.Gr
A SYSTEM
LATIN PROSODY AND METRE,
HE BE-ST Tnrr H OTITT I E 8,
ANCIENT AND MODERN.
CHARLES ANTHON, LL.D.,
PKOFCS8OK OP THE GREEK AND LATIX LANGUAGES IN COLUMBIA COi.I.EOl
KXW-YORK, AND RECTOR OF THE GRAMMAR-SCHOOL.
NEW YORK:
HARPER & BROTHERS, PUBLISHERS,
329 k 331 PEARL STREET,
FBANKLIN SQUARE.)
1856.
Entered, according to Act of Congress, in the year i841, by
CHARLES ANTHON,
in the Clerk's Office of the Southern District of New- York
TO THE
REV. MATTHEW SIMPSON, D. D.,
President or the Asbnry University in the Stale of Indiana,
THIS WORK
IS RESPECTFULLY DEDICATED,
AS A
TRIBUTE OF ESTEEM FOR PROFESSIONAL TALENT, AND FOR THB
DEEP INTEREST WHICH HE HAS EVER EVINCED IN
THE CAUSE OF TRUE LEARNING.
PREFACE.
THE author of the present work prepared several years
ago a Treatise on Latin Prosody and Metre, which met with
no unfavourable reception, and proved, as he has reason to
believe, a somewhat useful guide to the young prosodian.
This volume having been for some time completely out of
print, the author has been induced to write a new work on
the subject : one that may not only be more worthy of his
increased experience as an instructer, but may furnish also
more detailed information on various points that were neces-
sarily omitted in the previous treatise. In collecting ma-
terials for this purpose, the author has derived important aid
from the Latin Prosodies of Dr. Carey and Professor Ram-
say, especially the latter, and he has introduced into the pres-
ent work whatever appeared of value in these and other
publications of a similar nature. He does not entertain a
doubt, therefore, but that the young scholar will find in the
present volume everything that may be needed by him, not
only at the commencement, but also throughout the several
stages, of his academic career.
Very few rules are given, it will be perceived, for the struc-
ture of Latin verse. This forms a distinct department of
study, for which, at some future day, a separate work will be
prepared ; and this intended work will also contain the Essay
0:1 Leonine and Macaronic Versification, which was to have
ft rmed part, of the present volume, but which has been omit-
ted through fear of making the work too large a one.
The Latin metrical rules of Alvarez were given in the pre-
vious work, but are omitted in this. The practice of giving
Latin rules in matters of prosody is fast going out of use,
and deservedly so ; and, besides this, the rules of Alvarez
are in numerous instances decidedly erroneous.
Columbia College, Aug. 2, 1841.
PROSODY.
SECTION I.
I. PROSODY treats of the Quantity of Syllables, and of the
different Species of Verse. 1
II. A Syllable is composed of one or more Letters ; as,
/, in, in-tus.
III. Letters are divided into Vowels and Consonants.
IV. Vowels, or open, free-coming sounds, are six in
number : A, E, I, O, U, Y.
V. From the vowels are formed eight Diphthongs : JE,
AI, AU, El, EU, (E, OI, UI.
VI. Consonants, or sounds produced by the compression
of the organs of speech, are divided into Mutes and Semi-
vowels.
VII. The Mutes are eight : B, C, D, G, K, P, Q, T.
VIII. The Semivowels, which have a sound more open
than that of the mutes, are likewise eight : F, L, M, N, R,
S, X, Z.
IX. Of the semivowels four are Liquids, L, M, N, R,
and they are so called because ihey flow, as it were, from
the lips with less exertion than is required in articulating
the other consonants.
X. Two other of the semivowels are Double Letters, X
and Z ; the letter X being equivalent to CS, GS, or KS ;
and the letter Z to DS or SD.
1. The Doctrine of Accent will be found under a separate head, im-
mediately after that part of the present work which treats of Versifica-
tion.
B
11 QUANTITY OF SYLLABLES.
XI. The letter H is to be regarded, not as a true conso
I ant, but as a mere breathing.
XII. The letter J was altogether unknown to the an-
cients. It appears that, among the Romans, the letter I
exercised a double function, being sometimes purely a
vowel and sometimes a consonant, answering very nearly
to our y. The character j was introduced, in later times,
into those words where i had the power of a consonant,
and therefore of itself, when not followed by another conso-
nant, could not have lengthened a short vowel. The letter J,
accordingly, is not, as some maintain, a double consonant. 1
XIII. The letter V, in like manner, represented among
the Romans, on some occasio'ns, a consonant, on others a
vowel sound. And the character U, to indicate it when a
vowel, is, like the letter J, a modern invention.
XIV. The letter Q was precisely equivalent in sound to
C or K, all three being pronounced hard:
SECTION II.
QUANTITY OF SYLLABLES.
I. By the Quantity of a syllable is meant the duration or
continuance of the voice in pronouncing it.
II. A syllable is either Short, Long, or Common.
III. A short syllable is sounded rapidly, and consists of
what is technically termed one time, like the a in the Eng-
lish word orator, or the e in the Latin word legere ; and is
thus marked, legere.
IV. A long syllable is slowly pronounced, and occupies
twice the time employed in sounding a short one ; as in the
a of the English word mediator, or of the Latin word sedare ;
and is thus marked, sedare.
V. A common or doubtful syllable may be made long or
1. Consult remarks on page 27, with regard to such forms as ejtu,
cujrj, fee
ORIGIN OF PROSODTAL RULES. 15
short, at the option of the poet ; as, papyrus or papyrus ;
fuerimus or fuerimus. 1
VI. The quantity of .a syllable is either natural, that is,
dependant on the intrinsic nature of the vowel itself, as the
re of rcsisto, in which the e is short by nature ; or accident-
al, as the re in restiti, which becomes long because it hap-
pens to be followed by two consonants.
VII. The quantity of syllables is determined by certain
established rules, or, when these fail, by the authority of
the poets.
VIII. In polysyllables, or long words, the last syllable
except one is called the penultima, or, more briefly, the
penult ; and the last syllable except two, the antepenultima,
or antepenult.
SECTION III.
OF THE ORIGIN OF PROSODIAL RULES.
I. Rules in Prosody originate from a careful examination
of the works of the best Latin poets. 2
II. In making such examination, we perceive that, with
a very few exceptions, the quantity of the same syllable in
the same words is always the same ; and, by classifying
those which are analogous, we arrive by induction at cer-
tain fixed principles, which are imbodied in rules.
III. In a great many other cases, however, we are una-
ble to detect any fixed principle, and must rest satisfied
with saying that we have the authority of the poets for
making such syllables long or short. 3
1. Consult remarks on page 29, Observation 1, with regard to a
short vowel before a mute followed by a liquid.
2. For some remarks on the relative value of the Latin poets as met-
rical authorities, consult Appendix.
3. All prosodial rules, in fact, are based upon the authority of the po-
ets ; but, according to the usage of prosodians, those syllables only are
said to be long or short by authority which cannot be reduced to rules.
hi VOWEL BKFOttfc. ANOTHER VOWEL.
SECTION IV.
RULES.
1. A VOWEL BEFORE ANOTHER VOWEL.
/ I. A vowel before another vowel, or a diphthong, is
short ; a.s,fuit, deus, tineas, eocinii^.
Tibull. Quisfuit, horrendos primus qui protulit enses?
Virg. Ipse ctiam exirmas laudis succensus amore.
The letter H being merely a note of aspiration or breath-
ing, is not regarded in such cases as the present ; and there-
fore, when h stands between two vowels, the preceding
vowel is short ; as, riihil, niihi.
Pers. De nihilo nihil, in nihilum nil posse reverti.
Virg. Musa, mihi causas memora ; quo numine lasso
EXCEPTIONS.
Exc. 1. Fio has the i long in all its tenses, except in
those where it is followed by er ; as, fio, flebam, flam,
but fierem, fieri. 1
1. It has been conjectured that the old form of Jio was/eio, and that
the first syllable was long in all the tenses without distinction. Some
of the parts, however, of these tenses in which er occurs, could not have
been used at all in Dactylic verse, if the first syllable had been always
long : thus, fleres, fierent, cannot stand in any place of a Dactylic
verse, and not even fierem, fieri, without an elision. Hence, when
the e of the diphthong was dropped, the writers of heroic verse intro-
duced this change into the quantity of those tenses where it was most
necessary, preserving the proper and original quantity in the rest. This
opinion receives much support from the fact, that the Comic writers, who
lived before the prosody of the language was very accurately defined, and
whose verse required no such modification of these words, constantly
used fieret, &c., with the first syllable long ; as, " Injurium 'st nam si
esset unde id fieret." (Tercnt., Ad., 1, 2, 20.) On the other hand, in
the works of the Christian poets, such as Prudentius, Arator, Tertul-
lian, &c., not only the first syllable in fierem, &c., but \nfio, &c., also,
is made short. Thus, we have the following Archilochian heptameter
in Prudentius : " lamque tuus fieri man das, fio Cyprianus alter."
(Perist., 13. f>9.) A less accurate mode of explanation is given by
Vossius (Aristarck., 2, 13), founded upon a passage in Priscian.
(Hamsay's Latin Prosody, p. 23.) Of the use of ei in earlier Latinity,
where, at a later day, the long i was employed, we have numerous ex-
amples. Thus, on the Duilian column we find castreis, socieif, classeis,
VOWEL BEFORE ANOTHER VOWEL. 17
Ovid. Omnia jam f lent, fieri qua posse negabam,
Virg. Anchiscs, fleret vento mora ne quaferenti.
Exc. 2. The genitives and datives singular of the fifth de-
clension make e long before i ; as, diei, speciei. But it
is found short in spei, and common in fidei or jidei,
and in rei or rei. 1
Virg. Nunc adeo, melior qucfiiiam pars acta diei.
Senec. Credi periisse. Vix spei quidquam est super
(Iamb, trim.)
Lucret. Nee jacere indu manus via qua munitajidei.
Statius. Quis morum fideique ? modus nunquamne mrili.
Lucret. Praterea rei qua corpora mittere possit.
Horat. Curia?- nescio quid semper abest rei. (Choriamb.)
naveis, numei ; and in the S. C. de Bacch. we have quei, virei, sibci,
eeis, vobcis, &c. Consult also Orell., Inscript. Lat. Select., n. 626,
3308, 3673, 4848, &c.
1. According to some, the old nominative form of th^ fifth declension
was diets, specieis, Jideis, making in the genitive din-is, speciei-is,
Jidn-is, which case afterward dropped the *, and became dieii, specieii,
fideil, and eventually diei, speciei, fidei, the i of the diphthong being
dropped. (Ramsay's Latin Prosody, p. 22.) The explanation given
by Bopp, however, is in every way preferable. This eminent philologist
makes the original form of the nominative to have been die-is, specie-is,
fole-is, and the genitive to have dropped its characteristic ending s, and
to have terminated like the old locative in i; thus making die-i, specie-i,
&c. If we write die-is in Greek characters, 6iri-i<;, and call to mind, at
the same time, the close relation that subsists between the Ionic 77 and
the Doric a, we will perceive at once a remarkable analogy between the
first and fifth declensions in Latin, and why so many words are found
belonging, as it were, to each ; as, planitia and planilie-s, canilia and
camtie-s, &c. In this way, too, we can explain the old form of the
genitive singular in the first declension, by supposing the nominative to
have been, for example terra-is, musa-is, whence came the genitive
terrd-i, musd-i, by dropping the * and retaining the locative ending,
while the Doric a in the nominative terra was displaced by the yEolic a.
(Bopp, Ve.rgleiche.nie Grammatik, p. 141, seqq. Id ib., p. 217, seqq.)
It remains now to account for such forms as Jidei, rei, &c. In all prob-
ability we have here merely an attempt to reduce the old forms with the
long penult to the operation of the general rule. It is worthy of re-
mark, too. that the examples of fidei all occur in writers of the lower
age, with whom it is very common. The subject of the identity of all
the Latin declensions, which has been merely glanced at in this note,
may be seen more fully discussed in Bopp's work just referred to, and in
Slruve's treatise, "Ueber die Lateinische Declination," &c., p. 3S, seqq.
B2
18 VOWEL BEFORE ANOTHER VOWEL.
Exc. 3. Genitives in ius have the i long in prose, but
common in poetry. Alterius, however, has the al-
ways short, all us always long. 1
Virg. Posthabita coluisse Samo ; hie illius arma.
Id. Tu faciem illius noctem non amplius imam.
Id. Ipsius Anchisae long&vi hoc munus habebis.
Id. Nunc ultro ad cineres ipsius et ossa parentis
Horat. Nullius addictus'jurare in verba magistri.
Virg. Non te nullius exercent numinis ira.
Id. Unius ob noxam et furias Ajacis O'ilei.
Id. Navibus infandum amissis unius ob iram.
Exc. 4. Such proper names as Ca'ius, Pompe'ius, Vul-
te'ius, as also Veius, and the like, have the a or e long
before i. 2
1. The authority for the remark that genitives in ius have the penult
long in prose, is furnished by Quintilian : " Prctlerea qua fiunt spalio,
tive cum syllaba longa. corripitur, vt unius ob noxam et
furias; extra carmen non dcprehendas." (Inst. Or., 1, 5, 18.) Bopp
considers the Latin genitive-ending ius analogous to the Sanscrit termi-
nation sya, the a being changed to M before the final s by a very usual
process in early Latin. (Compare the Sanscrit vrka-s with the Latin
lupu-s, and yung'mas with jungimus.) In accordance with this view of
the subject, the old genitive forms of illius, ipsius, nullius, &c., will
have been illi-ius, ipsi-ius, nulli-ius, &c., which, when changed to illius,
&c., preserved their proper quantity in prose, although the poets took
advantage of the circumstance of the i being followed by a vowel, to
bring them, when it suited their purpose, under the general rule.
(Bopp, Vergleich. Gramm., p. 220.) With regard to alterius, it may
be remarked, that although no instance can be found in any poet of its
occurring with the long penult, yet there can be no doubt of its having
been once employed. Indeed, Terentianus Maurus, the grammarian,
actually uses alterius on one occasion ; in the following Trochaic
tetrameter catalectic (cap. 3, de Fed., v. 1352):
" Sescuplo vel una vincct altering sinpuhim."
Priscian ascribes the short penult in alterius to the circumstance of the
genitive's exceeding the nominative by two syllables : " quod dual/us
tyllahis vincit genitivus nominativum" (lib. 6, c. 7, p. 695. ed. Putsch.).
A most singular explanation, certainly, and yet advocated by Vossius
(de Art. Gram., 2, 13, p. 150, ed. Focrtsch.). The reason assigned by
Scaliger for the long penult of genitives in ius, namely, that they were
originally written cms, is adopted by Ramsay, but is far inferior to
Bopp's explanation given above. (Scat., de Caus L. L., c. 43.)
2. According to Priscian (7, 5, p. 739). such words as Caius, Pom-
VOWEL BEFORE ANOTHER VOWEL. 19
CatulL Cinna est Caius, is sibi paravit. (Phalaecian.)
Mart. Quod peto da, Cai, non peto consilium. (Pentam.)
Ovid. Accipe, Pompei, deductum carmen ab illo.
Propert. Forte super portas dux Veins adstitit arcem.
" Exc. 5. The a is long in the penult of the old genitive and
dative of the first declension ; as, aulai, terrdi, &c.'
Virg. Aulai in media libabant pocula Bacchi.
Lucret. Terraique solum subigentes, cimus ad ortus.
Exc. 6. The verb aio is, in some of its parts,JprVnounced
with the first syllable forming a diphthong, and, conse-
quently, long ; while in other parts the a and t form
two short syllables. Thus, we have aio, uiunt, aiebam,
aiebas, aiebat, &c. ; and, on the other hand, Sis, ait, &c. 2
Horat. Servus ; habes pretittm, loris non ureris, aio.
Id. Plebs eris ; at pueri ludentes, Rex eris, aiunt.
Id. Felicem ! aiebam tacitus. Quum quidlibet ille.
Id. Non sum moechus Sis. Neque ego, hercule, fur, ubi
vasa.
Id. Nil Sit esse prius, melius nil caelibe vita.
Exc. 7. In ohe 3 and Diana* the first syllable is common ;
perns, Vulte'ius, &c., were originally written Caiius, Pompeiius, Vul-
leiius. On this supposition we may easily account for the long quantity
of the first syllable; and hence the vocatives Cai, Pompei, &c., are in
reality Cal-i, Pompel-i, which last undergoes another contraction, in
Horace, into Pompei. (Horat., Od., 2, 7, 5.) In like manner, Vullet-i
becomes, in the same poet, Vidta. (Epist., 1, 7, 91.)
1. The principle on which this exception is based has already been
stated, in note 1, page 17.
2. The verb aio was originally written aiio, and in all probability pro-
nounced al-yo. One of the i's being dropped, the a and remaining t
sometimes formed a diphthong, as in the original form, and sometimes
two short syllables. (Ramsay's Latin Prosody, p. 24.) Quintilian
informs us that Cicero actually wrote aiio. (Inst. Or., 1, 4, 11.
Consult Spalding, ad Inc.)
3. Ohe, the interjection, follows its primitive O, which, since it can-
not be elided, because words of this nature require a strong emphasis,
is made either long or short when it falls before a vowel. (Ramsay's
Lot. Pros., p. 25.)
4. Diana was originally Deiva lana, the lunar goddess, contracted
20
VOWEL BEFORE ANO1HER VOWEL.
in eheu 1 and Io 2 (the daughter of Inachus) it is long
The interjection io follows the general rule.
Mart. Ohejam satis est, ohe, libelle. (Phalaecian.)
Virg. Exercet Diana chores ; quam mille sccutee.
Id. Constiterunt , sylva alia Jovis lucusve Diana.
Tibull. Ferreus est, eheu, quisquis in urbe manet. (Pentam.J
Virg. Ad levem clypeum sublatis cornibus Io.
Ovid. Ante oculos Io, quamvis aversus, habebat.
Virg. Clamat io matres, audite, ubi quteque Latinos.
c. S.li
Exc. 8. In many words of Greek origin a vowel is long,
though immediately followed by another ; as, Her,
Achelous, Enyo, Troas, Troius, &c., the Latin quan
tity being controlled by that of the Greek words
whence they are derived. Others, however, would
appear to differ in quantity from their archetypes ; thus,
in Latin we have chorea and chorea, while in Greek we
subsequently into Deiana, and at last becoming Diana. The of the
diphthong being dropped, gave rise to the double quantity of Diana,
since it could be brought under the general principle of one vowel be-
fore another (Voss ., de Art. Gram., 2. 13. Varro, R. R, I, 37, 3.
Grcn., Thts., vol. 8, p. 31 1 Nigid. ap. Macrob., Sat., 1, 9. Creuzer,
Symbolik, par Guigniaut, vol. 2, pt. 1, p. 433.)
1. The interjection eheu is generally thought to have been abbreviated
by the transcribers from heu heu. (Heyne, ad Virg., Eclog., 2, 58.
Wagner, ad eund.)
2. As regards Io the proper name, and io the interjection, a differ-
ence of opinion exists. The ordinary doctrine is, that the first syllable
of both is doubtful, and so the rule is given in most works on Latin
prosody. A more correct view of the subject, however, teaches us that
the first syllable is always short in the interjection, and always long in
the proper name. In Silius Italicus (14, 517), the old reading was
" Qua,jue fertbatur ductor Sidonius, io," where the interjection occurs
with a long penult ; but the true lection is now given as " Quaque re-
hebatur Crantor Sidonius, Jo," where Io is the name of a ship. With
regard to Io, the daughter of Inachus, two lines are found which give
the word a short penult (Ovid, Her., 14, 103. Id., Ibis, 624). But in
the first of these the MSS. vary, and some commentators even regard
the term as an interjection ; while the other is drawn from a poem of
doubtful authenticiiv. and the text of which is notoriously corrupt. In-
deed, in this latter line, most editors now adopt the emendation of Hein-
sius, and read Ion for Io. In Greek, also, this word is always Iw.
Lot. Pros., p. 26.)
VOWEL BEFORE ANOTHER VOWEL. 21
find %opeia alone ; and so also platea, which is no-
thing more than the feminine adjective TrAarem. 1
Virg. Consurgunt vend, atque in nubem cogitur der.
Ovid. Opperiuntur aqua vultus Acheloiis agrestes.
Claud. Erubuit Mavors, aversaque risit Enyo.
Virg. Desidias cordi,juvat indulgere choreis.
Id. Pars pedibus plaudunt choreas et carmina dicunt.
Horat. Purae sunt platea, nihil ut meditantibus obstet.
Catull. Istos qui in platea modo hue tnodo illuc. (Phalse-
cian.)
Exc. 9. Those words which are written in Greek with
the diphthong ei (et), and in Latin with a single e or i,
have that e or t long ; as, jfcneas, Alexandria, Anti-
ockia, Caesaria, &c. 2
Virg. At pius jEneas, per noctem plurima volvens.
Propert. Noxia Alexandria, dolts aptissima tellus.
Exc. 10. On the same principle as stated in the prece-
ding exception, most adjectives in eus, formed from
4 Greek proper names, have the e long ; and it continues
long when resolved into e'i ; as, Cythereus and Cythe-
reius ; Pagaseus and Pagaseius ; Pelopeus and Pclo-
peius?
Virg. Parce metu Cytherea ; manent immota tuorum.
Ovid. Invocat Hippomenes, Cythf.reia, comprecor ausis.
Lucan. Eumenidum vidit vultus Pelopeus Orestes.
Ovid. Quid quod avus nobis idem Pelopeius Atreus /
1. Ramsay endeavours to explain these anomalies by supposing that
the Romans were in possession of authorities unknown to us. (Lai.
Pros., p. 27.) Prudentius uniformly makes it platea, but his authority
is of no value.
2. Many words of this class such as names of towns, temples, or
.monuments, are in reality only adjectives, with a noun understood; as,
'A2,eav(5pta (TroXtf), " Alexandria,' 1 '' or the city of Alexander;
ov (iepov), " Museum," or temple of the Muses ; Mavcufalov
ov), " Mausoleum," or sepulchre of Mausolus, &c.
3. Consult rerr.ar'^ on " Dizrcsis," p. 121.
22 VOWEL DEFORP ANOTHER VOWEL.
. Exc. 11. Greek genitives in eos, and accusatives in ea,
from nominatives in tus, have the penult short accord-
ing to the common dialect, but long according to the
Ionic ; as r Orpheos, Orphea ; Antheos, Anlhea (in
Greek, 'Op^eof, 'Op^ea ; 'Av6eo$ , 'Av0ea, by the com-
mon dialect) ; but Idomeneos, Idomenea; Ilioneos, II-
ionea (in Greek 'Idoj^vf/oc, 'Ido/zev^a ; 'I/UovTJoc, 'IA-
lovffa, by the Ionic). 1
Ovid. At non Chionides Eumolpus in Orphea talis.
Virg. Prospectum late pelago petit, Anthea si quern.
Id. Idomenea ducem, desertaque litora Creta.
Id. Ilionea petit dextra laevaque Serestum.
Exc. 12. Academia ought to be regarded as having a long
penult. 2 In the proper name Orion the second sylla-
1. We frequently find the quantity vary in words taken from the
Greek, because they appear in the original language under a double form,
which often depends on the dialect used by the poet. Thus, we have
Conopium and Conopeum, because we find in Greek KUVUTTIOV and Kw-
vcjTTttov. So also ~Eous and ~Eous, in Greek ^of and twof ; Malea
and Malca, from MaAeta and MaAen ; Nereides and Nereides, from NJ?-
pritdeg and N^peWef ; Rhea and Rhea, from 'Peia and 'Per/ ; Dadaleus,
Dctdaleus, Dadalns, from the triple Aatrfa/Utof, Aaiid/leof, Aai'(5a/^f,
&c. (Ramsay's Latin Prosody, p. 27.)
2. It is always long in the Greek authors, as the following examples
will prove.
'AA3.' elf 'A-Kadfylav tcariuv irro rcuf fiopiaic aKoOpst-ei. (Ana-
past. Tclram. Cat. Aristoph., Nub., 1001.)
. 'Avf '\Kodijfuaf Bopdopov EV irpoxoalf. (Pentam. Theocr., Ch.
ap. Brunch, Anal., vol. 1, p. 184.)
'Ajiadrifuaf ijicovaa, %6-/ov. (Anapaest. Dim. Epicrat. ap. Athen.,
2, p. 59, D.)
AVKEIOV 'A.Kadrj/jlav Qldeiov -irvfaf. (Iamb. Trim. Alex. ap.
Athen., 8, p. 336, E.)
The two Latin lines quoted in the text give the received quantity of
Academia in Cicero's time, the first being taken from the treatise " De
Divinatione" (1, 13), and the second from Laurea Tullius, one of Cice-
ro's freedmen, in an epigram preserved by Pliny (31, 2). This decides
the question as far as the practice of the golden age of Roman literature
is concerned. In opposition to all this array of authorities in Greek and
Latin writers, we have merely one from Claudian, and another from a
still later poet, Sidonius, who both use Academia with a short penult.
(Claud., de Coiu. M., \, 94 Sidort, Ap., 15, 120.^ Hermann thinks
that the penult of 'Axodj^'a was shart by nature, but lengthened by the
CONTRACTION. 23
ble is always long in every good Latin writer, although
it is common in Greek. On the other hand, the first
syllable is common in Latin, but always long in Greek ;
and the third syllable in the oblique case? is also com-
mon in Latin and long in Greek. 1
Cic. Inque Academla umbrifera nitidoque Lyceeo.
Laur. Tull. Atque Academias celebratam nomine villam.
SECTION V.
CONTRACTION.
RULE. Every syllable formed by contraction is long ; as,
cogo, contracted for coago or conago ; tibtcen, contracted for
tibidcen or tibncen, &c.
Virg. Bis gravidos cogunt fatus, duo tempora messis.
Ovid. Cur vagus incedit tola tibtcen in urbe.
BS. 1. Among the contractions of most frequent occur-
rence, the following may be here enumerated :
1. Two vowels uniting into one.
ambages, contracted for ambeages, 2 **
indago, " " induago, 3
allus, " " atiius,
dls, " " diis,
audis, " " audlis*
fiR, " " flie,
force of the accent, as the term was one in frequent and common use.
(Hcrm. ad Aristoph., Nub., 1001. Id. de Metris, c. 23. Class. Journ.,
vol. 6, p. 123.)
1. It is erroneous to say, as some do, that Orion has the second syl-
lable common. The true rule is given in the text. In Greek, however,
the t in 'Opiuv, gen. J Opid>vof, is of variable measure in Homer and the
epic poets, but short in the tragedians. (Spitzner's Gr. Pros., p. 112.)
2. Ambe was an old form of amb or am. Compare the Greek
a^tyi. (Varro, L. L., 7, 3.)
3. Indu or endo was an early form for in. Compare the Greek
ivdol.
4. The third conjugation is regarded now as the oldest, and the 1st,
2d, and 4ih as merely contracti-d conjugations derived from it. (Con-
sult Slmve, " Ueber die Lateinische Declination u.nd Conjugation,"
Konigtberg, 1823.)
CONTRACTION.
proles, contracted for profiles,
cdpia, " " coopia,
fructus, " fructuls,
anus, " anuis,
luxu, (dat.) " " luxui,
chely, " " chelye.
/' f . ' 2. The letter h dropped between two vowels.
ml, contracted for mlhi,
nil, " " nl/nl,
vemens, " " vehemens,
nemo, " nehdmo.
3. The letter v dropped between two vowels.
bubus, from boibus, and that from bovtbus.
" botcula, "
bucula, '
junior, '
jutum, '
jumentum, '
mobilis, '
momentum, '
malo, *
norms, ML '
oift;
prudens, '
4. Other letters similarly dropped.
denus, ' deenus, decenus, from decem.
blgae, biigas, biivga.
quadriga, quadrnga, quadrnugae.
juenior, "
juvenior,
from juvenu.
juatum, "
juvalum,
" juvo.
juamentum, "
juvamentum,
" juvo.
molbilis, "
mombilis,
" moteo.
moimcntum, "
movlmenium,
" mot co.
maolo,
mavolo,
u magis-wlo.
nocnus,
novenus,
" novcm.
oJmt,
obivit.
proidens, "
prdvidens.
OBS. 2. It sometimes happens, however, that when two
vowels meet in a compound word, one of them is elided or
struck out altogether, in which case the quantity of the re-
maining one suffers no change ; thus, in magnopere, com-
pounded of magnd opere, the o of magno is struck out alto-
gether, and the o in opere retains its natural quantity. So
in semdnimis, gravolens, suavolens, &c.
OBS. 3. In a few words, the ancients seem cither to have
blended the two vowels into one, or to have struck out one
DIPHTHONGS. 25
of them at pleasure, and hence the quantity of such syllables
is variable. Thus the participle ambitus, from ambio, has
the penult long, because it is regarded as a contraction for
ambeitus ; while ambitus, the substantive, has the same syl-
lable short, because, in this case, the e was supposed to be
elided before the i.
SECTION VI.
DIPHTHONGS.
RULE. A diphthong is long, whether in a Greek or Latin
word ; but pr& preceding a vowel in a compound term is
usually short. 1
Virg. Ultima Cumai venit jam carminis (etas.
Ovid. Exit et in Maias sacrum Florale Calendas.
Virg. Fertur equis duriga neque audit currus habenas.
Id. Caucasiasque refert volucres,furtumque Promethei.
Ovid. Quas ubi viderunt pr&acuta cuspidis hastas.
Virg. Nee tola tamen ille prior pr&eunte carina,
Id. Stipitibus duris agitur, sudibusve pr&ustisJP
1. The syllable pro, being originally prat or prae, the latter of the
two vowels is tacitly elided. Thus preeustus, pr&acutus, prazeo, be-
come pra'ustus, pra'acutus, praCeo, and the a. is necessarily short by its
position before the succeeding vowel. There is one example, however,
in Statius, where prae is lengthened in prairet, namely, " Pratmia, cum
vacuus domino prairet Arion" (Theb., 6, 519) ; but, besides this, other
instances of such lengthening are found only in the latest poets, as, for
example, prtzesse in Sidonius Apollinaris (Carm., 23), and praoplare in
Martianus Capella (lib. 1, pag. 9, ed. Grot.) ; but such instances are of
no value. (Consult Schneider, L. G., vol. 1, p. 103.) In some of the
cases where pro. occurs short in composition, we may obviate this by
having recourse to synaeresis ; this, however, will not answer on all occa-
sions. (Compare Max. Viet., p. 1965, ed. Putsch, and, on the other
side, Vossius, de Art. Gram., 2, 15, p. 161, ed. Foertsch.) Ovid, on
one occasion, shortens the diphthong in M&otis (Trist., 3, 12, 2), in
which he is imitated by Seneca (Oed., 474), but it is made long by him
elsewhere, as well as by other poets. The same poet has also Aeeta
(Heroid., 6, 103). In this and in Mceotis he appears to have followed
the Greek license. (Consult Buttmann, Ausf. Gr. Gr., 7, anm. 25.)
The old reading Dionaus in Catullus (29, 7) has long since been
changed. (Docring^ ad CatuU., 1. c. Vossius, de Art. Gram., 2, 33,
utr.)
C
26 DIPHTHONGS.
OBS. 1. Greek proper names in eus (genitive eos) always
have the eu a diphthong, in both Greek and Latin ; as,
Orpheus, gen. Orpheos, Pantheus, gen. Pantheos, &c.
OBS. 2. Frequently, however, a proper name in eus is
declined according to the forms of the second declension in
Latin, and then the diphthong disappears ; as, Orpheus,
gen. Orphei ; dat. Orpheo, <fcc.
OBS. 3. The combination yi also forms a diphthong in
some Greek words, since it answers to vi ; as, Orithyia
('OpEiOvia) ; Harpyta ("ApTTVia) ; Agyleus ('Aywevf ), &c.
Ovid. Orithyian amans fulvis complectitur a/if.
Virg. Et patrio insontes Harpytas pellere regiio.
Horat. Levis Agyiev. (Dactylic dimeter.)
OBS. 4. It is, strictly speaking, erroneous to rank such
combinations as we, ue, ui, uo, uu, among diphthongs ; on
the contrary, the letter seems to have been pronounced
in such cases like our to , as, lingua (equivalent to ling-wa),
sanguis (sang-uns), equus (ek-tcus) ; and so also in mono-
syllableS^ as, quis (equivalent to kwis), quod (ktcod), que
(kwe), fec.'
OBS. 5. In some cases, it is true, such combinations are
long : this, however, is not because they then form a diph-
thong, but because the vowel with which u happens to be
united is in itself long ; as, suadet (swadet), svetus (swetus),
&c. So also in the monosyllables : qui (km), huic (hwick),
quo (kwo), &c.
OBS. 6. In several words, again, u and the vowel by
which it is followed always form distinct syllables ; as,
sua, sues, suis, suos, suus, &c.
1. The whole question is ably discussed by Schneider, L. G., ol. 1,
p. 324, scqq.
,
POSITION. 27
f
SECTION VII.
POSITION.
RtiLE A vowel before two consonants either in the
same word, or in different words, or before a double conso-
nant, is long by position ; as, terra, respexit, gazd, &C. 1
Virg. Terra tremit : fugere ferae, ct mortalia corda.
Id. Libertas qua sera tanien respexit inermem.
OBS. 1. The letter h is not regarded in prost>dy,as a
consonant, but a mere breathing ; hence, in such words as
dbhinc, adhuc, &c., the first syllable is short. So also if a
word end in a consonant, and the succeeding word begin
with the letter h, there is no length by position. 2
Virg. Oro, siquis adhuc precibus locus, exue mentem.
Id. Tempora quaz messor, qua curvus orator haberet.
OBS. 2. A vowel is often found long before the consonant
j. This, however, is not because _;' is a double consonant ;
but the truth is, that the words in question were originally
all written with a double t ; as, for example, cuiius, huiius,
eiius, peiius, maiius, &c., and would seem to have been
1. The time occupied in pronouncing the two consonants, or the
double one, produces the length by position. A vowel thus situated is
in fact shorter than one which is long by nature ; but still in verse they
are regarded as both of equal length. (Kriigcr, Grundriss der Metrik,
P- 18.)
