April 10, 1903.] 



SCIENCE. 



567 



certain limits and tells iis what this valid- 

 ity is. The sizes of the probable errors 

 increase rapidly as we go down the list, 

 thus proving some degree of approxima- 

 tion to the theoretical curve, based on the 

 assumption that scientific merit and emi- 

 nence are dependent on a large number of 

 small causes and giving iis the data for 

 the construction of the actual curve. 



I am certainly under great obligations 

 to those who assist me in making the ar- 

 rangements ; some find it interesting, others 

 irksome ; all show a certain amount of reli- 

 ance on my discretion. The individual 

 lists will of course be used only for the 

 averages and probable errors, and no record 

 is kept of those who make them. I could 

 doubtless give this address a succes de 

 scandale by reading to you the order of 

 merit so far as ascertained, but I have no 

 intention of making public the list until 

 such time has elapsed that each may as- 

 sume that if the process were repeated he 

 would stand at the head. But while the 

 list may not be published, it is possible to 

 draw from it certain deductions of scien- 

 tific and practical value. The statistics 

 of the whole number of scientific men have 

 greater interest when compared with those 

 of the more eminent thousand. We can 

 tell whether the average scientific standard 

 in one part of the country, at a given uni- 

 versity, etc., is higher or lower than else- 

 where ; we can give quantitatively, the men 

 being weighted, the scientific strength of 

 a university or department. It would be 

 possible to determine more exactly than by 

 existing methods who should be a fellow 

 of the American Association or a member 

 of the National Academy. It is possible 

 to correlate age, education and other fac- 

 tors with scientific eminence. 



The selection of the thousand scientific 

 men who are thought to have done the best 

 work, and their arrangement in the order 

 of merit, are somewhat incidental to my 



main object, which is to secure a group 

 sufficiently large and homogeneous for sci- 

 entific study and for comparison with other 

 groups. The problems that are opened and 

 may ultimately be solved are numerous and 

 not unimportant. The old question of the 

 relative contribution of heredity and en- 

 vironment to the making of the individual 

 must for men be solved by a study of men. 

 The infant is more plastic and his sur- 

 roundings are more varied than is the case 

 with other animals. We may find that the 

 mathematician must be born, whereas the 

 naturalist can be made by a sea voyage. 

 The little scientist can doubtless be made, 

 but probably the great man of science must 

 be born. We have to determine what con- 

 ditions of both nature and nurture are 

 favorable for the production of usefulness 

 and greatness in scientific work. We 

 should like to know at what age the future 

 of a man can be foretold with a given 

 degree of probability, at what age he has 

 his most original ideas, at what age he does 

 his most efficient work, at what age he is 

 likely to become a public nuisance. We 

 want to know what conditions of health, 

 habits, family, employment, rewards and 

 the like are favorable for scientific per- 

 formance. In general, we should like to 

 find out how we scientific people differ 

 among ourselves and from others, to make 

 a natural history of scientific men and to 

 use the knowledge for the improvement of 

 the breed. 



I have made a small beginning in the 

 direction of getting a scientific description 

 of men of science. At the present convo- 

 cation of scientific societies I have arranged 

 an anthropometric laboratory in which cer- 

 tain physical, psychophysical and mental 

 tests are being made. These will show how 

 in some fundamental traits scientific men 

 differ among themselves and from other 

 groups. 



Certain other traits I am attempting ta 



