PART II. POLAR MAGNETIC PHENOMENA AND TERRELLA EXPERIMENTS. CHAP. VI. 715 



It might at first sight appear as if this were an indication that perhaps the sun is negative in the 

 [uatorial regions and positive in the polar. If so, it would suggest the thought whether a difference in 

 ectric tension might eventually be produced by rotation of the magnetic solar body in space. 



It is easy to make an estimate here. MASCART(') has calculated that the rotation of the magnetic 

 trth must give rise to an aggregate electromotive force of an order io r< volts, acting from the poles 

 the equator. 



If we wanted to make a similar calculation with regard to the sun, we must first of all have a 



ilue for the amount of the magnetic force near the sun. This is still unknown. HAI.E(-) is at present 



aking attempts to measure it. In the mean time SCHUSTER ( 3 ), with certain assumptions, has calculated 



a recent paper that the intensity in the sun should be 440 times greater than on the earth. If we 



ckon with an even magnetisation of the globes, this would make the sun's magnetic moment 440 X IO2 3 



.eater than that of the earth. 



Now we have shown (see p. 617) that if cathode-rays from the sun with a huge moment such as 

 lis were to reach the earth, they must have a magnetic stiffness answering to 



H Q > I0 



lilefrom the situation of the earth's auroral zones we may infer that the helio-cathode rays which pro- 

 uce aurora and magnetic storms have generally a value of 3 million C. G. S. And to this last result 

 ': are inclined to attach the importance of an experimental fact. 



On the other side we have calculated, from the retardation of up to 50 hours of the magnetic 

 sjrms in relation to a sun-spot's passage of the central meridian, that the magnetic moment of the sun 

 i from about 100 to 150 times as great as that of the earth, or of the order lo 28 C. G. S. 



We cannot of course from this conclude anything about the magnitude of the magnetic force near 

 te sun, for the sun is certainly no evenly magnetised globe. The general magnetisation of the sun is 

 jobably produced by electric currents in relatively thin layers round the solar equatorial regions inside 

 < outside the sun's surface. In this way the value of the magnetic force near the sun's surface may be 

 ilatively great, without any overwhelming magnetic moment for the sun being assumed ( 4 ). If we start with 



(') MASCART, Traite de Magnetisme Terrestre, p. 74. Paris, 1900 



(3) Mount Wilson Solar Observatory; Annual Report, 1912, p. 179. 



(**) SCHUSTER, Proc. Phys. Soc. of London, 1912, p. 127. 



( 4 ) The rays emitted from the sun will certainly to some extent serve to increase the sun's magnetism. 



We have shown that if the magnetic moment of the sun is of an order lo 33 (see p. 617), all rays of which the pro- 

 duct #b?0<5Xi 5 i will return to the sun and fall down again, or must circulate about the sun, the negative rays 

 clockwise, the positive anti-clockwise, seen from above and assuming that the sun's magnetic north pole is uppermost. 



The figures 248 b and c show how such flexible rays are moving in almost cylindrical rings about the magnetic ca- 

 thode-globe. The radius of such a ring seems never to come up to 2.5 times the radius of the central sphere, as the 

 theory predicted (p. 617). 



Perhaps the "dusky veil" of the sun (see p. 670) is due to such a cylindrical ring of corpuscles moving about the sun. 

 Under all circumstances (even if the sun's magnetisation is the reverse of that here supposed), a part, and perhaps the 

 more considerable part, of the rays emitted by the sun will thus serve to magnetise the sun; but there are perhaps also 

 electric currents in the interior that act in the same way. We may even imagine the sun's magnetism to have originated 

 in this manner, if we start with the assumption that the initial velocity of the negative rays is greater than that of the 

 positive. 



Suppose that the sun had originally been non-magnetic, but rotated in the same way as it now does. It is evident 

 that the positive rays, of which the bearers may be assumed to be positive material corpuscles, will then be deflected by 

 the rotation of the central body, even if the electric forces at first tried to eject the particles normally from the surface of 

 the sun. Owing to gravitation (and by electric attraction if the sun were negative), the ejected particles now make their 

 way back to the sun. The total magnetic effect of nil the positive rays must then be that of a positive current circulating 

 in the same direction as the rotation. The sun would thus be north-magnetic above, that is to say provided no other, 

 greater forces have been acting in the reverse direction. The negative rays consisting of electrons with great velocities 

 would probably be deflected by the rotation of the central body in the same direclion as the positive rays; but the deflec- 

 tion would be less and the electrons would not at first return to the sun. Not until the sun was magnetic would they 

 be reversed, and then serve, as shown above, to augment the sun's magnetism. 



