Two grizzly bear studies : moth feeding ecology and male reproductive biology by Donnell Dee White, Jr A thesis submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy in Biological Sciences Montana State University © Copyright by Donnell Dee White, Jr (1996) Abstract: ABSTRACT Grizzly bears (Ursus arctos horribilis) consume adult army cutworm moths (Euxoa auxiliaris) from late June through mid-September on high elevation talus slopes in Glacier National Park (GNP), Montana. To better understand the importance of cutworm moths to grizzly bears in GNP, I identified the sex and age classes and minimum number of grizzly bears foraging at moth aggregation study sites, documented the timing and use patterns of grizzly bears foraging in these areas, quantified the effects of mountain climber presence on the behavior of foraging grizzly bears, determined temporal abundance patterns of moths at aggregation study sites throughout the summer, and determined body mass, total body moisture, total lipid, gross energy, and total nitrogen of moths collected throughout the summer. A minimum of 36 grizzly bears were observed 106 times feeding at 6 of 9 known army cutworm moth aggregation sites in GNP, from late-June through mid-September, 1992-1995. No bears were observed on moth sites in 1993. Bears fed on moth aggregations disproportionately more at elevations >2561 m, on slopes between 31 -45°, and on southwest-facing aspects. Lone adult and subadult grizzly bears appear to be underrepresented and overrepresented at the sites, respectively. Seasonal body weight and seasonal whole-body percentages of total moisture, total nitrogen, total lipid, and gross energy, varied linearly as a function of time. Grizzly bears foraging at moth aggregation sites are sensitive to disturbance from mountain climbers. Because alternative high-quality, late summer foods may not be available, human disturbance should be minimized at moth aggregation sites. I also evaluated testicular growth and seminiferous tubule development, and age of sexual maturity in 20 male grizzly bears killed in Montana and Wyoming between 1978 and 1992. Seminiferous tubule diameter did riot differ among the regions of each testicle sampled. Testicle mass was related linearly to age. Seminiferous tubule diameter was related non-linearly to age. Mean testicle mass, volume, and seminiferous tubule diameter were smaller in immature bears than in mature bears. Based upon the presence or absence of spermatozoa in the lumen of the seminiferous tubules, sexual maturity in grizzly bears from the continental United States is attained at approximately 5.5 years of age.  TWO GRIZZLY BEAR STUDIES: MOTH FEEDING ECOLOGY AND MALE REPRODUCTIVE BIOLOGY by Donnell Dee White, Jr. A thesis submitted in partial fulfillment o f the requirements for the degree of Doctor o f Philosophy in Biological Sciences MONTANA STATE UNIVERSITY-BOZEMAN Bozeman, Montana May 1996 3)31* yism 11 APPROVAL of a thesis submitted by Donnell Dee White, Jr. This thesis has been read by each member o f the thesis committee and has been found to be satisfactory regarding content, English usage, format, citations, bibliographic style, and consistency, and is ready for submission to the College of Graduate Studies. Harold D Picton (Signature) Date Ernest R. Vyse Approval for the Department o f Biology (Signature) Date Approved for the College o f Graduate Studies Robert L. Brown (Signature Date Ill STATEMENT FOR PERMISSION TO USE In presenting this thesis in partial fulfillment o f the requirements for a doctoral degree at Montana State University-Bozeman, I agree that the Library shall make it available to borrowers under rules o f the Library. I further agree that copying of this thesis is allowable only for scholarly purposes, consistent with "fair use" as prescribed in the U.S. Copyright Law. Requests for extensive copying or reproduction o f this thesis should be referred to University Microfilms International, 300 North Zeeb Road, Ann Arbor, Michigan 48106, to whom I have granted "the exclusive right to reproduce and distribute my dissertation in and from microform along with the non-exclusive right to reproduce and distribute my abstract in any format in whole or in part." Signature Date IV PREFACE The grizzly bear (Ursus arctos horribilis) in the contiguous United States once ranged from approximately the 100th meridian westward to California and south into Texas and Mexico (Storer and Tevis, 1955). The development o f unfavorable environmental conditions in the wake o f westward expansion and human development caused a rapid distributional recession (Lewis, 1961; Guilday, 1968). Between 1800 and 1975, grizzly bear populations in the contiguous United States receded from estimates o f over 50,000 to less than 1,000 bears (U S. Fish and Wildlife Service, 1993). Livestock depredation control, habitat deterioration, protection o f human life and property, commercial trapping, and sport hunting were leading causes (Stebler, 1972; Martinka, 1976; Brown, 1985). Currently, grizzly bears occur in less than 2% of their former range south of Canada (U S. Fish and Wildlife Service 1993). Occupied grizzly bear habitat is largely confined to 5 or 6 areas known to contain either self-perpetuating or remnant populations o f bears. These areas include portions o f 4 states—Wyoming, Montana, Idaho, and Washington. Grizzly bears may still occur in Colorado, although confirmed sightings have not occurred since 1979 (U.S. Fish and Wildlife Service 1993). The decreasing numbers and habitat o f grizzly bears resulted in the grizzly being listed as threatened under the Endangered Species Act on 28 July 1975. It is important to continue grizzly bear and habitat research to ensure adequate scientific knowledge on which to base conservation and management decisions (U.S. Fish and Wildlife Service 1993). In an effort to extend our understanding o f grizzly bear biology and ecology, I have conducted 2 separate studies focusing on grizzly bear foraging ecology and male reproduction. Therefore, this dissertation is organized into 2 parts. Part A deals with grizzly bear use o f high elevation moth aggregation sites in Glacier National Park, Montana, and part B deals with testicular histology o f male grizzly bears in Montana and Wyoming. These studies, like all studies, are not the sole work o f their author. I would like to express my appreciation to Michael Kruger, Michelle Richards, Brian Killingsworth, Kimberly Medley, Jeremy Cannon, Piney Hardiman, and Jennifer Grossenbacher for their hard work and assistance in the rugged and remote backcountry o f Glacier National Park. Their dedication and spirit o f cooperation made the Glacier Park project possible. It was a delight to be in the field with each o f them. I also thank Kirk Barnette, Jack Barringer, Bob Brastrup, Dr. Donald Heaney, Darrel Krum, John Maatta, Marko Manoukian, John Maki, Dave Phillips, and Judee Wargo, Montana State University-Bozeman County Agriculture Extension Agents, for collecting army cutworm moths for me in the autumn. Vl I have benefited both professionally and personally from the opinions, expert advice, and friendship o f my major professor Dr. Harold Picton. I also appreciate Dr. James Berardinelli for his willingness to teach me histological techniques and analysis. I thank my Graduate Committee members, Drs. Lynn Irby, Tom McMahon, Bill Kemp, Gordon Brittan, and Pete Burfening for their willingness to sit on my Graduate Committee and for their time in critically evaluating my dissertation. I am indebted to Katherine Kendall for asking me to study grizzly bears in Glacier National Park and for entrusting me with this exceptional project. I also thank Matthew Reid for his encouragement and for securing initial funding. Sincere thanks are extended to Peter Busch, President o f the Peter W. Busch Family Foundation, for his friendship and for funding both the Glacier Park and reproductive biology projects. The National Park Service, Glacier National Park, the National Biological Service, the Montana Department o f Fish, Wildlife and Parks, and the Mountain Research Center, Montana State University-Bozeman provided additional funding and support; I thank each o f them. Above all, I wish to thank my wife, Nancy. As in my other projects, she participated in every aspect o f the work to such an extent that in some ways the results are almost as much hers as mine. She was someone with whom I could discuss concerns, she was also a buffer for disappointments. When at times we had to be apart, my joy in the work diminished and I knew then how much she is the focus o f my life. -i Vll TABLE OF CONTENTS Page A. GRIZZLY BEAR USE OF HIGH ELEVATION INSECT AGGREGATION SITES IN GLACIER NATIONAL PARK, MONTANA............................................................. I Introduction.............................. I Life History O f The Army Cutworm Moth............................................ 3 Definitions......... ............ 