^46 Seccmd Report of the South African Institiitimi. 



Sir J. Herschel also farther stated, that the mean annual ba- 

 rometric fluctuation at Calcutta, on the average of between two 

 ^d three years' observations made by Mr Prinsep, examined by 

 him, appears to be much greater than that at the Cape, and, 

 rwhat is very remarkable, in a contrary direction, the maximum 

 of Calcutta corresponding to the minimum at the Cape. And 

 he attributes this to an actual bodily transfer of a portion of air 

 from hemisphere to hemisphere, by the alternate heating and 

 cooling of the two hemispheres, as the sun crosses from side to 

 side of the equator. The effect of this cause, which he consi- 

 ders to be general over the whole earth, will be to modify the 

 regular and constant effects of the Trades, by a set of periodical 

 winds differing materially in their character from local mon- 

 soons, and to this cause he is disposed to attribute the observed 

 annual oscillation of the extreme north and south limits of the 

 Trade winds. 



The northern hemisphere, he further observed, being, by 

 reason of its greater quantity of land, more superficially heated 

 than the southern, it should be expected that the mean pressure 

 beyond the southern tropic should exceed that beyond the nor- 

 thern, and he suggested this as a subject worthy of examination 

 by meteorologists properly situated in both hemispheres. 



Lastly, he observed, that severe gales, occurring whether in 

 summer or winter, appear to depend on causes entirely extra- 

 neous to the regular periodical fluctuations of pressure, and are 

 probably dependent on causes of a local and transient nature — 

 but that a correspondence of extraordinary seasons in distant 

 • parts of the globe, may be expected to accompany great occa- 

 sional deviations from the usual law of these fluctuations in any 

 given place, and that it is far from impossible that an assiduous 

 attention to this point may ultimately enable us to predict their 

 occurrence. 



The series of observations at the Port-Office being still in 

 progress, the foregoing results are not considered as final ; but 

 whatever modifications future years' observations may necessi- 

 tate, will be from time to time inquired into and reported. 



