surge proper and temporarily streaming from the target ships are accepted, the fact that Umbrella tails are consistently somewhat shorter than Wahoo tails may indicate another difference between the two base surges. Shorter tails, for instance, might be expected from a higher, thinner base surge torus presenting a smaller basal area to surface drag. Alternative explanations for the persistence of the dose rate record after surge transit include: (1) changes in wind speed and direction after downwind surge arrival, (2) continued surge expansion or increased eddy diffusion resulting in a thicker base surge torus, (3) pres- ence of a second radioactive aerosol moving above the surface base surge at a different speed or direction, or (4) the prolonged generation of a radioactive mist by some processof white water out-gassing or wind dispersal of foam. Full evaluation of all these possibilities was not attempted, although each was investigated briefly before the tail hypothesis was selected as the most probable on the basis of the observed gamma records. Other surface wind speeds (or directions) within the reported limits of variation (Table 3.1) not only fail to eliminate all observed tails but also frequently create apparent tails at crosswind locations. Furthermore, because these assumed changes are oniy applicable after the last photographically established base surge position, they must often occur after the downwind surge transit but before arrival of the upwind surge (Appendix F). The time of the presumed change would thus differ for the various coracle positions. Continuous radial expansion or greater eddy diffusion of the base surge partially eliminates the observed tails at some locations (particularly for Umbrella) but requires either longer tails or, in some cases, negative tails at other locations. More recent calculations using surge toroids having somewhat greater radii than those assumed in this report fit some downwind gamma records very well; therefore, this postulate is the most probable alternate to the tail theory described above. The postulated influence of the atoll reef should also be remembered when considering any later base surge expansion. A radioactive cloud moving at higher altitude also appears unlikely. For Wahoo, no such upper cloud was photographically detected nor is any Significant wind shear reported up to an altitude of 5,000 feet (base surge height is 2,000 feet). The Umbrella case is more favorable, sincea diffuse remnant of the central column was photographically detected; however, this remnant moved centrally with the base surge torus. Again, no significant wind shear is reported up to an altitude of 5,000 feet (base surge height is 2,000 feet), although some shear is apparent in surge photography (Reference 91). Finally, although a mist emanating from white water was observed for a period of 13 to 14 minutes after Umbrella (Reference 91) and may also have existed undetected after Wahoo, Such tertiary processes Shouid not contain as much radioactivity as a secondary process derived from base surge directly. Furthermore, the doSe rate record of such a radioactive mist would be expected to terminate gradually rather than abruptly as is observed for base surge. None of these alternative explanations can, however, be definitely eliminated or accepted without analysis that is beyond the scope of the project. During the discussion above, a number of structural differences between the base surges generated by the two underwater nuclear detonations have been indicated. The Wahoo base surge appears to be a double toroid with the primary and secondary surges separated by an essentially surge-free annulus (intersurge decrement). The center of the second torus con- tains a number of diffuse clouds, which could represent additional degenerate base surges. The source center (Appendix F) appears to be situated well behind the visible surge front, somewhere near the inner boundary of the primary base surge. The Umbrella base surge is much simpler in comparison, being a single torus with a large surge-free central decrement, although faintly visible remnants of a central column are observed to move with but above the base surge. The source center appears to be situated at or in advance of the visible surge front. Some indirect evidence suggests that the Wahoo surge may have a greater horizontal thick- ness than the Umbrella surge; this observation, however, could simply be a result of the former’s compound structure. Preliminary studies indicate that the Wahoo explosion bubble went through its first maximum expansion and surfaced just prior to its first minimum at an internal pressure somewhat greater than atmospheric. Conversely, the Umbrella explosion 133