BALTOl* ON THE ZERO OF HEAT. Qj 



joot of the aerial fpace, it will be found, that when the mer- 

 curial thermometer ftands at 133-§ Q , a proper equidifFerential 

 thermometer would ftand at 139° 3. Hence no inconfiflent 

 refults can be derived from the above data. 



Before the merits of the hypothefis in queftion can be de-The mercurial 

 cided, we mull: have an inftrument to meafure equal incre- ^Twreft '* 

 xrients of heat. The mercurial thermometer equally divided meafure of fieat* 

 is not that meafure. De Luc has fhewn that the true mean 

 heat betwent freezing and boiling, is feveral degrees below 

 122° per fcale. Even Dr. Crawford, who has manifefted 

 more defire than any other perfon to find the mercurial ther 

 mometer an accurate meafure of heat, has concluded from a 

 variety of experiments, all of which prove that mercury ex* 

 pands more in the higher than in the lower part of the fcale^ 

 that the thermometer is nearly accurate only. Now it has 

 been fhewn, that elaftic fluids obferve a general law of ex- 

 panfion, whether by heat. Or diminifhed preflure. Why 

 mould not liquids do the fame ? — I expe£t to be able in a E<pe<&itfo# 



mort time to demonftrate that water and mercury agree in a * a!i ila , u '<« 



, . r . r r c .? Wl11 be found * 



one ana the lame law or rate or expanuon from the two obferve the fame 



points at which they freeze and boil refpectively : and yet law of expan- 

 thefe two liquids are apparently the moil difcordant in this * 

 refpect. It would certainly be a moft extraordinary anomaly, 

 and inconfiflent with the fimplicity of the other laws of nature, 

 if mercury alone mould be excepted from a common law by 

 which all other liquids expand, not uniformly, but itfcreaf- 

 ingly, with equal increments of heat. The truth is, I be- 

 lieve, that the fame quantity of heat which raifes the mer- 

 curial thermoter 2° at the freezing point (of water), wil[ 

 raife it about 3° at the boiling point, A mercurial thermo- 

 meter graduated according to the hypothefis I have fug- 

 gefted, that the increments of heat are as the increments of v 

 the cube roots of the manometrical fpaces, (in which cafe it 

 no where differs more than 7° from the common equal gra- 

 duation for 1 80) does, I apprehend, in all cafes, indicate tn£ 

 true mean temperature, when hot and cold waters are mixed 

 together ; at ieaft, I have not been able as yet to find any de- 

 viation. 



"With refpect to the difcordance, exhibited by Seguin, in Segum's defer * . 

 the refults of experiments to determine the real zero, it may m,nat i on °* 

 arife from an error in the theory, or from errors in thevexpe- 



D 2 rimei% 



