utilization by plants. 1983. Edited by D. A. Tobb and Stan Pierpoint. Academic Press, New York. 358 pp. U.S. $49.50. Monitoring of air pollutants by plants: methods and prob- lems. 1982. Edited by L. Steubing and H. J. Jager. Proceed- ings at a workshop, Osnabrtick, Germany, 24 to 25 Sep- tember, 1981. Junk, The Hague. 172 pp. Dfl. 80. Plant extinction: a global crisis. 1983. By Harold Koopo- witz and Hilary Kaye. Stone Wall Press (distributed by Stackpole Books, Harrisburg, Pennsylvania). x + 240 pp., illus. U.S. $16.95. Principles of dispersal in higher plants. 1982. By L. van der Pijl. 3rd revised edition. Springer-Verlag, New York. 215 pp., illus. U.S. $27.50. Sampling methods and taxon analysis in vegetation science. 1983. Edited by R. Knapp. Handbook of Vegetation Science 4. Junk, The Hague. 364 pp. Dfl. 155. Selected climatic data for a global set of standard stations for vegetation science. 1982. By W. J. Muller. Junk, The Hague. 334 pp. Dfl. 200. Then and now: a photographic history of vegetation change in the Central Great Basin Desert. 1982. By G. F. Rogers. University of Utah Press, Salt Lake City. 128 pp., illus. U.S. $15. BOOK REVIEWS 471 Vascular plants poisonous to livestock in Canada. 1. A preliminary inventory. 1983. By Biosystematics Research Institute. Agriculture Canada, Ottawa. 33 pp. Vegetation dynamics in grasslands, healthlands, and Medi- terranean ligneous formations. 1982. Edited by P. Pois- sonet et al. Proceedings of a symposium, Montpellier, France, September 1980. Junk. The Hague. x + 286 pp. Dfl. 195. The vegetation of the subantarctic islands Marion and Prince Edward. 1982. By N. J. M. Gremmen. Junk, The Hague. 160 pp. Dfl. 110. Environment Acid deposition. 1983. Edited by S. Beilke and A. J. Elshout. Proceedings of a workshop, Berlin, 9 September, 1982. Reidel (distributed by Kluwer, Boston). c 250 pp. U.S. $32.50. Canada’s special places in the north: an environment Can- ada perspective for the 80's. 1982. By Environment Canada, Ottawa. 22 pp. + map. Conceptual issues in ecology. 1982. Edited by Esa Saarinen. Reidel, Higham, Massachusetts. vil + 374 pp. U.S. $15.95. Costa Rican natural history. 1983. Edited by Daniel H. Janzen. University of Chicago Press, Chicago. 832 pp., illus. Cloth U.S. $50; paper U.S. $30. Cost-benefit analysis and environmental regulations: polit- ics, ethics, and methods. 1982. Edited by D. Swatzman, R. A. Liroff, and K. G. Groke. Conservation Foundation, Washington. 196 pp. U.S. $11.50. Ecology of tropical savannas. 1982. Edited by B. J. Huntley and B. H. Walker. Springer-Verlag, New York. 669 pp., illus. U.S. $49. Effects of accumulation of air pollutants in forest ecosys- tems. 1983. Edited by B. Ulrich and J. Pankrath. Proceed- ings of a workshop, Gottingen, West Germany, 16 to 18 May, 1982. Reidel (distributed by Kluwer, Boston). xvill + 390 pp. U.S. $58.50. Encyclopedia of American forest and conservation history. 1983. Edited by Richard C. Davis. Collier MacMillan, Don Mills. 780 pp., illus. 2 volumes. $195. Environmental assessment in Canada: directory of univer- sity teaching and research 1982-1983. 1983. Edited by Bruce Rigby. Federal Environmental Assessment Review Office, Ottawa. 96 pp. Free. Environmental effects of off-road vehicles. 1983. Edited by R. H. Webb and H.G. Wilshire. Springer-Verlag, New York. 534 pp., illus. U.S. $49.80. 472 THE CANADIAN FIELD-NATURALIST Environmental impact assessment. 1982. Edited by Univer- sity of Aberdeen. Proceedings of a NATO Advanced Study Institute, Toulouse, France, 30 August to 12 September, 1981. Martinus Nijhoff, The Hague. xii + 439 pp. Dfl. 135. {Environmental standards: a comparative study of Cana- dian standard setting processes, and enforcement. 1982. By M. A. H. Franson, R. T. Fransonand A. R. Lucas. Envir- onmental Council of Alberta, Edmonton. 206 pp. Free. Environment Canada and the north: discussion paper. 1983. The Environment Canada, Ottawa. 74 pp. Free. Joint U.S.A.-U.S.S.R. ecosystem investigation of the Ber- ing Sea, July-August, 1977. 1982. By U.S. Fish and Wildlife Service, Washington. xx + 272 pp., illus. Landscape ecology: theory and applications. 1983. By Z. Naveh and A. S. Lieberman. Springer-Verlag, New York. c352 pp., illus. Letters from Alabama (U.S.) chiefly relating to natural history. 1983. By Philip Henry Gosse. Limited re-issue of 1859 edition. Overbrook House, Mountain Brook, Ala- bama. 352 pp., illus. U.S. $12.95. +Marine and coastal systems of the Quoddy Region, New Brunswick. 1983. Edited by Martin L. H. Thomas. Cana- dian Special Publication of Fisheries and Aquatic Sciences 64. Fisheries and Oceans Canada, Ottawa. x + 306 pp., illus. $17.95 in Canada; $21.55 elsewhere. The physical environment of the Faeroe Islands. 1982. Edited by G. K. Rutherford. Junk, The Hague. 160 pp. Dfl. 95. Planning for change: a course of study in ecological plan- ning. 1982. By James A. Lande. Teachers College Press, New York. xiv + 181 pp., illus. cloth U.S. $13.95; paper WES3 $5595: Species at risk: research in Australia. 1982. Edited by R. H. Groves and W. D. L. Ride. Springer-Verlag, New York. 216 pp., illus. U.S. $38. Tasek Bera: the ecology of a freshwater swamp. 1982. Edited by J. 1. Furtado and S. Mori. Junk, The Hague. vill + 413 pp. Dfl. 185. To the rescue: seven heroes of conservation. 1982. By Rita Vandivert. Frederik Warne, New York. 119 pp., illus. U.S. $9.95. Turbidity in the aquatic environment: an environmental factor in fresh and oceanic waters. 1983. By Charles G. Wilber. Thomas, Springfield, Illinois. x + 134 pp. U.S. $18.75. Water resources and land-use planning: a systems approach. 1982. Edited by P. Laconte and Y. Y. Haimes. Proceedings of a NATO Advanced Study Institute, Vol. 97 Louvain-la-Neuve, Belgium, 3 to 14 July, 1978. Martinus Nyhoff, The Hague, vi + 470 pp. Dfl 135. Miscellaneous Adaptability: the significance of variability from molecule to ecosystem. 1983. By Michael Conrad. Plenum, New York. c370 pp. U.S. $45. Building for tomorrow: putting waste to work. 1982. By Martin Pawley. Sierra, San Francisco. 1x + 192 pp., illus. U.S. $17.95. The genesis and classification of cold soils. 1983. By Samuel Rieger. Academic Press, New York. 240 pp. U.S. $32. The great evolution mystery. 1983. By Gordon Rattray Taylor. Harper and Row, New York. viii + 280 pp., illus. WES S595: Interaction and coevolution. 1982. By John N. Thompson. Wiley-Interscience, New York. x + 180 pp. U.S. $27.50. Modelling fluctuating populations. 1982. By R. M. Nisbet and W. S. C. Gurney. Wiley, Somerset, New Jersey. 416 pp. U.S. $53.50. Mountains of North America. 1982. By Fred Beckey. Sierra, San Francisco. 255 pp., illus. U.S. $35. Remote sensing applications in marine science and tech- nology. 1983. Edited by Arthur Philip Cracknell. Proceed- ings of a NATO Advanced Study Institute, Dundee, Scot- land, | to 21 August, 1982. Reidel (distributed by Kluwer, Boston). xii + 466 pp. U.S. $78. Thread of life: the Smithsonian looks at evolution. 1983. By Roger Lewin. Smithsonian Books (distributed by Norton, New York). 256 pp., illus. U.S. $27.50. Books for Young Naturalists Beetles. 1982. By Sylvia A. Johnson. Lerner, Minneapolis. 48 pp., illus. U.S. $8.95. The bird book. 1982. By Laura Storms. Lerner, New York. 30 pp. U.S. $4.95. Bluebird rescue. 1982. By Joan Rattner Heilman. Lothrop, Lee, and Shephard, New York. 48 pp., illus. U.S. $9. Book of beasts. 1982. By John May and Michael Marten. Viking, New York. 192 pp., illus. U.S. $12.95. Crabs. 1982. By Sylvia A. Johnson. Lerner, Minneapolis. 48 pp., illus. U.S. $8.95. A first look at spiders. 1983. By Millicent E. Selsam and Joyce Hunt. Walker, New York. 32 pp., illus. U.S. $7.95. The life of a hamster. 1982. By Jan Feder. Children Press, Chicago. 32 pp., illus. U.S. $6.95. 1983 Life on earth: biology today. 1983. By Donald M. Silver. Random House, New York. illus. cloth U.S. $7.99; paper U.S. $6.95. The living ocean. 1982. By Rupurt Ormond. Childrens Press, Chicago. 64 pp., illus. U.S. $11.95. Mysteries of migration. 1983. By Robert M. McClung. Garrard, Champaign, Illinois. 64 pp., illus. U.S. $7.22. Mystery creatures of the jungle. 1982. By Gardner Soule. Watts, New York. 132 pp., illus. U.S. $9.90. Nanuck the polar bear. 1982. By Georgeanne Irvine. Child- rens Press, Chicago. 20 pp., illus. U.S. $5.95. Nature gone wild. 1982. By Walter Olesky. Messner, New York. 63 pp., illus. U.S. $9.29. Pitcher plants: the elegant insect traps. 1983. By Carol Lerner. Morrow, New York. 63 pp., illus. U.S. $10.50. BOOK REVIEWS 473 The secret world of underground creatures. 1982. By Dorothy Leon. Messner, New York. 96 pp., illus. U.S. $8.29. Wonders of the desert. 1982. By Louis Sabin. Troll, Mah- wah, New Jersey. 32 pp., illus. Cloth U.S. $7.89; paper U.S. $1.95. Wonders of the forest. 1982. By Francene Sabin. Troll, Mahwah, New Jersey. 32 pp., illus. Cloth U.S. $7.89; paper U.S. $1.95. Wonders of the pond. 1982. By Francene Sabin. Troll, Mahwah, New Jersey. 