2. That the h never produces length by position when joined with a
consonant either preceding or following it, is expressly asserted by
Terentianus Maurus, v. 789, seqq.
" Sola nee vocalis usum, nee tuetur consona,
Tempus aut ministral ullum brevibus usquam syllabis."
To the same effect are the remarks of Vossius (de Art. Gram., 2, 15,
extr.). The opposite doctrine is incorrectly maintained by Velius
Longus (p. 2217, ed. Putsch), and also by some modern scholars ; as,
for example, by Earth (Advers., 21, 17). Some of the lines adduced in
support of this latter opinion have the syllable lengthened by the arsis ;
in others the reading is erroneous. Equally unfounded is the remark,
that the h is sometimes employed to prevent the hiatus in a line.
(Schneider, L. G., vol. 1, p. 180, not.)
OF THE INITIAL X AND Z.
prontmnced cui-yus, hui-yus, ei-yus, pci-yus, mai-yus, &c. ,
so that the first syllable was long in each, a quantity that
was retained after one of the two i's was dropped. 1
Of the initial SC, SM, SP, SQ, ST, $c.
RULE. A short vowel at the end of a word, when followed
by a word beginning with sc, sm, sp, sg, st, &c., is rarely, if
ever, allowed to remain short, in serious compositions, by the
poets who flourished after the time of Lucretius ; but they
generally avoid with care such a collocation. 2
Of the initial X and Z.
RULE. A short vowel at the end of a word, before a word
beginning with x or z, remains short. At least, there is no
evidence at all that it was ever lengthened. 3
Ennius. Pontibus instratis conjunxit litora Xerxes.
Virg. Jam medio apparet fluctu nemorosa Zacynthus.
1. Consult note 1, page 18.
2. The whole question is discussed with great ability by Schneider
(L. G., vol. 2, p. 694) and Ramsay (La/. Pros., p. 260, seqq.). The
canon of Dawes on the subject is particularly examined by the latter.
Proper names, in the use of which the poets have always very natu-
rally allowed themselves considerable license, and the names of stones,
trees, &c., when such cannot be used in the metre without transgressing
the rule in the text, are exceptions, of course. This applies to such
words as Smaragdus, Scamander, neither of which could be employed
in hexameter verse % at all without a license of this kind. And it is
worthy of notice, that these words are frequently found in good MSS.,
both Latin and Greek, without the S at all: as, Mdpaydof, Maragdus;
ILafiavdpof, Kamander (Dawes, Misc. Crit., p. 6-148, ed. Kidd.) ;
which would seem to indicate that it was, in certain cases, softened
down in pronouncing them. (Ramsay's Lat. Pros., p. 261, not.
Compare Knight, Proleg. in Horn., 79. Spitzner, Gr. Pros., t) 7, 6.
Tkiersch, Gr. Gr., $ 146, 8.)
3. The line from Ennius is merely conjectured by Columna to have
been written by that poet ; a conjecture, however, without any founda-
tion. (Ennii Fragm.,ed. Column., p. 129.) Terentianus Maurus (v.
1160) quotes it without naming the author, and subjoins another imme-
diately after as follows : " Sanguine turbalus miscebat litora Xerxes."
It is more than probable that both lines are his own composition. An-
other instance is sometimes cited from Lucan (2, G72) : " Talis fama
canit tumidum super acquora Xerxen ;" but the text is now altered so
as to read " super aquora Perscn." (Consult Cort., ad
loc.) The Greek poets do not appear to have ever indulged in this h-
MUTE AND LIQUID. 29
SECTION VIII.
MUTE AND LIQUID.
RULE. A short vowel before a mute followed by a liquid,
both of which are in the following syllable, is common in
poetry, but always short in prose ; as, volucris and volucris ;
pdtrem and pdtrem ; pharetra and pharetra ; cochleare and
cochleare ; Cyclops and Cyclops ; cycnus and cycnus, &C. 1
Ovid. Et primo similis volucri, mox vera volucris.
Virg. Natum ante ora pdtris, pdtrem qui obtruncat ad aras
Id. Virginibus Tyriis mos est gesture pharetram.
Id. Succinctam pharetra et maculosas tegmine lyncis.
Horat. Multa Dirc&um levat aura cycnum. (Sapphic.)
Id. Donatura cycni si libeat sonum. (Choriambic.)
OBS. 1 . This rule depends on three conditions, viz. : *
1. The liquid must follow the mute. If it stand before
the mute, the preceding vowel, though naturally short,
becomes always long ; as, fert , ferlis.
2. The mute and liquid must belong to the same syllable.
If they belong to different syllables, the preceding
short vowel becomes long ; as, db-luo, dd-nitor, quam-
ob-rem.
cense. With regard to z, numerous examples are found of a vowel re-
maining short before it. (Ovid, Her., 1, 87.SU. Ital, 1, 275. Id.,
2, 603. Martial, 2, 58, 1. Auson., Prof., 22, II. Id. ib., 13, 3.
Senec., Here. Fur., 916. Id., Agam., 433, &c.) In many of these
instances, however, the words commencing with z could not otherwise
appear in the verse, unless a short vowel were allowed to precede. This
same principle regulates the admissibility of such words into Greek
verse. (Horn., I 1 .., 2, 634. Id., Odyss., 1, 246. Id. ib., 9, 24.
Tfieocrit., 4, 32. Oppian, Hal., 1, 367, dec.) The dramatic poets,
however, never allowed themselves this license.
1. This rule applies more frequently to the liquids I and r than to m and
n. The latter are found for the most part in words derived from the
Greek. The position which a mute and liquid make is termed by
grammarians " debilis positio," or weak position. The license in po-
etry, therefore, consists, not in having the vowel short, but in making it
long. The natural quantity, on the other hand, prevails in prose
(Vossius, de Art. Gram., 2, 16. Schneider, Jj. G., vol. 2, p. 676,
C 2
30 MTJTE AND LIQUID.
3. The vowel must be short by nature. A vowel
naturally long is never rendered short by a mute
and liquid following ; as, matris, salubris, atri. 1
OBS. 2. The lengthening of a vowel before a mute and
liquid takes place most frequently in the arsis of a foot.
Examples of a similar lengthening in the thesis are com-
paratively rare. 2
OBS. 3. Even in the arsis, however, the poets do not
seem to have lengthened at pleasure any syllable that was
followed by a mute and liquid. A particular usage appears
to have prevailed on this head, which forbade them, for ex-
ample, to lengthen the vowel in arbitror, genctrix, locuples,
&c., while in other words, containing, in like manner, a
vowel before a mute and liquid, the long quantity was
more customary than the short. 3
OBS. 4. The mute and liquid are sometimes found in
the beginning of the next word ; as,
Virg. Excursusque breves tentant, et s&pe lapillos.
Auson. Jane nove, primo qui das tua nomina mensi.
1. To determine whether a vowel which we find long before two con-
sonants be naturally long, or only rendered so by that position, we must
look to the word in a different state, where the position does not occur.
In some few instances it is quite impossible to determine with any cer-
tainty the natural quantity of the vowel ; as, for example, in Venafrum,
hybrida, &c. (Schneider, L. G., vol. 2, p. 677, not.)
2. Schneider, L. G., vol. 2, p. 676.
3. Thus we find colubris only in Valerius Flaccus (6, 175), but fre-
quently colubras and colubris. So in Lucretius (4, 406) we have rubntm,
but everywhere else rubri, rubro, &c. Again, we have Kbri, libra, &c.,
in only a very few places ; as, for example, Moral., Serm. t 1, 1Q, 63.
Id., Epist., 2, 1, 217. Auson., Idyll., 11, 77. Id., Epist., 4, 98 ; but
very frequently libri, &c. In like manner, we find the oblique cases of
niger, with the short i, in Catullus (43, 2), Horace (Od., 1, 32, 11 ; 3, 6,
4 ; 4, 12, 11), Virgil (J5n., 8, 353), <kc., while those with the long i
occur far more frequently. With regard to migro, it is to be observed,
that it is found, together with its compounds, having the short penult in
many passages of Plautus, and once also in Terence, Lucretius, and
Manilius (Housing., Obs. Antibarb., p. 375); but Virgil, Horace, Mar-
tial, Juvenal, and other poets uniformly have migro. Hence it has been
supposed that migro at first bad the short i, and retained this down to
the time of Catullus, ^>ut afterward changed it to the long quantity ;
while, on the other hand, latro (another word which has given hw
REDUPLICATING PRETERITES. 31
SECTION IX.
REDUPLICATING PRETERITES.
RULE. Reduplicating preterites, or such as double the
first syllable, have both the first and second syllables short ;
as, cecidi (from cado), ceclni, didici, tetigi, <fcc.'
Virg. Inter cunctantes cecid.it moribunda ministros.
Id. Tityre, te patulas cecini sub tegmine fagi.
Ovid. Max didici curvo nulla subesse iholo. (Pentameter.)
Virg. Nee victoris heri tetigit captiva cubile.
EXCEPTIONS.
But cecidi, from cado, and pepedi, from pedo, have the
second syllable long.
Juv. Ebrius ac petitions, qui nullum forte cetidit.
Horat. Nam, displosa sonat quantum vesica, pepedi.
OBS. The rule does not, of course, apply to the second
syllable when it is long by position ; as in momordi, cucurri,
peperci, and the like ; still, however, the first or prefixed
syllable is short.
Mart. Docti lima momorderit Secundi. (Phalaecian.)
Propert. Cum vicina novis turba cucurrit agris. (Pentam.)
t o remark) substituted, in process of time, the long a for the short.
(Schneider, L. G., vol. 2, p. 679, scq.)
1. The first syllable is short as a matter of course, since it consists
of a short prefix. The second syllable follows the quantity of the ver-
bal root. Hence arise the two exceptions mentioned immediately after,
namely, cado and pedo, where the first syllable of the verbal root is
long ; and so also the forms momordi, cucurri, &c. The early form of
the perfect of cado must have been cecadi. (Consult Priscum, 10, 4,
p. 489, ed. Putsch. Pott, Etymol. Forsch., vol. 1, p. 19, seqq.
Kuhner, Gr. Gr., ol. 1, p. 84, seqq. Bopp, Vergleich. Gramm., p.
697,
32 PRETERITES OF TWO SYLLABLES.
SECTION X.
PRETERITES 9F TWO SYLLABLES.
RULE. Preterites of two syllables, their compounds, and
the tenses formed from them, have the first syllable long ;
as, vidi, veni, Jovi, fugi ; while, in the present tense, video,
venio,foveo,fugio, have the first short. 1
Virg. Ut vtdi, tct peril, ut me mains abstulit error.
Id. Respexit tamen, et longo pajf tempore venit.
Id. Fdvit humum ; cape saxa manu, cape robora pastor
Id. Vipera delituit, calumqrue exterrita fugit.
EXCEPTION.
Seven dissyllabic preterites, however, have the first syl-
lable short, namely, bibi, dedi,ftdi (from fmdo), scidi
(from scindo), steti, stiti, and tuli?
1. According to the theory of Grimm (Deutsche Grammatik, vol. 1,
p. 1056), those verbs which change a short vowel in the root, or present
tense, into a long c in the perfect, had originally a reduplication.
Pango, or, rather, pago, makes peptgi, but compingo makes compegi.
This proves the analogy of the two forms ; and on the model of pago,
peplgi, contracted pcgi, we have
vemo,
video,
fovco,
fugio,
&c.,
vevem,
vividi,
fovovi,
fufugi,
&c.,
veent,
vndi,
foovi,
fuugi,
&.C.,
vem,
vidi,
foci,
&c.'
It is observed, in confirmation of this remark, that such verbs have in
many instances a reduplication, or, what is allied to it, an internal in-
flection, in the cognate languages ; thus fugi resembles irfyvya ; and
venio, veni, the Maeso- Gothic verb, which is analogous to " come" and
"came," &.c. (Pritchard, Origin of the Celtic Nations, p. 151.) It
must be borne in mind, however, that the remarks here made about the
mode of forming rem, vidi, &c., do not apply to such preterites as lust,
rlsi, misi, &c., from ludo, rideo, mitto, &.C., the preterite in these
verbs having been formed by the insertion of *; as, lud*i, ridsi, mittsi,
and the consonant or consonants before the having been subsequently
dropped for the sake of euphony. (Pritchard, 1. e. Ramsay, Lat.
Pros., p. 100. Struve, Ueber die Lateinische Declination vnd Conju-
gation, p. 155.)
2. These seven are in reality no exceptions at all, but belong to the
preceding rule, since they are all reduplicating preterites, some of which
have dropped the first syllable, instead of contracting the first two into
one. Thus tuli and scidi were anciently tetuli and sciscidi. The form
SUPINES OF TWO SYLLABLES. 33
Horat. Lusisti satis, edisti satis, atque bibisti.
Virg. Hie mild responsum primus dcdit tile petenti.
Horat. Demersa exitio ; diffidit urbium. (Choriambic.)
Statius. Gaudia, florentesque manu scidit Atropos annos.
Virg. Explicuit legio, et campo stetit agmen aperto.
[d. Constitit atque oculis Phrygia agmina circumspe.xit.
Td. Cui mater media sese tulit obvia sylva.
OBS. Abscldi, from caedo, has the middle syllable long ;
but abscidi, from scindo, has it short.
Lucan. Abscidit nostrte multum sors invida nostrae.
Id. Abscidit impulsu ventorum adjula vetustas.
SECTION XL
SUPINES OF TWO SYLLABLES.
RULE. Supines of two syllables, and the parts of the
verb supposed to be derived from them, hav? the first syl-
lable long ; as, visum, lusum, motuip. ; visus, visurus, &c '
Virg. Terribiles visu forma ; Letumque Labor que
Id Lenaeos, ea visa salus morientibus una.
Id. Nascitur et casus abies vlsura marines.
Horat. Lusum it Maecenas, dormitum ego Virgiliusque.
Vhg. Quos ego sed motos prasstat componere jluctus.
tetuli occurs in Plautus (Amphitr., 2, 2, 84, 168. Mentzchm., 4, 2, 25,
66. Panul., 3, 1, 58. Rud. prol., 68); in Accius and Caecilius (ap.
Non, , 839); in Catullus (Carm., 63, 47, 52; 66, 35); in Terence
(Antr., 4, 5, 13 ; 5, 1, 13); and in the grammarians Diomedes (lib. 2,
p. 435) and Priscian (10, 6, p. 497, seq.). Of the form sciscidi Priscian
gives Oiamples from Afranius, Attius, Naevius, and Ennius (lib. 10, p.
890. Oompare Aul. Gell., 7, 9). On the same principle, fidi would
be originally fiftdi, from fido. The form libi is an actual reduplication
from bio, the same as the Greek mu. Bibo in the present arose from
the digarmnatized form IIIFQ. As for steti and sfili, they are merely
different forms of the reduplication of sto, just as dedi and didi from do.
(Ramsay's Lot. Pros., p. 100. Struve, Ueber die Lat. Decl., &c., p.
160. Pot', Etymol. Forschungen, vol. 1, p. 188.)
1. The Sanscrit infinitive is preserved in the Latin first supine ; as,
paRtum, alitum ; samtum, cinctum. (Kennedy, Researches, p. 256.
Wilkins's Sanscrit Grammar, p. 123.) The old form of visum ap-
pears to have been vidsum ; of lusum, ludsum ; of motum, movitum
(moilum) ; of fletum, flemtum ( fleitum) ; of minutum, minuitum ; of
34 SUPINES OF TWO SYLLABLES.
EXCEPTIONS.
In the following dissyllabic supines the first syllable is
short ; namely, dtum, from cieo ; datum, from do ;
itum, from eo ; titum, from lino ; guitum, from queo ;
ratum, from reor ; rutum, from ruo ; satum, from sero ;
situm, from sino ; and futum, from fuo, whence fu-
turns.
Horat. Puppes sinistrorsum cites. (Iambic.)
Virg. Intraro, gentique mea data mania cernam.
Lucret. Nee repentis itum quojus vis cumque animantis.
Ovid. In te fingebam violentos Troas ituros.
Virg. Ardentes auro ct paribus Kta corpora guttis.
Id. Nos abiisse rati, et vento petiisse Mycenas.
Virg. Impulerat torrens, arbustaque diruta ripis.
Id. Deinde satis fluvium inducit rivosque sequentes.
Horat. Aurum irrepertum et sic melius situm. (Alcaic.)
Id. Quid sitfuturum eras fuge qu&rere. (Alcaic.)
OBS. 1. There is some doubt with regard to the quantity
of statum. In Lucan, Martial, Statius, and Claudian, we
find stdtura, constdtura, obstdtura, pr<Bstatura, with the a
long; while 'the derivatives, stdtim, status both substan-
tive and adjective, stdtio, stdtor, are used by Catullus,
Ovid, and others, with the first short. It seems best, how-
ever, to follow the authority of Priscian, and make the first
syllable of the supine long. 1
OBS. 2. Citum, from cieo of the second conjugation, has
the first syllable short; whence citus, "quick;" concitus
and excitus, " aroused ;" but citum, from do of the fourth
acHtum, acmtum, <fec. Contraction rendered the syllable long. But
futum, rutum, &c., are formed by syncope, and therefore continue
short.
1. " Steti vero statum supinum penultima producta debet facere."
(Priscian, 9, p. 863, ed. Putsch.) The variation in quantity, noticed
in the text, seems to have arisen from the prosodial difference in statum
and slitum, as we see exemplified in prastitum and prastatuin, which
are both attached to prasto as its supines. (Ramsay's Latin Pros., p.
>93. Vossius, de Art. Gramm., 2, 22.)
POLYSYLLABIC SUPINES. 35
conjugation, has the first syllable long ; whence dtus
" excited ;" accitus, concltus, excitus, and indtus. 1
Virg. Altior insurgens et cursu concitus keros.
Ovid. Necfruitur somno vigilantibus exctta cuns.
Lucan. Unde ruunt toto condta pericula mundo.
Id. Rupta quies populis, stratisque exdta juventus.
OBS. 3. Ruo has ruitum and rutum in the supine. Its
compounds form the supine in utum having the penult
short, whence dirutus, erutus, obrutus. 2
Ovid. Diruta sunt aliis, uni mihi Pergama restant.
Virg. Nee mihi cum Teucris ullum post eruta bellum.
OBS. 4. It must be borne in mind, as regards the deriva-
tives of itum, that the participle is ambitus, but that the
substantive is ambitus.
Ovid. Jussit et ambita circumdare littora terras.
Horat. Et proper antis aquae per amoenos ambitus agros.
SECTION XII.
POLYSYLLABIC SUPINES.
RULE 1. Supines in utum, of more than two syllables,
and the parts of the verb formed from them, have the pe-
nult long; as, solutum, volutum, minutum, argutum?
1. But scltum is always long, whether it comes from scio or scisco.
Stilus, from scio, signifies " skilful," " graceful," &c. ; but scltus,
from scisco, "ordained," "decreed;" whence we have plebis-scitum,
" a decree of the commons."
2. It has been supposed, and with great appearance of probability,
that tn, in the supine of ruo, was pronounced as one short syllable,
runtum ; whence diruntum, enritum, &c., which afterward became
dirutum, erutum, on the dropping of the i. (Ramsay's Lat. Pros.,
p. 103.)
3. Solutum and volutum do not come from solvo and volvo, where v
is a consonant, but from the earlier forms soluo and voluo. The supines
in question will, according to this view, have been originally solmtum,
voluitum, changed afterward by contraction to solutum, volutum. The
same principle of contraction operates, as already remarked, in the case
of other supines in utum. Thus minutum is contracted from minuitum ,
urgutum from arguitum ; tributum from tribuilum, &c.
35 POLYSYLLAUIC SUPINES.
Virg. Lumina rara micant, somno vinoque soluti.
Id. Ecce autem flammis inter tabulata volutus.
Lucret. Conveniebat enimfruges quoque seepe minutas.
RULE 2. Supines in itum, from preterites in twt, and the
parts of the verb formed from them, likewise have the penul
long ; as, cupivi, cupitum ; petivi, petitum ; condivi, conditum,
<fec.'
Ovid. Mens videt hoc, visumque cupit, potiturque cupito.
Virg. Telaque trunca viri et bis sex thoraca petitum.
Horat. Ne male conditum jus apponatur ; ut omnes.
RULE 3. But supines in itum, from preterites not in ivi,
have the short ; as, monui, monltum ; tacui, taciturn, &c. 2
Virg. Discite justitiam moniti, et non temnere divos.
Id. Quis te, magne Cato, taciturn, out te, Cosse, relinquat
Id. Conditus in nubem, medioque refugerit orbe.
OBS. The rule last given, however, does not extend to
polysyllabic compounds from supines of two syllables.
1. Here also, as in the previous rule, contraction operates to lengthen
the penult. Thus we have, oldest form petnitum (pethtntum), by throw-
ing out the v (or rather w>), pelntum, and, by still farther contraction,
pi Alum. So, also, cupimtum (cupiicttum), cupntum, cupilum; condirn-
lum (condlwHum), condntum, conditum, &c.
2. In the case of supines in utiim and Itum, the contraction is effect-
ed by a blending of the two vowels into one long ; but in supines in
ilum, a syncope operates, or, in other words, one of the vowels is drop-
ped, and the remaining one is left with its short quantity. Thus,
monuitum, monltum ; lacuitum,- taciturn, &c. Recensitum is often
adduced as an exception to the rule given in the text. It is only, how-
ever, a deviation in appearance, being formed, not from reccnsui, but
from the old perfect recensivi. The simple verb censeo made censui
and ccnslvi in the perfect, censum and censitum in the supine. Hence
we find, in an old inscription, censita sunt for censa sunt ; and, in some
of '.be writers on the civil law, censiti for censi ; so, also, the noun
censor is a contraction from ctnsitor. and occurs in this latter form in
other inscriptions which have come down to us, as well as in the wri-
tings of the ancient lawyers. (Ulpian, Dig., lib. 50, tit. 15, leg. 4.
Inscript. ap. Grut., p. 417, n. 5. Ib., p. 439, n. 5. Orell., Inscript.
Lat., n. 208, 3044, 3652.) Analogous to this is the verb pono, which
made in the perfect posim as well as posui. Plautus uses posirimus
(Vidul. ap. Prise., 10, 7, p. 499, ed. Putsch); Apuleius apposivcrunt
(ap. eutid.); Lucilius opposivit (Fragm., p. 211, ed. Bip.) ; Catullus
defoxiril (34, 8), &c. (Compare Burmann, ad Anihol. Lot., 2, 130, 16.)
DERIVATIVES. 37
These follow the quantity of the simple supines from
which they are formed; as, itum, obitum; datum, abdttum;
condttum, reddltum ; satum, insifum, &c., except cogmtum
and agnitum, from notum. 1
SECTION XIII.
DERIVATIVES. 2
RULE. Derivatives follow the quantity of the words from
which they are formed.
OBS. 1. This rule applies strictly to the modifications
which words undergo in declension, comparison, and con-
jugation, in so far as those syllables are concerned which
are not affected by the inflections.
Thus, since the first syllable in arils is short in the nomi-
native, it remains short in arietis, arieti, &c., since its form
does not change with the inflections ; but the same cannot
be said of the last syllable, which is long in the nominative,
although in the oblique cases the corresponding syllable is
short: arietis, arieti, &c. So also from mitis comes mitia;
diirus, durior, durissimus, &c.
In like manner, the first syllable in lego being short, it
remains short in all the tenses which are formed from the
present ; as, legebam, legam, legerem, &c. ; and, on the
other hand, the first syllable ' in the preterite legi being
long, it will be long in legeram, legerim, legissem, &c., and
all other parts of the verb formed from the preterite.
OBS. 2. The rule applies to all words which are clearly
and distinctly formed from other words, by the addition of
certain terminations or suffixes, according to well-estab-
lished analogy.*
1. Many of the Latin supines are simple contractions, made long by
the rule of position ; thus, legitum, leg'lum, lectum ; rumpilum, rump 1 '
turn, ruptum ; nubitum, nub'tum, nuptum ; scribilum, scriVtum, scrip-
turn ; docitum (dokitum), doc'tum, doctum, &c.
3. Ramsay's Latin Prosody, p. 9, seqq.
3. Still, however, there are some exceptions to this. Thus, from
D
DERIVATIVES.
Thus, from animus w* have animosus,
ndtura
rosa
viola
sanguinis
pulveris
labor
" naturalis,
" rosetum,
" vwlarium,
" sanguineus,
" pulverulentus,
" laboriosus.
OBS. 3. But when two words are merely connected to-
gether by derivation from a common root, we cannot, even
when they resemble each other in structure, with any cer-
tainty infer that the quantity of the corresponding syllables
will be the same ; for, although this happens much oftener
than otherwise, yet the exceptions are too numerous to ad-
mit of the principle being broadly stated.
OBS. 4. Among the exceptions alluded to in the prece-
ding remarks, the following are most deserving of attention.
1. Several kindred verbs which have two forms, one
active and the other neuter, or which differ otherwise
in meaning, differ also in quantity. 1 Thus,
placare,
sedare,
legare,
dtcare,
labare,
placere,
sedere,
legere,
dtcere,
labi.
sidere, sedes,
2. Words which differ in meaning, but which are spelt
in the same way, often differ in quantity. This arose,
probably, from the pronunciation being purposely va-
ried, to prevent confusion. Thus,
moles we have molestus ; from rego, regula ; from luceo, lucerna, &c.
With regard to these and others of the kind, we may use the language
of Varro : " Cum in vestitu, adificiis, sic in supellectile, cibo, ctuereis
omnibus, quiz us** td vitam sunt adsumpta, dominetur incufualitas ; in
sermone quoque, qui est usiis causa constitutus, ca non repudianda."
1. Exceptions, however, to this remark not unfrequently occur.
Thus, we have da.ra.re and cldrere ; rigare and rigere ; fugire and
fugere ; jarf-e and jaeere, &c.
COMPOUND WORDS. 39
duds,
from
duco,
but
duds,
from dux.
regis,
M
rex,
"
regis,
" rego.
legis,
u
lex,
"
legis,
" lego.
voces,
Cf
vox,
u
voces,
" voco.
OBS. 5. The entire class of verbs in tm'o, called desid-
eratives, have the u short, though derived from the future
participle in urus, of which the penult is invariably long ;
as, parturit, esurit, cwnaturit, nupturit, &c.
OBS. 6. The following deviations in quantity, on the
part of other derivatives, are also worthy of notice.
homo, but humanus.
humus, humilis, humare, but humor, humidus, humeus, &c.
persono, but persona.
tego, but tegula.
macer, mdceo, modes, but macero.
lateo, but laterna.
sopor, soporus, soporifer, soporo, but sopio, sopitus.
quater, but qualuor.
dlcere, but dtcax.
vado, but vddum.
SECTION XIV.
COMPOUND WORDS.
RULE. Compound words retain, in general, the quantity
of the simple words from which they are formed.
Thus, in perlego, relego, the middle syllable is short, be-
cause it is short in the simple lego. But in the perfects
perlegi, relegi, it is long, because lengthened in the simple
legi.
Again, attigi, concidi, diffidi, ebibi, rescidi, have the pe-
nult short, because the corresponding vowel is short in
their primitives, tefigi, cecidi,ftdi, libi, and scidi.
Obtttum, from obUno, has the short penult, because the
vowel is short in the simple litum. But ollitus is from
Mini t cor.
40 PREPOSITIONS IN COMPOSITION.
OBS. The quantity of the simple words is preserved in
the compounds though the vowel be changed. Thus,
concido, e.xcido, incido, occido, recido, from cado, shorten the
penult ; and, in like manner, eRgo, seligo, &c., from lego.
On the other hand, concido, excido, incido, recido, occido,
from cado, have the penult long. So also allido, from
leedo ; exquiro and requiro, from qu&ro, &c.
EXCEPTIONS.
The following shorten the penult, though the corre-
sponding vowels in the simple words are long ; as,
dejero, pejero, from juro,
pronuba, innuba, " nubo,
maledlcus, causidicus, &c., " dico,
semisopitus, " sopio,
cognitum and agriitum, " notum.
OBS. 1. The participle ambitus, as has already been re
marked, has the penult long ; but the noun ambitus has it
short. 1
OBS. 2. The second syllable in connubium is naturally
short, but it is occasionally lengthened by the poets in the
arsis of the foot.
Virg. Connubio jungam stabili, propriamque dicabo.
Id Hectoris Andromache, Pyrrhin' connubia servos ?
SECTION XV.
PREPOSITIONS IN COMPOSITION.
RULE 1. The prepositions a, e, de, di, and se, in compo-
sition are long, except di in dtrimo and disertus. s
1. Besides ambio, ambitum, a simple derivative from ambe, there'
probably was also amb-eo, amb-itum, a compound from eo.
2. The old form of dirimo was disimo, and hence the inseparable
preposition in dirimo and discrtus is in fact dts, not di. In disertvt
the original form appears to have been dissertus (the participle of
dissero), and one of the two *'s was subsequently dropped.
PREPOSITIONS IN COMPOSITION. 41
Virg. Amissos longo socios scrmone requirunt.
Prop. Conjugis Evadne miseros elata per ignes.
Virg. Deducunt socii naves, et littora complent.
Id. Tergora diripiunt costis et viscera nudant.
Lucret. Dissidio potis est sejungi seque gregari.
Virg. Cede deo dixitque et proelia voce diremit.
Mart. Causas inquit agam Cicerone disertius ipso.
HRuLE 2. Re is short ; as, relinquo, refer o ; but the im-
personal verb refert ("it concerns," &c.), from the noun
res, has the first syllable long. 1
Ovid. Propellit Boreas, astus et unda refert. (Pentam.)
Virg. Pr&terea nee jam mutari pabula refert.
\L
RULE 3. The prepositions ab, ad, in, ob, per, and sub,
are short in composition before vowels, since they are
short in their simple state ; as, abeo, ddoro, ineo, obeo,
perambulo, subigo.
1. The impersonal refert is only mentioned here in order to guard
against the possibility of a mistake in pronunciation, since it is not a
compound of the inseparable preposition re, but comes, according to
some, from rci, the dative, according to others from re, the ablative, ol
the noun res, and the verb/ero. Verrius Flaccus, the ancient gramma
rian, as cited by Festus, was in favour of the dative : " Refert cum
dicimus, errare nos ait Verrius, esse enim rectum rei fert, dativo scilicet,
non ablativo casu ; sed esse jam usu posseesum." (Festus, p. ccvii., ed.
Seal.) Reisig, on the contrary, maintains that refert comes from the
ablative re and the verb/ert, and makes refert mea, for example, equiv-
alent to re fert mcd, " it brings something to bear in my case." In
the same way some modern scholars imagine that interest mea is to be
explained, and they advance the opinion that inter and prater in early
Latin governed the ablative. Hence they account for the quantity of the
final syllable in intered, prcetered. (Reisig, Vorlesungen, p. 640, ed.
Haase. Benary, Romische Laictlehre, vol. 1, p. 37. Hartung, iiber
die Casus, p. 84. Bopp, Vergleich. Gramm., p. 215. Schmid, df
Pronom., p. 79.) An opposite doctrine, however, with regard to the
etymology of refert, is maintained by Perizonius (ad Sanct. Min., 3, 5.
vol. 1, p. 581, ed. Bauer), and after him by Scheller (Lot.
Dcutsch. Worterb., vol. 4, col. 9188, seqq.). both of whom deduce it
from refero, and regard the change of quantity merely as an expedient
for distinguishing between refert and refert, with their different signifi-
cations.
D 2
42 PREPOSITIONS IN COMPOSITION.
Virg. Tityrus hinc dbcrat ; ips& te Tityrc pinus.
Id. Bella gero : ct quisquam numen Junonis adoret f
Id. Et lituo pugnas insignis obibat et hasta.
Id. Prima levcs ineunt si quando proelia Parthi.
Horat Recte necne crocum floresque perambulet Atta.
Virg. Arvina pingui, subiguntque in cote secures.
_
OBS. 1. Ob in composition sometimes drops the b before
a consonant, in which case the o remains short ; as,
omitto.
Horat. Pleraque differat, et pr&sens in tempus omit tat.
OBS. 2. Trans frequently drops the last two letters in
composition, but preserves its proper quantity ; as, trano
(from transno) ; traduco (from transduce) ; trado (from
transdo).
Virg. Illafretus agit ventos, et turbida tranat.
Id. Atque satas alio vidi traducere messes.
Id. Tradit equum comiti, paribusque resistit in -armis.
RULE 4. Pro is short in Greek words ; as, Propontis,
Prometheus ; but in Latin words it is usually long ; as,
prodo, procudo, prdcurvus, &c.'
Ovid. Fas quoque ab ore freti longteque Propontidos undis.
Virg. Caucasiasque refert volucres, furtumque Promethei.
Id. Proderc voce sua quemquam out opponere morti.
Id. Maturare datur ; durum procudit orator.
Id. Exoritur proeurva ingens per littora fletus.
OBS. 1. There are, however, some Latin words in which
1. " It is impossible," observes Ramsay, " to agree with Dr. Carey
in supposing that pro was in reality always doubtful, and lengthened or
shortened as might suit the convenience of the poets. Since we find
BO many words in which it is uniformly long, a few in which it is always
short, and not above two or three at most in which it is doubtful, 'such
an hypothesis must be pronounced extravagant." (Ramsay's Latin
Prosody, p. 89.)
a, e, t, w, AND y, IN COMPOSITION. 43
pro is uniformly short, viz., the compounds of cello, fanum,
fari, fateri, festus, fugio, fundo, fundus, nepos, neptis
torvus ; as, procello, procella, profari, profano, prof anus,
profiteri, profestus, profugio, profugus, profundo, profundus,
pronepos, proneptis, protervus, protervitas, to which add
proficiscor, profectus, profeclo. 1
y OBS. 2. The following have the pro doubtful, namely,
propago (both noun and verb) and propino. To which
some, without sufficient grounds, add procumbo, procuro,
prdpello, which have the first syllable always long in the best
writers ; and profari, profundo, in which it is always short. 2
Virg. Sylvarumque alias presses propaginis arcus.
