VARIATION. 



Bill of all llie places on the globe with which we are ac- 

 qiiaiiited, none exhibit fuch wonderful phenomena in the 

 Variation as the coafts of China, Corea, Tartary, Japan, 

 and Kamtfchatka northward. 



In 17S7, from Macao to N. lat. 41°. E. long. 136° from 

 Paris, the quantity of wefterly variation never exceeded 2° ; 

 and from this laft point to N. lat. 5l|°. E. long. 142!°, 

 where the variation was only 53' E., the quantity never 

 exceeded 3°. 



In 1804, in N. lat. 52°. E. long. 143° from Paris, there 

 was no variation ; i5'^ farther to the E., and 4° to the S., 

 5° 20' of wefterly variation was obferved ; and in 1779, in 

 N. lat.69°55'. E. long. 195° 14' (Bhenng's Straits), there 

 was found 35° 37' E. variation. 



From obfervations made between 1700 and 1775, there 

 appeared three places or points on the globe where the 

 change in the variation was much greater than elfewhere. 



Thefe iNere, JitJ}, in the middle of the Indian ocean, from 

 10° to 15° S. lat., and from 64° to 69° E. long., where the 

 change was 11° and 11° 45'; fecondly, in the Ethiopian fea, 

 from 5° N., to 20° or 25^ S. lat., and from 10' to 15° 

 \>r 20° E. long., the change in the variation was about 

 io°; and thirdly, at 50° N. lat., and between 17° E. and 

 iO° W. long., the change was nearly 1 1°. In thefe different 

 places, the variation has fince continued to increafe at nearly 

 tlie fame rate. 



During the fame interval it was alfo afcertained, that there 

 were four places or points on the globe where the variation 

 has undergone no change. 



Thefe viere,jirjt, from tlie eaftern point of Africa to the 

 fartheft of the Bermuda iflands ; fecondly, the environs of 

 the ifle of Madagafcar and part of Zanguebar ; thirdly, that 

 part of the ocean which is to the S. and S.E. of the Sunda 

 iflands, between them and New Holland ; 2nd fourthly, in the 

 fame fea, about 4° S. lat., and 97° E. long., that is, in the 

 middle of the fpace comprifed between the weftern angle 

 of New Holland and the fouthern point of Africa. In 

 all thefe places, the variation did not vary perceptibly during 

 56 years. And it may be remarked, that obfervations made 

 ilnce the above period, have not (hewn any change worthy 

 of notice in its quantity at the above-mentioned places. 



Some intelligent fea-officers are of opinion, that in the 

 weftern part of the Englifh Channel, the wefterly variation 

 has begun to decline, whilft others aflert, that the variation 

 h ftill increafing in the Channel, and as far weftward as W. 

 long. 15°, in N. lat. 51° ; at which place they fay the va- 

 riation amounts to 30° W. Neither of thefe opinions, how- 

 ever, can, according to Mr. Bain, be relied on as correft, 

 though each may have been deduced from obfervation. If 

 the head of the (hip is on the eq/l point of the compafs at 

 the time of obfervation, from 20° to 23*^ of variation will 

 be obferved ; but on the other hand, if the (hip's head is at 

 the ivefi at the time, the obferved variation will amount to 



30° oi" 33°- 



The circumftance above alluded to, of the apparent varia- 

 tion being fo much influenced by the local attrattion of the 

 Ihip, is certainly of the higheft importance ; and the means 

 which Mr. Bain has adopted of making it generally known 

 are highly laudable : at the fame time, we cannot but feel 

 confiderable doubt, after this faft is once pointed out, of the 



accuracy of many of the obfervations Rated in the preced- 

 ing pages, as feveral of thefe were made by perfons wholly 

 unconfcious of fuch an influence, which feems to have been 

 entirely unknown till Mr. Wales, the aftronomer, who failed 

 with Capt. Cook, firft noticed the phenomenon ; and hit ob- 

 fervations have been fince confirmed by Capt. Flinders ; and 

 Mr. Bain, in the work we have above alluded to, has added 

 many additional fafts to thofe before known, and to which 

 we (hall have again occafion to refer ; but in the firft place, 

 it will be proper to infert a general table of variations in 

 different latitudes, as given in the Philofophical Tranf- 

 aftions for 1757, with additional obfervations of the above 

 author. 



It may not be amifs to add here, as belonging to the hif- 

 tory of this fubjeft, that we owe the firft variation chart to 

 Dr. Halley. Previoufly to this period he had coUefted, and 

 made, a multitude of obfervations on the variation of the 

 needle in many parts of the world, and was enabled to draw 

 on a Mercator's chart, lines (hewing the variation of the com- 

 pafs in the places through which they palled. But as the 

 deviation of the magnetic meridian from the true one was 

 then, as now, fubjeft to continual alteration, this chart was 

 foon found ufelefs. 



However, in 1744, Mountain and Dobfon publilhed a new 

 variation chart, adapted to that year ; which being well 

 received, they publilhed a fecond, adapted to 1756; and a 

 third in the following year : the laft we know of. 



Nicholfon ftrongly recommends the employment of the 

 variation as a means of finding the longitude at fea ; but 

 navigators are long fince convinced of its inadequacy. Van- 

 couver, fpeaking of this fubjeft, fays, " This very able fea- 

 man, Nicholfon, ftill wedded to formerly -adopted opinions, 

 ftrongly recommends the variation of the compafs as a means 

 for afcertaining the longitude at fea ; yet, had we been no 

 better provided, we might have fearched for the Cape of 

 Good Hope, agreeable to his propofitions, to little effedt ; 

 for when we were in lat. 35° 7' S., with 20° 16' W. varia- 

 tion, we had only reached the long, of 6° 30' W. ; and again, 

 when in lat. 35° 22' S., with 22° 7' W. variation, we had 

 only advanced to the long, of 11° 25', inftead of being, ac- 

 cording to Mr. Nicholfon's hypothefis, in the firft inftance, 

 nearly under the meridian of the Cape of Good Hope, and 

 in the fecond, under that of Cape Aguillas ; and it was not 

 until we had nearly 26° W. variation, that we approached 

 the meridian of the Cape of Good Hope. The obferva- 

 tions for the variation were made with the greateft care and 

 attention ; and though generally confidered as correft, they 

 differed from one to three, and fometimes to four degrees ; 

 not only when made by different compalTes, placed in differ- 

 ent fituations on board, and the (hips on different tads, but 

 by the fame compafs in the fame fituation, and at moderate 

 intervals of time ; the difference in the refults of fuch ob- 

 fervations, at the fame time, not preferving the leaft degree 

 of uniformity. Hence the alfertion amounts nearly to an 

 abfurdity, which ftates, ' that with 20° to 20° 10', or 20° 30' 

 wefterly variation, you will be certain of fuch and fuch longi- 

 tudes ;' and it is greatly to be feared, that navigators who rely 

 on fuch means for afcertaining their fituation in the ocean, will 

 render themfelves hable to errors that may be attended with 

 the moft fatal confequences." 



A Table 



