1 86 CARNEGIE INSTITUTION OF WASHINGTON. 



It will be seen from the above results that I personall}' can not 

 form a very high expectation of the amount of results of first-class 

 value which would be obtainable by forming an institute to deal 

 with the existing masses of observations. 



(4) Nevertheless, if we reject 50 percent of existing observations 

 as worthless, if we frankly "scrap" them, I still think something 

 of service might be done with the remainder under certain conditions. 



(a) If the right man were available. This is the chief difficulty. 

 He must be a man of wide appreciation of many branches of science, 

 otherwise a special man will be wanted for each branch — astronomy, 

 meteorology, physics, medical science, sociology, etc. Even were 

 the money forthcoming for this multiplicity of workers, I doubt 

 whether the men themselves are to be found. If Professor New- 

 comb's institute is carried out, the right man for director will be a 

 man of very exceptional attainments, falling little short of scientific 

 genius. I doubt if one man of this type could be procured. It is 

 certain that .several could not. 



(^) The right man must have been rightly trained. He is to be 

 occupied in drawing logical conclusions from other persons' obser- 

 vations and data. He must therefore in the first place be an adept 

 in scientific method ; he must be a first-class mathematician, statis- 

 tician, and a trained calculator and computator. 



(c) The right man must be rightly supported. He must have a 

 competent staff of workers under him, and be to a considerable 

 extent a man of aifairs. He will have to reject after examination 

 whole masses of observations and data as unsuitable, and his pro- 

 ceedings will be questioned and criticised. Unless he is a man of 

 weight and tact, he will soon be in an impossible position relative 

 to the mediocre observers whose data he is to manipulate. For 

 example, he proposes to deal with the weights of the human viscera 

 in health and disease. He collects all the available data, but issues 

 his report and conclusions, silently passing by the measurements of 

 some well-known physician or hospital, because they have been 

 made in a manner which renders them of no real scientific value. 

 The result would be certainly controversy, po.ssibly uproar, and the 

 director of the in.stitute would have to fight a series of pitched bat- 

 tles before his reputation as a censor and official ' ' scrapper ' ' was 

 finally established beyond di.spute. He might survive this initial 

 state of affairs if he had the support of the best scientific minds in 

 the country ; but unless he was a strong man he would take the 

 easier course, and simply add another long series of reports on a// 



