ON THE FACTS OF EARTHQUAKE ΡΗΔΟΝΟΜΈΝΑ, 69 
affected during or after earthquakes in South America; and Erman has 
shown the same for Asia. 
At the moment of the two first shocks of the earthquake of Cumana of 
1799, Humboldt says there was a strong electrical explosion at a great 
height, and a few minutes before a violent blast of wind, succeeded by rain. 
The barometer was a little lower than usual, but the progress of the horary 
oscillations was in no way interrupted; the shock took place just at the mo- 
ment the height was a minimum. (Per. Nar. vol. iii. p. 319.) 
Cotte thinks the barometer rises during earthquakes. Hoffman (Pog- 
gen. Annal. xxiv. 56) says it fell at Palermo. 
It has been asserted, that at Cape Francois a water barometer fell 23 inches 
at the moment of the shock of 1770; this would correspond to about one- 
fifth of an inch of mercury. No general or well-authenticated facts of falls 
or rises of the barometer on such occasions could be traced however by 
Humboldt, and if there be any, he is disposed to attribute them to evolutions 
of gaseous matter from the shaken earth. (Per. Nar. vol. ii. p. 225.) 
It however must be borne in mind, that at the moment of shock, if it have 
any vertical element of motion w hatever, some motion must be produced by 
mere inertia, in the mercurial column, and this fact does not seem hitherto 
to have met any attention from earthquake observers. 
Von Hoff says, “ Since the condition of the barometer on the occurrence 
of an earthquake has been attentively observed, before, during, and after the 
shock, both at the place where the earthquake took place and at others more 
or less distant from it, these observations seem to convince us that no fixed 
rule with respect to the behaviour of the barometer during an earthquake has 
been up to the present time proved.” There are examples of falls of the baro- 
meter before or during earthquakes, and there are also instances of the mer- 
curial column rising during similar occurrences, as well as others where 
perfect absence of motion prevailed during very violent shocks. The same 
observations have been made with respect to volcanic eruptions. We are 
indebted to Herr Kries (De Nexu inter Terrze Motus, &c. et Statum Atmo- 
sphere,’ Mém. cour. Leipsic, 1832) for an excellent memoir on these baro- 
metrical changes, supported by numerous examples. As the view given by 
him does not come down further than the year 1826, we add here some ex- 
amples of more modern date, and also some not in Kries’s collection. 
Examples of Falls of the Barometer preceding Earthquakes. 
1720, Ist July.—A great fall of the barometer two days before the earth- 
quake in the Erzgebirge. 
Ἷ 1744., 3rd June.—Before the earthquake in North America, the mercury fell 
3 lines. 
Bi 1826, 23rd June.—A¢é the moment of the shock at Trient, there was a fall of 
1 inch 8 lines. 
1828, 29th January.—Immediately after the shock in the Swabian Alps, the 
ς΄ barometer fell 3 lines. 
a 1828, 8th February.—At the same place, on the day following the earth- 
4 quake, there was a fall of 3 lines. 
_ 1828, 23rd February.—Before the earthquake in Belgium, there was a very 
‘ great fall of the barometer through the whole of Germany and even 
further. It rose however during the shock. 
_ 1828, 22nd March.—On the repeated shock, a much more widely-spread 
᾿ς low position of the mercury. 
3 1830, 9th September.—On na occasion of the earthquake in the Swabian 
