INTRODUCTION Data documenting the quantity of fallout reaching the earth's surface is derived, in part, from measurements of samples collected passively at specific locations. Knowledge of the sampling accuracy is necessary to provide reliable values of fallout per unit area. Wind effects constitute the major consideration in representative sampling since particle collection depends critically on the characteristics of the airstream above the instruments. Winds are generally streamline in nature but when they encounter the physical obstruction of the collecting instruments, varying degrees of turbulence and other flow disturbances are produced. Under the influence of these flow conditions above the instruments the trajectories of falling particles become distorted and displaced. These effects will depend on the intensity of the flow disturbances and the physical nature of the particles involved. The net result is that the quantity of particles falling into the collector will differ significantly from the quantity which would have fallen through the area occupied by the instrument. As a consequence, a biased or non-representative sample is collected. In studies of precipitation collection, an analogous situation, the problem of biased collection due to wind effects has long been recognized. It has been found that the amount of rainfall collected varies inversely with the height at which the collectors are positioned due to increasing wind effects. Horizontal windshields have been attached to experimental collectors in attempts to minimize air flow disturbances. Studies have shown that when many identical instruments are arrayed adjacent to each other, their fallout collections vary with respect to their positions relative to the wind direction.2 Fallout sampling in Project 2.63 at Operation REDWING3 was conducted by more openly and systematically spaced arrays or groups of similar collecting instruments. Each array was accommodated in an elevated and circular wind-shielded platform designated as the standard platform. These standard platforms were located at a land station as well as on several ships and anchored barges. The purpose of the windshield was to standardize the air flow pattern over the standard platforms and to minimize wind bias effects. The results of the fallout sampling indicated that for a single platform array the quantities of fallout collected around the platform showed considerable variation. However, it was noted that in certain cases this variation followed a geometry which was oriented to the wind direction. This variation of collection or sampling bias was caused by the particular