1968, someone suggested to Michigan State graduate student Robert D. Regan that an interesting topic for a dissertation would be the use of satellite to compile a detailed map of the irregularities in the earth's field. did a hand calculation, Regan recalls, I said, 'That's absurd!' ' The strength of the earth's field at the surface, after all, is only about 50,000 gammas, roughly equivalent to that of a toy horseshoe magnet. The irregularities, or anomalies, to be charted represented no more than 10 or 12 gammas-variations, in other words, of less than one-fortieth of one percent-to be reliably plotted from collected at satellite altitudes. Yet Regan, now with the U.S. Geological Survey at Reston, Va., together with Joseph C. Cain of NASA (now with USGS) and W. Minor Davis, also of USGS, have done it. Studies of their map, already under way, will shed light on a variety of geophysical questions such as tectonic movements and mineral concentrations. The map, recently published in the JOURNAL OF GEOPHYSICAL RESEARCH (80:794), was compiled from gathered by three U.S. satellites in the Orbiting Geophysical Observatory series: OGO-2, 4 and 6, launched in 1965, 1967 and 1969 respectively. All three were in polar orbits, so that the whole planet passed beneath them; hence their official name, POGO. Such mapping has been done before, Regan notes, but with considerably less resolution. Data from Cosmos 49, for example, a Soviet probe of a decade ago, were organized into map form, but there were only 11 days' worth of usable observations, which yielded simply too few measurements for detailed results. each square of the earth's surface measuring 5 degrees of latitude by 5 degrees of longitude there were only from I to 23 Cosmos 49 observations, or data points.' The PoGo's, by comparison, yielded from 300 to 450 measurements per 5-degree square. There have been previous attempts using POGO data, but with longer wavelengths, which limits the possible resolution. To keep distortions to a minimum, Regan and his colleagues started by throwing out most of the POGO data. They ruled out everything gathered when solar activity was having too great an influence, as extrapolated from an indicator of planetary activity called the Kp index. To avoid distortions in polar and auroral zones, they limited their initial effort to from less than 50 degrees north and south of the equator. Because the extretnely weak anomalies are even weaker at extreme altitudes, gathered from more than 700 kilometers up were ruled out. Finally, measurements made between 9 a.m. and 3 p.m. local time were eliminated to avoid the effects of diurnal changes in the field. More than half of the PowO measurements were eliminated, but a rich harvest remained for mapping: 393,452 observations, taken at 7-second (about 50-kilometer) intervals along a given orbit, resulting in about 15 observations for every square degree of surface to be covered. From this mass of numbers, already a map of sorts, the overall field of the itselfthe magnetism associated with compasses and north, Regan observes-then had to be subtracted. Graphically plotted by a computer, the resulting data, at long last, produced the map. The sharp-edged contours, emphasized further by eight shades of color, make the features of the magnetic earth seem clear and well-defined. Yet the unimaginably tiny variations that comprise the map are in fact elements in what is probably the subtlest portrait of the planet ever made. Subtle, but real. The variations in the map seem to have little to do with the outlines of the continents, but some of them do represent variations in the material in the earth's crust. We are quite confident that most of the anomalies are actually related to significant differences in the upper 50 kilometers of the earth, says Regan. In particular, we suspect that areas showing the greatest range in anomalies on the map will turn out to be areas that have experienced the most intense geotectonic activity, such as crustal movement, faulting and volcanism, in the past. The so-called Bangui anomaly in Africa, for example, shows up clearly on the map (it was one of the first crustal anomalies to be discovered from satellite observations), and lies over the