208 



KNOWLEDGE 



[November 1, 1898. 



will take place without violence and without producing 

 any heat, for it is obviously paradoxical to suppose the 

 temperature to rise ; if it did, dissociation would again 

 take place. Consequently there will be no expansion, but, 

 on the contrary, a contraction, as the compound molecule 

 will require less space than its constituents. 



My view, therefore, is that chemical combinations taking 

 place under the solar conditions, namely, the cooling of 

 gaseous mixtures from temperatures above their dissocia- 

 tion points, will simply result in contractions of volume, 

 and I cannot see that these will be of the nature of 

 explosions. 



Of course it may be said that explosive combinations 

 might occur when the component gases are not mixed, but 

 cool separately to a temperature considerably below the 

 dissociation pomt. A subsequent accidental mixing would 

 then, no doubt, set free the energy due to their chemical 

 union. But in this case also true explosion would not 

 occur, as the gases could not be dift'used into each other 

 and then Ignited as in terrestrial explosions. Under solar 

 conditions ignition would always occur immediately the 

 gases came into contact, and would continue, only in the 

 region of contact, as a surface flame until one or other 

 element was entirely consumed. 



When, on the other hand, the combined elements are 

 re-heated in the lower strata of the solar photosphere and 

 re-dissociated, expansions will occur, which possibly, if 

 confined by some means, would be sufficiently violent to 

 account for the observed outbursts. But that these ex- 

 pansions would not necessarily be of a violent character 

 seems probable, if we may compare the dissociation of 

 compounds in the sun with the change in vapour density 

 of certain substances which can be experimented upon in 

 our laboratories. Thus nitric peroxide at low temperatures 

 has a vapour density approximating to the formula N^O^, 

 but on heating a change takes place independently of the 



law of Charles, the double molecule (^q") being dis- 

 sociated into NO, + NO, ; that is, at a given temperature 

 and pressure (above the dissociation temperature) the 

 vapour will occupy twice the volume it would have done 

 in the former condition. But this expansion is by no 

 means sudden and violent, occurring at the moment the 

 dissociation temperature is reached ; for according to the 

 accepted kinetic theory of gases, at any temperature, 

 indi^ddual molecules are regarded as moving with widely 

 different velocities, the temperature indicating the mean 

 velocity, and therefore dissociation begins at a far lower 

 temperature than the theoretical dissociation point, and 

 it will not be completed until a much higher temperature 

 has been attained. It follows from this, that as the vapour 

 or gas cannot augment its temperature through the required 

 range instantaneously, the dissociation and consequent 

 expansion will take time, and not therefore be explosive. 



A serious objection, however, to the theory of chemical 

 combinations and dissociations in the sun appears to mc 

 to be the absence of spectroscopic evidence either of the 

 compounds formed or of the combining elements. Under 

 the conditions I have explained above, one would expect 

 to find m the outrushing gases a mixture of dissociated 

 elements. What we actually do find is a mixture of calcium 

 and hydrogen and helium. Of the last-named element of 

 course nothing is known, but the other two at any rate 

 are both positive elements, and therefore unlikely to form 

 combinations or to possess any strong aiiinity for each 

 other. It is true that in the comparatively rare metallic 

 eruptions other elements (metals) are present, but these 

 are usually seen at the base of an ascending column of 

 heated gas, and seem to be merely thrown up from below 



by the violence of the outrushing calcium and hydrogen — 

 these latter elements, with helium, being the only invari- 

 able constituents of all classes of prominences. 



Thus it appears to me that chemical forces such as we 

 are acquainted with fail to account in a satisfactory manner 

 for these extraordinary explosions and outrushes from the 

 solar photosphere, of which I have on several occasions 

 had the good fortune to be an eye-witness. 



J. EVERSHED. 



To the Editor of Knowledge. 



Dear Sir, — The few words with which I concluded my 

 lecture to the British Association at Nottingham were, I 

 need scarcely say, not meant to herald a theory of the sun, 

 and it would be unbecoming on my part to enter into a 

 detailed discussion of solar physics. But, as you wish it, 

 I will just say in respect to Mr. Evershed's letter that 

 in reflecting on the past condition of the earth I never 

 pictured a homogeneous mixture of hydrogen and oxygen 

 developing by cooling into explosive or even tlame-like 

 combustion. All the probabilities seem to be against such 

 a distribution of the gases. In a heterogeneous distri- 

 bution, flames would, on Mr. Evershed's own showing, 

 become possible at a certain temperature. The steam 

 might be dissociated in hotter layers and the gases carried 

 again into heterogeneous circulation, just as steam mixed 

 with an inert gas is partly dissociated on passing through 

 a hot tube. These two points were all I contended for. 



As to the other matters raised in Mr. Evershed's letter, 

 I will only say that whilst not presuming to argue in 

 favour of chemical explosions as a source of solar eruptions, 

 I would advise caution to those who argue their impossi- 

 bility. Such reasoning as that oflered by Mr. Evershed 

 would never allow us to admit (apart from our experi- 

 mental knowledge) of such phenomena as the super-heating 

 or super-cooling of water or steam, or the existence of any 

 false equilibria. Iron changes its molecular condition at 

 a certain temperature. When cooled down from higher 

 temperatures the reverse change, which is attended with 

 evolution of heat, should set in gradually. As a matter of 

 fact it does not — we have the phenomenon of " recalescence " 

 — the iron suddenly glows again. This is Mr. Evershed's 

 paradox realized. False equilibria abound in chemistry ; 

 what is called chemical affinity seems with our present 

 knowledge most capricious, and I think he would be a rash 

 man who would deny the possibility of super-cooled chemical 

 systems. There is reason to believe that chemical affinity 

 varies periodically with temperature. Ozone which is formed 

 at ordinary temperatures is destroyed at a moderate heat, 

 and, it appears, formed again at much higher tempera- 

 tures. Chemical phenomena, in short, cannot be predicted, 

 as has been shown again and again, by a priori thermo- 

 chemical reasoning. This being so, I feel, as I expressed 

 myself at Nottingham, that one must not rashly exclude 

 chemical action as a factor in solar phenomena. 



Arthur Smithells. 

 [I agree with Prof. Smithells that Mr. Evershed's a 

 priori method of reasoning is dangerous and must not be 

 relied upon for the prediction of phenomena which cannot 

 be verified by experiment. The appearances observable 

 on the sun seem to point to the existence of rapid disturb- 

 ances very like explosions. I notice that Mr. Evershed, 

 who is a practical observer, speaks of the solar disturbances 

 as e.rplosiioiti, a word which implies rapid change of volume, 

 if not chemical action. There are great difficulties in 

 accounting for the rapid motions observed on any other 

 supposition. Our terrestrial winds are produced by 

 differences of temperature causing unequal expansions of 



