but rather from the historical standpoint than as the manifestation of national psychology. The artist unfamiliar with the topic of early America will find the paper a mine of information on preColumbian American historical relations, if a little confusing since the mass of material is poorly organized and unsystematic. breathes the enthusiasm and vivid interest of the author. The Peruvian specialist also will find in it a tremendous fund of suggestion, more or less valuable. The article hardly pretends to be a properly developed scientific monograph, and the author does not attempt to prove any thesis, merely drawing on his great fund of knowledge of American archeology, and noting resemblances with every part of America on every possible topic. Some of the resemblances and analogies thus pointed out will not be accepted as valid by conservative archeologists, but they have their value as suggestions for consideration. They are too numerous to select and comment upon in particular. A large part of the paper, concerning the prehistoric relations of Mexico and Central America, while of great interest, seems hardly in place in a work on Peruvian Art.

The footnotes are many and full, but an unfortunate error in references destroys the correlation of footnotes 24 to 45, leading to the greatest confusion. The absence of references in text to plates and figures is also a lamentable omission.

Two tables of chronology complete Lehmann's text. Table A attempts a correlation of epochs, forms of thought, artistic styles and representative tribes and peoples. It is a valuable piece of work, one of the best of several such chronological tables which have been published by Peruvianists in the last few years. While accurate stratigraphical researches in Peru are still pitiably few and one of the great desiderata of American archeology, yet sufficient detailed studies have now been made to render the preparation of such chronological tables no longer premature but rather distinctly welcome as tentative hypotheses for discussion. The actual dates in terms of years given in the present table are well within possibility, but any assignment of actual dates on such slight evidence as Peruvian archeology offers at present is an attempt at which more conservative students have heretofore hesitated.

Table B is larger, fuller and more complex, a rather daring temporal and geographical cross-section of the entire field of LatinAmerican archeology from the Toltecs and Huaxtecs to the Calchaqui, involving considerations of language, artefacts, artistic styles,

1

migrations, traditions and mythology. While the many suggestive resemblances, culled from the author's unexcelled fund of knowledge of Latin-American archeology, are highly inspiring, the table as a whole suffers from over-complexity and would have been more impressive if the field of research had been more restricted. As before, the dates are reasonably conservative, and the table as a whole. probably as good and complete as the all-too-meager data permit.

"The Art of Old Peru" is a work primarily for artists and students of primitive art, but of maximum importance as well to archeologists, both because of its wealth of otherwise unpublished objects and of suggestive discussion.

J. ALDEN MASON

ASIA

The Panchatantra Reconstructed. F. EDGERTON. American Oriental Series, vols. 2 and 3. American Oriental Society: New Haven, Connecticut, 1924. 408 and 405 pp.

The purpose of this work is to reconstruct the parent work from which are derived those collections of tales known as the Tantrakhyayika, the Hitopadesa, the Kathāsaritsāgara, and the rest. The method is essentially the same as is pursued in collating manuscripts; there is this difference, however, that the derivatives do not strive to reproduce word for word the lost original; they retell and rearrange the original to suit the editor's taste. The author may follow the original fairly closely, or he may, like Somadeva, turn the whole into verse and retain only the substance. It is not therefore merely a question of eliminating clerical errors; but of determining what parts the existing tales reproduce verbally from the original, and what parts have been rewritten.

To this task Prof. Edgerton has brought a stupendous industry. He has compared every sentence of his ten derivatives, and further has had to collate the various manuscripts of the same derivative in order to establish his text. This minute analysis has not, however, impaired his judgment and he frankly admits that his reconstruction. is only approximate. How approximate is clearly shown by the use of italics and brackets. The italics "indicate matter of which we cannot be sure that it literally corresponds to the original text." When we consider the amount of italics, often whole pages together,

with a sprinkling of certainties, one cannot help wondering whether the result justifies the enormous effort. In scholarship we must keep the end steadily in view. The author does indeed seek to justify his labours by pleading that "it must be worth while to recreate the original form of a work that has enjoyed such enormous popularity." To recreate it with certainty, perhaps; but since the exact text is largely doubtful the reconstruction will never be used as an authority; it is not likely ever to be quoted in support of historical studies; the existing derivatives will always be referred to because they are books that were actually written. Even if a certain reconstruction were possible it would scarcely possess the interest that would attach to a reconstruction of the original Mahabharata or Ramayana. The author is prepared to admit that the original work was as late as the fifth century and not earlier than the Christian era. The interval separating it from its derivatives is not so great as to make a profound difference in the opinions and customs of the times. In point of fact we shall find that the variations bear on words and phrases rather than on ideas that interest the student of customs. This is not to be wondered at since animal fables are singularly poor in information about customs, and I doubt whether the student could cull from Aesop or La Fontaine anything which he could not gather in less time and more fully elsewhere.

