VOL. XLVIII.^ PHILOSOPHICAL TRANSACTIONS. 459 



The next trial of this machine was on board a small sailing vessel, in company 

 with Dr. Knight, and Mr. William Hutchinson, an experienced seaman. The 

 expedition was on the river Thames, and some leagues below the Nore. The 

 intention of the trial here, was to find in general, how far it agreed with the 

 log, and how it would behave in the swell of the sea; a comparison with the 

 measure of a real distance being here impracticable, on account of the tides and 

 currents. The method of trial was this: the whole log-line was suffered to run 

 out, being 357 feet between the first knot and the end. The person who hove 

 the log gave notice, at the extremes of this measure, that the person who at- 

 tended the dial of the machine might stop the spindle at the beginning and end; 

 while a third observed, by a second's-watch, the time taken up in running these 

 357 feet. By these means, we were enabled to ascertain the comparative velocity 

 moved, and the number of turns of the plate at each trial, corresponding to 357 

 feet by the log ; which, if the machine and log were both accurate, ought to 

 have been always the same. The particulars of these experiments are contained 

 in the following table. 



Turns of the Seconds of time during the run- Turns of the Seconds of time during the run- 



pUte. ning out of 357 feet oflog-line. plate. ningoutofss? feetof loj-line 



83 In the river at anchor by the tidel24 70 Before the wind at sea .... 56 



82 The sarae rei)eated 134 70 The same 52 



81 Sailing in the river 98 66" Before the wind in the river 55 



79 In the river at anchor by the tidel35 6i The same 53 



76 Sailing in the river 115 64 The same 60 



74 At sea upon a wind 64 64 The sarae 43 



74 The same repeated ... 69 63 At sea upon a wind 53 



71 Sailing in the river 71 62 The same ....•• 52 



70 The same 66 62 Sailing in the river 45 



70 Before the wind at sea 77 



It appears from these trials, made in diflferent positions of the vessel with 

 regard to the wind, both in the river and at sea, as well by the tides at anchor, 

 as in sailing, that the turns of the plate corresponding with the space of 357 feet 

 by the log, were from 62 to 83 ; and the times in which this space was run, were 

 from 45 to 135 seconds, the greater number of revolutions answering to the 

 greater number of seconds, or slower movement of the vessel. On finding this 

 considerable disagreement between the log and plate, when swift and slow mo- 

 tions are compared, Mr. S. did not suppose that they proceeded from a retarda- 

 tion of the plate in swift motions, but from the hauling home of the log in slow 

 ones. For instance, the log, to do its office accurately, ought to remain at rest 

 in the water, whatever be the motion of the vessel. But even the keeping the 

 line straight, and much more the suffering the log to haul the line off the reel, 

 as practised by many, will make the log in some measure follow the vessel, and 



3 N 2 



