Synapsida

A Brief History of Zokors

2026-05-10T14:11:00.004+01:00

The rodents are the largest group of mammals, in terms of number of species, and they are divided into several families. Some of these are familiar, such as the mice, voles, squirrels, and gophers, but others are much less so, at least to the majority of people in the West. This includes, for example, the zokors.

There are a couple of reasons why zokors are not as well-known as some other rodent groups. For one thing, they only live in northeast Asia, an area that it's fair to say doesn't receive much attention in the Western world. For another, even if you did live there, you wouldn't see them very often, because they are burrowing animals that don't like coming to the surface if they can help it. Rather like moles, you might see the mounds of earth they leave, but not often the animal itself.

How the zokors fit into the wider rodent family tree has undergone significant revision over the last few decades. Back in the 20th century, they were considered a subfamily in the mouse family, but then, far more rodents were in those days than is the case now, the older broad definition of "murid" having been refined. Then, at the dawn of the current century, early molecular studies showed that they weren't mice at all but a kind of hamster. Which was a bit surprising, since, while they look a bit like hamsters superficially, they are are different in almost every other respect.

In 2004, however, a flaw was discovered in the original genetic analysis. Specifically, it turned out that the "zokor" whose genetics had been analysed was, in fact, a Siberian hamster. Reanalysis of an actual zokor showed that it belonged to a different rodent family, the blind mole-rats. Some zoologists have argued since then that they're unique enough to be given their own family, but today, they are generally placed as one of three subfamilies with the blind mole-rat family, technically referred to as the spalacids.

Zokors are moderately large rodents, between about 15 and 25 cm (6 to 10 inches) in length. They burrow through the soil using the claws on the front feet, which is a different method than that employed by the mole-rats, which use their teeth. Like the blind mole-rats, they try to avoid venturing above ground, and they have very small eyes and no external ears. (Blind mole-rats, it should be noted, are not the same thing as the African mole-rats, which include the naked sort). They survive by eating roots and underground seeds, and are thought to be quite common in those areas where they live.

There are six living species of zokor, three belonging to the genus Eospalax, and living in central China, and three others living in parts of Siberia, Mongolia, and Manchuria. These last three are usually placed together in the genus Myospalax, from which the subfamily takes its technical name (Myospalacinae), although one species sometimes gets its own, Siphneus.

But we also know of fossil species and, using these, we can piece together something of the history of the group. It's sometimes the case, when there are a small number of living species, that they represent the last gasp of some group that was once more diverse and perhaps wider spread. But that's not the case with the zokors, which don't ever seem to have spread far beyond their current home, and to have lived much the same specialised lifestyle as they do now.

Blind mole-rats as a whole are thought to be a very early branch within the mouse-like rodents, splitting off before, for example, the hamsters and voles separated from the true mice and their kin. This would place their origin around 20 million years ago, but the oldest known fossils that we can specifically call zokors date back to only 9 million.

These belong to Prosiphneus, which lived during the Late Miocene across much of eastern Siberia and northern China and even reached into southwestern Tibet, close to the border with India This is a much wider range than any living species of the group, but the world was warmer then than it is now, with relatively mild conditions likely persisting across the region before they became split into more isolated pockets around 5 million years ago at the dawn of the Pliocene. This breakup of formerly hospitable terrain may have led to the geographic fragmentation of the family we see today, where each species lives in relatively restricted areas.

This change in the climate would have isolated early zokors into at least two different regions, perhaps divided by the Takla Makan desert and the surrounding Kunlun and Tien Shan mountain ranges. One group, represented by the Pliocene genus Pliosiphneus, remained in China, where it is likely that the very first zokors had appeared a few million years earlier. That died out by the end of the epoch, but one isolated group had survived, most likely after becoming trapped in Mongolia by the changing climate of the surrounding regions, and evolved into the genus Episiphneus. Common during the Ice Ages, it eventually moved back south again, giving rise to the modern Chinese zokors, which include the most well-studied species.

Over on the other side, the Pliocene sees the appearance of Siberosiphneus, initially in the western reaches of Siberia. Indeed, the oldest known species of the genus lived on the banks of what is now the Irtysh River, on the eastern side of the Ural Mountains, and thus about as far west as you can get without leaving Siberia. This new species has several features in common with its presumed ancestor Prosiphneus, indicating that it may represent one of the first members of its lineage, shortly after the split with the Chinese species.

This, or something like it, would have given rise to the modern species living further north. During the Ice Ages, this 'Siberian' lineage spread further afield, perhaps driven by the expansion and retreat of the glaciers. While Siberia itself had relatively little in the way of massive ice sheets, being too dry to accumulate enough snow to form them, the permafrost tundra would have shrunk and returned as the interglacials came and went, and permanently frozen subsoil is obviously not going to be encouraging for an animal that would have had to dig through it and still find enough fleshy plant roots to eat.

So it's at this time that the Siberian zokors would have spread east, leading to the appearance of the two modern species in Mongolia and Manchuria. To back up this pattern of sudden dispersal, we also have fossils that appear all-but indistinguishable from the modern Siberian species discovered on the western side of the Urals - and thus in Europe, not Asia - where they would have lived during the Middle Pleistocene. They probably didn't remain there long, perhaps just during an interglacial, but that they got there at all is surprising for something long assumed to be unique to Asia, especially when there was a mountain range in the way.

[Photo by "Avustfel", from Wikimedia Commons.]

Strange Carnivores of Madagascar

2026-05-03T11:40:00.000+01:00

Madagascar is the fourth-largest island in the world, almost three times the size of Great Britain, or half again the size of California. We should not be surprised, therefore, that it has a considerable amount of native wildlife. However, it's also noteworthy that it has been an island for a very long time - certainly for far longer than Britain has been.

In fact, Madagascar became an island around 91 million years ago, over 20 million years before even the likes of Tyrannosaurus appeared on the scene. Moreover, this was when it split away from what was then the island of India, with its break from Africa being almost twice as far back in time. But, even if we take that younger date for the beginning of its isolation from any sort of 'mainland', it's well before any of the sort of mammals we would recognise today had evolved. There's no equivalent here of mammoths nipping across the English Channel.

Yet Madagascar is not devoid of native mammals. Obviously, island isolation doesn't cause any problems for marine mammals that might be found off the coast, and it's not much of a problem for bats, either. But there is far more than this, animals that have somehow made the trip from elsewhere to this large and (at the time) empty land, probably by clinging to some stormswept piece of driftwood. It seems remarkable, but you've got a very, very long time to get it right, and you only need to do it once for any given group of animals.

Lemurs are probably the most famous example of this, but Madagascar also has rodents, small insectivorous mammals, and even carnivores. This last group, however, have posed a significant puzzle to biologists struggling to fit them into the wider carnivoran family tree. It's been known since the early 19th century that, due to the structure of their ear, they are all more related to the "cat" side of the order than to the "dog" one... but it's less obvious how they are related to each other, or to anything on the mainland.

The first species were formally described in the 18th century, before the concept of animal "families" had been developed. They were both placed in the genus Viverra, along with other small, bitey, vaguely cat-like animals, such as the civets I have been looking at recently. But, as more species were named, it became clear that they really appeared quite different from one another and, in some cases, from pretty much anything else.

For around 150 years, different naturalists and taxonomists came up with different schemes, placing some species in one group and others elsewhere, or just throwing up their hands and leaving the weirder ones on their own. Some were placed with the civets, some with the mongooses, and one was regularly classified as a cat, albeit one thought to have branched off from all other cats very early on.

At the heart of this, at least once evolution had become widely accepted, was just how many times new species of carnivorous mammals had reached Madagascar from elsewhere. It's something that, strictly speaking, only needs to have happened once... but that doesn't mean it did, not if their descendants are so obviously different from one another. A lack of fossils on the island, which has few deposits of the right age, didn't help matters.

In 1945, George Gaylord Simpson published an influential classification of mammals that, among many other things, answered this question: it had happened three times. The Madagascan carnivores belonged to three different groups that had, over the course of millions of years, each arrived separately on the island. Of the seven species known at the time, four were mongooses, two were palm civets, and the other one was a viverrid, but odd enough to be given its own subfamily. Aside from debates about whether mongooses were distinct enough to be given a family separate from that of the viverrids, this scheme was followed for the remainder of the century.

It was, as we now know from molecular and genetic studies, wrong.

In 2003, a genetic analysis compared the Madagascan carnivorans with each other and with similar animals elsewhere. As has been confirmed several times since, it turned out that all of the species really did share a single common ancestor that had somehow crossed over from Africa. There had only been one event, and all those features that had confused earlier researchers were down to a mix of rapid adaptation to new niches and cases of parallel evolution.

We now know that the closest relatives of these animals are, in fact, the mongooses, but the split is so far back (about 19 million years ago) that we can't reasonably place them in the same family. Not only that, but the Madagascan carnivores turn out to be more closely related to hyenas than to what had been the other prime contender, the palm civets. An old name for the group was dredged up from a 19th-century classification and raised to family level.

The Madagascan carnivores now had their own family: the Eupleridae.

Given all of this, it should not be surprising that it's impossible to describe what a "typical euplerid" looks like - there is no such thing. With every other carnivoran family, we could reel off a list of features that define them, even if there might be a few species here and there that don't quite fit the usual pattern. But, here we can't; Simpson may have been wrong about them belonging to three different groups, but it isn't as if he didn't have his reasons. That's what it looks like, with some closely resembling mongooses, some civets, and some... not.

There are, in fact, only two unique features that we can say unite the euplerids: they all descend from a single common ancestor, and they all live on Madagascar, and nowhere else. And that's it, which is unusual, especially when you can't even tell the first one is true without conducting genetic analysis.

Of course, there are some features that they have in common; they're just not unique. For instance, the structure of the bones around the ear does follow the "cat-like" pattern seen in their relatives, which at least tells us which half of the carnivoran family tree they belong to.

We can also make generalisations about their teeth; each half of each jaw has three incisors, a canine tooth, three or four premolars and one or two molar teeth. This is very similar to the pattern in mongooses, but the actual shape of the teeth varies remarkably between species, so this isn't really telling us very much beyond the fact that they are suited for a broadly carnivorous diet.

They are all forest-dwelling animals, because that wasn't really anything else to live in before the arrival of humans sometime between the third and sixth centuries AD. Since then, many of the old forests have been cut down to make way for farmland, especially in the central highlands, with the result that euplerids are no longer found through much of the interior. However, the forests that do survive are still reasonably extensive, and vary more than one might think, due to the differing weather patterns on the east and west coasts. The various species, therefore, each have their own preferences for forest type.

The euplerid family is comparatively small, with, depending on who you're asking, anything from eight to ten species. Which, to be fair, is the same as the number of living bear species and is not spread across half the globe as they are, but still. It is divided into two subfamilies, one of which itself contains unusually varied forms.

The other contains those that were long considered to be mongooses. While we now know that isn't what they really are (although, admittedly, they are closer to them than to any other non-euplerid), they are still regularly referred to as such. There has, however, been a trend in recent years to use a new word for them, taken from the native Malagasy language: "vontsira".

Next time, I will be taking a look at them.

[Photo by Rob Foster, from Wikimedia Commons.]

Covid and the Landscape of Fear

2026-04-26T13:22:00.000+01:00

Left to their own devices, animals will wander about wherever there is food or other useful resources to be found. They would be limited only by the physical properties of their landscape, such as the inability to cross a wide river or the presence of a barren mountain range. Reality, of course, is never quite this simple.

The earliest use of the term "landscape of fear" that I could find on a quick search referred to a study about how female college students felt while walking particular routes alone after dark. Put in this context, one can well understand why people don't necessarily pick the simplest or most direct route to their objective, and why there are certain places one might simply wish to avoid altogether.

The idea, however, was soon extended to ecology, whereby the distribution of animals across a landscape is determined, not just by physical factors or the direct action of predators, but by fear. The idea is that particular prey animals might become less common in a particular area, not because predators are killing them in those areas, but because they fear that they might.

So, depending on how the predator you are most concerned about hunts, you might avoid open areas where you can easily be seen, or denser ones where they can easily hide. Puma/cougar/mountain lions, for instance, kill most of their prey near forest edges, and animals such as elk adjust their own movements accordingly. At the very least, prey animals will approach such areas cautiously, remaining vigilant, and thus spending less time feeding or resting than they might elsewhere.

While the general idea had been around for decades, it became extended to the modern biological concept of the "landscape of fear" around 2010, where perceived predation risk, as indicated by any of various measures, is plotted against the physical geography of an area. When we are studying how animals use particular habitats, we need to take fear into account just as much as the availability of useful resources.

But we can extend the concept further. The landscape in question need not be physical; it could also be temporal. Consider those college students again. They might avoid certain places, or be sure to travel in groups or with other protection if they have to (i.e. remaining vigilant), but they might also, so far as possible, just not walk the streets after dark.

This, in the technical parlance of ethology, is "diel behaviour". That is, how your activity changes throughout the day. For example, what is considered the founding study to use the landscape of fear concept showed how elk in Yellowstone Park avoid certain areas because they are afraid of wolves. The assumption was that the elk just found those areas scary, as well they might, and so stayed away from them. But it turns out that elk are quite happy to forage for food in places frequented by wolves if they think that the wolves are asleep.

The landscape of fear can therefore be said to change, not just slowly as habitats change with rivers drying up or whatever, but from hour to hour throughout the day. There are places you might be happy to visit during the day that you wouldn't want to go at night.

Perhaps the clearest example of this comes not from fear of predation, but fear of humans. Even where wild animals are not at direct risk of being hunted by humans, they tend to avoid us, and that's just as much a landscape of fear as anything caused by regular predators. The details can be complex, but a common trend is that animals that have to live near humans become more nocturnal.

That's the opposite of what we would usually do if we found a particular place threatening, but that's rather the point. Humans are mostly active during the day, so if you want to avoid them, then, so far as you possibly can, it is better to be active at night. This is something seen across a wide range of animal species, although the details naturally vary.

But, if the landscape of fear changes over the course of a mere day, how long does it take animals to adjust to changes in human activity? Animals are, after all, capable of learning and, while it would likely depend on the animal species in question, we can easily imagine them becoming more or less wary as the human activity around them waxes and wanes. For instance, a study at a recreation area in Utah showed that mule deer and raccoons are less active at weekends, especially near camping grounds.

If we're talking about changes in human activity levels, however, it's hard to argue that the largest disruption to daily life in modern times was anything other than the COVID pandemic. It's well known that many animals changed their behaviour once they realised that humans were not travelling about so much, and some scientific studies have attempted to quantify that.

One, for example, found that stone martens in Poland were more active in the early evenings during lockdown than they normally would be. Stone martens are nocturnal anyway, so would have had little motivation to exploit quieter days, but they could make use of more of the night if humans weren't around after dark in the evening.

Such studies continue to come out, such is the time taken to conclude data analysis, write everything up formally, and get it published and available. Most tend to focus on either urban areas, where we might expect the effect to be greatest, or recreational areas, where animal monitoring equipment may already be set up and that should have been almost totally deserted during lockdown. One new one, however, looked at the more complex matter of the interface between the two.

This study concerned the Jasper Ridge Biological Preserve operated by Stanford University in the San Francisco Bay Area. It's a 1,193 acre (4.8 km²) preserve described as an "outdoor classroom" for nature studies and education, lying on the traditional land of the Muwekma Ohlone tribe, and thus also referred to by the native name of 'Ootchamin 'Ooyakma. Crucially, it is almost entirely surrounded by suburban housing, close to the university. However, in addition to being a wild area in itself (there are cougars and rattlesnakes, for example), there is a narrow corridor linking it to the nearby wilderness of the Santa Cruz mountains. All of which gives it a different character from a typical city park.

On the 12th March 2020, all of the usual guided tours to the Preserve were cancelled, and four days later, so were all educational and research visits, with access restricted only to essential personnel. Educational services resumed in spring 2021, meaning that the former daily visits were paused for an entire year. Camera traps placed to monitor the activity of the local wildlife, however, remained running throughout.

In fact, they found relatively little effect. Of the six mammal species they were able to evaluate, only brush rabbits (Sylvilagus bachmani) - a type of cottontail found only along the West Coast - changed their behaviour to become notably more active during the day. Jackrabbits shifted the peak of their activity to be less active in the evening after the lockdown ended than they had been during it, but since they were already active then before it started, they don't seem to have noticed the beginning of the period, only its end.

The other four species studied, foxes, bobcats, cougars, and black-tailed deer, changed their activity little or not at all.

The researchers conclude that this may be because the animals in this particular area were already used to human activity, and moreover, human activity that didn't involve hunting and killing them. That is, they already felt secure moving about whenever they wished, and the fact that the humans that hadn't been bothering them in the first place weren't there any more didn't make much difference to them.

They also speculate that the brush rabbits being different had little to do with human activity. They are, of course, relatively small herbivores and had a great deal to fear from the foxes and bobcats. And, while those species were no more active during the day than they had been before, they were actually more active during their usual hunting times at night - perhaps feeling more confident in entering the area in general, rather than at particular times of day. And, with more predators around at night, it made sense for the brush rabbits to stay hidden then and cut back on their nocturnal forays.

But otherwise... well, it seems that there is more than one way to cut back on the landscape of fear.

[Photo by Eric Kilby, from Wikimedia Commons.]

Viverrids: The Odd Ones Out

2026-04-19T13:38:00.000+01:00

Binturong

Most viverrids have a similar appearance: slender, long-snouted animals, many of them spotted or brown-coated. One, however, stands out as particularly distinctive. This is the binturong (Arctitis binturong) and it's the largest species of "civet", with a total body length ranging from 61 to 96 cm (24 to 38 inches) and weighing between 9 and 20 kg (20 to 44 lbs), similar to small dog such as a Jack Russell or border terrier. Moreover, they appear somewhat larger than they are because, unlike other civets, they have a thick shaggy coat that obscures their actual body.

They are mostly black in colour but with grizzled fur that makes them look grey, and prominent white whiskers. The tail is long and prehensile, and the feet have large, rough pads, giving them an enhanced grip. In other respects, however, they are civet-like, with the same general pattern of teeth and other detailed anatomical features, including a simple perineal scent gland.

They live across southeast Asia, reaching as far west as Bengal and eastern Nepal, and across many of the nearby islands, to Palawan (one of the Phillipine islands) in the north and western Java in the south. There are no universally recognised subspecies, but the island and mainland populations are genetically distinct and may well warrant such a status, and there is evidence that those on Palawan may be different yet again.

Binturongs are, like other civets, forest-dwelling animals. While they are happy to inhabit open woodlands or even plantations if left to their own devices, today, they are more often found in dense jungle, preferring the lowlands, but also seen as high as 2600 metres (8,500 feet). This is due, of course, to human interference, with binturongs being popular as pets or display animals, and, in some parts of their range, harvested for their pelts or as bushmeat. There have also been attempts to use them to make coffee, using the same method as in palm civets.

This has led to their elimination from the more populous, lowland regions of Vietnam, and numbers are declining in Cambodia and Laos as well. This, together with the fact that the amount of forest it would even be capable of living in is also declining, leads to it being classified as a "vulnerable" or "threatened" species, but not yet an endangered one.

Binturongs are more closely related to palm civets than the regular sort, which may help explain their preference for trees. Nonetheless, while they do often climb, and the rough pads on their feet probably help with this, they spend at least as much time on the ground, where they walk on the flats of their feet like a bear, instead of on their toes like a cat. When they do climb, it is slowly and carefully, using the flexible bushy tail for balance.

Rather like cats, however, they spend a lot of their time snoozing, getting in over twelve hours a day if they can. Although their activity pattern is somewhat irregular, and they can be up and about at almost any time of day or night, they are most active at dawn and dusk. They are solitary animals, with home ranges of around 6 km² (2½ square miles) recorded in some studies, but these overlap sufficiently that it seems very unlikely that they are at all territorial.

Like other civets, they are thought to be omnivorous, eating insects and other small animals alongside fruit. However, the evidence for this is fairly limited, with the few analyses conducted on their dung in the wild finding little, if any, animal matter, and a diet that consists almost entirely of figs. Certainly, their movements seem constrained by whether or not there are fig trees nearby, and they are often seen climbing them. In experiments, fruit seeds seem to survive the passage through their gut without difficulty, suggesting that they may be a significant disperser of fig seeds in the wild.

Perhaps the best-known fact about binturongs is that they smell of buttered popcorn. This is due to a chemical present in their scent glands and their urine, and is thought to be important in communication between the animals, especially with regard to reproductive availability.

At least in captivity, they breed year-round, with an 82-day oestrus cycle, although more young are born between January and March than in other months. The female can remain sexually receptive for a full 15 days of the cycle, longer than is the case for many other animals. In another twist on the common pattern in mammals, females are larger than males, and it's clearly she that calls the shots. Most of the time, binturongs make high-pitched whines and squeaks, growling or hissing when annoyed, but prior to and during the act of mating, the female purrs, presumably to encourage the male. Mating takes place in trees, and the pair will engage in several bouts before they finally go their separate ways.

After a 92-day pregnancy, the female gives birth to a litter of tiny blind young, each weighing about 300g (11 oz.). Litters of up to six have been reported, although as the female only has four teats, it's likely that not all the members of larger litters survive.

The binturong diverged from other palm civets relatively early on, perhaps around 12 million years ago, during the Middle Miocene. However, there is another species that branched off on its own path even earlier than that, around 18 million years ago, making it the oldest single-species lineage within the entire family. That doesn't necessarily make it the most "primitive", since it will have changed during that time just as everything else has, but it does indicate that whatever it was doing worked well enough for it that it had little reason to innovate by forming new species.

This is the small-toothed palm civet (Arctogalidia trivirgata), which lives in broadly the same area as the binturong, but is somewhat more restricted, reaching Bengal in the northwest, and the western tip of Java in the southeast. while being absent from many parts of Myanmar, Thailand, and Cambodia in between. Perhaps because it is relatively inconspicuous, it is less hunted than many of its larger relatives, and remains common in those areas where it is found.

Small-toothed palm civet

It is on the small side for a civet, although not remarkably so, with a brownish or grey coat and, in most (but not all) individuals, a white streak on the snout. Its key distinguishing feature is the presence of three black stripes down the back, each formed from a series of tightly packed spots. Unusually, only the females have a scent gland, which is small and located next to the vulva, rather than reaching up to the anus as it would in most other species.

Very little is known of the species, with most scientific reports consisting of little more than a sighting of one up in the trees going about its daily business. Still, even from these, we can at least say that it is found in evergreen tropical forests, preferring the lowlands where possible. They are nocturnal, apparently solitary, and spend most of their lives up trees, where they can clamber about with ease, using their unusually long tail for balance. This is one of the reasons why we don't see them much, since even the camera traps used to survey the nocturnal animals in a forest are set at ground level, where the civets rarely venture.

