Abstract

This paper traces some of the main developments in the study of earthquakes and their scientific investigation from 1755 (the year of the Great Lisbon Earthquake: GLE) to 1855. The GLE was widely reported and discussed, though at that time there was no systematic and accurate collection of seismic data so that the event did not in itself lead to significant scientific advances. But an idea is given of the attempts as regards Portugal and Spain to explain the GLE in the terms of the day. In 1760, John Michell described methods for ascertaining (in principle) the position of what would today be called the GLE's epicentre and its focal depth. His attempted explanation of the quake is described. The Calabrian Earthquake (1783) was followed by more systematic studies of its effects, showing how the centre of damage could be identified and estimates made of zones of equal damage (isoseismal zones). The undulatory nature of seismic displacements was recognized by Michell and others, but some observers in Italy thought they detected "vorticose" motion - an idea supported by the clockwise and anticlockwise rotation of the stones of two obelisks disturbed by the Calabrian Earthquake. The association of earthquakes with volcanoes received ongoing discussion through the century following the GLE and electrical explanations were also popular, particularly in Italy. The connection of volcanoes with land elevation or subsidence attracted the attention of Lyell and Darwin. The idea of isoseismal maps was adumbrated by von Buch in relation to the Silesian Earthquake of 1799 and a simple isoseismal map was drawn for the Rhineland Earthquake by Egen (1828) and a simple intensity scale proposed. Von Humboldt described earthquakes and volcanoes he had studied in South and Central America, but failed to establish any systematic system for their recording, and unhelpfully he gave rise to the notion of "craters of elevation" to account for the formation of volcanoes. Through the first half of the nineteenth century, extensive efforts were made to catalogue historical data on earthquakes' timing, location, and intensity, and their concomitant astronomical and meteorological circumstances, but initially few useful patterns could be discerned. There was no network of seismic stations, and the pendulum instruments for earthquake detection and recording were largely ineffective. The early development of seismoscopes/seismographs is described, but none worked satisfactorily in the period under discussion (except for Mallet's method for detecting artificial seismic disturbances). In the 1840s, William Hopkins published mathematical analyses of crustal deformations and earthquake phenomena and the transmission of seismic waves. He recognized two kinds of wave, which travelled at different velocities, and on that basis he proposed methods for determining the focal position of an earthquake. But the wave velocities were not known accurately and, though valid in principle, his method, utilizing the different travel-times for the two kinds of waves, could not be applied immediately. Studying the Visp Earthquake (1855), Georg Volger (with August Petermann) drew two isoseismal maps and proposed a numerical intensity scale, but it was not generally applicable since Volger allocated a value of '0' to the region of maximum intensity and '6' to the areas where motion was just discernible. Robert Mallet's work in the early 1850s was fundamental and marked the beginning of modern seismology (his term). Using artificial explosions and accurate clocks, he measured (longitudinal) wave velocities in soft sediments and hard granite, finding that velocities were higher for the latter. His catalogue of earthquakes and his plot of their distribution worldwide yielded a map that matches modern maps of plate boundaries. Mallet was stimulated by Lyell's drawings of the rotated Calabrian obelisks, and he showed that such movements could be produced by seismic waves, and "vorticose" motions need not be invoked. Soon after 1855, improved seismic detectors and recorders were devised and the systematic seismic investigations began. The period discussed in the present paper could be said to belong to the "pre-paradigm" stage of seismology.