In larger colonies, females not only mate with several males in sequence, but the large majority of their litters also contain the DNA of more than one of these. The females may be exerting some choice-they sometimes show a preference for males from outside their group, thus preventing inbreeding, and also for those males from their own group that are able to defend the largest territories.

The female's practice of offering herself to several males one after the other may have another purpose: protecting her future offspring from infanticide. Each male has observed her being mated by others, but cannot know which of her kittens are his and which not. Likewise, since each of the larger males has mated with several females, none has any incentive to kill any litters.

Most urban males today are faced with a new and different problem: how to locate females that are capable of breeding. Nowadays, an increasing proportion of pet cats are neutered before they are old enough to reproduce. Animal welfare charities not only promote the spaying of all females before they have even one litter, they also attempt to seek out and neuter the feral colonies that remain.21 The urban tomcat is unlikely to locate, let alone defend, a harem of reproductive females. In built-up environments, most tomcats probably adopt the same roaming lifestyle as their wild forebears, hoping to stumble on a young female whose owners have delayed spaying either because they want her to have a litter or simply because they are unaware that today's reliable nutrition enables females to mature more quickly than ever before and can come into season when they are as young as six months old.

Judging by my admittedly casual observations, tomcats seem unable to distinguish between neutered females that form the large majority in many towns and cities, and the few, mostly young, females that are still reproductively intact. Roaming toms still visited annually in late winter to check out my two females ten years after they had been neutered (having produced one and three litters respectively). Spayed females may be difficult to distinguish from intact females between seasons, certainly not by the way they behave and possibly not even by their odor, to which tomcats are presumably very sensitive.

Urban toms thus face a needle-in-a-haystack problem: they are surrounded by hundreds of females, only an otherwise indistinguishable few of which will ever present them with the opportunity to sire any kittens. The tomcats must therefore roam as widely as possible, endlessly straining their senses for the yowl and odor of the rare female that is coming into season. Such toms are shadowy animals; some are theoretically "owned"-though their owners rarely see them-and many feral. Because they make themselves inconspicuous except when they have located a prize female, there are probably far more of them than most people realize. When it first became possible to obtain a cat's DNA fingerprint from just a few hairs, my research team attempted to locate every litter born in homes in a couple of districts of Southampton, UK. From what we'd read, we expected to find that just a few "dominant" tomcats had sired most of the litters in each district; instead, we found that out of more than seventy kittens, virtually all litters had different fathers, only one of which we were able to locate. We found no evidence for littermates having different fathers, which implied that most estrous females had attracted only one male. Apparently, by inadvertently "hiding" the few reproductive queens that remain in a sea of spayed females, the widespread adoption of neutering is making it difficult for even the fiercest, strongest tomcat to do much more than search at random, thereby giving all the males in the area an even chance of reproducing.

Cats of both s.e.xes have shown remarkable flexibility in adapting their s.e.xual behavior to the various scenarios that we humans have, over the course of time, imposed on them. One reason that they have coped so well is simply that female domestic cats are very fecund, capable of producing as many as a dozen kittens per year. Even when conditions for reproduction are difficult, most free-living female cats will leave at least two or three descendants, enough to maintain the population even if many of her litters do not survive to adulthood.

Cat society has even adapted to life at high densities, despite origins in territorial rivalry. Logically, that heritage should result in mayhem, totally unsuitable for raising defenseless infants. Females seem to have resolved this problem by accepting the advances of several males each time they come into season, thereby making each male uncertain which kittens are his, and so banishing all thoughts of infanticide. When cat colonies are smaller, females choose to live in small family groups, and may bond with only one male, whom they accept as being both the most suitable father for their kittens and the most effective at driving away marauding rivals. When they are deprived of the company of their female relatives, either because food is insufficient for more than one cat, or due to human intervention, females are equally capable of bringing up litters without any a.s.sistance whatsoever. This remarkable flexibility must have contributed to the cat adapting to such a wide range of niches.

Male cats have also adapted remarkably well, though in a different way to females. Since in theory one male cat can sire many hundreds of kittens, the number of males in any one area is never likely to be the factor that limits the size of the cat population. Nevertheless, each individual male has no interest in the survival of his species, only in producing as many offspring of his own as he can. When breeding females are few and far between, a male must search as widely as possible for mating opportunities, pausing only to grab enough food to keep him on the move. If he can be the first and ideally only male to reach a female at the crucial moment, she is unlikely to be choosy about his quality. If he happens to be in an area where there are several females, and some of these live in groups, it may benefit him to form a pair bond or harem-although this seems not to inhibit him from searching for additional females elsewhere, when he judges that he can get away with it. He will also keep an eye out for other groups of females that he might try to take over in the future, and it is most likely males that are in this frame of mind that commit the occasional act of infanticide. However, if he finds himself living in a vast colony with many potential rivals as well as many potential mates, he will learn to curb his natural aggression, this being the only sensible strategy that will allow him to father some kittens without sustaining mortal injury.

These tales of s.e.x and violence seems far removed from the cozy world of cat and devoted owner, but these are of course the ways in which the next generation of cats is created. Only about 15 percent of cats in the Western world come from planned mating, the great majority arising from liaisons initiated by the cats themselves. Since most pets are now neutered before they become s.e.xually mature, most cat owners are only dimly aware of the shenanigans that undoctored cats get up to; spayed females behave as if permanently stuck between cycles of reproductive activity, and males that are neutered young never develop characteristic tomcat behavior, instead behaving more like neutered females. Tolerance of other cats is improved by neutering, but only up to a point, and family bonds established between brothers and sisters, or mothers and their offspring, are still apparent if the cats continue to live in the same house. Unrelated cats that are brought together by their owners to live in the same household, or that meet on the boundaries between adjacent gardens, still often display the natural antipathy that they have inherited from their wild ancestors. However, unlike those ancestors, today's domestic cats can also establish close bonds with humans, bringing a new dimension to their social networks.

CHAPTER 8.

Cats and Their People The relationship between cat and owner is fundamentally affectionate, surpa.s.sed in its richness and complexity only by the bond between dog and master. Cynics often suggest that cats trick people into providing food and shelter through false displays of affection, and that cat owners project their own emotions onto their cats, imagining that the love they feel for their pet is reciprocated.