The few examinations of the diet of wild animals suggest a roughly even mix of fruit and insects, although one is recorded to have eaten a squirrel. In 2015, a pair were observed mating, a chance encounter by the researchers aided by the fact that, while the pair partly concealed by the leaves of their tree, it was daylight, when we would normally expect them to sleeping. The contrast with the binturong's technique could hardly be more extreme, with the male clearly dominating the female. After her angry hisses failed to ward him off, he overpowered her, wrestling her down with his forepaws, and then holding her neck with his teeth while he aggressively mated with her every couple of minutes over the course of about half an hour. After which, they both fell asleep, so it might have been less traumatic than it sounds.

From the few animals held in captivity, we know that litters are generally small, with only two or three young. They are born with comparatively pale fur, and are initially blind and helpless, opening their eyes after 11 days and being weaned at two months.

That brings me to the end of the viverrids, a relatively small family of mammals in terms of number of species. However, when I started this series, I mentioned that the family used to be much larger, including other animals that we now know are not closely related. Next time, I will turn to take a look at some of those that ended up being split off, a group most of whose members are probably even less well-known than the civets...

[Photos by Donar Reiskoffer and Tontan Travel, from Wikimedia Commons.Cladogram adapted from Nyakatura and Bininda-Emonds, 2012.]

Eocene (Pt 3): Sword Teeth and River Pigs

2026-04-12T16:59:00.001+01:00

Anoplotherium

Today, by number of species, the most successful group of large mammalian herbivores are the terrestrial artiodactyls - broadly speaking, the "cloven-footed mammals". Except for Australia, they are a key element of almost any mammalian fauna. Their first great burst of diversity occurred during the Middle Eocene, over 40 million years ago, when many new forms appeared across the Northern Hemisphere, some unique to particular places, others more widespread.

Most, however, were nothing we would recognise today. There were as yet no cattle, antelopes, deer, or pigs, and many of the creatures that did live that far back left, so far as we can tell, no living descendants.

One group that may be an exception are the anthracotheres. These migrated to Europe from Asia and have been proposed as possibly including the ancestor of hippopotamuses (although this has been questioned). This is in part due to a physical resemblance between the two. For example, they did not have hooves and, while they had the basic anatomical arrangement seen in cloven-footed animals, they walked on all of their toes. There was a full set of five of these on their front feet, and four on the back, albeit with those that would form the hoof in later animals taking most of the weight.

Anthracotherium itself, first discovered in France, but known to have lived across Asia as well as Europe, was perhaps the most hippo-like of all the Eocene anthracotheres. It was about the size of a modern pygmy hippo, roughly two metres (6 feet) in length with a heavy build and short thick legs. However, the head was long and narrow, perhaps more horse-like than anything else despite the short neck. The teeth were strong with sharp cusps, suggesting a mixed diet of leafy browse, fruit, and grazing. Elomeryx was another common anthracothere in Europe, smaller than its relative and possibly more agile. Its teeth were more adapted to eating tough plants, and like its larger relative, it was probably semi-aquatic, reinforcing the similarity to hippos.

The choeropotamids have had a complex taxonomic history, sometimes being grouped together as relatives of the anthracotheres, and sometimes split into different groups or given different names. As currently defined, it probably isn't a real evolutionary unit, but a collection of similar forms, with much of the debate being about where the dividing lines should go.

Choeropotamus itself, which by definition must belong to the family, was a pig-like creature in both size and general shape, and, like the anthracotheres, had a long narrow head. Distinguishing features include the lower canine teeth taking the form of incisors, and the tooth immediately behind forming a tusk that almost resembles a regular canine. Taking this and other features of the teeth into consideration, it was probably omnivorous, perhaps with a diet similar to modern pigs. The name translates as "river pig", although whether it was genuinely semi-aquatic, as was thought when it was first described in the 19th century, is highly debatable.

We only have the skull and ankles of Choeropotamus, limiting our knowledge of its overall form. We have rather more of Amphirhagitherium, known from England, France, and Germany. It was smaller than Choeropotamus, and had a relatively slender build, accentuated by the fact that, unlike many artiodactyls today, it still had a long tail. It is likely to have had a similarly omnivorous diet to its larger relative, but with more of a focus on herbs than leafy bushes. It has the same tusks as well, which might indicate at least some meat in its diet, but which could, as in Choeropotamus, also have been used in fights for mates or for defence against predators.

Depending on the taxonomic definition used, the choeropotamids may be unique to Europe. They are not alone in this, with two other significant families being found only on the continent. The anoplotheres (literally "unarmed beasts") have an unusual mix of advanced and primitive features. The shape of their teeth resembles that of ruminants, which would enable them to eat relatively tough plants. However, all ruminants (and, indeed, pigs and choeropotamids) have a diastema, a gap between the front and back teeth that allows the cheeks to manipulate vegetation as it is being eaten. Anoplotheres did not, and still had relatively normal canine teeth as well.

Diplobune is one of the smaller members of the family, although still weighing in at around 20 kg (44 lbs). It had three toes on each foot, still relatively elongated, and shaped such that it probably still partly walked on its soles, rather than tip-toe, as hoofed mammals do. This unusual shape was formerly interpreted as indicating that the animal was semi-aquatic, using its feet to paddle, or at least to support itself on soft muddy river banks. However, the internal structure of its ears is thought to rule out any possibility of underwater hearing, as well as indicating that it was rather slow-moving. A flexible index finger is an unusual feature; one interpretation suggests that this might have been used as a thumb, implying that the animal might have been able to climb trees to get at the leaves there. If true (which it may well not be), this would certainly be unusual for an artiodactyl.

Anoplotherium is the largest known anoplothere, as well as the first to be named, all the way back in 1804. It was at least the size of a sheep, and probably quite a bit larger, with most estimates suggesting that it weighed at least 230 kg (510 lbs). It has been suggested that the shape of the snout indicates the presence of a long tongue, similar to that of giraffes, and presumably used to pull at small branches and leaves in the same way.

It had just two toes on each foot (and was formerly interpreted as being related to camels on these grounds) and retained a long muscular tail, perhaps reminiscent of that of a kangaroo. The shape of muscular attachments elsewhere on the skeleton and of some of the joints suggests that it may have been able to stand on its hind legs, allowing it to browse from the lower branches of trees. The limbs are too short for it to have been able to do this very effectively, so it likely braced itself with its forelimbs while doing so but even this would have enabled it to reach food that no similar European animal of the time could have done.

The xiphodontids were another group unique to Europe. Appearing around 40 million years ago in the Late Eocene, they belonged to a group regarded as more advanced than the earlier forms. Xiphodon was the size of a small gazelle, perhaps standing around 80 cm (2'7") at the shoulder. It was, however, one of the largest, with others, such as Haplomeryx, perhaps only reaching half that height.

Its placement as a relatively advanced form is due to the shape of the teeth; the name means "sword-tooth" and refers to the sharp crescent-shaped cutting blades on the molars that would have been ideal for cutting up leaves. Like the anoplotheres, they lacked a diastema, and they had distinctly cloven feet, holding the side toes off the ground. Xiphodon, the only genus for which we actually have limb bones, had long legs and a slender build that would have made it a fast runner, perhaps resembling miniature llamas, although they were not close relatives.

The closely related amphimerycids were, perhaps, even more changed from their ancestral form. They had similar teeth, and walked on only two toes of each foot, but they had also developed the cannon bone seen in living ruminants, further adapting the limb for running. Only two genera are known, both unique to Europe. Both were tiny animals, at least by the standards of artiodactyls. Amphimeryx, which was the larger of the two, is thought to have weighed around 1.5 kg (3 lbs 6 oz.), about the size of a rabbit, while Pseudamphimeryx was even smaller. They had large eyes and a narrow snout, somewhat resembling the modern chevrotains.

Despite the similarity of the limb bones, it's unlikely that the amphimerycids were related to, let alone ancestral to, the modern ruminants; if anything, they may be closer to the camels. The similarities are probably just a case of parallel evolution. Nonetheless, it is likely that the first ruminants did appear towards the end of the Eocene although, since the defining feature is the four-chambered stomach, which obviously wouldn't fossilise, we don't know for sure.

The best candidate for an early ruminant from Europe is Gelocus, a small hornless animal which had limbs resembling those of musk deer and could, conceivably, be related to their ancestor. Bachitherium is another possibility. It was larger, standing around 60 cm (2 feet) tall at the shoulder and, while otherwise similar to Gelocus, had tusk-like upper canine teeth, but no incisors in the upper jaw - also a point of similarity with musk deer. Other differences rule out a direct relationship, and it was probably a side-branch in evolution, the only known member of its family, and may or may not have been an actual ruminant.

The earliest known artiodactyl is Diacodexis. Although some of the earliest fossils are known from Portugal, some of similar age are also known from North America, as well as Asia. It seems unlikely that a single genus could spread so rapidly early on, so it's entirely possible that the primitive nature of the animals has made it difficult to distinguish different genera on either side of the Atlantic. Either way, however, we cannot, at present, say on which continent the artiodactyls originally appeared, since the identity of their immediate ancestors is not clear, either.

On the other hand, it seems likely that the first artiodactyls resembled Diacodexis, since whether it is one genus or two (or more) there is no doubt that it lived very early on - almost at the beginning of the Eocene - and is distinctly primitive. It did already have the trend towards putting its weight on the third and fourth toes, as cloven-footed animals do, despite still having five toes on the fore-feet and four on the hind ones. It was small, about the size of a rabbit, with a short but slender snout ideal for rooting around in leaf litter, while the teeth suggest that it still had a relatively omnivorous diet. The hind-limbs were noticeably longer than the front ones, increasing the resemblance to a rabbit and likely indicating that it was capable of jumping. Analysis of the ears implies that it was best at hearing high-frequency sounds.

There were, however, many other herbivores in the island forests of Eocene Europe, and next time I will take a look at some of them.

[Photo by "ghedoghedo" from Wikimedia Commons.]

Hedgehogs versus Killer Robots

2026-04-04T14:40:00.002+01:00

Hearing is a key sense for just about any mammal species. While humans, and primates in general, place particular importance on vision, this isn't true for many other mammals. Granted, that's usually because of the significance of scent, which we humans are particularly bad at, but hearing can hardly be discounted, even for us.

The range of hearing, however, varies dramatically between different species. For humans, the typical range is from about 20 to 20,000 Hertz, which covers about 10 octaves (middle C is 261 Hz) although the upper range drops off with age, and there is some leeway under perfect conditions. Many mammals can hear outside this range, with smaller animals, in particular, being able to hear higher notes. This can be of more than academic interest, since it means that the soundscape a particular animal lives in may not be the same as our own, and that can have a bearing on conservation - we might think somewhere is quiet and peaceful, but other species might not agree.

Most research on the effects of human-created noise on animals has been conducted on whales and birds, which are likely to be particularly affected. If we want to extend it to other animals, it could be useful to know just what their hearing range is, and one recent study asked this question specifically of hedgehogs (Erinaceus europaeus).

I have discussed hedgehogs before, but here's a quick summary. There are about 18 species of living hedgehog found across Eurasia and Africa, but here we are talking about the one simply called "the hedgehog" in British English, because it's the only one found in that country - and, indeed, across the rest of western Europe. They are not at all closely related to porcupines, which are rodents, with the spines being a case of parallel evolution. Instead, the hedgehog family (which also includes some spineless species that aren't normally called "hedgehogs") is most closely related to the shrews, partly explaining their similar diet of insects and worms.

The European hedgehog is not an endangered species, but its population has declined dramatically in recent years. In the UK, for example, numbers are thought to have fallen by about 20% between 2010 and 2020, and similar figures apply to other countries. Premature deaths are largely due to road traffic, but poison bait intended for rats can be another factor, as can robotic lawnmowers.

It was the last of the above that motivated the recent hearing study. The researchers were not interested in the general effect of human-created noise on hedgehogs (that they try to cross roads suggests this may concern them less than it might) but on whether lawnmowers, and similar garden devices such as strimmers, could be made safer for hedgehogs. They reasoned that fitting the devices with ultrasonic emitters creating an annoying sound would warn the animals, allowing them to flee before being killed.

For this to be acceptable, the sound emitted has to be too high-pitched for humans to hear, or nobody would ever use them. Ideally, however, you don't want it to alarm cats or dogs either, and that puts more constraints on the frequency. A cat, for example, can hear sounds up to at least 65,000 Hz, roughly two octaves higher than the highest-pitched sound a human can detect. If hedgehogs can't beat that, ultrasonic screamers won't be of much use to families with pets.

The study was conducted on 20 injured, sick, or orphaned hedgehogs delivered to a wildlife rescue sanctuary in Denmark. After recovering from whatever their original complaint was, and having their ears examined to rule out hearing problems, they were sedated and fitted with electrodes under the skin to monitor the activity in their auditory nerves and brainstems. Afterwards, they got to relax on warmed heat pads (of the sort you might use for your cat) before being fed and eventually released into the wild.

All of this is designed to minimise the stress and discomfort of the animals being studied. But it does obviously have limitations. Most notably, although it should show what sounds hedgehogs are capable of hearing, it doesn't demonstrate what they respond to. If, for example, a sound is relatively quiet, the animal might ignore it even if it can hear it; in birds, the difference between what they can hear and what they'll respond to is about 20 to 30 dB, which is quite significant.

To test that, we would need to keep the hedgehogs in captivity for much longer, but they become particularly stressed if that happens. Which, even leaving aside any issue of unintended animal cruelty, could affect the results; a stressed animal will not respond in the way that a comfortable one will. So we have to do what's practical, with the hedgehog's interests in mind.

The result of this was that hedgehogs' peak hearing sensitivity was around 40 kHz. This is definitely ultrasonic - a full octave beyond even the highest pitch that most humans can hear. But it is similar in pitch to a dog whistle and so clearly audible to them (regardless of breed and size, if you're wondering) and even more so to cats. Still, while this demonstrates that hedgehogs can hear ultrasound, which we already knew, it's merely the peak brainstem response. That is to say, it's the pitch at which you can play the sound at the lowest volume and still have the animal able to hear something. Raise or lower the pitch, and you'll have to up the volume before the hedgehog will hear you, but it still might.

There has to be some limit, some highest pitch the hedgehog can hear if you blast out the sound loud enough. The study was not able to find out what that was, but the good news is that the highest tone they could produce still registered in the hedgehog's auditory nerves. This was 85 kHz, roughly three notes on the scale above the highest tone a cat can hear. The sound had to be quite loud at that point, so the ceiling probably doesn't go much higher, but the point is that it should be high enough.

You should, in other words, be able to create a sound at, say, 70-75 kHz that will alarm a hedgehog and save it from a lawnmower, but not frighten a cat.

Furthermore, the researchers were also able to examine the ears of a hedgehog that had arrived at the sanctuary with unsurvivable injuries from a rat trap, and had had to be euthanised. This showed that the malleus bone of the middle ear was connected to the bony ring around the eardrum by a tough, fibrous structure. This would increase the overall stiffness of the chain of bones in the middle ear, and is associated with ultrasonic hearing in other mammals, such as rodents and bats.

What we can't tell from this sort of study is why hedgehogs would want to hear ultrasound in the first place. They are solitary animals, so communicating with each other, while not impossible, isn't likely to be as important as it is in, say mice. Indeed, so far as we know, they don't make many sounds of any kind, although it's conceivable that we just haven't been listening at the right frequencies.

Another possibility is that it enables them to hunt for insects that make sounds in this range. This has previously been suggested for four-toed hedgehogs (Atelerix albiventris), a species native to the savannah and open grasslands south of the Sahara. We don't know that it would also be true of the European sort, but it's plausible, and it doesn't rule out other uses for the same sense.

Either way, the research team now plans to see if they can use this information to develop ultrasonic repellents to fit on those killer robot lawnmowers.

[Photo by Michael Gäbler, from Wikimedia Commons.]

The Thrush-beaked Consort Bird

2026-03-28T14:35:00.008+00:00

A modern thrush

This post will be the latest as of 1st April, so it's that time again...

About 40% of all living mammal species are rodents. Over a third of those that aren't are bats, leaving just 38% belonging to any other group - primates, cloven-hoofed mammals and all the rest. These are, at least in terms of speciation, clearly very successful body plans.

But the bias is even stronger with birds. A whopping 60% of living bird species belong to just one order: the Passeriformes. The chances are that, if you think of a "typical bird" the image that pops into your head is of a passerine. There are about 140 different families of passerine (the exact number being a matter of taste among ornithologists), of which 123 constitute the songbirds.

Songbirds are distinguished by advanced musculature in their syrinx, an organ analogous to the larynx (or "voicebox") in mammals. Unlike the larynx, which birds also have, it's located at the base of the trachea, near the lungs, rather than up at the top in the throat, but it's this that explains why songbirds are so good at... well, singing. So far as we can tell, this more complex syrinx only evolved once, so all songbirds belong to the same suborder within the passerines, and it's one that's obviously of huge importance.

Even so, that leaves 17 families - three of which have over 100 species each - that are passerines, but not songbirds. They do have the syrinx, because all birds do, but it's less complex, and any birdsong they produce is strictly limited. They're mostly tropical, so not necessarily familiar to those of us living further north, but we're talking about antbirds, tyrant flycatchers, and the like.

Passerines are typically small birds, with the raven being the largest by body weight, and most being much smaller than that. (The raven is, perhaps surprisingly, a songbird... just not a very musical one). The distinguishing feature of passerines as a whole, songbirds included, lies in the structure of their legs.

Like most birds, the feet of passerines are anisodactylous, which is to say, they have four toes in total, three pointing forward and one (what would be the big toe in a human) pointing backwards. Where they differ is that they have a mechanism for locking their tendons in place that allows them to sleep while perched. The conventional wisdom is that their grip on a branch is maintained by their own body weight pressing down on the lock. It's probably not quite that simple, but, nonetheless, the common English term for passerines is "perching birds".

Given their importance, it could be argued that the appearance of the passerines was the most significant event in the evolution of birds, other than the development of flight itself. However, we know relatively little about the early history of the passerines, or what the earliest ones might have looked like. That's largely because they are so small and fragile, being unlikely to leave even partially intact fossil remains. So, while molecular evidence and biogeography tell us that they most likely evolved somewhere in the Southern Hemisphere during the Early Eocene, with the songbirds probably first appearing in Australia shortly thereafter, the details remain obscure.

A recent discovery, however, sheds some light on one small corner of this puzzle... and it comes from a bird that wasn't originally thought to be a passerine at all.

In 2010, a fossil bird unearthed in Denmark was formally described for the first time, being given the name Morsoavis sedilis. The fossil was unusually complete, including the whole of one leg and part of another. It lived 54 million years ago, during the Early Eocene, and, unsurprisingly given that age, could not be placed in any living family of birds. Nonetheless, the evidence at the time suggested that it was a shorebird, a cousin to the bird families that contain, among others, the seagulls and plovers.

But there was an oddity; the hindleg had some features in common with those of passerines, suggesting that it may have been able to perch, rather than being adapted for wading as modern shorebirds are. Over the following years, more fossils of the same and closely related species were discovered near London and in Virginia in the US. Our best guess as to where these birds should fit in the larger family tree switched around a bit, but in 2023, Morsoavis and two of its presumed relatives were placed in a new family, the Morsoavidae.

Crucially, this family was identified, not as belonging to the shorebirds, but as a stem group of the passerines. Classifying it as a "stem passerine" means that it is not thought to have any living descendants, and is not a passerine itself, but is closer to them than it is to anything else alive today. In other words, this newly described family of extinct birds is thought to represent an early branch from the ancestors of the true passerines that ultimately failed.

Moreover, living as early as it did, it was probably a very close relative of those passerine ancestors, and so could give us, if not proof, at least hints as to what they may have been like.

Last year, a fourth genus of birds was added to the family, based on a fossil recovered from a lagerstätte in Wyoming. Lagerstätten are sites with exceptional preservation of not just bones, but also of the outlines of soft body parts, due to their burial in fine sediments without any oxygen, killing off most of the bacteria that would normally cause decomposition.

In this case, the fossil was almost complete, including the remains of the feathers. These latter are sufficient to indicate that it probably had a short tail and does not appear to have had a crest on its head. Its formal description gives it the name Consoravis turdirostris, which translates as "thrush-beaked consort bird", the first part having been chosen so as to sound similar to that of two of the other genera in the family - Morsoravis ("bird from Mors") and Soroavis ("sister bird"). It has been dated to 51.6 million years ago, during the Early Eocene.

A particularly key part of this new fossil is the preservation of the feet, which are notably incomplete in all the known fossils of Morsoravis. This enabled the researchers to conclusively prove that it isn't a shorebird, while various other aspects of the skeleton suggest that the "stem passerine" attribution is probably correct. However, it turns out that the foot was not anisodactyl, as any real passerine foot should be.

Instead, it's "semi-zygodactyl". This is a pattern seen in owls and ospreys, among other birds, and it means that the fourth toe can be rotated backwards on a bony flange so that it can face in either direction as the bird wishes. Most of the other stem passerines known have fully zygodactyl feet (that is, the fourth toe can only ever point backwards, so there are two toes to the front, and two to the back), so the morsoravids are closer to the modern pattern than some of their relatives from the same time period.

That they don't appear to be any closer in other respects suggests that this modification of the feet is a case of parallel evolution from an ancestor that, like parrots and woodpeckers, was fully zygodactyl. The authors suggest that, like ospreys, Consoravis and its relatives relied more on their feet for manipulating food than their beaks.

That beak, as the scientific name suggests, resembles that of a thrush. This isn't hugely helpful in itself, since thrushes are omnivores with diets that vary across the 194 living species. However, other aspects of the skull and legs suggest, according to the authors, that, while Morsoravis would have been suited to poking into the soil to pull out worms and other invertebrates, Consoravis used the grip of its feet to climb about in tree branches in search of fruit and insects, while also occasionally descending to the ground to forage for invertebrates among the leaf litter.

[Photo by Dion Art, from Wikimedia Commons.]

Viverrids: Long-tailed Palm Civets (and coffee...)

2026-03-22T18:02:00.003+00:00

Common palm civet

While the palm civets do, arguably, form a single group of related animals, they are spread between two subfamilies that diverged at least 23 million years ago, at the dawn of the Miocene epoch. This is only about 5 million years after the split between the true civets and the genets, so it's fair to say that the two subfamilies are distinct. The subfamily that includes the animal originally known in the West as a "palm civet" is that of the paradoxurines.

The animal in question is now known simply as the common palm civet (Paradoxurus hermaphroditus). The alternative name of "Asian palm civet" is also often seen, although it has the disadvantage of not clearly distinguishing it from all the other palm civets that live in Asia. Which, since there are no palm civets outside of Asia, isn't really narrowing it down much.

