5» Method of tneafuring a Ship's Way at Sea, *^ 



feparated to plunge and clear the tube when neceflary (fee note G). The variation of the 

 needle from a polar direftion is another courfe of error in navigation, but the corre(ftion by 

 azimuths and amplitudes affords a perfedt remedy for it, and when the quantity of variation 

 is difcovered, the dead reckoning may be corre£led thereby by well known means ; alfo 

 the variation may be found in a more fimple manner at all times of the day, by annexing 

 any of the univcrfal fun-dials to the compafs box, in fuch a manner that the meridian of 

 the dial may be parallel to its fide, and by adjufting it to the latitude of the place ; for if 

 fuch a dial be turned until the ftiadow points to the hour found by a lately corredted watch, 

 its meridian, and the fide of the box to which it is attached, muft be in a true north and 

 fouth diredtion, and the deviation of the north point of the card from this meridian will 

 be the angle of variation. (See note E) 



Currents are the fources of the greateft errors to which navigation is liable ; the method 

 of finding their fet and drift by the courfe and rate at which a funken body feparates from 

 a drifting boat is erroneous, being grounded on a fuppofition, that this body meets with 

 ftill water, whereas it is believed, that fuch a mafs of water moves in a current together, 

 that our experiments only penetrate its furface ; however, this method feems to be fufficient 

 to determine the exiftence of a current, and the rhumb on which it flows, becaufe every 

 ftratum of moving water runs with a different degree of velocity, an-d this will be the cafe 

 even if the funken body fhould defcend into ftill water, or into a contrary current (and 

 fince interfering currents have never been obferved, we may conclude, that this experi- 

 ment will, in all cafes, prove the exiftence of a current, and {hew the line of its motion, 

 though it remain uncertain to which extremity of that line it may flow. 



If the preceding methods of determining courfe and diftance can be relied on, the error 

 arifing from a current which does not flow in a parallel of latitude may be afcertained by 

 attributing the error of dead reckoning to the line of the currents motion, and then by 

 adding or fubtra£l:ing, as the cafe may be, fuch a fupplementary diftance in the direftion 

 of that line, as will make the dead reckoning correfpond with the latitude by obfervation, 

 which will give the true place of the (hip ; and if the currents be regular and periodical, as 

 is fuppofed, numerous experiments made in this way at various times and places, might 

 afford materials for the conftrudlion of tables, or hydrographic charts, pointing out the 

 drift fet and periods of thefe currents, which would enable the navigator to naakc the ne- 

 ceffary allowance for their effcft in all cafes. 



Alfo if the true direftion of the wind can be afcertained in a current, the lines of 

 its motion being known, the fet and drift may alfo be known, becaufe if the (hip lie to,j 

 the apparent wind will be a compofition of the true ^^ind, and of the wind arifing from the, 

 relative motion of the air produced by the punning water (fee note F), which, in moft cafes, 

 bears a very fenfible proportion to the true wind (fee Encyclopaedia Britt. Vol. XVII. 

 p. 204) i therefore if the veffcl fail along the current line, varying the rate of her motion 

 ttntil the wind vane lie exadly in the direction of the true wind, there will be no current 

 3 wind,. 



