To take an excellent example of the services which experimental physiology can render not to science only--which would, indeed, be quite sufficient to justify them--but to humanity, I will cite the experiments of Paul Bert with relation to elevated atmospheric pressures. There are certain workmen who are obliged to work under water, at a depth of 20 to 30 yards, for the construction of piers and bridges, or the exploration of sunken vessels. Now, it had long been observed that some of these men died suddenly on returning to the surface. Experimental physiology was able to discover the cause of that sudden death. When a man (or an animal), after having been subjected to several times the normal atmospheric pressure, is suddenly released from this pressure, the nitrogen dissolved in the blood is disengaged suddenly: this produces gaseous embolism, that is to say, bubbles of gas are formed, which block the blood-vessels and prevent the blood circulating in the capillaries. Knowing this, the death of men working at a pressure of four atmospheres could then be avoided by releasing them slowly, that is by bringing them slowly back to the normal atmospheric pressure. Is it barbarous to attach more importance to the death of these men than to the death of the few dogs and mice that served to establish this law? I was able to demonstrate that, if the temperature of the air is very high, as in the hottest days of summer, dogs that are muzzled die rapidly of hyperpyrexia (_i.e._ high fever), for they are no longer able to cool themselves by panting. It is true that this experiment cost the lives of a few dogs, but has it not saved many others by pointing out that dogs should not be muzzled under certain conditions? It goes without saying I am not speaking of the theoretical consequences of this experiment. Artificial respiration, which can restore to life the apparently drowned, is one, of the conquests of experimental physiology; for we have been able to determine the best method and the essential conditions (for artificial breathing) by experiments of a very precise nature. Is it nothing to know how to restore to life the apparently drowned?