In the second place, an artery, the crural or carotid for instance, must be opened with the view of observing the varieties of colour in the blood projected from it. A small artery should not be chosen. From such a one the course of the blood would be suspended by the slightest accident; and on the other hand, the larger arteries expend in a little time too large a quantity of blood; this inconvenience may be remedied, by adapting to these vessels a tube of a small diameter, or a stop-cock. All things being thus prepared, on a dog, for instance, let us see what are the phenomena which take place, when the colour of the blood is altered. In my indication of these, however, I shall speak only of what I have seen, and by no means pretend that in man their duration should be similar or uniform, or even that in animals of the same species, under the different circumstances of sleep, digestion, exercise, and passion, &c. if it were possible in such way to repeat them, they should be alike. The instability of the animal functions, as I have said, is extreme; they cannot be submitted to calculation; they remain indeed the same, but their variations as to plus or minus are innumerable. Let us now return to our subject: 1st. If the cock of the pipe be shut immediately after the animal has inspired, the blood begins to be altered in colour at the end of about thirty seconds.--At the end of a minute its colour is dark; at the end of a minute and half or two minutes, it is perfectly similar to venous blood. 2dly. If the cock of the pipe be shut immediately after the animal has expired strongly, the blood receives its tinge of black some seconds the sooner. 3dly. If the air of the lungs be pumped out entirely with a syringe, the blood will suddenly pass from red to black.[94] In such case it appears that the artery immediately throws out a black stream, after it has expelled the red blood which it previously contained. There is no gradation. The blood is expelled by the arteries, such as it is in the veins. 4thly. If, instead of making a vacuum in the lungs, we inflate the air cells to the full, the blood is a longer time in becoming black, a minute at least, and is not completely black before the end of three minutes.--This will vary according to the quantity of air injected. From all these experiments it follows, 1st, That the length of the interval, during which the blood retains its red colour, is in direct proportion to the quantity of air contained in the lungs; 2dly, That as long as there remains any quantity however small of respirable air in the cells of the lungs, the blood will preserve more or less of its crimson colour; 3dly, That this colour diminishes in proportion as the respirable air diminishes; and 4thly, That the blood is exactly similar to that of the veins, as soon as the whole of the vital air in the extremities of the bronchiæ has been exhausted.