We may cause the water in the flask to boil at a temperature of more than 100° C. (212° F.) by fitting to the free end of the platinum tube a glass tube bent at right angles, which we plunge to any depth in a deep vessel filled with mercury (Fig. 7). Whilst the water is boiling under pressure, we must separate the tube which is plunged in the mercury; the water in the flask will continue to boil at the ordinary pressure. We must then leave the flask to cool. It will gradually become filled with air that has been heated to a high temperature, more than sufficient to burn up all the organic particles of dust which that air could have contained. When the flask is cold we must close the tap and detach it, and proceed to prepare other similar flasks. It will be advisable to close the tap when the temperature of the flask is still a few degrees above that of the surrounding atmosphere: this precaution will cause the air in the cooled flask to have a lower pressure than that of the external atmosphere. During the interval which must elapse between the preparation and the use of a flask, it is a good thing to keep the free branch of the tap inclined towards the ground, to secure the inside of its tube from the deposit of external particles of dust. Whether this precaution be adopted or not, we must take care to heat this branch in the flame of a spirit lamp just before we bring our flask into requisition. If we have to study blood, we must take it from a living animal—a dog, for example. We must expose a vein or an artery of the animal, and make an incision into which the end of the free tap-branch, which has previously been heated and allowed to cool, must be introduced and fixed by a ligature in the vein or artery. On opening the tap, the blood will rush into the flask; it must then be closed, and the flask placed in an oven at a certain temperature. We have successfully accomplished these manipulations, thanks to the kind help of our illustrious colleague and friend, M. Claude Bernard. The operation is nearly the same in the case of urine. The end of the free branch of the tap is introduced into the passage of the urethra; the tap is turned at the moment when the urine is emitted, which is then allowed to pass into the flask, until it is a third or half filled. The following were the results of our experiments:—Blood underwent no putrefactive change even at the highest temperatures of the atmosphere, but retained the odour of fresh blood, or acquired the smell of lye. Contrary to what we might have expected, the direct oxidation of the constituents of blood by slow combustion was rather sluggish. After subjecting our flasks to a temperature of 25° C. or 30° C. (77° F. to 80° F.) in an oven for several weeks, we observed an absorption of not more than 2 or 3 per cent. of oxygen, which was replaced by a volume of carbonic acid gas of about an equal bulk.[29]