32 pp., illus. Cloth U.S. $7.89; paper U.S. $1.95. *assigned for review tavailable for review Index to Volume 97 Compiled by W. Harvey Beck Abies balsamea, \7 Accipiter gentilis, 188 Acer glabrum, 445 spicatum, 24 Achillea millefolium, 24 Actaea rubra, 24 Actia interrupta, 7 Agaricus campestris, 10 xanthodermus, |1 Agropyron spicatum, 445 subsecundum, 172 Agrostis perennans, 24 Ainley, M. G., reviews by, 246, 356, 466 Alaska, 66, 110, 333, 428 Alaska, Characteristics of Gray Wolf, Canis lupus, den and rendezvous sites in southcentral, 299 Alaska, Home range size, movements, and habitat use intwo Moose, Alces alces, populations in southeastern, 79 Alaska, Movements of collared Caribou, Rangifer tarandus, in relation to petroleum development on the Arctic slope of, 143 Alaska, Sphagnum at Prudhoe Bay, 115 Alberta, 111, 216, 323, 452 Alberta, Denning and foraging habits of Red Foxes, Vulpes vulpes, and their interaction with Coyotes, Canis latrans, in central, 1972-1981, 303 Alberta, Nodding Thistle, Carduus nutans: an addition to the vascular flora of, 328 Alberta, The tropical fish fauna in Cave and Basin hotsp- rings drainage, Banff National Park, 255 Alces alces, Moose, Effects of, on aquatic vegetation in Sibley Provincial Park, Ontario, 57 Alces alces, Moose, populations in southeastern Alaska, Home range size, movements and habitat use in two, 79 Alectoria, Bryoria, Usnea et Ramalina sur la Céte Nord et la Cote Sud du Golfe Saint-Laurent, Distribution sur les coniféres des lichens appartenant aux genres, 26 Alectoria sarmentosa sarmentosa, 28 Alewives, Alosa pseudoharengus, in Lake Ainslie, Nova Scotia, Food habits of young anadromous, 423 Alnus crispa, \7 incana ssp. tenuifolia, 173 rugosa, \7 sinuata, 445 Alopecurus alpinus, 282 Alopex lagopus, Arctic Foxes, in a petroleum development area, Winter movements of, 66 Alosa pseudoharengus, Alewives, in Lake Ainslie, Nova Scotia, Food habits of young anadromous, 423 Amanita flavorubescens, 10 muscaria, |1 Ambloplites rupestris, 47, 218 Ambystoma laterale, 33 Amelanchier alnifolia, 444 sanguinea, 24 American Association for the Advancement of Science: Arc- tic Division, 119 Amerorchis rotundifolia, 173 Anaphalis margaritacea, 24, 445 Anas acuta, 188, 453 platyrhynchos, 453 strepera, 453 Anas acuta, Pintail, brood and attempted predation by hawks, Extensive overland movement of, 216 Anas clypeata, Northern Shoveler, along the northern coast of Ontario, Breeding records of, 113 Anas rubripes, Black Ducks, during the breeding season, Survival of female, 62 Ancistrocerus antilope, | a. antilope, 2 catskill, | c. catskill, 3 Andromeda glaucophylla, 197 Anoplopoma fimbria, Sablefish, caught off Quatsino Sound, British Columbia, A golden-yellow colored, 106 Antennaria ekmaniana, 287 rosea, 175 Aquila chrysaetos, 189 Arabis arenicola, 285 Aralia nudicaulis, 17 Arctagrostis latifolia, 282 Arctostaphylos uva-ursi, 24, 313 Ardea herodias, 188 Ardea herodias, Great Blue Heron, reproduction at Boot Island and other Nova Scotia colonies, Comparison of, 275 Arenaria humifusa, 284 longipedunculata, 174 Arenaria interpres, 375 Armeria maritima ssp. arctica, 170 Armstrong, E. and D. Euler. Habitat usage of two woodland Buteo species in central Ontario, 200 Armstrong, E. R. and D. L. G. Noakes. Wintering biology of Mourning Doves, Zenaida macroura, in Ontario, 434 Arnica alpina spp. angustifolia, 287 latifolia, 445 Artemisia biennis, 170 furcata, 175 Artemisia rupestris ssp. woodii, sage, in south-western Yukon, Range extension of the, 449 Asio otus, 190 Aster macrophyllus, 24 Aulocomnium palustre, 196 Auplopus caerulescens, | c. subcorticalis, 7 Austin-Smith, P. J. and G. Rhodenizer. Ospreys, Pandion haliaetus, relocate nests from power poles to substi- tute sites, 315 Aythya affinis, 188 collaris, 188 474 1983 Badger, Taxidea taxus, in southwestern Ontario, Distribu- tion of the, 107 Ballard, W. B. and J. R. Dau. Characteristics of Gray Wolf, Canis lupus, den and rendezvous sites in southcentral Alaska, 299 Ballard, W. B., R. Farnell, and R. O. Stephenson. Long distance movement by Gray Wolves, Canis lupus, 333 Bass, Largemouth, Micropterus salmoides, Fish predation and other distinctive features in the diet of Nogies Creek, Ontario, 47 Bass, Rock, 47, 218 Smallmouth, 48, 218 Bat, Little Brown, 452 Bat, Myotis, injured by an American Kestrel, Falco sparve- rius, Ring-billed Gull, Larus delawarensis, predation on, 452 Bateman, M. C., 391 Bateman, M. C. A second record of the Deer Mouse, Pero- myscus maniculatus, from Newfoundland, 117 Bats, Little Brown, Myotis lucifugus, on Prince Edward Island, Distribution and ectoparasites of, 320 Beacham, T. D. Variability in size and age at sexual maturity of Witch Flounder, Glyptocephalus cynoglossus, in the Canadian Maritimes region of the Northwest Atlantic Ocean, 409 Bear, Black 313 Bears, Polar, Ursus maritimus, Small game hunting behav- iour of, 93 Beaubien, E. G., 452 Beaver, 312 Beckmannia erucaeformis ssp. baicalensis, 172 Bees in the Eastern Townships region, Quebec, Nesting biology of solitary wasps and, | Beese, W. J. White Malaxis, Malaxis monophyllos var. diphyllos, an addition to the orchids of Canada from the Queen Charlotte Islands, British Columbia, 215 Behaviour, Small game, hunting, of Polar Bears, Ursus mari- timus, 93 Berberis aquifolium, 445 Betula papyrifera, \7 Biogeography, Island, of seed plants in Lake Nipigon, Ontario, 16 Bird, D. M., review by, 239 Bombycilla cedrorum, 191 Bondrup-Nielsen, S. Ambivalence of the concealing pose of owls, 329 Bosmina, sp., 423 Boychuk, R. W., 453 Braham, H.W. Northern records of Risso’s Dolphin, Grampus griseus, in the Northeast Pacific, 89 Brant, 373 Branta bernicla, 373 Braya purpurascens, 285 Breeding, Avian, and occurrence notes from the Sutton Ridges area of northeastern Ontario, 187 Breeding birds at Polar Bear Pass, Bathurst Island, North- west Territories, Densities of, 371 Breeding in Labrador, Caspian Terns, Sterna caspia, 448 Breeding record, Presumed, of Brewsters x Brewster’s Warbler ( Vermivora chrysoptera x pinus) in Ontario, 458 INDEX TO VOLUME 97 475 Breeding records of Northern Shoveler, Anas clypeata, along the northern coast of Ontario, 113 Breeding season, Survival of female Black Ducks, Anas rubripes, during the, 62 Bridgland, J. and J. M. Gillett. Vascular plants of the Hayes Sound region, Ellesmere Island, Northwest Terri- tories, 279 British Columbia, 89, 428 British Columbia, A golden-yellow colored Sablefish, Ano- plopoma fimbria, caught off Quatsino Sound, 106 British Columbia and southwest Yukon, Variation in Red Fox, Vulpes vulpes, summer diets in northwest, 311 British Columbia, Nesting of the Thick-billed Murre, Uria lomvia, in, 450 British Columbia, White Malaxis, Malaxis monophyllos var. diphyllos, an addition to the orchids of Canada from the Queen Charlotte Islands, 215 Britton, D. M., review by, 361 Brodo, I. M., review by, 136 Bromus ciliatus, 24 vulgaris, 445 Brown, G. S., 423 Brown, P. W., 454 Browne, P. M., D. A. Duffus, and R. W. Boychuk. High nesting density of ducks on an island in Saskatche- wan, 453 Bryoria, Alectoria, Usnea et Ramalinia sur la C6te Nord et la Céte Sud du Golfe Saint-Laurent, Distribution sur les coniféres des lichens appartenant aux genres, 26 Bryoria capillaris, 27 furcellata, 27 fuscescens, 27 lanestris, 27 nadvornikiana, 27 salazinica, 27 simplicior, 27 tenuis, 27 trichodes americana, 27 t. trichodes, 27 Bucephala clangula, 188 Bufo americanus, 33 Bullhead, Brown, 48 Yellow, 48 Bunting, Snow, 373 Burtt, E. H., Jr.and W. Chow. ‘Facing in’ is not general to all gulls nesting on cliffs, 222 Buteo jamaicensis, Red-tailed Hawk, and scavenging by Rough-legged Hawk, Interspecific food piracy of Rough-legged Hawk, B. lagopus, on, 330 Buteo lagopus, Rough-legged Hawk, on Red-tailed Hawk, B. jamaicensis, and scavenging by Rough-legged Hawk, Interspecific food piracy of, 330 Buteo lineatus, 62, 200 platypterus, 200 swainsoni, 216 Buteo species in central Ontario, Habitat usage of two wood- land, 200 Calamagrostis canadensis, 24 purpurascens, 282 Calcarius lapponicus, 375 476 Calidris alba, 374 bairdii, 373 canutus, 373 fuscicollis, 373 maritima, 373 California, 428 Callitriche verna, 184 Calvatia fragilis, 10 Cameron, R. D., 143 Campanula aurita, 175 uniflora, 287 Canis lupus, Gray Wolf, den and rendezvous sites in south- central Alaska, Characteristics of, 299 Canis lupus, Gray Wolves, Long distance movement by, 333 Canis lupus, Wolf, Responses of two groups of Mountain Goats, Oreamnos americanus, to a, 110 Cannings, R. J., 450 Cantharellus cibarius, 11 Carbyn, L. N. and D. Patriquin. Observations on home range sizes, movements and social organization of Lynx, Lynx canadensis, in Riding Mountain National Park, Manitoba, 262 Cardamine bellidifolia, 285 pensylvanica, 24, 182 pratensis var. angustifolia, 285 Carduelis flammea, 191 hornemanni, 375 Carduus nutans, Nodding Thistle,: an addition to the vascu- lar flora of Alberta, 328 Carex amblyorhyncha, 170, 279 atrofusca, 283 bigelowii, 283 capillaris, 283 capitata, 173 disperma, 24, 173 franklinii, 173 geyeri, 445 glacialis, 283 lasiocarpa, 196 limosa, 196 maritima, 283 membranacea, 283 misandra, 283 nardina vat. atriceps, 283 pauciflora, 196 paupercula, 196 rariflora, 173 rostrata, 173, 196 rupestris, 284 scirpoidea, 279 sp., 385 stans, 282 ursina, 284 Caribou, Rangifer tarandus, in relation to petroleum devel- cpment on the Arctic Slope of Alaska, Movements of collared, 143 Carp, 48 Carp, Cyprinus carpio, in the Bay of Quinte watershed, Ontario, Seasonal feeding of, 293 Carpentier, A. G. Presumed breeding record of Brewster’s x Brewster’s Warbler (Vermivora chrysoptera x pinus) in Ontario, 458 THE CANADIAN FIELD-NATURALIST Vol. 97 Carpodacus purpureus, 191 Cassiope tetragona, 287 Castor canadensis, 312 Cartharus guttatus, 191 ustulatus, 191 Catling, P. M., 102 Ceanothus velutinus, 445 Cerastium arcticum, 283 regelii, 284 Ceratophyllum demersum, 170, 294 Certhia americana, 190 Cervus elephus, Elk, foraging related to forest management and succession in Idaho, 443 Chaemaedaphne calyculata, 197 Charadrius vociferus, 189 Charr, Brook, 99 Chen caerulescens, 373 Chickadee, Boreal, 190 Chlidonias niger, 189 Chordeiles minor, 190 Chorus Frog, Striped, 33 Chow, W., 222 Chrysis coerulans, 5 Chrysomela sp., 7 Chubb, Creek, 218 Cicuta mackenzieana, \75 Circus cyaneus, 216 Clangula hyemalis, 373 Clavaria cinerea, 10 vermicularis, 10 Claytonia sarmentosa, 174 Clearweed, Dwarf, Pilea pumila (Urticaceae) — new to Nova Scotia, 118 Clepsis clemensiana, 5 Clethrionomys rutilus, 73 Clintonia borealis, 24 Clitocybe dealbata, 10 subconnexa, 10 Coad, B. W., 334 Cochlearia officinalis, 285 Cody, W. J., reviews by, 137, 241 Colgan, P., review by, 246 Colpodium vahlianum, 283 Congress, The XIX International Ornithological, 461 Contopus borealis, 190 Cook, F. R. Editor’s report for 1982, 229 Coptis occidentalis, 444 trifolia, 24 Corallorhiza trifida, 24 Cornus canadensis, |7 stolonifera, \7 Cortinarius sp., 11 Corvus brachyrhynchos, 331, 453 Corydalis pauciflora, 174 Coyotes, Canis latrans, in central Alberta, 1972-1981, Denning and foraging habits of Red Foxes, Vulpes vulpes, and their interaction with, 303 Crane, Sandhill, 189 Creeper, Brown, 190 Crossbill, White-winged, 191 Crow, Common, 331, 453 1983 Cuerrier, J.