94 Mr. Nixon on the Theory of i lie [Aug. 



with moisture counterpoises 0*00 1 in. of mercury, allowance 

 being made for the diminution of gravity in the vertical line. 



Dividing 26208 by ttTsa^ ^^ hviy^ 2*35, the correction in feet for 

 one degree of difference of the attached or interior thermome- 

 ters of the two barometers, subtractive from the calculated 

 height, when, as is generally the case, the instrument at the 

 upper station is inferior in temperature to the other. Substitut- 



^^^4^1 f*^^ AHQ ^^ ^^ niean rate of expansion per degree of humid 

 air, we alter the 2*35 feet in conformity for other temperatures 

 of the air. 



The mean value of forty-one degrees of difference of the inte- 

 rior thermometers (the mean of the detached ones being 50°), 

 being equal to 100 feet, were the artist to divide forty-one 

 degrees of the scale into 100 equal parts, making the zero of the 

 new scale to correspond with 0^ F. and numbering the divisions 

 upwards to designate them SiS feet j we should note at the two 

 stations the number of feet opposite the summit of the column 

 of mercury in the thermometer in lieu of the degrees, and deduct 

 their difference (the upper barometer being coldest), from the 

 altitude calculated with the observed pressures.* When the 

 difference of level of the stations is great, the two detached 

 thermometers may differ considerably, and some Httle correction 

 should in strictness be apphed when the mean temperature of 

 the air differs from 50°. The difficulty of ascertaining the mean 

 temperature of the mercury is, however, so great, and the value 

 of its dilatation so variously given by different experimenters, 

 that it would be mere affectation of exactness to regard it. Its 

 value with the proper signs affixed is subjoined. 



Difference of the detached Thennometers, 

 Mean Temperature of the Air. 



Deluc imagined that it was necessary to reduce the heights 

 of the mercury, the temperatures of both barometers being the 

 same, to their lengths at one fixed temperature. The error 

 being pointed out. Dr. Maskelyne demonstrated that the correc- 

 tion m feet would be constant for any given temperature of the 



♦ It is to be hoped that our artists will not neglect to substitute for the centigrade 

 scale one of this description, which will enable tlic observer to dispejise with every cal- 

 culation whatever on account of the unequal temperatures of the barometers. M'^ith tlie 

 tables we shall furnish, tlie scale of Fahrenheit will be much more convenient than the 

 one with the zero at tlie freezing )K)int. 



