Let us now turn to another point of view. The laws that direct observation gives us are never more than results. Take for example the law of Mariotte. For most physicists, this is only a consequence of the kinetic theory of gases; the gaseous molecules are animated with considerable velocities, they describe complicated trajectories whose exact equation could be written if we knew according to what laws they attract or repel each other. By reasoning on these trajectories according to the rules of calculating probabilities, we succeed in demonstrating that the density of a gas is proportional to its pressure.
The laws governing the observable bodies are therefore only consequences of the molecular laws.
Their simplicity would only be apparent and hide an extremely complex reality since complexity would be measured by the very number of molecules. But it is precisely because this number is so great that differences of detail would be mutually compensating and that we would believe in harmony.
And the molecules themselves are perhaps worlds; their laws are perhaps only results, and to find the reason, it would be necessary to go down to the molecules of molecules, without knowing where we will end up.
The observable laws then depend on two things, the molecular laws and the arrangement of the molecules. It is the molecular laws that enjoy immutability since they are the true laws and the others are only appearances. But the arrangement of the molecules can change and with it the observable laws. And that would be a reason to believe in changing laws.