1825.] Barometrical Measurement of Heights, 169 



surrounds that tube. This ratio is termed (for brevity) the 

 capacity. 



If we suppose the zero (see last figure) of the scale of an 

 Englefield barometer placed in a vacuum to coincide with the 

 surface of the mercury x within the tube, depressed from capil- 

 larity some little below its level / in the cistern, it is evident that 

 the height of the mercurial column on the admission of the 

 atmospheric air will always be observed in defect, and in one 

 constant ratio. Admitting the bore of the tube to be uniform, 

 it is unnecessary to be acquainted with this ratio (equal to that 

 of the capacity); we have merely to suppose the mercury 

 increased in density in some unknown degree^ and the calcula- 

 tions will give at once the true altitude without the previous 

 troublesome corrections for capacity (and capillarity). 



Example, 



Correct pressures. 



30-600in Log. -4857214 



-^. 20-400 -3096302 



Differences -1760912 - 



Observed pressures (capacity j'o), 



30-000 in... Log. -4771213 



20-000 -3010300 



Differences -1760913 



When a variation of temperature occurs, the heights of the 

 columns augment and diminish, not only without interfering 

 with the level of the cistern, but the height of the mercury 

 therein is itself subjected to a simultaneous elevation and 

 depression in proportion to its varying depth.* Setting aside 

 this latter cause of error, as being too trivial to be regarded, it 

 must, however, be admitted that the reduction of the columns 

 for temperature should be made on their observed heights aug- 

 mented in the ratio of the capacity, or that the expansion per 



degree (tt^^ should be proportionally increased. The capa- 

 city being rarely greater than one-fortieth, the error in a differ- 

 ence of 16 degrees of the attached thermometers will be no more 

 than one foot. 



The adjustment in a vacuum being scarcely practicable, let 

 the artist measure the height of the column of mercury (as 

 usual) from the level of the cistern, and subtract the correction 

 for capacity minus that for capillarity. The scale being after- 



* Admitting tlie tube and cistern to preserve their diameters unaffected by tempera- 

 ture. 



