During the summer of 1967, three high-frequency, high-gain, and highly portable seismographs were operated at seventy-eight sites throughout Iceland. Over 990/0 of the more than 1000 events recorded were found to lie in nine regions with radii of less than about 5 km. Although most of the events were not greater than 4 km deep, six were of the order of 5 to 15 km deep, and one may have been as much as 40 km deep. One large earthquake swarm was recorded from Myrdalsjiikull in south-central Iceland, where four events less than magnitude 5 were reported by the U. S. Coast and Geodetic Survey (U.S.C.G.S.) in early 1967. No earthquake greater than magnitude 4.5 has been reported since 1958 from the other regions of high microseismicity, suggesting that these microearthquakes were not simply aftershocks. Three events of magnitude 4 to 5 did occur, however, in each of two seismic regions after the initial recording period. Thus, some of the microearthquakes may have been foreshocks. A close correspondence was found between areas of major hydrothermal activity and high microearthquake activity. The highest activity recorded was in the Krafla volcanic region in northeastern Iceland, which has not been active since 1746. This activity had a b value of 0.83 ± 0.16 over 1% units of magnitude. The focal mechanisms were consistently similar and gave a solution with one nearly vertical nodal plane striking north-south. Eight of the nine zones of microseismirity lie on an east-west line near 64·N. When considered in relation to adjoining active seismic zones, reported historic seismicity of Iceland, and the location of the mid-Atlantic ridge and areas of active rifting and volcanism, the existence of a transform fauIt is suggested, following the methods used by Sykes (1967) to outline such faults on the sea floor with larger earthquakes. Magnetic data and some geologic features support this hypothesis. Seismic refraction data are not in disagreement with it. Many of the tectonic features of Iceland do not, however, readily fit into this framework. If sea-floor spreading is active in Iceland, it is more complicated in detail than previously suggested.