The common palm civet is, however, undeniably common. It was first scientifically described all the way back in 1777 by Prussian naturalist Peter Simon Pallas, although it's not known where he obtained his specimen. It could have been many places, since the common palm civet is widespread and would have been known to many European colonists at the time. In fact, it's found across pretty much the whole of southern Asia east of Pakistan, occupying almost the whole of India and Nepal, plus Southeast Asia and southern China as far north as Guangzhou. It is also common on islands, including Sri Lanka, western Indonesia, and the Philippines. For that matter, they have been introduced to islands further east, notably Sulawesi, which somewhat muddies the picture.

Superficially, it resembles many other civet-like animals, being around the same size and shape. It has numerous spots across its body, which merge into three parallel lines running down its back. The face has black and white mask-like markings, the exact shape of which varies considerably between individuals. The scent glands are much less developed than they are in most other viverrid species, present, but without much in the way of detailed structure. Notably, the tail is unusually long, not much less than the entire rest of the head and body put together at 33 to 66 cm (13 to 26 inches). It often, but not always, has a white tip.

Palm civets are forest-dwelling animals, but as their wide range might indicate, they are tolerant of many different types of forest. They are found among palm trees at least half of the time, but that may be because such forests are particularly common in that part of the world. Otherwise, any sort of forest will do, and they seem surprisingly willing to put up with human disturbance. Under some conditions, they actually prefer it and inhabit plantations alongside secondary growth and deeper, more primeval jungles.

A limiting factor may be the need for trees to sleep in during the day. They prefer those with dense foliage or creepers that can hide them, so while they may be common at night along roadsides, they also need denser woods nearby. On the other hand, many common palm civets sleep during the day in attics and similar human structures that offer good concealment.

Compared with true civets, palm civets eat a lot of fruit, to the extent that this constitutes the great majority of their diet. They do eat insects and small mammals such as mice and wild gerbils, but less so than one might expect, with one study finding such remains in less than one in four of their scats, while fruit seeds were virtually universal. They may switch to animal prey primarily when fresh fruits are scarce, although this happens less frequently in the tropics than it would further north.

The fruit in question varies across their range. In India, for instance, papaya seems particularly common, and they will also drink the sap from tapped "toddy" trees on coconut plantations. Their preference for relatively large-seeded fruits may make them an important seed disperser, especially in disturbed habitats where other wild frugivores may be uncommon. For example, rambutan, a relative of the lychee, is often described as a fruit eaten by monkeys, but, in fact, macaques tend to spit the seeds out, while palm civets swallow them whole and deposit them hours later up to 270 metres (295 yards) away.

Another fruit commonly eaten, in places such as Nepal and Indonesia, is that of coffee plants. These are actually berries, similar in form to cherries, with the seeds being what we make the drink from. This, of course, leads to what may be one of the more popularly known facts about civets in general: the production of kopi luwak, or "civet coffee". This comes almost exclusively from the common palm civet (and not from any of the "true civets") and is produced from coffee beans that have passed through the animal's digestive tract.

While this is something they eat in the wild, they don't restrict themselves to the arabica beans that humans generally prefer, and it certainly isn't something they do in large volumes. As a result, most kopi luwak is produced on battery farms, where the palm civets are force-fed suitable beans and their dung collected. The digestive process affects the chemical composition of the beans, reducing some of the compounds that make them bitter, and increasing the proportion of fat, which likely gives a fuller flavour to the end product.

So it's true that the resulting coffee tastes different from the regular sort and, from most people's perspectives, probably better. It's just whether or not you want to drink coffee sifted from the dung of an animal that almost certainly won't have been kept in humane conditions.

Brown palm civet

Palm civets are solitary, with adults sleeping alone, and rarely using the same nesting site on consecutive nights. Even so, they aren't especially territorial, with multiple animals having been seen peacefully eating together at the same tree before going their separate ways. As with other viverrids, they communicate using the scent from their perineal glands, which allows them to distinguish sex and the specific identity of individuals they are familiar with. Mating takes place throughout the year, with the female giving birth to a litter of up to five young two months later. Like those of cats, the newborn young are blind and helpless.

It has been suggested, on the basis of mitochondrial DNA, that the common palm civet represents three species: one on the mainland, one in peninsular Malaysia, Sumatra, and Java, and the other further east in Borneo and the Philippines. This does not seem to have been widely accepted yet, but it is becoming more popular, so that may well change soon. Two other species, however, have long been regarded as separate.

The brown palm civet (Paradoxurus jerdoni) is distinguished by its relatively uniform brown colour. It lives in the Western Ghats, a mountain range along the southwest coast of India, where it is found from almost the southern tip of India in Kerala to at least as far as Goa, and likely a little way beyond. It lives in rugged and mountainous woodlands, not venturing below 280 metres (900 feet) elevation.

One study found that they prefer to sleep through the day in fruit trees, notably mango and roseapple, often using abandoned squirrel nests to make themselves comfortable. About 90% of their diet consists of fruit, most commonly small berries, although they also eat some larger fruits such as Ceylon olives (which are unrelated to actual olives). Although they do occasionally enter plantations, they do not do so as frequently as the common palm civet, and their rugged homeland, while not immune to human encroachment, is perhaps less so than the lowlands. It also seems relatively numerous across a mountain range that's rather longer than many in the West may realise, and so isn't considered a threatened species.

Golden palm civet

We know even less about the golden palm civet (Paradoxurus zeylonensis) beyond the fact that it is golden to ruddy in colour, and lives in inland areas of Sri Lanka. They seem to be willing to use a range of different forest types, from sea level to low mountains and, like the common palm civet, tolerate those that have been altered by human activity. They are less likely to be seen near human settlements, but that may just be because the more numerous common palm civets keep them out. They may, however, be better at climbing, having been seen moving along telephone wires.

The masked palm civet (Paguma larvata) is much better known. It is another widespread species, being found across southern and central China, much of Southeast Asia, along the foothills of the Himalayas as far west as Pakistan, and on Borneo, Sumatra, and Taiwan. However, it is not in these areas that it has been most thoroughly studied, but rather in Japan, where it is not native. It has lived in the country since at least the late 19th century, with genetic analysis showing that the animals arrived there from Taiwan, but are sufficiently varied that they must have been introduced to the islands multiple times.

The masked palm civet is slightly larger than its relatives, weighing in at 4 to 5 kg (9 to 11 lbs) and has a grey, unspotted body. What makes it distinctive is the mask, which is more prominent than on the common palm civet, with a white stripe down the centre of the face, black patches to either side, and smaller white spots under the ears and eyes. The tail, while long, is not quite so much, in proportion to the body, as in the Paradoxurus species.

It mostly inhabits warm evergreen forests, but also utilises deciduous forests in some parts of its range, especially in the Himalayan foothills. While it is found at a range of altitudes, in Indonesia it is more common on mountain slopes than in the lowlands; this may, of course, be a reflection of where humans are least likely to be rather than a genuine preference.

Masked palm civet

It is thought to be more omnivorous than its close relatives, with perhaps a third of its diet consisting of rodents and other small animals. In the more northerly parts of its range, where the climate is more temperate, it switches from fruit in the summer and autumn to animal prey in the winter and spring when the former is less available.

Like the others, however, it is a solitary and nocturnal animal. In southern China, individuals have been reported to occupy home ranges of about 190 hectares (470 acres) each, and it's similar even in the much richer jungles of Borneo. It's likely that this degree of travel means that the masked palm civet, like the common one, is a seed disperser, if perhaps not quite of the same level of importance. They are less active in winter, probably conserving their energy in a time of food shortage.

Interestingly, while adults stay apart, juveniles learning to forage for themselves may not only rely on their mothers to help them out, but may seek out help from other adult females (possibly their aunts), grooming them in return for support in a way that is more typically seen in group-living animals. The manner of their mating has also been described as unusual, consisting of repeated sessions of prolonged thrusting intercourse with only a single ejaculation.

Masked palm civets are significant in a less positive way, as well, in that they were probably the source of the SARS epidemic in humans, which jumped species in a market in China in a similar way to the currently most favoured theory of the origin of the later Covid pandemic. This may have been rendered more likely by the fact that genetic evidence shows wild individuals are often introduced to civet farms, potentially bringing in viruses from outside.

This brings me almost to the end of the list of recognised palm civets. However, there are two other species in the same subfamily as those listed above, one of which is sufficiently odd that it isn't called a palm civet at all. Next time, I will conclude my survey of the viverrids by looking at these two...

[Photos by Mike Prince, David V Raju, Mahendra Prasad Peiris, and Rejoice Gassah, from Wikimedia Commons. Cladogram adapted from Nyakatura et al 2012.]

The Patagonian Homunculus

2026-03-15T15:53:00.003+00:00

In 1891, Argentinian naturalist Florentino Ameghino, the founding father of South American palaeontology, described a new species of fossil primate. Naming it Homunculus ("little man"), he recognised that it resembled a lemur almost as much as a monkey and must therefore be very primitive, but he was unable to categorise it further.

To be fair, all he had at the time was a section of the lower jaw. Over the next seven years, working with his brother Carlos, he uncovered a few further specimens. These included part of a skull and some limb bones, but the exact details of what they had discovered remained obscure, beyond the fact that it was a primate of some sort.

It didn't help that, after 1898, the next discovery of a fossil belonging to the genus didn't happen until the 1980s. And that was only a few isolated teeth. Since then, nothing until the current century.

Yet one thing we did know was that this fossil was important. It lived during the Early Miocene, between 18 and 16 million years ago, and all of its remains have been discovered in southern Argentina, roughly between the Santa Cruz and Gallegos rivers. This is significant for a couple of reasons.

For one, it's relatively early. There are older primate fossils on the continent, with the very oldest dating back a full 26 million years to the Oligocene. But Homunculus is the most common known species from this far back, with most of the rarer fossils appearing to belong to close relatives. Tremacebus, for example, the oldest known Miocene monkey from South America, was originally identified as a new species of Homunculus. So it's a good place to start if we want to know what the others were like.

But the second reason is that the Gallegos River is remarkably far south. Very nearly 52°S, to be precise. This is further south than any other primate species (except humans). For comparison, the southernmost point of Africa is just north of 35°S, and, due to continental drift, South America was about three degrees further south than it is now, so the contemporary figure may have been more like 55°S.

The climate in this part of southern Patagonia today is best described as cold semi-desert or steppeland. The vegetation is essentially barren grassland, without much in the way of shrubs, let alone trees. In the height of summer, temperatures reach around 15°C (59°F), and frost is common in winter. This is obviously not the sort of place we would expect monkeys to live. However, it turns out that the time when Homunculus was alive is slap in the middle of the warmest part of the Miocene, and back then, southern Patagonia had a comfortable, damp climate and was covered in deciduous forests.

So what exactly is Homunculus? It's undoubtedly a monkey, and related in some way to the living monkeys of South America. Today, these are divided into five families: marmosets, capuchins, titi monkeys, spider monkeys, and night monkeys. Over the decades, various attempts have been made to fit Homunculus into one of these families, most commonly the titi monkeys.

However, there has long been a counter-theory suggesting that it was actually a "stem platyrrhine". That is to say, it belongs to a lineage of early South American monkeys that left no living descendants and only resemble some of the modern forms through convergent evolution. If this is right, and the evidence seems to be swinging in its favour, Homunculus is related to, but outside, all of the living families, probably having split off from their common ancestor before they split from each other.

In 2015, a new fossil of Homunculus was discovered at a coastal site about 50 km (30 miles) north of the mouth of the Rio Gallegos. It was the most complete skeleton of the animal discovered, making it possible to analyse both the skull and the limbs of a single individual for the first time. (For comparison, only one other similarly complete skeleton of a Miocene-age South American fossil monkey has ever been discovered: Cebupithecia, a saki monkey from Colombia).

The results of that analysis were published last year.

We already knew, from the shape of the teeth, that Homunculus probably ate fruit as the major component of its diet. However, by looking at the shape of the skull, the new analysis was also able to come up with estimates for the size of the jaw muscles, which turned out to be unusually large. The researchers think that it's unlikely this was due to a need to increase bite force, and, using comparisons to fossil lemurs, suspect that it was the length of the muscle fibres that was increased. This, combined with the fact that Homunculus had a relatively long jaw for a monkey, suggests that it had a wide gape - presumably to more easily eat large fruits.

An unusual feature of the teeth that has often been remarked on is how worn down they are on all of the fossils so far discovered. That suggests a hard, abrasive diet, which doesn't really fit with fruit. It's thought likely here that the fruit Homunculus was eating was often covered in gritty dust, since the area it lived in was downwind of a series of active volcanoes in the southern Andes. Whether eruptions would have been common enough to create such an effect might, perhaps, be questionable, but the researchers point out that fruit is only common at certain times of the year outside the tropics and, even if the climate was warmer, there would still have been distinct seasons this far south. This means that the monkey would have been forced to rely on other foods at regular intervals; leaves are the most obvious candidate, and they tend to be more abrasive than soft fruit.

While even this fossil did not include the hands or feet, it did include enough of the limbs and, crucially, of the hips and shoulders, to give a good idea of how Homunculus moved. It was, as expected, an arboreal animal, and the way that its joints flexed suggests that it would have been particularly adept at climbing thick tree trunks as well as moving among the branches. It would also have been able to leap from tree to tree, but it may be that, this far south, on the fringes of the forest, trees were often far enough apart that it had to descend to walk across the ground to reach new ones.

Over millions of years, however, those trees would have become further and further apart and the fruit, perhaps, less nutritious. In many respects, including the size, Homunculus was similar to the many American monkeys that followed it, even if it wasn't a close relative of them. Replicating its pattern proved useful for later species, but it itself could not hold out as the lands of southern Patagonia became increasingly cold and dry.

By the time it became the treeless steppeland it is now, Homunculus would have been long gone.

[Image from Perry et al. 2025, available under the Creative Commons CC BY 4.0 license.]

Black Bears and the Uncertain Apex

2026-03-08T16:01:00.005+00:00

The concept of the food pyramid is a central one in ecology. The idea is that since consumption cannot ever be 100% efficient, every type of animal must necessarily be less common than whatever it is that it feed on, at least in terms of its total biomass. Plants are more common than herbivores are more common than small carnivores are more common than larger carnivores.

The actual picture is more complicated than this. Many "carnivores" are at least partly omnivorous, and they often eat large herbivores more regularly than they eat small carnivores. Plus, we also need to consider the detritivores and parasites. But the general pattern holds, and at the top of the pyramid, we find the apex predators.

It's possible to argue as to what exactly constitutes an apex predator. The general idea, however, is that they feed on other animals without being preyed upon themselves. At least among terrestrial mammals, an average body mass of more than about 15 kg (33 lbs) is generally about enough that predators need to manage their own population (through competition, territoriality, infanticide, etc.) rather than having to worry about something larger and scarier managing it for them.

Even so, the non-human animals typically described as "apex predators" in most parts of the world are quite a bit larger than this. Such animals dominate the "landscape of fear", influencing what other animals choose to live and feed where, and, even if they aren't eating them directly, will have at least some competitive effect on the smaller predators nearby. Not always in the way you might think, either; one study found that jackals are more common where lions are also common, but only if there are many other predators around as well - if it's just lions, they will avoid them.

So what exactly counts, and what doesn't?

The American black bear (Ursus americanus) is one of the more common large mammalian predators. Of course, one could question the definitions in that statement, not least because black bears are omnivorous, and meat isn't even an especially large part of their diet. Nonetheless, they do have significant effects on the other animals around them. In Virginia, they were found to be responsible for around half of all deaths of white-tailed deer fawns, and, in Yellowstone, they have been reported to eat elk calves, on average, once every eight days.

Understanding how black bears affect the other animals around them can be important for a couple of reasons. For one, they are relatively tolerant of human presence, often wandering into semi-rural residential areas and being more likely to come into conflict with humans. For another, their population has increased over the last thirty years or so, making them one of only two carnivoran mammals that are more common now than they were in the late 20th century. (The other is the giant panda, which was starting from a much lower point, and isn't a predator).

There are currently thought to be just under a million American black bears in the world, more or less evenly split between the US and Canada, with a much smaller number in Mexico. Their population probably dipped to its lowest point around 1900 or so, and has risen erratically and unevenly since then. Certainly, they are no longer found in as many places as they were in the 18th and 19th centuries - they once inhabited the whole of the US except the southwestern deserts, Hawaii, and the colder parts of Alaska, and reached from Hudson Bay almost to Mexico City. But, with increasing numbers where they are found, understanding their ecology isn't important because they are endangered, but because of what could be wide-reaching effects.

A review published last year confirmed that, despite being omnivorous, American black bears do meet many of the criteria used to define apex predators in other studies. For example, cougar/puma/mountain lions have been described as apex predators in part because they are the primary predators of white-tailed and mule deer in areas where they live together. In eastern Canada, black bears are responsible for a full 94% of the deaths of young reindeer... and, yes, maybe they do eat a lot of honey, nuts, and berries, but that's probably not much consolation if you're a mother reindeer.

Having said which, the review was unable to find much evidence that black bear numbers affected the local deer population. But it did identify cases in which they influenced other mammalian predators, which could be argued as evidence for their "apex" role. For instance, in the western US, they scare away coyotes, just as cougars do.

In fact, cougars, a fairly clear example of an apex predator, are themselves affected if black bears are in the neighbourhood. This is largely through kleptoparasitism, a fancy word for stealing somebody else's dinner. That is, the cougar kills something, and then the bear turns up, scares it off, and eats the carcass. No hunting (or, in the strict sense, predation) required, but the end effect is the same. And this does affect the cougars; when black bears are around, they are able to spend, on average, less time at each fresh kill, get less to eat and, in consequence, have to hunt again sooner than they would otherwise. So the cougars actually end up killing more deer than they would if there were no bears nearby.

Wolves seem to be less affected, possibly because, being pack hunters, they are more efficient at taking down deer, and perhaps defending carcasses when the need arises. While the presence of brown bears can, indeed, reduce the kill rate of wolf packs, in the case of the smaller and more peaceable black bears, it's probably they who have to change their habits when wolves are around.

Of the four main areas where most studies of American black bears have taken place, wolves are present in two - the Great Lakes and Yellowstone Park. Here, perhaps, they have less direct effect on the local ecosystem, not because nothing is eating them, but because other predators (including brown bears, in the case of Yellowstone) are more influential. But, in the other two main study sites - Newfoundland and the Great Smoky Mountains National Park on the Tennessee/North Carolina border - there is nothing else of quite the same size to compete with.

This underlines the fuzzy nature of the concept of "apex predator". It's reasonable to argue that black bears, being omnivorous and eating more plant matter than meat, don't really count. Because they are relatively harmless to humans who leave them alone, unless they have cubs to defend, they rarely fit our usual concept of the term. And, in the western and northern US, where there are also wolves, and in Alaska and limited parts of the northwest, where they share the landscape with grizzlies, that could well be fair enough.

But in the east, where there are no longer wolves and have never been brown bears or cougars, black bears may be as close to an apex predator as you're going to get. Certainly, it seems that they have a direct, top-down effect on the other animals around them, ungulates and smaller carnivores alike.

[Photo by "Cephas" from Wikimedia Commons.]

Viverrids: Half-Weasel Palm Civets

2026-03-01T16:25:00.003+00:00

Banded palm civet

The word "civet", as currently used in English, is a rather broad term, referring to a wide range of vaguely similar-looking animals. Not all of these are even members of the "civet family" as we currently understand it, and even those that are don't form a natural biological group within that family.

The word was originally Arabic (pronounced something like "zabad") and would have referred to the animals that medieval Arabs were familiar with, which, given how far they traded, would have included both African and South Asian species. These are still regarded as "true civets" today, but the word now also appears in the name of the "palm civets", long thought to be merely a variant of the true sort.

There are at least ten different species commonly called palm civets, one of which of which we now know belongs outside the civet family. The other nine belong to two closely related subfamilies, which were split apart as early as 1864. One of these is that of the hemigalines, a word that literally translates as "half-weasels".

The subfamily is named for the banded palm civet (Hemigalus derbyanus). When it was first described in 1837, it was placed with the other palm civets, but it was given its own genus by French zoologist Claude Jourdan just a few months later. It is he, therefore, who can take credit for the "half-weasel" part of the name.

First identified in Borneo, this lives across that island, and on Sumatra and the Mentawai Islands just to the west. It is also found on the Malay Peninsula, from continental Malaysia to the peninsular parts of southern Thailand. Its presence across the border in southern Myanmar may be sporadic; after over a century without any being definitively spotted, a camera trap snapped a picture of one in 2022.

The banded palm civet is much smaller than any of the true civets, due in part to the light, slender build that likely inspired Jourdan's name for it. They reach around 50 cm (20 inches) in length, not counting the tail, but typically only weigh 2 kg (4½ lbs), about the same as a chihuahua. The fur is grey to brownish over most of the body, and lacks markings on the animal's underside. The name, however, comes from the presence of five to eight distinct bands of dark brown fur across the back, which appear triangular from the side, with the points facing downwards.

They live in dense jungles, and, while they can be found as high as 1,600 metres in the mountains of Borneo, they prefer lowland forests over highlands. Similarly, while they can be found in disturbed forests, and even oil palm plantations, they are more common in primeval forests relatively untouched by man. They also tend to avoid rivers, although that's possibly because humans don't.

Analysis of their stomach contents reveals that they eat little but insects and earthworms. The only vegetable matter they have been reported to consume consists of a few dead leaves, such as might be found lying about on the forest floor; they are probably just eaten along with the insects hiding among them. This suggests, and camera trap images tend to support, the idea that they primarily hunt on the ground among fallen leaf litter. However, they have retractable claws and are quite adept at climbing trees, so this may not be their only option.

Hose's palm civet

Like other viverrids, they are nocturnal, usually sleeping through the day in the cavities of fallen logs, although, again, they have also been discovered snoozing up trees, on branches up to 8 m (26 feet) above the ground. They are solitary, marking their territories with rubbings from their perineal scent glands. These are less well-developed than those of true civets, consisting of a pair of deep folds covered by silky hair. Nonetheless, banded palm civets are capable of spraying their contents like a skunk when alarmed, although they remain on all fours when doing so, simply lifting their tail.

They give birth to one or two young per litter. Initially blind and helpless, they can climb trees by four weeks and hunt for their own food at six.

Hose's palm civet (Diplogale hosei) is named for the collector of the first specimen, colonial administrator and amateur zoologist Charles Hose. It is physically similar to the banded palm civet, but is even more lightly built, with a typical weight of just 1.3 kg (3 lbs). Notably, it lacks any trace of the banded pattern, instead having a uniform dark brown colour over most of the body, with white underparts and white markings on the face.

It lives only in northern Borneo, with the great majority of sightings coming from the states of Sarawak and Sabah in Malaysia. Based on its habitat requirements, it is assumed to also live to the south, across many of the highlands of Indonesian Borneo, but sightings are there are very rare - if not quite non-existent. The only living specimen to be examined by anything other than a camera trap photo was found in a national park in Brunei, although they may also live in other parts of that country, too.

