[ 98 ] 



First. The fele£\jon of the materials which compofe gun- 

 powder. 



Secondly. The flrongeft and moft durable proportion of thofe 

 materials. 



Thirdly. The beft mode of intermixing and combining 

 them. 



Lastly, I fhall add fome general obfervations. 



The qualities of nitre are not cafily afcertained by thofe rules 

 which chymifts have prefcribed for determining its purity ; their 

 deviations are frequent and fometimes material in the compofition 

 of gun-powder, whofe bafis this fait conftitutes. The method I 

 have generally adopted for deteding the impurity of nitre, is to 

 drop a flrong folution of Sacch. Sat. into a phial of diftilled 

 water, faturated with falt-p?tre ; which, if it retained any con- 

 flderable portion of marine fait or magnefia, affumed a turbid 

 milky appearance : The lunar folution is too powerful a teft for 

 any nitre I have met with : But it does not always follow that 

 the pureft nitre produces the ftrongefl powder : The beft I have 

 feen is the Ruffian, yet the manufacturers in that country are not 

 very felicitous about the magnitude of the cryftals, the white- 

 nefs of the fait, nor even its freedom from heterogeneous fubftances, 

 though with us thofe qualities are accounted eflential. In Ruffia 

 I am informed they feldom refine their nitre more than twice ; 

 and having analyzed fome very excellent Ruffian powder, I found 

 the falt-petre contained a conliderable portion of marine filt and 



magnefia. 