8 Study Area....................................... 8 Methods.................................................................................... io Moth Site Characteristics.................................. 10 Bear Observation Techniques.....................................................................11 Bear Activity Budgets..................................................................................12 Scat Analysis................................................. ,............................................ 12 Moth Capture Techniques...................................... 13 Nutritional Analysis o f Moths.......................:......................................... 14 Bear Disturbance........................................................................................ 15 Statistical Analysis........... .......................................................................... 15 Results...................................................................................................................... 16 Moth Site Characteristics............................................................................16 Bear Use O fM oth Sites.............................................................................. 17 Bear Activity Patterns..................... 23 Chronology OfM oth Migrations.............................................................. 24 Nutritional Value OfM oths....................................................................... 25 Migration Potential......................................................................;.............34 Moth Density In A Talus Slope................................................................. 34 Potential Climber Disturbance....................................................... 3.5 Discussion.................................................................................................................. 36 ' Grizzly Bear Use Of Army Cutworm Moths........................................... 36 Bear Activity Patterns................................................................................. 39 Nutritional Value O f M oths........................................................................40 Migration Potential......................................................................................43 Moth Use OfTalus Slopes......................................................................... 45 An Energetic Mosaic Foraging Hypothesis..............................................46 Disturbance Potential.... ...............................................................................50 Vlll Management Implications........................................... :............................52 B. REPRODUCTIVE CHARACTERISTICS OF THE MALE GRIZZLY BEAR IN THE CONTINENTAL UNITED STATES.......................................................................................................... 55 Introduction....................................................................................... 55 Methods................................................. 57 Results...................................................................................................................... ... Discussion.......... ........ 65 LITERATURE CITED................................................. 68 LIST OF TABLES Table Page 1. Minimum number o f different grizzly bears observed on moth aggregation sites in Glacier National Park, Montana, 1992, 1994-1995........... ..................... .................................................. 19 2. Observed and expected numbers o f grizzly bears at moth aggregation sites in Glacier National Park, Montana, 1992-1995................................................................. 20 3. Number o f grizzly bear observations and proportionate representation of bear classes at moth aggregation sites in Glacier National Park, Montana, 1992-1995................................................... 20 4. Average content o f bear scats (/? = 280) collected at moth aggregation sites in Glacier National Park, Montana, 1992-1994................................................................... 21 5. Chemical composition o f army cutworm moths collected in 3 counties in Montana, 1994-1995. Numbers are percentages based on dry weight o f moths................................. 33 6. Age (yr), date killed, and testicular characteristics o f 20 male grizzly bears from Montana and Wyoming, 1978-1992.................................... 61 7. Mean (+/- SE) testicle mass, volume, and seminiferous tubule diameter (STD) for immature and mature grizzly bears in Montana and Wyoming, 1978-1992........... 64 8. Mean (+/- SE) testicular mass and volume, and seminiferous tubule diameter (STD) for mature grizzly bears killed during May through mid-July and mid-July through November in Montana and Wyoming, 1978-1992...... 65 LIST OF FIGURES Figure Page 1. Distribution o f the army cutworm moth in North America. Modified from Burton et al. (1980)...................................................................... 4 2. The life cycle o f the army cutworm moth Euxoa auxiliaris. Redrawn from Kendall (1981).......................................................... .................... 6 3. Location map of Glacier National Park, Montana and the location o f counties in Montana (shaded areas) where adult army cutworm moths were collected in late summer and early autumn, 1992-1995....................... 9 4. Temperatures on the surface and at 10, 20 and 30 cm depths within a southwest-facing talus slope (2150 m) in Glacier National Park, Montana, 3 August-10 September 1995. MST = Mountain Standard Time........................................................................... 18 5. Observed and expected frequencies o f individual grizzly bears or family groups excavating for army cutworm moths in Glacier National Park, Montana, by elevation (bottom), slope (middle) and aspect (top). A plus or minus sign denotes observed frequencies significantly greater or less than expected, respectively (P < 0.05).........................................................................22 6. Diurnal activity patterns o f 18 adult grizzly bears relative to time o f day on moth aggregation sites in Glacier National Park, Montana, 1992 and 1994-1995. MST = Mountain Standard Time............... 23 7. Seasonal occurrence o f army cutworm moths at an alpine light trap (2150 m) located adjacent to a moth aggregation site in Glacier National Park, Montana, 1994 (circles) and 1995 (squares)................................... 25 8. Seasonal occurrence o f army cutworm moths at sex-attractant traps located adjacent to a wheat field in 6 counties in north- central and south-central Montana, 1992-1995. Pondera county, closed squares; Liberty county, closed triangles; Hill county, closed squares; Chouteau county, closed circles; Fergus county, open triangles, and Carbon county, open circles.:.................. 26 9. Wet mass o f moths captured at an alpine light trap (2150 m) located adjacent to a moth aggregation site in Glacier National Park, Montana, 1994 (bottom), 1995 (top)....................................... 27 10. Total body moisture o f moths captured at an alpine light trap (2150 m) located adjacent to a moth aggregation site in Glacier National Park, Montana, 1994 (bottom), 1995 (top)................................................................ :......................................................28 11. Total lipid (ether extract) o f moths captured at an alpine light trap (2150 m) located adjacent to a moth aggregation site in Glacier National Park, Montana, 1994 (bottom), 1995 (top)........................ 29 12. Gross energy content o f moths captured at an alpine light trap (2150 m) located adjacent to a moth aggregation site in Glacier National Park, Montana, 1994 (bottom), 1995 (top)........................ 