In a recent article in Science, Wayne and colleagues analyzed the sequence of mitochondrial genes and used these sequences to establish genetic resemblance among 67 dog breeds, 27 geographically distinct gray wolf populations, several jackals and several coyotes.  The study was undertaken in an attempt to clarify the relationship between these various canids (1).  I will here outline only the impact of these findings on the estimated age as well as the origin of some sighthounds and their relationship with other dog breeds.  The data show that superficial similarities or dissimilarities in overall appearance of breeds may not be meaningful at the genetic level which groups very different looking dog breeds.  In addition, the apparently so diverse looking various dog breeds are genetically no more different from each other than the similar looking gray wolf populations.

Dogs are domesticated wolfs.
       Dogs, wolfs and coyotes, although distinct natural species, have the same number of 78 chromosomes and can all interbreed and produce fertile offspring (2, 3).  Thus dogs could potentially be the result of a hybridization between wolf and coyote or jackal out of which humans selected some of the early domestic dogs.  And a rather wide variety of speculations how this may have happened have been produced over the years.  The new study provides clear evidence that all dogs tested are related to wolves and both wolves and dogs are more distantly related to coyotes and jackals.  Moreover, the analysis used only genes inherited from the mother.  This simply means that probably female wolves were at the origin of all the 400 domestic dogs currently recognized worldwide.  This analysis does not yet fully exclude the idea that a species closely related to the wolf could have been the father of the original ‘protodog’.  Like in human genetic analysis where we now know about a molecular Adam (based on the male specific y chromosome analysis) and a molecular Eve (based on mitochondrial gene analysis, only inherited from the mother), further studies on the male specific ‘y’ chromosome of dogs are needed to clarify the issue of the ancestral dog ’Adam’.

Domestication happened a long time ago.
       Recognizable dog bones buried with human bones were found about 14,000 years ago, not long before humans developed first settlement centers.  This has been taken by many as evidence that domestication of wolves may not have started much earlier.  However, wolf-like bones have been found together with human bones as early as 400,000 years ago (3) indicating a substantial delay between first association of wolves with humans and the earliest recognizable dog fossil.  Using the diversity of the analyzed genes, the new study suggests an age of about 135,000 years for the splitting of the wolf and the dog line.  This suggests that the earliest domestication of a wolf could have happened much earlier than previously believed.  However, the calibration of this ‘molecular clock’ could be off a bit so we may for the moment assume that domestication started about 50,000 to 100,000 years ago.  Transforming the wolf morphologically into a dog could have been a slow process at first and some of the wolf-like bones found in human fossil sites may have been actually bones of the ‘protodog’ or partially domesticated wolf. 
      Why is this important.  Obviously if the domestication of the wolf happened so early in human history, people were more interested in a hunting companion and guard dog for the camp and not so much in a herding dog as they had not yet reached an agricultural stage with the need for such a dog.  Given that sighthounds are hunting dogs, it stands to reason that they may be close to the ancestry of domestication and may have been hunting companions of humans for a long time (3).  Clearly, the oldest known pictures of some 6,000 years ago show all sighthound like dogs (e.g., Egyptian drawings).  Obviously there is a need for some time to transform a wolf-like ancestor into a recognizable sighthound.  This time may have been as long as 100,000 years according to the evidence provided by the data of Wayne et al (1).  The current evidence suggests that dogs, in particular hunting dogs (perhaps looking like sighthounds), may have been associated with man during the last 100,000 years or almost from the earliest days of modern humans.