Id. Sed truncis oleae melius, propagine vites.
Id. Sit Romano potens Itala virtute propago.
Lucret. Propagare genus possit vitamque tueri.
Id. Ecficis ut cupide generatim saecla propagent.
Mart. Crystallinisque murrhinisque propinat. (Scazon.)
Id. PraBstare jussi, nutibus propinamus. (Scazon.)
SECTION XVI.
^ A, E, I, U, AND Y, IN COMPOSITION.
/ RULE 1. If the first part of a Latin compound end in o,
r that vowel is long ; as, trdno, trdduco, trddo ; 3 but if it end
in e, the e is in general short ; as, trecenti, nefas.*
Virg. Expertes belli juvenes ; ast Ilva trecentos.
Juv. Credebant hoc grande nefas et morte piandum.
1. Ramsay's Latin Prosody, p. 84.
2. In the two lines from Virgil, quoted immediately after, in which
propago is used in its primitive sense, of the sucker or layer of a tree
or shrub, the pro is long ; in all other places it is employed in the figu-
rative sense of progeny, race, stock, and has the first syllable uniformly
short. (Ramsay's Lat. Pros., p. 88.)
3. These, as has already been remarked, were originally written
transno, transduco, transdo ; and hence the quantity of the a in the
initial syllables.
4. The quantity of the a in Greek compounds is to be learned from the
rules of Greek prosody. Thus a is sometimes shor ; as, adipsos,
sometimes long ; as, Nedpolis, genealogus.
44 0, P, , U, AND y, IN COMPOSITION.
Or.s. 1. Verbs compounded of facio or Jlo have the
almost uniformly short, and not common, as is generally
maintained. In the following the e is always short : Cale-
facio, calefio, calefacto ; labefacio, labefio, labefacto ; made'
facio, madefio ; pavefacio, pavefio ; rubefacio, rubefio ; stu-
pcfacio, stupefio ; tremefacio, tremefio ; tumefacio, tumefio. 1
OBS. 2. The only verbs in which any doubt exists re.
specting the quantity of the e, are patefacio, putrefacio,
tepcfacio, and liquefacio ; and yet even in these the vowel
is much more frequently short than long. 2
OBS. 3. The exceptions to the rule of e short in compo-
sition are, for the most part, based upon peculiarities of
derivation. Thus the following have the e long, namely,
nequis, 3 nequa, nequod, &c., nequitia, nequam, nequaquam,
nequidquam, nequando ; videlicet, 4 vencfica, 5 semodius, 6 se-
mestris,'' sedecim, s &c. Martial, however, makes the first
syllable of selibra short in several instances, and never
long.
RULE 2. If the first syllable of a compound terminate in
1. Ramsay'' s Latin Pros., p. 94.
2. Patefacio has the e long only in two instances, both of which oc-
cur in Lucretius (4, 346, and 6, 1000) ; to which some add a line quo-
ted from Ennius by Isidorus : " fade patefecil radiis rota Candida
ccelum." The verb putrefacio is found with the long vowel in one
passage of Lucretius (2, 898) ; and so likewise one solitary instance of
the long e is found in lepefacio. (Catullus, 64, 361.) Two instances of
long e occur in liquefacio. (Catull., 90, 6. Or., Met., 7, 161.)
3. The difference in quantity between necesse, nefas, nefandus, ne- ' t
fastus, nefarius, nequeo, and ncquis, nequam, nequitia, &c., is proba-
bly owing to the following circumstance, that in the former class of
words the ne was formed by apocope from the conjunction nee, and so
retains its original quantity ; whereas in the latter it is either the adverb
ne, which is always long, or else the c of nee was in these cases retain-
ed in pronunciation, though omitted in writing.
4. From ridcre and licet, just as scilicet comes from scirc licet, and
ilicct from Ire licet.
5. From venenum and facio.
6. By contraction from semimodius.
7. Semcstris, " half monthly," is by contraction from semis and
mensis ; and semcstris, " half yearly," from sex and mentis. In either
case, therefore, the e is long.
8. From tcxdtcim.
a, e, t, u, AND y, IN COMPOSITION. 45
i or , the vowel is generally shortened ; as, omnipotens,
causidicus, biceps, ducenti, quadrupes.
Virg. Turn pater ommpotens, rerum cui summa potestas'.
Mart. Et te patronum causldicumque putas. (Pentam.)
Virg. Tollit se arreclum quadrupes, et saucius auras.
RULE 3. When y terminates the first member of a Gretk
compound, that vowel is short ; as, Thrasybulus, Euryp'j-
lus, Polydamas, polypus; unless rendered common by a
mute and liquid, or long by position ; as, Polycletus, which
has the y common, and Polyxena, in which it is long.
Auson. Anna superveheris quod, Thrasybule, tua. (Pentam.)
Ovid. Vel, cum De'iphobo, Polydamanta roga. (Pentam.)
EXCEPTIONS.
Exc. 1. But in ludimagister, lucrlfacio, lucrlfio, and com-
pendifacio (which are properly not compounds, but
each a combination of two distinct and complete
words), the i is long ; and the same may be said of
agricultura, though the i is short in the compound
agricola. 1
Exc. 2. The masculine idem, 2 as also bigce, quadriga, 3
siquis, siqua, siquod, 4 scilicet, 5 bimus, 6 trimus, quadri-
1. Tubtcen, according to the general rule, has the i short, whereas in
tiblcen the middle syllable is long, because, as has already been remark-
ed, it is a crasis of two short vowels into one long, from the original
tiKicen or tilnacen. In liiicen (by syncope from lituicen) the penult MI
short.
2. The old form for the masculine idem was isdem (is-dcm), where the
i was long by position, a quantity retained after the s was dropped.
The neuter, on the other hand, has the short vowel (i-dem), with which
we may compare the Sanscrit i-dam. (Bopp, Vergleich. Gramm.,
p. 500.)
3. Bigee and quadriga: have already been explained, being mere con-
tractions from bijuga, quadnjuga. (Vid. page 24.)
4. In slquis, slqua, siquod, the long i follows the. analogy of si.
5. Scilicet is from scire licet, just as llicet is from ire licet.
6. Bimus, trimus, quadrlmus, &.C., are, according to Scaliger and
Vossius, contractions from biennis, triennis, quadricnnis. ( Voss., Ely-
mol. lMt. t s. v.)
46
a, e, ', w, AND y, IN COMPOSITION.
, quivis, quldam, quilibet, 1 tantldem, 2 bidu:tm, tridu*
urn, 8 quotldie* and the other compounds of dies, have
the i long.
Virg. Omnibus idem animus, scelerata excedere terra.
Mart. Si totus tibi triduo legatur. (Phalaecian.)
Phaedr. Nam vita mort i propior est quolidie. (Iambic.)
Terent. Quotidianas vita consuetudinem. (Iambic.)
Phaedr. Idem faclurum melius se postridie. (Iambic.)
Eim. Tantldem, quasi feta canis, sine dentibu 1 latrat.
Exc. 3. In ubicunque and ubivis, the t, as in the primi-
tive ubi, is common ; while in ubique and ibidem the
middle syllable is generally long; though, strictly
speaking, it should also be regarded as common. 5
1. In qulvis, qutdam, and quilibet, the i follows the quantity of the
simple qul. The termination dam, in quidam, is compared by Bopp
with the similar termination in the Sanscrit i-dam. (Vergleich. Gramm.,
p. 500.)
2. Tantidcm has the short penult in Varro (ap. Non., c. 7, n. 105).
" Hospcs quid, miras nummo curare Serapim 1
Quid ? quasi non cures tantldem Aristoteles."
Supposing the reading to be correct, we may conjecture that tantldem
was formed by crasis from tan/ndcm, but tanfidem, on the other hand,
by syncope. The word indidem (from indcidem) affords an argument in
favour of this supposition. (Carey, Lat. Pros., 10, p. 49.)
3. Originally, perhaps, bisdunm, trtsduum.
4. Quolidie has the t everywhere long, except in Catullus (68, 139),
where we have quottdiana. This, however, is either a corrupt reading,
as some maintain (Docring, ad Catull., 1. c. var. lect.), or else, in scan-
ning, we must read quotlidiana, and pronounce it quottid-ydna, making
the line a Spondaic one.
5. The middle syllable in ubique should be considered common, for
the reason that the corresponding syllable in ubicunque and ubivis is re-
garded as such, they being all derived from ubi. The authorities, how
ever, to prove that the middle syllable of ubique was short as well as
long, are very rare. Wasse cites two lines from Plautus (Bacch., 5, 1,
1. Cos., 2, 3, 38), in which he maintains that ubique is found with the
short penult ; and he. then goes on to remark : " Totum discrimen in
hoc rcrti ridetur, quod ubi, cum dcsinal in ancipitcm, quotiet adjicitur
particula copulativa, ultimam corripiat ; quoties cxpletiva, producat :
apud reccntiores quidem, discretionis causa ; non item apud vetustis-
simos." The difference in quantity here alluded to. though it may not
hTe any very strong arguments in its favour, may yet be recommended
"discretionis causa," and will be found not inconvenient. Foi exam-
ple, in reading Sail us l, where ubique, in the sense of et ubi, frequently
IN COMPOSITION. 47
SECTION XVII.
IN COMPOSITION.
RULE 1. In words of Greek origin, where the first part
of the compound ends in o-micron (or short 0), that vowel
is short ; as, schoendbates, bibliopola, aredpagus, Cleopatra ;
unless rendered common or long by position ; as, chiro-
graphus, Hippocrene, Phildxenus, Nicdstratus.
Juv. Augur, schoenobates, medicus, magus, omnia novit.
Mart. Non habeo, sed habet bibliopola Tryphon. (Pentam.)
Lucan. Hesperios auxit tantum Cleopatra furores.
RULE 2. But if the first part of the compound word end
in an o-mega (or long o), the o is long in Latin ; as, Mind-
taurus, gedmetres, gedgraphus, lagdpus.
Virg. Mindtaurus inest, Veneris monumenta nefandas.
Mart. Si meus aurita gaudet lagdpode Flaccus.
OBS. In compound Latin words, the o is sometimes long ,
as, alidquin, quandoque, ndlo ; and sometimes short ; aa
quandoquidem, hodie, duodeni. 1
Horat. Mendosa est natura, alidquin recta, velut si.
Id. Indignor quandoque bonus dormitat Homerus.
Propert. Hanc utinam faciem ndlit mutare scnectus. '
occurs, we may pronounce the word with the penult shortened, ubique,
reserving the long penult for ubique, when it signifies "everywhere," "in
every place." With regard to ibidem, though found long in the best
writers, it nevertheless occurs with the penultitna short in Juvencus and
Mamercus, whose authority, though by no means equal to that of Hor-
ace and Virgil, is still not to be overlooked in a case of this kind. Plau-
tus also shortens the penult of the same word in several instances.
(Merc., 2, 3, 99. Most., 2, 2, 51. Trin., 1, 2, 166. Captiv., 4, 2,
94. Bacch., 2, 3, 79. Stick., 2, 3, 12.)
1. Since quando has the last syllable common, it is more than proba-
ble that the o was likewise common in both quandoque and quandoqui-
dem, although no instances can be adduced of a short penult in the case
uf the former, or a long antepenult in that of the latter. In like man-
ner, as duo has the o common, the same vowel may have been common
likewise in dvodeni.
48 INCREMENTS OF NOUNS.
Virg. Dicite, juandoquidem molli consedimus herba.
Horat. Non dices hodie, quorsum hac tarn putida tandem.
Virg. P?r duodena regit mundum sol aureus astra.
SECTION XVIII.
INCREMENTS OF NOUNS.
I. By the increments of nouns is meant the syllable or
syllables by which an oblique case exceeds the nomina-
tive.
II. When a word of one or more syllables increases, the
penult is regarded as the increment. Thus, the re in re-
gis, from rex, is the increment, the last syllable never be-
ing considered such in any word.
III. When there are more increments than one, which
seldom happens but in the plural, they are to be reckoned
in the retrograde order, beginning with the penult.
IV. If a noun has one syllable, in an oblique case, more
than in the nominative, it is said to have one increment or
increase ; as,
i
rex, re-gis,
i
sermo, ser-mo-nis.
V. The quantity of the increment of all the other oblique
cases is regulated by that of the genitive ; as, sermoni,
sermoncm, sermonibus, &c., in which the o is long, because
the o of sermonis is long. There is but one exception to
this remark, namely, bobus, from bos, bovis ; but this is, in
reality, a contraction from bovtbus. 1
VI. Nouns in general have but one increase in the sin-
gular ; but iter, jecur (when its genitive isjecinoris), supel-
lex, and the compounds of caput ending in ps, have two
increments. Thus,
1. Bovibus is the same as bowibus, which by contraction become
bowbus, or, when written with the long o, bobus.
INCREMENTS OP THE FIRST DECLENSION. 49
1 2
iter, i ti ne-ris.
jecur, je- ci -no-ris.
supellex, supel-lec-ti-lis.
anceps, an- ci pi-tis.
VII. The dative and ablative plural of the third declen-
sion, in ibus, have generally two increments ; as,
1 2
ser mo-ni-bus.
But the words mentioned in the previous paragraph have
' three increments in these two plural cases. 1 Thus,
1 2 3
*' ti neri-bus.
je ci-noribus.
supel-lf.c- ti - It-bus.
an ci pi ti-bus.
SECTION XIX.
INCREMENTS OF THE FIRST AND SECOND DECLEN-
SIONS.
FIRST DECLENSION.
RULE. The vowel a, in the old increment of the first
1. The uncommon increase of these words is owing to their having
come originally from nominatives, now obsolete, which consisted of a
greater number of syllables than the nominatives to which they are at pres-
ent assigned. Iter has for its regular formation in the genitive ileris; and
Charisius (p. 108) adduces examples of this form of the genitive from Pa-
cuvius and Hyginus. Priscian (p. 695) finds this same form in Nsvius.
The ablative itere occurs in Lucretius (5, 652). Be.sides iter, however,
there was another form for the nominative, namely, itiner, from which
comes the ordinary genitive itineris, and the other cases similar to it.
(Compare Charis., p. 16, 34, 63, 109. Priscian, p. 646, 659.) This
nominative itiner is found in Lucretius (6, 338). The remarks just
made respecting iter apply equally to jecur, supellex, and anceps
Thus, besides jecur, there was another form for the nominative, jecinur
or jecinor, whence came jecinuris or jecinoris. (Charts., p. 34.
Priscian, p. 707.) So also supellex borrows its oblique cases from an
adjective, supellectilis, supellectile (Charis., p. 34, 67. Priseian, p.
724) ; while anceps obtains them from an old nominative ancipes, which
made ancipitis in the genitive, just as miles makes militis. This
nominative ancipes occurs in Plautus (Rud., 4, 4, 114). In like man-
ner, proEceps borrows its genjtive pracipitis, and other oblique cases,
from the old form pracipes.
E
50 INCREMENTS OF THE SECOND DECLENSION.
declension, is always long ; as, auld'i, aura'i, longai, pic-
fa.. 1
Virg. Aulai in media libabant pocula Bacchi.
Id. jEthereum sensum, atque aurai simplicis ignem.
SECOND DECLENSION.
RULE. The increments of the second declension are
short ; as, miser, miseri ; vir, viri ; satur, saturi ; put,
pueri. 2
Virg. Non ignara mali, miseris succurrere disco.
Id. Arma viri ! ferte arma ! vocat lux ultima victos
Tibull. Turbaque vernarum, saturi bona signa coloni.
EXCEPTION.
Iber and its compound Celliber have the penult of the
genitive long ; as, Iberi, Celtiberi?
Lucan. Quiqueferos movit Sertorius exsul Iberos.
Mart. Vir Celtiberis non tacende gentibus. (Iambic.)
OBS. The increment in ius has already been noticed in
section IV.
SECTION XX.
INCREMENTS OF THE THIRD DECLENSION.
INCREMENT IN a.
RULE. The increment in a of nouns of the third declen-
1. The principle on which this long penult depends has already been
explained. Consult page 17, note 1.
2. In strictness these are no im rements at all, since miser, rzr, satur,
pucr, &.C., are merely contractions from forms in us ; as, miserus, virus,
saturuf!, puerus, &c. The vocative puere in Plautus (Asin., 2, 3, 2)
can only come from a nominative puerus. Other examples of puerus
and pucre are given by Priscian (p. 697 and 738). So, again, the form
vira, of which Festus speaks, could only come from virus. (Fest., p.
411, cd. Amstel., s. v. Querquetulanee. Serv. ad Mn., 12, 468.)
3. These forms, again, like those mentioned in the previous note, are
not actual increments ; on the contrary, the nominatives Iber and Celti'
ber are merely contractions from Jbf.ni and Cdtiberus, in Greek
INCREMENTS OF THE THIRD DECLENSION. 51
sion is chiefly long ; as, vectig'al, vectigdlis ; pax, pads
calcar, calcdris, &c.
Horat. Jurgatur verbis, ego vectigdlia magna.
Ovid. Jariefac aster nos pdcem pacisque ministros,
Mart. Accipe belligerae crudum thordca Minervas.
EXCEPTIONS.
Exc. 1. Masculine proper names in al and ar (except
Car and Nar) increase short ; as, Hannibal, Hanni-
bdlis ; Hamilcar, Hamilcaris ; l Caesar, Casdris.
Sil. Hannibdlem Fabio ducam spectante per urbem.
Id. Cut sasvum arridens narrabis Hamilcaris umbris.
Propert. Gallum per medios ereptum Caesdris enses.
Exc. 2. The adjective par and its compounds ; the sub-
stantive par ; the noun sal, whether neuter or mascu-
line ; and also hepar, nectar, baccar, vas (vddis), mas,
anas, lar, an&jubar, have the increase short'.
Virg. Ardentes auro et pdribus Ufa corpora guttis.
Horat. Damnati populo pdria, atque epulum arbitrio Arri et.
Virg. Vela dabant laeti, et spumas salts cere ruebant.
Mart. Ipsa merum secum portat, et ipsa salem. (Pentam.)
Horat. Quinta parte sui nectdris imbuit. (Choriambic.)
1. Ennius and other early authors wrote Hannibdlis, Hamilcaris,
Hasdrubdlis, with the long penult. Aulus Gellius informs us, that the
grammarian Valerius Probus followed this same pronunciation, and as-
serted that Plautus, Ennius, and many other writers of that period, as
we have already remarked, gave the penult long. He adds, however,
that Probus cited merely one verse from the "Scipio" of Ennius, be-
longing to the trochaic class : " Quique propter Hannibalis capias con-
riderani." (Aul. Gell., 4, 7.) It cannot be denied that this is the more
accurate pronunciation, since Ennius and the others lived nearer to the
Carthaginian times ; still, however, the authority of their successors, who
shortened the penult, is the rule to be followed in modem days. It may
be remarked in favour of the long penult, that this harmonizes with the
etymology of the names in question ; since both Hannibal and Hasdru-
bal obtain the latter part of their form from the Oriental Baal, where the
two vowels coalesce into one long : while Hamilcar is to be adduced
in part likewise, from Milcar, where the final syllabic is also long.
(Consult Gcscnius, Phasn. Man., p. 407. Gronov. ad Gell., I c.)
52 INCREMENTS OF THE THIRD DECLENSION.
V r irg. Errantes hederas passim cum baccdre tellus.
Ovid. Sacra Bonos, mdribus non adeunda, Dea. (Pentara.)
Avien. Latipcdemque andtem cernas excedere ponto.
Tibull. Sed patrii servate Lares, aluistis et idem.
Virg. It portis, jubdre exorto, delecta juventus.
Exc. 3. Greek nouns in a and as increase short ; as
pocmu, poemdtis ; lampas, lampddis.
Horat. Non quivis videt immodulata poemdta judex.
Ovid. Undique collucent prascinctcB lampddes auro.
Exc. 4. Nouns ending in s, preceded by a consonant,
and many nouns in ax, increase short in the genitive ;
as, trabs, trdbis ; Arabs, Ardbis ; fax, facts ; arcto-
phylax, arctophyldcis (and many other compounds of
(frvkat;) ; climax, climdcis ; dropax, dropdcis, &C. 1
Virg. Auratasque trdbes, veterum decora alia parentum.
Horat. Otia diviliis Ardbum liberrima muto.
Virg. Jamque faces et saxa volant, furor arma ministrat.
Mart. Psilothro faciem lavas, et dropdce calvam.
OBS. The proper name Syphax is commonly supposed
to have in the genitive Syphdcis and Syphdcis. This,
however, is incorrect, as the- short quantity turns merely
on an erroneous reading in Claudian, and the true form of
the genitive is consequently Syphdcis. 2
INCREMENT IN 6.
RULE. The increment in e, of the third declension, is
1. These nouns in x are in fact nothing more than nouns ending
in a preceded by a consonant, the letter x being equivalent to cs, gs,
or ks.
2. The line of Claudian occurs in the poem " De Bella Gildonico"
(v. 91), and is as follows : " Compidimus dirum Syphacem, fractumqut
Metello." The passage, however, of which this line forms part, in-
volves a palpable historical error, if the reading Syphacem be retained;
and Hannibalem has therefore been substituted by Barth ; an emenda-
tion approved of by Broukhusius (ad Propcrt., 3, 9, 61) and other schol-
ars, although the common reading is retained by Artaud in Lemaire'i
Collection.
'
INCREMENTS OF THE THIRD DECLENSION. 53
mostly short ; as, grex, gregis ; pes, pedis ; teres, teretis ;
mulier, mulieris.
Ovid. Nobiliumque greges custos servabat equarum.
Fur. Pressatur pede pes, mucro mucrone, viro vir.
Virg. Incumbens tereti Damon sic c&pit olivae,
Afran. Hcec sunt venena formosarum mulierum, (Iambic.)
EXCEPTIONS.
Exc. 1. Iber, Iberis, and genitives in enis, have the pe-
nult long ; as, ren, rents ; Siren, Sirenis ; except that
of hymen, which increases short.
Prise. Quern juxt a, terras habitant Orientis Iberes.
Horat. Si latus out renes morbo tentantur acuto.
Ovid. Monstra maris Sirenes erant, qua voce canora.
Exc. 2. Ver, mansues, locuples, h&res, merces, quies, lex,
rex, plels, vervex, seps, and halec, increase long ; as,
verts, mansuetis, locupletis, haredis, mercedis, quietis,
legis, regis, plebis, vervecis, septs, halecis.
Virg. Vere novo gelidus canis cum montibus humo*
Horat. Tu me fecisti locupletem, Vescere, sodes.
Mart. Edcnt haredes, inquis, mea carmina quando.
Id. Turba gravis pad, placid&que inimica quieti.
Propert. Illorum antiquis onerentur legibus aures.
Horat. Forlunam et mores antique plebis et idem.
Mart. Capparin, et putri cepas halece natantes.
Exc. 3. Hebrew and other foreign names in el increase
long ; as, Daniel, Danielis ; Michael, Michaehs. 1
Alcim. Magnum ilium Dominum atque deum Danielis ado-
" Exc. 4. Greek nouns in es and er increase long ; as,
1 . Hebrew names in el follow the analogy q the long vowel in that
language.
E 2
54 INCREMENTS OF THE THIRD DECLENSION.
tapes, tapetis ; trapes, trapetis; lebcs, lebetis ; soter,
soteris ; crater, crateris ; except aether and aer, which
increase short. 1
Sid. Ap. Jpse per attonitos bacca pendente trapetas.
Ovid. Viginti fulvos operoso ex are lebetas.
Virg. Crateras magnos statuunt, et vina coronartt.
Id. Quacumque ilia levem fugiens secat osthera pennis.
Id. Si nigrum obscuro comprenderit aera cornu.
INCREMENT IN t AND y.
RULE. The increment in t or y, of the third declension,
is for the most part short ; as, stips, sfipis ; nemo, neminis ;
pollex, polUcis ; chlamys, chlamydis ; chalybs, chalybis.
Ovid. Die, inquam, parva. cur stipe qu&rat opes. (Pentam.)
Virg. Qualem virgineo demessum poltice florem.
Ovid. Anchisae sceptrum, chlamydem pharetramque nepoti.
Virg. Insula inexhaustis chaly'bum generosa metallis.
EXCEPTIONS.
Exc. 1. Genitives in inis or ynis, from nouns of Greek
origin, have the penult long ; as, delphin, delphinis ;
Ph'orcyn, Phorcynis ; Salatnis, Salammis.
Virg. Orpheus in silvis, inter delphmas Arion.
Id. Laomedontiaden Priamum Salarmna petentem.
F^xc. 2. The following also have the long penult in the
genitive: Dis, Dltis ; glis, gliris ; vibex, vibicis ;
gryps, gryphis ; Samnis, Samriitis ; Quiris, Quintis ;
Nesis, Nesidis.
Virg. Nodes atque dies patet atri janua Ditis.
Mart. Somniculosos ille porrigit glires. (Scazon.)
Pers. Si puteal multa cautus vibice jlagellas.
Virg. Huic horret thorax Samriitis pellibus ursaR.
1. Greek nouns in es and cr which increase long in the genitive,
have T] in the penult ofi^hat case in Greek. On the contrary, aether and
air have e in the genitive ; as, aiOepof, uepof.
INCREMENTS OF THE THIRD DECLENSION. 55
Luc. Tradite nostra viris, ignavi, signa, Quirites.
Stat. Silvaque qua fixam pelago Nesida coronat.
INCREMENT FROM IX AND yx.
RULE. Nouns in ix or yx have, for the most part, the pe-
nult of the genitive long ; as, bombyx, bombycis ; perdix,
perdicis ; pernix, pernlcis ; coturnix, coturnlcis ; lodix, lo-
dicis.
Propert. Nee si qua Ardbico lucet bombyce puella.
Seren. Seu f el perdicis parili cum pondere mellis.
Virg. Progenuit pedibus celerem et pernicibus alis,
Ovid. Ecce coturntces inter sua pr&lia vivunt.
Mart. Lodices mittet docti tibi terra Catulli.
EXCEPTIONS.
Nix, Cilix, strix, fornix, histrix, choenix, varix, salix,
Jilix, larix, coxendix, pix, calix, calyx, Eryx, Styx,
lapyx, Phryx, onyx, and some others, have their in-
crease short, as have likewise some proper and gen-
tile names ; such as Ambiorix, Vercingetorix ; Bitu-
rix, Caturix,
Virg. Perque nives alium perque korrida castra secuta est.
Luc. Armenios CiUcesque feros, Taurosque subegi.
Propert. Et stngis inventce per busta jacentia pluma.
Calph. Venit ; et hirsuta spinosior histnce barba.
Seren. Saepius occultus victa coxendice morbus.
OBS. 1 . Mastix, mastichis, " a gum," increases short ;
but mastix, mastigis, " a whip," or " scourge," has the in-
crement 1 ong.
Seren. Pulegium, abrotonum, nitida cum mastiche coctum.
Horn. A 2f apa (JHAvrfaag ifiaaev fidarlyt (jxzeivy.
Prudent. Nunc mastlgophoris, oleoque et gymnadis arte.
OBS. 2. Appendix is generally considered as increasing
56 INCREMENTS OF THE THIRD DECLENSION.
short in the genitive ;' and perpendiculum, a noun of kin-
dred origin, has the penult short in Ausonius (Parental.,
5,8).
OBS. 3. Bebryx and sandix have the increment common.
Val. Flacc. Bebrycis et Scythici procul inclementia sacrL
Sil. Ital. Possessus Baccho sava Bebrycis in aula.
Propert. Illaque plebeio, vel sit sandicis amictu.
Gratius. Interdum Libyco fucantur sandice pinna.
INCREMENT IN O.
RULE. O, in the increment of the third declension, in
words of Latin origin, is for the most part long ; as, sol,
solis ; vox, vocis ; victor, victoris ; and all other verbal
nouns in or ; ros, roris ; dos, ddtis ; statio, statidnis ; and
all other feminine verbals in io ; Cato, Catonis, and other
Latin proper names in-o.
Ovid. Regia solis erat sublimibus alta columnis.
Tibull. Ille liquor docuit voces inflectere cantu.
Virg. Nee victoris heri tetigit captiva cubile.
Lucan. Ire vetat, cursusque vagos statibne moratur.
Id. Et mala vel duri lacrymas motura Catonis.
EXCEPTIONS.
Exc. 1. Nouns in o or on, taken from the Greek o>v,
preserve in Latin the same quantity of the increment
which they have in the Greek. If that increment be
an omicron, it is short ; if an omega, it is long.
Thus, sindon, aedon, Agamemnon, Jason, Philemon,
&c., increase short; whereas Laco, Plato, Solon,
Sicyon, &c., increase long.
Mart. Cultus sindone non quotidiana. (Phalaecian.)
Paulin. Si confers fulicas cycnis, et aedona parra.
Horat. Pythagoran, Anytique reum, doctumque Platona.
Mart. Et gratum nautis sidus fulgere Laconum.
1. It is so given, for instance, by Scheller (Lai. Dcutsch. WiJrferi.,*
e.) and Freund ( Worterb. der Lot. Sprache, s. .*
INCREMENT'S OF THE THIRD DECLENSION. 57
OBS. Sidon, Orion, JEgoson, have the increment common,
and so likewise has Britto, " a native of Britain." Saxo,
Seno, and some other gentile nouns, increase short.
Sil. Ital. Stat,fucare colos nee Sidone vilior, Ancon.
Virg. Atque equidem Teucrum memini Sidona venire.
Ovid. Quorum si mediis Bceoton Oriona quceras.
Lucan. Ensiferi nimium fulget latus Orionis.
Claud. Htec centumgemini strictos JEg&onis enses.
Ovid. JEg&ona suis immunia terga lacertis.
Juv. Qua nee tcrribiles Cimbri, nee Brittones unquam.
Mart. Quam veteres braccce Brittonis pauperis, et quam.
F fc. 2. Genitives in oris, from Latin nouns of the neu-
er gender, have the penult short ; as, marmor, mar-
moris ; corpus, corporis ; ebur, eboris.
Ador, however, which is of the masculine gender,
makes adoris and adoris ; it being found short in
Ausonius, and both long and short in Gannius, ao
old poet quoted by Priscian. 1
Auson. Max ador, atque adoris de polline pultificum far.
Gann. Illam sponte satos adoris stravisse maniplos.
Id. Emicat in nubes nidoribus ador adoris.
Exc. 3. Greek proper names in or, and appellatives, a
rhetor, increase short.
Val. Flac. Ingemit et dulcifrater cum Castore Pollux.
Ovid. Et multos illic Hectoras esse puta. (Pentam.)
1. Hence we have adoreus in Virgil, and adorea in Horace and Clau-
dian. It is possible that the variation of quantity in adoris and adorit
may be connected with a difference of gender. Priscian considers advt
an irregular noun, which, though ending in or, still shortens the penulf
of the genitive, and which terminates in or, though of the neuter gender.
For this he is censured by Scaliger, who maintains that ador properly
makes only adoris in the genitive, and that adoris comes from the obso-
lete nominative adus, of the neuter gender. (Seal, ad Fest., vi. Pm-
cian, 6, 9, 49 vol. 1, p. 251, td. Krehl.) Scaliger, however, is himself
in error when he derives ador from the Greek udvp. It is to be traced
to the Sanscrit ad, " to eat." (Lindemann ad Fest., p. 302.)
58 INCREMENTS OF THE THIRD DECLENSION.
Mart. Peleos et Priami transit vel Nestoris &tas.
Id. Dum modo causidicum, dum te modo rhetor a jingis
Exc. 4. Os, oris ; and adjectives of the comparative de-
gree, have their increment long ; as, majoris, pejoris,
melioris.
Virg. Componens manibusque manus, atque oribus ora.
Id. Nate dea, nam te majoribus ire per altum.
Horat. Mutius, indignum quod sit pejoribus ortus.
Exc. 5. The compounds of TTOV$, as, tripus, polypus,
CEdipus ; and also memor, arbor, lepus, bos, compos,
and impos, increase short.
Juv. Stantibus oznophorum, tripodas, armaria, cistas.
Mart. Phineas invites, Afer, et GEdipodas. (Pentam.)
Virg. Strata jacent passim sua quaque sub arbore poma.
Mart. Mavis, Rufe, coquum scindere quam leporem. (Pen-
tam.)
Exc. 6. Cappadox, Allobrox, pracox, and other nouns
which have a consonant immediately before s in the
nominative ; as, scobs, scrobs, ops, inops, ^Ethiops, Ce-
crops, Dolops, increase short ; except Cyclops, Cercops,
and hydrops.
Horat. Mancipiis locuples, eget (Kris Cappadocum rex.
Senec. Materna, letum pracocis mali tulit. (Iambic.)
Juv. A scrobe vel sulco redeuntibus altera ccena.
Virg. Non hcec humanis opibus, non arte magistra.
Ovid. Virginibusque tribus gemino de Cecrope natis.
Manil. Et portentosos Cercopum ludit in ortus.
Ovid. Tela reponuntur manibus fabricata Cyclopum.
INCREMENT IN U.
RULE. The increment in u of the third declension is for
the most part short ; as, murmur, murmuris ; furfur, fur-
furis ; dux, duds ; praesul, praesulis ; turtur, turturis.
PLURAL INCREMENT OF NOUNS. 59
Virg. Aspice, ventosi ceciderunt murmuris aurae.
Seren. Furfuribusque novis durum miscebis acetum.
Pedo. Consule nos, duce nos, duce jam victore caremus.
EXCEPTIONS.
Exc. 1. Genitives in udis, uris, and utis, from nomina-
tives in us, have the penult long ; as, palus, paludis-;
incus, incudis ; tellus, telluris ; virtus, virtutis.
Virg. Quum primum sulcos (Equant sata ? quique paludis.
Mart. Turn grave percussis incudibus osra resultant.
Virg. Vix e conspectu Siculcs telluris in altum.
Id. Mittatur Pallas, quern non virtutis egentem.
