Carnivores and Behaviour:
Revision Notes

Compiled as a final-year zoology student at the University of Edinburgh, based on the information given in lectures.

History and phylogeny of Carnivora

Defined nowadays as the descendants of a common ancestor, rather than on the basis of shared characteristics.
Carnivore body plan is typical of a quadrupedal mammal - no significant lengthening of limbs or loss of digits.
Carnivora vary enormously in size - the Kodiak bear is 10,000 times the size of the least weasel.
Carnivora arose in Laurasia (the northern continents) and radiated to other parts of the world (Martin 1989).
Not all members of order Carnivora are flesh-eaters (e.g. panda eats bamboo, kinkajou eats flowers and fruit).
Pinnipeds (seals and sea lions) are now considered monophyletic, due to similarity in flipper structure (Wyss 1988) and molecular evidence.
Phylogenetic trees can be combined using a 'super matrix' of the characters used in different trees (though this is likely to be full of holes) or a 'super-tree' based upon the conclusions of each tree. Newer and more reliable studies should perhaps be weighted more highly when constructing super-trees. (Sanderson et al 1998)

Phylogeny of carnivores

(based on MacDonald 2001)

Fissiped families

Canidae (dogs) - built for cursorial hunting and long-distance chases; can regurgitate food for young.
Mustelidae (weasels etc) - have long, thin bodies for getting into small spaces.
Mephitidae (skunks) - a New World family that deter predators with noxious compounds.
Ursidae (bears) - are omnivorous (have chewing teeth rather than carnassials), lack whiskers, hibernate in northern areas, and can live off fat reserves in extreme environments. Includes giant panda.
Procyonidae (raccoons) - mostly insectivorous and New World. May include red panda (Slattery & O'Brien 1995).
Viverridae (civets) - molecular evidence suggests that this 'family' is paraphyletic (Yoder et al 2003).
Herpestidae (mongooses) - includes meerkats.
Hyaenidae (hyenas) - includes the termite-eating aardwolf.
Felidae (cats) - obligate flesh-eaters with reduced dentition (hardly any molars); hunt by stealth.

Pinniped families

Phocidae (true seals) paddle with back flippers and steer with front ones.
Otariidae (eared seals - fur seals and sea lions) propel themselves in the water with forelimbs, and can pull hind limbs under body for use on land.
Odobenidae (walrus) combines features of both groups.

British fissipeds

(Corbet & Harris 1991, The handbook of British mammals; Macdonald & Barrett 1993, Mammals of Britain and Europe; Yalden 1999, The history of British mammals)

Canidae

Fox, Vulpes vulpes - Urban foxes are a potential rabies vector, if the disease ever returns to Britain.
Wolf, Canis lupus - extinct since 18^th century.

Mustelidae

Pine marten, Martes martes - has elongated legs; hunts in trees for birds and red squirrels. Restricted to Scotland - could be reintroduced to England, but conflicts with bird conservation.
Stoat, Mustela erminea, and weasel, Mustela nivalis - long slender bodies, short legs; hunt down burrows. Weasel is highly sexually dimorphic - driven by sexual selection (facilitating polygyny) and environmental selection (males and females specialise in different prey). Female stoats are mated during infancy - allows early reproduction and successful colonisation. (Macdonald 2000)
Polecat, Mustela putorius - mainly restricted to Wales, but recently spread back into England. Hybridisation with feral ferrets is a problem.
Feral ferret, Mustela furo - introduced 1100. Originated from M. putorius or Steppe polecat M. eversmanni.
Mink, Mustela vison - elongated body; a vicious predator of birds and small mammals. Released from fur farms. Doesn't compete with otter (separate niches); might compete with polecat but currently living in separate areas.
Badger, Meles meles - legally protected. May act as a vector for TB in cattle.
Otter, Lutra lutra - population recovering after being badly affected by river pollution.

Felidae

Wildcat, Felis silvestris - Hybridisation with domestic cat is a big problem. It is difficult to define a "wildcat". There is no baseline data on DNA. True wildcats have heavy tails with thick black tips and broad striping; skeletons differ from that of domestic cat. Wildcats are legally protected but feral tabby cats (which look similar) aren't.
Feral cat, Felis catus - introduced.
Lynx, Lynx lynx - extinct since early postglacial period.

