98 REPORT — 1886. 



X. The Advantages to the Science of Terrestrial Magnetism to he obtained 

 from an expedition to the region vnthin the Antarctic Circle, By Staif 

 Commander Ettrick W. Creak, B.N., F.B.S. 



In Ganss's paper on the general theory of magnetism, published in 

 England in 1839, will be found the following conclusions: — ■ 



(1) 'It is clear that the knowledge of T (or the component of the 

 horizontal magnetic force directed towards the west) on the whole earth, 

 combined with the knowledge of X (or the component of the hoi'izontal 

 force towards the north) at all points of a line running from one pole of 

 the earth to the other, is sufficient for the foundation of the complete 

 theory of the magnetism of the earth.' 



(2) ' Finally it is clear that the complete theory is also deducible 

 from the simple knowledge of the value of Z (or the component of the 

 magnetic force directed towards the centre of the earth) on the whole 

 surface of the earth.' 



Accepting these conclusions as thoroughly sound, and in no measure 

 altered since they were written by other investigators, let us now inquire 

 into the question how far are we prepared by observation of the earth's 

 mao-netism for a calculation of this kind. Thanks to the activity of 

 observers in many lands and over many seas during the years 1865-85, 

 we have been supplied with the necessary observations, which have been 

 utilised for compiling charts on a large scale of the normal values of the 

 declination, horizontal force, and vertical force for the epoch 1880 from 

 which the values of X, Y, Z may readily be obtained for a large portion 

 of the earth's surface. 



These elements for the zone contained between the parallels of 60° N". 

 and 50° S. are (except for some portions of Northern Asia and Central 

 Africa) accurate ; from 60° to 70° north latitude and 50° to 60° south 

 latitude they are less accurate. North of the parallel of 70° N. and south 

 of 60° S. are two portions of the earth of which our knowledge is far 

 more limited ; but whilst we have had comparatively recent observations 

 ill Arctic regions, nearly the whole of the Antarctic regions have re- 

 mained unvisited for magnetic purposes since the memorable survey 

 conducted by the late Sir James Ross in 1840-43, so that the charts are 

 correspondingly weak in those latitudes. 



A reference to the accompanying map shows that the Challenger, 

 whilst on her voyage from the Cape of Good Hope to Melbourne, crossed 

 the Antarctic circle about the meridian of 79° E. The magnetic observa- 

 tions during that period combined with those made at Sandy Point, 

 Magellan Strait, since 1868, and some declination observations recently 

 made in the Soiith Pacific between New Zealand and Cape Horn in lati- 

 tudes between 50° and 60° S. give ample evidence that considerable change 

 of the magnetic elements has occnrred since Ross's voyage. Of the extent 

 of these changes our information is so limited that the old survey is of 

 but little use in enabling us to complete the charts of 1880 with the 

 requisite amount of accuracy, and therefore the X, Y, Z required for 

 Gauss's method of theoretical investigation are still wanting. 



Although it is true that Gaass has shown the method by which 

 mathematicians may, from an accurate knowledge of the magnetic ele- 

 ments over an extended area in both hemispheres, calculate, nearlj^ those 

 of the remaining portions, yet supposing this to have been done — and it 



