HARVARD UNIVERSITY e Library of the
Museum of
Comparative Zoology
am
Wikis on)
ay a i
+ neh ay
Wi ot
Jee Ae ae i ; Matai one
i y }
We i ely. NE
Aa “
1,10 ‘hitb
THE CANADIAN FIELD-NATURALIST
Volume 102
1988
THE OTTAWA FIELD-NATURALISTS" CLUB LIBRARY
OTTAWA CANADA NOV 04 1988
HARVARD UNIVERSITY
The CANADIAN FIELD-NATURALIST
Published by THE OTTAWA FIELD-NATURALISTS’ CLUB, Ottawa, Canada
CSO ~
SS
Volume 102, Number 1 January-March 1988
The Ottawa Field-Naturalists’ Club
FOUNDED IN 1879
Patron Her Excellency The Right Honourable Jeanne Sauve, P.C., C.C., C.M.M., C.D., Governor General of Canada 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
Edward L. Bousfield Claude E. Garton Stewart D. MacDonald Hugh M. Raup Irwin M. Brodo W. Earl Godfrey George H. McGee Loris S. Russell William J. Cody C. Stuart Houston Verna Ross McGiffin Douglas B. O. Savile William G. Dore Louise de K. Lawrence Hue N. MacKenzie Pauline Snure
R. Yorke Edwards Thomas H. Manning Eugene G. Munroe Mary E. Stuart Clarence Frankton Don E. McAllister Robert W. Nero Sheila Thomson
1988 Council
President: Bill Gummer Barry Bendell Doreen Duchesne Vice-Presidents: Jeff Harrison Ronald E. Bedford Eileen Evans Kenneth Strang Daniel F. Brunton Peter Hall Recording Secretary: Roy John William J. Cody Shane Jordan Corresponding Secretary: Barbara A. Campbell Kathleen Conlan Catherine O’Keefe Treasurer: Frank Valentine Francis R. Cook E. Franklin Pope Peter Croal Wright Smith
Barbara Desrochers Paul B.M. Ward Elliane M. Dickson
Those wishing to communicate with the Club should address correspondence to: The Ottawa Field-Naturalists’ Club, Box 3264, Postal Station C, Ottawa, Canada K1Y 4J5. For information 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, P.O. Box 3443, Station D, Ottawa, Ontario K1P 6P4; (613) 996-1755; Assistant to Editor: Lise Meyboom; Editorial Assistant: Elizabeth Morton; Copy Editor: Louis L’Arrivée
Business Manager: William J. Cody, Box 3264, Postal Station C, Ottawa, Ontario K1Y 4J5 (613) 996-1665
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: Anthony J. Erskine William O. Pruitt, Jr. C.D. Bird Charles Jonkel Stephen M. Smith Edward L. Bousfield Donald E. McAllister Constantinus G. Van Zyll de Jong
Chairman, Publications Committee: Ronald E. Bedford All manuscripts intended for publication should be addressed to the Editor.
Subscriptions and Membership
Subscription rates for individuals are $20 per calendar year. Libraries and other institutions may subscribe at the rate of $35 per year (volume). The Ottawa Field-Naturalists’ Club annual membership fee of $20 includes a subscription to The Canadian Field-Naturalist. All foreign subscribers (including USA) must add an additional $3.00 to cover postage. 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: Leatherback Turtle, Dermochelys coriacea, on dock at Fisherman’s Market, St. John’s, Newfoundland, after getting entangled in fishing gear, and shortly before release. Photograph courtesy of Jon Lien. See Goff and Lien pp. 1-5.
The Canadian Field-Naturalist
Volume 102, Number |
January-March 1988
Atlantic Leatherback Turtles, Dermochelys coriacea, in Cold Water Off Newfoundland and Labrador
GREGORY P. GOFF! and JON LIEN2
'Fisheries Research Branch, Department of Fisheries and Oceans, St. John’s, Newfoundland AIC 5X1 Present address: Marine Sciences Research Laboratory, Memorial University of Newfoundland, St. John’s,
Newfoundland AIC 5S7
2Newfoundland Institute for Cold Ocean Science and Department of Psychology, Memorial University of
Newfoundland, St. John’s, Newfoundland AIB 3X9
Goff, Gregory P., and Jon Lien. 1988. Atlantic Leatherback Turtles, Dermochelys coriacea, in cold water off Newfoundland and Labrador. Canadian Field—Naturalist 102(1): 1-S.
Encounters with 20 Leatherback Turtles (Dermochelys coriacea) in waters off Newfoundland and Labrador between 1976 and 1985, primarily through incidental catches in inshore fishing gear, are reported. Leatherback Turtles were found in July-September and were associated with seasonally high water temperatures. Healthy animals which occur in coastal Newfoundland waters are probably not strays, and encounters to date may indicate the turtles’ regular use of
this habitat.
Key Words: Leatherback Turtle, Dermochelys coriacea, incidental fishery catches, Newfoundland.
The marine Leatherback Turtle, Dermochelys coriacea, is the largest known extant reptile, and reaches weights in excess of 680 kg (Rhodin et al. 1981). It is a cosmopolitan species that enters Canadian waters off both the Atlantic and Pacific coasts.
The Leatherback Turtle is suitably adapted to cold water. Friar et al. (1972) recorded a 25.5°C body temperature for a captive Leatherback Turtle in 7.5°C seawater. They possess vascular counter- current heat exchangers in their flippers (Greer et al. 1973), and have thick subcutaneous insulation. Standora et al. (1984) measured this species’ endothermic ability to respond to a drop in ambient temperature by metabolically increasing its heat production.
The Leatherback Turtles’ intrusions into cold water are seasonal (Pritchard 1971). Bleakney (1965) compiled records of 88 Leatherback Turtles in New England and Nova Scotian waters from June to October between 1899 and 1964 and concluded that these northward travels were made by healthy turtles of various ages and sexes in order to feed on large northern jellyfish (Cyanea capillata arctica). Based on preferential feeding by leatherbacks on this species, Lazelle (1980) argued
that the high medusan-producing waters off New England are critical leatherback habitat.
Bleakney (1965) recorded only two Leatherback Turtles in the cold northeastern coastal waters around Newfoundland and Labrador. Steele (1972) reported a specimen in Conception Bay, Newfoundland. Threlfall (1978) reported a single animal found near Nain, Labrador.
Despite recent increases in its world population estimates, the leatherback is still considered an endangered species (Pritchard 1982). Many adults are slaughtered annually on nesting beaches and egg collection is still practised. Once considered secure away from nesting beaches, this pelagic species now suffers some commercial fisheries- related mortality in Pacific (Balazs 1982) and Atlantic (Lien 1980) net fisheries.
This paper reports encounters with leatherbacks in Newfoundland coastal waters from 1976 to 1985 and ocean temperatures in the vicinity of the sighted turtles.
Methods
In 1979 a toll-free phone line was initiated by which fishermen could report whales and sharks entrapped in their fishing gear. This service was
2 THE CANADIAN FIELD-NATURALIST
widely advertised throughout Newfoundland and Labrador and has been described in detail elsewhere (Lien 1980). As they became more familiar with the service, fishermen began to report unusual fish, seals, squid and turtles caught in their nets, as well as abnormal water conditions and fishery problems (Lien et al. 1985).
While this method of reporting is frequently used, it does depend on fishermen to volunteer information. Under-reporting of whale and shark entrapments varied from 25% in the first several years to about 10% in recent years (Lien et al. 1985). Reports of ‘unusual’ catches apart from whales and sharks were not common in 1979-1980. They have increased since that time. There is no way of knowing the likelihood that sightings or catches of turtles would be reported. The Leatherback Turtles which are reported are therefore a minimum estimate of the actual numbers which have been encountered around the Newfoundland coast.
Prior to establishment of the phone reporting system, occasional reports of Leatherback Turtles were given to field officers of the Department of Fisheries and Oceans or Memorial University of Newfoundland.
In some instances of a reported turtle, an observer was sent to the site to verify the catch and
Vol. 102
the species. Alternately, photographs of entrapped specimens were often available. When a catch was reported by a vessel at sea, identity was determined by radio conversation. Specimens of recently dead turtles were obtained when possible for measure- ment and dissection.
Ocean temperatures in the vicinity of encoun- tered turtles were approximated using data obtained by the long-term temperature monitoring program of the Department of Fisheries and Oceans (Dobson 1982, 1983, 1984; Dobson et al. 1985). This monitoring program uses Ryan Model J thermographs deployed at a 5 to 10 m depth in numerous sites around the Newfoundland coastline. Temperatures at the thermograph located nearest an encountered turtle are presented for the date of the encounter. Maximum temperatures for the month and season of the encounter were also compiled.
Examination of several dead specimens provided morphometric measurements to indicate the animals’ size.
Results
Locations of 20 Leatherback Turtles encountered between 1976 and 1985 in northwest Atlantic waters near Newfoundland and water temperatures on the encounter date are presented in Table |. Locations of
TABLE |. The sightings and incidental catches of Leatherback Turtles in Newfoundland (1976-1985).
Date Location
21 September 1976 Western Bay, C.B. 3 October 1976 Lourdes 6 July 1977 Trepassey
2 September 1981 13 September 1981 28 September 1981 21 October 1981
15 August 1982
Petty Harbor St. Bernard’s, F.B. Bauline South York Harbor
Port aux Basque
20 August 1982 Bauline, C.B. 25 July 1983 St. Brides, P.B. 14 August 1983 Lumsden
20 March 1984 13. August 1984 26 August 1984
Sunnyside, T.B. Jerseyside, P.B. Southern Hbr., P.B.
4 August 1985 Flatrock 5 August 1985 Burin 5 August 1985 Flatrock
10 August 1985 14 August 1985 23 September 1985
Bonavista, B.B. Harbour Grace, C.B. Happy Adventure, B.B.
Water Temp Capture Condition (rey Method at release _- salmon net alive — herring net alive — gillnet alive 15 trawl line alive 14 free swimming alive Ik) gillnet alive — found dead dead _ gillnet dead 11 gillnet alive 14 gillnet alive 13 gillnet dead 0 free swimming alive 14 trawl line dead 14 found dead dead 12 free swimming dead 9 gillnet alive 12 free swimming alive 10 gillnet alive 11 crab pot line dead 12 gillnet alive
1988 GOFF AND LIEN: LEATHERBACKS OFF NEWFOUNDLAND AND LABRADOR 3
63° 62° 61° 60° 59° 58° 57° 56° 65° 54° 53° 520 51° 50,0
@ REPORTED 1976-1985 : Wi PUBLISHED RECORDS PRE —1976 56° aot A SIGHTINGS NEAR 0°C 56°
LABRADOR SEA
550 550
54° 54°
oe ~ LABRADOR * 53°
3 ATLANTIC 520 ee mw ee nee me ee ee ee es oe _—-=. 52°
OCEAN
sot QUEBEC yw fy 51°
50° 50°
49° 49°
GULF
Anticosti Ny OF
ST. LAWRENCE
48° 48°
47°
46° 46° 63° 62° 6I 60° 59° 58° Sie 56° 55° 54° 53° 52° 51° 50° FiGuRE 1. Locations of encounters with Leatherback Turtles in Newfoundland waters. Reports prior to 1976 are from Squires (1954), Steele (1972), Threlfall (1978), and Bleakney (1965).
these encounters, along with locations of leather- Of 20 Leatherback Turtles encountered, 14 were backs previously reported in waters off Newfound- entangled in fishing gear (70%), 4 were observed land, are presented in Figure 1. swimming freely (20%), and 2 were reported dead
4 THE CANADIAN FIELD-NATURALIST
Vol. 102
TABLE 2. Summary of reported encounters with Leatherback Turtles by Newfoundland fishermen 1976-1985.
Condition Total N N Released Mortality Entangled in fishing gear 14 10 4
a. gillnets 11 9 2
b. trawl or crab pot lines 3 I 2} Free-swimming 4 3 1 Found dead 2 — 2, Totals 20 13 7
(10%). Ten of the turtles entangled in fishing gear were released alive (70%); the remainder were found dead or were killed in the process of being removed from the gear. One free-swimming turtle was shot (Table 2).
August and September were the months when turtle encounters were most likely. Ocean temperatures on the day Leatherback Turtles were encountered (mean + S.D. = 12.6 = 1.9°C) were on average within 2.2 degrees C of the monthly maximum temperature (mean 14.7 + 1.6°C) at that location (Table 1). Average annual maximum temperature recorded 14.4 + 1.3°C.
One notable exception to this association with seasonal high water temperatures was one turtle reported in Trinity Bay on 20 March 1984. It was alive and observed by fishermen throughout an entire day swimming in open water leads among ice where water temperatures was approximately 0°C. It is not known if this animal survived.
The sizes of turtles in Newfoundland waters are indicated by the measurements in Table 3. No tags were found on any animals examined.
Discussion
Recent records of Leatherback Turtles in waters off Newfoundland confirm their regular and seasonal occurrence in this area. Most of the turtles encountered are reported in the eastern portion of
the province where the amount of inshore fishing activity is highest. The increase in reports in recent years reflects increased use of the toll-free phone system and does not represent any real increase in catch per fishing effort. Still, the number of turtles reported very likely represents a minimal proportion of the turtles encountered and an even smaller proportion of the actual number of turtles present in these waters.
Ocean circulation on the Newfoundland continental shelf is such that a large proportion of the turtles reported here came under the influence of the Labrador Current (Petrie and Anderson 1983), a southward-flowing mass of cold arctic water. The average ocean temperature near the locations of turtle encounters between 1981 and 1985 was 12.6°C, near the warmest ocean temperatures (mean = 14.4°C) recorded in those areas for the year. It is likely that Leatherback Turtles move northward in warm Gulf Stream water and only venture into the Labrador current water at inshore Newfoundland when it is very near yearly maximum temperatures.
Examinations of dead specimens and the reported behaviour of free-swimming. turtles confirm that these animals north of 48 degrees latitude are healthy. The presence of 6-7 cm of adipose tissue under the carapace indicates their good condition and may represent a further
TABLE 3. Sex and measurements (in cm) of four Leatherback Turtles from waters off Newfoundland.
Specimen
Je le Is
Measure Oct. 1981 Sex — Carapace length
over the curve 165 Carapace width —
curved maximum 120 Margins of left front flipper (anterior/ posterior) 123/92
Aug. 1983 Aug. 1985 Aug. 1985 135 146 140 Be 110 84 = 100/75 94/73
1988
adaptation for periodic intrusion into cold environments.
The destinations and routes of the turtles encountered in northwestern Atlantic waters are not established. It is unlikely that they are strays because of their excellent condition. Carapace lengths, which range from 135 to 165 cm, suggest these animals are mature (Rhodin 1985), but their eventual breeding sites and the nature of their migratory paths remain open to speculation. They should be considered regular migrants to Newfoundland waters.
The mortality which results from entanglement with fishing gear is fairly low; about 70% of the entangled turtles between 1981 and 1985 were released alive. Two of four entangled turtles perished naturally in the gear; two others were killed. It is possible that the other two turtles found dead were killed in fishing gear. Mortality may have decreased recently through educational efforts which give fishermen encouragement to release live animals (Lien et al. 1985).
Acknowledgments
We would like to thank Lois Batemen, Jack Temple, Gary Cowen, Alan Burger, Gary Stenson, Chris Harvey-Clark, Heidi Obserheide and Ellison and Sadie Barfett who aided with various aspects of field work. Field personnel from the Newfound- land Department of Fisheries and from Fisheries and Oceans Canada also provided assistance on many occasions. Several companies and agencies also aided our work, including Ocean Harvesters, Beothic Fisheries and the Port aux Basques Provincial Bait Depot. Support for the Memorial University of Newfoundland toll-free phone system and its entrapment program came from Fisheries and Oceans Canada and is gratefully acknowledged. Without the cooperation and assistance of fishermen we would have little information on Leatherback Turtles; we express sincere thanks for their help. Thanks are also due to C.C. Davis, G. Stensen and K. Breck who criticized drafts of this paper and to Francis Cook for encouraging us to write it.
Literature Cited
Balazs, G. H. 1982. Driftnets catch leatherback turtles. Oryx 428-430.
Bleakney, J.S. 1965. Reports of marine turtles from New England and eastern Canada. Canadian Field—Naturalist 79(2): 120-128.
Cook, F.R. 1981. Status report on the Leatherback Turtle Dermochelys coriacea. Committee on the Status of Endangered Wildlife in Canada (COSEWIC). 18 pp.
GOFF AND LIEN: LEATHERBACKS OFF NEWFOUNDLAND AND LABRADOR 5
Dobson, D., and B. Petrie. 1982. Long-term tempera- ture monitoring program 1981. Canadian data report of hydrography and ocean sciences No. 6. 297 pp.
Dobson, D., and B. Petrie. 1983. Long-term tempera- ture monitoring program 1982: Newfoundland Region. Canadian data report of hydrography and ocean sciences No. 11. 335 pp.
Dobson, D., and B. Petrie. 1984. Long-term tempera- ture monitoring program 1983: Newfoundland Region. Canadian data report of hydrography and ocean sciences No. 21. 411 pp.
Dobson, D., B. Petrie, and P. Stead. 1985. Long-term temperature monitoring program 1984: Newfound- land Region. Canadian data report of hydrography and ocean sciences No. 34. 333 pp.
Friar W., R.G. Ackman, and N. Mro- sovsky. 1972. Body temperature of Dermochelys coriacea: warm turtle from cold water.
Greer, A.E., J.D. Lazelle, and R.M. Wright. 1973. Anatomical evidence for a counter- current heat exchange in the leatherback turtle (Dermochelys coriacea). Nature 224: 131.
Lazelle, J.D. 1980. New England waters: critical habitat for marine turtles. Copeia 1980 (2): 290-295. Lien, J. 1980. Whale collisions with fishing gear in Newfoundland. Final report to Fisheries and Oceans
Canada — Newfoundland Region. 316 pp.
Lien J., S. Staniforth, and L. Fawcett. 1985. Teaching fishermen about whales: the role of education in a fisheries management and conservation problem. Pp. 231-240 in Marine parks and conservation: Challenge and promise, Volume |. Edited by J. Lien and R. Graham. NPPAC, Toronto.
Petrie, B., and C. Anderson. 1983. Circulation on the Newfoundland continental shelf. Atmosphere-Ocean 21(2): 207-226.
Pritchard, P. C. H. 1971. The leatherback or leathery turtle Dermochelys coriacea. International union for conservation of nature and natural resources monograph No. |. 39 pp.
Pritchard, P. H.C. 1982. Nesting of the leatherback turtle Dermochelys coriacea in Pacific Mexico, with a new estimate of the world population status. Copeia 1982(4): 741-747.
Rhodin, A. G. J. 1985. Comparative chondro-osseous development and growth of marine turtles. Copeia 1985(3): 752-771.
Rhodin, A.G.J., J.A. Ogden, and G.J. Cono- logue. 1981. Chondro-osseus morphology of Dermochelys coriacea a marine reptile with mammal- ian skeletal features. Nature 290(5803): 244-246.
Squires, H. J. 1954. Records of the marine turtles in the Newfoundland area. Copiea 1954(1): 68.
Standora, E. A., J. R. Spotila, J. A. Keinath, and C. R. Shoop. 1984. Body temperatures, diving cycles, and movement of a subadult leatherback turtle, Dermochelys coriacea Herpetologica 40(2): 169-176.
Steele, D. H. 1972. A leatherback turtle (Dermochelys coriacea) caught in Conception Bay. Osprey 3(6): 44- 46.
Threlfall, W. 1978. First record of the Atlantic Leatherback Turtle (Dermochelys coriacea) from Labrador. Canadian Field—Naturalist 92(3): 287.
Received 27 June 1986 Accepted 4 May 1987
Effects of the Herbicide 2,4,5-T on the Habitat and Abundance of Breeding Birds and Small Mammals of a Conifer Clearcut in Nova Scotia
B. FREEDMAN!, A. M. POIRIER!, R. MORASH!, and F. SCOTT?
'School for Resource and Environmental Studies and Department of Biology, Dalhousie University, Halifax, Nova Scotia B3H 4J1 2Nova Scotia Museum, 1747 Summer St., Halifax, Nova Scotia B3H 3A6
Freedman, B., A. M. Poirier, R. Morash, and F. Scott. 1988. Effects of the herbicide 2,4,5-T on the habitat and abundance of breeding birds and small mammals of a conifer clearcut in Nova Scotia. Canadian Field- Naturalist 102(1): 6-11.
After the silvicultural herbicide spraying of a conifer clearcut in Nova Scotia, the foliage cover of shrub-sized plants was reduced by an average factor of 41% compared with the pre-spray condition, while during the same period an unsprayed reference plot increased in cover by a factor of 29%. Shrub-sized plants on the sprayed plots were also reduced in stem density (by 31%) and basal area (67%). The total cover of ground vegetation was also reduced on the spray plots (by an average factor of 19%), while on the reference plot it increased by 16%. However, vegetation on the sprayed plots was still abundant, and the habitat changes were not sufficient to cause an important difference in the abundance, species richness, and diversity of breeding birds and small mammals between sprayed and unsprayed plots. The total bird density was 71 pairs/ 10 ha on the unsprayed clearcut plot, 76 and 77 pr/ 10 ha on two plots sprayed at the prescribed silvicultural rate of 3 kg 2,4,5-T/ha, and 50 pr/ 10 ha on a plot sprayed at 6 kg/ha. The abundance of small
mammals on these same treatment-plots was 12.8, 14.3, 13.5, and 11.7 per 100 trap-nights.
Key Words: forestry, herbicide, 2,4,5-T, birds, small mammals.
The herbicides that have been commonly used for silvicultural purposes in the 1980s in Canada (i.e. 2,4,5-T, 2,4-D, and glyphosate) have a small toxicity to birds and mammals under operational spray conditions (Way 1969; Kenaga 1975; Morrison and Meslow 1983; Anonymous 1984). However, herbicides have an indirect effect on the habitat of at least some species of wildlife, because they cause changes in: 1) the absolute and relative distribution of biomass among plant species; 11) the horizontal and vertical structure of vegetation; and ili) the abundance and quality of the invertebrate and plant food resource (Newton and Norris 1976; Borrecco et al. 1979; Ware 1980; Freedman 1982; Morrison and Meslow 1983, 1984a,b).
Because the impacts of herbicide spraying on the breeding birds and small mammals of regenerating cutovers have seldom been studied (we are aware of no studies in Canada, and only a few from the United States: Beaver 1976; Savidge 1978; Morrison and Meslow 1983), we initiated such work in Nova Scotia. Here we describe the abundance of breeding birds and small mammals, and the changes in their habitat, one year after spraying the phenoxy herbicide 2,4,5-T in a conifer release program.
Methods The study site is near Stewiacke in central Nova Scotia, at 45°12’ N, 63°23’ W. The site was
formerly occupied by a conifer stand, but it had been clearcut three years prior to our study, planted with Black and Norway spruce seedlings (see Appendix | for all binomials), and required a silvicultural herbicide treatment to reduce the abundance of “weeds” and release the planted and natural conifer regeneration. The clearcut was about 20 hectares in size, roughly rectangular, and appeared uniform in vegetation. It was divided into four rectangular treatment-plots. One plot (4.9 ha) was an unsprayed reference plot, two plots (4.1 and 5.1 ha) were sprayed with 2,4,5-T at the prescribed silvicultural rate (3 kg 2,4,5-T/ha, hereafter abbreviated as “1 X”), and one plot was treated at double that rate (“2X”). The plots were sprayed on 11 September 1983 by the Nova Scotia Department of Lands and Forests using a skidder- mounted apparatus (3-nozzle spray cluster, 1700 litre capacity tank with internal stirring mechanism).
Vegetation surveys were done in mid-August of 1983 (prior to spraying) and 1984 (one-year post- spray). Shrub-sized plants were measured in 12 evenly-spaced, 5 m X 5 m permanent quadrats per treatment-plot. For each leaf-bearing stem, the species was noted and the diameter was measured at 25 cm above the ground surface. Calculations were made of stem basal area (SBA; m2/ha) and density (stems/ha).
1988
Ground vegetation was surveyed in| mX1m quadrats located in each corner of the twelve 5mX5m quadrats per plot (i.e. 48 ground vegetation quadrats per treatment-plot; however, note that vegetation was not surveyed on the 5.1 ha 1X treatment-plot). For each species, cover was estimated as the proportion of the ground surface occupied by a perpendicular projection of the plant foliage, attempting to take overlap into account (Grieg-Smith 1964). The statistical significance of pre-spray differences in vegetation among the treatment-plots was tested by analysis of variance. Significant differences from the ANOVA were assessed using Duncan’s multiple range test (Ott 1977). In this analysis, the pre-spray treatment- plots were considered to be replicates with internal sub-sampling. Since the spray treatments were not replicated, it was impossible to separate location (block) effects from treatment effects (Hulbert 1984) once the plots were sprayed. Therefore, our post-spray analysis of vegetation data involved the use of relatively simple statistical tests to draw out major points of biological interest (Chatfield 1985). For each treatment-plot, paired t-tests were used to determine whether the vegetational differences between years were statistically significant. This procedure is used to determine the statistical significance of differences between dependent samples (Ott 1977), as in this study where vegetation was sampled in permanent quadrats. For the cover data only, arcsine square root transformations were used (Sokal and Rohlf 1974).
Breeding birds were censused using the spot- mapping method (Williams 1936; IBCC 1970; Robbins 1978). Each plot was surveyed ten times (7-8 dawn and 2-3 dusk censuses) between 22 May and 28 June 1984. The density of breeding pairs was expressed as pairs/10 ha for species having = 0.5 territory/plot. Bird species diversity was calculated as H’=-Xpi-In p; (Shannon and Weaver 1949), where pj is the relative density of species 1. Because of the sampling design, which involved a whole-plot census of essentially unreplicated treatments, we did not test for the Statistical significance of differences in avian density among treatments.
Small mammals were surveyed with Victor snaptraps baited with peanut butter and rolled oats. Three traps were placed within a | m radius at stations located at 10 m intervals along a single transect through the length of each treatment-plot. The number of sampling stations ranged from 26
FREEDMAN, POIRIER, MORASH, AND SCOTT: EFFECTS OF 2,4,5-T q
to 30. Sampling was done for three consecutive nights during each of three intervals in 1984 (28-30 June, 31 July - 2 August, 28-30 August). Abundance was standardized as the number of captures per 100 trap-nights. One-way ANOVA calculations were based on the number of captures per night and sampling station. H’ was calculated using relative density to estimate pi.
Results and Discussion 1. Vegetation.