Ursidae

Brown bear, Ursus arctos - extinct by 10^th century.

British pinnipeds

Harbour (common) seal, Phoca vitulina - widely distributed in north Atlantic and north Pacific.
Grey seal, Halichoerus grypus - exists as four separate populations in north Atlantic.
Both species are sedentary (non-migrating) phocids.
Occasional visitors to British waters: bearded seal, hooded seal, walrus.
Grey seal pups have white, non-insulating fur. Harbour seal pups, by contrast, can swim from birth - they are typically born in summer when water is warmer.
Male harbour seals attempt to defend aquatic territory around pupping sites. There is low variance in male reproductive success, and no mate fidelity (number of full sibs no greater than predicted by random model).
During breeding seasons, male harbour seals balance competing demands of foraging (most intense during pre-mating period) and encountering females (there is weight loss during mating period due to less foraging).
Male grey seals defend females, not territory. They fast for up to six weeks (they are bigger than harbour seals, with more endurance). There is evidence for polygyny (degree may depend upon topography and ease of female defence), and mate fidelity (both through site preference and partner preference).
Mother grey seals fast during lactation. Mother harbour seals are smaller and don't have sufficient body resources to fast for the whole lactation period (Coltman et al 1997).

Jaws and dentition

(Biknevicius & Van Valkenburgh 1996, in Carnivore behaviour, ecology and evolution)

Long canine teeth are used to deliver a stabbing/suffocating bite.
Blade-like carnassial teeth (upper 4^th premolar and lower 1^st molar) are used for shearing.
Canoids use post-carnassial teeth for crushing bone; feloids lack molars and use pre-carnassial teeth for crushing.
Long canines of sabre-toothed cats (e.g. Smilodon) would have been vulnerable to breakage if stabbed deep into prey; sabretooths probably killed with sabre slashes and incisor battery.
Carnivores typically have large temporalis, pterygoid and digastricus muscles (involved in opening/closing jaws), but small masseter muscles (involved in chewing).
Meat eaters (e.g. lion: I 3/3, C 1/1, PM 3/2, M 1/1) - stabbing canines and incisors, powerful carnassials with blade parallel to jaw for maximum strength, cheek teeth well-adapted for grinding bones and chewing flesh.
Insectivores (e.g. aardwolf: I 3/3, C 1/1, PM 3/1-2, M 1/1-2) - canines reduced and incisor-like, carnassials less prominent with blade at an angle to jaw, cheek teeth not adapted for grinding, teeth generally weak and small.
Omnivores (e.g. brown bear: I 3/3, C 1/1, PM 4/4, M 2/3) - incisors moderately sized with less need for tearing flesh, long canines for stabbing and killing small prey, reduced carnassials, large and versatile molars for grinding.
Fish eaters (e.g. harbour seal: I 3/2, C 1/1, post-canines 5/5) - definite incisors and canines for killing bites, conical post-canines adapted for chewing fish rather than crushing bones, fewer incisors, no carnassials.

Mating systems

(Davies 1991, in the book Behavioural Ecology)

Types of mating system: monogamy, polygyny, polyandry, promiscuity (indiscriminate mating)
Females should invest in obtaining resources; males should invest in obtaining mates.
Female dispersion should be influenced by resources; male dispersion should be influenced by female dispersion.
Mammals (unlike birds) should be polygynous or promiscuous, unless there is a need for paternal care.
Factors affecting mating systems: dispersion, economics of female defence by males, need for paternal care, phylogenetic inertia, life history (K-selected or r-selected), ecological factors promoting group formation, habitat constraints, reproductive specialisations, outbreeding mechanisms.
In lion and hyena groups (as in most mammals), males emigrate. In African hunting dog packs, females emigrate.
In most fissipeds, litter size depends on conditions. Pinnipeds, by contrast, almost always have one pup at a time.
In solitary carnivores, female range depends on food abundance, especially at the most critical time of year.
Small, exclusive ranges are expected when food resources are stable and evenly-distributed; large, overlapping ranges are expected when there is spatial and temporal variation in food availability.
Male spatial organisation should be related to food resources outside the breeding season, and to female distribution during the breeding season. Male ranges are larger than predicted on the basis of food requirements, overlapping many female ranges (Sandell 1989, in Carnivore behaviour, ecology and evolution).
In pinnipeds, ecological factors facilitate and constrain polygyny, and hence sexual dimorphism. Parturition occurs on land, is highly seasonal and synchronised (with an annual cycle, except in walrus), males are attracted to parturition sites for access to females, and there is no paternal care of young.
Most pinnipeds breeding on land (e.g. sea lions, elephant seals) or on land-fast ice (e.g. Weddell seal, ringed seal), where suitable sites are stable but not abundant, are polygynous (due to crowding) and spend longer periods on shore. There is more polygyny in the Antarctic, in the absence of land predators (polar bears).
Pinnipeds breeding on pack ice (e.g. hooded seal), which is unstable but abundant, show increased synchrony, shorter time on substrate, and less polygyny. (Exception: walrus is a polygynous ice-breeder, perhaps partly due to phylogenetic inertia.) Land-breeding harbour seal populations form larger groups and are more sexually dimorphic than ice-breeding populations, suggesting greater polygyny. (Le Bouef 1986, in New Scientist)

