Tuesday, November 06, 2007

An Ancient Breed of Dog?




Genetic research has given legs to something most canine historians knew was true: most breeds are not very old, never mind what Kennel Club breed enthusiast may claim. The supposedly "ancient" Ibizan hound and Pharaoh hound, for example, turn out to have been made up within the last 100 years or so -- no doubt bred to look like the drawings and sculptures of sleek, slender-necked canines with pointy ears and long snouts that were seen on the Pharonic tombs whose disovery were all the rage at the time of Carter. As for the Norwegian elkhound, which supposedly dated back to Vikings dog, it turns out to have originated no farther back than the past few hundred years.

Using 96 distinct patterns in the genes called "microsatellites," genetic researchers compared dogs within breeds, and breeds with one another. In the May 21, 2004 issue of the journal Science, the team concluded that almost every breed was surprisingly distinct genetically. A few suprising gleanings:

  • Ancient dogs included a very motley assortment of dogs found all over the world: the Alaskan malamute, the Siberian husky, the Samoyed, basenji, Saluki, Afghan, Lhasa apso, Pekingese, Shar-Pei, Shih Tzu and Akita.

  • German shepherds, which might have been expected to be in the either the ancient group (due to their resemblance to wolves) or the herding group were found to actually be more closely related to to mastiff-type dogs, such as the bull mastiff, the bulldog and the Rottweiler.

  • Herding dogs, included such obvious members as the collie and the sheepdog, but also the greyhound.

  • Terriers and scent-tracking hounds, such as spaniels and setters and were deemed to be of relatively recent European origin. This should hardly surprise anyone -- almost all of the terrier "breeds" were created after 1800 and most after 1860 and the beginnings of the Kennel Club and livestock shows. For a detailed pictorial history of terriers, see A Pictorial History of Terriers.

For more information on canine genetic research, see the >> Genome News Network's Dog Page

.

6 comments:

Camera Trap Codger said...

Right interesting, Patrick. I must get that article, and wonder where the Indian pi dog fits in. It's not a breed per se, as much as a variety of mongrel, but it surely predated the British and picked up some new genes from colonial hunting dogs. BTW, there are some interesting descriptions of strange but lost native breeds on Indian dogs in old issues of the Bombay Natural History Society Journal. I am told that the huting dog of the Chin people in western Burma is distinctive, and they do look different from the run of the mill pi dogs you see there. There might still be some distinctive populations of dogs tucked away in Asia's remote valleys.

gabboon said...

nothing to do with your current article. I haven't even read it. Don't even know what its about. I'm just having a bad day and want nothing more than to go hunting without having anyone, except my hawk or dog, expect me to be any where at any time. Figured you and your readers would understand.

PBurns said...

The pye dog (or pi dog or pariah dog) is actually found all over the world and is actually the case statement for a simple point, which is that a dog is not a wolf. If dogs are left to breed randomly forever, you do not get wolves, you get pariah dogs. Dog and wolves are very different at the level of communication, sex, and demonstrations of social rank (among other things). More on that in a longer piece for another day, but I would argue that to the extent we have a living example of the ancient proto-dog alive today, it is the coyote-like pye dog found all over the world.

The oldest *definitive* record of dogs that I know of are out of Egypt, but it clear that they got basenji-like dogs from the south as well as molosser-like breeds from Asia. The greyhound-saluki-afghan-sight-hound dogs seem to have sprung from the Middle East before recorded history, and they are not too much changeg. Some, like the Tazi, have a single estrus like a wolf (most dogs have two).

P

Christopher said...

P -

Glad to see you posted about the study I e-mailed you about back in April. It's one of my favorites. I still get hate mail from an AKC hound woman from a yahoo ShowDog list that decided to go toe-to-toe with me over how ancient her breed was (hint, not that old).

Let's just say one of her defenses after I dropped that study on her was that DNA dating is like Carbon dating and can be contaminated ... something about a church fire and the Shroud of Turin. Unintentionally some of the funniest stuff I've read.

Any way, in case any of your readers are interested in having their working terriers DNA becoming part of the study, let me include a snippet from that e-mail:
===
Another interesting bit of news to pass along. The NHGRI Dog Genome project:

http://research.nhgri.nih.gov/dog_genome/

...is still collecting DNA samples from volunteer dogs within the AKC registry. Not a surprising choice given the sheer number of dogs within a single registry. This of course gives a skewed genetic perception of certain breeds that don't have a representative sample within the AKC, notably working dogs that have much greater numbers outside the AKC registry than within, or domestic breeds that have predominantly overseas breeding lines in the AKC. But hopefully the researches aren't ignorant of that fact.

In reading the research they have published (they outed the Ibizan and Pharaoh Hounds as Victorian concoctions and were able to cluster hundreds of breeds by genetic similarity...I have their actual published article if you'd like to read it) I predict that the bombshells will continue to drop from this research and I wanted to get involved.
-------------

Cheers,

C

PBurns said...

Actually, this post was written back in in 2004, and all I did was change the date :) I do that sometimes as there are something like 1,500 posts on this blog now, and a good number of them are recyclable. Plus, I had comments turned off for two years while the blogger people figured out how to keep spammers off, so now people can yell at me if they want.

For the record, the genome dating stuff may be flawed if wolves have been "back bred" into the gene pool as seems likely with certain kinds of wolf-looking dogs such as chows, huskies, samoyed, malamutes and akitas.

P

Steve Bodio said...

I have dogs with parents in the study.

Of course there is also the "breed" conundrum. Most of the really old breeds are "landraces" formed by both natural and human selection, and not closed studbook AKC breeds.

As an example, the study says Afghans "and" salukis. But Afghan, saluki, tazi, taigan are all just local variations of one big interbreeding population that reaches from North Africa to Mongolia (well, not modern show Afghans, but they are barely dogs!) I would breed my tazis to good native working specimens of any. For a look at variation in Afghanistan go here. Everything from smooth to shaggy...