Exc. 2. The following also increase long : fur, furis ;
lux, lucis ; Pollux, Pollucis ; and frugis, from the old
nominative frux. But intercus, pecus, and Ligus, in-
crease short.
Virg. Quid domini faciant, audent cum taliafures?
Tibull. Luce sacra requiescit humus, requiescit orator.
Ovid. Pollucem pugiles, Castora placet eques. (Pentam.)
SECTION XXI.
PLURAL INCREMENT OF NOUNS.
The penult of the genitive or dative plural is called the
plural increment of a noun when either of these cases con-
tains more syllables than the nominative plural ; as, muses,
musarum ; ambo, amborum, ambbbus ; res, rerum, rebus ; in
the first of which the syllable sa, in the second bo, in the
third re, are the respective plural increments. 1 So also bi
in nubium, nubibus ; quo in quorum ; qui in quibus.
PLURAL INCREMENTS IN O, 6, 0, i, U.
RULE. In the increase of the plural, a, e, and o are long,
t and u short ; as, quorum, hdrum, Musarum ; rerum, rebus ;
horum, quorum ; quibus, trii vs, montibus, lacubus.
60 INCREMENT OF VERBS.
Virg. Quorum qua: forma pulcherrima, De'inpeam.
Ovid. Cum tamen a turba rerum requieverit harum.
Virg. Jupiter, et rebus nox abstulit atra colorem.
Id. At Capys, et quorum melior sentenlia menti.
Id. Montibus in nostris solus tibi certet Amyntas.
Ovid. Pr&mia de lacubus proximo musta tuis. (Pentam.)
OBS. Bubus has already been explained under section V
SECTION XXII.
INCREMENT OF VERBS.
I. The second person singular of the present tense, in-
dicative active, is the measure by which to estimate the in-
crement of verbs. If any tense or person of a verb do not
contain a greater number of syllables than the above stand-
ard, the verb is said, in that tense or person, to have no in-
crement. Thus, in amat, amant, ama, amem, amans, the
verb amo has no increment, because they all contain only
two syllables, like amas.
II. If, however, a tense or person exceed the given
standard, then, if that excess be by one syllable, the verb
is said to have in that part a single increment ; if by two
syllables, a double ; if by three, a triple ; if by four, a four-
fold increment. Thus, in aMAmus there is a single incre-
ment, which is the penult, for the final syllable is never
called an increment ; in aMABAmws there is a double in-
crement ; in aMAVERIwiu* a triple increment ; and in au-
DIEBAMIni a fourfold increment.
III. In determining the increments of deponent verbs,
we may imagine an active voice, 1 and obtain from this the
requisite standard for the regulation of the increments ; or
we may be guided by analogy, and estimate them by means
1. Most deponent verbs had, in fact, in earlier Latin, an active voice,
which was dropped during the more cultivated period of the language.
Consult Struve, iiber die Lot. Decl. und Conj., p. 80. Reisig, Vorle-
sunken, p. 243, $ 150.
INCREMENT OF VERBS. 61
of other verbs of the same conjugation which have an activo
voice. Thus, for the verb largior, we may either form an
imaginary active, largio, largis, of the fourth conjugation,
or be guided by the tenses of audior, which has a real ac-
tive.
IV. The final syllable, as has just been observed, is
never regarded as an increment. The first syllable, how-
ever, becomes one when the standard tense is a monosyl-
lable. Thus, in the case of do and^Zeo, the tenses by which
we are to estimate their respective increments are das and
fles, and consequently in damns, dabam, dare ; flemus, fle-
bam, fare, the initial syllables are the increments of the
verbs.
VERBAL INCREMENT IN a.
RULE. The vowel a is long in the increments of verbs
of every conjugation ; as, stdbam, stares, properamus, doce-
bdmus, audiebdmini, &c.
Virg. Trojaque. nunc stares, Priamique arx alta maneres.
Ovid. Serins ant citius sedem properamus ad unam.
Horat. Pugndbant armis, qu<R post fabricdverat usus.
Mart. Festinavit Arabs, festindvere Sabcei.
Ovid. Ipse guberndbit residens in puppe Cupido.
Id. Clam tamen intrdto, ne te mea carmina lasdant.
Virg. Et cantdre pares, et respondere pardti.
Mart. Esse videbdris,fateor, Lucretia nobis.
EXCEPTION.
The first increase of the verb do is short ; as, damns, da-
bunt, dare ; and hence the pronunciation of circumda-
mus, circumddbunt, circttmddre ; vennmddbo, venumddre,
&c., with the penult short. 1
1. This violation of analogy on the part of do, to which Priscian al-
ludes (9, 6 vol. 1, p. 453, ed. Krehl.}, may be accounted for by suppo-
sing that the verb was originally conjugated, do, dere, didi, dltum, as we
have it in the compound dcdo (de-do). It would thus have belonged at
F
62 INCREMENT OF VERBS.
Virg. His lacrymis vitam damns, et miserescimus ultra.
Mart. Mille ddbam numos ; noluit accipere. (Pentam.)
Tibull. Quamvis magna ddret, quamvis majora ddturus.
Virg. Taurino quantum pos sent circumddre tergo.
Ovid. Multa rogant utenda ddri, data redder e nolunt.
OBS. The second increase of do, not being excepted, fol-
lows the general rule, and is long ; as, ddbdmus, ddbdtis,
ddbdmur, ddbdtur, ddbdmini.
Virg. Nam quod consilium, out qu<B jam for tuna ddbdtur.
VERBAL INCREMENT IN .
RULE. The vowel e is long in the increase of verbs ; as,
flebam, rebar, lacereris, docerem, legerunt.
Ovid. Flebat Aristaus, quod apes cum stirpe necatas.
Virg. Sic equidem ducebam animo, rebarque futurum.
Mart. D&dale Lucano cum sic lacereris ab urso.
EXCEPTIONS.
Exc. 1. But e before r in the^ir^ increase of every pres-
ent and imperfect of the third conjugation, and also in
the terminations beris and bere, is short ; as, cognos-
cere, legere, legerem, legeremus, legeris, celebraberis,
celebrabere.
Virg. Jam legere, et qua sit poteris cognoscere virtus.
Id. Semper honore meo, semper celebrabere donis.
OBS. 1. But in the second increment, where the word
terminates in reris or rere, the e is long ; as, diripereris,
loquereris, loquerere, proscquerere.
Ovid. Cum consternatis diripereris equis. (Pentam.)
Mart. Hoc tibi Roma caput, cum loquereris, erat. (Pentam.)
Claud. Jungebam Phrygios, cum tu raperere leones.
^
first to the third conjugation, and afterward have been transferred to
the first, with a change of e to a. Such forms, therefore, as circumddre^
renumdare, &c., were originally circumdere, venumdere.
INCREMENT OF VERBS 63
OBS. 2. The forms velim, velis, velit, &c., have the e
short.
Horat. Musa, velim memores : et quo patre natus uterque.
Mart. Esse velis, oro, serus conviva Tonantis.
Exc. 2. The vowel e before ram, rim, ro, of every con-
jugation, is short ; as, amaveram, amaverim, amavero ;
feceram, fecerim, fecero ; and the quantity remains the
same in the other persons ; as, amaveris, amaverit,
amaverimus, amaveritis ; fecerimus, feceritis.
Ovid. Fecerat exiguasjam sol altissimus umbras.
Catull. Dein cum millia multa fecerimus. (Phalaecian.)
OBS. This rule does not apply to those syncopated tenses
which have lost the syllable ve ; as, Jleram, fierim > Jlero ;
the e in these contracted forms retaining the same quantity
which it possessed previously to the syncope ; viz., j?<?(ve)-
ram, Jle(ve}rim, fie(ve)ro.
Virg. Implerunt monies, flerunt Rfipdopeia arces.
Ovid. Nerunt fatales fortia fila de&. (Pentam.)
Exc. 3. The poets sometimes shorten e before runt, in
the third person plural of the perfect indicative active. 1
Virg. Obstupui, steteruntque comes, et vox faucibus hoBsit.
Id. Matri longa decem tulerunt fastidia menses.
florat. Di tibi divitias dedcrunt artemque fruendi.
Tibull. Nee cithara, intonsas profueruntve comas. (Pentam.)
Mart. Nee lua defuerunt verba Thalasse mihi. (Pentam.)
Sil. It. Terruerunt pavidos accensa Ceraunia nautas.
VERBAL INCREMENT IN t.
RULE. The vowel * in any of the increments of verbs is
short, whether such increment be the first, second, third, or
fourth ; as, linquimus, amabtmus, docebimmi, audiebammi,
J. Consult remarks under the article Systole, page 127.
64 INCREMENT OF VERBS.*
&c., and venlmus, comperimus, reperlmus, &c., of the per
feet tense.
Virg. Linquimus Ortygiae portus, pelagoque volamus.
Horat. Infra se positas ; extinctus amabitur idem.
Manil. Victuros agimus semper, nee vivimus unquam.
Claud. Vicimus, exputimus ; facilis jam copia regni.
Plaut. Quapropter id vosfactum suspicamim ',? (Iamb.)
EXCEPTIONS.
Exc. 1. But the i is long in nollto, noUte, nolimus, notitis ,
oeUmus, velltis ; malimus, malltis ; simus, sitis ; and
their compounds, posslmus, adslmus, proslmus, &c.
Mart. Nolito fronti credere, nupsit heri. (Pentam.)
Calp. Credere, pastores, levibus nolile puellis.
Mart. Ne nimittm simus, stultorum more, molesti.
Calp. Possitis, ter quisque manus jactate micantes.
Exc. 2. The penult of the preterite in ivi, of any conju-
gation, is long-; as, petlvi, audlvi.
Virg. Cessi, et sublato montem genitore petivi.
Id. Adventumque pedum, flatusque audlvit equorum.
Exc. 3. In the first increase of the fourth conjugation
whenever a consonant immediately follows, the i v
long ; as, audimus, auditis, audite, audlrem, audire, au
dimur, audltur, audlrer, auditor, audiri ; to which add
the contracted form of the imperfect, audlbam, and the
old form of the future, audibo, which we uniformly
find in Ibam and ibo, from eo, as well as in quibam and
quibo, from queo.
Senec. Audlmur, en, en, sonitus Herculei gradus. (Iamb.)
Virg. Montibus audiri fragor, et resonantia longe.
Horat. Alterius sermone meros audiret honores.
Virg Ferre rotam, et stabulo frenos audite sonant cs.
Id. Tu ne cede malis sed contra audentior ito. _
INCREMENT OF VERBS. 65
/irg. Jungimus hospitio dextras, et tecta subimus.
id. Nutrlbat, teneris immulgens ubera labris.
PropjTt. Lenlbunt tacito vulneru nostra sinu. (Pentam.)
OBS. 1. Venimus, comperlmus, reperimus, &c., of the pres-
ent tense, fall under the previous exception, and are long ;
whereas venlmus, comperlmus, reperimus, &c., of the perfect
tense, have the penult short, as has been mentioned, ac-
cording to the general rule.
OBS. 2. When a vowel, and not a consonant, immediately
follows the i in the first increase of the fourth conjugation,
the i becomes necessarily short by position ; as, audiunt t
audiebam, audiam, audlar, audiens, &c.
OBS. 3. The i in the penult of the first and second per
sons plural of the indicative future perfect, or second future,
and the perfect of the subjunctive is doubtful. 1
Lucret. Quas ob res ubi viderimus nil posse creari.
Catull. Dein cum millia multa feccrimus . (Phalsecian.)
Ovid. Videritis Stellas illic ubi circulus axem.
Id. HCBC ubi dixeritis, servet sua dona, rugate.
Id. Accepisse simul vitam dederitis in unda.
Id. Et maris lonii transieritis aquas. (Pentam.)
Id. Consulis ut limen contigeritis, erit. (Pentam.)
Virg. Egerimus, nosti ; et nimium meminisse necesse est.
1. We have given the rule as equally applicable to both the indica-
tive future perfect and the perfect subjunctive. It holds good, however,
more certainly of the former than the latter tense. With regard to the
perfect subjunctive, it is frequently impossible to distinguish it from the
future perfect, since in very many cases where the. one is employed, a
very slight modification of the sense would render the use of the other
equally appropriate. The only example discovered by prosodians where
this tense undoubtedly occurs in such a position as to determine its
quantity, is in the line from Virgil cited above, and which is found in JEn.,
6, 514. The old grammarians are at variance on this subject. Dioine-
des (p. 331) and Agroetius (p. 2267) assert that the penult of rimus and
ritis in the future perfect is long, and in the perfect subjunctive is
short. Probus, on the other 'hand (p. 1412 p. 1434), affirms that the
syllable is long in both tenses ; and both Probus and Servius (ad Virg.,
I. c.) expressly declare, that the penult of egerimus, in the passage quoted,
was shortened by Virgil " metri necessitate.'-' (Voss., de Art. Gram,,
2,31.)
F 2
66 FINAL SYLLABLES.
VERBAL INCREMENT IN AND U.
RULE. In the increase of verbs o is always long, but u
is generally short ; as, facitote, habctote ; sumus, possumus,
mlumus.
Ovid. Cumque loqui poterit, matrem facitbte salutet.
Id. Hinc quoque presidium laesae petitote figura.
Horat. Nos numerus sumus, etfruges consumere nati.
Virg. Dicite, Pierides : non omnia possumus omnes.
Horat. Si patriae volumus, si nobis, vivere cari.
EXCEPTION.
But u in the penult of the future participle in rus is al-
ways long; as, periturus,facturus, amaturus.
Virg. Si periturus abis, et nos rape in omnia tecum.
Id. Tarda venit, seris factura nepotibus umbram.
SECTION XXIII.
FINAL SYLLABLES.
The quantity of final syllables is ascertained, in some
cases, by position ; as, prudens, pracox ; in others by their
containing a diphthongal sound ; as, musas, penntB ; but in
most by special rules, which follow.
FINAL a.
RULE 1. A in the end of words not declined by cases is
long ; as, circa, citrd, contra, extra, frustrd, intrd, &c. ; to
which add the imperative of the first conjugation ; as, mem-
ore, amd. 1
1. In imperatives of the first conjugation the final a is long, because
formed by contraction from ae. Thus, memorae, memord ; amae, amd,
&c. (Struve, uber die Lot. Declin. und Conj., p. 135.) With regard
to circa, citra, contra, <fec., various opinions exist. Scheller views them
as old ablatives singular feminine of the first declension, with an ellipsis
of parte, or opera, or some other equivalent term. (Scheller, Lat.-
Deutsch. Worterb., s. T>.) Ramsay, on the other hand, regards most
of them as old imperatives of verbs of the first conjugation, of which
frustro, intro, supero (or supro), are still in use ; while, as respects such
forms as an ed, posted, intered, pratered, &c., he agrees with the author
FINAL SYLLABLES. 67
Horat. Circa mite solum Tiburis et mania Catili. (Chori-
ambic.)
Ovid. Dextera diriguit, ncc citrd mota, nee ultra.
Virg. Contra non ulla est oleis cultura, neque ilia.
Horat. Laudet ametque domi, premat extra limen iniquus.
Id. Frustra, nam scopulis surdior Icari. (Choriambic.)
Virg. Musa, mihi causas memora ; quo numine lasso.
Plaut. Si auct oritat em postea defugeris. (Iambic.) 1
EXCEPTIONS.
But eia, ita, puta 2 (the adverb), and quia, 3 have the final
vowel short ; and also the names of letters ; as, alpha,
beta, &c., which latter follow the quantity of the Greek.
Val. Flac. Ferret ad aurigeras caput arboris : Eia per ipsum.
Virg. Incolimus ; sed vos, sifert ita corde vohintas.
of a paper in the "Journal of Education" (vol. 1, p. 106), who supposes
them to be formed from ante earn, post earn, &c., the correlatives ante-
quam, postquam, still retaining the final letter. (Ramsay's Lat. Pros.,
p. 38.)
1. Posted occurs with the final vowel short in Ovid (Fast., 1, 165);
but the difficulty may be obviated either by writing post ed, or by pro-
Bouncing the word as a dissyllable, postyd, making it, in this latter case,
have the long final vowel.
2. Puta. is short only when taken adverbially. When it stands as an
imperative the a is long. Some doubt, however, has been attempted to
be thrown upon the quantity of the final letter in puta when an adverb.
It is found short in Persius (4, 9), and also in Martial (3, 26, 5 ; 9, 96,
5 ; 1 1, 95, 2), in accordance with the exception in the text, and it has been
allowed to stand by several editors, where others have given puto. Ca-
saubon and Konig, for example, retain puta in the text of Persius, and
Priscian, the ancient grammarian, as cited by the former, observes, " Ita
solum a correptum habet, quamvis quidam puta adverbium esse accipiunt,
ideoque Persium id corripuisse, ut, hoc, puta, non justum," etc. (Prise.,
lib. 15 vol. 1, p. 617, ed. Krehl.) Servius likewise, in his commentary
on Virgil (JEn., 2, 651 vol. 1, p. 167, ed. Lion), after observing that
adverbs in a are long, expressly excepts puta and ita. Still, in all the
instances cited above, many MSS. give puto, which leaves the matter
involved in some degree of uncertainty. .
3. Dr. Carey, on the authority of a single line in Phaedrus, pronounces
the final vowel in quia doubtful. The line is as follows : " Ego primam
tollo nominor quid ho." (Phadr., 1, 7.) But almost all editors agree
in considering the line corrupt, and most of them read " Ego primam
tollo quoniam nomine* leo."
68 FINAL SYLLABLES.
Virg. Sed quid non aliter vires dabit omnibus aquas.
Juv. Hoc discunt omnes ante Alpha et Beta puella.
RULE 2. A in the end of words declined by cases is
short, except the ablative singular of the first declension,
and Greek vocatives from nominatives in as ; as, ancho~a
de prord, jEnea, Palld. 1
Virg. Anchord de prora jacitur ; slant littore puppes.
Id. Quid miserum, jEned, laceras ? jam parce sepulto.
OBS. 1. Greek names in es and e are frequently changed
by the Latins into a ; as, Atrida for Atrides ; Oresla for
Orestes ; Circa for Circe. In nouns of this class, the final
a in the vocative is short. 2
OBS. 2. Some prosodians make the final a in numerals
either long or short. The true quantity, however, is the
long one ; as, trigintd, quadragintd, sexagintd?
FINAL 6.
RULE. E final is for the most pan short; as, nate,fuge,
eripe, ille, sine, pane, &c.
1. The final a in the ablative singular of the first declension is long,
because contracted from at. The old dative and ablative singular of the
first declension had this latter ending; as, terral for terrac. ; tcrrai for terra.
(Plank ad Enn., Med., p. 80. Grutcr, Inscr., 2, 12. Id., 2nd., p. 84.)
2. The short a in these forms is obtained from the ^Eolic a (Com-
pare the Greek forms ve<j>e^.ij'yEpf~a., iiriroTa, evpi>ona., and the Latin
pocla, from iroi^rrjf.) In Virgil (JEn ., 3, 475) we have Anchisa in the
vocative, which some deduce from a Doric nominative Anchisas. It 'u
better, however, to ascribe the length of the final vowel, in this instance,
to the force of the caesura or arsis.
, 3. To support the position that the final a in numerals is common, the
following line from Manilius has been adduced : " Ter trigintd quadrum
paries per sidera reddant." (Manii, 2, 322.) Bentley, however, in-
sists that the reading here is erroneous, especially since triginta ociurs
six lines lower down, in the same poet, with the final vowel long. In
all the oldest MSS. of the classics, numbers were expressed by marks,
. not by words ; hence, when the transcriber found LXXXX in his copy,
he ignorantly, according to the English critic, rendered it by tcr trigintu
instead of nongentat, which latter reading Bentley of course restores.
Some other examples of numerals with the short final vowel arc adduced
from Martial and Ausonius, but the MSS. here give different readings.
The rule, therefore, laid down above is undoubtedly the true one.
FINAL SYLLABLES. 69
Virg. Hcu fuge, note Dea, teque his ait eripe flammis.
Id. Hie miki ante olios fortunatusque laborum.
Id. Qucis sine nee potuere seri nee surgere messes.
Id. Pane simul tecum solatia rapta Menalca.
EXCEPTIONS.
Exc. 1. E final is long in all cases of the first and fifth
declensions ; as, JEgle, Thisbe, Melpomene, die, Jide,
fame, re ; and in adverbs derived from the latter ; as,
hodie, pridie, quare, &C. 1
Virg. JEgle Na'iadum pulcherrima, jamque videnti.
Ovid. Saspe ut constiterant, hinc Thisbe, Pyramus illinc.
Horat. Quern tu Melpomene semel. (Choriarabic.)
Virg. Forte die solemnem illo rex Areas honorem.
Id. Libra die somnique pares ubifecerit horas.
Horat. Effare; jussas cum fide poenas luam. (Iambic.)
Virg. Amissis, utfama, apibus mqrboque fameque.
Id. Pro re pauca loquor ; nee ego hanc abscondere furto
Horat. Muneribus servos corrumpam ; non hodie si.
Id. Quare per divos oratus uterque penates.
1. The final e is long in all cases of nouns of the first declension, be-
cause answering to the Greek 77. It is long in all cases of the fifth de-
clension, because it is a contracted syllable. (Schneider, L. G., vol. 3,
p. 355, seqq ) Under this same exception falls the ablative fame (Virg.,
JEn., 6, 421), the noun fames having been, according to Aulus Gellius
(9, 14), originally of the fifth declension, fames, famei; like plebes, pie-
lei. The vocatives Ulysse and Achille have also the final e long. These
are Greek forms. The yEolo-Doric tribes changed the termination evf
into 7/f, and said 'Op<j>rie for 'Optyevc, 'Odvoorif for 'Odvaaeiif, 'A^t'A/l^f
for 'AxM^vf, &c. (Maittaire, Gr. D., p. 183.) The Latins, in imi-
tation of these, used Ulysses and Achilles, with some others, as nouns
of the third declension, making in the vocative Ulysse, Achille, &c.,
with the final e long, because answering to the Greek 77. Another Latin
form, and one of more frequent recurrence in poetry, is that in eiis, of
the second declension ; as, Ulyeseiis, Achilleiis, making in the genitive
Ulyssei, Achille'i, contracted into Ulyssi, Ar.hilli. The vocative of such
a form will be Ulyssee, Ar.hiU.ee. We may suppose Achille, in Proper-
tius (4, 11, 40), to be formed from it by apocope, if we retain the com-
mon reading, " Quique tuas proavus, frcgit, Achille, domos." But
Heyne has given the true lection : " Qui tumidas proavo frcgit Achille
domos," which makes Achille the ablative, with the final vowel short, as
a matter of course. (Heyne ad JEn., 6, 840.)
70 FINAL SYLLABLES.
Exc. 2. E final is long in the contracted nominative and
accusative plural of the third declension, in words
transplanted from the Greek; as, cete, mele, pelage,
tempe, &c., the plural vowel in such words answering
to the Greek eta, or long e. 1
Sil. Ital. Dum cete ponto innabunt, dum sidera ccelo.
Lucret. At Musasa mele per chordas organici qua.
Id. At pelage multa et late substrata videmus.
Catull. Tempe qua silvas cingunt super impendentes.
Exc. 3. Verbs of the second conjugation have e final
long in the second person singular of the imperative
active ; as, gaude, salve, vale, &c. But cave has the
last syllable either long or short. 2
Propert. Gaude, quod nulla est aeque formosa, doleres.
Virg. Salve magna parens frugum, Saturnia tellus.
Id. Cat. Vale, Sabine,jam valete formosi. (Scazon.)
Horat. Cave, cave ! namque in malos asperrimus. (Iambic.)
Id. Tu cave ne minuas, tu ne majus facias id.
Ovid. Neu cave defendas, quamvis mordebcre dictis.
Exc. 4. Adverbs in e, formed from adjectives in us, have
the final e long; as, placide, probe, late, longe, &c., ex-
cept bene, male, inferne, and superne.
1. Thus, Kfirea, contracted K^nj ; pefaa, fishf ; irf^d-yea,
Ttfiwta, reftTTT], &c.
2. The second person singular of the present imperative active, in
verbs of the second conjugation, is, like the corresponding tense in verbs
of the first, a contracted form. Thus, gaudee, gaude ; salvee, salve ;
valec, vale, &c. The double quantity in care or cave arises from the
following circumstance, that anciently two forms of the verb were in use,
one belonging to the second, and the other to the third conjugation ; just
as we find both ferpco and ferco ; fulgeo and fidgo ; oleo and olo, &c.
(Struve, titer die Lot. Dec!., &c., p. 189. Voss., de Art. Gramm., 2,
25.) Besides care or cave, we find it frequently asserted that vale, vide,
responds, and salpe have the last syllable common ; but it will be dis-
covered, on examination, that there is little, if any, evidence to prove
this. The question will be found discussed in Ramsay's Latin Prot-
ody, p. 44, seqq.
FINAL SYLLABLES. 71
Mart. Excipe sollicitos placide, mea donj, libellos.
Catull. Suffenus iste, Vare, quern probe nosti. (Scazon.)
Virg. Directaqitv acies, et late fluctuat omnis.
Juv. jEquora transiliet, sed longe Calpe relicta.
Virg. Si bene quid de te merui,fuit out tibi quidquam.
Id. Insequitur, cumulosque ruit male pinguis arena.
Lucret. Ne tibi sit fraudi quod nos inferne videmus.
Id. Remorum recta est, et recta superne guberna.
But adjectives neuter of the third declension, used as ad-
verbs, retain the final e short ; as, sublime, suave, dnlce,
facile, &c.
Virg. Cantantes sublime ferent ad sidera cycni.
Id. Ipse sed in pratis aries,jam suave rubenti.
Exc. 5. The adverbs ferme,fere, and ohe have the final
vowel long. Fere, however, has the last short in the
later writers.
Juv. Rarus enimferme sensus communis in ilia.
Hoiat. Vina fere dulces oluerunt mane Camcena.
Id. Importunus amat laudari donee ohe jam.
Auson. Nam tecumfere totus ero quocumque recedam.
Exc. 6. Monosyllables in e are also long ; as, de, me, te,
se, and ne (lest or not) ; except the enclitics que, ve,
ne, and the syllabic additions pte, ce, te, de ; as, in su-
apte, nostrapte, hosce, tute, quamde. 1
Virg. De cado tactas memini pradicere quercus.
Id. Me me, adsum quifeci; in me convertite ferrum.
1. The lengthening of monosyllables which consist of, or terminate in
e vowel, depends upon an established principle of metrical harmony, since
they would be nearly lost in the reading if the voice did not dwell upon
them, and make them necessarily long. In the case of enclitics and syl-
labic additions, however, the principle does not apply. These are con-
nected so closely with the preceding word, that they form but one word
with it in the rapidity of pronunciation, and are no longer considered as
separate monosyKables.
72 FINAL SYLLABLES.
Virg. Te vcniente die, te decedente canebat.
Id. Ne pueri, ne tanta anitnis adsuescite bella.
Id. Arma virumque cano, Troja qui primus ab oris.
Id. Si quis in adversum rapiat, casusve Deusve.
Id. Tantane vos generis tenuit Jiducia vestri ?
Enn. O Tite tute Tati tibi tanta tyranne tulisti.
FINAL t and y.
nna l i g f r tne most part long ; as, frumet. fi,
scribendl, nulli, partiri, fieri, &c.
Virg. Paullatim et sulcis frumenti quareret herbam.
Horat. Garrulus, atque piger scribendi fcrre laborem.
Proper^. Nulli curafuit externos quasrere divos.
Virg. Ne signare quidem out partiri limite campum.
Id. Pastorcs, mandatjieri sibi talia Daphnis.
EXCEPTIONS.
"
Exc. 1. I final is short in nisi and quasi. 1
Virg. Nee veni, nisi fata locum sedemque dedissent.
Ovid. Quid quasi natali cum poscit munera libo.
Exc. 2. The final t and y are short in Greek neuters ;
as, gummt, sinapi, moly ; in the dative singular of Greek
nouns ; as, Palladi, Thetidi, Phyllidi ; in Greek voca-
tives ; as, Adorn, Aleon, Tiphy, Tethy, chely (but no^
in Tethy, the contracted dative for Tethyi) ; and in
datives and ablatives plural in si ; as, heroisi, Dryc
Troasi.*
1. Quasi occurs with the i long in Lucretius (2, 291, and 5, 728),
and in Avienus (Phacn., 554, 1465, 1567, 1654); but the final vowel
in all these instances may be considered as made long by the arsis, es-
pecially since we find quasi twice in Lucretius (4, 1011, and 6, 972).
Nisi also has the final vowel long in the following Phalaecian line from
Statius (Syfo., 4, 3, 59) : " His parvus, Lcchia nisi vetarent ;" but the
MSS. here are hopelessly corrupt, and scarcely two editors read the pas-
sage in the same way. The Bipont edition has " His parrus, Lecheo
nihU vetante," which is retained in that of Lemaire.
2. It often happens that in such Greek datives as The/irfi, Panda,
Tynddndi, &.C., the final vowel is lengthened by the arsis, since other-
FINAL SYLLABLES. 73
Ovid. Moly vacant superi ; nigra radice tenetur.
Stat. Palladi litoreae celcbrabat Scyros honwem.
Ovid. Semper Adorn, met, repetitaque mortis imago.
Id. Quid tibi cum patria, navita Tiphy, mea. (Pentam.)
Id. Troasin invideo, qua si lacrymosa suorum. 1
Exc. 3. In mihi, tibi, sibi, ibi, and ubi, the final i is com-
mon. 2
Virg. Non mihl si lingua centum sint, oraque centum.
Id. Tros Tyriusque mihl nullo discrimine agetur.
Id. Haud obscura cadens mittet tibi signa Bootes.
Id. Cuncta tibi Cererem pubes agrestis adoret.
Id. Jam sibi turn curvis male temper at unda carinis.
Horat. Quanta quisque sibi plura negaverit. (Choriambic.)
Virg. Aut ibiflava seres mutato sidere farra.
Id. Ter conatus ibl collo dare brachia circum.
Id. Nosque ubi primus equis oriens afflavit anhelis.
Horat. Instar veris enim vultus ubi tuus. (Choriambic.)
OBS. 1. The quantity of the final vowel in uti is involved
in some uncertainty. Most prosodians make it long, a
quantity which it is always found to possess ; and so, too,
it is always long in veluti. If, however, any stress is to be
wise forms like these could not find a place (on account of their containing
three short syllables in succession) in dactylic verse. Instances of such
lengthening occur in Catullus (64, 21), Propertius (3, 8, 29, seq.), Va-
lerius Flaccus (1, 190), Ovid (Heroid., 20, 60), Id. (R. A., 711), &c.
1. In this example, the n added to Troasi is placed there merely to
prevent the hiatus at the meeting of the two vowels, and makes no dif-
ference whatever in the quantity. It is like the v E^E^KVOTLKOV of the
Greeks.
2. These words originally ended in the diphthong ei, as mihei, libei,
sibei. ibei, ubei, and under this shape they are frequently found in inscrip-
tions (compare note 1, page 16) and MSS., especially those of Lucre-
tius. One of the vowels of the diphthong being subsequently dropped,
they would sometimes appear as mike, tibe, sibe, &c., and sometimes as
mihi, tibi, sibi, &c. In the former case, the final e being short in Latin
words, except under particular circumstances, the last syllable would be
made short by the poets ; in the latter case, the final i being long in
Latin words, the syllable would retain its original quantity, as it probably
always did in prose. (Ramsay's Lat. Pros., p. 60.)
6
74 FINAL STLLABLES.
laid upon the fact that the i is always short in sicufi, #
nam, and utique, the final vowel in uti ought rather to be
regarded as common.
OBS. 2. In neculi, sicubi, ubinam, and ubivis, the i of ubi
is always short ; but if we are to be guided by the quantity
of the final letter of ubi out of composition, as also by the
circumstance of the i being long in ublque (" everywhere"), 1
but common in ubicunque, we shall, in all probability, be
more correct in making the i of ubi common also in the
compounds first mentioned.
, Exc. 4. Cui, when a dissyllable, generally has the i
1 (Ji short.
Sen. Mittat et donet cuicumque terras. (Sapphic.)
Mart. Sed norunt cui serviunt leones. (Phalaecian.)
Id. Drusorum cut contigere barboe. (Ditto.)
Id. Et credit cui Postumilla dives. (Ditto.)
- 7 / OBS. Cui is commonly considered as forming a mono-
/ syllable in poetry. Instances, however, occur, in which
it may be regarded as a dissyllable, even in hexameter
verse, without any injury to the metre, but with advantage,
rather, to the smoothness and harmony of the line ; a? in
the following, among others :
Juv. Cantabat patriis in montibus : et cui non tune.
Virg. At puer Ascanius cui nunc cognomen lulo.
Id. Munera vestra cano. Tuque O cui prima ferentem .
Id. Incipe parve puer : cui non risere parentes.
FINAL o.
RULE. O final is common, though more generally long
than short.
Horat. Quando pauperiem, missis ambagibus, horres.
Mart. Quando moree dulces, longusqut a Casare pulvis.
Virg. PraRterea duo nee tuta mihi valle reperti.
1. Consult note 5, page 46.
/
FINAL SYLLABLES. 75
Auson. Europam Asiamque duo vel maxima terras.
Mart. Capto tuam, pudet lieu I sed capto Pontice c&nam.
Gall. Obruta virgo jacct : servat quoque nomina turns.
Ovid. Victa jacet pietas ; et virgo cade madentes.
Mart. Miscuit, Elysium possidet ambo nemus. (Pentam.)
Virg. Ambo florentes aetatibus, Arcades ambo.
OBS. 1. O final in nominatives of the third declension is,
with very few exceptions, long in the writers of the Augus-
tan age and their predecessors. In proper names, how-
ever, o final is common even in the best writers ; as, Car-
thago, Pollio, Scipio, Curio, Vinio. 1
OBS. 2. O final in verbs is very rarely shortened by wri-
ters of the Augustan age and their predecessors, except in
seio, nescio, puto, volo, which are for the most part used
parenthetically. 2
EXCEPTIONS.
