Meteorological jOhservations. 267 



Observations at Canton. 



If we now turn to the observations at Canton, we shall find the 

 mean half yearly results for the same months, for ten years, as 

 follows : viz. 

 Mean pressure of ten half years, November to April 



inclusive, . . . . 30. 140 inches. 



Mean pressure of ten half years. May to October 



inclusive, .... 29.868 inches. 



Thus, at Canton the mean pressure of the winter half year 

 exceeds that of summer, 0.272 inch, or more than one fourth of 

 an inch. The latitude of Canton is 23° 07' N., being nearly in 

 parallel with the north side of Cuba and part of the Bahama Isl- 

 ands : Ion. 113° 14' E. The mean height of the barometer for 

 the ten years observed at Canton, is 30.005 inches ; but if we 

 add a correctiou for an assumed elevation of 40 feet above tide, 

 it will be 30.051 inches. We may infer, therefore, that there is 

 little error in the adjustment of the scale of inches in this ba- 

 rometer. 



Assuming the same mean annual pressure for Canton as at New 

 York, the following comparison of the half yearly results at these 

 two places may be instituted. 



Winter half year. Summer half year. 



Mean at New York, 30.133 30.089 



Mean at Canton, as adopted, 30.246 29.974 



Excess at Canton, 0.113 in. Excess at N. Y., 0.115 in. 



The inequality at these places for the two periods, being the 

 same in kind, but differing in degree by more than one ninth of 

 an inch ; the inequality being greatest at Canton. 



If the results at St. Johns, Newfoundland, be compared in like 

 manner with those at Canton, they will appear as follows : 



Winter half year. Summer half year. 



Mean pressure at Canton, 30.246 in. 29.974 in. 



Mean do. at Newfoundland, 30.039 30.184 



Excess at Canton, 0.207 in. Excessat N. F., 0.210 in. 



These results are different in kind and the inequalities greater 

 than those between Canton and New York : the mean inequali- 

 ties being more than one fifth of an inch. 



