PRECAUTIONS NECESSARY. 



to be compared one with another, it is therefore necessary to allow 

 for the difference of temperature, and this is usually done by 

 calculating at both places what the height of the column would be 

 if the temperature were that of melting ice, or 32. 



9. It is of extreme importance in the use of barometers to be 

 always sure that no portion however small, of air or other elastic 

 fluid, is contained in that part of the tube which is above the 

 summit of the mercurial column, for if any such fluid be there, 

 it will react upon the mercurial column and will depress it, 

 so that its altitude, instead of expressing the pressure of the 

 atmosphere, will express that pressure diminished by the pressure 

 of the air or other elastic fluid which is above the column in 

 the tube. 



Now it happens that nothing is more easy than to ascertain 

 whether any such fluid is there. It is only necessary suddenly to 

 incline the tube from the vertical position, so as to cause the 

 mercury to be forced up to the top by the atmospheric pressure, in 

 consequence of the vertical height of the column being rendered 

 less than that which balances the weight of the atmosphere. If 

 in such case the mercury striking the top of the tube renders the 

 blow audible by a distinct, sharp, and well-defined sound, it may 

 be concluded that the top of the tube is free from air, for if air be 

 present, even in the smallest quantity, it will react like a cushion 

 or buffer, so as to soften the blow of the mercury, and deprive the 

 sound of that sharp and distinct quality. 



10. The changes of altitude incidental to the barometric column 

 are so minute that various expedients have been resorted to, to 

 render them more easily and accurately observable. 



More sensible indications would be obtained by adopting a 

 barometer of a lighter fluid than mercury. Thus, water is 13 

 times lighter than mercury, and, consequently, a water barometer 

 would exhibit a column 13| times greater than that of 

 mercury. 



Such a column would, therefore, measure about 34 feet, and 

 a change which would produce a variation of about the tenth of 

 an inch in the column of mercury, would produce a variation of an 

 inch and a third in the column of water. 



But to the use of water, or any other liquid save mercury, 

 for barometric purposes, there are numerous and insuperable 

 practical objections. Independently of the unwieldy height of 

 the column, which would render it impossible to transport the 

 barometer from place to place, all the lighter liquids would pro- 

 duce vapour in the upper part of the tube, which would vitiate 

 the vacuum, would react against the barometric column, and 

 disturb its indications. The consequence of this has been, that 



181 