As a result, it's a very mysterious animal, presumably nocturnal and perhaps with a greater fear of man than its banded cousin. Almost all of the sightings that have been made were in dense, primeval jungle far from human habitation. Aside from avoiding humans, this may also be because it prefers highland areas, not venturing below 325 metres (1,000 feet) and typically being found at least half as high again. The highest elevation at which they have been found 1,700 metres (5,600 feet) but that's likely because nobody has looked any higher in such a remote part of the world.

It's thought that, where both species are found, the banded palm civet occupies the lowlands and Hose's species the hills and mountains, thus ensuring the two do not compete for limited resources. The little we know of the habits of the latter suggest that it has a similar diet and that it, too, tends to forage on the ground; that one living specimen did not try to climb branches while she was held in captivity.

The closest relative of the banded species, however, is Owston's palm civet (Chrotogale owstoni). Named for its collector, a yachtsman with an interest in wildlife, it was first described as recently as 1912, from a specimen found at the mouth of the Songhai River in Vietnam. It is largely restricted to that country, where it is found in the northern and central regions, but individuals have been sighted in Laos east of the Mekong, and the Vietnamese border regions of southern China. It seems reasonable to suppose that it might also exist in parts of eastern Cambodia, but there are no definitive sightings, possibly due to a lack of camera traps in the area.

The fur is greyish to light brown, with a hint of orange in the adults. It is marked by four or five darker bands, similar in shape to those of the banded palm civet, making the two species look very similar. Owston's species, however, has additional black spots on the neck, flanks, and forelimbs, and is typically larger, with an average body length of around 60 cm (24 inches) rather than 50.

It lives in damp jungles with only limited dry seasons which, in this part of the world, prevents it from going too far west of the Annamite Mountains forming the Vietnam/Laos border, doubtless explaining its restricted range. Within the mountains, it can reach relatively high elevations, having been sighted at 2,600 metres (8,500 feet), but it also lives on the eastern coastal plains. The limited information on its diet suggests that it largely eats earthworms, and so, like the other hemigalines, must spend most or all of its time on the ground.

Formerly common, the species is thought to have undergone a dramatic population decline since the mid 1990s due to an increase in the demand for bush meat as Vietnam became more affluent. While there is some demand for the animal specifically for use in traditional medicine, most of the hunting is thought to be indiscriminate, a particular problem in the region. While there may be refuges in particularly rugged rocky areas and, at least as of 2004, the species retained a strong level of genetic diversity, the overall decline, combined with its restricted range, means that it has been officially listed as an endangered species since 2016.

Sulawesi palm civet

Moving east, we come to the Sulawesi palm civet (Macrogalidia musschenbroekii), known to occur in at least four different parts of that island. Having said which, it was only discovered on the southeastern peninsula as recently as 2003, so it's entirely possible that it lives elsewhere, given the inaccessibility of some of the options. For a long time, it was considered to belong to the other subfamily of palm civets until genetic studies in the 2010s confirmed it to be a hemigaline.

In fairness, it does have several differences from its banded relatives. These are not, however, particularly apparent on first sight. It's the largest of the hemigalines, and is sometimes known as the "giant palm civet" despite being no larger than a regular civet at around 90 cm (3 feet) plus tail, and weighing about 5 kg (11 lbs): it's actually the largest native predator on the island The coat is chestnut-brown, fading towards the underparts and marked with vague darker patches along the back.

A closer look, however, reveals that the more detailed anatomy is different from that of the banded civets. Its ankles can rotate to point its feet backwards, a common adaptation in animals that want to climb down trees headfirst so they can see where they are going. Similarly, the retractable claws have sheaths, as they do in cats, but not in other hemigaline palm civets. Only the female has scent glands, and these are very simple in form, just a small pit; in males, this is reduced to an apparently non-functional dimple.

As the form of the ankles might indicate, it's an effective climber, and, unlike its relatives, it spends most of its time in the trees. Here, it eats an almost equal mixture of arboreal rodents and palm fruit, a far cry from the pure insectivory of the other hemigalines. Other than a requirement for trees, they are relatively adaptable in their habitat choices, being as likely to be found in secondary forest as the primeval sort, and even being found in farmland.

Other than this, we know relatively little. They are nocturnal and solitary and estimates from that newly discovered southeastern population suggested that each animal had a home range of around 150 hectares (370 acres).

Otter civet

Their closest relative turns out to be another unusual hemigaline, one that's sufficiently odd that it isn't normally called a "palm civet" at all. This is the otter civet (Cynogale bennetti), which lives in much the same area as the banded palm civet, but is even more restricted to the lowlands. It is a relatively bland light brownish colour without any significant markings, but it is more its body shape that makes it unusual.

For one, the tail is much shorter than in other hemigaline species, where they are often the same length as the entire body. The ears are shorter and rounder, and the snout is also rounder, unlike that of most other viverrids. The nostrils are located on the top of the nose, rather than facing forward, and they can be closed by muscular action. The whiskers are unusually large and long. The feet are webbed and the scent gland is unusually small and simple, especially in the male.

Aside from the lack of a long, muscular tail, all of these features are reminiscent of those of otters and it's easy to see why it was given the name. So much so, in fact, that when the only known specimen of a supposed second species, Lowe's otter civet (C. lowei) from Vietnam, was genetically analysed, it turned out to be an actual otter. Previously regarded as a high priority for conservation efforts, it was removed from the list of endangered species in 2008 on the not unreasonable grounds that it never actually existed.

We can reasonably assume from the anatomical specialisations that the otter civet is semi-aquatic, but there are so few observations of it that we know little of the details. Certainly, it is often seen in peat swamps and similar wetlands, but it is also known from regular jungles and bamboo forests so it may be more adaptable than we'd assume. One camera trap image caught one climbing a tree, which isn't something you would expect a regular otter to do, although how common this is is difficult to say.

They are thought to eat small aquatic invertebrates but, again, much hard evidence is lacking. They do, however, have unusually sensitive whiskers, with which they are thought to probe along river bottoms in search out prey. Although there are no reliable estimates of their population, the presumption that they rely on swamps, which are being drained and/or polluted in the area, and the knowledge that they avoid areas frequented by loggers, means that it's highly likely to be declining. As a result, they are currently (if somewhat tentatively, given the lack of information) listed as another endangered species.

However there are, as noted, above, two subfamilies of palm civets. They are related, but are thought to have diverged at least 20 million years ago, so they are quite clearly distinct. Next time, I will be looking at some of the species in that other group.

[Photos by JJ Harrison and "Stavenn" from Wikimedia Commons, pictures by Joseph Smit and A.B. Meyer, in the public domain. Cladogram adapted from Veron et al. 2017.]

Beavers in the Wetlands

2026-02-22T14:13:00.005+00:00

The North American beaver (Castor canadensis) is one of two living species of beaver, along with its Eurasian counterpart. They are, of course, rodents - they are related to gophers - and relatively large ones at that. They are found across all but the most treeless parts of the US and Canada, as well as the border regions of northern Mexico; they have even been introduced to Finland, Belgium, and Argentina. They are a relatively common species across much of this range, something that has been helped in recent decades by restrictions on hunting.

Nonetheless, while not endangered themselves, they can be key to maintaining ecosystems, not least because they are one of the few nonhuman species that substantially modifies the land around them. Their ability to alter wetland habitats by dam-building has been identified as a key factor in maintaining other species at greater risk, such as amphibians in the Rocky Mountains. On top of which, their habit of cutting down trees affects the composition of the forests in which they live.

For these reasons, it can be useful to know what factors affect beaver population density and how the relevant habitat will be changed as a result. Which turns out to be a more complex question than it might appear at first glance.

Part of the reason that so many different species of animal (and plant) exist is that each adapts to a particular environment, or exploits the same environment in a slightly different way. What's good for one animal isn't necessarily good for another, even if they are otherwise similar. The species evolves to be good at exploiting whatever its particular favoured resource may be, and if some become particularly good at exploiting some other resource, then, even absent any physical separation, they adapt to that and may become a different species.

The particular resources an animal species may require can vary considerably. Food supply is obviously a key one, but there may be physical requirements as well, such as suitable places to sleep or hide from predators. But, whatever they are, the animal wants to seek them out, and we'd therefore expect that the more of them are in a given area, the more individuals of that particular species will live in it.

This has been formalised scientifically as the "ideal free distribution" model, which states that, assuming no physical barriers are preventing free movement, population density increases as habitat quality increases. This is, however, a bit like those economic theories that assume markets always act in a perfectly rational manner, largely because that "no barriers" part is quite significant, and we can't always rely on a direct correlation between resource availability and population density.

Furthermore, if the animal lives in communal groups, as beavers do, an increase in population density could happen for either of two reasons (or both). A better habitat could support more groups of the animal, each group requiring a smaller area to support itself. Or the groups could become larger, with more individual animals occupying the same area. A key limitation here, if the animal is territorial, is how much effort it has to put into defending its borders against rivals and, therefore, how much it helps to have those borders be shorter.

This is one of the questions that remains open for North American beavers, with conflicting results from studies looking into it. In general, population density of beavers seems to be related to the availability of aquatic vegetation such as cattails and water lilies, since this forms over half of their diet, but likely also to the types of trees that are available, since they rely on those for food, too, as well as constructing dams from them.

Poplars, aspen, and willows seem to be the most popular trees, but they often use ash trees where poplars are less common. The latest study to look at how habitat affects beaver population density was conducted in an area where this is the case - Plaisance National Park in Quebec. This is a relatively small park at 28 km² (11 square miles), occupying the north bank and the islands within a stretch of river between Ottawa and Montreal.

Other than the river itself, it consists of woodland with numerous patches of swampy ground. The trees are about a quarter ash and a quarter maple, with a random mix of oak, elm, linden, and other species making up the remainder. While poplars are present, they are not numerous, so the beavers rely on the ash, being spread broadly across the flat countryside, where they have prospered since the Park was established in 2002. (The beavers, in contrast, largely ignore the maple trees).

Beavers were fitted with GPS collars so that they could be monitored as they moved across the Park. This should have given a reasonable estimate of the area each family occupied, but the researchers confirmed this by the simple practice of walking across the area, examining the scent-marked mounds that beavers leave to mark out the edges of their territories.

Determining the size of each family group is, however, a little more complicated, since only a small proportion of the beavers will get the GPS collars. Here, the researchers used drones equipped with cameras to watch the beaver lodges and see how many animals came in and out around dawn and dusk. This, it should be noted, takes quite a bit of time, since you need numerous trial flights before the study proper to make sure that the beavers realise the drones are harmless and don't try to hide from them.

The study showed that most beaver families occupied an area of between 10 and 20 hectares (25 to 50 acres). There was considerable variation, with some families occupying areas smaller or larger than typical, but statistical analysis was unable to find any clear pattern relating to the number of ash trees or the availability of tasty water plants. However, the better the quality of the habitat, the more beavers there were in each family group, with the smallest just having three, and the largest, seventeen.

This tells us that, at least in this specific part of the world, beavers take advantage of high-quality land by raising larger families but occupying the same general area that they would have anyway. They could defend smaller territories, sharing the landscape around them with more neighbouring families, and raising the overall population that way, but they don't.

This could be because there are only so many places one can build a good dam, so that the ability to occupy the landscape is limited by the locations you have to start out from. A similar study on Eurasian beavers, for instance, seemed to show a "first-come, first-served" effect, where the first beavers to arrive in a given area get the best bit of terrain and spread out around it, while late comers end up with smaller territories dotted around it.

In badgers, on the other hand, we don't see the same effect. Here, the larger the badger group, the larger the area it will occupy, even though they wouldn't have to if the habitat quality (availability of earthworms, in their case) really were better. This may be due to differing behaviour patterns, with badgers being less cooperative than beavers; it's in the interest of male badgers to have as big a territory as they can get away with, so that other males who might sneak in to mate with the local females are kept as far away as possible.

Coyotes show a different pattern again: the better the habitat (e.g. the more deer there are to eat in it) the smaller the territory of any given group. This, again, is probably due to their social structure. Coyotes have to defend their territory from rivals, so, if they can keep it small and still find enough to eat in it, they're going to be better off.

Beavers, in comparison, are more tolerant, a phenomenon known as the "dear enemy" effect. What this means is that they tend to leave their known neighbours to their own devices and only get defensive when a stranger turns up. It could also be the case that beaver society is structured such that the dominant male can let his younger relatives help look after infants, he can devote more time to protecting his territory - so a bigger family is more of a help, freeing up his time.

Since about 2002, emerald ash borers - invasive beetles from northeastern Asia - have been detected across most of the eastern US and neighbouring parts of Canada, where they cause significant damage to the native ash trees. Since, in the case of the area in the study, these trees were key to the beaver's population density, this could alter that, leading to smaller families occupying the same areas. Having said which, that obviously won't apply in beaver colonies where ash trees aren't common in the first place, and it wouldn't if they just get replaced by poplars, which the beavers prefer.

Especially in biology, things can often be more complex than they first appear.

[Photo by "finchlake2000", from Wikimedia Commons.]

Eocene (Pt 2): In the Jungles of Europe

2026-02-15T16:33:00.002+00:00

Heterohyus

During the Eocene epoch, Europe was very different from the way it is today. For one thing, it was still separated from Asia, with the Turgai Strait running between the two, roughly from what would now be the Black Sea to the Arctic Ocean. But, were you to look at a map without any modern day context, you probably wouldn't call it a continent, because it wasn't a single landmass, but a chain of large islands.

The biggest of these lay to the north, encompassing Scandinavia and the lands to the east. Just south of that, the second-largest was the one that would later become Britain, France, Germany, and some of their smaller neighbours. A smaller, but still sizable, Iberian island lay to the southwest, and a collection of low-lying ones occupied the south and east, with the more mountainous parts of that region having yet to form.

But, even if you didn't know the changes in the geography, if you could simply travel back in time to the Early Eocene and look around you, it wouldn't feel much like Europe. The continent was closer to the equator than it was now - northern Germany was about where Milan is today - but, even ignoring that, the world as a whole was much hotter. This is part of the reason for the islands, since there were no ice caps back then, but it means that our hypothetical time traveller would be, in almost any part of the landmass, standing in a jungle.

The temperatures went up and down as the lengthy epoch wore on, so the vegetation wasn't the same throughout it but, for the most part, there would be a lot of dense woodland, plenty of rich plantlife, and a lot of rain. Fossilised plants from the Middle Eocene of Germany, for example, resemble those now common in Southeast Asia. The wildlife that lived in these forests, however, would often have been unfamiliar.

Even early on, however, there were a few exceptions. For instance, the oldest known rodent fossils in Europe date to the very beginning of the Eocene. They belonged to a group called the ischryiomyids, and they appear to have entered the continent from North America, using a temporary land bridge running from Greenland to Scotland. Examples include the mouse-sized Microparamys and the marmot-sized Pseudoparamys, with the former proposed as a possible ancestor of the first dormice. Masillamys, from Germany and France, was quite common during the Middle Eocene, and has some limb adaptations that suggest it may have been good at digging, in much the way that modern voles are.

Ailuravus is also usually classified as an ischriomyid. It was native to Germany, although its own ancestors are also thought to be North American, and is known from some exceptionally well-preserved fossils, including impressions of the fur. From these, we know that it was about 40 cm (16 inches) in length, not counting the long bushy tail. It had sharp claws and teeth suitable only for soft foods, which, taken together with some fossilised stomach contents, imply that it climbed trees and fed on their leaves.

A different group, the theridomorphs, became common in the Late Eocene, diversifying into numerous species. Roughly squirrel-like in form, and presumably also arboreal, most, such as Treposciurus from Spain, were still only able eat soft food such as fruit. A few later forms, however, such as Elfomys from France and Switzerland, had begun to develop more resistant, grinding teeth that would have allowed them to eat tougher and more fibrous food.

In addition to plenty of fruit and leaves, jungles are also rich in insects, so it should be no surprise that many of the other small mammals living among the branches and in the undergrowth were insectivores of one kind or another. These included the amphilemurids, currently thought to be an early side-branch in the evolutionary line that would later lead to shrews and hedgehogs.

Macrocranion is perhaps the best-known example, also being known from North America, probably having emigrated in the opposite direction to the rodents. It was about the size of a small squirrel, but with a more rat-like shape, with small eyes and large ears. The mobile snout has invited comparisons to elephant shrews, although this is almost certainly a case of parallel evolution. It most likely lived on the forest floor, with insects forming the bulk of its diet, although some larger species may have had a more varied predatory diet.

Its close relative Pholidocercus was, however, perhaps rather more peculiar. It was similar in size to a hedgehog, and had similarly spiny fur, although otherwise, the shape was more rat-like. Significantly, it had a horny plate on its head like an armadillo, which it may have used to help it burrow into the ground. The tail was also encased in small bony scales, a feature seen in some modern rodents from Africa, but not in anything else from its branch of the mammalian family tree.

Other small mammals of the day are more difficult to place. Apatemys and its close relatives Heterohyus and Carcinella belong to a group that survived the arrival of the rodents and other competitors at the dawn of the Eocene, having lived in Europe since the previous epoch, and that would go on to reach the next one before dying out. They were peculiar animals, about the size of a long-tailed rat, but living in trees and possessing unusually long front teeth and elongated second and third fingers.

It is assumed that they had a similar lifestyle to the modern aye-aye, gnawing into wood to extract grubs and pull them out with their fingers. Analysis of the skull of Carcinella suggested that it had a brain that was, about as large as we would expect for a modern mammal of its size... which is to say, larger than that of most of its contemporaries. They don't appear to closely related to anything alive today, with one study suggesting that they may have branched off from the common ancestor of rodents and primates... so quite a long way back, then.

The pantolestans are similarly difficult to place, perhaps being related to carnivoran mammals, but, again, having diverged very early on. Buxolestes is a typical example, a small semi-aquatic animal known from Germany, France, and England. It had strong claws that may have helped it dig burrows, a muscular otter-like tail, and limbs that appear adapted for swimming. The teeth were large and somewhat flattened, suggesting that it may have fed on clams and snails, crushing their shells, although the one fossil well enough preserved to still have food in its mouth had been feeding on fish.

The tree-climbing herbivorous Merialus may be related, although how closely is unclear. It had unusually large and sharp incisor teeth, possibly for defence against predators, but possibly for clipping fruit from trees (or both, of course).

Leptictidium may not even be a placental mammal at all, in the sense of being descended from the last common ancestor of living placentals, although it was certainly closely related - and some studies have placed it either with the primate/rodent branch or with the early African mammals. Either way, it was another odd-looking animal. It had a long, mobile, snout and a long tail, but what really would have stood out was that its hind legs were much longer than its forelegs.

Normally, we would assume this meant that it moved by jumping, like a kangaroo or certain rodents. However, its ankles weren't structured for absorbing impacts on landing, which has led to debates as to how often it could really have done this. Instead, the assumption is that it must have been bipedal, using its long legs to run rather than leap, although it probably did a bit of both. Living mammals just don't do this (arguably, humans come closest), making it more like certain kinds of dinosaur than anything else. Analysis of the bones around the inner ear suggests that it had a highly developed sense of balance, and would have been very agile. It was likely insectivorous, although, at around 21 cm (8 inches) in height, it could easily have eaten small lizards or other mammals as well.

During the Eocene, however, Europe still had native marsupials. Or, at least, metatherians, members of the branch that today consists entirely of marsupials, since the term can only strictly be applied to descendants of the last common ancestor of all the living forms, and the ancestors of the European animals probably split off just before that common ancestor lived. Examples include Peratherium and Amphiperatherium, which were both around the size and shape of an opossum and likely had a similar, ground-dwelling, omnivorous lifestyle.

But, if we're talking about small, insect-eating mammals, there was another group present in Europe at the time that remains significant today. Bats first appeared very early in the Eocene, with some of their oldest known fossils coming from Europe. This isn't necessarily where they originated, however, since there are equally old fossils way across in Wyoming, and ones not much younger in Africa, Asia, South America and Australia, perhaps reflecting the benefits of flight when it comes to rapidly spreading across the world.

Even the earliest known bats would be instantly identifiable as such today, their form having changed relatively little since and leaving the question of what exactly they evolved from an ongoing mystery. Nonetheless, very early examples such as Archaeonycteris do have some differences from the modern sort, most notably in that they still had claws on their index fingers, rather than keeping them only on their thumbs. It lived across Europe, with some of the oldest fossils being from Portugal, and also reached as far east as India. The last known example comes from Dorset in England, living at the end of the Middle Eocene.

It had a wingspan of around 35 cm (14 inches) and so was fairly large for a bat, as well as having a long tail. The shape of the wings suggests it flew in open spaces, rather than navigating between densely packed tree trunks, a habit that resembles that of most modern European species. It did not, however, belong to any living family, and the best bet is that it was a member of an early branch that has since died out, and possibly one that predates the last common ancestor of the living animals.

Palaeochiropteryx may have been more closely related to modern bats. It was smaller, with a wingspan of perhaps 25 cm (10 inches), but, more significantly, had a wing shape that suggests it was capable of slow, agile flight. This would allow it fly close to the ground, dodging trees and bushes in the dense undergrowth of the semitropical jungles in which it lived. Being preserved in the same high-quality deposits, it is another animal for which we have fossilised stomach contents so we know that, like most modern bats, it ate insects - one had eaten a diet of moths, while another had consumed caddisflies.

Eating moths would suggest that the animals were already nocturnal, while the shape of their inner ear indicates that they could also echolocate, albeit perhaps not quite as effectively as modern species. Vielasia, from southern France, shows another key adaptation familiar from living forms: it is the oldest known bat known to have lived in caves.

All of these animals are relatively small, and while a time traveller would surely note some of them running about on the tree branches, their eye would likely be more quickly drawn to the larger animals of the day. Next time, I will take a look at some of herbivores that lived in the jungles of Eocene Europe...

[Photo by Chris Woodrich, from Wikimedia Commons.]

Viverrids: Genets of Central and Southern Africa

2026-02-08T17:45:00.003+00:00

Rusty-spotted genet

The genets are one of the most speciose of the carnivoran genera, with at least fourteen species living across Africa and, in the case of the common genet, just beyond it. However, the majority of the species have been little studied. As small, nocturnal carnivores often dwelling in hard-to-reach places, it is relatively hard to do so, and they lack the cachet that applies to larger, more glamorous animals such as cheetahs, hyenas, or even wild cats.

It probably doesn't help that there is not a great visual or (so far as we know) behavioural difference between them. In many cases, the easiest way to tell them apart is simply to note where they were found. There are subtle differences between them, to be sure, but you may have to look quite closely, and the variations are often matters of degree. Nonetheless, let's see what I can say as I take a look at the seven confirmed species that I didn't cover in the last one.

