” 1614 ” ” _s_│ 5 47│224│ 509│ 512│ 253│ 57│ 6│ │ 1613 ” 671 ” ” _t_│ 10 57│170│ 234│ 150│ 44│ 7│ │ │ 672 ” 180 ” ” _u_│ 10 33│ 61│ 52│ 20│ 4│ │ │ │ 180 ” 35 ” ” _v_│ 6 10│ 12│ 6│ 1│ │ │ │ │ 35 ────────────────────────────────────────────┼───────┼───┼────┼────┼────┼────┼───┼──────┼──────────── Total 10,000 Fathers (or _Mothers_) │ 34 168│655│1623│2522│2522│1623│655│168 34│ 10,004 │ ╰─┬─╯ │ │ │ │ │ │ │╰─┬─╯ │ ” 100 ” │ 2 │ 7│ 16│ 25│ 25│ 16│ 7│ 2 │ ────────────────────────────────────────────┴───────┴───┴────┴────┴────┴────┴───┴──────┴──────────── _Note._—The agreement in distribution between fathers (or _mothers_) and sons (or _daughters_) is exact to the nearest whole per centage. The slight discrepancy in the ten-thousandths is mainly due to the classes being too few and too wide; theoretically they should be extremely numerous and narrow. The result is shown to the nearest whole per thousand in the table up to “V and above,” to the nearest ten thousands. They may be read either as applying to fathers and their sons when adult, or to mothers and their daughters when adult, or, again, to parentages and filial couplets. I will not now attempt to explain the details of the calculation to those to whom these methods are new. Those who are familiar with them will easily understand the exact process from what follows. There are three points of reference in a scheme of descent which may be respectively named “mid-parental,” “genetic” and “filial” centres. In the present case of both parents being alike, the position of the mid-parental centre is identical with that of either parent separately. The position of the filial centre is that from which the children disperse. The genetic centre occupies the same position in the parental series that the filial centre does in the filial series. “Natural Inheritance” contains abundant proof, both observational and theoretical, that the genetic centre is not and cannot be identical with the parental centre, but is always more mediocre, owing to the combination of ancestral influences—which are generally mediocre—with the purely parental ones. It also shows that the regression from the parental to the genetic centre, in the case of stature at least, would amount to two-thirds under the conditions we are now supposing. The regression is indicated in the diagram used to illustrate this paper, by converging lines which are directed towards the same point below, but are stopped at one-third of the distance on the way to it. The contents of each parental class are supposed to be concentrated at the foot of the median axis of that