158 EARTHQUAKES. 



of Cauquenes fell from 118° to 92° F. Subsequently, 

 however, they again rose. 



Fumaroles are similarly disturbed. Thus, at the time 

 of the earthquakes of Martinique (September, 1875), the 

 fumaroles there showed an abnormal activity.^ 



Wells often appear to be acted upon in the same 

 manner as springs. 



At the time of the California earthquake (April, 1855), 

 the level of the water in certain wells was raised ten to 

 twelve feet. 



A consequence of the earthquake at Neufchatel, in 

 1749, was to fill some of the wells with mud.'^ At 

 Constantinople, on September 2, 1754, wells became 

 dry.3 



Explanation of the above phenomena, — That the 

 water in springs and wells should be caused to rise at 

 the time of an earthquake, admits of explanation on the 

 supposition of compressions taking place similar to those 

 which cause the rise of water in fissures. That the water 

 in wells and springs should be rendered turbid, is partly 

 explained on the supposition of more or less dislocation 

 taking place in the earthy or rocky cavities in which they 

 are contained or through which they flow. 



At the time of a large earthquake it is extremely 

 probable that there is a general disturbance in the lines of 

 circulation of subterranean waters and gases throughout 

 the shaken area. By these disturbances, new waters may 

 be brought to the surface, two or more lines of circulation 

 may be united, and the flow of a spring or supply of 

 a well be augmented. Fissures, through which waters 

 reached the surface, may be closed, wells may become dry, 

 or springs may cease to flow, hot springs may have their 



» Comj)te Rendu, 1875, p. 693. ^ (j^^^, j/^^_ yoi. xix. p. 190. 



' Phil. Trans, vol. xlix. p. 115. 



