PSYCHOLOGICAL SCIENCE 91 



hypothesis, he finds that they agree with observed 

 facts. The hypothesis is verified in one case. 



Next it is shown sufficient to explain also the 

 motion of the planets round the sun in ellipses. It 

 is verified in a second case. It proves able to account 

 for all the complicated perturbations of the moon's 

 orbit, and even to disclose the existence of a hitherto 

 unknown planet from its disturbing effects on others. 



Each new step in this process of induction gives the 

 mind a new feeling of security in further applications 

 of the theory. It becomes more and more probable 

 that the next test will again prove its power ; more 

 and more probable, indeed, that the theory is 

 of universal validity, that it represents a general 

 law of nature. Thus by induction we pass from 

 particular to particular and from particular to 

 general. 



Nevertheless, we never reach logical certainty ; 

 we can only approach certainty more nearly, as 

 instances of the successful application of the theory 

 multiply under our hands. There is no such thing 

 as certainty in natural science ; it is an affair of 

 probabilities. A radium atom may behave as a 

 permanent institution for a thousand years, and 

 in one surprising moment explode into fragments 

 at the last. 



Yet, in science no less than in practical life, we 

 are dependent on induction at every step. In 

 practice it is necessary to assume that because the 