-P., Adaptation of Atlantic Salmon, Salmo salar L., to a restricted freshwater environment, 439 Culaea inconstans, 255 Cwynar, L. C. Range extensions of vascular plants from the northern Yukon Territory, 170 Cyanocitta cristata, 321 Cyprinus carpio, 48 Cyprinus carpio, Carp, in the Bay of Quinte watershed, Ontario, Seasonal feeding of, 293 Cypripedium passerinum Rich. (Sparrow’s Egg Lady- slipper, Orchidaceae) on the north shore of Lake Superior, An ecological study of, 268 Cystopteris fragilis, 282 Daborn, G. R., 423 Dace, Longnose, 255 Daphnia retrocurva, 423 Darter, Blackside, 177, 218 Fantail, 177, 218 Greenside, 177, 218 Johnny, 177, 218 Rainbow, 177, 218 Darters (Pisces: Percidae) in the Thames River watershed of southwestern Ontario, Predation by fish and Com- mon Mergansers on, 218 Darters, stream, (Pisces: Percidae) in the Thames River watershed of southwestern Ontario, Habitat segrga- tion by, 177 Dau, J. R., 299 Day, R. T. A survey and census of the endangered Furbish Lousewort, Pedicularis furbishiae, in New Bruns- wick, 325 Deer, White-tailed, Odocoileus virginianus, fawns, Some effects of winter shelter conditions on, 391 Dekker, D. Denning and foraging habits of Red Foxes, Vulpes vulpes, and their interaction with Coyotes, Canis latrans, in central Alberta, 1972-1981, 303 Dendroica magnolia, 191 palmarum, 191 virens, 191 Deschampsia caepitosa, 24 flexuosa, 24 pumila, 283 Diaptomus minutus, 423 Dick, J. A., 187 Dicranum polysetum, 197 Diervilla lonicera, 24 Diet of Nogies Creek, Ontario, Largemouth Bass, Micro- pterus salmoides, Fish predation and other distinctive features in the, 47 Diets in northwest British Columbia and southwest Yukon, Variation in Red Fox, Vulpes vulpes, summer, 311 Dilabio, B., review by, 236 Dipogon sayi, | S. Sayi, 7 Dispersal of White Spruce, Picea glauca, seed, Observations of primary, 104 Disporum oreganum, 444 Distribution and ectoparasites of Little Brown Bats, Myotis lucifugus, on Prince Edward Island, 320 INDEX TO VOLUME 97 477 Distribution, Coastwide, and ocean migration patterns of stream- and ocean-type Chinook Salmon, Oncorhyn- chus tshawytscha, 427 Distribution of the Badger, Taxidea taxus, in southwestern Ontario, 107 Distribution, winter, of Rosy Finches, Leucosticte arctoa, in Montana, Dynamics of the, 307 Doerr, J. G. Home range size, movements and habitat use in two Moose, Alces alces, populations in southeastern Alaska, 79 Dolman, T. M. Photoperiod as an environmental cue for hibernation in juvenile Richardson’s Ground Squirrels, Spermophilus richardsoni, 323 Dolphin, Risso’s, Grampus griseus, inthe Northeast Pacific, Northern records of, 89 Douglass, R. J., L. G. Fisher, and M. Mair. Habitat selec- tion and food habits of Marten, Martes americana, in the Northwest Territories, 71 Dove, Mourning, 189 Doves, Mourning, Zenaida macroura, in Ontario, Wintering biology of, 434 Dowitcher, Short-billed, 189 Draba alpina, 285 cinerea, 285 corymbosa, 286 fladnizensis, 286 glabella, 286 lactea, 286 nivalis, 286 oblongata, 286 subcapitata, 286 Drosera intermedia, 196 rotundifolia, 197 Dryas integrifolia, 287 Dryopteris fragrans, 282 Duck, Ring-necked, 188 Ducks, Black, Anas rubripes, during the breeding season, Survival of female, 62 Duffus, D. A., 453 Dugle, J. R., 120 Dulichium arundinaceum, |82 Duncan, D. C. Extensive overland movement of Pintail, Anas acuta, brood and attempted predation by hawks, 216 Dupontia fisheri, 283 Dupuis, P., 332 Eagle, Bald, 188 Golden, 189 Eberhardt, L. E., R. A. Garrott, and W. C. Hanson. Winter movements of Arctic Fox, A/opex lagopus, in a petro- leum development area, 66 Ecological study of Cypripedium passerinum Rich. (Sparrow’s Egg Lady-slipper, Orchidaceae) on the north shore of Lake Superior, An, 268 Ecology, breeding, of the Horned Grebe, Podiceps auritus, in southwestern Manitoba, 401 Ecology, breeding, Fungus fairy rings in soil: etiology and, 9 Ecology of the Muskrat, Ondatra zibethicus, at Luther Marsh, Ontario, The, 377 478 Ectoparasites of Little Brown Bats, Myotis lucifungus, on Prince Edward Island, Distribution and, 320 Edge, T. A. and B. W. Coad. Reduction of the pelvic skeleton in the Threespine Stickleback, Gasterosteus aculeatus, in the lakes of Québec, 334 Editor’s report for 1982, 229 Edwards, Y., review by, 362 Eedy, W. Summary of book review activities: volume 96, 1982, 230 Eider, King, 373 Elaeagnus commutata, 313 Eleocharis acicularis, 185 compressa, 24 palustris, 182 sp., 385 uniglumis, 173 Elk, Cervus elaphus, foraging related to forest management and succession in Idaho, 443 Elodea canadensis, 182, 294 Empetrum nigrum, 287, 313 Empidonax alnorum, 190 Englert, J. and B. H. Seghers. Habitat segregation by stream darters (Pisces: Percidae) in the Thames River watershed of southwestern Ontario, 177 Englert, J. and B. H. Seghers. Predation by fish and Com- mon Mergansers on darters (Pisces: Percidae) in the Thames River watershed of southwestern Ontario, 218 Environment, freshwater, Adaptation of Atlantic Salmon, Salmo salar L., to a restricted, 439 Environmental cue for hibernation in juvenile Richardson’s Ground Squirrels, Spermophilus richardsoni, Photo- period as an, 323 Equisetum arvense, 282 fluviatile, 182 variegatum, 282 Epilobium angustifolium, 24, 445 arcticum, 287 latifolium, 287 Equisetum, sp., 385 Ereaux, B., 449 Erigeron compositus, 287 elatus, 175 Eriocaulon septangulare, 183 Eriophorum russeolum ssp. rufescens, \73 tenellum, 197 triste, 284 Erysimum pallasii, 286 Esox lucius, 218 masquinongy, 48 Etheostoma blennioides, 177, 218 caeruleum, 177, 218 flabellare, 177, 218 nigrum, 177, 218 Euler, D., 200 Eurycea bislineata, 33 Eurycea bislineata, Ywo-lined Salamander, in Ontario, Northern range extension of the, 116 Eutrema edwardsii, 174, 286 Exeristes comstockii, 3 THE CANADIAN FIELD-NATURALIST Vol. 97 Falco rusticolis obsoletus, 95 sparverius, 189 Falco sparverius, American Kestrel, Ring-billed Gull, Larus delawarensis, predation on bat, Myotis, injured by an, 452 Farnell, R., 333 Feeding, Seasonal, of Carp, Cyprinus carpio, in the Bay of Quinte watershed, Ontario, 293 Ferguson, M. A. D., review by, 243 Ferguson, R.S. and S. G. Sealy. Breeding ecology of the Horned Grebe, Podiceps auritus, in southwestern Manitoba, 401 Festuca baffinensis, 283 brachyphylla, 283 Finch, Purple, 191 Finches, Rosy, Leucosticte arctoa, in Montana, Dynamics of the winter distribution of, 307 Fisher, L. G., 71 Flounder, Witch, G/yptocephalus cynoglossus, in the Cana- dian Maritimes region of the Northwest Atlantic Ocean, Variability in size and age at sexual maturity of, 409 Flycatcher, Alder, 190 Olive-sided, 190 Food habits of Marten, Martes americana, in the Northwest Territories, Habitat selection and, 71 Food habits of young anadromous Alewives, Alosa pseudo- harengus, in Lake Ainslie, Nova Scotia, 423 Food piracy of Rough-legged Hawk, Buteo lagopus, on. Red-tailed Hawk, B. jamaicensis, and scavenging by Rough-legged Hawk, Interspecific, 330 Foraging related to forest management and succession in Idaho, Elk, Cervus elephus, 443 Forster, R. A., 321 Fox, Red, 313 Fox, Red, Vulpes vulpes, summer diet in northwest British Columbia and southwest Yukon, Variation in, 311 Foxes, Arctic, Alopex lagopus, in petroleum development area, Winter movements of, 66 Foxes, Red, Vulpes vulpes, and their interaction with Coyotes, Canis latrans, in central Alberta, 1972-1981, Denning and foraging habits of, 303 Fragaria vesca, \7 virginiana, 24 Fraser, D. and H. Hristienko. Effects of Moose, Alces alces, on aquatic vegetation in Sibley Provincial Park, Ontario, 57 Fraser, D. and J. K. Morton. Aquatic plants in Lake Superior Provincial Park in relation to water chemistry, 181 French, T. W. and G. L. Kirkland, Jr. Taxonomy of the Gaspé Shrew, Sorex gaspensis, and the Rock Shrew, S. dispar, 75 Frog, Green, 33 Mink, 33 Northern Leopard, 33 Striped Chorus, 33 Wood, 33 Gadwall, 453 Galium triflorum, 24 1983 Gambusia affinis, 255 Garrott, R. A., 66 Gasterosteus aculeatus, Threespine Stickleback, in two lakes of Québec, Reduction of the pelvic skeleton in the, 334 Gaultheria hispidula, 24 Gavia immer, 187 stellata, 375 Geocaulon lividum, 173 Geothlypis trichas, \9| Geum macrophyllum ssp. perincisum, 174 Gilbert, F. F., 377 Gilbert, F. F. and M. C. Bateman. Some effects of winter shelter conditions on White-tailed Deer, Odocoileus virginianus, fawns, 391 Gillett, J. M., 279 Gillett, J. M., review by, 136 Glaser, P. H. A patterned fen on the north shore of Lake Superior, Minnesota, 194 Glyptocephalus cynoglossus, Witch Flounder, in the Cana- dian Maritimes region of the Northwest Atlantic Ocean, Variability in size and age at sexual maturity of, 409 Goat, Mountain, 313 Goats, Mountain, Oreamnos americanus, to a Wolf, Canis lupus, Responses of two groups of, 110 Godfrey, S. and D. F. J. Hilton. Nesting biology of solitary wasps and bees in the Eastern Townships region, Quebec, | Godwit, Marbled, 95 Goldeneye, Common, 188 Goodyera repens, 24 Goose, Canada, 453 Snow, 373 Gopher, Pocket, 305 Goshawk, 188 Gots, B. L., 161 Grampus griseus, Risso’s Dolphin, in the Northeast Pacific, Northern records of, 89 Gray, P. A., reviews by, 236, 244 Grebe, Horned, Podiceps auritus, insouthwestern