ON EARTH TBEMORS. 327 



after-effect the light-point Tvas slowly travelling across the paper. At 

 7 P.M. waves begin to be visible, which are beautifully distinct between 

 8.44 and 10.40. The mean period of a wave is about 9 minutes, and 

 the average range is 0"'l. The motion disappears gradually, and the 

 curve soon reassumes its ordinary appearance as a dark line of uniform 

 breadth. 



This is the only case in which earth pulsations were distinctly ob- 

 served during the experiments in 1889. The curves at Wilhelmshaven 

 certainly present very large irregularities, and systems of waves of all 

 sorts of periods, from a few minutes up to nearly one hour, appear ; but 

 all these disturbances bear a more irregular character. At Potsdam, 

 again, the light-jaoint was rather large, and consequently a broad line was 

 drawn, and it may be that this was the reason why no trace of earth 

 pulsations was ever noticed again, for, when the waves are small, the 

 light-point must not be beyond a certain size in order that the waves 

 may be visible. 1 may mention, however, that the curves which I ob- 

 tained when the glycerine arrangement was still in use showed traces 

 of pulsations on many days. The borders of the black line have the 

 appearance of a zigzag, like the teeth of a saw, a tooth on one side 

 corresponding with an interval between two teeth on the other. Thus, 

 it is quite possible that, for the sake of representing such motions as 

 earth pulsations, it would be useful to damp the swing of the pendulum 

 in order to avoid the effect of other microseismic movements, which cause 

 the pendulum to oscillate, and thereby help to increase the breadth of 

 the curves. 



On January 5, 1890, earth pulsations first appeared on a photograph 

 taken at Orotava. The period of the waves is much smaller than in the 

 case described above, and they are so close together that it requires a 

 magnifying glass to distinguish and to count them. Beginning at 

 6.37 A.M., 103 half- oscillations are distinctly seen, which are extremely 

 regular as far as the period is concerned, which is 45 seconds, or 1^ 

 minute for a complete oscillation. The range of motion is variable, and 

 it is easily recognised that several systems of waves are superimposed on 

 each other. The total range of motion is about 0"'15 on an average. 

 Afterwards the waves are less distinct ; larger deflections interfere and 

 destroy the regularity. 



Soon after this observation 1 heard of some curious phenomena that 

 had been observed at Madeira. The sea had risen and fallen at short 

 intervals on the south coast, whilst on the Desertas Islands large land- 

 slips had occurred which caused the breakage of a cable in the neighbour- 

 hood. When examining more closely the times of observation, I found 

 that there was certainly no direct connection between these phenomena 

 and the disturbance observed at Orotava on Januarv 5. 



Later on it appeared that the latter was not at all extraordinary, for 

 on many days similar disturbances were observed, always at the same 

 time of the day, viz., the time in which the pendulum is in its western 

 elongation. On many days it was possible to count a few waves, but 

 with one exception they were too close together to allow such a detailed 

 investigation to be made as on January 5. This happened on April 7 

 when thirteen waves, with a period a little more than 7^ minutes each, 

 appeared on the photograph, nearly as distinct as in the case observed 

 at Potsdam. It is interesting to note that, on the system of these 

 waves, which have a range of 0"'07, other smaller waves are visible, 