Exc. 1. All cases in o of Greek nouns, written in the
original with an o-mega, or long o, have the final vow-
el long; as, nominative, Id, Ino, Clio; genitive, An
droged ; accusative, Atho, Clotho, &c.
Propert. Id versa caput primos mugiverat annos.
Auson. Clio gesta canens transactis tempora reddit.
Virg. Inforibus letum Androged : turn pendere pcenas
Pedo. Quondam ego tentavi Clothoque duasque sorores.
1. It was not until the age of Lucan that the practice of shortening o
final in nouns of the third declension became general. In the writings
of this poet we find cardo, pulmo, tiro, turbo, &c. ; and in Martial and
his contemporaries it is perhaps oftener short than long. (Ramsay's
Lot. Pros., p. 55.)
2. No example occurs in Lucretius, Virgil, or in the Odes of Horace,
of the final o in a verb being left short, except in scio and nescio, which,
as well as puto, volo, rogo, credo, do not form real exceptions, for these
words were either used parenthetically, or in colloquial formula enun-
ciated rapidly. The shortening of the final o in verbs is very rare in
Catullus, in Tibullus, in Propertius, and in Ovid ; it gradually becomes
more common in the writers who follow them, and when we come down
to the age of Statius and Martial it is to be found on every page, (flam-
ay's Lot. Pros., p. 56. Lennep ad Ov., Ep., 15, 32.)
76 FINAL SYLLABLES.
Exc. 2. Monosyllables in o are long ; as, O, do, sto,pro,
&C. 1
Virg. decus, o fama merito pars maxima nostra.
Id. Do quod vis ; et me victusque volensque remitto.
Id. Pro molli viola, pro purpureo narcisso.
Exc. 3. O final is long in the dative and ablative singu-
lar of the second declension ; as, viro, vento, auro, sic-
co, <fcc. 2
Virg. Cui se pulchra viro dignetur jungere Dido.
Ovid. Nutritur vento, vento restinguitur ignis.
Propert. Auro pulsa Jides, auro venalia jura.
Virg. In sicco ludunt fulicae ; not asque paludes.
^*
Exc. 4. O final in the gerund is perhaps never found
short, except in writers subsequent to the Augustan
age. 3
Virg. Frigidus in pratis cantandb rumpitur anguis.
Ovid. Et voluisse mori, et moricndo ponere sensus.
Juv. Plurimus hie <eger moritur vigilando, sed ilium.
Auson. Qua nosti meditando velis inolescere menti.
Exc. 5. Adverbs formed from adjectives have the final o
for the most part long ; as, multd, raro, crebro, consul-
to, &c. 4
1. Compare note 1, page 71.
2. The final vowel in the dative and ablative singular of the second
declension is the result of contraction, and therefore long. The primi-
tive termination was oi. (Struve, uber Declin., &c., p. 14.)
3. Two passages oppose this doctrine, which, however, is generally
recognised by scholars. One is from Tibullus (3, 6, 3) : " Aufer et
ipse meum pariter medicando dolorcm," where Heyne reads medicande,
from Broukhusius. Disscn also gives this same lection. On the whole
question respecting the shortening of the final syllable in gerunds, consult
Heyne ad Tibull., I. c. BroukUus. ad Tibull., 1. c. Hcins.ad Ot> , Ep.,
9, 125. Burmann ad Anthol. Lot., vol. 1, p 298 Perizon. ad Sanr.t.
Min., vol. I, p. 148, ed. Bauer. Ramsay's Lot. Pros., p. 57.
4. These, in fact, retain the quantity of the dative singular, from which
they are formed.
FINAL SYLLABLES. 77
Juv. Pcena autem vehemens, et multi savior Hits.
Ovid. Adde quod iste tuus, tarn raro prcelia passus.
Horat. Est mihi purgatam crebro qui personet aurem.
Id. Extenuantis eas consulto ; ridiculum acri.
*
Exc. 6. O final is never, perhaps, found short in ergo,
ideo, immo, porro, postremo, sero, vero, except in writers
subsequent to the Augustan age. 1
Virg. Ergo non hiemes illam, nonflabra neque imbres.
Horat. Ergo Quintilium perpetuus sopor. (Choriambic.)
Propert. Ergo velocem potuit domuisse puellam.
Ovid. Ergo dum Stygio sub terris gurgite labor.
Lucan. Ergo pari voto gessisti bella juventus.
Juv. Impune ergo mihi recitaverit ille togatas.
Mart. Sed tamen esse tuus dicitur, ergo potest. (Pentam.)
Horat. Ac ne me foliis ideo brevioribus ornes.
Mart. An ideo tantum veneras ut exires ? (Scazon.)
Catull. Frustra ? immo magno cumpretio atque malo. (Pent.)
Mart. Adeo bene emit ? inquis : immonon solvit. (Scazon.)
Id. Vendere, nil debet, foznerat immo magis. (Pent.)
Catull. Sed dicam vobis, vos porro dicite multis.
1. Some of the ancient grammarians, and almost all the modern ones,
make ergo, when it signifies " on account of" have the final syllable
long, and short when it means " therefore.'" The distinction does not
appear to be a correct one, for the two meanings are in fact the same,
and the word in either case is merely Ipyw, the dative of Ipyov. The
line quoted by Dr. Carey, from the Ciris, to prove that ergo, " therefore,"
occurs in good writers with the final syllable short, cannot be received
as authority, since the Ciris, which few suppose to have been the work
of Virgil, is notorious for its corrupt text. The line is as follows : " Ergo
mctu capiti Scylla est inimica paterno" (v. 386). Barlh reads " Ergo
turn capili," and Heinsius " Ergo ilerum capiti," which latter emenda-
tion is adopted by Heyne. The passage sometimes cited from Proper-
tius (3, 7, 1), " Ergone sollicita jtu causa, pecunia, vita es" is given in
the best MSS. and editions as follows: " E:gb sollicita tu causa, pccu-
nia, vita cs." One instance, however, occurs in Ovid (Her., 5, 59),
where ergo has the o short, according to the received reading. It is 23
follows : " Votis ergo meis alii rediture redisti." It is v< ry probable,
however, tha' some error lurks here in the text, since Ovid everywhere
else makes tne final syllable of ergo long. (Hcins. ad Ov., Trist., 1, 1,
87. Ramsay's Lat. Pros., p. 58.)
G2
78 FINAL SYLLABLES.
Juv. Multos porro vides, quos s&pe elusus, ad ipsum.
Id. Et Scauros et Fabricios postremo severos.
Till all. Heu sero revocatur amor seroque juventus.
Juv, H<BC animo ante tubas. Galeatum sero duelli.
Mart. Sero dedit pasnas. Discerpi noxia mater.
Virg. Hie vero victus genilor se tollit ad auras.
Stat. Tu potior, Thebane, queri, nos vero volentes.
> ~
Exc. 7. O final is always short in the following words
in good writers : cito, ego, 1 modo the adverb, and its
compounds dummodo, postmodo, quomodo, tantummodo,
together with the numeral octo.
Horat. Quicquid praecipies esto brevis, ut cito dicta.
Ovid. Nee cito credideris, quantum cito credere l&dat.
Virg. S&pe ego, quumjlavis messorem induceret amis.
Lucret. Non modo non omnem possit durare per avom.
Virg. Hie inter densas corulos modo namque gemellos.
1. Ego is said to have the final o common. " The fact, however, is,"
observes Ramsay, " that there are many hundred instances, in writers of all
ages, in which ego is found with the last short, and three or four at most,
in decent metrical authorities, where it is found long ; but even here, in ev-
ery case, if I mistake not, under suspicious circumstances. It may serve
to set at rest the question with regard to the final o in ego and modo (the
adverb), if I state that I have marked 532 examples of ego with the o
short in Ovid alone, 91 in Propertius, 90 in Horace, 64 or 65 in Virgil,
53 in Tibullus, 27 in Catullus, and five in Lucretius, in all 862 ; while
in the same authors I have been unable to find more than two with the
long o ; one of these (Catvll., 19, 1) is from a poem which, though often
placed among the works of Catullus, is found in no MS. of that author,
and is now left out by the best editors ; the other from Ovid (Her., 13,
135), in a line where the MSS. afford half a dozen different readings. I
am aware that other examples are to be found in old editions, but these
have all disappeared upon a careful examination of the MSS. ; as, for
example, Propert., 1, 8, 31 ; 4, 2, 3, &c. Such being the evidence, I
feel justified in reversing the judgment pronounced by Broukhusius (ad
Properl., 1. c.), Drakenborch (ad Sil. Ilal., 17, 357), and Ruperti (ad
Sil., I. c.) in favour of the o final in ego being common, and in laying
down the rule as I have givn it. With regard to modo (the adverb), I
have marked 363 examples in Ovid, 48 in Propertius, 22 in Horace, 13
in Virgil, six in Catullus, two in Lucretius ; in all of these (451) the
final o is short, against which there is one in Lucretius where it is length-
ened by the arsis. The same holds good of its compounds, with the
single exception of quom id in Catullus (10, 7)." (Ramsay's Lot. Pro*.,
p. 60, seq.)
FINAL SYLLABLES.
79
Horat.
Ovid.
Lucret.
Horat.
Id.
Tibull.
Horat.
Id.
Manil.
Juv.
Mart.
Herculis ritu modo dictus O plebs. (Sapphic.)
Nam modo, vos animo, dukes reminiscor amid.
Dummodo ne totum corrumpas luminis orbem.
Fcenum habet in cornu, longe fuge ; dummodo risum.
Postmodo, quod mi obsit, dare certumque locuto.
Postmodo qua votis irritafacta velit. (Pentam.)
Cum victors sequor. Maecenas quomodo tecum.
Proximus esse. Velis tantummodo, qua tua virtus.
Sed regione Nepoz vix partes odd trahentis.
Sic crescit numerus, sic fiunt octo mariti.
Vix octo nummis annulum unde ccenaret. (Scazon.)
X FINAL .
RULE. U final is long ; as, cornu, metu, partu, Panthu,
vitatu, diu. 1
Horat. Cornu decorum, leniter atterens. (Alcaic.)
Virg. Parce metu Cytherea, manent immota tuorum.
Id. Eumenidesque satte ; turn partu terra nefando.
Id. Quo res summa loco, Panthu, quam prendimus arcem ?
Horat. Aiebat sapiens vitatu, quidque petitu.
Virg. Phoe.be diu, res siqua diu mortalibus ulla est.
^ EXCEPTIONS.
Exc. 1. Indu, the old form of in, and nenu for non, both
Lucre tian words, have the u short. 2
1. Final u in the dative and ablative singular of the fourth declension
is the result of contraction from ui, and therefore long. (Struve, iiber
Declin., &c., p. 36. Burmann ad Propert., p. 119.) Hence metu is
for metui, and partu, for partui, or, rather, partue. Some of the old
grammarians maintained that neuter nouns in u had the final vowel short
in the nominative, accusative, and vocative singular, but long in the
other cases. This doctrine, however, is condemned by Priscian (vol. 1,
p. 351, ed. Krchl). The u in Panthu represents the diphthong ov in
the original Greek word. Diu is an old ablative from dius, as is evident
from the common phrase diu noctuque. (Ramsay's Lat. Pros., p. 62.)
2. Indu appears to have come from the ^Eolic tvSov for evdov. Nenu
is said to have been the parent of the Latin non. According to Wake-
field, the more correct orthography for indu is cndu when it stands
singly, and indu when compounded. (Wakef. ad Lucret., lj 83, et 2,
1005.) His authority, however, is of no great weight, especially as the
80 FINAL SYLLABLES.
Lucret. Necjacere indu manus, via qua munita fidei.
Id. Nenu queunt rapidei contra constare leones.
Exc. 2. The u continues short in those words whicn
naturally end in short us, and are only deprived of the
s by the more ancient mode of pronunciation, in order
to preserve the syllable from becoming long by its po-
sition before a consonant at the beginning of the follow-
ing word ; as, plenu 1 for plenus, bonu 1 for bonus, &c.'
Ennius. Ille vir haud magna cum re, sed plenu 1 fidei.
Id. Suavis homo,facundu\ suo contentu\ beat us.
4 FINAL y.
RULE. Y final is short ; as, moly, chely, Coty, Tiphy?
Ovid. Moly vacant superi ; nigra radice tenetur.
Stat. Cedamus chely, jam repone cantus. (Phalaecian.)
Ovid. O Coty, progenies digna parente tuo. (Pentam.)
Id. Ars tua, Tiphy,jacet si non sit in aquore fluctus.
^ FINAL I, d, t.
RULE. Final syllables ending in b or d are short, as also
those ending in t pure, that is, t immediately preceded by
a vowel ; as, db, ad, quid, illud, et, at, amdl.
Ovid. Jpse docet quid agam. Fas est et db hoste doceri.
Virg. Dixit : at illafurens, acrique incensa dolore.
Tibull. Luce sacra requiescdt humus, requiescdt orator.
EXCEPTIONS.
Exc. 1. But if t be preceded by another consonant, or
the t or d by a diphthong, the syllable must of course
remain long ; as, dst, amdnt, out, haud.
Ovid. Ast ubi blandiliis, agitur nihil horridus ira.
^Eolic change of c into i is well known. (Maitt., Dial., p. 208, ed.
Sturz.)
1. Consult remarks under "Ecthlipsis."
2 The final y answers to the short final v in Greek. This rule is in
part repeated from page 72.
FINAL SYLLABLES. '81
Virg. Aut onera accipiunt venientum, ow. agminc facto.
Id. Hdud obscura cadens millet libi signa Bwtcs.
Exc. 2. Those third persons singular of the perfect tense,
active voice, which contract ivit or nt into it, or avit
into at, have the final syllable necessarily long ; as,
petit for petilt or petiint ; obit for obnl or obivit ; irritat
for irritavit ; disturbat for disturbavit.
Ovid. Flamma petit ahum, propior locus aera cepit.
Juv. Magnus civis obit et formidatus Othoni.
Lucr. Irritat animi virtulem, ecfringere ut arcta
Id. Disturbat urbes, et terras motus abortus.
FINAL C.
RULE. C final has the preceding vowel long ; as, iliic,
illuc, ac, sic, hue, the adverb hie, the ablative hoc.
Virg. Illic, officiant laetis ne frugibus herbas.
Catull. lonios fluctus postquam illuc Arrius isset
Horat. Si sapiat vitet simul dc adoleverit aetas.
Virg. Sic oculos, sic illc manus, sic oraferebat.
Catull. Hue, hue adventate ; meas audite querelas.
Virg. Classibus hie locus, hie acies cerlare solebant.
Ovid. Aut hoc, out simili carmine notus eris. (Pentam.)
EXCEPTIONS.
Exc. 1. Nee, donee, 1 and the imperative /ere 2 are short.
Ovid. Parve, nee invideo, sine me liber ibis in urbem.
1. Donee is merely an abbreviation of donicum, a word of frequent oc-
currence in Plautus, and itself evidently an adjective of the neuter gen-
der.
2. Vossius says that fac is always long, and cites the following lines
in support of his opinion :
Has fac Armenios, hcec est Danaeia Persis. (Ov., A. A., 1, 225.)
Durius incedit,fac ambulet, omne papilla. (Id., R. A., 337.)
Heinsius, however, upon unexceptionable MS. authority, restored in the
first Has facito for Hos fac, and in the second fac inaitbulet for/ac am-
bulet. In almost all cases where fac is followed by a vowel, the MSS
vaiy between fac and/ace. (Ramsay's Lat. Pros., p. 33.)
82 FINAL SYLLABLES.
Ovid. Donee eris fclix mullets numerabis amicos.
Lucret. Non possunt ; fdc enim minimis e partibus esse.
Mart. Signa rarius, out semelfac illud. (Phalaecian.)
Exc. 2. Hie the pronoun is common, but much, more
frequently long than short.
Virg. Solus hlc inftexit sensus animumque labantem.
Id. Hie vir hlc est, tibi quern promitti sapius audis.
Id. H<BC finis Priami fatorum, hie exilus ilium.
Ovid. Atque ait, Hlc, hlc est, quern ferus urit amor. (Pent.)
Exc. 3. The neuter hoc is also common, but no example
can be quoted, except from the comic writers, in which
it is found short. 1
Ovid. Dicendum tamen est, hoc est, mihi crede, quod <egra.
Id. Hoc deus et votes, hoc et mea carmina dicunt.
Plaut. Heus ! ecquis hie est ? ecquis hoc aperit ostium ?
(Iambic.)
Id. Quid hoc hie clamoris audio ante &des meas ? (Iambic. )
FINAL I.
RULE. L final has the preceding vowel short ; as, Han-
nibal, 2 semel, nihil, procul.
1. The facts respecting the usage of the poets in the case of Ate and
hoc are given in the text. The opinions expressed by the old gramma-
"rians respecting the quantity of these words differ widely from each other.
Velius Longus and Priscian seem to think that hie and hoc are both nat-
urally short, and that in all passages where they are found long they
ought to be written luce, hocc, and regarded as abbreviations of hicce,
hocce. Terentianus Maurus, Marius Victorinus, Probus, Charisius, and
Martianus Capella, on the other hand, assert, that in these words c has
the same force in pronunciation as a double consonant ; that, conse-
quently, hie and hoc ought always to be long, and that Virgil was guilty
of an inaccuracy in changing the pronunciation and quantity of hie in the
two passages from the JSneid cited in the text. Vossius says that hoc
is short in the nominative and vocative ; but he is unable to bring any
better authority than that of two anonymous poets in the collections.
(Priscian, vol. 1, p. 564, cd. Krehl. Velius Longus, p. 2219, ed. Putsch.
Marius Victorinus, p. 2471. Probus, p. 1390. Charts., p. 4, stq.
Terent. Maur., v. 1657. Mart. Capell., lib. 3. Voss., Art. Gramm.,
2, 29.ClassicalJournal, vol. 9, p. 339. Ramsay's Lot. Pros., p. 35.)
2. Consult note 1, page 51> where it would appear probable thit the
FINAL SYLLABLES. 83
Juv. Hannibal, et slantes Collina turre mariti.
Virg. Quum semel hasserunt arvis aurasque tulerunL
Virg. Versando terram experti, nihil improbus anser.
Id. Arboris acclinis trunco, procul area ramis.
EXCEPTIONS.
Exc. 1. Sal, 1 sol, 2 and nil contracted from nihil, are long.
Stat. Non sal, oxyporumve, caseusve. (Phalaecian.)
Auson. Sal oleum panis, mel, piper, herba, novem. (Pent.)
Ovid. UUerius spatium medio sol altus habebat.
Claud. Nil opis external cupiens, nil indiga laudis.
Exc. 2. Hebrew names ending in / have the final sylla-
ble generally long ; as, Daniel, Raphael, Ismael?
Tert. Quum magnus Daniel, qualis vir, quanta potestas !
Fortun. Qualiler out Raphael occursum impendent almae.
Victor. Nee tamen Ismael, Agar de semine natus.
earlier quantity of Hannibal, and other similar Carthaginian names, was
Hannibal.
1. There is great doubt whether sal ought to be regarded as an ex-
ception to this rule. It appears to be nothing more than an abbreviation
of the old nominative sale, still extant in a line of Ennius preserved by
Aulus Gellius (2, 26) : " Cceruleum spumat sale conferta rate pulsum."
Dr. Carey, therefore, thinks that it was in reality short, and that Statius
and Ausonius made it long merely by poetic license. The apocope could
never of itself lengthen sal from sale, since even those nouns in al, which
had the a long in die before the apocope took place, thence became short ;
as, cervical, tribunal, vectigal.
2. Cicero's derivation of sol from solus would supply us with a suffi-
cient reason for the long o in the former, if the etymology were really
worth anything. His remark is as follows : " Cum sol diclus sit, vel
quia solus ex omnibus sideribus est tantus, vel quia, cum est exortus,
obscuratis omnibus solus apparet" (N. D., 2, 27). So also Boethius
(Cons. Phil, 5, metr. 2):
" Quern, quia rcspicit omnta solus,
Verum possis dicer e solem."
The Lalin sol is rather to be traced to sauil, one of the three Gothic
forms for " sun," and both sol and sauil are related to the Sanscrit st*-
ria. (Grimm, Deutsch. Gramm., vol. 1, p. 611. Pott, Etymol. Forsch.,
vol. 1, p. 130.)
3. The Hebrew words have in the last syllable, in the original tongue,
the long vowel isert.
84
FINAL SYLLABLES.
FINAL m.
RULE. When a word ends in m, and is immediately fol-
lowed by a word beginning with a vowel or h, the poets of the
Augustan age and their successors generally elide the m by
a figure termed ecthlipsis, and also cut off the vowel prece-
ding the m by another figure termed synalcepha : as, monstr'
horrend' informe for monstrum horrendum informe, &C. 1
Virg. Monstrum ftorrendum informe ingens cui lumen ademp-
tum.
EXCEPTIONS.
Exc. 1. The older poets, or those prior to the Augustan
age, frequently preserved the final m from elision, and
made the preceding vowel short.
Ennius. Insignita fere turn millia militum octo.
Id. Dum quidem unus homo Roma tola superescit.
Lucil. Pr<Ktext<K ac tunicas, Lydorum opu 1 sordidum omne.
Vomerem atque loceis avertit seminis ictum.
Nam quod jluvidum est, e levibus atque rotundis.
Sed dum abest quod avemus, id exsuperare videtur.
Lucret.
Id.
Id.
OBS. 1. An instance of m being retained before a vowel
occurs even in Horace (Sat., 2, 2, 28) :
Quam laudas, pluma ? cocto num adest honor idem ?
1. Consult remarks on Ecthlipsis and Synalcepha, among "Figures of
Prosody." In strictness, no grammatical figure, such as ecthlipsis (lic-
BTu^ig, i. e., "a dashing out"), lakes place here, but the whole is a mere
matter of pronunciation. The final m was never fully sounded among the
Romans, as Priscian expressly remarks : " M obscurvm in extremitatc dic-
tionum sonat." Quintilian also, who in one part calls m a "quasi mu-
giens littera" (12, 10, 31), observes in another passage, "M littera, quo-
ties ultima est, et vocalem verbi sequenlis ila contingit ut in earn transire
posxit, etiamsi scribitur tamen parum exprimitur, ut multum ille, et quan-
tum erat ; adeo vt pane cujusdam nova littera sonum reddat ; neque
enim eximitur sed obscuratur, et tantum aliqua inter duas vocales velut
nota. est tie ipsat coeant" (9, 4, 40). It would appear, therefore, that tho
Romans did not^ive to such a syllable as om or nm a full labial sound,
with a close compression of the lips, but uttered the m with a slight na-
sal sound, such as the French give it, for example, in the word faim,
and as the Portuguese enunciate it even in Latin words. It would seem
that even in Hebrew the final mem was not very clearly enunciated ; at
least, sin-h is the opinion of Gesenius (Hcb-. Grtimm. Anm., $78, 2 a)-
FINAL SYLLABLES. 85
OBS. 2. But the best and purest writers seem in general
to have retained this practice only in words compounded
of com (or con) and of circum ; as, comes, comedo, circumago,
circumeo.
Ovid. Tu tibi dux comiti ; tu comes ipsa dud. (Pentam.)
Juv. Luctantur paucas, comedunt coliphia paucae.
Id. Quo le circumagas ? qua prima out ultima ponas.
Stat. Circumeunt hilares, et ad alta cubilia ducunt.
FINAL n.
RULE. N final has the preceding vowel for the most part
long, both in Latin words and in those of Greek origin ; as,
non, en, ren, splen, Siren, Hymen, Pan, Titan, quin, stn,
&C. 1
Virg. De grege non ausim quicquam deponere tecum.
Id, Dixerit, Hos calamos tibi dant (en accipe) Musts.
Ser. Et trita illinitur : vel splen apponitur haedi.
Catull. Hymen, O Hymentse ! Hymen, ades, O Hymenae !
Tibull. Lacte madens illic suberat Pan ilicis umbra.
Lucret. Flammiger an Titan ut alentes hauriat undas.
Ovid. Non potuit mea mens, qiiin esset grata, teneri.
Phaedr. Quern si leges, laetabor ; stn autem minus. (Iamb.)
To these add Greek accusatives in an from nominatives
in as, 2 and accusatives in en from nominatives in e or es ;
as also all Greek genitives plural in on; as, jEnean, Tire-
1. Rarnsay gives the rule of n final as making the previous vowel
short. We have thought it more advisable, however, to retain the old
form of expression. In Greek nouns, such as Siren, Hymen, &c., there
is a long vowel, in the original, in the final syllable.
2. There is some doubt with regard to the accusative in an from short
a in the nominative, bince some examples occur in which it is made long.
In all of these, however, the syllable is in the arsis, and we may there-
foie safely pronounce it to be naturally short. Thus, we have in Ovid
(Trist., 2, 395), " Qui Icgis Electran et egentem mentis Orestem ;" and
again (Met., 4, 756), " Protenus Andromcdan, et tanti pramia facti."
But then, on the other hand, we have Ori/hyian (00., Met., 6, 707);
Ossan (Property 2, 1, 19. OB., Fast., 1, 307) ; Iphigenian (Ov., E. P.,
3, 2, 62, &c. Ramsay's Lai. Pros., p. 67). Consult Exc. 4 undet
'.his rule.
H
86 FINAL SYLLABLES.
sian, Panelopen, Anchisen, Cimmerian, Chalybon, Meter*-
orphoseon, &c.'
Virg. Et stBvum jEncan, agnovit Turnus in armis.
Id. "Occurrit ; veterem Anchisen agnoscit amicum.
Catull. Jupiter ! ut Chalybon omne genus pereat ! (Pentam.)
Tibull. Cimmerian etiam obscuras accessit ad arces.
EXCEPTIONS.
Exc. 1. An, forsan, forsitan, in, tamen, viden\ satin 1 , 2 are
short.
Horat. Quis scit an adjiciant hodiernas crastina summtK.
Virg. Mittite ; forsan et hcec olim meminisse juvabit.
Id. Forsitan et Priami fuerint qua fata requiras.
Ovid. Ludit in humanis divina potentia rebus.
Virg. Hie tamen ille urbem Patavi sedesque locavit.
Tibull. Vota cadunt : viden', ut trepidantibus advolet alis ?
Terent. Satin 1 id est ? Nescio, hercle : tantum jussu 1 sum.
(Iambic.)
Exc. 2. Nouns in en, which increase short in inis in the
genitive case, have the final syllable short in the nom-
inative ; as, nomen (nominis), flumen (Jluminis), legmen
(tegminis), augmen (augminis).
Ovid. Nomen Arionium Siculas impleverat urbes.
Virg. Casperiamque colunt, Forulos, et jlumen Himellte.
Id. Tegmen habent capiti ; vestigia nuda sinistri.
1. As the Greek genitive plural ends in uv, the Latin on formed from
this is of course long. The later Latin poets, however, make errors al-
most continually in words borrowed from the Greek, which in the latter
language end in uv. Thus, we have in Prudentius (Perislcph., 2, 505)
the following iambic dimeter : " Dum dttmon inviclum dei," where the
on in dcEinon is erroneously shortened, the Greek form being Saipuv. So,
again, in the same writer (Psyehom., 857), the following hexameter oc-
curs : " Hie chalcedon Tubes perfunditur ex hyacinthi," where chalcedon
has the final syllable short, although the Greek form is xa/.Ktjdur.
2. Viden" is a colloquial form of vidlsne, and the change of quantity
is supposed to have resulted from the former being employed as a short
interrogation. So talW for satisnc is of very frequent occurrence in tho
omir writers. (Ramsay's Lot. Pra . p. 61.)
FINAL SYLLABLES. 87
Exc. 3. The final syllable on is short in the singular
cases of Greek nouns, which have those cases written
in the original with an omicron or short o , as, nom-
inative, Ilion, Erotion, Pelion ; accusative, Cerberon,
Menelaon, Rhodon.
Ovid. Ilion et Tenedos, Simo'isque et Xanthus, et Ide.
Mart. Pallida nee nigras horrescat Erotion umbras.
Ovid. Cerberon abstraxit, rabida qui percitus ira.
Id. Tu fore tarn lentum credis Menelaon in armis?
Horat. Laudabunt alii claram Rhodon, out Mytilenen.
But Greek accusatives in on, of the Attic dialect, havir g
an omega or long 0, in the original, are long ; as,
Athon, Androgeon, Peneleon, Demoleon.
Exc. 4. Greek accusatives in an, of the feminine gender,
are also short ; as, Maidn, Iphigenidn, Orithyian. 1
Ovid. Maidn et Electram Taygetamque Jovi. (Pent.)
Id. Nescio quam dicunt Iphigenian iter. (Ditto.)
Id. Orithyian amansfulvis amplectitur alis,
Exc. 5. Greek accusatives in in and yn are likewise
short ; as, Thyrsln, Daphnin, Pann, Tketm, Ityn.
Propert. Thyrsin et attritis Daphnin arundinibus. (Pentam.)
Ovid. Tantaque nox animi est, Ityn hue arcessite, dixit.
FINAL r.
RULE. R final has the preceding vowel for the most part
short ; as, calcdr, audiar, oleaster, iter, glorier, color, robur,
caditur.
Ovid. Crescit, et immensum gloria calcdr habet. (Pentam.)
Id. Trans ego tellurem, trans latas audidr undas
Virg. Infelix super at foliis oleaster amaris.
Id. Angustum formica terens iter, et bibit ingens.
Ovid. Fratre magis, dubito, glorier, anne vtro. (Pentam.)
\. Consult note 2, page 85.
88 FINAL SYLLABLES.
Virg. Sen plures color Hie vias et caeca relaxat.
Id. Vomis et inflexi primum grave robur aratri.
Id. Casditur et tilia ante jugo levis altaque fagus.
EXCEPTIONS.
Exc. 1. Cur is long, and also Ndr,far,fur, and ver. 1
Horat. Malta quidem dixi, cur excusatus abirem.
Virg. Sulfurea Ndr albus aqua, fontesque Vdini.
Ovid, Far erat, et puri lucida mica salis. (Pentam.j
Mart. Callidus effracta numos fur auff.ret area.
Ovid. Et ver auctumno, brumae miscebitur cestas.
Exc. 2. Greek nouns in er, which in the original end in
TIP, and which increase in the genitive, have the final
syllable of the nominative long ; as, aer (afjp, depog),
(Ether (aiOrip, aWepog), crater (Kparrjp, Kparrjpog), &c.
But pater and mater (narrip, rcarpog ; ^rjTrjp, jtw/rpdf)
have the final syllable short.
Lucret. Inde mare, inde acr, inde (ether ignifer ipse.
Ovid. Summus inaurato crater erat asper acantho.
Virg. Est, mihi namque domi pater, est injusta noverca
Id. Non jam mater alit tellus viresque ministrat
OBS. Hector, Nestor, and Castor, however, though com-
ing from "Earcop, Nearwp, and Kdffrwp, have the final syl-
lable shott.
Ovid. Hector erat : turn colla jugo candentia pressos.
Id. Cum sic Nestor ait, vestro fuit unicus asvo.
Horat. Infamis Helena Castor offensus vice. (Iambic.)
Exc. 3. Iber is long, but Celtiber has the final syllable
long in Catullus and short in Martial.
1. Cur is merely a contraction from quur, and consequently long.
(Priscian, vol. 1, p. 45, ed. Krchl.) Fir apparently gets its long quan-
tity from the Greek (j>up. Far, if we may judge from its genitive /arm,
was originally written fdrr, being long by position. Ver is from the
Greek fjp (a contraction from lap) with the digamma prefixed.
FINAL SYLLABLES. 89
Lucan. Si tibi durus Iber, aut si tibi terga dedissct.
Catull. Nunc Ccltiber es : Celtiberia in terra. (Scazon.)
Mart. Ducit ad auriferas quod me Salo Celtiber oras.
OBS. 1. Par and lar are usually accounted long ; and so,
indeed, they are found, the former very frequently, the lat-
ter in one instance in Ovid (Fast., 5, 141) ; but it would
sf em more consistent with accuracy to call them common. 1
OBS. 2. The quantity of cor has also been made a matter
' of dispute. The best opinion, however, is in favour of its
being accounted short. 2
FINAL as.
RULE. Final as is long ; as, terras, tempestas, tractas,
veniebds. 3
Virg. Turbabat cceZo, nunc terras ordine longo.
Id. Forte sua Libycis tempestas appulit oris.
Horat. Tractas et incedis per ignes. (Iambic.)
Ovid. Dure quid ad miseros veniebas exults annos.
1. The reasons that have been assigned in support of this latter opin-
ion are as follows: 1. Par and lar increase short; and all other nouns
in ar, which have a short increment, have the final syllable short. 2.
Even those which from are (with the a long) are reduced by apocope to
ar, have the ar short ; as, calcar, pulvindr, torcular. 3. Valerius Pro-
bus says, " Nominations singularis, R litera finilns, omni genere ....
brevem habet." (Putsch, Gram. Lat., col. 1393) ; and Servius (ad
&n., 3, 91) remarks, " Omnia monosyllaba ad artem non pertinent."
4. The compounds of par are found short in Prudentius (In Symm., 8,
5), Avienus (Fab., 23, 8), and Martianus Capella (6, 55), whose author-
ity (though not sufficient to outweigh that of earlier writers) may be al-
lowed to have some weight in a doubtful or probable case, when sup-
ported by reason and analogy. (Carerfs Lat. Pros., p. 140, 3d cd.)
2. It is shortened, for example, by Cicero (Tusc. Quasi., 3, 26), by
Ovid (Trisl., 5, 8.Ep. ex Pont., 1, 3, 32. Met., 5, 384), by Mar-
tial (10, 15), and by Paulinus (tie Cels. Ob., 379). In opposition to all
these authorities, the following line has been cited from Ovid (Her., 15,
79) : " Molle meum levibus cor est violabile telis." Burmann, however,
conjectures, " Molle mihi levibusque cor cst violabile telis ;" and a Frank-
fort MS. has " Molle meum levibus'que," &c.
3. In terras, and other accusatives plural of ttie first declension, as is
long because contracted from acs. In nominatives like tempestas, it is
long because the old form was tempcstdls. In tractas arid tie like, it is
long because contracted from ais.