Manitoba, Breeding ecology of the, 401 Gregory, P. T., review by, 237 Gregory, R. S., G. S. Brown, and G. R. Daborn. Food habits of young anadromous Alewives, Alosa pseudoharen- gus, in Lake Ainslie, Nova Scotia, 423 Grosbeak, Rose-breasted, 322 Ground Squirrel, Arctic, 312 Ground Squirrels, Richardson’s, Spermophilus richardsoni, Photoperiod as an environmental cue for hibernation in juvenile, 323 Grouse, Sharp-tailed, 189 Grus canadensis, 189 Gull, Bonaparte’s, 189 Glaucous, 373 Herring, 222 Thayer’s, 373 Gull, Ring-billed Larus delawarensis, predation on bat, Myotis, injured by an American Kestrel, Falco spar- verius, 452 Gulls, Ivory, Pagophila eburnea, nesting on the Brodeur Peninsula, Baffin Island, N.W.T., 332 INDEX TO VOLUME 97 479 Gulls nesting on cliff, ‘Facing in’ is not general to all, 222 Gulo gulo, Wolverine, in Lake St. John area, Quebec, 457 Guthrie, J. E. and J. R. Dugle. Gamma-ray irradiation of a boreal forest ecosystem: the Field Irradiator-Gamma (FIG) facility and research programs, 120 Gyrfalcon, 95 Habitat segregation by stream darters (Pisces: Percidae) in the Thames River watershed of southwestern Ontario, 177 Habitat selection and food habits of Marten, Martes ameri- cana, in the Northwest Territories, 71 Habitat usage of two woodland Buteo species in central Ontario, 200 Habitat use intwo Moose, Alces alces, populations in south- eastern Alaska, Home range size, movements and, 79 Haliaeetus leucocephalus, 188 Halimolobos mollis, 286 Hallworth, B. and M. Mychajluk. Nodding Thistle, Carduus nutans: an addition to the vascular flora of Alberta, 328 Hamilton, J. G. and P. M. Powles. Fish predation and other distinctive features in the diet of Nogies Creek, Ontario, Largemouth Bass, Micropterus salmoides, 47 Hanson, W. C., 66 Hare, Snowshoe, 313 Hare, Snowshoe, Lepus americanus, population cycles, Spatial trends in Canadian, 151 Harms, V. L., review by, 241 Harms, V. L. The Swamp Saxifrage, Saxifraga pensyl- vanica, a rare plant in Canada, newly discovered in Saskatchewan, 91 Harrier, Northern, 216 Hawk, Broad-winged, 200 Red-shouldered, 62, 200 Swainson’s, 216 Hawk, Red-tailed, Buteo jamaicensis, and scavenging by Rough-legged Hawk, Interspecific food piracy of Rough-legged Hawk, B. lagopus, on, 330 Hawk, Rough-legged, Bureo lagopus, on Red-tailed Hawk, B. jamaicensis, and scavenging by Rough-legged Hawk, Interspecific food piracy of, 330 Hawk-owl, Northern, 189 Healey, M. C. Coastwise distribution and ocean migration patterns of stream- and ocean-type Chinook Salmon, Oncorhynchus tshawytscha, 427 Helvella connivens, 10 crispa, \1 Hemichromis bimaculatus, 255 Hendricks, P. and J. E. Swenson, Dynamics of the winter distribution of Rosy Finches, Leucosticte arctoa, in Montana, 307 Heron, Great Blue, 188 Heron, Great Blue, Ardea herodias, reproduction at Boot Island and other Nova Scotia colonies, Comparison of, 275 Hibernation in juvenile Richardson’s Ground Squirrels, Spermophilus richardsoni, Photoperiod as an environ- mental cue for, 320 Hierochloe alpina, 283 odorata, 24 480 Hill, N. and S. P. Vander Kloet. Dwarf Clearweed, Pilea pumila (Urticaceae) — new to Nova Scotia, 118 Hilton, D. F. J., | Hippuris vulgaris, 184 Hoefs, M., D. Russell, and B. Ereaux. Range extension of the sage, Artemisia rupestris, ssp. woodii in south- western Yukon, 449 Hogan, G. G. Interspecific food piracy of Rough-legged Hawk, Buteo lagopus, on Red-tailed Hawk, B. jamaicensis, and scavenging by Rough-legged Hawk, 330 Holroyd, G. L. and E. G. Beaubien. Ring-billed Gull, Larus delawarensis, predation on bat, Myotis, injured by an American Kestrel, Falco sparverius, 452 Houston, C.S. and P. W. Brown. Recoveries of Saskatchewan-banded White-winged Scoters, Mela- nitta fusca, 454 Houston, C. S. Birds first described from Hudson Bay, 95 Houston, C. S., reviews by, 130, 138, 359 Hristienko, H., 57 Hutchins, H. E., 208 Hydrocotyle americana, 24 Hygrocybe turunda, new to Nova Scotia, Two Sable Island fungi, Peziza ammophila and, 102 Hydrophorus niveus, |1 russula, || Hyla crucifer, 33 Hypericum ellipticum, 182 Ictalurus natalis, 48 Idaho, Elk, Cervus elaphus, foraging related to forest man- agement and succession in, 443 Ihssen, P. E., review by, 130 Inocybe lacera, 10 Interaction with Coyotes, Canis /atrans, in central Alberta, 1972-1981, Denning and foraging habits of Red Foxes, Vulpes vulpes, and their, 303 Tris versicolor, 182 Irwin, L. L. and J. M. Peek. Elk, Cervus elaphus, foraging related to forest management and succession in Idaho, 443 Isoetes echinospora, 183 Jaagumagi, R., review by, 238 Jaeger, Long-tailed, 375 Parasitic, 375 Pomarine, 375 James, R. D.,J. A. Dick, S. V. Nash, M. K. Peck, and B. E. Tomlinson. Avian breeding and occurrence notes from the Sutton Ridges area of northeastern Ontario, 187 Jay, Blue, 321 Jewelfish, 255 Jobin, L., 26 Jones, D. M. and J. B. Theberge. Variations in Red Fox, Vulpes vulpes, summer diets in northwest British Columbia and southwest Yukon, 311 Jones, G. S. and H. H. Thomas. Distribution and ectopara- sites of Little Brown Bats, Myotis lucifugus, on Prince Edward Island, 320 Juncus albescens, 284 alpinus s. lat., 173 biglumis, 284 THE CANADIAN FIELD-NATURALIST Vol. 97 Kalmia polifolia, 197 Kamstra, J. Northern range extension of the Two-lined Salamander, Eurycea bislineata, in Ontario, 116 Keddy, C. J., P. A. Keddy, and R. J. Planck. An ecological study of Cypripedium passerinum Rich. (Sparrow’s Egg Lady-slipper, Ochidaceae) on the northwest shore of Lake Superior, 268 Keddy, P. A., 268 Kennedy, A. J., review by, 365 Kennethiella trisetosa, 3 Kestrel, American, 189 Kestrel, American, Falco sparverius, Ring-billed Gull, Larus delawarensis, predation on bat, Myotis, injured by an, 452 Killdeer, 189 Kingsbird, Eastern, 190 Kinglet, Golden-crowned, 190 Ruby-crowned, 190 Kirkland, G. L., Jr., 75 Kittiwake, Black-legged, Rissa tridactyla, in the Prairie Prov- inces and north-central United States, Occurrences of the, 111] Knot, Red, 373 Kobresia myosuroides, 284 simpliciuscula, 284 Kuyt, E., 111 Lactarius resimus, |\ Lady-slipper, Sparrow’s Egg, (Orchidaceae), Cypripedium passerinum Rich. on the north shore of Lake Superior, An ecological study of, 268 Laflamme-Levesque, M.,J. M. Perronet L. Jobin. Distribu- tion sur les coniféres des lichens appartenant aux genres Bryoria, Alectoria, Usnea et Ramalina sur la Cote Nord et la Céte Sud du Golfe Saint-Laurent, 26 Lagopus lagopus, 93 mutus, 373 Lanius excubitor, 191 Lappula occidentalis, 175 Larix laricina, 197 Larus argentatus, 222 hyperboreus, 373 Philadelphia, 189 thayeri, 373 Larus delawarensis, Ring-billed Gull, predation on bat, Myotis, injured by an American Kestrel, Falco spar- verius, 452 Lazell, J. D., Jr. A River Otter, Lutra c. canadensis, of record size, 225 Ledum groenlandicum, 24, 197 Lepomis gibbosus, 47 Lepus americanus, 73, 313 Lepus americanus, Snowshoe Hare, population cycles, Spa- tial trends in Canadian, I5] Lesquerella arctica, 286 Leucopaxillus giganteus, || Leucosticte arctoa atrata, 303 arctoa tephrocotis, 307 Leucosticte arctoa, Rosy Finches, in Montana, Dynamics of the winter distribution of, 307 Limnodromus griseus, 189 Limosa fedoa, 95 1983 Linnaea borealis, \7 Lintack, W. M.and D. R. Voigt. Distribution of the Badger, Taxidea taxus, in southwestern Ontario, 107 Lobelia dortmanna, 183 Lock, A. R. Caspian Terns, Sterna caspia, breeding in Lab- rador, 448 Longcore, J. R., 62 Longspur, Lapland, 375 Linicera canadensis, 24 hirsuta, 24 utahensis, 444 Looman, J. Water Meal, Wolffia arrhiza (Lemnacea) in Saskatchewan, 220 Loon, Common, 187 Red-throated, 375 Lophodytes cucullatus, 188 Lousewort, Furbish, Pedicularis furbishiae, in New Bruns- wick, A survey and census of the endangered, 325 Lovejoy, D. A., reviews by, 129, 235 Lovig, D., 104 Loxia leucoptera, 191 Lutra c. canadensis, River Otter, of record size, A, 225 Luzula arcuata ssp. unalaschensis, 173 confusa, 284 nivalis, 284 rufescens, 173 Lycoperdon curtisii, 10 perlatum, 10 Lycopodium annotinum ssp. annotinum, 172 selago, 282 Lynx canadensis, Lynx, in Riding Mountain National Park, Manitoba, Observations on home range sizes, move- ments and social organization of, 262 Lynx, Lynx canadensis, in Riding Mountain National Park, Manitoba, Observations on home range size, move- ments and social organization of, 262 Lyophyllum immundum, \1 MacCrimmon, H. R., 293 MacCrimmon, H. R.,R. W. Pugsley and B. L. Gots. Naturali- zation of the Rainbow Smelt, Osmerus mordax, in Lake Simcoe, Ontario, 16] Macrae, D. A., 293 Macronyssus crosbyi, 320 Maianthemum canadense, 24 Maine, 62, 392 Mair, M., 71 Malaxis monophyllos var. diphyllos, White Malaxis, an addition to the orchids of Canada from the Queen Charlotte Islands, British Columbia, 215 Malaxis, White, Malaxis monophyllos var. diphyllos, an addition to the orchids of Canada from the Queen Charlotte Islands, British Columbia, 215 Mallard, 453 Manitoba, 94, 120 Manitoba, Breeding ecology of the Horned Grebe, Podiceps auritus, in southwestern, 401 Manitoba, Observations on home range size, movements and social organization of Lynx, Lynx canadensis, in Riding Mountain National Park, 262 Marasmius oreades, \0 INDEX TO VOLUME 97 481 Marten, Martes americana, in the Northwest Territories, Habitat selection and food habits of, 71 Martes americana, Marten, in the Northwest Territories, Habitat selection and food habits of, 71 Martin, Purple, 95 Massachusetts, 225 Maunder, J. E. Amphibians of the Province of Newfound- land, 33 Mayfield, H. F. Densities of breeding birds at Polar Bear Pass, Bathurst Island, Northwest Territories, 371 McAllister, D. E., reviews by, 133, 134 McCorquodale, D., reviews by, 463 McNair, D. B. and R. A. Forster. Heterospecific vocal mimicry by six oscines, 321 McNeill, J., review by, 242 MeNicholl, M. K., review by, 131 Megachile