S U M M A R YWe examined the seismicity of central Greece between 1890 and 1988, using macroseismic and instrumental data, to ask two questions: (1) does the seismicity of this period reveal all the major tectonic structures that are known to be active?; and(2) what are the likely strains associated with the seismicity over this period? Many known active structures have been effectively aseismic for the last hundred years, and even the inclusion of all known large events earlier than 1890 reveals no activity associated with the NE coast of Evia, Gulf of Argos, or graben NE of Mt Parnassos. It is clear that even 100 years' data are inadequate for either a reasonable assessment of seismic risk or for a confident estimation of maximum magnitude. However, we are aware of no eai thquakes in central Greece during the last 200 yr that were larger than Ms7.0. It is probable that the maximum magnitude is restricted by the maximum length of fault segments, which appears to be around 15-20 km. The earthquakes of M, 3 5.8 during 1890-1988 can account for a N-S displacement of around 45-70 cm (with maximum and minimum estimates a factor of two greater and smaller than this) across part of a 1890-1900 triangulation network in central Greece that was resurveyed in 1988. The contribution of smaller events may increase this displacement by about 50 per cent. This cumulative seismic displacement is similar to that estimated from the geodetic work (about 100 cm), but a detailed comparison of the two sets of observations will be reported elsewhere. A re-evaluation of all the important earthquakes of 1890-1988 in central Greece is presented in the Appendix, which summarizes information of use to both earth scientists and engineers.
INTRODUCTIONIn this paper we examine the seismicity of central Greece (36.3"-39.SUN, 21.0°-25.5"E) over a hundred year period, from 1890-1988. We attempt to assess the location, magnitude, and likely style of faulting for all earthquakes with M s a 5 . 8 , and then to estimate the total strain across central Greece in this period. There were two principal reasons for carrying out this study. The first was to see whether the seismicity over one hundred years correlates better with the active structures revealed in the geomorphology than does the seismicity over the last 25yr for which good instrumental data are available. The second was to estimate the overall strain accounted for by the seismicity in the period 1890-1988, in order to compare it with the strain measured geodetically over the same time interval.