Rinewooo.—On Red Sunsets. 891 
in the southern hemisphere it performed two and seven-eighths (21) revolu- 
tions in 294 days, showing that the initial velocity at starting has only very 
slightly fallen off in even latitude 45° south. So in the following discus- 
sion, I will adopt a mean diurnal velocity for the dust-cloud of 2,083 miles, 
or 87 miles an hour to the westward. 
As I showed at the beginning that if the atmosphere be considered as 
part and parcel of the earth, a particle of it at a certain height will cover a 
greater distance in a certain time than that part of the earth immediately 
beneath would: so if we know the rate per hour that a certain thing 
apparently moves to the westward, or seems to lag behind the diurnal revo- 
lution, we ean ascertain the height. We know that it lags behind at the 
rate of 2,088 miles a day, which added to the circumference of the 
world gives the circumference of a circle 26,908 miles + 8:1416, gives a 
diameter of 8,565 miles, or 664 greater than that of the earth, or at a height 
of 882 miles above the surface. Or putting it this way,—we may assume 
that at the latitude of Krakatoa the earth has an hourly velocity of 1,034 
miles, and that any matter ejected thence into the upper regions of the 
atmosphere would retain the same rotary velocity as it had before, viz., 
1,084 per hour to the eastward ; but we have material under our observation 
which cannot keep its zenithal position at starting, by 87 miles per hour, 
showing it to be at an elevation of 332 miles. 
Now the spectroscope tells us that the red colour is produced through 
dust of almost ultra-microscopic fineness, and in some specimens of this dust 
that have already fallen, the microscope shows the existence of salt crystals, 
which fact in itself almost proves it to be of volcanic origin, and not 
meteoric or cosmic dust. Now Professor Helmholtz states that ‘the 
reflecting medium, whatever it was, over Berlin on the last three nights of 
November, was about 40 miles above the earth ;" and if we work on these 
data, we have a circle whose diameter is 80 miles greater that of the earth, 
or a circle of 7,981 miles, which x 3°1416 gives a circumference of 25,078, 
or 248 miles more than that of the earth, which divided by 24 shows an 
excess of about 104 miles per hour above the surface velocity of rotation. 
But we want to account for an excess of 87 miles per hour, so if we accept 
Professor Helmholtz's statement, we must only suppose that at the altitude 
of 40 miles there is an easterly current, or one moving to the westward, of 
77 miles per hour. For, assuming as we do from the foregoing tables and 
caleulations that the earth rolls from under the cloud at the rate of 87 miles 
per hour, and that, unless we admit of an easterly current, we cannot stop 
short of that enormous height of 332 miles, unless we suppose that the 
power of gravitation has only a feeble hold on those most minute dust 
particles at the altitude of 40 miles, where the air has not the many 
thousandth part of the density it has on the surface of the globe, 
