EXPERIMENTS on WHINS TO NE and L A VA, £^ 



from the one to the oiher. He has fiiice expreffed the fame 

 fentiments in one of his late publications, (Theory of the Earth, 

 vol. I. p. 25 I.J, where he cenfures thofe who '' judge of the 

 " great operations of the mineral kingdom, from having kin- 

 " died a fire, and looked into the bottom of a little crucible." 



BoT, notwithflanding my veneration for Dr Hutton, I 

 could not help differing from him on this occafion ; For, grant- 

 ing that thefe fubftances, when in fufion, were adled upon by a 

 heat of ever fo great intenfity, it is certain, neverthelefs, that 

 many of them muft have congealed in moderate temperatures, 

 fince many are eafily fufible in our furnaces ; for it is impoflible 

 that a fubftance lliould congeal at a higher point than that at 

 which it may afterwards be melted. If, then, thefe phenomena 

 depend upon the circumftances of congelation, the imitation of 

 the natural procefs is an objedl which may be purfued with rationr- 

 al expe<5lation of fuccefs ; and, could we fucceed in a few examples 

 on a fmall fcale,. and with ealily fufible fubftances, we fhould 

 be entitled to extend the theory, by analogy, to fuch as, by their 

 bulk, or by the refracflory nature of their corapofition, could 

 not be fubjedled to our experiments. It is thus that the aftro*- 

 nomer, by obferving the efl[e(5ls of gravitation on a little pendu- 

 lum, is enabled to eftimate the influence of that principle oa 

 the heavenly bodies, and thus to extend the range of accurate 

 fcience to the extreme limits of the folar fyftem. 



Encouraged by this reafoning, I began my projeifled feries 

 of experiments in the courfe of the fame year (1.790), with very 

 promifing appearances of fucceis. I found that I could com- 

 mand the refult which had occurred accidentally at the glafs- 

 houfe ; for, by means of flow cooling, I converted bottle glafs, 

 after fufion, into a fl;ony fubftance, which again, by the ap- 

 plication of ftrong heat, and fubfequent rapid cooling, I refto- 

 red to the ftate of perfedl glafs. This operation I pei-formed re- 

 peatedly with the fame fpecimen, fo as to afcertain that the cha- 



radeE- 



