32 



SCIENCE 



[N. S. Vol. XXX. No. 757 



ai every 20° of dip from + SO" to — 80°, with 

 interchange of instruments on each dip, to elimi- 

 nate station difference. Curves were shown repre- 

 senting the corrections thus determined for the 

 four needles of a Dover land dip circle recently 

 purchased, which is found to be an exceptionally 

 good one. The corrections range from + 0'-5 to 

 — 2'.8. The similarity of the curves for the four 

 needles suggests that part of the error is inherent 

 in the dip circle rather than in the needles. The 

 station difference showed a regular variation with 

 the dip, following approximately a sine curve. 

 Possible variations in method and sources of 

 error were discussed. 



Ship dip circles in general exhibit a consider- 

 ably larger variation in the correction with vary- 

 ing dip and force than the land instruments. 

 The Carnegie Institution Marine Collimating 

 Compass: J. A. Fleming, of the Carnegie In- 

 stitution of Washington. 



The experience gainea on board the Magnetic 

 Survey yacht Galilee indicated the necessity for 

 developing an instrument of greater precision for 

 determining magnetic declination at sea. The in- 

 strument exhibited was devised by Mr. W. J. 

 reters, who is in charge of the ocean magnetic 

 survey work of the institution. 



The compass is of the liquid type. Four 

 spherical mirrors are attached to the buoyant air 

 chamber at intervals of ninety degrees, their op- 

 tical centers being in the horizontal plane through 

 the point of suspension. In the focus of each 

 mirror there is a scale of thin, blackened German- 

 silver wire. The optical arrangements are such 

 that homoeentrie rays from the scales are reflected 

 through the dilute alcohol in which the compass 

 is mounted and issue as parallel rays into the air 

 through windows exhibiting the scales each as 

 seven luminous points one degree apart. The 

 windows are segments of spherical shells, the cen- 

 ters of which are at the point of suspension so 

 that the optical conditions are not altered by the 

 rocking of the bowl and other motions at sea. 

 The bowl swings in a perforated gimbal ring sup- 

 ported in a spindle-bearing cylinder with gradu- 

 ated base, thus making it possible to quickly 

 orientate the optical systems. The usual form of 

 binnacle stand with suitable alterations is used. 

 The method of observation calls for the meas- 

 urement of the angle, a, between the collimator 

 and a celestial body of known altitude, h. As the 

 angle, c, between the zenith and the collimator is 

 ninety degrees, the fundamental formula takes the 

 form cos A = cos a ■ see h where A is the mag- 



netic azimuth of the celestial body. (For small 

 values of A more accurate results may be obtained 

 by use of formula involving the tangents of half 

 the sum and difference of a and h.) Errors in c 

 due to rocking may be eliminated by suitable 

 arrangement of a series of observations. Lack of 

 perfect adjustment of collimator producing error 

 in c of do which may be easily measured and 

 the resulting error in A corrected by dA = 

 — cot B ■ do, where B is the angle between the 

 vertical plane and the great circle passing through 

 the collimator and the celestial body. Owing to 

 the trigonometrical relations involved, the condi- 

 tions are not good for values of a less than 45° 

 or more than 135° for the usual upper limit of 

 altitude of 15°; it is for this reason that four 

 collimators are used, as then suitable selection 

 may be made. 



A special instrument for measuring the angle 

 a has been constructed, although the usual form 

 of pocket sextant may be used for this purpose. 

 This instrument, which may be called a circle of 

 reflection, makes use of the law that the incident 

 and reflected rays lie in the same plane normal 

 to the mirror surface; the construction is such 

 that the angular motion of the mirror is equal 

 to the angle measured. 



The advantages are: motion is practically re- 

 stricted to the oscillation of the magnets; the 

 celestial body observed upon and compass scale 

 are seen simultaneously and all the observed 

 quantities refer to the instant of observation; 

 there are no movable parts subject to wear; 

 graduation errors are limited to twenty-eight 

 divisions, a number so small that each may be 

 separately examined; error of eccentricity affects 

 only the distance between mirror and correspond- 

 ing window, hence the focal distance and conse- 

 quently the scale values, but does not alter the 

 constant for the middle of each scale. By the 

 attachment of a suitable standard centrally to the 

 bottom of the bowl horizontal intensity deter- 

 minations by the method of deflections introduced 

 by Dr. L. A. Bauer can be made and from the 

 same measures accurate values of declinations 

 may be derived. 



The instrument is not intended for navigational 

 purposes, but it is hoped that it may give much 

 greater precision in the determination of declina- 

 tion at sea for use in the more careful study of 

 the laws governing the distribution of magnetism 

 over the globe. 



R. L. Faris, 

 Secretary 



