August 13, 1891] 



NATURE 



;55 



forced in, owing to the previous heating having caused a Jarge 

 percentage of combination, and hence the production of a partial 

 vacuum. Even after taking the rigid precautions to insure 

 purity above dcsciibed, no definite quantitative rule connecting 

 the time and percentage of conibination has been discovered, 

 experiments performed simultaneously upon similarly treated 

 mixtures yielding widely different results ; showing that the 

 irregularities of glass surfaces, even after removal of their air- 

 films, are quite sufficient to modify very sensibly the conditions 

 under which combination occurs. 



The additions to the Zoological Society's Gardens during the 

 past week include an Egyptian Gazelle [Gazellu donas) from 

 North Africa, presented by Mr. S. C. Saunders ; a Ring-tailed 

 Coati {Nasua rufa) from South America, presented by Mr. 

 Edward J. Brown ; two Herring Gulls {Lams argcntattis), 

 British, presented by Mr. T. A. Cotton ; two White-bellied Sea 

 Eagles {Haliaettts leiicogaster) from Australia, presented by Mr. 

 Hugh Nevill, F.Z.S. ; a Lesser Sulphur-crested Cockatoo 

 {Cacaina sulphurea) from Moluccas, presented by Miss Partridge ; 

 three Barbary Turtle Doves {Tiirlur risorius) from North Africa, 

 s presented by Miss D. Bason ; an Indian Cobra {iVaia tripudians) 

 I from India, presented by Mr. H. E. Lindsay ; two Harnessed 



f Antelopes ( Tragelaphus scripUis 6 9 ) from Gambia, a 



[ Paradoxure {Paradoxurus aureus) from Ceylon, two Grey 

 ' Ichneumon {Herpestes griseus) from India, four grey Parrots 

 (Psittacus erithacus) from West Africa, deposited. 



OUR ASTRONOMICAL COLUMN. 



' The Spectrum of j3 Lyr/e. — A study of twenty-nine photo- 



s graphs of the spectrum of ^ Lyra; has led to some interesting 

 ', results, noted by Prof. E. C. Pickering in Astronoinische Nach- 

 5 richten, No. 3051. The spectrum of this star contains, in 

 * addition to the absorption lines, several bright lines, the most 

 ' conspicuous of which are about A\ 486, 443, 434, 410, 403, and 

 \ 389, to me a three-figure reference. The lines near A 443 and 

 A 403, are two of the most prominent lines in the spectra of the 

 Orion stars, and the remaining four coincide with the hydrogen 

 lines F, G, h, and a. From the investigation it appears that 

 these bright lines change their positions, so that sometimes they 

 have a greater wave-length than the corresponding dark lines, 

 whilst at other times the reverse is the case. In some of the 

 photographs several bright lines are double, and the dark lines 

 are also not free from changes. This naturally led to the in- 

 quiry as to whether the changes were connected with the varia- 

 tions of the star's brightness. Starting from a minimum of 

 brightness there is a maximum at 3d. 5h., a secondary minimum 

 at 6d. iih., another maximum at gd. i6h., and then the prin- 

 cipal minimum is again reached after a total period of I2d. 22h. 

 The point of interest is that the fourteen plates in which the 

 wave-length of the bright lines was increased were taken during 

 the first half of this period of variation — that is, before the 

 secondary minimum ; whilst on the eleven plates taken during 

 the second half of the period the displacement was towards the 

 blue end of the spectrum. And since the photf^graphs extend 

 over more than four years, there can he litile doubt that the 

 displacements are intimately connected with the variations of 

 the star's brightness. One of the explanations s-uggested by 

 Prof. Pickering to account for the observed phenomena is that 

 the bright lines are emitted by an object revolving in a circular 

 orbit round the principal star, with a maximum velocity of about 

 300 miles per second, and completing its circuit in a period of 

 1 2d. 22h. The corresponding periastron distance is about 

 50,000,000 miles. If this be so, j3 Lyra: is a binary of the /8 

 Auriga; type, but differing from it in the fact that the component 

 stars have unlike spectra. The phenomena could also be pro- 

 duced by a meteor stream, or by an object like the sun, rotating 

 I in 1 2d. 22h., and having a large protuberance on it extending 

 over more than 180" of longitude. The study of the additional 

 photographs which are being taken will doubtless elucidate the 

 ' matter. 



