always spelt vertue or vertu and vertuous, as was common in Milton's time; but virtue does occur (e.g. P. R., II. 431, 455), and virtual, which occurs but twice, is both times so. He has the odd form aries at least once for arise, though the latter is normal with him. The word rime, though spelt so in the prose preface to Par. Lost, is spelt once rhyme in the poetry (Lyc. 11), and rhime in the only other case where it occurs there (P. L., I. 16). We have die and di'd for "colour" and "coloured;" but, to make up, there is a slip once or twice into dye and dy'd for our verb of mortality. Plowman occurs twice, and each time in that form; but in the MS. of the Sonnet to Cromwell we have plough'd. Bough and boughs or boughes are normal in the text, but once at least there is bowes. The adjective foul, for " unclean," which is a frequent word, occurs in Par. Lost first as fowl (1. 33), and the very next time (1. 135) as foul. The word flower is very unstable. I find it, in the singular, in no fewer than six forms, flower, flowr, flowre, flour, floure, and flouer; and it is about the same in the plural. Similarly we have tower in three forms,-tower, towre, tour (all three forms occurring within eight lines of each other, P. L., XII. 44-52); and so with shower, hour, and other similar words. Seize he spells four ways,-seize, sieze, sease, seise. (7) As a promiscuous assemblage of examples of occasional consonantal spellings different from ours, take warr, dinn, lipps, mortall, celestiall, faithfull, musicall, committ, compell, farewel, mattin, sollemne, etc. On turning the leaves these are easily found also as war, din, lips, mortal, celestial, faithful, musical, commit, compels, farewell, matin, solemn, etc. So endles, darknes, sweetnes, etc., are found also as endless, darkness, sweetness, etc. (ripenesse in Sonnet III. in Milton's own MS. appearing as ripenes in the same when printed under his own eye); musick, majestick, etc., are found also as music, majestic, etc.; lincked is found also as linked; sulfurous as sulphurous. Patriark and patriarch are found in two consecutive pages (P. L., XII. 117-151); murtherer is found, but also murder, and murd'rous; chrystal and chrystall, but also crystal; autority, but also authority and authoritie.

Ample proof has now been furnished, not only of the general fact that Milton's spelling, like the spelling of most of his contemporaries, was unstable and variable, but also of the more special fact that, in the cases where he varied his

In

spelling, it was most frequently a mere accident, a mere turn of the wrist, whether he should give us a spelling that we now think odd or the one now adopted and authorised. fact, though we have used the phrase "Milton's spelling," it is impossible to say what Milton's spelling really was. There is an extant mass of his own manuscript, containing the drafts of a portion of his earlier English Poems. There, certainly, so far as the mass goes, we have Milton's own spelling. But then the spelling there differs in numberless particulars from the spelling of the same pieces when printed in 1645. The spelling in the volume of that year may be called Milton's own too, inasmuch as he had then the use of his eyesight, and it is to be taken for granted that he revised the proofs. But which is most Milton's spelling,—that of the MSS. so far as they go, or that of the printed volume? Farther, for all the later poetry, including Paradise Lost, Paradise Regained, and Samson Agonistes, we have neither a spelling set up by the printers from Milton's own manuscript, nor a spelling passed by Milton's personal revision after the printers, but only the discordant spellings of different printers, set up from the discordant spellings of no one knows how many different amanuenses to whom a blind man had dictated, and revised of course not by the blind man himself, but only by the readers of the printing offices, or by friends reading the proofs aloud for his benefit, with perhaps a shot of correction now and then from his own mouth when his quick ear detected anything wrong. The spelling of the First Book of Paradise Lost in the original edition differs incessantly from that of the same Book in the preserved MS. copy in the hand of an amanuensis from which the printers set up the text. In the very first page we find blissful seat in the printed edition substituted for blissfull seate in the press-copy, mortal for mortall, loss for losse, brook for brooke, soar for soare, pursues for persues, chiefly for cheifly, dark for darke, etc.

Suppose that, in this absolute impossibility of getting at a spelling for the poems throughout that could in any sense be called Milton's own spelling, there were to be a rough vote that the spelling of the original printed editions, just as it is, might pass for Milton's. What, even with that violent solution of the difficulty, ought to be the policy in a modern edition of the poems? Apart from the interest that might

VOL. III.

M

attach to an exact orthographic reprint of the original editions regarded as a bibliographical curiosity, the sole purpose it would serve would be to exhibit that very phenomenon of variability of spelling which we have been illustrating. But, in view of all else that we expect and require in a modern edition of Milton, would it be worth while to refabricate a collective edition of the poems expressly to exhibit the phenomenon of the variability, in Milton's case as in others, of the seventeenth century spelling,-nay, not its variability only, but its reasonless flutterings round and round our present spelling, with constant returns to it, and an evident disposition to poise upon it finally? Because Milton's original editions give us flower, but also five variations from it, flour, flowre, flour, floure, and flouer, did he mean to tie down his readers in all time coming to the sextuple spelling rather than the single? At the utmost, would he not have asked, in the interest of the history of English orthography, that the fact that the sextuple spelling was allowed in his day should be remembered in a footnote or the like, begging posterity at the same time to fix him to one of the spellings in the text if they found reason for it, on the single condition that they should not tamper at any point with sound or meaning, vocable or metre? In short, does not common sense decide that a modern edition of Milton's Poems for general use ought to consist, like our copies of the authorised English Bible, or our standard editions of Bacon and Shakespeare, of the most authentic text from the original editions spelt in conformity with our present orthography, except in cases where an archaic form ought to be preserved for some . etymological or phonetic significance which our present spelling would conceal?

