Method for Determining the Gases Dissolved in Water. 549 
that the water can be examined immediately after collection, I 
proceed as follows :—The flask is filled by passing a tube to the 
bottom and allowing a current of the water to flow through the 
flask until the contents have been several times renewed, or by 
immersing the flask in the water and sucking out the air with 
the usual precautions. The cock is then turned off and the water 
in the bulb and neck emptied out. The flask is now partly 
immersed in water, the temperature of which is a few degrees 
lower than that in the flask, the condenser communicating with 
the Sprengel pump is attached to the flask and the pump is set 
in action. The connexion between the flask and condenser may 
be made perfectly air-tight in a moment by smearing the conical 
end of the tube that fits loosely into the neck of the flask with 
a cement made by melting together three parts of beeswax 
and eight of resin; the tube being placed in position the neck of 
the flask is warmed with a fragment of glowing charcoal or a 
hot wire until the cement melts. Too much heat must not be 
employed, and it must be applied only where the cement is 
intended to adhere or it will spread and make a mess, Joints 
made in this way are perfectly air-tight, and answer just as well 
as if the tubes were fused together, while they have one great 
advantage—the tubes can be separated in a moment by gently 
warming the joint; of course they are not applicable if it is 
intended to heat the apparatus beyond the softening point of 
the cement (about 45°C.) While the apparatus is being exhausted 
cold water is allowed to circulate slowly through the condenser, 
and as soon as the metallic click of the falling mercury indicates 
that all the air is removed, an inverted tube filled with mercury 
is placed over the opening of the fall tube of the pump to receive 
the gases. The cock of the flask is now turned on enough to 
allow the gases to escape slowly. If the operator desires to expel 
the gases at the lowest possible temperature, the pump should be 
kept working continuously, if, on the contrary, the pump is not 
kept working the gases will escape into an atmosphere of 
gradually increasing tension, and considerable elevation of 
temperature must be resorted to before the liquid parts with the 
last traces of dissolved gas. Supposing the object in view is 
simply the collection of all the dissolved gas, it is obviously most 
