drift electrons would therefore move eastward instead of westward. In the vicinity of the boundary, this would tend to cause a decrease of ion density. Furthermore, the ccmbination of decrease in atmospheric density with height and the conservation of mass would result in a greater downward velocity of air molecules at higher altitudes than at lower levels. The ions that are moved eastward by the descending air would be swept eway more rapidly at higher than at lower levels, and the effective height of the F2 layer might even be lowered. Tie ionosphere recorder would not, therefore, register the large apparent increase in height that would be observed ‘directly below the rising heated air mass. Of course, turbulence near the boundary, and tpiling up" of ions would complicate the picture. The above discussion is offered as an explanation of the events following Shot 4. The yield of this shot was smaller, which would have resulted in a smaller volume of heated air. Furthermore, because of the later date, seasonal changes in upper atmosphere winds might have caused a different horizontal distribution of blast-wave energy. The net result may have been that, for this shot, the ionosphere recorder was located outside of the boundary of the heated air mass, hence no pronounced change in the height of the F2 layer occurred. (The large apparent increases in level observed at about plus one hour appear to have been due to a horizontal motion of the point of ree~ flection as described in Section 3.3.3, and may have been the early stages of the wave motion which gave rise to the distant effects de= scribed in Section 3.4.) In the case of Shot 5, the occurrence of a G condition (see Section 3.3.4) made it impossible to obtain a complete picture of the F2elayer phenomena. Another mechanism which may be of importance in explaining the results is the "cyclone" action due to the rising air mass. Air moving in horizontally to replace the ascending air would be deflected by the Coriolis force caused by the earth's rotation, and a counterclockwise whirl would result in the northern hemisphere, just as in the case of an ordinary tropical or extratropical cyclone. The winds of this whirl would have a northwardly directed component east of the shot point (where one of tke ionosphere recorders was located). This component would lie along the earth's magnetic field, and, inasmuch as electrons can be moved freely parallel to the field, the F2-layer ions would tend to move northwards and downwards. This may be the mechanism by means of which the F2-layer conditicns return to normal (i.e., F2-layer ionization is moved in from the south by this process). This would explain the occurrence of the lower of the two maxima de-= scribed in Section 3.3.1 (see also Fig. 3.12) and would also explain the gradual descent of the F2 layer as conditions return to normal, From the above discussion, it is obvious that the F2-leyer phenomena that follow a large atomic blast are quite complicated and probably cannot be com letely explained with the limited data available. Data covering a wider range of distances and azimuths would be ree guired in order to obtain a complete picture, 68 SECRET

Select target paragraph3