countrymen; wedded to their own form of Mohammedanism, and anathematizing all others.

They are the very Venetians as well as the Swiss of North Africa, travelling everywhere, penetrating from Timbuctoo to Asia Minor, serving in all sorts of capacities, connected with every caravan in Africa on the highway from its central and unknown regions to Morocco, Tunis, Algiers, and Egypt; possessing vast herds of camels, which are let out everywhere; with a free and republican form of government, highly artificial, but coloured in some respects by a theocracy; the young men nearly all abroad, but invariably returning with a competency in their old age to their poor and barren, yet cherished country. They are reserved and cold, but integrity characterizes their commerce, truthfulness their conversation, and morality their domestic life. In fact, as a French officer, who was expatiating on the contrast between them and the Arabs, once exclaimed to me, 66 They are the very Protestants of Mohammedanism."

CHAPTER XII.

HISTORY AND INSTITUTIONS OF THE M'ZAB.

History of the M'zab Herodotus - Greek and Roman traditions and reports Gætuli and Melano-Gætuli — Touareg Distinctness of language-Arab tradition - Jewish tale - Legends among the M'zab - Historical facts commence only with emigration from Oran — Temporary colony at Waregla-Berryan and Guerrara - Customs -Religious peculiarities-No dervishes - French interference Form of excommunication - Shame wiped out in blood - Poor-laws -Public works - Taxation - Industry - Obstinacy-Freemasonry. THE early history of the M'zab, as of the other sedentary inhabitants of the Sahara, is lost in the mists of tradition. Although they all preserve traditions of their early settlements, yet, in comparison with most of the races which now inhabit Europe and Northern Africa, they may be looked upon as aborigines. Herodotus (iv. 197) speaks of four races as inhabiting Africa: "Two of these are aboriginal, and two not. The Libyans and Ethiopians are aboriginal; the former inhabiting the north, the latter the south, of Libya. The Phoenicians and Greeks are foreign settlers." Among these races Herodotus could not include the Arab tribes, who now form the whole nomad population of Northern Africa, but who immigrated at a later period. The dwellers in cities on the northern coast, the Moors, settled in the country still more recently.

Pliny more exactly defines the Gætulians as those peoples who inhabited the country between Mauritania and the river Niger, placing the Ethiopians, or negroes, south of this (v. 4). His Gætulians then must be comprised under the Libyes of Herodotus.

Sallust (Jug. 181) distinguishes between the Gætuli and Libyes, placing the former in the southern portion of the Sahara; and, again, Ptolemy subdivides them into Gætuli and Melano-Gætuli-these latter probably being the mingled race of whites and negroes (iv. 6). If any of the aboriginal blood remains, it must evidently be sought among the stationary, not the nomad tribes of North Africa, excepting always the Touareg, whose wild habits and wilder domains have resisted all Arab aggression.

The Touareg may very possibly be the MelanoGætuli of Ptolemy; for though Lyon describes the Touareg of Fezzan to be nearly white, yet I conceive him to speak by comparison, as all those I have seen in the west, though answering in all other respects to his description, in being tall, handsome, and of dignified and independent mien, are certainly of a very dark brown, almost black, but without the slightest approach to the negro physiognomy.

The Libyans then of Herodotus, and the Northern Gætuli of Pliny and Ptolemy, must be sought among the Berbers or Kabyles, M'zab, Wareglans, and inhabitants of the Wed R'hir chain of oases.

These four speak distinct dialects, but all are branches of the same, or Berber family. The Touareg, whom Heeren affirms to speak the original Berber, I found to be quite unintelligible to Kabyles, who are able, without much difficulty, to interchange some ideas with Mozabites and Wareglans. On the affinities of these languages I do not venture to pronounce, but leave that to philologists: I can only state from experience, that the four vernacular dialects of the Sahara are more or less mutually intelligible; while Kabyles, Wareglans, and Mozabites alike, in conversing with

the Touaregs, always appeared to prefer the Arabic, a language foreign to both, or else made use of professed interpreters. The type of the Touaregs too is very distinct from that of any inhabitants of the oases. May we not, therefore, set down the Touareg as the Melano-Gætuli of Ptolemy and the Gætuli of Sallust; and the inhabitants of the oases as the Libyes of Herodotus and the later writers?

It is seldom easy to draw an accurate line between conterminous tribes; but habits, language, and latterly religion, seem to have preserved the Touareg from the slightest admixture with their northern neighbours. The locality of the Garamantes is so clearly defined to be south of Tripoli and the Syrtes, that they may be excluded from the consideration; but Dr. Shaw (1,143) holds the M'zabs, Wareglans, and Wed R'hir to be the representatives of the Melano-Gætuli, a supposition which I can scarcely conceive the Doctor would have maintained had he been personally acquainted with these nations.

The Kabyles, or Berbers, are now generally admitted to be the descendants of the ancient Numidians, driven by successive waves of conquest into their mountain fastnesses; and if we take the inhabitants of the oases, very different from them in physiognomy, as the representatives of the Libyes of Herodotus and the Gætuli of the Romans, we have all the interior races of Western Africa, as described by classic authors, satisfactorily accounted for. There is a marked difference, however, between the type of the M'zab and the other settled tribes. They are generally taller, the cheek-bone is more prominent, the lips not so thick, the eyes more closely set, and the nose larger, not so aquiline, and very broad at the tip. While it would be

difficult to distinguish between the inhabitants of other oases, different as they are from either Arab, Kabyle, or Touareg, a glance will suffice to point out unmistakably the son of M'zab.

It is the less difficult to draw a very distinct line between the aboriginal and the immigrant races of the Sahara, from the fact, that not only have their habits remained perfectly distinct, but also, that, while many Arabic expressions have been engrafted into the primitive languages, their structure has remained essentially Berber. It does not seem possible to maintain that any of those tribes which now speak any of the languages distinct from the Arabic can have had any considerable admixture of Arabian blood. The wandering habits of the nomads, and their constant intercourse with the mother country, have combined with the Koran to preserve the tongue tolerably pure, the Mogreb Arabic being, even in Western Morocco, at worst a corruption of Arabic. There are tribes claiming Arab descent who speak Berber dialects, but their language seems decisive in negativing their claim. It is scarcely conceivable that those who above all others pride themselves on speaking the tongue of the Prophet, should ever, while surrounded by their brethren, have relinquished it for the language of the conquered people; and this, independently of their physiognomy, appears an irresistible argument against any claim of the M'zab to be held as descendants of Ishmael.

Respecting their origin there are three traditions current. One, that of their Arab neighbours, is that they are a section of the Berber race of the Atlas, driven out of their country on account of their schism from the common faith of Islam, and who thereupon fled to the country they now inhabit, protected and