88 Mr Toplis on Hydrostatical Pressure 



particularly that of Lyell, vol. ii., p. 249, &c, 5th ed. It is 

 ohserved by Ulloa and Humboldt, that the subterraneous 

 roarings which accompany the earthquakes and the amazingly 

 great quantities of water that issue from the earth in the dry- 

 est places during their shocks, also the eruptions from volca- 

 nos, and numerous other phenomena, indicate that all the soil 

 of the plains in the vicinity of the Andes is undermined. This 

 may have been occasioned by the constant percolation of wa- 

 ter forced gi'adually forwards, where there is the least opposi- 

 tion, by the weight of the high columns contained in the 

 mountains. It is stated by Raynal, in his Histoire Philoso- 

 phique, that numbers of rich mines in Peru are ruined by the 

 water which descends in a slope from the Cordilleras to the 

 South Sea. 



The waters of seas, rivers, lakes, wells, and springs are ge- 

 nerally affected by earthquakes, in some cases to a very con- 

 siderable distance ; for instance, on the day of that at Lisbon, 

 November 1. 1755, within a short interval, the waters of the 

 sea in the West Indies, on the shores of Madeira, of lakes in 

 Scotland, &c. &c, were unusually agitated. This may have 

 been occasioned by the oscillations of the earth being com- 

 municated laterally, causing it to heave and subside under 

 the waters. The strata were observed to be shaken in many 

 of the Derbyshire mines from the effects of that earthquake. 

 According to Bischof, in his Memoirs on Volcanos and Earth- 

 quakes, the commissioners who were employed to make ob- 

 servations on the earthquakes in the district of Pignerol near 

 Turin, relate that the very day, April 2. 1808, when one of 

 the most violent shocks was felt, the masting engine at Toulon 

 was elevated more than an inch. 



Boussingault asserts, Annal. de Chim. et de Phys. t. lviii., 

 p. 83, that the most memorable earthquakes in the New 

 World, which ravaged the towns of Latacunga, Riobamba, 

 Honda, Caraccas, Laguayra, Merida, Barquisimeto, &c, do 

 not coincide with any well-established volcanic eruption. The 

 oscillation of the surface, owing to an eruption, is as it were 

 local ; whilst an earthquake which is not subject (at least ap- 

 parently) to any volcanic explosion, extends to incredible dis- 

 tances, in which case it has also been remarked that the 



