Correlation and Application of Statistics to Problems of Heredity 31 



Galton was wont to illustrate the beauty of the "pre-ordained niche" on 

 a marshalled series of bean pods which he had many years before prepared. 

 This series is reproduced on Plate II. Unfortunately the tips of some of the 

 pods have bent back, but the general scheme survives. 



(ii) The Phenomenon of Regression, a great Hindrance to the 

 Establishment of Breeds (pp. 495-6). 



" It will be seen from the large values of the ratios of regression how speedily all peculiarities 

 that are possessed by any single individual to an exceptional extent, and which blend freely 

 together with those of his or her spouse, tend to disappear. A breed of exceptional animals, 

 rigorously selected, and carefully isolated from admixture with others of the same race, would 

 become shattered by even a brief period of opportunity to marry freely. It is only those breeds 

 that blend imperfectly with others and especially such of these as are at the same time prepotent, 

 in the sense of being more frequently transmitted than their competitors, that seem to have a 

 chance of maintaining themselves when marriages are not rigorously controlled — as indeed they 

 never are, except by professional breeders. It is on these grounds that I hail the appearance 

 of any new and valuable type as a fortunate and most necessary occurrence in the forward pro- 

 gress of evolution." 



Galton admits that the precise manner in which a new type comes into 

 existence is unknown, but suggests that a multitude of petty causes may 

 contribute to reshape the grouping of the germinal elements and so lead to 

 a new and fairly stable position of equilibrium, which admits of hereditary 

 transmission. In favour of this view he cites the frequent experience of 

 "sports," useful, harmful and indifferent and therefore without ideological 

 intent. These, he considers, have various degrees of heritable stability, and 

 form fresh centres towards which some at least of the offspring have a 

 tendency to revert. He considers that such sports, by refusing to blend 

 freely, may be transmitted almost in their entirety. 



"On the other hand, if the peculiarity blends easily, and if it was exceptional in magnitude, 

 the chance of inheriting it to its full extent would be extremely small...*. I feel the greatest 

 difficulty in accounting for the establishment of a new breed in a state of freedom by slight 

 and uncertain selective influences, unless there has been one or more abrupt changes of type, 

 many of them perhaps very small, but leading firmly step by step, though it may be along a 

 devious track, to the new form." 



* Galton gives in a footnote the percentage of sons who are as tall or taller than their fathers. 

 I have recalculated this table on somewhat better data than Galton had available {Biometrika, 

 Vol. ii, p. 381). It now runs as follows: 



The considerable changes from Galton's percentages arise from the facts : (i) that the sons in 

 our data had a mean stature 1" greater than their father's, (ii) that our regression was -516 

 against Galton's -333. 



