RECORD AND DISCUSSION OF ATMOSPHERIC PRESSURE. 107 



January, 1855. 20th, 21st, 22d, 23d, and 24tli. The temperature readings for these days have bein 

 taken from page 422 of the second volume of the Narrative. 

 25th, 2Gth, 2Tth, 2Sth, 29th, 30th, and olst. The barometer and temperature readings have been 



taken from the same page of the Narrative. 

 The hist horizontal column of the reduced barometer readings is derived from the preceding column 

 by subtracting 0.024 inch, so as to allow for the introduction of the means of the last seven days. 

 February, March, and April, 1855. For these months the original record could not be found. The 

 daily and monthly means of the atmospheric pressure (and corresponding temperature), will 

 be found in Appendix No. XII. of the second volume of the Narrative. 



Diurnal Variation of the AfmosjiJierlc Pressure. — The following table exhibits the 

 diurnal change of the barometric pressure, for each month of the year and for the 

 "W'hole year, as made out from the preceding abstract. For the months between 

 September and January (inclusive), the mean from the two sets is given. 



Owing to the small amplitude, the comparatively short period of observation, 

 and the magnitude of the occasional disturbances, the law of the diurnal variation 

 is apparently subject to considerable fluctuations ; and it has, therefore, only been 

 attempted to express the figures in the last vertical column, or the mean variation, 

 analytically. Using Be^sel's formula, the variation can be expressed by the for- 

 mula: — 



Inches. Inch. I"<'1'- 



29.76-5 + 0.0034 sin {6 + 290°) + 0.0022 sin (2 B + 204°); 

 the terms containing 3 6, etc., becoming too small to have any real value. The 

 angle counts from noon, and is expressed in degrees at the rate of 15° g-n hour. 