30 13. Total nitrogen (crude protein) o f moths captured at an alpine light trap (2150 m) located adjacent to a moth aggregation site in Glacier National Park, Montana, 1994 (bottom), 1995 (top)............................................................................................................. 31 14. Activities o f 18 adult grizzly bears foraging on a moth aggregation site undisturbed or disturbed by mountain climbers in Glacier National Park, Montana, 1992.......................................... 35 15. Distribution o f potential moth aggregation sites in Glacier National Park, Montana........................................................................................ 54 16. Age variation in testicle mass for male grizzly bears from Montana and Wyoming, 1978-1992.................................................................. 62 17. Growth curve o f seminiferous tubule diameters (STD) in male grizzly bears from Montana and Wyoming, 1978-1992........................ 63 ABSTRACT Gnzzly bears (Ursus arctos horribilis) consume adult army cutworm moths (Euxoa auxiliaris) from late June through mid-September on high elevation talus slopes in Glacier National Park (GNP), Montana. To better understand the importance of cutworm moths to grizzly bears in GNP, I identified the sex and age classes and minimum number o f grizzly bears foraging at moth aggregation study sites, documented the timing and use patterns o f grizzly bears foraging in these areas, quantified the effects o f mountain climber presence on the behavior o f foraging grizzly bears, determined temporal abundance patterns o f moths at aggregation study sites throughout the summer, and determined body mass, total body moisture, total lipid, gross energy, and total nitrogen o f moths collected throughout the summer. A minimum o f 36 grizzly bears were observed 106 times feeding at 6 o f 9 known army cutworm moth aggregation sites in GNP, from Iate-June through mid-September, 1992-1995. No bears were observed on moth sites in 1993. Bears fed on moth aggregations disproportionately more at elevations >2561 m, on slopes between 3 1 - 45°, and on southwest-facing aspects. Lone adult and subadult grizzly bears appear to be underrepresented and overrepresented at the sites, respectively. Seasonal body weight and seasonal whole-body percentages o f total moisture, total nitrogen, total lipid, and gross energy, varied linearly as a function o f time. Grizzly bears foraging at moth aggregation sites are sensitive to disturbance from mountain climbers. Because alternative high-quality, late summer foods may not be available, human disturbance should be minimized at moth aggregation sites. I also evaluated testicular growth and seminiferous tubule development, and age o f sexual maturity in 20 male grizzly bears killed in Montana and Wyoming between 1978 and 1992. Seminiferous tubule diameter did riot differ among the regions o f each testicle sampled. Testicle mass was related linearly to age. Seminiferous tubule diameter was related non-linearly to age. Mean testicle mass, volume, and seminiferous tubule diameter were smaller in immature bears than in mature bears. Based upon the presence or absence o f spermatozoa in the lumen o f the seminiferous tubules, sexual maturity in grizzly bears from the continental United States is attained at approximately 5.5 years o f age. PARTA GRIZZLY BEAR USE OF HIGH ELEVATION MOTH AGGREGATION SITES IN GLACIER NATIONAL PARK, MONTANA Introduction Grizzly bears (Ursus arctos hofribilis) feeding on insect aggregations excavated from alpine talus slopes have been documented in several areas in the northern Rocky Mountains: the Mission Mountains (Chapman 1954, Chapman et al. 1955, Servheen 1983, Klaver et al. 1986), Scapegoat Wilderness (Sumner and Craighead 1973, Craighead et al. 1982), and mountains o f the Rocky Mountain East Front (Aune and Kasworm 1989), Montana; and in the AbsarOka Mountains, Wyoming (Mattson et al. 1991, French et al. 1994, O'Brien and Lindzey 1994). Grizzly bears consume army cutworm moths (Euxoa auxiliaris) (Lepidoptera: Noctuidae) and ladybird beetles {Coccinnella and Hippodamia spp.) (Coleoptera: Coccinellidae) in the Mission Mountains and army cutworm moths in the Scapegoat Wilderness, Rocky Mountain East Front, and Absaroka Mountains. Additionally, Ustinov (1965) has recorded bears eating aggregations o f caddis flies (Trichoptera) along the shores o f Lake Baikal, Russia, and Gurney (1953) observed bears feeding on fossil grasshoppers (Orthoptera) melted out o f glaciers. 2 Anecdotal accounts o f grizzly bears frequenting high elevation (>2000 m) talus slopes in Glacier National Park (GNP), Montana, are common in Park records. The first recorded report o f bears observed above timberline in GNP was made by climbers on Mt. Cleveland (3190 m) in 1933. These climbers found "signs o f a grizzly within 1000 feet o f the summit" and the climbers considered this "the most remarkable thing in the whole trip." Though the climbers were the ChiefNaturalist o f GNP and a seasonal staff member, they had no plausible explanation for what the bear was doing at that elevation. Moth remains were confirmed in bear feces (scats) collected from several o f GNP s tallest mountain peaks in the early to mid-1980's (Katherine C. Kendall, unpub. data). From 1989 - 1991, 5 radiotelemetered grizzly bears spent several weeks feeding on army cutworm moths at 2100 - 2800 m elevation on mountains along the east side o f GNP. These bears initiated feeding on moths as early as the third week o f June and continued until mid to late August or early September (Daniel Carney, U.S. Fish and Wildlife Service, Browning, Montana, unpub. data). In this study, I investigated grizzly bear use o f army cutworm moths on high elevation talus slopes in GNP. Prior to my investigation, information on grizzly bear use o f army cutworm moths was limited to 4 studies; 3 in the Greater Yellowstone Ecosystem (Mattson etal. 1991; French et al. 1994; O'Brien and Lindzey 1994) and I in the Mission Mountains, Montana (Klaver et al. 1986). 3 My goals were to determine the significance o f high elevation army cutworm moth aggregations to the grizzly bear population in Glacier National Park and to quantify what physical, biological, and social parameters influence grizzly bear use o f these areas. To meet these goals I pursued 5 specific objectives: (I) to identify the sex and age classes and minimum number o f grizzly bears foraging at moth aggregation study sites, (2) to document the timing and use patterns o f grizzly bears foraging in these areas, (3) to determine temporal abundance patterns o f moths at aggregation study sites throughout the summer, (4) to determine total moisture, total nitrogen, total lipid, gross energy, and body mass o f moths collected throughout the summer and early autumn, and (5) to quantify the effects o f mountain climber presence on the behavior o f foraging grizzly bears. Life History O fThe Armv Cutworm Moth The army cutworm moth is distributed throughout the semiarid region of the Great Plains, extending as far east as Kansas, south to southern New Mexico, west to California, and north to central Alberta (Fig. I). Since 1929, the army cutworm has been reported in all states west o f the Mississippi River except Louisiana (Burton et al. 1980). The army cutworm moth is also widespread throughout Alberta and is commonly abundant as far north as the Peace River District (Strickland 1942, Burton et al. 1980). 4 Figure I . Distribution o f the army cutworm moth in North America. Modified from Burton et al. (1980). 