the Aethiopic culture is patriarchal. Frobenius quotes Herodotus as showing that the former culture of North Africa was matriarchal, but thinks that pure matriarchy no longer exists either there or elsewhere. ("Reines Matriarchat als Kulturform gibt es aber auch auf andern Erdteilen nicht mehr.") In the Hamitic culture-cycle at the present time he finds only what he regards as remnants or survivals of the former matriarchate. The woman tends and milks the cattle, she builds the hut or shelter, and works leather. The man is the hunter and warrior. The woman chooses her husband; and rivals for the hand of a woman fight a duel to decide who is to have her. Virginity in a bride is insisted on, but married women have considerable sexual freedom. In what Frobenius regards as the patriarchal cycle, on the contrary, women are not permitted to have anything to do with the cattle, and leather is worked by the men. The custom of marriage by capture is found, and the widow belongs to the husband's relatives (levirate, etc.). Virginity is not esteemed. In all this it is difficult to see exactly what sort of a picture of the matriarchate Frobenius has in his mind. He would probably not be prepared to deny, however, that such institutions as matrilineal descent of the group, matrilineal inheritance of property, matrilineal succession to chieftainship and rank, and matrilocal marriage, belong rather to the matriarchate than to the patriarchate. But when we consider the distribution of these customs we find that they are associated with the area of his Aethiopic culture cycle, with tribes which are entirely or mainly agricultural. On the contrary the strongest development of patriarchal institutions is found amongst people who depend largely on their cattle, some of them being purely pastoral. On the basis of the theories of Frobenius it is difficult to see how we are to account for the fact that the nearest approach to matriarchy is found amongst people of the Aethiopic (patriarchal) cycle, while the most definitely patriarchal tribes are those in which cattle (which belong to the Hamitic matriarchal cycle) are the most important element in their culture. Some of the difficulties might be got over by supposing that there are two distinct cattle cultures in Africa, in one of which women tend and milk the cattle and often work in leather, while in the other everything that concerns the cattle is entirely confined to men. The former might then be regarded as corresponding to the Hamitic culture of Frobenius. But the latter would not correspond to his Aethiopic cycle. There is evidence that there are at least two distinct species of domestic cattle in Africa south of the Sahara, and it is generally supposed that both of these are Asiatic in origin, and that one came into Africa long after the other. They are both distinct from the Mauretanian ox which belongs to the region from which the Hamitic culture may be supposed to have spread. This does not solve our difficulties, however, for we still have the problem of the matriarchal agricultural people of the forest region. And again it might well be held that in Africa there are two agricultural cultures and not one. One of these belongs to the forest and depends on the banana and manioc. The other belongs to savannah or steppe regions and depends on grains (millet, sorghum, etc.). The scheme of Frobenius is too simple to account for all the facts of the distribution of culture in Africa, but this must inevitably be the failing of any such scheme. The history of culture has been a very long and very complex process, and any hypothetical reconstruction of it must therefore be complex also. It is probable that a complete theory would have to be just as complex as the facts that it sets out to explain. Any comparatively simple theory will have to leave a great deal unexplained. The archaeological stratum with which Frobenius connects his Hamitic culture consists of the paintings and engravings of prehistoric times that are found in northwest Africa. In the work entitled "Hadschra Maktuba" reproductions of these from photographs and drawings are now being published. The whole work is to consist of six parts, the first of which has made its appearance. The plates are excellent and reveal to us an art of which the finest specimens are as good as anything in prehistoric art except the very best of the European paintings. The art here represented would certainly seem to provide a link between the late palæolithic art of Europe and that of the Bushmen of South Africa. But it is possible that it may possess an even greater importance than this. Evidence is accumulating that early Egyptian culture was influenced from the west and it is quite possible that further investigation of the early cultures of Africa Minor may throw much light on the beginnings of civilization in the Nile Valley. Those who are interested in the study of prehistoric art certainly cannot afford to be without "Hadschra Maktuba." A. R. RADCLIFFE-BROWN The Bagesu and Other Tribes of the Uganda Protectorate. JOHN ROSCOE. Mr. Roscoe's volume is the third and last contribution of the Mackie Ethnological Expedition to Central Africa, the two earlier volumes dealing with the Bakitara or Banyoro and the Banyankole. The region covered lies between Mount Elgon and Lake Albert, and southwestward around the boundary of the Uganda Protectorate to the Kigezi country bordering on the new Belgian state, Ruanda. As this information was obtained only incidentally during his travels, these descriptions of eleven tribes are brief and somewhat random. Mr. Roscoe's record is as always reliable and a welcome addition toward a systematic survey of a new area. Several of the brief accounts include kinship terms and vocabularies. Occasionally Mr. Roscoe offers from his very full knowledge some general remarks on the distribution of traits (e.g., p. 51). We note that cultural similarities