The description of vegetation changes presented here is relatively superficial. A detailed, species- specific analysis is on file. [R. Morash, B. Freedman, and C. Stewart. 1986. Initial impacts of silvicultural herbicide spraying on the vegetation of regenerating clearcuts in central Nova Scotia. Research Report to the Canadian Forestry Service. On file at Department of Biology, Dalhousie University, Halifax, N.S.]
After spraying, most of the live shrub stems on the unsprayed plots had a reduced leaf cover because of defoliation by the 2,4,5-T (Table 1). For example, compared with the 1983 pre-spray condition, the average foliage cover of Red Maple decreased in 1984 by a factor of 70% on the 1X plot (p = 0.02) and by 84% on the 2X plot (p = 0.02), whereas it increased by 25% on the unsprayed plot (p< 0.01). Because of plant growth on the reference plot, the SBA of Red Maple increased by 13% (n.s.) between the 1983 and 1984 samplings, while stem density was little changed. However, because of herbicide-caused mortality the SBA of foliage-bearing stems of Red Maple on the 1X spray plot decreased by 47% (p = 0.05) and stem density by 25% (n.s.), and on the 2X spray plot SBA decreased by 19% and density by 25% (both n.s.). Birches increased in foliage cover by 80%, in SBA by 80%, and in density by 22% on the unsprayed plot (all p < 0.001), whereas because of mortality these decreased by 4% (n.s.), 51% (p = 0.005), and 62% (p = 0.004) respectively on the 1X spray plot, and by 74% (p< 0.001), 62% (p = 0.04), and 67% (p = 0.009) on the 2X plot. Red Raspberry increased in cover by 22% (p = 0.03) and in stem density by 35% (p=0.005) on the unsprayed plot, but on the 1X spray plot these decreased by 20% (p = 0.04) and 24% (p = 0.004) respectively, and on the 2X plot by 63% (p <0.001) and 28% (p = 0.005). Considering all shrub species in aggregate, on the unsprayed plot foliage cover increased by 29% (p< 0.001), live SBA by 34% (p < 0.001), and stem density by 29%
8 THE CANADIAN FIELD-NATURALIST
Vol. 102
TABLE |. Selected characteristics of the vegetation of the treatment-plots prior to the spraying of 2,4,5-T, and in the first post-spray year. The data are average values (+ S.E.) of n = 125 * 5 m quadrats for SBA and density of shrub-
sized vegetation, and n = 48 | X | m quadrats for foliage cover.
Spray Pre-spray First-year Post-Spray
Treat- SBA Density Cover SBA Density Cover
ment Vegetation (m2/ ha) (103/ha) (%) (m2/ha) 103/ha) (%)
OX Acer rubrum 111+0.60 140+ 5.4 8.3 + 6.2 1:25 220.75) 13:1 e13:8" #1042163 Betula spp. 0.43+0.09 26.3 + 6.0 3y 1 ae (0), 9/ #0.77 £0.16 #320466 #5.6+ 1.3 Rubus spp. 0363 22 OM14) 8521 ss1993) S17 Dee 45 #0.86£0.16 115422 #20.9+2.8 All shrub species 2.4540.70 147424 29 Bie Tel #3.28+0.78 #190430 #37.847.0 All pteridophytes Dif dret= Ted: 25.4 + 6.1 All monocots 8:5) ae 2.5 #11.0 + 2.5 All dicots 43.4+ 6.4 #56.1 + 6.6 All plants 89.1 + 8.3 #103 + 8
IX 9 Acer rubrum 1.53 + 0.61 OF a= 226 148s 6.2 #0.81 + 0.34 6841.7 #45+1.9 Betula spp. *1.0420.27 51.54 12.4 S3EL O08 #0.51 40.16 #19.6 + 4.2 Syl sexist Rubus spp. 1026 2104 174 = 19 = *300a= 3:2 #0.96+ 0.09 #133413 #24.1 43.2 All shrub species 403 cE\0l05 gia 258i 2288) +52 Dee 0 #2.5140.36 #176211 #35.543.9 All pteridophytes +32 Ose eZ #4.6 + 1.4 All monocots SEOs 29) #10.6 + 2.5 All dicots 69.8 + 6.5 asi) ae 8)-) All plants 90.8 + 7.4 #183 425.2
2X = Acer rubrum 1.29 + 0.98 8.7444 12.2+5.0 1.05 + 1.01 65+4.0 #1.940.9 Betula spp. 0.454013 20.445.3 Pei ae (Ves #0.17+0.10 #68440 #0.740.3 Rubus spp. POE ONS al Sorte I 39D #0.80+0.14 #111217 #24642.8 All shrub species 2.99 + 0.88 195222 545 i= 16:0 #2.1640.97 #134420 #27.242.9 All pteridophytes 14.8 + 3.3 #19.0 + 4.5 Al monocots 40+ 0.8 #6.0 + 1.0 All dicots 70.3 = 5.9 #44.1 + 3.0 All plants Sail ae (oy? HM Bits) AE Ola)
*significantly different (p < 0.05) from the 0X treatment in the pre-spray sampling: F-tests #significant change (p < 0.05) from pre-spray condition; paired t-tests.
(p = 0.005), while on the 1X plot these decreased by 32% (p=0.004), 38% (p=0.01), and 30% (p < 0.001) respectively, and on the 2X plot they decreased by 50% (p < 0.001), 28% (p = 0.02), and 31% (p = 0.003).
In the first growing season after spraying, the cover of ground vegetation averaged 103%, 78%, and 74% respectively, on the unsprayed, 1X sprayed, and 2X sprayed treatment-plots (Table 1). Compared with their 1983 pre-spray condition, these represent a relative increase in cover of 16% (p < 0.001) on the unsprayed plot, a decrease of 14% (p = 0.03) on the IX spray plot, and a decrease of 23% (p < 0.001) on the 2X plot. Pteridophytes are not susceptible to 2,4,5-T. The average cover of pteridophytes was changed marginally on the unsprayed plot between 1983 and 1984, while on both spray plots it increased by 28% (both
p = 0.04). Monocotyledonous plants are also not susceptible to 2,4,5-T. Monocot cover increased on all plots, by 29% on the unsprayed plot, 20% on the 1X spray plot, and 50% on the 2X plot (all p< 0.001). Most dicotyledonous plants are susceptible to damage from 2,4,5-T. The cover of herbaceous dicot plants increased by 30% (p < 0.001) on the reference plot, and there was also a small increase of 4%-7% (both n.s.) on the two sprayed plots. Although the 2,4,5-T caused much mortality of dicot herbs, there was also substantial regeneration of this group of plants in the first post-spray growing season. Various species of Asteraceae were especially prominent in the post-herbiciding regeneration. In the first year after spraying, the cover of Asteraceae increased by a factor of 27% (p = 0.006) on the 1X plot, and by 37% (p = 0.004) on the 2X plot, while on the reference plot it increased by 49% (p < 0.001).
1988
FREEDMAN, POIRIER, MORASH, AND SCOTT: EFFECTS OF 2,4,5-T 9
TABLE 2. Breeding birds on the 2,4,5-T spray plots. Data are in pairs/10 ha; richness d= the number of species;
diversity = -Xpi- In pi
Species unsprayed American Woodcock
Ruby-throated Hummingbird
Olive-sided Flycatcher
Alder Flycatcher 2.0 American Robin 1.0 Mourning Warbler 3) Common Yellowthroat 24.4 Song Sparrow 4.1 Lincoln’s Sparrow 2.0 White-throated Sparrow 19.3 Northern Junco 9.2 American Goldfinch 4.1 TOTAL DENSITY ED
SPECIES RICHNESS* 17 SPECIES DIVERSITY* 1.76
1X spray A 1X spray B 2X spray 1.3 1.3 163 2 1.3 0.9 2:5 3.8 2.6 24.6 26.9 18.8 7.4 3.8 Il7/ 6.1 2.6 3.4 Dal 20.5 16.2 4.9 9.0 Sel 7.4 5) 1.7 76.2 76.9 50.4 11 19 12 E73 1.84 1.60
*includes transients and species with < 1/2 territory per plot.
2. Breeding Birds.
There were small differences in the density of breeding birds among the herbicide treatment- plots in the first post-spray growing season (Table 2). All of the treatment-plots had the same dominant species of bird, and the relative abundance of these were similar. For example, the Common Yellowthroat accounted for 34% of the total bird density on the unsprayed plot, 32% and 35% on the two 1X spray plots, and 37% on the 2X plot, while the White-throated Sparrow was 27%, 29%, 27%, and 32% respectively. All of the common breeding species had a smaller absolute density on the 2X plot, which averaged 29%-34% fewer individuals than on the reference plot or on the two 1X spray plots. However, no bird species were eliminated from the 2X plot. Overall, in the first post-spray year the habitat changes caused by the 2,4,5-T treatment appears to have caused no more than a minor difference in the breeding birds of the various treatment-plots of this conifer clearcut.
Few other studies have reported the effects of silvicultural herbicide spraying on birds. As in our study, Morrison and Meslow (1984a,b) reported only moderate differences in the breeding birds of herbicide-sprayed and unsprayed clearcuts in Oregon. A much larger change in the breeding bird community, particularly in species composition, takes place when mature forest is clearcut (Franzreb 1978; McArthur 1980; Freedman et al. 1981; Welsh 1981; Morgan and Freedman 1986).
3. Small Mammals.
The overall abundance and diversity of small mammals did not differ (p > 0.05) among the treatment-plots in the first post-spray year (Table 3). Variable results have been reported among the few studies that have examined the effects of silvicultural herbicide treatments on small mammals. Borrecco et al. (1979) found no effect on the overall abundance of small mammals, but reported a change in species composition. Both Savidge (1978) and Kirkland (1978) found an increase of total abundance, but little change in species composition. Spencer and Barrett (1980) found that the abundance of Meadow Voles decreased by one-half after herbicide spraying.
Conclusions
Shrub-sized angiosperm plants were greatly decreased in abundance as a result of mortality caused by the 2,4,5-T spray treatment. The primary intent of the herbicide treatment was, in fact, to achieve this ecological effect. The ground vegetation of the spray plots suffered somewhat less mortality, but there were large changes in species composition because of differential susceptibility of taxa to the 2,4,5-T. In spite of the large changes in the structure and plant species composition of the vegetation of the sprayed plots, an important effect on the abundance and species composition of breeding birds and small mammals was not apparent.
10 THE CANADIAN FIELD-NATURALIST Vol. 102 TABLE 3. Small mammal abundance (no./ 100 trap-nights) for the various treatment plots.
Species unsprayed IX spray A IX spray B 2X spray Masked Shrew 6.8 8.9 ee led Smoky Shrew 1.0 1.0 1.0 0.0 Short-tailed Shrew 0.9 1.4 233 122 Pygmy Shrew 0.4 0.7 0.0 0.1 Arctic Shrew 0.0 0.0 0.0 0.1 Red-backed Vole 1.0 0.5 0.5 0.3 Meadow Vole 0.9 1.5 eS 1.6 White-footed Mouse 0.6 0.1 0.6 0.5 Deer Mouse 0.6 0.1 0.6 0.1 Meadow Jumping Mouse 0.6 0.1 0.0 0.3 TOTAL 12.8 14.3 13:5 ed SPECIES DIVERSITY 1.68 1.36 1.45 1.30
Note: 562 individuals were captured in 5128 trap-nights (11% capture efficiency).
Acknowledgments
F. Lavender, C. Stevens, and A. Waite assisted with the fieldwork. Logistic support was provided by the Nova Scotia Department of Lands and Forests. This research was supported by an operating grant to B.F. from the Natural Sciences and Engineering Research Council of Canada.
Literature Cited
Anonymous. 1984. Pesticide background statements. Volume |. Herbicides. U.S.D.A. Forest Service, Agricultural Handbook Number 633. Washington, D.C.
Beaver, D. L. 1976. Avian populations in herbicide- treated brush fields. Auk 93: 543-553.
Borrecco, J.E., H.C. Black, and E.F. Hooven. 1979. Response of small mammals to herbicide-induced habitat changes. Northwest Science 53: 97-106.
Chatfield, C. 1985. The initial examination of data. Journal Royal Statistical Society 148: 214-253.
Franzreb, K. E. 1978. Tree species used by birds in logged and unlogged mixed coniferous forest. Wilson Bulletin 90: 221-238.
Freedman, B. 1982. An overview of the environmental impacts of forestry, with particular reference to the Atlantic Provinces. School for Resource and Environmental Studies, Dalhousie University. Halifax, Nova Scotia.
Freedman, B., C. Beauchamp, I. A. McLaren, and S. I. Tingley. 1981. Forestry management practices and populations of breeding birds in a hardwood forest in Nova Scotia. Canadian Field-Naturalist 95: 307-311.
Greig-Smith, P. 1964. Quantitative plant ecology. Butterworths. London, England.
Hurlbert, S. 1984. Pseudoreplication and the design of ecological field experiments. Ecological Monographs 54: 187-211.
IBCC. 1970. International Bird Census Committee recommendations for an international standard for a mapping method in bird census work. Audubon Field Notes 24: 722-726.
Kenaga, E. E. 1975. The evaluation of the safety of 2,4,5-T to birds in areas treated for vegetation control. Residue Reviews 59: 1-19.
Kirkland, G.L. 1978. Population and community responses of small mammals to 2,4,5-T. U.S.D.A. Forest Service, Research Note PNW-314. Pacific Northwest Forest and Range Experiment Station. Portland, Oregon.
McArthur, L. B. 1980. The impact of various forest management practices on passerine bird community structure. Ph.D. thesis, West Virginia University. Morgantown, West Virginia.
Morgan, K., and B. Freedman. 1986. Breeding bird communities in a chronosequence of hardwood forest succession in Nova Scotia. Canadian Field-Naturalist 100: 506-519.
Morrison, M. L., and E. C. Meslow. 1983. Impacts of forest herbicides on wildlife: toxicity and habitat alteration. Transactions of the 48th North American Wildlife and Natural Resources Conference. Pages 175-185.
Morrison, M. L., and E. C. Meslow. 1984a. Response
of avian communities to herbicide-induced vegetation
changes. Journal of Wildlife Management 48: 14-22.
Morrison, M. L., and E. C. Meslow. 1984b. Effects of
the herbicide glyphosate on bird community structure,
western Oregon. Forest Science 30: 95-106.
Newton, M., and L. A. Norris. 1976. Evaluating short and long-term effects of herbicides on nontarget forest and range biota. Down to Earth 32: 18-26.
1988
Ott, L. 1977. Anintroduction to statistical methods and data analysis. Duxbury Press, North Sciviate, Massachusetts.
Robbins, C. S. 1978. Census techniques for forest birds. Pages 142-163 in Management of southern forests for nongame birds. U.S.D.A. Forest Service, General Technical Report SE-14. Southeastern Forest Experiment Station. Asheville, North Carolina.
Savidge, J. A. 1978. Wildlife in a herbicide-treated Jeffrey Pine plantation in eastern California. Journal of Forestry 76: 476-478.
Shannon, C. E., and W. Weaver. 1949. The mathemati- cal theory of communication. University of Illinois Press, Urbana, Illinois.
Sokal, R., and F. Rohlf. 1969. Biometry. W. H. Freeman and Company, San Francisco, California. Spencer, S. R., and G. W. Barrett. 1980. Meadow vole (Microtus pennsylvanicus) population responses to vegetational changes resulting from a 2,4-D application. American Midland Naturalist 103: 32-46.
Ware, G. W. 1980. Effects of pesticides on nontarget organisms. Residue Reviews 76: 173-201.
Way, J.M. 1969. Toxicity and hazards to man, domestic animals, and wildlife from some commonly used auxin herbicides. Residue Reviews 26: 37-62.
Welsh, D. A. 1981. Impact on bird populations of harvesting the boreal mixedwood forest. Pages 155- 167 in Boreal mixedwood symposium 0-P-9. Canadian Forestry Service, Sault Ste. Marie, Ontario.
Williams, A. B. 1936. The composition and dynamics of a beech-maple climax community. Ecological Monographs‘6: 317-408.
Received 25 July 1985 Accepted 2 December 1987
FREEDMAN, POIRIER, MORASH, AND SCOTT: EFFECTS OF 2,4,5-T 11
Appendix 1. Binomials of organisms mentioned in the manuscript.
A) PLANTS
Black Spruce, Picea mariana (Mill.) BSP.
Norway Spruce, Picea abies L.
birches, Betula L. spp.; esp. White Birch, B. papyrifera
Marsh raspberries, Rubus L. spp.; esp. Red Raspberry, R. strigosus Michx.
Red Maple, Acer rubrum L.
B) MAMMALS
Masked Shrew, Sorex cinereus Kerr
Smoky Shrew, Sorex fumeus Miller
Arctic Shrew, Sorex arcticus Kerr
Pygmy Shrew, Microsorex hoyi (Baird)
Short-tailed Shrew, Blarina brevicauda (Say)
Deer Mouse, Peromyscus maniculatus (Wagner)
White-footed Mouse, Peromyscus leucopus (Rafinesque)
Red-backed Vole, Clethrionomys gapperi (Vigors)
Bog Lemming, Synaptomys cooperi Baird
Meadow Vole, Microtus pennsylvanicus (Ord)
Meadow Jumping Mouse, Zapus hudsonicus (Zimmermann)
Woodland Jumping Mouse, Napaeozapus insignis (Miller)
C) BirRDsS
American Woodcock, Philohela minor (Gmelin)
Ruby-throated Hummingbird, Archilochus colubris (Linnaeus)
Olive-sided Flycatcher, Contopus borealis (Swainson)
Alder Flycatcher, Empidonax alnorum Brewster
American Robin, Turdus migratorius Linnaeus
Mourning Warbler, Oporornis philadelphia (Wilson)
Common Yellowthroat, Geothlypis trichas (Linnaeus)
Song Sparrow, Melospiza melodia (Wilson)
Lincoln’s Sparrow, Melospiza lincolnii (Audubon)
White-Throated Sparrow, Zonotrichia albicollis (Gmelin)
Northern Junco, Junco hyemalis (Linnaeus)
American Goldfinch, Spinus tristis (Linnaeus)
Migratory Patterns of the Wapiti, Cervus elaphus, in Banff
National Park, Alberta
L. E. MORGANTINI and R. J. HUDSON
Department of Animal Science, University of Alberta, Edmonton, Alberta T6G 2P5
Morgantini, L. E., and R. J. Hudson. 1988. Migratory patterns of the Wapiti, Cervus elaphus, in Banff National Park, Alberta. Canadian Field—Naturalist 102(1): 12-19.
Migratory behavior of the Wapiti along the eastern boundary of Banff National Park was studied over a three-year period. Most of the Wapiti population which summers on alpine ranges in the northern half of Banff National Park was found to migrate on to three winter ranges outside the park. Minimum distances between summer and winter ranges varied from 26 to 68 km. The entire yearly migratory cycle consisted of 52 to 138 km of mountain travel and a minimum total elevation change of 2000 m. Wapiti exhibited predictable movement patterns and a tendency to return to the same ranges each year. This migratory pattern is interpreted as a vestige of the dispersal of Wapiti from the Bow River valley in Banff National Park after their re-introduction in the years 1917 and 1920.
Key Words: Wapiti, Elk, Cervus elaphus, migration, Banff National Park, Alberta.
In mountainous regions of North America, most Wapiti, Cervus elaphus, populations migrate between seasonal ranges (Altmann 1952; Craigh- ead et al. 1972). Migratory behavior may vary from local movements of 2 to 4km (Anderson 1958; Dalke et al. 1965) to migrations of more than 100 km (Skinner 1925; Anderson 1958). Within the same population, migratory and non-migratory behavior may be present (Martinka 1969; Boyd 1970). Some herds exhibit spring and summer migrations, while others remain on winter ranges until early summer (Knight 1970).
Within the diversity of migratory behavioral patterns, Wapiti show considerable fidelity to seasonal ranges year after year (Murie 1951; Altmann 1952; Anderson 1958; Knight 1970). In the Yellowstone National Park area, most animals return to different winter ranges, even though mingling occurs on summer ranges, and thus maintain distinct herd entities (Craighead et al. 1972). The same migratory routes may be used each year (Altmann 1952; Anderson 1958).
While migratory behavior of Wapiti popula- tions in the United States has been well documented, knowledge of migrations in the Canadian Rocky Mountains is limited to observations of seasonal range use and distribu- tions obtained incidentally during studies of herbivore interactions, population dynamics, or Wolf, Canis lapus, predation (Cowan 1950; Flook 1970; Carbyn 1974; Stelfox 1976).
The objective of this study was to determine Wapiti distribution and movements in the Panther, Red Deer and Clearwater rivers region and to assess whether present migratory patterns
12
can be related to the dispersal of Wapiti from the Bow River valley in Banff National Park after their re-introduction in the years 1917 and 1920.
Study Area
The study was conducted along the Canadian Rocky Mountains, in west-central Alberta (Morgantini 1988). The study area includes over 4000 km? of mountain terrain, 80% of which is within the boundary of Banff National Park. It includes four major river valleys: the Red Deer, Clearwater, Panther and Pipestone rivers. Elevation ranges between 1500 m on valley floors and 2600 m on alpine sites.
Three ecoregions are identified (Holland and Coen 1985; Stelfox 1981). The alpine ecoregion occurs at elevations above 2300m, and is characterized by the absence of trees and presence of cold harsh climatic conditions. Plant communi- ties are those typical of alpine heath tundra. The subalpine ecoregion ranges between 1600 m and 2300 m. Forests are dominated by Engelman Spruce (Picea engelmanii) and Subalpine Fir (Abies lasiocarpa). White Spruce (Picea glauca) and Lodgepole Pine (Pinus contorta) are found at lower elevations.
A third ecoregion, Montane, is restricted to alluvial meadows along the main river valleys 5 to 10 km outside Banff National Park: the Ya Ha Tinda Ranch along the Red Deer River, the “Corners” along the Panther River, and Harrison Flats along the Clearwater River. Due to the sheltering effect of the surrounding mountains, these meadows have mild winters. The vegeta- tional mosaic is characterized by rolling, Rough
1988
Fescue (Festuca scabrella) grasslands, which are kept largely snow free by strong westerly winds.
Methods
Most of the data on Wapiti movements and distribution was collected during extensive field surveys carried out on foot or on horseback throughout the study region. The study extended from December 1976 to November 1979. Every month an average of 15 days was spent in the field, for a total of 537 field-days. Field work consisted in locating Wapiti herds and in continuously monitoring, from day to day, their movements throughout the field period. Animal movements and distribution were photo-documented in the field and later transferred to 1:21 000 scale aerial photographs.
The identification of herds and the location of seasonal ranges were initially facilitated by the presence of 11 cows (six years old and older) equipped with neck collars. These animals had been collared in 1971 and 1973 by the Alberta Fish and Wildlife Division as part of an uncompleted study on Wapiti movements in the Ya Ha Tinda Ranch area along the Red Deer River (Rosin and Paulsen 1973. Alberta Fish and Wildlife unpub- lished report. 14 pp.). In order to further facilitate the identification of different herds and to follow the animals during their long seasonal movements, an additional four cows (five years old and older), four yearlings (two males and two females) and three calves (one male and two females) were trapped and radio-collared.
Ground surveys were complemented by aerial surveys. During the summers of 1977 and 1978, seven aerial surveys (27.4 hours) were carried out with a Bell 206 helicopter. The study area was surveyed by flying all the major and secondary valleys in a pattern to allow maximum coverage of alpine-subalpine ranges and of meadows and forests in the region. In the winters of 1976-77, 1977-78, and 1978-79, aerial surveys were conducted by the Banff National Park Warden Service and by Alberta Fish and Wildlife Division inside and outside the National Park, respectively (Banff National Park files; Alberta Fish and Wildlife files).
Results
During the three-year study 652 groups, comprising a total of 14758 animals, were observed (Table 1). Collared animals were present in 199 observations for a total of 248 sightings. Most observations were recorded in the Red Deer River watershed, which supported an estimated
MORGANTINI AND HUDSON: MIGRATORY PATTERNS OF WAPITI 13
population of 600 animals. Wapiti along the Little Pipestone and the Pipestone rivers were found to be part of the herd that winters along the Red Deer River. The Clearwater and the Panther river watersheds each supported about 200 individuals. The entire yearly migratory cycle in the region is summarized in Figure |. It involved 52-138 km of mountain travel (Table 2), and a minimum total elevation change of 2000 m.
Winter distribution (December- April)
Between December and April Wapiti were mostly found outside Banff National Park. During this study, Wapiti did not winter in the Pipestone River drainage. Within the Red Deer River drainage, large cow-calf-juvenile herds (50-400 individuals) wintered on the open grassland in the Ya Ha Tinda Ranch region, while smaller cow herds and bulls were observed in the surrounding areas. Only a few Wapiti remained through the winter in Banff National Park. Along the Clearwater River Wapiti made extensive use of several open meadows and south-facing slopes 2- 12 km outside the National Park boundary. The use of ranges in the National Park was limited.
Within the Panther River watershed, Wapiti wintered in significant numbers both inside and outside Banff National Park. Outside the National Park, most observations were recorded in the “Corners” region. In Banff National Park Wapiti wintered throughout the Panther River valley. A large number of bulls and a few cows were found wintering on higher subalpine meadows along the Panther-Red Deer river divide. Movements across the boundary of the National Park were also detected. However, it could not be determined whether they reflected normal movement patterns or whether they were caused by recreational activities (snowmobiling, etc.) in the “Corners” area.
During special winter hunting seasons (January- February 1977 and 1978), Wapiti returned to Banff National Park and heavily used small grassland meadows along the Red Deer River valley. Other herds moved from the “Corners” and from the Ya Ha Tinda Ranch on to surrounding high elevation ranges (Morgantini and Hudson 1985). However, in three to five days following the hunting seasons, the animals re-established their habitual range outside the National Park.
Summer Distribution (July-August)
During July and August 1977, 1978 and 1979, a total of 1417 Wapiti were counted in 118 observations. Collared animals were present in 66 groups.
14 THE CANADIAN FIELD-NATURALIST Vol. 102
one oe Zia vanes = anges,
Miles 5 0 5 5 0 5 Kilometers
=z. Migrations to ~ Summer Ranges
Location of
study wa
Lake Louise Ae
FiGureE |. Wapiti migrations from winter to summer ranges in Banff National Park (1977-1979).
In the summer, Wapiti were mostly found within the boundary of Banff National Park and were widely dispersed over some 1600 km? of mountain terrain. Out of 22 collared animals, 10 were always found to summer in the Pipestone-Lake Louise area, seven in the Red Deer River watershed, one
along the Panther River, and two in the Clearwater River region. Two animals were never located and were presumed dead.