5 The life history o f the army cutworm moth is remarkable and required decades for biologists to piece together. Because o f the abundance o f the moths in the spring and autumn on the Great Plains, Gillette ( 1904) and Johnson (1905) incorrectly proposed that the army cutworm moth was bivoltine. Cooley (1916) reared army cutworm moths in outdoor cages on the Plains during the summer in Montana, but few individuals survived. This led Cook (1927) to propose that adult army cutworm moths aestivated (i.e., were inactive) under matted vegetation and other debris on the Plains to escape the heat o f summer. This "heat-escape hypothesis" also explained why army cutworm moths vanished during summer on the Plains. Simple but elegant experiments conducted by Pepper (1932) determined that army cutworm moths could be kept alive during the summer on the Plains by storing them at cool temperatures. . In May o f this same year. Pepper (1932) observed a swarm of army cutworm moths flying southwesterly near Bozeman, Montana. Pepper was the first to hypothesize that these moths migrated to high elevations for the summer. Pruess (1967) developed a more detailed and cohesive theory o f army cutworm moth migration based on 4 observations: I) the absence o f adult army cutworm moths on the Plains during the summer in Nebraska, 2) the long distance flight ability o f this moth (as demonstrated by experimentation in the mid-1960's), 3) numerous records o f the moths at high elevations during the summer, and 4) moths captured on the Plains in the autumn had higher body fat reserves than those collected in spring indicating that they fed during the summer and did not aestivate as previously accepted. Pruess 6 and Pruess (1971) further documented nocturnal unidirectional migration o f army cutworm moths at North Platte, Nebraska, and Laramie, Wyoming, using telescopes focused on the moon. This and the work that has followed has clarified the life history o f this complex animal (Fig. 2). The army cutworm moth is holometabolous. The eggs are laid in the Figure 2. The life cycle o f the army cutworm moth. Redrawn from Kendall 1981. 7 soil o f the Great Plains in the autumn and develop to the first or second larval instar before hibernation (Johnson 1905, Cooley 1916, Strickland 1916, Burton et al. 1980, D. Kendall 1981). In spring, the larvae commence to feed on a variety o f plants such as alfalfa and small grains (Burton et al. 1980, Morrill 1991). I f development is not interrupted by low temperatures and the larva continue to feed, the larval stage may last for only 10 days (Rockbume and Lafontaine 1976). The larval stage can, however, last longer: 25 to 32 days in Kansas (Walkden 1950) and 43 to 63 days in Montana (Cooley 1916). When abundant and short o f food, the larvae will move en masse to adjacent fields, thus the name "army" cutworm. After a total o f 6 or 7 instars from egg to last molt, pupation occurs in underground cells (Snow 1925, Seamans 1927, Burton et al. 1980). In early summer the newly emerged adult moths enter a migratory phase and fly west into the Rocky Mountains (Pepper 1932, Walken 1950, Chapman et al. 1955, Pruess 1967, Hardwick and Lefkovitch 1971, Burton et al. 1980) where they spend the summer. It is here, while inhabiting the interstia o f talus slopes that the moths are excavated and consumed by grizzly bears. They do not become reproductively active until autumn when they return to the plains. On the plains the females enter a settling phase and oviposit approximately 2000 eggs per individual into the soil (Walkden 1950, Burton et al. 1980). The army cutworm moth is univoltine (Cooley 1908, Seamans 1927, Burton et al. 1980, Kendall 1981). 8 Definitions Several terms used loosely in the ecological literature are defined below to clarify my use o f the terms: 1. Migration is the act o f moving from one spatial unit to another (Baker 1982). 2. Remigration is the act o f migrating to a spatial unit that is environmentally similar to a spatial unit that has been visited before. 3. Movement is defined as simply a change in position. Study Area GNP is located in northwestern Montana adjacent to the Canadian border (Fig. 3). Two mountain ranges dominate the park: the Livingston Range, located on the west side o f the park, extends north from the Canadian border southward to Lake - McDonald; and the Lewis Range, on the east side, extends the entire length of the park. The continental divide bisects the park following the crest o f the Lewis Range northward to about 18 km south of the Canadian border where it turns westward to follow the crest o f the Livingston Range into Canada (Carrara and McGimsey 1981). About one third o f GNP s 410,000 ha occurs above timberline. Much o f the remainder o f the park is forested with scattered meadows, which occur most frequently on the eastern side o f the park. Reliefis precipitous. Deep glaciated valleys and basins divide large, rugged mountain massifs. Many valleys are 1800 m below their surrounding summits (Carrara and McGimsey 1981). Elevations in the park range from 948 m at 9 Figure 3. Location map o f Glacier National Park, Montana and the location of counties in Montana (shaded areas) where adult army cutworm moths were collected in late summer and early autumn, 1992-1995. the Middle Fork-North Fork Flathead River confluence to 3,190 m atop Mount Cleveland (Finklin 1986). The mountains o f GNP presumably originated with the uplift and folding of sedimentary rocks o f the Precambrian Belt Supergroup during the late Mesozoic. These strata are mostly reddish brown and greenish gray argillites with some quartzites (Carrara and McGimsey 1981). Glaciers shaped the terrain to its present appearance during the Pleistocene (Alden 1953, Dyson 1966, Alt and Hyndman 1991). 10 The climate is Continental with Pacific Maritime modifications, particularly on the western side o f the park (Dightman 1967, Carrara and McGimsey 1981). The alpine climate is characterized by frequent strong (>66 km/h) winds, typically westerly in winter (December-February) and southwesterly during summer (July-August) (Finklin 1986). Precipitation generally increases with elevation, but decreases rapidly with horizontal distance near and beyond the eastern edge o f the park (Finklin 1986). Mean July temperatures range from 15 to 17°C in the lower valleys (Finklin 1986). Summer afternoon temperatures usually decrease with increasing elevation, at an average lapse rate o f 7.5 to 8°C per 1000 m (Finklin 1986). Methods Moth Site Characteristics I visited 7 alpine moth aggregation sites in GNP before, during, and after seasonal bear use, which occurred from late June to mid-September. Data collected at moth aggregation sites included measurements and descriptions o f site characteristics (e.g., elevation, slope, aspect) and bear sign (e.g., dig dimensions and dig-site features, bed dimensions and locations). To quantify the thermal environment o f a talus slope, I placed automatic temperature loggers (HOBO TEMP, Onset Computer Corporation, Pocasset, MA) , programmed to measure temperature every 30 minutes for 36 days at the surface and 11 at 3 depths within a southwest-facing talus slope: 10 cm, 20 cm and 30 cm. The thermometers were buried 3 August 1995 and removed 10 September 1995. Bear Observation Techniques I observed bear foraging activity at alpine moth aggregation sites using variable power spotting scopes or telescopes at distances o f 250 m to 2 km. Moth aggregation study sites were chosen for observability and accessibility, intensity o f bear use, minimal bear disturbance by researchers, and researcher safety. Observation posts were chosen on the basis o f bear observability and bear avoidance, researcher concealment, and prevailing wind direction. Bears were identified as to species and sex/age classes using established guidelines for field identification o f bears (Burkholder 1959, Meehan 1961, Woodgerd 1964, Craighead et al. 1974, Martinka 1974, Egbert and Stokes 1976, Atwell et al. 1980, French et al. 1994, O Brien and Lindzey 1994). Subadults were distinguished by size and relative body proportions. Repeated observations o f the same bears on a talus slope allowed me to identify many bears individually. Pelage color and shedding patterns, size, conformation, and deformities were the distinguishing characteristics used. Sex, age, family groupings, and behavior traits were used as distinguishing factors, which led to a high confidence in the identification o f individuals. Location o f each bear observed foraging on moths was recorded on U S. Geological Service 7.5 minute series topographic maps using Universal Transverse Mercator (UTM) coordinates, 12 Bear Activity Budgets I determined activity budgets during moth feeding episodes using behavior scans during direct observations and by reviewing 16 mm film records. Moth aggregation sites were scanned for a 15 minute time period for each hour o f observation to determine the activity patterns o f adult female, lone adult, and subadult grizzly bears observed feeding on moths. At 10 second intervals throughout the 15 minute sampling period, the category o f activity o f a focal bear was recorded (Altmann 1974). Behaviors were classified as: foraging on moths, moving, sleeping, loafing-awake, foraging on vegetation, nursing, and defense (i.e., head down and swaying back and forth, ears laid back on sides o f head, charges, etc.). The percent occurrence for a , given behavior was determined by dividing the number o f instantaneous observations recorded for that behavior by the total number o f behavioral observations. Other data collected included number, temporal use patterns, and intraspecific behavioral interactions. Scat Analysis Bear scats were collected from moth aggregation sites and their locations (UTM coordinates) recorded. Scats were analyzed for percent frequency, volume, and ingested volume o f items consumed. Percent volume ingested was estimated by using correction indexes for differential digestibility o f items (Hewitt 1989). These methods . ; ' were the same as described by Mattson et al. (1991). 