“…In such records of complex signals, however, it should be noticed that the distance dependence of signal level is liable to be flattened so as to show a high level at a shorter distance. Figure 16 shows a frequency distribution with distance of 50 MHz radio contacts of 12,500 instances [ Gerson, 1955], that of 3,228 TV-DX reports [Smith, 1957], and the above estimated probabilities on 50 and 70 MHz signals under appropriate parameters. They agree, approximately, with each other, but the sporadi½-E events appear skewed toward shorter distances than estimated, which might, at least in part, be due to the geographical boundaries of the tests.…”