ON COMPARING AND REDUCING MAGNETIC OBSERVATIONS. 77 



'VU. Report to the Secretary of the Committee of the British Association 

 appointed for the purpose of considering the best means of comparing and 

 reducing magnetic observations. 



In the suggestions which I submitted to the Committee last year I 

 proposed that the method of spherical harmonics employed by Gauss 

 should be applied to the periodic variations of terrestrial magnetism. I 

 have in the course of the past year examined by this method the principal 

 term of the solar diurnal variation ; and I hope that the results obtained 

 will induce the Committee to continue the examination in greater detail 

 of the same variation, and also the other periodic changes. Assuming 

 that the Committee is willing to adopt a course of action which must 

 necessarily lead to results of primary importance, I venture to submit to 

 them more definite proposals. 



If we had eight magnetic stations separated as far as possible we 

 should be able to obtain a fairly complete expression of the general distri- 

 bution of magnetic potential on the surface of the earth. For the expan- 

 sion in spherical harmonics, including terms of the third order, involves 

 fifteen constants, while the horizontal components of magnetic force at 

 each of the eight stations would give us sixteen quantities to determine 

 them. As far as I can judge at present, periodic variations are not much 

 affected by local circumstances, and therefore harmonics of the first three 

 orders will in all probability be sufficient, as a first approximation at any 

 rate. After obtaining the expression for the variable part of the poten- 

 tial, we can easily, with the help of the vertical component of magnetic 

 force, determine the question whether the cause of the disturbance has 

 its seat inside or outside the surface of the earth. I think, therefore, that 

 we should endeavour to obtain a complete record during a period of about 

 ten years of the elements of terrestrial magnetism at eight stations, and 

 that these should be reduced in exactly the same manner, under the super- 

 intendence of the Committee. 



As it is important to proceed without delay, we must choose stations 

 at which self- registering instruments are at present in existence. I think 

 the following will be the most suitable : Lisbon, Greenwich or Kew, St. 

 Petersburg, Bombay, Mauritius, Melbourne, Zi Ka Wei (China), Toronto, 

 or Washington. It is a matter of regret that South America is not re- 

 presented, but I do not think the deficiency is sufficiently great to justify 

 delay. 



I should propose, then — 



(1) To write to the chief observers at these stations at once, asking them 

 for such information as will enable us to judge whether the instruments 

 are in good condition, whether sufficient precautions are taken to elimi- 

 nate temperature variations, and whether they are willing to pay par- 

 ticular attention to the magnetic instraments for a period of ten years. 



(2) To ask these observers how far they are willing to undertake 

 the reduction of their observations according to a scheme submitted to 

 them by the Committee, and, in case they cannot undertake this, whether 

 they are willing to forward to the Committee the necessary records. 



(53) To obtain an estimate of the cost of the reductions, which will 

 have to be undertaken by the Committee. 



