yA 
507 A. EF. Verrill—The Bermuda Islands. 95 
The most marked and remarkable feature in the meteorology of 
February was the unusual persistence of the northwest winds. 
According to the meteorological tables kindly furnished to me by 
Mr. T. G. Gosling, of Hamilton, northwest winds are recorded 
forty-two times in February, 1901, as contrasted with nineteen times 
in 1900. They were continuous for four days, from the 5th to the 
9th, and again six days, from the 13th to the 19th. North and north- 
east winds were also frequent. These northerly winds were usually 
accompanied by a fall of six to nine degrees in the temperature of 
the air, as contrasted with southerly and southwesterly winds. That 
these persistent northerly winds caused currents of cold northern 
waters to impinge upon the shores of the Bermudas can scarcely be 
doubted. Moreover they might easily have caused an upward flow 
of the cold waters that rest against the submerged slopes of the 
islands at the depths of forty-five to sixty fathoms and more, for 
the surface currents, set in motion by the long-continued northerly 
winds, would inevitably also cause an upward flow of the colder 
waters of the submerged slopes, as I have many years ago proved to 
be the case on our own coast. By these combined effects, it is easy 
to understand how the body of shallow warm waters around the 
Bermudas could quickly have been cooled sufficiently to kill the more 
sensitive species of tropical fishes. These would naturally be those 
that habitually live in shallow water and among the sheltered places 
near the shores, where the water is usually warmest. 
I was told by elderly and intelligent persons, who have always 
lived in Bermuda, that no such instance of the death of fishes in 
large numbers had occurred there within fifty to sixty years, or so 
far back as they could recollect. Nor can I find any record of any 
similar event in the early annals of Bermuda. 
Several instances of the death of vast numbers of fishes on the 
Gulf Coast of the southern United States, and especially on the 
west coast of Florida, are on record. The actual causes of the fatal- 
ities in that region are not fully known. In view of the instance 
recorded above, and the famous case of the death of the tile-fishes, 
etc., beneath the inner edge of the Gulf Stream, in 1882, it is not 
improbable that the Florida cases were also due to periods of 
unusually low temperature, acting upon tropical fishes that were 
living at or near their extreme northern ranges. Thus a slight fall 
in the temperature of the water, below their critical point, might 
have been sufficient to kill them, as in the case at Bermuda and in 
that of the tile-fish. 