inermis, | mendica, | Melandrium Taylorae, 174 Melanitta fusca, 188 fusca deglandi, 454 perspicillata, 188 Melanitta fusca, White-winged Scoters, Recoveries of Sas- katchewan-banded, 454 Melanitta nigra, Black Scoter, population moulting in James and Hudson bays, An estimate of, 147 Melanoleuca humile, 10 Melittobia chalybii, 7 Mentha arvensis, 24 Menyanthes trifoliata, 196 Merganser, Common, 188 Hooded, 188 Red-breasted, 188 Mergansers, Common, on darters (Pisces: Percidae) in the Thames River watershed of southwestern Ontario, Predation by fish and, 218 Mergus merganser, 188, 218 serrator, 188 Mertensia ciliata, 445 paniculata, 24 Micropterus dolomieui, 48, 218 Micropterus salmoides, Largemouth Bass, Fish predation and other distinctive features in the diet of Nogies Creek, Ontario, 47 Microtus pennsylvanicus, 73, 305 xanthognathus, 73 Middleton, J., reviews by, 132, 357, 362 Migration patterns, ocean, of stream- and ocean-type Chi- nook Salmon, Oncorhynchus tshawytscha, Coast- wide distribution and, 427 Miller, G. D. and D. R. Wooldridge. Small game hunting behaviour of Polar Bears, Ursus maritimus, 93 Minnesota, A patterned fen on the north shore of Lake Superior, 194 Minuartia rubella, 284 Mitella nuda, \7 Molly, Sailfin, 255 Moneses uniflora, \7 Monodon monoceros, Narwhals, in Arctic Canada, Two recent ice entrapments of, 459 Monolepis Nuttalliana, 174 482 Monotropa uniflora, 24 Montana, Dynamics of the winter distribution of Rosy Finches, Leucosticte arctoa, in, 307 Montia sibirica, 444 Moose, 313 Moose, Alces alces, Effects of, on aquatic vegetation in Sibley Provincial Park, Ontario, 57 Moose, Alces alces, populations in southeastern Alaska, Home range size, movements and habitat use in two, 79 Morris, J. M. J., review by, 363 Morton, J. K., 181 Mosquitofish, 255 Mouse, Deer, Peromyscus maniculatus, from Newfound- land, A second record of the, 117 Movement, Long distance, by Gray Wolves, Canis lupus, 333 Movement of Pintail, Anas acuta, brood and attempted predation by hawks, Extensive overland, 216 Movements and habitat in two Moose, Alces alces, popula- tions in southeastern Alaska, Home range size, 79 Movements and social organization of Lynx, Lynx canaden- sis, in Riding Mountain National Park, Manitoba, Observations on home range sizes, 262 Movements of collared Caribou, Rangifer tarandus, in rela- tion to petroleum development on the Arctic Slope of Alaska, 143 Movements, Winter, of Arctic Foxes, Alopex lagopus, ina petroleum development area, 66 Murre, Thick-billed, Uria /omvia, in British Columbia, Nest- ing of the, 450 Muskellunge, 48 Muskrat, 313 Muskrat, Ondatra zibethicus, at Luther Marsh, Ontario, The ecology of the, 377 Mustela erminea, 73 Mychajluk, M., 328 Myodopsylla insignis, 320 Myotis, bat, injured by an American Kestrel, Falco spar- verius, Ring-billed Gull, Larus delawarensis, preda- tion on, 452 Myotis lucifugus, 452 Myotis lucifugus, Little Brown Bats, on Prince Edward Island, Distribution and ectoparasites of, 320 Myriophyllum alteriflorum, 185 exalbescens, 182, 294 farwellii, 182 spicatum, 175 tenellum, 182 Nagtegaal, D. A. A golden-yellow colored Sablefish, Ano- plopoma fimbria, caught off Quatsino Sound, British Columbia, 106 Najas flexilis, 184, 294 Narwhals, Monodon monoceros, in Arctic Canada, Two recent ice entrapments of, 459 Nash, S. V., 187 Nelson, J. S. The tropical fish fauna in Cave and Basin hotsprings drainage, Banff National Park, Alberta, 255 Nesting biology of solitary wasps and bees in the Eastern Townships region, Quebec, | THE CANADIAN FIELD-NATURALIST Vol. 97 Nesting density of ducks on an island in Saskatchewan, High, 453 Nesting of the Thick-billed Murre, Uria lomvia, in British Columbia, 450 Nesting on cliffs, ‘Facing in’ is not general to all gulls, 222 Nesting on the Brodeur Peninsula, Baffin Island, N.W.T., Ivory Gulls, Pagophila eburnea, 332 Nests from power poles to substitute sites, Ospreys, Pandion haliaetus, relocate, 315 New Brunswick, 75, 223 New Brunswick, A survey and census of the endangered Furbish Lousewort, Pedicularis furbishiae, in, 325 Newfoundland, A second record of the Deer MOuse, Pero- myscus maniculatus, from, 117 Newfoundland, Amphibians of the Province of, 33 Nighthawk, Common, 190 Noakes, D. L. G., 434 North, N. R., 113 Northwest Territories, Densities of breeding birds at Polar Bear Pass, Bathurst Island, 371 Northwest Territories, Habitat selection and food habits of Marten, Martes americana, in the, 71 Northwest Territories, lvory Gulls, Pagophila eburnea, nest- ing on the Brodeur Peninsula, Baffin Island, 332 Notemigonus crysoleucas, 47 Notropis cornutus, 48 Nova Scotia, 315 Nova Scotia colonies, Comparison of Great Blue Heron, Ardea herodias, reproduction at Boot Island and other, 275 Nova Scotia, Dwarf Clearweed, Pilea pumila (Urticaceae) —new to, 118 Nova Scotia, Food habits of young anadromous Alewives, Alosa pseudoharengus, in Lake Ainslie, 423 Nova Scotia, Two Sable Island fungi, Peziza ammophila and Hygrocybe turunda, new to, 102 Nucifraga columbiana, Clark’s Nutcracker, on conifer seeds during the postfledging period, Dependence of, 208 Nuphar variegatum, 57, 183 Nutcracker, Clark’s, Nucifraga columbiana, on conifer seeds during the postfledging period, Dependence of, 208 Nyctea scandiaca, 373 Nymphaea odorata, 183, 294 Odocoileus virginianus, White-tailed Deer, fawns, Some effects of winter shelter conditions on, 391 Oldsquaw, 373 Ondatra zibethicus, Muskrat, at Luther Marsh, Ontario, The ecology of the, 377 Ontario, 9, 147, 181, 268 Ontario, Albino Eastern Garter Snakes, Thamnophis sirtalis sirtalis, from, 456 Ontario, Avian breeding and occurrence notes from the Sut- ton Ridges area of northeastern, 187 Ontario, Breeding records of Northern Shoveler, Anas cly- peata, along the northern coast of, 113 Ontario, Distribution of the Badger, Taxidea taxus, in southwestern, 107 Ontario, Effects of Moose, Alces alces, on aquatic vegetation in Sibley Provincial Park, 57 1983 Ontario, Habitat segregation by stream darters (Pisces: Perci- dae) in the Thames River watershed of southwestern, 177 Ontario, Habitat usage of two woodland Buteo species in central, 200 Ontario, Island biogeography of seed plants in Lake Nipi- gon, 16 Ontario, Largemouth Bass, Micropterus salmoides, Fish predation and other distinctive features in the diet of Nogies Creek, 47 Ontario, Naturalization of the Rainbow Smelt, Osmerus mordax, in Lake Simcoe, 161 Ontario, Northern range extension of the Two-lined Sala- mander, Eurycea bislineata, in, 116 Ontario, Predation by fish and Common Mergansers on darters (Pisces: Percidae) in the Thames River watershed of southwestern, 218 Ontario, Presumed breeding record of Brewster’s x Brewster’s Warbler (Vermivora chrysoptera x pinus) in, 458 Ontario, Seasonal feeding of Carp, Cyprinus carpio, in the Bay of Quinte watershed, 293 Ontario streams, The occurrence and origin of Tiger Trout, Salmo trutta x Salvelinus fontinalis, in, 99 Ontario, The ecology of the Muskrat, Ondatra zibethicus, at Luther Marsh, 377 Ontario, Wintering biology of Mourning Doves, Zenaida macroura, in, 434 Oreamnos americanus, Mountain Goats, to a Wolf, Canis lupus, Responses of two groups of, 110 Oregon, 428 Oryzopsis asperifolia, 24 Osmerus mordax, Rainbow Smelt, in Lake Simcoe, Ontario, Naturalization of the, 161 Ospreys, Pandion haliaetus, relocate nests from power poles to substitute sites, 315 Ottawa Field-Naturalists’ Club, A Publication Policy for The, 231 Ottawa Field-Naturalists’ Club Annual Business Meeting, Notice of The, 227 Ottawa Field-Naturalists’ Club Auditor’s report, 340 Balance sheet, 340 Minutes of the 103rd annual business meeting, 337 Report by Council, 344 Statement of Centennial project revenue and expendi- ture, 343 Statement of income and expenditure, C. F.-N., 342 Statement of income and expenditure, O. F.-N. C., 341 Ottawa Field-Naturalists’ Club awards, Honorary member- ship and, 462 Ottawa Field-Naturalists’ Club awards, The, 228 Ottawa Field-Naturalists’ Club, Call for nominations for the Council of The, 228 Ottawa Field-Naturalists’ Club, Notice of motion to amend the Constitution of The, 227 Otter, River, Lutra c. canadensis, of record size, A., 225 Ouellet, H., review by, 358 Owl, Long-eared, 190 Northern Hawk-, 189 Snowy, 373 Owls, Ambivalence of the concealing pose of, 329 Oxyria digyna, 284 INDEX TO VOLUME 97 483 Pachistima myrsinites, 445 Pogophila eburnea, \vory Gulls, nesting on Brodeur Penin- sula, Baffin Island, N.W.T., 332 Pandion haliaetus, Ospreys, relocate nests from power poles to substitute sites, 315 Papaver radicatum, 285 Partridge, Gray, 305 Parus hudsonicus, 190 Patriquin, D., 262 Paxillus involutus, || Peck, M. K., 187 Pedicularis capitata, 287 hirsuta, 287 lanata, 287 sudetica, 287 Pedicularis furbishiae, Furbish Lousewort, in New Bruns- wick, A survey and census of the endangered, 325 Peek, J. M., 443 Peeper, Spring, 33 Pelecanus erythrorhynchos, 95 Pelican, White, 95 Penstemon spp., 444 Perca flavescens, 47 Perch, Yellow, 47 Percina maculata, 177, 218 Perdix perdix, 305 Peromyscus maniculatus, Deer Mouse, from Newfound- land, A second record of the, 117 Perron, J. M., 26 Petasites palmatus, 175 Peziza ammophila and Hygrocybe turunda, new to Nova Scotia, Two Sable Island fungi, 102 Phalaris arundinacea, 24 Phalarope, Red, 373 Phalaropus fulicaria, 373 Phenacomys intermedius, 73 Pheucticus ludovicianus, 322 Phippsia algida, 283 Physocarpus opulifolius, 24 Picea glauca, \7 mariana, \7, 196 Picea glauca, White Spruce, seed, Observations on primary dispersal of, 104 Picoides tridactylus, 190 Pilea pumila (Urticaceae), Dwarf Clearweed, — new to Nova Scotia, 118 Pintail, Anas acuta, brood and attempted predation by hawks, Extensive overland movement of, 216 Pintail, Northern, 188, 453 Pinus strobus, 24 Planck, R. J., 268 Plantago major var. Pilgeri, 