fi4 Sir William Siemens on the 
we shall be safe in estimating its mean density as equal to that 
of nitrogen, or equal to 14; and assuming the metallic sea 
below to consist of both the heavy and light metallic vapours 
in reasonable proportions, we shall probably be near the truth 
in assuming its mean vapour-density as being equal to that of 
iron, or equal to 56. The photospheric current would there- 
fore sweep over an ocean only four times denser than itself, 
which, under terrestrial conditions, would give rise to very 
active mechanical admixture; but this tendency is counter- 
acted upon the solar surface by a force of gravity 27 times 
greater, exercising a separating influence analogous to that 
upon which the action of the ingenious cream-separator de- 
pends, in which the force of gravitation is replaced by cen- 
trifugal action. To get the parallel of terrestrial conditions, 
we should have to assume the solar ocean to be 4 x 27 = 108 
times denser than the atmospheric current sweeping over 
it. The light metallic vapours will be held in suspension in 
the vast metallic ocean in the same way as oxygen and car- 
bonic anhydride are retained in sea- water. 
Mr. Cook thinks that the preponderance of hydrogen and 
CO in meteorites is inconsistent with the idea that the me- 
teoric gases have been absorbed in space; and he puts forward 
a theory making them the result of decomposition of solid 
hydrocarbon and oxides by heat while passing through our 
atmosphere perhaps more than once. A shell of fused mag- 
netic oxide of iron is supposed to be formed, retaining these 
gases bottled as it were ; but it may fairly be objected that 
such a fused shell, if it were really formed, would crack in 
cooling and allow the gas to escape through innumerable 
fissures. It is sufficient for us to know, I think, that the me- 
teorites contain all the constituents of our atmosphere except 
hydrogen, which latter occurs in our atmosphere combined 
with oxygen, whereas in space and in the solar photosphere 
we have evidence of its separate existence or in combination 
with carbon — a difference which is to my mind the necessary 
consequence of dissociation of carbonic acid and aqueous 
vapour in space, and of ignition and slow combustion, effecting 
their oxidation upon our earth. 
I cannot agree with Mr. Cook in supposing that a terrestrial 
polar inflow current would be inconsistent with the direction 
of the " return trade " winds, of which it would simply form 
part, the direction of both being necessarily from the north- 
east, in the northern hemisphere. 
Mr. Cook has evidently not followed me in my description 
of the equatorial outflow as a balanced current depending 
upon solar gravitation only for its density and rate of flow ; 