Large herds (30-70 animals) ranged on high subalpine and alpine meadows at the headwaters of several tributary creeks of the Red Deer,
1988
MORGANTINI AND HUDSON: MIGRATORY PATTERNS OF WAPITI 15
TABLE |. Summary of Wapiti observations and estimated population sizes in the study region (1977-1979).
River Estimated Winter Drainage Pop. size (Dec.-Apr.) Red Deer* 600 175 (8144) Clearwater 200 47 (343) Panther 200 38 (916) TOTALS 1000 260 (9403)
No. of observations
Spring Summer Fall (May-June) (July-Aug.) (Sept.-Nov.) 122 80 ay) (154) (994) (751)
31 13 11 (613) (122) (149) 24 24 30 (387) (301) (184) 177 117 98 (2854) (1413) (1084)
*The animals that summer in the Pipestone River were found to be part of the herd that winters in the Red Deer River
region. ( ) = total number of animals counted.
Panther and Clearwater rivers. Approximately 200 to 250 Wapiti, more than 30% of the Ya Ha Tinda Ranch winter herd, summered in the Pipestone- Lake Louise area. However, small herds of cows (1-10 animals) and bulls, and signs of their activity, were found throughout the National Park wherever favorable habitat was available. Occasionally, Wapiti could also be observed along the main river valleys while travelling from or to their summer ranges.
Between 1977 and 1979 less than 10% of the total winter population of the region summered outside Banff National Park. They consisted largely of small herds of cows and calves and isolated bulls. The highest number (30 animals) was observed in the Clearwater watershed.
In the summer of 1977, a herd of 34 animals (5 bulls, 21 cows with two collared individuals and 8 calves) remained on the Ya Ha Tinda Ranch until the third week of July. Following human harassment (hikers and trail riders), the animals moved first on to the surrounding slopes and ridges where they remained until the first week of August. After continuing harassment by 4X4 vehicles, the animals left the region and moved on to alpine ranges, 24 km distant, in Banff National Park.
Spring and Fall Migrations (May-June and September- November)
Wapiti exhibited well-defined seasonal migra- tions between winter and summer ranges. The minimum distance travelled from winter ranges varied from 26 to 68 km (Table 2).
Spring migrations consisted of an initial gradual shift from winter ranges to ranges located farther west along the major river valleys (spring “intermediate” ranges). This movement was later followed by a rapid altitudinal migration to high elevation ranges. The timing of spring migrations showed great variation. Even though some animals were observed leaving their winter range in early May, most moved during the second half of May and in early June.
The location and level of utilization of intermediate ranges depended on the date the animals left their common winter range and on the distance between winter ranges and individual summer ranges. Within the Red Deer watershed, animals that summered close to the Ya Ha Tinda Ranch winter range remained along the Red Deer River valley until early July. In contrast, the segment of the population with summer ranges in the Pipestone-Lake Louise area continued its
TABLE 2. Distance of summer ranges in Banff National Park of 18 Wapiti collared on the Ya Ha Tinda Ranch (1977-
1979). Linear distance (km) N Average Range Su} 21 36.7 17.5-57.5 14.1
Minimum travel distance* (km) Ss) D). 17.9
Average 49.8
Range 26.5-67.5
Note: During the study two animals used different summer ranges.
* Minimum travel distance along river valleys.
16 THE CANADIAN FIELD-NATURALIST
gradual movement westward along the Red Deer and the Pipestone rivers. Some of these animals left the Ya Ha Tinda Ranch in the middle of May and by early June had established their spring “intermediate” ranges (55 km distant) along the Pipestone River. Others never established actual intermediate ranges, but, having left their winter range at a later date, used the entire month to gradually shift on to summer ranges in the upper Pipestone River.
Spring migrations overlapped the calving season (25 May-5 June). Calving was observed through- out the region, both on winter and on intermediate ranges, and it appeared to slow spring movements. Due to the proximity of summer and winter ranges along the Clearwater and the Panther rivers, intermediate ranges and winter ranges coincided for those few Wapiti that spent the summer outside Banff National Park.
Fall migrations towards winter ranges occurred between September and November. Initially, they consisted of a rapid shift on to lower elevation ranges along the major river valleys (intermediate ranges). This movement coincided with early snowfall (Table 3). Temporary returns to higher elevation ranges during warm fall weather, one to two days following snowfall, were also observed. By the end of September, most of the Wapiti population was found on intermediate ranges. During the month of October, instead of gradually shifting onto winter ranges, Wapiti concentrated on intermediate ranges just inside Banff National Park. These sites acted as major “staging areas” for
Vol. 102
large herds before their late fall-early winter (15 November-15 December) movements onto winter ranges outside the National Park.
Along the Clearwater River, Wapiti tended to move outside the National Park earlier.
Range fidelity
Throughout the study region, Wapiti seemed to exhibit predictable movement patterns between ranges and a general tendency to return to the same ranges each year. The relatively low percentage of the total population collared (4% of the Red Deer River herd) does not allow a quantitative assessment of range fidelity. Nonetheless, the frequency of return of collared animals to specific summer ranges and their return to a common winter range are a clear indication of a well- developed habitual behavior (Table 4). This behavior was further manifest in the use of the same migratory trails connecting different seasonal ranges.
Between 1977 and 1979 11 out of 18 collared Wapiti returned to the same summer ranges. For eight of these animals, spring movements to their traditional ranges involved more than 60 km of travel across trails which allowed access to summer ranges closer to their wintering grounds.
In winter, 16 out of 18 collared Wapiti returned to their common winter range for four successive years. Two males, trapped and collared as 10- month-olds in March 1978, returned to the Ya Ha Tinda Ranch only the following winter. After moving to their habitual summer ranges in 1979,
TABLE 3. Elevational distribution of Wapiti before, during, and after early snowfalls in
1977 and 1978.
Elevation (m)
Total snowfall No. of
Date ( <25)S4D!) at 1650 m observations 1977
9-10 September 2181 + 253a 0.0 6 (105) 13-16 September 1746 + 102be 15.0* 1393) 17-26 September 2028 + 192ad 0.0 19 (131) 27-30 September 1741 + 18be 6.4** 15 (148) 1978
11-17 September 2030 + 194 0.0 8 (66) 18-19 September 1877 + 101 13.0* 10 (78) 20-28 September 2077 + 142 0.0 11 (134)
Values within columns followed by different letters are significantly different at P< 0.05
(T-test).
*Snowfall over the entire period.
**Snowfall on 27 September.
( ) Total number of animals observed.
1988
they were not relocated. Both animals were shot by hunters outside the National Park — one in the fall of 1980 along the Clearwater River, the other in the fall of 1981 between the Panther and the Red Deer rivers.
Based on these data and on field observations, the Wapiti population in the region can be subdivided into three separate herds, each associated with a major river valley (i.e. Panther, Red Deer and Clearwater). Some mingling between herds on summer ranges was detected. In early fall the great majority of the animals that had shifted to another drainage were observed to move back to their habitual ranges. Further mingling between the Panther River and the Red Deer River herds appeared to occur in the late fall and winter, especially during intensive hunting harassment.
Movements of Wapiti from and to areas adjacent to the study region were observed. While no evidence was ever found of movements from the Red Deer herd across Pipestone Pass into the Sifleur River, travel through Clearwater Pass into the Sifleur River valley was detected. Mingling also occurs between the Red Deer River and the Bow River herds as indicated by the presence in 1977 of two collared Wapiti from Kootenay National Park in the Molar Creek area and on the Ya Ha Tinda Ranch. During 1978, two more Wapiti not collared during this study were found within the Red Deer River herd. They had been trapped in Jasper National Park and released by the Alberta Fish and Wildlife Division in 1974 along the Red Deer River east of the Front Ranges 25 km from the Ya Ha Tinda Ranch.
TABLE 4. Number of collared animals that returned to the same ranges in successive years (1977-1980).
No. of
successive Number of animals that returned years to the same ranges Summer ranges Winter ranges
0 2 1 3 2 2 2 0 3 11 0 4 * 16
Note: The table does not include two animals that were
never found and presumably died during the first winter and 2 animals that were shot by hunters outside Banff National Park. *Systematic monitoring of summer ranges was carried out for only three years.
MORGANTINI AND HUDSON: MIGRATORY PATTERNS OF WAPITI 17
Discussion
The distribution of Wapiti observed during this study is consistent with observations recorded by Banff National Park wardens between 1949 and 1976 (Banff National Park files). The stability of regional movements and distribution is further confirmed by 11 ground surveys carried out by L.E.M. between 1980 and 1983, and by more recent Banff National Park wildlife surveys (Banff National Park files). Seasonal distribution and movement patterns in the study region are also consistent with knowledge of Wapiti behavior in mountain environments (Adams 1982).
The winter concentration of Wapiti on the Ya Ha Tinda Ranch reflects the availability of open winter ranges. The area represents less than 4% of the entire region and, due to its mild winter weather and mostly snow free conditions, is ideal winter range. In comparison, the “Corners” area along the Panther River and Harrison Flats and its adjacent south-facing slopes along the Clearwater River offer significantly less winter range and indeed support smaller numbers of animals (Morgantini and Bruns 1984).
In this typical northern mountain environment, Wapiti appear to respond to seasonal environmen- tal changes with shifts from low elevation winter and intermediate ranges (1500-1600 m) to high elevation summer ranges (2100-2400 m), and vice versa (Morgantini 1988). In the study region the location of summer ranges and range fidelity indicate that tradition (learned behavior) plays a major role in shaping Wapiti distribution. The migration of approximately 200 animals from the Ya Ha Tinda Ranch to summer ranges north of Lake Louise and in the upper Pipestone River cannot be explained solely in terms of adaptation to seasonal environmental conditions. Migrating to summer ranges north of Lake Louise, for instance, involves an initial travel of 40 km along the Red Deer River valley to an elevation of 2100 m, then a downward movement of 25 km to an elevation of 1700 m and a final 2-3 km climb towards high elevation meadows. All along the route there are well-established trails to summer ranges significantly closer to the Ya Ha Tinda Ranch.
There are two possible explanations for this long-range migratory behavior. Yearly westward movements may reflect the original gradual dispersal and colonization of the region by animals wintering on acommon winter range. Conversely, the entire migratory pattern may be a vestige of the original dispersal of Wapiti from the Bow River Valley. Historical evidence and this study tend to support the latter interpretation.
18 THE CANADIAN FIELD-NATURALIST
In the early 1900s, Wapiti had almost disappeared from the Canadian Rocky Mountains as a result of severe winters and indiscriminate hunting by white and native people (Millar 1915; Stelfox 1964; Soper 1970). The present population is believed to have originated from the release in Banff National Park, mostly along the Bow River valley, of between 245 and 251 animals from Wyoming between 1917 and 1920 (Lloyd 1927; Green 1946). It has been suggested that this introduced stock interbred with the few remnant native Wapiti and with Wapiti moving into Banff National Park from British Columbia (Holroyd and Van Tighem 1983: 416). The Wapiti population rapidly increased and colonized adjacent valleys. Historical records show a gradual dispersal southeast into the Cascade River valley (1925), along the Panther River (1927) and Snow Creek (1931) (Banff National Park files). These records and the existence of well-established trails from the Panther and the Dormer rivers into the Wigmore-Cascade area suggest that Wapiti dispersing from the Bow River valley reached the Panther River by following the Cascade River and then Wigmore Creek. The shift on to the Red Deer River valley may have occurred in the summer across Snow Creek or in winter through the lower Dog Rib Creek.
The presence of Wapiti from Kootenay National Park in the Red Deer River herd and the well- established migratory pattern from the Pipestone to the Red Deer River point to a second dispersal route. In the north-eastern section, Wapiti were first reported in 1930 along Mosquito Creek, in 1936 along the Saskatchewan River and in 1942 along the Pipestone River (Banff National Park files). It is here suggested that at the time some Wapiti from summer ranges in the upper Pipestone River travelled east along Little Pipestone River into the upper Red Deer River valley. The winter range on the Ya Ha Tinda Ranch may have been encountered by chance during downward move- ments along the Red Deer River valley. This movement may have been facilitated by the presence of a few Wapiti remnant from the native population.
In comparison with the Panther and the Red Deer rivers, access to the Clearwater River from the Bow River Valley is limited. Dispersal and colonization may have occurred from the Sifleur River or from the lower Red Deer River valley.
The dispersal of Wapiti from their site of re- introduction and the present day migratory pattern in the region conform to the theory of seasonal return migrations within familiar areas as developed by Baker (1978, 1982). The familiar area is defined as “the portion of the lifetime range from
Vol. 102
any point in which an animal is capable of finding its way to any other point” (Baker 1978:378). It is initially established during the course of successive exploratory movements (Baker 1978) and is largely maintained or extended through social communi- cation within family units (cow-calf; Murie 1951) or through association of inexperienced with experienced animals.
In the study region, exploratory and/or dispersal (Horn 1978) movements may have gradually led to the establishment of fairly separate familiar areas. The apparent mingling between herds inhabiting different watersheds and the occasional shifting of animals from one herd to the other may reflect a continuing process of extension of the familiar area of the individuals involved. The movements of herds from summer ranges outside Banff National Park on to range in the park following human harassment suggest that at least some Wapiti are familiar with a region signifi- cantly larger than the one they are inhabiting.
In conclusion, Wapiti, in their seasonal migrations in the study region, appear to use the same routes that the species followed during its dispersal from the Bow River valley after its re- introduction in 1917 and 1920. This dispersal and the establishment of regular seasonal movements may have been facilitated by the presence of remnant Wapiti in the region.
Acknowledgments
We acknowledge the cooperation of the Alberta Fish and Wildlife Division for financial and logistic support, Parks Canada for providing radiotelemetry equipment, and the personnel of the Ya Ha Tinda Ranch for their cooperation. The Banff National Park Warden Service provided invaluable field support. This study was particu- larly made possible by the interest and cooperation of many individuals who helped to ensure its success. We especially thank the following individuals: Slim Haugen, Earl Hays and the late Gordon Patterson of the Ya Ha Tinda Ranch, Perry Jacobson, Dale Loewen, Gordon Antoniak and John Wackerle of Banff National Park, Keith Baker, formerly of Parks Canada, and Eldon Bruns of the Alberta Fish and Wildlife Division.
Literature Cited
Adams, A. W. 1982. Migration. In Elk of North America. Ecology and management. Edited by J. W. Thomas and D.E. Toweill. Stackpole Co., Harrisburg, Pennsylvania. 698 pp.
Altmann, M. 1952. Social behavior of elk, Cervus canadensis nelsonii, in the Jackson Hole area of Wyoming. Behavior 4(2): 116-143.
1988
Anderson, C.C. 1958. The elk of Jackson Hole. Bulletin 10. Wyoming Game and Fish Commission, Laramie. 184 pp.
Baker, R. R. 1978. The evolutionary ecology of animal migration. Hodder and Stoughton. 1012 pp.
Baker, R. R. 1982. Migration. Paths through time and space. Hodder and Stoughton. 248 pp.
Boyd, R. J. 1970. Elk of the White River Plateau, Colorado. Technical Bulletin 25. Colorado Division of Game, Fish and Parks, Denver. 126 pp.
Carbyn, L. N. 1974. Wolf predation and behavioural interactions with elk, and other ungulates in an area of high prey diversity. Canadian Wildlife Service Report. Ottawa. 233 pp.
Cowan, I. McT. 1950. Some vital statistics of big game on overstocked mountain range. Transactions of the North American Wildlife Conference 15: 581-588.
Craighead, J.J., G. Atwell, and B.W. O’Gara. 1972. Elk migrations in and near Yellow- stone National Park. Wildlife Monograph No. 29. The Wildlife Society, Washington D.C. 48 pp.
Dalke, P. D., R. D. Beeman, F. J. Kindel, R. J. Robel, and T. R. Williams. 1965. Seasonal movements of elk in the Selway River Drainage, Idaho. Journal of Wildlife Management. 29(2): 333-338.
Flook, D. R. 1970. A study of sex differential in the survival of Wapiti. Canadian Wildlife Service Report, Service Report, Series No. 11. Ottawa. 71 pp.
Green, H. U. 1946. The elk of Banff National Park. Banff National Park Report. Banff, Alberta. 32 pp. Holland, W.D., and G.M. Coen. 1983. Ecological (biophysical) land classification of Banff and Jasper National Parks. Volume 1: summary. Alberta Institute
of Pedology. Publication No. M-83-2.
Holroyd, G. L., and K. J. Van Tighem. 1983. Ecological (biophysical) land classification of Banff and Jasper National Parks. Volume III: The wildlife inventory. Canadian Wildlife Service Publication. Edmonton, Alberta. 691 pp.
Horn, H. S. 1978. Optimal tactics of reproduction and life-history. Pp. 411-429 in Behavioural ecology. An evolutionary approach. Edited by J. R. Krebs and N. B. Davies. Blackwell Scientific Publications, Oxford. 494 pp.
MORGANTINI AND HUDSON: MIGRATORY PATTERNS OF WAPITI 19
Knight, R. R. 1970. The Sun River elk herd. Wildlife Monograph No. 23. The Wildlife Society, Washing- ton, D.C. 66 pp.
Lloyd, H. 1927. Transfer of elk for re-stocking. Canadian Field-Naturalist 41(6): 126-127.
Martinka, C. J. 1969. Population ecology of summer resident elk in Jackson Hole, Wyoming. Journal of Wildlife Management. 33(3): 465-481.
Millar, W. N. 1915. Game preservation in the Rocky Mountains Forest Reserve. Department of the Interior, Canada. Forestry Branch Bulletin No. 51. 69 pp.
Morgantini, L. E. 1988. Adaptive strategies of Wapiti in the Canadian Rocky Mountains. Ph.D. thesis, University of Alberta, Edmonton, Alberta. 196 pp.
Morgantini, L. E., and E. Bruns. 1984. The assessment of three elk winter ranges in Alberta: an appraisal. Pp. 106-116 in Western States and Provinces Elk Workshop Proceedings. Edited by R. W. Nelson. Alberta Fish and Wildlife Division. Edmonton, Alberta. 218 pp.
Morgantini, L. E., and R. J. Hudson. 1985. Changes in diets of wapiti during a hunting season. Journal of Range Management 38: 77-79.
Murie, O. J. 1951. The elk of North America. Stackpole Co., Harrisburg, Pennsylvania. 376 pp.
Skinner, M. P. 1925. Migration routes in Yellowstone Park. Journal of Mammalogy 6: 184-192.
Soper, J.D. 1970. The mammals of Jasper National Park, Alberta. Canadian Wildlife Service No. 10. Ottawa. 80 pp.
Stelfox, J. G. 1964. Elk in northeast Alberta. Land, Forest, Wildlife 6(5): 14-23.
Stelfox, J. G. 1976. Range ecology of Rocky Mountain Bighorn Sheep. Canadian Wildlife Service Report Series No. 39. Ottawa. 50 pp.
Stelfox, H. A. 1981. Ecological land classification: Red Deer-James River. Resource Evaluation and Planning Division, Alberta Energy and Natural Resources, Report No. T/11-No.8. 112 pp.
Received 10 October 1985 Accepted 23 April 1987
Breeding Performance of Black-legged Kittiwakes, Rissa tridactyla, at a Small, Expanding Colony in Labrador
T. R. BIRKHEAD! and D. N. NETTLESHIP?
'Department of Zoology, University of Sheffield, Sheffield, United Kingdom S10 2TN 2Canadian Wildlife Service, Bedford Institute of Oceanography, P.O. Box 1006, Dartmouth, Nova Scotia B2Y 4A2
[ Address for reprint requests]
Birkhead, T. R., and D. N. Nettleship. 1988. Breeding performance of Black-legged Kittiwakes, Rissa tridactyla, ata small, expanding colony in Labrador. Canadian Field—Naturalist 102(1): 20-24.
The breeding performance of Black-legged Kittiwakes, Rissa tridactyla, was studied at the Gannet Islands, Labrador, in 1981 to 1985. In 1981-1983 median clutch size varied between |.7 and 2.0 eggs and productivity averaged 1.1 to 1.3 chicks per breeding pair. In 1984 and 1985 however, very few pairs (10%) attempted to breed. In 1985 at least, this was part of a more general geographic trend in south-eastern Canada. Breeding by kittiwakes at the Gannet Islands was first recorded in 1972; between that date and 1984 the breeding population increased from 16 to 119 pairs.
Key Words: Black-legged Kittiwake, Rissa tridactyla, breeding distribution, breeding performance, Labrador.
The Black-legged Kittiwake, Rissa tridactyla, has a disjunct breeding distribution in eastern North America, with most of the population occurring in either the northern (i.e., SE Baffin Island, north to Jones Sound) or southern (Newfoundland and the Gulf of St. Lawrence) parts of the species’ range (Brown et al. 1975). Kittiwakes probably started to breed in Labrador only recently; the first breeding record was of 16 occupied nests on Outer Gannet Island (54°00’N, 56° 32’W) in 1972 (Nettleship and Lock 1974). Six years later (1978) there were 48 occupied nests on Outer Gannet and three nests on one (GC4: see Methods for details) of the nearby Gannet Islands (53° 56’N, 56° 32’W). In 1979 there were 40 nests on Outer Gannet and 10 on the same Gannet Island (D. N. Nettleship, unpublished).
During the breeding seasons of 1981-1983 we conducted studies of seabirds (mainly alcids) breeding at the Gannet Islands. Here we present information on the breeding performance of Black-legged Kittiwakes at the Gannet Islands, together with some further information on the kittiwakes on Outer Gannet Island. Short visits were made to the Gannet Islands and Outer Gannet in 1984 and 1985; observations from those periods are also reported.
Methods
Daily observations of kittiwake nests were made from a blind located about 50 m from the breeding area, between late May (1982 and 1983) or early June (1981) and late August. All kittiwake nests were located in a cove on one island (GC4: see below), and in 1981 and 1982 all nests could be
20
observed from the blind. In 1983, however, 10 nests were built in areas not visible from the blind. Using methods similar to those described by Birkhead and Nettleship (1980) we recorded the timing of breeding (date of first egg of each clutch), clutch size, and breeding success (number of young fledged from each nest). A chick was considered to have fledged once it made its first flight at age 35 days or more (see Swartz 1966). In 1984 and 1985 visits were made on 2-15 July and 3-15 July, respectively, during which periods nests were counted and their contents recorded at the Gannet Islands. At Outer Gannet also, nests were counted in both years. Code designations used to identify the six islands comprising the Gannet Islands archipelago (e.g., GC1l, GC2, GC3, GC4, etc.) are those described in Birkhead and Nettleship (1987).
Results
During the study the number of kittiwakes breeding at the Gannet Islands (not including Outer Gannet) continued to increase (Table 1), with 26 nests (18 clutches) in 1981, 37 nests (31 clutches) in 1982, and 52 nests (at least 42 clutches) in 1983, 63 nests in 1984 and at least 58 in 1985. Up to 1983 all prospecting kittiwakes were seen in the vicinity of the breeding colony on GC4, but in 1983 prospectors were seen on other islands and three nests were built on GCI. We did not check the contents of these nests, but judging from the birds’ behaviour we assume that no eggs were laid. In 1984 there were 5 nests on GC]; no count of nests was made on this island in 1985. At Outer Gannet the population apparently stabilized, with 57 nests in 1983, 56 in 1984, but in 1985 only 40 nests were
1988
BIRKHEAD AND NETTLESHIP: BLACK-LEGGED KITTIWAKES IN LABRADOR 21
TABLE |. Changes in the numbers of Black-legged Kittiwakes (pairs of nest-site holders) at the Gannet Islands and Outer Gannet Island since they were first discovered breeding there in 1972.
Gannet Islands
Total % nests Year nests with eggs 1972 0 0 1978 3 ? 1979 10 ? 1981 26 69 1982 37 84 1983 52 81 1984 63 10 1985 58 5
'No data available.
Outer Gannet Island
Total % nests Total nests with eggs nests 16 ? 16 48 ? 51 40 50
zB : F Sy) z 109 56 2 119 40 u 982
2Values may be low; 5 nests recorded on GCI in 1984 were not examined in 1985.
counted. An area on Outer Gannet which had contained 16 nests in 1984 held only the remains of nests in 1985.
Kittiwakes were present at the Gannet Islands colony each year when observations began. First and median egg dates were 20 and 21 June in 1981 (N = 18), 15 and 18 June (N = 31), and 4 and 8 June in 1983 (N = 42). The median laying date advanced by 13 days over these three years. The differences in laying dates between years were not related in any obvious way to environmental conditions. In both 1981 and 1983 air temperatures and the timing of ice break-up were similar, whereas in 1982 (and 1984 and 1985: see below) temperatures were relatively low and ice break-up late [early June] (Birkhead and Nettleship 1987).
Mean clutch sizes (and the numbers of 1-, 2- and 3- egg clutches) were 1.67 (6, 12, 0; N = 18) in 1981, 1.87 (6, 23, 2; N = 31) in 1982, 2.02 (4, 33, 5; N = 42) in 1983. Sample sizes are too small for detailed Statistical analysis, but a comparison of the proportion of l-egg clutches versus 2- and 3-egg clutches (combined) showed no significant differences during 1981-1983 (x2= 5.03, 2d.f., NS). However, over that three-year period the proportion of l-egg clutches decreased (33%, 19% and 9% in 1981-1983, respectively), and the proportion of 3-egg clutches increased (0%, 6% and 12%, respectively). The overall increase in mean clutch size over those three years was probably related either to the timing of breeding, the change in age-structure of the population (cf. Coulson and Thomas 1985), or both. In 1984, 52 (90%) of the 58 nests examined on GC4 contained no eggs at a date when in previous years all birds would have laid. Four nests contained | egg, and 2
had 2 eggs. Similarly, in 1985, 55 (95%) of the 58 nests examined had no eggs, and 3 (5%) contained a single egg. As observations were limited we cannot exclude the possibility that most birds had laid and lost their eggs by the time observations were made, but it seems more likely that most birds failed to lay at all (see below).
In 1981-1983 the mean intervals between the laying of first and second eggs of 2- and 3-egg clutches did not differ significantly between years (1981, 2.43d + 0.55 S.D., N = 7; 1982, 2.66 + 1.2, N = 9; 1983, 2.29 + 0.08, N = 28; Fy 4, = 0.7, NS). Nor was there any inter-year difference in mean incubation periods (1981, 27.3d + 0.9S.D., N = 7; 1982°275 22 le nIN = 13; 1983, 27.1 22 1025 N= 16: F, 33 = 0.4, NS).