We cannot dismiss these claims lightly, but surely we feel such affection for cats with good reason. Ferrets, which can be just as effective at pest control as cats, have never found a place in the hearts of the majority-although they do, of course, have their fans. Our emotional bond with cats does not stem from grat.i.tude for mere utility; in fact, many of today's cat owners find themselves disgusted by their cat's hunting prowess, while continuing to love them as pets. So it is indeed possible that we humans are somewhat credulous, drawn in by some quality cats possess that encourages us to anthropomorphize their behavior.

The most obvious reason we might think of cats as little people is the humanlike qualities of their facial features. Their eyes face forward, like ours and unlike those of most animals-including ferrets-which point more or less sideways. Their heads are round and their foreheads are large, reminding us of a human baby's face. Infants' faces are powerful releasers of caring behavior in humans, especially in women of childbearing age. The effects of cats on humans can be remarkable: for example, scientists have found that simply viewing pictures of "cute" puppies and kittens temporarily enhances people's fine-motor dexterity, as if preparing them to care for a fragile infant.1 Our preference for baby-faced animals is exemplified by the "evolution" of the teddy bear. Originally naturalistic depictions of brown bears, teddy bears changed over the course of the twentieth century, gradually becoming more and more infantile: their bodies shrank, their heads-especially their foreheads-grew, and their pointed snouts were transmogrified into the b.u.t.ton nose of a baby.2 The "selection pressure" that caused these changes did not come from the children that played with the bears-children of four and under are equally happy with a naturalistic bear-but instead the adults, mostly women, who bought them. Cats' faces did not have to evolve to appeal to us; they have always had the right combination of features to appeal to people. That's not to say they have achieved their maximum "cuteness": freed from the constraints of biological practicality, cat images have continued to evolve, reaching their apotheosis in the j.a.panese cartoon figure "h.e.l.lo Kitty," whose head is bigger than her body, and whose forehead is larger than the rest of her head.

Cats have a built-in visual appeal that we find attractive, but while this may persuade us that it might be pleasurable to interact with them, it cannot possibly be sufficient to sustain an affectionate relationship. Indeed, pandas look very appealing-for much the same reasons-and their image has helped raise millions of dollars for the World Wildlife Fund which adopted the panda as its logo more than fifty years ago. The WWF acknowledges the key role that the panda's apparent cuteness has played in its success, with fundraising campaigns such as "The Panda Made Me Do It."3 Yet no one who has encountered a panda would a.s.sert that it would make a great pet. They simply don't like people-or each other, much. Thus superficial cuteness, while important, cannot be the only quality that endears us to cats.

Cats owe their success as pets not just because of their looks, but because they are open to building relationships with humankind. As they became domesticated, they evolved an ability to interact with us in a way that we find appealing, and it is this that has enabled them to make the transition from vermin killer to treasured companion. Our need for cats as pest controllers has waned, but despite this they have increased in numbers, with the majority now pets first and foremost.

We feel affection for our cats, but what do they feel for us? Wildcats generally regard people as enemies, so the answer must lie in the way cats have changed during domestication. Cats have not traveled as far down this road as dogs. Dogs have largely worked with man, herding, hunting, and guarding, and have therefore evolved a unique ability to pay close attention to human gestures and facial expressions. Cats have worked independently of man, going about their pestcontrolling tasks alone and acting on their own initiative; their primary focus had to be on their surroundings, not on their owners. Historically, they have never needed to form as close an attachment to humans as dogs have. Nevertheless, even at the earliest stages of domestication, cats needed humans to protect and feed them during the times when the vermin they were supposed to eradicate were in short supply. The cats that thrived were those that were able to combine their natural hunting ability with a newfound capacity to reward people with their company.

Cats' attachment to people cannot be merely utilitarian; it must have an emotional basis. Since we now know that cats have the capacity to feel affection for other cats, why should they not feel the same emotion toward their owners? Domestication has enabled cats to extend their amicable social bonds to include not only members of their own (feline) family, but also to members of the human family that takes care of them. Because cat society is nowhere near as sophisticated as canine society, we cannot expect the same degree of devotion from every cat as dogs typically display toward their masters. The lasting loyalty feral cats can show toward members of their own family is the raw material on which evolution has worked, resulting in a new capacity to form affectionate bonds toward humans as well.

Every time I'm tempted to think that the skeptics might be right, and that cats only pretend to love us, I think back to one of my own cats, Splodge. He was a neutered, long-haired mutt-cat, as standoffish as his mother and sisters were outgoing and affectionate. He liked to sit on his own in a corner of the room, never on anyone's lap. If visitors came to the house, he would get up with an air of supreme reluctance, stretch himself, and slowly leave the room. He wasn't afraid of people; he just didn't like being disturbed. Yet, Splodge did like a couple of select people. One was a research student of mine who came to the house to play games with him as part of her doctoral studies. After the first few times, he came running as soon as he heard her voice at the door, presumably recalling the fun he'd had on her previous visit.

The other target of Splodge's affections was, thankfully, me. Whenever he noticed I had taken the car to work, he would sit in the front garden all day, even if it was raining, and wait for me to come home. On seeing my car return home, he would come running across the driveway and sit by the car. As soon as I opened the car door, he would push his way in, purring loudly. After a brief but excited tour of the car's interior, he would stand with his hind legs on the pa.s.senger seat and his front paws on my leg, and rub his face on mine. It is hard to argue that such a display was not driven by a deeply felt emotion, and by all appearances, that emotion must have been affection. This affection was unlikely motivated by a desire for food or reward-my wife usually fed Splodge, yet he never behaved in this way toward her.

The most convincing indicator that cats feel genuinely happy when they're with people was found by accident more than twenty years ago. Scientists were investigating why some wild felids were proving especially difficult to breed in captivity, guessing that many of the females were so stressed by being kept in small enclosures that they were unable to conceive.4 To develop a method for a.s.sessing stress, the scientists relocated two pumas, four Asian leopard cats, and one Geoffroy's cat from their familiar enclosure into an unfamiliar one. These are all territorial species, so the loss of their familiar surroundings should have caused them considerable anxiety. The scientists measured how much stress hormone (cortisol) appeared in each cat's urine, and, as they expected, it increased dramatically on the first day after the move, settling gradually over the next ten days or so as the cats adjusted to their new surroundings.

