Given below is a table showing the number of ice bergs south of 

 latitude 48° N.; the annual total and the number in each month 

 from 1900 to 1937. 



1900. 

 1901. 

 1902. 

 1903. 

 1904. 

 1905. 

 1906. 

 1907. 

 1908. 

 1909. 

 1910. 

 1911. 

 1912. 

 1913. 

 1914. 

 1915. 

 1916. 

 1917. 

 1918. 

 1919. 

 1920. 

 1921. 

 1922. 

 1923. 

 1924. 

 1925. 

 1926. 

 1927 

 1928 

 1929 

 1930 

 1931. 

 1932. 

 1933. 

 1934. 

 1935. 

 1936. 

 1937. 



May 



July 



Sept. 



Oct. Nov 



222 

 1,024 



50 

 39ft 

 1,019 

 550 

 731 

 468 



54 



38 

 199 

 317 

 445 

 748 

 523 

 236 



11 

 109 



515 



1,351 



475 



13 

 514 

 216 

 576 

 875- 



22 

 470 



38-year average, 423. 



An inspection of the above table shows the months in wliich bergs 

 are most numerous south of Newfoundland and the annual totals 

 bring out strikingly the great variation in the yearly amount of ice. 

 The correlation factor between numbers of bergs and amounts of 

 field ice is very liigh (Smith, 1931, p.. 185). In other words, years 

 which show large berg counts will also be rich in field ice. It would 

 seem then that on the basis of annual totals a direct comparison could 

 be made between any two seasons. This is not true. From the 

 above table or from any similar statistics no idea can be obtained of 

 the distribution of the ice. This is an important factor in determining 

 the relative danger from ice to shipping in any year. To illustrate 

 this the season of 1935 was a very severe year bringing 875 bergs south 

 of latitude 48° N., (over twice the average) and a correspondingly 

 great amount of field ice, but the coast of Newfoundland was remark- 

 ably free of field ice and the east and south coast ports were open the 

 entire year. In contrast, during the 1937 season, which had only 

 470 bergs or a little over half the number in 1935 and proportionately 

 less field ice, the Newfoundland coast was at certain times unap- 



