148 



TABLES OF OBSERVATIONS, ^'C. 



REMARKS. 



The experiments on George's Bank were too few to be 

 satisfactory. Besides, several of the temperatures in the 

 foregoing Table were not properly on George's Bank, but 

 to the southward of it, where the Gulf Stream often extends, 

 and the water is consequently always much warmer. It 

 will be found to be uniformly the case, that the water is 

 much colder on banks, than on soundings shelving gradu- 

 ally from the land. In summer, the difference of tempe- 

 rature of the water, on and off soundings, on the coast of 



