An apparatus was devised with the primary object of determining the atmospheric nitrogen and oxygen dissolved in sea water from on board ship in fine weather. It was found that the apparatus could be used at sea in fine weather, but owing to the difficulty of manipulation when any sea was running there was a possibility of error which necessitated duplicate determinations being made, each of which took about twenty minutes. In the Laboratory, on the other hand, the apparatus gives rapid and reasonably accurate results, and may be of use for the estimation of oxygen or other dissolved gases in fluids which contain organic matter, rendering Winkler's method of oxygen determination impossible, and where there is an insufficient quantity of fluid and time to carry out extraction of the gases by means of a mercury pump.

The apparatus was made as shown diagramatically in Fig. I, consisting of a glass bulb filled with mercury, with the necessary means of evacuating it and trapping any air leak which may occur in the process. It is mounted in a weighted teak box with hinged door and can be let down by a line to any desired depth in the sea, the tap t opened by a messenger and the charge of sea water (4.59 c.c. in this particular apparatus) drawn in, as shown in Fig. II. The apparatus is then drawn up, tap t and screw pinch cock s closed, and the bulb evacuated by lowering the mercury reservoir and opening tap q (Fig. III).