INTRODUCTION TO THE EIGHTH EDITION. 



The great activity in our field of research, tlie increasing num- 

 ber of laborers in it, and the rich gleanings from it, have made 

 necessary a recast of this work. With new lights and better ma- 

 terials, I have not hesitated to tear down and build up anew when- 

 ever they seemed to require it. The results that are embodied in 

 Plate I. alone of this edition would, had the data for it been col- 

 lected by a force specially employed for the purpose, have de- 

 manded constant occupation from a fleet of ten sail for more than 

 one hundred years. The co-ordinating of these observations after 

 they were made, and the bringing of them to the present con- 

 densed form, has involved a vast amount of additional labor. Of- 

 ficers here have been engaged upon the work for many years. 

 This patient labor has been rewarded with the discovery of laws 

 and the development of truths of great value in navigation and 

 very precious to science. 



It would be presumptuous to claim freedom from error for a 

 work like this : true progress consists in the discovery of error 

 as well as of truth. But I may be pardoned for saying that the 

 present edition of this work will be found to contain more of truth 

 and less of error than any of its predecessors, simply because it is 

 founded on wider research, and based on the results of more abun- 

 dant observations than they. Indeed, it could not, or, rather, it 

 should not be otherwise; for, as. long as we are making progress 

 in any field of physical research, so long must the results continue 

 to increase in value ; and just so long must what at first was con- 

 jecture grow and gain as truth, or fade and fall as error. 



How much more valuable the sum total of the contribution now 

 offered to the stock of human knowledge concerning the physics 

 of sea and air than were the offerings of previous editions, is for 

 the reader, not for the writer, to decide. The fact seems now to 

 be clearly established that the atmosphere is very unequally di- 



