though often enough thrown in between the parts of a discourse, are very rarely “thrown in between the parts of a sentence.”  They more frequently occur at the beginning of a sentence than any where else; and, in such cases, they do not come under this narrow definition.  The author, at the head of his chapter on interjections, appends to this definition two other examples; both of which contradict it in like manner:  “Oh!  I have alienated my friend.”—­“Alas!  I fear for life.”  Again:  Interjections are used occasionally, in written, as well as in oral discourse; nor are they less indicative of the emotions of the writer, than of those “of the speaker.”

30.  I have thus exhibited, with all intentional fairness of criticism, the entire series of these nine primary definitions; and the reader may judge whether they sustain the praises which have been bestowed on the book,[69] or confirm the allegations which I have made against it.  He will understand that my design is, here, as well as in the body of this work, to teach grammar practically, by rectifying, so far as I may, all sorts of mistakes either in it or respecting it; to compose a book which, by a condensed exposition of such errors as are commonly found in other grammars, will at once show the need we have of a better, and be itself a fit substitute for the principal treatises which it censures.  Grammatical errors are universally considered to be small game for critics.  They must therefore be very closely grouped together, to be worth their room in this work.  Of the tens of thousands who have learned for grammar a multitude of ungrammatical definitions and rules, comparatively few will ever know what I have to say of their acquisitions.  But this I cannot help.  To the readers of the present volume it is due, that its averments should be clearly illustrated by particular examples; and it is reasonable that these should be taken from the most accredited sources, whether they do honour to their framers or not.  My argument is only made so much the stronger, as the works which furnish its proofs, are the more esteemed, the more praised, or the more overrated.

31.  Murray tells us, “There is no necessary connexion between words and ideas.”—­Octavo Gram., Vol. i, p. 139.  Though this, as I before observed, is not altogether true, he doubtless had very good reason to distinguish, in his teaching, “between the sign and the thing signified.”  Yet, in his own definitions and explanations, he frequently confounds these very things which he declares to be so widely different as not even to have a “necessary connexion.”  Errors of this kind are very common in all our English grammars.  Two instances occur in the following sentence; which also contains an error in doctrine, and is moreover obscure, or rather, in its literal sense, palpably absurd:  “To substantives belong gender, number, and case; and