KELLOGG! (continued) scavenging by dust, ond as * tentative explanetion I think ve can invoke the fact that we have very strong internal cirgulation seff up imeiiately, and at the mories this afternoon it vill desnstrate this think, and the strong internal circulationalééds causes the dust colum td move into the firedall in this way, and the air gets in first. ‘The redidactive material almcct immediately epreats itself out into the doughnut padt of the cloud, and later the dust vhich contimues to pour into the gushrods merely circles sround the Goughaut, and never reaches the redionctive cord, which is now here and here, and this, os I say, is a tentative explanatfon, enl we simply felt that we verentt happy about observations until we hed isome idea of how this occurs. Jicw if we cen heve the lights.... We're going to move on to what ve've done about raincdt, and raincus is a more difficult thing to talk about, because as I've mentioned, we don't have any well documented cases of raincut, where we can m a hypothesis and cay, vell, 1t heppened in such and such a way on such nd such a date, that these vere the conditions, therefore out hypothesis s out of not as the case my de. %2 be borne We bave almost no information on raipout except ecme inforantion fron the Harverd people, Dr, Bell,| DaTRAD, who have mide careful observations of the resdicactivity however the thing that's lecking there to some extent1 i rain, and area) aixen detailed informtion about the structure of the rain storafs tt dows. So we've had to rely on a theoretical approach ig b droweet ta what ome would oxpect from rain. Firot, and I think the most cbvious approach, would bd to say vint happene vben @ falling drop falls through « eloul of pilcdctive particles. We appracched this by looking at the Langsuir theory for Sllections by a Departmont -« ~ OM gtr 3/