2 THE RECENT SUNSETS AND SUNRISES. 
hollow glassy vesicles, such as may be supposed to be 
produced by a sudden discharge of very high pressure steam 
through a layer of melted lava ; and that these vesicles, from 
their extremely small actual weight, and the relatively large 
surface that they expose to the atmosphere in comparison 
with their weight, must be many months at least before they 
can fall down through the atmosphere to the surface of the 
earth, and during that time they will be liable to be carried 
by the currents of air to great distances over the earth. 
Taking a direct line westward from the volcano (the 
direction in which the great equatorial currents of the upper 
regions of the atmosphere travel), the special atmospheric 
phenomena were observed (as illustrated in the chart, Plate I.) 
in two days after the eruption at Seychelles Island towards 
Madagascar ; in six days at the Gold Coast on the west of 
Africa; and in a week at Trinidad in the West Indies. This last 
place is at a distance of about 12,000 miles from the volcano 
(half round the earth), travelled in a week, giving an average 
rate of 70 miles an hour, which is within the rate of observed 
velocities in the upper regions of the atmosphere. The rate 
of rotation of the earth’s surface at the equator being about 
1000 miles an hour (8000 miles diameter, or 24,000 miles 
circumference travelled in 24 liours\ the rate of rotation of 
the surrounding atmosphere at different heights from the 
earth’s surface may be considered to range between the limits 
of 1000 miles an hour at the surface of the earth, and 
nothing where the relatively stationary inter-planetary atmo¬ 
sphere is reached ; and at the height of the supposed stratum 
of volcanic matter, which was estimated at as much as 40 
miles by Helmholtz from observations in the special sunsets 
seen at Berlin, the lagging behind of the atmosphere from its 
slower rotation, would give the effect upon the earth of a 
westward current ; and 70 miles an hour for such a current 
would amount to only a small retardation from the 1000 
miles an hour surface velocity (only 7 per cent). 
The slower lateral dispersion northwards and southwards 
of the stratum of matter producing the special atmospheric 
phenomena appears to follow from the circumstances of the 
successive appearances at different distances north and south, 
as illustrated in the chart, reaching Madras and Ceylon a 
fortnight after the eruption, Cape of Good Hope a month 
after, and England three months after. The special pheno¬ 
mena that have been named at present (mainly from Mr 
Norman Lockyer’s interesting communication to the Times) 
as having been observed at the several places, are as 
follows : — 
