Eleventh regional seminar on earthquake engineering, granada, september 1984.

Some of the problems of combining macroseismic (descriptive) and instrumental information on modern and historical earthquakes are discussed, together with the relationships that have evolved between these two fields of study. The paper emphasises the dangers of transferring macroseismic information straight into computer databanks, without taking full account of the levels of confidence attached to the information and its suitability for earthquake risk analysis.