175 Plectrophenax nivalis, 373 Plethodon cinereus, 33 Pleuropogon sabinei, 283 Pleurozium schreberi, 197 Plover, Black-bellied, 374 Pluvialis squatarola, 374 Poa abbreviata, 283 alpigena var. colpodea, 283 arctica, 283 glauca, 24, 283 484 harizii, 283 palustris, 24 Podiceps auritus, Horned Grebe, in southwestern Manitoba, Breeding ecology of the, 401 Poecilia latipinna, 255 Pogonia ophioglossoides, 197 Polelmonium pulcherrimum, 446 Polygonum amphibium, 185 amphibium ssp. laevimarginatum, 173 caurianum, 174 sp., 24, 385 viviparum, 284 Polytrichum juniperinum var. affine, 196 Population cycles, Spatial trends in Canadian Snowshoe Hare, Lepus americanus, 151 Population moulting in James and Hudson bays, An esti- mate of the Black Scoter, Melanitta nigra, 147 Populations in southeastern Alaska, Home range size, movements and habitat use in two Moose, Alces alces, 79 Populus balsamifera, \7 tremuloides, \7 Porcupine, 313 Potamogeton alpinus, 184 amplifolius, 184 crispus, 294 epihydrus, 183 filiformis, 57, 172 foliosus, 182 gramineus, 184 illinoiensis, 182 natans, 185 pectinatus, 182, 294 praelongus, 184 richardsonii, 182, 294 robbinsii, 184 spirillus, 185 Potentilla anserina, 25 hyparctica, 287 nivea ssp. chamissonis, 287 norvegica, 25 palustris, 25 pennsylvanica, 175 pulchella, 287 rubricaulis, 287 vahliana, 287 Powles, P. M., 47 Powles, P. M., H. R. MacCrimmon, and D. A. Macrae. Seasonal feeding of Carp, Cyprinus carpio, inthe Bay of Quinte watershed, Ontario, 293 Predation by fish and Common Mergansers on darters (Pisces: Percidae) in the Thames River watershed of south- western Ontario, 218. Predation by hawks, Extensive overland movement of Pin- tail, Anas acuta, brood and attempted, 216 Predation, Fish, and other distinctive features in the diet of Nogies Creek, Ontario, Largemouth Bass, Micro- pterus salmoides, 47 Predation on bat, Myotis, injured by an American Kestrel, Falco sparverius, Ring-billed Gull, Larus delawaren- sis, 452 THE CANADIAN FIELD-NATURALIST Vol. 97 Prescott, J. Wolverine, Gu/o gulo, in Lake St. John area, Quebec, 457 Primula mistassinica, 25 Prince Edward Island, 331 Prince Edward Island, Distribution and ectoparasites of Little Brown Bats, Myotis lucifugus, on, 320 Pringle, J. S., reviews by, 135, 360 Procyon lotor, 63 Progne subis, 95 Proulx, G. and F. F. Gilbert. The ecology of the Muskrat, Ondatra zibethicus, at Luther Marsh, Ontario, 377 Province of Quebec Society for the Protection of Birds, Inc., The Alfred B. Kelly Memorial Fund of the, 227 Prunus pensylvanica, 25 Pseudacris triseriata, 33 Ptarmigan, 313 Rock, 373 Willow, 93 Puccinellia angustata, 283 Nuttalliana, 172 phryganodes, 283 Pugsley, R. W., 161 Pumpkinseed, 47 Pygmephorus mahunkai, 320 Pyrola chlorantha, 175 grandiflora, 287 secunda, |7 virens, 25 Québec, 26 Quebec, Nesting biology of solitary wasps and bees in the Eastern Townships region, | Québec, Reduction of the pelvic skeleton in the Threespine Stickleback, Gasterosteus aculeatus, in two lakes of, 334 Quebec, Wolverine, Gulo gulo, in Lake St. John area, 457 Quinney, T. E. Comparison of Great Blue Heron, Ardea herodias, reproduction at Boot Island and other Nova Scotia colonies, 275 Raccoon, 63 Ramalina dilacerata, 30 farinacea, 30 roesleri, 30 thrausta, 30 Ramalina sur la Céte Nord et la Céte Sud du Golfe Saint- Laurent, Distribution sur les coniferes des lichens appartenant aux genres Bryoria, Alectoria, Usnea et, 26 Rana clamitans, 33 pipiens, 33 septentrionalis, 33 sylvatica, 33 Range extension of the sage Artemisia rupestris ssp. woodii in southwestern Yukon, 449 Range extension of the I wo-lined Salamander, Eurycea bis- lineata, in Ontario, Northern, 116 Range extensions of vascular plants from the northern Yukon Territory, 170 1983 Range size, Home, movements and habitat use in two Moose, Alces alces, populations in southestern Alaska, 79 Range sizes, Home, movements and social organization of Lynx, Lynx canadensis, in Riding Mountain National Park, Manitoba, 262 Rangifer tarandus, Caribou, in relation to petroleum devel- opment on the Arctic Slope of Alaska, Movements of collared, 143 Ranunculus confervoides, 174 hyperboreus, 285 nivalis, 285 pensylvanicus, 25 Pygmaeus, 285 reptans, 185 scleratus ssp. multifidus, 174 sulphureus, 285 Redhead, S. A. and P. M. Catling. Two Sable Island fungi, Peziza ammophila and Hygrocybe turunda, new to Nova Scotia, 102 Redpoll, Common, 191 Hoary, 375 Reed, A. and P. Dupuis. Ivory Gulls, Pagophila eburnea, nesting on Brodeur Peninsula, Baffin Island, N.W.T., 332 Reeves, R. R., review by, 357 Regulus calendula, 190 satrapa, 190 Reproduction at Boot Island and other Nova Scotia colonies, Comparison of Great Blue Heron, Ardea herodias, 275 Responses to two groups of Mountain Goats, Oreamnos americanus, to a Wolf, Canis lupus, 110 Reznicek, A. A., review by, 465 Rhamnus alnifolia, 25 Rhodenizer, G., 315 Rhinichthys cataractae, 255 Rhynchospora alba, 196 Ribes glandulosum, 17 hudsonianum, 25 lacustre, 25 oxyacanthoides, 25 spp., 444 triste, 25 Ringelman, J. K. and J. R. Longcore. Survival of female Black Ducks, Anas rubripes, during the breeding sea- son, 62 Rissa tridactyla, Black-legged Kittiwake, in the Prairie Prov- inces and north-central United States, Occurrences of the, 111 Rosa acicularis, 25 Ross, R. K. An estimate of the Black Scoter, Melanitta nigra, population moulting in James and Hudson bays, 147 Ross, R. K. and N. R. North. Breeding records of Northern Shoveler, Anas clypeata, along the northern coast of Ontario, 113 Rubus idaeus, \7 parviflorus, 444 pubescens, \7 Russell, D., 449 INDEX TO VOLUME 97 485 Russula aerugina, 10 aurantiolutea, || decolorans, 10 densifolia, 10 Sablefish, Anoiplopoma fimbria, caught off Quatsino Sound, British Columbia, A golden-yellow colored, 106 Sagina intermedia, 284 Sagittaria latifolia, 182 sp., 385 Salamander, Blue-spotted, 33 Eastern Redback, 33 Two-lined, 33 Salamander, Two-lined, Eurycea bislineata, in Ontario, Northern range extension of the, 116 Salix arctica, 284 humilis, 17 interior, 173 Phylicifolia, 17 scouleriana, 445 Salmo salar L., Atlantic Salmon, to a restricted freshwater environment, Adaptation of, 439 Salmo trutta, 99 Salmo trutta x Salvelinus fontinalis, Viger Trout, in Ontario streams, The occurrence and origin of, 99 Salmon, Atlantic, Sa/mo salar L., to a restricted freshwater environment, Adaptation of, 439 Salmon, Chinook, Oncorhynchus tschawytscha, Coastwide distribution and ocean migration patterns of stream- and ocean-type, 427 Salvelinus fontinalis, 99 Salvelinus fontinalis x Salmo trutta, Tiger Trout, in Ontario streams, The occurrence and origin of, 99 Sambucus racemosa, \7 Sanderling, 374 Sandpiper, Baird’s, 373 Purple, 373 White-rumped, 373 Sanguisorba officianalis, 175 Sarracenia purpurea, 197 Saskatchewan, High nesting density of ducks onan island in, 453 Saskatchewan, The Swamp Saxifrage, Saxifraga pensyl- vanica, a rare plant in Canada, newly discovered in, 91 Saskatchewan, Water Meal, Wolffia arrhiza (Lemnaceae) in, 220 Saskatchewan-banded White-winged Scoters, Melanitta fusca, Recoveries of, 454 Saxifraga caespitosa, 286 cernua, 286 flagellaris ssp. platysepala, 286 foliosa, 286 foliosa var. foliosa, 174 hirculus var. propinqua, 286 oppositifolia, 286 rivularis, 286 tenuis, 286 tricuspidata, 286 486 Saxifraga pensylvanica, The Swamp Saxifrage, a rare plant in Canada, newly discovered in Saskatchewan, 91 Saxifrage, Swamp, Saxifraga pensylvanica, a rare plant in Canada, newly discovered in Saskatchewan, The, 91 Scaup, Lesser, 188 Scheuchzeria palustris, 196 palustris var. americana, \97 Schizachne purpurascens, 25 Scholarships, Alfred B. Kelly Ornithological, 461 Scirpus sp., 385 subterminalis, 183 Scoter, Black, Melanitta nigra, population moulting in James and Hudson bays, 147 Scoter, Surf, 188 White-winged, 188 Scoters, White-winged, Melanitta fusca, Recoveries of Saskatchewan-banded, 454 Scotter, G. W., 350 Scotter, G. W. Publications of J. Dewey Soper, 353 Sealy, S. G., 401 Seghers, B. H., 177, 218 Semotilus atromaculatus, 218 Senecio spp., 444 Sergeant, D. E. and G. A. Williams. Two recent ice entrap- ments of Narwhals, Monodon monoceros, in Arctic Canada, 459 Sheep, Dall, 313 Shih, C. T., reviews by, 133, 134 Shiner, Blackchin, 52 Common, 48 Golden, 47 Shoveler, Northern, Anas clypeata, along the northern coast of Ontario, Breeding records of, 113 Shrew, Gaspé, Sorex gaspensis, and the Rock Shrew, S. dispar, Taxonomy of the, 75 Shrew, Rock. Sorex dispar, Taxonomy of the Gaspé Shrew, S. gaspensis, and the, 75 Sorex gaspensis, Gaspé Shrew, and the Rock Shrew, S. dispar, \axonomy of the, 75 Sorex spp., 73 Sparganium angustifolium, 183 chlorocarpum, 182 fluctuans, 185 sp., 385 Spatt, P. .D. Sphagnum at Prudhoe Bay, Alaska, 115 Spermophilus richardsoni, Richardson’s Ground Squirrels, Photoperiod as an environmental cue for hibernation in juvenile, 320 Sphagnum angustifolium, 196 fimbriatum, 115 fuscum, 196 girgensohnii, 115 magellanicum, 196 majus, 196 rubellum, 196 Sphagnum at Prudhoe Bay, Alaska, 115 Spinturnix americanus, 320 Spiraea betulifolia, 444 Spruce, White, Picea glauca, seed, Observations on primary dispersal of, 104 Squirrel, Red, 305, 313 THE CANADIAN FIELD-NATURALIST Vol. 97 Stahevitch, A. E., review by, 135 Stellaria humifusa, 285 longipes, 285 Stephenson, R. O., 333 Stercorarius longicaudus, 375 parasiticus, 375 pomarinus, 375 Sterling, K. B., reviews by, 247, 464, 467, 468, 469 Sterna caspia, Caspian Terns, breeding in Labrador, 448 Sterna paradisaea, 373 Stevens, W. E. and G. W. Scotter. Joseph Dewey Soper, 1893-1982, 350 Stickleback, Brook, 255 Stickleback, Threespine, Gasterosteus aculeatus, in two lakes of Québec, Reduction of the pelvic skeleton in the, 334 Streptopus roseus, 25 Suillus granulatus, || Surnia ulula, 189 Swenson, J. E., 307 Symmorphus