By way of comparison, they also a.n.a.lyzed the urine of eight domestic cats that were kept in zoo-type enclosures. Four of these were known to be very affectionate toward people, whereas the other four were somewhat unfriendly. All eight were given a daily veterinary examination, something many cats find mildly stressful, and as expected, this caused an increase in the amount of cortisol excreted by the four less sociable cats, showing that they had indeed felt stressed. When in their usual environment, the level of stress hormone in the four affectionate cats' urine was slightly higher than in the other four, suggesting that they specifically did not like being caged; when the additional daily handling by the veterinary staff began, their stress levels went down. Thus, these cats appeared to be mildly upset when left alone, but contact with people-even contact that the average pet cat might object to-had a calming effect. Although it is probably going too far to suggest that these cats were suffering from "separation anxiety," they seem to have been happiest when they were receiving attention from people.

With no access to a hormone a.s.say lab, owners can judge their cats' emotions only by their actions, and cats do not wear their hearts on their sleeves as dogs do. For most animals besides dogs, we can more logically examine the way they communicate as attempts to manipulate the behavior of other animals. Apart from animals that spend their entire lives in permanent social groups, evolution selects against total transparency. If every other member of its species is potentially a rival, it is not in an animal's best interest to shout with joy every time it finds something tasty to eat, a safe place to sleep, or an ideal mate. Conversely, if an animal is in pain or unwell, it attempts to hide all outward signs of discomfort for fear that this weakness will encourages a rival to drive it away. In solitary animals such as cats, exaggerated displays of emotion are therefore unlikely because evolution would have selected out these behaviors. Individuals that give little away leave more offspring than those whose behavior betrays their fortune, both good and bad.

Honest indicators of an animal's strength and health, such as a peac.o.c.k's tail or (in my opinion) a tomcat's pungent urine, usually evolve only when the interests of one set of animals-in both these instances, the females-are distinct from those of another set-the males. For their part, females evolve a way they can discriminate between males that are successful hunters from those that are not, one that the males cannot fake. The male peac.o.c.k's huge tail is the cla.s.sic example, requiring its bearer to be supremely healthy; a less-than-perfect tail clearly indicates to peahens that they should look elsewhere for a mate. Likewise, any tomcat that produces urine that stinks of the breakdown products of animal protein must be a good hunter, with no room for faking.

On the other hand, using a signal is a low-cost way of getting something-a means of obtaining something from another animal without a fight. For example, when kittens are tiny, their mother has no choice but to feed them: it is the only way they can survive, and she has already invested two months of her life in them. As they get older, it is in her best interest to persuade them to take solid food, but they often try to continue to take milk from her-and one way they do this is by purring.

Here we have the first indicator that purring, the cla.s.sic sign of a contented cat, might also mean something else. Purring is a rather quiet vocalization (see box on page 194, "The Rumble: How Cats Purr"), audible only over short distances, which almost certainly evolved as a signal produced by kittens, not between adult cats. In the cat's wild relatives, the only close-quarters interactions that occur, apart from mating, are those between mother and kittens. The kittens begin purring when they start to suckle, and the mother sometimes joins in. And as they start to suckle, the kittens knead their mother's belly to stimulate the flow of milk.

Cats don't purr only when they are in physical contact: some will also purr when they're walking around, trying to get their owner to feed them. This purr takes on a more urgent quality, particularly effective at getting the owner's attention. Some cats continue to purr even when their body language, for example a fluffed-up tail, says they are getting angry with their owner.5 Not so obvious to the average owner, adult cats purr also when there are no people in the vicinity. Studies performed with remote radio microphones show that cats may purr when they're greeting another cat that they're friendly with, when they're grooming or being groomed by another cat, or simply resting in contact with another cat. Occasionally, cats have also been heard purring when they were in deep distress, perhaps following an injury or even during the moments before death.

The Rumble: How Cats Purr Purring is an unusual sound for an animal to make; although it appears to come from the region of the voice box, it is (almost) continuous. For this reason, scientists once thought it to have been produced by the cat somehow making its blood rumble through its chest. A close examination of the sound reveals that it changes subtly between the in-breath and the out-breath, and almost stops for a moment in between. To see this in action, read the chart, known as a sonogram, from left to right: the higher the spikes, the louder the noise. The in-breaths are shorter and louder, and the out-breaths are longer.

Sonagram of purring: reading from left to right, the in-breaths are shown by the dense "spikes," with quieter out-breaths in between The rumbling sound is produced as the vocal chords are vibrated by a special set of muscles, like a low-pitched hum. The difference from a hum is that the basic sound is not produced by air pa.s.sing across the vocal cords and making them vibrate, but rather by the vocal cords themselves banging together, like an old-fashioned football rattle, ratchet, or gragger. Often the cat will also hum at the same time, but this is possible only on the out-breaths, reinforcing the sound of the purr and giving it a more rhythmic quality.

Some cats can add a further meow-like sound to their purr, making it seem more urgent to the human listener.6 These cats often use their "urgent purr" when soliciting food from their owners, reverting to the normal purr when they are more content. This version of the purr has not yet been recorded from kittens or from ownerless cats, so it may be that, like the meow, this is something individual cats learn as an effective way to get something they want.

Purring therefore seems to convey a general request: "Please settle down next to me." In the gentlest possible way, the purring cat is asking someone else, whether cat or human, to do something for it. When they purr, kittens are asking their mother to lie still for long enough to allow them to feed-something they cannot take for granted. As in the wild, where purring evolved, there may come a time when she will feel too hungry and exhausted to stay with them, and she may have to choose between feeding them and going hunting to feed herself. Although scientists have not studied this systematically, when adult cats purr to one another, they are presumably asking the other cat to remain still. I have heard my cat Libby purring while grooming her mother, Lucy, in a rather aggressive manner, sometimes placing one paw across her mother's neck as if to hold her down.

