64 Mr. W. Ellis. Relation between Diurnal Range of 



" On the Relation between the Diurnal Range of Magnetic 

 Declination and Horizontal Force and the Period of Solar 

 Spot Frequency." By WILLIAM ELLIS, F.R.S., formerly 

 of the Royal Observatory, Greenwich. Received March 3, 

 Read March 10, 1898.' 



(Second Paper.) 



In a paper communicated to the Royal Society in the year 1879, 

 and printed in the ' Philosophical Transactions ' for 1880, I com- 

 pared the diurnal range of magnetic declination and horizontal 

 force, as observed at the Royal Observatory, Greenwich, during the 

 vears 1841 to 1877, with the corresponding numbers of sun-spot 

 frequency as determined by the late Dr. Rudolf Wolf, of Zurich. As 

 I then said, I conceived that the long series of Greenwich observa- 

 tions, made throughout on the same general plan and with instru- 

 ments of the same kind, might be applied as a valuable independent 

 test of the reality of the relation generally understood to exist 

 between the phenomena in question. And the comparison appeared 

 to be distinctly confirmatory thereof. For it was to be observed 

 that, although the sun-spot period, commonly called the eleven- 

 year period, varied in length to the extent of several years, the 

 corresponding magnetic periods varied in a similar manner. Still, 

 in a case of this kind, in which the cause of the phenomena 

 observed has not been determined or a'scertained, it becomes im- 

 portant indisputably to prove the accuracy of the observed facts and 

 of the inference to w r hich they lead. And if further observation 

 shows that the phenomena continue to progress collaterally, the 

 circumstance must eventually be accepted as indicating that between 

 the two phenomena there exists a more or less direct relation which, 

 in any theoretical consideration of the subject, could not be ignored. 

 The previous paper, as mentioned, includes results to the year 1877, 

 but the material since accumulated, available now to 1896, happens 

 to be especially interesting, and, contrasting in some respects with 

 the earlier portion, is worthy of being made known, the series 

 as a whole forming one continuous chain of evidence that much 

 strengthens the argument for relation. Apart, however, from any 

 individual opinion on the matter, it is well that, so long as the 

 phenomena observed remain without explanation, the facts thereof 

 should be carefully set forth. I propose, therefore, to discuss the 

 question anew for the whole period 1841 to 1896. 



It is unnecessary to say anything in explanation of the results 

 given in the previous paper. I will therefore proceed to describe 

 the new work, extending from 1878 to 1896. As before, the mean 



