SCIENTIFIC METHOD 69 



tion, and it is part of the scientific method to 

 make them and test them. While there seems 

 to be no doubt that some scientific conclusions 

 have arisen in the mind of the investigator as 

 if by a flash of insight, in the majority of cases 

 the process of discovery is a slower one. The 

 scientific imagination devises a possible solution 

 an hypothesis and the investigator proceeds 

 to test it. He makes intellectual keys and then 

 tries whether they fit the lock. If the hypothesis 

 does not fit, it is rejected and another is made. 

 The scientific workshop is full of discarded keys. 



It need hardly be said that whether the* hy- 

 pothesis is reached imaginatively or laboriously, 

 whether it is suggested by induction from many 

 particulars or as a deduction from some previously 

 established conclusion, it has to be tried and 

 tested until it rises to the rank of a theory. t 



TEST EXPERIMENTS AND CONTROL EXPERI- 

 MENTS. The distinction between observation 

 and experiment is not of much importance. In 

 the former we study the natural course of events; 

 in the latter we arrange artificially for certain 

 things to occur. The method of experiment 

 saves time and we can make surer of the condi- 

 tions. In studying the effect of electric discharges 

 on living plants, it would be worse than tedious 

 to wait for the lightning to strike trees in our 

 vicinity, so we mimic the natural phenomena in 



