Seismometer

Seismometer is an instrument for measuring shakings, tremors, and tiltings of the earth (see EARTHQUAKE). In its earlier and ruder forms it was merely a seismoscope or seismograph, and its indications or records could not be interpreted quantitatively. Such, for example, is Babbage's bowl of treacle, in which the liquid, tending by inertia to remain steady as the bowl moves with the ground, leaves a high tidal mark as an evidence of the earthquake. Probably, but by no means certainly, such an instrument would indicate the direction of the largest motion occurring in an earthquake. It could not, however, record the succession of oscillations that make up the shock, or give really comparative records of different earthquakes. Similar limitations exist in all forms of liquid seismoscopes, such as Mallet's and Palmieri's, in which mercury moving in glass tubes forms the 'steady body.' A complete seismometer must, indeed, be capable of recording the entire earthquake motion in time, so that the amounts and rates of motions in all directions can be readily estimated from the records. The practical realisation of such an instrument we owe to the labours of Ewing, Gray, and Milne, who, working simultaneously and more or less in concert, brought to bear upon the problem of Japanese earthquakes a rare combination of scientific knowledge and mechanical skill. Gray and Milne's Seismometer (see Philosophical Magazine, 1887) is perhaps the most complete of its kind, and is identical in principle with Ewing's Bracket Seismometer, which was the earliest instrument constructed for recording on the same sheet both the horizontal and vertical motions of small earthquakes. In these instruments there are three weights suspended by brackets, so that each has somewhere within it a 'steady point' with reference to one of the three directions, N. and S., E. and W., up and down. To these weights suitable levers are attached, which trace out on a smoked surface any desired magnifications of the relative motions of the earth and the corresponding steady points. Thus we obtain records of the three components of an earthquake motion; and from these the whole motion can be reconstructed and a model made of the complex motion of an earth particle, as has been done by Professor Sekiya of Japan. When the vertical motion is large, so that the earth's surface is thrown into distinct waves, these bracket seismometers fail to act as such. The precise meaning of a 'steady point' may be best obtained from consideration of a long pendulum with a heavy bob. We may suppose a pencil fixed to the bob and bearing lightly upon a sheet of paper resting on the ground. In an earthquake the ground moves, and with it the pendulum support. But the pendulum bob, because of its inertia, remains steady during the first motion, of which, consequently, we get a tracing on the paper. The subsequent tracing is a combination of the real earthquake motion and the slow swing of the pendulum itself. To get a perfectly steady point we should have to use an infinitely long pendulum. By a combination of an inverted unstable pendulum with an ordinary stable pendulum of convenient size, the stability may be reduced to neutrality, and a very satisfactory steady point obtained. This plan, first suggested by Principal Forbes, is very effectively realised in Ewing's Duplex Pendulum Seismometer. The record is a superposition of the whole horizontal motion, and gives definitely little more than the maximum displacement in direction and magnitude. For earth tremors and earth tiltings much more delicate instruments are needed than have been described. Of these Bertelli's Tronometer is the best known. For large and disastrous earthquakes the havoc wrought is the only seismoscope that can be depended upon.