Sexual behaviour and reproduction

An animal's reproductive strategy is linked to its life history. Covarying factors include: longevity, size, number of litters, litter size, growth rates, and investment in offspring.
There is a continuum of life-histories between small, short-lived animals with frequent, large litters, and large, long-lived animals with infrequent small litters.
Stoats have an annual breeding cycle, whereas weasels may have two litters in good years (Macdonald 2000).
Births are timed to maximise survival of offspring; this affects the timing of mating.
Number of young reared depends on female's condition. In some species litter size is linked to prey abundance - spotted hyenas' conception rates are lowest at times of year when prey is least abundant (Holekamp et al 1999).
Embryonic diapause (delayed implantation) occurs in polar bears and various mustelids and pinnipeds. It appears to have evolved once in the common ancestor of these groups, but has been secondarily lost in some species.
Delayed implantation in some mammals (e.g. kangaroos) is facultative - it occurs during the period of lactation, with female 'constantly pregnant' - but in carnivores it is obligatory. (Mead 1993)
Delayed implantation dissociates mating and reproduction, and allows mating to be done at a convenient time. It has fitness costs (loss of fecundity), especially in warmer climates and in smaller species. In mustelids, delayed implantation occurs in bigger species and colder climates (Thom et al 2004), e.g. in stoat but not in weasel.
Delayed implantation 'guarantees pregnancy' - it is valuable when it is vital to give birth early, and allows subordinates in a group to hedge their bets. It is useful in colonising species.
In stoats, embryos implant in spring in response to increasing photoperiod (Macdonald 2000); in badgers (in which gestation period is longer), embryos implant in autumn in response to decreasing photoperiod.
Polar bear embryos are implanted in autumn; tiny cubs are born and suckled in mid-winter while mother is hibernating and emerge in spring (Pond 1989).
In pinnipeds, delayed implantation allows birth and mating to be done at the same time (so animals only risk coming ashore once), while maintaining an annual cycle.
Female spotted hyenas have 'false penises', which make mating and birth awkward and dangerous. Why incur this handicap? Hyena societies are very competitive, with strong selection for female aggressiveness, which has been achieved through androgenisation. False penises are used in social displays. (Frank 1997)

Olfactory communication

(Stoddart 1976, Mammalian odours and pheromones)

