» RODN re a et : : ~ ase ct. Y a . wee west dia Liasyg wasecca Lily phere as it has to be by the alternative theory, the movement of its several parts some’ hours later would hawe been quite different from what they were. As far as my information goes, there is little doubt that these derivatives moved with the winds in the leve: between about 35,000 feet and 60,000 feet. On the other hand, as the sunset observation: show, the plume did penetrate the stratosphere, and moved with, and was deformed by, the winds appropriate to that region; it may well have reached 134,000 feet. Now let us consider the alternative hypothesis based. on unspecified measurements: that the base of the mushroom cloud was at 50,000 feet and the top, apart from the plume, was at 120,000. Of course, I do not kmow the nature of the measurements, but some of | them, at least, were probably of the type usually made on such occasions, the interpretation of which has in the past led to considerable error in the estimate of bomb cloud heights. Angular measurements alone, whether with theodolite or sextant on the ground or by bubble sextant from aircraft, are subject to the same erroneous interpretation. If the clowl, on reaching some limiting height, advances not vertically but laterally toward the observer, there will be a very rapid increase in vertical angle, which if interpretec as vertical movement, will lead to gross over estimates in the height of the top. This error was made with GEORGE shot during Operation GREENHOUSE. On the basis of the mis— interpretation the height of that cloud was announced as over 85,000 feet, whereas the true height as measured from photographs long afterwards was considerably less than this. Owing to the enormous lateral extent of the MIKE cloud the chances of this kind of misinterpretation have been greatly enhanced. For example, if I had made this interpretatic of our own angle measurements, crude as they were, I would have had to conclude that with in a very few mimites after the detonation, the mushroom cloud had reached to an infinite distance from the surface of the earth. Even radar measurements are not free from this objection, since it is difficult to be sure what part of the cloud is being measured: is it the active plume or is it the edge of the mushroom cloud? Is it perhaps a knot of reflecting material buried somewhere in the cloud? I cannot answer these questions, since the original data are not available to me, but I suggest, in view of the importance of the topic, that they be asked. The only reliable way of directly measuring the cloud height is to triengulate all parts of the explosion product by means of a network of special cameras. The difficulty with MIKE was that ho one seems to have anticipated the great size of the final cloud or the fact that it would set up so much secondary natural cloud and precipitation that complete triangulation would be difficult - it is hard to recognize the same perts of the cloud different photographs, This apart from the fact that very special wideangle lenses ought to be used on all cameras, and thet this, epparently, was not done. I venture now to suggest certain theoretical reasons why the mushroom cloud lay in the troposphere and not in the stratosphere, chiefly because I think these may help in Bolving your operational problem. It seems to have been generally assumed that the MIKE cloud differed in no essential way from that formed by conventional shots, except, of course that it was much larger. My own hypothesis on this is, that MIKE was so large that qualitatively different geophysical effects were produced and that it is erroneous to extrapolate from the Nevada shots, say, or even from previous tests in the Marshall Islands, A conventional air-drop, for examphe, results in the formation of a fireball, which, after being transformed into a ring-vortex, ascends to the tropopause. The vortex ring or mushroom head consists of original bomb material plus water, etc., entrained into it by turbulent mixing; the stem is largely secondary although it too contains radio