The corpuscular theory is, on the contrary, an example of a good hypothesis in physics, where it plays a part analogous to that of the inertia of bodies in mechanics.[139] The innermost structure of bodies is unknown to us. But we have a right to introduce all the hypotheses which can help us in our research, and in particular the hypothesis of atoms, so long as we do not understand it as something representing a reality. The ethers and the fluids tend to “explain” the physical phenomena by the nature of the agent which produces them. It is here that these hypotheses bear the mark of the metaphysical spirit. To understand the appearance and especially the persistence of these hypotheses, it is not enough to consider them in themselves. We must get back to the history of physics, and compare it with that of the other fundamental sciences. Was it possible for physics to pass suddenly from the period in which phenomena are referred to causes and essences, to the positive period where they are conceived as simply subject to laws? A period of transition was necessary. The scholastic entities, before disappearing, became semi-materialised. They were transformed into fluids. What is heat conceived as existing apart from a hot body, light independent of a luminous body, electricity separated from an electric body? They are the old entities in a new garment, more easily grasped, in spite of their “equivocal corporeity.” They gradually lead to the more and more exclusive consideration of phenomena and of laws, until, in their turn, they disappear. Astronomy went through the same phases before Physics. In it we have also seen hypotheses which cannot be verified come to facilitate the transition from the theological to the positive state. Such was the conception of Descartes who explained the celestial motions by the system of vortices. Those famous vortices introduced the idea of a mechanism where Kepler himself had only dared to conceive the incomprehensible action of souls and genii. Then Newton came, who preserved the idea of mechanism, while giving up the vortices. In vain did the Cartesians fight against his entirely positive conception. Their arguments in favour of fluids and ethers were as plausible as those of the physicists of our own time. But we have ceased to listen to them. Having become entirely positive, astronomy no longer seeks anything but the laws at work in the phenomena observed. Every accessory hypothesis aiming at anything else has no further interest for us.