202 COSMIC PHILOSOPHY. [pt. i. 



scientifically coordinated any one class of phenomena dis- 

 played in the heavens, they had previously coordinated a 

 parallel class of phenomena displayed upon the surface of the 

 earth." 1 



This criticism is a very incisive one. It destroys this part 

 of Comte's classification not only from the historical, but 

 also from the logical point of view. It shows that the study 

 of astronomy depends upon that of terrestrial physics, and 

 should therefore come after, and not before it. In fact the 

 whole science of astronomy, as at present constituted, con- 

 sists of two portions, — the theory of gravitation and the 

 theory of nebular evolution. The first of these, as we have 

 just seen, is a mere extension to celestial phenomena of cer- 

 tain laws of terrestrial physics. The second depends upon 

 the study of terrestrial phenomena in a yet greater degree, 

 since it involves the knowledge not only of gravitation, but 

 also of radiant heat, and of the conditions of equilibrium of 

 gases and liquids. 2 



If now we compare physics with chemistry, we shall find 

 a similar ambiguity in Comte's results. It is easy to say 

 that chemistry was not organized into a science until toward 

 the close of the eighteenth century, while physics was 

 organized at the beginning of the seventeenth : but what do 

 we now mean by physics ? If we mean merely the science 

 which generalizes the phenomena of weight, our proposition 

 is indisputable ; but unfortunately it is of little use in 

 supporting the Comtean classification. For Comte, as we 

 have seen, includes under the general head of physics, not 



1 Spencer's Essays, 1st series, p. 179. 



2 1 leave this as it stood five years ago, when this chapter was written. 

 The numerous and wonderful disclosures of spectrum-analysis, not only giving 

 us unlooked-for information concerning the physical constitution of the stars, 

 but even throwing new light on their movements, make it desirable, perhaps 

 to enlarge the scope assigned to astronomy in the text. But such a modifica- 

 tion of the form of statement would show only the more forcibly how closely 

 the study of astronomy depends on the study of terrestrial phenomena. The 

 greatest step recently taken in science is thus an additional argument against 

 the validity of Comte's conception. 