Purring conveys information to those with an ear for it, but not necessarily the emotional state of the cat. Of course, it may occur when the cat is happy; indeed, this may be the norm. But often, a purring cat-whether a kitten suckling from its mother, or a pet enjoying being stroked by its owner-purrs not to show that it is contented, but instead to prolong the circ.u.mstances that are making it so. On other occasions, the purring cat may be hungry or mildly anxious because it is unsure how its owner or another cat is going to react; it may even be experiencing fear or pain. Under all these circ.u.mstances, the cat instinctively uses the purr to try to change the situation to its advantage.

Mark Twain lightheartedly acknowledged that a purring cat might be conning him when he observed, "I simply can't resist a cat, particularly a purring one. They are the cleanest, cunningest, and most intelligent things I know . . ."7 To say that cats are cunning probably overstates their mental abilities: they do not deliberately and consciously deceive their owners and each other through purring. Rather, each cat simply learns that purring under certain circ.u.mstances makes its life run more smoothly.

With the purr in mind, we can see that the way our cats behave toward us can be widely misinterpreted. Science has demonstrated that a signal many owners interpret as a sign of love may sometimes mean something else. Insofar as we know, purring is not central to affectionate relationships in cat society, apart from its original role in the bond between a mother cat and her kittens. However, other signals that we tend to overlook may in fact be more genuine displays of affection. Relationships between adult cats seem to be sustained mainly through mutual licking and rubbing, so we should examine whether these also indicate affection when our cats direct such behavior toward us.

Many cats lick their owners on a regular basis, but scientists have not yet investigated why this should be. Cats that do not lick their owners may have been put off because their owners resisted being licked in the past; the cat's tongue is covered with backward-facing spines that work well at disentangling fur, but can feel harsh on human skin. Conceivably, some cats may lick their owners because they like the taste: some researchers speculate that they lick us for the salt on our skins, but cats don't seem to have a strong preference for salty flavors.8 The most likely explanation is a social one, that the cat is trying to convey something to its owner about their relationship. The question is, what that something is. It may vary from one cat to another, as also seems to be the case when one cat grooms another (allogrooming).

The reason must be basically affectionate, because two cats that don't like each other never groom each other, although grooming can apparently reunite two cats that have recently quarreled. Cats licking their owners may sometimes be attempting to "apologize" for something the cat thinks it has done wrong-possibly something the owner has not even noticed, but that has some significance for the cat. However, a cat that licks its owner's hand with one paw placed over her wrist may be attempting to exert some sort of control over her. Until we know more about why cats groom one another, we can only speculate on why they groom us.

Cat owners also engage in a tactile ritual with their cats, of course, when they stroke them. Most owners stroke their cats simply because it gives them pleasure, and because the cat also seems to enjoy it, but stroking may also have symbolic meaning for the cat, possibly subst.i.tuting for allogrooming in some and possibly for rubbing in others. Most cats prefer to be stroked on their heads than any other part of the body, precisely the area toward which cats direct their grooming; studies show that fewer than one in ten cats likes to be stroked on the belly or around the tail.

Many cats do not simply accept their owner's stroking pa.s.sively; rather, they regularly invite their owners to stroke them, perhaps by jumping on their laps or rolling over in front of them. These rituals may not have any underlying significance, simply being mutually agreed exchanges that particular cats and their owners have learned will lead to pleasurable interactions. But while the owner has to initiate the actual stroking, most cats then indicate precisely where they wish to be stroked by offering that part of the body, or by shifting their position to place it under the owner's stroking hand.9 By accepting our petting, cats are doing more than enjoying themselves: in their minds, they are almost certainly engaging in a social ritual that is reinforcing the bond with their owner.

Some scientists have speculated further, that the cat is also deliberately inviting its owner to take up its scent. Cats may prefer to be stroked around the cheeks and ears because those areas are equipped with skin glands that emit perfumes that appeal to other cats, and the cat wishes the owner to take on the smell of these specific glands.10 Cats have similar glands on areas of skin that they don't generally like their owners to touch, such as at the base of the tail, so this theory implies that the cat does not want its owner to smell of these other areas. The subtle smell of the cat, virtually undetectable to our noses, will inevitably be transferred onto the owner's hand by the act of stroking, but this exchange may not have much social significance for the cat. If it did, cats would presumably be constantly sniffing our hands; while this does sometimes happen, cats do not do so obsessively. More likely, the primary function of the stroking ritual lies in its tactile component.

While touch is very important to cats, a common visual signal, the upright tail, is probably the clearest way cats show their affection for us. In the same way that an upright tail is a sign of friendly intentions between two cats, so it must be when directed at a person. When cats raise their tails to another cat, they usually wait to see whether the other will reciprocate before approaching, but this is obviously impossible when the recipient is human. Presumably each cat learns enough about their owner's body language to be able to work out, first, whether they've been noticed-they tend not to raise their tails until they have been-and second, whether the owner is ready to interact. Or at least, most cats do: my long-haired cat Splodge sometimes startled me by approaching while my back was turned, and jumping up to rub his head against the side of my knee.

Since this tail-up signal seems unique to the domestic cat, we do not know whether it evolved first as a signal to be directed at people, and then became useful to maintain amicable relationships with other cats, or vice versa. The latter seems more likely, however: because the raised tail has its origins in kitten-mother interactions, all cats are presumably born with an innate sense of its significance, and adults are therefore able to extend its use for interactions with other cats. The alternative explanation seems more far-fetched, since we would have to a.s.sume that the first people to domesticate the cat found this gesture so appealing that they deliberately favored cats that did it every time they met them.

As when two cats meet, a cat approaching its owner with tail raised will often rub on its owner's legs. The form that the rub takes seems to vary from cat to cat, and despite years of research I am still uncertain whether there is any significance in which part of the body the cat uses to rub. Some rub just with the side of their head, others continue the rub down their flank, and some routinely make contact with head, flank, and tail. Many simply walk past without making any contact at all. A few, like Splodge, jump as they initiate the rub, so that the side of the head makes contact with the owner's knee, and the flank caresses the owner's calf.

