60 
only series of determinations hitherto published of the 
amount of carbonic acid contained in the air above the ocean 
are those of Lewy. At the request of the French Academy, 
Lewy, in 1848, collected the air of the Atlantic at different 
times during a voyage from Havre to Santa Marta, and 
determined the proportion of its three principal constituents 
in the eudiometric apparatus of Regnault and Reiset. The 
mean composition of the air of the Atlantic from Lewy’s 
analyses is in 10,000 volumes, 
Carbonic Acid 4*630 
Oxygen 21 02 *750 
Nitrogen 7892*620 
10000*000 
These numbers differ but slightly from those usually given 
as representing the average composition of the atmosphere 
of the land, but on examining the experimental details of 
Lewy’s analyses it appears that the air of the day was found 
to be considerably richer in carbonic acid and oxygen than 
the air of the night. The following are the means of each 
series : — • 
The Day (7 Expts). The Night (4 Expts). 
Corbonic Acid 5*299 3*459 
Oxygen 2105*801 2097*412 
This remarkable difference appeared to become greater as 
the middle of the ocean was approached, where the air of 
the sea was less liable to be mixed with that of the land. 
These singular diurnal variations in the composition of the 
atmosphere of the sea are ascribed by Lewy to the evolution, 
and consequent admixture with the superincumbent air, 
during the day time, of dissolved gases from the heated 
surface-layers of the sea, such gases being considerably richer, 
as is well known, in oxygen and carbonic acid than ordinary 
air. During the night, on the other hand, the disengagement 
of gas is supposed to be arrested. Morren, and subsequently 
