354 



NATURE 



[August 13, 1903 



detection of such variations in the spectra of hydrogen, 

 helium, and lithium, and has obtained indications that the 

 energy of the shorter waves is relatively increased with in- 

 crease of temperature, assuming that the temperature in 

 Geissler tubes rises with increased potential and current 

 strength. It is considered probtible that further laboratory 

 experiments combined with photometric or photographic 

 estimates of the intensities of the stellar lines may result 

 in a fairly accurate knowledge of the temperatures of some 

 of the stars ; great progress will have been made if the 

 temperatures can only be ascertained within one or two 

 thousand degrees. 



Observations of the Minima of Mira. — In No. 3888 of 

 the Astronomische Nachrichten, Prof. A. A. Nijland records 

 his observations of the last minimum of Mira, which took 

 place during December. Plotting his observations on a 

 curve, he found that the actual minimum occurred on 

 December 17, 353 days after the minimum of December 29, 

 1901, the magnitude on that date being 8-70 on the Harvard 

 photometer scale. 



The following table shows the differences between the 

 dates of minima as predicted by Guthnick (Astronomische 

 Nachrichten, No. 3745) and those actually observed :— 

 Observed Guthnick O-G 



1901 Feb. 16 ... 1901 March 6 .. - 18 days 

 „ Dec. 29 ... 1902 Jan. 31 ... -33 „ 



1902 „ 17 ... „ Dec. 28 ... -II ,, 



The Size of Stellar Systems. — In an editorial article 

 in the Observatory for August, a table is given which com- 

 pares the dimensions of various stellar systems with those 

 obtaining in the solar system. As the writer states, these 

 are not generally known or not remembered, therefore he 

 has tabulated a few of the more interesting and approxi- 

 mately known data, which must, however, only be taken 

 as approximations owing to the uncertainty of the original 

 data from which they are computed. 



Object 



Earth 



Saturn 9 '5 



Procyon 17-3 



Uranus 19*2 



Sirius 2I-I 



o Centauri 23-3 



Castor 27-5 



Neptune 30-1 



02 Eridani 34 5 



(B and C) 



rj Cassiopeioe i 447 



6 Ursse Maj 63*0 



61 Cygni 68 o 



Polaris ! 250 



Motion across the line 

 Separation of components j of sight, in millions of 

 miles per annum 



In astro- 

 nomical units 



Aldebaran. 



Eridani 



(A and B) 



282 



455 



In millions 

 of miles 



93 



883 

 1,608 

 1,782 

 1,962 

 2,167 

 2,557 

 2,792 

 3.207 



" 3.Q47 

 5,860 



6.324 

 23,250 

 26,226 

 42,315 



372 



316 

 465 



140 



2,000 



580 

 1,300 

 1,116 



133 



170 



2,000 



Recently Determined Stellar Parallaxes. — No. 10 of 

 the Publications of the Groningen Astronomical Laboratory 

 contains the details of the observations and reductions of 

 parallax for the stars and clusters " h and x Persei," " 745 

 Groombridge, " and " 61 Cygni and the surrounding stars." 

 The photographs from which the parallactic values were 

 determined were obtained by Prof. A. Donner, and have 

 been reduced by Prof. J. C. Kapteyn and Dr. W. de Sitter. 



In the summary given for the cluster h and x Persei, 178 

 stars are included, and it will be possible, when it has been 

 decided, from observations of their proper motions, whether 

 or not the individual stars actually belong to the cluster, 

 to determine the parallax of this cluster with extreme 

 accuracy. 



The parallax of 745 Groombridge relative to stars of the 

 mean magnitude 90 was found to be +o"o83 + o"o24, and 



NO. I/63. VOL. 68] 



on consideration of the star's magnitude (8-2) and its annual 

 proper motion (o"-64), -|-o"o68 was accepted as the most 

 probable value of this parallax. 



The final value of the parallax of 61 Cygni relative to 

 the four comparison stars (mean magnitude =74) is given 

 as -f-o"-326±o"o35 ; the plates from which this result was 

 obtained do not confirm the existence of any real difference 

 of parallax between the two components. 



No. 1 1 of the same Publications contains a discussion on 

 " The Luminosity of the Fixed Stars " by Prof. J. C. 

 Kapteyn. 



EXPERIMENTS IN RADIO-ACTIVITY, AND 

 THE PRODUCTION OF HELIUM FROM 

 RADIUM.^ 



(i) Experiments on the Radio-activity of the Inert Gases of 



the Atmosphere. 

 (^ F recent years many investigations have been made by 

 ^-^ Elster and Geitel, Wilson, Strutt, Rutherford, Cooke, 

 Allen, and others on the spontaneous ionisation of the gases 

 of the atmosphere and on the excited radio-activity obtain- 

 able from it. It became of interest to ascertain whether 

 the inert monatomic gases of the atmosphere bear any 

 share in these phenomena. For this purpose a small 

 electroscope contained in a glass tube of about 20 c.c. 

 capacity, covered in the interior with tin-foil, was em- 

 ployed. After charging, the apparatus if exhausted re- 

 tained its charge for thirty-six hours without diminution. 

 Admission of air caused a slow discharge. In similar 

 experiments with helium, neon, argon, krypton, and xenon, 

 the last mixed with oxygen, the rate of discharge was pro- 

 portional to the density and pressure of the gas. This 

 shows that the gases have no special radio-activity of their 

 own, and accords with the explanation already advanced 

 by these investigators that the discharging power of the 

 air is caused by extraneous radio-activity. 



Experiments were also made with the dregs left after 

 liquefied air had nearly entirely evaporated, and again 

 with the same result ; no increase in discharging power is 

 produced by concentration of a possible radio-active con- 

 stituent of the atmosphere. 



(2) Experiments on the Nature of the Radio-active 

 Emanation from Radium. 



The word emanation originally used by Boyle (" sub- 

 stantial emanations from the celestial bodies ") was re- 

 suscitated by Rutherford to designate definite substances of 

 a gaseous nature continuously produced from other sub- 

 stances. The term was also used by Russell (" emanation 

 from hydrogen peroxide ") in much the same sense. If 

 the adjective " radio-active " be added, the phenomenon of 

 Rutherford is distinguished from the phenomena observed 

 by Russell. In this section we are dealing with the eman- 

 ation, or radio-active gas obtained from radium. Ruther- 

 ford and Soddy investigated the chemical nature of the 

 thorium emanation (Phil. Mag., 1902, p. 580) and of the 

 radium emanation (ibid., 1903, p. 457), and came to the 

 conclusion that these emanations are inert gases which 

 withstand the action of reagents in a manner hitherto 

 unobserved except with the inembers of the argon family. 

 This conclusion was arrived at because the emanations 

 from thorium and radium could be passed without alter- 

 ation over platinum and palladium black, chromate of lead, 

 zinc dust, and magnesium powder, all at a red-heat. 



We have since found that the radium emanation with- 

 stands prolonged sparking with oxygen over alkali, and 

 also, during several hours, the action of a heated mixture 

 of magnesium powder and lime. The discharging power 

 was maintained unaltered after this treatment, and inas- 

 much as a considerable amount of radium was employed 

 it was possible to use the self-luminosity of the gas as an 

 optical demonstration of its persistence. 



In an experiment in which the emanation mixed with 

 oxygen had been sparked for several hours over alkali, a 

 minute fraction of the total mixture was found to discharge 

 an electroscope almost instantly. From the main quantity 



1 By Sir William Ramsay, K.C.B., F.R.S., and Mr. Frederick Soddy. 

 Received at the Royal Society July 28. 



