were comprehended originally many more relations, and that prepositions are used only to define the meaning of them with greater precision. These instances are sufficient to show the sort of errors which grammar introduces into language. We are not considering the question of its utility to the beginner in the study. Even to him the best grammar is the shortest and that in which he will have least to unlearn. It may be said that the explanations here referred to are already out of date, and that the study of Greek grammar has received a new character from comparative philology. This is true; but it is also true that the traditional grammar has still a great hold on the mind of the student. Metaphysics are even more troublesome than the figments of grammar, because they wear the appearance of philosophy and there is no test to which they can be subjected. They are useful in so far as they give us an insight into the history of the human mind and the modes of thought which have existed in former ages; or in so far as they furnish wider conceptions of the different branches of knowledge and of their relation to one another. But they are worse than useless when they outrun experience and abstract the mind from the observation of facts, only to envelope it in a mist of words. Some philologers, like Schleicher, have been greatly influenced by the philosophy of Hegel; nearly all of them to a certain extent have fallen under the dominion of physical science. Even Kant himself thought that the first principles of philosophy could be elicited from the analysis of the proposition, in this respect falling short of Plato. Westphal holds that there are three stages of language: (1) in which things were characterized independently, (2) in which they were regarded in relation to human thought, and (3) in relation to one another. But are not such distinctions an anachronism? for they imply a growth of abstract ideas which never existed in early times. Language cannot be explained by Metaphysics; for it is prior to them and much more nearly allied to sense. It is not likely that the meaning of the cases is ultimately resolvable into relations of space and time. Nor can we suppose the conception of cause and effect or of the finite and infinite or of the same and other to be latent in language at a time when in their abstract form they had never entered into the mind of man...If the science of Comparative Philology had possessed “enough of Metaphysics to get rid of Metaphysics,” it would have made far greater progress.