@phdthesis{oai:az.repo.nii.ac.jp:00005193,
 author = {髙田, 隼人},
 month = {2019-02-12, 2018-03-22},
 note = {The Japanese serow (Capricornis crispus) is typical forest dwelling ungulate and predominantly inhabit deciduous broad-leaved forest of Japan Archipelago. Its behavioral traits are thought to be adapted to forest environments, especially in terms of their diet and social organization. Although some populations of the Japanese serow inhabit alpine or subalpine open cliffy habitat including meadow, its behavioral traits in these habitats is largely unknown. Moreover, anti-predator behavior and habitat selection of the Japanese serow is also largely unknown. Therefore, further ecological research is required. The Japanese serow which maintains primitive traits is important materials to consider evolution of Caprinae. Starting from the premise that the first caprids were small forest dwellers, not unlike serows, caprids evolution seen as an adaptive radiation into open environment such as alpine habitat. It is thought that gregarious habitst was evolved from solitude habits, along with it adaptive radiation. Therefore, to compare the ecology of the Japanese serow in the forest habitat and the alpine meadow habitat and to elucidate the variation of their ecology helps to infer the process of social evolution of the caprids. In this paper, in order to evaluate the relationship between habitat features and foraging ecology, anti-predator behavior, habitat selection, and social organization, I have examined behavior of the Japanese serow in different two populations, which lives in closed forest habitat (termed Asamasanso) and in alpine meadow habitat (termed Kazankan), over 6 years in the Mt. Asama area, central Japan.
1. Foraging ecology 
The diet of the Japanese serow was analyzed quantitatively in two populations, Asamasanso consisted closed forest habitat and Kazankan consisted of alpine meadow, 
by direct observation of feeding behavior during two periods, 2012-2014 at Asamasanso and 2016-2017 at Kazankan. Analyses of 5868 bites at Asamasanso indicated that serows fed mainly on leaves and twigs of deciduous broad-leaved trees, which formed 71.9% of the diet throughout the year, followed by ferns (14.5%) and forbs (12.8%). Graminoids were rarely fed by serows throughout the year (0.8%). Serows selected deciduous broad-leaved trees (in spring to autumn) and forbs (in spring) and avoided graminnoids (in all season) as a food plants. The results suggest that the Japanese serow inhabiting forest is a browser throughout the year. Analyses of 2438 bouts at Kazankan indicated that diets of serows were drastically changed according to seasons. Serows fed mainly on grasses in spring (71.6%), on forbs in summer (88.2%) and autumn (71.2%), and on various food plats such as dwarf bamboo, coniferous trees, and deciduous broad leaved trees in winter. The results suggest that serows in alpine meadow are generalist herbivores and seem to eat what is available. Its food habit is quite different from serows inhabiting forest habitat, is rather similar to food habits of other caprids inhabiting alpine zone.
2. Anti-predator behavior 
We have examined the response to human observers of the Japanese serow in relation to visibility, terrain type, and group size based on direct observations conducted during 2012-2014 at Asamasanso. From a total of 221 sightings, freezing behavior was observed more frequently in the closed habitat (14.3% of 196 sightings) and during the growing season (22.8% of 92 sightings) than in the open habitat (0% of 25 sightings) and during the non-growing season (5.2% of 129 sightings). Freezing behavior was also frequently observed in solitary animals (15.2% of 184 sightings) than in groups (0% of 37 sightings). These results suggest that the function of this freezing behavior is concealment from 
predators. Japanese serows fled more frequently in the open habitat (72%), during the non-growing season (58.7%), and flat terrain (89.7%) than in the closed habitat (18%), during the growing season (41.3%), and steep terrain (44.2%). These results suggest that Japanese serow adopt cryptic anti-predator strategy to avoid detection by predators in the dense forest habitat and that the Japanese serow are adept at evading predation in steep terrain, and that steep slopes are also used to escape danger. Their anti-predator behavior changes according to surrounding environments such as visibility and terrain.
3. Habitat selection 
We have examined the habitat selection four VHF-marked serows at within home range-scale in relation to terrain, visibility, plant biomass, and foraging efficiency during 2012-2014 at Asamasanso. Serows avoided for habitat with flat terrain and with good visibility and selected steep terrain and specific plant communities. Serows select for plant communities with abundant deciduous broad-leaved trees which is main food plants of serows and with high foraging efficiency. The results suggest that serows avoid for habitat with more high risk of predation and select for habitat with more safety and high benefit from foraging. Even plant communities with abundant foods, flat terrain was not selected by serows. Thus, most important factor of habitat selections of the Japanese serow within home range-scale is thought to predator avoidance rather than food requirements.