Some more nervous cats often prefer to perform their rubs on a physical object nearby, such as a chair leg or the edge of a door. Splodge's great-great niece Libby was a cla.s.sic example. Even confident cats will sometimes do this when they don't know the person well, even though they are perfectly happy to rub on their owner's leg. Indeed, most times this happens, the cat is probably just redirecting its rub onto an object because it is confident that the object, unlike a familiar person, will not push it away. However, sometimes when this happens it looks as though the cat is also scent marking the object with the glands on the side of its head. Scent is certainly deposited: when I've invited such cats to rub on posts covered in paper, those pieces of paper excite a great deal of interest when presented to cats in other households. However, the redirected rubbing behavior is not performed in precisely the same way as when a cat is deliberately scent marking an object. The difference can easily be seen if the same cat is presented at head height with the blunt end of a pencil, mimicking the protruding twigs that are many cats' favorite targets for "bunting" with their heads.

Rubbing can only be a sign of affection. Because many cats rub most intensely when they are about to be fed, they have been accused of showing nothing more than what we British call cupboard love. However, few cats confine their rubbing to occasions when they are expecting to receive a tangible reward. When two cats rub on each other, they exchange no food or any other currency; after the rub, each usually continues with what it was doing before. Such an exchange of rubs is a declaration of affection between the two animals-nothing more, nothing less.

Cats also rub on animals other than cats and humans, even animals that do not understand the significance of the ritual and are unlikely to give anything in return. Splodge used to perform the tail-up/rubbing ritual on our Labrador retriever, Bruno. Bruno was already a couple of years old when Splodge arrived in our house as an eight-week-old kitten; although he had not been brought up with cats, Bruno was too laid back to think about chasing them, so Splodge regarded him as a friend from the very start. Of course, Bruno never fed Splodge-quite the opposite: as a typical Lab he was all too eager to finish up any uneaten cat food-so the cupboard love explanation cannot possibly apply. Again, the only plausible significance must be social.

Scent must become transferred from cat to owner during rubbing, but this does not seem to hold any particular significance for the cat. Most owners certainly seem oblivious to this possibility-although apparently not Mark Twain, who once observed, "That cat will write her autograph all over your leg if you let her."11 If the primary motivation for rubbing was to leave scent behind, cats should constantly try to sniff people's legs to discover if any other cat had left its scent there. Of course, our habit of changing our clothes regularly cannot help, but the resulting confusion might lead to more sniffing rather than less. All the evidence points to rubbing, like stroking, as a primarily tactile display.

When two cats rub each other, they don't do so in equal measures, and a similar asymmetry seems to apply when cats rub on humans and other animals. When two cats of a different size approach each other, the smaller cat usually rubs on the larger, which usually doesn't reciprocate. When Splodge rubbed on Bruno, who was substantially larger, Splodge's instincts probably told him not to count on any particular response. Likewise, when our cats rub on us, our greater height, along with their knowledge that we are in control of most of the resources in the house, probably lead them to expect the rub to be one-sided. They are showing their affection for us in a way that doesn't demand a response-which is just as well, because unless we bend down and stroke them, we generally don't reply, at least not at the time.

Cats presumably find rubbing on us rewarding in its own right-if they didn't, they'd probably give up doing it-and like most of their attempts to communicate with us, they learn how to do it gradually. Kittens spontaneously rub on older cats with which they are friendly, and continue to do this as they get older. However, after they arrive in their new homes at around eight weeks of age, young cats (especially females) may take several weeks or even months to start rubbing on their new owners, as if they need time to work out how best to use this behavior to cement the relationship. Once the habit is formed, however, it seems to become fixed.12 Cats are more than intelligent enough to learn how to get our attention when they need to. Many use purring to persuade us to do something they want us to do, and a few invent their own personal rituals, such as jumping onto laps, or walking along the mantelpiece dangerously close to valued ornaments. However, the meow comes nearest to being their universal method of attracting our attention.

Purring is too quiet and low-pitched to be of much use for this sort of summons. Cats also have a greeting call that they use toward one another, a brief, soft "chirrup" sound; for example, mothers use this when returning to their kittens.13 Some cats will also use this chirrup to greet their owners: my cat Splodge used to greet me with this sound when he came back from a roam around the garden. Knowing a bit about cat behavior, I would try to chirrup back at him-something he evidently appreciated, as this exchange became something of a ritual between us.

The meow is part of the cat's natural repertoire, but they rarely use it to communicate with other cats, and its meaning in cat society is somewhat obscure. Feral cats occasionally meow when one is following another, perhaps to get the cat in front to stop and partic.i.p.ate in a friendly exchange of rubs. However, feral cats are generally rather silent animals, nowhere near as vocal as their domestic counterparts. While all cats are apparently born knowing how to meow, each has to learn how to use it to communicate most effectively.

The meow is much the same wherever the cat happens to live, confirming that it is instinctive. Every human language has a linguistic representation of this type of call: The English cat "mews," the Indian cat "myaus," the Chinese cat says "mio," the Arabian cat "naoua" and the Egyptian cat "mau." To ill.u.s.trate how difficult it is to interpret the cat's language, her "mew" is spelled in thirty-one different ways [in English alone], five examples being maeow, me-ow, mieaou, mouw, and murr-raow.14 Domestication appears to have subtly modified the sound of the cat's meow. All Felis silvestris wildcats can make a meow sound, wherever they live, whether the north of Scotland or South Africa. Meows performed by southern African wildcats, Felis silvestris cafra, are lower-pitched and more drawn out than typical domestic cat meows. When researchers played recordings of these wildcat meows to cat owners, the owners rated them significantly less pleasant than domestic cat meows.15 During the course of domestication, humans may have selected for a cat with a meow that is easier on our ears. However, it's equally possible that the meow of the direct ancestor of the domestic cat, Felis (silvestris) lybica, was different from that of its southern African cousin.

Feral cat meows are not as guttural as wildcat meows, but not as sweet-sounding as those of pet cats. Keep in mind that feral cats are genetically almost identical to pet cats, which suggests that their calls-as much else in cat behavior-are profoundly affected by each individual's early experiences of people. Pet kittens don't start meowing until after they're weaned, so as they become old enough to meow, they likely try out a range of meows on their owners, quickly finding that higher-pitched versions produce a more positive reaction. As with so much of cat behavior, differences between the wild meow and the domestic meow seem to be partly genetic and partly learned; domestication has enhanced cats' ability to learn how to use their meow, but may have also altered its basic sound.

