20 THE INTELLECTUAL RISE IN ELECTRICITY. 



to the stone and in this way perceived its attraction. 

 Iron, however, is never found in a metallic state in 

 nature, except in meteorites. Excluding this infinitesimal 

 supply, the metal is obtained from its ores, by means 

 usually involving the development of intense heat, so 

 that to devise modes of attaining the necessary tempera- 

 tures, let alone the even more complex mental work 

 of contriving apparatus and processes for separating the 

 metal, requires advanced powers of observation and inven- 

 tion. Hence modern ethnological and geological author- 

 ities unite with Lucretius 1 and other ancient writers in 

 affirming that the Age of Iron has always followed that 

 of brass or bronze. So far, therefore, as establishing the 

 probable time of the discovery of the attractive force of 

 the lodestone is concerned, it is immaterial whether we 

 consider that the phenomenon was first remarked as an 

 effect of outcropping magnetite upon iron brought near to 

 it; or as one exerted by fragments of magnetite in an iron 

 mine upon other fragments of the same substance, or 

 upon extracted iron. In any case, the observation of the 

 fact seems necessarily to have followed the advent of an 

 Iron Age, and therefore may not extend indefinitely back 

 into prehistoric times. 



On the other hand, with regard to the directive tendency 

 of the lodestone a different conclusion is reached. To sus- 

 pend an elongated piece of the stone and see it turn itself 

 in a definite direction; or to do this repeatedly and with 

 different pieces and thus learn that the phenomenon is 

 true of this particular stone and not of other stones, obvi- 

 ously involves no necessary knowledge of its attractive 

 effect on iron. Therefore, if we admit the possibility of 

 sufficient intelligence in the race then living, we may con- 

 jecture that an acquaintance with magnetic polarity may 

 have existed among the earliest peoples of which we 

 have any tradition. I shall show hereafter that reason 

 for such conjecture is by no means absent, which if ac- 



1 De Natura Rerum, v. 



