error, had he not himself, in the first volume of his work, most clearly pointed out the true idiom, and guarded others against the very mistake into which he has himself fallen. We will quote his own words :

In clauses introduced by if, or as if, implying a negation of the proposition expressed, present time, in English, is denoted in the conditional clause by the preterite tense, and past time by the pluperfect. When I say, "If I have the book, I will send it," the meaning is clearly dubitative; and the expression implies, that I am uncertain whether I have it or not. When I say " If I had the book, I would send it," the meaning is, that I have it not; and the conditional clause, here equivalent to a negation, is expressed in the preterite tense, though the same thing be implied, as in the preceding sentence. If past time is to be denoted, I say, "If I had had the book, I would have sent it." Here the pluperfect is employed. Thus also with " as if."-He fights as if he "contended," or were contending, for his life." Present time is signified, and the two actions are evidently contemporary; yet the former verb is in the present, and the latter in the preterite tense. "He fought as if he had contended," or " had been contending, for his life." Here also the two actions are contemporaneous, and past time is implied; yet the former verb is in the preterite sense, and the latter in the pluperfect. This is not the case in Latin. The first of the two sentences would be rendered thus, Pugnat quasi pro vita contendat. The actions, being contemporary, are each expressed in the same tense; and time present being meant, the verbs are put in the present tense. The second sentence would be thus rendered, Pugnavit quasi contenderet. Here also the actions are represented as contemporaneous and past, and the verbs are each in the preterite When the actions are not contemporary, the prior is expressed in the preterite tense, if the other be expressed in the present; and if both actions be past, the subsequent action is expressed in the preterite, and the one preceding it in the pluperfect." -vol. i. p. 52.

tense.

To the above admirable exposition of the Latin and English idioms, we will add, that the same remarks are applicable to many interrogative sentences. Thus, vellesne,' Would you have wished?' 'Quid facerem,' What ought I to have done?' But quid faciam' (subj.) What shall I do?' The passage from Horace, referred to in page 340 of vol. ii., is another specimen, in addition to those already given, where Dr. Crombie has neglected his own principles. Stated, indeed, as he there has it-' Dispeream ni summosses omnes'it seems to include two tenses inconsistent with each other: but, in fact, the tense of submosses is not dependent upon dispeream ni. Even supposing those words to be omitted, submosses is still in the tense and mood required. The whole sentence, as it stands in Horace, the reader will recollect to be,

Haberes

Magnum adjutorem, posset qui ferre secundas,
Hunc hominem velles si tradere: dispeream ni

Submosses omnes'

6

which may be rendered, You would have had a powerful friend to have seconded you, if you had been so kind as to have introduced your humble servant to the family; as I hope to live, you would, ere this, have had the field clear of all your rivals.' By pointing out the advantages that would have arisen from such a course, he gives Horace a strong hint to take such a step now. According to the ordinary translation, the request is more direct; but this is altogether inconsistent with the pluperfect submosses. Again, the very first sentence in the translation from Livy, which constitutes the exercise upon the potential mood (p. 342), we find, No plebeian would offer violence to the daughter of a patrician; this libidinous exploit belongs exclusively to the patricians themselves'-given as the English for-Nemo plebeius patricia virgini vim afferret; patriciorum ista libido est.' Where the ista, by the bye, is neglected in the English, it ought to have been, exclusively to you, patricians.' See also the exercise, p. 164, vol. i., in the translation of sequerentur and actum foret. There is another passage, in reference to si, in p. 359, vol. ii., which seems to us to lay down a wrong principle; we mean the assertion, that si, when used hypothetically, implying merely a supposition, and not a fact, though generally joined with the subjunctive mood, is frequently found with the indicative.' We should have thought the two words, subjunctive and indicative, had changed places through an error of the press, but that we find our author following up his precept by practice, and translating, if you do not believe me,' by si mihi non credatis.' See the Latin translation of the first exercise.

In p. 65, vol. i., there is another statement, which must have escaped the author's attention. But if we say "quod doceam," "because I teach," the verb is under the government of the conjunction quod; and were it not for this conjunction, the English being indicative, the verb would be put in the indicative mood. In this example, therefore, the verb is strictly in the subjunctive mood, this form being used, not because the sense requires it, for the English is indicative, but because it is subjoined to the conjunction quod."

After what Dr. Crombie has written upon the inconsistency of the English tenses in contingent sentences, and the decided superiority of the Latin in the same, it is not very consistent in him to test the accuracy of a Latin phrase by a comparison

with the English, and to say that the sense does not require the subjunctive mood, for the English is indicative.' But this is not our present question. We wish to know the authority for the assertion that quod governs the subjunctive mood. Cæsar and Cicero at any rate use it with the indicative, of course excepting those cases where it occurs in the obliqua oratio, or, what is nearly the same thing, where it expresses a reason alleged or felt by another. But Dr. Crombie is himself fully aware that in these cases the subjunctive mood inserts itself, no matter what conjunction be used, and that the mood therefore cannot be attributed to the conjunction itself. But again, we may correct the author by himself. In page 171 he quotes: Minus curo quod operarios ejecisti-quod bene vales gaudeo.' In the same page, indeed, he also produces two cases of quod with the perfect subjunctive; and affixes the authority of the three letters Cic. to them. It is an inconvenient practice to omit the precise reference, especially in works of criticism. The last passage there given, 'Nihil est quod succenseas,' should be referred to the use of the relative.