Cats can also modify their meows to suit different situations: some are coaxing, and others more urgent and demanding. They do this by altering their pitch and duration, or by combining the meow with another of their calls, perhaps a chirrup or a growl. Owners often say that they know what their cats want from the tone of their meow. However, when scientists recorded meows from twelve cats and then asked owners to guess the circ.u.mstances under which the meow had been uttered, few guessed correctly. Angry meows had a characteristic tone, as did affectionate meows, but meows requesting food, asking for a door to be opened, and appealing for help were not identifiable as such, even though they made sense to each cat's owner in the context in which they were uttered.16 Therefore, once cats have learned that their owners respond to meows, many likely develop a repertoire of different meows that, by trial and error, they learn are effective in specific circ.u.mstances. How this unfolds will depend on which meows get rewarded by the owner, through achieving what the cat wants-a bowl of food, a rub on the head, opening a door. Each cat and its owner gradually develop an individual "language" that they both understand, but that is not shared by other cats or other owners. This is, of course, a form of training; but unlike the formalities of dog training, the cat and the owner are unwittingly training each other.

If we can decode them, the meows that we inadvertently train each of our cats to use may provide us with a window into their emotional lives. Our universal recognition of the "angry meow" and the "affectionate meow" suggests that each has an underlying and invariant emotional component, as the names I've used for them imply. The cat uses the others, the "request meows," simply to attract its owner's attention. The context within which they occur provides clues on what the cat wants-whether it is sitting beside a closed door, or walking around the kitchen gazing at the cupboard where the food is kept. The meows themselves are probably emotionally neutral.

Meowing to be let indoors Cats demonstrate great flexibility in how they communicate with us, which rather contradicts their reputation for aloofness. Cats come to realize that human beings do not always pay attention to them, and so often need to be alerted with a meow. They learn that purring has a calming effect on most of us, as it did on their mothers when they were kittens. They learn that we like to communicate our affection for them through stroking, which fortuitously mimics the grooming and rubbing rituals in which friendly cats indulge with one another. They may even learn, through our lack of reaction, that we are oblivious to the delicate odor marks that they leave behind on our furniture and even our legs.

We could consider some of this behavior manipulative, but only to the extent that two friends negotiate the details of their relationship. The underlying emotion on both sides is undoubtedly affection: cats show this in the way they communicate with their owners, using the same patterns of behavior that they employ to form and maintain close relationships with members of their own feline family.

Owners who expect long, intense interactions with their cat are frequently disappointed. Unlike most dogs, cats are not always ready to chat, often preferring to choose a moment that suits them. Cats are also nervous of any indication of a threat, however imaginary that might be; as such, many do not like being stared at, staring often being an indicator of impending aggression if it comes from another cat. The most satisfying exchanges between cat and owner are often those that the cat chooses to initiate, rather than those in which the owner approaches the cat, which may regard such uninvited advances with suspicion.17 Do cats think of us as surrogate mothers, equals, or even as kittens? Biologist Desmond Morris considers that at least two of these might apply, depending on the circ.u.mstances. When cats bring freshly caught prey home to "present" to their owners, Morris a.s.serts that "although usually they look upon humans as pseudoparents, on these occasions they view them as their family-in other words, their kittens."18 Mother cats do bring prey back to the nest, but presumably their behavior must be activated by some combination of hormones and the presence of kittens. Male cats and females without kittens don't do this, nor do female cats attempt to treat their owners as their kittens in any other way.19 A much more likely explanation for the unwelcome "gift" deposited on the kitchen floor is that the cat has simply brought its prey home, intending to consume it at its leisure. The place where it caught the prey will almost certainly contain scent marks indicating that other cats may be nearby, so what better way of avoiding an ambush than to return to the protection of its owner's home? However, when the cat gets there, it seems to remember that while mice are good to catch, they are not nearly as tasty as commercial cat food-hence the prey is abandoned, to its owner's revulsion.

It seems implausible that cats think of human beings as their kittens, given the size difference between us. On the other hand, it is logical to a.s.sume that they regard their owners as mother subst.i.tutes. Much of the cat's social repertoire appears to have evolved from mother-kitten communication. Our cats' behavior indicates that they take our greater size and upright stance into account when they interact with us. For example, many cats routinely jump up on to furniture to "talk" to their owners, and many others would probably like to do so, but recall that their owners haven't approved in the past. That they rub on us without necessarily expecting a rub in return, and the apparent exchange of grooming when they lick us and we stroke them, suggests that while they do not regard us as their mothers, they do acknowledge us as being in some way superior to them. Perhaps this is simply because we are physically larger than they are, so we trigger in them behavior that they would under different circ.u.mstances direct toward a bigger or more senior member of their feline family. Perhaps it is because we control their food supply, or at least (for a cat that hunts outdoors) the tastiest options available, mimicking the situation in which a few individuals can limit access to food for other cats in a large feral colony. Both of these a.n.a.logies refer to situations that wildcats never encounter and that have existed only since cats began to become domesticated, so the underlying behavior has presumably only evolved during the past 10,000 years. Hence, we must a.s.sume that their relationship with us is still in flux. A definitive answer on how cats perceive us thus remains elusive; for the time being, the most likely explanation for their behavior toward us is that they think of us as part mother subst.i.tute, part superior cat.

"Presenting" a dead vole An affectionate relationship with people is not most cats' main reason for living. Our cats' behavior shows us that they are still trying to balance their evolutionary legacy as hunters with their acquired role as companions. They form strong attachments not just to the people they live with, but also to the place where they live-the "patch" that encompa.s.ses their supply of food. Most domestic dogs, in stark contrast, bond to their owners first, other dogs second, and their physical surroundings third. This is why it is easier to take a dog on a vacation than a cat: most cats must feel uneasy when they're uprooted from their familiar surroundings, and certainly behave as if they do. They generally prefer to be left at home when their owner goes away.

Considered logically, well-fed neutered cats should not feel the need for a territory of their own, neither for s.e.xual purposes nor for nutrition. Most cats that have gotten into a daily routine of being fed high-quality food by their owners do not hunt. Those that do hunt are not particularly enthusiastic-after all, they don't need the nourishment-nor do they usually consume the prey they catch, which is generally less tasty than commercial food. Nevertheless, most do still patrol an area around their homes, if their owners allow them to do so. In urban areas, many do not stray far-in one study, some only ventured about twenty-five feet from the cat-flap, and none more than fifty-five yards. In rural areas, this increased to between roughly twenty and one hundred yards, depending on the cat.20 But what motivates these cats to pursue what appears to be an unnecessary remnant of their ancestral territorial behavior?

