PHYSICAL SCIENCE 25 



in individual experiments appeared : but, on the 

 average, his results showed that the distances tra- 

 versed from rest were proportional to the squares 

 of the times of fall. 



He had now verified one of the consequences of his 

 hypothesis. In this case it was enough, because, if 

 that consequence be assumed, the original hypothesis 

 may be shown to follow necessarily. Thus the original 

 hypothesis was verified also, and it was certain that, 

 as he supposed, the velocity of fall of his ball in- 

 creased with the time. Repetition of the experi- 

 ments has but served to confirm this result, and 

 gradually, by this process of induction, complete 

 conviction has been forced on us that it is a general 

 law of nature that the speed of falling bodies in- 

 creases in proportion to the time of fall. 



In one respect all physicists are not so lucky as 

 Galileo. It cannot always be shown that the original 

 hypothesis is the only one consistent with the deduc- 

 tion from it which is chosen for submission to an 

 experimental test. If not, all verifiable consequences 

 of the original hypothesis must be deduced, and as 

 many of them as possible examined experimentally. 

 With each concordance the evidence in favour of the 

 original hypothesis is increased, till eventually a very 

 high degree of probability may be obtained. 



Galileo's method, extended to cover this further 

 case, is that by which advance has been made in 

 physical science from his day to ours. 