Dogs form four genetically distinct groups.
       When the gene sequences are organized in a logical fashion, called a cladogram, four distinct groups of dogs become apparent.  For sighthounds, only two of these groups are important (Fig. 1).  One of the groups has no known matching wolf gene sequence whereas the other groups have dogs with genes close to or even identical to those of some local wolf populations.  Of the five sighthound dogs tested thus far, Afghan, basenji, greyhound, Irish wolfhound, and whippet, four belong to the cluster with no known wolf association.  Incidentally, this cluster also contains most of the 67 dog breeds tested.  However, while Basenji, Greyhound and Irish wolfhound grouped close together (and are closest related to the Dingo, Siberian husky and New Guinea singing dog), the whippet grouped with many other breeds, in particular with terriers.  It is noteworthy that the whippet genes showed identity with fox terriers.  This suggest that the alleged cross of the early whippet with a terrier described in many books about the whippet (4) was either with a fox terrier dam, or a terrier genetically identical to the fox terrier.  Some other terriers such as Norfolk, airedale and West Highland terrier are no possible ancestors of Whippets but many more terrier breeds such as the Bedlington terrier need to be tested before a firm scientific base to rule out other terriers as being related to whippets is established.
       The Whippet is not the only sighthound that shows a genetic identity with other dog breeds, suggestive of a mixed origin or some later cross breeding.  In fact, all sighthounds included in this test had some genes in common with other dog breeds.  For example, the same genes as in the Greyhound are also found in such diverse dogs as the St. Bernard, miniature schnauzer, Irish setter and Tibetan terrier.  Even a dog so fairly recently imported from its African continent like the basenji shows genes identical to some apparently distinct breeds such as Chinese crested and chow chow.  Interestingly, some evidence suggests that the Chinese crested originated in Africa (5).  The shared genes of basenji and Chinese crested would be in agreement with such an idea.  A more likely similarity, at least in size, is apparent in the Irish wolfhound.  This breed has some genes identical only to genes found in the Pyrenean mastiff.  While surprising at first glance, we know that the Irish wolfhound was almost extinct in the last century and was recreated from a few surviving specimens that were crossed with other breeds (5).  One of the other breeds may have brought in the genes probably common to mastiffs and now also to the Irish wolfhound.
       The most complicated case is the Afghan hound.  While apparently distinct in its appearance, it shares a genetic sequence with wolves from Romania and western Russia.  However, the identical gene is also found in a large number of other dog breeds such as basset, bulldog, German shepherd, otter hound, toy poodle, golden retriever and water spaniel.  What these data simply show is that at least once in the last several thousand years some backcrosses between wolfs and some dogs have occurred and/or between the ancestors of the other currently recognized breeds.  For the Afghan hound it may mean that the suspected crosses (6) between south Russian herding dogs (which may have been crossed with Russian wolves) may have brought in the genes now common between wolves and Afghan hounds. 
      While the Sighthounds breed now and for the last 100 years true to their standards, they still show the mixing done over the last few centuries lingering in their background as genes identical with those of some other breeds and even some wolf populations.  Obviously we need to wait until all of the 400 worldwide recognized breeds are analyzed before we can make sense out of the pattern and develop a feeling what was mixed with each other and possibly backcrossed to wolves in the past.  In the very least, these data should make all of us a little more humble to imply explicit or implicit a purebred decent from ancient sighthound breeds.  This in turn may bring some more level headedness into an otherwise heated discussion about alleged ancestry of our beloved sighthound breed(s).  Clearly, much like wolf populations from different localities can be genetically distinct so can be sighthounds from different geographical regions and comparing them by looks alone may be as misleading in sighthounds as it is in the gray wolf, which all look alike but are genetically as diverse as all the apparently more diverse dog breeds. 
       Last, but not least, the only species other than wolf/dog with this highly disperse and largely mixed genetic background is the human.  It stands to reason that this similarity is not so be accident but that humans have applied the same interbreeding strategy to their long time companions as they apparently used and still use for themselves.  Nevertheless, local populations of dogs, wolves and humans can be identified by distinct gene combinations and we should now be careful in our dog breeds not to eliminate these distinctions in an age of rapid worldwide transportation of dog and human and possibly even more rapid mixing.

This diagram shows on the left the relationship of wolf and dog populations as revealed by the genetic analysis of Wayne and associates.  On the right are the names of the dog breeds and their closest related wolf populations listed.  Note that among sighthounds, Afghan hounds are uniquely associated with wolfs.  Note also that three sighthound breeds are closely related but also each with other dog breeds.  The whippet alone among sighthounds shows affinity with terriers.
 

 References:

1) Vila, C., Savolainen, P., Maldonado, J.E., Amorim, I.R., Rice, J.E., Honeycutt, R.L., Crandall, K.A., Lundeberg, J. and Wayne, R.K.  (1997) Multiple and ancient origins of the domestic dog.  Science 276: 1687 - 1689.
2) Wayne, R.K. (1995) Phylogenetic relationships of canids to other carnivores. In: Anatomy of the dog. H.E. Evans, ed. Saunders, Philadelphia,  15-21.
3) Clutton-Brock, J. and Jewell, P. (1995) Origin and domestication of the dog. In: Anatomy of the dog. H.E. Evans, ed. Saunders, Philadelphia, pp 21-31.
4) Bengtson, B. (1994) The Whippet. MIP Publishing, Montecito.
5) Fogle, B. (1995) The Encyclopedia of the dog. Dorling Kindersley, New York.
6) Miller, C.O. and Gilbert E.M. (1975) The complete Afghan Hound.  Howell Book, New York