Breeding performance of birds on GC4 for each year is summarized in Table 2. In general pairs laying larger clutches produced the greatest mean number of fledged chicks. Considering all clutch sizes, productivity (fledglings/breeding pair) differed rather little between years in 1981-1983 (Table 2), but the proportion of pairs rearing at least one chick to fledging was significantly lower in 1983 [26 out of 42 pairs (62%)] than in either 1981 [16/18 (89%)] or 1982 [26/31 (84%)] (x? = 6.9, 2 dete, (P'<0}05)) “The! reason for reduced productivity in 1983 is not clear, but occurred as a result of high egg loss. In June 1983, 10 out of 27 nests with eggs (37%) lost one or two eggs (14 in total) between 18:00 on 9 June and 08:00 on 10 June. Two days later 6 out of 30 nests with eggs lost a total of 8 eggs overnight. We are uncertain about the cause of egg-loss in 1983; we never saw potential avian predators such as Common Raven, Corvus corax, or Great Black-backed Gull, Larus
22
THE CANADIAN FIELD-NATURALIST
Vol. 102
TABLE 2. Population size and breeding performance of Black-legged Kittiwakes at the Gannet Islands (GC4), 1981-
1985. Productivity Clutch Eggs Chicks (fledglings/ size No. of Total hatched fledged breeding
Year (eggs) pairs no. eggs N % N % pair) 1981 0 8 - . - - - -
l 6 6 5 83.3 5 100.0 0.83
B 12 24 21 87.5 19 90.5 1.58
Totals! 26 30 26 86.7 24 92.3 1.33 1982 0 6 - - - - - -
I 6 6 4 66.7 4 100.0 0.67
2 23 46 42 91.3 33 78.6 1.43
3 2 6 6 100.0 2) 83.3 2.50
Total! 31 58 52 89.6 42 80.8 Ess 1983 0 10 - - - - - -
I 4 4 2 50.0 0 0 0
2 33 66 46 69.7 38 82.6 IBN)
3 5 15 13 86.7 9 69.2 1.80
Totals! 42 85 61 71.8 47 77.0 el 1984 0 52 0 - - - - - (2-15 l 4 4 ? ? ? ? - July) 2 2 4 ? ? ? ? ?
Totals! 6 8 2 ? ? ? (0.14)? 1985 0 55 0 - - - - - (3-15 l 3 3 0 0 0 0 0 July) 2, 0 0 - - - - -
Totals! 3 3 0 0 0 0 0
'Totals for pairs of nest-site holders that laid at least one egg. 2Eight eggs still present when observations ceased on 15 July; maximum productivity possible was 1.33 fledglings/
breeding pair.
marinus, near kittiwake nests. In that year, however, a single Short-tailed Weasel, Mustela erminea, was on GC4, and although the majority of nests appeared to be inaccessible it is possible that the weasel was responsible for the egg-loss. We were unable to measure breeding success in 1984 or 1985, but as the proportion of pairs that apparently laid was low, productivity was likely low in those years also.
We did not measure breeding success at Outer Gannet Island, but the following observation may be relevant. In 1983 Outer Gannet was visited on 23 July. We checked the contents of 57 nests; 55 were empty, one had two eggs, and the other had a single newly hatched chick. On the same date most nests on the Gannet Islands had large chicks. The reason for the delayed breeding and very low productivity on Outer Gannet in 1983 is not known. Nest contents were not examined at Outer Gannet in 1984 or 1985.
Kittiwake chicks on GC4 fledged after about 39 days each year in 1981-1983. First fledging was recorded on 18, 16 and 9 August in 1981 to 1983, respectively, and all chicks had fledged by the first week in September each year.
Discussion
The only other study of Black-legged Kittiwake breeding performance in low arctic waters of eastern North America was by Maunder and Threlfall (1972) at Witless Bay, Newfoundland, 800 km south of the Gannet Islands. It is difficult to make meaningful comparisons between these studies because observations were made in different years, and breeding parameters can vary markedly between years (Coulson and Thomas 1985). The timing of breeding was, not unexpect- edly, earlier by about two weeks in Newfoundland (mean laying date: 3 June and 29 May in 1969 and 1970, respectively). Mean clutch sizes were similar:
1988
1.9 at the Gannet Islands (in 1981-1983) and 1.85 in Witless Bay. However, there was a slightly higher proportion of 3-egg clutches in Labrador (7 of 93) than in Witless Bay (4 of 225), a difference that is statistically significant (x? = 4.91, 1d.f., P< 0.05). Inter-egg laying intervals, incubation and chick- rearing periods were similar between the two colonies. Maunder and Threlfall (1972) found egg mortality to be 28% and 27% in two years (1969, 1970), whereas in this study it was 13%, 10% and 28% in three years (1981-1983). Chick mortality was similar in both colonies, Witless Bay: 19% and 26% in the two years, and in Labrador: 21%, 19% and 23% in the three years. In Witless Bay the mean number of chicks fledged per nest was 1.07 and 1.0 in two years, compared with 1.33, 1.35 and 1.12 in 1981 to 1983 in Labrador. As the percentage chick mortality was similar at both colonies, the difference in breeding success and overall productivity is due to the higher egg mortality in Witless Bay. This is turn may be due to different study techniques; Maunder and Threlfall (1972) inspected nests by visiting them, whereas our technique involved observations from a distance and thus no (detectable) human disturbance.
The most noticeable difference apparent between Maunder and Threlfall’s (1972) study and ours was the almost total reproductive failure, or lack of breeding at the Gannet Islands in 1984 and 1985 (and at Outer Gannet in 1983). In 1985, failure to breed also was noted at colonies in Newfoundland, with about 20% non-breeding in Witless Bay (D.N. Nettleship, unpublished). There are no such data from Newfoundland in 1984. It seems likely that low productivity was not restricted to the Gannet Islands in 1985 (and 1984). Hunt et al. (1981) recorded similar phenomena at several Black-legged Kittiwake colonies in Alaska during the 1970s, and Springer et al. (1984) attributed such effects to a reduction in food availability associated with late ice break-up and low temperatures. As ice break-up was late in eastern Newfoundland and Labrador in both 1984 and 1985, the same explanation is plausible there.
The kittiwake population of Outer Gannet Island and the Gannet Islands increased rapidly, from 16 pairs in 1972 to 119 pairs in 1984. Studies of kittiwakes elsewhere indicate that such rapid colony growth, typical of small colonies (Coulson 1983), occurs mainly through immigration, rather than through the colony’s own reproductive output (Porter 1985). Nonetheless, breeding performance of kittiwakes at the Gannet Islands, at least during 1981 to 1983, was similar, or relatively high, compared with other studies (e.g.,
BIRKHEAD AND NETTLESHIP: BLACK-LEGGED KITTIWAKES IN LABRADOR 23
Coulson and White 1958; Barrett and Runde 1980; Galbraith 1983; Coulson and Thomas 1985).
Kittiwakes have long been known to occur as non-breeding summer visitors along the Labrador coast (see Nettleship and Lock 1974), and we saw flocks totalling tens of thousands of birds on several occasions during August. There are several possible explanations for why kittiwakes should recently have started to breed in Labrador. First, Labrador may be an “overflow”, either for the high arctic, or for the Newfoundland and Gulf of St. Lawrence populations. The status of kittiwake populations in the eastern Canadian arctic is unknown (Nettleship 1977), but populations in the southern part of their range (Newfoundland and the Gulf of St. Lawrence) have increased markedly during the last 30 to 40 years (Nettleship 1977, 1980). “Overflow” from this area therefore seems plausible, and the expansion into Labrador may simply be part of the increase in this population.
Another related possibility is that a change in the marine environment may have resulted in Labrador only recently providing suitable breeding conditions for kittiwakes. Further evidence for some sort of marine habitat change is the recent colonization of Labrador and eastern Newfoundland by Northern Fulmars, Fulmarus glacialis, (Nettleship and Lock 1973; Nettleship and Mongomerie 1974; Montevecchi et al. 1978); and Manx Shearwaters, Puffinus puffinus, (Storey and Lien 1985), both surface-feeding species with diets similar to Black-legged Kittiwakes.
Acknowledgments
This research was funded by the Canadian Wildlife Service and is associated with the programme “Studies on northern seabirds”, Seabird Research Unit, CWS, Environment Canada, Dartmouth, Nova Scotia (Report No. 198). We thank R. D. Elliot, S. D. Johnson, A. MacFarlane and E. Verspoor for their help in the field, and J. W. Chardine, A. J. Erskine and J. Porter for constructive comments on the manuscript. We also thank Petro-Canada Ltd. (in particular, Bill and Millie Elson, and Richard Morris) for their excellent logistic support in Goose Bay and Cartwright, Labrador.
Literature Cited
Barrett, R.T., and O.J. Runde. 1980. Growth and survival of nestling Kittiwakes Rissa tridactyla in Norway. Ornis Scandinavica | 1: 228-235.
Birkhead, T.R., and D.N. Nettleship. 1980. Census methods for murres, Uria species: a unified approach. Canadian Wildlife Service Occasional Paper Number 43: 1-25.
24 THE CANADIAN FIELD-NATURALIST
Birkhead, T. R.,and D. N. Nettleship. 1987. Ecological relationships between Common Murres, Uria aalge, and Thick-billed Murres, Uria lomvia, at the Gannet Islands, Labrador. I. Morphometrics and timing of breeding. Canadian Journal Zoology 65: 1621-1629.
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.
Coulson, J.C. 1983. The changing status of the Kittiwake Rissa tridactyla in the British Isles, 1969- 1979. Bird Study 30: 9-16.
Coulson, J. C., and C. S. Thomas. 1985. Changes in the biology of the Kittiwake Rissa tridactyla: a 31-year study of a breeding colony. Journal of Animal Ecology 54: 9-26.
Coulson, J. C., and E. White. 1958. The effects of age on the breeding biology of the Kittiwake Rissa tridactyla. Ibis 100: 40-51.
Galbraith, H. 1983. The diet and feeding ecology of breeding Kittiwakes Rissa tridactyla. Bird Study 30: 109-120.
Hunt, G.L., Jr., Z. Eppley, and W.H. Drury. 1981. Breeding distribution and reproductive biology of marine birds in the eastern Bering Sea. In The Eastern Bering Sea shelf: oceanography and resources, Volume 2. Edited by D. W. Hood and J. A. Calder. University of Washington Press, Seattle.
Maunder, J. E., and W. Threlfall. 1972. The breeding biology of the Black-legged Kittiwake in Newfound- land. Auk 89: 789-816.
Montevecchi, W.A., E. Blundon, G. Coombes, J. Porter, and P. Rice. 1978. Northern Fulmar breeding range extended to Baccalieu Island, Newfoundland. Canadian Field-Naturalist 92: 80-82.
Nettleship, D. N. 1977. Seabird resources of eastern Canada: status, problems and prospects. Pages 96-108 in Canada’s endangered species and habitats. Edited
Vol. 102
by T. Mosquin and C. Suchal. Canadian Nature Federation Special Publication No. 6., Ottawa.
Nettleship, D. N. 1980. Guide to the major seabird colonies of eastern Canada: identity, distribution and abundance. Canadian Wildlife Service “Studies on northern seabirds” Manuscript Report No. 97. 133 pp.
Nettleship, D. N., and A. R. Lock. 1973. Observations of Fulmars on ledges in Labrador. Canadian Field—Naturalist 87: 314.
Nettleship, D. N., and A. R. Lock. 1974. Black-legged Kittiwakes breeding in Labrador. Auk 91: 173-174. Nettleship, D. N., and R. D. Montgomerie. 1974. The Northern Fulmar, Fulmarus glacialis, breeding in
Newfoundland. American Birds 28: 16.
Porter, J. M. 1985. Recruitment to the colony and other aspects of the biology of the Kittiwake Rissa tridactyla. Ph.D. thesis, University of Durham, Durham. 133 pp.
Springer, A. M., D. G. Roseneau, E. C. Murphy, and M.I. Springer. 1984. Environmental controls of marine food webs: food habits of seabirds in the Eastern Chukchi Sea. Canadian Journal of Fisheries and Aquatic Sciences 41: 1202-1215.
Storey, A., and J. Lien. 1985. Development of the first North American colony of Manx Shearwaters. Auk 102: 395-401.
Swartz, L. G. 1966. Sea-cliff birds. Pages 611-678 in Environment of the Cape Thompson Region, Alaska. Edited by N. J. Wilimovsky and J. N. Wolfe. U.S. Atomic Energy Commission, Oak Ridge, Tennessee.
Tuck, L.M. 1961. The murres: their distribution, populations and biology — study of the genus Uria. Canadian Wildlife Service Monograph Series No. 1. Ottawa. 260 pp.
Received 24 April 1986 Accepted 27 January 1988
Winter and Early Spring Habitat Use by Snowshoe Hares, Lepus americanus, in South-central Alaska
JAMES G. MACCRACKEN!, WILLIAM D. STEIGERS, JR.2, and PATRICK V. MAYER
Agricultural Experiment Station, University of Alaska-Fairbanks, Palmer Research Center, 533 E. Fireweed, Palmer, Alaska 99645
!Present address: USDA Forest Service, Cordova Ranger District, P.O. Box 280, Cordova, Alaska 99574.
2Present address: SRD Box 9038-A, Palmer, Alaska 99645
MacCracken, James G., William D. Steigers, Jr., and Patrick V. Mayer. 1988. Winter and early spring habitat use by Snowshoe Hares, Lepus americanus, in south-central Alaska. Canadian Field-Naturalist 102(1): 25-30.
Snowshoe Hare, Lepus americanus, use was examined for 23 plant communities at differing elevation, aspect, and degree of slope on the north and south sides of the Susitna river in south-central Alaska. Based on fecal pellet occurrence, hares preferred White Spruce (Picea glauca) forest, alder (Alnus spp.) and willow (Salix spp.) plant communities, on slopes of 8 to >30°, at elevations from 630 to 750 m, with an eastern, southern, or southeastern aspect. Hare pellets consisted primarily of spruce, willow, and Labrador Tea (Ledum groenlandicum) fragments.
Key Words: Snowshoe Hares, Lepus americanus, habitat, foods, Alaska.
Snowshoe Hares, Lepus americanus, are the Recently, Van Horne (1983) pointed out that only leporid in the taiga of Alaska, and their evaluation of habitat preference of small mammals importance in that ecosystem has been recognized. _ based only on abundance estimates during periods Cyclic-like fluctuations in hare populations can be _ of high populations can be misleading. She cited dramatic, with densities ranging from highs of 800- examples of studies that found suboptimal habitat 1200 hares/km2 to lows of 50/km? (Keith 1974). temporarily holding greater numbers of individu- Wolff (1980) reported values of about 6 hares/ha als than optimal habitat during population highs. at a population high, to< 1/haatapopulationlow Factors that contribute to this condition are in interior Alaska. During population highs hares _ seasonal habitat use, temporal unpredictability of are a major food source for Lynx, Lynx the environment, habitat patchiness, social canadensis, and are also eaten by canids, dominance interactions, high reproductive mustelids, and raptors. At high populations, hares capacity of the species, and species that are habitat deplete their winter food resources (Wolff and generalists (Van Horne 1983).
Zasada 1979; Wolff 1980), which are sometimes At least four of these factors apply to Snowshoe shared with other herbivores (Wolff 1982). Hares in Alaska. Hares use certain habitats
Habitat use by Snowshoe Hares in Alaska has _ seasonally, the habitat can be extremely patchy, been evaluated only in the interior (O’Farrell 1965; hares have a high reproductive capacity, and they Wolff 1980). Wolff (1980) investigated hare use of appear to be habitat generalists. Thus, studies of a mature Black Spruce (Picea mariana) forest, a habitat use by Snowshoe Hares are prone to the burned Black Spruce stand, willow-alder (Salix- errors described by Van Horne (1983). However, Alnus) thickets, and an open willow stand. Other _ these problems can be avoided by studying habitat studies examining habitat use by Snowshoe Hares use by hares during a single season when in areas with similar plant communities were those populations are at low levels and the habitats are of Grange (1932), Bider (1961), Keith (1966), not saturated.
Conroy et al. (1979), and Pietz and Tester (1983). This paper reports on habitat use by Snowshoe
Hare abundance influences habitat use (Keith Hares in winter-early spring during a population 1966; Wolff 1980; Pietz and Tester 1983). Wolff low. It also examines the applicability of Wolff's (1980) found that hares exhibited seasonal (1978, 1980) results to an area of Alaska with differences in habitat use and that occupancy of greater habitat diversity. open habitats increased with increasing hare density. Open habitats are of poor quality for Study Area
Snowshoe Hares because of greater exposure to Our study area was located in the middle Susitna predation and less food (Keith 1966; Wolff 1980; River basin, near the confluence of the Oshetna Pietz and Tester 1983; Sievert and Keith 1985). and Susitna rivers in south-central Alaska. The
py)
26 THE CANADIAN FIELD-NATURALIST
broad, U-shaped glacial basin is generally oriented in an east-west direction between the Alaska Range to the north and the Talkeetna mountains to the south. Elevation ranged from about 330 m at the lowest portion of the river to over 2000 m on the mountain peaks. The wide range of environ- mental conditions in the basin has resulted in a complex mosaic of plant communities. Plant community composition and physiognomy have been strongly influenced by fire history, topo- graphy, soil moisture, aspect, and browsing by large herbivores. Precipitation in the basin averages about 51 cm and most of it falls during the frost-free period. Snow depths vary depending on site conditions. Wind-blown ridges can be free of snow, but drifts can exceed 3 m in thickness. Sixty to 75cm was about the average snow thickness in most areas.
Snowshoe Hares were extremely rare in the middle basin at the time of study. Hare sign and sightings were essentially non-existent except in a relatively small area near the confluence of the Oshetna and Susitna rivers. Based on this fact, we concluded that hares were at a population low in the middle Susitna River basin.
Methods
Habitat use by Snowshoe Hares was evaluated along four transects. Each transect was | km in length. Transects were paired and paralleled each other; there were two on the north side and two on the south side of the Susitna River. The paired transects ran perpendicular to the river and were approximately 100m apart. All four transects began in the basin above the river channel and extended to the river’s high water mark. At the high water mark of the river, the paired transects ran parallel to the river bank in established vegetation, in opposite directions for approxi- mately 100 m. The transects went through many different plant communities at different elevations, slopes, and aspects. Transects were surveyed twice between 15 May and | June 1983. On 15 May there were still patches of snow along all transects. By 1 June snow patches were confined to areas of northern aspect.
Each transect was examined by a different observer each time. At 10-m intervals along each transect (5-m intervals along the river), the observer stopped and recorded the plant community that dominated the immediate area. Plant communities were classified to level four of Viereck and Dyrness (1980) system — 1982 revision. Their classification system is hierarchial with five levels. Level one separates forest, scrub
Vol. 102
(dwarf trees-shrubs), and herbaceous communi- ties. Level two distinguishes between needleleaf and broadleaf forest, low and tall shrub, etc. Level three further divides forest communities into woodland (10-24% tree cover), open (25-59% cover), and closed (= 60% cover); a closed shrub community has > 75% cover, and an open shrub community from 25-74% cover. Level four of the system generally takes into account the dominant species, e.g. closed White Spruce (Picea glauca), open low willow, etc.
The elevation at each stop was estimated to the nearest 30 m with a pocket altimeter calibrated each day. Degree of slope was estimated with a clinometer in seven classes ranging from 0 to 30°. Aspect was estimated to the nearest of 8 divisions of the compass.
The observer also noted the presence or absence of Snowshoe Hare fecal pellets within a five meter radius of each stop; this provided an index to habitat use (Keith 1966; Wolff 1980; Pietz and Tester 1983; Litvaitis et al. 1985a). Hare pellets were collected from areas between stops during the first run of the transects and then from the stops during the second examination. Up to five pellets were collected at each stop where present.
Snowshoe Hare fecal pellets were used to estimate food habits. Only pellets that were intact and lying on top of the snow, litter, or moss layer were collected. This sample included older pellets from the previous seven months as well as fresh pellets. The pellets were pooled by transect and ground through a Wiley mill fitted with a 1-mm mesh screen. The ground material for each transect was thoroughly mixed and a random sample made into five microscope slides. Twenty fields per slide were examined at 100x magnification, and botanical composition of the slides was determined as described by Sparks and Malecheck (1968).
Habitat data were quantitatively expressed as a percentage of stops in each plant community, elevation, slope class, and aspect. Stops where hare pellets were present were quantified in the same manner with regard to plant community, elevation, slope, and aspect. Habitat and food habits data were averaged over transect pairs for the north and south sides of the Susitna River.
Chi-square goodness-of-fit tests were used to determine whether Snowshoe Hares used habitat in proportion to its availability with respect to plant community, elevation, slope, and aspect. Hare food habits were tested by species and category for differences between north and south sides of the river with t-tests. The categories tested
1988 MACCRACKEN, STEIGERS, AND MAYER: HARES IN ALASKA PL]
TABLE 1. Mean (SE) percentage of total stops and stops with Snowshoe Hare pellets in plant communities, elevation ranges, slope, and aspect along transects on north and south sides of the Susitna River.
Susitna River
North South Habitat Variable Total With Hares Total With Hares Plant Community Forest closed White Spruce 2(2) 3(3) open White Spruce 14(9) 30(28) 5(1) 20(17) woodland White Spruce 18(5) 11(0) 24(4) 10(1) open Black Spruce 1(1) 3(1) woodland Black Spruce 9(6) open spruce-birch 2(2) 4(4) woodland spruce-birch 2(2) 1(1) Scrub-Shrub woodland Black Spruce 3(3) 1(1) closed alder 3(2) 8(1) 6(1) 32(17) open alder : 7(5) 17(9) 7(3) 28(28) open Dwarf Birch 6(1) open low willow 14(3) 17(12) 10(2) 1(1) open Dwarf Birch-willow 16(6) 16(6) 3(3) open ericaceous shrub tundra 1(1) 3(0) open low alder 2(2) 15(1) 6(6) open Paper Birch scrub 1(1) (1) open low Buffaloberry-cinquefoil 4(4) Elevation range (m) 810-780 22(2) 14(1) 10(2) 779-750 27(3) 9(9) 6(1) 11(10) 749-720 7(3) 14(4) 14(2) 29(16) 719-690 9(1) 28(7) 18(2) 10(2) 689-660 12(1) 32(1) 9(1) 9(5) 659-630 12(6) 14(4) 29(3) 31(29) 629-600 11(5) 3(1) 10(3) Degree of slope 0-1 10(8) 5(5) 2(2) 2-3 23(1) 4(1) 19(4) 5(5) 4-7 19(4) 2(2) 27(4) 8(8) 8-10 8(3) 30(10) 24(3) 61(30) 11-15 10(3) 25(17) 17(5) 10(2) 16-30 16(8) 13(13) 6(4) 14(14) > 31 14(7) 26(18) 2(1) Aspect east 2(1) 2(2) 54(26) 69(30) southeast 43(9) 65(18) 5(5) south 27(4) 22(11) (1) southwest 28(4) 11(11) 1(1) west 1(1) northwest 1(1) north 21(21) 25(25) northeast 16(3) 6(6) n = mean number of stops 171 21 216 24
were trees, shrubs, forbs, grasses, and cryptogams. composition between opposite sides of the river. A similarity index (Wolff 1978) and rank-order Statistical significance was accepted at P = 0.05 correlation were used to compare total diet for all tests.
28 THE CANADIAN FIELD-NATURALIST
Results
Elevation of transects on both sides of the Susitna River ranged from 810 m to 600 m. (Table 1). Twenty three different plant communities were sampled; however, data from only 17 were used in the analysis.
Hare fecal pellets were present most often in White Spruce forest, alder stringers, and willow stands on both north and south transects (Table 1). A majority of hare pellets on the north transects were in open White Spruce, open alder, and open low willow communities. Along the south transects, stops with hare pellets were most often present in closed alder, open alder, and open White Spruce stands. Hares did not use plant communi- ties in proportion to their availability on either north (x2 = 106, P< 0.001) or south (y2= 266, P < 0.001) transects.
Hares preferred the steeper slopes (x2 = 128, P<0.001) and median elevations (x? = 120, P < 0.001) on the north and south (slope x? = 120, P < 0.001; elevation x2 = 35, P < 0.04) sides of the river (Table 1).
The majority of stops along the north transects with hares were on southeast and south aspects
Vol. 102
(x2 = 22, P< 0.005). On the south transects hares preferred east and north aspects (x?= 19, P< 0.01).
Snowshoe Hare fecal pellets contained primarily fragments from trees and shrubs (Table 2). The total number of individual pellets collected was 591 and 303 for north and south transects, respectively. On the north transects hare pellets contained 13% tree, 70% shrub, 9% forb, and 7% grass fragments. On the south transects trees were in 51% of the sample, shrubs, in 41%, forbs, in 6%, grasses, in 0.4% and cryptogams, in 0.4%. Spruce, willow, Labrador Tea (Ledum groenlandicum), and Dwarf Birch (Betula glandulosa) were the major forage species. Blueberry (Vaccinium spp.), horsetail (Equisetum spp.), and unidentifiable forbs and grasses were of lesser importance (Table 2).
Occurrence of spruce was significantly greater (P < 0.005) in fecal samples from the south side of the Susitna River. Frequency of occurrence of willows, American Red Raspberry (Rubus idaeus), alder, Dwarf Birch, and unidentifiable grasses were greater (P < 0.05) in samples from the north side. When forage species were included in categories only the frequency of occurence of trees
TABLE 2. Mean (SE) percent relative density of plant fragments identified in Snowshoe Hare feces collected from
north and south sides of the Susitna River.
Plant Species
Trees Spruce (Picea spp.)
Shrubs willow (Salix spp.) Labrador Tea (Ledum groenlandicum) Dwarf Birch (Betula glandulosa) alder (A/nus spp.) blueberry (Vaccinium spp.) American Red Rasberry (Rubus idaeus) Prickly Rose (Rose acicularis)
Forbs horsetail (Equisetum spp.) Coltsfoot (Petasites frigida) Bunchberry (Cornus canadensis) Unidentified forb
Grasses Holy Grass ( Hierochloe alpina) Bluejoint (Calamagrostis canadensis) Fescue (Festuca altaica) Unidentified graminoid Moss Lichen
Susitna River
North South 13.4(2.9) 51.8(1.6) 30.7(3.7) 16.7(1.7) 20.8(7.0) 10.7(3.8) 10.6(0.3) 6.9(0.4)
3.0(0.1) 1.9(0)
3.8(0.8) 3.4(0.3)
0.9(0.9) 0.7(0)
0.2(0.2) 1.1(0.3)
5.6(0.6) 3.3(1.1)
0.2(0.2) nae} .2(0.
