826 The American Naturalist. (October, | 
mechanism to the total respiration. But this was no easy 
thing todo. In the first place the animal must be confined 
in a somewhat narrow space in order that the air and water 
which are known to have been affected by its respiration may 
be tested to show the changes produced in it by the respira- 
tory process ; in the second place the water has so great a dis- 
solving power upon carbon dioxide that even if it were 
breathed out into the air it would be liable to be absorbed by 
the water; then some means must be devised to prevent the 
escape of the gases from the water as their tension becomes 
changed ; and finally, the animal in the water must be able to 
reach the air. A diaphragm must be devised which would pre- 
vent the passage of gases between the air and water, and at 
the same time offer no hindrance to the animal in projecting 
its head above the water. As a liquid diaphragm must be 
used it occurred to me that some oil would serve the purpose; 
but the oil must be of peculiar nature; it must not allow any 
gases to pass from air to water or the reverse, it must not be in 
the least harmful or irritating to the animal under experi- 
mentation, and finally it must itself add nothing to either air 
or water. Olive oil was thought of and later the liquid par- 
affins. The latter were found practically impervious to oxy- 
gen and fulfilled all the other requirements, but unfortunately 
they absorb a considerable quantity of carbon dioxide. Pure 
olive oil was finally settled upon as furnishing the nearest 
approximation to the perfect diaphragm sought.” 
The composition of the air being known, and a careful 
determination of the dissolved gases in the water having been 
made, the animal was introduced into the jar, and the water 
covered with a layer of olive oil from ten to fifteen millimeters 
thick. The top of the jar was then vaselined, and a piece of 
plate-glass pressed down upon it, thus sealing it hermetically. 
Two tubes penetrate this plate-glass cover, one connecting with 
the overlying air chamber and the other extending into the 
water nearly to the bottom of the jar. As the water and air 
were limited in quantity the shorter the time in which the ani- 
"See Wm. Thörner on the use of olive oil for the prevention of the absorption of 
carbon dioxide. Repetorium der analytischen Chemie, 1885, pp. 15-17. 
