ji4 Defetts of Thermometers. [Book II. 



aicertained 'by a ftill more accurate experiment of Dr. 

 Crawford, who contrived a method of combining the 

 boiling and freezing points together, and found that 

 the degree of heat communicated to the thermometer 

 was as nearly as poifible the arithmetical mean *. 



It is evident, from the nature of expanfion, that 

 thermometers might be conflructed of folid bodies. 

 Metallic thermometers have indeed occafionally been 

 made, and graduated for different purpofes ; but their 

 utility is ncceffarily very limited, fince folid bodies are 

 expanded with much more difficulty, and in a lefs de- 

 gree, than fluids. 



Though the mercurial thermometer is fo much 

 more perfect, and is capable of exhibiting much higher 

 degrees of heat than thofe which had been in ufe be- 

 fore the time of Fahrenheit ; yet as mercury boils at 

 600, that is, confiderably below the red heat of iron, 

 and as it is plain that no fluid can afford any true 

 meafure of heat beyond that point in which it is itfelf 

 converted into vapour, it is equally plain, that there 

 mud exifl fcveral degrees of heat which cannot pof- 

 fibly be exhibited by the mercurial thermometer. 

 Thefe degrees are very inaccurately defined by the 

 chemifts and artifls, according to the appearance, term- 

 ing them a red and white heat*, &c. To remedy the 

 inconveniences resulting from the want of a definite 

 llandard of heat above the point of boiling mercury, 

 leveral methods have been propofed, but there are 

 only two which I efteem worthy of notice. 



A very eminent philofopher, who may be termed 

 the father of the modern doctrines concerning heat, 

 propofes, in order to afcertain the heat of any given 

 furnace, for inftance, to heat fome body (the dimen- 



* Crawford on Heat, p. 47, 48. 



fions 