3.2(0.6) 2.5(0.4)
1.11.1)
0.7(0.7)
0.2(0.2)
5.2(0.4) 0.4(0.1)
0.2(0.2)
0.2(0.2)
1988
(P <0.005) and cryptogams (P < 0.05) differed significantly between north and south transects. However, when evaluated as a whole, Snowshoe Hare diets were 91% similar and positively correlated (r; = 0.63, P << 0.01).
Discussion
Snowshoe Hares used habitats disproportion- ately to their availability. Hares preferred White Spruce forest, alder, and willow communities with canopy covers from 25- > 75%. Hares avoided open plant communities such as woodland spruce forest, ericaceous shrub tundra, and low Buffaloberry (Shepherdia canadensis) — cinqu- efoil (Potentilla sp.). It should be kept in mind that open forest and scrub communities in Viereck and Dyrness (1980) classification system can have up to 59% and 75% cover of dominant species, respectively. Snowshoe Hares preferred dense forest and scrub plant communities when they were associated with the steeper slopes of the river channel, and eastern, and southeastern, or southern aspects. Most often spruce and scrub communities where Snowshoe Hares were present in this study had cover values near the upper limits of a category.
Snowshoe Hare fecal pellets were composed primarily of spruce and shrubs during winter-early spring on our study area. There were significant differences in the percentage of some forage species between north and south transects when evaluated individually. However, hare diets, based on all foods eaten, were 91% similar, indicating little difference between north and south sides of the river.
Wolff (1978) reported that Snowshoe Hares in interior Alaska consumed primarily spruce, willow, Labrador Tea, and alder in winter. Alder was not an important forage species in our study, although it was extremely abundant.
Habitat diversity of our study area was greater than that of Wolff’s (1980). We sampled 23 plant communities, but only 11 were of major importance based on availability and use by Snowshoe Hares. Our results of diet and habitat use by hares during winter-early spring are very similar to those of Wolff (1978, 1980). This indicates that Snowshoe Hares have similar gross habitat requirements in interior and south-central Alaska during winter when populations are at low densities. Based on the results of other studies in Canada (Bider 1961; Keith 1966) and the north- central United States (Conroy et al. 1979; Grange 1932; Pietz and Tester 1983), hares appear to have similar gross habitat requirements throughout
MACCRACKEN, STEIGERS, AND MAYER: HARES IN ALASKA 29
their northern range. Keith (1966) and Wolff (1980) concluded that habitats occupied by Snowshoe Hares during population lows and/or winter are critical habitat. These habitats support remnant hare populations during periods of low density, and thus play a major role in preventing local extinctions and in providing a nucleus for subsequent population increases (Wolff 1980).
In our study area Snowshoe Hares also preferred areas of specific elevation, slope, and aspect. Undoubtedly, there is an interaction among these factors and the plant community that exists at a specific site. Few published studies have examined habitat use by hares with regard to elevation, slope, or aspect except at very gross levels. However, Litvaitis et al. (1985b) included visual estimates of slope and aspect in their study of hare habitat use in Maine. Contrary to our results, they reported that slope and aspect did not influence hare abundance. In our study, the large changes in elevation, slope, and aspect along the transect may have increased the importance of these factors relative to other studies. Although vegetation cover and density are the single most important factors influencing habitat use by hares, further investigation of topographical habitat components seems warranted.
Acknowledgments
We thank D. Helm for assistance in this study. J.O. Wolff and J. A. Litvaitis reviewed and commented on an early draft of this paper.
Literature Cited
Bider, J. R. 1961. An ecological study of the Hare Lepus americanus. Canadian Journal of Zoology 39: 81-103.
Conroy, M.J., L. W. Gysel, and G. W. Dudderar. 1979. Habitat components of clear-cut areas for Snowshoe Hares in Michigan. Journal of Wildlife Management 43: 680-690.
Grange, W. B. 1932. Observations on the Snowshoe Hare Lepus americanus phaeonotus Allen. Journal of Mammalogy 13: 1-19.
Keith, L. B. 1966. Habitat vacancy during a Snowshoe Hare decline. Journal of Wildlife Management 30: 828-832.
Keith, L.B. 1974. Some features of population dynamics in mammals. Pp. 17-58 in Proceedings of the XIth International congress of game biologists. Edited by S. Lundstrom. National Swedish Environment Protection Board, Stockholm, Sweden.
Litvaitis, J. A., J. A. Sherburne, and J. A. Bissonette. 1- 985a. A comparison of methods used to examine Snowshoe Hare habitat use. Journal of Wildlife Management 49: 693-695.
30 THE CANADIAN FIELD-NATURALIST
Litvaitis, J. A., J. A. Sherburne, and J. A. Bissonette. 1- 985b. Influence of understory characteristics on Snowshoe Hare habitat use and density. Journal of Wildlife Management 49: 866-873.
O'Farrell, T.P. 1965. Home range and ecology of Snowshoe Hares in interior Alaska. Journal of Mammalogy 46: 406-418.
Pietz, P. J.,and J. R. Tester. 1983. Habitat selection by Snowshoe Hares in north central Minnesota. Journal of Wildlife Management 47: 686-696.
Sievert, P.T., and L.B. Keith. 1985. Survival of Snowshoe Hares at a geographic range boundary. Journal of Wildlife Management 49: 854-866.
Sparks, D. R., and J. C. Malecheck. 1968. Estimating percentage of dry weight in diets using a microscope technique. Journal of Range Management 21: 264-265.
Van Horne, B. 1983. Density as a misleading indicator of habitat quality. Journal of Wildlife Management 47: 893-901.
Vol. 102
Viereck, L. A., and C. T. Dyrness. 1980. A preliminary classification system for vegetation of Alaska. U.S.D.A. Forest Service, General Technical Report PNW-106. Pacific Northwest Forest and Range Experiment Station. Portland, Oregon. 38 pp.
Wolff, J. O. 1978. Food habits of Snowshoe Hares in interior Alaska. Journal of Wildlife Management 42: 148-153.
Wolff, J. O. 1980. The role of habitat patchiness in the population dynamics of Snowshoe Hares. Ecological Monographs 50: 111-130.
Wolff, J.O. 1982. Moose-Snowshoe Hare competition during peak Hare densities. Proceedings of the North American Moose Conference 16: 238-254.
Wolff, J.O., and J.C. Zasada. 1979. Moose habitat and forest succession on the Tanana River floodplain and Yukon-Tanana upland. Proceedings of the North American Moose Conference and Workshop 15: 213-244.
Received 6 June 1986 Accepted 17 April 1987
Viability and Germination of Herbaceous Perennial Species Native to Southern Alberta Grasslands
E. A. SMRECIU,! R. S. CURRAH,! and E. TOOP?
'University of Alberta Devonian Botanic Garden, Edmonton, Alberta T6G 2E1 2Department of Plant Science, University of Alberta, Edmonton, Alberta T6G 2P5
Smreciu, E. A.,R. S.Currah, and E. Toop. 1988. Viability and germination of herbaceous perennial species native to southern Alberta grasslands. Canadian Field—Naturalist 102(1): 31-38.
The viability and germination of seed lots of 41 herbaceous perennial species native to southern Alberta grasslands were examined. The practical value of tetrazolium (TTC) and x-ray photography in screening seed lots for viability was evaluated. Although a reliable, rapid test for a few taxa, TTC often gave erratic results which were inconsistent with germination percentages and/or the physical condition of excised embryos. X-rays were useful for determining the percentage of full seeds in samples of species having large seeds, but had insufficient resolving power for samples of smaller seeds. Stratification improved germination in most species of Compositae, all three species of the Scrophulariaceae, and both species of Cactaceae. Scarification improved germination in I1 of 12 species of Leguminosae, and, in combination with stratification, improved germination of Allium textile (Prairie Onion) of the Liliaceae. Seeds of species in both Ranunculaceae and Rosaceae generally germinated as well with as without pre- treatment.
Key Words: viability, germination, herbaceous plants, perennials, tetrazolium, x-rays, Alberta, grasslands, Compositae, Scrophulariaceae, Cactaceae, Leguminosae, Ranunculaceae, Rosaceae.
Over 150 native, perennial, herbaceous species Two viability tests were used. The first involved occur in undisturbed grasslands in Alberta. Many soaking seeds in a 0.1% solution of 2,3,5-tripheny] of these have potential for use in land restoration _ tetrazolium chloride [TTC] (Grabe 1970) for up to programmes in the province (Watson et al. 1980; 24 hours and examining embryos for evidence of Currah et al. 1983) but primarily due to the lack of dehydrogenase activity as indicated by the information on their ecology and reproductive 4¢velopment of a red colour throughout the biology few have been considered for this purpose. ore ce os se are ey eee As a first step in providing much needed data in Bids OVCre Btepa we S10 these areas, this study was undertaken to examine Kodak X-OMAT TL film plates and exposing
the viability and germination of seeds collected SAS ay ALS Ge eso NALIN tee
from wild populations, and to observe the effects epamined On alent table, Seeds) Were considered Atos viable if the embryo appeared well formed and of standard pre-treatments on germination of
: intact. seeds of 41 herbaceous perennial species native to : dene Alberta’s grasslands. To obtain germination counts, seeds treated with a slurry of the fungicide Arasan 75 (lg/ 100ml), were sown on moist filter paper in petri dishes and placed in germination cabinets at 22°C Seeds were collected from southern Alberta Pie daeric s0dane Dishes were examined daily plants (nomenclature follows Packer 1983) in for germinated seeds, which were counted and 1980, 1981, and 1982, dried at room temperature, ;emoved. During the counting, seeds were exposed and stored dry in paper envelopes at 3-6°C. to normal room light. Water was replenished as Germination and viability tests were conducted in necessary. Seeds were considered germinated once the spring and summer of the year following _ the radicle had emerged through the seed coat. collection. Only apparently full seeds were used in Pre-treatments included one or more of the viability and germination tests. Full seeds were following: stratification, mechanical scarification, selected by applying slight pressure to individual acid scarification, alternating temperatures, and seeds with forceps. If there was resistance, seeds water-soaking. Seeds were stratified by moist cold were considered full. Except where noted, (1 to 6°C) storage for 2-3 months in the dark. germination tests done in 1981 consisted of four Mechanical scarification was achieved by abrading replicates of 25 seeds each. Tests in following years the seed coats with sandpaper (GR = 80 or 100) consisted of four replicates of 100 seeds each. until scratches could be seen on the coat surface
Materials and Methods
31
32 THE CANADIAN FIELD-NATURALIST
under 40X magnification. A file was used to abrade seed coats of Opuntia polyacantha (Prickly Pear). For acid scarification, seeds were soaked in either concentrated H,SO, or HCI for approximately 45 minutes (the time necessary to cause the breakdown of the seed coat in H,SO,) and then washed for several hours with running water. Hot- water soaking involved placing seeds in water heated to 80°C and allowing them to gradually cool there for 24 hours at room temperature. This treatment is used to soften seed coats to allow water uptake and radicle penetration, and/or to leach inhibitors from the seed or its coverings. Seeds incubated at alternating temperatures were held for 8 hours at 5°C and for 16 hours at 30° C for periods of three to six weeks. Viability and germination data were compared using analysis of variance and the F-test.
Results and Discussion
Neither of the two standard viability tests examined here (X-ray and TTC) gave results consistent with germination results for all species tested (Table 1).
The usefulness of x-ray examination was limited to larger seeds, since very small seeds (e.g. Heuchera_ richardsonii (Alum-root) and Coryphantha vivipara (Ball Cactus) were beyond the resolution capabilities of the x-ray machine. For larger seeds (e.g. Opuntia polyacantha and Thermopsis rhombifolia (Golden Bean), x-rays were useful in detecting shrivelled or damaged embryos (Figures 1-2).
The reliability of TTC tests was questioned when apparently healthy embryos stained incompletely (Figures 3-4). These observations support the conclusion reached by Justice (1972) that TTC is unacceptable as a universal test for viability.
In this study, we were only able to validate the effectiveness of viability tests when actual germination was similar to, or exceeded, percentages predicted by the viability tests. When fewer seeds germinated than predicted by viability tests, we could only conclude that dormancy was not overcome by the conditions provided. High viability and low germination percentages were obtained with Coryphantha vivipara, Glycyrrhiza lepidota (Wild Licorice), Grindelia squarrosa (Gumweed), and Opuntia polyacantha. Pre- treatments required to enhance germination of seeds collected from wild populations can be highly variable, either within a single seed lot (from one population), or among different seed lots (from different populations) [Crocker and Barton 1953]. Some consistent trends in pre-treatments required for germination did occur within some of
Vol. 102
the families. The following discussion is restricted to those families for which two or more species were examined. Table 2 summarizes pre-treatment and germination results for all taxa examined.
Dormancy in legumes is generally maintained by the sclerified nature of the cells comprising the palisade layer of the seed coat (Crocker and Barton 1953: Barton 1965b; Villiers 1972; Rolston 1978; Werker 1980/81), and disruption of this layer is apparently necessary for germination (Brant et al. 1971). With the exception of Glycyrrhiza lepidota, scarification increased germination and/or decreased the time required for germination to occur. The effects of stratification varied from being detrimental in seed lots of Glycyrrhiza lepidota, Hedysarum alpinum (American Sweet- broom) and some seed lots of Astragalus pectinatus (Narrow-leaved Milk Vetch), to having no appreciable effect on Astragalus bisulcatus (Two-grooved Milk Vetch), A. crassicarpus (Buffalo Bean), A. drummondii (Drummond’s Milk Vetch), and Petalostemon purpureum (Purple Prairie Clover), to being beneficial in some seed lots of Astragalus striatus (Ascending Purple Milk Vetch), Oxytropis monticola (Yellow Loco- weed), and O. sericea (Early Yellow Loco-weed).
Most seed lots of Compositae (except Senecio canus (Prairie Groundsel) and one seed lot of Antennaria nitidus (Pussy-toes) had increased germination percentages or decreased time for germination after stratification. Seed lots generally varied in their degree of response to stratification (e.g. Gaillardia aristata (Gaillardia), Grindelia squarrosa, Heterotheca villosa (Golden Aster), and Liatris punctata (Blazing Star) (Table 2). These variations might be attributable to conditions under which seeds were formed (Barton 1965a, Austin 1972).
In the Rosaceae, Geum (Avens) species demonstrated consistently high germination percentages and apparently required no pre- treatment. Some dormancy was present in some seed lots of G. triflorum (Prairie Smoke), since a decrease in the time required for germination was observed following stratification. Sorensen and Holden (1974) also found that pre-treatments were not necessary for Geum triflorum seeds collected in South Dakota. These observations do not agree with those made with other rosaceous seeds (of different sub-families) which have been shown to require a long stratification period for germination (Mayer and Poljakoff-Mayber 1982).
Germination of species of Cactaceae varied according to the year in which seed lots were collected. Coryphantha vivipara seed lots collected
1988
SMRECIU, CURRAH, AND TOOP: VIABILITY AND GERMINATION 33
TABLE 1. Percent viability as determined by TTC and x-rays in a selection of seed lots of species native to Alberta’s
Maximum Observed
grasslands. X-ray
Taxa (%) Achillea millefolium 93 Allium textile 97 Anemone cylindrica 92 Anemone multifida 95 Antennaria nitida 98 Arnica fulgens 95 Astragalus bisulcatus 96 Astragalus drummondii — Astragalus gilviflorus 100 Astragalus pectinatus 96 Astragalus striatus 100 Besseya wyomingensis — Coryphantha vivipara — Eriogonum flavum ; 89 Gaillardia aristata 74 Geum aleppicum 100 Geum triflorum 99 Glycyrrhiza lepidota 81 Grindelia squarrosa 98 Haplopappus spinulosus 83 Hedysarum alpinum 99 Heterotheca villosa 92 Heuchera richardsonii — Hymenoxys richardsonii 96 Liatris punctata 96 Linum lewisii 99 Monarda fistulosa 95 Opuntia polyacantha 98 Oxytropis monticola 99 Oxytropis sericea 100 Penstemon nitidus 95 Penstemon procerus 98 Petalostemon purpureum 98 Solidago rigida 79 Thermopsis rhombifolia 95
— no data available
from the same site over a three year period responded differently to stratification according to the year of collection. In Opuntia polyacantha, scarification (acid or mechanical) did not affect germination, but some improvement was observed following stratification. A period of alternating temperatures was partially effective in promoting germination in this taxon but satisfactory germination percentages were never obtained in the laboratory even though viability tests indicated that 98% of seeds in a given seed lot were viable. In the Scrophulariaceae, some variation from year to year was observed in the germination of
AMIS
(%) Germination (%) 99 96 99 82 94 100 95 93 97 100 96 100 94 98 89 65 97 96 99 100
100 100 78 91
100 92 96 98 79 96 99 100 98 100 87 47 99 3 94 98 99 100 92 100 98 93 89 100 99 100 99 100 93 100 98 49 99 100 98 99 90 62 93 69 98 97 73 92 95 88
Penstemon (Beard-tongue) species. Both Penstemon procerus (Slender Blue Beard-tongue) and P. nitidus (Smooth Blue Beard-tongue) germinated better following several months of cold stratification, but neither species had germination percentages above 65%. Highest germination percentages for Besseya wyomingensis (Kitten- tails) were obtained following two to three months stratification.
Each of the three species examined in the Liliaceae reacted differently to pre-treatments. Non-stratified seeds of Yucca glauca (Soapweed) germinated as well as stratified seeds. Smilacina
34
THE CANADIAN FIELD-NATURALIST
wc
Vol. 102
1988 SMRECIU, CURRAH, AND TOOP: VIABILITY AND GERMINATION 35
TABLE 2. Effects of certain pre-treatments on seed lots of native, herbaceous perennials. (st — stratification, sc — acid or mechanical scarification, sk — soaking).
Range of Germination Range of Germination Values Among Seed Lots Values Among Seed Lots FAMILY/ Species for Untreated Seeds (%) for Treated Seeds (%) Comments CACTACEAE Coryphantha vivipara 0-80 st 0-92 Significant variation in germination among seed lots. Opuntia polyacantha 0-2 st 2mo 0-49 Highest germination st 3mo 1-45 with alternating sc 0-4 temperatures was 16%. scHCl 0-5 sc H,SO, 0 sk COMPOSITAE Achillea millefolium 16-50 st 59-96 Antennaria nitida 8-100 st 83-96 Response to stratifica- tion varied among seed lots. Arnica fulgens ~ 59-85 st 79-100 Gaillardia aristata 32-96 st 36-93 Stratification increased germination signifi- cantly in one seed lot. Grindelia squarrosa 4-15 st 0-73 Haplopappus spinulosus 78-90 st 91-98 Helianthus : subrhomboideus 0 st 48-59 Heterotheca villosa 84-100 st 84-100 Stratification decreased time for germination. Hymenoxys richardsonii 49-96 st 79-100 Liatris punctata 74-100 st 92-100 Stratification increased germination signifi- cantly in one seed lot. Senecio canus 79-86 st 69-75 - Solidago rigida 22-92. st 65-92 Stratification decreased time for germination. LABIATAE Monarda fistulosa 68-92 st 70-100 LEGUMINOSAE Astragalus bisulcatus 0-30 st 20-35 sc 64-98 sc/st 89-91 sk 32-44 Astragalus crassicarpus 8-14 st 6-15 sc 97-100 sc/st 77-83
(Continued)
FiGurRE |. X-ray photograph of Thermopsis rhombifolia seeds showing full seed (f), empty seed (e), and a shrivelled embryo (s). X1.8.
FiGURE 2. X-ray photograph of Opuntia polyacantha seeds showing full seed (f) and a shrivelled embryo (s). X2.1.
FiGuRE 3. Viable embryos of Hedysarum alpinum stained with TTC showing differential staining reaction: (a) completely stained embryo, (b) partially stained embryo, and (c) unstained embryo). X2.5.
FiGureE 4. Viable embryos of Liatris punctata stained with TTC showing differential staining reaction (a) completely stained embryo, (b) partially stained embryo, and (c) unstained embryo). X2.5.
36 THE CANADIAN FIELD-NATURALIST Vol. 102
TABLE 2. (Continued). Effects of certain pre-treatments on seed lots of native, herbaceous perennials. (continued) (st — stratification, sc — acid or mechanical scarification, sk — soaking).
Range of Germination Range of Germination Values Among Seed Lots Values Among Seed Lots FAMILY / Species for Untreated Seeds (%) for Treated Seeds (%) Comments Astragalus drummondii 2-6 st 5-9 sc 41-65 sc/st 33-44 Astragalus gilviflorus 25-40 sc 88-96 Astragalus pectinatus 15-40 st 5-10 Less than 2% of the sc 72-99 seeds collected in 1982 sc/st 82-98 germinated. sk 56-100 Astragalus striatus 7-30 st 1-21 sc 77-100 sc/st 75-91 sk 32-52 Gylcyrrhiza lepidota 19-47 st 0-3 sc 5-17 sc/st 4-18 Hedysarum alpinum 88-100 st 60-91 Scarification generally sc 82-100 decreased the time for sc/st 58-95 germination. sk 60-88 Oxytropis monticola 3-24 st 6-50 sc 64-100 st/sc 81-99 sk 32-48 Oxytropis sericea 9-23 st 7-40 sc 90-99 sc/st 73-92 Petalostemon purpureum 45-54 st 44-57 sc 95-97 sc/st 30-70 Thermopsis rhombifolia 2-12 st 5-19 sc 26-88 sc/st 30-79 sk 4-12 LILIACEAE Allium textile 3-8 st 16-31 sc 4-72 sc/st 27-82 Smilacina stellata 0-66 st 10-89 Yucca glauca 76-92 st 81-92 LINACEAE Linum lewisii 60-94 st 73-100 Stratification decreased time for germination. POLYGONACEAE Eriogonum flavum 29-88 st 37-98 PRIMULACEAE Dodecatheon conjugens 0-4 st 0-36 sc 0-2 RANUNCULACEAE Anemone cylindrica 58-100 st 60-87 Anemone multifida 4-93 st 74-92 No significant differ-
ence between stratified and non-stratified seeds.
(Continued)
1988
SMRECIU, CURRAH, AND TOOP: VIABILITY AND GERMINATION 37
TABLE 2. (Concluded). Effects of certain pre-treatments on seed lots of native, herbaceous perennials. (continued) (st — stratification, sc — acid or mechanical scarification, sk — soaking).
Range of Germination Values Among Seed Lots
Range of Germination Values Among Seed Lots
FAMILY/ Species for Untreated Seeds (%) for Treated Seeds (%) Comments ROSACEAE Geum aleppicum 99-100 st 100 Geum triflorum 95-100 st 88-100 Stratification some- times decreased time for germination. SAXIFRAGACEAE Heuchera richardsonii 8-93 st 1-31 SCROPHULARIACEAE Besseya wyomingensis 0-5 st 2mo 0-91 st3mo 68-89 sk 0 sc 0-1 Pensstemon nitidus 0-20 st 2mo 0-25 Stratification decreased st3mo 26-62 time for germination. sk 0-2 Penstemon procerus 4-52 st 10-69 Stratification decreased
stellata (Star-flowered Solomon’s-seal) germi- nated best following stratification. Since seeds were removed when the radicle emerged from the seed coat, it is unknown if this species has a double dormancy as has been reported in Smilacina racemosa (False Solomon’s-seal) (Crocker and Barton 1953). Allium textile germinated best following stratification and scarification.
Clearly, there are a number of different dormancy mechanisms to consider in the seeds of the species examined. Our survey indicates that, in general, a single stratification period will break the endogenous dormancy of Compositae, whereas scarification alone is a suitable treatment to break exogenous dormancy in most species in the Leguminosae. There should be few problems in growing native species of these two families for use in land reclamation or habitat enhancement projects. The dormancy mechanism in hard to germinate seeds (e.g. Opuntia polyacantha, Dodecatheon conjugens (Shooting Star) and Glycyrrhiza lepidota) requires closer examination.
Literature Cited
Austin, R. B. 1972. Effects of the environment before harvesting on viability. Pages 114-149 in Viability of seeds. Edited by E. H. Roberts. Chapman and Hall Ltd., London.
time for germination.
Barton, L. V. 1965a. General survey of dormancy types in seeds and dormancy imposed by external agents. Pages 699-720 in Encyclopaedia of plant physiology. Volume 15. Edited by W. Ruhland. Springer-Verlag, Berlin.
Barton, L. V. 1965b. Dormancy in seeds imposed by the seed coat. Pages 727-745 in Encyclopaedia of plant physiology. Volume 15. Edited by W. Ruhland. Springer-Verlag, Berlin.
Brant, R.E., G.W. McKee, and R.W. Cleve- land. 1971. Effect of chemical and physical treat- ments on hard seeds of penngift crownvetch. Crop Science 11: 1-6.
Crocker, W., and L. V. Barton. 1953. Physiology of seeds. Chronica Botanica Company, Waltham, Massachusetts.
Currah, R.S., A. Smreciu, and M. Van Dyk. 1983. Prairie Wildflowers: an illustrated manual of species suitable for cultivation and grassland restoration. Friends of the Devonian Botanic Garden, University of Alberta, Edmonton.
Grabe, D. F. Editor. 1970. Tetrazolium testing hand- book for agricultural seeds. Association of Official Seed Analysts. Mississippi State University. Mississippi.
Justice, O. L. 1972. Essentials of seed testing. Pages 301-370 in Seed biology. Volume III. Edited by T. T. Kozlowski. Academic Press, New York.
Mayer, A.M., and A. Poljakoff-Mayber. 1982. The germination of seeds. Third Edition. Pergamon Press, Toronto.
38 THE CANADIAN FIELD-NATURALIST
Packer, J. G. 1983. Flora of Alberta by E. H. Moss. Second edition. The University of Toronto Press, Toronto.
Rolston, M. P. 1978. Water impermeable seed dor- mancy. Botanical Review 44: 365-396.
Sorensen, J. T., and D. J. Holden. 1974. Germination of native prairie forbs. Journal of Range Management 27: 123-126.
Villiers, T. A. 1972. Seed dormancy. Pages 219-281 in Seed biology. Volume II. Edited by T. T. Kozlowski. Academic Press, New York.
Vol. 102
Watson, L.E., R.W. Parker, and D.F. Pols- ter. 1980. Manual of species suitable for reclamation in Alberta. Volume II. Land Conservation and Reclamation Council, Government of Alberta.
Werker, E. 1980/81. Seed dormancy as explained by the anatomy of embryo envelopes. Israel Journal of Botany 29: 22-44.
