314 FEELY AND BAZAN was injected and that subSequent sampling along any meridian should give a representative cross section of the atmosphere. Thus, although we must remember the limitations on the accuracy of the distributions we measure and the stratospheric burdens of *Sr we calculate from them, our estimates of the burden should generally be accurate to within 25%. The stratospheric distributions of "Sr calculated from a combina- tion of balloon and WU-2 data are shownin Figs. 10 to 13. The distribu- tion during January through April 1962 (Fig. 10) reveals the presence of fairly high *°Sr concentrations in the layer between the tropopause and 70,000 ft in the northern hemisphere. The concentrations of °°Sr in this layer which can be attributed to pre-1961 weapons tests were approximately equal to the total *°Sr concentrations in the same region in the southern hemisphere. The remaining **Sr was produced by the 1961 Soviet tests. It is noteworthy that in spite of the very high yields of some of the devices tested during that series, the injected debris was contained almost completely in the lower stratosphere. Of course, as we have previously stated, representative concentrations of the debris from these high-yield weapons were not present in the sampling corridor at the time. _ The distribution of *Sr during May to August 1962, presented in the lower half of Fig. 10, shows the effects in the northern polar strato- Sphere of the debris from the very high yield weapons and in the tropical stratosphere of the debris from the 1962 United States tests. The appearance of the debris from the high-yieid weapons caused an increase in the concentrations of *Sr in the 60,000- to 100,000-ft layer in the polar stratosphere but not enough of an increase to change our opinion that comparatively little debris had been injected above 70,000 ft. The concentrations of "Sr indicated in the tropical strato- sphere are actually quite uncertain. Sampling in the region below 70,000 ft was not intensive enough to ensure its representativeness during this period of weapons testing. The concentrations shown in the region above 70,000 ft are conjectural since debris from the United States tests had not yet reached either San Angelo or Mildura in quantity and no sampling was performed within the upper tropical strato- sphere where this debris had been injected. The *Sr distributions during September and October 1962 and November and December 1962 are shown in Fig. 11. Both of the indi- cated distributions are subject to relatively large errors because the 1962 U.S.S.R. tests were in progress during both intervals and even the debris from the earlier United States Dominic I series was apparently still not uniformly distributed in the zonal direction. The distributions indicated in the upper stratosphere, above 70,000 ft, and in the entire tropical stratosphere are quite uncertain.

Select target paragraph3