H 2
90 ' FINAL SYLLABLES.
EXCEPTIONS.
Exc. 1. Anas, "a duck," has the final syllable short.
Petron. Et pictus anas enotata pennis. (Phalaecian.) 1
Exc. 2. Final as is short in the nominative of Greek noun i
which form their genitive singular in dos (or in the Latin
dis) ; as, Areas, genitive Arcados or Ar cadis ; Pallas,
genitive Pallados or Palladis.
Mart. Cum quibus Alcides, et pius Areas erat. (Pentam.)
Ovid. Bellica Pallas adest, et protegit agide fratrem.
OBS. But Pallas,- genitive Pallantis, Calchds, genitive
Calchantis, and the like, follow the general rule, and have
xs long.
Virg. Tela manusque sinit : Hinc Pallas instat et urget.
Ovid. Quam postquam reddit Calchds ope tutus Achillis.
Exc. 3. Final as is also short in Greek accusatives plural
of the third declension ; as, herods, lampadds, delphinds.
Virg. Permistos herods, et ipse videbitur illis.
Tibull. Accendit geminas lampadds acer Amor. (Pent.)
Virg. Orpheus in sylvis, inter delphinds Arion.
FINAL 6S.
RULE. Final es is long ; as, spes, noctes, vides, pones*
Ovid. Una tamen spes est, qua me soletur in istis.
1. This line occurs in Petronius Arbiter (c. 93, 4), but B-jrmann con-
jectures avis for anas.
2. Ennius furnishes one instance of the Latin plural es being short, in
the following line : " Virgin? nam sibi quisque domi Romanus habct sas"
(Enn., Fragm., p. 32, ed. Column.). Cicero is said to give another in
a fragment of his poetical version of Aratus (v. 472) : " Obruitur Pro-
cyon ; emergunt allies vna ;" but Ernesti reads, " Obruitur Procyon ;
emergunt alite lapsu," &c. The common text of Ovid (Her., 10, 86)
also contains a reading which exhibits cs in the accusative plural short :
" Quis scit an hccc sactas insula tigrcs habet." Burmann, however,
gives the line as follows : " Quis scit, an Jute sitras tigridas insula ha-
bct J" 1 and observes, "Duo sunt qua- in hoc tcrsu offendu.nl. Prirno,
quod Latinc haud dicitur, Quis scit an habet, sed an habeat. Dcindt
quod posleriorem in tigres corripit." (Burmann ad OP., /. c.)
FINAL SYLLABLES. 91
Virg. Nodes atque dies patet atri janua Ditis.
Properl Hoc quodcunque vides, hospes, quam maxima Roma.
Horat. Pones iambis sivejlamma. (Iambic.)
EXCEPTIONS.
<2 -ij Exc. 1 . Nouns in es of the third declension, Avhich in-
crease short in the genitive, have es in the nominative
short ; as, hospes, caspes, ales, miles, praspes, &c. (in
the genitive hospttis, c&spitis, atttis, militis,' prcepetis).
Ovid. Vivitur ex rapto : non hospes ab hospite tutus.
Rutil. Exiguus regum rectores casspes habebat.
Virg. Namque volans rubrafulvus Jovis ales in athra.
Id. Myrmidonum Dolopumve out duri miles Ulixi.
Id. Acer, anlielanti similis, quern prcepes ab Ida.
, OBS. But aries, abies, paries, and Ceres, as also pes, with
its compounds, follow the general rule.
Virg. Creditur : ipse aries etiam nunc vellera siccat.
Id. Populus influviis, abies in montibus altis.
Horat. Votiva paries indicat uvida. (Choriambic.)
Virg. Flava Ceres alto nequidquam spectat Olympo
Manil. Desuper Auriga dexter pes imminet astro.
Horat. Omnia magna loquens : modo sit mihi mensa tripes et.
Virg. Stat sonipes, acfrenaferox spumantia mandit.
Id. Tollit se arrectum quadrupes, et calcibus auras.
Exc. 2. Es, in the present tense of the verb sum, is
short, together with its compounds poles, abes, ades,
obes, prodes, &C. 1
1. In many passages of Plautus, an enumeration of most of which
may be found in \Vasse (cap. 16, p. 226, seqq.), es from sum occurs
with a long quantity. These passages are too numerous to permit our
supposing the syllable lengthened by a mere license, or by the force of
the arsis, and it is therefore probable that, in the time of Plautus, es
from sum, corresponding as it did to the Greek elf, was actually long,
and was only shortened at a subsequent period. S''ch, at least, is the
opinion of Schneider (Gr. Lat., vol. 2, p. 757.) According to Vossius,
es, " thou eatest," til i second person of edo, is long, being contracted,
92 FINAL SYLLABLES.
Virg. Quisquis es, amissos hinc jam obliriscere Grows
Id. TV poles unanimcs armare in prazlia fratres.
Id. Tuque ades inceptumque una decurre laborem.
Exc. 3. The preposition penes has the final syllable short.
Horat. Quern penes arbitrium est, et jus, st norma loqucndi,
Ovid. Me penes est unum vasti custodia mundi.
Exc. 4. Es is likewise short in Greek neuters ; as, ca-
coethes, hippomenes, <tc. ; and in Greek nominatives
and vocatives plural of the third declension, from nouns
which increase in the genitive singular, but which do
not form that case in cos ; as, Tritones, Arcades, Tre-
es, Amazones, Troades, &C. 1
Juv. Scribendi cacoethes, et <Rgro in corde senescit
Stat. Armigeri Tritones eunt, scopulosaque cete.
Virg. Ambo florentes aztatibus, Arcades ambo.
Id. Pulsant, ct pictis bellantvr Amazones armis.
OBS. 1. But nominatives and vocatives plural in es, from
Greek nominatives forming the genitive singular in eos, are
long ; as, h&reses, crises, phrases, metamorphoses* &c.
OBS. 2. Where the Latin es represents the Greek 770, it
is of course long ; as in Alcides, Brontes, Palamedes, from
FINAL is AND yS.
RULE. Final t* and ys are short ; as, dulcis, lapis, Ks,
amabis. bibis, Thetis, Tethys, Itys, Capys?
Horat. Dulcis ineocpertis cultura potentis amid.
Tibull. Fac lapis his scriptus stet super ossa notis. (Pent.)
probably, from cdis. (Voss., de Art. Gramm., 2, 31.) Carey opposes
this, but on weak grounds.
1. Es here answers to the Greek er. and is short, as a matter of course.
2. Because cs here answers to the Greek tif.
3. Final ys corresponds to the Greek vj-, which is for the most put
ebon.
FINAL SYLLABLES.
93
Lucan. Ante Ms exactum quam Cynthia conderet orbem.
Mart. Et bibis immundam, cum cane,pronus aquam. (Pent.)
Ovid. Tethys et extremo satpe recepta loco est. (Pentarn.)
Virg. At Capys, et quorum melior sententia menti.
EXCEPTIONS.
Exc. 1. All plural cases ending in is have that syllable
long ; as, Musis, terns, nobis, vobis, illls, amaris. 1
Mart. Carmina quod scribis, Musis et Apolline nullo.
Lucret. Secernunt, calumque a terris omne retentant.
Nobis est ratio, solis lunaque meatus.
Abstuht omne Phaon quod vobis ante placebat.
Pinguia concipiunt, sive illis omne per ignem.
Strymonieeque grues, et amaris intuba Jibris.
Exc. 2. Final is is long in the second person singular of
verbs of the fourth conjugation ; as, sentis, fastidis, au-
dis ; to which add /is, from^zo. 2
Horat. Sentis, ac veluti stet volucris dies. (Choriambic.)
Id. Pocula, num esuriens fastidis omnia prater.
Propert. Non audis ? et verba sinis mea ludere, cum jam.
Horat. Lenior ac melior fis, accedente senecta?
Exc. 3. Glis, vis whether noun or verb ; veils and sis*
with their compounds, as quamvis, notis, mails, adsis.
possis* have the final is long.
1. Plural cases in is were anciently written with the diphthong ei;
as, Museis, terreis, &c.
2. The syllable is in verbs of the fourth conjugation is the result of
contraction, and therefore long. Thus, we have audits, contracted au-
dis ; sentiis. sentis, &c.
3. Sis is formed by contraction from ties. The old forms siem, tia,
net, occur frequently in Plautus.
4. In Juvenal (5, 10) some read possis with the final syllable short.
Ruperti, however, condemns this reading, and substitutes possit. So in
Ovid (Her., 12, 71), nescis is said to occur with the final syllable short,
but erroneously. It appears neither in the edition of Heinsius nor in
that of Burmann. The latter merely mentions it in a note, as a reading
which is in direct violation of the metre.
94 FINAL SYLLABLES.
Mart. HOEC tibi si vis est, si mentis tanta potestas.
Id. Bellas homo et magnus, vis idem, Cotta, vidert.
Id. Esse veils oro serus conviva Tonantis.
Horat. Cum sis, et prove scctum stomacheris ob unguerrn.
Propert. Quamvls tile sua lassus requiescat avena.
Juv. Et cut per mediam nolis occurrere noctem.
Horat. Magnas Grdcorum malts implere catervas.
Virg. Adsts, O Tegeae, favens ; oleasque Minerva.
Horat. Non possts oculis quantum contendere Lynceus.
Exc. 4. The adverbs forts, gratis, and ingrafts have the
final syllable long. 1
Horat. Ne biberis diluta, forts est promus et atrum.
Phaedr. Gratis anhelans, multa agenda nil agens. (Iambic.)
Lucret. Effugere hand potis est, ingrafts hasret et angit.
Exc. 5. Final is is long in those nouns which form their
genitive singular in entis, mis, or ttis, with the penult
long ; as, Simols (gen. Simoenlis), Salamts (gen. Sal
amlnis), Samnts (gen. Samnltis), Us (gen. lltis).
Ovid. Hoc that Simols; h<ec est Sige'ia tellus.
Lucil. Samnls in ludo ac rudibus causis satis asper
Exc. 6. The final syllable ris, in the second future of
the indicative and perfect subjunctive, is common ; as,
amaveris or amaverts. 2
Exc. 7. Final ys is long in such contracted plurals as
Erinnys for Erinnyes or Erinnyas. The following
line of Seneca (CEdip., 644) shows the use of the
1. Forts is in reality the ablative of fora, " a door," the same asforis
of the third declension. Gratis and ingrafts are contracted datives for
gratiis and ingratiis, which are found in the open form in the comic
writers.
2. Almost all the examples in which it is found long are in the arsis ;
but there is at least one instance in Horace which cannot be explained
upon this principle : " Si ture placans et horna" (Od., 3, 23, 3). A
numerous list of instances where ris occurs, either with the long or short
quantity, may be seen in Ramsay's Lot. Pros., p. 77.
FINAL SYLLABLES. 95
word, though it cannot be made any proof of the quan-
tity:
Et mecum Erinnys pronubas thalami traham.
FINAL OS.
RULE. Final os is long ; as, viros, pueros, custds, ventos,
jactatos.
Virg. Inter se cofisse viros, et cernere ferro.
Propert. Differat inpueros ista tropeea suos. (Pentam.)
Horat. Gustos amatorem trecento. (Iambic.)
Virg. Ventos et varium caeli praediscere morem.
Id. His accensa super, jactatos aquore toto.
EXCEPTIONS.
Exc. 1. Final os is short in compos, impos, os (" a bone"),
and its compound exos.
Ovid. Insequere, et voti postmodo compos eris. (Pent.)
Lucret. Exos et exsanguis tumidos perfluctuat artus.
Exc. 2. Final os is likewise short in Greek words, writ-
ten in the original with an omicron or short o ; as, Ili-
os, Tyros, Samos, Chios, Rhodos, epos. 1
Ovid. Turn cum tristis erat, dcfensa est Ilios armis.
Lucan. Et Tyros instabilis, pretiosaque murice Sidon.
Horat. Romas laudetur Samos et Chios, et Rhodos absens.
Id. Facta canit, pede ter percusso, forte epos acer.
FINAL US.
RULE. Final us .s short ; as, taurus, tempus, cultus, im-
probus, solibus, scindimus, intus.
Virg. Taurus, et adverse cedens Canis occidit astro.
Id. Tempus humo tegere, et jamdudum incumbere aratris.
Id. Conveniat, quae euro bourn, qui cultus habendo.
Id. Improbus, et duris urgens in rebus egestas.
\. But those words in which the Latin os represents the Greek wf re
tain their original quantity ; as, herds (f/put)> Minds (M f), & .
96 FINAL SYLLABLES.
Virg. Pulverulenta coquat maturis solibus astas.
Id. At prius ignotum ferro quam scindimus aquor
Id. Intus aquas dulces, vivoque sedilia saxo.
EXCEPTIONS.
Exc. 1. Monosyllables in us are long ; as, jus, plus, pus,
thus. 1
Pedo. Sed rigidumjus est et inevilabile mortis.
Mart. Emi hortos ; plus est : ins true tu ; minus ?st (Pent.)
Horat. Proscripti Regis Rupili pus atque vencnum.
Id. Angulus ille feret piper et thus ocius uva.
Exc. 2. Final us is long in nouns which increase with
long in the genitive ; as, virtus (gen. virtutis), tellus
(felluris), servitus (servilutis), palus (paludis). 2
Horat. Virtus indigno non committenda poetaB.
Prise. Divitias magnas hie tellus ipsa ministrat.
Phaedr. Brevi docebo. Servitus obnoxia. (Iambic.)
Virg. Cocyti, tardaque palus inamabilis undo.
Exc. 3. Final us is also long in the genitive singular,
and in the nominative, accusative, and vocative plural,
of the fourth declension ; as, genitive sing., manus ;
nom., accus., and voc. plural, manus. But nominative
and vocative singular, manus. 3
Pedo. Scilicet immunis si luctus unafuisset.
Lucret. Sensus ante ipsam genitam naturam animantis.
Virg. Saltus et saturi petito longinqua Tarcnti.
Exc. 4. Final us is also long in words transplanted from
1. This exception and the one immediately following belong in strict-
ness to the same head, namely, that of a long increment in the genitive.
2. Horace (Ep. ad Pis., 65) furnishes a solitary instance of pulus
with the final syllable short. Bentley proposes a different reading. The
line, however, is retained unaltered in the best editions. (For some re-
marks on the verse, consult Horat., cd. Anth.. p. 326, not. crit.)
3. In the genitive singular of the fourth declension the final us is 4
contraction from uis ; and in the nominative, accusative, and vocatiwe
plural from ucs. Both, therefore, are long, of course.
FINAL SYLLABLE OP A VERSE. 97
the Greek, in which us represents the Greek ov$ t
whatever the case may be ; as, Panthus (JlavfloDc),
Amathus ('Ajwaflovf), Mantus (Mavrovf), Didus (At-
rfovc), &c.
Virg. Panthus, Othryades, arcis Phcebique sacerdos.
Id. Est Amathus, est celsa mihi Paphos, atque Cythera.
Id. Fatidical Mantus et Tuscifilius amnis.
Varro. Didus atque suum misceri sanguine sanguen.
OBS. 1. Polypus has the final syllable short in Horace
(Epod., 12, 5), which it gets, not from the common Greek
form TroAvTTOff, but from the Doric TrwAvTrof, which will
account also for the lengthening of the initial syllable. 1
OBS. 2. The sacred name IESUS (in Greek IHSOT2)
is included in this exception, and has the final syllable
long.
SECTION XXIV.
FINAL SYLLABLE OF A VERSE.
The final syllable of every verse (except the Anapaestic
and Ionic a minore) may be either long or short, at the op-
tion of the poet ; that is, a long syllable may be used to
close a verse, though the measure require one that is short ;
or a short syllable may be used, though the measure re-
quire one that is long. Thus, in the first of the following
lines, the long syllable r& is made to stand in place of a
short ; and, in the second, the short syllable que stands in
lieu of a long.
Horat. Jam satis terris nims atque dir&. (Sapphic.)
Virg. Nesaee, Spioque, Thaliaque, Cymodoceque.
1. Athenaeus (7, 107) quotes the following line from Epicharmus,
where the Doric form occurs : HuXvTrol re, GTjmai. TE, KOL iroravai reu-
0('r5cf ; and another from Archestratus : Hul.VTroi Iv re Quay icai Kapia
elalv upiffToi. He then goes on to remark, Aupieif d' avrbv 6ia TOV u
KakoiJci: TTw/lvTrov, wf 'ETTi^ap/tof KOI ^i/nuvl6r]f 6' (j>tj' ,,TTU%,VTTOV
iitfifievoc." 'Arrt/coi 6e Trohvirovv. (Athcn., 7. -'vol. 3, p. 169, ed.
Sckweigk.) So the Greeks used both Oidixove, -o<5of, and Oidiirof, -OR
I
98 THE QUANTITY OF THE PENULT OF WORDS.
OBS. 1 . The principle on which the above rule depends
is not that the syllable in question undergoes, any actual
change of quantity, but simply, that, by reason of its posi-
tion at the end of the line, and the interruption which the
metre there sustains, the same strictness is not required as
in other syllables differently situated ; and hence the real
quantity of the syllable becomes so comparatively unimpor-
tant, that the poet has the license of which we are treating
allowed him. 1
OBS. 2. The exceptions in the case of jhe Anapaestic
and Ionic a minore measures will be explained when we
come to treat of those two kinds of verse.
SECTION XXV.
REMARKS ON THE QUANTITY OF THE PENULT OF
WORDS.
I. Patronymics in ides or ades usually shorten the penult;
as, Priamides, Atlantiades, &c. Unless they come from
nouns in eus ; as, Pelides, Tydldes, &c. 2
1. Compare the remarks of Hermann': " Quum in numeris tempora
omnia certa cssc ac dejinita dcbeant, facile intelligitur, in numeris ipsis
nihil tisquam posse anceps esse; itaque si qua inveniuntur ancipitet
syllaba, i. e., qua breves sint quum longa esse debeant, vel longa quum
dcbeant breves esse, cas, quod ad numerum attinel, pro talibus numerari,
quales debeant esse, etsi non sint tales. Id auiem nemo non videt sic
tanlum fieri posse, si qui sirU in numeris loci in quibus pravitas ista
mensura nihil out parum offcnsionis habcat. Hujus modi loci duo sunt.
Unus est in Anacrusi ex una brevi syllaba. Alter est in fine ordinis,
ubi, quoniam nihil scquitur, quod tcrminum ponat ccrtum, ac potiut
pausa quadam succedit, paritcr delitescit mensura pravitas. Unde vel
brevis syllaba longa locum tcnerc potcst, vel longa pro brevi esse."
(Herm., Elem. Doctr., 1, 9.)
2. In a paper on " Greek patronymics" (published in the European
Magazine for August, 1817), Dr. Carey, in remarking on the patronym-
ics 'Arpe57f. Hnfaidr/c., &c., and their corresponding Latin forms
(which he writes with ei instead of the long t alone, as), Atreidcs, Pe-
leides, observes, " I conceive, that wherever, in Greek or Latin poetry,
we find one of those patronymics in such a position as to allow the alter-
native of one long syllable or two short, we are, if not bound, at least
authorized, to pronounce the ei as two distinct syllables; thus producing,
in each of the following instances, a dactyl instead of the spondee, which
results from the ordinary mode of pronunciation ; ex. gr.,
THE QUANTITY OF THE PENULT OF WORDS. 99
II. Patronymics and similar words in ais, eis, itis, ois,
otis, ine, and one, commonly lengthen the penult ; as, Achd~
is, Ptolemais, Chryseis, jEne'is, Memphitis, Latois, Icario~
tis, Nerine, Arisione. Except Thebais and Phocais ; and
Nereis, which is common.
III. Adjectives in acus, icus, idus, and imus for the most
part shorten the penult ; as, jfEgyptiacus, academicus, lepi-
dus, legitimus ; also superlatives, as fort issimus, &c. Ex-
cept opacus, amicus, apricus, pudicus, mendicus, posticus, fi-
dus,infidus (bat perfidus, of per and fides, is short), bimus,
quadrimus, patrlmus, matrimus, opimus ; and two superla-
tives, Imus, primus. . ,
IV. Adjectives in emus have the penult long ; as, postre-
mus.
V. Adjectives in alis, anus, arus, irus, ivus, orus, osus,
vdus, urus, and utus, lengthen the penult ; as, dotdlis, ur-
bdnus, avarus, delirus, oBstivus, decorus, formosus, percrudus,
edurus, astutus. Except barbarus, opiparus.
VI. Adjectives in His, if derived from verbs, shorten the
penult; as, agilis, facilis, habilis, &c. But derivatives from
nouns usually lengthen it ; as, anilis, civllis, herllis, &c. To
these add exilis, subtilis ; and names of months ; as, Apri-
lis, Quinctllis, Sextilis : except huniilis, parilis, and also
similis. But all adjectives in atilis are short ; as, versati-
lis, volafilis, umbratilis, &c.
VII. Adjectives in inus, derived from inanimate things,
re, aval- avdpuv, ttal dl
Atreidas, Priamumque, et s&vum ambobus Achillem.
Thus also, instead of spondaic lines in the following instances (Iliad, 2.
9, and 17. 191),
'Aya/ie/wovof 'Ar
Oi Trport uffTV <j>pov K^MTO. revx
we should have verses of the regular form, with the dactyl in the fifth
place ; and the same remark applies to Hri?.eiuva, which often occurs in
the Iliad, and to various other patronymics, which it is not here neces-
sary to enumerate."
100 THE QUANTITY OF THE PENULT OF WORDS.
as plants, trees, stones, &c. ; from adverbs of time, or
from substantives denoting the four seasons of the year,
shorten the penult ; as, amaracinus, crocinus, hyacintKinus ;
cedrinus,faginus, oleag inus ; adamanimus, crystalltnus, sma-
ragdtnus ; crastlnus, diutinus, serotinus ; earinus, oponnus,
chimcrinus, therinus ; also annotinus, hornotlnus. To which
add bombycinus, clephantmits, which seem to refer rather to
the silk and ivory than to the animals themselves.
VIII. Adjectives in inns, derived from living things,
numeral distributives, proper names, and gentile nouns,
lengthen the penult ; as, agninus, caninus, leporinus ; binus,
trinus, quinus ; Albinus, Cralinus,jjTustinus ; Alexandrlnus,
Latinus, Venusinus, &c. To which add adjectives of
place ; as, coltinus, marinus, vicinus ; and those derived
from nouns denoting time ; as, matutlnus, vespertinus ; to-
gether with all other adjectives in inus not included in the
preceding rule : as,festlnus, libcrtlnus, inopinus,peregrinus,
suplnus, &c.
IX. Diminutives in olus, ola, olum, and ulus, ula, ulum,
always shorten the penult; as, urceolus,filiola, mustcolum ;
lectulus, ratiuncula, corculum, &c.
X. Adverbs in tim lengthen the penult ; as, oppidatim,
viritim, tributim. Except affatim, perpetim, and statim. 1
XI. Desideratives in urio shorten the antepenultima,
which in the second or third person is the penult ; as, esu-
rio, esuris, esurit. But other verbs in urio lengthen
syllable ; as, ligurio, liguris ; scaturio, scaturis.
1. But statim, signifying "on Ike pot," "steadily," "constantly,"
has the penult long. It occurs in Plautus (Amph., 1, 1, 84 ; Ib., 120)
and in Terence (Phorm., 5, 3, 7). It is said to be derived, in common
with statim, from the same verb, sto.
FIGURES OF PROSODY. f 101
SECTION XXVI.
FIGURES OF PROSODY.
I. CjESURA. 1
PART I.
I. The term Caesura is used in two different senses by
prosodians ; first, with reference to whole verses, and, sec-
ondly, as regards single feet.
II. In the former acceptation, it means the division of a
verse into two portions or members, affording a short pause
or rest for the voice in some convenient part, where that
pause may take place without injury to the sense or to the
harmony of the line ; as,
Virg. Tanta molis erat |] Romanam condere gentem.
Id. Errabant actifatis 1| maria omnia circum.
III. In this sense, however, it is usually, for distinction*
sake, sty led* the Caesural Pause, and is chiefly connected
with the consideration of hexameter verse. It will there-
fore be treated of more at large under that head.
IV. In its application to single feet, the caesura means
the division or separation which takes place in a foot, on
account of the syllables which compose that foot belonging
to different words ; as,
Virg. Pasto\res ovi\um tener\os de\pellere \ foetus.
In this verse the caesura, in its second acceptation, occurs
three times, viz., in the second foot, between res and ovi ;
in the third, between um and lener ; and in the fourth, be-
tween os and de. 2
V. There are three kinds of caesura in its second accep-
1. Casura, "a cutting off" '(in Greek ro/z??), from cado, "to cut."
2. Some give the name Caesura to the syllable that remains after the
completion of a fool ; as, res, um, and os, in the line given in the text.
The best prosodians, however, consider it more accurate to confine the
term to the separation or division which takes place in a foot, and to call
the residuary long syllable simply a long syllable, or a semifoot.
I 2
102 FIGURES OF PROSODY.
tation, namely, the Syllabic, the Trochaic, and the Mono*
syllabic.
VI. The Syllabic Caesura is that in which the first part
of the divided foot consists of the last syllable of a word ;
as,
Virg. Sylves\lrem tenu\i mus\am medi\taris av\ena.
Here the syllables trem (of sylvestrem), i (of tenui), and am
(of musam), are instances of this species of caesura.
The Syllabic Caesura may take place in a heroic verse
at what are technically called the triemimeris, penthemime-
ris, hephlhemimeris, and sometimes at the ennehemimeris. 1
Thus,
3 67
Virg. Si canilmus sylvlas sylv\as sint \ console \ dignas.
36 9
Id. Ille la\tus nive\um moll\i fult\us hya\cintho.
Here the caesural syllables mus (of canimus) in the first
line, and tus (of latus} in the second, are in the triemime-
ris, or third half-foot of the line ; the syllables (of sylvas)
and urn (of niveum) are in the penthemimeris, or fifth half-
foot ; the syllables <R (of sylvas) and i (of molli) are in the
hephthemimeris, or seventh half-foot ; and the syllable us
(oifultus} is in the ennehemimeris, or ninth half-foot.
VII. The Trochaic Caesura is that in which the first part
of the divided foot consists either of a long and short sylla.
ble, or, in other words, a trochee (~ "), remaining at the end
of a word, or of an entire word composed of one long and
one short syllable ; as,
Virg. Fortu\natus et \ tile de\os qui \ novit a\grestes.
I. These terms owe their origin to the practice, generally adopted by
the old grammarians, of measuring lines by half-feet. Thus, the tricm-
imcris is that portion of a verse (counted or measured from the beginning
of a line) which contains three half-parts (rpetf, "three," f/fii, "half,"
and f^epif, "a part"), i. e., three half- feet, or a foot and a half (two short
syllables being counted as one long). Again, the penthemimeris indi-
cates Jive half-parts (TTEVTE, "fine" j/fii, and //ep/f), or two feet and a
half ; the hephthcmimeris, seven half-parts (lirra, " seven," jjfu, and //c-
plf) ; and the ennehemimeris nine half-parts (ewea, " nine," rjfii, and
FIGURES OP PROSODY. 103
Here natus in the second foot, illc in the third, and novit in
the fifth, form each a trochee, and in each of these divis-
ions of the line a trochaic caesura takes place.
The Trochaic Caesura may occur in either of the first five
feet of a verse ; as,
Virg. Arma pro\cul curr\usque vir\um mi\ratur in\anes.
Id. Talia \ voce re\fert, o \ terque qua\terque be\ati.
Two successive trochees, however, in the second and third
feet must be avoided, since they give the verse a flippant
and undignified air. Thus,
Ennius. Ergo ma\gisque ma\gisque vir\i nunc \ gloria \ claret.
Propert. Et gravi\ora re\pendit in\iquis \ pensa qua\sillis.
In the third and fourth they are nearly as disagreeable ;' as,
Ennius. Prudent\em qui \ multa lo\quive ta\cereve \ posset.
But two successive trochees may agreeably occur in the
first and second feet ; as,
Virg. are ci\ere vir\os Mart\emque ac\cendere \ cantu.
Or in the fourth and fifth ; as,
Virg. Et glau\cas sali\ces casi\dmque cro\cumque ru\bentem.
VIII. The Monosyllabic Ceesura is that in which the first
syllable of the divided foot is a monosyllable ; as,
Virg. Hie vir hie | est tibi | quern prc\mitti \ saepius \ audis.
IX. Of the three kinds of caesura which have been here
enumerated, the syllabic seems to have been the principal
one in Latin versification ; and but few harmonious lines
can be found in which it is not introduced. Next in met-
rical effect is the trochaic.
1. In Homer, however, we have two remarkable instances of the use
of successive trochees in producing a beautiful onomatopoeia. Thus,
.U 6' uv\dvTa, KaT\uvTd, Kdp\avTu TE, \ 66x[ua \ r\ qlidov. (H.,
23, 116.)
tra ne6\ovSe Kv7<\ivdTo \ ^.aa? av\ai6^. (Od., 11,598.)
The following striking passage may also be cited from Virgil (jn., 1, 85):
Una Eu\rusque Not\usque rufrnt cre\berque pro\cdlis.
104 FIGDRES OF PROSODY;
X. On the caesura depend, in a very great degree, the
beauty and melody of a verse. While its presence serves
to give au-itnat ion to the line, and, by connecting together
the different words of which it is composed, imparts to it
an air of gracefulness and ease, its absence, on the other
hand, is marked by a total want of poetic harmony. A
line in which the caesura is either wholly omitted, or in a
great degree neglected, has, in fact, little to distinguish it
from common prose, and can only be admitted into Latin
poetry on occasions in which harmony is purposely avoid-
ed, as in many of the neglected hexameters of Horace.
XI. The following lines may serve to show the uncouth-
ness and inelegance attendant upon the neglect of the cae-
sura.
Ennius. Sparsis \ hastis \ late \ campus \ splendet et \ hor-
ret.
Id. Disper\ge kostes \ distrahe \ diduc \ divide \ differ.
Propert. Non me \ moribus \ ilia sed \ kerbis \ improba \ vicit.
XII. A caesura, however, is not indispensably requisite
in every foot of a verse. Although the most melodious
lines are certainly those in which the caesura is frequently
introduced, still a long uninterrupted series of them would
have any other than a pleasing effect ; and therefore it be-
comes advisable occasionally to omit the caesura in one or
more of the feet, and in this way to produce an agreeable
variety. The following rules have been laid down on this
subject, and deserve attention.
1. In the first foot of a verse the caesura may generally
be omitted ; as,
Virg. Pauperis \ et tugur\i con\gestum \ cespite \ culmen.
2. In the second foot the caesura is often omitted ; but,
wlren this omission takes place, the word which be-
gins the foot is generally of sufficient length to com
FIGURES OF PROSODY. 105
plete it, and leave a caesural syllable in the next foot ; l
as,
Virg. Squamea \ convolv\ens sub\lato \ pectore \ terga.
3. The caesura is not so frequently omitted at the pen-
themimeris as it is in the other feet ; and when it is
omitted in the third it always occurs in the fourth, and
generally in the second foot. When this omission of
the caesura at the penthemimeris takes place, the third
foot generally consists of the first two or three sylla-
bles of a word, which is finished in the next foot ; as,
Virg. Jussa mor\i qua [ sorti\tus non \ pertulit \ ullos.
4. In the fourth foot the caesura is not necessary, if there
be one at the penthemimeris ; as,
Virg. Pinguis et \ ingra\t(Z preme\retur | caseus \ urbi.
5. The syllabic and monosyllabic caesuras are seldom
introduced after the fourth foot, but the trochaic often
'occurs at the ennehemimeris, and in most instances is
conducive to the harmony of the line ; as,
Virg. S&pe le\vi somn\um sua\debit in\ire su\surro.
Id. Hinc al\ta sub \ rape can\et frond\ator ad \ auras.
6. When there is but one caesura in a verse, it is gener-
ally in the third foot, sometimes in the fourth, but nev-
er in the second ; as,
Virg. Quern mea \ carmini\bus meru\isset \ fistula \ caprum
7. In a pentameter verse, a syllabic caesura general!)
1. The frequent recurrence of the verb nescio as a dactyl, and of the
prepositions inter and intra as spondees, forming the second foot, appears
at first view inconsistent with this rule ; but, in reality, it is quite in accord-
ance with it. It has been clearly ascertained that the preposition and its
case were frequently pronounced with one accent as one word, and there
is reason to suppose that nescio was often connected in a similar manner
with the word which followed it. Thus, the words inter se were pro-
nounced, and consequently regarded in versification, as though they were
written intersc, and nescio quis as though written nescioquis. A similar
connexion is not unusual in English words ; thus, some body is pronoun-
ced somebody; no body, nobody; can not, cannot. (Bradley's Lot.
Pros., p. 49.;
106 FIGURES OF PROSODY.
takes place at the penthemimeris, and a trochaic in
the foot preceding the final syllable in the second hem-
istich or half-verse ; as,
Ovid. Nee quere\rer tard\os || ire re\licta di\es.
Id. Nil mihi \ rescriblas, U attamen | ipse ve\ni.
8. There is sometimes a monosyllabic caesura at the
penthemimeris of a pentameter, when the preceding
word is a monosyllable ; as,
Ovid. Magna tam\en spe.s \ est || in bonit\ate de\i.
9. The trochaic caesura is sometimes neglected in the
foot preceding the final syllable of a pentameter, and
the verse is concluded by a word of four or more syl-
lables ; as,
Ovid. Lis est \ cumform\a || magna pu\diciti\ae.
C JE s u R A.
PABT II.
f I. A syllable naturally short is occasionally lengthened
when it is caesural. 1 This, however, takes place chiefly in
hexameter verse, and more sparingly in elegiac and lyric
versification. Thus, in hexameters we have, among other
instances, the following :
Virg. Pectori\bus inhi\ans spir\antia \ consulit \ exta.
Id. Emicat \ Eurya\lus et \ muncre \ victor a\mici.
Horat. Cum gravi\us dors\o subi\it onus. \ Incipit \ ille.