Rate of urine marking in the vixen is highest: during the breeding season, on frequently-used paths, on outbound journeys, and near core of animal's range. Vixens often explore and re-mark previously-marked patches (overmarking). When markings from strangers are artificially placed within a vixen's range, she will investigate thoroughly, appear disturbed, and re-mark the spot. (Macdonald 1979)
Marking with faeces by badgers is done at shared latrines along territorial boundaries, and at unshared latrines in territory's 'hinterland'. Boundary latrines are used most frequently during the breeding season, and are mainly used by males (Roper et al 1993).
Spotted hyenas living in poorer areas with larger ranges mark the hinterland, whereas those in smaller ranges mark the boundaries - economics of defence.
There is a trade off between longevity and volatility of odour molecules.
Mammals, unlike insects, often use heavy, waxy secretions containing complex mixtures of compounds.
Hyenas can produce two types of anal gland secretion: a short-lived watery one and a long-lived waxy one.
Pheromones, defined as a specific molecule eliciting a specific response, are probably rare in mammals.
Odour-detecting membranes are on endoturbinales deep within nose. Dogs do a series of little sniffs to force air into sinuses that contain a lot of their olfactory epithelium. They exhale sideways to avoid re-breathing stale air.
Skin secretions come from sebaceous glands, apocrine glands (sweat glands associated with hair follicle) and eccrine glands (sweat glands not associated with hair follicles).
Sebaceous glands produce oily, long-lasting secretions, under the control of steroid hormones.
Apocrine and eccrine glands produce watery short-lived secretions under nervous/blood hormone control.
Other sources of odour: saliva, accessory gland at corner of eye, urine, vaginal secretions.
Animals may mark conspicuous landmarks, mark social companions (allomarking), or release odours into the air.
Functions of social odours: identity, reproduction and detection of oestrus, temporal information ("I was here"), territory and social status. (Gorman & Trowbridge 1989, in Carnivore behaviour, ecology and evolution)
Possible reasons for scent marking territories (Gosling 1982): deterring intruders, intimidating intruders, enhancing confidence of resident, informing intruders about status of the resident, orientating the resident, attracting or stimulating mates, assisting pair-bond formation, assisting population regulation, assisting optimal foraging, associating the resident with its territory (an intruder is unlikely to attack the resident, who will fight hard).

Group living

Living in groups is not the norm in carnivores: only 10-15% of fissiped species do so outside the breeding season.
Two main benefits of group living: assistance in killing large prey, and defence.
Additional advantages of groups - defending prey, social learning, locating unpredictable prey, division of labour, alloparental behaviour (babysitting), sexual competition, territory inheritance.
There is a positive relationship between group size and territory size in some species (coyotes, wolves, lions, spotted hyenas), but not in others (badgers, red foxes). Why? Resource dispersion hypothesis - animals choose minimum territory needed to support themselves, and if food is patchy a territory sufficient for some may accommodate others too. Prey renewal hypothesis - when prey is rapidly-renewing there is plenty to go round.
Babysitting in meerkats (Clutton-Brock et al 1998) - some (usually subordinates) guard burrow while others forage. Babysitters incur a cost - lost foraging opportunity. In larger groups, each individual spends less time babysitting.
Small carnivore species tend to be solitary, perhaps because they eat small food items that are difficult to share. (Exception: one group of related herpestids, in which group living has evolved.)
Medium-to-large species are likely to be solitary if they hunt small, indivisible prey (e.g. foxes, maned wolf) or if they hunt by surprising prey in low-visibility habitats (e.g. leopard, tiger).
Phylogenetic constraints may also play a part: felids tend to be solitary, while canids are more social.
Larger hunting groups take larger prey (e.g. in lions, hyenas, wolves) but each individual doesn't necessarily get a bigger portion. There is an optimum pack size that maximises an individual's food intake.
Success rate in hunting may also be important (e.g. jackals can only steal gazelle fawns by working in pairs).
Larger canids live in larger groups. Larger groups take proportionately larger prey. Diet breadth increases with body size in co-operative hunters, but is independent of body size in solitary hunters. (Moehlman & Hofer 1997)
In lions, co-operative hunting may only occur at times of scarcity or when tackling huge prey (though plenty of solitary cat species also tackle prey much bigger than themselves). Food intake per lion per day is highest when alone or in groups of 5 to 7. Lions may forage in non-optimal group sizes because of constraints imposed by protection of cubs (in cr�ches), territorial defence, and the risk of meeting larger groups (Packer et al 1990).
Reproductive success amongst a 'coalition' of male lions is less skewed when they are brothers.
In hunting dogs, modal pack size appears to optimise food intake per dog per kilometre chased, but variation in data is enormous, and the measure of cost may be wrong (benefit minus cost should be measured instead).
Jackals live in monogamous families with helpers. The number of helpers correlates with the number of young reared, but we shouldn't assume causation. Manipulation experiments are difficult to do on carnivores.

