400 EFFECT OF A CHANGE IN THE GULF-STREAM. 



the .... great valley now occupied by the chain of lakes." 

 In this case the Gulf-Stream would no longer be deflected 

 by the American coasts, but would pass directly up this 

 channel into the Arctic Sea ; and as a very great ocean current 

 must have its counter current, it is probable that there would 

 be a flow of cold water from the North between the coasts 

 of Norway and Greenland. The absence of the Gulf-Stream 

 would probably lower the January temperature of Western 

 Europe ten degrees, while the presence of a cold current from 

 the North would make a further difference of about three 

 or four degrees,* an alteration of the climate which would 

 apparently be sufficient to account for all the phenomena. 

 This theory Mr. Hopkins considers as no mere hypothesis, 

 but as necessarily following from the submergence of North 

 America, which has been inferred from evidence of a different 

 nature. 



In this case, of course, the periods of great cold in Europe 

 and in America must have been successive, and not synchro- 

 nous ; and it may also be observed, that in this suggested 

 deflection of the Gulf-Stream, Mr. Hopkins was contemplating 

 a period anterior to that of the present rivers. For if we are 

 to adopt this solution of the difficulty, an immense time would 

 be required. If, when the gravels and loess of the Somnie and 

 the Seine were being deposited, the Gulf-Stream was passing 

 up what is now the valley of the Mississippi, then it follows 

 that the formation of the loess in that valley and its delta 

 an accumulation which Sir C. Lyell has shown to require a 

 period of about 100,000 years would be subsequent to the 

 excavation of the Sornme valley, and to the presence of man 

 in Western Europe. 



The deflection of the Gulf-Stream from our coasts niisht, 



o 



however, be owing to another cause, namely, a subsidence of 

 the Isthmus of Panama ; in support of which suggestion may 



* Hopkins. 1. c. p. 85. 



