36 SECTIONAL ADDRESSES. 



should expect the projection of solar matter to take place radially, we 

 should similarly anticipate that when the earth is near the sun's equatorial 

 plane the number of magnetic storms would be rare. Such is the case. 

 Moreover, when the earth reaches a higher heliographic latitude north or 

 south, the probability of the earth being hit by a stream is much greater. 

 This is in accordance with observations. This also makes it clear why 

 there may be large sunspots without magnetic storms, but we should not 

 anticipate magnetic storms without solar disturbances of some kind, which, 

 if intense, would probably be associated with spots. > 



Simultaneity of Commencement of Magnetic Storms. 



Bauer concluded from observations made in 1902 and 1903 that 

 magnetic storms do not begin precisely at the same moment all over the 

 world, the velocity of propagation of the disturbances being in general 

 eastwards. On the other hand, Nippoldt had previously concluded that 

 the times of commencement of disturbances over the entire region involved 

 are not measurably different, and went so far as to suggest that they 

 might be of use in determining longitude. Rodes concluded that magnetic 

 storms do not begin at precisely the same moment all over the world ; 

 abruptly beginning ones which were investigated by him ajjpeared to 

 progress more often towards the west with a velocity such that it would 

 require about four minutes to encircle the earth at the equator. Rodes 

 put forward the hypothesis that the earth may enter a cloud of electrical 

 particles projected from the sun, in which case the storm will in general 

 be recorded by those situated in the foremost position of the earth's 

 translatory motion. It follows that since the orbital velocity of the earth 

 is such that it traverses a distance equal to its diameter in about 6^ minutes, 

 this also is the total time for the earth to become involved in a cloud. 

 At the Madrid meeting of the section on Terrestrial Magnetism and 

 Electricity of the International Geodetic and Geophysics Union, it was 

 decided that a systematic investigation of sudden commencements of 

 magnetic storms be made, and I believe that Prof. Tanakadate is much 

 of the same opinion as myself with regard to the necessity for similar 

 instruments and the most precise recording of time in order to obtain a 

 decisive answer to this question. It should not be difficult to obtain. 



Need for more Precise Data. 



This very hasty sketch of some theories relating to terrestrial magnetism 

 reminds me of Dr. Chree's remarks that the deductions from such theories 

 are just as hj^pothetical as the theories themselves, and I am very sensible 

 that this rapid survey is not only incomplete, but that no theory con- 

 sidered is completely satisfactory. Moreover, while fully realising that 

 they are vital links in any chain of evidence, I have avoided the companion 

 subjects of aurorse, atmospheric electricity and earth currents, because to 

 have considered them would have made my address far too long. I do, 

 however, wish to emphasise that data of a precise kind are much needed 

 to modify existing theories and to produce new ones, and I cannot do 

 better than conclude with a remark of Riicker's in this city thirty-two 

 years ago. Riicker said : ' If there be any who are inclined to ask whether 

 the careful study of terrestrial magnetism has led, or is leading, to any 