Received 6 June 1986 Accepted 7 December 1987
Characteristics of Sharp-tailed Grouse, Tympanuchus Phasianellus, Leks in the Parklands of Manitoba
RICHARD K. BAYDACK
Natural Resources Institute, University of Manitoba, Winnipeg, Manitoba R3T 2N2
Baydack, Richard K. 1988. Characteristics of Sharp-tailed Grouse, Tympanuchus phasianellus, leks in the parklands of Manitoba. Canadian Field—Naturalist 102(1): 39-44.
Characteristics of Sharp-tailed Grouse (Tympanuchus phasianellus) leks were studied in southwestern Manitoba from May 1983 through May 1985. Leks were situated an average of 2.2 km apart. Leks averaged 450 m? in size, with area per displaying male approximately 50 m2. Leks were higher in elevation than most surrounding terrain within 500 m. Lek display areas were = 0.5 m higher than display perimeters. Display areas were flat surfaces sloped < 1%. Vegetation height was less on display areas than on perimeter areas at all times of the year. Ground cover consisted of grass (70%), forbs (15%), bare ground (15%), and shrubs (< 1%). Visibility on display areas increased progressively from summer to fall to spring. Each lek had escape cover within 500 m and trees for perching within 400 m. Key environmental characteristics for Sharp-tailed Grouse leks are elevated sites with wide-viewing horizons and nearby female perching trees. These locations appear to maximize sound transmission for both sexes.
Key Words: Sharp-tailed Grouse, Tympanuchus phasianellus, lek, dancing ground, habitat measurement, Manitoba.
A dancing ground or lek is acommunal display Study Area and Methods area where males congregate for the purpose of The study was conducted in the Carberry Sand attracting and courting females, and to which Hills, southwestern Manitoba, approximately females come for mating (Wilson 1975). Bradbury 160 km west of Winnipeg. The area derives its and Gibson (1983) described alek matingsystemas "ame from a deltaic deposit of sand formed where one in which males contribute no parental care, the ancient Assiniboine River system emptied into male territories contain no limiting resources for &!acial Lake Agassiz. Surficial deposits range from females, females select a mate, and a mating arena 2PProximately 3 to 120 m, with topography nearly aig. level to gently undulating. Average annual
Several studies have provided qualitative a ee 18 3 cm, with pout 10% descriptions of Sharp-tailed= Grouse occurring as rain. Part of the study area included
T h ea eae Pete q Canadian Forces Base Shilo, an active military MMI hid Clase che training range, where vehicular disturbance and
Baumgartner (1939) described leks as open, grassy GOoue ea al wildtites oO Cound:
knolls or ridges, usually with sparse vegetation.
: The study area was aspen parkland vegetation crete Une) an sane (1731); EOE On type, consisting of Trembling Aspen (Populus display grounds in Wisconsin and Michigan,
: j tremuloides) thickets interspersed with open respectively, suggested that relatively open, prairie grassland. Major grasses included Blue elevated sites having low or sparse vegetation were Grama (Bouteloua gracilis), Prairie Junegrass preferred. Hart et al. (1950) found that most leks (Koeleria cristata) and Porcupine Needlegrass used by sharptails in Utah were on small knolls or (Stipa spartea). Predominant forbs included sages high hills, in a weed-grass cover type. In (Artemisia frigida, A. campestris, A. ludoviciana), Saskatchewan and Idaho, respectively, Pepper Three-flowered Avens (Geum triflorum), Field (1972) and Ward (1984) found that unhindered Horsetail (Equisetum arvense), Leafy Spurge visibility was characteristic of display sites. (Euphorbia esula) and Black-eyed Susan Kobriger (1965), Twedt (1974), and Sisson (1976) (Rudbeckia hirta). Shrubs were Prickly Rose provided quantitative descriptions of some Sharp- (Rosa acicularis), Creeping Juniper (Juniperus tailed Grouse lek components in Nebraska studies. orizontalis), willows (Salix spp.), Common
In this paper, I provide a comprehensive, Snowberry (Symphoricarpos albus) and Poison quantitative description of the environmental !vy (Toxicodendron radicans).
components which constituted Sharp-tailed The study area was searched systematically for Grouse leks in the aspen parkland of southwestern _ leks during spring 1983 and 1984 (Baydack 1986). I Manitoba. verified locations by walking to the designated site
39
40 THE CANADIAN FIELD-NATURALIST
and observing grouse and/or evidence of display activity, i.e. trampled vegetation, droppings, and/ or feathers.
Lek locations were plotted on 1:50 000 National Topographic Service maps of the study area. Twelve leks were selected for investigation in locations where agricultural influences would be minimized. Ten of the twelve study leks were active during the entire investigation. The remaining two were not used by sharptails after spring 1983; therefore, data from these two inactive leks were analyzed separately. Site characteristics were measured in spring (April - May), summer (June - August), and fall (September - November) 1983 and 1984.
Lek areas and general shapes were determined by eight evenly-spaced measurements from lek centre to edge. Lek centre, usually the most heavily trampled location on the lek, was determined by observing grouse activity. Lek edge was defined as the first S5-m distance along a transect from lek centre where visual evidence of display activity was no longer apparent. Four transect lines, oriented along and perpendicular to the longest axis, were staked from lek centre to edge (display area), and to a point 50m _ beyond lek edge (display perimeter). Sampling points were at 5-m intervals along each transect. Elevation, vegetation height, ground cover and visibility were recorded at each sampling point, and were compared for each lek between display areas and display perimeters using a t-test.
Elevation was recorded using a surveyor’s level and rod. Lek slope was derived using a best-fit, linear, unbiased regression plot. Elevation was also measured at all points =< 500 m from each lek which appeared to be higher than lek centre.
Maximum vegetation height was measured in dm using a surveyor’s rod. Ground cover (percent of shrubs, forbs, grasses, and bare ground) was measured using line-intercept (Canfield 1941) and 0.1-m2 plots (Daubenmire 1959).
Visibility was measured at sampling points using the cover board technique (Jones 1968). In addition, visibility from lek centre was estimated along each transect. An observer at lek centre, lying at grouse-eye level (approximately 20 cm), indicated the highest point on a surveyor’s rod where his vision was obstructed. This height was recorded at each sampling point and was used as an index of visibility from lek centre. A best-fit, linear, unbiased regression of visible height versus distance from lek centre was derived.
Distances to escape cover and advertising or perching trees were measured by pacing from each
Vol. 102
lek centre. Appropriate locations were determined by observing grouse behaviour during summer and fall 1983 and spring 1984.
Results and Discussion Spatial Distribution
Active leks were an average of 2.2 km apart over 87.5 km2, representing a density of 0.1/km/?. Inactive leks A and B were closer than average to an active lek. Ammann (1957), Lumsden (1965) and Sisson (1976) found that spacing between sharptail leks ranged from 0.8 - 2.4km. Lek distribution likely varies according to habitat type, habitat availability, and population density. The fact that the number of dancing grounds on a given area changes yearly as a result of population fluctuation (Lumsden 1965; Cannon and Knopf 1981) possibly explains the inactivity at formerly active leks A and B. Leks A and B may also have been transient or satellite locations.
Appropriate lek area varied from 100 to 1220 m2, with mean size about 450 m?. Lumsden (1965), Twedt (1974) and Sisson (1976) also noted that lek area was highly variable. Inactive leks A and B were larger than average, possibly indicative of changeable boundaries at transient locations.
Area per displaying male at the ten active leks was found to be 50 m2, which is within the territory size range of 14 - 170 m2 determined by Evans (1969) and Hjorth (1970) and is identical to the mean size reported by Hjorth (1970). If this mean territory size is found to be constant over North American Sharp-tailed Grouse range, population estimates might be derived from lek size measurements.
Leks were not oriented in a consistent compass direction, although oval-shaped, NW-to-SE orientations were most common. This characteristic is likely a result of study area topography.
Elevation
Lek elevations were generally higher than most surrounding terrain within 500 m. For individual leks, the number of sites within 500 m_ having higher elevation ranged from 2 to 9, with a mean of 4.7. Lek centres were = 0.5 m higher than display perimeters of leks. Display areas were flat surfaces, and a best-fit, linear, unbiased regression estimate for lek slope was -0.010 + .003, indicating an approximate 1% drop in elevation over display areas.
The predominant descriptor of open country avian leks has been an elevated site (Hjorth 1970). These locations afford improved visibility and unrestricted movement, important to increased
1988
BAYDACK: SHARP-TAILED GROUSE LEKS IN MANITOBA 4]
TABLE |. Mean of maximum vegetation heights (cm) for Sharp-tailed Grouse lek display area (DA) vs. display
perimeters (DP), Carberry Sand Hills, Manitoba, 1983-84.
Summer (June-August) 1983
Lek
number! DA t-test P DP DA l
2
3 35.3 0.02 46.0 38.2 4 32h1 5)
6 41.1 0.01 64.4 35.8 7 34.2 0.02 45.5 38.3 8 27.0 0.32 29.9 24.5 9 41.4 0.06 46.6 40.3 10 30.9 0.03 45.5 34.1 Mean 35.0 0.04 46.3 34.8 A 33.1 0.09 39.1
B 26.7 0.64 28.3 24.0
‘Numerals refer to active leks; letters refer to inactive leks.
observability of mates and decreased predation (Hjorth 1970; Wiley 1974). Bradbury and Gibson (1983) indicated that environmental considera- tions are necessary to fully explain behavioural models (hotspots or large clumps of males) of determination of lek dispersion. I suggest that a critical environmental determinant for sharptails is an elevated location.
Study leks were relatively level, with << 1% decrease in elevation over the display area. Twedt (1974) and Sisson (1976) also found the slope on leks in Nebraska was gentle, usually < 3.5%. A greater slope might increase visual obstruction, thus counteracting benefits of improved visibility afforded by increased elevation.
Vegetation Characteristics
Vegetation height on leks throughout the year was less on active display areas than on display perimeters (Table 1). Spring vegetation height on
Fall (September-November) 1983
Spring (April-May) 1984
t-test P DP DA t-test P IDI 9.7 0.07 16.2
11.4 0.08 19.6
0.04 48.5 13.3 0.05 21.8 0.08 39.8 8.3 0.06 17.9 8.1 0.06 16.7
0.01 86.7 14.3 0.04 2351 0.03 49.6 11.8 0.09 16.9 0.05 31.0 10.6 0.04 17.1 0.02 48.6 7.9 0.03 NW/Eg/ 0.71 35.8 8.1 0.03 17.4 0.03 48.6 10.4 0.08 18.4 11.9 0.18 16.7
0.91 24.3 14.3 0.84 16.6
display areas was less than at other times of the year, averaging 10.4 + 1.1 cm.
Lower vegetation height on display areas, especially during spring, was primarily a result of grouse trampling. Other researchers noted similar results in Nebraska (Kobriger 1965; Sisson 1976). Twedt (1974) reported spring vegetation heights at dancing grounds in Nebraska at 8.5cm, and Anderson (1969) indicated 12-15 cm for Greater Prairie-chickens (7. cupido) in Wisconsin. Hart et al. (1950) noted that leks in Utah were in the shortest cover type available.
Ground cover of display areas was composed of fewer shrubs and more bare ground than on display perimeters. Percent ground cover type did not vary (P>0.10) among seasons (Table 2). Ground cover was dominated by grasses (70%), followed by approximately equal amounts of forbs and bare ground (15%), and a small amount of shrubs (< 1%).
TABLE 2. Seasonal ground cover on Sharp-tailed Grouse lek display areas, Carberry Sand Hills, Manitoba, 1983-84.
Date! N2 Shrub Summer 1983 98 1.0 Fall 1983 145 1.6 Spring 1984 183 0.8
Ground cover type (%)
Forb Grass Bare ground 19.1 67.7 12 18.4 65.2 14.8 1S 68.0 S29
‘Summer = June-August; Fall = September-November; Spring = April-May.
2Number of 0.1 m2? plots (Daubenmire 1959).
42 THE CANADIAN FIELD-NATURALIST
Vol. 102
TABLE 3. Mean % visibility! for Sharp-tailed Grouse display areas (DA) vs. display perimeters (DP), Carberry Sand
Hills, Manitoba, 1983-84.
Summer (June-August) 1983
Fall (September-November) 1983
Spring (April-May) 1984
Lek
number? DA t-test P DP DA t-test P DP DA t-test P DP | Tall 0.01 22.8 2 69.2 0.02 3322 3 20.8 0.38 16.7 BED 0.32 323 76.3 0.09 SET 4 39.6 0.81 40.1 SS 0.18 44.4 5 65.2 0.01 42.8 6 17.1 0.04 6.3 8251 0.53 ONT 68.3 0.03 49.7 I 32.1 0.01 8.9 74.6 0.02 53.4 8 40.8 0.41 B73 62.9 0.55 60.0 84.0 0.02 76.4 9 1382 0.31 16.7 22.3 0.64 24.0 7S 0.05 64.3 10 49.3 0.55 46.4 52.9 0.52 56.8 81.5 0.07 VL Mean 28.9 0.33 22.1 Atlee 0.59 39.2 71.8 0.04 52.2 A 40.4 0.04 24.3 61.4 0.62 59.2 B 46.1 0.98 46.0 50.5 0.38 a7 67.3 0.49 63.7
'Percentage of 75 black and white squares on a cover board visible from grouse-eye level (20 cm) at 10-m distances
from each sampling point (Jones 1968). 2Numerals refer to active leks, letters to inactive leks.
The prevalence of bare ground and absence of shrub cover on leks relative to surrounding areas likely improves grouse visibility. Whether sharp
TABLE 4. Environmental characteristics of Sharp-tailed Grouse leks in the Carberry Sand Hills, Manitoba, 1983- 84.
Component Measurement
Next nearest lek 1.7-2.9km, X= 2.2 km
Orientation NWtoSE _
Area 100-1220 m2, X = 446 m2
Area/ displaying male 50 m2 over display area
Surrounding terrain Flat to undulating, > 0.5 m lower eleva- tion than lek centre < 1% over display area
Slope
Vegetation height in
spring 7- 13cm, X= 10.4 cm Ground cover in spring
shrub < 1%
forb 15%
grass 10%
bare ground 15% Visibility Unrestricted in all
Distance to escape cover
Distance to perching trees
directions, 70-80% on display area.
<= 500 m
<= 400 m
tails select sites for leks because of cover composition or cause it through trampling (especially of shrubs) is unclear. Visibility
Visibility on display areas increased progres- sively from summer (29%) to fall (41%) to spring (72%) (Table 3). Visibility on display areas was greater than on display perimeters in spring, but no different in summer or fall. Relatively high visibility on lek display areas during spring allows for improved observability of predators, mates, and other males. Reduced visibility due to vegetation growth occurs during summer and fall, periods of low or non-use. My results are similar to Ward’s (1984) findings in Idaho that grouse preferred sites with 70-80% visibility during spring.
Visibility from lek centre was reduced progres- sively at greater distances along each transect. A best-fit, linear, unbiased regression showed that the height at which an object was obstructed from view from lek centre increased by approximately 12 cm with each 10-m progression from lek centre. The reduction of visibility with distance from lek centre may indicate that the observation range for central males on each lek is confined to a certain distance, leaving peripheral males to function as sentinels for predator observation.
Surrounding Cover
Distance to escape cover from study leks ranged from 200 to 3000 m. At all leks, suitable cover was present < 500 m from lek centre. Although no
1988
specific distances were presented, Twedt (1974), Sisson (1976), and Ward (1984) noted the importance of nearby escape cover.
Distance to female advertising or perching sites from study leks ranged from 200 to 600 m and averaged 400 m. Perching sites were sometimes equidistant from two or more leks. These sites are, I believe, important determinants of lek location. Prior to being observed on leks in spring, female sharptails congregated during mornings in tall (6- 10 m) aspen trees and “gobbled”. The gobble sound was audible on nearby leks and stimulated male dancing activity; however, only the male ‘coo’ and ‘chilk’ notes from the dancing ground could be heard at perching trees.
All sounds audible to humans at the perching trees have been measured as low frequency sound waves (Kermott and Oring 1975; Sparling 1981). Given that the least attenuation occurs in grassland habitats for low frequency waves (Marten and Marler 1977), these sounds are likely serving a long-distance communication function. Since vocalizations are important to successful mating in all tetraonines and many other species, I hypothesize that sharptail leks may be located so as to maximize sound transmission, thereby optimizing mating opportunities for each sex. The presence of perching trees at a certain distance may contribute to the spatial distribution of leks, and may be an essential factor for females in selecting a lek and/or a male for mating.
The characteristics of Sharp-tailed Grouse leks which I have described quantitatively (Table 4) agree with the general, often qualitative, descriptions available. These data will be useful not only for species management, but also for better understanding of the lek mating system. Key environment factors for sharptail leks include elevated sites with wide-viewing horizons and nearby female advertising sites — locations which maximize sound transmission for both sexes. These environmental characteristics should be addressed as additional considerations to the behavioural models presented by Bradbury and Gibson (1983).
Acknowledgments
I thank D. A. Hein for his assistance during the development of this manuscript. A. J. Erskine and F.N. Hamerstrom also provided constructive comments. Earlier versions were reviewed by R. A. Ryder, J. A. Bailey, M. C. Baker, and C. E. Braun. I thank M. D. Wonneck, B. R. Minish, and G. M. Goodwin for field assistance. Funding and support for this project were provided by the Natural
BAYDACK: SHARP-TAILED GROUSE LEKS IN MANITOBA 43
Resources Institute at the University of Manitoba, Manitoba Department of Natural Resources, Canada Department of National Defence, the Natural Sciences and Engineering Research Council of Canada, and the Sigma Xi Scientific Research Society.
Literature Cited
Ammann, G. A. 1957. The prairie grouse of Michigan. Michigan Department of Conservation Technical Bulletin. 200 pp.
Anderson, R.K. 1969. Prairie chicken responses to changing booming ground cover type and height. Journal of Wildlife Management 33: 636-643.
Baumgartner, F. M. 1939. Studies on the distribution and habits of the sharp-tail grouse in Michigan. Transactions of the North American Wildlife Conference 4: 485-490.
Baydack, R. K. 1986. Sharp-tailed grouse response to lek disturbance in the Carberry Sand Hills of Manitoba. Ph.D. thesis, Colorado State University, Fort Collins, Colorado. 83 pp.
Bradbury, J. W., and R.M. Gibson. 1983. Leks and mate choice. Pages 109-138 in Mate choice. Edited by P. Bateson. University of Cambridge Press, Cam- bridge, United Kingdom.
Canfield, R. H. 1941. Application of the line intercep- tion method in sampling range vegetation. Journal of Forestry 39: 388-394.
Cannon, R. W., and F. L. Knopf. 1981. Lek numbers as a trend index to prairie grouse populations. Journal of Wildlife Management 45: 776-778.
Daubenmire, R. E. 1959. A canopy-coverage method of vegetational analysis. Northwest Science 33: 43-64.
Grange, W.B. 1948. Wisconsin grouse problems. Wisconsin Conservation Department Publication Number 338. 318 pp.
Hart, C. M., O. S. Lee, and J. B. Low. 1950. The sharp- tailed grouse in Utah. Utah Department of Fish and Game Publication Number 3. 79 pp.
Hjorth, I. 1970. Reproductive behavior in Tetraonidae, with special reference to males. Viltrevy 7: 185-596. Jones, R. E. 1968. A board to measure cover used by prairie grouse. Journal of Wildlife Management 32:
28-31.
Kermott, L.H., and L. W. Oring. 1975. Accoustical communication of male sharp-tailed grouse ona North Dakota dancing ground. Animal Behavior 23: 375-386.
Kobriger, G.D. 1965. Status, movements, habitats, and foods of prairie grouse on a sandhills refuge. Journal of Wildlife Management 29: 788-800.
Lumsden, H. G. 1965. Displays of the sharptail grouse. Ontario Department of Lands and Forests Technical Service Research Report Number 66. 68 pp.
Marten, K., and P. Marler. 1977. Sound transmission and its significance for animal vocalization. Behavioral Ecology and Sociobiology 2: 271-290.
44 THE CANADIAN FIELD-NATURALIST
Pepper, G. W. 1972. The ecology of sharp-tailed grouse during spring and summer in the aspen parklands of Saskatchewan. Saskatchewan Department of Natural Resources Wildlife Report Number 1. 56 pp.
Sisson, L. 1976. The sharp-tailed grouse in Nebraska. Nebraska Game and Parks Commission, Lincoln, Nebraska. 88 pp.
Sparling, D.W. 1981. Communication in prairie grouse. I. Information content and intraspecific functions of principal vocalizations. Behavioral and Neural Biology 32: 463-486.
Twedt, C.M. 1974. Characteristics of sharp-tailed grouse display grounds in the Nebraska Sandhills. Ph.D. thesis, University of Nebraska, Lincoln, Nebraska. 72 pp.
Vol. 102
Ward, D. J. 1984. Ecological relationships of Colum- bian sharp-tailed grouse leks in Curlew National Grasslands, Idaho, with special emphasis on effects of visibility. M.Sc. thesis, Utah State University, Logan, Utah. 63 pp.
Wiley, R. H., Jr. 1974. Evolution of social organization and life-history patterns among grouse. Quarterly Review of Biology 49: 201-227.
Wilson, E. O. 1975. Sociobiology — the new synthesis. Harvard Belknap Press, Cambridge, Massachusetts. 697 pp.
Received 30 June 1986 Accepted 17 June 1987
The Biological Flora of Canada 8. Aralia nudicaulis L., Wild Sarsaparilla
L. B. FLANAGAN! and J. F. BAIN2
‘Department of Botany, University of Toronto, Toronto, Ontario M5S IA1 ?Department of Plant Sciences, Macdonald Campus of McGill University, Ste. Anne de Bellevue, Quebec H9X 1C0
Flanagan, L. B., and J. F. Bain. 1988. The biological flora of Canada: 8. Aralia nudicaulis L., Wild Sarsaparilla. Canadian Field-Naturalist 102(1): 45-59.
Aralia nudicaulis L. is a rhizomatous, perennial herb with a short, thick caudex which bears the above ground shoot. The shoot consists of a single compound leaf which, when reproductive, subtends a short scape which bears an umbellate inflorescence. The plant is dioecious and exhibits sexual dimorphism for several characteristics. Aralia nudicaulis is common in mesic forests from Newfoundland to British Columbia where it usually forms large clones in undisturbed communities. Flowering occurs in late May and early June and the fruit, a berry-like drupe, matures in early August. Although the plant is of no contemporary economic importance it has been used for medicinal purposes in the past.
Key Words: Aralia nudicaulis L., Wild Saraparilla, ecological life history, sexual dimorphism, clonal plant, botany.
1. Name Aralia nudicaulis L.; Araliaceae. Wild Sarsaparilla, Wild Ginseng (Turner 1975); Salseparielle (Marie-Victorin 1964). The generic name is thought to be derived from the Indian name for the plant (Marie-Victorin 1964).
2. Description of the Mature Plant
(a) Raunkiaer life-form: Hemicryptophyte. Winter-deciduous, broad-leaved, clonal, perennial herb; reproduces by seeds and rhizomes, the latter long-lived and extensive. The plant behaves like an “underground or buried shrub”; unlike most true herbs, it has abscission layers for leaves and their leaflets.
(b) Shoot morphology: The above-ground portions of the Aralia nudicaulis shoot include only the twice- compound leaves and the fertile shoots, both of which develop from small, subterranean, or occasionally emergent, erect spur shoots which themselves are attached to the extensive underground rhizome system (Figure 1). The leaf blade is ternately, pinnately decompound, 3-6 dm long including the petiole, the primary divisions 3-5 foliolate, petiolulate, the petiolules up to 10 cm long (occasionally longer), the glabrous leaflets sessile or short-stalked (less than 2 cm), ovate to elliptic, unequally obtuse or acute at the base, acuminate at the apex, serrate, 3-6 cm long. The peduncle is erect, usually shorter than the petiole, commonly bearing 3 umbels.
(c) Root morphology: The extensive rhizome system, which occurs at a mean depth of 6 cm in the mineral soil (Flinn and Wein 1977), branches in a distichous pattern with each branch producing at its apex an erect ‘spur shoot’ or caudex (Figure 1). The caudex produces both leaves and flowering shoots, but individual caudices are not necessarily active every season. True roots are adventitious and develop from the nodes on the rhizome. Secretory canals may be found in the pith, phloem and cortex of both stem and root tissue (Graham 1966).
(d) Inflorescence: The umbels of the inflorescence contain numerous 5-6 merous, epigynous, functionally unisexual flowers. The pedicels are 5-15 mm long, the calyx is approximately 2 mm long at maturity with minute sepal-like lobes; the greenish-white petals are 1.5-3.0 mm long; stamens and styles are distinct; the fruit is a fleshy, 5-loculed, berry-like drupe, 3-6 cm in diameter, purplish-black when mature.
45
46 THE CANADIAN FIELD-NATURALIST Vol. 102
FiGuRE |. Developmental stages of Aralia nudicaulis: (a) the overwintering leaf bud at the top of the caudex; (b) developing leaf and inflorescence; (c) mature male reproductive ramet; inset — enlarged male flower; (d) mature female reproductive ramet (note smaller number of flowers compared to male); inset — enlarged female flower and fruit.
(e) Subspecies: None.
(f) Varieties and Forms: Two varieties, var. prolifera Apgar and var. elongata Nash, were described by Smith (1944) as fairly distinct and very local in distribution. Neither one is recognized in any regional flora consulted (e.g. Fernald 1950; Gleason and Cronquist 1963; Porsild and Cody 1980; Moss 1983; Scoggan 1979). Our examination of herbarium specimens did not reveal clear evidence of varietal or ecotypic differentiation.
1988 FLANAGAN AND BAIN: ARALIA NUDICAULIS WILD SARSAPARILLA 47
(g) Ecotypes: None has been described.
(h) Chromosome numbers:
n 2n Locality Reference
24 (48) Kelowna, British Columbia Taylor and Taylor (1977) 24 Delta, Manitoba Love and Love (1982) 24 North Carolina: Blair (1975)
Ashe Co.;
Durham Co.;
Jackson Co.;
Mitchell Co.;
Watuaga Co.;
Virginia: Giles Co. 24 New Hampshire: Bowden (1945)
Williams Co.
Chromosome counts to date suggest that the species is diploid throughout most of its range. The single report of tetraploidy is interesting and further studies should be undertaken to ascertain whether geographically distinct chromosome races exist. Reports of other Aralia species suggest that they too are diploid, although a similar polyploid series has also been reported for A. pseudoginseng Benth. (Sharma 1970).
3. Distribution and Abundance
Aralia nudicaulis is native to North America. It is acommon to locally dominant understory plant in forest regions from Newfoundland to British Columbia, north to Yukon (Scoggan and Cody 1979) and the Northwest Territories (Figure 2), south to Georgia in the east and in the west to Colorado, where it is found only at higher elevations (2000-2500 m above sea level).
