72 CARNEGIE INSTITUTION OF WASHINGTON. 



a chart of isabnormal temperatures. It was found that the earth as 

 a magnet acts Hke any other magnet as regards appHcation of heat. 

 Thus, wherever the earth's surface is relatively warm, on the average 

 for the year, there the magnetization of the earth shows a decrease, 

 and where, on the other hand, it is relatively cold, there it suffers an 

 increase. 



It was further found that — 



there is very close similarity between the residual permanent 

 magnetic field of the earth and that of the field of forces causing 

 the diurnal variation of the earth's magnetism ; and there ap- 

 pears to be more than a mere chance connection in this relation, 

 as is shown by the simultaneous studies of the vector diagrams 

 for various parallels as resulting from the two respective fields. 



Investigation II. Discussion and publication of the data on the magnetic 

 perturbations observed during the eruption of Mont Pelee, Martinique, 1902. 



The data obtained as the result of a circular letter sent by the 

 Superintendent of the U. S. Coast and Geodetic Survey to observa- 

 tories over the entire globe were turned over by him, as agreed, to the 

 department for discussion and publication. First, the investigation 

 of the notable magnetic disturbance coincident with the eruption 

 on May 8, 1902, was undertaken, the necessary information having 

 been extracted from the reports and magnetograms received from 

 twenty-six institutions distributed over the globe, and most impor- 

 tant results have been derived. 



It was found that the Mont Pelee magnetic disturbance of 

 May 8, 1902, resembled a cosmic one in two respects, viz : First, 

 that the time of beginning of the disturbance was practically 

 the same around the whole earth ; and, second, that any elec- 

 tric-current system capable of producing the observed phenom- 

 ena would have its seat chiefly outside the earth. 



Owing to the peculiar vaporous nature of the products of the 

 eruption, it would appear as though their violent ejection was ac- 

 companied by the formation of electric charges above the earth's 

 surface sufficient to disturb the entire potential of the earth. We 

 thus have had .shown us how a magnetic storm can be produced by 

 a tremendous explosion, and the further study may throw some 

 light upon the connection between terrestrial magnetic .storms and 

 solar eruptions, and on the mochis operandi of the operating forces. 

 The investigation is tnerefore being continued so as to include other 

 disturbances occurring at about the same time, and an examination 



