Effects of Hectt imdlpad oy Compression. SOI 



method previously used. I therefore deteruiincd in future 

 to adopt the following arrangement. . - |. 



A small lube of poFcelaiii {i }i, fig. 23.) was,. weighed by 

 means of a counterpoise of sand, or granulated tin ; then 

 the carbonate was firmly rammed into the tube, and the 

 whole weighed again : thus the weight of the carbonate, 

 previous to the experiment, was ascertamed. After the ex- 

 periment, the tube, with its contents, w-as again weighed; 

 and the variation of weight oljiaincd, Indepemlently of any 

 mutual action that had taken place between the tube and 

 the carbonate. The balance which I usc4,- turned, in 

 a constant and steady manner, with one hundredth of a 

 grain. When pounded chalk, was rammed into this tube^ 

 I generally left part of it free, and in that space laid a small 

 piece of lump chalk (/'), dressed to a cylinder, with the ends 

 cut flat and smooth ; and I usually cut a letter on each end, 

 the more effectually to observe the effects produced by heat 

 upon the chalk ; the weight of this piece of chalk being al- 

 ways estimated along with that of the powder contained in 

 the tube. In some experiments I placed a cover of porce- 

 lain on the muzzle of the little tube, (this cover being 

 weighed along with it,) in order to provide against the case 

 of ebullition : but as that did not often occur, I seldom took 

 the trouble of this last precaution. 



It was now of consequence to protect the tube, thus pre- 

 pared, from being touched during the experiment by any 

 substance, above all by the carbonate of lime, which might 

 adhere to it, and thus confound the appreciation by weight- 

 This was provided for as follows : Tiic small tube (fig. 23. 

 ik), with its pounded carbonate {k), and its cylinder of 

 lump chalk (i), was dropt into a large tube of porcelain 

 [pk, fig. 24). Upon this a fragment of porcelain (/), of 



such a size as not to fall in between the tubes, was laid. 



I' 

 Then a cylinder of chalk (m) was dressed, so .as 7)ear!y t© 



fit and fill up the inside of the large tube, one e.nd of it |»eiing 



rudely cut into the form of a cone. This mfls.? bcJpg tlK-n 



introduced, with its cylindrical ciid; , downward!!,; v^asni^de 



to press upon the fragment of porcelain {I). J liiCij dropped 



into the space («), between the corneal- pari oti.this- mass 



and 



