A CATALOGUE OF OBSERVATIONS OF LUMINOUS METEORS. 



413 



1'' m' , v" m" the apparent directions of two meteors directed from the radiant 

 E. Then the angle a' is evidently greater than a". 



" As far as I have heen able to examine the question, the arc subtended by 

 the difference of the angles a', a" may be more or less exactly measured by 

 comparing together the lengths of the major and minor axes of the radiant 

 region in those cases where it appears to have an elongated form. Projected 

 and measured upon a map of the stars, this arc occasionally amounts to 10°, 

 or 15°, independently of o° allowed for errors of observation, and for other 

 sources of inaccuracy." 



By the " arc subtended by the difference of the angles a', «"," the difference 

 from parallelism between the lines r' m, r" m" in the above paper is per- 

 haps intended to be signified; and this may amount occasionally to 15°, 

 It is plain, however, that the real difficulty connected with the long endu- 

 rance of particiilar radiant-points in a nearly fixed position is, to explain 

 why the elongation of the radiant region, or the difference from parallelism 

 actually observed, does not, in general, amount to a quantity nearly as large 

 as the difference between the angles «', a" . 



Thus, supposing A ff, B 6 to be the inner and outer limits of a current of 

 meteoric bodies moving in parabolic orbits round the sun, S, in the same 

 plane with the earth's orbit, a B, and 

 having the common axis B S s ; S a, 

 ■ perpendicular to B S, the earth's dis- 

 tance from the sun when it encounters 

 the inner limit of the stream, three 

 months before the time when it arrives 

 at B. The absolute velocity of the nie- 

 • teors where they encounter the earth is 

 everyn^here represented by the cUagonai 

 of a square, as a v, whose side a V re- 

 presents the velocity of the earth in its 

 nearly circular orbit. Now as the di- 

 rections of the circle and parabola at a 

 are, respectivelj^, in the side a V and 

 diagonal a v of the same square, tlie re- 

 lative velocity of the meteors at ri, with 

 respect to the earth, is in the direction 

 V V, parallel to rr S ; and this is also the 

 direction of the relative velocity, with 

 respect to tlie earth, of the meteors vriiich overtake the earth at B. The 

 radiant-point of the meteors at a, B would, in such a case, therefore have a 

 nearly invariable direction, or apparent position in the skj', E, E, ; and in 

 the intermediate interval of three months, during which the shower con- 

 tinues, it would only undergo very small changes of its place. 



