712 



The barometric heights* are indicated in 

 this table in hundredths of lines (antient mea- 

 sure of the French foot). In order to disengage 

 them from the influence of the temperature, or 

 reduce them to the freezing point, I have added 

 the mean temperature of the limit-hours in 

 which the mercury attains the maximum and 

 the minimum. The periodical movement of 

 the horary variations at Rio Janeiro are, as 

 every where else, 'where the mean of well-made 

 observations can be taken, of the greatest regu- 

 larity. Nor would it have been extraordinary 

 if among 1095 partial observations, published 

 by M. Dorta for the year 1785, more numerous 

 anomalies had been found; for Rio Janeiro 

 (lat. 22° 540 is like the Havannah <lat. 23° 9'), 

 Calcutta (lat. 22° 34'), Canton (lat. 23° 8') and 

 Macao (lat. 22° 12'), near the limit of the torrid 

 zone, where the perturbating influence of the 

 temperate zone begins to be felt. 



The doubts thrown on the regularity of the 

 horary variations of the coast of Brazil, are 

 spread still more to the east, as far as Macao, a 

 spot situated at an equal distance from the 

 equator, in the northern hemisphere. A series 



* We must add twenty-eight inches to every height, so 

 that the mean height of the barometer at Rio Janeiro, at 

 IO h ia the morning, in the month of July was 28 ln 4> li . 

 65. 



