THE METHODS OF DISCOVERY ARE CONCRETE. 457 



additional new discoveries by our making it the basis of 

 new hypotheses, experiments, and observations ; and the, 

 additional knowledge thus obtained may, by being in its 

 turn subjected to similar mental processes of comparison, 

 and inference, yield a still further amount of new truths ; 

 and so on repeatedly to an extent which appears to be 

 limited only by the laws of nature and the state of advance- 

 ment of other branches of knowledge at the time. 



The actual methods of making scientific discoveries 

 are nearly always concrete. The most usual method is by 

 studying a subject, then experimentally investigating it, 

 and drawing such conclusions as the results afford. During 

 such study new hypotheses or questions of nature are ima- 

 gined, means of testing those questionings are next invented, 

 and the requisite experiments and observations are then 

 made. Having made those experiments and observations, 

 the results of them are studied, classified, and compared 

 in every conceivable way, and as many conclusions drawn 

 from them as they will logically afford. From these con- 

 clusions we proceed to infer such explanation, cause, rela- 

 tion, or general lav or principle as they appear to warrant, 

 and test by further experiments if the supposed expla- 

 nation, cause, &c., are the true ones. In actual research 

 all these processes are continually alternating, i.e. we do 

 not conclude our study before we make any experiments, 

 nor make all the experiments before we draw any con- 

 clusions or suggest new hypotheses ; but we raise an hypo- 

 thesis, then make one or more experiments, then draw 

 conclusions and raise new hypotheses, make more experi- 

 ments, and so on. Most researches are made by a series of 

 methods, and each research is really a compound result of 

 a series of discoveries, each evolved by each succeeding 

 method. In accordance with this, we already possess, in 

 every orderly treatise on chemical analysis, an outline of a 



