CKOLOGY MOULTON. [87 



ried out the computations intrusted to him with accuracy and enthu- 

 siasm. The number of cases to be examined in order to get an 

 adequate understanding of the subject is very great, because of the 

 variety of circumstances which may modify the motion of the ejected 

 material. They are (i) the nearness of approach of the disturbing 

 sun, (2) the mass of the disturbing sun, (3) the eccentricity of the 

 orbit of the disturbing sun, (4) the longitude of the disturbing sun 

 from the perihelion at the time of the ejection of matter, and (5) the 

 velocity of ejection. In any combination of these circumstances 

 there are two cases, for the matter may be ejected either toward or 

 from the disturbing sun. 



If it is assumed that three values for each one of the factors which 

 modify the orbits of the ejected material will give a sufficiently accu- 

 rate idea of what may happen, it is found that there are.altogether 486 

 separate orbits to be computed. This gives some idea of the magni- 

 tude of the problem we have undertaken. For most of the factors 

 three values are by no means an excessive number. For example, it 

 will in all probability be necessary to consider the differences in the 

 orbits when the material is ejected some time before the disturbing 

 sun reaches perihelion, wmile it is at perihelion, and some time after it 

 reaches perihelion. However, it is not likely that it will be necessary 

 to examine the whole 486 orbits, for we shall presently have enough 

 data so that our conjectures will become more valuable and certain. 



So far we have computed 9 orbits. The results are of the most 

 varied and surprising character. In 4 of these 9 cases the ejected 

 materials were left moving in elliptic orbits, in 2 of them they were 

 sent away from both suns in hyperbolic orbits, in 2 of them they 

 w T ere captured by the disturbing sun, and in one of them the ejected 

 materials fell back into the sun after making a great loop. If we 

 add to this list the obvious case where the material is ejected with 

 a small velocity and falls back into the sun without being much per- 

 turbed, we shall have nearly every imaginable result. 



The velocities of ejection were in all cases moderate, never being 

 enough to carry the matter, if it were undisturbed, much beyond the 

 ortyt of Saturn. The fact that under these circumstances matter 

 can be driven beyond the gravitative controls of both suns is very 

 surprising, and, if I am not mistaken, an important result. It shows 

 the disintegrating effects of gravitational forces, whereas we have 

 heretofore regarded gravitation as uniformly tending toward con- 

 centration. Although we did not have this question of wide dis- 

 persion in mind in taking up these computations, it may be that in 



