On the nature of atmospherics—V

Abstract

Despite the defects and limitations of the methods described in the first three papers of this series (Watt and Appleton 1923; Appleton, Watt and Herd 1926), which were recognized from the first and towards whose elimination continued experimental work was directed, it was considered desirable to make a preliminary survey of the problem of the wave form of atmospherics without a waiting the results of this developmental work. Extensive series of observations were therefore made in the Red Sea and Indian Ocean, in Egypt, and in the Anglo-Egyptian Sudan, with "control" observations in south-eat England. In all of these series one or more of the present authors participated as observers. While no detailed discussion of the results is at present proposed, the general features of the survey are presented in Tables I-VI ( sec pages 289-291), in the form which was adopted for Table II of the second paper of this series. The authors' own criticisms of this early technique fell into two main groups: those arising from the impossibility, at the time of the early observations, of obtaining photographic records on any economically attainable basis of adequate sampling, and those arising from the lack of sufficient knowledge as to the place of origin of the disturbances studied. Some of the criticisms made by other workers, however, appeared to be based on misconception of matters discussed in the earlier papers. It is not now necessary to discuss these criticisms at length; one of the most important points at issue was very fully discussed at the Copenhagen meeting of the Union Radio Scientifique Internationale, and this discussion resulted in the complete acceptance of the argument of the papers, that the time constant of the input circuit, if sufficiently high, did not modify the apparent duration of the field change when the initial and final field values were sensibly the same. Ad hoc photographic testes by the authors, on large-aerial systems of the type used in the earlier work, in which the time constant was varied over wide limits had, in the interval, confirmed the view that sensible distortion of the recorded form only occurred when the time constant fell to the abnormally low value of about four times the duration of the field change to be delineated.