BY C0MPRES310!>r. J7 



These experiments, and many others made about the same T^e water in- 

 time, with the same success, clearly prove the efficacy of \va- creased the 



• 1 1 1 • pressure, and 



tor in assisting the compression ; and results opproacaing to ^^^ eflect was 



these in quality, obtained, in some cases, by means of a very merely nie- 



^ •' . . . „ r " 1 • ohanical. 



small air-tube, shew that the mfluencc of water on this occa- 

 sion has been merely mechanical. 



During the following summer and autum 1803, I was oc- 

 cupied with a ditlercnt branch of this subject, which I shall 

 soon have occasion to mention. 



In the early part of last year, ISO^, I again resumed the Other volatile 

 sort of experiment lately described, having in view principally compression ;^'^ 

 to accomplish absolute compression, in complete imitation of bat water is the 

 the natural process. In this pursuit, T did not confine myself ^'' ' 

 to water, but made use of various other volatile substances, in 

 order to assist compression ; namely, carbonate of ammonia, 

 nitrate of ammonia, gunpowder, and paper impregnated with 

 nitre. With these 1 obtained some good results, but none such 

 as to induce me to prefer any of these compressors to water. 

 Indeed, I am convinced, that water is superior to them all. 

 I found, in several experiments, made with a sirriple air-tube, 

 without any artificial compressor, in which a very low red-heat 

 had been applied, that the carbonate lost one or one and a half 

 per cent. Now, as this must have happened in a temperature 

 scarcely capable of inflaming gunpowder, it is clear, that such 

 loss would not have been prevented by its presence : whereas 

 water, beginning far below redness to assume a gaseous form, 

 will effectually resist any calcination, in low as well as in high 

 heats. And as the quantity of water can very easily be regu- 

 lated by weight, its employment for this purpose seems liable 

 to no objection. 



On the 2d of January 1804, I made an experiment with Experiment in 

 marble and chalk, with the addition of 1.1 grain of water. I t^e^s;,!^" ^l^^, 

 aimed at a low heat, and the pyrometer, though a little bro- sure chalk was 

 ken, seemed clearly to indicate 22''. Unluckily, the muzzle y^h heat^and 

 of the large tube, vvhich was closed as usual with chalk, was another part 

 placed uppermost, and exposed to the strongest heat. I found nme^tone at a 

 it rounded by fusion, and in a frothy state. The little tube lower. 

 came out very clean, and was so nearly of the same weight as 

 when put in, that its contents had lost but 0.074 per cent, of 

 the weight of the original carbonate. The marble was but 



Vol. XIV.— May. iSO(). D feebly 



