i8 AVOYAGETO Book 1. 



that ufed by Don George Juan ; and as this officer con- 

 tinued his obfervations to the end of the voyage with 

 the fame needle, the difference, which at firft was nega- 

 tive, on the variation changing its denomination became 

 pofitive 5 and from my changing inftruments, the differ- 

 ence on my fide continued always negative. The reafon 

 of this is, that the difference of the five firft obfervations 

 proceeded lefs from a real difference in the variation, 

 than from the poles of the needle, which was fo far from 

 anfwering exactly with the meridian-line on thecompafs- 

 card, that it inclined towards the N. W. ; the contrary 

 happened in the fecond compafs made ufe of, its inclina- 

 tion being towards the N. E. confequently, whatever the 

 angle of that inclination was, it occafioned a propor- 

 tionate diminution in the variation of a contrary fpecies. 



These obfervations, thus compared, fhew the errors 

 to which navigators are liable, for want of attention in 

 making choice of proper needles, which they fhould be 

 careful to procure, not only well made and exa(5f, but 

 alfo ftridly tried with regard to their inclination 

 to the true meridian, before they venture to de- 

 pend upon them in any voyage. In this point Spain is 

 guilty of a notorious negleft, notwithftanding it is evi- 

 dently the fource of a thoufand dangerous errors ; for 

 a pilot, in correding the cdurfe he has fteered, in making 

 vife of a compafs whofe variation is different from the 

 true, will confequently find a difference between the la- 

 titude by account and the latitude obferved and to 

 make the neceffary equation according to the rules moft 

 commonly received in failing on points near the meri- 

 dian, he muft either increafe or diminifh the diftance, 

 till it agrees with the latitude, whereas in this cafe the 

 principal error proceeded from the rhomb. The fame 

 thing happens in parts where it is apprehended there may 

 be currents ; which often occur in failing when the la- 

 titude by account, and that by obfervation, difagree ; 

 though in reality the water has no motion, the difference 

 proceeding entirely fram making ufe of another varia- 

 7 tion 



