twenty thousand to seventy thousand feet, and for the eighty thousand and ninety thousand foot levels. The above forecasts, along with all available observations, were -plotted as hodographs (wind vector diagrams) for further convenient computations. B. Air Particle Trajectory Forecasts. In addition to the wind forecasts themselves, auxiliary air particle trajectory forecasts were issued at H minus 24 and H minus 8 hours, These consisted of trajectories from the shot site for the period of H-hour to H plus 72 hours, and were for each ten thousand foot level, from ten thousand to sixty thousand feet. Revised tra- jectories for the same period were issued at H plus 6 and H plus 15 hours. The forecasts were used by radsafe personnel primarily to assist in analysis of the long-range fallout aspects and to. alert other units of the Armed Forces relative to the probable locations and altitudes of areas with airborne radioactive particles, C. Constant Altitude Balloon Flights. During CASTLE, a project was instituted whereby constant altitude balloons were used to determine the feasibility of tracking radioactive debris at selected levels. It was assumed that the movement of that part of the radioactive cloud debris remaining in the air at a particular level could be described by a balloon flown at the same level. This project was unsuccessful during CASTLE due to poor pre-launching checking procedure, an insufficient number of balloons, and improper type of balloon; however, this type of pro- 182