132 



KNOWLEDGE 



[July 1, 1892. 



corresponding to the wave-lengths absorbed by the outer 

 cool gas. If, however, a part of the gaseous constituents 

 condensed into liquid or solid particles in the outer and 

 cooler regions of the nebula, we should have the above 

 conditions complicated by the action of the glowing 

 particles on the molecules of the uncondensed vapours 

 surrounding them. 



The extreme faintness of the nebular glow is a fact that 

 must not be lost sight of in seeking for a possible 

 explanation of the facts observed. If the nebular matter 

 glowed with a millionth part of the brightness of the solar 

 photosphere, the nebulge would give us far more light than 

 they do. 



Let us assume with Prof. E. C. Pickering that the 

 stellar magnitude of our sun is -25-5 — in other words, 

 that our sun gives about forty thousand million times as 

 much light as a star of the first magnitude. According to 

 this estimate, the sun would need to be removed to a 

 distance where its diameter would appear to subtend an 

 angle of only -00964 of a second of arc to reduce its light 

 to equality with that of a star of the first magnitude. A 

 nebula which subtends a minute in diameter and gives 

 the light of a star of the eigth magnitude gives -g-jT-th 

 part of the light of a star of the first magnitvule, while its 

 diameter is more than six thousand times as great as the 

 diameter which our sun would appear to have at a distance 

 where its light would appear equivalent to that of a star 

 of the first magnitude ; consequently the solar photosphere 

 must, area for area, be more than twenty -two thousand 

 million times as bright as such a nebula. 



According to Prof. Langley, the sun's photosphere* is 

 5800 times brighter than the molten metal in a Bessemer 

 " converter " ; consequently, the nebula must glow with a 

 light which corresponds to less than one millionth of the 

 brightness of white-hot iron. Perhaps one might com- 

 pare the brightness of the nebula with the faint glow of 

 the trail left by a large meteor on entering our air. Such 

 trails seem to be due to glowing red-hot particles. The 

 vapour driven off from the meteor is for an instant 

 intensely luminous. It is no doubt under great pressure, 

 and rapidly expands, driving back the cold upper air. In 

 expanding, it cools till the greater part of the vapour is 

 precipitated into a glowing mist, which having no elasticity, 

 and not being able to do work by driving back the sur- 

 rounding air, can only cool by radiation. In this condition 

 the trail from a large meteor sometimes remains faintly 

 glowing for half an hour. The evidence seems to show 

 that at first the trail gives out bright lines in addition to 

 a continuous spectrum. The study of such trails may 

 throw much light on the constitution of the nebula. 



Hitters. 



[The Editor does not hold himself responsible for the opinions or 

 statements of correspondents.] 



THE EVOLUTION OF DOUBLE STAES. 

 To the Editor of Knowledge. 



Sir,— I am not sure that I quite follow the theory of 

 Mr. See on the evolution of binary stars, but I think 

 generally he maintains that these stars owe their higlily 

 eccentric orbits to the fact that their masses are much 

 more nearly equal than, for instance, those of the sun and 

 .Jupiter.^ Assuming this to be the case, we would expect 

 the orbits of very unequal double stars to be much less 

 eccentric than those of nearly equal ones ; and although 



* Sec Prof. C. A. Young's " Suu," p. 2-15. 



the mass of a star may not correspond with its magnitude, 

 there must on an average be a pretty close relation 

 between them. I accordingly examined the stars in 

 Mr. Gore's catalogue, taking in each instance the latest 

 orbit as founded on a larger number of observations than 

 its predecessors. I classed as equal the binaries wliose 

 magnitudes did not differ by more than one-half, and as 

 unequal those whose difference was not less than two 

 magnitudes. The following is the result of my com- 

 parison : — 



Unequal Binaries. 



Equal Binaries. 



Ecfeiiti'icitv. 



0.674 



0-632 



0.3086 



04346 



O-3029 



0-8715 



0-480 



0-2667 



0-5836 



0-7352 



0-493 



0-1698 



0-4244 



0-6000 



0-3052 



0-1849 



0-0768 



Star. Eccentricity. Star. 



■q Cassiopeijp 0-6244 /) Eridani 



40 Eridani 01 862 2 1037 



05 149 0-460 2 3121 



Siiius 4055 Ov 215 



a Centauri 0-5443 OS 234 



7 Coronse B. 0-3483 v Virgiuis 



A Ophiuchi 0-4424 42 Coma? 



i, Herculi? 0-463 y] Corona? B. 



70 Ophiuchi 0-4994 05 298 



99 Hereulis 0-7928 5 2091- 



S Cygni 0-827 u. Uraconis 



/3 Delphini 0-0962 I Sagittarii 



T Cygni 0-3475 y Corome Aust. 



02 489 0-343 I Aquarii 



85 Pegasi 0-35 5 1S19 



25 Can. Ven. 0-7221 7. 2173 



J Scorpii . 



No striking contrast will be found here, nor do either 

 set differ much in eccentricity from those where the 

 difi'erence in magnitude between the components ranges 

 between one-half and two. The eccentricities, however, 

 are in most cases very uncertain. Thus in Mr. Gore's 

 table those for Sirius range from 0-4055 (Mr. Gore's) to 

 0-945 (Mr. Mann's^, for Castor from 0-300 to 0-797, for 

 a Centauri from 0-5260 to 0-9689, o- Coronse Borealis 

 0-3088 to 0-7515, \ Ophiuchi 0-4424 to 0-8191, t Ophiuchi 

 0-0375 to 0.6055, &c. The following, however, are agreed 

 in either by all computers, or a large majority : y Virginis 

 (equal) 0-85 to 0-9, ? Urs» Majoris (medium) 0-38 to 0-43, 

 A Ophiuchi (unequal) 0-44 to 0-5, ^ Hereulis (unequal) 

 0-43 to 0-48, 70 Ophiuchi (unequal) 0-4 to 0-5, y CoronfB 

 Borealis (medium) 0-34 to 0-39, 2 3121 (equal) 0-26 to 

 0-88. The result, on the whole, does not seem to me to 

 favour the theory of tidal evolution. 



Truly yours, 



W. H. S. MoNCK. 



P.S. — In the solar system the rule appears to be that 

 the orbits of the smallest bodies are the most eccentric. It 

 may be worth noticing that the equal pairs (with computed 

 orbits) appear to have in all cases spectra of the second or 

 solar type. 



[By courtesy of the Editor I have been permitted to read 

 Mr. Monck's criticism, and will say in reply that he slightly 

 mistakes my meaning. Large relative masses imply the 

 development of large eccentricities only in so far as these 

 masses imply large initial moments of momentum of axial 

 rotation. For upon the moments of momentum of axial 

 rotation depends the increase in the orbital momentum and 

 mean distance, with which the increase of the eccentricity 

 is so intimately connected. Unless the sum of the moments 

 of momentum of axial rotation of the stars of a system is 

 greater Avhen the stars are equal than when unequal, it 

 does not follow that the eccentricity would become higher. 

 According to theory, therefore, we must not conclude that 

 (for example) rj Cassiopeia^ (masses 1 to 4) should have a 

 smaller eccentricity than ex. Centauri (masses equal). It is 

 also to be borne in mind that the eccentricity will depend 



Enlarged from a Photogr! 



