566 



KNOWLEDGE & SCIENTIFIC NEWS. 



[October, 1906. 



wire projecting- from the prime conductor, so that its 

 point is between the two discs on the rij;^ht hand side. 

 Attaching a small Leyden jar to the prime conductor a 

 very strong- spark, well suited for X-ray work, is ob- 

 tainable. The spindle may be made long enough to 

 carry further pairs of plates in the same way, with in- 

 creased output. To regulate the pressure of the 

 rubbers, they should be on hinged supports so as to be 

 quite free to yield to outward pressure, while a curved 

 springv piece of thin brass wire is made to clip lightly 

 the two supports and thus hold the rubbers pressing 

 "■cntlv asjainst the class. 



ASTRONOMICAL. 



By Charles P. Butler, A.K.C.Sc. (Lond.1, F.R.P.S. 



Mills Expedition to the Southern 

 Hemisphere. 



\r the close of the two-year period of actual observation 

 in October, 1905, by the D. O. Mills Expedition to Santiago, 

 Chili, Mr. Mills most generously offered to continue the 

 work for five additional years. He has also provided for 

 extensive improvements in, and additions to, the equipment 

 of the Observatory. rwo-prism and one-prism spectro- 

 graphs have been built, in order that radial velocity deter- 

 minations may be carried to the fainter stars. Professor 

 Wright's experience has made it practically certain that 

 rapid changes in focal length and other sources of dis- 

 turbance in the stellar images are due to rapid changes of 

 the temperature of the mirror during the first hours of the 

 night. Dr. Curtis has designed a refrigerating apparatus, 

 which has been sent to Cerro San Cristobal for trial. It is 

 hoped that by the use of this the apparatus may be kept at a 

 sufficiently constant temperature. 



Accelerated Motion of Jupiter's Great 

 Red Spot. 



Mr. VV. F. Denning, writing to the Observatory for 

 September describes some recent observations he has made 

 on the Great Red Spot on Jupiter during the present year. 

 Tovvards the close of the last opposition, March to May, 

 1906, his measures showed that the rate was as nearly as 

 possible conformable with that of System II., based on a 

 rotation period of 9h. 55m. 40.6s. On re-observing the 

 planet before sunrise on .Vugust 9, however, he immediately 

 saw that the Red Spot and Hollow were far in advance 

 of their predicted places. This was confirmed by the Rev. 

 T. E. R. Phillips on the morning of .August 11, and by 

 another observation obtained by Mr. Denning on August 

 16, showing that in a period of little more than three 

 months the markings have lost 16°, equivalent to 26™ 29^, 

 relatively to the zeromeridian of System II. In other 

 words, while the rotation between March 24 and May 4 

 was gh. 55m. 40.6s., it had become only gh. 55m. 33. Ss. 

 between May 4 and August 8. The latter period is the 

 same as that exhibited by the Red .Spot during 1879. 



Mr. Denning notes that the dark material (extending 

 over about 63" in longitude), forming the .South tropical 

 disturbance on Jupiter, must have been in central conjunc- 

 tion with the Red Spot in June last, and he thinks that it 

 may have been responsible for the marked increase of 

 velocity shown by the latter. Even then the recent accelera- 

 tion in the niotion of the Red Spot is far greater than has 

 ever been jireviously observed, and either the Hollow in the 

 South Equatorial belt or the Red Spot appear to have been 

 visible for the past 75 years. 



Recent Observations of Phoebe. 



In the iiSth Circular of the Harvard College Observatory, 

 Professor E. C. Pickering gi\-es the reductions obtained 

 from six photographs of Saturn's satellite, Phoebe, during 

 this year. They were taken duriii;; May .-uul June, with 

 the 24-inch Bruce telescope, at Arequipa, the plates being 

 exposed for about two hours each. The following table 

 shows the co-ordinates of the satellite at the times noted : — 



The System of Castor. 



Mr. II. I). Curtis gives a complete discussion of the Lick 

 Obs(r\al(iiy plates of the spectrum of a Geminorum, 

 Castor, with correlative comparisons w'ith the measure- 

 ments made at Pulkova, by Belopolsky. Most of the plates 

 were taken with the re-mounted Mills spectrograph, with 

 ^4,5oo central, the region of spectrum investigated ex- 

 tending from A4367.839 to A4629.S21. Both components 

 of the star are of the Sirian type, being cataloijued in the 

 flarvard classification as A and VHIa. The absorption is 

 somewhat more complete in a,, the fainter component, than 

 in O.J. The plates for measurement were exposed for 

 eighteen minutes to the brighter component, and forty-four 

 minutes to the fainter component. 



a I Geminorum : fainter component, south preceding; 



magnitude, ;1.7. — The binary character of this component 



of Castor was discovered by Belopolsky at Pulkova, in 



1896. The elements deduced from the Lick plates are : — 



Period — 2-928285 days. 



T = J.D 2416828-057 + 0-042 days. 



c — o'oi + 0-0066 



m" = i22"-939 



u = 102^-516 + 5^-120 



K = 31 76 + 0-22 



Velocity = — 098 km. + 0-15 km. 



a Sin i = 1,279,000 km. 

 ai Geminorum, brighter component: north following; 

 magnitude, J.7. — The v'ariation in the radial velocity of this 

 component of Castor was discovered by Curtis in October, 

 U)04. The final elements of the orbit arc : — 

 Period = 9-218826 days. 



T = J. D. 2416746-385 + 0-02I days, 

 f = 0-5033 + 0-0112 



M-'" = 39" 05053 + 0-00046 



w = 26fi5i ± 1-730 



K = 13-557 + 0218 



Velocity = +6 20 km. + 0-17 km. 



a Sin i = 1,485,000 km. 

 The author expresses the hope that a combination of the 

 spectrographic and visual results will eventually give a fairly 

 accurate value of the parallax, masses, and other physical 

 constants of this unique quadruple system ; at present the 

 elements of the visual orbit are so indeterminate that no 

 conclusion can be reached at present. It is important to 

 note that in Castor we have two systems whose orbital di- 

 mensions are probably of the same order of magnitude. The 

 brighter component has, however, the very great eccentricity 

 of 0.50, while the fainter pair revolve in orbits which are 

 practically circles (f=o-oi). This extraordinary difference 

 seems, by the generally accepted theories of stellar evolution, 

 to indicate that the brighter component is the older, and 

 that the fainter is, comparatively speaking, a binary of 

 relatively recent origin. The effect of tidal friction on a sys- 

 tem so eccentric as that of oi must be enormous, as at 

 periastron the stars are only about one third their apastron 

 distance apart. -Such an eccentricity has only been found 

 in those spectroscopic binaries which show variability in 

 their light. A special series of observations were accord- 

 ingly made with the smaller of the two Bruce double- 

 image photometers, attached to the twelve-inch equatorial. 

 No marked variation of regular period could be detected, 

 so that any existing must be very small. No irregularities 