13 Moth Capture Techniques I used 3 different methods to collect moths. A standard blacklight trap using an 8 watt Sylvania F8T5/BL fluorescent tube was the primary method used in the alpine to capture moths from late June through early September, 1994 and 1995. Each blacklight was powered by a 12 volt DC marine battery. The number o f moths captured per trap night was used as an index to moth abundance on moth aggregation sites throughout the summer. The killing agent was 10% formaldehyde and detergent. Traps were equipped with automatic light switches. Traps were emptied and a fresh battery exchanged approximately every 14 days. Moths were also excavated in the alpine by hand from I talus slope on 15 August 1992, 17 August 1994 and 21 August 1995. A 40 m x 40 m grid was established each year using a 10 m-long rope. Sixteen sampling stations with an area o f 0.25 m2 each were established and the number o f moths were determined at each station each year. The talus was excavated until sandy soil or snow/ice prohibited digging further. Most o f the moths were located within the top 15 cm. In 1992,1 visually selected 5 areas on the talus slope where density o f moths appeared to be especially high and sampled these areas using the same method. To compare nutritional composition o f moths captured in the alpine o f GNP with moths captured during remigration on the plains, sex-attractant traps (Unitrap, Phero Tech. Inc. British Columbia, Canada) baited with army cutworm lure (Terochem Laboratories LTD, Edmonton, Alberta, Canada) were used in 6 counties o f north and south-central Montana in late summer and early autumn (mid-August to mid-October or early November) (Fig. 3). Prior to placement in the field each trap was positioned near the top o f a wood stake (approximately 120 by 4 by I cm) and tied securely with plastic baling twine. At each trap site, 2 traps, spaced IOm apart, were placed adjacent to a wheat field. Locations with high road grade, deep ditches, or any other feature that might alter air flow pattern were avoided. At the trap site, the stake with the attached trap was driven into the ground to a depth such that the trap was 1.5 m above ground level. Traps were checked weekly and captured moths were removed, counted, and frozen in plastic bags. Trap construction, placement, and monitoring was conducted by Montana State University-Bozeman Agriculture Extension Agents. Nutritional Analysis O f Moths To assess the nutritional benefits o f moth foraging by grizzly bears, gross energy and total nitrogen (crude protein) were determined on moths collected with bomb calorimetry and macro-Kjeldahl methods, respectively. Total lipid (ether extract) was also estimated using a Soxhlet apparatus and total moisture was determined via weight difference before and after drying. Except for total moisture, nutritional analyses are reported on a dry weight basis. Moths were stored frozen in dry sealed containers to minimize dehydration prior to analysis. At least 20 moths were used for each analysis for each date. All analyses were conducted at the Analytical Laboratory, Montana State University-Bozeman. 15 To determine the number o f moths a bear could potentially consume in a day, I multiplied the mean number o f moths in a bear scat taken from a moth site by the mean number o f times a bear defecated in a day. Since little or no moth foraging occurred at night or before 0600, a day here refers to the 16 hour observation period bears were observable on a moth site. Bear Disturbance On 7 separate occasions, I observed mountain climbers disturb grizzly bears as the bears fed at moth aggregation sites. Potential effects o f disturbance was quantified by conducting activity budgets on bears as they reacted to climber presence. Statistical Analysis I used the log-likelihood ratio (G) to test goodness o f fit between observed and expected frequencies (Zar 1984) o f bear sightings among bear sex and age classes and elevation, aspect, and slope classes. Bonferroni confidence intervals were calculated to determine which classes differed significantly between observed and expected values (Byers et al. 1984), given a significant difference in overall frequency distributions. The expected distribution o f sightings among bear classes was derived from GNP employee bear sighting reports accumulated from January 1967 through December 1994 (Katherine C. Kendall, pers. commun., 1995). Lone adults, subadults, females with cubs o f the year, and females with yearling or 2.5 year old cubs 16 comprised 60.5, 13.4, 14.4, and 11.7%, respectively, o f 4576 bear observations made during this time period. Expected distributions o f observations for elevation, aspect, and slope classes were derived for 20,000 random points using the geographical information system digital elevation model for GNP. Random point determination was restricted to areas above 2141 m, the minimum elevation o f documented sites o f bears feeding on moths. Seasonal changes in nutritional content o f moths was analyzed using least-squares linear regression. In all analyses performed, I accepted Alpha < 0.05 as significant. . Results Moth Site Characteristics I identified a total o f 9 sites used by bears to feed on army cutworm moth aggregations. I visited 7 o f these sites a total o f 38 times from 1992 to 1995. Seven sites were located in glacial cirques on talus slopes immediately below steeper headwalls, 2 sites were at or near the summits o f their respective massifs. The talus slopes at the sites I visited were active and deposition o f rock debris commonly occurred during my visits, particularly in the early morning hours (0400 - 0600). None o f the talus interstitia were filled with debris. The angular rocks o f the talus slopes ranged from 2 to > 100 cm in size. Elevations at the 9 sites averaged 2621 + 198 m (2141-3129 m), and slopes were 37 + 9.5° (8 - 66°). Five o f the sites were on south aspects, and 4 were on western aspects. Talus slopes used for feeding by bears were essentially devoid o f vegetation except for strings o f vegetation associated with water seepage areas; even the lichen cover on rocks was sparse. As the moth sites were located at or near the summits o f their associated massifs, few alpine tundra-covered benches occurred above the moth sites. Below the moth sites, however, plateaus and benches occurred and were dominated by forbs, most commonly Erythronium grandiflorum, Polygonum bistortoides, and Dryas hookeriana. Graminoids, including Poa spp. and Carex spp. were also common on these moister, lower elevation benches. Temperatures o f talus slopes varied widely with depth. Mean surface temperatures o f a southwest-facing talus slope (2300 m) during 3 August-10 September 1995 fluctuated widely (e g., 3-22°C), reaching 22 °C in the early afternoon (Fig. 4). At 30 cm deep into the talus debris, however, temperature fluctuations were much reduced, ranging only a few degrees daily (Fig. 4), and averaging about 5°C over a 24 hour period. Bear Use OfM oth Sites Cutworm moths predominated the diet o f bears occupying talus slopes. I observed a minimum of 36 different grizzly bears feeding at 6 of 9 known moth aggregation sites in GNP from Iate-June through mid-September, 1992 - 1995 (Table I). No bears were observed on moth sites in 1993. Subadults, lone adults, adult females with cubs o f the year, and adult females with yearling cubs were observed. O f 106 total bear 1 8 Surface 10 cm 20 cm 30 cm 8 10 12 14 16 18 20 22 24 Hour of Day (MST) Figure 4. Temperatures on the surface and at 10, 20, and 30 cm depths within a southwest-facing talus slope (2150 m) in Glacier National Park, Montana, 3 August-10 September, 1995. MST = Mountain Standard Time. 19 observations (family groups were counted as a single observation), no black bears (U. americanus) were observed. Table I . Minimum number o f different grizzly bears observed on moth aggregation sites in Glacier National Park, Montana, 1992, 1994-1995*. Moth Site n, RFa Cb Y lc Y2d SA6 LAf Twin Peaks 16 4 3 3 I 0 5 Two Medicine 4 I I I 0 0 I Northern Exposure 4 I 3 0 0 0 . 