When Murray's Grammar was first placed in my hands, I found in it certain references to the Canons of Language in the larger edition. I questioned my teacher as to what it meant. 'It is a trick of the printer,' he answered, 'to induce you to buy the larger volume.' I do not believe this now. I believe that it was a necessary reference. An author who has written upon a given subject, naturally finds his own ideas coincident illustrations of his views, and honestly refers to them. In this book I have made a few references to previous works of mine, and it has struck me that nine ont of ten of the readers will set this down to artifice or egotism. Yet it is neither. I have referred only to avoid the full quotation of some necessary illustration of the argument. Yet few will penetrate to the fact, and most will be apt to infer a trick from appearances. CHAPTER VI. SCIENCE Whatever we know must be in the number of the primitive data, or of the conclusions which can be drawn therefrom.-- J.S. Mill To have reached, in the study of observed phenomena, the point of perception indicated in this motto, and to feel the full force of the remark, is to have imbibed the spirit of science---whose traits are dear distinctions, accurate classification, and strict reference to primitive data. The bases of all science are methodical facts. The first step to the perfection and enlargement of a science is the resolution of its propositions into axioms, and into propositions which are to be proved. Dr. Reid observes--'This has been done in mathematics from the beginning, and has tended greatly to the emolument of that science. It has lately been done in natural philosophy, and by this means that science has advanced more in 160 years than it had done before in 2,000. Every science is in an unformed state until its first principles are ascertained; after this it advances regularly, and secures the ground it has gained.' Classification is one of the first steps to Science. The maxim in government, divide and conquer, retains, when applied to science, all its wisdom without its machiavelialism. The young grammarian reduces the mass of words, that so threaten to confound his powers, to a few natural classes, and he conquers them separately with ease. 'The single power by which we discover resemblance or relation in general, is a sufficient aid to us in the perplexity and confusion of our first attempts at arrangement. It begins by converting thousands, and more than thousands, into one; and, reducing in the same manner the numbers tiros formed, it arrives at last at the few distinctive characters of those great comprehensive tribes, on which it ceases to operate, 'because there is nothing left to oppress the memory or the understanding.'* * Brown's Moral Philosophy, Lect, xvi.