to investigate people who have been subject to that training or who are in the process of training, thus ignoring the selective influence of the factor itself in original nature.  For instance, to study the value of high school training we compare those in training with those who have never had any; if the question is the value of manual training or Latin, again the comparison is made between those who have had it and those who haven’t.  To find out the influence of squalor and misery, people living in the slums are compared with those from a better district.  In each case the fact is ignored that the original natures of the two groups examined are different before the influence of the element in question was brought to bear.  Why do some children go to high school and others not?  Why do some choose classical courses and some manual training courses?  Why are some people found in the slums for generations?  The answer in each case is the same—­the original natures are different.  It isn’t the slums make the people nearly so often as it is the people make the slums.  It isn’t training in Latin that makes the more capable man, but the more intellectual students, because of tradition and possibly enjoyment of language study, choose the Latin.  It is unfair to measure a factor in the environment and give it credit or discredit for results, when those results are also due to original nature as well, which has not been allowed for.  It must be recognized by all those working in this field that, after all, man to some extent selects his own environment.  In the second place, it must be remembered that the environment will influence folks differently according as their natures are different.  There can be no doubt that environment is accountable for some individual differences, but just which ones and to what extent are questions to which at present the answers are unsatisfactory.

The investigations which have been carried on agree that environment is not so influential a cause for individual differences in intellect as is near ancestry.  One rather interesting line of evidence can be quoted as an illustration.  If individual differences in achievement are due largely to lack of training or to poor training, then to give the same amount and kind of training to all the individuals in a group should reduce the differences.  If such practice does not reduce the differences, then it is not reasonable to suppose that the differences were caused in the first place by differences in training.  As a matter of fact, equalizing training increases the differences.  The superior man becomes more superior, the inferior is left further behind than ever.  A common occurrence in school administration bears out this conclusion reached by experimental means.  The child who skips a grade is ready at the end of three years to skip again, and the child who fails a grade is likely at the end of three years to fail again.  Though environment seems of little influence as compared with