IV. EXTERNAL CHARACTERS.—_Height._ All breeders regard height as characteristic of race among animals. It is also one of the traits which are most striking in man. This character sometimes shows that it is very evidently dependent upon the conditions of existence. Sheltering and feeding somewhat carefully the mares of La Camargue, has been sufficient to raise the height of this excellent breed of horses. With man, M. Durand (de Gros), confirming an observation already due to Ed. Lartet, has shown that, in the Aveyron, the populations of the limestone cantons are sensibly taller than those of the granite or schistose cantons. He agreed with the statement of Dr. Albespy, that liming lands in the non-calcareous portions of this district has raised the height by two, three or even four centimetres (·78, 1·17, 1·57 inch) on the lands where the practice has existed for the longest time. But on the other hand, it is indisputable, that races of very different height live side by side, without its being possible hitherto to point out the cause of this diversity. The dwarfed Negroes, the Akkas and Obongos, seem to be placed under conditions precisely similar to those under which much taller neighbouring tribes live. I have given above 163 statures of human races. I have insisted with sufficient strength upon the consequences which follow from them; from the point of view of the gradation and intercrossing of characters. But we can extract from these numbers some other results which are not less interesting. The general mean given by these numbers will be 1^m·635 (5ft 4·37 inch). I regard it as a little too great, the measures being wrong rather for the short races, than for the tall. Nevertheless it cannot be very far from the truth, and may be accepted provisionally. It is seen from the table that the Roumanians and the Magyars represent, from this point of view, exactly the mean stature. The oscillations of the mean statures above and below this general mean, extend in the case of the Patagonians to +0^m·115 (4·53 inch), and in the case of the Bosjesmans to -0^m·265 (10·43 inch). The individual oscillations are +0^m·295 (11·71 inch) for the inhabitants of Tongatabou, and -0^m·495 (1ft 7·49 inch) or -0^m·635 (2ft 1 inch) for the Bosjesmans. We see from the table that the oscillations below the general mean, are less numerous than those above it. This result may be connected with the fact which I have just pointed out. Nevertheless it appears to me probable that the number of races above the mean stature, is greater than those below it. The difference in number is compensated for by the more than double extent of the oscillations below the mean.