PLACES IN THE STATE OF OHIO, ETC. 271 



by No. 1, was 71° 23' 



by No. 2, 71 22-5 



Mean, 71° 22' -75 



Time, 9 to 11, A. M.; temperature 70°, F. 



At Springfield, latitude 39° 53' N., longitude SS^' 46' W., March 

 29, 1838, the dip 



by Needle No. 1, was, 71° 26' 

 by « No. 2, 71 28 -75 



Mean, 71° 27-375 



Time 6 to 8 o'clock, A. M. ; temperature 53°, F. 



Professor Lloyd, in the account of his "Magnetical Observations in 

 Ireland," points out, very clearly, the fact that there is, in some dip- 

 ping needles, "a source of constant error, which remains uncorrected 

 by the various reversals usually made." He proposes to ascertain this 

 error, and "apply it as a correction to all future results within certain 

 limits." From my observations at Springfield, I became satisfied that 

 the discrepancy between the results with the two needles, 2*'75, arose 

 from a want of perfect roundness in the pivots of one needle ; for it 

 showed itself only at one of the reversals of polarity, and totally disap- 

 peared at other places, where the dip was either a little more or a lit- 

 tle less, so as to throw the pivot on another point of bearing. Such 

 mechanical errors would be expected ; yet when they are so small as 

 above, they are scarcely worth noticing, unless to point out their nature. 

 If the above view is correct, the " error" is far too limited in its opera- 

 tion to justify the application of a correction which had been made at 

 any one place, to observations made at another. When it amounts to 

 as much as "twenty minutes," it certainly shows a needle of bad me- 

 chanical qualities. 



At Urbana, latitude 40° 03' N., longitude 83° 44' W., March 30, 

 1838, the dip by needle No. 1, was 71° 30'-44 ; by No. 2, 71° 29'-44. 

 Mean, 71° 29''94. 



At Columbus, the seat of government for the state of Ohio, latitude 

 39° 57' N., longitude 83° 00 W., I had expected the dip to be nearly 

 VI. — 3 s 