4. Physical Habitat
(a) Climatic relations: The distribution of A. nudicaulis in North America extends from latitudes higher than 60° N to latitudes lower than 40°N. This broad geographic area includes a variety of climatic regions. However, A. nudicaulis is confined to rich, moist woodlands and, thus, is absent from dry and open areas within its total range. In boreal mixed-wood forests in northern Alberta where A. nudicaulis is abundant, mean annual precipitation is 440 mm (Strong and Leggatt 1981). Most of the precipitation (300 mm) comes during the summer months, May to September. The mean summer temperature is 12°C with a range from 10.5 to 14.0°C (Strong and Leggatt 1981). The mean number of growing degree-days above 5°C is 1190 with a mean frost-free period of 85 days. The influence of different precipitation (moisture), temperature and photoperiod regimes on the performance of A. nudicaulis has yet to be investigated.
(b) Physiographic relations: Aralia nudicaulis occurs in the understory of forests on sites that range from level to very steep (40°) river bank slopes. The plant may also be found in open, regenerating clear-cut forests (Corns and La Roi 1976) and roadsides adjacent to forests (Barrett and Helenurm 1981).
The parent material of most forest soils where A. nudicaulis is common consists of glaciolacustrine tills. In some areas the parent material may consist of sand which originated as deltaic deposits in glacial meltwater lakes and has subsequently been modified by aeolian processes (St. Onge 1972).
The soil types in forests where A. nudicaulis occurs are commonly well-drained Luvisols, Brunisols, and Podzols (Canadian Soil Survey Committee 1978).
(c) Nutrient and water relations: Several authors have considered A. nudicaulis to be an indicator of moist, rich, upland sites (Dix and Swan 1971; Moss 1983). Rowe (1956) has classified the plant as ubiquitous with respect to site moisture characteristics in the Canadian prairie provinces. However, La Roi (1967) has shown that A. nudicaulis is rarely found in muskegs or bogs where Picea mariana, Black Spruce, is the dominant tree species; this indicates that the plant does not grow well in wet organic soils.
48 THE CANADIAN FIELD-NATURALIST Vol. 102
Aralia
nudicaulis L.
FiGure 2. Distribution of Aralia nudicaulis L. in Canada. Based on specimens in the herbaria at Agriculture Canada, ‘Ottawa, Ontario (DAO), National Herbarium, National Museums of Canada, Ottawa, Ontario (CAN), University of Alberta, Edmonton, Alberta (ALTA), University of Montreal, Montreal, Quebec (MT), McGill University, Montreal, Quebec (MTMG).
5. Plant Communities
Aralia nudicaulis L. is an important member of forest understory vegetation in the Acadian, Great Lakes-St. Lawrence and Boreal forest regions (sensu Rowe 1972) of Canada. Although it has been collected as far north as the Yukon and northern British Columbia (Figure 2), it generally is not a significant component of northern boreal plant communities. In more central regions it usually forms large clones in undisturbed communities.
Table | presents comparative data from forest stands representative of different regions across the country, seven from the Boreal forest region (1-5, 7, 8), one from the Great Lakes-St. Lawrence forest region (6) and one from the Acadian forest region (9). The data are taken from La Roi (1967) and Strong and La Roi (1983).
The stand descriptions in Table | were chosen not only to represent a wide geographical range but also to represent communities dominated by different overstory species. Aralia nudicaulis is present in Jack Pine ( Pinus banksiana) (stand 1), Trembling Aspen ( Populus tremuloides) (stands 2 and 3), White Spruce ( Picea glauca)-Fir (Abies balsamea) and Black Spruce dominated stands. Other studies have described A. nudicaulis as common in birch ( Betula sp.)-maple (Acer sp.), Red Oak (Quercus borealis), Balsam Poplar (Populus balsamifera), Largetooth Aspen (P. grandidentata) and Sugar Maple (A. saccharum) communities (Amiro and Courtin 1981) as well as Beech (Fagus grandifolia)-maple forests (Maycock 1961).
Thus, A. nudicaulis does not show any clear association with overstory species across its range, nor does it appear to be affected by canopy composition changes during succession. Based on studies of a successional sequence from birch to fir in the forests south of James Bay, Carleton and Maycock (1980)
1988 FLANAGAN AND BAIN: ARALIA NUDICAULIS WILD SARSAPARILLA 49
TABLE 1. Aralia nudicaulis and associated species in forest stands* from west to east across Canada’.
Species #1 #2 #3 #4 #5 #6 #7 #8 #9
een eee Dae
Pinus banksiana 5 — Es ve wa8 Abies lasiocarpa = == 4 Pinus contorta = = oar Populus balsamifera Populus tremuloides = 6 ae Picea glauca = Bs Picea mariana = == = Betula papyrifera = = = Abies balsamea LS a ong Pinus strobus == ae Acer rubrum ie al Betula lutea : se Alnus crispa 5 4 — | 2 — 2 — Corylus cornuta = 5 2 a
Prunus spp. - ae Sambucus pubens = = ome 1 1 _ Acer spicatum 1
Pyrus decora _ 3 Il
1
IN | Be AWE WD Anr| nv |
= aio] oe AN= sf | pe] |
—e NNK NO —
Pyrus americana _ Rosa acicularis 2 2 — Ribes triste Amelanchier alnifolia os 1 = Lonicera dioica 1 Cornus stolonifera 1 Ribes lacustre 2 — 1 — Viburnum edule — 2 1 6 1 — l Ledum groenlandicum — — — 1 1 l
Jul uyene | ed
—
J, eNVe NS) |
Ribes glandulosum - “= ~ — 1 — Diervilla lonicera — _— _ — 1 — Rubus ideaus 3 3 - Rubus pubescens — 3 — Rubus strigosus | , — Amelanchier spp. 2 — l a — Lonicera canadensis — — a a 2, 1 _ —- 1 Amelanchier bartramiana 1 — Corallorhiza trifida | Pyrola asarifolia _— — — 3) Equisetum scirpioides Mertensia paniculata 6 Equisetum arvense _ — _ 1 Fragaria virginiana — — — 1 | 3 |
Lathyrus ochroleucus — — — Fragaria vesca Pyrola virens — — = Habenaria obtusata Equisetum sylvaticum 1 Pyrola secunda 1 Viola renifolia — — — 2 Petasites palmatus — — _ | 1 5 6
Moneses uniflora Linnaea borealis 2 1 I; — Mitella nuda — = — 5 Apocynum androsaemifolium 1 Cornus canadensis — 2) r 6 1 3 5 3 4 1 1 1 1
NW | eee
Goodyera repens — _— — Lycopodium annotinum — — — 1
parietal Lihat iteitanbe eats ane one een eee er Oe aS ESS
(Continued)
50 THE CANADIAN FIELD-NATURALIST Vol. 102
TABLE |. (Continued)
Species #1 #2 #3 #4 #5 #6 #7 #8 #9
Circaea alpina | |
Anemone quinquifolia | 1 1 — — Actaea rubra | ] — I — Galium boreale — — T
Galium triflorum 2 2 ]
Aralia nudicaulis 2 4 ie 6 6 D 4 4 1 Rubus pubescens 6 | 1 2 2 I Aster macrophyllus 5 1 1 — 1 Habenaria orbiculata I — Gymnocarpium dryopteris 6 1 2 4 1 — Vaccinium myrtilloides I 1 2 — 1 Arctostaphylos uva-ursi
Schizachne purpurascens 1 -~ l — 1 Athyrium felix-femina I 3
Cinna latifolia 2 | ] | _ I Vaccinium vitis-idaea 1 1
Vaccinium angustifolium I 1 — — Maianthemum canadense | a 2 4 2 6 2
Calamagrostis canadensis ~- Elymus innovatus _ Epilobium angustifolium — Pteridium aquilinum Trientalis borealis 2 I
l
| |
Streptopus roseus | Streptopus amplexifolius — Smilacina trifolia Osmunda claytoniana — Lycopodium obscurum 1 Clintonia borealis 5 1 Oxalis montana 3 Coptis groenlandica 1 1 2 2, 1
| ee w
Dryopteris austriaca I Viola incognita Gaultheria hispidula Solidago macrophylla Thelypteris phegopteris 2 — — Monotropa uniflora 1 — Lycopodium lucidulum 1 —
Jue leunanspee | enn |
meee EE NNUN-»w |
aStands | to 3 taken from Strong and La Roi (1983) (1 = stand #3; 2 = stand #10; 3 = stand #4). Cover classes r = 1%; 1 = 1-5%; 2 = 6-15%; 3 = 16-25%; 4 = 26-50%; 5 = 51-75%; 6 = 76-95%. Stands 4 to 9 taken from La Roi (1967). All are White Spruce stands (4 = stand #12; 5 = stand #19; 6 = stand #21; 8 = stand #28; 9 = stand #30) except #7, which is Black Spruce (7 = Black Spruce Stand #17). Abundance notation differs for trees, shrubs and herbs. Consult La Roi (1967) for details.
bStands | to 3 Alberta (55° 15’ N, 114°0’ W; 4 Alberta (54° 21’ N, 116°35’ W); 5 Western Ontario (48°54’ N, 90°25’ W); 6 Northern Michigan (46° 15’ N, 87°25’ W); 7 Northeastern Ontario (49° 14’ N, 80°39’ W); 8 Québec (49° 45’ N, 68°41’ W); 9 New Brunswick (46° 47’ N, 66°31’ W).
concluded that A. nudicaulis was a member of a species group including Cornus canadensis, Maianthemum canadense, and Clintonia borealis, which seemed indifferent to canopy change.
Aralia nudicaulis also appears to be relatively insensitive to soil moisture changes. Maycock (1961) described A. nudicaulis on Mt. St. Hilaire (Quebec) as very ubiquitous and present in stands in all segments of the moisture gradient (but see section 4c).
Although apparently common and shade tolerant as a boreal and cool-temperate forest understory species, A. nudicaulis is not common in open canopy communities such as either the open ‘barren community’ or the relatively open ‘birch transition’ community analyzed by Amiro and Courtin (1981).
1988 FLANAGAN AND BAIN: ARALIA NUDICAULIS WILD SARSAPARILLA 51
6. Growth and Development
(a) Morphology: The mature plant overwinters in a leafless state. The overwintering bud, which is located at the top of the caudex (Figure la), occurs in the leaf litter or upper horizon of the mineral soil. In the spring after air and soil temperatures have increased, the bud swells and leaf growth and development are initiated. If a shoot is reproductive, inflorescence development is concomitant with leaf development. The petiole and scape elongate before the leaflets unroll and the flowers open (Figure 1b). The leaflets are brown initially and do not mature until approximately one week after unrolling. Flower opening occurs centripetally within an umbel. In male flowers petals and stamens remain attached until after the pollen is shed (Figure Ic). After pollination the ovary of the female flower swells but then does not ripen for approximately one month. We have been unable to germinate seeds and know of no published accounts of seedling growth and development (see section 7c).
Barrett and Helenurm (1981) studied the relative growth rates of male and female shoots in the field. Growth rates were calculated for vegetative parts in both sexes based on three weekly harvests during leaf development. The relative growth rate for females was .310 + .098 g.g'.wk ', which was significantly higher than the .154 + .066 g.g '.wk ' recorded for male shoots. Differences in relative growth rate between the sexes were evident in the month of June. For the remainder of the season little shoot growth occurs and senescence rates are similar in male and female ramets (Barrett and Helenurm 1981).
The reproductive effort (expressed as a percentage of total shoot weight) and the absolute biomass of reproductive structures of male shoots are significantly higher than females in early June due to the larger number of flowers on male inflorescences. At peak flowering, the reproductive effort of male shoots is 17.144.4% compared to 10.1 +2.5% in females (Barrett and Helenurm 1981). However, the relationship is reversed within a two-week period as a result of the termination of female flowering and the initiation of fruit development. Reproductive expenditure in females increases during June and reaches a maximum of 23.3 + 8.9% in early July (Barrett and Helenurm 1981). For a six-week period, during which the fruit develops and matures, the female shoots incur a reproductive cost not experienced by male shoots.
The pattern of biomass allocation for vegetative ramets is listed in Table 2.
(b) Physiology: Almost nothing is known about the physiology of A. nudicaulis. The pattern of 4C assimilate distribution in A. nudicaulis was studied by Flanagan and Moser (1985a) to determine the extent of physiological integration among individual shoots in a clone. Most of the labelled carbohydrate exported from a shoot was translocated basipetally into the rhizome from which the shoot emerged. While the rhizome basipetal to a shoot accumulated the highest amount of labelled carbon because of its role as a major storage organ, the roots and new developing rhizomes adjacent to a shoot had the highest specific activity, indicating that they were strong sinks during growth and development. Changes in the normal translocation pattern were observed when one shoot was shaded before an adjacent shoot was labelled. There was an increased amount of carbohydrate translocated from an unshaded shoot to the root and rhizome components adjacent to a shaded shoot. The changed pattern of translocation after disturbance indicated the potential for physiological integration among shoots within a clone (Flanagan and Moser 1985a).
TABLE 2. Mean dry weight (mg) of vegetative components of Aralia nudicaulis L. grown in the greenhouse (n = 10).
Standard Component Mean Deviation Maximum Minimum Leaflets 1319.0 840.0 2726.8 166.7 Petiole 166.7 96.5 318.5 61.8 Caudex 1439.3 827.1 2743.4 201.1 Rhizome* 5585.1 1489.5 8431.9 3967.9 Root** 1085.1 709.4 2626.2 78.1
*Weight of total length of rhizone connecting two shoots. **Weight of all roots produced along a rhizome connecting two shoots.
52 THE CANADIAN FIELD-NATURALIST Vol. 102
(c) Phenology: In central Alberta, bud break occurs during May and early June. Some preliminary data suggest that bud break is initiated after daily minimum temperatures rise above 0°C. Rapid elongation and development of the leaf results in the plant reaching maximum height within two weeks of growth initiation. The next season’s bud is formed by the middle of July. The leaf remains fully expanded and mature for approximately 115 days (until mid-September) before leaf senescence occurs.
Floral development is concomitant with leaf development. Flowers may be open and functional before the leaf has completed development. Flowering normally occurs during late May and early June, with female flowers opening before male flowers (Moss 1960; Barrett and Helenurm 1981; Barrett 1984; Flanagan and Moser 1985b). After pollination, fruit, which ripens in late July or early August, is produced. Virtually all the fruit are dispersed by the end of August.
7. Reproduction
(a) Floral biology: Aralia nudicaulis is primarily dioecious with some rare inflorescences containing perfect flowers or both male and female flowers (Barrett and Helenurm 1981; Bawa et al. 1982). The female flowers have five long styles and five short stamens with non-functional anthers. After a flower opens the styles lengthen and diverge while the stamens and green petals fall off. Stigmas are receptive for six days after the styles diverge (Barrett and Helenurm 1981). The ovary of the female flower begins to swell after pollination and reaches full size in less than one week. The berry-like drupe remains green for approximately one month, ripening and turning black in late July or August.
The male flowers have five long stamens with bright white anthers and five short styles in a non- functional pistil. The stamens fall off after the pollen is shed. In most cases the male inflorescence withers and dies after the flowers have shed their pollen. In some male flowers the styles lengthen and diverge after the stamens have fallen off. In these cases, the ovary may swell and produce a fruit, but the fruit contains no seed, so that the flowers remain functionally male.
Male inflorescences have, on average, twice as many flowers as female inflorescences (99.6 vs. 49.8, t = 25.2, p = .001; Flanagan and Moser 1985b). The difference in flower number between the sexes results from females having fewer flowers per umbel than males, since the majority of inflorescences in both sexes have three umbels. Females, however, have a significantly higher frequency of two-umbel inflorescences while males have a higher frequency of four-umbel inflorescences (Flanagan and Moser 1985b). Flower number is correlated with flowering time within a season. Individual female ramets which flower early in the season have fewer flowers than later flowering ramets (Flanagan and Moser 1985b).
The sexes also differ in aspects of their flowering phenology. Female inflorescences begin flowering earlier and reach peak flowering before males (Barrett and Helenurm 1981; Flanagan and Moser 1985b). The peak flowering time of females is from 2 to 4 days earlier than males.
Aralia nudicaulis is insect-pollinated. Pollination is required for seed set because there is no apomictic seed production (Flanagan and Moser 1985b).
(b) Seed production and dispersal: Fruit production is generally high (90-100% fruit set) but seed production is lower than potential. On average, only two of the five seeds per fruit ripen. In Alberta, individual ramets produced an average of 100 seeds per year in 1983-84 (Flanagan and Moser 1985b). Lack of pollination is not the cause of low seed production. Addition of pollen in excess of natural pollination only slightly increased seed set per flower in one of two study seasons. Seed set was never higher than an average of 2.9 seeds per flower; this suggested that resource limitation reduces seed production (Flanagan and Moser 1985b). Individual ramets exhibit a differential seed set as a function of their flowering time within a season. Seed production was highest in ramets that flowered during the peak flowering period in one season and was highest in ramets that flowered during the later stages of flowering in another season. There was a weak negative correlation between ramet size and seed production (Flanagan and Moser 1985b).
Fruit of A. nudicaulis matures in late July and early August in natural populations of the plant. The fruit is eaten and the seeds are dispersed in the scats of several mammals (Edwards 1984) and forest bird species (Beal 1915).
(c) Seed viability and germination: There are few published reports of seed germination in A. nudicaulis. Nichols (1934) and Krefting and Roe (1949) both reported that seed germination is very low. Cold treatment has been shown to be a requirement for seed germination (Nichols 1934). Several attempts to break the dormancy of A. nudicaulis seeds using various treatments (gibberellic acid, scarification, cold
1988 FLANAGAN AND BAIN: ARALIA NUDICAULIS WILD SARSAPARILLA 53
treatment) have not been successful (Flanagan unpublished). The establishment of new genets by seed germination probably occurs only rarely. We have never observed seedlings in the field.
(d) Vegetative reproduction: The principal mode of reproduction in A. nudicaulis is by growth of the rhizome. Clonal growth can result in extensive areas being occupied by ramets of a single genet. The largest clone excavated by Edwards (1984) had a diameter of 7 m, which was considered to be a very conservative estimate of clone size because rhizomes enmeshed in the root systems of trees could not be followed. The clone excavated consisted of 27 ramets, only three of which were flowering. The distance between two ramets on a linear sequence of rhizome was rarely < | m and in some cases was > 3 m. The rhizome does not readily fragment, so that the connections between distant ramets may persist for several years. The 27 excavated ramets ranged in age from | to 26 years (Edwards 1984). Shoots are not always produced sequentially as the rhizome grows, since the ages of shoot caudices do not necessarily increase with increasing distance from the growing rhizome tip. This pattern suggests that new young shoots can differentiate from buds along the rhizome at various times during the development of a clone.
8. Population Structure and Dynamics
(a) Dispersion patterns: Ramets of A. nudicaulis have a clumped spatial pattern. Morisita’s index of dispersion for a wide range of quadrat sizes indicates that significant clumping occurs at a scale of 0.25 m both for total shoots and for flowering shoots alone (Edwards 1984). Flowering shoots show more clumping than all shoots taken together.
A group of nearest-neighbour shoots which form a local patch are unlikely to be from the same rhizome system. Excavation of clones indicates that ramets connected to the same rhizome system are widely separated (0.9 m to greater than 3 m, Edwards 1984) and occupy many different patches. The above- ground patches are a result of different rhizome systems producing shoots in the same area.
On a larger spatial scale (90 m2), male and female flowering ramets often occur in patches which are sexually segregated (Barrett and Thomson 1982). Male flowering ramets occur in patches with greater densities than females (Barrett and Thomson 1982; Bawaet al. 1982). The sexual segregation of flowering ramets is probably primarily related to the clonal nature of the plant (Barrett and Thomson 1982). Clones of a different sex may have initially been established in different areas of the forest. However, the negative association of ramets may also result from differences between the flowering behaviour of the males and females. Female flowering ramets are less likely to be found in shaded areas of a forest than are males (Barrett and Thomson 1982). This may indicate sexual differences in the light requirements necessary for flower initiation. The observed spatial segregation may also be related to differences in the frequency of flowering in the two sexes (see below).
The higher degree of clumping among reproductive shoots may in part be a result of moose herbivory patterns (Edwards 1984). The pattern of moose herbivory is patchy, and moose prefer flowering shoots to vegetative shoots (Edwards 1985). Herbivory reduces the capacity for flowering in the subsequent season; this would decrease the number of patches of flowering shoots and would result in a higher Morisita index value (Edwards 1984, 1985).
(b) Age distribution: Individual ramets of A. nudicaulis are very long lived. Age distributions from three boreal forest habitats in northern Alberta indicate that the mean age of a ramet is 19.5 years (Figure 3). Ramets ranged in age from 5 to > 40 years. This suggests that individual genets must be extremely old. Edwards (1985) has shown that reproductive shoots are significantly older (mean 12.5 years) than vegetative shoots (6.9 years). All but one of the flowering shoots in Edwards’ (1985) study was at least 5 years old.
(c) Size distribution: Female ramets exceed males in both total and vegetative biomass (Barrett and Helenurm 1981). However, in terms of petiole length, leaflet number and leaflet size, no differences between male and female ramets are apparent (Table 3).
Female flowering shoots are significantly larger than non-flowering shoots (Table 3). Male flowering shoots have longer petioles than non-flowering shoots, but the two are not significantly different with respect to leaflet number and performance index (PI; Table 3).
Non-flowering shoots have a very broad size distribution relative to females and males (Figure 4). The shoot with the largest PI measured in this sample was a non-flowering shoot. Therefore, the change from a flowering to non-flowering shoot is not directly controlled by reaching a size threshold.
54 THE CANADIAN FIELD-NATURALIST Vol. 102
30 G 20 1 10 fe) 6 30 B , 10 10
40 6 NF 30 . 2 20 600 1200 1800 2400 ShZ Ea PA)
] FIGURE 4. Frequency distribution of Perfor- mance Index in female, male and non-flowering shoots of Aralia nudicaulis in 1983. Measure-
T ments were made on plants in reference stand #3 10920 306 (40
of Strong and La Roi (1983). Performance Index is defined in Table 3.
S)
(o?) ie) SS)
3
NO. OF RAMETS a
FREQUENCY
S
AGE YRS)
FiGureE 3. Population age structure for Aralia nudicau- lis in three successionally related plant communi- ties: A) Stand 2; B) Stand 3; C) Stand 4 of Strong and La Roi (1983). N = 20 for each stand.
In shaded habitats A. nudicaulis produces smaller leaflets than in open areas. In extremely shaded habitats not only is leaflet size reduced but leaflet number is also reduced (Table 4).
(d) Growth and turnover rates: Male-biased floral sex ratios have been reported for several natural populations of A. nudicaulis in New Brunswick and Massachusetts (Barrett and Helenurm 1981; Bawa et al. 1982). The male-biased floral sex ratio results from differences in the frequency of flowering between the two sexes (Table 5). Of the Alberta ramets marked in 1983, only 4.8% of the females flowered again in 1984, while 39.1% of the males flowered again in 1984. Similar differences between the sexes in consecutive flowering pattern were observed in 1984-85, but higher percentages of both male and female ramets flowered in 1985. Approximately 10% of the flowering ramets marked in one season produced no leaf the following season.
1988 FLANAGAN AND BAIN: ARALIA NUDICAULIS WILD SARSAPARILLA 55
TABLE 3. Mean values for the leaf characteristics of flowering and non-flowering Aralia nudicaulis L. ramets during 1983*. Differences among ramet types were determined by nonparametric multiple comparison tests after Kruskal-Wallis analysis of variance. Within each row, values followed by the same letter are not significantly different
(P > 0.05). Ramet Type Characteristic Female Male Non-flowering Petiole Length (cm) 24.254 26.894 21.875 Leaflet Number 14.284 13.0746 12.53> Performance Index** 1040.84 1012.4a 852.76 n 60 60 60
*Measurements made on plants in reference stand #3 described in Strong and La Roi (1983) (Stand #1, Table 1).
** Performance Index = n (L,W, + L, W,)/2 where n = the number of leaflets per shoot, and L, W = the length and width (cm) of the two largest leaflets per shoot. PI is linearly related to leaf area (LA) and is approximately twice as large (PI = 1.8 LA + 13.05, r= 0.99, P< 0.001).
(e) Successional role: Aralia nudicaulis is an abundant understory herb in the boreal forest where extensive wildfires are frequent (Rowe and Scotter 1973). Aralia nudicaulis is present in a wide variety of plant communities ranging from young, post-fire, seral communities to old White Spruce-Fir communities. This may be attributed to the ability of the rhizome to survive many fires while protected in the mineral soil (Flinn and Wein 1977). Regeneration of shoots from buds on rhizome fragments may allow quick recolonization after fire. Fire may also provide suitable environmental conditions for seed germination. A common characteristic of many boreal forest understory species, including A. nudicaulis, is the lack of response to qualitative changes in, or affinity for, any forest canopy type (Carleton and Maycock 1980, 1981). However, in extremely low light regimes under mature White Spruce-Fir forests, reproductive performance (Barrett and Thomson 1982) and vegetative vigor are reduced (Table 4).
Aralia nudicaulis is also able to survive disturbances like forest clear-cutting. Corns and La Roi (1976) have noted the presence of A. nudicaulis in Lodgepole Pine, Pinus contorta, sites which had been clear- cut and scarified seven years before sampling.
9. Interaction with Other Species
(a) Competition: Aralia nudicaulis may be affected by reduced light intensity when associated with the shrubs Alnus crispa, Rosa acicularis, Shepherdia canadensis, Viburnum edule, or the fern, Pteridium aquilinum, which overtop it.
Competition for water and nutrients may occur with any of the many understory herbaceous plants which have shallow roots in the organic layer and upper levels of the mineral soil. Some of these species which are commonly associated with A. nudicaulis are Clintonia borealis, Cornus canadensis, Epilobium angustifolium, Lycopodium annotinum, Maianthemum canadense, Mertensia paniculata, and Trientalis borealis. The feather moss species Pleurozium schreberi, Hylocomium splendens and Ptilium crista- castrensis may also compete with A. nudicaulis for water and nutrients.
Understory species that bloom synchronously with A. nudicaulis and are bee-pollinated may compete for pollinator service. Any of the herbaceous angiosperms listed above could potentially compete for pollinators.
(b) Symbiosis: The major pollinators of A. nudicaulis are the bumble bees, Bombus vagans F. Smith and B. ternarius Say. Other flower visitors include andrenids, halictids, syrphids, small flies and thrips (Barrett and Thomson 1982; Bawa et al. 1982).
