In a recent paper, Chen et al. (Tellus 6, 1970) describe the results of measurements of fallout particles in the ground level air at Fayetteville (36 0 N, 94 0 W), Arkansas, observed after the December 24, 1967, Chinese nuclear explosion. The authors show graphically the daily values of the concentration of radioactive particles and try to interpret the variations in terms of meteorological conditions, particularly time of arrival of computed trajectories.Since I think that there are some aspects that need to be clarified, I would like to make a few comments.There are of course strong practical reasons for using 300 mb and 500 mb as those heights at which trajectories are computed. Before the contrary has been proved, it is however not clear how well it can be assumed, as in Chen's paper, that the cloud from the beginning is divided into two more or less distinct parts located at those heights, and then interpret the peaks in the observed radioactivity as the arrival of the "500-mb-cloud" or the "300 mbcloud". A more natural assumption would be that the cloud initially had a certain dimension -possibly larger than 500 to 300 mb-and that it is subsequently continuously deformed by the vertical shear of the horizontal wind. The horizontal extension of the cloud could then be approximately estimated by the distance between the endpoints of the trajectories at the top and the bottom-levels of the cloud. Variations on shorter time scales than the time of passage of the cloud could then be attributed to changes in meteorological conditions of a more local character, e.g. the passage of an upper level trough.