4. Social organization 
In order to elucidate the variations in social organization of the Japanese serow, I examined social organization tow populations inhabiting in different environments, closed forest habitat (Asamasanso) and alpine meadow habitat (Kazankan), by direct observation of social behavior of serows during two periods, 2011-2015 at Asamasanso 
(for 235days) and 2014-2017 at Kazankan (for 252days). Moreover, in order to elucidate Ecological influences on social organization, I examined causes of variability in social behavior in the Japanese serow by comparison among populations and within populations.
1) Social organization of closed forest habitat, Asamasanso 
Serows were usually solitary, but two to three animals temporally formed group. Groups were mostly of mother-kid units, or of male-female pairs. Adult of both sexes maintain stable home ranges (88.1ha for males and 44.8ha for females) throughout the year. Home ranges between adult males were overlapped except for core areas (CP50%), and tolerant or affinitive interactions were observed between adult males at home range overlapped areas. Thus, it is thought that adult males did not defend their whole of home ranges as a territory. Although home ranges between adult females were spacing, only tolerant or affinitive interactions were observed between adult females. Home range core area of one males coincided with home range of one female, and courtship behavior and mating behavior were observed only at these male-female pairs. The probability that males and females stay together was very high (75%) during rutting season. These results suggest that males mating tactics is a “following” strategy rather than territory, and mating system is female defense monogamy.
2) Social organization of alpine meadow habitat, Kazankan 
Serows were usually solitary, but two to four animals temporally formed group. Groups were mostly of mother-kid units, of male-female pairs, or of female-female pairs. Adult of both sexes maintain stable home ranges (8.1ha for males and 7.4ha for females) throughout the year. Home ranges between adult males were spacing, and warning interactions were observed between adult males at periphery of home ranges. Thus, it is 
thought that males defended their whole of home ranges as a territory. On the other hands, home ranges between adult females were largely overlapped, and tolerant or affinitive interactions were frequently observed between adult females. Home range overlaps between females were observed between two to three specific individuals, and their interactions were more tolerant than other females. Thus, it is thought that females organized social units between specific individuals. Observing the agonistic interactions between females indicated that adult females were organized in a stable linear hierarch within and among social units. Home range of one males coincided with home range of one to five females, and courtship behavior and mating behavior were frequently observed among these male-female. A mean of 75% of territorial males in each year were polygynous by covering two to five female ranges. These results suggest that male mating tactics is a “territorial” strategy, and mating system is resource defense polygyny.
3) Comparison among populations 
I used data for two populations of present study (Asamasanso and Kazankan) and for four populations inhabiting forest habitat of previous studies (Nibetsu, Shimokita, Asahi, and Kamikochi) to compare about social organization and habitat characteristics, such as winter forage availability and population density, among populations. For four populations inhabiting forest habitat (Asamasanso, Shimokita, Asahi, and Kamikochi), female serows had smaller ranges and population density was high where winter forage availability was relatively abundant. Male territorial and courtship behavior differed among four populations (Asamasanso, Kazankan, Nibetsu, and Shimokita) and was related to female home range size. Interactions between males were most agonistic and male ranges were spacing where females had small home ranges, and males defended a female rather than a territory where females had large home ranges. The rate of 
polygynous males differed among four populations (Asamasanso, Kazankan, Nibetsu, and Shimokita) and was related to female dispersion. The rate of polygynous males was high where females had small home ranges and overlap rate between female ranges was high. These results suggest that variation in social organization among the Japanese serow populations reflects female responses to the availability of food resources and male responses to the variable distribution and ranging behavior of females.
4) Comparison within populations at Kazanakn 
Relationships between grassland and female dispersion and male mating success were investigated for population at Kazankan. Female range size and overlap rate between female ranges were differed among individuals, and were related grasslands. Home range size of female, which established home range at grasslands, was small and overlap rate between females was high. It indicated that females more concentrated at grasslands rather than forest. Male mating success was differed among individuals, and was related to grassland. Males, which established territory at grassland, covered more female ranges. These results suggested that variation in social organization of the Japanese serow within populations at Kazankan reflect female responses to the grasslands and male responses to the variable distribution of females. Polygynous mating systems was thought to be facilitated by concentrate distribution of females, and concentrate distribution of females was thought to be facilitated by grasslands which supply abundant foods.
