I go REPORTS ON INVESTIGATIONS AND PROJECTS. 



initial velocities were sufficient to carry the materials, if undisturbed, 

 to the distance unity. The matter ejected toward S' was left moving 

 in an ellipse whose eccentricity was 0.98 and whose perihelion distance 

 was about equal to the radius of the sun. That which was ejected 

 in the direction opposite to S' was left moving in an hyperbola whose 

 eccentricity was about 1.5. 



The next case was the same, except that the material was ejected 

 toward S' with a velocity sufficient to make it recede, if undisturbed, 

 only to the distance 0.5. In this case the material was captured 

 by S', and when it had receded so that the disturbance of S became 

 small its eccentricity with respect to S' was 0.64 and its periastron 

 distance 26,000,000 miles. 



The next case treated was different, in that the matter was sup- 

 posed to be ejected when S' was at its perihelion at a distance 5 from 

 S. The velocities of ejection were such that the material would have 

 gone out to the distance 5 if it had not been disturbed. That which 

 started out toward S' was captured. When this fact was assured it 

 was not followed farther, and the eccentricity of its orbit was not 

 computed. The matter which was ejected in the direction opposite 

 toS' was left moving in an ellipse whose eccentricity was 0.53, whose 

 major semi-axis was 4.5, and whose perihelion distance was 2.09. 

 The orbits are illustrated in fig. 4. 



Up to the present orbits have not been found having quite so small 

 eccentricities as we should expect those bodies had which have united 

 to form our planets. But the work has gone so far that it has shown 

 us that they almost certainly exist. The very variety of the possi- 

 bilities disclosed will assist us, if this theory of the origin of our system 

 is true, in finding more exactly the antecedents of the planets. The 

 work at present seems to me to be very promising of fruitful results. 



It does not seem advisable now to enter into the details of explain- 

 ing the methods of computing we have used. They have varied with 

 the character of the motion and the magnitude of the disturbing 

 forces. Experience is suggesting modifications, and there is much 

 yet to be hoped for in the line of brevity. We shall also have to 

 examine more carefully the effects of the approximations made than 

 we have done so far. 



