Oct. n, 1883] 



NATURE 



567 



ing to the changes in the vertical intensity. We have 

 two sets of these instruments exactly similar in construc- 

 tion, of which one is kept in reserve. All the magnetical 

 instruments are erected in the magnetic observatory in 

 the manner shown in the diagram (Fig. 1). This observa- 

 tory is divided into three sections, and arranged as fol- 

 lows : — Furthest to the east is the variation chamber, 

 which has, in order to make room for both sets of instru- 

 ments, been made rather large, viz. 7-5 m. long, 5 m. 

 wide, and 2 "8 m. high. It has been constructed partly 

 under the earth's surface, the ground having been hol- 

 lowed to a depth of about 1 m., where the floor has 

 been laid, and the mould cast up along the outer 

 sides of the hut, which have first been covered with 

 birch bark, and in turn by turf. The roof has first been 

 covered with fireproof paper, and then with turf and 

 mould, which gives to the observatory the appearance of 

 a subterranean chamber. The object of this is, of course, 

 to keep the temperature in the room as constant as 

 possible, which we have in fact fully succeeded in doing, 

 as the diurnal variations have seldom exceeded 5 to 6° C. , 

 while averagely the temperature during twenty-four hours 

 has only varied from 2° to 3 . The lowest temperature 

 registered in the chamber during the winter was — 3° C. 

 during a high wind, while the highest during the summer 

 was 23° C. Of windows there are none ; but lighting is 

 effected by means of four petroleum lamps, which are kept 



Fig. 1. — Diagram showing the plan of the Magnetic Observatory, di d tv 

 variation instruments for the declination; H 3 ffij t variation instn 

 for the horizontal intensity ; K, I'm. variation instruments for the vertical 

 intensity; A", A'n. the reading telescopes; A, pillar for absolute mea- 

 surements; )jj> > , magnetic meridian. 



burning day and night, and to the faint heat emitted by these 

 I ascribe the circumstance that the average temperature 

 during the winter was several degrees above freezing- 

 point, and this under an open-air temperature of - 20° C. 

 From the great variation chamber a small ladder leads 

 through an aperture in the western cross wall up into a 

 narrow corridor 5 m. by 1-5 m., which opens into the 

 chamber for the absolute measurements. This is 2^5 m. 

 in length, width, and height, and from this a door leads 

 out into the open, while it is also provided with a small 

 window in the western wall. In the corridor is a parti- 

 tion with a door, in order to prevent as much as possible 

 any change in the temperature in the variation chamber. 

 The building is constructed of rough deals, while the use 

 of iron has, of course, been carefully avoided. The whole 

 has been joined by means of wooden pegs, and the roof- 

 paper fastened with copper nails. The lamps have been 

 hung on brass wire or placed on wooden tripods ; while 

 the hinges of the doors are of brass, and the handles of 

 wood. 



In the variation chamber eight pillars have been raised 

 of iron-free bricks and cement. These pillars rest on 

 large slates, which have been laid at a depth of about 

 7 decimetres below the floor, and run free through 



openings in the floor. The tops of the pillars have a 

 marble slab attached to them, on which the instruments 

 are placed, viz. one on each of the six pillars, and a 

 reading telescope on each one of the remainder. 



The needles in the six instruments are exactly alike in 

 size as well as in magnetic moment. They are made of 

 lamellar watch-spring, separated by three small bits of 

 brass, and are about 9 cm. in length. Above the needle, 

 and parallel with the same, a mirror is affixed, from which 

 the image of the scale (paper on spruce wood), placed 

 perpendicularly on the reading telescope, is reflected in 

 the focus of the telescope. Each scale is one metre in 

 length, but even this comparatively great length has, as 

 regards the intensity instruments, been found insufficient 

 during great perturbation, and we have been compelled 

 to lengthen the scale on the side most exposed by adding 

 an auxiliary scale. The scale is divided into millimetres, 

 the distance between mirror and scale being exactly 

 1719 mm , making the value of the angle of one part of 

 the scale exactly 1'. The reading telescopes of the two 

 sets are fixed on a common vertical axis, the one for hori- 

 zontal intensity being highest, the one for declination in 

 the centre, and the one for vertical intensity lowest. This 

 is also the order in which the instruments are read, a 

 reading of all three being easily effected in fifteen to 

 twenty seconds. 



The regular magnetic variation observations are, as 

 previously stated, effected every hour, seven readings 

 being taken of each instrument. At fixed terms, viz. on the 

 1st and the 15th of the month, readings are effected every 

 fifth minute, while, during a certain hour of these two 

 days, the variations of the declination and the horizontal 

 intensity are read every 20th second. Magnetic disturb- 

 ances, some even of great violence, have during the 

 twelve months been the order of the clay here. Thus 

 when perusing our "log," one will hardly find five days 

 in average in each month during which the needle has 

 remained completely at rest throughout twenty-four 

 hours. Little jerks or oscillations have constantly oc- 

 curred, particularly during the night, when the disturb- 

 ance has generally been greatest, and it has not been an 

 uncommon occurrence that the readings have fallen out 

 on the auxiliary scale, oftenest, however, as regards the 

 vertical intensity. 



The zero of the variation instruments is partly con- 

 trolled by direct simultaneous observations of both sys- 

 tems, compared once a week, and partly by absolute 

 measurements of the three terrestrial magnetic elements 

 effected several times during the month. The unifilar 

 magnetometer is employed for determining the declination 

 as well as the horizontal intensity, the latter embracing 

 both vibratory motions and deflections with the magnet 

 deflector at two different distances. The effect of the 

 torsion on the position of the declination needle is also 

 determined by deflections in the manner advocated by 

 Lamont in his " Handbuch der Erdmagnetismus," §91. 

 The inclinometer is provided with four needles, of which 

 three are in use, and during the last half year all three 

 have been used during inclinatory observations, which 

 have given results with a pretty uniform difference. As a 

 matter of course, every reading during absolute measure- 

 ments has been accompanied by a stringent and simul- 

 taneous reading of the corresponding variation instru- 

 ments. 



For the study of the aurora borealis we have employed 

 a theodolite which has been constructed by Prof. Mohn, 

 and finished by Herr C. Olsen. The theodolite has an 

 excentrically attached conic tube, which serves as drop. 

 The small ocular end is closed by a disk in which is a 

 circular hole a little larger than the pupil of the eye. 

 The objective end, which forms the basis of the conus, 

 is completely open, with a cross of thin steel wire sus- 

 pended therein. The one half of the curved surface of 

 the tube, the ocular part, is a solid brass pipe, while the 



