ilfr. Wedgwood’s additional Obfervat ions, &c. 391 
paring gage made for that purpofe, fo as to pafs exaClly too® at 
the entrance of the Converging canal of the meafuring gage. 
But the pieces thus formed have been found liable, in paffing 
through ftrong fire, to receive a little alteration in their figure, 
which produces an uncertainty with refpeCl to their fubfequent 
meafurement *: the two fides, inftead of continuing flat, be- 
come concave; the edges, both at top and bottom, projecting 
beyond the middle part, fometimes very confiderably, as at a 
and b , fig. 1, (Tab. XIV ) where AB reprefents a perpendicular 
feCtion of at) unburnt piece, and ab a like feCHon of the fame 
piece after it has undergone a heat of 160 degrees. This irre- 
gularity in the form, which is fcnfible only after paffing 
through the high degrees of fire, was obferved in forne of 
the early experiments, but was not then looked upon as being 
productive of any error. 
On more attentively examining this matter, it appeared, that 
when the clay is preffed into a mould, the furface in contaCt 
with the mould acquires a more compaCt texture than the inner 
part of the mafs ; — that this compactnefs reftrains, in force de- 
gree, its diminution in the fire; — and therefore, that when 
this furface, or lefs diminifhable cruft, is pared off from the 
two fides only, the piece may be confidered as having its upper 
and lower ftrata (AA and BB, fig. 1.) compofed of a lefs di«* 
minifhable matter than the intermediate part, the neceflary 
confequence of which ftruCture will be fuch a figure as we find 
the pieces to affume ; for if any .ftratum in the mafs fhrinks 
lefs than the reft, the extremities of that ftratum muft be left 
proportionably prominent. That this was the true caufe of 
the inequality, 1 was convinced by firing fome pieces unad~ 
jitfied, with all their furfaces entire, as they came from the 
mould ; for thefe pieces, after paffing through the fame ftrong 
3 ' foeg; 
