November 1, 1893.] 



K NOWLEDGE 



207 



Photosphere ■' " are of wide import ; for if they be valid 

 for the sun, they must be equally so for the great majority 

 of the stars. It seems impossible to stop short at the 

 conclusion that " the light of the solar photoi^phere is due 

 to the. brilliant incandescence of the most refractory 

 substances present in the sun, at a level where they are 

 just on the point of being driven into vapour." The further 

 inference presses itself upon us that " the most refractory 

 substances " in the stars behave similarly ; and these sub- 

 stances are likely to be, if not identical, at any rate closely 

 analogous in sun and stars. Hence, if photospheric 

 temperatures be determined by their boiling-points, a very 

 narrow range of diversity can be allowed to stellar emissive 

 power. All stars of normal constitution must radiate a 

 nearly equal amount of light and heat per unit of super- 

 ficial area. Binaries with known orbits afford, however, a 

 measure of what has been called their " density-brightness." 

 Supposing, in other words, their mean density to be the 

 same, the intrinsic lustre of their photospheres can be 

 determined. It is found to vary enormously ; but if the 

 condensation theory of photospheric formation be true, 

 there can be but little difference in this respect between 

 one star and another. If we admit the theory, then, the 

 observed variations must be laid to the score of density, 

 not of brilliancy. This, however, involves somewhat 

 improbable consequences. The whole bulk of the stars of 

 y Leonis, for instance, should, on the supposition in 

 question, be composed of matter fully seven thousand 

 times more rarefied than the average solar materials ; and 

 these stars being of the same spectral type with the sun, no 

 differences of absorption could help to explain away any 

 part of this vast discrepancy. Again, 8 Cygni sends us 

 one hundred times more light than we should receive from 

 the sun if it were so situated as to form with a satellite- 

 sun a system with the same period, and of the same 

 apparent dimensions as the system of that binary. The 

 surface of 8 Cygni, if of solar brilliancy, should accordingly, 

 in order to enclose an equal quantity of matter with that 

 constituting the sun, be one hundred times more extensive 

 than the solar surface ; that is to say, the star must be one 

 thousand times less dense than the sun. If these results 

 be inadmissible, then nothing remains but to revert to 

 photospheric diversities of lustre. 



From certain of the Algol-stars, however, a direct 

 measure of brilliancy may in time be derived. Thus, 

 spectroscopic determinations of relative velocities in the 

 system of Y Cygni will give, apart from any hypothesis, 

 the joint mass of its two equal components ; their bulk 

 can be inferred from the duration of their phases, once the 

 size of their orbits is known ; and should it prove possible 

 to ascertain their distance from the earth, their total 

 light — conseijuently, their photospheric brilliancy — will 

 follow at once. Faithfully yours, 



Agnes M. Clekke. 

 [I do not feel that we can safely assume that all stars 

 are composed of similar materials. The uniform diii'erence 

 in the colour of the larger and smaller components of 

 bmary stars seems to point to a difference of composition 

 rather than to a difference in the temperatures of the large 

 and small stars (see <>hl and Xeiv Antrononvj, p. 784). And 

 the densities of the planets of the solar system seem to point 

 to the conclusion that the different planets must be 

 composed of different materials ; otherwise it is difficult 

 to conceive why Mercury, which is smaller than the earth 

 and larger than the moon, and presumably hotter than 

 either of them, should be denser than either the earth or 

 the moon. The earth is about 5-6(j times as heavy as 

 water, or about twice as dense as the average density of 

 the rocks which compose its surface — a fact which might 



be accounted for on the assumption of uniform composition, 



by assuming that the matter composing the central part.s 

 of the earth is materially compressed by the enormous 

 weight of the overlying strata. The moon is about 3-46 

 times as heavy as water, and though presumably colder 

 than the earth owing to its larger surface compared with 

 its mass, it is conceivable that the moon might be com- 

 posed of similar materials less closely compressed under 

 the more feeble action of lunar gravity. But Mercury, 

 which has a diameter of about 3000 miles, is about 6-85 

 times as heavy as a similar sized globe of water — a density 

 which could hardly be accounted for, on the above assump- 

 tion as to uniform composition, even if it were assumed 

 that Mercury is much colder than the earth. 



If we were able to assert that all stars must be composed 

 of similar materials, and that the brightness of their 

 photospheres corresponds to the boiling temperatures of 

 the same substance, there might still be a difference in 

 the apparent brightness of the photospheres of stars, caused 

 by differences of absorption above the photosphere due to 

 differences in the density and mass of the stars, as well as 

 to differences in the explosive energy which carries matter 

 into the regions of the chromosphere and corona above the 

 photosphere. Possibly we are not even warranted in 

 assuming that such differences must give rise to a 

 difference in the type of stellar spectrum emitted. 



While generally concurring with Miss Clerke, it seems 



to me that we cannot safely assume that it is improhahle 



that one star is a thousand times as dense as another 



star— in fact, the evidence afforded by the disposition of 



the stars on the nebulous streams of the Hercules cluster 



seems to me to point to the conclusion that these stars 



are very little denser than the nebulous matter surrounding 



them, and consequently that their average density is 



probably less than a thousand millionth of the density of 



our sun. — A. C. Ranyakd.] 



— ►-*.* — 



EXPLOSIONS IX THE SUN. 



To the Editor of Knowledge. 



Dear Sir, — I have read your most interesting and sug- 

 gestive article on the solar photosphere in the October 

 number of Knowledge, and in reference to this will you 

 permit me to draw your attention to a point which I 

 should much like to see discussed in your valuable 

 periodical '? I refer to the suppose dexplosive combinations 

 of gases at the solar surface. 



It appears to be generally assumed that when two 

 (mixed) gases, having a strong chemical affinity, are 

 cooled from a very high temperature to the point of dis- 

 sociation, a violent explosion will be the result. Now I 

 understand by the term " explosion " a sudden, almost 

 instantaneous, increase of volume, accompanied (or caused) 

 by a great rise of temperature, the potential energy due 

 to the chemical aftinity having been suddenly converted 

 into kinetic molecular energy, or heat. But this can only 

 occur when the molecules of the combining gases are in 

 unstable equilibrium previous to the explosion. Thus 

 when hydrogen is mixed with an equivalent of oxygen at 

 ordinary temperatures the mixture is unstable, and on 

 applying a relatively minute force the gases explode, an 

 enormous amount of heat being evolved, so that the 

 resulting water vapour, which otherwise would only occupy 

 two-thirds the original volume, expands to many times 

 that volume. 



But suppose the two mixed gases to be at their dissocia- 

 tion temperature, then the force which we call chemical 

 affinity is exactly balanced by the molecular repulsion due 

 to the extremely high velocity of the molecules, and when 

 this velocity is slightly reduced by cooling, combination 



