729 



As we are ignorant of the mean temperature 

 of the epochas of the day and night when these 

 observations were made, on the table-land of 

 Kathmandu, the mean of the barometric heights 

 from 3 h in the afternoon till 4 h in the morning, 

 cannot be directly compared together ; but the 

 observations of M. Dorta * made at Brazil, (the 

 most numerous and complete which have 

 hitherto been published on the horary varia- 

 tions in the southern hemisphere), furnish the 

 possibility of a direct comparison. I have 

 added the mean temperature of the hours ex- 

 pressed in degrees of the thermometer of Fah- 

 renheit. In reducing to the temperature of 

 zero the barometric mean of the following 

 table, we find for 10 h in the morning 

 28 in 2.01 n ; for noon 28 in 1.57 u ; for 4 h in the 

 afternoon 28 in 0.97 n ; for 10 h evening 28 in l.Slii. 

 The extent of the variations is therefore from 

 10 h in the morning till 4 h in the afternoon, 

 2.34 ram ; that from 4 h in the afternoon till 10 u 

 in the evening 1.89 mm . The mean of noon is 

 0.1 7 mm , more elevated than the mean of day, 



* Mem de Acad, de Lisboa., Tom. ii, p. 397—398. M. 

 Dorta having made observations only every 2 hours, we 

 could not present the barometric heights of 9 h in the morn- 

 ing and 11 h in the evening, which I should have preferred. 

 The heights are expressed in inches and hundredths of lines 

 of the French foot, and are not yet reduced to the tempera- 

 ture of zero. 



