Improvement of the Barometer. 



money, led me first to enquire the cause, which I soon learned, 

 and then undertoojj to invent something which should effectually pre- 

 vent the evil. I soon succeeded, even beyond my most sanguine 

 expectations, so as even to render the instrument perfectly secure 

 against all accidents, except breaking, to which all instruments are 

 subject. With this security, I have not encumbered the barometer 

 with any thing on the outside, but the whole is confined to the cis- 

 tern, thereby retaining the instrument in the most portable form. 



The annexed engraving, represents a 

 vertical half section of the barometer. — 

 a, b, c, and d, is the cistern two inches 

 long and one inch in diameter, e and f 

 is a glass lube, open at both ends, and let 

 into the cistern above and below zero, 

 which in the barometer, is always chang- 

 ing its position. The original zero is 

 marked on this tube at g, with decimal 

 parts of the inch, extending above and 

 below, to be deducted from or added to 

 the height of the mercurial column, in 

 the large tube, h is zero, which when 

 made, on a level with the ocean, stood f 

 of an inch from the top of the cistern, 

 which immerses the top of the globe f of 

 an inch in mercury. The f of an inch 

 between original zero and the top of the 

 cistern, leaves sufficient space for the col- 

 umn in high altitudes to fall ; a circum- 

 stance which has never been properly at- 

 tended to, in constructing barometers ; 

 although probably no other barometer 

 will admit of so much space, without en- 

 dangering the instrument, i is the end 

 of the tube, with the column drawn to a 

 small point, which answers precisely the 



same purpose as the contraction in Mr. Gay Lussac's " improved 

 marine portable barometer." But in this case, the contraction at the 

 bottom of the tube, possesses other advantages than merely to pre- 

 vent the sudden rise and fall of the mercury ; for by placing the con- 

 traction at the bottom, we can draw the end of the tube to a small 



