TBANSACTIONS OF SECTION A. 489 



take those approximating to the line of flight. Those hetween G0° and 120° are 

 30 + 26 = 55, and those between 240° and 300° 29 + 30 = 59 ; for both 114, or more 

 than double those of the flank divisions of equal extent. Had I been able to 

 obtain much smaller segments than 60 degrees, the diflerences as contrasts would 

 likely have been much more striking I make no further comment than to ask, is 

 it likely that our sun's primary lies to the dexter side ? referring here to the sides 

 of the plate. 



Table. 



5. Oti Cometic Nehulce. By Professor A. W. Rucker, M.A., F.B.S. 



Mr. Lockyer has suggested (' Proc. Roy. Soc' No. 26G, vol. xliv. p. 10, 1888) 

 that comet-like nebulie may be caused by the passage of a very dense swarm 

 through a sheet of meteorites, the relative velocity of the two being considerable. 

 The author has therefore attempted to calculate the increase in the number of 

 collisions which take place in the rear of an attracting mass which passes through 

 a swarm of meteorites so sparsely scattered through space that the main eflecta of 

 the attraction are produced in a distance which is small compared with the length 

 of the mean free path. 



Assuming with Clausius that the particles have equal velocities equally 

 distributed in all directions, which are small compared with the relative velocity of 

 approach, the collisions will be most numerous within a cone the vertex of which 

 is the attracting body or nucleus, and which contains the lines which are parallel 

 to the relative velocities of the individual meteorites and the nucleus when at an 

 infinite distance apart. 



Let o) be the number of collisions per unit of time and volume at a point the 

 length of the perpendicular from which on the central line of the cone is ^, 

 intersecting it at a distance s from the nucleus. Let the radius of the circle 

 in which tlie cone nuts the plane through s perpendicular to its central line be R. 

 Then if v be the velocity of agitation of the meteorites at infinity, and V the relative 



velocity of approach, R = : — . Let n be the number of meteorites per unit of 



volume at infinity, and /i the acceleration due to the nucleus at unit distance, and 

 let 8 be the diameter of a meteorite. Within the circle of radius R the quantity 

 a is constant, and 



<i)=v^2n--aV/x?si/R-V- 



without the circle but near to it ; 



2^/27r8^nV?s4 / 1 R r, R^ • .,Tf^\ 



{'-=-V'-i; 



+ sin"'— > , 



If ^ is large relatively to z, and if ^ = r sin 6,z = )- cos 6, and u = 1/r, 



7ry«VMV2" 



V»(l - cos (9)71 -cos ^ + '4rtM 



These formulaj are only applicable on the hypothesis that the chances of a second 

 collision may be neglected, and the author illustrates them by a case in which 



