Sunday 2 August 2020

Were 'Cave Lions' Really Lions?

A theme that crops up on this blog every now and then is that some animal that many people probably assume is a single species is actually two or more. There are, for example, three species of zebra, three jackals, and seven different kinds of musk deer. The lion (Panthera leo) is not one of these animals; there really is only one species alive today - and it's been around for a long, long time.

I remember a few years ago, at London Zoo, I happened to be passing the lion enclosure when one of the keepers was giving a talk. The lions at the zoo were obtained from the Gir Forest in Gujarat, India, rather than being the African sort. The keeper repeatedly referred to "this species" of lion when, for example, he was indicating how much more endangered the Indian population is than the African one.

A zoo talk, to an audience including a high proportion of children, is probably not the best place to get into a detailed discussion of taxonomy. But, from a technical perspective, the word he actually needed would have been "subspecies". Because, while there may only be one species of lion, there are multiple subspecies. But just how many?

Rather like the term "species" itself, it's hard to come up with an absolute definition of "subspecies" that every researcher is going to agree on. But, generally speaking, in order to qualify as a subspecies, a population of animals must:
  • Be readily distinguishable from other population(s) of the same species
  • Nonetheless, happily interbreed with those populations whenever they live in the same place
  • Live in a different geographic area than they do (because, otherwise, they'd breed themselves out of existence)
It's the first point that tends to be main bone of contention. Just how distinguishable do they have to be? This question has resulted in a number of different schemes for listing lion subspecies down the years. Perhaps because lions are a popular and dramatic animal, there has been a tendency for many researchers to identify new subspecies, often on evidence that, from a more modern perspective, looks a little flimsy.

Nonetheless, the 3rd edition of Mammal Species of the World, a highly influential listing first published in 2005, lists no less than eleven subspecies of lion as having reliable evidence to back up their reality. Two of those were considered extinct, but that still leaves nine.

That, however, was 15 years ago and, since then we have discovered a lot more about the genetic variability of lions. Or, rather, just how little of it there is between supposedly different subspecies. A 2010 study examined the detailed shapes and sizes of various lion skulls to determine whether, even if they were genetically similar, you could at least reliably tell some of the subspecies apart by, you know, looking closely at the animal. In almost all cases, it turns out you can't.

Most of the subspecies were stripped away. A review of the cat family conducted by the International Union for the Conservation of Nature in 2017 put together all of the evidence on lion subspecies gathered over the previous decade or so. It all pointed in the same direction: there are just two living subspecies of lion.

In an ironic twist, they turned out to be the two subspecies thought to be extinct when MSW3 had been written back in 2005. The reason that this is possible is because of the way that scientific names are awarded. When a species - or subspecies - is named, one of the things you have to do is say where you found your defining example of the animal in question. If your subspecies later gets split in two, the population that lives in the area given in the original description gets to keep the name you originally gave it, and the other receives the new one.

Or, as in this case, if two or more subspecies are subsequently merged, then whichever one was named first is the one that keeps the name, and the other name is cast into the accursed outer darkness of synonymy. When Linnaeus originally described lions (the species) in 1758, he said that they lived on the Barbary Coast - which today would be the region running from Morocco to Libya. When lions were first divided up into subspecies, the ones living on the Barbary Coast became the "nominate" form, against which the others were compared.

Unfortunately, lions in this area suffered heavy persecution in the 19th and 20th centuries. The last definitely known example in the wild was shot dead in 1942, although there were some reported sightings after that, and one analysis suggests that they may not have died out entirely until the late '60s. There had been quite a few in zoos, but they were bred with other lions, so while some of their genes may live on mixed ancestry descendants, Barbary lions as such did not.

When the subspecies were trimmed down to two, however, one of them, perhaps inevitably, included the extinct Barbary population. And since it had (by definition) been named before any of the others had, the living populations were re-awarded the name of the extinct one. Barbary lions might be extinct, but their subspecies lived across a much larger area than we thought, so that isn't.

