This paper gives the preliminary results of observations made during the course of three years upon the diurnal variation of the level of a concrete pier supporting a seismograph. The pier itself is a solid block of concrete 1.3 m. square, standing 0.58 m. high above ground and extending 1.35 m. below ground, so that the total height of the pier is 1.93 m. Round this pier, at a distance of 15 cm. or so, is built a brick wall, which serves the double purpose of supporting the floor of the

more »

... raph-room and keeping the earth away from the sides of the pier. Therefore the indications of the seismograph may be taken as applying to the stratum 1.35 m. below the surface of the ground. The seismograph is of the horizontal pendulum form, the boom carrying a weight of nearly 6 kilograms. There is an adjusting screw for sensibility, and a graduated micrometer screw for the purpose of measuring the tilt. The indications of the pendulum are received upon a sheet of smoked paper travelling at the rate of 15 cm. an hour. The pendulum part of the apparatus (which was made to my design, with some modification, by the Cambridge Instrument Company) has served the purpose I had in view very well. There is room for a good deal of improvement, however, in the clockwork and drum. The apparatus is housed in a wooden hut of weather-boards, the walls being double and lined with cinders. There is not a window to the hut, a space being left under the eaves for ventilation. The pendulum is adjusted in the meridian to a mean period of 18 seconds, which, however, varies somewhat during the course of the year. A tilt of 1 second of arc corresponds to an average deviation of the extremity of the boom of almost exactly 2 mm. I n Table 1 , at the end, will be found the results in monthly averages.