are always interpreted as evidence of migration. On page 53 a list of 126 exogamous clans in a tribe whose total population is not more than several hundred is given without comment. If Mr. Roscoe's descriptions are unfortunately meagre, there is abundant descriptive material in Mr. Rattray's book. It is modestly described as a first year's gleanings by a newly established department in Ashanti, but includes observations made over a longer period. It is only to be regretted that it is marred by verbosity and the tendency of the author to obtrude his personal reactions. The material deals almost wholly with social organization, rites, and the drum language. Each individual belongs to both matrilineal and patrilineal exogamous named groups. This is looked on as literal inheritance of the blood from the mother, the spirit from the father. Marriage is between cross-cousins. Kinship terms are given at length, illustrated by pedigrees. Land tenure is explained in terms of its parallelism with English feudal tenure. All land is owned and is secured by services and fees to the overlord. It is transmitted in the female line, is not individually owned, and cannot usually be alienated. Hence it is clan property, in which the ancestral spirits have an interest. Measurements of a small series of natives are considered by Mr. Dudley Buxton in an appendix. He observes that the relatively low standard deviations for stature and cephalic index (males, 5.53 and 2.47; females, 5.04 and 2.73) point to a homogeneous population. The sub-title of Mr. Willoughby's book indicates its purpose: "A study of the relations of Bantu and Britons in those parts of Bantu Africa which are under British control." The discussion of social problems is temperate enough, even if described as "the foreign life that Britain has to handle," yet one wonders whether Mr. Willoughby will follow this book with one for the Bantu advising him what to do with the British foreigner. All this is prefixed by a generalized, but fairly adequate, description of Bantu life and thought. Mr. Willoughby does not specify however which tribes. he writes about and the new material mentioned in the foreword is difficult to discover. LESLIE SPIER OCEANIA The Morioris of Chatham Islands. H. D. SKINNER. (Memoirs of the Bernice P. Bishop Museum, Vol. ix, No. 1. Honolulu, 1923. 140 pp., 35 figs., 35 pls., 1 map.) This monograph, by Mr. H. D. Skinner, Lecturer in Ethnology at the University of Otago, New Zealand, comprises two distinct parts dealing with the inhabitants and culture of the Chatham Islands, which lie some 536 miles east of Port Lyttelton, on the South Island of New Zealand. The first part in the main consists in a compilation of information relative to the Moriori migration, physical type, sociology, religion, and material culture culture, from accounts of early visitors that are not available to most students. The second part presents the results of a study of art and artifacts observed by the author in available collections in England and New Zealand. In addition there are numerous items of information gleaned by the author himself during a short stay on the islands in 1919. The paper as a whole is, of course, chiefly of interest to students of Polynesian culture, but in certain features it should also be of value to others interested in material culture in general, and particularly to those engaged in the study of adze-head forms. For Polynesian historians the value of the work lies in the fact that, while the Chatham group was of little importance in the history of the area as a whole, all the information that can be brought together concerning its inhabitants and their culture is of the greatest importance in the ethnographic problem in Polynesia by reason of the fact that in these islands were preserved certain features characteristic of an early group of Polynesian settlers in New Zealand. Mr. Skinner rejects the heretofore generally accepted theory as to the origin of the Moriori, based on Maori traditions, which considered the inhabitants of the Chatham Islands to be survivors of a fugitive remnant of a non-Polynesian people called the Muruiwi who are said to have been established in New Zealand before the Maori came. He points out that a "comparison of physical characteristics shows that the mythical (non-Polynesian) Muruiwi are the direct antithesis of the Moriori," and concludes by saying that on "the cultural side the evidence is no less convincing" (p. 18) as to the Maori Polynesian derivation of the Moriori. In summarizing the observations on physical characteristics Mr. Skinner finds that the Moriori were slightly shorter and broader than the Maori. Their hair was black except for a reddish tinge which occasionally appeared, and straight or slightly waved. Their skin color approximated that of the Maori though perhaps it was a shade darker; while the nose form in general was of the usual Polynesian type (with a moderately elevated bridge and broad nostrils), though there was a frequent appearance of the high aquiline or hooked nose which appears also among the Maori as well as other Polynesians and Melanesians. Characterizing him as "the greatest authority on Maori linguistics," Mr. Skinner quotes Archdeacon Herbert Williams who has written that the Moriori dialect is not correctly to be described as "a sub-dialect of New Zealand Maori" but "has as much right to be considered independent as any of the known dialects of the Polynesian language." The author of the work under review then proceeds with commendable perspicacity to point out a possibility overlooked by Archdeacon Williams, namely that the Moriori may have affinity with the little known Kai-Tahu dialect of the Canterbury and Otago Maori in the South Islands of New Zealand. This dialect, which differs widely from "dictionary or northern Maori," has never been redacted. This is an example of a type of consideration that may well |