BioKEETON.- — On the Origin of Double Stars. l6l 



motion in space, and may, in extreme cases, prevent their escaping the mass 

 at alh It is this increased attraction after impact that is the force which 

 causes the two bodies to become associated. 



The following three lines of reasoning show how insignificant is the 

 work of sheering compared to the available energy. It is known that a 

 cannon ball, with a velocity of less than 2,000 feet per second, is capable 

 of penetrating a plate of h'on its own thickness. Now the velocity at 

 impact of two suns is at least 200 miles per second, or an energy in equal 

 mass of over 200,000 times as great as the cannon ball, but as the size 

 increases the ratio of volume to section increases also ; in fact, it is in the 

 ratio of the diameter, for the section varies as the square, and the volume as 

 the cube, of the diameter. So that were the density alike, its ratio would 

 be about a thousand million times as great in a sun as a cannon ball, or the 

 available energy is at least one hundi'ed million of million times greater 

 than required for sheering were both as hard as iron. 



It has already been suggested that sheering force has its limits in the 

 latent heat of fusion ; this is probably the amount of energy required to 

 separate the molecules from their fixed positions, but, in sheering, the 

 molecules require only to be separated in a single plane, — an insignificant 

 fraction of the whole in such a body as a cannon ball, and in a cosmical 

 body so insignificant a fraction as to be disregarded. 



But, even supposing it were required to separate all the molecules instead 

 of those of a single plane, it has been shown that the energy required to do 

 so is such an insignificant fraction of the whole as to be disregarded. In 

 dealing with impacts of bodies such as our sun, if the body be liquid or 

 gaseous of course there is no sheering force, and as in all of the collisions 

 of the bodies under discussion the energy is incomparably greater than that 

 necessary to volatihze the colliding parts, there is actually no sheering force 

 to prevent the escape of the other parts. 



From these several lines of reasoning it is evident that sheering force 

 may be absolutely disregarded, and that there is nothing in the impact 

 itself tending to destroy momentum in the non-colliding parts. The ratio 

 that is required to be cut off in order for the stars to become associated 

 depends largely upon the proper motion j)ossessed by the original bodies. 



If we take two such bodies as the sun as an illustration, their small 

 proper motion would allow them to become associated if as small a part as 

 one-thousandth were struck off each. In this case, however, the pah- would 

 move in orbits so highly eccentric as to almost, if not quite, graze at their 

 perihelion. If they struck off any ratio above this up to about one-half, it 

 is probable that they would still form binaries with increasingly circular 

 orbits, but this problem is so much influenced by the distortion of the 



21 



