326 Terrestrial Magnetism. 



one plus and the other minus, as it would be merged by both be- 

 ing phis or both minus ; and hence half of the greatest difference 

 by the two needles may be taken as the limit of instrumental 

 errors, which in the above observations would be only 0'.8625 of 

 a minute ; a quantity much smaller than I should have anticipa- 

 ted before making the examination. The instrument is evidently 

 a very perfect one, yet at certain points, when the dip arrives at 

 a particular quantity, probably from a want of perfect roundness 

 of the pivots, one needle will read constantly and uniformly more 

 than the other. Thus at Dubuque, where the dip is 73° 04'', 

 needle No. 1, read in the mean, constantly 2*1'' more than No. 2. 

 Even this error would ordinarily be considered very small. The 

 French have a saying that " the dipping compass is one very un- 

 grateful instrument." But with this line piece of workmanship 

 of Robinson's, I have repeatedly admired the beautiful manner 

 in which the reversals correct all of the errors, and the two nee- 

 dles, none of whose individual readings are alike, will ultimately 

 in the mean, give almost identical results. 



In my surveys, I did not expect the dip to be so little at Prairie 

 du Chien, nor so great at Madison, Wisconsin Territory, as I 

 found it. I went through with four suits of observations at the 

 former place, before I was satisfied of their correctness. But all 

 of the observations between Prairie du Chien and the four lakes, 

 agree in determining that the lines of equal dip along the Wis- 

 consin river, in advancing westward, incline rapidly to the north. 

 From a point about five miles south of Mineral Point, the line of 

 dip for 73° 16' passes to Prairie du Chien, in the direction of 

 west 22° 10' north. The curvature between Madison and Blue 

 Mounds, is probably still more to the north. At Dubuque, how- 

 ever, there was no evidence of such a northern inclination of the 

 lines of equal dip. At Columbus, Ohio, the dip appeared to be 

 so much less than I expected, that after determining it twice at 

 our station, suspecting local attraction, I removed to another a 

 mile distant, but the result was still the same. At the very points 

 marked by Prof Loomis, I anticipated the departure of the mag- 

 netical quantities from their general direction, and was especially 

 cautious in my examinations, but finally was compelled to record 

 their results in obedience to the authority of nature. I believe 

 there are so many anomalies in the elements of terrestrial magnet- 

 ism, that the only safe way in proceeding with our surveys, is to 



