But the maxima and minima show still more clearly how far this confusion would be carried, if individuals were compared. Hottentots and Australians, by their maxima of 83, would stand before Germans and Anglo-Americans, whose minimum is not so high. With much greater reason would they be placed in the midst of all the other races, which, by their means, are placed above them. This is not all. Between the highest and the lowest mean, between the English and Hottentots, or Australians, the difference in cranial capacity is only twenty-one cubic inches. The difference between the maximum and minimum of the Chinese is exactly the same. And it is much greater in nine other races, being more than double in the Germans and Peruvians.
Do we meet with facts like those resulting from the measurements of Morton in the species of a single genera of plants and animals? Certainly not; and this table is of itself sufficient to prove that the human groups are races, which have little uniformity owing to the absence of selection, and in no sense species.
III. Characters drawn from the face alone.—Similar conclusions to those furnished by the examination of the cranium are suggested by that of the entire face. It may be either broad or long; and in order to distinguish these two forms by special epithets, we may employ the terms euryopse, dolichopse (οψις, theatrical mask).
Since the face is much more irregular in form than the cranium, it gives rise to a far greater number of observations. Each one of its features would deserve our attention, were we writing a detailed work, and the more so, as such close study as this can only boast an existence of a few years. Failing space, I shall confine myself to pointing out the nature of the characters, and commenting upon some of the principal results.
In the living subject the length of the face is estimated from the commencement of the hair to the extremity of the chin. But measurements of this kind are difficult to procure when exotic races are in question. Skulls, therefore, have been examined. In the latter, the inferior maxillary bone is very often wanting, and even the teeth have, in too many cases, fallen out. The inferior limit of the length of the face could therefore be carried no further than the alveolar border of the superior maxillary bone. The point sus-nasal of M. Broca serves as the superior limit. The interval comprised within these limits is always less than the breadth measured across the zygomatic arches. In multiplying by 100 the length of the face and dividing it by the breadth, M. Broca has obtained the facial index. The following are some examples which I borrow from him with M. Topinard:
| Esquimaux | 73·4 | New Caledonians | 66·2 |
| Negroes | 68·6 | Parisians | 65·9 |
| Gallo-Bretons | 68·5 | Australians | 65·6 |
| Auvergnats | 67·9 | Tasmanians | 62·6 |
In spite of the small number of these examples, they might lead to remarks similar to those which I have already brought forward on several occasions, and which I believe it to be unnecessary to repeat.
The nose is one of the most striking features of the human face. Its general form and dimensions furnish some of the most special external characters in the distinction of races. But the morphological variations of this organ, presenting considerable difficulties, had long been neglected. M. Topinard filled this gap, and showed that it is possible, even upon casts, to take measurements suitable for indices. Nevertheless, it is the skull that, up to the present time, has contributed the clearest indications. The breadth of the nose taken at the opening of the nasal fossæ and multiplied by 100, compared with the length from the spine to the naso-frontal articulation, has furnished M. Broca with the terms of the relation expressed by his nasal index, the study of which has led him to important results.
Measurements, taken upon more than 1,200 skulls of all races, have enabled M. Broca to give 50·00 as the mean nasal index. In the entire number of races this index varies from 42·33 (Esquimaux) to 58·38 (Houzouanas). We see that the variation is only 16·05. The individual differences are much wider, extending from 72·22 (Houzouanas) to 35·71 (Roumanians), thus giving a maximum variation of 36·51.