the thermometer for high Degrees of Heat . 39*? 
rod of the thermometric clay tapering, from about four parts 
in diameter at top to three at bottom, which are nearly the 
proportions between the width of the piece when unburnt, or 
but juft ignited, and when it has fuffered a heat of 160 degrees* 
To the foregoing fources of inequality in the pieces, one 
more may be added, fmall cavities, or air* bubbles accidentally 
inclofed, which fometimes happened in the earlier experi- 
ments. In order to prevent thefe, particular attention is now 
paid by the workmen to what we call handing or flapping the 
clay, an operation by which its different parts are intermixed, 
and the mafs rendered of an uniform temper throughout. Tho 
workman takes a lump of the clay in his hands, perhaps of two 
pounds weight, and, breaking it in two in the middle, lays one 
part upon the other, and prefles them flat again, repeating this 
forty or fifty times, or perhaps oftener. Now, confidering the 
pieces at firft as two diffimilar mafles, with any number of air- 
bubbles inclofed ; each of thefe pieces being by the firft 
doubling divided into two, by the next into four, by the third 
into eight, and fo on in geometrical progreflion, each of the 
original mafles will be divided by the fiftieth repetition into 
upwards of eleven thoufand millions of millions of invifible 
laminae: — invifible, becaufe the lump of clay would, long 
before the laft doubling, be of one uniform colour, though at 
firft one-half of it had been black, and the other white. If 
therefore no air be inclofed between the pieces at the times of 
their being put together in this procefs, all the air which might 
have been in the mafs before would certainly be driven out ; 
and, to avoid as much as poflible the introduftion of any frefh 
portions of air, the two feparated pieces are each time made 
fmooth, and a little convex, on the furfaces that are to be 
brought together. 
By 
t f ft 
