DRAPER'S BAROGRAPH. 



Oy PLORIiVN CiVJOKI. 



The description of Draper's Barograph in the Scientific 

 American Suiyplcmejit, No. 209, 1880, contains the following 

 passage: "Heat has a slight effect on them (the springs), 

 causing them to lengthen about iV of an inch for 90 degrees 

 Fahr. ; .... OfJierwise this instrument gives the correction 

 for temperature (or reduction to 32 degrees) from the fad 

 that it tceighs the mercury instead of measuring its length, 

 which is affected by heat J' This statement appears to be 

 considered correct, for we are told that in the use of Draper's 

 Barograph no corrections are made for variations in tempera- 

 ture of the mercury. 



It is the object of this paper to point out that the state- 

 ment quoted in italics is erroneous. The instrument in 

 question consists of a tube 36 inches long, the upjjer portion 

 being of larger diameter than the lower. The tube is held 

 firmly in a fixed position and filled with mercury. Its lower 

 end dips into a movable cistern which is suspended on two 

 spiral springs. When the atmospheric pressure diminishes, 

 part of the mercury flows from the tube into the cistern; this 

 becoming heavier, stretches the steel springs, causing the ink 

 pencil fastened to them to mark downwards. If the pressure 

 increases, the opposite movement takes place. 



The error referred to can be established by the following 

 reductio ad ahsurdum: All the mercury in the instrument 

 is supported by the movable cistern except that which is in 

 the fixed tube above the surface of the mercury in the cistern. 

 Now, if we suppose the atmospheric pressure to be constant, 

 but the temperature of the mercury to vary (the temperature 

 of the springs remaining the same), then, if the quoted state- 

 ment is correct, the cistern should neither risp nor fall, else 



