METEORS OP AUGUST AND NOVEMBER. 97 



pode of the observer's direction are not compensated as on ordinary nights, but 

 there remains an outstanding uncompensated tendency towards a point more 

 or less deflected from this antipode, indicating the prevalence of a true mean 

 velocity, of greater or less magnitude, in a perpendicular to the observer's di- 

 rection. The meteors which do not appear to tend towards this extraordinary 

 convergent point are termed sporadic, and in such cases the term includes both 

 classes of sporadic and convergent meteors of ordinary nights. It seems to me, 

 in conformity with the opinion of Olbers, that the appearance of an extraordi- 

 nary convergent point is a better and surer criterion of an extraordinary display 

 of meteors than a moderate excess of numbers above the mean. Let us suppose 

 that an observer on the earth's surface, in his annual and rotary motion, falls 

 in with a portion of planetary space interspersed with these small bodies, sepa- 

 rated by intervals of not many hundred miles, yet so far separated as to exhibit 

 no tendency towards fixedness of relative position, like the particles of the sin- 

 gle bodies themselves, or like the solids, liquids, or even gases, connected with 

 the earth, — their mutual perturbations, owing to the smallness of their masses, 

 not being much greater than those of the principal planets of the system. Let 

 us suppose, however, that besides proximity in position in the system, the 

 separate bodies have also common elements of motion, the discrepancies in 

 the latter being of the same order as those of position relatively to the whole 

 system, whether from common circumstances connected with their origin, or 

 primitive projectile force. Then we should have all the phenomena of an 

 extraordinary display of meteors, namely, unusual numbers, and unusual po- 

 sition of the convergent point of their relative paths. This position could 

 readily be determined by the method already stated ; and the mean relative 

 direction in space of this flock of bodies could be ascertained by a single 

 observer, at one station, with a precision which, but for their common rela- 

 tions, no corresponding observations at any variety of stations could afford. 

 Then the only remaining element for the complete determination of the 

 mean elliptic elements of this small system, or cluster of asteroids, would be 

 the relative velocity of one of them, or their mean relative velocity, (the 

 single velocities being nearly common by hypothesis. ) In the absence of a 

 complete determination of this velocity, if it were presumable that this mean 

 value was analogous to that of Table I., then, since the mean error of the posi- 

 tion of the convergent point is very small, and may be neglected, a system of 

 elements derived from the adoption of the mean value in Table I. would have 



VIII. — z 



