May 16, 1913] 



SCIENCE 



739 



view. He begins to look at science and the 

 whole body of knowledge with anointed eyes, 

 and presently the entire structure assumes a 

 new aspect. The student is to be congratu- 

 lated if this is accompanied by a revolution in 

 his own mind, in his ways of thinking. If 

 these vital inward changes do not take place 

 there is usually little reason for his continu- 

 ing in graduate work. 



All graduate study which properly deserves 

 the name involves research either directly or 

 indirectly. It consists of three parts which are 

 to be developed simultaneously, not succes- 

 sively: (1) One acquires the detailed and 

 specific knowledge needed for research; (2) 

 one develops the spirit of inquiry and conse- 

 cration to the task of extending the bounds of 

 knowledge — the spirit which characterizes the 

 man of research; (3) one is inducted into the 

 actual labor of discovery, and thus begins to 

 experience what is perhaps the profoundest 

 pleasure of which our nature is capable. 

 Graduate study which lacks any one of these 

 three elements is essentially deficient; it is 

 not taken into account in our discussion below. 



But what is research? What gives to it its 

 central place of importance? What are the 

 materials upon which it feeds? Let us first 

 answer the last question. 



The man of research should be free to choose 

 his material wheresoever he will. A directing 

 authority would ultimately be fatal to his vi- 

 tality and destructive of all useful labor. But 

 he must exercise an intelligent choice. Out of 

 the myriads of facts in the universe selection 

 must be made. Some are irrelevant; and 

 these should be discarded. To determine the 

 number of sprigs of grass on the campus or to 

 count the lady-bugs on our planet is not re- 

 search. These facts — though facts they are — 

 have no permanent character; they do not lead 

 anywhere. 



True research consists of any one or more 

 of three kinds of work of equal rank, as fol- 

 lows: 



1. Ascertaining new facts of a permanent 

 character or drawing attention to new rela- 

 tions among facts already known. This re- 



quires the power to direct attention to things 

 which other people have overlooked, to sepa- 

 rate them from the mass of facts in which 

 they are imbedded and to study them first for 

 their own sake and then in relation to other 

 things. The man of research requires the 

 power to see the mosquito on the monument 

 and for the moment to forget the monument 

 for the sake of the mosquito. It is so often 

 the trivial thing which turns out to be im- 

 portant. It is of more concern to us to know 

 the mosquito which holds the power of life 

 and death than to contemplate the battle com- 

 memorated by the monument. 



2. Deriving the consequences of facts al- 

 ready known. No fact is thoroughly under- 

 stood until all its consequences are brought 

 into review or the possibility of doing this has 

 been clearly and definitely recognized. Indeed 

 it is only when this has been done that we can 

 be said to have ascertained that the thing is a 

 fact. 



3. Developing a body of theoretical doctrine, 

 with or without reference to facts to be ac- 

 counted for by it. Under this head come such 

 matters as the Mendelian theory of inherit- 

 ance, the electron theory, the mathematical 

 theory of electricity, projective geometry. 



Granting now this definition of research 

 and its fundamental relation to graduate study 

 as outlined above, the question arises as to 

 when the student should begin the actual work 

 of research. Should it be in the first year? 

 Or, should one await a longer period of prepa- 

 ration in order to be better fitted for it? Prob- 

 ably no other subject requires as much prepa- 

 ration for research as mathematics, because in 

 this the whole body of doctrine is closely con- 

 nected and interdependent. Many extensive 

 parts of it can be learned in essentially only 

 one order. One may compare it to a tree. 

 The trunk corresponds to the fundamental 

 parts of the subject, the branches are the sub- 

 divisions, the remoter twigs are boundaries of 

 present knowledge, and it is here that new 

 truth is principally to be developed. Before 

 one is ready for research he must ascend the 

 trunk, so to speak, and climb out along some 