Despite this, one species besides the common genet is very widely distributed, living across pretty much the whole of central and southern Africa. This is the rusty-spotted genet (Genetta fieldiana), which is also known by variations on the theme of "large-spotted genet" and, more rarely, as the "leopard genet". There is also confusion over the correct scientific name. In 1855, French zoologist Jacques Pucheran had named it G. rubiginosa, and, that was the one used for many decades.

Then, in 1981, a problem was discovered with this name. This is because when a species is named, you have to point to a particular specimen (ideally held in a museum or the like) and say "one of those - that's what I'm talking about". This is helpful if, later on, somebody splits the species into two; the one your specimen belongs to is the "real" one, and it's the other that gets the new name. This is exactly what happened in this case, so when it was proposed to split the species into the larger number that we have today, scientists went back to look at Pucheran's specimen to check which of their "new" species it belonged to.

At which point, they discovered that it was, in fact, a Hausa genet, a species that was already known and clearly identified as something separate. Well, you can't define a species by pointing to something entirely different, so they had to find the first time somebody had given a name to an actual member of the species. Which turned out to be G. maculata, coined by Charles Gray in 1830, although he had a different definition, which was why it hadn't stuck previously. But, as the best option available, we used that for the next quarter of a century, and it's still common in reference sources today.

Except that it doesn't work, either, because Gray had accidentally reused a name already given to an entirely different animal (a marsupial, of all things) by somebody else in 1792. After this was pointed out, and following a full six years of arguing, the name was officially scrapped, and the current one was applied in its place.

Whatever we want to call it, the rusty-spotted genet lives across a wide swathe of sub-Saharan Africa, reaching Eritrea in the northeast, Ghana and Burkina Faso in the northwest, and Nambia and eastern South Africa in the south. It primarily lives in forests and savannah, as one might expect given that range, and seemingly prefers open woodland with plenty of available water. They can be found at high altitude, for example on the slopes of Mount Kilimanjaro. In fact, it's quite happy in oil palm and pineapple plantations, and even in the suburbs, making it better known than its warier kin.

It's a medium-sized, slender, genet with a rather variable coat colour. While the name would suggest that its spots are rusty colour, and that's often true, they can also be much darker. They are typically larger than in many other species, varying in shape from oblong to rounded, and even here, there is variation. Some rare individuals are pure black, in a similar manner to "black panthers" just being all-black forms of leopards and jaguars.

Rodents are, perhaps unsurprisingly, their favourite food, with some studies failing to find any scats of the animal that don't contain rodent remains, even if many of them do contain other animals such as beetles and grasshoppers. Most of the rodents are mice, although the details will naturally vary across the wide range of their occurrence. For example, a Nigerian study showed Tullberg's soft-furred mouse as the single most common prey of the genet, and they only live along the coast of West Africa, so clearly they aren't part of the diet in, say, Ethiopia or Mozambique.

Like other genets, they are solitary outside the breeding season, wandering across regions of no more than 10 km² and marking their territory with dung piles and the scent from their anal glands. During the day, they sleep in patches of dense trees or bushes, and they tend to avoid the most open terrain when travelling at night. Although they often hunt on the ground, they can also climb trees to hunt among the branches, sometimes taking birds or bats. They stalk their prey, as cats do, pouncing on it once they are close enough, and dispatching it with a single bite.

The timing of the mating season varies significantly across their range, although it tends to be in the winter in southern Africa. The female initially avoids the male, running away from him until he can demonstrate his persistance, after which they lick and sniff each other and rub their cheeks before mating. This has been described as lasting about five minutes and is often accompanied by "meow" sounds. A litter of up to five young are born 70 to 77 days later, already furred, but with wider blotches than the adults, and, like cats, they are initially blind.

The servaline genet (Genetta servalina) lives in central Africa, from the Sanaga River in Cameroon to just south of the mouth of the Congo on the Atlantic coast across to Uganda and central Kenya in the east. Isolated populations were discovered in various mountainous regions of Tanzania in 2006 and 2012, and on the island of Zanzibar in 1998, so it's possible that there may be others beyond the regular range that have yet to be discovered.

Crested genet

They prefer forested areas, but is otherwise quite adaptable, being found in dense jungles, woodland savannah, and high-altitude bamboo forests. The stripe along the back is broken into long dashes, rather than being continuous, and the dark spots get progressively smaller as they approach the animal's underside.

In 1940, a subspecies of servaline genet was described living to the north of the Sanaga River. It has since been genetically confirmed as a separate species, and is now known as the crested genet (Genetta cristata). As often happens, since the two species were separated, notable differences have come to light. For one, the crested genet prefers drier woodlands, typically deciduous, rather than tropical evergreen, and they are found only in lowlands. This is likely reflected in the details of their diet, which overlaps more with the rusty-spotted species than with its close relative.

There are also physical differences. As indicated by their name, crested genets have a crest of fur down the middle of the back, which servaline genets do not, and the spots are typically larger. They are now thought to live further to the west than they were when initially described, occupying both the northern coast of Cameroon and much of southern Nigeria. A combination of habitat loss and a presumed small population mean that it is one of just two genet species (along with Bourlon's genet from Liberia and Sierra Leone) to be listed as "vulnerable" or "threatened", the category short of being truly endangered.

Moving further south, we come to the Miombo genet (Genetta angolensis), which lives from Angola to Malawi and Mozambique. The spots on the upper back merge into streaks, and fully black individuals are quite common in comparison with other species. The name comes from the miombo woodland that it inhabits - a type of savannah dominated by Brachystegia, tropical trees in the legume family related to the tamarinds of India.

Cape genet

The cape genet (Genetta tigrina) lives along the southern coast of South Africa, from Durban to just north of Cape Town. It physically resembles the rusty-spotted genet, and is therefore sometimes called the "South African large-spotted genet". Unlike that animal, however, it has a more consistent coat, with a pale greyish background and large spots with brownish centres and a darker outline. The stripe along the midline of the back is continuous, as it is in that (and most other) species, but includes a short crest of vertical hair.

It lives in the fynbos, a belt of scrubby heathland vegetation similar to that of the Mediterranean or coastal California. It tends to live near water, and to hunt in areas where bushes provide cover as well, presumably, as sleeping areas. They are tolerant of human activity, often venturing into suburban areas, where they snooze on building roofs, out of reach of dogs. This tolerance may be in part because the locals in that part of the world don't hunt the genets, and sometimes leave pet food out to encourage them to visit. Their natural diet consists of a mixture of rodents and shrews, along with beetles and grasshoppers, and an unusually high proportion of grassy vegetable matter. Their favoured rodent prey are climbing mice and, to a lesser extent, vlei rats.

The giant genet (Genetta victoriae) is the largest of the species, although perhaps not by as much as its name might suggest. It measures up to 61 cm (2 feet) in length, rather than 50 cm (1'8"), and weighs 3 kg (5 lbs 10 oz.) rather than 2.5 kg (5 lbs 8 oz.). Otherwise, it has large, irregular spots and a short crest down the middle of the back; the central stripe is discontinuous, but the darker hairs are so long that this isn't always obvious.

It lives in some of the densest parts of the Congo jungle, in the northern DRC and some neighbouring parts of Uganda and Rwanda. One could reasonably assume that it eats larger prey than other genets, with perhaps more rats than mice, but studies on its diet simply don't exist.

The most distinctive of all genets, however, is the aquatic genet (Genetta piscivora), which lives in much the same area as the giant species. So much so, in fact, that it was long placed in a different genus, before genetic data confirmed that, despite appearances, it is just a regular genet.

Aquatic genet

The most obvious difference is that, unlike every other genet, it has no spots at all, just a rich reddish brown coat, albeit with distinctive black and white markings on the face. Furthermore, it has teeth that are smaller than usual and bare palms on its forefeet. Both of these features are thought to be adaptations to catching fish, since the locals say it eats little else, and a dead one whose stomach contents were examined turned out to have recently eaten a catfish.

This fits with the fact that they are observed almost exclusively along waterways - although, in fairness, it might be hard to spot them elsewhere in the depths of the jungle. One observed briefly in captivity wandered along the edges of a pool, stabbing its paw vertically into the water and then lunging to grab fish with a single bite. It would also eat frogs and some aquatic insects, although it isn't clear how much of its diet these normally comprise, and it completely ignored the grasshoppers that any other genet would be happy to feed on. (It's said to feed on crustaceans too, but has never been observed doing so, and doesn't appear to have the relevant adaptations - but, given how few observations there are, you never know).

While the exact number of species has varied down the years, the animals listed in this and the previous post were long considered to be the only kinds of genet. However, genetic studies confirmed that there was one other species of animal that, if not strictly speaking a genet, was closely related to them, and now belongs in the same subfamily. Previously called the "African linsang", this means it is not a close relative of the real (Asian) linsangs, and deserves a name more appropriate to its African origins. Furthermore, in 2005, following some earlier suggestions, the two subspecies were raised to full species status.

West African oyan

Thus, we now have the Central African oyan (Poiana richardsonii) and the West African oyan (Poiana leightoni). The two do look much like regular genets, but their spots are more irregularly placed than in most true genet species, and they typically (but not always) lack the black stripe down the middle of the back. They are smaller, and much more slender, at around 35 cm (14 inches) in length and 600g (21 oz.) in weight, with the West African species being slightly shorter and lighter than the other.

The Central African species lives in much the same area as the servaline genet. Along the Atlantic coast, it is found from Cameroon to the mouth of the Congo, from which it stretches eastward to the Great Rift Valley, along the eastern border of the DRC and into western Rwanda and Uganda. The much rarer West African species lives some distance away, in eastern Liberia and the neighbouring parts of Côte d'Ivoire. Both of these places are heavily forested, and the oyans are thought to spend far more time in the trees than true genets.

The few reports of their activity indicate that they are, like genets, nocturnal, and that they sleep in trees during the day, with one West African individual reported to have built a nest about 2 metres (6 feet) above the ground. They are thought to eat birds, insects, and tree-climbing rodents, but we really don't have any studies to go into more detail than that. Which, beyond the fact that a lactating female of the the Central African species was once spotted in October, giving at least some vague hint as to when they might give birth, is pretty much the extent of our knowledge.

The genets and oyans taken together form a subfamily within the viverrids, alongside that containing the true civets. There are, however, four subfamilies in total, with both of the others going by the collective name of "palm civets". Next time, I will be taking a look at one of those subfamilies.

[Pictures by Bernard Dupont, Donovan Rosevear, and Justin Ponder, from Wikimedia Commons, and Richard Deckert and St George Mivart, in the public domain. Cladogram adapted from Gaubert et al. 2006.]

Rabbits in the Ice Ages

2026-01-31T16:26:00.000+00:00

The rabbit (Oryctolagus cuniculus) is one of the more familiar mammal species to most people in the developed world, not least because it has been partially domesticated. Moreover, wild and feral rabbits are found across wide parts of the world, living on every continent except Antarctica. They are adaptable animals, able to survive in a wide range of different habitats, from hot semidesert to forests and bleak moorland. So much so, in fact, that, in most parts of the world, they are a pest.

The rabbit is widespread because we humans have spread it, following the initial domestication event in France, no later than 800 AD. This was originally for meat and fur, with pet breeds appearing only from the late 18th century. Rabbits are now found, for example, on Middleton Island, a chilly speck of land 130 km (80 miles) off the south coast of Alaska, and on the Kerguelen archipelago in the Indian Ocean, which is about as remote a place as it's possible to get.

These animals are "feral", in the sense that they live wild, but are descended from domesticated stock. But, obviously, they must have come from somewhere in the first place, and that place is Europe. (In contrast, the cottontail rabbits of North America proved behaviourally unsuited to domestication, although the ancient Mexicans gave it a try). More specifically, the rabbit is truly native only to Spain, Portugal, and southern France, with the Romans having introduced the pre-domesticated form across the rest of Europe, where most people would think of it as wild today.

In fact, wild rabbits are rare in their native range today, due partly to land development but more importantly, two waves of lethal disease - myxomatosis in the 1950s, and RHD in the '90s. As a result, you may be surprised to learn that the rabbit is, officially, an endangered species. It's just that we don't count all those feral forms in that classification.

What can we say about the earlier history of the rabbit? Rabbits have been regarded as a keystone species - one whose presence is central to maintaining their native ecosystem - so that uncovering their history can also tell us something about the world in which they lived. In their case, that's both by the way that they affect the local plantlife by feeding on it, and thus affecting other creatures that rely on the plants, but also by being a key food source for predators. A great many medium-sized predators feed on rabbits, and they are particularly important to Iberian lynx and Spanish imperial eagles, both of which are threatened species today.

Genetic evidence has confirmed that there are two living subspecies of European rabbit. The nominate form (O. c. cuniculus) is that to which the original scientific name was applied in 1758. That was, of course, by Linnaeus, and since he was Swedish, it's the subspecies that is found in Sweden. Originally native to southern France and neighbouring parts of Spain, this is the one that became domesticated and to which virtually all modern rabbits belong.

The other subspecies (O. c. algirus) is found only in wild populations in southern Spain, Portugal, and some parts of North Africa, although it has also been introduced to places like the Canary Islands. It was never domesticated, presumably because once the other one had been, there was no point in starting again from scratch. Based on their degree of genetic divergence, the two subspecies are thought to have last shared a common ancestor around two million years ago, which therefore puts a minimum age on the species as a whole.

Trying to unravel how rabbits changed and spread over that time period relies on skeletal and fossil remains, and that turns out not be as easy to do as one might hope. Trying to distinguish different forms relies on the size of the skeletons and, in particular, on the precise shape of the third premolar tooth in the lower jaw. However, both of these things can change in response to changes in the climate and associated foodstuffs or to local genetic effects with little wider meaning.

A recent study examined the remains of over 1,200 fossil rabbits from across their native range, comparing with the modern species. The remains date from across the Middle and Late Pleistocene - the last four Ice Ages and the three interglacials between them. When we look at modern rabbits, there is evidence that they partially follow Bergmann's Rule, which states that, for suitably close relatives, the colder the climate, the bigger the animal. I say "partially" because there are exceptions, but, in general, rabbits living at more northerly latitudes grow larger than those further south, perhaps because a larger body makes it easier for a mammal to retain body heat.

The study confirmed that this held true in prehistoric times, too. The oldest rabbits in the study, living 650,000 years ago were particularly large, and this is the height of the Günz/Cromer/Don glaciation, when the ice sheets reached unusually far south. 220,000 years ago, during the height of the last-but-one interglacial, they are noticeably smaller and there is another significant increase from 70,000 years ago as the Last Ice Age approaches and cold coniferous forests spread south.

At the start of the Middle Pleistocene, rabbits were restricted to Spain, where local conditions were relatively mild. The first evidence for them crossing the Pyrenees dates to around 560,000 years ago, and the fossil site in that case is only in the northern foothills, so it may have been another 80,000 years before they reached into the French lowlands and were, at least temporarily, prevented from heading further east by the Rhône. Surprisingly, they don't seem to have reached Portugal until even later, 425,000 years ago. This is about the same time that they reached Italy, although they subsequently went extinct there.

Rabbits then retreated and returned in various waves as the climate worsened, then improved again. The last such migration occurred as the Last Ice Age was thawing, seeing rabbits spread along the Mediterranean coast, and finally crossing the Rhône.

During the Ice Ages, rabbits, like many other animals, would have been isolated in "refugia", relatively small patches of suitable habitat surrounded by those less favourable. In this case, the study identified at least four during the Last Ice Age: two along the Mediterranean coast of Spain, one stretching from central Portugal to southern Spain, and the other in southern France.

The last one would have been temporarily destroyed when the ice reached its worst, as evidenced by the fact that we have rich fossil deposits from the time in southern France, and they don't include rabbits at all - it was more a time for mammoths and the animals that lived alongside them. The limitation here may well have been the expansion of the tundra, the permanently frozen subsoil preventing them from digging burrows.

If the divergence date between the two subspecies is correct, however, this can cover only the last third of the rabbit's existence. It's history before that, during the Early Pleistocene, remains a mystery, not least because the oldest confirmed fossils for the species, from Andalusia and Valencia, only date back 700,000 years. However, a lot may depend here on how we are defining the species. Presumably, over the course of two million years, it must have changed a fair bit, and we wouldn't necessarily recognise those earlier forms as belonging to the same species from their bones alone, even if, genetically speaking, they actually did.

Crucially, there is a candidate: Gibert's rabbit (Oryctolagus giberti). First identified in 2008 from remains in southeastern Spain, it is now known to have lived across both Spain and Portugal, and to have reached at least southeastern France from around 1.8 to 0.8 million years ago. This means that it died out just before the oldest fossils of the modern rabbit are found, so if it's a direct ancestor, then either it's not a distinct species or, perhaps more likely, the date for the split between the living subspecies is off by a fair margin - perhaps due to mixing of genes with ancient stock at some point.

Today, the European rabbit is the only member of its genus. But other species have existed in the past. The Valdarno rabbit (Oryctolagus valdarnensis) lived at around the same time as Gibert's species, between 2.1 and 1.2 million years ago. However, it lived in Italy, and so isn't a likely candidate for the ancestor of the modern form.

The only other known Ice Age species is Lacoste's rabbit (Oryctolagus lacosti), known from France around 2 million years ago, at the beginning of the Pleistocene. The oldest species that can be assigned to the genus, however, is the Layna rabbit (Oryctolagus laynensis) from the Late Pliocene of Spain 3.5 to 2.5 million years ago, apparently dying out when the Ice Ages started. It's thought that these two species evolved from an even older, as yet unknown, species that may have lived across much of Europe during the comparatively warm Early Pliocene.

This would push the origin of the rabbit genus back about as far back as the last common ancestor of the hares and jackrabbits. If this hypothetical species existed, it would have been the first true rabbit.

[Photo by N. P. Holmes, from Wikimedia Commons.]

Running Hyenas of Greece

2026-01-25T16:40:00.000+00:00

Chasmaporthetes

Mentioning a "hyena" today likely brings to mind the image of a muscular, aggressive, scavenging creature, probably with a spotted coat. This, the spotted hyena (Crocuta crocuta), is, however, only one of three species alive today, with the two lesser-known ones being smaller and less aggressive. In fact, the hyena family includes a fourth living species as well, although this isn't always referred to as a "hyena" because it's really only dangerous to termites and can't crack bone like "true" hyenas can.

Four living species isn't very many for a family of mammals but, like many other such small groups, there is a long fossil history that includes a great many extinct forms. These varied in form even more than the living species do. At one extreme are animals larger and stronger even than the living spotted hyena, while at the other (all living very early on) are small tree-climbing animals that looked more like civets.

Somewhere in between are the "running hyenas".

The best-known of these is Chasmaporthetes, which lived across Africa and much of the Northern Hemisphere from around 5 to 0.8 million years ago. We know from its remains that it had large fles-cutting teeth and limbs adapted for chasing down prey; it may not have had the speed of a cheetah, but it was likely a good deal stronger and may have hunted in a similar manner.

Two other genera are typically grouped with it, on the assumption that they had a similar lifestyle. However, we have far fewer remains for either of them, especially when it comes to the limb bones. Their teeth and jaws were at least similar to those of Chasmaporthetes, hence the assumed similar lifestyle, but it's hard to prove. Of the two, Lycaena is the earlier, but Hyaenictis is perhaps the least known.

Hyaenictis was first described by French palaeontologist Albert Gaudry in 1861, from a fossil unearthed at Pikermi, just outside Athens. It was almost a hundred years before remains attributable to the genus were discovered anywhere else, and most such sites have only been identified in the last 20 years.

We currently know of at least three species, which may have differed from one another as much as say, lions do from tigers or coyotes from wolves. H. graeca is the original named by Gaudry, and known only from Greece, where it lived around 7 million years ago, towards the end of the Miocene. The other two definitive species include a 9-million-year-old form from Spain, and a 5-million-year-old one from South Africa and Kenya. Fossils that are harder to place and that may or may not belong to the known species have been found at other sites, mostly in Kenya, although there is also a possible example from Ethiopia.

With so few limb and body bones, the question of Hyaenictis's running ability remains unresolved. However, there have been considerable advances in our knowledge and in statistical and measurement techniques over the decades since many of the existing fossils were last described. Last year, researchers published a detailed re-examination of most of the known fossils, along with six newly discovered ones from Greece, aiming to see if there was anything new this could tell us about the creature they once belonged to.

Because all we have are partial skulls and jawbones, a large part of the resulting paper consists of a detailed discussion of the precise shape of the teeth. Much is made, for example, of the absence of a metaconid on the m1, and there's a line about the imbrication of the dp3... the sort of thing that mammal palaeontologists spend a lot of time talking about, but that would require a large chunk of this post to explain, and that's unlikely to make an interesting read if I did.

The upshot of a lot of this, however, is that Hyaenictis shows an intermediate development between the tooth form of the older Lycaena and the later Chasmaporthetes. This doesn't prove that each directly evolved into the other in a chain since there may, of course, be other forms we don't know about, but it does show a general evolutionary trend within the group towards a more specialised animal. Presumably, this would also be reflected in aspects of their lifestyle, since Lycaena's own ancestors were probably more like jackals than the fast-running pursuit predators that survived into the Ice Ages.

However, one of the reasons that the study of tooth shapes is so key to mammalian palaeontology is that they can tell us a lot about what the animal in question ate. In the case of hyenas, a key adaptation is the presence of teeth towards the back of the jaw that are large enough and strong enough to crack bone. The relevant teeth are modified in Hyaenictis, being considerably more robust than those in, say, lions, but not quite as much as in the living hyenas. For example, they don't have quite as wide a base - making them better for piercing than crushing - and the microscopic structure of their enamel indicates that it wasn't as strong and resilient as we might expect.

The authors conclude that Hyaenictis could crush bones, cracking them open to get at the marrow as a living hyena would. But, equally, this can't have been as common and effective a habit as it is in the modern animals, and Hyaenictis therefore probably ate a higher proportion of soft meat, wasting less time on getting at hard-to-reach nutrition.

Another source of evidence comes from looking at the sites where the fossils were discovered. From other studies, for example, we know that, at the time Hyaenictis lived there, the area around Athens was a warm savannah landscape, dominated by grass but with patches of trees. The sites at which the Spanish species were found were more heavily wooded, consisting of subtropical forest. The pattern is repeated at the sites in Africa, with woodland of various kinds predominating wherever Hyaenictis was found.

That's interesting, because, if it really was a fast-chasing animal, we'd expect a more open environment. Cheetahs don't live in jungles, but out on the plains where whatever they are chasing has limited opportunities for escape, and the predator doesn't have to keep swerving to avoid trees or heavy undergrowth. The only limb bones we do have for Hyaenictis come from a juvenile in South Africa; they look to be more suited for speed than strength, but that may just be because it was young and slender, saying little about the adult form.

On the other hand, the authors point out that wolves are pursuit predators, and they often live in woodland. Perhaps Hyaenictis may not have been as pure a chasing animal as cheetahs, or as we have good reason to suppose the later Chasmaporthetes was, but that doesn't mean that it wasn't at least developing in that direction.