Much the same happened with the second subspecies, originally named for an extinct population living near the Cape in South Africa, but which turned out to include several other, decidedly alive, populations further north.

The genetic history of lions reveals that, around 124,000 years ago, the ancestors of today's lions were split into two populations, probably by the expansion of the central African jungles during a warm gap between the Ice Ages (lions being savannah animals, not forest-dwelling ones). The one living in the south and east, from Ethiopia to South Africa, became the "southern lion" (Panthera leo melanochaita), while the other, spread much more widely across the continent, became the "northern lion" (Panthera leo leo).

At some point, probably less than 20,000 years ago (which is not all that long ago, on this sort of timescale), some lions from the northern subspecies headed out of Africa and into southern Asia. Virtually all of these were subsequently wiped out, but a remnant population survives in the Gir Forest of India. So those lions in the zoo don't even represent an Asiatic subspecies in the modern scheme of things.

Which doesn't make them any less endangered, of course.

But if the ancestors of Asiatic lions didn't reach Asia until 20,000 years ago, where does that leave "cave lions"? There are multiple fossils of what certainly look like lions found from Spain to eastern Siberia, some of them dating back at least 60,000 years. Some went so far as to cross the Bering Land Bridge into the Americas, leaving fossils in Alaska, and even parts of Canada. Clearly these cannot be Asiatic lions, so what were they?

A couple of facts about cave lions are significant here. One is that we know, from their fossils, that were noticeably larger than modern lions - something that may have helped them in a colder climate. Secondly, we also know, because contemporary humans helpfully left us paintings of them on cave walls, that the males apparently didn't have manes.

That, admittedly rather significant, point aside, though, they certainly looked just like modern lions - the cave paintings are that good. The paintings also show them hunting in prides, which no other species of cat does, although it's possible that that, at least, was artistic licence. This leaves us with two possibilities.

Firstly, cave lions might represent some third subspecies of lion, now long-extinct, that left Africa before the modern Asiatic lion did. Secondly, of course, they might not actually be lions at all, but just some other large cat that, apart from the manes, looked remarkably like them. It's apparently even been suggested that they might have been a stripe-less subspecies of tiger, which frankly sounds even stranger than a mane-less lion to me, but there you go.

Fortunately, there's a way to resolve this. The remains in question are so recent, geologically speaking, that it's actually possible to extract usable amounts of DNA from them. Which means that we can do the same sort of genetic analyses on extinct cave lions that we used to determine that there are only two living subspecies of lion.

This has been technologically possible for over ten years now, but it's something that's improving all the time. The most thorough such investigation was published earlier this week, and it adds considerable new detail and precision to what the earlier, smaller scale, studies had already revealed.

The study used 31 cave lion fossils and samples of 14 modern lions, comparing their genes and using their known ages (they're recent enough to carbon date, so this should be pretty accurate) to calculate the average rate of genetic change down the centuries. They then extended this rate to calculate how long it must have been since the various specimens last shared a common ancestor and derive a picture of not just what splits occurred in the lion family tree, but when they happened.

The conclusion was that cave lions and modern lions diverged 1.85 million years ago, with no evidence of interbreeding since. This is around the very beginning of the Ice Ages and much, much earlier, than the split between the two living subspecies of lion around 0.12 million years ago. In fact, it's not that much more recent than the estimated split between lions and leopards a little over 2 million years ago.

Given this, it's hard to argue that the cave lion (Panthera spelaea) isn't an entirely separate species from the modern lion. On the other hand, cave lions would have been more closely related to living lions than to any other living animal, so one could plausibly consider them to still to qualify as "lions" in the same way that the three living species of zebra all count as "zebras".

