36 
Day (mean of 7 expts.) Night (mean of 4 expts.) 
Carbonic acid 5”299 3"459 
Oxygen 2105-801 2097-412 
and this variation appeared to increase in proportion as the 
middle of the ocean was approached. 
This remarkable phenomenon, of the variation in composi- 
tion of the air above the tropical oceans, may doubtless be 
accounted for, without any reference to the direct action of 
infusoria, by the heating effect of the sun on the sea water 
and the consequent disengagement during the day of gas pro- 
portionately rich in carbonic acid and oxygen. During the 
night, on the other hand, as this source of action is removed, 
the disengagement may be assumed not to occur ; and, follow- 
ing Lewy, one may perceive that this difference would be- 
come more appreciable and easier to trace in air at great 
distances from any continent than in air collected nearer the 
coasts, and consequently, liable to be mixed with the air of 
the land. 
Although the precision of these results is certainly remark- 
able, they still require confirmation. The air was collected 
in glass tubes of about 100 cbc., and analysed eighteen or 
twenty months after in the eudiometric apparatus of Reg- 
nault and Reiset. The fact pointed out by Regnault that 
air which has remained for any great length of time in glass 
tubes invariably exhibits notable diminutions in the amount 
of carbonic acid, since the glass absorbs a portion of this gas; 
and the difficulty generally experienced in accurately noting 
contractions so minute as the absorption of the carbonic acid 
from a small volume of atmospheric air, are circumstances 
which may possibly influence the reliability of the results. 
The kind permission of the Honorable Board of Trinity 
House has enabled me during the vacation of last summer to 
make some additional experiments in this direction on board 
the “ Bahama Bank ” Light-vessel, situated in the Irish sea, 
latitude 54° 21' and longitude 4° IT, seven miles W.N.W. of 