Virg. Graius hom\o infect\os linq\uens profu\gus hymen\teos.
1. In ordinary language we ascribe this to the force of the casura ;
but the true principle on which the rule depends is simply this ; that the
pressure of the voice is exerted on the syllables in question, and produ-
ces the same effect as if they were pronounced with double length. The
stress of the voice in dactylic versification always falls, it must be re-
membered, on the first part of the foot. This part is called the arsis
(apfftf), or " elevation," the tone of the voice being here somewhat ele-
vated ; while the remaining portion is termed the thesis (-diais ), or " de-
pression," the voice being in this part of the foot comparatively depress-
ed. (Compare the remarks of Maltby on the JUus Mctricus, ch. 3, p.
xii. Lex. Pros.)
FIGURES OF PROSODY. 107
In the second line of the elegiac distich we have the fol-
lowing :
Tibull. Quicquid ag\at sang\uis \\ est tamen \ ille tu\us.
Propert. Vinceris \ out vin\cis || h&c in am\ore rot\a est.
Ovid. In liquid\Km redi\lt \\ csthera | Mortis eq\uis.
In lyrics the following occur :
Horat. Si nan \ peri\ret || immisera\bilis.
Id. Caeca tim\et ali\unde \ fata.
Id. Perru\pit Acheront\a Herculeus \ labor.
Id. Sifygit adamant\inos.
Id. Angu\lus ri\de( ubi \ non Hym\etto.
II. Instances even occur in which, on account of the in-
fluence of the caesura (to adopt the common phraseology),
the final m remains unelided before a vowel in the begin-
ning of the word following, and forms, with its own prece-
ding vowel, a long syllable. The following are among the
number :
Propert. O me \ feli\cem ! O \ nox mihi \ Candida et \ O tu.
Tibull. Et tant\um vene\ratur vir\um hunc | sedula \ curet.
Luc. Scit non \ esse cas\am. O \ vitas \ tuta fa\cultas.
Manil. Emerit\us ccel\um et \ Claudia \ magna pro\pago.
II. ELISION.
Elision is the cutting off of the final vowel or diphthong,
or of the two final letters, of a word, and is divided into
Synalapha and Ecthlipsis.
SYNALCEPHA.
. I. Synaloepha is the elision of a vowel or diphthong at
"I the end of a word, when the following word begins with a
vowel, a diphthong, or the letter h ; l as,
1. Quintilian applies the term Synalcepha, in one place (1, 5, 18), to
what is commonly called Synaresis ; as, Phatlion for Phaethon: and in
another (9, 4, 36), to what is usually styled Ecthlipsis ; as, Prcesidi 1 est
for Presidium est. As a general definition, however, he terms owak-
108 FIGURES OF PROSODY.
Virg. Conticuere omnes, intentique ora tenebant.
Id. Dardanidae e muris : spes addita suscitat iras.
Id. SCBVUS ubi siacid<K telo jacet Hector, ubi ingens.
Id. Humida solstitia atque hy ernes orate Serenas.
Which lines, in scanning, are read as follows :
Conticuer* omnes, intentiqu 1 ora tenebant.
Dardanid" e muris : spes addita suscitat iras.
Scevus ub" 1 JEacidas telo jacet Hector, ub j ingens.
Humida solstiti 1 atqu 1 yemes orate Serenas.
This, however, is done only in scanning, and not in wri-
ting, nor in the most approved mode of pronouncing a
verse. In the two latter cases, the lines are always writ-
ten, and generally pronounced, without any elision. 1
II. Synaloepha affects not only a single syllable, but also
two syllables sounded as one by synaeresis ; as,
Virg. Stellio ; et lucifugis congesta cubilia blattis.
Catull. Et earum omnia adirem furibunda latibula. Gal.
Here, in the first line, stellio et is pronounced slell yet;
and, in the second, omnia adirem becomes omn yadirem.
III. In addition to the elisions already mentioned, the
earlier Latin poets were in the habit of frequently eliding
the letter s, in words ending in is and us, when followed
by a word beginning with a consonant, and thus permitting
the vowel, which would otherwise have been long by posi-
tion, to remain short ; 2 thus,
oi$ai " coeuntcs literal (9, 4, 36). The word Synalcepha is from the
Greek ffvval.oidij, and, according to its etymology, refers, not so much
to the elision of one vowel before another, asto the blending of two
vowels or syllables into one. On this account, some have considered
the term Synalcepha, as commonly used, an improper one, and recom-
mend that, instead of Synalcepha. and Ecthlipsis, the general term Eli'
sion be substituted.
1 . The best opinion with regard to the ancient mode of pronouncing
Latin verse is this, that much of the apparent harshness of elisions was
removed, by giving the elided syllable so slight and imperfect a sound
that it could hardly be distinguished, and, consequently, interrupted but
little the measure of the verse.
2. Ennius, Lucilius, Lucretius, and the comic writers furnish the most
!
FIGURES OF PROSODY. 109
Enn. Te nunc sancta precor Venus, et genetrix patrifs)
nostri.
Id. Versibu(s) quos olim Fauni Vatesquc canebant.
Lucret. Ut quasi transactis sape omnibu(s) rebu(s) profun-
dant.
Catull. At fixus nostris tu dabt(s) supplicium. (Pentam.)
IV. We may notice here a peculiar species of elision or
abbreviation not unfrequently employed in short, quick
questions, by which the vowel is dropped in the interroga-
tive particle ne before- a consonant ;' thus,
Catull. Ten 1 provincia narrat esse bellam ? (Phalaecian.)
Propert. Vidistin' toto sonitus procurrere ccelo ?
Virg. Hectoris Andromache Pyrrhin' connubia scrvas ?
Horat. Men' moveat cimex Pantilius ? out cruciet quod.
V. It has already been remarked, under the rules for final
n, that in such forms as vidln' ut, the s is dropped before n,
and the vowel shortened. Both this and the last-mentioned
abbreviation are combined in viden' (for videsne) and vin'
tu (for visne tu).
Catull. Nostra verba. Viden' ? faces.
Horat. Tempore dicam ; hodie triccsima sabbata ; vin' tu.
These expressions, as might be expected, are very com-
mon in the comic writers. 1
EXCEPTIONS.
Exc. 1. Synaloepha never takes place in the words O,
numerous examples of this license. Some instances also occur in Cicero's
translation of Aratus, though comparatively few in number. About the
time, however, when Cicero wrote his Orator, the custom of dropping
the s before a consonant appears to have been fast going out of use ; at
least the following passage from the work just referred to is in favour of
such an opinion : " Quin etiam, quod jam subrusticum cidetur, olim au-
tcm politius, eorum verborum quorum ccedem erant postremcE du<t htertz
que sunt in optumus postremam literam detrahebant, nisi vocalis insc-
qucbatur. Ita non erat offensio in versibus quam nunc fugmnt poctce
noti ; ita enim loquebantur ; qui est omnibu' princeps, non omnibus
princeps, et vita ilia dignu' locoque, non dignus." (Cic., Orat.> 48,
161. Compare QuintiL, V, 4, 38. Mar. Viet., p. 2472.)
1. Ramsay's Lat. Pros., p. 114.
K
110 FIGURES OF PROSODY.
heu, ah, ai, io, proh, vae, vah, hei, and the like interjec-
tions, which sustain the voice, and retard the pronun-
ciation, on account of the feeling or passion which
they express ; as,
Virg. O pater, \ o homin\um div\umque as\terna pot\estas.
Id. Heu ubi \ pacta fid\es ? ubi \ qua jur\are sol\ebas.
Tibull. Ah ego \ ne poss\im || tanta vid\ere mal\a. (Pent.)
Ovid. Ipsc su\os gemit\us foli\is in\scribit et \ at at.
Id. Et bis i\o Areth\usa i\o Areth\usa voc\avit.
Exc. 2. Elision is sometimes apparently neglected in
the case of a long vowel or diphthong ; and this gen-
erally happens when the unelided vowel or diphthong
is in the caesural syllable. Hence arise the following
rules :
First. When a long vowel or diphthong occurs in a
caesural syllable, and the word hat follows begins
with either a vowel or diphthong, then the long vow-
el or diphthong sometimes apparently remains une-
lided; but, in fact, the long vowel or diphthong
parts, by means of elision, with one of its two short
component vowels, and the remaining short vowel,
being in the caesura, is lengthened again by the stress
of the voice.
Secondly. When a long vowel or diphthong occurs, not
in a caesural syllable, and the word that follows be-
gins with a vowel or diphthong, then the long vowel
or diphthong sometimes apparently remains uneli-
ded ; but, in fact, the long vowel or diphthong parts,
by means of elision, with one of its two short com-
ponent vowels, and the remaining short vowel, not
being in the caesura, continues short.
FIGURES OF PROSODY. Ill
ILLUSTRATIONS OF EXCEPTION ffECOND.
Rule First.
Virg. Et succ\us pecor\i, et \ lac sub\ducitur \ agnis
Id. Slant et \ juniper\l et \ caslanefe hirs\ut(B.
Id. Ter sunt \ cona\tl im\ponere \ Pelio \ Ossam. 1
Id. Amphi\on Dir\caus in j Act&\o Ara\cyntho.
Ovid. Et celer \ Ismen\us cum \ Phocai\co Erym\antho.
Rule Second.
Cat\:ll. Uno in \ lectulo \ erud\itul\i ambo. (Phalaecian.)
Virg. Credimus ? \ an qui am\ant ips\i sibi j somnia \ fin-
gunt ?
Id. Nomen et \ arma loc\um serv\ant te am\ice neq\uivi.
Id. Et long\um form\ose val\e vale \ inquit Ilola.
Id. Insulce | Ioni\o in magn\o quas \ dira Cel\ano.
Id. Te Coryd\on o Al\exi trah\it sua \ quemque vol\uptas.
Horat. Et Esq\uili\nv al\ites. (Iambic.)
OBS. The only exception in Virgil to the remarks con-"^
tained in the preceding paragraph, occurs in the following
line :
Glauco | et Pano\peos et \ Ino\o Meli\certce. (Georg., 1,437.)
Here the o in Glauco is not elided, but is allowed to remain
long, though not in caesura. The line, however, in all
probability, contains a false reading, and for Glauco we
should substitute Glaucoque. 2
*
1. In this line, the o in Pelio furnishes an illustration of rule second.
2. Ramsay's usual accuracy deserts him, when he regards this line as
transplanted without change from some one of the Greek poets, who
were much less strict, in matters of this kind, than the Latins. It hap-
pens, unfortunately for this remark, that the Greek, in that event, would
be, ThavKu K.OL TlavoTreiri IS 1 'Ivciu Me/lt/cepr?;, in which case the last
syllable of T/lavKu would be long, as a matter of course, without any li-
cense whatever. Aulus Gellius (13, 26) informs us that Virgil borrow-
ed the line from the Greek poet Parthenius, in whom it ran as follows :
TAav/cu nal Njjpfji nal 'Ivucj MeAt/ceprfl. (Compare Macrob., 5, 17.
Fulv. UTS., Virg. cum Grate. Script. O'/., p. 130.) This form of the
112 FIGURES OF PROSODY.
Exc. 3. Very rarely is a short vowel left unclided. This
happens only twice in Virgil, and in both cases there
is a pause after the word ending with the short A'owel,
so that in repeating the line the effect would not bo
disagreeable. 1 Thus,
Addam \ cerea \ pruna : hon\ns erit \ huic quoque \ porno.
(Ed., 2, 53.)
El ve\ra incess\u patu\it Ded. \ Ille ubi \ matrem. (jEn., 1,
405.)
OBS. For some remarks on elision at the end of a verse,
when the next verse begins with a vowel or diphthong, con-
sult the article Synapheia.
ECTHLIPSIS. 2
I. Ectklipsis is the eliding of the consonant m at the
end of a word, when the following word begins with a
vowel or the letter h ; and after the m is removed by ecth
lipsis, a synalcepha immediately operates, and cuts off the
rowel which preceded the m. Thus,
Catull. Omnia tecum una perierynt gaudia nostra.
Id. Sternum hoc sanctos f&dus amicitia. (Pent.)
In scanning which lines we must read them as follows :
Omnia tec 1 una perierunt gaudia noslra.
JEterrf hoc sanctas fozdus amicitia.
Greek line, however, still remains hostile to Ramsay's opinion. Wag-
ner, on the other hand, maintains that the true reading in Virgil is Glau-
coquc, as we have remarked in the text. His words are as follows :
" Nullus dubilo, quitm nusquam longam syllabam in to posilu produxc-
rit Virgilius, pessumdata etiam soni clcgantia, quin tcripscrit Glauco-
que. Apparel autcm nusquam facilius excidere potuisse hanc copulam,
per simplex q scilicet cxprimi solitam, quam hoc loco, praxcdcntibvs du-
abus ipsi Q similibus litcris, altera quidcm simillima, C ct O. Sic in
Medic. JBn., 1, 69, a m. pr. scriptum est SUBMERSASQ. DRUE pro
submersasq. obrue, &c. (Wagner, Quasi. Virgil., 11,3.)
1. Riimsay''s Lai. Pros., p. 11G.
2. The subject of Ecthlipsis has been in part anticipated in the re-
marks respecting m final, p. 84, seqq.
FIGURES OF PROSODY. 113
II. Ecthlipsis sometimes, by the aid of Synapheia, strikes
out a syllable at the end of a line, when the next line be-
gins with a vowel, and no long pause intervenes ; as,
Virg. Jamque iter emensi, turres ac lecta Latino\rum
Ardua cerne.bant juvencs, murosque subibant.
H re the second line, is to be read r 'Ardua cernebant, &c.,
the last syllable of Latinorum being cut off from the first
verse and prefixed to the second.
KXCEPTIOXS.
Exc. 1. The elision of m is sometimes neglected, espe-
cially by the older poets, and in this case the syllable
ending in m, when not a caesural one, is short ;' as,
l^ucret. Nam quod jluvidum est, e levibus atque rotundis.
Id. Sed dum abest quod avemus, id exsuperare videtur.
Exc. 2. In some instances the caesura operates to pre-
serve the final m unelided, in which case the syllable
thus saved from efision- becomes long ; as,
Propert. O me felicem ! O nox mihi Candida ! et O tit.
Before quitting the subject of elision, it may not be amiss
to add the following rules, which regulate its use.
1. A verse in which there are more than two elisions, is
most commonly deficient in harmony ; as, for exam-
ple, the following pentameter line :
Catull. Quam modo qui me unum atque unicum amicum habuit.
2. Elisions may generally be introduced into a verse
without diminishing its harmony, when the final vow-
el of a word is the same as that which begins the next
word, and when the elided vowel is either naturally
short, or is followed by a long syllable ; as,
1. This and the succeeding exception have been partially anticipated
in the remarks relative to final m.
K2
114 FIGURES OF PROSODY.
Virg. Ipse ego cana legam lenera lanugine mala.
Id. Turn casia atque aliis intexens suavibus herbis.
3. An elision has seldom a good effect, either when it
occurs in the first syllable of a verse, or in the end of
the fifth foot of an hexameter, or immediately after the
penthemimeris in a pentameter, or in a word ending
with a long vowel or a diphthong before a word be-
ginning with a short vowel ; as,
Horat. Nam ut ferula ccedas meritum majora subire,
Juv. Loripedem rectus derideat ^thiopem Albus.
Catull. Troja nefas ! commune sepulcrum Europae Asiaque.
Id. Me misero eripuisti omnia nostra bona. (Pentam.)
IIL SYN.ERESIS.
I. Synaeresis is the contraction of two syllables into.one,
and is resorted to by the poets in order to make a word ad-
missible into hexameter or other verse, which could not
otherwise be introduced without some degree of inconve-
nience or dissonance, or perhaps not at all.
II. Hence we have, among others, the following instan-
ces of this species of poetic license.
1. CONTRACTION OF CO.
aurea is pronounced as aura,
cered " " cera.
alvedria " " alvaria.
respondeamus " " respondyamus (4 syllables). 1
anteactos " " antyactos (3 syllables). 8
antedmbulo " " antyambulo (4 syllables).
antehdc " " antyac (2 syllables).
Virg. Aurea compomit sponda, mediamque locavit.
1. The y is to be pronounced here like the same letter when it stands
at the beginning of an English word.
2. This word and the two that follow may otherwise be pronounced
ant'actos, antyambulo, ant'ac, and the vowel c may be supposed to have
been actually elided before 1 w e one that succeeds.
FIGURES OF PROSODY. 115
Herat. Abdiderant furtim terris, et imagine eerea.
Virg. Sen lento fuerint alvearia vimine texta.
Lucret. Quid respondeamus nisi justam intcndere litem.
Tibull. Cum memor anteactos semper dolor admonet annos,
Mart. Anteambulones et togatulos inter. (Scazon.)
Horat. Antehac nefas depromere C&cubum. (Alcaic.)
2. CONTRACTION OF 61.
aeret is pronounced as #n.
aranei
aram.
baltei "
" baltt.
ferrei "
u fern.
aureis "
" auris.
anteirent "
" antlrent.
anteis "
" ant is.
anteit "
" antit.
Virg. Centum aerei claudunt vectes aternaque ferri.
Lucret. Nee nebulam noctu, neque aranei tenuiajila.
Virg. Exanimem, rapiens immania pondera baltei.
Id. Ferreijue Eumenidum thalami et Discordia demens,
1^, Atria, dependent lychni laquearibus aureis.
Id. Qui candore nives anteirent cursibus auras.
Horat. Te semper anteit saeva necessitas. (Alcaic.)
3. CONTRACTION OF 60.
alveo is pronounced as alvo.
aureo " " auro.
Typhoeo " " Typhoo.
Virg. Deturbat, laxatque foros, simul accipit alveo.
Id. Vulcanum alloquitur, thalamoque hose conjugis aureo
Id. Inarime Jovis imperils imposta Typhoeo. 1
1. The nominative of this proper name is Typhoeus (three syllables),
never Typhaef .
116 FIGURES OF PROSODY.
4. CONTRACTION OF 1C.
vindetmdtor is pronounced as vindemyator (4 syllables). 1
Formiano " " Formyano (3 syllables)
Horat. Vindemiator, tt invictus, cui saspe viator.
Catull. Formiano saltu nonfalso Mentula dives.
5. CONTRACTION OF U.
connubTis is pronounced as connubyls (3 syllables).
dcnarns " " denar]/is (3 syllables).
Paoniis " " Paonyis (3 syllables).
Ttentis " " Tacnyls (2 syllables).
Virg. Connubiis arvisque novis operetta juventus.
Mart. Denariis tibi quinque Martialem. (Phalaecian.)
Virg. Paeoniis revocatum herbis et amore Dianas?
Id. Puniceis ibant evincti tempora taeniis.
6. CONTRACTION IN 10.
connubib is pronounced as connubyb (3 syllables).
Idomenios " " Idomenybs (4 syllables).
Virg. Connubio jungam stabili propriamque dicabo.
Catull. Idomeniosne petam monies ? at gurgite in alto.
7. CONTRACTION IN t.
f -omontonum is pronounced as promontoryum (4 syllables).
Antium " " . Antyum (2 syllables).
omnium " " omnyum (2 syllables).
mensium " " mensyum (2 syllables).
lacrymantium " " lacrymantyum* (4 syllables).
1. The y, as has already been remarked, is to be here pronounced
like the same letter at the beginning of an English word.
2. In scanning this line some erroneously make Pctoni a dactyl. The
second syllable, on the contrary, is lon, the corresponding Greek form
being Tlaiuviof. Such a form as P&onits could only refer to the Pae-
onians (Tlaiovaf) of Thrace. (Consult Heyne ad Virg., ^En., 7, 779,
not. crit.)
3. Genitives in ium are frequently written without inserting the i,
FIGURES OF PROSODY. 117
Ovid. Inde legit Capreas promontoriurnywe Minerva.
Id. Et tellus Olrc&a, et spissi litoris Antium.
Lucret. Imbecillorum esse Gquum misererier omnium.
Ovid. Cum tua sint cedantque tibi confinia mensium.
Virg. Exclusi, ante oculos lacrymantiumjue ora parentum.
8. CONTRACTION IN 00.
cooluerint is pronounced as coluerint.
cooperiant " " coperiant.
cooperuisse " " coperuisse.
Lucret. Tandem cooluerint ea qua conjecta repente.
Id. Cooperiant maria ac terras immensa superna.
Id. Per terras omneis atque oppida cooperuisse.
9. CONTRACTION IN 6.
decsse is pronounced as desse.
deest " " dest.
deer at " " derat.
deero . " " dero.
deerrarunt " " derrarunt.
deerraverat " " derraverat.
Lucret. Talibus in rebus communi deesse saluti.
Virg. Deest jam terra fugas : pelagus Trojamne petamus
Id. Divitis uber agri Trojave opulentia deerit.
Lucret. Deerrarunt passim motus ab sensibus omnes.
Virg. Vir gregis ipse caput deerraverat, atque ego Daph-
II. Synaeresis may sometimes be referred to synaloepha,
or, in other words, the first vowel in several instances may
rather be considered as elided, than as uniting with the fol-
as, parentum, serpentum, mensum, &c. The poets, however, use the
open form also, when it can be conveniently introduced ; as, parentium,
terpentium, &c. ; nor must it be supposed that the contraction is purely
a poetic license, since it is found in the best MSS. of prose authors.
LRamsay's Lot. Pros., p. 124.)
118 FIGURES OP PROSODY.
lowing one to form one syllable. Thus many words com-
pounded with semi followed by a vowel, or by the letter h
with a vowel, such as semianimis; semianimus, semiadaper.
lus, semihians, semihomo, semiustus, &c., lose by elision the
final vowel of semi, and are to be pronounced in verse,
semianimis, semianimus, semiadaper tus, semihians, sem'homo,
sem'ustus, &C. 1
Virg. Semianimesywe micant digiti ferrumque retractant.
Lucret. Languida semianimo cum corpore membra videres.
Ovid. Obliquum capiat semiadaperta latus. (Pentam.)
Catull. Semihiante labello. (Choriambic.)
Virg. Semihominis Caci fades quam dira tenebat.
Id. Fama est Enceladi semiustum fulmine corpus.
III. Words which contain an h between two vowels oc-
casionally drop the h, and contract the two vowels into one
syllable ; thus, we have vehemens, vehementi, vehementer,
vehementius, and also vemens, vementi, vementer ; soprehensi
and prensi, dehinc and deinc, mhtlum and nilum, mihi and
ml, prohibeat, pronounced prtnbeat (three syllables), &c. 2
Virg. Transit equum cursu, frenisque adversa prehensis.
Id. Ingentes tollent animos, prensique negabunt.
Id. Cervici subnecte dehinc ubi libera colla.
Id. Eurum ad se Zephyrumque vocat, dehinc taliafatur.
Lucret. Nam sive est aliquid quod prohibeat efficiatque.
IV. More violent than these are ostrea, eadem, eodem.
e&dem, eosdem, torreat, debrsum, sedrsum, sebrsus, omnia,
vtetis, Gain, onundi, patrm, flmtant, pronounced as dissyl-
lables ; precantta, uizyges, duellica, prczbptarit, as trisylla-
bles ; prout, a monosyllable ; all of which might be, and
most of them often aie, employed without being contracted. 3
1. Among these some would place grarfolcns and suav'olens, but these
ought to be considered as two distinct words, grave olens and suave
olens. (Ramsay's Lat. Pros., p. 123.)
2. Vemens occurs in Horace, Epist., 2, 2, 120 ; vcmcnti in Lucre-
tius, 3. 153 ; vementer in the same, 2, 1023, and 4, 821.
3 Ramsay's Lat. Pros., p. 124.
FIGURES OP PROSODY. 119
Horat. Sudando, pinguem vitiis albumque neque os'.rea.
Virg. Una eademque via sanguis animusque sequ'jntur.
Propert. Hoc eodern ferro stillet uterque cruor. (Pentam.)
Lucret. Uvescunt, eeedem dispansas in sole liquescunt.
Id. Eosdem habuit secum quibus, est elata capillos.
Id. Quod sitis exurat miseros atque arida torreat.
Id. Pondera quantum in se est deorsum deducere pvg-
nent.
Id. At neque seorsum oculei, neque nareis, nee manus
ipsa.
Id. Seorsus item sapor oris habet vim, seorsus odores
Virg. Bis patrice cecidere manus; quin protenus omnia.
Horat. Qui sudor vietis, et quam malus undique membris.
Propert. Et qui nunc nulli, maxima turba Gabii. (Pentam.)
Lucret. Denique ccelesti sumus omnes semine oriundi.
Stat. Nocturnique canum gemitus, ubi lumina patrui.
Lucret. Per malos volgata trabeisque trementia fluitant.
Virg. Pr&ferimus manibus vittas et verba precantia.
Ovid. lazyges et Colchi, Metereaque turba Getaeque.
Lucret. Lanigera pecudes, et equorum duellicaproZgj.
Catull. Omnibus his Thesei dulcem praeoptarit amorem.
Horat. Pasco libatis dapibus. Prout cuique libido est.
V. I is sometimes considered as a consonant, and in
these cases had, it is probable, the sound of the English y
in young, yes, &c. Thus, abiete, ablefibus, anete, parieti-
bus, were pronounced in dactylic verse as abyete, dbyetibus,
aryete, paryettbus, the first syllable being considered long
by position. 1
Virg. jEdificant, sectaque intexunt abiete costas.
Id. Abietibus^wuenw patriis et mantibus aquas.
Id. Custodes sufferre valent. La \at ariete crebro.
Id. H&rent parietibus scalas, postesque sub ipsos.
VI. Upon the same principle, although without the plea
1. Ramsay's Lat. Pros., p. 131.
120 FIGURES OF PROSODY.
of absolute necessity, we find jluvloriim in a passage of
Virgil (Georg., 1, 482), where it must be pronounced jluv~
yonim; and Nasidieni in Horace (Sat., 2, 8, 1), which
must be pronounced Nasidyeni. In Propertius likewise,
we have abtcgni (4, 1, 42), to be pronounced abyegni, and
abiegna (3, 19, 12), to be pronounced abyegnte.
Virg. Fluviorum rex Eridanus, camposque per omnes.
Horat. Ut Nasidieni juvit te ccena beati?'
Propert. L&serat abiegni venter apertus equi. (Pent.)
Id. Induit abiegnae cornua falsa bovis. (Pent.)
VII. In the same way the poets took advantage of the
double power of u, and made it a consonant in words where
such a change, was necessary or convenient. In this way
tenuia, tenulus, tennis, tenue, tenuem, tenui, tenues, genua,
curruum, fortmtus, pitultd, become tenvia, tcnvius, tenvis,
&c., genva, currvum, fortvitus, pitvita, &c. 2
Virg. Vclleraque utfoliis depectant tenuia Seres.
Lucret. Qua neque mobilius quidquam neque tenuius cxstat.
Virg. Tenuis ubi argilla el dumosis calculus arvis.
Id. Genua labant, vastns quatit aeger anhelitus artus.
Id. Per compos pascuntur equi ; qu& gratia curruum.
Juv. Non quasi fortuitus, sed ventorum rabie, sed.
Horat. Pr&cipue sanus, nisi quum pituita molesta est.
Horat. Nee fortultum spernere cespitem. (Alcaic.) 3
Catull. Mucusque et mala pMIta nasi. (Phalaecian.)
1. NasuRene, without contraction, occurs in Martial (7, 8): " Aut
tigila aiit dormi, NdsitKene tUri." (Pentam.)
2. By combining the processes described in this and paragraph V.,
icnuiorc is pronounced tenv-ydrc. (Ramsay's Lai. Pros., p. 132.)
3. Fortultus, without contraction, occurs also in Phaedrus (2,*4, 4).
On the other hand, in Manilius (1, 182) we have fortvitiu.
FIGURES OF PROSODY. 121
IV. DLERESIS. 1
I. Di&resis is defined to be the division of one syllable
into two ; as, aurai for auras, suadent for suadent, miluus for
milvus, sylua for sylva, soluo for solvo, <kc.
Virg. JEthereum sensum, atque aura'i simplicis ignem.
Lucret. Atque alias alii irrident, Veneremque suadent.
Phaedr. Columbae s&pe quum fugissent miluiim. (Iambic.)
Horat. Nunc mare, nunc syluae. (Dact. Trim. Cat.)
Tibull. Nulla queat pcsthac nos solu'isse dies. (Pentam.)
II. In strictness, however, no such figure as diaeresis ex-
ists, nor did the poets ever assume the power of stretching
out words, although they sometimes contracted them. The
examples usually quoted of this so-called figure, including
those given under the preceding paragraph, are in reality
archaisms, or antiquated forms of expression, which were
adopted by the poets for the sake of convenience or orna-
ment.
III. As the Ionic dialect in Greek frequently resolves the
diphthongs ei and q into ijt, the Roman poets occasionally
availed themselves of that license i^ words of Greek origin,
written in the latter language with either of those diph-
thongs ; as,
Ovid. Flebilis indignos, Elegeia solve capillos.
Stat. Quas inter vultu petitions elegeia propinquat.
Ovid. Invocat, Hippomenes, Cytherela, comprecor ousts.
Virf. farce metu Cylhereia ; manent immota tuorum.
Ovid. Quid quod avus nobis idem Pelopeius Atreus.
Lucan. Eumenidum vidit vultus Pelopeius Orestes.
Propert. Pleiadum spisso cur edit imbre chorus. (Pent.)
Ovid. Pleiades incipiunt humeros relevare paternos.
Id. Pleias enixa est, letoque del imperat Argum.
Virg. Pleias, et Oreani spretas pede repulit amnes.
1. Aio/pfffif, "a separation," "a taking or drawing asunder," from
6ta. and alpea,
L
122 FIGURES OF PROSOD1
So also Phabeius and Pk&bJius ; Theseius and Thesei-
tis ; T/treicius, Threcius, and Thrdcius ; Pegaseius and Pe-
gaseus, &c.
V. PROSTHESIS.' APH^RESIS. S
I. Prosthesis is defined to be the addition of a letter or
syllable to the beginning of a word ; as, gnatus for natus,
gnavus for navus, &c.
II. In strictness, however, no such figure as prosthesis
exists. On the contrary, the forms which are cited as in-
stances of it are in reality archaisms, or antiquated modes
of expression.
HI. Apharesis is defined to be the cutting off of the first
letter or syllable of a word ; as, maragdus for smaragdus,
conia for ciconia.
IV. Aphaeresis is, in truth, as imaginary a figure as pros-
thesis. The forms usually cited as instances of it are ei-
ther archaisms or colloquial shortenings. ~^-
VI. SYI^OPE. 3 EPENTHESIS. 4
I. Syncope is defined tojbe the dropping of a letter or
syllable out of the body of a word ; as, aspris for asperis,
caldior for calidior, circlos for circulos, oraclum for oraculum,
periclum for periculum, &c.
Virg. Improvisum aspris veluti qui sentibus anguem.
Horat. Caldior est ? acres inter numeretur. Opinor.
Virg. Ac primum laxos tenui de vimine circlos.
Catull. Oraclum Jovis inter cestuosi. (Phalffician.)
Virg. Vota me.tu duplicant matres, propiusque periclis.
II. We must not imagine, however, that forms such &a
and KOTTTU.
FIGURES OF PROSODY. 123
these were all peculiar to the poets, or that syncope was a
figure the use of which was confined to tins class of wri-
ters. Many of these shortened forms, on the contrary, oc-
cur also in prose. Thus, for example, vinda is found in
Cicero (Ep. ad Fam., 16, 18). Quintilian (1,6) says that
calfacit was more common in his time than calefacit ; while
from the contracted periclum we have the verb periclitor,
and so on.
/- III. The words most frequently contracted by syncope
are the praeter tenses of verbs; as, scripsti for scripsisti,
dixti for dixisti, promisse for promisissc, illuxe for illuxisse :
the participles of compound verbs ; as, repostum for repos-
iturn : genitives plural; as, deum for deorum, amantum for
amantium ; and words which have a in the penult before
,.- the consonant I ; as, vinclum for vinculum. &c.
IV. Epenthesis is defined to be the insertion of a letter
i or syllable into the body of a word ; as, pluvi,fuvi, annum,
genuvi, which is done, according to the grammarians, in
order to lengthen the -short u of plui, fui, annul, genui. All
these so-called instances of epenthesis occur in Ennius.
V. Epenthesis is, however, an imaginary figure, like those
of which we have already made mention. " If we study,"
observes Ramsay, " the etymological formation of the Latin
verb, we soon discover, that, in a very large class, the per-
fect tense is distinguished by the addition of v to what
Bopp calls the crude form ; we shall also readily perceive
that the v is frequently dropped, as in amasti, amarunt, au-
dieram, &c. We shall then have no difficulty in recogni-
sing the true old perfects of pluo, fuo, &c., in pluvi, fuvi,
&c. ; and, as might be expected, they are found in the very
earliest specimens of the language, and scarcely, if ever,
appear after the time of Plautus. It would be just as rea-
sonable to assert that a v had been inserted in audiveram or
abivi, in order to lengthen the short i of audii and abii, as
to advance the same proposition with regard to pluvi,fuvij
124 FIGURES OF PROSODY.
and the rest. Moreover, we have the express testimony of
Priscian, that preterites in ui had the u long in the oldest
writers, especially those derived from the present in uo ; as,
eruo, erui ; arguo, argui ; annuo, annui ; and he quotes from
Ennius, ' Annuit sese mecum decernere ferrof in all of
which it is clear that the long quantity of the u pointed out
the recent disappearance of the w." 1
VII. Apocope. 2 PARAGOGE.*
I. Apocope is defined to be the omission of the final vow-
el or syllable of a word ; as, men 1 for mene.
II. In such instances as the one just cited, where a quick
colloquial interrogation occurs, the figure may be allowed
to stand. In other cases, however, its existence is purely
imaginary ; as, for example, in the following line from Lu-
cretius :
Disjectare solet magnum mare transtra, guberna,
f
where guberna is said to be by apocope for gubernacula.
III. Paragoge is defined to be the addition of a letter or
syllable to the end of a word ; as, amarier for amari, doce-
rier for doceri, legier for legi.