There are a couple of other twists to this story, though. One is that the same study showed a split within the cave lion lineage around 578,000 years ago. The two resulting lines are distinct enough that the authors regard them as representing different subspecies of cave lion. One, consisting of animals that were smaller on average, lived only in eastern Siberia, Alaska, and Canada, and seems to have been the first to die out. The other was found across the whole of Eurasia, eventually replacing the first one in eastern Siberia once it died out there.

The other question is where exactly this leaves the American lion (Panthera atrox), an animal sometimes also referred to as a "cave lion". This lived across much of what is now the United States and Mexico (and maybe further south) at around the same time that the "true" cave lion was living in the north. They weren't included in the 2020 study, but a more limited one in 2009 indicated that their split from the true cave lions took place 340,000 years ago, probably when they got separated from one another by ice sheets. 

So are they some other subspecies of cave lion? They can't be one of the two identified in the new study, because they are much larger, but if the dates of the two studies both hold up (which they might not) the only other possibility is that there were once three different species of cave lion.

Which may, or may not, have been actual lions, depending on your definition...

[Photo by Sumeet Moghe, from Wikimedia Commons.]


  1. There were lions in Europe -- Greece, for example -- in early historic times. Where do they fit in? I would guess they were a population of P. leo leo, the Barbary-to-India subspecies, but was this checked?

    1. Not from Greece, but they checked a couple from Iran, and they belonged to the northern subspecies, as you'd expect. Certainly there's nothing to indicate that the European forms would be anything else, given how late lions seem to have left Africa.

  2. Without mentioning the lion, I regret that there is no regularly updated checklist of mammals on internet except the outdated mammals species of the world website based on eponymous book (2005). All zoological group but mammal has their updated checklist : IOC, Tif,Clements for birds, amphibian species of the world, amphibiaweb for Amphibians, Reptile database for Reptiles,or fishbase for fishes, however mammals, nothing, nada.. it's not normal

    1. True enough. The IUCN is pretty good for cats specifically, but not necessarily for everything else (although probably better than MSW3, these days), and I often have to pull a number of resources together while writing these posts.

    2. And unfortunately, I don't know where to look for an updated list of update mine. Even wikipedia didn't add the list of new mammals described this year. If anyone could work on such a project, it would be really cool

  3. WP has an article for Panthera fossilis, aka P. leo fossilis or P. spelaea fossilis, apparently an older European lion. How do they fit in?

    1. That's not mentioned in the paper on the two subspecies, but it probably lived earlier than any of the samples they looked at (they were mostly interested in fossils recent enough to carbon date, and fossilis is quite a lot older than that) so the answer is that we don't know.

  4. Even after the 2009 DNA study of cave lions and American lions you mentioned, there was still some doubt about whether American lions were more closely related to lions or to jaguars. In 2011, Wired published an article asking American Lion, or Giant Jaguar? It referenced a 2009 study by Per Christiansen and John Harris in the Journal of Vertebrate Paleontology that concluded Panthera atrox was the sister taxon to jaguars, not cave lions, based on skeletal, particularly skull, morphology. A 2010 article in National Geographic took Christiansen and Harris's side.

    This is not a new idea. Brian Switek, who wrote the Wired article, documented that it went back to John Merriam and Chester Stock in 1932 and George Gaylord Simpson in 1941, who thought that Joseph Leidy, who named the species, was mistaken in identifying it as a lion. While this was not the dominant interpretation when I was a research associate at the Los Angeles County Museum of Natural History and specifically at the George C. Page Museum from 1985 to 1989, so I knew Harris personally and reported indirectly to him — P. atrox as a lion was — it was never rejected, either.

    By the time Switek wrote his article, the display specimen was relabled as "Naegele’s giant jaguar." Apparently, that one of their own curators thought it was more of a jaguar than a lion was enough to sway the people in charge of exhibits. According to a blog post from 2014 by Max Greene, so was the generosity of a donor to the museum, which is why it was named after Naegle. Both Switek and Greene, like you, think DNA trumps bones, so they fall on the lion side of the controversy, as do I.