t 4§4 ] 
accuracy in making experiments, had, after trial, 
judged to be inefficient. And it was not till about 
the latter end of the year 1734, that I again refumed 
them on the following occafion. A gentleman de- 
firous to make fome experiments concerning the ex- 
panlion of metals, employ’d me to make him an in- 
ftrument like one invented by Mr. Mufchenbroek for 
that purpofe, which he calls a pyrometer. Upon 
looking over Mr. Mufchenbroek’s experiments, I not 
only found the difference between the expanfion of 
fome of the metals much greater than I expected, 
but, as I thought (if they were to be depended upon) 
fufficient to anfwer my former purpofe. This led 
me to conlider the ftrutture of the inftrument, which 
Mr. Mufchenbroek made ufe of in his trials, and 
upon examination, I thought it liable to fome objec- 
tions, which I imagined would make the refult of 
experiments made by his inffrument very uncertain. 
I therefore endeavour’d to contrive one of a different 
conft ruction, that might be more to be depended 
upon. Such an inftrument I fome time afterwards 
completed, and had the pleafure to find it fo far 
met with the approbation of l'everal very worthy 
members of this Society, that, at their particular de- 
lire, I drew up a defeription of it, which was read, and 
the inftrument itfelf fhewn to the Society on the 8 of 
April 1736*: and though it was not in every re- 
fpedt 
• This appears from the minutes of the Royal Society ; tho’ the 
defeription and manner of that inftrument was, by fome miftake, 
placed fix months later in the Pbilof. Tranf. Vol. XXXIX. N. 443. 
for October 1736. The other dates, mention’d in this paper, re- 
lating to that inftrument, are likewife juftified by authentic vouch- 
ers, which have been produced by Mr. Ellicott. 
