388 PREFACE TO 



observation, and still more in every series of observations, it is 

 no less true, as a familiar fact, that observations made by one 

 man, without conscious reference to any theory whatever, may 

 be perfectly available to another with reference to theories of 

 which the first never heard or dreamed. Colonel Reid's theory 

 of storms, for instance, was worked out, I am told, not in the 

 West Indies among the hurricanes, but at the Admiralty among 

 the ships' logs. And though Bacon would never have denied 

 that many results of theory go to the correct keeping of a 

 ship's log, who can doubt that a collection of logs kept during 

 hurricanes would have been accepted by him as a most valuable 

 contribution to a history of the winds, and a good specimen of 

 the very thing he wanted ? It would be easy to add more 

 instances ; but I suppose nobody will deny that, in this sense, 

 observation and theory can be carried on apart and by different 

 persons. And if it be objected that the observers will never hit 

 upon all the facts which are necessary to suggest or establish 

 the theory, unless their observations be renewed again and 

 again under directions devised by the theorist with special re- 

 ference to what he wants to know, I reply by asking what is 

 to prevent the renewal of them, under directions so devised, as 

 often as necessary ? a thing (I may observe) which Bacon him- 

 self distinctly intended. " Illud interim," he says, after giving 

 an example of a " topica particularis " in the De Augmentis, 

 f( quod monere occospimus iterum monemus, nempe ut homines 

 debeant topicas particulares suas alternare, ita ut post majores 

 progressus aliquos in inquisitione factos, aliam et subinde aliam 

 instituant topicam, si modo scientiarum fastigia conscendere 

 cupiant." Now if the directions, judicious to begin with, be 

 judiciously varied and repeated as the inquiry proceeds, an 

 immense mass of observations of the greatest importance to 

 science might surely be collected in this very way. Nay, in 

 subjects which have their phenomena spread far and wide over 

 the world (like winds, seasons, and oceanic or atmospheric cur- 

 rents), it is in the gradual accumulation of observations so made 

 that our only hope lies of ever coming to understand their laws 

 at all ; and if we cannot cause them to be collected under direc- 

 tion and design, we must wait till they accumulate by acci- 

 dent. For it is manifestly impossible that in such subjects as 

 these, philosophers should provide themselves with all the facts 

 which they want unless they can use the help of those whp 



