original work has now become scarce, and the second edition which now appears was entrusted to Mr. Crooke. Had Sir H. Risley lived, he would no doubt have revised the work thoroughly, but his lamented death in 191 1 soon after his retirement from India prevented all such plans, and Mr. Crooke, in the absence of any notes left by Sir H. Risley, has reprinted the text almost without alteration, simply bringing the statistics up to date in accordance with the Census of 191 1, and making a few obvious corrections. Some interesting additions have also been made to the plates, but the series is still far from representative of the People of India as a whole. Mr. Crooke has added an intro- duction, including a biographical sketch of Sir H. Risley and some remarks on controversial points which have become pro- minent since the publication of the first edition.
One of the most important of these questions is the influence of caste isolation on physical characteristics. This influence, which was perhaps unduly exaggerated at the time of the 1891 and 1901 Censuses, now tends to be treated with undue neglect, for, whatever may be the influences tending to miscegenation, those tending to isolation are still very strong. Another tendency is to neglect anthropometry as a test of race, and to trust rather to environment as the true cause of modifications. Here again great caution is needed in accepting any sweeping theories. The classi- fication of the races of India adopted by Sir H. Risley in the 1901 Census and in this work has also been the subject of much criticism, and the supposed extension of the Dravidian race in Northern and Central India can hardly now be maintained as advocated by Sir H. Risley. The theory of a Scythian origin for the broad-headed element among the Marathas in the Western Deccan can also hardly be maintained. If the far more extensive ' Scythian ' settlements of Northern India, still recognizable as tribal entities among Rajputs, Jats and Gujars, have left no influ- ence upon head-shape, it seems improbable that the small settle- ments in Western India should have produced such a result. Nor can it be regarded now as certain that all the invaders classed together as ' Scythian ' were of one race. The investigations of Sir A. Stein in the Takla-Makan Desert (analysed by Mr. T. A. Joyce in ^^ Journal of the Anthropological Institute in 1912) show