ON EAETH TREMORS. 329 



period could always be accurately determined by counting the number of 

 the waves within the space of a lew hours, and a period of a little less than 

 three minutes was found in most cases. But it is evident that this is not 

 a constant, for occasionally waves with longer periods are observed, and 

 again in some parts of the curve the borders show traces of waves of very 

 short period, like a fine fringe. 



The size of these waves is always about the same as above, and they 

 present no such variations as in Puerto Orotava. They generally are 

 visible during several succeeding hours, but often, too, a portion of the 

 curve, which is as undisturbed as a straight line, is broken by a short 

 series of waves. 



Towards the end of the winter these earth pulsations seem [to have 

 again entirely disappeared. No change of any kind was made in the 

 instrumental arrangements, and the curves are quite as distinct as before. 

 Perhaps a more careful inspection, which will be made when the observa- 

 tions have come to a close, will show traces where they have now been 

 overlooked ; but certainly nothing like the two cases mentioned above has 

 occurred during the last four or five months. 



At Tokio Professor Milne has observed earth pulsations, which differ 

 from the above, because their period is only a few seconds. But in a 

 photograph which he has lately sent me I find a zigzag line of quite the 

 same form as described above ; thus it is certain that waves of longer 

 period occur in Japan also. 



The study of earth pulsations may prove to be of importance for 

 many branches of scientific research. For instance, they may explain 

 some curious discrepancies which are occasionally noticed in astrono- 

 mical observations. If it should be found by future observations that 

 waves representing an angular value of a few tenths of a second are a 

 frequent occurrence, astronomers will be forced to take them into account, 

 and will no longer be able to rely on the present method of determining 

 the level in all delicate researches. 



(3) Earth Tremors. — A third appearance of interest in the curves is 

 what is generally called earth tremors. Perhaps they are nothing else 

 than earth pulsations of short period, but I prefer to think that they are 

 principally due to the swinging of the pendulum, produced by small 

 vibrations and pulsations, which now increase and now again retard its 

 motion. Such tremors generally last many hours, and sometimes days, 

 and the curve, which is often like a dark black line drawn by a ruling pen, 

 takes the appearance of a succession of small earthquake figures. When 

 roughly comparing the results obtained at Potsdam and Wilhelmshaven, I 

 found that strong tremors were nearly always observed at both places 

 simultaneously, and that they bear a relation to the force of the wind. When 

 strong wind is blowing tremors may always be expected to occur, and 

 sometimes they appear to outrace the wind, for they are also noticed in 

 calm weather w^hen there is wind at a distance. The intensity of tremors 

 is not always proportional to the strength of the wind. No systematical 

 comparison, however, has yet been made, because of the many interrup- 

 tions at Wilhelmshaven and the short duration of the observations. 



At Puerto Orotava tremors were quite insignificant ; in fact, there is 

 scarcely a trace of them, except during the hours when earth pulsations 

 occur. This is all the more remarkable because heavy winds are con- 

 stantly blowing against the high mountain chain in the island, but per- 

 haps they would have been more noticeable if observations had been 



