60 Papers from the Department of Marine Biology. 



of the cup C, and shaken violently laterally while in the vertical posi- 

 tion, for 2 minutes exactly. In this way the sea-water was brought into 

 approximate equilibrium with the gas pumped out of it. The appara- 

 tus was clamped again and the lower cock turned so that the sea-water 

 passed into the trap T, the screw being adjusted so that the water 

 meniscus was brought exactly to the upper entrance to the hole in the 

 lower stopcock. One minute was allowed for the sea-water to drain 

 down; then the lower stopcock was reversed and the leveling bulb 

 raised so that the gas in the gas burette B was brought to atmospheric 

 pressure. One minute was allowed for the moisture caught between the 

 glass and the mercury to rise, and then the sea-water on top of the 

 mercury was carefully measured and recorded. It should be about 

 0.05 c.c. The leveling bulb was then raised sufficiently to make the 

 mercury meniscus rise half the height of the water column. The lower 

 stopcock was closed and the leveUng bulb brought back to the same 

 level as the mercury in B. 



The apparatus was now removed from the clamp and placed hori- 

 zontally, with the trap T upwards. By little jerks the mercury was 

 forced into B, thus displacing the gas and water into the wide part of 

 the apparatus. It was shaken in this position 1 minute, and returned 

 to the clamp, care being taken that no water worked around from T, 

 and rose into B when the lower cock was opened again. On opening 

 this lower cock, the mercury meniscus should not rise or fall more than 

 a small fraction of a millimeter. The leveling bulb was now carefully 

 adjusted with the screw so as to bring its meniscus on exactly the same 

 level as the mercury meniscus in the burette B. This is not at first an 

 easy matter, but may be assisted by placing the leveling bulb directly 

 behind the burette and placing 2 specks of dust on the mercury in it, 

 so that it is possible to bring the meniscus in B and the two specks of 

 dust in the leveling bulb in line. 



A cyhndrical bulb should be used, as a pear-shaped bulb is liable to 

 lead to error from refraction. The water meniscus is now carefully 

 read, so as to determine the total gas volume, and recorded. It is a 

 great advantage to have the graduations on B run at least half-way 

 around the tube so as to avoid parallax errors, as an error of 0.001 c.c. 

 means an error of 0.1 c.c. CO2 per liter. We used a thennometer lens 

 for reading the meniscus. About 0.5 c.c. of half-normal C02-free 

 NaOH was placed in C and allowed to run down into B a little at 

 a time while tapping the apparatus. 



Complete absorption of CO2 usually took place in about 5 minutes, 

 at 20° or 10°, but at 30° the vaseHne floated out of the stopcock grease 

 and over the surface of the alkali, and it sometimes required 30 minutes 

 for complete absorption. Half-normal NaOH has about the same 

 absorption coefficient for the elementary gases as the average sea- 

 water, but it may be possible to use a stronger solution without making 