albomarginatus, | cristatus, | c. cristatus, 5 Synaptomys borealis, 73 Tamiasciurus hudsonicus, 73, 305 Taraxacum phymatocarpum, 287 sp., 25 Taxidea taxus, Badger, in southwestern Ontario, Distribu- tion of the, 107 Taxus canadensis, 25 Tern, Arctic, 373 Black, 189 Terns, Caspian, Sterna caspia, breeding in Labrador, 448 Thalictrum sparsiflorum, 174 Thamnophis sirtalis sirtalis, Eastern Garter Snakes, from Ontario, Albino, 456 Theberge, J. B., 311 Thistle, Nodding, Carduus nutans: an addition to the vascu- lar flora of Alberta, 328 Thomas, H. H., 320 Thomomys talpoides, 305 Thrush, Hermit, 191 Swainson’s, 191 Thuja occidentalis, 17 Timoney, K. P. Island biogeography of seed plants in Lake Nipigon, Ontario, 16 Toad, American, 33 Tomlinson, B. E., 187 Toohey, J. 1. Fungus fairy rings in soil: etiology and chemi- cal ecology, 9 Tricholoma irinum, \1 terreum, |1 Trientalis borealis, 17 Triglochin palustris, 172 Tringa flavipes, 189 melanoleuca, 189 Trisetum sibiricum ssp. sibiricum, 172 spicatum, 283 Troglodytes troglodytes, 190 Trout, Brown, 99 1983 Trout, Tiger, Salmo trutta x Salvelinus fontinalis, in Ontario streams, The occurrence and origin of, 99 Turnstone, Ruddy, 375 Tympanuchus phasianellus, 189 Typha latifolia, 25, 182 sp., 385 Tyrannus tyrannus, 190 Uria lomvia, Thick-billed Murre, in British Columbia, Nest- ing of the, 450 Ursus maritimus, Polar Bears, Small game hunting be- haviour of, 93 Urtica dioica, 25 Usnea, Bryoria, Alectoria, et Ramalina sur la Cote Nord et la Cote Sud du Golfe Saint-Laurent, Distribution sur les coniféres des lichens appartenant aux genres, 26 Usnea filipendula, 29 loingissima, 29 sorediifera, 29 trichodea, 29 Utricularia cornuta, 196 intermedia, 185, 196 minor, 185 Purpurea, 182 vulgaris, 184 Vaccinium angustifolium, 25 myrtilloides, 25 oxycoccos, 197 spp., 313 uliginosum var. alpinum, 287 Valisneria americana, 294 Vallee, A. and R. J. Cannings. Nesting of the Thick-billed Murre, Uria lomvia, in British Columbia, 450 Vander Kloet, S. P., 118 Vander Wall, S. B. and H. E. Hutchins. Dependence of Clark’s Nutcracker, Nucifruga columbiana, on conifer seeds during the postfledging period, 208 Variation in Red Fox, Vulpes vulpes, summer diets in northwest British Columbia and southwest Yukon, 311 Vermivora chrysoptera, 458 pinus, 458 Vermivora chrysoptera x pinus, Brewster’s Warbler, in Ontario, Presumed breeding record of Brewster’s x, 458 Veronica Wormskjoldii ssp. alternifolia, 175 Viburnum edule, 25 Viola macloskeyi, 25 nephrophylla, 25 renifolia, 25 Vireo flavifrons, 321 griseus, 32) olivaceus, 321 solitarius, 191, 321 Vireo, Red-eyed, 332 Solitary, 191, 322 White-eyed, 321 Yellow-throated, 322 Voigt, D. R., 107 Vole, Meadow, 73, 305 Red-backed, 73 Yellow-cheeked, 73 INDEX TO VOLUME 97 487 Vulpes vulpes, Red Fox, summer diets in northwest British Columbia and southwest Yukon, Variation in, 311 Vulpes vulpes, Red Foxes, and their interaction with Coyotes, Canis latrans, in central Alberta, 1972-1981, Denning and foraging habits of, 303 Warbler, Black-throated Green, 191 Blue-winged, 458 Golden-winged, 458 Magnolia, 191 Palm, 191 Warbler, Brewster’s x Brewster's, (Vermivora chrysoptera x pinus) in Ontario, Presumed breeding record of, 458 Warner, B. G., review by, 138 Washington, 428 Wasps, solitary, and bees in the Eastern Townships region, Quebec, Nesting biology of, | Water Meal, Wolffia arrhiza (Lemnaceae) in Saskatchewan, 220 Waxwing, Cedar, 191 Weller, W. F. Albino Eastern Garter Snakes, Thamnophis sirtalis sirtalis, from Ontario, 456 Whitten, K. R. and R. D. Cameron. Moments of collared Caribou, Rangifer tarandus, in relation to petroleum development on the Arctic Slope of Alaska, 143 Williams, G. A., 459 Witzel, L. D. The occurrence and origin of Tiger Trout, Salmo trutta x Salvelinus fontinalis, in Ontario streams, 99 Wolf, Canis lupus, Responses of two groups of Mountain Goats, Oreamnos americanus, to a, 110 Wolf, Gray, Canis lupus, den and rendezvous sites in south- central Alaska, Characteristics of, 229 Wolffia arrhiza (Lemnaceae), Water Meal, in Saskatchewan, 220 Wolverine, Gulo gulo, in Lake St. John area, Quebec, 457 Wolves, Gray, Canis lupus, Long distance movement by, 333 Woodpecker, Three-toed, 190 Woods, R., review by, 245 Woodsia alpina, 282 glabella, 282 Wooldridge, D. R., 93 Wren, Winter, 190 Yellowlegs, Greater, 189 Lesser, 189 Yellowthroat, Common, 191 Yukon, Range extension of the sage Artemisia rupestris ssp. woodii in south-western, 449 Yukon Territory, Range extensions of vascular plants from the northern, 170 Yukon, Variation in Red Fox, Vulpes vulpes, summer diets in northwest British Columbia and southwest, 311 Zammuto, R. M., reviews by, 129, 130, 133 Zapus hudsonius, 73 Zasada, J. C., and D. Lovig. Observations on primary dis- persal of White Spruce, Picea glauca, seed, 104 Zenaida macroura, 189 Zenaida macroura, Mourning Doves, in Ontario, Wintering biology of, 434 Zizania aquatica, 294 488 THE CANADIAN FIELD-NATURALIST Vol. 97 Index to Book Reviews Botany Benson, L. The Cacti of the United States and Canada, 137 Bovin, B. Survey of Canadian herbaria, 240 Buckley, A. R. Trees and shrubs of the Dominion Arbore- tum, 360 Dobson, F. Lichens: an illustrated guide, 136 Douglas, G. W., G. W. Argus, H. L. Dickson, and D. F. Brunton. The rare vascular plants of the Yukon, 241 Given, D. R. and J. H. Soper. The Arctic-Alpine element of the vascular flora at Lake Superior, 465 Mohlenbrock, R. H. The illustrated flora of Illinois: flower- ing plants, Basswoods to Spurges, 136 Muenscher, W. C. Weeds, 242 Soper, J. H.and M. L. Heimburger. Shrubs of Ontario, 135 Thannheiser, D. Die Kustenvegetation Ostkanadas, 135 Tryon, R. M.and A. F. Tryon. Ferns and allied plants: with special reference to tropical America, 361 Environment Ager, D. V. The nature of the stratigraphical record, 138 Beechey, T. J. and B. L. Raad (comps.). Nature conserva- tion day, 243 Burnham, K. P., D. R. Anderson, and J. L. Laake. Estima- tion of density from line transect sampling of biologi- cal populations, 246 ~ Gehlbach, F. R. Mountain islands and desert seas: a natural history of the U.S. — Mexican borderlands, 466 Giesy, J. P., Jr. (ed.). Microcosms in ecological research, 362 Judd, W. W.A bibliography of the natural history of Middle- sex County, Ontario, to the year 1980 with an histori- cal introduction, 138 Marhsall, I. B. Mining, land use, and the environment, |: a Canadian overview, 363 Radford, A. E., D. Kay, S. Otte, L. Otte, J. R. Massey, P. .D. Whitson, and contributors. Natural heritage: classification inventory, and information, 245 Swanson, G. A. (co-ord.). The Mitigation Symposium: a national workshop on mitigating losses of fish and wildlife habitats, 244 Vontobel, R. (ed.). Man and wildlife in a shared environ- ment, 362 Zoology Brigham, A. R., W.U. Brigham, and A. Gnilka (eds.). Aquatic insects and oligochaetes of North and South Carolina, 238 Cade, T. J. Falcons of the world, 239 Chu, Y. T. and C. W. Meng. A study of the lateral-line canals system and that of Lorenzini ampullae and tubules of elasmobranchiate fishes of China, 134 Davis, D. E. (ed.). Handbook of census methods for terres- trial vertebrates, 133 Farber, P. L. The emergence of ornithology as a scientific discipline: 1760-1850, 356 Fay, F. H. Ecology and biology of the Pacific Walrus, Odobenus rosmarus divergens \\liger, 357 Ferguson, R. S. Summer birds of the Northwest Angle Pro- vincial Forest and adjacent southeastern Manitoba, Canada, 130 Gardner, K. A. Birds of Oak Hammock Marsh Wildlife Management Area, 359 Glasgow, W. M. Fisheries and wildlife resources and the agricultural land base in Alberta, 463 Goodwin, C. E. A bird-finding guide to Ontario, 236 Harrison, C. Anatlas of the birds of the western Palaearctic, 131 Hazard, E. B. The mammals of Minnesota, 235 Martinka, C. J. and K. L. McArthur (eds.). Bears: their biology and management, 129 Monson, G. and A. R. Phillips. Annotated checklist of the birds of Arizona, 359 Ogilvie, S. C. The Park Bison, 463 Powell, R. A. The Fisher: life history, ecology, and behavior, 236 Preston, W. B. The amphibians and reptiles of Manitoba, 237 Remmert, H. Arctic animal ecology, 132 Ryman, N. (ed.). Fish gene pools, 130 Smith, M. H. and J. Joule (eds.). Mammalian population genetics, 129 South China Sea Fisheries Institute, et al. The fishes of the islands in the South China Sea, 133 Van Gelder, R. G. Mammals of the National Parks, 235 von Blotzheim, G., Urs N., and K. M. Bauer. Handbuch der Vogel Mitteleuropas: volume 9, Columbiformes — Piciformes, 358 Wakeley, J. S. (ed.). Wildlife population ecology, 130 Williams, J.G. and N. Arlott. A field guide to the birds of east Africa, 357 Wisher, L. Eastern Chipmunks: secrets of their solitary lives, 464 Miscellaneous Astbury, E. E. (comp.). Casey A. Wood (1856-1942): a bio- bibliography, 466 DeLatte, C. Lucy Audubon: a biography, 469 Ewan, J. and N. D. Ewan. Biogeographical dictionary of Rocky Mountain naturalists: a guide to the writings and collections of botanists, zoologists, artists, and photographers, 1682-1932, 468 Grenne, R. H. Sampling design and statistical methods for environmental biologists, 364 Lawson, M. Dammed Indians: the Pick-Sloan Plan and the Missouri River Sioux, 1944-1980, 467 Lord, J. B. Aleksander Tamsalu, 1891-1960: a botanist in exile, 247 Peck, R. M. Acelebration of birds — the life and art of Louis Agassiz Fuertes, 246 The Ottawa Field-Naturalists’ Club Special Publications 1. Autobiography of John Macoun : A reprint of the 1922 edition of the fascinating life story of one of Canada’s outstanding early naturalists, with a new introduction by Richard Glover and bibliographical essay, footnotes, and index by William A. Waiser, plus three maps of John Macoun’s western travels. Individuals $12.50 plus $2 postage and handling Libraries $15.00 plus $2 postage and