695 



mount ; or, (which is more theoretically exact, 

 supposing changes that are unperceived by 

 our senses, and alike rapid on both sides of the 



summit), we may indicate ^ =4 h 45', as 



the real epocha of the minimum. I have learnt 

 from long experience, that there is often more 

 regularity in the period with respect to the 

 hours of the apparent maximum and minimum, 

 than in the duration of the stationary state. 

 The apparent minimum was attained in South 

 America, for instance, very uniformly during 

 whole months, at from 4 h to 4 h 15' in the after- 

 noon, but at the same season the barometer rose 

 visibly, sometimes from 5 h , sometimes from 6i h . 

 I have therefore, in my tables, placed the signs 

 of the maximum and the miiiimum (+ and — ) 

 near the hours when the mercury appears to 

 have attained the concave and convex summits 

 of the curve. It would have been impossible 

 for me to express by the half-sum of equal 

 heights, the moment that corresponds thoreti- 

 cally to the real summit, my occupations not 

 having permitted me to do more for discover- 

 ing the extent of the variations, than observe at 

 the hours when the barometer attains its appa- 

 rent maximum, or minimum. According to this 

 remark, the assertion of Dr. Balfour, that the 

 mercury has a prevailing tendency to descend 

 from 10 h in the morning till 6 h in the evening, 



