If so, all Europe (the British Isles included) might have had as its aborigines a family older than the oldest members of the Indo-European stock; a family of which every member may now be extinct, or a family of which remains may still survive. Where are such remains to be sought? In two sorts of localities-- {140} 1. Parts _beyond_ the limits of the area occupied by the so-called Indo-Europeans. 2. Parts _within_ the limits of the so-called Indo-Europeans; but so fortified by nature as to have been the stronghold of a retiring population. What are the chief parts coming under the first of these conditions? _a._ The countries beyond the Indo-Europeans of the Scandinavian and Slavonic areas, _i.e._ the countries of the Laplanders and Finnlanders. _b._ The countries beyond the Indo-Europeans of the Iranian stock, _i.e._ the Dekkan, or the country of those natives of India (whatever they may be) whose languages are not derived from the Sanscrit. What are parts coming under the second of these conditions? _a._ The Basque districts of the Pyrenees, where the language represents that of the aborigines of Spain anterior to the conquest of the Roman. _b._ The Albanians.--Such the doctrine of the _continuity_ of an _ante_-Indo-European population, from Cape Comorin to Lapland, and from Lapland to the Pyrenees. There is _some_ philological evidence of this: whether there is _enough_ is another matter. This view, which on its _philological_ side has been taken up by Rask, Kayser, and the chief Scandinavian scholars, and which, whether right or wrong, is the idea of a bold and comprehensive mind, as well as a powerful instrument of criticism in the way of a provisional theory, has also been adopted on its _physiological_ side by the chief Scandinavian anatomists and palaeontologists--Retzius, Eschricht, Niilson, and others. Skulls differing in shape from the Celtic skulls of Gaul, and from the Gothic skulls of Germany and Scandinavia, have been found in considerable numbers; and generally in burial-places of an apparently greater antiquity than those which contain typical Celtic, or typical Gothic crania. Hence there is some _anatomical_ as well as philological evidence: whether there is enough is another question. * * * * * {141} PART III. SOUNDS, LETTERS, PRONUNCIATION, SPELLING. -------- CHAPTER I. GENERAL NATURE OF ARTICULATE SOUNDS. s. 196. To two points connected with the subject of the following Chapter, the attention of the reader is requested. I. In the comparison of sounds the ear is liable to be misled by the eye. The syllables _ka_ and _ga_ are similar syllables. The vowel is in each the same, and the consonant is but slightly different. Now the words _ka_ and _ga_ are more allied to each other than the words _ka_ and _ba_, _ka_ and _ta_, &c., because the consonantal sounds of _k_ and _g_ are more allied than the consonantal sounds of _k_ and _b_, _k_ and _t_.