546 REPORT — 1894. 



mined by himself with those deduced from the self- registering apparatus of the 

 observatory. 



The discrepancies between the so-called standards, which were thus brought 

 to light, were quite startling, and prove the necessity for au investigation as to 

 their causes. 



Magnetlcians had long been aware that the instruments used by travellers 

 should be compared at the beginning and end of a journey with those at some 

 fixed observatory, to make sure that the comparatively rough usage to which thoy 

 are subjected has not affected their indications. But Dr. Van Eijckevorsel's 

 expedition first drew general attention to the fact that there are serious differences 

 between the standard observatory instruments themselves. 



The importance of a careful comparison between them was at once recognised. 

 The Magnetic Sub-Committee of the International Meteorological Conference, held 

 at Munich in the autumn of 1891, resolved that it is ' necessary that the instru- 

 ments employed for absolute measurements at the different observatories should 

 be compared with each other and the results published.' As far as I am aware, 

 nothing has been done to give effect to this resolution, but the necessity for such 

 an international comparison is urgent. The last few years have been a period of 

 unexampled activity in the conduct of local magnetic surveys. To cite instances 

 from the north-west of Europe only, observations have recently been made on a 

 more or less extended scale in the United Kingdom, France, Holland, North 

 Germany, and Denmark. 



It will be absurd if these surveys cannot be collated and welded into a homo- 

 geneous whole, because we are in doubt whether the indications of our standard 

 instruments for the measurement of declination and dip differ by five or six 

 minutes of arc. 



If, however, an official international comparison of the magnetic standards iiii 

 use in different countries is instituted, it is probal^le that only one observatory in 

 each country will take part in it. 



It may fairly be left to each nation to determine for itself the relations between 

 the results of measurements made in its own institutions. Apart, therefore, from 

 all other reasons, we in England would only be able to make the best use of an 

 international comparison if we had beforehand set our own house in order, and 

 were able at once to extend the results of experiments made at Kew or Greenwich 

 to Stonyhurst, Yalentia, and Falmouth. 



This we are not at the present moment in a position to do. As far as I know, 

 nobody has ever carried a magnetometer backwards and forwards between Kew 

 and Greenwich to test the concordance of the published results. During the 

 recent survey single or double sets of observations have been made at Stonyhurst, 

 Falmouth, and Valentia, with instruments which have been compared with Kew ; 

 but these measurements, though amply sufficient for the purposes of our research, 

 were not numerous enough to serve as a firm basis for determining the discrepancies 

 between the various standards, so that the exact relations between these important 

 sets of apparatus are still unknown. 



The first point, therefore, to which I wish to draw the attention of the Section 

 is the necessity for a full primary comparison between the standard magnetic 

 instruments in use at our different observatories. 



But, if this were satisfactorily accomplished, the question would arise as to 

 whether it should be repeated at regular intervals. We have at present only 

 a presumption in favour of the view that the standards which we know are 

 discordant are nevertheless constant. A single instance may suffice to show how 

 necessary it may be — at all events in the case of outlying and isolated observatories 

 — to put this belief to the test. 



In the most recent account of the work of the observatory of the Bombay 

 Government at Colaba, the dips are discussed for the period of twenty yeais 

 between 1872 and 1892. During this interval the adjustment of the agate plates 

 upon which the dip needle rolls has thrice been modified. In 1877 the plates 

 were renewed. In 1881 and 1887 the dip circle was taken to pieces and rebuilt. 

 In the intervals the dip as determined by several needles, but always with this 



