Emporia, more probably the former. Now, as there is express evidence that a little to the west of Marseilles the Ligurian and Iberian areas met, the likelihood of the word belonging to the latter language is considerable.

It is increased by the circumstance of two-thirds, if not more, of the Keltic portion of Gaul being Iberian. Posidonius places the centre of the Keltic country in Provence, near the spot where the Roman settlement of Narbo was built: an Iberian locality. The Kelts of Herodotus are in the neighbourhood of the city called Pyrene; a word which carries us as far westward as the Pyrenees, although its meaning is different. As far as they extended beyond the present provinces of Roussillon and Languedoc, they extended westwards; beyond—according to Herodotus—the Pillars of Hercules, and as far as the frontier of the extreme Kynetæ. Aristotle knew the true meaning of the word Pyrene, i.e., that it denoted a range of mountains; and he also called Pyrene “a mountain of Keltica.” By the time of Cæsar, however, a great number of undoubted Gauls were included under the name Celtæ: in other words, the Iberian name for an Iberian population was first adopted by the Greeks as the name for all the inhabitants of south-western Gaul, and it was then extended by the Romans so as to include all the populations of Gallia except the Belgæ and Aquitanians. The word Celtæ also passed for a native name—“ipsorum lingua Celtæ, nostra Galli appellantur.” Upon this Prichard reasonably remarks, that Cæsar would have written more accurately had he stated that the people whom the Greeks called Κἑλται were Galli in the eyes of a Roman.

But the Greek form for Galli is Γἁλ-αται, a form suspiciously like Κἑλτ-αι. I admit that this engenders a difficulty, since it shows the possibility of the two words being the same. At the same time it can be explained. The ατ in Γἁλ-αται is non-radical. It is the sign of the plural number, as it is in Irish at the present moment; whereas the τ in Κελτ-αι is a part of the root.

And now I have given the additional reason for believing that the so-called Kelts of Spain were no Kelts at all in the modern sense of the word, but only Iberians; and I further suggest the likelihood of the word meaning mountaineer, or something like it, in which case the Kelts of South Gaul must be supposed to be (as they are made by Herodotus and Aristotle) the Pyrenean Iberians, the Celtiberi and Celtici being also the Highlanders of the great central range of Spain, of Gallicia, and of Alemtejo. This, however, is only a suggestion.

Perhaps the point is not very important. Whether we look to the amount of their civilization, to their national temper as shown in the defence of their independence, or to the extent to which they contributed to the literature of the Latin language, there are no very striking differences between the Gaul and the Iberian. Personal heroes like Viriathus and Vercingetorix occur on both sides; whilst Gaul resisted Cæsar by instances of endurance behind stone walls scarcely inferior to the display of obstinate valour at Numantia.

The Gothic conquest of Spain was succeeded, in the eighth century, by one of equal, perhaps, greater, importance. The line it took was from south to north; so that its direction was different from that of the Goths. It was also made by a southern population. The Arabs who effected the first invasion under Musa, were the Arabs of an army; i.e., almost wholly males; probably, too, they were pretty pure in blood. Afterwards, however, larger swarms came over from Africa; and it cannot be doubted that, along with these there were females and families of mixed African as well as of pure Arab descent. The areas which were successively appropriated by these invaders are not exactly those that we expect, à priori. Murcia, or the March, was less modified by the conquest than Valencia and other countries northwards. It was held in a sort of imperfect independence by Theodemir, and under the name of Tadmor, into which that of the Gothic king was metamorphosed by the Arabs, long continued to be the most Gothic part of south-eastern Spain.

In contrast to Grenada, and in consonance with what we expect from their geographical position, were the northern provinces of Asturias, Biscay, Navarre, and Galicia—Galicia, in respect to its ethnology, belonging almost as much to Portugal as to Spain. Into Asturias the arms of the Arab conqueror never penetrated: so that the original nationality was preserved in the kingdom of Oviedo, under the successors of Pelagius or Pelayo. Were these brave and independent mountaineers Goths or Romans? or were they original Iberians? And if of mixed blood, in what proportion were the different elements? They seem to have been second in purity of blood to the true and Proper Basques only. They were somewhat more Romanized than the latter, as is shown by their language; but both were equally free of Gothic admixture. This view rests partly on the previous details of their history, and partly on the names of the kings who succeeded Pelayo. They are not Gothic, like Euric, Wallia, or Roderic, nor yet Latin, like Pedro; but truly and properly Spanish (with the exception, perhaps, of Frivila), as Alonzo, Ordonio, Sancho, &c.; Spanish in the same way that Edward and Richard are German, or Arthur and Owen, Keltic. Pacheco, perhaps, is the truest Iberian designation. It occurs in Cæsar, as Paciecus. When the Arabs conquered Spain, their peculiar civilization was but partially developed. It grew up, to a great degree, within Spain itself.

The Arab elements belonged to the same class with the Phœnician, though to a different section of it. So did the Jewish, which were introduced earlier, and, if not of equal amount, were, at least, of longer duration. The Jews brought with them the oldest civilization in the world. But they were important physical influences as well. They came with their families, and, consequently, were less thrown upon the necessities of intermixture than the majority of the Arabs. The intermixture, however, was in both cases considerable. As long as the Arian kings of the Gothic stock held their sway, the Israelite was tolerated and something more. His industry was protected, and his earlier familiarity with letters and the civilizing influences of commerce respected. The prejudices against intermixture were chiefly on his side. Orthodoxy, however, introduced persecution. Some of its earliest enactments forbid Christian wives and Christian mistresses to Jews, a sure proof of the previous prevalence of an opposite custom. In the Mahometan parts of the Peninsula, the toleration was considerable throughout. Lastly must be noticed the great extent to which the pride in his real or supposed purity of blood characterizes the Hidalgo. This would not have been the case if purity of blood were the rule, and an Arab or Jewish cross the exception. The reign of Ferdinand and Isabella was signalized by the double ejection of the Jews from the Peninsula in general, and the Arabs from their last possession, the kingdom of Grenada. Such ejectments are never complete. Each, however, of these was one of remarkable magnitude.

The Normans, who settled on so many of the coasts of southern Europe, made a smaller impression on the Iberian peninsula than elsewhere. Still they must be recognised as an element.

Such is the basis of the Spanish stock, and such the chief superadded elements—Iberic in the first instance: then Phœnician, Greek, Roman, Gothic, Vandal, Alan (?), Jewish, Arab, and Norman, to say