110 ADDITIONAL OBSERVATIONS OF THE 



Magnetic Dip at Hudson, Ohio. Latitude 41° 15' iV.; Longitude bJu 26m. W. 

 Place of observation the same as formerly. 



Date. 





Hour 





Needle. 



No 



Readings. 





Dip. 



1839, Oct. 



7th, 



1—3, 



P 



. M. 



No. 1, 





40 



72^ 



' 49'.1 





ti 









No. 1, poles reversed, 





40 





45.0 





ii 









Mean of No. 1, 





80 





47.0 





ii 









No. 2, 





40 





43.8 





it 









No. 2, poles reversed. 





40 





46.0 





(t 









Mean of No. 2, 





80 





44.9 





(£ 









Mean of both needles. 





160 



72 



45.9 



In comparing the preceding observations, some discordances will be perceived. 

 The dilference of the readings in the two magnetic states of the needles is, by- 

 no means, constant. When the marked end of needle No. 1 was a north pole, 

 the average observed dip was 4'.4 greater than when it was a south pole, show- 

 ing a small but real inequality in the weight of the arms. Correcting the dif- 

 ferences for this constant effect, the remainders are quite anomalous, and are 

 subject to no obvious law. Their mean value is a little less than four minutes, 

 and the greatest amounts to ten minutes. These anomalies may be ascribed 

 to various causes, such as errors of observation ; erroneous entries of readings ; 

 influence of currents of air, which, although the observations were all made 

 under a glass cover, are quite appreciable when the wind is fresh; dust and 

 moisture, and, perhaps, loose ferruginous particles collecting upon the needle, 

 and producing a temporary inequality in the weight of the arms : this source 

 of error I always endeavoured to guard against by carefully wiping each needle 

 before the observations, yet it is doubtful if, even with this precaution, its effect 

 was wholly eliminated. Dust and moisture, adhering to the pivots of the nee- 

 dles and to the agate supports, might also have a sensible influence; and, finally, 

 it is, perhaps, credible that there may have been an appreciable diurnal change 

 in the dip during the interval of the observations. Similar remarks apply to 

 the observations with needle No. 2 in its different magnetic states. The mean 

 results of the two needles at a single locality usually differ somewhat; and in 



