GIG TRANSACTIOXa OF SECTION A. 



The magnetic and other physical observations conducted on the shore close to 

 the ship extend over a period of nearly two years. 



The ' Discovery ' ' vv'inter quarters ' were to the south-east of the south magnetic 

 pole, the mean absolute declination being about 152° E. From the photographic 

 records by the Eschenhagen magnetometers the general character of the regular 

 diurnal variation of declination is clearly shown, the movement beino- five or 

 six times as large as that at Kew; even during the months when the sun is 

 continuously below the horizon the diurnal range remains large — at least double 

 that at Greenwich at midsummer. 



Tlie seasonal change in the type of the diurnal variation is very small. In 

 temperate climates generally the changes of declination during the day are con- 

 spicuously larger and more rapid than during the night. At ' winter quarters ' 

 there is no such marked difference. 



The extreme positions are reached at about 9 A.M. and 7 p.m., and are totally 

 different from those customary. The absolute declination observations indicate an 

 apparent decrease of about 26' between mean of observations May-Deceoiber 1902 

 and mean May-December 1903, the horizontal force observations an apparent 

 increase of 0-0013 C.G.S. between mean April-December 1902 and mean April- 

 December 1903, and the inclination an apparent decrease of about 7' between 

 mean May-December 1902 and mean May-December 1903. The horizontal force 

 at 'winter quarters' was about a third of that at Kew, but the range of the 

 diurnal inequality about 50 per cent, greater than at Kew. The type also varies 

 very little with the season, and is totally different from that at temperate 

 stations. 



Vector dlagi-ams have been made for ' winter quarters ' for the whole year, 

 and for the seasons midwinter (May to July), equinox (March, April, September, 

 October), and midsummer (November-J.anuary). The differences between 

 these diagrams and the vector diagrams for Kew and Falmouth are remarkable. 

 There is much less asymmetry than in England, the direction of motion is anti- 

 clockwise, and there is very little difference in type for the different seasons. 

 The most striking peculiarity in the diurnal inequality in all the elements is the 

 large size of the fundamental or 24-hour Fourier ' wave ' as compai'ed with the 

 harmonics of shorter period. 



' In order to determine the position of the south magnetic pole by means of all 

 the declination observations taken at " winter quarters," sledge journeys, and at 

 sea, the results have been plotted (by Captain L. W. P. Chetwynd, R.N.) and the 

 direction of the magnetic meridian as indicated by the observations extended 

 towards the magnetic pole. These lines of direction intersect within a space 

 triangular in form. The radius of the circle inscribed in the triangle measures 

 about thirty-eight geographical miles, and the centre of the circle indicates the 

 probable position of the pole, and is in lat. 72° 50' S., long. 156° 20' E. 



' All the inclination results were plotted on a chart and lines of equal inclina- 

 tion drawn, from which the probable position of the pole is indicated to be in 

 lat. 72° 52' S., long. 156° 30' E. The agreement between this position and that 

 determined by the declination results is remarkable, and may be considered as 

 corroboration of the results. 



' The mean of the two positions, viz., lat. 72° 51' S., long. 156° 25' E., is, in all 

 probability, a close indication of the centre of the polar area.' ^ 



The results of the pendulum observations, taken with three separate quarter- 

 meter invariable pendulums, give values all slightly in excess of the theo- 

 retical. 



Accepting 981,200 (centimetres/second'^) as the value at Kew, the probable 

 mean value deduced from all observations combined for * winter quarters ' is 

 982,985, tie theoretical value being 982,963. 



The most interesting outstanding feature is the large difference between the 

 resulta obtained at ' winter quarters ' during the winter months, July, August, 

 September, on the one hand and those obtained during February, the end of the 



' Mag^netic Results,' Physics, vol. i., Royal Society. 



