THE AIM AND ACHIEVEMENTS OF SCIENTIFIC METHOD. 101 



and when it arises is seen instinctively to depend upon the 

 " pull " in the cord which is drawing the bodies together. Wk 

 seem to perceive, in fact, an alternative method of measuring . 

 the " force " in the cord ; namely, by the rate of transference of 

 momentum from one body to another which the cord in a given 

 state of elongation mediates. The soundness of this method is 

 capable of an indirect verification. Imagine a number of 

 precisely similar cords to be suspended vertically and to be 

 stretched, each to the same extent, by loads of different 

 materials attached to their lower ends. Then we are bound to 

 think of the elongations of the cords as being caused by equal 

 " forces," the equal weights of the suspended loads. If at the 

 same instant the loads are all set free and allowed to fall 

 (in vacuo}, we are bound once more to think that the circum- 

 stances of their motion are determined by the same " weight " 

 which caused them to elongate the cords to the same degree in 

 each case. If the forces in the stretched cords could be 

 measured unambiguously not only by the weights which 

 produce the given elongation but also by the rate of change of 

 momentum of a body drawn along freely by a similar cord 

 stretched to an equal degree ; then it would be reasonable to 

 expect that the different loads now supposed to be falling 

 freely would exhibit equal changes of momentum. But since 

 their weights are equal their masses are equal also. It remains 

 to be shewn, therefore, that the falling masses gain velocity at 

 the same rate irrespective of the material of which they are 

 composed. As is well known, there is abundant experimental 

 evidence that this is the case. 



Making, finally, an easy abstraction from the complications 

 introduced by the extended character (and consequent rotatory 

 movements) of the bodies in our experiments, we are now 

 prepared to hold with much confidence that when two 

 "Articles" are involved in a "dynamical transaction" the 

 rates of change of momentum along the straight line joining 

 them are at any moment equal and opposite, and that this rate 



