Results are reported of statistical studies of 2-hour power spectra of nonoverlapping H component magnetic record segments. The records were obtained in January 1972 at an antarctic magnetometer station (Siple) near L = 4 and at three stations (L • 4.4, •4.0, •3.2) at approximately the same magnetic longitude in the northern conjugate area. The K index measured near the longitude of the stations was <5 during the interval studied. (The explicit dependence of magnetic power on geomagnetic activity is examined in part four of this series (Lanzerotti and Surkan, 1974).) It is found in the present paper that during more disturbed periods, higher power levels are accompanied by steeper spectral slopes; this finding implies a concentration of energy at the lower frequencies at these times. Relatively higher power levels are observed during the local afternoon and late evening than during other periods of the day. On the basis of power levels and spectral slopes, Siple (L • 4) is more nearly conjugate to a northern station at L • 4.4 than to a station at L • 4 during the 1971-1972 December solstice period. This finding may arise from hemispherical differences in ionospheric conductivities and, as such, may be dependent upon season. Because of the difficulty of characterizing and/or parameterizing large numbers of chart-recorded observations the background geomagnetic disturbances measured at a ground-based location have not been studied in recent years as extensively as the observed wavelike variations. Frequently, analog chart records of the earth's field are parameterized by determining the largest excursion in some field component during a specified time interval [e.g., Frey et al., 1971]. The well-known, and extensively used, K indices are an example of such a parameterization [Mayaud, 1967]. Quantitative techniques for the determination of the K parameter, using the output of an automatic standard observatory [Alldredge and Saldukas, 1964], have been made more sophisticated by Joselyn [1970]. Quantification of the magnitude of geomagnetic disturbances (in terms of physical quantities such as magnetic power in given frequency bands) would facilitate detailed studies of the spatial extent of geomagnetic events as well as their conjugacy. In the early 1960's, much work on magnetic conjugacy was performed by comparing chart recordings of field variations taken at opposite ends of a line of force (e.g., reviews by Wescott [1966] and Campbell [1968]). In recent years, relatively little research of this type has been carried out, even though conjugacy (or the lack of it) and its spatial extent are not well understood. In one of the few multistation experiments, Walker [1968] examined hourly range data from Byrd (in the auroral zone) and its conjugate area using five stations in an east-west array in the northern hemisphere. In his analysis, Walker examined the diurnal variation of correlations in the hourly range data as well as the coherence as a function of frequency over a short period of time during a magnetic storm. ...
