A third case may be presented--that in which the weight of the body is equal to the thrust of the water. Weight and thrust are then in mutual equilibrium. No force invites the body either to descend or to rise, and it remains balanced in the midst of the liquid, wherever it happens to have been placed. This state of indifferent equilibrium is, however, possible only if the weight of the body remains rigorously constant. The slightest augmentation of the weight immediately causes the body to descend, while the slightest diminution sends it up. From this source arise the difficulties that are met in the construction of submarine boats, when their ascent or descent is obtained by means of air chambers, which are filled with water or emptied of it according to the requirements. The equilibrium of these engines is always precarious, and this explains why none of them, from that of Van Drebbel in 1620 to the experiments of Goubet in 1895, have given really practical results in the matter of stability of immersion. When Galileo, following Aristotle, had demonstrated the ponderability of the air, and Torricelli had proved that atmospheric pressure was a result of that property, it was immediately thought that the principle discovered by Archimedes might be extended to the air, and Otto von Guericke gave an experimental demonstration of it by the invention of the baroscope. From this period it seems, then, that the discovery of aëronautics was possible. If the weight of the volume of air displaced is greater than that of the body, the latter should take an ascensional movement in the atmosphere, as a cork does when plunged into water; and it is evident that for a body to satisfy such conditions we have only to fill a very light envelope with a gas less dense than the ambient air. But the study of gases was still in its infancy in the seventeenth century, and it required the labors of Mortrel d'Élement and Hales, at the beginning of the following century, to teach physicists how to collect and retain them.