The Polarization Theory of the Solar Corona.— In 

 the Publications of the Astronomical Society of the Pacific, 



NO. II 37, VOL. 44] 



vol. iii. No. 16, i?9i, Prof. Frank H. Bigelow gives some 

 further results of bis investigations of coronal forms, and arrives 

 at some new results. It can be shown that in the case of 

 repulsion of matter in a spherical rotating body like the sun, 

 two poles of repulsion are formed, and the body is polarized 

 about an axis. Within the body the lines of force are parallel 

 to the axis of polarization, and their curvature outside the surface 

 may be calculated. Applying these considerations to the similar 

 coronal forms exhibited in the eclipse photographs of July 1878 

 and January and December 1889, Prof. Bigelow finds that the 

 axis of polarization is at the surface of the sun about 4^° from 

 ' the axis of rotation, and taking the radius of the sun as 866,500 

 • miles; the length of the axis to which the lines of force are 

 parallel is 1,729,700 miles. Its direction is fixed, and in 1878 

 the north and south coronal poles had the positions, north pole 

 = 20i°'2, south pole = 30i°'6, when referred to the ascending- 

 node of the sun's equator on the plane of the ecliptic. If 

 138 + 349°'85, 151 + 3ii°-40i and 12 + 3i2°-55 be taken as 

 the number of revolutions and the angular excess during the 

 three intervals between the dates of the above eclipses, the mean 

 daily motion in longitude at the latitude of the coronal pole, 

 85° 5, i^ found to be 13°"I3307. From this the following periods 

 of the sun's rotation in latitude 85°'5 is deduced — 



Sidereal period 27-41 I7ld. = 27d. 9h. 52m. 525. 

 Synodic period 29-6358od. = 29d. I5h. 15m. 33s. 

 The formula proposed to express I he rotation-period in different 

 solar latitudes is X = 862' - 76' sin /, where X is the mean 

 daily motion in minutes, and / the latitude. With these ele- 

 ments it is possible to predict the positions of the coronal poles 

 at any epoch, and in consequence the relative form of thecoiona 

 at the time, as seen from the earth. A comparison of the calcu- 

 lated results and photographs, obtained during some recent eclipser, 

 displays a striking concordance. The investigation " also serves 

 to strengthen the conviction that the sun-spots are probai ly 

 formed by the descent of material from the extremities of the 

 coronal streamers, in a vertical direction upon the sun." 



Observations of the Motion of Sirius.— -At the Berlin 

 Academy of Sciences on June 4, Prof. Vogel communicated 

 some observations of the motion of Sirius in the line of sight. 

 Using the iron spectrum as the term of comparison with the 

 spectrum of the star, it was found that the velocity of approach 

 on March 22 was i'96 geographical miles per second with 

 respect to the sun. With hydrogen comparison lines the velocity 

 found was 1'73 miles per second. 



Return of Encke's Comet.— A telegram from the Lick 

 Observatory to Prof. Kriiger, announces that Encke's periodic 

 comet has been observed on its return by Mr. Barnard on 

 August I "9958 G. M.T., in the position R.A. 3h. 55m. 20 6s., 

 Decl. 29° 59'-i N. 



ON SOME TEST CASES FOR THE MAXWELL- 

 BOLTZMANN DOCTRINE REGARDING DIS- 

 TRIBUTION OF ENERGY} 



(i) ATAXWELL, in his article {Phil. Mog., i860) "On the 

 ^^^ Collision of Elastic Spheres,"' enunciates a very re- 

 markable theorem, of primary importance in the kinetic theory 

 of gases, to the effect that, in an assemblage of large numbers of 

 mutually-colliding spheres of two or of several different magni- 

 tudes, the mean kinetic energy is the same for equal numbers of 

 the spheres irrespectively of their masses and. diameters ; or, in 

 other words, the time-averages of the squares of the velocities of 

 individual spheres are inversely as their masses. The mathe- 

 matical investigation given as a proof of this theorem in that first 

 article on the subject is quite unsatisfactory ; but the mere enun- 

 ciation of it, even if without proof, was a very valuable contribu- 

 tion to science. In a subsequent paper ('.'Dynamical Theory of 

 Gases," Phil. Trans, for May 1866) Maxwell finds in his equa- 

 tion (34) (" Collected Works," p. 47), as a result of a thorouj^h 

 mathematical investigation, the same theorem extended to in- 

 clude collisions between Boscovich points with mutual forces 

 accoiding to any law of distance, provided only that not more 

 than two points are in collision (that is to say, within the dis- 

 tances of their mutual influence) simultaneously. Tait confirms 

 Maxwell's original theorem for colliding spheres of difl"erent 



' Paper read at tlie Rojal Society by Sir William Thomscn, D.C.L., 

 P.R.S., on June 11, 1891. 