Malloch and Malloch (1981) have shown that endomycorrhizae are commonly associated with the roots of A. nudicaulis but the fungi involved were not identified.
(c) Predation and parasitism: Grazing by Moose (Alces alces) on A. nudicaulis may be intense in some aeas. On Isle Royale, Michigan, Edwards (1985) has shown that the proportion of A. nudicaulis shoots
56 THE CANADIAN FIELD-NATURALIST Vol. 102
TABLE 4. Mean values for the leaf characteristics of 30 Aralia nudicaulis ramets in three successionally related plant communities in the boreal forest near Hondo, Alberta. Differences between stands were determined by Kruskal-Wallis and nonparametric multiple comparison tests.
Reference Stand*
2 3 4 Leaflet size 19-2 atte 53.8 b 24.2 ¢ (L x W = cm?) No. of leaflets 12.6a 13.2a 6.9b pi** 1052.6a 746.4 b 195.1 c¢
*Reference Stand # from Strong and La Roi (1983). Stands 2 to 4 are progressively more shaded (see Ross et al. (1986)). **Performance Index is defined in Table 3. *** Within each row, values followed by the same letter are not significantly different, P= 0105.
eaten increased with shoot density from 11% in quadrats with less than nine shoots m to 77% in quadrats with over 20 shoots m~. Reproductive shoots were grazed more often (63% affected) than vegetative shoots (33% affected). Other grazers included Snowshoe Hares (Lepus americanus), and meloid beetles (Epicauta murina; Edwards 1985).
The fruit of A. nudicaulis is eaten by Black Bears (Ursus americanus Pallas), Red Foxes (Vulpes vulpes L.), Wolves (Canis lupus L.; Edwards 1985) and several forest bird species including the thrushes Catharus ustubatus and C. guttatus (Beal 1915).
A rust, Nyssopsora clavellosa (Berk.) Arth. attacks A. nudicaulis throughout most of its Canadian range (Savile 1975). A fruticolous smut, Mundkurella mossii, of A. nudicaulis is known mainly from Alberta and Saskatchewan. It is not clear whether the smut is perennially systemic in the rhizome or only annually systemic in individual flowering shoots (Savile 1975).
10. Evolution and Migration
Species of the genus Aralia are found in Asia and Malaysia as well as North America. Numerous Aralia fossils have been described (Berry 1903), including some of Tertiary age from regions of Alaska where Aralia is no longer present. The evidence suggests, therefore, that ancestral Aralia species existed in the Tertiary mesophytic forest which was circumpolar at that time.
Aralia nudicaulis is not closely related to other North American species (Harms 1898). No evidence of hybridization has been discovered. Aside from the recent report of tetraploidy (Taylor and Taylor 1977), all information suggests that the species is a very homogeneous, well-defined assemblage.
11. Response Behaviour
(a) Fire: Aralia nudicaulis may survive the effects of many forest fires because its rhizome occurs at a mean depth of 6 cm in the mineral soil (Flinn and Wein 1977). Species with regenerative organs located primarily in the mineral soil will have the greatest survival rate during fire. New leaves of A. nudicaulis produced from buds on rhizomes have been observed in the field the season after a forest fire occurred.
(b) Grazing: Those shoots of A. nudicaulis whose leaves had been experimentally clipped produced significantly fewer fruits than intact shoots during the experimental season (Edwards 1985). Five of 42 clipped reproductive shoots and 28 of 37 clipped vegetative shoots produced a new leaf during the season clipping was performed. However, in all but one case the newly produced leaf was smaller than its intact matched shoot. In the following year, none of the clipped shoots flowered. A relatively high percentage of previously marked, intact reproductive (29%) and vegetative shoots (11%) flowered in the following year. Clipping had no effect on survivorship of shoots (Edwards 1985).
(c) Flooding: High levels of soil moisture are not tolerated by A. nudicaulis. Maintaining A. nudicaulis in the greenhouse in continuously wet soil results in rotting of the rhizome and subsequent death of the plant.
1988 FLANAGAN AND BAIN: ARALIA NUDICAULIS WILD SARSAPARILLA 57
TABLE 5. Demographic behavior of Aralia nudicaulis flowering ramets*. Flowering ramets were marked in 1983 and classified into the three classes shown below in 1984. A similar procedure was followed for a new group of flowering ramets marked in 1984 and followed to 1985. The response of the sexes is significantly different (chi-square = 38.3, 2df, P< 0.001 for 1983-84; chi-square = 17.5, 2df, P< 0.001 for 1984-85).
Leaf and Inflorescence Leaf No Leaf Total
1983-84
Male 26 42 6 74 (35.1%) (56.8%) (8.1%)
Female 8 140 16 164
(4.8%) (85.4%) (9.8%)
1984-85
Male 51 34 11 96 (53.1%) (35.4%) (11.5%)
Female 32 715 13 120 (26.7%) (62.5%) (10.8%)
*Observations made.on plants in reference stand #3 described in Strong and La Roi(1982) (Stand #1, Table 1).
(d) Insecticide: In New Brunswick aerial insecticide spraying (fenitrothion) for spruce budworm can affect the fecundity of entomophilous plants. Aralia nudicaulis showed significantly lower fecundity in sprayed relative to unsprayed areas (Thaler and Plowright 1980). This difference in fecundity is thought to be related to the mortality of insect pollinators, particularly bumble bees, in the sprayed area.
12. Relationship to Man
Although Aralia nudicaulis is of no contemporary economic importance, it was apparently of some use (mostly medicinal) in the past. The Bella Coola made a refreshing beverage by boiling the rhizomes in tall wooden boxes until the water was reddish-brown. After the coming of Europeans this “tea was sweetened with sugar. It was also taken as a medicine for stomach pains (Turner 1975). Boiled and powdered roots were used by some North American Indians as a cough remedy (Lewis and Elvin-Lewis 1977), while Fernald (1950) and Marie-Victorin (1964) state that the roots have been used in folk medicine as a substitute for officinal sarsaparilla (from Smilax sp.). The latter author also mentions that one can make from the fruits “un vin de ménage aromatique”.
Acknowledgments We thank H. Addy, B. David and W. Moser for help in collecting some of the data presented here, S. Wolff for drawing Figure 1, and G. La Roi for providing many useful comments on the manuscript.
Literature Cited
Amiro, B. D., and G. M. Courtin. 1981. Patterns of vegetation in the vicinity of an ecologically disturbed ecosystem, Sudbury, Ontario. Canadian Journal of Botany 59: 1623-1639.
Barrett, S.C. H. 1984. Variation in floral sexuality of diclinous Aralia (Araliaceae). Annals of the Missouri Botanical Garden 71: 278-288.
Barrett, S.C. H., and K. Helenurm. 1981. Floral sex ratio and life history in Aralia nudicaulis. Evolution 35: 752-762.
Barrett, S.C. H., and J. D. Thomson. 1982. Spatial pattern, floral sex ratios, and fecundity in dioecious Aralia nudicaulis. Canadian Journal of Botany 60: 1662-1670.
Bawa, K. S., C. R. Keegan, and R. H. Voss. 1982. Sexual dimorphism in Aralia nudicaulis L. Evolution 36: 371-378.
Beal, F. E. L. 1915. Food habits of the thrushes of the United States. U.S.D.A. Bulletin 280: 1-23.
Berry, E. W. 1903. Aralia in American paleobotany. Botanical Gazette 35: 421-428.
Blair, A. 1975. Karyotypes of five plant species with disjunct distributions in Virginia and the Carolinas. American Journal of Botany 62: 833-837.
58 THE CANADIAN FIELD-NATURALIST Vol. 102
Bowden, W. M. 1945. A list of chromosome numbers in higher plants. I: Acanthaceae to Myrtaceae. American Journal of Botany 32: 81-92.
Canadian Soil Survey Committee, Subcommittee on Soil Classification. 1978. The Canadian system of soil classification. Canada Department of Agriculture Publication 1646. Supply and Services Canada, Ottawa.
Carleton, T. J., and P. F. Maycock. 1980. Vegetation of the boreal forests south of James Bay: non-centred component analysis of the vascular flora. Ecology 61: 1199-1212.
Carleton, T. J., and P. F. Maycock. 1981. Understory—-canopy affinities in boreal forest vegetation. Canadian Journal of Botany 59: 1709-1716.
Corns, I. G. W., and G. H. La Roi. 1976. A comparison of mature with recently clear-cut and scarified lodgepole pine forests in the lower foothills of Alberta. Canadian Journal of Forest Research 6: 20-32.
Dix, R. L., and J. M. A. Swan. 1971. The roles of disturbance and succession in upland forest at Candle Lake, Saskatchewan. Canadian Journal of Botany 49: 657-676.
Edwards, J. 1984. Spatial pattern and clone structure of the perennial herb Aralia nudicaulis L. Bulletin of the Torrey Botanical Club 111: 28-33.
Edwards, J. 1985. Effects of herbivory by moose on flower and fruit production of Aralia nudicaulis. Journal of Ecology 73: 861-868.
Fernald, M. L. 1950. Gray’s manual of botany. Eighth edition. American Book Co., New York.
Flanagan, L. B., and W. Moser. 1985a. Pattern of '4C assimilate distribution in a clonal herb, Aralia nudicaulis. Canadian Journal of Botany 63: 2111-2114.
Flanagan, L. B., and W. Moser. 1985b. Flowering phenology, floral display and reproductive success in dioecious Aralia nudicaulis L. (Araliaceae). Oecologia 68: 23-28.
Flinn, M. A., and R. W. Wein. 1977. Depth of underground plant organs and theoretical survival during fire. Canadian Journal of Botany 55: 2550-2554.
Gleason, H. A., and A. Cronquist. 1963. Manual of vascular plants of northeastern United States and adjacent Canada. Willard Grant Press, Boston.
Graham, S. A. 1966. The genera of Araliaceae in the southeastern United States. Journal of the Arnold Arboretum 47: 126-136.
Harms, H. 1898. Araliaceae. Pp. 1-62 in Die naturlichen Pflanzenfamilien III 8. Edited by A. Engler and K. Prantl.
Krefting, L. W., and E.I. Roe. 1949. The role of some birds and mammals in seed germination. Ecological Monographs 19: 269-286.
La Roi, G. H. 1967. Ecological studies in the boreal spruce-fir forests of the North American Taiga. 1. Analysis of the vascular flora. Ecological Monographs 37: 229-253.
Lewis, W. H., and M. P. F. Elvin-Lewis. 1977. Medical botany. Wiley and Sons, New York.
Love, A., and D. Love. 1982. Pp. 344-360 in IOPB chromosome number reports LX XV. Taxon 31.
Malloch, D., and B. Malloch. 1981. The mycorrhizal status of boreal plants: species from northeastern Ontario. Canadian Journal of Botany 59: 2167-2172.
Marie-Victorin, Frére. 1964. Flore Laurentienne. Deuxiéme édition, revised by E. Rouleau. Les Presses de VUniversité de Montréal, Montréal.
Maycock, P. F. 1961. Botanical studies on Mont St. Hilaire. Canadian Journal of Botany 39: 1293-1325.
Moss, E. H. 1960. Spring phenological records at Edmonton, Alberta. Canadian Field-Naturalist 74: 113-118.
Moss, E. H. 1983. Flora of Alberta. Second edition, revised by J. G. Packer. University of Toronto Press, Toronto.
Nichols, G. E. 1934. The influence of exposure to winter temperatures upon seed germination in various native American plants. Ecology 15: 364-373.
Porsild, A. E., and W. J. Cody. 1980. Vascular plants of continental Northwest Territories. National Museums of Canada, Ottawa.
Ross, M.S., L. B. Flanagan, and G. H. La Roi. 1986. Seasonal and successional changes in light quality and quantity in the understory of boreal forest ecosystems. Canadian Journal of Botany 64: 2792-2799.
Rowe, J. S. 1956. Uses of undergrowth plant species in forestry. Ecology 37: 461-473.
Rowe, J.S. 1972. Forest regions of Canada. Department of the Environment, Canadian Forest Service, Ottawa. Publication no. 1300.
Rowe, J. S., and G. W. Scotter. 1973. Fire in the boreal forest. Quaternary Research 3: 444-464.
Savile, D. B. O. 1975. Mundkurella mossii, a smut of Aralia nudicaulis. Mycologia 67: 273-279.
Scoggan, H. J. 1979. The flora of Canada. National Museums of Canada. Ottawa.
Scoggan, H. J.,and W. Cody. 1979. Interesting vascular plants from the southern Yukon Territory. Canadian Field- Naturalist 73: 163-170.
Sharma, A. K. 1970. Annual report, 1967-1968. Research Bulletin, University of Calcutta (Cytogenetics Lab.) 2: 1-50.
Smith, A. C. 1944. Araliaceae. Pp. 3-41 in North American flora. Volume 28B. New York Botanical Garden.
St. Onge, D. A. 1972. Sequence of glacial lakes in north-central Alberta. Geological Survey of Canada Bulletin 213. Ottawa.
Strong, W. L., and K. R. Leggatt. 1981. Ecoregions of Alberta. Energy and Natural Resources, Edmonton, Alberta. ENR Technical Report No T/4.
1988 FLANAGAN AND BAIN: ARALIA NUDICAULIS WILD SARSAPARILLA 59
Strong, W. L., and G. H. La Roi. 1983. Rooting depths and successional development of selected boreal forest communities. Canadian Journal of Forest Research 13: 577-588.
Taylor, R. L., and S. Taylor. 1977. Chromosome numbers of vascular plants of British Columbia. Syesis 10: 125-138.
Thaler, G. R., and R. C. Plowright. 1980. The effect of aerial insecticide spraying for spruce budworm control on the fecundity of entomophilous plants in New Brunswick. Canadian Journal of Botany 58: 2022-2027.
Turner, N. J. 1975. Food plants of British Columbia Indians. Part 1: Coastal Peoples. British Columbia Provincial Museum, Victoria.
Received 8 August 1986 Accepted 8 April 1987
Notes
Nesting of King Eiders, Somateria spectabilis, and Snowy Owls, Nyctea scandiaca, near Cape Churchill, Manitoba
TIMOTHY J. MOSER! and DONALD H. RUSCH
Wisconsin Cooperative Wildlife Research Unit, University of Wisconsin, Madison, Wisconsin 53706 'Present address: Indiana Department of Natural Resources, 3900 Soldier’s Home Road, West Lafayette, Indiana
47906
Moser, Timothy J., and Donald H. Rusch. 1988. Nesting of King Eiders, Somateria spectabilus, and Snowy Owls, Nyctea scandiaca, near Cape Churchill, Manitoba. Canadian Field-Naturalist 102(1): 60-61.
We report the nesting of King Eiders (Somateria spectabilis) (second and subsequent records for Manitoba), and Snowy Owls (Nyctea scandiaca) (first records for Manitoba since 1936), near Cape Churchill, Manitoba.
Key Words: Snowy Owl, Nyctea scandiaca, King Eider, Somateria spectabilis, nests, breeding range, tundra,
Manitoba.
Our purpose is to report nests of King Eiders (Somateria spectabilis) and Snowy Owls (Nyctea scandiaca) near Cape Churchill, Manitoba (58° 46’N, 93°16’W). Our observations were made in the “coastal tundra” zone (Wellein and Lumsden 1964) of the Hudson Bay Lowlands from April to August 1981-84 and from 12 June to | July 1985. The study area lies 58 km ESE of Churchill, within 5 km of Hudson Bay and more than 15 km beyond the treeline. This area was included in recent studies of avifauna by Jehl and Smith (1970) and Cooke et al. (1975).
KING EIDER. Adult drakes or pairs were observed from 7 to 29 June 1983 and from 20 May to 25 June 1984. Observations spanned only seven- day periods in June and July of 1981 and 1982. On 22 June 1983 a female, judged to be a King Eider by its color and bill morphology, flushed from a nest containing one egg. An adult drake King Eider had been circling the nest site during our approach (approximately 5 min). The egg measurements (62.9 X 44.8 mm), and down and contour feathers collected from the nest, most resembled those of King Eiders (Bent 1925; Cramp and Simmons 1977; Harrison 1978). The nest subsequently was found destroyed. On 29 June another drake King Eider behaved similarly around a hen and nest containing four eggs. This nest was also subsequently destroyed.
On 13 June 1985 a female King Eider flushed from a one-egg nest. A drake King Eider was also
60
present. On 19 June 1985 a King Eider female flushed from the same nest that contained six fresh eggs; no male was observed. Egg measurements (x = length = 62.5 mm, x width = 43.0 mm), down coloration, and photographs of the female indicate their identity as S. spectabilis. This nest was found destroyed, apparently by avian predators, on 25 June. King Eider nesting has been recorded farther north in the North-west Territories (Godfrey 1966) and there is evidence of breeding to the south in Ontario (Alison 1975). There is only one previous nesting record in Manitoba, however (Abraham and Cooke 1979). King Eiders were reported as rare migrants or visitors by Jehl and Smith (1970) and Cooke et al. (1975). Our observations suggest that King Eiders may now nest relatively frequently near Cape Churchill. Common Eiders (Somateria mollissima) nest annually on our study area at low densities (<< 0.2 nests/km_?).
SNowy OwL. On 12 June 1984 we visited a Snowy Owl nest containing six eggs which we estimated had been incubated 17 days. Both adults were present. Eggs were still intact on 25 June and three young near fledging age were observed from a helicopter on 5 August. We frequently sighted owls 4 km to the north and P. Majewski and J. Reynolds (personal communication) saw two owls copulating in the area on 27 May, which suggested the presence of another breeding pair.
On 12 June 1985, one adult flushed from a beach ridge nest containing seven eggs, one of which was
1988
infertile (or its embryo had died very early). On 20 June one adult flushed from the same nest that contained four owlets and three intact eggs. Prey items at the nest included three Snow Goose goslings (two blue phase and one snow phase) (Chen caerulescens). During a visit on 24 June one adult flushed from the nest and feigned injury approximately 100 meters from the nest. The nest contained three owlets, one pipped egg and one intact egg. A partially consumed male Willow Ptarmigan (Lagopus lagopus) was near the nest. On | July no young were found at the nest and were presumed to have been killed by predators. Another nest was found on a hummock in a sedge meadow on 24 June 1985. One adult and four owlets were observed on 24 and 27 June and | July 1985. A third nest found in 1985 was visited only once. The nest, on ahummock in a sedge meadow, contained two eggs and was auended by two adults on 23 June.
Snowy Owl nesting in Manitoba was last reported in 1936 (Shelford and Twomey 1941). Owls were observed during the nesting season near Cape Churchill in 1968 (Jehl and Smith 1970) and by us in 1981 but showed no evidence of breeding. One of us (D. H. R.) found no evidence of owls nesting during breeding season visits to the study area in 1971-1980. Local Collared Lemming (Dicrostonyx groenlandicus) populations appeared very high in 1984, and to a lesser degree in 1985, based on the frequency of sightings of Lemmings, remains of winter nests, and fecal piles. In 1984 and 1985 we also observed increases in the abundance and nesting effort of other species (Moser and Rusch 1988) that prey heavily on microtine populations. Snowy Owls and other predators may also be responding to the additional prey base provided by recent increases in the numbers of Snow Goose broods using the area south of Cape Churchill.
Snowy Owls were commonly observed by us from 9 May to 15 July in 1981 and 4 May to 8 June in 1982. Our earliest observation was 21 April 1983.
NOTES 61
Acknowledgments
We thank S. DeStefano, M. A. Hay, D. L. Orthmeyer, M. Gillespie, M. C. Brittingham and E. Santana and the many agency and university personnel who participated in field work on Cape Churchill. Support for the study of Canada Geese on Cape Churchill was provided by agencies of the Mississippi Flyway Council.
Literature Cited
Alison, R.M. 1975. The King Eider in Ontario. Canadian Field-Naturalist 89: 445-447.
Bent, A.C. 1925. Life histories of North American wildfowl. (Order: Anseres Part II). United States National Museum Bulletin 130. 376 pp.
Cooke, F., R.K. Ross, R.K. Schmidt, and A.J. Pakulak. 1975. Birds of the tundra biome at Cape Churchill and La Perouse Bay. Canadian Field—Naturalist 89: 413-422.
Cramp, S., and K. E. L. Simmons. 1977. The birds of the western palearctic, Volume |: Ostrich to ducks. Oxford University Press, Oxford. 722 pp.
Godfrey, W. E. 1966. The birds of Canada. National Museum of Canada Bulletin 203. 428 pp.
Harrison, C. 1978. A field guide to the nests, eggs and nestlings of North American birds. Collins, Cleveland. 416 pp.
Jehl, J. R., Jr., and B. A. Smith. 1970. Birds of the Churchill region, Manitoba. Manitoba Museum of Man and Nature, Winnipeg, Special Publication 1. 87 pp.
Moser, T.J., and D.H. Rusch. 1988. Notes on uncommon birds and mammals near Cape Churchill, Manitoba. Blue Jay 46: 52-54.
Shelford, V.E., and A.C. Twomey. 1941. Tundra animal communities in the vicinity of Churchill, Manitoba. Ecology 22: 47-69.
Wellein, E. G., and H. G. Lumsden. 1964. Northern forests and tundra. Pp. 67-76 in Waterfowl tomorrow. Edited by J. P. Linduska. United States slags! of Interior, Washington.
Received 25 July 1985 Accepted 2 September 1987
62 THE CANADIAN FIELD-NATURALIST
Vol. 102
A Replacement Clutch in Wild Gyrfalcons, Falco rusticolus, in
the Northwest Territories
K. G. POOLE
Department of Zoology, University of Alberta, Edmonton, Alberta T6G 2E9 Present Address: Wildlife Management Division, N.W.T. Department of Renewable Resources, Yellowknife,
Northwest Territories X1A 2L9
Poole, K. G. 1988. A replacement clutch in wild Gyrfalcons, Falco rusticolus, in the Northwest Territories. Canadian
Field—Naturalist 102(1): 62-64.
A pair of Gyrfalcons (Falco rusticolus) in the Northwest Territories laid a replacement clutch after abandonment of its initial set of eggs as a result of disturbance. Renesting in wild Gyrfalcons is uncommon.
Key Words: Gyrfalcon, Falco rusticolus, breeding, renest, Northwest Territories.
The practice of producing a second or replacement clutch when the first is destroyed or abandoned is common among most raptors (Newton 1979). The majority of falcons, for example, will renest if the initial clutch is lost within the first 2 wk of incubation (Cade 1982). However, renesting has rarely been reported in falcons larger than the Peregrine (Falco peregri- nus) (Newton 1979; Morrison and Walton 1980, Allen et al. 1986). Additionally, within a species the frequency of renesting tends to decrease at higher latitudes (Cade 1960; Newton 1979). This paper reports on an observation of renesting in Gyrfalcons (F. rusticolus) in a population in the Northwest Territories (N.W.T.) (approximately 68°N, 107° W), and reviews the available literature on renesting in this species.
A Gyrfalcon population of between 14 and 18 territorial pairs was studied in an area of 2000 km? of rugged tundra from 1982 to 1986. For details on survey techniques and methods, and description of the study area, see Poole and Bromley (1988).
The renesting incident described here resulted from nest abandonment unintentionally induced by human disturbance. On 22 May 1985, a Super-8 time-lapse movie camera unit (after Temple 1972), set to take one frame every 3 min, was placed approximately 3m from Gyrfalcon Site 113 containing four eggs. The film recorded activity at the nest ledge for | wk, and showed that incubation effectively ended at camera placement. The daily minimum ambient temperature ranged from -4 to 0°C. Brief individual visits by both adults were recorded up to 5d after camera placement, but these visits never totaled more than 3% of the frames in any day. Three periods of incubation, one by the female and two by the male, were recorded on 24 May, but only lasted for one or two frames. On 29 May the camera unit was removed
and the cold eggs collected. Both adults were present in the nest area.
In mid-July the pair was found nesting approximately 450 m southwest of the original nest in Site 1197. Although the adults were not banded, there is little doubt that the pair were the same birds. Detailed notes on and photographs of the plumage of most Gyrfalcons in the study area have been taken; in this case a white female and a grey male with a dark malar bar. Two female chicks were raised to at least banding age from a clutch of three eggs. The third egg, of unknown fertility, did not hatch. The site was last visited on 7 August when the chicks, estimated at 22 and 24d of age, were banded and measurements taken.
The date of initiation of laying of the first clutch was likely between 8 and 13 May, the range for the other eight Gyrfalcon sites productive in 1985, as determined by direct observation of hatching or back-dating from estimated age of nestlings. A 35- d incubation period and 2 d between laying of eggs has been assumed (Cade and Weaver 1976; P. Trefry, Canadian Wildlife Service, personal communication). The age-estimation is believed to be accurate to+3d, and has been refined by documenting the development of known-age chicks on the study area (unpublished data). Therefore, abandonment of the initial clutch likely took place 9 to 14 d after initiation of laying, and 5 to 10d after the start of continuous incubation (beginning with the penultimate egg).
Ratcliffe (1980) pointed out that most Pere- grines will initiate replacement clutches if incubation has not proceeded longer than 7 to 10d. Beyond 10d the probability of renesting decreases rapidly, although there are documented cases of successful replacement clutches started after the initial clutch had been incubated to term (Morrison and Walton 1980).
1988
Similar back-dating techniques placed the initiation of the replacement clutch at 7 June. Thus, approximately 16d (the recycling period) elapsed between termination of incubation and effective abandonment, and the laying of the first egg of the replacement clutch. This compares with the 14-d period reported for most falcons by Cade (1982). Prairie Falcons (F. mexicanus), closely related taxonomically to the Gyrfalcon (Cade 1982), had a mean recycling time of 16 d (n= 12) (Morrison and Walton 1980). Ratcliffe reported a mean recycling time of 19 to 20d for 43 cases of relaying in Peregrine Falcons in England. A mean recycling period of 16d (range 15 to 17d, n= 5) was noted in captive Gyrfalcons (P. Trefry, personal communication).
Renesting in Gyrfalcons has been documented in captivity (Platt 1977; P. Trefry, personal communication), but little has been published on this behaviour in wild birds. In an extensive review of replacement clutches in North American raptors, Morrison and Walton (1980) did not mention the Gyrfalcon. Referring to wild birds, Cade (1960: 206) stated “if a first clutch is lost, Gyrfalcons can lay a second set with a reasonable chance for survival of the young”, but only cited one case, in the Alaskan Range (latitude approximately 63°N). Platt (1976) reported two successful Gyrfalcon renests on the Yukon North Slope (69°N). Kuyt (1980) suggested renesting in