8 4 8 



NA TURE 



[December 23, 1922 



surface, produced, in the yellow stars, at a tempera- 

 ture of about 6000 C. ; but the ionisation from this 

 cause is the same in the two types of star. In order 

 to explain the greater brightness of the giant stars, it 

 is necessary to suppose that they contain another source 

 of ionisation which is peculiar to them, or more im- 

 portant than it is in the dwarf stars. The principal 

 supplementary cause appears to me to be a penetrating 

 radiation emitted by the interior layers of the star ; this 

 radiation would be stronger in giant stars, which have 

 greater masses, and therefore higher internal tempera- 

 tures. In an example cited by Eddington 7 the tempera- 

 ture at the centre reaches 4,650,000° C, the mass being 

 only one and a half times that of the sun. The wave- 

 length of maximum energy for a black body at this 

 temperature is 6 A.U. — corresponding to an X-ray near 

 the ultra-violet, and not very penetrating ; but, accord- 

 ing to the theory, the radiation extends much further 

 towards the very short wave-lengths, the penetration 

 of which is much greater ; and the intensity of these 

 extreme radiations increases with the temperature of 

 the star. The emission of exceptionally penetrating 

 radiations by giant stars is therefore admissible. Fur- 

 ther, in the case of giant stars which are in the phase 

 of increasing temperature, the atoms are dissociated, 

 and their breaking-up is accompanied by an intense 

 emission of a-, [3-, and y-rays. It should be remarked 

 that recently certain rays of radium have been ob- 

 served, much more penetrating than any previously 

 known, the source of which must be in the very nucleus 

 of the atom. 



In the yellow stars, all these radiations, of very high 

 frequency and of great penetrating power, form, in 

 reality, only a very small part of the total radiation ; 

 but their remarkable electrical properties assign to them 

 an important role in the electrical phenomena of stellar 

 atmospheres. It is not, however, intended to assert 

 that the existence of the radiations is proved, but it is 

 very probable. As a matter of fact, we have a very 

 imperfect knowledge of the properties of the material 

 in the interior of a star near the surface and in the 

 atmosphere ; and, as often happens in astronomy, the 

 deductions rest on wide extrapolations. In forming 

 conclusions, great care must be exercised. 



There has been a great deal of discussion on the 

 nature of the solar surface. In my opinion, setting 

 aside every theory and every explanation, the solar 

 surface is a simple fact of experience ; it is a surface 

 of discontinuity, with a clear-cut boundary, such that 

 the light emitted by the interior is much more intense 

 than that given by the exterior. I give the name 

 " atmosphere " to all that is outside this surface. The 

 word " surface," however, should not be understood 

 strictly in its geometrical sense : it implies, actually, 

 a relatively thin luminous stratum which, at our 

 distance from it, appears to have no thickness. The 

 solar surface has often been described as a cloud, 

 made up of incandescent liquid or solid particles. If 

 this were so, in all the yellow stars having the same 

 temperature, whether giant or dwarf, the pressure of 

 the gases at the surface should be the same ; but it 

 has been objected that we have no knowledge of any 

 matter which remains liquid at a temperature of 

 6ooo° C. The attractive optical theory of Schmidt also 



7 Astrophysical Journal, 48, pp. 205-214, 1918. 

 NO. 2773, VOL. I IO] 



has been advocated : when thoroughly examined, how- 

 ever, it is found not to be applicable to the sun. Let 

 us say simply that, from a cause still imperfectly 

 understood, solar matter, probably gaseous, acquires 

 suddenly, in a stratum called the surface, the emissive 

 power of a solid body, and there are good reasons for 

 believing that the pressure of the gas in this stratum 

 varies little from one yellow star to another, so long 

 as the temperature of the strata is the same. 8 These 

 considerations support the idea of the very penetrating 

 emission postulated in the giant stars. 



III. These special rays, remarkable for their pene- 

 tration and their electrical action, have been known 

 or suspected only for a few years ; but their importance 

 is already declaring itself, and I think that they will 

 furnish the key to several of the still numerous enigmas 

 presented by the celestial bodies. 



The matter of the sun, then, probably emits X-, 

 ultra-X-, and corpuscular rays, with an intensity which 

 increases from the surface to the centre. In the spots, 

 which are in general cavities, the emission is strongest 

 in the centre, and, because of its greater penetration, 

 is able to persist in spite of local absorption and the 

 diminution of the ordinary light. Similarly, if the 

 earth gives rise to a radiation of this kind, its in- 

 tensity should be greater at the poles than at the 

 equator. 



These radiations should be borne in mind especially 

 in considering the nebulae — in particular, the gaseous 

 and planetary nebulas. A nebula with a stellar nucleus 

 may be considered as a star the atmosphere of which 

 is extraordinarily developed and contains special gases, 

 such as nebulium. The conditions are then, on a very 

 large scale, those of the yellow giant stars examined 

 above, the atmospheres of which are particularly bright ; 

 and the same causes may be held to account for the 

 luminosity in the nebulous atmosphere. Moreover, 

 the nucleus, being of the Wolf-Rayet type, is one of 

 the hottest stars : it is conceivable that the maximum 

 emission takes place, for the nucleus in the X-region, 

 and for the nebula, properly so called, in the visible 

 region. The luminosity is produced by radiations of 

 very short wave-length, but with a habitual tendency 

 towards longer wave-lengths. Lastly, the nucleus may 

 contain a large proportion of radio-active bodies. These 

 ideas were put forward in 1902, and Russell has recently 

 developed similar hypotheses. 9 



If a nebula has no nucleus, we may suppose that 

 there are radio-active bodies disseminated in the space 

 which it occupies. Similarly, in the lower part of our 

 atmosphere, a considerable fraction of the ions formed 

 per second is due to the gaseous emanations of radium 

 and rhodium spread abroad in the air. If there were 

 a greater proportion of radio-active bodies, the gas 

 might become luminous. 



To sum up, the penetrating radiations are interesting 

 in the highest degree, and it is important that we 



8 If the pressure at the surface is les> in tlir giant stars, the average 

 density "i winch is smaller, we < an explain p.irtlv the stronger ionisation in 



: 1 tai I" thi very interesting theory of M. N. Saha. This th ry deals 



with effects due to temperature alone, and the point of view is different. 

 In a giant star the pressure gradient is evidently less steep, but the average 

 pressure in the middle stratum, and especially in the upper stratum, may 

 be very nearly the same as in a dwarf star, ft should be noted that the 

 greater proportion of the positive ions of calcium in the upper atmosphere 

 may also be explained simply by the repulsion due to the positive charge 

 on the star. 



9 Deslandres, Comptcs rmdus, 134, pp. 1134 and i486, 1902; Russell, 

 Proceedings of the U.S. National Academy of Sciences, 5, No. 10, p. 410. 



