382 FKOM NEBULA TO NEBULA 



from the sun, not from the earth; and she should have 

 been as diminutive as Eros, besides. 



THE BOLIDE THEORY. By way of an alternative hy- 

 pothesis, Professor Grove K. Gilbert, whilom president 

 of the Philosophical Society of Washington, in an ad- 

 dress before that body in 1892, expounded the notion that 

 the earth may primevally have possessed a ring like that 

 of Saturn, and that the lunar asperities were produced by 

 its collapse. The reader will find an excellent presenta- 

 tion of this view in Popular Astronomy, March, 1917, by 

 Donald Putnam Beard, from which I quote : 



ORIGIN. If we conceive the moon as an edifice which had 

 its foundation in a ring or shoal of meteors encompassing the 

 primeval earth, and similar to the giant planet Saturn (the 

 meteoric constitution of whose rings was spectroscopically dem- 

 onstrated by Keeler in 1895), and if we imagine this shoal grav- 

 itating together and building up our satellite by accretion, no vio- 

 lence is done the essential principles of Laplace's immortal Neb- 

 ular Hypothesis. Meteors replace molecules, that is all, as long 

 ago pointed out by the late C. A. Young. The mechanical be- 

 haviour of a meteor swarm containing individual masses and en- 

 dowed with the ordinary velocities of meteors would be precisely 

 similar to a nebulous mass of continuous gas. 



The mathematical analysis of the mechanical conception of 

 a Saturnian ring is not in place in a discussion of this nature, but 

 by imparting to the postulated meteors in the swarm orbits not 

 widely variant from that of the moon's, and in a similar direction, 

 their initial velocities at impact were small as compared with those 

 created by the moon alone. Since the course of these moonlets 

 were parts of curved orbits with the moon at their focus, they 

 cannot justly be considered as straight lines. By restricting these 

 meteors to a thin plane ring, and assuming a fairly equable dis- 

 tribution through this plane, the distribution of impact angles de- 

 duced by Gilbert yields a curve in which 58 per cent deviate from 

 the vertical less than 20 ; 70 per cent less than 30, while 80 

 per cent fall within 40 from the true vertical. To the vertical 

 infalls consequent upon this condition is due the prevalent cir- 

 cularity of the craters and obviates a resort to R. A. Proctor's 

 improbable suggestion of an elastic return to circularity. 



Laboratory experiments with a lead disk 5.5 inches in dia- 

 meter and about 0.5 inches thick as a target, into which .22 cali- 

 ber bullets of the same material were fired, demonstrate experi- 

 mentally the effects produced by the impacting moonlets upon the 