0 Pinchot Creek 6 I I 0 0 0 4 Medicine Grizzly 3 I 0 I 0 0 I Triple Divide 3 0 0 0 0 3 0 36 8 8 5 I 3 11 *Only 2 bears (a female and a 1.5 year old cub) in 1994 and 3 bears (a female and 2 1.5 year old cubs) in 1995 were observed on moth sites. aRF = reproductive females (i.e., females accompanied by cubs). bC = cubs-of-the-year. cY l = 1.5 year old cubs. dY2 = 2.5 year old cubs. eSA = subadults. fLA = lone adults. Frequencies o f observed and expected grizzly bear classes differed significantly (G = 19.80, d f = 3, P < 0.05) (Table 2). This difference was attributable to overrepresentation o f subadults and underrepresentation of lone adults. The frequency of observations among bear classes did not differ (G = 0.17, d f = 2, P = 0.92) between the time periods analyzed (15 June - 15 July, 16 July - 15 August, 16 August - 15 September) (Table 3). 20 Table 2. Observed and expected numbers o f grizzly bears at moth aggregation sites in Glacier National Park, Montana, 1992-1995. Observed Population expected Subadult 31a Ub Lone adult 34a 65b Female with cub(s) o f year 24a 15a Female with yearling(s) 17a 13a Values in rows followed by a different letter are significantly different (P < 0.05). Table 3. Number o f grizzly bear observations and proportionate representation of bear classes at moth aggregation sites in Glacier National Park, Montana, 1992-1995. Bear class proportion No. o f bear observations Subadults Lone adults Females with young 15 June -1 5 July 18 0.333* 0.333* .0.333* 16 July - 15 August 48 0.33 0.35 0.31 16 August -1 5 September 40 0.3 0.35 0.35 Values are proportions o f total bear observations, by time period. * Sample sizes <24. Based upon examination o f 280 bear scats collected from moth aggregation sites in 1992, 1994, and 1995, bears ate mainly army cutworm moths (Table 4). Moths appear to be highly digestible; essentially all moth parts are digested except for their exoskeletons. Graminoids were also eaten, particularly, Poa spp. and Carex spp., but little grazing was observed in adjacent mesic areas. 2 1 Table 4. Average content o f bear scats (n — 280) collected at moth aggregation sites in 1992 - 1994 in Glacier National Park, Montana. % Frequency % Volume %Ingested Volume8 Moths 78.3 67.2 95.2 Debris (mostly rocks) 58.4 16.3 Grass/Sedge 23.1 15.3 3.8 Ants (Formicidae) 5.8 0.9 0.9 Vaccinium globulare 6.5 0.3 0.1 Tngested volume was estimated by applying correction factors o f Hewitt (1989), excluding debris. Bear excavations were oval in shape and 0.1 - 0.8 m, typically 0.2 - 0.5 m deep. Mean total volume o f talus debris excavated for 30 pits was 0.22 m3. When 2 or more bears or family groups fed at a moth site, they were separated by 50 - 100 m, with the exception o f paired subadults. Bear use was concentrated within areas o f about 200 x 200 m. I found up to 100+ scats in these areas; others were undoubtedly covered as bears dug. Bears observed digging for moths (n = 106) were not distributed randomly with respect to elevation, slope, or aspect (Fig. 5). Fifteen bedding sites were found at 3 moth sites. These beds were excavated in the open talus feeding sites (/? = 5) or adjacent to large boulders (« = 10). These beds were typically shallow depressions (0.05 - 0.1 m) shaped from scree or snow. Few scats were found around the beds. H Observed I I Expected SW W Aspect 16-30 31-45 46-60 >60 Slope (deg.) 75 -I R n - 2141-2350 2351-2560 2561-2770 2771-2980 Elevation (m) Figure 5 Observed and expected frequencies o f individual grizzly bears or family groups excavating for army cutworm moths in Glacier National Park, Montana, by elevation (bottom), slope (middle) and aspect (top). A plus sign denotes observed frequencies significantly greater or less than expected, respectively (P < 0.05). 23 Bear Activity Patterns I determined the type o f activity engaged in by 18 adult grizzly bears for 104 15-minute intervals. Bears foraged heavily on moths from 0600 - 1200, commonly sleeping on site between 1300 - 1800, then foraging for moths again in the early evening (Fig. 6). While foraging for moths, a bear typically stayed in I location for up < D E o co ■c 0 CL 2 CL 1 600 800 1000 1200 1400 1600 1800 2000 Time of Day (MST) Figure 6. Diurnal activity patterns o f 18 adult grizzly bears relative to time of day on moth aggregation sites in Glacier National Park, Montana, 1992, and 1994-1995. MST = Mountain Standard Time. to 15 minutes before moving to a new location, which was usually within a 1-2 m o f the previous excavation. A mean o f 67% o f each day was spent foraging for moths. 24 Moving on, off, or within the site tended to be more common after 1200. Little movement o f the bears occurred before 1200. Defensive behaviors were observed only during the early morning hours (Fig. 6). The amount o f time during the summer devoted to moth foraging varied widely among bears observed. I observed I female with a yearling cub use I moth aggregation site for at least 32 days. Five lone adults were observed foraging for moths on 3 different moth aggregation sites for at least 28, 25, 22,21, and 20 days each. Two females with cubs-of-the-year were observed for only 3 days each. Because it was not possible to observe every moth aggregation site every day, these numbers should be considered conservative estimates o f the length o f time grizzly bears utilize moth aggregation sites in GNP. Chronology O fM oth Migrations Army cutworms moths arrived in the alpine o f GNP in early July in 1994 and in late June in 1995 (Fig. 7). Moth abundance peaked in late July in 1994 and in mid July in 1995. I did not capture any army cutworm moths after 10 August in 1994 or after 30 , July in 1995 (Fig. 7). The earliest moths were captured on the plains was in early August in Pondera County, Montana (Fig. 8). Peak seasonal abundance, however, did not normally occur until after the first week o f September in most years in Pondera, Chouteau, Liberty, I Date Figure 7. Seasonal occurrence o f army cutworm moths at an alpine light trap (2150 m) located adjacent to a moth aggregation site in Glacier National Park, Montana, 1994 (circles) and 1995 (squares). Hill, and Fergus counties o f north-central Montana (Fig. 8). In Carbon county in south-central Montana (Fig. 2, Fig. 8), peak moth abundance occurred after the first week o f September in 1993, but not until the first week o f October in 1994 (Fig. 8). Nutritional Value OfM oths Cutworm moths showed a marked increase in wet mass, total moisture, total lipid, and gross energy, over the course o f the summer. Upon arrival in the alpine of GNP, army cutworm moths weigh about 0.10 g each. When they leave the alpine to remigrate to the plains in late summer, they weigh about 0.16 g each, a 60% increase 26 O 60- O 40- O 20- I m) talus debris. Given the 50 nonuniform distribution o f moths on a talus slope, it is useful to view a moth aggregation site as an energetic mosaic: high energy patches interspersed within a much larger low energy matrix. Because the energy distribution on a talus slope can be mapped and grizzly bear foraging behavior can be easily observed and mapped, grizzly bear use o f moth aggregation sites may provide an adequate model to empirically test the relationships between forage quantity and quality and the seasonal movements and habitat use by grizzly bears. It should be noted that using total digestible energy, protein content, or some other measure o f grizzly bear forage quality is potentially misleading and could possibly lead to a misunderstanding o f grizzly bear foraging strategies and habitat use. Foods may be chosen on the basis o f vitamin or mineral content or other criteria (Hamer and Herrero 1983, Hamilton and Bunnell 1986). Presence o f protective or defensive agents could cause avoidance o f certain food items by grizzlies (Robbins 1983) as well. In considering the factors that influence grizzly bear selection o f habitat, one would do well to heed the warning given by Himstein (1984): "The temptation to fall back on common sense and conclude that animals are adaptive, i.e., doing what profits them most, had best be resisted, for adaptation is at best a question, not an answer." Disturbance Potential Human influence on the selection o f food and feeding sites in a protected