These series of studies show that the Japanese serow has adaptive traits to the forest environment as well as adaptive traits in mountainous environments like other caprids inhabiting open cliffy habitat, and suggests that foraging ecology and social organization of the Japanese serow is more variable than previously believed., ニホンカモシカ（以下、カモシカ）は落葉広葉樹林を主な生息環境とし、森林環境に適応的な生態を持つと考えられている。ただし、カモシカは高標高の草原にも生息するが、このような環境における生態は未解明である。また対捕食者行動や生息地選択についても未解明であり、生態研究が求められる。また原始的な形質を持つカモシカはヤギ亜科の進化を検討する上で重要な材料である。ヤギ亜科は森林から開放的な草原環境への適応放散と同時に単独性から群居性の社会を進化させたと考えられている。このため、森林と草原におけるカモシカの生態を比較し、その変異を解明することはヤギ亜科の社会進化の過程を推察する一助となる。そこで本研究の目的は、カモシカの採食生態、対捕食者行動、生息地選択、社会構造と生息環境の関係性を解明し、その適応的意義を検討することとした。
１．採食生態
　長野県浅間山の浅間山荘周辺および火山館周辺の2地域で直接観察によりカモシカの食性を調査した。浅間山荘は主にカラマツからなる森林環境であり、火山館は亜高山帯風衝草原からなる草原環境である。浅間山荘ではバイトカウント法により採食割合を、火山館では採食率を算出した。また各環境の食物供給量を調査し、浅間山荘では食物の選択性を評価した。浅間山荘においてカモシカの採食行動を合計5868バイト観察した。浅間山荘におけるカモシカの採食割合は落葉広葉樹が調査期間を通じて最も高く（71.9%）、次いで広葉草本（12.8%）およびシダ類（14.5%）が高く、グラミノイド（0.8%）はほとんど採食されなかった。食物の選択性は、落葉広葉樹および広葉草本を選択し、グラミノイドを忌避することが示された。このことから、森林環境におけるカモシカはブラウザーであることが示唆された。火山館におけるカモシカの採食行動を合計2438バウト（分）観察した。火山館におけるカモシカの食性は季節に応じて著しく異なり、春はイネ科（71.6%）、夏および秋は広葉草本（夏：88.2%、秋：71.2%）を主に、冬はイネ科と広葉草本に加えてササ類、落葉広葉樹、カラマツ、シラビソなど多様な食物を利用した。火山館では食物供給が激的に季節変化するため、これに応じて食性が季節変化したと考えられた。また、火山館で確認された食性の著しい季節変化やイネ科草本の頻繁な利用は近縁である山岳性のヤギ亜科の持つ採食生態と類似した。これらにより、ブラウザーであると考えられてきたカモシカが環境に応じて食性を柔軟に変化させることが初めて示された。
２．対捕食者戦略
　浅間山荘において直接観察によりカモシカの対捕食者行動を調査した。調査は人に対する個体の行動を記録し、各行動が地形や見通しに応じてどのように変化するかを評価した。合計221回カモシカと遭遇し、フリーズ、逃避、歩行移動、採食・休息の行動がそれぞれ28回、134回、23回、36回観察された。フリーズ行動は見通しの悪い展葉期および閉鎖環境と単独でいる時により発現した。フリーズ行動は捕食者からより発見されづらい状況において発現することから、捕食者からの隠蔽の効果があると考えられた。より危険な状況で起こる逃避行動は地形が平坦なほど、見通しが良いほど発現した。このため、カモシカの対捕食者戦略において急峻な地形および見通しの悪い環境が特に重要であると考えられた。これらのことからカモシカが森林環境に適応的な隠蔽的対捕食者戦略と他のヤギ亜科に見られる急峻な地形を利用した対捕食者戦略を併せ持つことが示唆された。
３．生息地選択
　浅間山荘においてラジオテレメトリおよび直接観察による調査を実施した。ラジオテレメトリにより得られた各個体の測位点を用いて、どのような地形、見通し、植物群落を選択もしくは忌避するかを評価した。また、直接観察により得られた各植物群落における採食効率（バイト/分）および落葉広葉樹の供給量が群落選択性に与える影響を評価した。カモシカは平坦で見通しの良い環境を忌避し、急峻な地形および特定の植物群落を選択した。植物群落の選択性では落葉広葉樹の供給量が多く採食効率の高い群落をより選択した。このため、捕食の危険が高い平坦で見通しの良い環境を忌避し、急峻な地形にある採食効率が高く食物が豊富な群落を好むことが示された。また、食物の豊富な群落であっても平坦な地形では選択されないことから、カモシカの生息地選択において重要なのは捕食回避であり、その制約の中で採食の利益を高める生息地選択をおこなっていると考えられた。
４．社会生態
