42 



Mr. F. Galton. 



[Jan. 21, 



the author explain in any way the difference between his own result 

 and that of Maxwell. 



[The method followed by Maxwell is liable to be vitiated to a very 

 sensible degree by small errors of level of the movable disks, especially 

 when they are closest to the fixed disks. The final adjustment is 

 stated to have been that of the fixed disks, and no special precautions 

 seem to have been taken to secure the exact horizontality of the 

 movable disks. By a calculation founded on the equations of motion 

 of a viscous fluid. I find that at the closest distance (about the one- 

 sixth of an inch) at which the fixed and movable disks were set, an 

 error of level of only 1° 8' would suffice to make the internal friction 

 appear 8 per cent, too high. 



In Mr. Tomlinson's reductions no allowance has at present been 

 made for the effect of the rotation of the spheres or cylinders about 

 their own axes, which is not quite insensible, as it would be in the 

 case of a ball pendulum. The introduction of a correction on this 

 account would slightly diminish the values resulting from the experi- 

 ments, especially in the case of the sphere, where it would come to 

 about 4 per cent. — G. Gr. S.] 



January 21, 1886. 



Professor STOKES, D.C.L., President, in the Chair. 



The presents received were laid on the table, and thanks ordered 

 for them. 



The following Papers were read : — 



I. "Family Likeness in Stature." By Francis Galton, 

 F.R.S. With an Appendix by J. D. Hamilton Dickson, 

 Fellow and Tutor of St. Peter's College, Cambridge. 

 Received January 1, 1886. 



I propose to express by formulae the relation that subsists between 

 the statures of specified men and those of their kinsmen in any given 

 degree, and to explain the processes through which family peculiarities 

 of stature gradually diminish, until in every remote degree of kinship 

 the group of kinsmen becomes undistinguishable from a group 

 selected out of the general population at random. I shall determine 

 the constants in my formulae referring to kinship with a useful 



