6o6 



NATURE 



[April 2^, 1889 



nothing of strictly scientific interest in the book ; but we 

 may say that Mr. Baddeley has the great merit of always 

 trying to see things with his own eyes, that in his judg- 

 ments of men and places there is no trace of any kind of 

 British prejudice, and that his style is fresh and interest- 

 ing. Among the subjects of which he has something 

 to say are Bulgaria, Buenos Ayres, Constantinople, and 

 Tunis. 



LETTERS TO THE EDITOR. 

 [ The Editor does not Iwld himself responsible for opinions ex- 

 pressed by his correspondents . Neither can he undertake 

 to return, or to correspond with the writers of, rejected 

 manuscripts intended for this or any other part of NATURE. 

 No notice is taken of anonymous communications. ] 



Large Fireball. 



On the night of Monday, April 15 last, at I2h. 26m., a 

 meteor of very exceptional proportions was observed from many 

 parts of the country. The full moon was shining at the time, and 

 near its meridian passage, but the brilliancy of the fireball was 

 such that it vividly illuminated the sl^y and landscape with a 

 flash which many people mistook for sheet lightning. Several 

 observers describe the meteor as larger and considerably more 

 brilliant than the moon, and at Swindon and Ramsbury a detona- 

 tion was heard. At the former place the meteor " appeared so 

 close that people thought it descending upon the town ; it startled 

 the rooks out of the trees, and suddenly illuminated the country 

 round like the electric light." 



At Worthing the meteor is described as falling almost per- 

 pendicularly from west-north-west to due north. At Clapham 

 it seemed to take a slanting course from the south towards 

 the west. At Bath the meteor was seen in the east moving 

 horizontally, at a considerable altitude, from right to left. Prob- 

 ably therefore the body was situated over the region of Berks, 

 but the data are altogether too imperfect to admit of trustworthy 

 inferences either as to its position or height above the earth's 

 surface. 



It will be important if other observers can furnish accounts in 

 which the position and direction of this fireball are more 

 definitely given. The phenomenon was clearly one of un- 

 common character, but its apparition occurred at such a late 

 hour that comparatively few persons must have witnessed it. 



Bishopston, Bristol, April 20. W. F. Denning. 



Variable Stars and the Constitution of the Sun. 



In Nature of March 21 (p. 492), Mr. A. Fowler has given 

 an excellent account of my theory of the constitution of the sun, 

 but he has not succeeded so well in describing my theory of the 

 variable stars. I may here draw attention to some of the cardinal 

 points not sufficiently noticed by Mr. Fowler : — 



(i) The proof that the chemical combination at the external 

 layers of the stars must be intermittent with regular intervals. 

 This proof, pp. 8, 9, and 10, is mentioned as the very basis of 

 my theory of the sun and stars. 



(2) The fact that the intermittent eruptions of heat, if produced 

 in this manner, cannot become visible through some perceptible 

 increase of the heat of the star. 



(3) The high probability that in the case of some red stars the 

 vapours, noticed by the spectroscope in their external layers, are 

 cooled to their dew-point, and so, with the smallest radiation of 

 heat, are made ready to change into clouds, which suddenly 

 withdraw the invariable inner light from our view. 



(4) The hi><h probability, too, that only those stars will be 

 variable in which the external layers are cooled down to their 

 dew-point, and that the intermittent eruptions of heat become 

 visible because at intervals they cause the evaporation of the 

 clouds, which surround the invariable inner light of the star 

 during the minimum. 



Therefore only those stars of Class III. will be variable, whose 

 external vapours are cooled to their dew-point. The others also 

 have their periodical eruptions of heat, but these are inperceptible 

 to our eye. 



The changes of the variable stars, therefore, are never associated 

 with important changes of temperature. It would be possible 



for them to take place without the least change of temperature if 

 the calories produced by an eruption of heat were entirely used 

 for the evaporation of the clouds. And so even an increase of 

 ternperature of i° would be sufficient to make a seemingly 

 extinguished orb glitter again as a new star, whilst a similar 

 decrease of temperature would suffice to restore the veil, which, 

 steadily growing thicker, would make it invisible again, perhaps 

 for centuries together. 



