One or two points call for comment. It will be seen that at the limits of each group there are indices which should numerically be placed in the next group. For example, the New Caledonians and the Papuans of the archipelago at the extreme south-east of British New Guinea have mesorhine instead of platyrhine indices; this is due directly or indirectly to a crossing with Polynesians. Of this we have direct evidence for the New Caledonians, and I have adduced evidence' for a migration from the Melanesian Archipelago into the south-east of New Guinea; but these Melanesians had already been subjected to Polynesian influence. The nasal index of the Polynesians is at the extreme lower end of mesorhiny. The Berbers of Biskra have doubtless had their nose broadened by Nigritic mixture, for the pure Berbers are among the most leptorhine of men. The Lombardians have an exceptionally high index for Europeans. By their exceptionally low index the Eskimo are sharply separated from the Mongolic and American races; this fact is in harmony with their very dolichocephalic cranium (index about 72) and long face. The exact relationship of these interesting people is not yet definitely established.

The nasal index of the primitive Andaman Islanders (50.9) and that of the nearly equally primitive Veddahs (52.5) are smaller than one might expect, as these races are very unspecialised groups of mankind. The new-born

1 A. C. Haddon, "The Decorative Art of British New Guinea," Cunningham Memoir X., Roy. Irish Acad., 1894.

French infant, according to Broca, has an index of 59.2, and is, consequently, about as platyrhine as the Bushmen of South Africa.

One word of warning is necessary: although the nasal indices in the living and in the skull agree very well as a whole, it must be distinctly understood that there is no necessary relation between them. It is impossible even to approximately calculate the one index from the other.

The learned Broca' has made a comparative study of the nasal index in the skulls of Egypt and France, and in both cases he has proved that the nasal index, despite numerous crossings, was perpetuated for centuries without important changes.

2

We need not enter into a discussion concerning the ethnical relationships of the earliest population of Egypt; it is generally admitted that they were fundamentally a branch of the great Mediterranean race, but there is no doubt that several ethnic elements entered into their composition. The two types to which I have already referred, and which Pruner-Bey first described as the fine and coarse types, are found in the subterranean galleries of Sakkara, which belong to the IVth Dynasty (about 4000 B.C.), and equally occur in the collections of skulls of the ancient Empire. Whatever may be the origin of the fine type, it is only the coarse type that participates in the characters of the Nigritic peoples of Nubia. This prehistoric mixture must have been already ancient at the beginning of the Pharaonic period, for the characters of the Egyptian race have since been maintained with a remarkable stability; whence we may conclude that the effects of this crossing had at that early period already

1 P. Broca, "Recherches sur l'indice nasal," Rev. d'Anthropologie, i., 1872, p. I.

? Pruner-Bey, "Recherches sur l'origine de l'ancienne race Égyptienne," Mem. Soc. d'Anth, i., 1861.

arrived at this condition of stability, and this could only be realised at the end of a great number of generations.

The nasal index of the skulls of the IVth Dynasty had a mean of 47.93, and in this figure one may recognise the influence of the platyrhine races of Nubia. Let us now see what happened to the nasal index in later times.

It is known that in the Pharaonic times the valley of the Nile was invaded in turn from the south, the east, and the west, and thus received influences from the Nubian or Ethiopian race, from the Syro-Arab or Semitic race, and from the Lybian or Berber race.

The Persian, Macedonian, and Roman conquests further introduced new Asiatic or European elements, and the Semitic element was restored more fully and persistently by the Arab conquest. In spite of all these mixtures, the nasal index of the existing Kopts does not differ perceptibly from that of the ancient Graves.

About the IVth Dynasty' (3998-3721 B.C.), before all these invasions, the mean nasal index was, as we have seen, 47.93; among the Kopts of the nineteenth century it is 47.15. Is this because all the people who were successively established in Egypt had the same nasal index? By no means; for the index of the victorious Ethiopians was greater, whilst that of the Syro-Arab peoples and of the Berber tribes was, on the contrary, very small. MM. Hamy and Broca obtained twenty-two skulls from an ancient interment in the Island of Elephantine, opposite to Assouan, at the foot of the First Cataract, near the border of Upper Egypt and Nubia; the nasal index was 55.17.

It is known that the VIth Dynasty (3503-3322 B.C.) had

1 The dates of the dynasties are those given by Flinders Petrie in his History of Egypt, 2d ed., i., 1895, p. 252.

* Broca says "VIth Dynasty," but according to Flinders Petrie (History of Egypt, i., p. 69), the Vth Dynasty (3721-3503 B.C.) was of Elephantine origin.

its origin in Elephantine. So much for the Ethiopians. As to the Syro-Arabs, there is no doubt they had a very small nasal index, since that of recent Arabs is only 45.57, and that of Syrians 45.87. There remain the Lybian and Berber peoples, who established themselves in the Delta during the XIXth Dynasty, and who later, under Psammetik, gave a dynasty to Egypt. It is admitted that these peoples have a remarkably low nasal index. Broca measured ten Kabyle (Berber) skulls that had an index of 44.28, and some Guanches, who belonged to the same race, had an index of 44.25.

At the XIth Dynasty (2821-2778 B.C.), after the Dynasty of Elephantine, which was the VIth, and which lasted for over two hundred years, the index rose to 48.43. Under the XVIIIth Dynasty (1587-1327 B.C.), which followed close the long domination of Syro-Arab pastoral kings, it was maintained at 48.77; under the dynasties that followed, until the Macedonian period, it descended to 47.28. These are insignificant oscillations, and quite as insignificant is the change which has followed the Arab conquest of the seventh century, as the existing Kopts have an index of 47.15. One sees then, that if the mixtures of races have been able to exercise a slight influence on the nasal index of the Egyptians, this influence has only been temporary.

MEAN NASAL INDEX OF EGYPTIANS AT VARIOUS PERIODS.

The persistence of the nasal index has not been less re

markable among the peoples who, from the most remote periods, have occupied the soil of France.

For the Mammoth Period, Broca has only two skulls from Eyzies, whose indices were 48.98 and 45.09. The mean index of these would be 47, but, as Broca points out, the true average index is the index of the means and not the mean of the indices; as the nasal heights in these cases were 24 and 22 (mean 23), and the breadths 51 and 49 (mean 50), the index of the means is 46.

The neolithic series is not homogeneous, but we may conclude that the peoples of the Polished-Stone Age, so far as is known, were leptorhine.

MEAN NASAL INDEX OF ANCIENT AND MODERN POPULATIONS OF FRANCE.

The low index remains constant through the Bronze Age. The fifteen Gaulish skulls of the Iron Age measured by Broca belonged to at least a century before the conquest by Cæsar. The Roman invasion did not modify the index.

Most of the Merovingian skulls came from a cemetery at Chelles. The interment belonged to somewhat different dates, but it is probable that most of the skulls belong to the second half of the seventh century. Eleven more ancient skulls, that were found in a lower layer, have a mean index of 51.52; the more recent series of forty-four