Regular food certainly seems to reduce the area cats patrol. Socialized cats without a permanent owner, who cannot rely on a regular food supply, travel significantly farther from their "home" base, perhaps a couple of hundred yards. This is still nowhere near as far as unsocialized feral cats travel, even those that have been neutered and are therefore no longer roaming in search of mates; these cats may range for a mile or more. Neutering of males once they become adults does not substantially reduce the distance they travel, as if they are still in the habit of locating as many receptive females as possible, even though they no longer have the ability to do anything if they did find one. It is clear that any cat that realizes it cannot rely on a regular source of food instinctively increases the size of its range to compensate.

That the ranges of well-fed, owned cats are so small suggests that many are not deliberately hunting at all. If the opportunity presents itself, they may grasp it; but without hunger goading them on, they may not do so with single-mindedness. Nevertheless, the cat's brain does not link hunting and hunger together tightly, and for good evolutionary reasons. A single mouse provides few calories, so a wildcat must kill and eat several each day. If cats waited to start each hunting trip until they felt hungry, they'd unlikely get enough to eat. So, even a well-fed cat, seeing a mouse within catching distance, is likely to grasp the moment-or, rather, the mouse.

Since most pet cats do not hunt seriously, it seems curious that they spend so much time outdoors, apparently sitting and doing nothing or meandering between the same locations they visited the previous day. When my cat Splodge was about eighteen months old, I borrowed a lightweight radio-tracking transmitter and fixed it to an elasticated "safety" collar so I could locate him wherever he went.21 I already knew that he spent about a third of his time in our garden or on the roof of a nearby garage. Once the collar was on, I learned that he crossed the garden behind the apartment block next door into a strip of woodland beyond, about an acre in area. An older male cat lived in the apartment complex, accounting for Splodge's disinclination to remain there for more time than absolutely necessary: I'd already seen the two cats engaged in standoffs on several occasions, so they were undoubtedly aware of each other. Splodge rarely ventured beyond the trees-to my great relief, since there was a busy road not far beyond. He would sometimes remain in the same location for hours at a time, usually one of a few favored vantage points such as a branch of a fallen tree, before moving on to another site or returning home. He rarely seemed to be hunting: occasionally, he caught a mouse or a young rat, but would let birds fly past him without batting an eyelid. I often wondered, and still do, what was going through his mind as he maintained his surveillance of the same small area, day after day, year after year.

Even the best-fed cats seem inclined to pursue this outdated territorial agenda. They clearly feel the need to maintain s.p.a.ce of their own, and many are prepared to fight other cats to hang onto it, even though they no longer need to hunt to survive. Their instinct to hunt is dampened, though not entirely eliminated, by their never going hungry. Both sides of this behavior are understandable from an evolutionary perspective. Hunting puts cats at risk, so they possess a mechanism that dampens the need to hunt actively when they have not gone hungry for a long time. However, few of today's pet cats are more than a small number of generations away from feral cats that have had to live on their own resources, and for whom a productive territory was a necessity. Going back a few generations further, when commercial cat food was neither universally available nor nutritionally complete, virtually every cat would have had to hunt for much of its food, and would therefore have to defend an area in which it had exclusive access to prey. Too few generations have elapsed for this instinctive need to have disappeared-however archaic it may seem to a pet cat's owner.

Each cat's desire to establish and then defend a territory inevitably brings it into conflict with other cats. In the countryside, pet cats live mostly in cl.u.s.ters, dictated by our habit of living in villages rather than evenly spread across the landscape-which would presumably be their preference, since that would match their own ancestral pattern. In this situation, they reduce conflict with other cats by each foraging out in different directions, the pattern of territories resembling the petals of a flower with the village or hamlet at its center. The somewhat relaxed notions of "ownership" of cats in rural areas, with farm cats becoming pets and vice versa, also mean that if two cats find themselves living too close together for comfort, one can usually find a vacant niche nearby.

Most cat owners nowadays expect their cat to live wherever the owners choose. Perhaps not fully understanding the cat's need to form an attachment to its physical environment, owners a.s.sume that it is enough to provide food, shelter, and human company, and that if they do, the cat will have no reason not to stay put. In reality, many cats adopt a second "owner," and sometimes migrate permanently.22 Surveys I have performed in the UK confirm this picture: many cats stray and get "lost" even from apparently high-quality homes. We see some clues as to what happens to many of these cats from the significant proportion of owners-as many as one-quarter in some areas-who, when asked where they obtained their cat, reply, "He just turned up one day." These were not feral cats: that they were so keen to adopt a new home shows that they had recently been someone else's pet, and were desperate to find a new owner. A few of these cats may have been genuinely lost, still searching unsuccessfully for their original home, but most were probably migrants, looking for a better place to live than their original owner could provide. For the few instances of this that I could trace back to the original home, the cat had evidently left not because it wasn't being fed properly, or was unloved; something else must have gone seriously wrong for the cat to abandon the certainties of a regular food supply and, in most cases, feelings of attachment towards its original owner. The most likely explanation is that these cats could not establish an area where they could feel relaxed, safe from challenges from other cats.

The threat could have come from the next-door cat, or even from another cat living in the same household. Within the home, opportunities for conflict are rife. Just because two cats have the same owner does not mean they are going to get along. Many pay heed to the cardinal rule of cat society: proceed with caution when meeting any cat that has not been a part of your (cat) family for as long as you can remember. Many cat owners seem oblivious to this principle, blithely a.s.suming that when they obtain a second cat, the two will quickly become friends. Although this is generally true of dogs, cats are more likely to merely tolerate one another (see box on page 212, "Signs that Cats in a Household Do or Don't Get Along with One Another"). To reduce conflict, they often set up separate, if overlapping, territories within the house, but may continue to sc.r.a.p with each other sporadically. In surveys of owners with two cats, roughly one-third report that their cats always avoid each other if they can, and about a quarter say that they fight occasionally. The two cats will probably come to respect the others' favorite places to rest-the larger or original cat will generally take the prime spots-but tension may remain if both cats are fed in the same room or if there is only one litter tray. They may also compete for the cat-flap, if one cat claims it as being within its core territory. Feeding each cat in a different room and providing several litter trays in different locations (not the rooms used for feeding) can often make the situation more tolerable for both cats.