Mr. Fowler is mistaken in saying that I do not seem to be 

 aware that Algol is one of the hottest stars in the heavens, and 

 that its spectrum is the same at maximum as at minimum. On 

 the contrary, on p. 15, I have stated the exceptional case of 

 Algol, and, seeing the impossibility of making it agree with my 

 theory as in the case of those Algol-stars which are red at 

 minimum, I ascribed the variability of Algol to the periodicity of 

 its spots. Moreover, I added that this seeming contradiction t 

 my theory was oniy a physical peculiarity of little important 

 For the spots, too, are caused by periodical eruptions of heat i: 

 clouds. The only difference is that the clouds — I mean on the 

 sun and Algol — are photospheric, and by vaporization cause dark 

 spots which diminish the light, whilst in the cooler red stav<, 

 the clouds form a dark veil round the star, and therefore by 

 vaporization increase its interior light. 



In this defence of my theory, gradually passing from the vari 

 able stars by means of Algol to the sun, I must observe that m\ 

 theory by no means suggests, as Mr. Fowler thinks, that the 

 sun should have more spots in the Polar regions than near the 

 equator. I only say that the spots must be found in parallel 

 zones ; of the breadth of those zones I say nothing. The spot- 

 can only be produced in places where the temperature and the 

 chemical compositions work together to produce eruptions of 

 heat. As the places of equal chemical composition and of 

 equal temperature are only possible in the photosphere in two 

 parallel zones of equal latitudes on opposites sides of the 

 equator, it is plain that the spots must be produced there. 



I conclude with an expression of gratitude to the Editor of 

 Nature and to Mr. Fowler for the trouble they have taken in 

 noticing my theory. A. Brester, Jz. 



Delft, April i. 



In reply to Dr. Brester I have to remark, in the first place, 

 that I made no attempt to give all the details of the theory, 

 limitations of space not permitting. One of my principal argu- 

 ments against the theory was that, if it were true, all cool stars 

 should be variable, and I still see no reason to alter my views. 

 The observations of the red stars by Duner show that the 

 spectra of some of the stars which are not variable are identical 

 with some of those which are. For example, the spectra of 120 

 Schj., and D.M. -l- 47"'229i, which are not variable, are exactly 

 like those of x Cygni and R Leonis. The compositions and 

 temperatures of the gaseous surroundings of these bodies are 

 therefore similar, and there is no reason, from Dr. Brester's 

 pomt of view, why one should be variable more than another, 

 since, if they are cooling, they all start cooling under exactly 

 equal conditions. (It may fairly be assumed that the spectra of 

 the variables have been generally taken at maximum.) The 

 cooling to dew-point is therefore not in question in the variable 

 any more than in the apparently invariable stars. 



The high probability that by far the greater number of vari- 

 ables are uncondensed meteor-swarms which are increasing in 

 temperature, as demonstrated by Mr. Lockyer, is also obviously 

 against any theory of variability which assumes a state of 

 cooling. 



The same objections which apply to the red stars apply also to 

 the "unimportant" case of Algol; there are many other stars 

 with identical spectra, and therefore temperatures, which 

 exhibit no variability at all. 



With regard to the sun, I remarked that the theory would 

 suggest that spots should be most numerous at the poles, for the 

 reason that it would be there where the atmosphere in the 

 neighbourhood of the sun would be coolest, and where, there- 

 fore, chemical combinations would be most likely to take place. 

 To this Dr. Brester replies that his theory only requires that 

 spots should be formed in equal zones on opposite sides of the 

 equator, and says nothing about the breadth of the zones. Of 

 course, if it be assumed that the substances present in Polar 

 regions are not such as to form combinations competent to pro- 

 duce spots, the difficulty is overcome, but an explanation 

 depending upon such an assumption is far from satisfactory. 



London, April 5. A. FoWLER. 



