THERMOMETER. 



for the relation between % and ©, 



99 



equttion exhibiting the relation between % and T, we hare, 



log. « — 



log. z - 



made by fir George Shuclcburgh {ubi fupra) compared 

 with the refult of M. de Luc's rules. 



20000000 



4'-7'5S = 



899 



99 



92.804 = 



loooo X 899 

 = the height of the thermometer in boiling 



2000 X ICX) 



e 



ICXD 



water, above melting ice, in degrees of Bird's Fahrenheit, 

 when the height of the barometer in tenths of an Englilh 

 inch, is -i. And thus M. de Luc's formula, for the varia- 

 tion of the boiling point, is adapted to Englifli inftrumciUs, 

 and reduced to Engliih meafures of length. 



For s write 287.7525, the length of 27 French inches m 



tenths of an Engliih inch, and , the height of De Luc's 



° too 



boiling point above melting ice, in degrees of Bird's Fahren- 

 heit, comes out 177.989. Hence M. de Luc's boiUng pouit 

 falls upon 209.989 of Bird's fcale, i. e. upon 210 very nearly, 

 or infenfibly more than two degrees below Bird's pouit 

 of boiling. But as 899 is a troublefome divifor, the 

 computation will be more eafy and expeditious, by wntmg 



for — 5-^— log. s, i. Then s -(- — - s - 92.804 = 



lOOOOOO 



900 



very nearly. Upon thefe principles Dr. Horfley has 



100 



computed the table following, for finding the heights to 

 which a good Bird's Fahrenheit will rife, when plunged in 

 boiling water, in all dates of the barometer, from 27 to 31 

 Englifh inches ; which will ferve, among other ufes, to direft 

 inftrument-makers in making a true allowance for the effeft 

 of the variation of the barometer, if they are obliged to 

 finifh a' thermometer, when the barometer is above or below 

 30 inches ; though it is beft to fix the boiling point when the 

 barometer is at the height prefcribed. 



Equation of the Boiling Point. 



The nombers in the firft column of this table exprefs 

 heights of the quickfilver in the barometer in Englifli 

 inches and decimal parts : the fecond column (hews the 

 equation to be apphed, according to the fign prefixed, to 

 212" of Bird's Fahrenheit to find the true boiling point 

 for every fuch ftate of the barometer. The boiling point 

 for aD intermediate dates of the barometer may be had 

 with fufBcient accuracy by taking proportional parts, by 

 means of the third column of differences of the equations. 

 ( See Phil. Tranf. vol. Ixiv. part i. art. 30. See alfo an 

 excellent paper on this fubjeft by Dr. Mafkelyne, in the 

 Phil. Tranf. vol. Ixiv. parti, art. 20.) In the following 

 table wc have the refult of fifteen different obfervations 



Sir George Shuckburgh has alfo fubjoined the follow- 

 ing general table for the ufe of artids in condrufting the 

 thermometer, both according to his own obfervations, and 

 thofe of M. de Luc. 



The Royal Society, fully apprized of the importance of 

 adjuding the fixed points of thermometers, appointed a 

 committee of feven gentlemen to confider of the beft method 

 for this purpofe ; and their report is pubhlhed in the Phil. 

 Tranf. vol. Ixvii. part ii. art. 37. From a variety of expe- 

 riments and obfervations, relating to this fubjetft, the com- 

 mittee have deduced the following pratftical rules, which 

 they recommend in adjuding the fixed points of ther- 

 mometers. The mod accurate way of adjuding the toiling 

 point is, not to dip the thermometer into the water, but to 

 expofe it only to the fteam, in a veffel clofed up in the 

 manner reprefented in Plate XVL Pneumatics, ^g^ 6. where 

 A B i a is the veffel containing the boiling water, D d the 

 cover, E a chimney made in the cover intended to carry 

 off the deam, and M m the thermometer paffed through a 

 hole in the cover. In the purfuit of this method the fol- 

 lowing particulars muft be regarded : the boiling point 

 mud be adjufted when the barometer is at 29.8 inches ; 

 unlefs the operator correfts the obferved point in the 



manner 



