118 OBSERVATIONS ON BASALT, &C. 



when examined by a microfcope, appear to be fafciculi of 

 flender prifms, nearly rectangular, terminated by planes per- 

 pendicular to the axis ; they are extremely brilliant ; their 

 colour is greenifh-black; they are harder than glafs, and 

 fufible at the blow-pipe ; they are fufpended by the action of 

 a magnet. They are arranged nearly fide by fide, but not 

 accumulated in thicknefs, fo that they prefent the appearance 

 of broad thin laminae ; they crofs one another at all angles, 

 but always on nearly the fame plane; and the lamina thus 

 formed is often three or four lines long, and from a line to a 

 line and a half broad, but extremely thin.* 



It feems obvious, that an equalized temperature would have 

 rendered the whole fimilar to the fubftance laft defcribed; and 

 it may be fairly inferred, that by a continuance of heat, the 

 minute cryftals would have been augmented in their dimen- 

 fions, by the accefiion of molecules fiill engaged in the bafis, 

 or by the union of feveral cryftals, till they acquired fufficient 

 magnitude for their nature to be abfolutely determined by the 

 ufual modes of inveftigation. It is probable, however, if fuch 

 precautions had been taken as might have fecured this degree 

 of perfection in the ulterior refult, that the mafs would only 

 bave exhibited an uniform afpecl, and that the interefting ini- 

 tial phenomena would not have been difcovered.f 



There 



* It may be obferved, that the cavities which exifted in the glafs 

 are not obliterated during the fubfequent proceffes, though their 

 interior furfaces undergo fome change. The minute globules fiift 

 formed often become prominent, and project into the cavities. 

 Thefe minute points are foon obliterated by the large curves of the 

 fibrous fpheroids, which give a mamellated form to the interiors of 

 the cavities ; and, when the cryftals are generated in the mafs, 

 they (hoot into fome of the cavities, and line them with their bril- 

 liant laminae. 



f In this and the fucceeding paragraphs, the word molecule is 

 ufed in the fenfe affigned to it by Hauy and Dolomieu, and is un- 

 derftood to reprefent the peculiar fblids, of definite compolition and 

 invariable form, the accumulation of which, forms the cryftals of 

 mineral fubftanccs. Such molecules, preferving their form and 

 their effential characterises, may be extracted from moft cryftals 

 by mechanical divifion, and may be fubdivided as far as our fenfes 

 can recognife them. Though we cannot by mechanical means di- 

 rectly divide them into their elementary particles, we are enabled 



to 