Play

(Bekoff & Byers 1998, Animal play: evolutionary, comparative and ecological perspectives)

'Play' is easy to recognise but hard to define.
Play often includes repetitive or exaggerated actions, sudden switches in behaviour, and/or self-handicap.
Play is a characteristic of more intelligent animals (certain mammals and birds).
Rats play more obviously than mice do - does this reflect larger brain size?
Types of play: locomotor play, social play, object play (latter is usually the last to emerge).
Function of play (Smith 1982): Physical training? Building competitive social skills (fighting)? Building non-competitive social skills (social bonding, acquisition of rank, communication)? Building cognitive skills?
Male offspring typically play-fight more than females, but this may because they get less attention from mother.
Play incurs costs: time and energy, risk of injury, risk of predation due to lack of vigilance. For example, in young fur seals play occupies only 6% of the time, but 80% of deaths due to predators occur during play (Harcourt 1991).
Artificial experiments into play are hard to do because it is difficult to stop animals playing without restricting them in other ways. However, natural experiments exist. Lee (1994) observed no difference between monkeys that grew up under easy conditions and played a lot, and those that grew up under harsh conditions and played less.
Kittens weaned earlier play precociously, suggesting that play is important (Bateson et al 1990).
Play may be important for synaptogenesis - the development and refinement of neural connections (Furlow 2001).
Play may facilitate the development of 'coping strategies' for dealing with situations (Spinka et al 2001).

Allometry

Y = b(M^a) where M is body mass, Y is another measurement, a (exponent of allometry) and b (coefficient of allometry) are constants. Parameters estimated from straight line: log Y = a(log M) + log b. (Huxley 1972)
In positive allometry, a > 1; in negative allometry, a < 1; in isometry, a = 1.
For carnivore skull mass or brain mass vs. body mass, a < 1. For heart mass vs. body mass, a = ~1.
For mammal skeleton mass vs. body mass, a > 1, due to a heavier body's greater need for support.

Development of behaviour - the mother-offspring relationship

(Bateson 1994)

Young may be altricial (highly dependent on mother at birth) or precocial (born at a more advanced stage).
Milk is thicker and fattier in pinnipeds than in fissipeds, although between pinnipeds there are differences, depending on maternal strategy (Oftedal 1984).
In domestic cat, adult males are larger than females, but males' growth spurt occurs late in development, so the rearing of sons is not significantly more strenuous for a mother than the rearing of daughters.
There is a conflict of interest between youngsters and their siblings (e.g. competition for nipples to suckle).
There is also a conflict between the offspring (which want a greater share of the mother's resources) and the mother (who wants to maximise her overall lifetime reproductive success).
Lactation can be a very costly process (much more costly than gestation), especially for larger mammals.
Mothers typically gain weight during pregnancy and lose it during lactation (exception: the dwarf mongoose, which loses weight during pregnancy, perhaps due to a trade-off between carrying weight and greater mobility).
In badgers, body condition scarcely deteriorates after undergoing pregnancy without lactation, but deteriorates dramatically after pregnancy and lactation (Woodroffe & Macdonald 1995).
Energy intake of mother cat increases by 100-200% at the time of peak milk yield, but she still loses weight.
Gut length may increase during lactation (e.g. in rats) to meet the demands of greater food intake.
A mother cat with larger litters produces more milk, but there is a limit to her capacity to do this, so growth rate is faster among kittens in smaller litters (Deag et al 2000).
Later in the lactation period, mother cats increasing deny milk to kittens, by changes in posture (e.g. use of 'blocking posture'), or in lab experiments by retreating to a shelf that kittens can't reach.
As lactation progresses, feedings are increasingly initiated by kittens rather than by mother.
Weaning is a process during which there is a sharp drop in rate of parental investment in offspring (Martin 1984).
Growth curves of offspring kink upwards at the time of weaning (especially in bigger litters), since growth is no longer constrained by the mother's capacity for milk production. Weaning is inevitable since there comes a point at which milk can no longer satisfy the youngsters' increasing energy requirements.
The role of conflict in weaning may have been overstated - it is not in young carnivores' interests to exhaust their mother completely during lactation because she still has to bring back solid food and teach them to hunt.
Mothers bring back dead prey at first, then start to bring back live prey for kittens to practice with. Later on, kittens begin to participate in the hunt for themselves.
Kittens' play becomes increasingly associated with predatory behaviour as they grow older (Caro 1981).

More notes and essays

Carnivores and Behaviour: Revision Notes