230 



The Journal of Heredity 



under how similar conditions they 

 existed; so that to Galton there seemed 

 "no escape from the conclusion that 

 nature prevails enormously over nur- 

 ture, when the diflfcrcnccs of nurture 

 do not exceed what is commonly to be 

 found among persons of the same rank 

 in society and in the same country." 



This kind of evidence was a good 

 start for eugenics, but as the science 

 grew, we outgrew such evidence. We 

 no longer wanted to be told, no matter 

 how minute the details, that "nature 

 prevails enormously over nurture. ' ' We 

 wanted to know exactly how much. We 

 refused to be satisfied with the state- 

 ment that a certain quantity was large; 

 we demanded that it be measured or 

 weighed. So Galton, Karl Pearson and 

 other mathematicians devised means of 

 doing this, and then Professor Edward 

 L. Thorndike of Colimibia University 

 took up Galton's problem again, with 

 the new methods, from whose conclu- 

 sions we think there can be no appeal. 



THE COEFFICIEXT OF CORRELATION 



The tool used by Professor Thorndike 

 was the coefficient of correlation, which 

 shows the amount of resemblance or 

 association between any two things 

 that are capable of measurement, and 

 is expressed in the form of a decimal 

 fraction somewhere between and the 

 unit 1. Zero shows that there is no 

 constant resemblance at all between the 

 two things concerned, that they are 

 wholly independent of each other, 

 while 1 shows that they are completely 

 dependent on each other — a condition 

 that rarely exists, of course. ^ For 

 instance, the correlation between the 

 right and left femur in man's legs is .98. 

 The nearer our fraction a]jproaches 

 unity, the greater is the resemblance 

 and the smaller it is, the less is the 

 resemblance; while a coefficient of .9 is, 

 of course, three times as great as a 

 coefficient of .3; and so on. 



Thorndike picked out in the New 

 York schools fifty pairs of twins about 

 the same age and measured the closeness 



^ What I say here refers to positive correlations, which are the only kind I cite in this paper. 

 Correlations may also he negative, lying between and — 1 : for instance, if we measured the 

 correlation between a man's lack of appetite and the time that had elapsed since his last meal, 

 we should have to express it by a negative fraction, the minus sign showing that the greater his 

 satiety, the less would be the time since his repast. 



of their resemblance in eight physical 

 characters, and also in six mental 

 characters, the latter being measured 

 by the proficiency with which the sub- 

 jects performed various tests. Then 

 children of the same age and sex, picked 

 at random from the same schools, were 

 measured in the same way. It was 

 thus possible to tell how much more alike 

 twins were than ordinary children in 

 the same environment. 



"If now these resemblances are due 

 to the fact that the two members of any 

 twin pair are treated alike at home, 

 have the same parental models, attend 

 the same school and are subject in 

 general to closely similar environmental 

 conditions, then (1) twins should, up 

 to the age of leaving home, grow more 

 and more alike, and in our measure- 

 ments the twins 13 and 14 years old 

 should be much more alike than those 

 9 and 10 years old. Again (2) if 

 similarity in training is the cause of 

 similarity in mental traits, ordinary 

 fraternal pairs not over four or five 

 years apart in age should show a resem- 

 blance somewhat nearly as great as 

 twin pairs, for the home and school 

 condition of a pair of the former will 

 not be much less similar than those of a 

 pair of the latter. Again, (3) if training 

 is the cause, twins should show greater 

 resemblance in the case of traits much 

 subject to training, such as ability in 

 addition or in multiplication, than in 

 traits less subject to training, such as 

 quickness in marking off the A's on a 

 sheet of printed capitals, or in writing 

 the opposites of words." 



thorndike's conclusions 



The data were elaborately analyzed 

 from many points of view. They 

 showed (1) that the twins 12-14 years 

 old were not any more alike than the 

 twins 9-11 years old. although they 

 ought to have been, if environment had 

 any power to mould the character during 

 these so-called "plastic years of child- 

 hood." They showed (2) that the 

 resemblance between twins was two or 