They also raise the point that Hyaenictis is relatively rare, with far fewer fossils (especially adults) from the sites where it is known than for the larger, more muscular, hyena Adcrocuta. This raises a possibility that has previously been considered for sabretooth cats: where more than one form of a generally similar predators live in the same area, they will inevitably compete. Living spotted hyenas will chase off leopards and cheetahs from their kills given half a chance, and lions will do the same to them, so Adcrocuta might have competed with Hyaenictis.

If that's right, the smaller running hyenas might have been forced into the woodlands because the larger ones drove them away from where they would prefer to be. It's hard to know whether this was true, and it depends in part on where Adcrocuta itself wanted to hunt, but it's a distinct possibility. At a time when many kinds of hyena lived in the same area, they couldn't all have been winners.

[Photo by Jonathon Chen, from Wikimedia Commons.]

Viverrids: Genets of Northern Africa (and beyond)

2026-01-18T17:30:00.001+00:00

Common genet

At the dawn of scientific taxonomy in 1758, the group of animals that would later become the civet family included four species that we still recognise today. Two of these - skunks and mongooses - were later moved elsewhere, and one of the others was, of course, the defining species of the family, the large Indian civet.

The other was the common genet (Genetta genetta). Frédéric Cuvier split the genets off from the civets proper in 1816, recognising just one other species - which molecular evidence has since shown isn't a viverrid at all, although this wasn't obvious even in the late 20th century. His basis for the distinction was that genets have a much smaller scent gland than civets, although he conceded that the secretions it produced were, in his words, "très manifeste".

We now recognise over a dozen species of genet, although exactly how many is a matter of debate. The usual number quoted today is either fourteen or fifteen, although a paper presented at a conference on biodiversity in 2005 proposed seventeen. Genets are native to Africa, being found across the whole of the continent other than the Sahara Desert. For the most part, they are similar in form and difficult to distinguish beyond which part of Africa they happen to have been found in, although genetic evidence shows that the ancestors of the living species began diverging over 8 million years ago, during the late Miocene.

Of all of these species, the common genet is the most widespread. It is found across the Sahel (the band between the Sahara and the equatorial rainforests), and most of eastern and southern Africa. It is also the only genet species to live north of the Sahara, being found in Morocco, northern Algeria, Tunisia, and a few coastal regions of Libya. Moreover, it's also found in Yemen and along the west coast of Saudi Arabia and, perhaps more significantly, in Spain, Portugal, and southern France.

The animal is clearly not native to Europe, but has been present in the area for so long that, in that original species description in 1758, it was described as coming from a location near Madrid. In fact, they have been on the continent for centuries, with the oldest known remains dated to the 8th century. This suggests that they may have been brought across by the Muslim conquerors of Andalusia, although there is a suggestion that at least some might date back centuries further to Carthaginian times.

In recent years, they are reported to have crossed the Rhône, presumably due to the construction of new bridges, and reached the Italian border. Cold winters are thought to be the main reason they haven't travelled any further north than Nantes, although this could potentially change in future decades. Either way, even if they aren't strictly native to the region, the bulk of research on the animals has been conducted on European populations.

Like most genets, the common species is about the size of a housecat, but, being much more slender, only weighs about half as much. They have a pointed snout, erect, rounded ears, and a long tail. The background colour of their fur varies from pale grey to a reddish-yellow with paler underparts. The body and limbs are marked with rows of black spots and the tail is striped. There are also distinct facial markings, including white spots above and below the eyes. In Europe, a few albino and all-black individuals have been noted, although it's unclear why this is apparently not also the case in Africa.

The hair is typically long, although it less so in the subspecies living in the warmest of climes. Notably, a black stripe of especially long hairs runs down the middle of the back. In fact, an alarmed genet acts very much like a cat, arching its back, erecting the hairs of the stripe into a crest, sticking out those on its tail into a "bottle-brush", and hissing loudly.

Given the vast range it inhabits, it's unsurprising that the common genet can adapt to a wide range of habitats. They avoid true deserts and dense jungles, but otherwise, so long as it's warm enough, they will live almost anywhere. Having said which, they prefer woodland, and where they live in drier places, they will at least focus on places with bushes and on ravines, both of which provide safe places to sleep.

Similarly, the common genet has a wide diet, which varies somewhat across its range. While the details depend on the exact area, between 70 and 80% of their diet consists of small mammals, such as mice, shrews, and even moles. In Europe and along the Mediterranean coast of Africa, this mostly means wood mice (Apodemus sylvaticus), which can form up to two-thirds of the mammalian component of their diet, although across most of Africa, they must be eating something else. The remaining 20 to 30% of their diet is roughly evenly divided between small birds and large insects (beetles, grasshoppers). Rarer foods include rabbits, bats, crayfish, and a few plants.

Like other genets, the common species is nocturnal and solitary, spending the day resting under cover. In Spain, each male has been reported to use an area of about 113 hectares (280 acres) over the course of a year, while females use 72 (180 acres); this will surely vary in other parts of the world, depending on the availability of food and water. Genets mark their territory with midden piles, often shared between neighbours. These are placed in prominent locations, often on top of rocks or in bushes. Occasionally, they are placed in trees, although genets are generally reluctant to use their climbing skills more than they have to.

In addition, they also scent mark using their perineal glands, spraying onto vertical objects while doing a handstand on their fore-paws, in the same manner as civets. Additional, smaller, scent glands allow them to deposit markings by rubbing their flanks along objects; their use of this method varies across the year and may be related to reproductive status. Breeding is seasonal, although when that season is varies across their range. The young two or three in each litter are, like those of cats, born blind but already furred.

Several subspecies of common genet have been named over the year and there remains disagreement as to how many are really valid. One of them, the "feline" or "South African" genet, is often regarded as a separate species (Genetta felina) and is the reason for the "fourteen or fifteen" given in the species count I gave above. It lives in South Africa and Namibia where, in the absence of wood mice, its preferred prey include the four-striped grass mouse and the African pygmy mouse.

One part of Africa where the common genet is not found is the tropical margin of West Africa, where the forests are too dense for it to be comfortable. Several different species of genet are currently recognised in this area, but they have been little-studied compared with their common relative.

One of the more common is the pardine genet (Genetta pardina). The word "pardine" means "leopard-like" and here refers to the animal's spots. These are pale and brownish, but offset from the light grey background fur by black rings, creating an effect similar to the ring-like spots of a leopard.

The pardine genet is also larger than the common sort, weighing around 3 kg (6½ lbs) instead of just 2 kg (4½ lbs). They live across much of West Africa, from Senegal to Ghana; it has been suggested that they are not found east of the Volta River, although there have been claims of genets with similar spots as far east as Benin. While they are primarily forest-dwellers, they are adaptable and at least somewhat tolerant of human disturbance, being found in plantations and even some suburban areas. One killed in Côte d'Ivoire had recently eaten a brush-furred rat (which is relatively small as rats go), but that's about the extent of our knowledge of its diet.

The Hausa genet (Genetta thierryi) is more widespread than its close pardine relative, although it is said to be less common in those areas where it lives. It is found from Senegal all the way across to Cameroon, and so clearly isn't bothered by the Volta - or indeed, the Niger River. It is distinguished by lacking the dark rings around the spots, giving it a much paler and more yellowish appearance, but is also only about half the weight of the pardine genet. Like its cousin, it lacks the crest of black erectile hair along the back possessed by the common genet. It prefers relatively open woodland and savannah.

Johnston's genet (Genetta johnstoni) was first named as a separate species in 1904, and has more consistently been considered as such than the more widespread Hausa and pardine forms. It is somewhere between the two in appearance and size, having spots with faint rings that sometimes merge into stripes towards the rump. It does, however, have the crest of raisable black hairs along the back that those two lack.

It is another forest dweller, being restricted to dense rainforest, and is especially common in wetlands such as wooded swamps. In line with this, individuals are reported to frequently sleep in trees through the day, unlike the more ground-dwelling common species. It has traditionally been regarded as native to a region stretching from southern Guinea to western Ghana, although a single report from a camera trap in 2011 showed one close to the Guinea/Senegal border about 260 km (160 miles) north of what had previously been thought of as their furthest limit.

Hausa genet

The remaining two species in the region are even less known. In the case of Bourlon's genet (Genetta bourloni) that's because it was only split off from the pardine genet in 2003. Living in Liberia and Sierra Leone, and to a limited extent across their borders into Guinea and Côte d'Ivoire, aside from some differences in the exact form of the skull, it is distinguished by being greyer and having darker spots, with more black on the tail and limbs. If there are any other differences from the pardine genet, we don't yet know what they are.

The king genet (Genetta poensis) has been recorded from four disparate locations along the Atlantic coast: Liberia, Ghana, Bioko Island in Equatorial Guinea, and around the mouth of the Congo River. The spots are elongated and almost rectangular, and, while it has a clear black streak down the back, this does not form a crest. Other than the fact that all of the locations where it has been spotted are tropical rainforest, that's literally all we know. Indeed, we can only confirm it exists from the ten specimens held in museums and nobody has recorded even seeing one since 1946. That could be because it's rare, perhaps even extinct, but it's also possible that it just looks so similar to the other species of the area that, in the absence of detailed surveys, nobody has remarked on it.

Over on the other side of the continent, we have the Ethopian genet (Genetta abyssinica). This is native to northern Ethiopia, Eritrea, and Djibouti, and likely just across their borders into neighbouring countries. The spots on the back are merged into five parallel dark stripes, and a thin white streak runs down the middle of the wider black strip along the middle of the back. It's also one of the smallest genets, typically weighing less than 2 kg (4½ lbs).

It inhabits highland regions, mostly forested, and is commonly associated with tree heath (a type of bush), goatweed, and Abyssinian rose. However, it has also been sighted in lowland areas, so it may have a broader tolerance for different habitats than this would suggest. There have been few detailed studies on it, but there are enough to identify that its favoured prey includes grass rats, which are native to high altitude grasslands in the area. Other studies indicate a solitary lifestyle, leaving the same sort of middens that common genets do, among other points of similarity.

That covers seven (or eight) of the fourteen (or fifteen) species of genet. The others live, on average further south, and I will turn to them next time.

[Photos by "Foto-Ardeidas", "Gecko-kus", and "Paradox usik", from Wikimedia Commons.]

Just Cold Enough

2026-01-11T15:52:00.003+00:00

Northern meadow jumping mouse

Hibernation is one of the most effective means that mammals can employ to withstand harsh winter temperatures. Many mammals, especially larger ones, use other means, such as growing a winter coat, caching food, or simply moving somewhere warmer, but, if the goal is to reduce energy requirements when you can't get out to feed, hibernation is the most effective means of doing so.

During hibernation, bodily metabolism slows right down, so that an animal may need as little as 1% of its usual calorie supply to stay healthy. This has negative consequences, so that the animal does need to wake at intervals to stave them off, and how often this happens varies from species to species. In this respect, true hibernation can be distinguished from shorter, often daily, bouts of torpor by the fact that each "sleep" can last for weeks or even months.

If you're going to hibernate, however, you need somewhere safe to do it. Bats use caves that predators are unlikely to to enter, clinging to the ceiling out of reach even if they do, but most other animals use a burrow or similar small shelter. The place that an animal uses to hibernate is called a hibernaculum and, since they are going to spend a long time there with little, if any, ability to move elsewhere, it's important for them to pick one that's just right.

One of the most important features of a hibernaculum is the temperature it is likely to maintain through the winter. The whole point is to stay warm, and to do so while using as little of your own body heat as possible. If a hibernaculum is too cold, the animal will be forced to burn its fat reserves to stay alive, which is the last thing you want to do when you can't get out to feed.

Thus, a good hibernaculum not only has to be safe from predators or other threats, but also have the right degree of insulation. There has been a good body of research on how snow cover can affect the insulation properties of hibernacula for animals such as Arctic ground squirrels. This will likely be significant if climate change alters the amount of snow available but it's likely not the only factor, since the depth below the soil is likely also relevant, especially where snow is less common.

The southern meadow jumping mouse (Zapus luteus) lives in southern Colorado, central New Mexico, and eastern Arizona. It was considered a subspecies of the more widespread "meadow jumping mouse" (Zapus hudsonius, now the "northern meadow jumping mouse") until 2017 and, since being split off, may potentially count as an endangered species.

Jumping mice are not members of the mouse family proper, although they do look rather similar. Instead, they belong to their own family, the Zapodidae, which was itself only raised to family status in 2013, following recognition of how different it was from the jerboas and birch mice that it had previously been included with. Even before that, however, it was recognised as more distant from the true mice and rats than, say, hamsters, voles, and blind mole-rats due, in part, to the presence of two extra sets of molars in the jaw. In addition, like the jerboas, they have elongated hind legs adapted for jumping and long tails for balance.

The southern meadow jumping mouse hibernates for eight to nine months each year, spending most of its life in a state of torpor. However, because it has only recently been recognised as a species in its own right, we don't know a lot about its hibernacula, or, indeed, its resting places more generally. Studies on the northern and western (Z. princeps) meadow jumping mice show that they dig their own burrows, rather than using those prepared by other species, or, for that matter, sheltering in natural crevices or fallen logs. We do know something of how they select their desired habitat, but, prior to last year, the only study to look at this included a description of just one hibernaculum.

That's a good start, but it's only a start. Which is why a new study has taken a wider view, acknowledging that the species is threatened, not least because it relies on rivers and streams, and the few that exist in the American Southwest are mostly being developed for ranching (and, to a lesser extent, urban and recreation areas, plus some threats from forest fires and general drought).

The study looked at two populations of jumping mice, in Colorado and Arizona. Both lived in high-altitude river valleys dominated by willows and grasses and surrounded by higher forested slopes with juniper, pine, fir, and spruce trees. It consisted of humanely trapping mice, luring them in with oats and crushed peanuts, then fitting them with microchip tags and tiny radio collars before releasing them. They could then track them to their dens, following them daily through the winter to see when and where they hibernated, monitoring the properties of their hibernacula from November to June - two weeks after the start and end of the maximum hibernation period.

In total, the researchers captured 28 mice, but sometimes the collars fell off, or the battery went flat, or the animals just disappeared, leaving a total of 11 hibernacula. While two were on flat terrain, all of the others were on north-facing slopes. All were in vegetated areas with at least 30% tree canopy cover, and were, on average, 44 metres (145 feet) from the nearest river or irrigation ditch.

They consisted of single tunnels with a low slope running to a sleeping chamber 8 cm (3 inches) across at about 30 cm (12 inches) below ground level and without any food caches or side tunnels. The chamber was lined with insulating material, although what the mice used depended on what was available. Where there were oak trees, for example, the mice used fallen leaves from them, while, in areas dominated by conifers, they relied instead on locally gathered moss.

The depth of the tunnels is probably sufficient to ensure that temperatures remained stable throughout the winter, while the north-facing slopes accumulate more snow which may also help in that regard. What was notable, however, was that the hibernacula were colder than randomly selected sites nearby at the same soil depth. This suggests that the mice are searching for cold places - the shading tree cover, for example, may help with this, although detailed soil conditions may also be relevant.

Since the ideal temperature for a hibernaculum is one that matches the body temperature reached during torpor, jumping mice must chill themselves unusually far down. In fact, the average temperature in the sleeping chambers was just 1ºC (34ºF). Analysis of soil at different depths around the burrows showed that, at 10 cm below the ground, temperatures got much colder in December and January, commonly reaching -5ºC (23ºF), so at the coldest times of the year, the depth helps to insulate them, while in the spring it can get noticeably warmer, which is also undesirable before the animals wake up. Placing temperature probes deeper down, to 50 cm, made little further difference, so the mice are likely picking depths (and possibly soil conditions) to get everything as stable as they can without wasting effort by digging too far.

Since this is something of a preliminary study, we don't know how low the body temperature of a jumping mouse gets when it hibernates, but this would suggest it's probably around 1ºC. Although this isn't uniquely low for a rodent, it's lower than we would expect for ones that live and can't huddle together, with only a couple of species of ground squirrel known to regularly tolerate conditions colder than this.

All of this aids our understanding of what habitat conditions these animals need, and may go some way to explaining why their numbers are declining. They don't just dig anywhere, or even in just any location near water, but they carefully select places with just the right conditions to keep the temperature at just the level of chill that they prefer.

[Photo by Ryan Hodnett, from Wikimedia Commons.]

Staying Away from the Boys

2026-01-04T15:56:00.001+00:00

Many large, hooved mammals live in herds. This has obvious advantages of safety in numbers, spreading the work of looking out for predators, as well as benefiting from the knowledge of more experienced herd members as to the best places to find food or shelter. The structure and composition of these herds vary from species to species, but one common trait is that males and females often form single sex herds that travel apart for much of the year.

This is referred to as "sexual segregation", and was first formally described by Charles Darwin in The Descent of Man. It isn't unique to mammalian herd animals, being seen in everything from fish shoals to bird flocks, as well as in non-hooved mammals (dolphins, bats, primates, etc.) Most zoological research, however, has tended to focus on large cloven-hooved mammals, such as deer and antelopes.

Among these animals, it's notable that sexual segregation seems especially common where males are much larger than females. There are many possible explanations as to why this might be, and it would be foolish to assume that there is one single answer that applies to all species, or even that multiple answers all apply to the same extent. Nonetheless, we can look at particular species to learn what we can about them, especially if they are of interest to conservationists.

Perhaps the most iconic herd animal of North America is the bison (Bison bison). They are not considered an endangered species these days, having recovered from the effects of past overhunting over the course of the 20th century. They are, however, reliant on conservation efforts to stay that way and the free-ranging population may be smaller than you would think - it was estimated as no higher than 13,000 in 2017 with just five populations being considered truly viable. This is only just short of what would be required to list them as a "threatened species", and, while their population is not currently declining that we know of, it is "critically depleted", in part because bison currently occupy just over 1% of the land area that they did in Native American times.

Returning that figure to its natural level is obviously not an option, short of abandoning every city in the US. However, it does underscore the fact that bison require ongoing conservation just to stay as they are. Understanding how they use the landscape in which they live is a part of that, and if males and females aren't using the same parts of that landscape, it's useful to know why.

At least three possible reasons present themselves. It could be down to differing food requirements, with females seeking out areas with high-quality food to help with pregnancy and lactation, while males, being larger and not having infants to tend for, just look for wherever food is most plentiful, even if it isn't top quality. Alternatively, predators could be the driving force. Under this explanation, females use places that are safer for themselves and their calves, whether because there are just fewer predators about or because it's easier to see them coming. Meanwhile, the strong, muscular males brave more dangerous regions so long as they have good food to eat.

Thirdly, the different sizes of the sexes may mean that they have different patterns of activity through the day. One possibility that comes under this heading, for example, is that males often spend a lot of time aggressively sparring with one another, and females might prefer to avoid the hassle and attendant physical risk. But it could just be that they get up and travel at different times, which would obviously make herd cohesion difficult regardless of what else is happening.

A recent study looked at the herd of free-ranging bison on Antelope Island in Utah. This lies in the Great Salt Lake, just off the coast from Salt Lake City. With an area of just over 100 km² (40 square miles), this is uninhabited but does have campsites and other tourist facilities, primarily at the north end. It's open and often rugged country, dominated by sagebrush and grass, with just a few juniper and maple trees on the western side. It has long been home to mule deer, bighorn sheep, and pronghorn "antelope", with no predators larger than coyotes, which aren't much of a threat to adult bison.

Bison were first introduced to the island in 1893, having been imported from Wyoming. The herd currently stands at around 700 individuals, around a third of them immature. Surveys of the animals' behaviour were conducted over two years, focused on the mating and calving seasons. As the researchers point out, bison are good for this sort of thing because it isn't hard to spot a really big animal walking about on the open plains, compared with, say, small deer in a forest.

The study showed that while the bison gathered into mixed-sex herds during the rut, they tended to stay apart while the females were calving. The first part of that is pretty much a given; the sexes have to mingle during the mating season. The question is more about why they segregate at other times of the year.

The idea that they might be doing so because the females find the males aggressive and dangerous was supported by the fact that the all-male groups were extremely small, with an average of three individuals each. That's probably because even male bison find other male bison unnecessarily aggressive and would rather not get into fights with them when mating opportunities aren't at stake. All-female groups, where this isn't an issue, varied in size throughout the year, reaching an average of 27 adults each during the calving season... quite a bit higher than "three".

However, if segregation were down purely to wanting to stay out of each other's way, we would expect that the male and female groups would be evenly distributed across the island. They wouldn't be in the same place at the same time, but, over the course of a few days or weeks, they would be in the same place at different times. They would go where the food was, and that wouldn't change from day to day. They'd just pick somewhere else if their first choice was already taken.

But, while there was some overlap around the middle of the island, in general, male bison stayed in the north, and females in the south. This suggests that there is something different about those ends of the island that favour one sex over the other, if they're going to be apart anyway.

One possibility is elevation, with the highest hills on the island being at the southern end. It's not unusual for herd animals to give birth on higher, more rugged, ground, since it provides some degree of protection from predators amid steep terrain and shielding slopes. Bison are no exception, although, in this study, many of the females did use the lower, shoreward terrain in the south of the island, too, so that might be only a minor factor.

On the other hand, if they are trying to avoid predators, another reason to stick to the south is that most of the campsites, along with things like the visitors' centre and the most-used hiking routes, are all up at the northern end, where a bridge and causeway connect to the mainland. From a bison's perspective, humans might as well be predators, and it's well-known that many hooved animals will run away if we approach too close. The fact that all-female herds are larger when calving than during the rut would also support the idea that they are focusing on being protective.

We see much the same behaviour in yak.

Food quality seemed to play less of a role, at least among these particular bison. Having said which, the northern part of the island does experience some die-back of high-quality food at certain times of the year, which would benefit the males with their larger, more efficient digestive systems. Furthermore, those all-female groups that used the lowlands tended to congregate around patches of alfalfa and reeds along the southern shores, which could give them a benefit, even if their warier sisters in the hills were losing out.

This, the researchers suggest, gives us some clues as to how we can better help conserve bison. We could, for instance, limit visitor access to the southern parts of the island during the calving season, making it easier for female bison to avoid us at that time of year. We could also plant seeds of grasses and other nutritious plants along the relatively barren southwestern shoreline, giving females and their young more to feed on. Significantly, we wouldn't have to do the same in the north, where human activity is already greatest because the males are more resilient, both in terms of diet and adaptation to humans.

Male and female bison may not be different species, but sometimes it may help our conservation efforts if we treat them as if they were.

[Photo by Dallas Penner, from Wikimedia Commons].

Prehistoric Mammal Discoveries 2025

2025-12-14T16:40:00.001+00:00

Computer reconstruction of the skeleton of the Miocene
right whale Megabalaena, published this year

As another year draws to a close, it's time to take another dash through the mammalian palaeontological findings of the last twelve months. As always, this will be a rapid survey of some discoveries that I find particularly interesting without any detailed discussion, but hopefully covering as wide a range of prehistoric mammals as I can.

