1883.] Burning Mountains . 331 
computation, as may be said of a period of activity coming 
the first or second week in November. From December 1st 
until January 2nd volcanic activity seems to know but little 
respite, although it seems probable that outbreaks have ar- 
rived more frequently during the week following the 24th 
and about the 30th of the month. 
If we now recal to mind the eruption of Etna just trans- 
pired, it might be an open question whether we should refer 
it to the influence of the full moon or to an influx of warm 
temperature. It will be remembered that the full moon 
arrived on March 23rd, and that the eruption began on 
March 20th and continued until April 4th. According to 
M. Ragona’s second table a period of warmth should have 
been likewise looked for in Italy, beginning on March 20th 
and ending April nth ; and, according to his first table, two 
periods of warmth should have come about this time, — one 
from March 27th to 29th, and another from March 31st until 
April 17th. Theoretically, his first table would with great 
probability assign the increase of warmth as the cause of 
the eruption ; but it must be remarked that, in this country, 
warm days did not set in until March 30th, and they termi- 
nated on April 6th, when the temperature in the shade fell, 
so that the eruption preceded the temperature change by 
nine days. The state of matters at Catania I am unable to 
divine : on March 21st it was reported in the newspapers 
that great cold prevailed there, and that the sky was cloudy, 
but when the change was first felt I do not know. 
On the whole I think statistics show that the eruptions of 
Etna and Vesuvius come most frequently at the winter sol- 
stice, or about December 21st ; at the vernal equinox, or 
about March 20th ; and somewhat less frequently at the 
summer solstice, or about June 21st : there are fewer in July, 
and about the autumnal equinox, or September 22nd, they 
are again frequent. Apart from the anomaly of the paucity 
of eruptions in July, these dates will be found indicated as 
warmer periods in the several seasons. As a corollary one 
might ask, Are thermometer fluctuations greater at the 
periods of sun-spots and no sun-spots when seismic activity 
is most violent ? 
