vir.] DISCUSSION OF THE DATA OF STATURE. 97

Statures of the Sons would be, if they were on the average identical with those of their Mid-Parents. Most obviously A B does not agree with C D ; therefore Sons do not, on the average, resemble their MidParents. On examining these lines more closely, it will be observed that A B cuts C D at a point M that fairly corresponds to the value of 68- inches, whether its value be read on the scale at the top or on that at the side. This is the value of P, the Mid-Stature of the population. Therefore it is only when the Parents are mediocre, that their Sons on the average resemble them.

Next draw a vertical line, E M F, through M, and let E C A be any horizontal line cutting ME at E, MC at E, and MA at A. Then it is obvious that the ratio of EA to EC is constant, whatever may be the position of E C A. This is true whether E C A be drawn above or like F D B, below M. In other words, the proportion between the Mid-Filial and the Mid-Parental deviation is constant, whatever the Mid-Parental stature may be. I reckon this ratio to be as 2 to 3 : that is to say, the Filial deviation from P is on the average only twothirds as wide as the Mid-Parental Deviation. I call this ratio of 2 to 3 the ratio of " Filial Regression." It is the proportion in which the Son is, on the average, less exceptional than his Mid-Parent.

My first estimate of the average proportion between the Mid-Filial and the Mid-Parental deviations, was made from a study of the MS. chart, and I then reckoned it as 3 to .5. The value given above was

if.