Large Herbivores

Land mammals don't come much larger than mammoths, and such animals, and their mastodon relatives, have shaped the world around them through their mere presence and the food they chose to eat. A study on southern mammoths (Mammuthus meridionalis) uncovered at a million-year-old site in northern Spain showed that the climate at the time was already Mediterranean, a warm gap within the Ice Ages, causing them to feed more on grasslands than trees. While hyenas had fed on some of the carcasses, they were not alone, as the evidence also shows clear signs of butchering with stone tools.

Elsewhere, a new study found that the South American gomphothere, Notiomastodon, which had entered the continent when the Panama Land Bridge formed, had a diet rich in fruit. When they went extinct as the climate warmed 10,000 years ago, their disappearance spelt trouble for the plants with large, fleshy fruit that had relied on them for seed dispersal.

But if Spain had a reasonably modern Mediterranean climate one million years ago, that obviously didn't last. Although we know they must have existed earlier, until this year, the oldest known cold-adapted mammals from Spain dated to no more 190,000 years ago, during the second-to-last Ice Age. This year, however, a reindeer tooth was described from the country that could date back as far as 300,000 years, during the Ice Age prior to that - it's the most southerly known reindeer fossil ever discovered.

On the subject of deer, a new analysis of the DNA of the Toronto Subway deer (Torontoceros hypogaeus) discovered during the construction of the eponymous transport system suggested that, while undeniably a distinct species, it might be more closely related to the living white-tailed deer than previously thought. It lived in open habitats, and was driven to extinction at least partly by the spread of the forests that now cover the area.

Studies of isotope ratios in the short-legged hippo-shaped rhinceros Teleoceras from Nebraska showed that they lived in much in the same areas throughout their lives, not migrating, or even moving as far as we would expect when they reached adulthood and left their parents. This may have been because they relied on habitats with plenty of water to wallow in, and they just couldn't find them elsewhere. Over in eastern Siberia, the longest ever fossilised horn from a woolly rhino (Coelodonta antiquitatis) was described; it was over 164 cm (65 inches) long, far larger than those of any living rhino.

Fossils are not necessarily of things like bones and horns, however. This year, the oldest known fossilised cowpat was described from 20 million-year-old deposits, also in Nebraska. At least it looks like a cowpat - it's about the right size and shape, and it's obviously dung - but cows didn't exist that far back, and, honestly, neither did anything closely related to them. So, probably a ruminant of some sort, but what kind will likely always remain a mystery.

Carnivores

Europe is devoid of big cats today, unless you count the occasional lynx. This was not always so, however, and a publication this year expanded the range of the European jaguar (Panthera gombazagoensis) to Poland and to Suffolk in England between 700,000 and 350,000 years ago. Another large cat, the famous sabretooth Smilodon fatalis, was discovered further south than ever before - in Uruguay.

We already knew about leopards (the modern species) in the Pyrenees during the Ice Ages, but their fossils are rare. A new study of their overall distribution and history showed that they became steadily larger as the Pleistocene wore on, and also that males and females were more similar in size than in living members of their species. Moreover, they showed a preference for mountainous habitats, and their body shape shifted towards that of modern snow leopards prior to their local extinction.

The largest mammalian carnivore during the Ice Ages in North America was the giant short-faced bear (Arctodus simus), with some individuals estimated to have weighed up to 950 kg (2,000 lbs). However, some fossils were significantly smaller than this and had been suggested to represent a distinct subspecies. Genetic analysis this year was unable to find any significant differences between the larger and smaller specimens except for one: all the large ones were male, and the small ones female...

And, well, yes, I should probably mention this, since it was all over the news. This year, a company claimed to have returned the dire wolf (Aenocyon dirus) from extinction, in the form of three, rather cute, white puppies. Almost nobody outside the company with any expertise agrees that this is really what they have done, simply adding a few dire wolf-like genes to what are rather obviously grey wolves.

Cetaceans are also carnivores, although descended from creatures that were not. An analysis of the skull of the very primitive dolphin-sized Protocetus showed that it had already developed the large brain that cetaceans are known for today but, more importantly, that it still had a keen sense of smell. Modern whales and dolphins have no use for smell but Protocetus likely spent at least some of its time on land, giving it a very different lifestyle.

Among more recognisable whales, Idiophorus is either the most primitive known sperm whale or a close relative of their last common ancestor. Recovered from Miocene age deposits in Patagonia, it had a long snout shaped like a wine bottle (so hardly like a modern sperm whale) and was around 6.6 metres (22 feet) in length. The key fossil was thoroughly re-examined this year for the first time since it was uncovered in the 19th century, revealing that it was a deadly predator feeding on relatively large vertebrate prey, unlike other known sperm whale species living in the area at the time.

At the other end of the Americas, and a few million years earlier, Fucaia was a Canadian whale belonging to a now-extinct group. Analysis of a well-preserved fossil this year showed that, like Idiophorus, it was an agile and active predator, although, partly because it was rather smaller, its hunting style was probably most like that of a modern sea lion. Despite this, its closest living relatives include the toothless krill-feeding baleen whales.

Other Placentals

Glyptodonts were gigantic relatives of armadillos, their heavy armour plating making them the mammalian equivalent of tanks. A study this year provided the first analysis of damage to that armour in three South American species, adding to those already performed on more common ones. It also confirmed that, like certain dinosaurs, they likely did wallop each other with the clubs on their tails, likely while competing for mates or territory.

Ground sloths have a long evolutionary history, much larger relatives of the small(ish) tree-dwelling animals that are their only remaining close relatives. A new analysis shows how they gradually increased in size as the climate shifted, but that, probably to the surprise of few people, it was not climate change that eventually drove them to extinction, but the arrival of humans.

Look at any reconstruction of a ground sloth, and it will inevitably show a large, somewhat shaggy animal. But it had been argued that, like elephants and rhinos, the largest ground sloths may not have had much fur at all, being so big that they did not lose much body heat anyway. It was a minority view, but geochemical analyses and simulations published this year seem to put the nail in its coffin. Given the relatively cool environments in which they lived, the simulations suggest that even the largest species would have needed a pelt of fur 10mm (0.4 inches) thick, and some would have needed fur three to five times that long. This, despite the fact that palaeothermometric analysis in the same study showed that they had unusually low body temperatures for their size.

Several studies this year took a look at the development of increasing brain size during primate evolution - primarily monkeys and apes. They did not all come to the same conclusions, which, while they don't necessarily contradict one another, either, suggests that there was probably a lot going on rather than one single answer. One study pinpoints visual processing as the main factor, with visual parts of the brain becoming disproportionately large before the rest caught up, something that would make sense for an animal that had to navigate through tree branches.

Another points to socialisation, with parts of the brain responsible for some of our most complex behaviour starting to enlarge early on. A third points out that brain enlargement correlates, to some extent, with the elongation of our thumbs (which are obviously very different from those of most non-primates), implying that fine manipulation of objects may have been another key element in the drive for greater intelligence.

Going further back in time, another study this year challenged the traditional view of primates originating in tropical jungles. Instead, it suggests that the very first primates appeared in cold and temperate habitats in the north, the changeable climates of such areas promoting the evolution and dispersal that allowed them to reach the tropics later on.

Among the rodents, analysis of a new fossil and comparison to others already known shed light on the origins of the group that includes the gophers. The fossil belongs to an unusually large relative of the ancestor of gophers and kangaroo rats, and seems to have already been a skilled digger. This supports existing theories that the ancestral gopher looked like, and had a similar lifestyle to, today's pocket mice.

An unusually complete fossil free-tailed bat found north of Marseilles in France dates back over 30 million years to the Early Oligocene. The shape of the wings and muscular anchors in the body shows that it was fast and capable of sustained flight, probably catching insects in the air as it flew over what was then a large lake.

Conoryctes was a medium-sized burrowing mammal living in North America just 3 million years after the extinction of the non-avian dinosaurs. It's probably a placental mammal, but it has been argued that it may actually be an early relative that's simply closer to true placentals than to marsupials. A study of the microscopic structure of its bones shows that it must have grown rapidly, reaching adult body size in a single year and that it would have been weaned at about the age we would expect for a true placental of its size. So even if, by some chance, it wasn't literally a placental mammal, it lived very much like one.

Marsupials and More

The best-known and distinctive marsupial species are probably the kangaroos and wallabies. All but one living species belong to a subfamily that spread through Australia starting around 11 million years ago. The oldest known fossil belonging to this subfamily belongs to Dorcopsoides, which may hold some clues as to why the group became so succesful on their home continent. A new analysis published this year found that it was a more efficient hopper than had previously been thought, suggesting that it was already adapted to open environments, something that was just starting to become more common at the time, as the Australian outback began to dry out.

It may have helped that, as another new analysis of the wear patterns on their teeth showed, Ice Age kangaroos had a wider diet than was apparent from their anatomy alone. On the downside, if they were adaptable, the fact that so many kangaroo species went extinct around 40,000 years ago is probably not primarily due to the change towards a colder climate...

While fairly detailed DNA analysis is possible for mammoths and other Ice Age animals preserved in permafrost, it is much harder for animals of similar age in warmer climates. A new study this year therefore used the molecular structure of collagen from sinews and other tissue to confirm the relationships between some of the marsupials that went extinct in the last 100,000 years. Perhaps the most significant finding was that the closest living relative of the extinct carnivorous marsupial lion Thylacoleo may be the koala, rather than it being an earlier relative of the koala-wombat common ancestor.

On the subject of carnivorous marsupials, these were at least as prominent in South America in prehistoric times, dominated in particular by the large sparassodonts. A new analysis of the skulls of South American marsupials this year showed that, while most had similarly shaped (elongated) brains to modern marsupials, that of the sabretooth marsupial Thylacosmilus was more rounded, due to its shorter, cat-like face. More significantly, revised estimates of its body weight showed that its brain was smaller than previously thought, in the range of some of its earliest ancestors, its size having evolved less than expected.

Similar analysis of an armbone belonging to the mysterious mammal Kryorectes confirmed that it was, as previously thought, a monotreme. Living alongside Australian dinosaurs a whopping 106 million years ago - 40% further back in time than Tyrannosaurus rex - the study also showed that it was probably a semiaquatic burrowing animal. The platypus lifestyle may be truly ancient.

Finally, the multituberculates were neither placental, marsupial, nor monotreme but a fourth grouping that survived well into the Age of Mammals. An analysis this year of their habitats and distribution across North America suggested that, at least on that continent, they were restricted to damp temperate forests dominated by redwoods and swamp cypress and that they finally disappeared when those forests dramatically declined as the world cooled at the end of the Eocene. This contradicts the popular theory that it was competition with newly arrived rodents that killed them off, since they were apparently living elsewhere at the time.

Synapsida is taking a break for the holiday period and will return on the 4th January

[Image from Yanaka et al. 2025, available under CC-BY-4.0]

Age of Mammals: The Eocene (Pt 1)

2025-12-07T14:10:00.000+00:00

When Scottish geologist Charles Lyell first created the system of epochs we now use for dividing the Age of Mammals, he designated four of them. This was in 1833, so he did not know the true age of the Earth, let alone the timespans of the epochs he was naming - he was basing them purely on geological strata and the types of fossil seashells found within them. We now know, however, thanks to the wonders of radiometric dating techniques, that the oldest of the four epochs he defined spanned over half of the Age of Mammals, longer than the other three put together.

Two epochs have been carved out at either end since, so the Eocene is not quite as long now as it was when Lyell named it. However, it remains the longest of the seven epochs since the extinction of the non-avian dinosaurs, still occupying a third of that entire stretch. As currently defined, it runs from 56 to 34 million years ago. Compared with the entire age of the Earth, that's not very much, but from the point of view of most mammalian palaeontology, that's unusually long.

If we imagine the history of the Earth as a single year, the Eocene would run from 5 p.m. on 27th December to 7 a.m. on the 29th. That's over a day and a half, and, at least as importantly, you've still got nearly three days to go by the time it's over. It's probably only because it's that far back in time that we are happy to keep it that long, when more recent epochs span just a couple of million years. Like the other epochs, we do subdivide it, but, here, even the subdivisions are relatively long. While there are detailed subdivisions for specific purposes, the most commonly used are the standard scheme of "Early, Middle, and Late" and a more formal division into four "stages", the latter splitting the Middle Eocene into two and giving each of the bits fancier names.

Given the length of time involved, it should not be surprising that the continents shifted about significantly during the epoch. But although we could easily identify them all if viewing the planet from space, there were also significant differences from the forms they take today.

Europe, for instance, genuinely was a separate continent, separated from Asia by the Turgai Strait on the eastern side of the Ural Mountains. The closure of this strait is one of the markers of the end of the epoch, but for much of the Eocene, an island continent stretched from the Urals to Scandinavia. South and west of that was an archipelago of large islands. At times, the Balkans and Anatolia formed a peninsula stretching westward from Asia south of the Turgai Strait (which connected, via the Caspian Sea and much of what is now eastern Europe, to the Baltic), but at others, they were at the eastern end of the archipelago. With the Alps not yet fully formed, shallow seas meant that Italy was also an island, separated from a larger one in Central Europe to the north.

On the other hand, during the Early Eocene, it may have been possible to walk from France to Canada. This is partly because the English Channel may have been dry at the time, but also because the North Atlantic had not fully finished opening. A land bridge ran from Scotland to Greenland - probably a distance of less than 500 km (300 miles) at the time - and the Davis Strait west of Greenland may have been at least partially closed.

At the same time, India was an island, heading north towards Asia and pushing up the Himalayas in the process. It collided with the continent around 52 million years ago, during the Early Eocene, but only at the western end, around what is now Pakistan. For the remainder of the epoch, a bay stretched across the Gangetic Plain, and fluctuations in sea level may have meant that the last water passage between the two did not close off until the following, Oligocene epoch.

The Drake Passage between South America and Antarctica probably opened up, at least as a shallow strait, during the Early Eocene. Perhaps more significantly, Australia and Antarctica were still a single landmass, only separating in the Late Eocene with the creation of the Tasmanian Passage.

These differences are further amplified by the fact that the Earth was warmer then than it is now. Without ice caps, the sea levels were much higher, creating islands from what are now peninsulas. Indeed, the official marker for the start of the Eocene is a sudden and dramatic change in the isotopic composition of carbon, indicating a rapid rise in CO₂ levels.

What caused this is a matter of debate. Theories include widespread volcanic activity and a cometary impact, but one of the more popular is that a pre-existing rise in temperatures reached a tipping point. At the very beginning of the Eocene, things became just hot enough to liberate gases from the frozen methane clathrates on the seafloor. Once that happened, a vast reserve of greenhouse gases was released in one, sudden, titanic belch.

Global warming accelerated as carbon was pumped into the atmosphere. It took 200,000 years for the Earth to recover by locking the carbon back into newly formed sediments. This time is known as the Palaeocene-Eocene Thermal Maximum, and it's the hottest the Earth has ever been since the extinction of the non-avian dinosaurs.

Worldwide temperatures were at least 5°C (9°F) higher than they are today. There were tropical jungles in Montana and England, complete with citrus trees, cashews, avocados, and lianas, and there were even denser rainforests and mangrove swamps across the American southeast. Alaska was covered in broadleaf forest, and much of Siberia was subtropical.

If anything, the changes were even more noticeable beyond the polar circles, with alligators swimming in the rivers of Ellesmere Island in the Canadian Arctic and forests of beech trees in Antarctica. This, bear in mind, is despite the permanent darkness of midwinter at these latitudes, regardless of the temperature. Indeed, we have fossilised tree rings from Greenland showing that it experienced the same seasons then that it does now, so it isn't as if the Earth had a different tilt. Evidently, so long as it was warm enough, the trees could happily go dormant for months at a time when the sun never rose.

The equatorial regions were, perhaps surprisingly, the least affected, with their temperatures increasing only marginally. This is likely because the global warming was primarily due to methane, not carbon dioxide. The former promotes the growth of stratospheric clouds over the poles that trap the heat there, but does relatively little where it is already hot. Either way, the fact that warm air can hold more moisture meant that the world's climate was much wetter than it is now, promoting rainforests far beyond their present reach.

Nonetheless, all of this did come to an end. Even once the PETM was over, temperatures initially remained well above where they are now, and the decline was slow, although not entirely steady. The cooling accelerated suddenly around 49 million years ago, at the end of the Early Eocene. One theory for why this happened is the so-called Azolla Event, when freshwater ferns in the Arctic Ocean built up to the point that they rapidly sequestered carbon as they died, bringing the CO₂ levels down to not much higher than they are now.

After stabilising, there was a brief reversal of the cooling trend around 40 million years ago during the Eocene Climatic Optimum, when CO₂ levels rose again, possibly due to volcanism. This lasted almost a million years before temperatures resumed their downward trend, culminating in the "icehouse" of the Oligocene.

The name "Eocene" means "dawn of the recent", referring to the fact that it is the first time that the fossil seashells Lyell was studying at least begin to approach their modern forms. When we look specifically at mammals, however, it's arguably less appropriate, since recent types of mammal are comparatively rare. The Eocene is defined in large part by the fact that most mammals were of ancient kinds that no longer exist, not by the ones we are familiar with today.

Furthermore, the fact that this is both and early and a long-lasting epoch within the Age of Mammals means that there are almost no living families of mammals that stretch back beyond it. It was the time of the first shrews, moles, hedgehogs, rabbits, armadillos, rhinos, and monkeys, among many others. Indeed, the Early Eocene is marked in part by a rapid diversification of mammals, such that even the living orders - the larger groups into which we collect the families - often have their origins here. The oldest fossils of carnivorans, cloven-footed mammals, whales, and bats all date back to the Eocene.

Moreover, these were generally in a minority at the time. Carnivorous mammals existed, but very few of them belonged to the main living group, the carnivorans, with other, vanished, animals taking their place. For mammals, this was often a time of beginnings and (to modern eyes) of oddities.

Over the next couple of years or so, I will be looking at as many of them as I can. Starting with mammals that lived on the European archipelago...

[Painting by Larry Felder, in the public domain.]

Viverrids: Civets Great and Small

2025-11-29T16:50:00.004+00:00

Small Indian civet

Most species of true civet weigh somewhere between 7 and 9 kg (16 to 20 lbs), somewhat larger than a domestic cat, and similar to a King Charles spaniel or a Highland terrier - albeit of a more slender shape. The large Indian civet (Viverra zibetha) is at the upper end of this range, the second-largest species of true civet. The name of the large Indian civet, however, tends to imply that there must also be a smaller version in the same general area, and so there is.

The small Indian civet (Viverricula indica) is the smallest of the true civets, between 48 and 68 cm (19 to 27 inches) long, not including the tail, and weighing just 2 to 4 kg (4½ to 9 lbs) - more of a toy poodle or a Pomeranian in terms of dog weight. Otherwise, it looks much like its larger cousin, although it has block spots like a Malay civet, rather than the fainter blotches of its larger namesake, and does not have the crest of erectile hair running down its back. The smaller size caused it to be placed in a separate genus from the other civets as early as 1838, and modern genetic analysis has shown that this is fair; the two parted company about 12 million years ago, while the other Asian civets are much more recent than that.

The small Indian civet is both common and widespread, being found across the whole of southern Asia from eastern Pakistan to Shanghai and Taiwan, and reaching Sri Lanka and Singapore in the south and perhaps more surprisingly, Java and Bali. At least twelve subspecies have been proposed to exist across this range, but a genetic study in 2017 was able to confirm only four of them. This included both the Javanese and Balinese populations, but the researchers weren't able to collect samples from mainland China, among a few other places, so the true story may be more complicated.

The animal has also been introduced to other Indonesian islands, and to places as far away as Madagascar. This is, as so often with civets, because of the chemicals in its perineal scent glands, which can be used for making perfume. The glands are well-developed, with distinct lips that are poked out when it sprays - targeting, as cats do, almost any vertical object. They are present in both sexes, but larger in the males.

They are woodland animals, preferring dry forests to humid jungles. They may be found on the fringes of rainforests, but do not venture far inside. Other common habitats include scrubland, bamboo stands, and wooded riverbanks where the vegetation is not too dense. Studies in Shanghai indicate that some will enter urban environments, making use of parks and other patches of greenery, but that they restrict their activities to the hours around midnight, when humans (and also stray cats) are least likely to be around. Elsewhere, while they are always nocturnal, their activity is more spread out through the night, peaking in late evening and early morning when there is more light to see by.

This also happens to be when many local rodents are most active, and one can see the advantages of that to a small omnivore. In fact, rodents are perhaps their most common prey, with their remains being found in 90% of their scats according to some studies. Insects and worms are also a significant part of their diet, and, while they seem to prefer animal food, they do eat leaves and fruits. On the other hand, birds, fish, and snails are only a minor component of their food, something eaten when the opportunity presents itself but not actively sought out. They also seem to ignore rice, even when they hunt in paddies, which must be a help to the local farmers.

They are solitary, with males in particular marking their territory with their spray. Individuals have been reported to have home ranges of between 2 and 3 km² (500 to 750 acres). Although they can climb trees if they have to, they sleep during the day under dense shrubs or, if they get lucky at finding a pre-made one, in holes in the ground. They are not especially vocal animals, although they do make alarm calls when stressed and females have been recorded screaming at males in captivity when the latter get aggressive.

The only other call they are known to make is a "da-da-da" sound used by males to entice females in heat. This happens in the spring, and occasionally also in autumn, lasting 12 to 15 days. Males scent mark just about everywhere when they are aware of a nearby receptive female, although the latter only do so in their immediate vicinity, probably hoping the male comes to them. Courtship is aggressive, with the female initially fighting and screaming before she calms down and encourages the male to mate. In one study, she gave a call when she had had enough, which the male respected, but then got bitten for his trouble anyway.

Pregnancy lasts about 67 days, during which time, and for two months after giving birth, the female stops scent marking altogether, probably wary of giving away her position while vulnerable. She gives birth to a litter of two to five furred, but blind kittens, whose eyes open after five days.

African civet

At the opposite end of the scale, the largest of the true civets is the African civet (Civettictis civetta). Ranging from 67 to 84 cm (26 to 33 inches) in length, not counting the tail, they typically weigh between 10 and 15 kg (22 to 33 lbs), although larger individuals have been recorded. They have a similar spotted pattern to some Asian civets, but the background coat is brown or reddish, rather than grey, and the erectile crest down the back is more prominent. Despite living on a different continent, they are more closely related to most Asian civets than the small Indian species is.

They are also perhaps the most widespread of the civet species, being found across the whole of Africa south of the Sahara and north of the southern arid belt; Lesotho is the only sub-Saharan African nation they are not recorded in at all, although they inhabit only the more northerly parts of Nambia, Botswana, and South Africa. Within this broad region, they inhabit a wide range of woodland and savannah habitats, avoiding only the densest jungles, unless rivers or logging roads make access easier.