The translation makes excellent reading; in fact it might be criticized for making better reading than the original. The author has aimed at lightening the style, but in so doing destroys one of the characteristics of the classical Sanskrit, its fondness for interminable compound adjectives; they express what the best ages of any literature would put into a separate sentence, if they did not omit it altogether. By making them into separate sentences the author gives the reader who does not know Sanskrit the impression that the original was much more natural and popular than it really was. For instance paragraph 2 on page 271 appears in the translation as four sentences; but in the original it is only one, the last three sentences of the translation representing as many compound adjectives, one of them consisting of twenty-seven syllables!

By an awkward convention in the text contracted vowels are shown by the same sign as an elision. Thus ceti, for ca iti, is written ce'ti, as if the first i were elided, whereas it has fused with a into e. Contraction could more suitably have been expressed by a circumflex.

The orthography is modernized: enuf for enough, huskt for husked,

thru for through, to all of which no exception can be taken, but one would expect thot to rhyme with not.

The type and general get-up are excellent.

ANURADHAPURA, Ceylon.

A. M. HOCART

Growth of Chinese. S. M. SHIROKOGOROFF and V. B. APPLETON. Reprinted from The China Medical Journal, May, 1924. 14 pp., 4 tables, 6 figs.

These observations on weight and stature of more than a thousand Chinese children of the Eastern Provinces are a distinct addition to the scanty data from China. Averages with maximum and minimum instances in each age class are given. It is of importance to have some other expression of variability as well, or the seriations for each age from which these might be calculated, but these are not included. I have calculated the following increments and rates of growth.

Weight Stature Weight Stature Weight Stature Weight Stature (Pounds) (Inches) (Pounds) (Inches) || (Percent) | (Percent) | (Percent) (Percent)

The values are quite irregular, but suggest that growth in weight is slight after the seventeenth year in males and after the sixteenth year in females, with a similar phenomenon in stature beginning a year earlier in both cases. I cannot agree with the authors that the ages 6 to 15 are "critical periods" for stature of males at which the rate of growth changes. The linear equations for the growth curves, and the maxima and minima for each age class calculated from these, which the authors add, do not give any insight into the phenomena. LESLIE SPIER

The Primitive Culture of India. Colonel T. C. HODSON. James Furlong Fund, vol. 1, pp. 1-133. London: The Royal Asiatic Society, 1922.

This book contains a series of five lectures delivered by Colonel T. C. Hodson at the School of Oriental Studies, University of London, in the year 1922. So small a book, dealing as it does with the cultural expression of millions of people spread over a territory of considerable extent and environmental variation, is bound to be a generalization. Almost every phase of material and immaterial culture is touched upon. Agricultural methods, metallurgy, pottery-making, wood and stone-working, fire-making, dress, dwellings, linguistics, myths, taboos, education, kinship systems, marriage regulations, ceremonies, cosmogonic beliefs, the life-cycle, all these and their related topics are mentioned. In describing these features the author refers to the practices of various tribal groups in Upper Burma, Northern and Southern India, in Ceylon and in the Andaman Islands.

At no time are we told just what tribes have a specific custom for the author is not interested in cultural distributions or relationships. Though he admits that there are gradations within the "lower culture" he makes no differentiation in cultural status between those tribes that practice terrace farming or jhumning and those that have no planting at all but live on products of the jungle, by hunting and fishing. Similarly as regards other elements of material culture, the people that use metals, those that have only stone or wooden implements, those wearing no clothing and those dressing elaborately, those erecting leaf shelters and those constructing plank houses, rambling forest dwellers and gregarious citizens, all fall in the stratum of civilization termed primitive.