EMPIRICAL KNOWLEDGE, EXPLANATION, d-c. 1G9 



combination of laws, will occur to a reasoner. Whatever 

 road a traveller takes, be is sure to arrive somewhere, but 

 unless he proceed in a very systematic manner, it is very 

 unlikely that he will reach every place to which a network 

 of roads will conduct him. In like manner there are 

 many phenomena which were virtually within the reach 

 of philosophers by inference from their previous knowledge, 

 but were never discovered until accident or systematic 

 empirical observation disclosed their existence. 



That light is propagated with a certain uniform but 

 very high velocity, was proved by Roemer, by observation 

 of the eclipses of Jupiter s satellites. Corrections could 

 henceforward be made in all astronomical observations 

 requiring it, for the difference of absolute time at which 

 an event happens, and that at which it becomes evident to 

 us. But no person happened to remark that the motion 

 of light compounded with that of the earth in its orbit 

 would occasion a small apparent displacement of the 

 greater part of the heavenly bodies. Fifty years elapsed 

 before Bradley empirically discovered this effect, called by 

 him aberration, when examining his accurate observations 

 of the fixed stars j . 



When once the relation between an electric current 

 and a magnet had been detected by Oersted and Faraday, 

 it ought, theoretically speaking, to have been possible for 

 them or any other person to foresee the diverse results 

 which must ensue in different circumstances. If, for in 

 stance, a plate of copper were placed beneath an oscillating 

 magnetic needle it should have been seen that the needle 

 would induce currents in the copper, but as this could not 

 take place without a certain reaction against the needle, it 

 ought to have been seen that the needle would come to 

 rest more rapidly than in the absence of the copper, let 

 this peculiar effect was accidentally discovered by Gambey 

 Laplace, Precis &amp;lt;le 1 liistoire do 1 Astronomic, p. iO|. 



