PHYSICAL SCIENCE 17 



light bar be balanced over an edge, the weight at 

 the end of the shorter arm must be greater than 

 that at the end of the longer arm in the same pro- 

 portion as the greater length is to the shorter, or the 

 product of the weight and its distance from the point 

 of support is the same on each side. 



But, if his proof of this proposition be examined 

 it will be found that he assumes a property, that of 

 the centre of gravity, which is, in reality, equivalent 

 to the principle of the lever. Moreover, some of 

 his axioms can only be justified by experiment. We 

 may, in fact, just as well investigate the law of the 

 lever directly by the experimental method, and use it, 

 when established, as the basis of further deductions. 



Nevertheless, Archimedes' attempt to explain the 

 lever in terms of phenomena which he regarded as 

 better known, was an immense step in advance, and 

 is, in truth, the type of all scientific explanations, 

 which can consist only in expressing the less known 

 in terms of the better known. To the mind of 

 Archimedes and his contemporaries, his axioms 

 appeared better known than his conclusion, and his 

 proof was therefore to them a true explanation. 



To us, steeped in the experimental method, it 

 seems better to base the subject on the law of the 

 lever, a law which lends itself readily to experimental 

 verification, than on the principle of the centre of 

 gravity, which, in its general form, cannot be demon- 

 strated so easily by experiment, though it may 



B 



