6 THE ROYAL SOCIETY OF CANADA 



shown than those which occurred during the first two months of the 

 past winter. Inspection of maps of mean atmospheric pressure 

 show that during both December and January the cyclonic areas 

 in both Pacific and Atlantic were abnormally pronounced and those 

 in the Pacific encroached to an unusual degree into the region con- 

 stituting Northern Alaska and the Yukon. 



Now I have before stated that we do know something of the con- 

 ditions most favourable for cyclonic development on the eastern 

 coast of America and it is permissible to assume not only that some- 

 what similar conditions lead to cyclonic development on the east 

 coast of Asia, but further that any increase in the warmth of the ocean 

 currents will lead to intensification of the cyclonic areas. 



The Canadian Service now has under investigation the question 

 of possible changes in the temperature and position of the ocean 

 currents in both Atlantic and Pacific. From time to time it is re- 

 ported that the warm waters of both the Gulf Stream and Japan 

 current are found further north or further south than normal and 

 there seems to be good foundation for such reports. Any variation in 

 these is probably due to the variations in the strength of the Trade 

 Winds during previous months. 



In Nature, 1905, Dec. 21, Dr. W. N. Shaw finds connexion 

 between the strength of the Trade winds at St. Helena and the rain- 

 fall of England. It seems a reasonable hypothesis that the connexion 

 is through the medium of the ocean currents. 



It is an established fact that the Trade winds of the tropics 

 do vary in average strength from year to year and this variation will 

 certainly be in agreement with variations in barometric gradient: 

 whence this change? It is only within the past few years that observa- 

 tion has proved that the solar constant is a variable. Dr. Abbott in his 

 address before the Philosophical Society of Washington on January 

 3rd, 1914, says — "During the whole solar constant campaign from 

 1902 to 1913, about 700 measurements of the solar constant of radia- 

 tion have been obtained, all but three of the values ranging between 

 1.80 calories and 2.10 calories. The range of these numbers is mainly 

 attributable to the actual fluctuation of the sun itself, though part, es- 

 pecially in Washington work, is due to accidental errors of measure- 

 ment. The mean value from 690 measurements is 1 . 933 calories 

 per square centimetre per minute. It is believed that this number 

 represents the average value of the solar constant of radiation for the 

 epoch 1902 to 1913 within 1 per cent." 



"Besides the short irregular fluctuation of solar radiation above 

 mentioned as having been shown by the simultaneous measurements 

 at Mount Wilson and Bassour, Algeria, it appears that a long period 



