466 



SCIENCE 



[Vol. LVI, No. 1452 



between tlie two classes of plienomena, his con- 

 clusions as to the immediate relation of the 

 two have not met with universal acceptance. 

 Very few physicists have given any attention 

 to the study of earth-currents in recent times, 

 and most of these have been influenced largely 

 by their theories as to the cause back of both 

 earth-currents and terrestrial magnetism. Some 

 physicists, of whom Balfour Stewart is per- 

 haps the most important, have gone so far as 

 to reach conclusions the reverse of those of 

 Airy and Weinstein, and to attribute the earth 

 currents to the magnetic variations. The fact 

 that the diurnal variations in magnetic 

 direction and intensity which would seem 

 to follow from the changes in the earth- 

 currents regularly lag behind the latter cer- 

 tainly indicates that the current does not imme- 

 diately act upon the magnetic needle, as in the 

 ease of the current and needle of a galvano- 

 meter, and as is 'maintained by Weinstein in 

 the quotation given above; but that the varia- 

 tions of the needle follow changes in the mag- 

 netic field of the earth, and that whatever 

 effects are produced upon the needle by earth- 

 currents must be due to changes which these 

 currents have produced in the magnetic field 

 of the earth at that place. 



Looked at in this way, a time lag between 

 the changes in current and the magnetic varia- 

 tions growing out of them does not seem im- 

 probable. Ewing has shown that in the case 

 of a soft iron wire placed in a magnetizing 

 coil an appreciable time is required for the 

 wire to become fully magnetized. Rayleigh 

 found that with a piece of annealed iron wire 

 only 1.6 mm. in diameter and 17 cm. long the 

 time for complete magnetization in a weak 

 field was from 15 to 20 seconds. Hopkinson, 

 after an estensive study of the rate at which 

 magnetic induction penetrates iron, says: 



Suppose a magnet such as we have here con- 

 structed, but of the dimensions of the earth, and 

 that some almighty electrician reversed its cur- 

 rents in the copper coils, the magnetizing force 

 being 2 or 3. It would take some thousands of 

 millions of years before the rate of change of 

 induction at the center of its core would attain a . 



and it is plain that the variation in the mag- 

 netic elements at a given place are not due to 

 changes in the whole magnetic field of the 

 earth, as in that case they would occur simul- 

 taneously over the whole earth instead of 

 moving around with the sun, as they do, but it 

 does not seem incredible that a change in the 

 direction or magnitude of an earth-current at a 

 given place should require two or three hours 

 to produce its maximum effect upon the mag- 

 netic elements at that place. In such a con- 

 tingency, it seems almost certain that the rate 

 of change of the magnetic field at a given time 

 would be proportional to the current intensity 

 at that time. If we assume this to be the ease, 

 it is possible to compute from the curve of 

 magnetic variation the direction and relative 

 intensity of the current which is the cause of 

 this magnetic variation. 



Thus, when a current flows from north to 

 south below a compass needle it deflects the 

 north end of the needle toward the west. The 

 north-south component of the earth current at 

 Berlin was usually toward the south, and was 

 much stronger in the day time than at night. 

 Its greatest intensity was usually just before 

 noon, and its least intensity in the morning 

 and evening. In Figure 1, the curve A is 

 Weinstein's curve for average diurnal variation 

 in the north-south component of the observed 

 earth-current for the year 1884. Curve B 

 gives the average diurnal variation in the west 

 component of the earth's magnetic field at 

 Vienna for the same period. It will be seen 



maximum. 



It is perhaps not allowable to compare the 

 earth magnetically with a sphere of soft iron. 



