Truth and error, like all such mutually antagonistic concepts, have only an absolute reality under very limited conditions, as we have seen, and as even Herr Duehring should know by a slight acquaintance with the first elements of dialectics, which show the insufficiency of all polar antagonisms. As soon as we bring the antagonism of truth and error out of this limited field it becomes relative and is not serviceable for new scientific statements. If we should seek to establish its reality beyond those limits we are at once confronted by a dilemma, both poles of the antagonism come into conflict with their opposite; truth becomes error and error becomes truth. Let us take, for example, the well-known Boyle's law, according to which, the temperature remaining the same, the volume of the gas varies as the pressure to which it is subjected. Regnault discovered that this law does not apply in certain cases. If he had been a realist-philosopher he would have been obliged to say, "Boyle's law is mutable, therefore it does not possess absolute truth, therefore it is untrue, therefore it is false." He would thus have made a greater error than that which was latent in Boyle's law, his little particle of truth would have been drowned in a flood of error; he would in this way have elaborated his correct result into an error compared with which Boyle's law with its particle of error fastened to it would have appeared as the truth. Regnault, scientist as he was, did not trouble himself with such childish performances. He investigated further and found that Boyle's law is only approximately correct, having no validity in the case of gases which can be made liquid by pressure when the pressure approaches the point where liquefaction sets in. Boyle's law therefore is shown only to be true within specific bounds. But is it absolute, a final truth of last instance within specific bounds? No physicist would say so. He would say that it is correct for certain gases and within certain limits of pressure and temperature, and even then within these somewhat narrow limits he would not exclude the possibility of a still narrower limitation or change in application as the result of further investigation. This is how final truths of last instance stand in physics, for example. Really scientific works as a rule avoid such dogmatic expressions as truth and error, but they are constantly cropping up in works like the Philosophy of Reality, where mere loose talking vaunts itself the supreme result of sovereign thought.