Social Environment and Eugenics 99 



puting the percentage of great men produced. 

 A state at the time of the census selected may 

 be in the first stage of a very rapid growth, 

 and its population may then be so relatively 

 small that the percentage of noted men com- 

 puted on it as a base will be disproportionately 

 large. It may be suggested that the noted men 

 should be sorted out according to the date of 

 birth, and handled statistically in several 

 groups, but this would be an almost impossible 

 task in such a mass of data. It has, however, 

 been attempted in part in connection with the 

 names from Who's Who in Science, where a 

 restriction to two decades has been made. In 

 the preparation of the data the question was 

 carefully worked out as to just what effect the 

 taking of a state at an earlier or a later stage 

 of its history would have upon the standing 

 of the state, but the conclusion was reached 

 that by two simple precautions all possible dan- 

 ger of serious error might be avoided. In the 

 first place, several different censuses might be 

 used, since an early census was found to favor 

 new states and a late one older states. In the 

 second place the study might be limited in the 

 main to states that have passed the first stages 

 of their growth, as, for example, those that are 



