Experiments carried out in Alaska with the object of observing long-delayed radio echoes (LDE) are described. The results are subjected to critical sonogram analysis, together with data from previous experiments carried out at Stanford and elsewhere. It is concluded that, although strong supporting evidence for sporadic LDE phenomena has accumulated over the last 50 years, and a variety of plausible mechanisms have been proposed, stringent analysis of records such as those discussed in the paper almost invariably reveal alternative possible explanations for them. Further detailed experimentation is required on mechanisms, for which the phenomenon seems best established, e.g., round-the-world echoes and other forms of guided propagation.