I 134 ) 



liiftfc, are very hard, and in their fra(Stures 

 affed: the conchoidal figure ; they move 

 the magnetic needle at the diftance of more 

 than a Hne, and give fparks tolerably freely 

 with fteel. The petrofiliceous fubftance 

 contains a few feltfpars, and a confiderable 

 number of fhoerls. Thefe globes of lava 

 are of various fizes, feme of them being a, 

 foot in diameter. They are detached, and 

 are never found in ftrata, but only in large 

 accumulated heaps. 



To what caufe can we afcribe the divifioa 

 of this lava, and its conformation in the 

 manner defcribed ? I at firft imagined that 

 its figure might be the confec^uence of the 

 agitation of the Tea, when its waters reached 

 to a greater height ; as thefe accumulations 

 of globes are now fome poles above its level. 

 In faft, in my maritime excurfions round 

 the other Eoliaa ifles, and at Etna, 1 have 

 frequently met occafionally with fimilar balls 

 of lava, which clearly indicated that they 

 had been rounded by being continually 

 rolled by the waves of the fea, in the famq 



manner 