　生息環境の違いによるカモシカの社会構造の変異を検討するため、浅間山荘および火山館において直接観察による社会構造の調査を実施した。また、生息環境が社会構造に与える影響を検討するために、既存研究の結果を用いて個体群間での比較検討および火山館の個体群では個体群内での比較検討をおこなった。
　浅間山荘
　カモシカの群れサイズは基本的に単独であり、群れ構成は母子ペアおよびオス－メスのペアがほとんどを占めた。オスメスともに安定した定住的な行動圏を持ち、年間平均行動圏サイズはそれぞれ88.1 haと44.8haだった。オス間の行動圏重複率は高く、行動圏のコアエリア（CP50%）以外の大部分が重複した。また、行動圏の重複部でオス間の許容的な交渉が観察されたことから、行動圏全体をなわばりとして防衛しないことが示唆された。一方、メス間の行動圏はスペーシングしていたが、行動圏の境界部において許容的な交渉が観察された。オスの行動圏のコアエリアは特定の1頭のメスの行動圏と重複しており、その個体間でのみ雌雄の同一行動および性行動が確認された。また、交尾期の雌雄同一行動率は75%（N=28）と非常に高い値を示した。オス間のなわばり性が認められず、交尾期の雌雄同一行動率が高いことから、浅間山荘におけるカモシカの配偶システムはメス追随型の一夫一妻性であることが示唆された。
　火山館
　カモシカの群れサイズは基本的に単独であり、群れ構成は母子ペア、オス－メスのペアおよびメス－メスのペアがほとんどを占めた。オスメスともに安定した定住的な行動圏を持ち、年間平均行動圏サイズはそれぞれ8.1 haと7.4haだった。オス間の行動圏はスペーシングしており、互いに警戒しあう緊張的な交渉が観察されたことから、行動圏全体をなわばりとして防衛していることが示唆された。一方、メス間の行動圏重複率は高く、許容的な交渉が頻繁に観察された。メス間の行動圏は特定の個体間で特に重複が大きく、そのような個体間で許容的な交渉が頻繁に観察されたことから、特定のメス間で社会単位を形成していることが示唆された。社会単位は他のメスと行動圏をほとんど重複させない単独メスと2-3頭のメスで行動圏を大きく重複させるメスグループに分類された。メス間の交渉の対戦結果から個体間には直線的な順位関係があることが示唆された。オスの行動圏は複数のメスの行動圏と重複しており、1頭のオスの行動圏に対して行動圏が大きく重複するメス（つがい）の数は平均2.4頭、最大5頭だった。1オス1メスおよび1オス複メスのつがいの割合はそれぞれ25％と75%だった。これらのことから、火山館におけるカモシカの配偶システムはなわばり型の一夫多妻性であることが示唆された。
　個体群間比較
　個体群比較には浅間山荘および火山館の2個体群と既存の森林に生息する4個体群（仁別、下北、朝日、上高地）のデータを使用した。森林に生息する4つの個体群（浅間山荘、下北、朝日、上高地）において、冬季の落葉広葉樹の供給量が多いほどメスの行動圏サイズが小さくなることが示唆された。浅間山荘、火山館、仁別、下北の4つの個体群において、メスの行動圏サイズと行動圏重複率はオスのつがいメス数に効果を与えており、行動圏サイズが小さく行動圏重複率が高いほどつがいメス数が多かった。このことから、食物が豊富でメスが集中して分布するほど一夫多妻性が促進されることが示唆された。また、メスの行動圏サイズが大きいほどオス間の行動圏は重複し、雌雄同一行動率は高くなる傾向にあった。このことから、メスの分布様式に応じてオスは配偶戦術を変異させており、メスの行動圏が大きい場合はメス追随型、メスの行動圏が小さい場合はなわばり型の戦略をとることが示唆された。
　個体群内比較
　各メスの行動圏内に占める草原の割合と行動圏サイズおよび行動圏重複率の関係を検討した。また、各オスのつがいメス数と各オスの行動圏内に占める草原面積の関係を検討した。メスの行動圏内の草原割合が多いほど行動圏サイズは小さく、行動圏重複率が高かった。このため、食物の豊富な草原環境はメスの行動圏サイズの減少および重複を促進していると考えられた。さらに、オスのつがいメス数は行動圏内の草原面積が大きいほど多く、メスが集中する草原をなわばりとして占有するオスほど一夫多妻的傾向が強くなることが示唆された。これらのことから、草原環境がメスの集中分布を、またメスの集中分布が一夫多妻性を促進したと考えられた。
　これらの一連の研究により、カモシカは森林環境に適応的な形質と同時に近縁種にみられる山岳環境に適応的な形質を持つことが示された。また、これまで保守的であると考えられてきたカモシカの採食生態および社会生態が生息環境の違いにより変異し、高標高の開放的な草原環境においては近縁種と似通った採食生態および社会生態を持つことが示された。また、メスの集中的な分布や一夫多妻性は多量にまとまって食物を供給する草原により促進されることが示唆された。これらの結果は、ヤギ亜科において森林から開放的な山岳環境への適応放散と同時に柔軟な採食生態や群居性の社会が進化したとする説を支持した。食物の供給状況の変異によるメス間の排他的関係の緩和が、群居性社会発達への一つの契機となる可能性が示唆された。},
 school = {麻布大学},
 title = {浅間山におけるニホンカモシカ（Capricornis crispus）の行動生態},
 year = {}
}