270 On Definitions, 



ers, refer to qualities which do not precisely or perfectly exist in 

 these substances. The best definition given of a fluid is, that it is 

 a substance which admits a free passage to other bodies among its 

 parts. But no liquid is actually found corresponding to this defini- 

 tion, and water, the liquid contemplated in these sciences, makes so 

 powerful a resistance to the motion of other bodies among its parts, 

 that some of the most improved instruments we possess, as the 

 steam-boat, are founded upon this principle. To render the dem- 

 onstrations in hydrostatics perfectly satisfactory, water should also 

 be entirely non-elastic, and incompressible, neither of which quali- 

 ties it perfectly possesses. Nor is air perfectly elastic, vi^hich is re- 

 quisite to complete the principles of pneumatics. 



In the departments of knowledge just mentioned, matter is con- 

 sidered in masses. In chemical science, the affinities which govern 

 its principles are understood to act either on the ultimate particles 

 of matter, or on such portions of it as are altogether imperceptible 

 to the senses. When gravity acts on a piece of iron, it acts on the 

 whole mass : but if an acid act upon the same metal, it acts only on 

 the portion with which it absolutely combines ; and in doing so, it acts 

 upon portions which are not discoverable even to the microscope. 



Definitions in Chemistry are either of the names of simple quan- 

 tities, or of compound quantities, or of instruments and operations. 



It is an important observation, that the definition of a simple quan- 

 tity, in order to be complete, must contain a distinct enumeration of 

 all its properties. These are its distinguishing characteristics, if any 

 one of which be wanting, it is not the substance we mean. A defi- 

 nition of gold, which should omit its solubility in aqua regia, would 

 manifestly be iniperfect ; for if another yellow metal should be 

 found, having the specific gravity of 19.3, and possessing the other 

 properties of gold, except that it was not soluble in aqua regia, it 

 would then appear not to be gold, but some new species of metal, 

 differing from gold in this particular quality. The same conse- 

 quence would follow from any other characteristic property of that 

 metal being found wanting in any supposed specimen that might be 

 produced. A definition, therefore, which should not distinguish this 

 new and hitherto unknown species of metal from that usually styled 

 gold, would be an imperfect and defective definition. The defini- 

 tion must, therefore, comprehend all the known properties of the 

 substance in question. But to avoid excessive and inconvenient 

 length, every property may be expressed by a single term, and the 

 full development of these terms, afterwards annexed in detail. 



