322 EAETHQUAKES. 



the close accord there is between the results obtained at 

 dififerent towns during successive months. 



At Florence, before a period of earthquakes there is 

 an increase in the amplitude and frequency of vertical 

 movements. These vertical movements do not appear 

 to coincide with the barometrical disturbances, but they 

 appear to be connected with the seismic disturbances. 



They are usually accompanied with noises in the tele- 

 phone, but as the microphone is so constructed as to 

 be more sensitive to vertical motion than to horizontal 

 motion, this is to be expected. This vertical motion would 

 appear to be a local action, inasmuch as the accompanying 

 motions of an earthquake which originates at a distance 

 are horizontal. 



Storms of microseismical motions appear to travel 

 from point to point. 



Sometimes a local earthquake is not noticed in the 

 tromometer, whilst one which occurred at a distance, 

 although it may be small, is distinctly observed. To 

 explain this, Bertelli suggests the existence of nodes. 

 Similar conclusions were arrived at by Eossi when experi- 

 menting on different portions of the sides of Vesuvius. 

 Gralli noticed an augmentation in microseismic activity 

 when the sun and moon are near the meridian. Gra- 

 blovitz found from Bertelli's observations a maximum two 

 or three days before the syzygies, and a minimum three 

 days after these periods. He also found that the prin- 

 cipal large disturbances occurred in the middle of periods 

 separating the quadrature from the syzygies, the apogee 

 from perigee, and the lunistigi period from the nodes, 

 whilst the smallest disturbances happened in the middle 

 of periods opposed to these. 



P. C. Melzi says that the curves of microseismical 

 motions, earthquakes, lunar and solar motions, show a 

 concordance with each other. 



