254 ANNUAL REPORT SMITHSONIAN INSTITUTION, 19 60 



which was made by projecting pictures onto a white sphere and then 

 photographing the mare from a point directly above it. The lines in- 

 dicate the positions of the Bay of the Rainbow or Sinus Iridum, a cir- 

 cular area before it, and the entire mare. Radiating from the circular 

 area are many ridges, particularly toward the southwest: there are 

 also many grooves in tlie surface, as can be seen on good photographs. 

 Gilbert noted all these features, except for what seems to be an obvious 

 relationship of Sinus Iridum to the rest of the pattern. The most 

 reasonable explanation of all these details is that they are related and 

 were produced by one event. The whole pattern is unsymmetrical, 

 and hence one concludes that the colliding object arrived from the 

 northeast at a substantial angle from the vertical, plowed a deep hole 

 in the moon, and spread out some of its own substance and some of the 

 Imiar substance in a wide fan-shaped area and to great distances. The 

 object probably came in through what is now Sinus Iridum and either 

 produced this bay or destroyed one wall of a walled plain. Alterna- 

 tively, a second collision of substantial size occurred in precisely the 

 same region after the formation of Mare Imbrium. Two large colli- 

 sions at the same spot seem most improbable, and hence it is most 

 likely that one great collision produced the entire pattern. 



A few of the ridges and grooves extend through the Jura Mountains 

 to the northeast of Sinus Iridum. The principal pattern spreads out 

 in a semicircular fan from the edge of the Carpathian Mountains on 

 the east, to Plato to the west of Sinus Iridum. Momitainous ridges 

 which point toward the circular collision area are found in and be- 

 yond the center of the moon's disk. High-density objects plowed 

 through the walls of Ptolemaeus and Alphonsus and produced great 

 grooves in the surface: these objects can reasonably be assumed to 

 liave been metallic iron-nickel. In Oceanus Procellarum there are 

 many short ridges orientated in the general direction of the collision 

 area. It is not clear whether some of the individual ridges and 

 grooves belong to the system or not, but the overall pattern is entirely 

 convincing. 



The circumsurface velocity of the moon is 1.7 km. per second, and 

 since tlie objects fell at a point some 1,000 to 1,500 km. from the point 

 of collision, their velocity must have been of about this value. The 

 broad pattern indicates that the object Avas moving at a velocity loAver 

 than that of sound in tlie material of which it is composed — in this 

 case, some 5 to 7 km. per second. An object moving with relatively 

 high velocity — say some 30 km. i)er second — would bury itself before 

 the rear side received a signal that contact was m.ade and would ex- 

 plode to give a symmetrical pattern. But if we assume a velocity 

 less than that of sound — say some 2.4 km. per second, which is the 

 escape velocity of the moon — the unsymmetrical pattern can be under- 



