In reference to this change of population, Mr. Taylor in one place uses the significant phrase, “The hills contain the ethnological sweepings of the plains.” Very true; but the effect of this on the ethnological character of the population of the places is various, and in the application requires much caution. It is right, for instance, to say generally that the Celtic language has everywhere in Europe retreated from the plains into the mountainous districts; but the people often still remain where the language has retreated, as the examination of any directory in many a district of Scotland, where only English is now spoken, will largely show. In Greece, in the same way, many districts present only Greek and Sclavonic names of places, where the population, within recent memory, is certainly Albanian. Inquiries of this nature always require no less caution than learning; otherwise, as Mr. Skene observes, what might have been, properly conducted, an all-important element in fixing the ethnology of any country, becomes, in rash hands and with hot heads, a delusion and a snare.[1]

But the science of language, when wisely conducted, not only presents an interesting analogy to geological stratification; it sometimes goes further, and bears direct witness to important geological changes as conclusive as any evidence derived from the existing conformation of the earth’s crust. How this comes to pass may easily be shown by a few familiar examples. The words wold and weald originally meant wood and forest, as the Anglo-Saxon Dictionary and the living use of the German language—wald—alike declare; but the wolds at present known in Yorkshire, Gloucestershire, and other parts of England, are generally bare and treeless, and in bad weather very cheerless places indeed. If, then, “there is nothing arbitrary in language,” and all local names tell an historical tale, it is certain that, at the time when those names were imposed, these same sites were part of an immense forest. The geologist, when, in the far-stretching bogs east of Glencoe, and near Kinloch Ewe, and in many other places of Scotland, he calls attention to the fact of layers of gigantic trees lying now deeply embedded under the peat, adduces an argument with regard to the primitive vegetation of our part of the world not a whit more convincing. The same fact of a lost vegetation is revealed in not a few places of England which end in the old word hurst, signifying a forest. Again, there is a large family of places in and about the Harz Mountains, in Germany, ending in ode, as Osterode, Hasselrode, Werningerode, and so forth. Now most of these places, as specially Hasselrode, are now remarkably free from those leagues of leafy luxuriance that give such a marked character to the scenery of that mountain district. It is certain, however, that they were at one time in the centre of an immense forest; for the word rode, radically the same as our rid, and perhaps the Welsh rhydd, Gaelic reidh, simply means “to make clear” or “clean,” and teaches that the forest in that part had been cleared for human habitation.

Once more: it is a well-known fact in geology that the border limit between sea and land is constantly changing, the briny element in some cliffy places, as to the north of Hull, systematically undermining the land, and stealing away the farmer’s acreage inch by inch and foot by foot; while in other places, from the conjoint action of river deposits and tidal currents, large tracts of what was once a sea-bottom are added to the land. The geological proof of this is open often to the most superficial observer; but the philological proof, when you once hold the key of it, is no less patent. In the Danish language—which is a sort of half-way house between high German and English—the word oe signifies an island. This oe, in the shape of ay, ea, ey, or y, appears everywhere on the British coast, particularly in the West Highlands, as in Colonsay, Torosay, Oransay, and in Orkney; and if there be any locality near the sea wearing this termination, not now surrounded by water, the conclusion is quite certain, on philological grounds, that it once was so. Here the London man will at once think on Bermondsey and Chelsea, and he will think rightly; but he must not be hasty to draw Stepney under the conditions of the same category, for the EY in that word, if I am rightly informed, is a corruption from hithe, a well-known Anglo-Saxon and good old English term signifying a haven; and generally, in all questions of topographical etymology, there is a risk of error where the old spelling of the word is not confronted with the form which, by the attritions and abrasions of time, it may have assumed.

These observations, which at the request of the author of the following pages I have hastily set down, will be sufficient to indicate the spirit in which the study of topographical etymology ought to be pursued. Of course, I have no share in the praise which belongs to the successful execution of so laborious an investigation; neither, on the other hand, can blame be attached to me for such occasional slips as the most careful writer may make in a matter where to err is easy, and where conjecture has so long been in the habit of usurping the place of science. But I can bear the most honest witness to the large research, sound judgment, and conscientious accuracy of the author; and feel happy to have my name, in a subsidiary way, connected with a work which, I am convinced, will prove an important addition to the furniture of our popular schools.

College, Edinburgh,

February 1875.

LIST OF ABBREVIATIONS

A DICTIONARY OF PLACE-NAMES

A