3&4 Mefnblr on tie Supply arid 



Ihouths of ttie curved pipes may be brought down to the ftir- 

 face ot the tvater; this gives a facility to the difcovery of any 

 heavier gas \vhich mav be mixed with one which is more 

 light, as the {liiid, whofe fpecitlc gravity is greater, will be 

 found on the furface of the \rateri 



I fliall now defcribe the changes efibfttd on the moft fixed 

 and refraftory fubftances by llie flame of the hydrogen and 

 oxygen gafes. 



In order to avoid a tedioiis recurrence to an awkward 

 phrafe, I fhall generally, in the fuhfwpient part of this paper, 

 diftinguifli the flame of the hydrogen and oxygen gafes by 

 the appellation of gafeous fiame. 



By expofiire to the galeoiis flame, either on fupports of 

 filver or of carbon, barytcs, alumine, and lilex, were com- 

 pletely fufed. 



The produfts of the fi'fion of alumine and (ilex were fub* 

 fiances very fimilar to each other, and much refembling white 

 enamel. 



The refult of the flifion of barytes \vas a fubftance of an 

 afh -coloured caft, which after long expofure fometimes ex- 

 hibited brilliant yellow fpccks. If it be certain that barytes 

 is an earth, thefe fpecks mufl: have been difcoloured particles 

 of the filver fupport, or of the pipes from which the flame 

 ifliied. 



Lime and magnefia are extremely difllicult to fufej not only 

 becaufe they are the moll rcfraClory fubflances in nature, but 

 from the diflFiculty of preventing thom from being blown on 

 one fide by the flame s nevcrlhelefs, in fume iiiftances by ex- 

 pofure on carbon to the gafeous flame, iinall portions of thefe 

 earths were converted into black vitreous mafl'es. Poflibly 

 the black colour of thefe products of fufion may have been 

 caufed by iron contained in the coal; for in the high tem- 

 perature of the gafcous tlame, a powerful attradlion is reci- 

 procally exerted by iron and the earths. 



Platina was fufed, by expofure on carbon, to the combuf- 

 tion of hydrogen gas and atmofpheric air. But the fufion of 

 this metal was rapidly acconiplilhed by the gafeous flame, 

 either when expofcd to it on carbon or upon metallic fup- 

 ports. 



A fmall quantity of this metal in it.-s native granular form, 

 being ftrewed in a filver fpoon, and pafliid under the gafeous 

 flame, the track of the flame became marked by the con- 

 glutination of the metal ; and, when the heat was for fome 

 time continued on a fmall fpace, a lump of fufed platina be- 

 eame immediately formed. 



About two penny-wciglits of the native f^rain? of platina, 



when 



