November U, 1910] 



SCIENCE 



683 



the salaries increase with distinction and 

 roughly measure it, placing it about three 

 times as high in the upper hundred as in the 

 lower third of the list. It is also the case 

 that the range of merit in the curve of dis- 

 tribution covered by the first hundred is al- 

 most exactly equal to the range covered by the 

 second and third hundreds, and each of these 

 is equal to the range covered by the remain- 

 ing seven hundred." It may not be possible 

 to fix a zero point at which scientific merit 

 begins, but it can plausibly be placed at a 

 point below the first thousand, about equal to 

 the range of merit covered by the other three 

 groups. In this case the merit of those 

 toward the bottom of each of the three groups 

 in the thousand — the first hundred, the second 

 and third hundreds, and the last seven hun- 

 dred—would be as 3:2:1. 



In order, therefore, to sum up in one figure 

 the strength of a university or department, 

 weights have been assigned to the men on this 

 basis — a man in the lower four hundred 

 being the unit, those in the other hundreds 

 were assigned ratings as follows : VII. and 

 VI. = 1.2; V. = 1.4; IV. = 1.6; III. = 1.9; 

 n. =: 2.2 ; and I. = 3. The first hundred were 

 subdivided, the lower fifty being assigned 2.5, 

 and the upper twenty-fives, respectively, 3 and 

 4. These ratings scarcely measure the real 

 value of the men to society; they are nearly 

 all paid less than they are worth, and the 

 greater the performance of a man the more 

 out of proportion is the payment for his ser- 

 vices. They do, however, give with tolerable 

 accuracy the value attached to men in our 

 competitive system. A university can obtain 

 a man of the first rank for from $5,000 to 

 $7,500, or a man in the lower hundreds of the 

 list for from $2,000 to $2,500. It is further 

 the case that a moderate alteration in the 



to mathematical physics, though he might have 

 earned large sums in the time devoted to these. 

 His technical work was doubtless worth far more 

 to society than he was paid for it, but it was 

 worth less than his scientific research. In his 

 three lines of work he was paid inversely as the 

 value of his services. 



"Cf. Science, N. S., 24, p. 707. moG. 



weights adopted would not considerably alter 

 the comparative results. 



The scientific strength of our strongest in- 

 stitutions rated in the manner described, to- 

 gether with the gain or loss in a period of 

 four years is shown in Table XI. Thus Har- 

 vard has a total scientific strength equivalent 

 to 146 men in the lower part of the thousand 

 and has made a gain equivalent to 16.3 such 

 men in the course of about four years. In 

 general the figures in this table correspond 

 with those in the preceding table, but they 

 tell us more. They take account not only of 

 the number of men gained or lost, but also of 

 the rank of these men and of the changes 

 which have taken place through men improv- 

 ing their standing or failing to maintain it. 



TABLE XI. THE SCIENTIFIC STEENGTH OF THE 

 LEADINQ INSTITUTIONS 



Harvard 



Chicago 



Columbia 



Hopkins 



Yale 



Cornell 



Wisconsin 



Geol. Survey .... 



Dept. Agric 



Mass. Inst 



Michigan 



California 



Carnegie Inst . . . 



Stanford 



Princeton 



Smithsonian Inst. 



Illinois 



Pennsylvania. . . . 

 Bur. of Standards 

 Clark 



If only the number of men is considered, 

 Columbia and Chicago are equal and Harvard 

 has made a larger gain than Chicago within 

 the past four years. But Chicago has in- 

 creased the number of men in the first hun- 

 dred by two and in the second hundred by five. 

 When we count up the total scientific 

 strength, we find that Chicago is in advance 

 of Columbia by the equivalent of 15.3 men 

 and has gained more than Harvard by the 

 equivalent of 1.7 men. Wisconsin and Illi- 



