Parallel evolution in canids is why you can’t trust the fossil record

I have to admit that I am bit of a Joe Rogan Experience fan. I generally watch the podcasts that are about politics, hunting, and animals. I’m not really into pugilism stuff.

I actually came to Steven Rinella’s work through Joe Rogan’s program, and a few years ago, he mentioned something about Dan Flores and his work on the “American Serengeti,” which is actually a book by Flores that I have not read.  It is about the megafauna of the North American Great Plains, and it is a topic I’m somewhat interested in.

But as you know from reading this blog, I am a big coyote fan. I have had an experience with a male coyote in the woods, which I blogged about right after it happened I blogged about right after it happened, and I’ve written some more literary accounts of this encounter (but not for public consumption yet).

I then heard that Flores had a book about coyotes that came out, and I decided to read it.

And I didn’t like it.

I found that he adhered way too much to the paleontology of canids and pretty much ignored all the latest molecular data. At one point in the book, he makes the comparison that the genetic difference between a coyote and wolf is like the genetic difference between a human and an orangutan.  I think that assertion comes from an mtDNA study from 1993, which was the first to say that dogs were wolves and that “red wolves” had no unique mtDNA haplotypes.  It posited a 4% difference in the mtDNA sequence from wolves and coyotes, which is pretty accurate.  (Ironically, this study comes from Robert Wayne of UCLA, whom Flores largely discounts in his interview with Rogan at about 22 minutes.)

But mtDNA studies are notorious for leading people astray when we’re dealing with closely related species that can and do hybridize. For example, initial studies on mtDNA in European wolves found no evidence of dog hybridization, but because virtually all matings between dogs and wolves in the wild involve a male dog mating with a female wolf, the influence of dog genes in European wolves never could be accurately measured. Mitochondrial DNA is inherited only from the mother, and thus, it misses a lot of genetic information.

More recent full-genome analyses have revealed a greater than 99 percent genetic relationship among wolves, coyotes, and domestic dogs. That’s not at all equivalent to the genetic distance between humans and orangutans. In fact, we know that domestic dogs, coyotes, and wolves readily hybridize and produce fertile offspring, and no hybrids between a human and orangutan has ever been documented.

Flores pretty much rejects off-hand the more recent genome-wide studies that have found red and Eastern wolves to be hybrids with wolves and coyotes because Robert Wayne and his colleagues do not use morphological studies or pay much attention to the fossil record of canids in North America.  And the Fish and Wildlife Service adheres to the red and Eastern wolf paradigm.

I’m going to defend Wayne and his colleagues here.  You really need to be careful about morphological studies in canids.  That’s because canids can evolve quite rapidly, and there is a great tendency toward parallel evolution in the family.  I can remember when it was seriously discussed that the bush dogs of South America were a potential close relative of the dhole, based solely upon their “trenchant heel dentition.”  We now know that the bush dog is very much in the South American canid clade, probably a close relative to the maned wolf.  Until very recently, it was believed that the diminutive coyote-like golden jackals of Africa were the same species as the golden jackal of Eurasia, but a recent mtDNA study suggests a much great variance– enough to consider them separate species.  The similarities between the two forms of golden jackal likely resulted from parallel evolution.  The African “golden jackals” are actually much more closely related to wolves and coyotes, and the name “golden wolf” has been suggested for them.

This tendency to evolve rapidly is something we see in the domestic dog. Every single kennel club critic blog posts photos of dog breeds from different periods to show how much breeds change through selective breeding.  Nature selectively breeds, too, and dogs in the wild can rapidly change to fit new niches.

These issues are going to confound virtually every study on canid evolution.  This is one reason why we have nothing resembling a consensus on dog domestication. It is very hard to figure out when a sub-fossil wolf is a dog or is too much like a wolf to be a dog.

This is why I trust molecular studies far more than paleontology, and it is why I think the Fish and Wildlife Service is largely misguided in trying to hold onto the red wolf paradigm. It is possible that a recent wolf and coyote hybrid is going to look a lot like an ancient wolf-like canid, and the amount of convergence between the two can be enough to fit character-based analysis that paleontologists and anatomists use.

