378 ANNUAL OF SCIENTIFIC DISCOVERY. 



in diameter instead of one eighteenth of an inch. He thought the Weston 

 meteorite was several feet m diameter. 



Professor Gould thought that Mr. Jones's observations threw more light on 

 the nature of the zodiacal light than all before him. He added that if, in- 

 stead of a ring, the zodiacal light were but an extension of the earth's atmos- 

 phere, it might not need planets to sustain it. 



Professor Hackley thought that if the extent of the zodiacal light varied, 

 the fact of their being 35 apart would throw no light on the question of the 

 size. 



OX CERTAIN ANOMALIES PRESENTED BY THE BINARY STAR, 

 70 OPHIUCHI, BY CAPTAIN W. S. JACOB. 



This pair has been long known to astronomers as a binary system, but the 

 exact orbit is yet in doubt, although nearly a whole revolution has been com- 

 pleted since it was first observed by Sir W. Herschel in 1779. All the orbits 

 that have been computed fail at certain points in representing the observed 

 positions, and those which best represent the angles fail entirely as regards the 

 distances. The most remarkable point is, that even in those orbits which 

 agree best with observation, the errors in the angles assume a periodical 

 form, retaining the same sign through a considerable space of time. An 

 orbit has been computed with a period of 93 years, in which the errors are-j- 

 from 1820 to 1823 with one exception from 1823 to 1830, doubtful in 1830 

 to 1832. and from 1833 to 1842 all-)-, after which they continue for the most 

 part . This sort of error must depend upon some law ; it might arise from 

 a change in the law of gravitation, but may be accounted for more simply by 

 supposing the existence of a third cpaque body perturbing the other two. 

 Such bodies have already been suggested to account for irregular motions 

 of apparently single stars, such as Sirius and Procyon. The body in this case, 

 if supposed to circulate as a planet round the smaller star, need not be very 

 large, as the deviation from the ellipse does not exceed about 0"'l. As- 

 suming the small star to describe a secondary ellipse in which a 0"'08, 

 e=rO"'15and w 200, and applying corresponding corrections to the com- 

 puted positions, the average errors in the angles is reduced from 50' to 37', 

 and in the distances measured subsequent to 1837, from 0"'14 to 0"'ll. or by 

 about \ ; while the maximum errors are also reduced in about the same pro- 

 portion. There is, therefore, prima facie evidence for the existence of such a 

 body, and it is desirable that the fact should be still further tested by careful 

 observation. The subject possesses an additional interest at the present 

 time, in reference to the opinion brought forward in the Essay " Of the Plu- 

 rality of Worlds," on the impossibility of double stars having attendant planets. 

 Proc. British Association. 



ON THE VARIABLE STARS CANCRI AND ALGOL. 



M. Argelander has recently been devoting attention to the star Cancri, 

 which has been found to undergo periodical fluctuations of brightness analo- 



