HEAT OR CALORIC. 59 



The thermometer ought therefore to be graduated, when the ba- 

 rometer is at the medium pressure, or a proper allowance should be 

 made for the variation.* 



(p.) Correspondence, of thermometers. 



All thermometers, accurately made upon these principles, will cor- 

 respond, however different in size or form.f 

 1.) Choice of fluids. 



[ercury from its mobility, cleanness, beauty, nearly equable ex- 

 pansion by heat, great sensibility to that agent, and the wide differ- 

 ence between its boiling point, -f 656 J and 39 its freezing point, 

 is generally used ; oil is viscid and water very limited in its range, be- 

 sides its unequal contraction between 40 and 32. 



Alcohol tinged with carmine, is used for intense cold, but cannot 

 be used for heats above 176,)| nor quite so high, on account of its 

 unequal expansion near the boiling point ; while in sensibility it is 

 much inferior to mercury. 



!r.) Imperfections of the thermometer. 

 t does not give the result instantly ; there is some loss of tempera- 

 ture before the effect can be observed ; it gives no information as to 

 the absolute heat, reckoning from the real zero ; it indicates only rel- 

 ative heat, or heat compared with some known degree, just as marks 

 may be placed on the links of a chain, whose terminations are con- 

 cealed. We know not the beginning or the end of heat ; but this is 

 not the fault of the thermometer : the range of the thermometer is 

 necessarily limited between the freezing and the boiling points of 

 the fluid with which it is filled. 



&.) Uses of the thermometer. 

 o accurate knowledge of the laws of heat could have been ob- 

 tained without it ; hence the observations of the ancients on heat are 

 of little value. 



For philosophical purposes, it is indispensable. It is of use to a 

 physician, in observing the phenomena of disease, as of fever and 

 inflammation and in experiments on animal life, &ic. 



* See Phil. Trans. 1777, for the rules of the Royal Society; also Phil. Trans, abr. 

 IV. 1. for Newton's rules. See also Martine, on heat and thermometers, and Eng- 

 iish Jour. Science, Vol. VII. p. 183, Chevalier Landriani. 



t For various causes of disagreement, see Cordier's Essay on Temp, of the Earth, 

 p. 143. 



t This point is stated by Irvine to be 672 of Fahr. (Murray, I. 153.); 662 Petit 

 and Dulong ; 656 Crighton, Glasgow ; mean of the three, 663^. Hen. 9th ed. 

 Vol. I. p. 101. 



Its boiling point is higher than that of any permanent fluid, and its freezing 

 point lower than that of any fluid, except alcohol and ether. Between 32 and 212, 

 its expansion is almost perfectly uniform, and, although at higher temperatures its 

 expansion goes on in an increasing ratio, glass has, within the above limits, the same 

 ratio, and therefore there is no practical error. Turner, p. 28. 



ij This is its boiling point when its specific gravity is 820, water being 1000. 



