Karl Pearson 
33 
Royal Instikition, Feb. 9, 1877. He had found it very difficult to collect human 
material for two generation.s and after careful consideration selected sweet pea seeds. 
These seeds were both measured and weighed and actually observations were taken 
on foliage and length of pod although as far as I am aware Galton never published 
the reductions of the latter. As he himself writes in 1885 : " It was anthropological 
evidence that I desired, caring only for the seeds as means of throwing light on 
heredity in man. I tried in vain for a long and weary time to obtain it in sufficient 
abundance, and my failure was a cogent motive, together with others, in inducing 
me to make an offer of prizes for family records, which was largely responded to, 
and furnished me last year with what I wanted*." 
The title of Galton's R. I. lecture was Typical Laws of Heredity in Man. Here 
for the first time appears a numerical measure r of what is termed 'reversion' and 
which Galton later termed ' regression.' This r is the source of our symbol for the 
correlation coefficient, which was really the first letter of ' reversion ' not of 
'regression.' The main results are given in a mathematical appendixf. Galton 
works with the modulus — i.e. our V2cr — probably because the tables of the prob- 
ability integral were then given in the modulus as argument. But we can at once 
convert into more customary notation. Thus we find the now familiar result 
W ■■= Vl — 
or, translating his symbols : 
Variability of family = Vl — x variability of general population. 
Galton had already reached the idea of homoscedasticity in the arrays of offspring. 
" I was certainly astonished to find the variability of the produce of the little seeds 
to be equal to that of the big ones ; but so it was and I thankfully accept the fact, 
for had it been otherwise, I cannot imagine, from theoretical considerations, how 
the typical problem could be solved " (p. 10). 
Next Galton supposes the mean taken of both parents and notes that the 
" variability of the parentage," what he would have called later the mid- 
parentage, = variability of either parent. He has not yet reached the idea of 
reducing one sex to the standard of the other, and the result is only true, if we have 
to deal with characters not sexually differentiated. 
Now we come to the test point |: 
" Reversion " — Galton tells us, p. 10 — " is the tendency of the ideal mean filial 
type to depart from the parental type, reverting to what may be roughly and 
perhaps fairly described as the average ancestral type. If family variability had 
been the only process in simple descent that affected the characteristics of a sample 
the dispersion of the race from its mean ideal type would indefinitely increase with 
* Address to Anthropological Section, 7). A. Ri'iwrt, 1885, p. 1207. 
t Boyal Institution of Great Britain, Friday, February 9, 1877. 
t Let the reader remember that these words were spoken just 40 years ago, and that they waited 12 
to bring forth fruit ! 
Biometrika xiii 3 