area such as GNP, primarily involves supplementation (and subsequent attraction) and 51 disturbance (and subsequent loss o f feeding opportunities). Glacier National Park is a favorite destination for mountaineers. About one third o f the Park's 410,000 ha occurs above timberline. Subsequently, many access routes to these alpine vistas take climbers near or through moth aggregation sites. This potential for human disturbance o f foraging grizzly bears concerns managers. Although difficult to quantify, human disturbance may reduce food intake through interruption o f foraging bouts or by displacement from feeding areas, and may increase energy expenditure from additional time moving. Reduced food intake and increased energy expenditure could potentially affect the ability o f bears to acquire nutrient reserves for successful reproduction. Furthermore, nutrient reserves acquired during the late summer and early autumn may influence winter survival. Bear sensitivity to human presence has been documented in several areas; Brown bears are known to avoid feeding areas when more than 4 people were present at McNeil River Falls, Alaska (Stonorov and Stokes 1972). Given the history o f human intrusiveness in this area, these bears may be more habituated to humans and consequently should display less o f a sensitivity to human presence than grizzly bears feeding on moths in GNP. Even at distances greater than 1000 m, bears may be aware o f human presence due to wind currents and noise resulting in a disruption of normal feeding activities. , Because o f potential conflicts between bears and humans, Klaver et al. (1986) recommended closure o f McDonald Peak in the Missions Mountains, Montana, 52 commensurate with bear arrival (usually late July) and departure (I October). Closure began in 1981 and remains in effect to date. The goal o f this action was to reduce human-bear interactions and conflicts, bear habituation, disturbance o f females with cubs, and to prevent teaching cubs human habituation tendencies (Klaver et al. 1986). The plan received strong public support, in part due to an effective public relations and education program. The importance o f climber disturbance can be expressed as amount o f lost energy due to interrupted feeding time. If a grizzly bear can consume 40,000 moths per day, or 2500 moths for every hour o f feeding time, every minute a bear does not eat due to disturbance costs the bear 42 moths. At 0.5 kcal per moth in late August, disturbance costs the bear 21 kcal per minute, not counting the energy expended in defensive behaviors and leaving the moth site. Management Implications Gnzzly bear habitat loss and population declines occurring concurrently with increasing human populations have been well documented in Europe and North America (U S. Fish and Wildlife Service 1993, Roben 1980, Cowan 1972, Mysterud 1980, Buchalczyk 1980, Markov 1980, Roth 1976, Elgmork 1978, Grachev 1976, Kaal 1976, Vereschagin 1976, Pearson 1975, Curry-Lindahl 1972, Stebler 1972, Zunino and Herrero 1971, Guilday 1968, Buss 1956, and Storer and Trevis 1955). Human activities that occur within grizzly bear range have and will continue to have 53 effects, however subtle, on grizzly bears (Aune and Kasworm 1989). The recommendations included here are suggested ways to ensure that grizzly bears continue to have access to moth sites and that they can feed on these sites uninterrupted by humans. I f grizzly bears in the Northern Continental Divide Ecosystem in general and Glacier National Park in particular are to survive in the long term, their critical biological needs and important seasonal foraging areas must be identified and given high management priority. High elevation army cutworm moth aggregation sites may prove to be a natural food that is important in maintaining grizzly bears in Glacier National Park and careful management o f these sites could therefore be critical for their conservation. Although present rates o f disturbance o f grizzly bears on moth sites are probably low, increased human activity could affect the energetic budgets o f foraging bears. Establishing controlled-human-use buffer zones around moth sites used by bears, determining alternate climbing routes for mountain peaks with moth sites on them, and establishing minimum flight elevation regulations for aircraft would not only reduce the chances for bear-human conflicts but would also decrease human disturbance on foraging grizzly bears. A geographical information system was used to obtain a Park-wide perspective o f potential moth site locations. Using elevation (2131-3114 m), slope (21-54°), and aspect (0-86°, 165-360° (East = 0°, North = 90°)) o f the 9 moth study sites, I 54 determined other locations in the Park that have similar geographical and geological characteristics. According to my model, 10.5% o f Glacier Park (about 43,000 ha) is potential army cutworm moth habitat (Fig. 15). Whether these areas are occupied by moths or if grizzly bears are feeding at these sites is unknown. Figure 15. Distribution o f potential moth aggregation sites in Glacier National Park, Montana. 55 PA RTB REPRODUCTIVE CHARACTERISTICS OF THE MALE GRIZZLY BEAR IN THE CONTINENTAL UNITED STATES Introduction Bears in northern temperate environments have a well-defined breeding season extending from May until early July. During the breeding season, testicular weights are higher than any other time o f year. By September, testicles regress in size and by late September azoospermia occurs, although spermatozoa may be present in the epididymides (Erickson et al. 1964, 1968; Pearson 1975; Horan et al. 1993; Garshelis and Hellgren 1994). Testicular regression continues throughout the non-breeding season. Although the pattern or extent o f regression differs among individual bears, by I October, the diameter o f the seminiferous tubules is markedly decreased (Erickson et al. 1964, 1968; Pearson 1975; McMillin et al. 1976; Reynolds and Beecham 1980; Tsubota and Kanagawa 1989). By mid- to late October or early November, testicular regression is nearly complete; by mid-November the testicles are infiltrated by adipose tissue and loose, fibrous, connective tissue (Erickson et al- 1964, 1968; Pearson 1975). The tubules at this time are small and show no lumen, and the germinal epithelium consists only o f Sertoli cells (Erickson et al. 1964, 1968). 56 Testicular weights in the early winter months are the lowest found in the mature bear throughout the year (Erickson et al. 1964, 1968; Pearson 1975). Testicular recrudescence, which is well underway before emergence from winter denning (Erickson et al. 1964, Palmer et al. 1988, Garshelis and Hellgren 1994), generally involves seminiferous tubule enlargement and increased Leydig cell activity (Erickson et al. 1964, 1968; Pearson 1975). Active Leydig cell development and spermatogenesis, with the presence o f spermatozoa in the epididymides, have been noted in bears killed in late May and early June to the middle or end o f July (Erickson et al. 1964). Fully formed spermatozoa are present in seminiferous tubules and epididymides at least I month before and several months after the seasonal period o f GStrus in females in the European brown bear (U. a. arctos, Dittrich and Kronberger 1963), American black bear (Erickson et al. 1964), Alaskan brown bear (U. a. middendorffi, Erickson et al. 1968), grizzly bear (Pearson 1975), and the Hokkaido brown bear (U. a. yesoensis, Tsubota and Kanagawa 1989). Although we have a general understanding o f male brown bear (U. a. spp.) gonadal activity, most o f our knowledge regarding the breeding activity o f male brown bears is based on empirical observations. Few studies have characterized male brown bear reproductive biology, including testicular growth and spermatogenesis. Currently, no data are available for testicular growth and age o f sexual maturity o f grizzly bears in the continental United States. 