KINETIC OR MECHANICAL VIEW OF NATCRE, 59 



mathematically what is nifaiit l>v a lluid. The chief 

 property of a tluid, as compared with a soliil body, is 

 the perfect mobility of its parts, the absence of rigidity. 

 Tluis there were two possible kinds of Huids — those 

 which retained their bidk or volume, whilst ottering no 

 resistance to change of shape, and those whicii tried to 

 expand, and could be compressed by means of external 

 forces. These latter were called gases. In dealing wiib 

 the former, incompressibility had to l)e defined mathe- 

 matically, as also perfect mobility. These properties 

 constitute what is called a perfect fluid. Sucii perfect 

 tluids do not exist in nature ; Imt the methud of 

 reasoning was to begin with an ideal, simple case, and 

 approach the explanation of natural phenomena by a 

 process of correction, introducing more and more com- 

 plications. The phenomena of the How of liquids, 

 practically by far the most important, could be studied 

 to a great extent by means of the simplest form of the 

 hydrodynamical conception, and up to I he middle of the 

 century such problems, as well as those of the propa- 

 gation of small displacements under the action of external 

 forces, — notably the motion of waves,— formed the prin- 

 cipal problems whicli were treated mathematically. The 

 idea of the friction of fluids, also called viscosity, had been 

 excluded in the definition of a fluid, inasmuch as friction 

 opposed the notion of perfect mobility of the parts, which 

 was the mathematical definition of a fluid. Now it is a 

 matter of experience that in all liquids wiib which we 

 are acquainted friction can produce rotational moliou, 

 such as whirls and eddies ; it was also found that other 

 forces, such as magnetic forces, are, under cerl^un con- 