African civets are highly omnivorous, with analysis of their diets in different areas, or even in the same area at different times of the year, showing it varying from 65% plant-based to 65% animal-based. They also do not seem to discriminate much, eating a wide range of fruits, leaves, and roots alongside rodents, birds, insects, and millipedes. Indeed, they eat enough fruit that they are important distributors of seeds, with clumps of seedlings germinating in hollows they have rested and defecated in and at their shared latrine sites.

The home range of African civets has been reported to be relatively small, around 0.8 km² (200 acres), perhaps because they live in fertile areas where they do not need to travel far to find food. They scent mark their latrine sites, which are situated along trails close to their resting sites, as well as prominent vertical objects such as tree trunks. The perineal scent gland is, like that of other civets, well-developed, formed of muscular, paired hair-lined sacs and distinct lips. Anything up to 2g (~ 1/8 of a tablespoon) of civet oil may be deposited on each visit, although, in practical terms, collecting this from civet latrines in the wild as an alternative to farming the animals yields no more than half this. Males have larger glands than females and produce a stronger-smelling scent.

African civets also possess small anal scent glands, marking their dung directly, rather than via spraying. The purpose of these remains unclear, although it seems likely that they must provide some additional information to the species that the main glands cannot.

There is no distinct mating season, with mothers able to give birth to two to three litters each year. Mating seems less aggressive than in the small Indian civet, although the female snaps angrily at the male at the beginning of courtship, before eventually inviting him to chase her as an immediate prelude to the act. Litters of up to four are born after a 73-day pregnancy, and, from the limited observations conducted, apparently with their eyes open. They can stand on all fours at five days, which is also quicker than is the case with many other carnivores.

The African civet is the only true civet living in Africa today, although at least one close fossil relative is known to have lived there in the last few million years (that is, since Africa and Asia shared a land bridge). It is, however, far from the only viverrid on the continent, and next time I will be looking at some of the others.

[Photos by "Tomio 34456" and Nikolai Usik, from Wikimedia Commons. Cladogram adapted from Gaubert et al. 2006.]

Climate, Cloud Forests, and Cotton Rats

2025-11-23T16:40:00.005+00:00

The Cricetidae is the single largest family of mammals in terms of the number of species, at least according to the current count from the American Society of Mammalogists. While the name translates as "hamster family", the great majority of species are not themselves hamsters. In fact, there are five subfamilies of cricetid: the hamsters, the voles, and no fewer than three with members that basically look like mice or rats... even though the true mice and rats belong to the second-largest family, the Muridae.

The largest of these subfamilies is the Sigmodontinae, consisting of mouse and rat-like animals primarily native to South America, although one species lives as far north as Virginia, and several others reach Arizona and New Mexico. The group is named for the S-shaped pattern on the molar teeth and was originally coined as the genus name for the cotton rats in 1825.

There are over four hundred known species of sigmodontine, with more being discovered all the time, and it's reasonable to ask why that number should be so high. Partly, it's down to a successful body plan, the same thing that helped the true mice achieve their similar dominance. The sigmodontines, however, also had an advantage that their northern counterparts, the neotomines (deer mice and packrats), did not: South America.

For millions of years, South America was an island continent, cut off from the outside world much as Australia is today. Few animals could reach it from elsewhere. It did have rodents, in the form of the guinea pigs and their relatives, but what it didn't have was anything much like mice. That changed as North America got closer.

The first sigmodontines may have crossed into South America four million years ago, when the Panamanian land bridge formed. It's also possible that it happened a little earlier, since Central America was a chain of islands for a while, potentially close enough to allow small land animals to be swept from one to the other in tropical storms. But, either way, when they got to the new continent, they found it ripe for the taking.

They rapidly diversified, moving into habitats where there was nothing else like them to act as competition. Across South America, there are jungles, pampas, mountains, swamps, deserts, and grassy plains. All of these could provide new lands for the incoming rodents, separating them off from their neighbours by geography or by specialisation to a particular environment. The number of species rocketed from what must have been a small initial sample to the great range we have today. Sigmodontines live just about anywhere, the result of a great - and rapid - success story.

The very rapidity of this diversification creates difficulties in determining the details of how it happened. With so many species and larger groups being roughly the same age, it's difficult to determine what order things happened in, and therefore how the details of geography and environment shaped their evolution. Over the last decade or so, there have been at least fourteen different patterns proposed, based on genetic studies, analysis of body form, and so on. Only one of these seems to be supported by a more recent study, and even that was only able to look at 11% of the species and around 40% of the generic diversity.

That's actually not too bad, considering the huge number we're talking about. But scientific techniques advance, and new species have their genetics sampled. It's not just biological science, either; advances in geological science can improve our picture of how things like the climate changed over millions of years, allowing us to match evolutionary events with the ecological changes that drove them. All of which results in the latest analysis, and the largest so far, including 38% of the known species and 84% of the genus-level diversity. It won't be the last word, but it's a move in the right direction.

When we do these sorts of studies, we analyse the genetic differences between different species, calibrating them with fossils of known age to determine not just in what order, but when, splits between species or larger subgroups occurred. By checking to see where the species live, we can also get some idea of where their common ancestor is likely to have lived, adding this information to the picture.

One of the things we want to know is when and where the very first split within the group occurred, effectively allowing us to pinpoint its origin. In this case, the study showed that the sigmodontines originated 10.5 million years ago, in the Late Miocene. Moreover, it seems have happened in South America. This is significant because it had often been thought that the group originated in the north before the crossing, partly due to the oldest known sigmodontine fossils coming from Mexico, and partly because the few North American species, such as the hispid cotton rat, belonging to the oldest branch in the family.

The reasoning had been, therefore, that these North American species were descendants of the sigmodontines that had never left their home. But the new study, including more cotton rat species from the south, shows that this is unlikely, and that the northern species descend from some group that crossed back again much later. That Mexican fossil, for example, dates to about five million years after the study suggests the group first appeared, so says little about its ultimate origin.

This, of course, favours the "island-hopping" theory about the arrival of the group in South America, rather than them having crossed over by land alone. However, it does leave a long gap between their putative arrival and the beginning of the rapid diversification, which took off in earnest, according to this new study, 6.6 million years ago, at least four million years after the apparent colonisation of the continent.

Why the wait? Well, 6.6. million years ago is in the middle of a cooling period as the Miocene came to a close and lines up very closely with the collapse of the Pebas Mega-Wetland. This was a vast series of lakes, or possibly just one single body of water, that existed for much of the Miocene east of Ecuador and northern Peru, connected by a long channel through present-day Colombia to the Caribbean. This effectively turned the northern part of the Andes into a peninsula, separated from the Amazon by a great bay, or at least an area of swamps and wetland.

So, it could be that the early sigmodontines were trapped on this peninsula, and only expanded beyond when the climate became drier and changes in geography helped drain the wetland. When that happened, it seems that the sigmodotines emerged into the rest of the continent in a great burst, heading into areas that changing weather patterns were making wetter and more forested.

This fits with the time that cloud forests, long present in the north of the continent, and in Central America, began to reach further south as the central Andes increased in height. Other groups of organisms also underwent expansions at the same time, including hummingbirds and wild gentian flowers. Unlike these, and unlike their closest relatives, the Central American tylomyines, the sigmodontines took advantage of the expansion of the cloud forests to reach beyond them, colonising new lands in the south.

Even so, the study suggests that, at this time, the rodents lived in comparatively warm, damp forests, even if they had moved down slope into the Amazon and the woodlands of what are now northern Argentina and Uruguay. Beginning in the mid-Pliocene, around 4.7 million years ago, that began to change as individual lineages began to undergo their own smaller bursts of diversification, one at a time, expanding into less forested habitats. A subgroup called the leaf-eared mice were among the first, entering the semi-arid parts of the central southern continent in the Gran Chaco region. This continued until about 3.1 million years ago, by which time the whole of the continent had been colonised, as far as Tierra del Fuego, where the olive grass mouse is the most southerly species of wild land mammal in the world.

Again, this was a time of shifting climates, in this case, towards drier weather. Many other mammal species died out at this time, as grasslands took over from forests. This it seems, gave the sigmodontines another chance to expand, if more slowly and sporadically than before, partly due to being forced to adapt to changing conditions, and to potential competitors dying out and ceding their territory when they failed to do the same.

The sigmodontines, it seems, were in the right place at the right time.

[Photo by Yamil Hussein, map by Armin Reindl, both from Wikimedia Commons]

Splitting the Troop

2025-11-16T13:23:00.002+00:00

Primates are, for the most part, social, group-living animals. This underlies many aspects of human behaviour and likely played a role in our development of intelligence. Often, these groups have a fission-fusion structure, where new members come and go, but, in other species, they can be long-lasting and stable. Either way, just as with nation-states or tribal societies among humans, nothing lasts forever. Groups die and new groups form.

This can be due to disaster or misfortune, but it can equally well be due to success. If a group becomes too large, there may no longer be enough food in the local area to keep it healthy, or parasites or disease may spread too rapidly within it. Or it may simply become too large for dominant individuals to control. In fission-fusion societies, this may lead to a temporary break-up into local subgroups that otherwise remain in contact. Sometimes, however, the pressure is too great and the only solution is for a new group to form.

Perhaps surprisingly, such permanent break-ups are most often due to competition between the females, not the males. This is because primate groups, like those of most social mammals, are matrilineal. Males leave home on reaching adulthood, but females stay with their mothers so that, regardless of the dominance of any incoming males, it's their lineage that remains at the core of the group.

In primates that live in small groups, such as lemurs, there is generally a fight between female relatives over resources, and the loser leaves to establish her own family. For species that prefer larger groups, however, things cannot be so simple. If we'd expect even smaller groups of that species to have multiple females, then there has to be a social break, with at least some females having to decide on their new allegiance - they can't head out on their own, but they can't all stay where they are, either.

In rhesus monkeys, for example, what seems to happen is that, as the group gets larger, some members spend more time hanging out with their best friends. These bonds are developed through social grooming - removing parasites and so on from each other's fur - and, if you can't groom everyone, you'll end up with preferred partners. As the group grows, such cliques become more significant and well-defined, until eventually the group splits along these pre-established lines. Basically, everyone gets to stay with whichever new band they think are the cool kids.

None of this is to say that males are never responsible for group break-ups, especially in more "patriarchal" species. In langurs and gorillas, for example, where there is typically only one dominant male in each group, if a younger individual becomes old enough and can't displace the existing patriarch, he takes a small group of females and leaves to create his own harem. He might ensure that they are all close relatives, or might not have any choice in the matter, but often, he acts despotically, forcibly breaking up pre-existing social ties to get what he wants.

Nor is this rigidly tied to every species. Formosan macaques (Macaca cyclopis), for example, can have either female or male-led break-ups, although the former are apparently more common.

Olive baboons (Papio anubis) have some of the most complex social structures of any non-human primate. One of six currently recognised species of baboon, they are the most widely distributed, living in savannah and woodland habitats from the southern edge of the Sahara to the northern edge of the dense jungles of the Congo, across the width of Africa from Mali and Guinea in the west to Ethiopia and Kenya in the east. Troops, which are based on female kinship, can be large, with some having over a hundred members.

Laikipia, in central Kenya, is a savannah region including some wildlife conservation areas. Baboon troops have been studied in the area since 1971, building up information on their behaviour and maternal relationships. (Paternal relationships are less important in baboons, as they are in most primates and, in any event, paternity is harder to test for). In recent years, an invasive cactus has begun growing in the area, proving so tasty and nutritious that the baboon population in the area has increased, leading to the classic situation where the existing troops are becoming too large to manage.

Between 2011 and 2016, two troops lived in the area, including one that had been in existence since at least the start of the studies forty years prior. Following the fission-fusion pattern, there was some movement of individuals between the two groups starting in May 2014. It began with a single male baboon, whom the researchers named 'Yohan' moving from the larger, older troop to the smaller one. He stayed there for three months and then moved back.

In January 2015, he returned to the smaller troop, this time bringing three females with him. One of the females stayed, probably because her mother already lived in the smaller troop, but the other two went back home in June. Then, in October, they came back again, accompanied by two other adult females, their young children, and two subadult females. A month later, another adult female headed across to join them, making the two troops comparable in size.

Finally, in June 2016, Yohan took all of these females with him and left to found his own brand-new troop. This allowed the researchers to look back through their data to see what had precipitated the foundation process and how and why the female baboons had decided to join Yohan rather than stay where they were.

As expected, when they did this, the researchers were able to show that the females joining the new troop had groomed one another more often than they had groomed other baboons in their own troop. Even when they lived with the smaller troop before the split, they remained somewhat apart from them, socialising less than might be expected. They were also more likely to be closely related to one another, perhaps explaining why they socialised more in the first place. In this respect, leaving to form a new troop was less of a stress than it might have been.

But the male, Yohan, was also a major factor. He had sired many of his companions' infants and, leaving aside the possibility that they just happened to like him (for whatever that means when you're a baboon), this would have given the females concrete reasons to stay as his partners. Male olive baboons, unlike those of some closely related species, are unlikely to kill the children of their rivals. However, they do back up the mothers of their children when the latter are picked on by other members of their troop, protect them from predators, and generally do everything they can to help them. It's to their benefit, after all; young female baboons who have good relations with their fathers can live for up to four years longer than those who do not - and that means more grandchildren for the father.

Thus, when Yohan left the original troop, the females who had given birth to his children faced a tough decision. If they stayed behind, they and their children would lose his protection, but, if they left, they would be breaking long-term ties with their female friends and relatives. The fact that the females that ended up staying with him were related (on their mothers' sides) and were already more friendly with each other than with other troop-mates probably made this an easier choice than it might have been.

Thus, while we don't know what triggered Yohan's decision to strike out on his own in the first place, we can see how he persuaded so many females to come with him and join in his new venture.

It's also interesting to note that the split wasn't as total as the researchers had initially supposed. Certainly, the new troop lived apart from the old one, but they were not entirely isolated. They remained relatively close by, and occasionally met up with and groomed members of their former troop. Perhaps they had formed at least some friendships during their stay, or their prior moves had made them less xenophobic - or they were just that way to start with. It may have helped that, in this particular area, there was plenty of food to go around, but not many sleeping sites, so differing troops would often end up close together without having to compete.

Regardless, at least in this case, splitting off to stay with a favoured partner and form a new troop didn't mean that the baboons had to totally abandon all of their former companions.

[Photo by "Zenith 4237", from Wikimedia Commons.]

Viverrids: Civets of Southeast Asia

2025-11-09T17:00:00.002+00:00

Large Indian civet

Our definition of the Viverridae family has, as I noted previously, undergone some ups and downs over the centuries. It was first named, in 1821, for the genus Viverra, which, when it was first described, had contained five species. Even by 1821, four of those had been moved elsewhere - and three are no longer even in the family. But, by the rules of scientific naming, unless we scrap the family entirely, Viverra must remain within it. Which leaves one species that, in a sense, defines the family, and against which everything else in it is compared.

That species is the large Indian civet (Viverra zibetha). Thus, even though "viverra" literally means "ferret" in Latin, this means that we can reasonably call the Viverridae "the civet family", as I will be doing from here on in.

When it was described in 1758, this animal was said to live "in India", hence the modern name. That's certainly true, but we now know that it's a relatively small part of its range. Within India, it primarily lives in the far east of the country, in Assam and its neighbouring smaller states, while it's also found across Bangladesh (which, of course, was part of India until 1947). But it's also found in Nepal and Bhutan and across the whole of mainland Southeast Asia to Vietnam and peninsular Malaysia.

The details of its presence in southern China are less clear; while it is certainly still found in at least some areas, it is far more restricted than it used to be even a few decades ago, and it has likely been driven out of at least some provinces, with the subspecies native to Hainan potentially being extinct. In fact, the decline across the region was significant enough that the large Indian civet was on the verge of being listed as a threatened species in 2008. Since then, large populations have been discovered in places such as Thailand and Cambodia, and the species as a whole is now considered reasonably safe.

While most civets are described as being about cat-sized, the large Indian civet is, well... larger. It weighs around 8.5 kg (18 lbs), similar to a dachshund or a Jack Russell, and has an average body length of about 80 cm (2' 7"). It has the typical slender form and pointed nose of the family. The fur is grey and marked with blurred black spots or irregular blotches, with black-and-white rings on the tail and white markings on the face. A crest of black hairs runs down the full length of the back and can be raised to make the animal look larger - the hairs can easily be 7 cm (3 inches) in length, so this is quite noticeable. Like cats, they have retractable claws.

However, despite it being the most widespread of the Asian civet species, and reasonably common in at least some areas, we know surprisingly little about the large Indian civet beyond the basic facts. It simply hasn't been studied very much.

We know that they live in evergreen forests and that they are more tolerant of humans than one might expect, such that they are found, for example, in plantations. Despite this, although they can climb trees if they have to, they spend most of their time on the ground. While some are found close to sea level in places like Bangladesh, they are more common in hilly forests up to around 2,500 metres (8,200 feet) elevation - although one was spotted in Nepal much higher than this.

They are nocturnal, apparently being most active in the evening and the first half of the night rather than the early hours of the morning. They are omnivorous, with a wide diet that may help explain their resilience outside of China, where they were widely hunted in the second half of the 20th century. Grasses are the most common plants eaten, but they also eat fruits and may play a significant role in seed dispersal. At least in central Thailand, they particularly favour the berries of the golden shower tree but this likely varies depending on what's available. Their animal prey includes a high proportion of insects, along with lizards, rodents and almost anything else they can find in the leaf litter of the forest floor. They do not appear put off by venomous animals, such as scorpions and centipedes.

They are solitary animals, with one individual recorded as having a home range of 12 km² (4½ square miles), although they can share latrine sites, which they likely mark with scent to advertise to potential partners or warn off rivals. From what we can tell in captivity, they give birth twice a year to litters of up to four, whose eyes open at around nine days.

Malay civet

The Malay civet (Viverra tangalunga) is rather better known. It is also a common species within its area and lives in a similar habitat, albeit further to the south. The two species overlap in peninsular Malaysia, but the Malay civet is also widespread across Borneo, Sumatra, and many of the smaller islands between them. It is also found in the Philippines, Sulawesi, and Java, but these populations are not thought to be native, having been introduced to the islands by humans.

The Malay civet is smaller than its Indian cousin, being about 65 cm in body length, with a tail about half that, and weighing around 5 kg (11 lbs). It is marked with numerous black spots, which are far more distinct than on the larger species, but it otherwise has a similar coat pattern. They also have similar habitats, including the tolerance for humans that allows them to inhabit palm plantations as well as more primeval jungles. Within these forests, they apparently prefer shady areas, preferably with some fallen logs and dense undergrowth - all factors that are likely to increase the ground-dwelling invertebrates that form much of the animal component of their diet.

They are nocturnal, with activity peaking just after sunset and again just before dawn. Individuals may travel around 9 km over the course of a night, sleeping under cover during the day. However, their home ranges are far smaller than those reported in the (admittedly limited) studies on the large Indian species. The largest recorded home ranges reach 214 hectares (530 acres), around a sixth the size of that apparently used by its larger relative. Most, however, are far smaller, with some evidence that those living in unlogged and undisturbed forests require less space to acquire enough food. Around 100 hectares (250 acres) seems typical, and these ranges overlap even between members of the same sex, suggesting that they are not territorial - although they do avoid one another when they can.

They also scent mark and use latrines, although there is less evidence as to whether or not they share them. While there is almost certainly informative scent in the dung piles, Malay civets also mark by rubbing their backsides onto tree trunks or on patches of the ground. We have less information on their breeding habits, although they are probably similar to those of their relatives.

Large-spotted civet

The closest living relative of the large Indian civet, however, is the large-spotted civet (Viverra megaspila). This also lives in Southeast Asia, but does not extend to India or Bangladesh. Instead, it is found from southern China and northern Myanmar in the north down to roughly the Thailand-Malaysia border in the south. Or, at least, it was, since it has only been seen once in China since 1998, and they have probably vanished from the parts of Vietnam, Thailand, and Laos. That would leave them in two discontinous regions, one in western Myanmar and southern Thailand, and the other in Cambodia, southern Vietnam and southern Laos.

This is partly due to its preferred habitat. Unlike the two species I have previously mentioned, it's a lowland animal, rarely found above 350 metres (1,150 feet) and never above 800 metres (2,600 feet). This is prime logging land even when it isn't being converted into farmland or (less frequently) for industrial or residential use. Not only does this mean that its habitat is being rapidly destroyed, but it is also regularly hunted. It's not necessarily a favoured food in the area, but it isn't avoided, either, and so inevitably ends up in the bushmeat markets. This was what led to the decline of the large Indian civet in southern China as well, although until around 2000, that was also hunted for perfume from its scent glands that can now be made more cheaply by artificial means. But highland refuges, and perhaps a greater degree of adaptability, mean that that remains relatively numerous.

The large-spotted civet is somewhere between the large Indian and Malay civets in size. It is distinguished by having large blotchy dark patterns on the fur rather than the small or indistinct markings of the other two species. In other respects, it closely resembles the large Indian species, complete with the crest of erectile black hair along the back. Other than a preference for deciduous forests over evergreen ones and the lowland habitat, its behaviour is likely similar to that of its cousin. However, while we know, for example, that it is nocturnal, almost the only studies on the species relate to its low abundance and patchy distribution.

It has been formally listed as an endangered species since 2016, and when it's so difficult to even find the animal, there isn't much you can learn about it when all you have are a few stuffed specimens.

It used to be thought that the large spotted civet also lived in a third area, far to the west, along the Malabar Coast of Kerala and far southern Karnataka in southwestern India. And possibly, it does, but there are two reasons to think otherwise.

Firstly, in 1996, despite a remarkable similarity in appearance and largely due to the great distances involved, the Indian population was promoted to full species status as the Malabar civet (Viverra civettina). That took the name originally given to it in the 19th century, from which it had been demoted to subspecies status in the 1930s. If this is a species, it inhabits lowland jungles and swamplands along the coast but it has not been universally recognised and may not really be distinct. Indeed, it has been argued that it isn't even real, with the few sightings attributed to misidentification and the collected skins being of uncertain origin, and thus potentially imports from Southeast Asia.

Certainly, nobody has ever photographed one or captured a live specimen. The last recorded instance was in 1989, and that was a skin that we can't be confident was really collected in the area, and so could just be a large-spotted civet. Which brings us to the second reason: even if those 19th-century sightings were real and the animal existed then, if perhaps only as a subspecies, it's probably extinct by now.

Fortunately, the two remaining true civet species are far better off, especially now that we no longer need to kill them for perfume. Next time, I will be looking at those.

[Photos by "Tontantravel" and Kalyan Varma from Wikimedia Commons, drawing by John Gerrard Keulemans, in the public domain.]