33 



into the arctic, north of Europe, in 1922. 



In a general nay, the waters of the central parts of the open 

 basins can be described as extremely stabl3, in their physical 

 character, froui year to year, and over long periods of years, 

 pared to the atmosphere. The close correspondence between te.>- 

 peratures and salinities recorded at several stations in mid-Atlantic 

 by the "Challenger" in 1377-73, and at nearby localities by the 

 "Michaal Sars" in 1910, the "Bache" in 1914, illustrates this funda- 

 mental constancy. Around the oceanic fringes, however, and egpo- 

 cially toward the outer boundaries of the several currents, condi- 

 tions are far less constant, not only seasonally, but as a result 

 of wide scale, but irregular, expansions or contractions in the 

 currents, or of shifts in their relative locations. The most v;idely 

 heralded event of this sort that has corco under human observation 

 in historic times (because its effects or accompaniments both on 

 land and in the sea were destructive) was the abnormal development 

 of the warm current from the north along the west coast of South 

 America in the winter of 1925, accompanied cither by a slackening 

 of the cold Humboldt current (or upwelling) which normally bathes 

 these shores, or, perhaps, by its diversion offshore. During that 

 same winter a westward deviation of the cool Benguela current was 

 reported as similarly accompanied by an expansion tov;ard the south 

 of the warn Guinea current along the west coast of South Africa.^ 

 Spectacular events of the same sort have also taken place in high 

 latitudes, within the memory'" of men now living. Between 1892 and 

 1897, for exan:ple, there occurred what has been described as a 

 "tremendous outburst" of ice from the Antarctic, sending so many 

 floes and icebergs northward into the southern ocean that the 

 traffic between South America, Africa and Australia had to be 

 diverted to more northern tra^cks. A similar outburst of Arctic ice 

 in 1903 is fresh in the memories of Scandinavian fishermen, for it 

 was followed by temporary failure of the cod and herring fisheries 

 along the whole length of Norway, north to south. In that year, 

 Barents Sea was full of pack ice up to May, while ice came closer to 

 the Murman and Finmark coasts than ever before. On the other hand, 

 a great expansion of warm Atlantic water was reported to have tak.en 

 place into these northern seas in the summer of 1922, 



It is true that departures from the norm.al so noticeable as 

 these are rare events, and up to recently it has only been these 

 major departures that have forced themselves on general attention. 

 It has long been known, hov/ever, that smaller fluctuations do take 

 place from year to year in the boundaries and extensions of the warm 

 North Atlantic drift. Similarly, the International Ice Patrol has 

 found that the interrelationships of the Labrador and Gulf Streaxi 

 currents around the Grand Banks are not alike in any two successive 

 years, either in the seasonal schedule, or in the volmnes, te.apara- 

 tures, or velocities of the two currents. And these differences are 

 reflected not onl3'' in the yearly variations in the amount of ioe 

 drifting down past the Grand Banks, but in the tracks followed b:/ 

 the individual bergs. In fact, wherever ocean currents have oo..\e 

 under continuous observation for a period of years, they have been 

 found to var^f, more or less, in a non-periodic, and up to date in an 

 unpredictable way. 



