200 CARNEGIE INSTITUTION OP WASHINGTON. 



a center-to-center encounter, a radial dispersion of matter trans- 

 verse to the line of collision would probably follow, returning from 

 which the material would again collide and, after a series of oscil- 

 lations, would gradually settle down into a pulsating gaseous mass.* 

 Here again the system would become gaseous at the outset, and 

 probably develop nothing of the typical meteoritic kind, except 

 possibly such sporadic elements as might be projected beyond the 

 control of the system. If the collision were eccentric, a rotatory 

 motion would doubtless be superposed upon the radial motion, and 

 the case would fall under either the gaseous or the orbital system 

 or under a combination of the two. 



In the line of my own suggestion f that stellar bodies passing close 

 by one another, but not colliding, may suffer disruption through their 

 differential attractions on one another, aided by internal elasticity, 

 on the principles developed by Roche, Maxwell, and others, I have 

 been unable to find anj^ plausible grounds for postulating a conversion 

 into a meteoritic nebula of the collisional type. 



In the case of such a disruption, the scattered constituents must 

 apparently be given a rotatory movement in a common direction 

 and in the orbital plane of the two bodies initiating it. The 

 dj'namics of the system are, therefore, from the outset, definitely of 

 a rotatory or revolutionar5^kind, and the case falls under the orbital 

 or planetesimal system rather than under the meteoritic system. 



It appears, therefore, that neither explosion, nor collision, nor 

 tidal disruption is likely to give rise to a distinctively meteoritic 

 swarm of the kind defined, and I have been unable to discover any 

 other source that can be assigned on definite grounds with a work- 

 able probability. Individual meteorites and rotatory and revolu- 

 tionary assemblages of dispersed elements, as well as true gaseous 

 nebulae, may be supposed to arise from the catastrophes named, but 

 apparently these catastrophes are not appropriate agencies for pro- 

 ducing fragmental swarms of the distinctively meteoritic type. 



I have made some study of meteorites to see if their characters 

 have any decisive bearings on the mode of their origin. 



Among the distinctive and significant characters of meteorites are 

 their fragmentary forms, their brecciated structures in part, their 

 occasional slickensided surfaces, their veins, the glassy nature of a 



*A case of this kind is described by Kelvin, Popular Lectures and Addresses, 

 I, p. 413. 



t On the Possible Function of Disruptive Approach in the Formation of Me- 

 teorites, Comets, and Nebulrp. A.strophvs. Jour., Vol. XIV, 1900. pp. 17-40. 



