352 



NA rURE 



[March i6, 1922 



Our Astronomical Column. 



The Partial Solar Eclipse of March 28. — 

 This eclipse is a successor, after twelve lunations, of 

 the large eclipse of last April. On this occasion the 

 central hne has moved southwards, crossing Brazil 

 and the Sahara. The whole of the British Isles enjoy 

 a partial ecHpse, the magnitude of which diminishes 

 from 0-20 at Greenwich to zero at the Shetlands. At 

 Greenwich the eclipse begins at 1.19 p.m. at the 

 lowest point of the disc, attains its greatest phase 

 at 2.15, and ends at 3.8. The only observations of 

 importance that can be made are the timing of 

 the contacts and the watch for lunar mountains at 

 the edge of the moon. 



Photography of the Ultra-violet Solar Spec- 

 trum. — MM. Charles Fabry and H. Buisson give an 

 account of their work in this field in the Astrophys. 

 J ourn.- {December) . They state that existing material 

 on this region is unsatisfactory ; Rowland's map is 

 disturbed by a spedtrum of another order ; that of 

 Higgs stops at \3000 ; and Cornu's map was made by 

 hand from Simony's photographs. Fabry and Buis- 

 son's photographs were taken at Marseilles in May 

 and June 1920. They used two quartz spectrographs, 

 the prism edges in them being perpendicular to each 

 other, and state that they thus ehminated diffuse 

 light of longer wave-length, which is one of the chief 

 difficulties in this region. Another difficulty, due to 

 the rapid increase in exposure-time as the wave-length 

 grows shorter, was overcome by using an occulting 

 screen, which was moved by hand at a calculated rate 

 along the spectrum during exposure ; the extreme 

 exposures were in the ratio of i to 1000. They state 

 that they were able to estimate the varying amount 

 of ozone present in the air by its absorbing effect on 

 these short waves. They give the mean amount 

 of ozone in the air as 0-4 c.c. per cubic metre ; 

 as this is much larger than the amount near the 

 ground, they conjecture that it is mainly at a great 

 height (say 50 km.). Two reproductions are given of 

 the spectrum from \2898 to X3150 ; the definition is 

 good and numerous lines are shown, the wave-lengths 

 of which are promised shortly ; they include the 

 lines designated rST/U. The region X2965 to X3030 

 appears on both photographs, the agreement being 

 good, but with differences in the relative intensities 

 of lines. 



The authors note that the intensity of spectrum at 

 the sun's limb is about half that at the centre, this 

 factor remaining nearly constant throughout the 

 region studied. They conclude that the fairly abrupt 

 termination of the spectrum at X2898 is due to the 

 terrestrial, not the solar, atmosphere. 



A Criticism of Majorana's Theory of Gravita- 

 tion. — -The Astrophys. Journ. for December contains 

 a criticism of this theory by Prof. H. N. Russell. It 

 will be remembered that the theory, which was based 

 on the apparent alteration in the weight of a mass 

 of lead when surrounded by mercury, asserts that 

 gravitation is subject to absorption by intervening 

 matter. Prof. Russell shows that a sphere composed 

 of homogeneous spherical layers would still attract 

 external bodies according to the inverse square law ; 

 as though its mass were concentrated at the centre ; 

 thus it would not give rise to any motion of perihelia, 

 as some have supposed. However, the apparent mass 

 of the sphere, as measured by its attraction on ex- 

 ternal bodies, w^ould be less than its true mass, the 

 deficiency increasing as the central condensation of 

 the matter in the sphere increases. It is then proved 

 that a massive planet would be considerably nearer to 

 the sun than a planet of small mass with identical 

 period. Jupiter, in particular, would be nearer to the 

 sun by i per cent, than the accepted distance; but 



NO. 2733, VOL. 109] 



this is inadmissible, since it would produce an error 

 of 7' in the geocentric place when the planet was in 

 quadrature. 



Another test applied is the difference between the 

 heights of the tides when the sun and moon respec- 

 tively are above and below the horizon. It is shown 

 that the theory would give results totally at variance 

 with observation. Lunar theory supplies a further 

 test, since the attraction of the sun on the moon 

 would produce an acceleration greater by i-6 per cent, 

 than that experienced by the earth. Prof. Russell 

 concludes that Majorana's theory must be abandoned ; 

 he suggests tentatively that the phenomena observed 

 by Majorana might be explained on the assumption 

 that the presence of a large mass itself diminishes the 

 masses of neighbouring bodies : " the space-curvature 

 produced by one mass of matter might be modified 

 by the superposition of that due to another." He 

 admits, however, that this theory also gives rise to 

 difficulties. 



The Definition of a Nova. — The Rev. J. G. 

 Hagen directs attention to the advisability of formu- 

 lating a definition of a nova which takes into account 

 our present knowledge of the nebulous material sur- 

 rounding new stars, and also class O stars and 

 planetary nebulae {Astrophysical Journal, vol. 54, 

 No. 4, p. 229). Seeliger, so long ago as 1886, put 

 forward the idea that temporary or new stars 

 originated from collisions between stars and cosmic 

 nebulas, and this theory is now generally adopted. 

 Dr. Hagen's view is that by assuming that the 

 nebulae are cometary — that is, become luminous at 

 the approach of stars — many facts of observation are 

 explained. Thus the observations account for nearly 

 all the peculiarities of new stars, especially the range 

 of magnitude, shape of the light-curve, character of 

 the spectrum, and the rarity of occurrence ; they 

 also show a physical similarity to the O stars and 

 planetary nebulae. 



The collision between the star and the nebula is 

 discussed and the possible results of four types of 

 encounters are given. Then it is shown that a transit 

 without contact might produce a nova with a rela- 

 tively small range of variation. If the star grazed 

 the nebula and the occurrence was repeated occa- 

 sionally an irregular variable would be the result. 

 The passage of the star through the nebula might 

 result in a nova of the ordinary type and in a star 

 with a nebulous envelope like an O star. If the 

 star were captured by the nebula a variable of the 

 8 Cepheid type or a planetary nebula would be 

 evolved. The definition of a nova is thus given : "A 

 cometary nebula brought temporarily into close 

 proximity or contact with a bright star." 



The Stellar Magnitude of the Ringless 

 Saturn. — Mr. J. van der Bilt investigates this subject 

 in Bull. Astron. Instil, of Netherlands, No. 6. Capella, 

 Vega, Procyon, Spica, Pollux, Regulus, and Polaris 

 were used ' as comparison stars, their magnitudes 

 being revised by the author's observations. The 

 value obtained for Saturn in opposition (Ringless) is 

 0-84 mag., that given by Miiller being 0-877. The 

 chief value of this work Hes in the deduced value 

 of the planet's albedo. Similar determinations made 

 when the rings are wide open will give a determina- 

 tion of their albedo, and may give a clue to the 

 constitution of the ring-particles. Mr. van der Bilt 

 has also investigated the effect of phase angle on 

 magnitude, and found the change for 1° of Satumi- 

 centric angle between sun and earth to be 0-050 

 magnitude, which is in good accord with the mean 

 of other observers. 



