Monthly Monthly Range 



January, .. .. 

 February, . . , 



May,' 



June, 



July, 



August, 



September, . . 

 October, .... 



December, . . 



29.40 



29.71 



29.31 



29.19 



29.74 



29.55 



29.82 



29.80 



29.45 ] 



29.60 



29.19 1 



29.70 



30.75 29.19 1.5 



30.80 29.4211.38 



The mean of the barometrical observations for 1835 and 1836, 

 as deduced from the above tables, is 30.131 inches.* 



The barometer used in 1835 and the early part of 1830, was of 

 the syphon construction, and no allowance has been made for ca- 

 pillarity or variation of level in the cistern. The calibre of the 

 tube is estimated at .28 inches, and the relative «reas of the cross 

 sections of the tube and cistern as 1 to 30. The scale was ad- 

 justed by measurement with common boxwood instruments, and is 

 deemed 'to be a little too high above the cistern, and consequently 

 reading off a little too low. Since May, 1836, the observations 

 were taken from a new barometer, constructed by Roach, under 

 special directions, which has a glass cistern surrounding the bot- 

 tom of the tube, with solid glass bottom which shows the level of 

 the mercury in the cistern and the point of zero adjustment, and 

 which is not subject to the variations in the adjustment which re- 

 sult from a flexible bottom of soft leather. The mercury was 

 boiled in the tube, and the measurement was carefully made on 

 the basis of a new twelve inch ivory scale of accurate workman- 

 ship, made by Elliott, of London, and which is presumed to be a 

 copy of the English standard foot. An experiment was tried with 

 an open syphon tube and cistern, in order lo ascertain the true 

 point for the zero level, including capillarity, in order to avoid the 

 trouble of this correction. The instrument was graduated at a 

 temperature of about 68° Fahrenheit, and the relative capacities 



