SECTION 9. - OBSER VA TION. 59 



sorption into the human system. Scientific observation therefore 

 commonly presupposes a rapidly growing arsenal of ingenious 

 and delicate instruments or other adjuncts, often devised or 

 perfected by the men of science themselves. 



An illustration drawn from Herschel and Jevons exemplifies 

 the importance of circumspect observation. Shells had been 

 encountered on high mountains, and seven hypotheses had been 

 propounded to account for the shells. Some men contended 

 that the shells had been left behind by the retreating waters 

 of the "deluge"; Voltaire argued that pilgrims had dropped 

 them there; others thought they were freaks of nature, or that 

 they were due to fermentation, to the influence of the celestial 

 bodies, or to birds feeding on shell-fish; and a seventh group 

 of persons claimed that they were the remains of living forms 

 covered by accumulations of debris of various kinds, and sub- 

 sequently exposed or detached. Now the first six hypotheses 

 were practically gratuitous surmises, since they were evidently 

 not derived from anything like wide observation. And if the 

 seventh one was merely a fortunate hit, not grounded on, nor 

 to be succeeded by, exhaustive observation, it was to all intents 

 and purposes as unprofitable as the other six. Only the pro- 

 longed, extensive, minute, and accurate observations of such 

 men as Sir Charles Lyell and Archibald Geikie, have enabled 

 geologists to propose valid generalisations, and to proceed 

 deductively with some effect. "Comparison must be made 

 with facts purposely selected so as to include every variety 

 of case, not omitting extreme ones', and in sufficient number 

 to afford every reasonable probability of detecting error", and 

 all conclusions need to be "in exact accordance with numerous 

 observations purposely made under such a variety of circum- 

 stances as fairly to embrace the whole range of the phenomena 

 which the theory is intended to account for". (Herschel, Dis- 

 course, [219-220.].) 



In earlier stages of a science when comparatively much 

 material has been accumulated, but yet far too little to permit 

 the detailed testing of a sweeping conjecture the formation 

 of a large hypothesis is only admissible when its purely pro- 

 visional character is stressed. Within these limits it has marked 

 advantages. If, however, the apparent and partial consistency 

 of the hypothesis with the imperfectly known facts is regarded 

 as of itself conclusive evidence for its correctness, we are 

 likely to be imposed on by a mirage, and the progress of 

 science is arrested. A pertinent illustration of the above is 

 offered by the story of the highly ingenious and purely hypo- 

 thetical phlogiston theory, which we accordingly quote: 



'The theory of phlogiston was originally broached as a theory of 

 combustion. According to this theory, bodies such as coal, charcoal, wood, 

 oil, fat, etc., burn because they contain a combustible principle, which 

 was assumed to be a material substance and uniform in character. This 



