The value to seismology of a global net of seismometers has long been recognized. Indeed, the first proposal for such a net came in 1895, only 6 years after the discovery that earthquake waves could be recorded at long distances, and the first net of standardized instruments was in existence by 1898 [Dewey and Byerly, 1969; Wood, 1942; Milne, 1899]. A true network, involving not only standardized high-quality instruments but also the exchange and ready availability of data, did not come until the establishment of the World-Wide Standard Seismograph Network (WWSSN) in the middle 1960's [Oliver and Murphy, 1971]. The importance of this to general seismology cannot be overstated; not only has it improved the quality of traditional research areas, but the availability of the data has suggested new types of investigations. In the decade since the WWSSN was built, two small networks have been set up: the High-Gain Long-Period instruments (HGLP) [Savino et al., 1972] and, very recently, the Seismic Research Observatories (SRO) [Peterson and Orsini, 1976]. Both of these networks use careful shaping of the instrument response to improve the detection of surface waves from very small events, which is important in lowering the magnitude threshold for discriminating earthquakes from explosions.