belyse Magnetens Skala, maatte en særskilt Tlteodolit an- 

 vendes til Bestemmelse ai' Azimut. Theodoliten opstilledes 

 i 1876 paa sit eget Stativ med sin Kikkert i samme Højde 

 som Magneten og saa nær denne som muligt. Da dette 

 voldte meget Bryderi, lod jeg forfærdige paa Hortens me- 

 kaniske Verksted et Underlag ai' Messing, der kunde lægges 

 paa Maunctometret og læstes til dette ved Hjælp af de 

 samme Indretninger, som anvendtes veil Deflektionsstangens 

 Befæstigelse. Paa den ene Side af dette Underlag an- 

 bragtes Theodoliten, med Fodskruerne i smaa dertil af- 

 passede Huller, og paa den anden Side en Modvægt af 

 Bly. Det var med denne Indretning altid let at faa se 

 og kunne indstille Magnetskalaens Midtstreg paa Theodo- 

 litens Vertikalfilament. ( (bservationerne udførtes i Regelen 

 paa følgende Maade : 



( )])hængningstraadens Torsion ophævedes. Theodoliten 

 nivelleredes. og indstilledes med Filamentet paa Skalaens 

 Midtstreg i Magneten, hvoretter Theodolitens Nonier af- 

 læstes (Magn. I). 



Theodoliten drejedes, saaledes at dens Kikkert pegede 

 lidt vestenfor Solen, i Solcentrets Højde, og fastklemtes. 

 Tidspunkterne for Overgangen af forangaaende og etter- 

 følgende Rand af Solen over Vertikaltilamentet noteredes 

 efter Kronometer. Denne Iakttagelse gjordes enten med 

 Blændglas foran Okularet eller ved at projicere Solens og 

 Filamentets Billede paa en hvid Skjærm. I mange Til- 

 ladde var. der to Iagttagere. af hvilke den ene observerede 

 Solrandenes Passage og raabte -'Ku'' i det Øjeblik. de 

 tangerede Filamentet. medens den anden observerede og 

 noterede tilsvarende Øjeblikke efter Kronometret. Nonierne 

 aflæstes. 



Kikkerten lagdes om gjennem Zenit (Nadir), drejedes 

 L80° om Vertikalaxen og begge Solrandes Passage obser- 

 veredes i denne Stilling efter Kronometret, ligesom Noni- 

 erne atfestes. 



Magneten omlagdes, idet den drejedes 1.80° om sin 

 Længdeaxe, og Filamentet i Theodolitkikkerten indstilledes 

 atter paa Magnetskalaens Midtstreg. og Nonierne atlæstes. 

 (Magn. II). 



Naar Omstændighederné tillot! det. observeredes atter 

 Magneten i Stillingen I. Stundom blev den omlagt flere 

 ( lunge. 



Ligesaa dyttedes, ved saadanne Lejligbeder, Theodo- 

 liten ved den følgende Sats saaledes, at Fodskruerne kom 

 hver i andre Huller, hvorved deus Cirkels Nulpunkt for- 

 andredes 120°. Det viste sig imidlertid, at dens Delings- 

 fejl ikke var saa betydelige, at denne Vexling af Stilling 

 var nødvendig lor at opnaa den forlangte Nøjagtighed af 

 c. 1 Minut. 



I flere Tilfælde bestemtes Azimut af en Mire, til 

 hvilken Observationen af Magneten i begge Stillinger blev 

 knyttet. 



Kronometrets Stand blev i de Heste Tilfælde be- 

 stemt ved Solhøjder tanne med Sextant paa Observations- 



having only a concave mirror by whieh to illuminate the 

 scale of the magnet, the azimuth bad to be determined 

 with a separate theodolite. In 1876, the theodolite was 

 mounted on its own stand, with the telescope facing the magnet, 

 and as near it as possible. This. however, prbving exces- 

 sively troublesome, I procured from the Horten Mechanical 

 Works a brass support, which. when placed upon the 

 magnetometer, could be attached in the manner adopted for 

 tixing the deflection-rod. On one side of this brass support 

 was mounted the theodolite, with the foot-screws fitting 

 into holes made for the purpose, the other side being given 

 a counterpoise of lead. With this arrangement, the scale 

 of the magnet could be easily sighted. and the middle 

 division bisected by the vertical. wire of the theodolite. 

 The observations were generally tåken as follows: — 



The torsion of the suspension thread having been 

 first removed. the theodolite was levelled. and the wire 

 made to bisect the middle division of the scale of the 

 magnet, after which the veruiers of the theodolite were 

 read off (Magn. I). The theodolite was then moved in 

 azimutb and altitnde till its teloscope pointed a little to 

 the west of the sun. at the altitude of the sun's centre, 

 and clamped in that position. The times for the transit 

 of the preceding and following limbs of the sun across the 

 vertical wire. were noted by a chronometer. This obser- 

 vation was tåken either with a coloured glass before the 

 eye-piece or by projecting the image of the sun and the 

 wire on a white screen. Frequently. there were two Obser- 

 vers, in which case one observed the transit of the solar 

 limbs. calling out at the moment they were tangent to. the 

 wire, while the other observed and noted the corresponding 

 readings of the chronometer. Then. after reading off the 

 verniers. the telescope was turned through the zenith 

 (nadir), moved 180° on its vertical axis. and the transits 

 of both solar limbs observed in that position by the chro- 

 nometer. tbe verniers being also read off. 



The magnet was now inverted, being turned 180° 

 about its longitudinal axis. and tbe wire of the telescope 

 of the theodolite again made to bisect the middle division 

 of the scale of the magnet, after which the verniers were 

 read off (Magn. II). 



Circumstances permitting, the magnet was again ob- 

 served in position I. Occasionally, it was inverted several 

 times. 



Moreover, on such occasions the theodolite was so 

 moved previous to the following set of observations, that 

 the foot-screws should correspond with different holes, and 

 change the zero of its circle by L20°. Its errors of divi- 

 sion, however. did not prove so considerable as to render 

 imperative such change in position for attaining the desired 

 accuracy of about one minute. 



In several cases, the azimuth of a mark was deter- 

 mined from the observations of the sun. aud the observa- 

 tions oi' the magnet, in both positions, conne.cted with the 

 direct observations of the mark. 



As a. rule. the error of tbe chronometer was foirnd 

 Ironi solar altitudes, tåken with tlie sextant at tbe place 



