48 THE PHYSICAL GEOGRAPHY OF THE SEA. 



the top ? By running out under this floor or field, rising up over 

 the edges, and flowing back to the middle ? I think not ; on the 

 contrary, I suppose the warm water would rise up here and there 

 in streaks, and that the cold would go down in streaks or seams. 

 The process would be not unlike what we see going on in a fount- 

 ain which is fed by one or more bubbling springs from below. 

 We can see the warm water rising up in a column from the ori- 

 fice below, and in winter the water on the top first grows cool and 

 then sinks. Now imagine the fountain to be a long and narrow 

 stream, and this orifice to be a fissure running along at the bottom 

 in the middle of it, and feeding it with warm water. We can well 

 imagine that there would be a seam of water rising up all the way 

 in the middle of the stream, and that a delicate thermometer would, 

 in cold weather, show a marked difference of temperature between 

 the water as it rises up in this seam, and that going down on either 

 side after it has been cooled. 'Now if we make our imaginary 

 stream broader, and place at a little distance another fissure par- 

 allel with the first, and also supplying warm w^ater, there would 

 be between the two a streak of cooler w^ater descending after hav- 

 ing parted with a certain degree of heat at the surface, and thus 

 we would have repeated the ribbons of cold and warm water which 

 the Coast Survey has found in the Gulf Stream. 



59. The hottest water in the Gulf Stream is also the lightest ; 

 as it rises to the top, it is cooled both by evaporation and expo- 

 sure, when the surface is replenished by fresh supplies of hot wa- 

 ter from below. Thus, in a -winter's day, the waters at the sur- 

 face of the Gulf Stream ofl" Cape Hatteras may be at 80°, and at 

 the depth of five hundred fathoms — three thousand feet — as act- 

 ual observations show, the thermometer will stand at 57°. Fol- 

 lowing the stream thence' ofi" the Capes of Virginia, one hundred 

 and twenty miles, it will be found — the water-thermometer having 

 been carefully noted all the way — that it now stands a degree or 

 two less at the surface, while all below is cooler. In other words, 

 the stratum of water at 57°, which was three thousand feet below 

 the surface off" Hatteras, has, in a course of one hundred and twen- 

 ty or one hundred and thirty miles in a horizontal direction, as- 



