Variation and Dip of the Magnetic Needle. 157 



belong chiefly to the older fossiliferous rocks below the coal meas- 

 ures. Granite is rare, but the gneiss and mica slate system is well 

 developed and occupies a considerable area in the northeast. The 

 carboniferous system is absent. The new red sandstone system 

 is very partially developed in the southeast. The oolitic and 

 cretaceous systems are entirely wanting. So also is the tertiary, 

 according to the usual acceptation of the term. 



Aet. XVIII. — The Variation and Dip of the Magnetic Needle 

 at Nantucket, Mass. ; by William Mitchell. 



I am not aware that the dip of the magnetic needle at Nantuck- 

 et has ever been published ; nor am I aware that this element or 

 that of the variation, has, till recently, been obtained in a manner 

 having claim to accuracy; yet a glance at the map of the Amer- 

 ican coast, shows it, at once, to be an important magnetic position. 

 In the year 1834, I obtained the variation of the needle by a 

 series of observations of the sun's rising and setting amplitude, 

 and also of his azimuth at equal altitudes before and after noon, 

 and the result was 8° 27' westerly. During the years 1837 and '8, 

 it was repeatedly obtained, and near the close of the latter, stood 

 at 9° 02' 19". In the summer of 1842, I established a meridian 

 line on an open plain with stations fourteen hundred feet asunder, 

 and remote from all visible local magnetic influences, and, with 

 a long delicately formed needle, carefully suspended, made a 

 great number of observations, principally during the months of 

 August and September, the means of which showed a variation 

 of 9° 9'. The amount of variation for the month of September, 

 1843, was obtained by the following method, which though far 

 more laborious, is, when performed with care, decidedly the most 

 satisfactory. From each station of the meridian line, the direc- 

 tion of the needle was referred by means of sight vanes, to a 

 movable mark equidistant with the opposite station of the meri- 

 dian, and the angle thus indicated by each settled position of the 

 needle, was carefully measured with a sextant, properly adjusted, 

 allowance being in each case made for the parallax of the instru- 

 ment. An equal number of observations was made before and 

 after noon at each station, and the mean of all, viz. 9° 09' 59", 