For any cat that is allowed outdoors, other cats in the neighborhood will also be a source of conflict and stress. When a cat comes into a new household-whether or not it was previously cat-free, it will most likely find that it has been parachuted into the middle of a mosaic of existing cat territories. If there are already cats on two, three, or even all four sides, the owner's garden will almost certainly "belong" to one or more other cats; for cats, garden walls are highways, not boundaries to be respected. The new cat will have to establish its right to roam in "its own" garden, and it will be able to do this only by standing up to the other cats that previously had undisputed use of the s.p.a.ce. Challenges over territorial boundaries may continue for years: in one survey, two-thirds of owners reported that their cat actively avoided contact with other cats in the neighborhood-and, frankly, the other third probably hadn't looked out of their windows often enough. One-third reported that they had witnessed actual fights between their own cat and their neighbor's.

Signs that Cats in a Household Do or Don't Get Along with One Another23 Cats that see themselves as part of the same social group generally *hold their tails upright when they see one another *rub on one another, either when walking past or alongside one another *regularly sleep in contact with one another *play gentle, "mock-fighting" games with one another *share their toys Cats that have set up separate territories within the house will tend to *chase or run away from one another *hiss or spit when they meet *avoid contact with one another: one cat may always leave the room when another enters *sleep in widely separated places; often, one will sleep high up, perhaps on a shelf, to avoid another *sleep defensively; the cat has its eyes closed and looks as if it is asleep, but its posture is tense and its ears may twitch *apparently restrict one another's movements on purpose-for example, one cat sitting for hours by the cat-flap, or at the top of the stairs *watch one another intently *look unusually tense when they're in the same room *interact separately with their owner-for example, they may sit either side of the owner to avoid physical contact with one another Surprisingly few owners seem particularly bothered about such conflicts until they begin to affect their cat's health: a bite can turn into an abscess requiring veterinary treatment. More ominously, the cat may become so stressed, or its movements so restricted, that it begins to urinate or defecate in the house (see box below, "Signs that a Cat Is Failing to Establish a Territory Outside Its Owner's House"). Even if the various cats in an area eventually arrive at a truce, owners may inadvertently reignite the conflict by temporarily removing their cat, perhaps boarding it elsewhere for a couple of weeks while they go on holiday. Encouraged by the signs that the cat may have left for good, such as fading scent-marks, absence of sightings, one or more of the neighbors' cats may start to encroach into what was previously that cat's territory. When that cat returns, it may have to reestablish its rights all over again.

Signs that a Cat Is Failing to Establish a Territory Outside Its Owner's House24 *Not leaving the house even when encouraged to do so *Waiting to be let out by the owner rather than using its cat-flap (because there might be a rival cat ready to ambush it on the other side) *Neighbor's cats entering the house through the cat-flap *Leaving the house only if the owner is in the garden *Excessive time watching out of windows *Running away from windows and hiding when another cat is spotted in the garden *Running into the house and immediately to a place of safety, far from the access point *Tense interaction with the owner, including rough play *Urinating and defecating in the house, by a cat that usually goes outside to do this but feels too insecure to do so *Spraying urine (scent-marking) in the house, especially if near to access points such as doors and cat-flaps (more likely in male cats than females) *Other signs of psychological stress, such as excessive grooming Increasingly, cat owners are avoiding such problems by keeping their cats indoors, although their motivation may be more to protect their pet from traffic, disease, or potential thieves (especially if it is a valuable pedigree animal) than from social stress. Restricting the cat to one relatively small area for its entire life can induce stresses of its own. Although the practice of keeping cats indoors has been common among apartment-dwellers for more than thirty years, we have little systematic research into whether domestic cats find this confinement stressful. To see how we should expect indoor cats to behave if they were stressed by confinement, we must therefore look further afield, to their wild ancestors.

Wild felids often react badly to being confined. Both "big cats" such as lions and "small cats," such as jungle and leopard cats, are p.r.o.ne to the habit, once commonplace in zoos all over the world, of pacing to and fro in their cages.25 Of other types of animal, only bears are as badly affected, and like most of the cat family, they are also solitary territorial carnivores. We do not fully understand why these animals pace, but the reasons probably arise from a mixture of frustration at not having access to a large enough hunting territory, even though their nutritional needs are being more than satisfied, and "boredom": well-fed carnivores do instinctively sleep for much of the time, but many seem to crave mental stimulation when they are awake. Changing the way the wild cat is fed can provide the latter: rather than simply providing one daily meal that can be wolfed down in seconds, zookeepers now provide food several times each day, and the cats must make an effort to get at least some of it-for example, by feeding bones that have to be cracked open, or placing the food in puzzle feeders that the cat has to work at for an extended period of time.

When considering whether domestic cats kept indoors are likely to suffer, we might first examine whether they show signs of objecting to being spatially restricted, and whether they show signs of "boredom." Repet.i.tive pacing is surprisingly rare in domestic cats, considering that this is the most common abnormal behavior observed in cats kept in zoos. This difference may have already evolved prior to domestication; when in captivity, the domestic cat's wild ancestor (Felis silvestris) is more p.r.o.ne to "apathetic resting" (taking no notice of its surroundings) than to pacing. The difference could also be a consequence of domestication. Whichever the culprit, domestic cats seem to have lost much of their ancestors' spontaneous "drive" to roam. We do not know why this loss would have benefited cats: for most of their 10,000-year history, they have had to hunt for their living.

Domestication may have given the cat a much greater flexibility in its territorial behavior. Wild felids generally feed on prey dispersed over the landscape around them, and therefore have always needed a large territory: they may need to venture farther afield if food is scarce, but in all their evolutionary history, they would never have encountered a situation where food was so locally plentiful that they could afford not to go out foraging day after day. Domestic cats, by contrast, have adapted to hunting in and around what are, by comparison, very small areas-human settlements-while retaining the ability to expand their hu