COEVAL with primeval man apparently existed a type of dog equally primitive, though upon this head even palaeontology does not give us much assistance. None the less, fragments of bone unearthed from some prehistoric cave show us carved thereon some rude resemblances to the Canidae of those far-off days. At that time, too, as now, we are led to suppose that the dogs were more or less associated with man from the fact that the remains of both have been found together. Then such an animal was a necessity of the time: now it is largely a luxury, yet. far more deserving of the encomium bestowed upon it by Cuvier than it was when he described it as "the completest, the most singular, and the most useful conquest ever made by man."

Geologically considered, the dog does not belong to a very remote past, despite the fact that, in the opinion of experts, it was contemporaneous with men of the Flint Age, some 30,000 years B.C. Historically considered, it is of course far nearer to our own times, the earliest authentic records not dating back more than five thousand years. Doubtless, then the animal was used by its owner as a means not only of providing food, but also of acting as a defence against the existent wild animals. From that period, therefore, if not from an earlier one, we are justified in assuming that the dog was domesticated, or at least semi-domesticated.

How the Domestic dog (Cam's familiaris) first originated has puzzled some of the greatest naturalists of our own and other times. It is not proposed here to attempt to cut the Gordian knot with regard to its fons et origo. To do so would be to assume a knowledge and a power that the writer cannot claim. What is intended is, as far as possible, to give the place the dog has occupied in history, and endeavour to connect the past with the present.

Still, before doing so, there will be no harm in taking a cursory glance at some of the theories of the more practical naturalists with regard to the origin of the Domestic dog. The evidences of the existence of prehistoric dogs, as already suggested, are of the scantiest, and often of the rudest, and it is by reason of this fact that so much in connection with primeval dogs is left to absolute conjecture. However, so wonderfully polymorphic is the Domestic dog as met with to-day, that one is constantly beset with the thought as to its origin. It must be confessed that it is difficult to believe that such an atom of dog flesh as the Chihuahua dog, the Japanese Spaniel, the Pug, the Greyhound, the Spaniel, and the huge-framed St. Bernard all sprang from one species of wild true dog. Yet, as Mr. St. John Mivart, in his "Monograph of the Canidae," suggests, it is possible, and we certainly see no reason to doubt it. Some naturalists incline to the theory that certain varieties of the Domestic dog, showing a good deal in common, sprang from different species possessing such characteristics, and that these were simply awaiting development at the hands of man. Against this must be placed, as Mr. St. John Mivart points out, the fact that no such races exist in Nature. "They can hardly all have existed," he says, "and become extinct, for two reasons: first, palaeontology affords us no evidence that such has been the case; secondly, . . . the dog family is not one the species of which tend readily to disappear, as is shown by the long, persistent efforts needed to exterminate the wolf even in the most civilised parts of the habitable globe. Therefore the Domesticated dog cannot well be the product or a variety of wild true dog once widely diffused, but now entirely extinct. . . . That the various breeds known to us may nevertheless have originated from one form must be admitted to be possible, when we consider the changes that have taken place in old breeds, and the new forms that have been called forth in the historical period."

Of those naturalists who assert that more than one species have contributed the elements that have resulted in the production of the Domestic dog, Darwin is one. He intimates that all Domestic dogs are descended from two species of wolf - Canis lupus and Cants latrans. These were both savage when hunting gregariously, but were readily amenable to man's influence when dealt with singly. Certain it is that there are many characteristics common to both the wild and the domesticated Canidoe of to-day. The Australian Dingo, that pest of the sheep and the stock farmer, is as savage as any of its more ancestral prototypes when hunting, as is its wont, in flocks. Yet that it is capable of at least semi-domestication has been demonstrated beyond the shadow of a doubt. Again, students of dog-form will readily see the very close resemblance that there is between the wolves and many present-day varieties of the Domestic dog. Of this no better example can be cited than the Esquimaux dog, described and illustrated elsewhere in this volume. Between this dog and the Grey Arctic Wolves there is a very great resemblance - so great a resemblance, in fact, that a pack of the dogs were once mistaken for wolves by the well-known Arctic traveller Sir John Richardson. Yet another example may be given: the dogs of the Hare Indians differ but very slightly from the Prairie Wolf, or Coyote (C. latrans), one of the species that Darwin suggests as a remote ancestor of the Domestic dog.

As bearing upon the subject, the remarks of that eminently practical naturalist Mr. A. D. Bartlett (the late Superintendent of the Zoological Society's Gardens), as given in the Zoological Society's Proceedings for 1890, are most interesting. He says: "The extraordinary and wonderful number of well-marked breeds of the Domesticated dog, and their variations of size, form, and colour, render any attempt to account for their origin a task of some difficulty; but as many wild dogs appear to be descendants of Domestic dogs, it is necessary to endeavour to account for the origin of the Domestic race. There can be no doubt that the Esquimaux dogs are reclaimed or domesticated wolves. All wolves, if taken young and reared by man, are tame, playful, and exhibit a fondness for those who feed and attend to them. The same may be said of all the species of jackals. This being so, it is highly probable that both wolves and jackals were for many ages found in the company of man, and that, owing to this association, the different species of these animals may have bred together and become mixed. A mixed breed would at once develop a new variety. A variety once commenced would in all probability in a few generations undergo many changes, especially if any well-marked variety should occur. Nothing would be more natural than to suppose that the owners of this variety would endeavour to increase its number, especially if it were found to possess useful qualities.