In page 70, Dr. Crombie enters upon a discussion whether what is called in some grammars the future subjunctive, really belong to the subjunctive or indicative mood. The question seems to us one of no great importance, and we shall be satisfied with the fact, that, with the exception of the first person, it is impossible to distinguish the so-called future from the perfect subjunctive, for the difference of quantity is utterly imaginary. The fact is, that all the tenses of the subiunctive mood, not excepting even the imperfect, have in certain constructions a future meaning; and it is, therefore, not very surprising that faciam should be at once a simple future indicative, and a present subjunctive, or that fecerit should also be common to the two moods. We will propose a sentence to Dr. Crombie for his opinion. The following, though not strictly from any Latin author, he will perhaps allow to be in correct idiom: Nunc reus est apud Crassum Divitem Vettius de vi; et quum erit damnatus, est indicium postulaturus: quod si impetrarit, judicia fore videntur.' If we understand Dr. Crombie correctly, he will contend that in this passage impetrarit is necessarily a perfect future indicative, and that it has nothing to do with the subjunctive mood. Let us now introduce that singular arrangement of tenses, which the Romans used in their epistolary writing, so as to adapt the matter to the period when the letter is read. The exact words of Cicero, without any real change of meaning, are: Nunc reus erat apud Crassum

Divitem Vettius de vi: et quum esset damnatus, erat indicium postulaturus: quod si impetrasset, judicia fore videbantur.' Here damnatus esset and impetrasset appear decidedly to have an indicative power, and yet no one will deny that they are also connected with the subjunctive mood.

In the otherwise admirable chapter upon the obliqua oratio, vol. ii. p. 250, there seems to be some inaccuracy from an inattention to the two distinct forms into which that mode of expression divides itself-the present and the past oblique. The distinction is so well marked at times in Cæsar's Gallic War, and the sudden transition from the past oblique to the present adds so much to the liveliness of the speech, that it is well worth while to attend to the difference. Thus in the forty-fourth chapter of the first book, part of which is quoted by Dr. Crombie, the speech begins in the past oblique-Multa praedicavit: Transisse, &c., which is continued until the threat is naturally expressed in the more lively language of the present-Si iterum experiri velint, iterum paratum sese decertare-and this time is continued to the paragraph ending in defenderit. From se prius to usos esse, we have again the past. At Debere once more the present recurs, but only again to give place to the past, which runs through the conclusion of the speech from Id se ab ipsis to confecturum. It is of course only in the subjunctive moods of the Latin that the distinction can be marked; but our own language has throughout a set of tenses proper to each. Thus the phrase -Non minus libenter sese recusaturum populi Romani amicitiam quam adpetierit'-which is translated, he would refuse the friendship of the Roman people, no less willingly than he had courted it,-would be more correctly rendered by 'he will refuse,' and 'he once courted it.' To justify the other translation, we ought to have had adpetiisset, and in the earlier part of the sentence remitteretur, together with subtraherentur, for the corresponding presents, as Cæsar gives them.

Another point which seems to us worthy of notice, is the habit of explaining the use of different cases by supposing certain prepositions to be understood. There seems to be altogether a want of philosophy in this; for what after all are the cases of a noun but compounds, of which one element marks the simple meaning of the noun, and the other expresses its relation to the sentence, or part of the sentence in which it appears? The m attached to the end of Roma, in order to express what is barbarously called the accusative case, is in substance and energy itself a preposition, though its position after the noun will not allow us to use the name. Had it been called a post-position, it would perhaps have

met with more respect. Be this as it may, the final m really corresponds to our word to, which we happen to place before our noun, and Roma-m means, without any aid of words understood, to Rome. Similar reasoning will of course apply to the other cases, and there is not only no occasion, but it is worse than useless to explain Venit hora tertia by supplying in-Mansit paucos dies by means of per-urbe capta by ab -eo ita loquente by in-Multo labore by cum-Die quarto by in-all of which pseudo-explanations occur in different parts of the Gymnasium. Another plan is to appeal to certain mystical words, which appear like magicians in the grammatical world, changing the whole into a kind of fairyland. At one moment by the wand of a paragoge, ego is metamorphosed into egomet. Then by command of an Atticism, a vocative macte is called upon to play the part of a nominative. At another time epenthesis wields the magic sceptre. But to be serious, can any one really believe that the use of five or six hard words, borrowed if possible from the Greek tongue, can afford any just explanation of grammatical difficulties? In the case of macte, we see no other alteration than what has taken place in the pronouns ille, iste, ipse. The final s of the nominative being omitted in pronunciation, was dropped in writing, and ipsus, istus, illus, losing their terminal consonant, the short vowel was as usual represented by that convenient and indistinct little sound, a short e. That the vocative has grown in the same way out of the nominative we will admit, but not the converse. The loss of the final s, which is the true characteristic of the nominative, is not confined to these words. Nearly all the nouns of the first declension, and many of the third, have undergone the same change. In Greek, too, the same variation may be observed, especially in Homer, as for instance the VEQEλnyɛpéta Zɛús. But it will, perhaps, be better not to venture into this quarter, or else some objector will overwhelm us with his explanations of Boeotice, Poetice, Aeolice, &c.-a new class of monsters from which Latin criticism is fortunately free.

Not unlike the use of the hard Greek terms above is the frequent introduction of the expression 'elegantly.' Thus 'the ablative absolute is elegantly turned into another case.'Cum and dum are elegantly omitted.'- Ne is elegantly used for ut non.'-' Nequis elegantly for ut nemo.'-Ni for si non'—and lastly, (but not because we could not carry the list farther,) the pluperfect subjunctive is sometimes used elegantly in oblique sentences.' From expressions of this kind, a student will naturally infer that the non-elegant