May 25, 19 1 8] 



NATURE 



DiSTANCK OK THE Orion Neblla. — An -interesting I 

 estimate of tlie distance of the Orion nebula has been 

 made by Prof. W. H. Pickering (Harvard Circular 

 No. 205). From a consideration of the brightnesses 

 and distribution of the stars in the nebula and in the 

 surrounding region, it is concluded that practically 

 all the stars within the nebula are of type B, and 

 that there are no stars in the nebula fainter than 

 fifteenth magnitude. Since we are looking very nearly - 

 along the axis of the great spiral nebula which 

 stretches over nearly the whole length of Orion, and 

 is connected with the great nebula, all the stars asso- 

 ciated with it must be at approximately the same 

 distance from us. Russell has shown that only very 

 massive stars can attain the colour of type B; and 

 assuming 105 as the mean magnitude of the stars 

 within the nebula, while the average absolute magni- 

 tude of such stars may be taken as — 10, it follows 

 that the distance of the nebula is 6520 light-years, or 

 that the parallax is 00005". Among the interesting 

 results which follow, the mean diameter of the bril- 

 liant Huygehian region is found to be 63 light- 

 years, and the distance between the extreme stars of 

 the trapezium 068 light-year. It is also cal- 

 culated that Rigel is 2,ioo,.ooo times as bright as the 

 sun, thus far exceeding Canopus, for which Walkey 

 estimated a brightness of 5o,txK) times that of the sun. 



TERRESTRIAL MAGNETIC OSCILLATIONS. 



THE Pfiper referred to below ^ is an important con- 

 tribution to our knowledge of oscillations in the 

 magnetic elements, especially those of shorter period 

 t(.rm€d "pulsations" by van Bemmelen. The records 

 were obtained in an underground chamber near the 

 Marine Biological Laboratory at Misaki,, between 1910 

 and April, 1914, with a special set of very sensitive 

 magnetographs, designed by Prof. Tanakadate. The 

 magnetographs, which recorded the north (N), 

 west (W), and vertical (V) components, show several 

 original features. The V instrument, which worked 

 very satisfactorily, had the magnet carried by hori- 

 zontal quartz fibres. The sensitiveness of the instru- 

 ments was about 0157 per i mm., and the time-scale 

 about 3i mm. to the minute. 



The original object was to ascertain whether seismic 

 movements were accompanied by magnetic waves. 

 No certain connection was established, but many in- 

 teresting records of pulsations were obtained. The distri- 

 bution of pulsations throughout the twenty-four hours 

 varied markedly with the period, waves with periods less 

 than seventy seconds having their maximum frequeAcy 

 during the day, and those with periods longer than 

 ninety seconds having their maximum during the 

 night. Periods shorter than thirty seconds were rare. 

 Pulsations in V were almost facsimiles of those in 

 N, except that they were of smaller amplitude and 

 had a retardation of phase. As the periotl became 

 longer, the ratio borne by the amplitude of the V to 

 that of the N pulsation increased, while the difference 

 in phase diminished. The hour of the day seemed 

 without direct influence on the value of the ratio. The 

 relation between the pulsations in N and V\', on the 

 other hand, depended largely on the hour of tlie day. 

 Regarding north and west as the positive directions 

 in the two cases, it was found that agreement in phase 

 between N and W pulsations was most frequent in 

 the early morning, whilst direct opposition in phase" 

 was most frequentr in the evening. Cases in which 

 the N pulsation was largely dominant were most 

 frequent near noon and near midnight. 



Generally there was a marked tendency in the vector 



"On Rapid Periodic V.-iriations of Tcrrestrml Magnetism." By 

 I rahiko Terada. lournal of the College of Science, Imperial University 

 of Tokyo, vol. xxxvii., 1917, Art. o. Pp. 85 + 5 plates. 



in the horizontal plane to rotate, after the fashion • 

 first described by R. B. Sangster for longar-period 

 movements. -According to the author, in pulsations at 

 Misaki, clock-wise rotation is most frequent between 

 sunrise and noon, and again between sunset and mid- 

 night, anti-clock-wise rotation predominating in the 

 intermediate hours. One interesting feature, which 

 the author thinks may possess considerable signifi- 

 cance, is a tendency when pulsations start abruptiv 

 for N to show a rapid rise. He is disposed to attri- 

 bute pulsations to fluctuations in the electrical cur- 

 rents in the upper atmosphere, to which the regular 

 diurn;il magnetic variation is now generally ascribed. 

 If, as he thinks most likely, pulsations arise simul- 

 tantausly and not successively at different stations, 

 the currents in the upper atrnosphere probably fluc- 

 tuate in intensity as well as in position. This might, 

 he thinks, arise from vertical oscillations in limited 

 portions of the upper atmosphere. A variety of mathe- 

 matical problems relating to oscillating linear electric 

 currents are worked out. The plates at the end con- 

 tain numerous interesting examples of pulsations. 



C. Chree. 



GLOBULAR STAR CLUSTERS. 



MR. HARI.OW SHAPLEY'S preUminary work 

 on the distances of the globular clusters at- 

 tracted much attention two years ago. He has since 

 then diligently piu-sued the subject, and gives an 

 interesting summary of the progress of his researches 

 in Pubns. -\str. Soc. Pac, February, 1918. 



His methods are :— (i) To determine the photographic 

 and photo-visual magnitudes of the cluster stars by 

 photographs on ordinary and panchromatic plates. 

 The colour-indices of the stars are thus determined 

 and their spectral types inferred. The fact that stars 

 are found in the clusters quite as blue as the B stars 

 in our neighbourhood leads to the assumption that 

 light absorption is negligible. The distances can then 

 be inferred, making assumptions on the absolute 

 magnitudes of stars of different spectral types 



(2) The work of Miss Leavitt, Hertzsprung, and 

 Shapley shows that the absolute magnitude of Cepheid 

 variables is a function of the period of light variation. 

 A curve is given in the article, from which the fol- 

 lowing values have been measured : — 



Since the cluster variables conform mainly to the- 

 Cepheid type, this affords a very accurate means of 

 obtaining the distances of clusters. Mr. Shapley 

 notes that the long-period Cepheids are the most 

 luminous of all stars. The longest observed period 

 is about 130 days, absolute magnitude' — 6-8 (indicating 

 about 50,000 times the luminosity of the sun). 

 Cepheid variables are also notable for their rapid 

 motion, whach appears to average more than 

 100 km. /sec. 



(3) Bv the above methods the average absolute 

 magnitude (photographic) of the brighter stars of the- 

 different clusters (twenty-five stars selected from each 

 cluster, rejecting the five brightest) is found to be 

 — 1-5. Making this assumption for other clusters, 

 we can estimate their distance without waiting, for 

 more detailed researches. 



(4) There is found to be a fairly close correlation 

 between distance and apparent diameter, indicating 

 that the linear diameter of a cluster is a function of 

 its distance. With diameter 1-4' corresponds distance 



NO. 2534, VOL. lOl] 



