number of observations is insufficient for applying this method. As stated before, the distribution of measurements is such that the parental types are more frequent than the average; for this reason the latter has no real biological significance. It must be
Fig. 9.—Breadth of Nose. Sioux.
considered merely as a convenient index of the general distribution.
Among the eastern Ojibwas I was able to make a classification into three groups: Indians, three-quarter bloods, and half bloods. In this case (Fig. 6) it will be noticed that the influence of the white admixture is very slight in the three-quarter bloods. The maximum frequency of the breadth of face remains at 150 millimetres, and we observe that a small increase in frequency takes place at 140 millimetres.
From the breadth of face I turn to the consideration of the height of face i. e., the distance from the chin to the suture between the nasal bones and the frontal bone (Fig. 8). This measurement is subject to considerable variations, on account of the difficulty of determining the initial points of the measurement with sufficient accuracy. This accounts for the irregularity
Fig. 10.—Breadth of Nose. Eastern Ojibwas.
of the curves. It appears clearly that the face of the half blood is shorter than that of the white. I am not able to say if this phenomenon is due to a general shortening, or if the nose, the jaw, or the teeth contribute most to this effect..The difference between full blood and half blood is much smaller than in the case of the breadth of face.
