210 COSMIC PIIILOSOrilY. [pt. I. 



instruments; the reply is that they enable us to generalize 

 phenomena which before were too inconspicuous to be 

 generalized. 



Again, other things equal, phenomena that are frequent 

 have been scientifically explained sooner than unusual phe- 

 nomena. " Rainbows and comets do not differ greatly in 

 conppicuousness, and a rainbow is intrinsically the more 

 involved phenomenon ; but chiefly because of their far 

 greater commonness, rainbows were perceived to have a 

 direct dependence on sun and rain while yet comets were 

 regarded as supernatural appearances." 



In like manner the more concrete, relations have been 

 formulated before those that are more abstract. If we were 

 to adhere rigorously to Comte's principle of decreasing 

 generality, we should have to place the infinitesimal calculus 

 before algebra, and algebra before arithmetic. But the order 

 of development has been just the reverse, — from arithmetic, 

 the least abstract department, to calculus, the most abstract. 



Lastly I would suggest a circumstance, not mentioned by 

 Mr. Spencer, namely that, other things equal, the sciences 

 must advance according to the ratio between the complexity 

 of the phenomena with which they deal and the multiplicity 

 of our means for investigating those phenomena. I shall 

 presently describe our three chief implements for extorting 

 the secrets of Nature — observation, experiment and com- 

 parison ; showing that in general, as phenomena become 

 more and more complicated, our ability to make use of these 

 implements increases. In astronomy we have only observa- 

 tion to help us; but astronomic phenomena are comparatively 

 simple, so that here we have a highly-developed science. In 

 biology we can use all three implements ; and so, in spite of 

 the complexity of vital phenomena, we have here a tolerably 

 well-organized science. But in meteorology, we have to deal 

 with very complex phenomena, and still have no resource 

 save in steadfast observation. Hence meteorology is still a 



