spherical particles 1 to 25 microns in diameter on their surfaces (Figures 3.13G and 3.15). Shot Tewa particles were almost entirely free from spherical particles of this kind, although a few with diameters less than 1 micron were discovered when someof the irregular particles were powdered and examined with an electron microscope. A few larger isolated spherical particles were also found in the Zuni fallout (Figures 3.13, Band H). Such particles varied in color from orange-red for the smallest sizes to opaque black for the largest sizes. While these particles were too small to be subjected to petrographic or X-ray diffraction analysis, it was possible to analyze a numberof larger particles collected during Shot Inca which appeared to be otherwise identical (Figure 3.19). The Inca particles were composed primarily of Fe,Q, and calcium iron oxide (2 CaO.Fe,0,) but contained smaller amounts of Fe,O, and CaO. Some were pure iron oxide but the majority contained calcium oxide in free form or as calcium iron oxide (Reference 24). Most of the spheroidal particles consisted of coarse-grained calcium hydroxide with a thin surface layer of calcium carbonate (Figure 3.16). Nearly all contained at least a few grains of calcium oxide, however, and some were found to be composed largely of this material (Figure 3.18) —-5 to 75 percent by volumie. Although melted, particles of this kind probably underwent much the same chemical changes as the irregular particles, the principal difference being that they were incompletely hydrated. They varied in appearance from irregular to almost perfect spheres and in color from white to pale yellow (Figure 3.13, C, H, and K). Many had central cavities, as shown in Figure 3.16 and were in some cases open on one side. Because of their delicacy, the agglomerated particles could not be thin-sectioned and had to be crushed for petrographic and X-ray diffraction analysis. They were found to be composed primarily of calcium hydroxide and some calcium carbonate. It has been observed that similar particles are formed by the expansion of calcium oxide pellets placed in distilled water, and that the other kinds of fallout particles sometimes change into such aggregates if exposed to air for several weeks. The particles were flaky in appearance, with typical agglomerated structures, and a transparent white in color (Figure 3.13, D, I, and J); as verified by examination of IC trays in the YAG 40 laboratory immediately after collection, they were deposited in the forms shown. The densities of 71 yellow spheroidal particles, 44 white spheroidal particles, and 7 irregular particles from Shot Zuni were determined (Reference 25) using a density gradient tube and a bromoform-bromobenzene mixture with a range from 2.0 to 2.8 gm/cm*. These results, showing a clustering of densities at 2.3 and 2.7 gm/cm}, are summarized in Table 3.8. The yellow spheres are shown to be slightly more dense than the white, and chemical spot tests made for iron gave relatively high intensities for the former with respect to the latter. No density deter- minations were made for agglomerated particles, but one black spherical particle (Table 3.7) was weighed and calculated to have a density of 3.4 gm/cm’. The subject of size distribution has been covered separately in Section 3.2.4, and all infor- mation on particle sizes is included in that section. Radiochemical Characteristics. Approximately 30 irregular, spheroidal and ag- glomerated particles from Shot Zuni were subjected to individual radiochemical analysis (Ref- erence 26), and the activities of about 30 more were assayed in such a way that certain of their radiochemical properties could be inferred. A numberof particles of the same type were also combined in several cases so that larger amounts of activity would be available. are tabulated in Tables B.7 and B.8. These data Radiochemical measurements af Sr®*, Mo**, Ba‘‘°-La° and Np?** were made. (All classified information such as the product/fission ratio for Np?**, which could not be included in Reference 26, and the limited amount of data obtained for Shots Tewa and Flathead were received in the form of a private communication from the authors of Reference 26.) For the most part, conventional methods of analysis (References 27 and 28) were used, although the amounts of Np? and Mo®(actually Tc??™) were determined in part from photopeak areas measured on the singlechannel gamma analyzer (Section 2.2 and Reference 29). The total numberof fissions in each sample was calculated from the number of atoms of Mo” present, and radiochemical results were expressed as R-values using Mo"? as a reference. (R-values, being defined as the ratio 50