SHAPE AND COLOR Particles of many shapes and colors have been identified. Descriptive details vary with the investigators and are qualitative only. Table 1 summarizes the findings for balloon, tower, ground surface, and under- ground shots at the Nevada Test Site. Most of the radioactivity is associated with glassy, fused material, particularly with rounds. Some more detail with regard to color is provided in Table 2, which shows the color distribution of round particles from various types of shots from the Plumbbob and Teapot series. Additional data of this type from other shots and series are available, but these data would not add anything material and therefore have been omitted. The data show that there is no clear height-~of-burst dependence of the color distribution, as expected. In general, the particles from tower shots are darker than those from air or balloon shots, probably as a result of the large amounts of iron and/or soil present. The color distribution of the particles from the one underground (cratering) shot mentioned is similar to that of the tower shots, although the sample renders this conclusion unclear. DENSITY OF RADIOACTIVE PARTICLES The particles for which density data have been found are mostly larger than 100 um, and were selected on the basis of their radioactivity content. Their representativeness may be questioned. The data are tabulated in Table 3. Whereas the density of airburst particles is mostly in the 3-4 g/cm? range, the densities measured on particles from near-surface, surface, and subsurface bursts are smaller, rarely exceeding 3.0 g/cm3. Often the densities are even less than the density of the "host rock," sometimes significantly so. Even for the balloon shot, a type closest to an airburst, the particle densities are not especially high. This is apparently caused by vesiculation. Vesiculation is probably also the cause of the large variability of the densities. It is further noted that the great majority of particles were larger than 100 um. We have arranged the table in such a way as to facilitate the detection of regularities, if any exist: by type of shot, by substrate, by scaled height of burst. There is no apparent dependence of the measured densi- ties on the scaled height of burst in tower shots over coral or over alluvium, nor in surface bursts on coral. Particles from tower shots over alluvium appear to have somewhat lower densities than those from tower shots over coral. This is commensurate with the lower density of 554