Also, the Fish and Wildlife Service is government, and in the US, government moves quite slowly.  I think it is going to take some time before the molecular data finally corrects these errors, but it doesn’t stop them from being errors.

The comparison of full genomes of wolves and coyotes that came out last summer pretty much ended this debate. Unless you’re going to argue over fossils, which is a dubious undertaking, I don’t think we can say that red wolves, Eastern wolves, or coyotes are what we thought they were.

Granted, Flores probably had the book at the publisher’s by the time this study came out, but the fact that he adheres to the old paradigm because Wayne and Wayne’s colleagues didn’t look at the fossils is pretty troubling.

If I were to rewrite Flores’s taxonomy, I would argue that coyotes have nothing to do with Canis edwardii.  That species was an early North American wolf that went extinct, and it could have been related to virtually any species in the genus Canis, including really divergent things like black-backed jackals.

The comparative genome study found that the most recent common ancestor of the wolf and coyote didn’t live 3.2 million years ago, as Flores asserts. Instead, it lived around 50,000 years ago, and it probably was living in Eurasia at the time. This animal was probably an archaic form of Canis lupus or maybe Canis mosbachensis.

When this animal crossed in North America, ancient North American wolves already dominated the landscape. There were also coyote-like forms of wolf, which likely weren’t coyotes at all.  The packing hunting wolf niche was already occupied by dire wolves and ancient North American dholes, so this radiation of the Eurasian wolf had to become more of a generalist to survive. The larger wolves, like the dire wolf, and the various forms of large predatory cat killed this ancestral coyote, and over time, it evolved into a smaller jackal-like canid.  This is how the coyote likely evolved the fission-fusion strategy of existence that Flores writes about. When the numbers are high, coyotes form stable packs and have relatively few young. They hunt mid-sized prey. When numbers are lower, they hunt rodents and lagomorphs, and female coyotes actually have a hormone change when the numbers are low and produce more ova during their estrus cycles. The females mate at 10 months instead of 22 months, and with more ova produced and more bitches breeding, the population can easily recover from a dire wolf or Smilodon attack. This is also why killing coyotes can actually force their numbers up, and it is one reason our intense persecution of coyotes has resulted in them spreading North, South, and to the East,

This is something that would have evolved in a mid-sized canid in the presence of many other large predators.  The fission-fusion strategy has just recently been confirmed in the Cape subspecies of black-backed jackal, which is another smaller canid that has evolved around large predators.

The Cape black-backed jackal is sort of the coyote of Southern Africa.  It is  generalist predator and scavenger, and it actually does cooperatively hunt small antelope species. It also kills sheep and goats.

It is not, however, closely related to wolves or coyotes. It is a very divergent form of Canis, which may actually be given its own genus (Lupelella) in the near future.  It has evolved coyote-like strategies for survival entirely in parallel with the coyote of North America.

This tendency toward parallel and convergent evolution in wild dog species is something that really messes up paleontology and morphological studies, and that is why the genome-wide studies are such compelling evidence. I’m dead-certain that many dinosaur specialists would love to have genomes from descendants of T. rex or the triceratops.

But those animals, like the ancient wolf- and coyote-like canids of North America, have left no descendants.

And what we likely have is a very diverse Holarctic wolf species that includes mid-size convergent jackals, massive megafauna-hunting wolves in taiga of Canada, the desert wolves of South Asia, and the all the weird domestic dogs that we have now.

That’s every bit as amazing as the older paradigm. Of course, I’m a bit of a rogue for suggesting that we include coyotes in the wolf species, but it seems to be right if we hare to adhere to cladistic classification.

This poor understanding of genetic studies actually ruined what could have been a great book on coyotes.

If you’ve ever looked into a coyote’s eyes, it is like looking into the eyes of a very bright dog.  They have so many dog-like mannerism that is hard not to see the similarity.

But you’re actually looking into the eyes of a super wolf.  This is the wolf that took all we could throw it at, and it thrived beyond our wildest expectations.

In Anthropocene, the meek do inherit the earth.


Natural History

Posted in Pet Health Articles | Tagged , , , , , , | Leave a comment

Leave a Reply

Your email address will not be published. Required fields are marked *

*

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>