732 ANNUAL REPORT SMITHSONIAN INSTITUTION, 1912. 



circulating water of the engines at a point within a few feet of the 

 intake mouth, situated about 16 feet below the water line. The wires 

 passed through the engine room to Chief Engineer McQuitty's office, 

 where the recorder was situated. The course followed was south of 

 Cape Race to the south of Ireland. 



The result of these experiments brings out several points of impor- 

 tance. The iceberg effect was clearly obtained when passing, early 

 in the afternoon, into the ice track. Figure 5 shows this very well, 

 even in water nearly at 32° F. The ship's course was altered here 

 to avoid a large berg, and when abeam of two immense icebergs, situ- 

 ated approximately 10 miles on either side of the ship, the micro- 

 thermometer showed its minimum reading, and as the bergs were left 

 behind the temperature rose rapidly to a maximum, fell off again, and 

 then rose steadily. No further ice was encountered until early the 

 following morning, when several icebergs were passed in water measur- 

 ing approximately 37° F. 



The sudden change of temperature on passing out of the Arctic 

 current into the Gulf Stream was clearly marked. Here a rise of 

 temperature of nearly 10° was recorded in a little over half an hour. 

 The great steadiness of the temperature of the Gulf Stream was re- 

 markable, since for hundreds of miles the variations were not more 

 than a quarter of a degree. The complete absence of any diurnal 

 variation of temperature was clearly marked. 



The tests have shown that large variations of sea temperature are 

 caused only by some abnormal condition. Thus land affects it, ice- 

 bergs produce characteristic disturbances, and current boundaries are 

 clearly shown. The existence of a submarine hot spring was indicated 

 when passing over the great wall of the continental shelf about 400 

 miles from the Irish coast. The bottom of the ocean rises quickly 

 here from about 3 miles to one-third of a mile. Just over this wall 

 the temperature rose rapidly to a sharp peak about 1 J° warmer than 

 the surrounding sea, and immediately fell again. Figure 6 shows this 

 disturbance very clearly marked. It shows the peak superimposed 

 on the small variations characteristic of the open sea. 



Nearing the Irish coast the variations of the sea temperature 

 became more marked, and this commenced at a distance of 200 miles 

 from the shore. Figure 7 illustrates this very well, where the vari- 

 ations may be seen to be much larger than in the previous plate. 

 Figure 8 shows the effect of drawing near the Fastnet lighthouse, 

 which was passed at a distance of about 4 miles. Here the tempera- 

 ture rose as in the iceberg effect, and then fell as the ship was abeam 

 of the nearest point of land. The temperature then is seen to rise, 

 as land was left behind, through the Irish Sea. The approach to 

 Lund}^ Island caused the temperature to rise rapidly again and then 



