Re my idea before that perhaps the Melosh eject-from-earth notions could be saved by the "swarming" idea that an ejected mountain, even if fragmented into pebbles, might still make it up thru Earth's atmosphere because atmosphere would not "notice" the fragmentation had occurred until too late... ...it occurs to me that this actually was tested by experiment. Specifically, the 30 June 1908 "Tunguska event" in a sparsely populated part of Siberia. http://en.wikipedia.org/wiki/Tunguska_event This was a large explosion estimated 5-to-30 megatons based on the damage (huge amount of flattened forest, charring, etc). This event must have been caused by a meteor weighing order 1 million tons. This would have been far more than enough t make it to ground level essentially un-slowed-down by air drag, so almost all the energy should have been dissipated on impact, not in air. However, there was no or essentially no blast crater, i.e. there was no ground impact. It was a multimegaton "air burst" at 5-10km up. There was a lot of dust but no larger ground hitting object has been found, although these is some unconfirmed speculation a solid mass of a few tons (tiny compared to original) did hit and now is buried below a lake/mud. The only explanation is that this meteor was made of weak material which exploded and fragmented in mid-air, and the fragments were small enough (enough surface to volume ratio, interiors thus uninsulated) that they could be totally vaporized by the air. Point is, IF there were a "swarming" effect where the atmosphere "did not notice" the meteor had fragmented and the drag was therefore comparable to if unfragmented, then it still would have hit the ground in a huge slam. It did not. I conclude that at least in the particular circumstances of this case, the swarming hypothesis is disproven by test. This also proves that at least for the particular composition of this object, materials strengths were weak enough to cause a massive fragmentation even though the stresses were presumably a lot smaller than would have been suffered in a Melosh impact-blast-ejection scenario. Obviously, such fragmentation does not always happen, since we know that huge meteors made of strong stuff can and have made it down to the Earth's surface in one piece. But it can happen, and did in the only known case in recorded human history. This --- (1) swarming theory refuted in only experiment, (2) large unfragmented theory refuted in only experiment --- makes it more difficult for the Meloshian "eject life in rocks" theory to work.