1858.] Meteorological Observations on Parisnath Hill. 27 



that of the air, in the latitude 23° 15', both being compared with 

 water, we obtain the value of p — 10535.8 (Log p = 4.0226691.) 

 If the height of a column of air H = 1, then the height of a column 



of Mercury of the same weight and of an equal basis is - ; 



a has the same value as in the last equation (Log. a = 

 2.4359260) 



t stands for the real mean temperature of the air. 



b stands for the mean barometrical pressure between the sta- 

 tions. 



B is the normal barometrical pressure at 0° C. B = 29 .9218. 



If we put — — = A, the equation becomes 

 pB 



S = A— L- b 



a-f t 



and introducing the numerical values we find A = 41.95234 (Log. 

 A = 1.6227562). 



The values for t were taken from the table of real mean tempera- 

 tures. The mean pressures (b) were found by taking the pressure at 

 the lower station as the first, and the pressure at the upper station 

 as the last term of a geometrical progression, and the height in feet 

 as the number of terms. The sum of all terms, divided by the 

 height in feet would approach the mean pressure sufficiently near 

 for our purpose.* To obtain the sums, it was necessary first to 

 fiud the common ratio, r, of the progression of each separate hour. 

 In the Appendix I have given the Logarithms of the hourly values 

 of r as also the values of r — 1, and the resulting mean pressures. 



The final coincidence of the results of the calculation ofS by 

 means of this formula, with the observed barometrical differences, 

 will depend on the correctness of the determination of H, t and b. 



* I have, in calculating the mean barometrical pressures, neglected to reduce 

 the height of the barometer at the upper station to the gravity at the level of 

 Calcutta. The correction would alter the mean pressure so little (less than 

 0.001 inch) that it would form only an insignificant part of the error, which, as 

 I have ascertained, must be allowed in the value obtained by the method I 

 employed, amounting to about + 0.005 inches. The correction would have no 

 appreciable effect on the value of S. 



E 2 



