237 



KDooiledge & Seientlfle jlems 



A MONTHLY JOURNAL OF SCIENCE. 



Conducted by MAJOR B. BADEN-POWELL, F.R.A.S., and E. S. GREW, M.A. 



Vol. II. No. ii. 



[new series.] 



OCTOBER, 1905. 



[stJSne's' Hall ] SIXPENCE NET. 



CONTENTS.— See Page VII. 



The Two New 

 SaLtellites of Jupiter. 



By A. C. D. Crommelin. 



It has been my privilege to chronicle in these columns 

 two very sensational astronomical discoveries in the 

 last seven years. The first was the minor planet, Eros, 

 which proved to be our closest planetary neighbour; 

 the second was Phoebe, Saturn's ninth satellite, notable 

 for its immense distance from its primary, and still 

 more for its retrograde motion. The zeal and skill of 

 American astronomers has been rewarded with three 

 more discoveries in the satellite world during the last 

 few months, all of which present some points of special 

 interest. 



Till 13 years ago it was entirely unsuspected that 

 any further mysteries lay hidden in the Jovian family. 

 The four Galilean satellites had been known for nearly 

 three centuries, and formed a symmetrical system of 

 worlds, comparable with our moon in size, revolving 

 in almost circular orbits near the plane of their 

 primary's equator. The discovery of a fifth member 

 of the family by Prof. Barnard in 1892 excited great 

 interest; this was a very minute world, but resembled 

 the others in the shape and plane of its orbit. Its chief 

 mathematical interest lay in the rapid motion of the 

 perijove produced by its proximity to Jupiter's equa- 

 torial protuberance. 



It was doubtless the discovery of Phoebe that 

 suggested the search for very distant satellites of 

 Jupiter, which Prof. Perrine undertook last winter with 

 the Crossley reflector, and which proved successful be- 

 yond expectation, resulting in the discovery of two 

 more tiny members of the system. It must be con- 

 fessed that the Lick observers were somewhat tardy in 

 distributing information on the subject to Europe, so 

 that we were for a time in uncertainty as to whether 

 the new worlds were really satellites, and not minor 

 planets, which happened to be hovering in Jupiter's 

 vicinity. However, there is now no doubt at all that 

 VI. is a true satellite, and scarcely any doubt in the 

 case of VII. 



Dr. Frank E. Ross has deduced approximate ele- 

 ments of their orbits from the observations extending 

 up to March last. As satellite VI. has been again ob- 

 served at Mount Hamilton at the end of July I have 

 used the new observations to correct his elements of 



this satellite, but as no recent observation of VII. has 

 been reported, his elements are given unchanged. 



Prof. Perrine had previously announced that the 

 motion of VII. was probably retrograde. Dr. Ross, 

 however, finds that direct motion is much more 

 probable, though the matter is not absolutely certain 

 till the satellite is re-observed or till some images of it 

 are found on Harvard photographs of the neighbour- 

 hood of Jupiter taken some years ago. Several of 

 these photographs are available for the search. * 



The most extraordinary features of these orbits are 

 their high inclinations to both the equator and orbit of 

 their primary and to each other. The satellites of 

 Mars, Jupiter {5 inner), Saturn (7 inner), and Uranus 

 (probably) move almost exactly in the equatorial plane 

 of their primary, while our moon, Japetus, and Phoebe 

 deviate from this towards the orloit plane of the 

 primary. Neptune's satellite, indeed, appears to be 

 inclined at a considerable angle to both planes, but a 

 repetition of this anomalous feature in the hitherto 

 symmetrical Jovian system was quite unexpected. The 

 near approach to equality in the mean distances is also 

 curious, and the fact that the two orbits interlock, like 

 two links in a chain; in this respect they recall the orbits 

 of Mars and Eros. Their great distance from Jupiter 

 compared with the other satellites is also remarkable, 

 and suggests that they were not original members of 

 the system but have been added later. The capture 

 hypothesis is attractive, but there are grave mathemati- 

 cal difficulties to be overcome before it can be adopted. 

 It would seem that a planet cannot capture a body in 

 such a way as to make it travel in a closed path round 

 itself, but only round some other body, e.g., the vari- 

 ous members of Jupiter's comet family have been com- 



•Observations of VII, in August have now been reported. They 

 confirm the direct orbital motion ; but appear to show that the 

 orbit is considerably more eccentric, and the period shorter, than 

 the values given by Ross. 



