70 



Ocean Magnetic Observations, 1905-lG 



The adopted formulse for the logarithms of the intensity constants, resulting from least- 

 square adjustments of all available data, are given in Table 18. 



Table 18. — Inlensily Constants for Sea Dip-Circle 189. 



Discussion of I^ea Dip-Circle Corrections. 



Owing to various mechanical imperfections, unavoidable even in the best construction, 

 values of inclination observed with the dip circle are subject to errors which can not be 

 eliminated by multiplying observations. At any one station the correction for a particular 

 needle and ciixle is found to be constant within the error of observation, except when deterio- 

 ration of the pivot, e. g., by wear or corrosion, causes changes, frequently quite erratic, with 

 time. The usual practice has been to determine the correction for each needle and circle at 

 some base-station by comparisons with standardized instruments. 



For a limited range of dip it is generally found that such corrections are sufficiently 

 close for magnetic-survey purposes, so long as the observed inclinations do not differ from 

 the base-station value by more than 5° to 10° at the most. But when an instrument is 

 used through a lai'ge range of inclination, as was the case on the Galilee, the corrections 

 determined at one base-station can not be assumed to hold. The data obtained from the sea 

 as well as the land work of the Department of Terrestrial Magnetism indicate that the cor- 

 rections vary with inclination and total intensity, and that the variation is more pronounced 

 for sea dip-circles than for land dip-circles. The variation is due probably in part to slight 

 magnetic impurities in the metal of the instrument, and in part to iiTegularities of the pivot, 

 different parts of which are brought in contact with the agates for different inclinations. 



For inclinations observed in the plane of the magnetic meridian according to the abso- 

 lute method, including reversal of polarity, the outstanding error caused by slight magnetic 

 impurities will arise from (a) magnetic effects due to the fixed parts of the instrument, and 

 from (6) magnetic effects due to the movable parts of the circle, viz, the arm carrying the 

 microscopes and verniers, or an equivalent arrangement. Because of (a), the actual hori- 

 zontal component H will be changed into H + h + iH, where h is the effect caused by 

 pennanent magnetization and iH that caused by induction effects from whatever source. 

 The first effect is hkely to be negligible in view of the usual careful tests for magnetic 

 material before acceptance of an instrument, and its rejection if the presence of such 

 material is revealed. Therefore, the entire effect, /; + iH, arising from (a), may be made 

 equal to a constant proportional part of the absolute horizontal intensity, say yH. To 

 consider the effect arising from (fe), resolve it into three rectangular components, one 

 along the longitudinal axis of the movable arm, that being also parallel to the longitudinal 

 axis of the needle, the second nonnal to the face of the needle, and the third perpendicular 

 to the longitudinal axis of the arm and in the plane of inclination. Only the last component, 

 say xF, will affect the inclination. Hence 



H' = H + AH = H + yH + xF sin I 



From similar consideration of the vertical intensity, Z, it follows that 



Z' = Z + AZ = Z + zZ-xF cos I 



