the Aninuds of the Deep Seas. 20 1 



tlian the subjacent strata by cooling or increase of its amount 

 of salt, penetrate. 



In my aquarium a downward current, whicli readily carried 

 organic bodies alun^ with it, was produced when the difFerence 

 between the superficial and bottom temperatures had scarcely 

 attained half a dej^rce (K.). In the seas of high latitudes, in 

 autumn aiul winter, ditfercnccs of temperature between the 

 upper and lower strata of tiie water will certainly occur, sulK- 

 ciently great to cause descending currents. 



In the year ISGU I was enabled l)y the captain of the pilot 

 schooner stationed at the mouth of the Ems near the island of 

 Borkum to make some measurements of temperature there, 

 which I may adduce as evidence of the correctness of this 

 assertion. On the 10th September 1SG9, all the strata of 

 ■water there (to a de})th of 13 fathoms) had acquired a tempe- 

 rature of 13° 11. ( = 01^° F.). From the 13th September this 

 began to sink, and in the following manner : almost on each 

 consecutive day the superficial stratum was about half a degree 

 (R.) colder than the bottom sti-atum, until on the 25t]i De- 

 cember a temperatiu-e of only 1° R. (=34^° F.) was found at 

 a depth of 7 fathoms, and at the surface only ^° (=33^° F.). 

 When sea-water begins to freeze, its refrigeration has de- 

 scended to -2°R. ( = 27^° F.). This low tcnq)erature Avas 

 observed in all strata of water in the North Sea at the north- 

 eastern point of Sylt, on the 14th of February 1870*. 



When the temperature of sea-water diminishes, its density 

 increases. Therefore about the middle of September, a de- 

 scending cuiTcnt must have been produced in the mouth of 

 the Ems, and continued until all the strata had accjuircd an 

 equally low temperature. There can be no doubt that in all 

 seas of high latitudes, with a great alteration of temperature 

 in autumn and winter, such descending currents move down 

 from the shore-regions towards the deeps. In the North- 

 Atlantic Ocean they must occur both on the European and 

 North-American coasts far to the south. This appears from 

 the summaries and charts lately published by A. Petermann 

 on the Gulf-stream and the state of thcrmomctrical know- 

 ledge of the Atlantic Ocean and land-district in the year 



• In connexion with, thia I may call attention to a distinction between 

 fresh and salt water which is frequently overlooked. Ordinary sea-water 

 (containing from 3'2 to 3'4 per cent, of salts) only attains its greatest 

 density when it is cooled below its freezing point ( — 2' R.). On becom- 

 ing colder, therefore, it sinks until it meets with a stratum of water of 

 it*j own density, or imtil it reaches the bottom. If it freezes on the way, 

 the fresh water separated as ice rises to the surface, and the sea-water, 

 which lias become richer in salts, and therefore heavier, continues to sink. 



