Mr. J. Croll on the Physical Cause of Ocean-currents. 171 



If a very large proportion of the heat possessed by the Gulf- 

 stream be not derived from the southern hemisphere, these 

 figures show that the Atlantic, from the equator to the tropic of 

 Cancer, should be as cold as from the tropic of Cancer to the 

 North Pole. But independently of this, a mere glance at a 

 chart of ocean-currents will show that the Gulf-stream is chiefly 

 fed by a current from the southern hemisphere. Without such 

 a transference of heat it would be impossible to account for the 

 N. Atlantic being actually 5 degrees warmer than the S. Atlantic. 



Again, Dr. Carpenter remarks : — " In computing the heat 

 imparted by the sun to the equatorial area from which the Gulf- 

 stream is fed, Mr. Croll assumes that the heat, being wholly 

 taken up by the water of the ocean, is transferred by its cur- 

 rents towards the polar regions; whilst of the heat which falls 

 upon the land, a very large proportion is lost by radiation, 

 passing off into the stellar spaces" (§ 102). 



But this cannot in any way affect the correctness of the result 

 of my computation of the amount of heat conveyed by the Gulf- 

 stream. What I have maintained in my papers is, that the 

 quantity of heat conveyed by the winds from intertropical land 

 is trifling to that which is conveyed by currents from inter- 

 tropical seas. Dr. Carpenter says that " the heat lost by eva- 

 poration from the sea must be far greater than that lost by 

 radiation from the land." According to the laws of radiation 

 and absorption, all the heat received from the sun by the land 

 must be either reflected or radiated from its surface, with the 

 exception of the small portion which is communicated to the 

 air in contact with that surface. In fact it is by radiation that 

 the sea as well as the land loses the greater part of its heat — the 

 only difference in the two cases being, that heat radiated from 

 the sea is absorbed more readily by the air than heat radiated 

 from the land, and consequently produces a greater influence 

 on climate. 



Dr. Carpenter continues : — " Mr. Croll leaves almost entirely 

 out of the question the N.E. transportation of an enormous 

 amount of heat from the general surface of the Atlantic by the 

 agency of the aqueous vapour thus raised ; although the im- 

 portance of this agency has been insisted on by the most emi- 

 nent authorities in meteorology " (§ 103) . 



Here again, however, my estimate of the heat conveyed by 

 the Gulf-stream., or the effects which it produces, cannot pos- 

 sibly be affected by the above consideration. It seems to be 

 forgotten in this objection, that, were it not for the Gulf-stream, 

 the quantity of heat which could possibly be derived from the 

 Atlantic would be so much the less by an amount equal to that 

 conveyed by the stream. Besides all this, there may be other 



