EXPLANATION OP OCEAN TEMPERATURE CHART No. 2. 



DRY TOETUGAS LIGHT-HOUSE, FLORIDA. 

 Observer: Robert H. Thosipson. 



Location of station.^The Dry Tortugas liglit-house is built on Loggerhead Key, the westernmost island of the 

 Tortugas Reefs. These reefs are a westerly extension of the Florida Reefs, and are located about 1 degree from Key 

 West. Loggerhead Key is about three-fourths of a mile long, and less than one-fifth as wide. It is situated on the 

 southeasterly side of an elongate bank, bearing the same name, and trending in a northeasterly and southwesterly 

 direction. The bank is about 5 miles long, and has an average width of three-fourths of a mile, inside of the 3-fathom 

 line. The Southwest Channel, with depths of 10 to 13 fathoms, separates Loggerhead, Bank from the Bird, Garden, 

 and Long Key Bank on the east. Strong tidal currents set through it, running northeast on the flood and southwest 

 on the ebb. The channel between the Tortugas and Cuba is about 90 miles wide and is occupied by the Gulf Stream, 

 the axis of which approaches much nearer the Cuban coast than the Tortugas. The depth of water in this channel 

 exceeds 1,000 fathoms in some places, being greatest in its southern part. The 10-fathom curve passes close by the 

 eastern side of Loggerhead Key, and the 100-fathom curve is distant only about 20 miles to the southward. 



(jfeographioal position of tlie Zifirft^-Tiouse.— Latitude, 24° 38' 04" N". ; longitude, 82° 55' 42" W. 



Depth of water.— Thoi depth of water where the observations were taken is 5 feet at mean low tide. 



Range of temperature.— Axx, 21°.5 (67° to 88°.5); surface, 20°.5 (65°.5 to 86°). 



The depth of water where the observations were taken is probably too little, and the locality too much sheltered 

 to afford satisfactory results respecting the temperature of the open waters surrounding the Keys. That such is the 

 case will appear evident on comparing the temperature chart for the Tortugas with those for Carysfort Reef and 

 Fowey Rocks. On the first mentioned, the more direct influence of the air upon the water temperature is made 

 apparent by the manner in which each fluctuation in the curves of air temperature is repeated in the curves of water 

 temperature. At the two more northern stations, although the depths are no greater, the curves of water temperature 

 present fewer angles, probably due to the more open exposure of the places of observation. 



The range of temperature given above is for 1881, that year having afforded the greatest extremes in temperature 

 of any plotted. There is comparatively little difference between the air and surface temperatures at any period, and 

 great uniformity in the curves of surface temperature during the summer and early fall mbnths of all the years 

 excepting 1834, in which year both the air and surface temperatures were constantly from .3 to 10 degrees lower 

 between April and the middle of October. The collateral observations fail to explain the cause of this variation, the 

 records for direction and force of the winds showing that essentially the same conditions in those particulars prevailed 

 during 1884 and 18S5, though both of those years differed more or less from the three preceding ones. 



Between March 21 and April 10, 1831, and between November 26 and December 6, 1882, the temperature of both 

 the air and water fell far below that for the corresponding periods of other years, affording the lowest observations 

 recorded. The cause of these extreme variations, which are so clearly brought out by the curves of temperature on 

 the chart, may be explained by the fact that during those two periods the winds were almost constantly from the 

 north and northwest, and blowing fresh ; the prevailing winds for the months in question are generally northeasterly. 

 Low temperatures were recorded during the same periods at Carysfort Reef and Fowey Rncks, the next stations to 

 the north.- 



Tahle showing the direction of the winds, iy quadrants, for each month of the year, ieing the means of five years' obseivations 



178 



