[ 139 ] 



XX. A Sensitive Mercury Barometer. 

 By Frederick Guthrie*. 



THE object aimed at is to get a barometer which shall com- 

 bine great sensibility with sufficient compactness. The 

 barometers depending upon the weight of liquid columns are, 

 unless the liquid is mercury, of unwieldy length. Descartes 

 suggested a modification of the mercurial barometer in which 

 the air-supported column of liquid consists of two liquids, the 

 lower one being mercury and the upper one water holding 

 tartar-emetic in solution to ensure the expulsion of air (fig. 1). 

 Xow that we know various liquids, notably hy- iv. 1. 

 drocarbons, which have no sensible vapour-ten- 

 sion* at atmospheric temperatures, it is surprising 

 that this form has not been reintroduced. It 

 appears, however, to be little known. It was 

 suggested to me by a friend ; and I made one 

 containing glycerine and used it with success for 

 some months before I was aware of its having 

 been suggested by Descartes. The sensibility 

 of such a barometer would obviously be, if the 

 upper liquid were without weight, directly pro- 

 portional to the ratio between the sectional areas 

 of the cylindrical chamber and the upper tube 

 (if also the open limb were of infinite area). 

 But, the upper liquid having weight, the limit of 

 sensibility is the comparative density of the 

 mercury and liquid (say 16 : 1). Accordingly 

 this limit is reached when the cylindrical cham- 

 ber has four times the diameter of the upper 

 tube. 



By inclining the top part of the supported co- 

 lumn a theoretical increase of sensibility is obtained ; but prac- 

 tically, on account of the dragging on the column and for other 

 reasons, this device is not in use for exact measurements. The 

 multiplication of motion caused by applying a float connected 

 with a wheel, as in the common weather-glass, is for similar 

 reasons to be discarded for such purposes. Aneroid barome- 

 ters are of exceedingly convenient form, but are of course 

 saddled with the objection that the metal chamber is never 

 perfectly elastic, so that they require frequent comparison with 

 standard mercurial ones. 



My friend the late B. F. Duppa devised a barometer in 

 which the cistern or open limb was dispensed with, and was 

 replaced by an open horizontal capillary tube. The free sur- 



* Communicated to the Physical Society, Jan. 20, 1877. 



