LEYELmG, AND USE OF BAROMETEE. 



75 



Taking the 2d formula and correcting h the upper reading for 

 difference of temperature of mercury by (i -f 0,0001. T'— t',) gives 

 5831.014 feet, here we have lost, as we have reduced by the differ- 

 ence of temperature of mercury which was lower at the upper sta- 

 tion, and consequently indicated a greater altitude. 



If now a correction for dilation of atmosphere is made by 



'T4- 1 



1 + 0,0021 



—32 



we obtain 6192.21 feet, this correction necessarily increases the dif- 

 ference of the readings and consequently the altitude. The third 

 correction of 1 — (0.0028 Cos. 2 Lat.) reduces the result only one 

 hundreth of a foot. This example gives 106.5 feet for every one- 

 tenth of an inch of difference of barometers, and it will be observed 

 that the result differs by but 0.78 feet in one hundred from the first 

 formula, which is arranged for a mean of 55"^ Fah. and by 3.70 in 

 one hundred from the second, which is for a mean of 32o Fah. 

 The complete formula with three cori-ections reads: 



b(i+o,oooi.(T'-t'))X 1 +( 0>0021 ( -p- 



=k log 



•32)): 



1— (0,0028 Cos. 2 Lat. 



The following form is convenient: 



B=28,94 

 b=23,13 - 



l+(0, 0001X15°)=!, 0015 



Difference=0, 096861 log= 



^==60200 Iog.= - - 



l + (0 ,0021X29,5°)=1 ,06195 log.= - 



1— (0,0028XCos. 89«50')=0, 99999 log= 



a:=6196 ,20 feet from sum of logs.= 



log.=l. 461499 



log.=l, 363988 

 log .=0,000650 



Sum^ 

 Diflerence= 



\ 1,364638 



0,0096861 



2,986148 

 4,779596 

 0,026102 

 1,999999 



3,791844 



The aneroid barometer is coming into general favor perhaps 

 more from convenience than any other reason, as it is imposible to 

 read as small fraction from the index as from the vernier of the 

 mercurial barometer. 



From observations made in two different rooms at temperatures 

 of 54^ and Qo° respectively, a difference was found amounting to 

 ten feet, but since this was equal to the probable error of observa- 



