ACADEMY OF SCIENCES] SCIENTIFIC ACTIVITIES 187 



departed from the equator as widely as the few thousand last-formed craters do, the impacts of 

 those falling in high latitudes, north or south, would be so oblique that the resulting craters would 

 be plainly oval instead of nearly circular. A third difficulty is that the supposed influence of 

 moonlet impacts on the moon's rotation, by which the incidence of impacts would be made more 

 nearly vertical, would be greatly weakened if not wholly overcome in the late period when the 

 last few thousand impacts occurred, by the earth's action on the supposedly prolate moon above 

 referred to. 4 



But apart from these difficulties, there are certain supposedly "sculptural" features of lunar 

 scenery which do not appear to have been recognized by others than Gilbert, and which give 

 independent support to certain aspects of the moonlet theory. These are, on the one hand, the 

 grooves which furrow the rims of certain craters or score the tracts between the craters; and, 

 on the other hand, the ridges or oval hills which appear to have been added to the surface. 

 They have locally accordant directions, so that the grooves in crater rims are arranged like parts 

 of a system of parallel chords across the rim circle; but when platted on a chart of the moon, 

 the directions diverge from "a point near the middle of the plain called Mare Imbrium [north- 

 west of the center of the moon's visible hemisphere], although none of them enter that plain." 

 This sculptural system therefore has nothing to do with the "white streaks" which diverge in 

 radial systems from certain craters, especially from Tycho Brahe in the southern part of the 

 moon's, face, and which, following Wtirdemann, are regarded as great splashes or spatterings of 

 whitish matter caused by impacts. Furthermore, "associated with the sculpture lines is a 

 peculiar softening of the minute surface configuration, as though a layer of semiliquid matter 

 had been overspread, and .... obliterated the smaller craters and partially filled some of the 

 larger"; and such Gilbert believed to be the fact. Hence, he thought that — 



a collision of exceptional importance occurred in the Mare Imbrium, and that one of its results was the violent 

 dispersion in all directions of a deluge of material — solid, pasty, and liquid. Toward the southwest the deluge 

 reached nearly .... a distance of 900 or 1,000 miles . . . Northward and northeastward it probably extended 

 to the limb. Thus . . . were introduced the elements necessary to a broad classification of the lunar surface. 

 A part was buried by liquid matter whose congelation produced smooth plains [maria]. Another part was over- 

 run by a flood of solid and pasty matter which sculptured and disguised its former details. The remainder was 

 untouched, and probably represents the general condition of the surface previous to the Imbrian event. 



Gilbert's confidence in this catastrophe is further indicated by the association with it of 

 certain great furrows, " comparable in magnitude with the Grand Canyon of the Colorado," 

 and believed to result from a terrific plowing of the lunar surface by fragments of the Imbrian 

 moonlet, after they had described a trajectory of over 1,000 miles. His confidence is also 

 illustrated by a description of certain "rills," which he imagines to have been "yawning chasms 

 three-fourths of a mile wide at top and several miles in depth" before the Imbrian deluge; when 

 "the swift tide rushed over them a small portion may have been arrested and engulfed, but the 

 chasms were not filled till the torrent stopped. Then that which spanned them sank down, 

 coming to rest a short distance below the edges and so forming the visible floors." 



The coherence of these explanations is attractive, but in the present connection the most 

 remarkable things about them are, first, the amount of time that was found, during the most 

 distressing period which the Geological Survey ever passed through, for making late at night 

 the observations on which the explanations were based; and, next, the distance to which the 

 explanations that were framed to account for the observations led their cautious inventor into 

 an extremely speculative field. In so far as his ventures concerned the explanation of existing 

 features of the moon's surface, it may be believed that Gilbert was exceptionally well equipped 

 by his minute and appreciative study of existing features on the earth's surface; in this respect 

 he had a great advantage over astromomers ; and in so far as his ventures touched concepts of an 

 astronomical nature in connection with his theory of moonlets, he was exceptionally well pre- 

 pared by his unusual capacity for mathematical and physical discussion; in this respect he had 

 advantage over most if not all geologists. One who was well acquainted with this lunar excur- 



» It has lately been suggested that the circular form of lunar craters might be determined by the explosion of material vaporized by meteoric 

 impact, whatever the direction of a meteor's approach. 



20154°— 26— 19 



