Fe in ada 
pee rere 3 < me 
Parr I. Seot.i.§2.] VOLCANIC ACTION. 211 
escape of vapours from the upper surface of the lava column of a 
volcano; for it is not so much to the lava itself as to the expansive 
vapours impregnating it that the manifestations of volcanic activity 
are due. Among the Vesuvian eruptions since the middle of the 
17th century, the number which took place in winter and spring was 
to that of those which broke out in summer and autumn as 7 to 4. 
But there may be other causes besides atmospheric pressure concerned 
in these differences; the preponderance of rain during the winter 
and spring may be one of these. According to Mr. Coan, previous to 
the great Hawaian eruption of 1868 there had been unusually wet 
weather, and to this fact he attributes the exceptional severity of the 
earthquakes and volcanic explosions. But at most the effects of 
varying atmospheric pressure can only slightly modify volcanic 
activity. Eruptions like the great one of Cotopaxi in 1877 have in 
innumerable instances taken place without, so far as can be ascertained, 
any reference to atmospheric conditions. 
Kluge has sought to trace a connexion between the years of 
maximum and minimum sun-spots and those of greatest and feeblest 
voleanic activity, and has constructed lists to show that years which 
have been specially characterized by terrestrial eruptions have — 
coincided with those marked by few sun-spots and diminished 
magnetic disturbance.. Such a connexion cannot be regarded as 
having yet been satisfactorily established. Again, the same author 
_. has called attention to the frequency and vigour of volcanic 

explosions at or near the time of the August meteoric shower. But 
in this case, likewise, the cited examples can hardly yet be looked 
upon as more than coincidences. : 
Occasional Periodicity of Eruptions.—The case of Kilauea, in 
Hawaii, seems to show a regular system of eruptive periods. Dana 
has pointed out that outbreaks of lava have taken place from that 
volcano at intervals of from eight to nine years, this being the time 
required to fill the crater up to the point of outbreak, or to a depth 
of 400 or 500 feet. But the great eruption of 1868 did not occur 
until after an interval of 18 years. The same author suggests that. 
the missing eruption may have been submarine.? 
General sequence of Events in an Eruption —The approach of 
an eruption is not always indicated by any premonitory symptoms, 
for many tremendous explosions are recorded to have taken place 
in different parts of the world without perceptible warning. Much 
in this respect would appear to depend upon the condition of 
liquidity of the lava, and the amount of resistance offered by it to 
the passage of the escaping vapours through its mass. In Hawaii, 
where the lavas are remarkably liquid, vast out-pourings of them 
have taken place quietly without earthquakes during the present 
century. But even there the great eruption of 1868 was accom- 
panied by tremendous earthquakes. | 
1 Ueber Synchronismus und Antagonismus, p. 72. 
? On the periodicity of eruptions, see Kluge, Newes Fahrb. 1862, p. 582. 
P 
5 
2 