Are there any peculiarities of Milton's spelling which are really significant, and ought therefore to be either (1) noted, or (2) preserved? There are, and we proceed to take account of these :

Mee, hee, shee, wee, yee.-That Milton had an intention in spelling these pronouns sometimes with a single e and sometimes with a double may be inferred from the fact that, in the Errata prefixed to the first edition of Par. Lost, he directs the word we in Book II. 414 to be changed into wee. On turning to the passage, it is seen that the reason was that the word we there has to be pronounced emphatically.

But, in fact, his own texts are not consistent with this principle, and the duplicated vowel is practically needless.

Then for than.-Though, as far as I have observed, the original texts keep to then, as writings of that date generally do, it seems unnecessary to recur to a spelling so strange to our present habits,-the rather because our form than was used in Milton's time, and is a good old one in pre-Elizabethan English.

Hunderd and Childern.—Among the Errata prefixed to the first edition of Par. Lost is the direction "Lib. I. v. 760 for hundreds r. hunderds"; which may be taken as vouching that Milton's ear preferred the latter pronunciation. Perhaps one ought to have obliged him here, especially as in the only three other occurrences of the word in his poetry (Arcades 22, Sonnet XIII. and Par. Reg., III. 287) it is hunderd or hunder'd. But hundred or hundreth is the old English form; Milton himself has hundreda in Latin; and people who still pronounce hunderd are accustomed to the spelling hundred. The form childern for children occurs four times in Par. Lost, and is worth noting; but, as we have childrens for children's in the same poem (1. 395), and children twice in Comus, once in Par. Reg., and once in Sams. Ag., there is no need to revive childern.

Furder and Fardest.-Milton, I think, never has the form farther in his poetry, and never the form furthest; but out of fifteen times in which he uses the word further he prints it three times furder, and in seven occurrences of farthest it is thrice fardest. No reason can be detected in the several cases for the change from the th to the d; and, as the th is most frequent with himself, that may be the rule.

Wardrope. The word wardrobe occurs twice (Lyc. 47, and Vac. Ex. 18). In the first case it is spelt wardrop in print, but wardrope in the Cambridge MS.; in the second wardrope. This may have been a pronunciation of the time; but it is erroneous, ungraceful, and not worth keeping.

Terf or terfe for turf.-This is one of the spellings of Milton that have escaped notice. It can hardly be accidental, for it occurs wherever the word is used in the poetry,—i.e. four times in all. But the phonetic difference between terf and turf is not appreciable, and turf is the genuine old form in writers from Chaucer to Milton.

Alablaster for alabaster.-The word occurs three times

twice with the I (Com. 660, and P. L., IV. 544), and once without it (P. R., IV. 548). As the proper word is alabaster, and is as old as Chaucer in that form, the insertion of the l was but a temporary freak.

Perfet and Imperfet: Verdit.-The word perfect occurs thirty-one times in the poetry, thirty times as the adjective, and only once as the verb (P. L., XI. 36). In eleven occurrences of the adjective the spelling is perfect, as now; in the remaining nineteen occurrences of the adjective, and in the single occurrence of the verb, the spelling is perfet. The spelling perfect predominates in the Minor Poems, occurring five times, while perfet occurs but twice (Com. 203, Lyc. 82); in the first two occurrences of the word in Par. Lost it is perfect (1. 550, II. 764), but uniformly through the rest of the poem, or sixteen times, it is perfet; in Par. Reg. it occurs five times, with a relapse into perfect in the first four, but a return to perfet the last time (IV. 468); and in Sams. Ag. it occurs but once, and then in the form perfet.

-The negative adjective occurs four times in all,—three times in Par. Lost, as imperfet (IX. 338, 345, and XII. 300), and once as imperfect (Vac. Ex. 3). There seems not the least doubt, therefve, that Milton preferred, at least occasionally, the French form (parfait, imparfait) to the direct Latin (perfectus, imperfectus). The French form indeed seems to have been the older; for we have parfit, parfite, and parfitly in our texts of Chaucer. All in all, as

Milton's oscillation between the two forms is curious, both might have been kept in the text; but, if there is to be uniformity, the predominance of perfect in the Minor Poems, the setting out with it in Par. Lost, and the return to it in Par. Reg., co-operate in its favour with present custom.There are no such reasons additional to present custom in the similar case of the French form verdit for verdict. Milton has the word twice only (S. A. 324, 1228); and in both cases the original gives verdit.

Show or shew.-At present either spelling of the word is legitimate, though show is the more common. There is little doubt, however, that shew is the more ancient spelling, that the word was pronounced correspondingly (like shoe), and that the spelling show came in with the fixing of pronunciation to our present practice. It is, accordingly, a very interesting word in Milton. If I am right in my