solution will contribute substantially to knowledge of stratospheric circulation. Two obvious experimental approaches are direct investigation by high altitude sampling and the use of selected atmospheric tracers. Someof the fallout evidence having bearing on the question of high stratosphere holdup of nuclear debris from past tests are presented and discussed below. Cloud Height Considerations The development and rise of nuclear clouds and the influence of meteorological factors on their height 9 10 : tas . : and shape have been discussed by Machta and Kellogg. Approximate stabilization cloud heights for equatorial surface bursts of several yields, based on results of Kellogg, 1 ° are given in Table 1. “7: Time of stabili- Table 1 Approximate Stabilized Cloud Heights for Equatorial Surface Bursts Height, Center (ft) Vertical Extent (ft) Height, Top (ft) 10 MT 82, 000 46, 900 105, 000 1 MT 63, 000 28, 000 77,000 100 KT 40, 000 16, 000 48, 000 50 KT 34, 000 12, 000 40, 000 Yield zation apparently increases somewhat with yield, but is usually taken as 6 minutes, For megaton yield shots, horizontal growth continues to 10 minutes or longer at a rate too rapid to be due to natural wind shear. 10 The observational data for megaton burst clouds are inadequate to reliably assess the vertical extent and heights of cloud tops at times beyond the first few minutes. Subsequent vertical extent due to meteorological mixing on a time scale of weeks and months is unknown. Considerations about debris condensation, particle growth and cloud distribution have been discussed by the Rand Corporation! and by Lapple. 2 Lapple argues convincingly that the radioactive debris cannot be restricted to a discrete small portion of the cloud volume, but must be distributed generally throughout the cloud, However, in stratospheric clouds for high-yield surface shots, the upper portion of the cloud should exhibit higher activity concentration, a greater proportion of the very fine particles which persist in the stratosphere, and some enrichment in fission products like sr’? and cg l3? which condense late from gaseous or volatile precursors. Subsequent upward migration of the more concentrated zone of the debris cloud on a time scale of hours, days, and longer, may occur under the influence of local heating from residual radiation and from debris absorption of radiation from external sources. The evaluation of these as well as meteorologi- cal factors which may modify the vertical distribution of nuclear debris clouds following stabilization appears to be a neglected subject. U-2 aircraft, used in the High Altitude Sampling Program, 2,8 provide excellent means of sampling up to levels of 70,000 feet. However, it is clear from the cloud height data, Table 1, and from other considerations mentioned above, that a substantial proportion of debris from megaton-yield tests was injected to higher levels. 78