Contributions to Astro-Meteorology . 43 



group will not have reached these orbits in its descent to the 

 sun. The whole extends along 267° of anomaly on the para- 

 bolic orbit. The breadth of the group at its centre, in the direction 

 of the radius vector, will be 96 metres, and in the direction per- 

 pendicular to the plane of the orbit 37 kilometres. Its density 

 at this place will be 400 millions of times the density of the 

 original group. 



If the primitive form of the group was other than spherical, 

 the same reasoning would apply and similar results follow. 



The apparent diameter of the assumed globe, as seen from the 

 sun, is about 0"*1. But if the original cloud be supposed to 

 have an apparent diameter of one minute, the resulting stream 

 would require 636 years to pass the perihelion. Some of the 

 nebulas have apparent diameters exceeding that of the sun. A 

 globe of such a size (19.24") would be transformed into a para- 

 bolic stream which would require more than 20,000 years to 

 pass the perihelion. The cross section of this stream would be 

 much greater than in the preceding case, but yet not so great 

 but that the earth might cross it in a few hours, or at most in one 

 or two days. We may, then, without extravagant hypotheses 

 account for the existence of meteoric currents which have been 

 observed for hundreds or thousands of years, like that of August. 



For facility of reasoning, the original position of the supposed 

 spherical group was placed at the aphelion of a long ellipse. 

 But results similar in kind, and comparable in degree, would re- 

 sult had the group been supposed at any point of a conic section 

 of very great length. 



If the orbit were an ellipse, the original form of the cloud would 

 never be regained. At each perihelion passage the length of the 

 stream would be increased until it formed a closed circuit. The 

 stream would be at first periodic, bat finally constant. If the 

 orbit were open, there would be a single passage only. 



The supposed cloud was deemed to exert no sensible attrac- 

 tion upon its particles. To justify such an assumption, M. 

 Schiaparelli makes an estimate of the density of the August 

 ring. For the mean distance from each other of the bodies of 

 the August group, he obtains by computation about 100 (geo- 

 graphic) miles. This is obtained by quite arbitrary supposi- 

 tions, but the result is very nearly that which is obtained by 

 computations from the best data in our possession (see Silliman's 

 Journal, vol. xxxix. p. 207, combined with observations at 

 Hartford in 1863, when six observers counted 153 different 

 meteors in a half hour *) . 



* The corresponding mean distance from each other of the members of 

 the November group, where we crossed it last year, is 30 or 40 miles. 



