MINERALS. 
9 
IT is very rarely, and indeed fcarcely ever, that the Species can be 
fufficienily determined, fince in thefe the generation proceeds not 
from the egg; but the multiplied variety of irregularly fportive 
nature, is at once the calamity of the fcience and the foundation of 
metallurgy. He therefore that fliall ralhiy endeavour to multiply 
the fpecies, is not lefs abfurd than him vvho combines fubltances 
totally different in nature. Nor does their matrix diltinguilh the 
different fpecies, more than their natural fituation and foil do the 
plants of the earth. The numerous diverfities of Hones, therefore, 
arc principally varieties;^ in the arrangement of Which, without 
caution, it Is eafy to fall into error. 
THE Ifiident has three modes of inveftigating this Kingdom: Phy. 
Jtcal, which defeends through the obfeure generation of minerals : 
N<^tural, which confiders their fuperficial and vifible (Irudlure ; 
Cbemicaly w'Hich afeends through their deflrudive analyfis. Jn 
this then, as in every thing elfe, he will mod fafely fo low the 
middle courfe, and by clofely following his ariadnean i bread, he 
will not, like an empyric, confound the fymptoms with the cure, 
nor bring forward the doubtful progeny of a long loft anceftry ; 
much lefs will his terrified imagination raife up fanciful fpedfres 
in the dark, or perfuade him that the Phoenix of the poets may be 
regenerated from its own afhes ; but he will le.irn, what names 
are repugnant to things, and w'hat are convenient ; and how to 
de'fine characters by their diagnoftics, and not merely by their ety- 
inology. But here let me paufe, left iti endeavouring to remove 
oofeurity; I myfelt become obfeure. 
VOL. VIL _ B 