handling 2. Transactions of The Ottawa Field-Naturalists’ Club and The Ottawa Naturalist — Index. Compiled by John M. Gillett A complete author, title, and subject index to the predecessors of The Canadian Field- Naturalist, the first thirty-nine volumes of the publications of The Ottawa Field-Naturalists’ Club. $25 plus $2 postage and handling Centennial Bird Record Songs of the Seasons More than fifty eastern North American birds and amphibians are presented in full stereo- phonic sound as recorded in the wild by wildlife recording expert F. Montgomery Brigham. $9.11 (postage and handling included but Ontario residents must add 7% sales tax Please send orders to: The Ottawa Field-Naturalists’ Club Box 3264 Postal Station C Ottawa, Ontario, Canada K1Y 4J5 489 Advice to Contributors Content The Canadian Field- Naturalist is a medium for the publi- cation of scientific papers by amateur and professional natu- ralists or field-biologists reporting observations and results of investigations in any field of natural history provided that they are original, significant, and relevant to Canada. All readers and other potential contributors are invited to sub- mit for consideration their manuscripts meeting these crite- ria. For further information consult: A Publication Policy for the Ottawa Field-Naturalists’ Club, 1983. The Canadian Field- Naturalist 97(2): 231-234. Manuscripts Please submit, in either English or French, three complete manuscripts written in the journal style. The research reported should be original. It is recommended that authors ask qualified persons to appraise the paper before it is sub- mitted. Also authors are expected to have complied with all pertinent legislation regarding the study, disturbance, or collection of animals, plants or minerals. The place where voucher specimens have been deposited, and their catalogue numbers, should be given. Latitude and longitude should be included for all individual localities where collection or observations have been made. Type the manuscript on standard-size paper, if possible use paper with numbered lines, double-space throughout, leave generous margins to allow for copy marking, and number each page. For Articles and Notes provide a biblio- graphic strip, an abstract and a list of key words. Generally words should not be abbreviated but use SI symgols for units of measure. Underline only words meant to appear in italics. The names of authors of scientific names should be omitted except in taxonomic manuscripts or other papers involving nomenclatural problems. Authors are encouraged to use “proper” common names (with inital letters capitalized) as long as each species is identified by its scientific name once. The names of journals in the Literature Cited should be written out in full. Unpublished reports should not be cited here but placed in the text. Next list the captions for figures (numbered in arabic numerals and typed together on a separate page) and present the tables (each titled, numbered consecutively in arabic numerals, and placed on a separate page). Mark in the margin of the text the places for the figures and tables. Extensive tabular or other supplementary material not essential to the text, typed neatly and headed by the title of the paper and the author’s name and address, should be submitted in duplicate on letter-size paper for the Editor to place in the Depository of Unpublished Data, CISTI, National Research Council of Canada, Ottawa, Canada K1A 0S2. A notation in the published text should state that the material is available, at a nominal charge, from the Depository. 490/ / The Council of Biology Editors Style Manual, 4th edition (1978) available from the American Institute of Biological Sciences, is recommended as a guide to contributors. Webs- ter’s New International Dictionary and le Grand Larousse Encyclopédique are the authorities for spelling. Illustrations — Photographs should have a glossy finish and show sharp contrasts. Photographic reproduction of line drawings, no larger than a standard page, are preferable to large originals. Prepare line drawings with India ink on good quality paper and letter(don’t type) descriptive matter. Write author’s name, title of paper, and figure number on the lower left corner or on the back of each illustration. Reviewing Policy Manuscripts submitted to The Canadian Field- Naturalist are normally sent for evaluation to an Associate Editor (who reviews it himself or asks another qualified person to do so), and at least one other reviewer, who isa specialist in the field, chosen by the Editor. Authors are encouraged to suggest names of suitable referees. Reviewers are asked to give a general appraisal of the manuscript followed by specific comments and constructive recommendations. Almost all manuscripts accepted for publication have undergone revi- sion — sometimes extensive revision and reappraisal. The Editor makes the final decision on whether a manuscript is acceptable for publication, and in so doing aims to maintain the scientific quality and overall high standards of the journal. Special Charges Authors must share in the cost of publication by paying $60 for each page in excess of five journal pages, plus $6 for each illustration (any size up toa full page), and up to $60 per page for tables (depending on size). Reproduction of color photos is extremely expensive; price quotations may be obtained from the Business Manager. When galley proofs are sent to authors, the journal will solicit on a voluntary basis a commitment, especially if grant or institutional funds are available, to pay $60 per page for all published pages. Authors must also be charged for their changes in proofs. Limited journal funds are available to help offset publica- tion charges to authors with minimal financial resources. Requests for financial assistance should be made to the Editor when the manuscript is accepted. Reprints An order form for the purchase of reprints will accompany the galley proofs sent to the authors. Gi U46 p TABLE OF CONTENTS (concluded) Notes (concluded) Wolverine, Gu/o gulo, in Lake St. John area, Quebec JACQUES PRESCOTT Presumed breeding record of Brewster’s x Brewster's Warbler, Vermivora chrysoptera x pinus, in Ontario A. GEOFFREY CARPENTIER Two recent ice entrapments of Narwhals, Monodon monoceros, in Arctic Canada D. E. SERGEANT and G. A. WILLIAMS News and Comment Second announcement: The XIX International Ornithological Congress — Alfred B. Kelly Ornithological Scholarships — Raptor collisions with utility lines: a call for information — Honorary membership and 1983 Ottawa Field-Naturalists’ Club awards Book Reviews Zoology: Fisheries and wildlife resources and the agricultural land base in Alberta — The park buffalo — Eastern chipmunks: secrets of their solitary lives Botany: The Arctic-Alpine element of the vascular flora at Lake Superior Environment: Mountain islands and desert seas: a natural history of the U.S.-Mexican borderlands Miscellaneous: Casey A. Wood (1856-1942): a bio-bibliography — Damned indians: the Pick-Sloan Plan and the Missouri River Sioux, 1944-1980 — Biogeographical dictionary of Rocky Mountain naturalists: a guide to the writings and collections of botanists, zoologists, artists, and photographers, 1682-1932 — Lucy Audubon: a biography New Titles Index to Volume 97 Compiled by W. HARVEY BECK Advice to Contributors Mailing date of the previous issue 6 March 1984 457 458 459 461 463 465 466 466 470 474 490 THE CANADIAN FIELD-NATURALIST Volume 97, Number 4 1983 Articles Densities of breeding birds at Polar Bear Pass, Bathurst Island, Northwest Territories HAROLD F. MAYFIELD 371 The ecology of the Muskrat, Ondatra zibethicus, at Luther Marsh, Ontario GILBERT PROULX and FREDERICK F. GILBERT 37 T/ Some effects of winter shelter conditions on White-tailed Deer, Odocoileus virginianus, fawns F. F. GILBERT and M. C. BATEMAN 39] Breeding ecology of the Horned Grebe, Podiceps auritus, in southwestern Manitoba ROBERT S. FERGUSON and SPENCER G. SEALY 40] Variability in size and age at sexual maturity of Witch Flounder, Glyptocephalus cynoglossus, in the Canadian Maritimes region of the northwest Atlantic Ocean TERRY D. BEACHAM 409 Food habits of young anadromous Alewives, A/osa pseudoharengus, in Lake Ainslie, Nova Scotia ROBERT S. GREGORY, GAIL S. BROWN, and GRAHAM R. DABORN 423 Coastwide distribution and ocean migration patterns of stream- and ocean-type Chinook Salmon, Oncorhynchus tshawytscha M. C. HEALEY 427 Wintering biology of Mourning Doves, Zenaida macroura, in Ontario EDWARD R. ARMSTRONG and DAVID L. G. NOAKES 434 Adaptation of Atlantic Salmon, Sa/mo salar, to a restricted freshwater environment J.-P. CUERRIER 439 Elk, Cervus elaphus, foraging related to forest management and succession in Idaho LARRY L. IRWIN and JAMES M. PEEK 443 Notes Caspian Terns, Sterna caspia, breeding in Labrador A. R. LOCK 448 Range extension of the sage Artemisia rupestris spp. woodii in southwestern Yukon M. Hoers, D. RUSSELL, and BETH EREAUX 449 Nesting of the Thick-billed Murre, Uria lomvia, in British Columbia ANNE VALLEE and RICHARD J. CANNINGS 450 Ring-billed Gull, Larus delawarensis, predation on bat, Myotis, injured by an American Kestrel, Falco sparverius GEOFFREY L. HOLROYD and ELISABETH G. BEAUBIEN 452 High nesting density of ducks on an island in Saskatchewan PHILIP M. BROWNE, DAviID A. DUFFUS, and RONALD W. BOYCHUK 453 Recoveries of Saskatchewan-banded White-winged Scoters, Melanitta fusca C. STUART HOUSTON and PATRICK W. BROWN 454 Albino Eastern Garter Snakes, Thamnophis sirtalis sirtalis, from Ontario WAYNE F. WELLER 456 concluded on inside back cover ISSN 0008-3550 ae i: oe A ‘ca oy i } o i a 5 fi ak : i > } . a aS Teg x 2 5, aN oS 2 5 3 2044 072 176 274 be, aE ear iste reer eats a4 erates Ly, Wb Bea Mwy WA ea aN, Uae i)