57 Nearly all grizzly bears killed in the continental United States are necropsied at the Montana Department o f Fish, Wildlife and Parks (MDFWP) Wildlife Laboratory in Bozeman, Montana. Due to the lack o f information on male grizzly bear reproductive physiology and the fortuitous, although limited, availability , o f male reproductive tracts, I sought to: (I) evaluate testicular growth and seminiferous tubule development in grizzly bears <15 years o f age, and (2) estimate age o f sexual maturity in male grizzly bears from Montana and Wyoming. Although my conclusions are limited by small sample sizes, I present the only information available on testicular histology o f male grizzly bears in the continental United States. Methods Testicles from 20 male grizzly bears were collected from bears killed by Federal and State wildlife personnel or hunters in Montana and Wyoming from 31 May 1978 to 27 August 1992. Age o f each bear was determined by counting the cementum annuli o f extracted premolars (Willey 1974; Coy and Garshelis 1992). As testicle collections were fortuitous, it was not possible to obtain specimens for all ages and seasons. % Usually within a few hours after death, the testicles o f each bear were removed from the scrotum, dissected free o f the tunica vaginalis, fixed in neutral-buffered formalin for several days, then stored in either 10% formalin or 70% ethanol at the MDFWP Wildlife Laboratory for < 10 years. Tissue shrinkage and suboptimal staining 58 are known to occur by prolonged storage in fixative (Gale Callis, Dep. Vet. and Molecular Biol., Mont. State Univ., Bozeman, pers. commun., 1996). No corrections for shrinkage were made on my data. Each testicle was weighed without the epididymis to the nearest 0.001 g using a Mettler PJ 300 digital balance. Testicle length and diameter were measured without the epididymis with vernier calipers to. the nearest 0 .1 cm. Using a dial scale, body mass was determined to the nearest pound and converted to kg; body length was measured with a tape to the nearest inch and converted to cm. One of the 2 testicles from each bear was selected (right and left side were unknown) and cut transversely into 3 approximately equal sections that represented the top, middle, and bottom. Two 3-mm3 blocks were cut from each section (6 blocks o f tissue/testicle), dehydrated in ethanol, and embedded in paraffin. Five sections (5 um thick) were cut from each block o f embedded tissue at 3 locations within the block (i.e., front, middle, and back) and placed on separate glass microscope slides. Within each block, the 5 sections from each location within each block were separated by 150 um o f tissue. Sections were removed from the paraffin and stained with Shandon hematoxylin I and eosin Y. Three o f the 5 sections from each location within each block (front, middle, and back) were chosen randomly and the diameters o f seminiferous tubules were measured using a Zeiss 20 light microscope (100x) and an image analyzer (AUSKey 2.0, Animal Ultrasound Services, Inc., Ithaca, N.Y.). Tubular diameters were estimated by taking 59 the mean o f 3 linear measurements at different axes that intersected at the center o f the tubule. This procedure was repeated for each slide from each block taken from the top, middle, and bottom o f each testicle. This represented 18 slides from each testicle with 30 seminiferous tubule diameters measured/slide. Seminiferous tubules from each grizzly bear were examined visually under a Zeiss 20 light microscope (400x) for the presence o f fully-formed spermatozoa. A bear was considered sexually mature if there were spermatozoa in the lumen o f the seminiferous tubules. To test whether seminiferous tubule diameter differed between the top, middle, and bottom o f each testicle sampled, data for seminiferous tubule diameters were analyzed by analysis o f variance for a split-split plot design using the General Linear Models (GLM) procedure o f SAS (SAS 1987). Variables in the model included: part of testicle (top, middle, and bottom), block (2 blocks, 3-mm3) and slide (front, middle, and back), and all interactions. The error term for testing part o f testicle was block within testicle and the error term for testing block was slide within block. Pearson correlation coefficients were calculated between age, body mass, body length and selected testicular traits in all animals using SAS. I used least-squares linear regression analysis to investigate the relationship between age and mean testicle mass for each bear. The effects o f age on seminiferous tubule diameters were examined by fitting a nonlinear regression model to these data using the PROC NLIN procedure o f SAS (1987). The proportion o f grizzly bears 4.5 years old that lacked fully-formed 60 spermatozoa found in the lumen o f seminiferous tubules and > 5.5 years old that had fully-formed spermatozoa were analyzed using contingency chi-square. Testicular volume was estimated using the formula for a prolate spheroid (Beyer 1976). Data for testicle mass and volume, and mean seminiferous tubule diameter were examined by an analysis o f variance using the GLM procedure (SAS 1987). The model included age class and date o f kill. Means were compared using the LSD procedure (SAS 1987). Data for mature bears were grouped by season (the breeding season [May through mid-July] and the nonbreeding season [mid-July through November]) and analyzed using a f-test. Results There were no effects (P > 0.10) o f blocks and sections, and there were no interactions (P > 0.10) o f these variables among the other independent variables. Therefore, data for seminiferous tubule diameters were pooled to yield a single estimate for each bear (Table 6). Age, body mass, and body length were correlated positively (P < 0.01) with testicle mass (r = 0.76, 0.62, and 0.69, respectively), testicle length (r = 0.70, 0.65, and 0.74, respectively), testicle diameter (r = 0.69, 0.68, and 0.77, respectively), and mean seminiferous tubule diameter (r = 0.60, 0.54, and 0.62, respectively). 6 1 Table 6. Age (yr), date killed, and testicular characteristics o f 20 male grizzly bears from Montana and Wyoming, 1978-1992. Age (yr) Date killed3 (M-D) Testicle mass'3 (g) A B Testicle length (cm) A B Testicle diameter (cm) A B Mean STDc (urn) 1.5 7-7 5.26 5.53 4.2 4.3 1.4 1.4 80.4 1.5 9-19 4.83 4.4 4.5 4.2 1.31 1.3 83.8 2.5 5-31 10.77 9.86 5.3 4.9 1.75 1.3 106.9 2.5 8-4 10.21 10.62 4.7 4.8 1.85 1.9 144.1 2.5 8-27 13.76 13.68 4.8 4.5 1.78 1.8 105 3.5 7-30 18.55 17.67 5.3 5.2 2.74 2.6 141.5 3.5 9-13 11.12 11.98 4.5 4.3 2.01 2.1 94.6 3.5 9-24 21.5 23.71 5.3 5.4 2.93 3 112.9 4.5 7-7 5.83 5.89 3.6 3.5 1.66 1.7 108.1 4.5 8-17 19.83 19.44 5.5 5.6 2.8 2.8 120.4 4.5 11-8 14.61 14.08 5.4 NDd 2.39 2.5 181.5 5.5 7-25 26.45 26.66 5.8 6.3 2.96 2.9 125.9 5.5 7-25 22.61 21.1 5.7 5.6 2.93 2.9 169.3 5.5 8-12 31.47 30.76 6.2 6.5 3.03 3 157.2 7.5 5-18 36.53 ND 6.3 ND 3.34 ND 151.1 7.5 8-18 27.77 ND 5.8 ND 3.03 ND 146.6 7.5 10-19 18.45 ND 6 ND 2.71 ND 133.3 11.5 9-20 31.15 31.74 6.1 6.1 3.18 3.2 167.2 12.5 6-26 45.41 46.6 6.8 7 3.54 3.5 212.6 14.5 7-31 52.28 53.08 .7.6 7.6 3.63 3.5 167.3 aDate when killed (M = Month, D = Day). bTesticles for each bear labeled A and B because side was unknown. cSTD = Seminiferous tubule diameter. dND = No data. 62 Testicle mass was related linearly to age (r2 = 0.80, P = 0.05; Fig. 16), whereas the relationship between age and seminiferous tubule diameter was non-linear. Seminiferous tubule diameter increased rapidly in bears between 1.5 and 5.5 years o f age, then increased less rapidly in bears older than 6.5 years (Fig. 17). Seminiferous tubule diameter appeared to asymptote at about 10-12 years o f age (Fig. 17). Y = 3.22X + 3.39 r2= 0.80, P < 0.001 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 Age (yr) Figure 16. Age variation in testicle mass for male grizzly bears from Montana and Wyoming, 1978-1992. 63 240 -| 210 - 180 - 150 - -0.20XY= 130(1-e ) + 56.0 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 Age (yr) Figure 17. Growth curve o f seminiferous tubule diameters (STD) in male grizzly bears from Montana and Wyoming, 1978-1992. The youngest bear in which fully-formed spermatozoa were found in the lumen of seminiferous tubules was 3.5 years-old and killed in July. However, 2 other 3.5 year-olds (both killed in September) and three 4.5 year-old bears (killed in July, August, and November) did not have spermatozoa in their tubules. Therefore, I considered the grizzly bears in our study that were < 4.5 years-old to be reproductively immature. Only I o f the 11 bears that were < 4.5 years-old had spermatozoa in the seminiferous tubules, whereas 8 o f the 9 bears > 5.5 years-old did. The single bear > 4 .5 years o f age without spermatozoa was a 7.5 year-old bear killed in October. The proportion o f bears < 5 .5 years-old with and without spermatozoa 64 was significantly different than the proportion in bears >5.5-years-old (A’2 = 13.5, d f = I). Testicular mass, testicular volume, and seminiferous tubule diameter were smalle