84 REPORT — 1886. 



currents in the moist conducting strata of the earth's surface and in 

 the upper atmospheric regions, which currents, as well as the magnetic 

 change which causes them, will, of course, influence a suspended magnet. 



23. There is still left the question, Why should there be small and 

 rapid changes of the earth's magnetism ? In reply to this it is argued 

 that we must regard the earth as we would any other magnet, the only 

 difference being one of size. Now, there are at least two known causes 

 which may operate upon a magnet to change its magnetic state. These 

 Sire first, mechanical disturbance; and secondly, a change in the electric 

 currents in whose field the magnet is placed : and it is asserted that the 

 changes which take place in the magnetism of the earth should be studied 

 from these two points of view. It is the second of these causes that has 

 hitherto been chiefly investigated, and magneticiaus have succeeded in 

 showing that disturbances vary with the state of the sun's surface, witlt 

 the time of the year, and with the hour of the day. Possibly, however, 

 considerations connected with the first of these causes might seem best 

 to explain the second portion of the preliminary results obtained by 

 Messrs. Stewart and Carpenter (see Appendix VI. to this report). 



24. In conclusion, perhaps the strongest objection to this hypothesis is 

 that which questions the possibility of electric currents being produced 

 in the upper atmospheric regions. It may be said that while undoubtedly 

 rarefied air is a conductor of electricity, yet it is not a good conductor ; 

 and where can we look for sufficient potential to drive such currents 

 throuo-h these regions ? To this it may be replied that as a matter of fad 

 we know that there are visible electric currents in the upper atmospheric 

 regions which occur occasionally at ordinary latitudes, and which are 

 very frequent if not continuous in certain regions of the earth. These are 

 known as the Aurora which, both with respect to the time of its occurrence 

 and to the disposition of its beams, manifests a close connection with the 

 phenomena of terrestrial magnetism, occurring at ordinary latitudes only 

 when there are great magnetic disturbances, and the disposition of its 

 beams having a distinct reference to lines of magnetic force. Besides, 

 con.oiderations of a mathematical nature induce us, as we have already 

 seen, to suppose that the solar-diurnal variation is due to electric currents 

 in the upper atmospheric regions. We are, therefore, justified in asserting 

 that there is no impossibility in conceiving a set of electrical currents 

 intimately associated with certain phenomena of terrestrial magnetism to 

 exist in the upper regions of the earth's atmosphere. 



IX. Examples of the Application of a Modified Form of Sabiiie's Method of 

 Reduction of Hourly Observations of Magnetic Declination and Hori- 

 zontal Force to a Single Quarter s Registrations of the Magnetogra-phs at 

 the Colaba Observatory, Bombay. By Chakles Chambers, F.R.S. 



On the invitation of Dr. Balfour Stewart, the Secretary of the Com- 

 mittee appointed by the British Association to consider the best means of 

 comparing and reducing magnetic observations, I submitted last year, 

 for the Committee's consideration, some remarks which had the honour 

 to find a place in their Report to the Association. Amongst the sugges- 

 tions which I then made was one for a trial application of a somewhat 

 elaborate process of reduction, the results of which it was anticipated 

 would, in some respects, be as definite and informing when derived from 



