382 Prufessor G. E. Darwin [Jan. 25, 



It was next necessary to find out how often the stones came into 

 collision, how far they travelled from one collision to the next, and 

 whether the collisions could be frequent enough to impart to the whole 

 nebula the gaseous property demanded by the nebular hypothesis. 



Even a microscopic animal in our atmosphere is not aware of the 

 individual impacts of molecules on his body, and his sensation is still 

 that of gaseous pressure. Bat it must clearly be a giant who would 

 not be aware of the individual blows of meteorites in a meteoric 

 nebula, but would only realise their average effects. 



It would not be easy to explain the exact reasoning by which it is 

 possible to determine how large the giant must be in order to act as 

 a judge of the gaseous property of the meteoric swarm, nor of how a 

 comparison of his dimensions with the texture of a meteoric swarm is 

 to be made, and it must suffice to say that the comparison is best 

 clothed in a form which may appear something quite different, but 

 which is really substantially the same. 



It may be stated, then, that a meteoric nebula would behave suffi- 

 ciently like a gas to allow the nebular hypothesis to be true, if the 

 average path of a meteorite between two collisions were only a short 

 portion of that curved orbit which it would describe under the action 

 of gravitation if it could move through the swarm without ever 

 colliding with another stone. 



These explanations led on to the numerical values derivable from 

 calculation, on the hypothesis that the solar nebula, consisting of 

 1 lb. iron stones, was distributed in a swarm extending half as far 

 again as the present distance of the planet Neptune from the Sun. 



It appeared, then, that at the middle of the swarm a meteorite 

 would, on the average, come into collision every 13 hours, and would 

 travel 140,000 miles between collisions ; at the distance of the small 

 planets called the asteroids, it would collide every 17 hours, and 

 would travel 190,000 miles between ; at the distance of Uranus the 

 collisions would be at intervals of 25 days, and the path 6,000,000 

 miles ; and lastly, at the distance of Neptune, the interval would be 

 190 days, and the path 28,000,000 miles. 



It may also be show^n that the path described between collisions 

 forms a larger portion of the whole curved orbit of a meteorite the 

 further we go from the middle of the swarm. Even at the distance 

 of the planet Neptune the collisions were, relatively speaking, so 

 frequent that, on the average, gravity only sufficed to draw the 

 meteorite aside from the straight path by l-66th of the length of path it 

 had traversed, before it was deflected into a fresh orbit by collision with 

 another stone. The fraction 1-6 6th was then the numerical value of 

 the criterion of the aiDplicability of quasi-gaseous properties to the 

 swarm, and this fraction is so small that it may be concluded that the 

 swarm passes the proposed test. 



It followed, therefore, that if meteorites possess a virtual elasticity, 

 a swarm of meteorites provides a gas-like medium of fine enough 

 structure to satisfy the demands of the nebular hypothesis. 



