600 SCIENTIFIC THOUGHT. 



expounder. Persons are grouped according to some 

 characteristic, and the number of persons forming the 

 group is set down under that characteristic. This is 

 the raw material from which the statist endeavours 

 to deduce general theorems in sociology. Other 

 students of human nature proceed on a different 

 plan. They observe individual men, ascertain their 

 history, analyse their motives, and compare their ex- 

 pectation of what they ' will do with their actual con- 

 duct. . . . However imperfect this study of man may 

 be in practice, it is evidently the only perfect method in 

 principle. ... If we betake ourselves to the statistical 

 method, we do so confessing that we are unable to 

 follow the details of each individual case, and expecting 

 that the effects of widespread causes, though very differ- 

 ent in each individual, will produce an average result on 

 the whole nation, from the study of which we may 

 estimate the character and propensities of an imaginary 

 being called the Mean Man. Now, if the molecular 

 theory of the constitution of bodies is true, all our 

 33. knowledge of matter is of a statistical kind. A con- 

 knowiedge stituent molecule of a body has properties very different 



of nature. 



from those of the body to which it belongs. The 

 smallest portion of a body which we can discern con- 

 sists of a vast number of molecules, and all we can 

 learn about the group of molecules is statistical in- 

 formation. . . . Hence those uniformities which we ob- 

 serve in our experiments with quantities of matter con- 

 taining millions of millions of molecules are uniformities 

 of the same kind as those explained by Laplace and 

 wondered at by Buckle, arising from the slumping to- 