IV. Paragoge is also, in strictness, an imaginary figure,
A and amarier, doccrier, legier, &c., are merely archaisms, or
antiquated modes of expression, mistaken for a poetical or
grammatical figure.
VIII. TMESIS.*
I. Tmesis is defined to be the separation of a word into
two parts, for the purpose of inserting another word be-
tween them ; as,
1. Lot. Pros., p. 160. Compare Voss.. de Art. Gramm., 2, 13.
2. 'kTTOKOiri), " a cutting off," from UTTO and KOTTTU.
3. Tlapayuyy, " a leading or bringing beside," from Trapd and uyu.
In military phraseology, " the extending a line of troops beyond the flank,
and presenting a new line or front."
4. Tfu'jaif, " a cutting." from riiivu, " to cut."
FIGURES OF PROSODY. 125
Virg. Talis Hyperboreo Septem subjecta trioni.
Lucret. Languidior porro disjectis dis^we supatis.
Id. Ceelera de genere hoc, inter quaquomque pretantur.
Id. Conlaxat, rarejuc facit lateramina vasi.
II. Here the words septemtrioni, dissipatis, interpretantur,
and rarefacit, are supposed to be divided by the figure tme-
sis. The truth appears to be, however, that, in the earlier
forms of the language, many words which were used sep-
arately were in process of time compounded together ; and
hence, just as we should expect, these separations are much
more frequent in Lucretius and the older writers than in
those who succeeded them. 1
IX. ANTITHESIS. 2 METATHESIS. 3
I. Antithesis is defined to be the putting of one letter for
another; as, olli for illi, faciundum for faciendum.
II. Antithesis is, in fact, no figure at all. The forms com-
monly assigned as instances of it, are merely old modes of
spelling words, which are found under this shape on many
monuments, some of them of a late date, and certainly many
of them not antiquated even in the age of Augustus. 4
' III. Metathesis is defined to be a transposition of letters ;
as, Evandre for Evander, Thymbre for Thymber, corcodilus
for crocodilus.
Virg. H(EC mea magna fides? At non Evandre pudendis.
Id. Nam tibi, Thymbre, caput Evandrius abstulit ensis.
Phaedr. A corcodilis ne rapiantur, traditum est. (Iamb.)
1. Ramsay's Lot. Pros., p. 161. Compare the remark of Hermann
" Qiiam tmesin vacant grammalici, ea et natura ct usu prior fuit quana
avvdtcig. Sed nimirum, quoniam recentior consuetude, ad quam omnia
exigebant, avvdeaiv usurpabat, tmesin istam ut discessionem a communi
usu animadvertere jubebant." (De Emend. Rat., G. G.,p. 117.) Con-
sult also, on the subject of tmesis in early Greek, the remarks of Mat-
thise (G. G., vol. 2, p. 1398, 3d Germ. ed.\ Kiihner (G. G., vol. 2, p.
313), and Nagelsbach (Anmerk. zu Ilias, Excurs. 16).
2. 'Avn'0e<Kf, " a substitution," from avri and rid?]fti.
3. Mera0<rtf, " a transposition," from /J.STU and riQrmi.
4. Ramsay's Lai. Pros., p. 162.
L2
126 FIGURES OF PROSODY.
IV. The instances, however, that are given of metathesis
are merely antiquated forms of expression. Thu.s, Evandre
and Thymbre are nothing more than vocatives regularly
formed from old nominatives in us ; and corcodilus is mere-
ly another form for crocodilus, obtained through the Greek."
X. SYSTOLE. 2
7 Li I I. Systole is defined to be the shortening of a syllable,
which, from its natural quantity or from position, ought to
be long.
II. No such figure, however, as systole in reality exists ;
and we must not suppose that the poets could shorten syl-
lables at their own caprice.
i i j ) III. Under systole are placed, for example, those com-
pounds ofjacio which sometimes drop the^'. Thus,
For dbjicit, abjici, we find a&icit, abici.
adjicit, adjici, " adicit, adici.
conjicit, u conicit.
injicit, " inicit.
objicis, objicit, " obicis, obicit.
subjicit, subjici, " subtcit, subtci.
IV. These, however, are by no means cases of systole or
poetic license, but are, in reality, old forms of the language. 3
1. The Greeks appear to have said both Kponodtdof and
as they said icpadia and Kopila, KpaTcpoc. and Kaprcpoc, and as the Scotch
use indifferently frith and firth. (Ramsay, L. P., p. 162.) Gudius,
indeed, maintains that the form corcodilus is more correct than crocodi-
lus. His words are as follows : " Testari' possum, in optimis et sep-
tingentorum, ct amplius annorum membranis, non tantum apnd poetas,
iis in locis, ubi metri ratio id postulat, scd etiam in prosa apud oratores
et philosophos reperiri Corcodilus. Neque his obstabit, quod belluse is-
tius Niliacae nomen a Graecis Latini acceperint, qui Kpoic66ei3.oc scrip-
serint. Sic enim saepc vocabula, quae a Grascis habenl Latini, immuta-
runt. Sic lard'us a ppadvf, etc. Eodem modo ndprfpof et tsparcpof.
Non negandum vero est; poetas, quoties prima foret corripienda, ad im-
itationem Graecorum scripsisse Crocodilus. Sic Hor., Epod., 12, 11:
Postea unice tandem obtinuit ut scriberetur Crocodilus." (Gud. &d
Phecdr., 1, 25, 4.)
2. SvtTToX);, " a drawing together," from <n>v and ffre^Au.
3. Ramsay's Lot. Pros., p. 13& Pareus, Lex. Cnt.Noris,
FIGURES OF PROSODY. 127
V. Under systole we also find Tanked the shortening of
the penultimate syllable in the third person plural of per-
fects ; as, dederunt, tulerunt, steterunt, &c. It is probable,
however, that we have here the ancient pronunciation ; at
least the analogy of the language is in favour of it. 1
VI. To systole is also attributed the quantity of the ini-
tial syllable in aperio, omitto, hodie, and the like, which,
being compounded, as is said, of ad and pario, ob and mitto,
hoc and die, &c., ought to have the first syllable long. It
will be sufficient, however, to observe, in reply, that the
quantity of these and similar words is invariable, and must
therefore have been the result of the ordinary pronunciation,
and not of any poetic license. 2
VII. A similar remark maybe made with regard to videri
for videsne. In the latter of these two forms the syllable es
olaph. Pis., diss. 4, c. 4. Aul. Gell., 16, 7. The last-named writer
blames Laberius for using coicior, on the ground that it was " obsoletum,
ant ex sordidiore usu vulgi depromptum."
1. By some these anomalies have been attributed to the mistakes of
transcribers, who, it is supposed, may have written erunt instead of erant,
or in some cases crint, bftth which terminations are occasionally found
in MSS. Thus, in the following line from Virgil's Eclogues (4, 61),
" Matri longa. dccem tulerunt faslidia menses," we find three readings
in the MSS., tulerunt, tulerant, and tulerint. If we examine, however,
with care the different examples adduced, it will be found that in not a
few of these the indicative perfect cannot be struck out without great vi-
olence ; nor can we fail to perceive that a transcriber, when copying a
MS., if he came to such a quantity as tulerunt or steterunt, which he
might consider anomalous, would be much more likely to change it to
tulerint or lulerant, than to transform one of these into tulerunt, in vio-
lation of all ordinary rules. (Ramsay's Lot. Pros. p. 105.) Others,
again, have boldly contended that the poets who used these forms were
inadvertently guilty of a breach of prosody. There might be some plau-
sibility in this assertion, if the instances of this pretended violation of rule
were only few in number. It happens, however, that in Lucretius there
are no less than 15, in Virgil five, in Propertius five, in Ovid 19, &c.
It is idle, therefore, to talk in such a case of "inadvertent" errors. The
supposition in the text, that the penult of the perfect, in the third person
plural, was originally short, seems much more probable, and coincides also
with the general rule of c short before r, or, in other words, with the
analogy of the language ; and, in confirmation of this opinion, we may
cite the words of the grammarian Diomedes : " Fere in tertio ordine
plerumque veteres tertia persona fmitiva tcmporis perfccti, numeri plu-
ralis, E mediam vocalem corripiunt, quasi legerunt, emerunt," &c.
2. Ramsay's Lat. Pros., p. 156
128
FIGURES OF PROSODY.
is uniformly long ; whereas in viden? the e is as uniformly
short. The reason probably is this, that in videii, as well
as in satin' and ait/*', the pronunciation was so rapid that the
voice was not permitted to dwell upon any of the syllables. 1
XI. DIASTOLE. 2
I. Diastole (or Ectasis 3 ) is defined to be the lengthening
of a syllable naturally short. Like systole, however, it is
merely an imaginary figure.
II. To diastole is attributed the lengthening of the first
syllable in 'Italia, Pnamides, 'Arabia, &c., from ~Italus>
Priamus, "Arabs, &c. But the true reason is, that such
forms as Ildfid, Priarmdes, Arabia, <fec., could not have
been used in dactylic verse unless the first syllable were
lengthened. This license was therefore assumed by the
Roman poets, in imitation of the Greeks. 4
III. To diastole is also attributed the lengthening of the
first syllable in certain compounds of Re. Thus, we find
rellgio, reUglone, religionum, religiosa.
reliquiae, retiquiarum, reliquias.
reddere, recidimus.
relatum, remotum.
rcducit, reducunt, reducere.
To these must be added the three preterites reperit, repulit,
retulit, and the tenses formed from them, repereris, repere-
rit ; repuleris, repulerint ; retuleram, retulerat, retuleris, re-
tulerit, <fec. 4
IV. Various explanations have been given of this anom-
aly. According to some, such words as religio, reliquiae,
recidere, could not find a place in dactylic verse, unless the
first syllable were made long ; but, although this is true as
far as ihese Avords are concerned, it will by no means ap-
1. R(unsay''s Lai. Pros., p. 156.
2. Aiaarol.r), "a separation," "a drawing out," from did and trre'/Uu
3. 'E/crc<Tif, " a stretching out," from t/c and reivu.
4. Ramsay's Lai. Pros., p. 139, scq.
FIGURES OF PROSODY. 129
ply to relatus, remains, reducit, &c., which are generally
found with the first syllable short ; nor will it explain the
quantity of reperit, repulit, retulit, which is invariable. 1
V. Other scholars, following the old grammarians, con-
tent themselves with doubling the consonant after re in all
cases where that syllable is long, and write relligio, rel-
liquias, redduco, remmotus, &c.^but this at best is but an
evasion of the difficulty.
VI. A more plausible explanation is grounded upon the
supposition that the ancient form of re was red, as it appears
in reddo ; for it cannot be said that the d is here inserted
for the sake of euphony, as may be urged in regard to such
words as redeo, redimo, and the like. In this way the ori-
ginal form of refero, recido, removeo, reduco, &c., was red-
fero, redcido, redmoveo, redduco, &c. ; and, although the d
was afterward dropped, the poets, especially the earlier
ones, considered themselves entitled to make use of either
form, as best suited their convenience.
VII. This may be. the true account of the matter in most
instances, but it certainly does not apply to the preterites
reperit, repulit, retulit, which ought always to be written
with the consonant doubled, being in reality contractions
for the reduplications repeperit, repepulit, retetulit. This is
the account given by Priscian 2 with regard to repperit, and
it applies equally to the other two. 3
VIII. The first syllable in quatuor, when made long, has
also been regarded as an instance of diastole, since quat-
uor is found short in Ennius, and its derivatives qualer,
quaterni, quadrupes, &c., have the a short. But Gesner
and Vossius correctly maintain, that Virgil should be made
the standard of pronunciation with regard to the word in
1. The old reading in Catullus (66, 36), " Si reditum retulisset is,
aut in tempore longo, et" is now charged in the best editions to " Si
reditum tetulisset," &c.
2. Vol 1, p. 509, ed. Krehl.
:i. Ratnsan'x Lai. Pros , p. 140.
130 f FIGURES OF PROSODY.
question, and that the first syllable should be considered
long, as he uniformly makes it.
XII. SYNAPHEIA. 1
I. Synapheia is the connecting of verses together, so as
to make them run on in continuation, as if the matter were
not divided into separate verses.
II. By this arrangement, the initial syllable of a succeed-
ing verse has an influence on the final syllable of the pre-
ceding, affecting it by the concourse of consonants, by ec-
thlipsis, and by synalo3pha.
III. Synapheia chiefly prevailed in anapaestic and Ionic
a minore verse, in which strict attention was paid to its ob-
servance. Its occurrence in other kinds of verse was occa-
sional and limited, and in these kinds it was seldom resort-
ed to by good writers.
IV. The following anapaestic lines furnish examples of
the effect of synapheia :
Prticeps \ sylvas || montes\que fugit ||
Citus Act\&on, || agiUs\que magls ||
Pede per \ sdltus || et sax\a vagus \\
Mctult | molas || Zephyrts \ plumas. \\ (Seneca.)
The short final syllables of fugit, magls, and vagus here
become long by position before their own final, and the
initial, consonants in the lines that immediately follow.
V. Among other instances of synapheia, the following
may be enumerated :
Virg. Jnseritur vero etfatu nucis arbutus Aomlda
Et steriles platani molds gessere valentes.
Virg. Jactemur, doceas: ignari fiominumque Zocorw/nlque
Erramus, vento hue, et vast is fluctibus acti.
Ilorat. Dissidens plebi, numero beato\mm
Eximit Virtus, populumque falsis.
.
1. Swa^tto, " a junction," from avv and airru.
FEET. 131
Horat. Cur facunda parum deco\ro
Inter verba cadit lingua silentio ?
In the first of these examples, the first line ends with horn,
and the i, being regarded as the final syllable of the verse,
becomes long. The syllable da combines with Et in the
* second line, and by synapheia and synaloepha we have d'et
as the initial syllable of this line.
In like manner, the second line of the second example
commences with qu'erramus, the second line of the third
with r'eximit, and the second of the fourth with r'inter. In
r'eximit, it will be observed, an ecthlipsis operates.
VI. In most cases of synapheia (except in anapaestic and
Ionic a minore measures) there is little or no pause at the
end of the line.
VII. The rule of synapheia is carefully observed by the
Greek tragic writers in their anapaestic systems ; and, as
far as our authorities go, by the Latins also. The law of
synapheia was discovered by Bentley. 1
SECTION XXVII.
OF FEET.
I. A Foot is a combination of two or more syllables, with
due regard to their times or quantities.
II. Feet in metre are so called, because by their aid the
voice, as it were, moves along through the verse in a meas-
ured pace. 2
III. Feet are divided into simple and compound. Of the
simple feet, four are of two syllables, and eight of three.
The compound feet are sixteen in number, and contain each
four syllables.
1. Dissertation upon Phalaris, p. 190, cd. Dyce. For farther pbser-
vations respecting synapheia, consult remarks on the Anapasstic and
Sapphic measures.
2. " Pes vocalur, sive quia in percussione metrica pcdis pulsus poni-
lur tolliturque ; seu quia, ut nos pedibus ingredimur atque progredimur,
ita et versus per hos p< les metricos procedit et scandit." (Mar. Viet.,
p. 2485.)
132 FEET.
SIMPLE FEET OF TWO SYLLABLES.
I. The SPONDKE (Spondceus) consists of two long sylla
bles, as omnes, and derives its name from o~ovSi], ' a liba~
tion," because it was much used in the slow, solemn chant
which accompanied a sacrifice. 1
II. The PVRRHICH (Pyrrhichius) consists of two shoit
syllables, as deus, and was so called from the martial Pyr-
rhic dance (irvppixT]), which was performed in quick time,
and in the measure adapted to which this foot predomina-
ted. 2 \
III. The TROCHEE (Trochaeus) consists of one long and
one short syllable, as servat, and takes its name from rpi-
#a>, " to run," or from rpo^og, " a u-heel" in consequence
of the tripping character which it communicated to the
verses in which it prevailed. 3 It was also called by the
Greeks %opeiog (from %opof, " a dance"), and by the Lat-
ins Chorius or Choraeus, because it is well adapted to dan-
cing. 4
IV. The IAMBUS (Iambus) consists of one short and one
1 . 'O CTTrovdoiOf , e/c dvo /naKpiJv avyneipcvof e/c/ljyft? de ovrof , 6n kv
Tai c~ovdaif , uf LTTOIOVVTO Trpbf rovf fieovf , TOIOVTU ftirptf) k^puvro'
K. T. /I. (Scliol, HephcEst., p. 82, 4.) According to Diomcdes (fib. 3,
p. 473), this foot was called by Numa Pompilius Pontificius, from its
being employed in the music 'of the Salic hymns. (Consult Terent.
Maur., ed. Sanlen., p. 62.)
2. Hofa/j.iKTi 6e tioxei elvat TJ irvppix'n- "Evo-xhoi -yap avrrjv iratdef
bpxovvrai. Tujovf <5e del T& Tro^e/iO elf TO SIUKEIV, KOI elf TO i)TTup.e-
vovf <j>evyeiv, fH]6e peveiv, ftTjff aldeioftai Kanovf elvai. (Atkcnaus, 14,
p. 629, c.) Another derivation of the name is from Pyrrhus, the son of
Achilles, " gui crebris et cilis exsultaliombus bis breviter prominentem
clypeum genibus incumbens, et, per hunc, terrorem hostibus immittcns,
infer ebalur, sicut versus illustrat, VTraanidia Trpo6i6<JvTOf." (Diomed.,
3, p. 471.) The Pyrrhich was also called yyepuv, " the leader" because
by some it was ranked first among metrical feet. Another name was
Ai6pot;f, which the Latins rendered by Bibrevis. It was likewise call-
ed IJapta/ifiof (Par iambus'), " quod minus habeat unum tempus ab lam-
bo : Trapu enirn Graeci minus dicunt." (Mar. Viet., p. 2489. Compare
Schol. ad Hephast., p. II.)
3. Mar. Victorm , p. '2487. Schol., Hephast ., p. 82. Plntius, p.
2\^5.~Dwmcd.. )>. 474. Sa.ntcn..ad Ttrcnt. Muiir.,p. 69.
4. Tho names Chorins (or Qkoretui) and Trvchaux were also given
tu ti.e tr:bracb.
FEET. 133
long syllable, as pws. Most of the old grammarians would
derive it from lambe, the name of a damsel, the slave of
Eleusinian Celeus, who amused Ceres by dancing and
song when mourning the loss of her daughter Proserpina.'
SIMPLE FEET OF THREE SYLLABLES.
I. The MOLOSSUS (Molossus) consists of three long syl-
lables, as dclectant, and derived its name, according to the
scholiast on Hephaestion, from Molossus, son of Pyrrhus
and Andromache, who- repeated hymns, in which this was
the prevailing foot, at the shrine of Dodona in Epirus.*
II. The TRIBRACH (Tribrachys} consists of three short
syllables, as melius, whence its name is derived (rpelg,
" three" and j3pa#?;c, " short"}. It is also called Chorius,*
and sometimes Trochaus*
III. The DACTYL (Dactylus) consists of one long and
two short syllables, as carrriina, and derives its name from
dd/CTV/lof, " a finger" because each finger consists of one
long joint and two short ones. 5
IV. The ANAP.EST (Anap&stus) consists of two short
syllables and one long, as animos, and it is thus denomina-
1. Plotius, p. 2625. Schol., Nicand. Alex., v. 130, p. 139. Schol.,
Hephast., p. 81. Eustath. ad Od., 11, 277. Other derivations may be
found in the notes of Santenius to Terentianus Maurus, p. 67, seqq.
2. Others say that it was named from the Molossi in Epirus, who
used it in their war-songs. (Diomcd., p. 475.) It was known also by
several other names, such as Vortumnus, Extensipes, Hippius, Chaon-
ius. (Diomed., I. c.)
3. Schol., Hephast., p. 158.
4. QuintiL, 9, 4. Terent. Maur., v. 1446. Sanlen., ad loc.
5. Plotius, p. 2625. Cledonius, p. 1885. Victorinus, p. 2488, quo-
ted, among other authorities, by Santen (ad Terent. Maur., v. 1406).
Diomedes gives another derivation for the name : " ab Idens Dactylis.
Hi namque in insula Creta Jovem cuslodiendo, ne vagilu se parculus
proderct, lusus excogitate gcnere, clypeolis aneis inter se concurrentes,
tinnilu cris illisi rhylhmica etiam pedis Dactyli compositione celavere
vocem infantis." To the same effect the scholiast en Hephaestion re-
marks : j) U.TTO TL>V &a.Krvhuv, ov 7: i^adofisvrj TTJS y^f dvijuev ?f 'Pea,
ol Js/ca ovref, KOL TOVTU rw fisl.ei adovn^ KOL bp^ovufvoL, TOV
Kpovov JIKUTUV. (Schol , Hephffst., p. 82.) The dactyl was also call-
ed Herons, from i's being particularly employed in relating the exploit*
of heroes. (Aristi*., R.'ift., 3, 8. Cic., de 0/at., 3, 47, scq.)
M
134 FEET.
ted from the eerb dvoTrato), " to strike in reverse order" be-
cause the order of its syllables is directly the reverse of that
observed in the dactyl. 1 Hence the anapaest is also called
'AvTidaKTvkog (Antidaclylus) by the Greeks, and Retroac-
tus by the Latins. 2
V. The BACCHIUS (Bo/t^tof) consists of one short sylla-
ble followed by two long ones, as dolores, and it is so call-
ed from its having been frequently used by worshippers in
the hymns to Bacchus. 3
VI. The ANTIBACCH!US ('AvTifiaK^etof) 4 consists of two
long syllables followed by a short one, as pelluntur, and it
takes its name from its opposition to the Bacchius.
VII. The AMPHIMACER ('A/t$|uaKpo?), or CRETIC (Creti-
cus), is composed of one short syllable between two long,
or, in other words, of a long, a short, and a long, as cdsti-
tas. Both names are mentioned by Quintilian, who makes
the second, however, the more usual one. The former of
these two appellations comes from d[jupi, " on both sides"
and paKpog, " long," while the latter is said to have been
given to this foot from its resembling in time the blows
struck by the Corybantes in Crete on their brazen shields,
to drown the cries of the infant Jove, lest these should
reach the ears of Saturn. 5
VIII. The AMPHIBRACH (Amphibrachys) consists of o'ne
long syllable between two short, or, in other words, a short,
1. " Dictus irapa rot avarraleiv, Kara TO avaTra/.iv avriicpoveiv npbf
rov Ao/crvAov, quia recurrcndo repercutiens Dactylum sono reciproco ob-
loquilur ei per antistrophen." (Diomed., p. 475.)
2. Diomed., 1. c. Quintil., 9, 4, 81.
3. A great deal of confusion exists among the old grammarians with
regard to the bacchius and antibacchius. Terentianus Maurus, for el-
ample, makes the former consist of two long syllables followed by a
short ( ~), and the antibacchius the reverse of this (~ ). The
common distinction, however, as given in the text, has the authority of
Quintilian (9, 4, 82) in its favour. (Consult Santcn. ad Terent. Maur.,
I. c.)
4. Called also Palinbacchius.
5. Compare note 5 page 133, where a similar etymology is given for
the term dactylus.
.FEET. 135
a, long, anil a short; as. amdre. It derives its name from
d[i<f>i, " on both sides" and j3pa%vg, " short." 1
COMPOUND FEET.
I. The DispoNaEus, or Double Spondee, is composed of
four long syllables, or, in other words, of two spondees ;
as, inflmtls.
II. The PROCELEUSMATICUS consists of two pyrrhichs, or
of four short syllables ; as, hominibus. It is said to have
taken its name from Kshevopa, the word of command given
by the ballet-master, in double quick time, to accelerate the
step. 2
III. The DIIAMBUS, or Double Iambus, consists of two
iambi ; as, severitas.
IV. The DITROCEUEUS, or DICHOR^ECS, consists of two
trochees ; as, permanere.
V. The IONICUS MAJOR (or a Majore) consists of a spon-
dee and a pyrrhich, that is, of two long syllables followed
by two short ; as, calcanbus.
VI. The IONICUS MINOR (or a Minore) consists of a pyr-
rhich and a spondee, that is, of two short followed by two
long ; as, proper abdnt.'
The two feet just described are called Ionic, either from
their having been favourites with the lonians, or from Ion,
their inventor. 3 One is called lonicus Major, or a majore,
because it begins with the greater quantity, that is, with
two long syllables ; and the other is called lonicus Minor,
1. The amphibrach is also called Amphibrems. (Diomed., 3, p. 475.)
Victorinus styles it Mesites (p. 1956).
2. 'Hywro 6s KO& EKaarov %6pov elf avrjp, o? hediiov TOtf dA/loif ra
T?iq opx^OEUC oy^uetd irpuTOf, d6ot/>opuv rdf TroAe/w/cdf ical avvrovovf
KiVT/creir h> roif irpoKe7i.Eva/j.aTiKoif 6$ ra iroTJ^a pv6fj.ol^. (Dion. Hal.,
Ant. Rom., 1. 7, p. 476.) For some learned observations on the proce-
leusmaticus, consult Santen. ad Terent. Maur., v. 1460.
3. Gesner thinks that the Ionic mode of dancing was adapted to these
two feet, whence their name. (Thesaur. L. L., s. .) Plotius, on the
contrary, refuarks, " lonici ab lone inventore suo dicti" (p. 2626).
136
FEI;T.
or a minore, because it begins with the less quantity, that
is, with two short syllables. 1
VII. The CHORIAMBUS consists of a choraeus or trochae-
us and an iambus, that is, of two short syllables between
two long ; as, nobititas. 2
VIII. The ANTISPAST (Antispastus) consists of an iam-
bus and a trochee, that is, of two long syllables between
two short ; as, secunddre. It derives its name from dvrt-
andtt), " to draw asunder" two long syllables being separa-
ted or drawn asunder by two short ones. 3
IX. The EPITRITUS PRIMUS, or First Epitrit, is compo-
sed of an iambus and spondee, and consists of one short
syllable and three long ; as, salutantes.
X. The EPITRITUS SECUNDUS, or Second Epitrit, is
composed of a trochee and a spondee, and consists of a
long, a short, and then two long syllables ; as, cbncitatl.
XI. The EPITRITUS TERTIUS, or Third Epitrit, is com-
posed of a spondee and an iambus, and consists of two
long syllables, followed by a short and a long ; as, commu-
nicant.
XII. The EPITRITUS QUARTUS, or Fourth Epitrit, is
composed of a spondee and a trochee, and consists of three
long syllables and one short ; as, incantare.
1. The Greek names for these two feet are 6 UKO fieiCovoc. J lu
and 6 an' VXaaaovoc, 'luviKOf. The greater Ionic was also called Per-
sicus, because Persian histories were written in this measure. Hepai-
df <5e, 6tu TO raf iaropiac; ruf TlepaiKuf TOVTU TV fitrpy yc
(Schol., Hephasf., p. 82.) The minor Ionic was likewise styled Hypo-
cyclius ("fnoKVK?.iof. Schol., Hcphast., I. c.).
2. The choriambus was also called Cyclius (Kti/cAtof), and Hypobac-
chius ('YTrofiuKxeiof), and likewise Bacchus a Trochao (Ba/c^etof KOTO,
Tpoxalov). It was probably used in the circular dances around the altar
of Bacchus, whence the names just mentioned, which are given by the
scholiast on Hephaestion (p. 83). Compare Santen. ad Tcrent. M^nur.,
v. 1487.
3. The antispast was also calleu* Spondcfocus (27rov<5am/cof) and Ba/c-
Xeioc, KOTU 'iafi6ov. (Schol., Hcpluest., p. 83.) This foot appears to
have been a favourite with the prose writers. Vossius states (fnst.,
Oral., 4, 4) that he counted in Cicero's oration for Archias twenty-six
instances of an ai.t.spast at fur i-r-d of a rianso. 'Compare Profits, p.
14'92, and Dimni..!:*. \< "n~i
FEET. 1 37
These four last-mentioned feet derive their name from
the word i~i~pi~og, which denotes a combination or pro-
portion containing some particular number, together with
a third part thereof. Thus, in the present instance we
have three long syllables, which may be regarded in one
sense as the basis of the foot, and besides these we have
another syllable, a short one, from the relative position of
which last in each of the four feet are derived the names
of first, second, third, and fourth epitrits. In other words,
we have three syllables, with a third of these, namely, one
syllable, added thereto (rpfT^-eTri). 1
XIII. The P^:ON PRIMUS, or First Paeon, is composed
of a trochee and a pyrrhich, and consists of one long syl-
lable followed by three short ; as, confuere.
XIV. The P.EON SECUNDUS, or Second P<eon, is compo-
sed of an iambus and a pyrrhich, and consists of a short
and a long, and then two short syllables ; as, resolvere.
XV. The PJEON TERTICS, or Third P<son, is composed
of a pyrrhich and a trochee, and consists of two short syl-
lables followed by a long and a short ; as, soctdre.
XVI. The P.EON QUARTUS, or Fourth Peon, is compo-
sed of a pyrrhich and an iambus, and consists of three short
syllables and one long ; as, celeritas.
The P&on, according to some authorities, 2 derived its
name from Paeon, an ancient poet, its alleged inventor.
According to other and better accounts, however, this foot
was so denominated from its having been particularly used
in the hymns to Apollo. 3 Hence also, besides the appel-
1. Cicero renders e^irpiTOf, which Plato uses in his Timaeus (p. 36,
a., ed. Stcph.), by the Latin sesquitertius (Tim., c. 7), a term likewise
employed by Augustinus (de Mus., 2, p. 76). Aulus Gellius (18, 14)
defines an epitritus as follows : " Epitnius est qui habet totitm aliqucm
numerum, et ejusdcm partem lerliam." (Compare Macrob., in Sotnn.
Scip., 2, 2.) The scholiast on Hephaestion has also the following:
'Em'rptrof upidfidf Aeyerai, 6 upi6/nov riva l^uv kv iavru, nai TO rpirov
TOVTOV. *
2. Plotius, p. 2626. Isidor., Orig., 1, 16, p. 830.
3. OI fif 'AiroAXwva <5e ypa^ovre f , rai nodi rotJ-ry, rci Tlaiuvi,
M 2
138 FEET.
lation of Paeon, it received that of Paan. 1 The paeon is
directly opposed to the epitrit. In the latter there is one
short syllable with three long ones ; in the former, one long
with three short. Thejirst, second, third, and fourth paeons
are so named from *.he relative situation of the long sylla-
ble in each.
To the list of co.npoxind feet are sometimes added the
following :
I. The DOCHMIUS (4o%/uof), composed of an antispast
and a long syllable, and consisting of five syllables, namely,
a short, two long, a short, and a long ; as, aberraverant. 2
II. The MESOMACER (M.eo6[i,aKpog), composed of a pyr-
rhich and a dactyl, and containing also five syllables,
namely, two short, a long, and two short ; as, avldisstmus*
The following table exhibits the feet, both simple and com-
pound, in the order in which they have just been described.
SIMPLE FEET OF TWO SYLLABLES.
t 1. SPONDEE
\ 2. PYRRHICH ~ ~
( 3. TROCHEE ~
f 4. IAMBUS ~
rat. (Schol., Hephtest., p. 12.) Compare the remark of another scho-
liast : Toi>f iraiavcLf pdovref kxpuvro avroif [Ilatucri] ol ird^aioL.
(Schol. , Hermog., p. 394.)
1. Arislot., Rhet., 3, 8. Cic., de Oral., 1, 59. Santen. ad Terent.
Hour., v. 1532.
2. The Dochmius takes its name from <56;tyutof, " obliquus" in allu-
sion to the irregularity that marks the order or succession of its compo-
nent times ; or, in the words of an ancient writer, 6iu TO TtoiKihov /cat
avouoiov, K<U TO pi) KOT' evdv tieupeladai TT/V frvdftOTroitav. (Aristid., de
Mus., 1, p. 39.) This foot is sometimes called Dochimus, but incor-
rectly. (Erncsti ad Cic., Oral., 64. Santen. ad Terent. Maur., v.
1551, p. 130, seq.)
3. From /ieerof, "mddle," and //a/cpof, "long," the name being de-
rived from the position of the long syllable, in the middle, between two
hor* on each side.
FEET. 139
SIMPLE FEET OF THREE SYLLABLES.
1. MOLOSSUS .......
2. TRIBRACH
3. DACTYL ,
4. ANAPAEST ........
5. BACCHIUS ........
6. ANTIBACCHIUS . . . . .
7. AMPHIMACER
8. AMPHIBRACH
COMPOUND FEET.
FOUR UlTH THE SAME FOOT DOUBLED.
1. DlSPOND^US
2. PROCELEUSMATIC ;s
3. DlTROCH-ffiUS
4. DlIAMBUS .
FOUR OF CONTRARY FEET.
1. GREATER IONIC ....
2. MINOR IONIC . . . . .
3. CHORIAMBUS
4. ANTISPAST
FOUR tN WHICH LONG TIMES EXCEED.
1. FIRST EPITRIT
2. SECOND EPITRIT
3. THIRD EPITRIT . . . .
FOURTH EPITRIT
FOUR IN WHICH SHORT TIMES EXCEED.
'l. FIRST P^EON
2. SECOND PJEON
3. THIRD P^EON . . . . .
4. FOURTH
TWO OTHER COMPOUND FEET, OF FIVE SYLLABLES.
1 DOCHMIUS . . . . . . . ~ w
2 MESOMACEE -^ ~, - _ ^
140 METRE.
SECTION XXVIII.
OF METRE.
I. METRE, in the general acceptation of the word, signi-
fies a combination of verses belonging to the same or to
different classes, which succeed each other in a fixed or-
der. When we speak of Dactylic metre, Iambic metre, &.C.,
the word metre is employed in this general sense, and is
then synonymous with verse.
II. But a metre, in the technical and restricted sense,
signifies either a single foot in a verse, or a combination of
two consecutive feet, according to circumstances.
HI. In Latin poetry, the metres employed are the follow-
ing:
1. DACTYLIC.
2. ANAPJESTIC.
3. IAMBIC.
4. TROCHAIC.
5. CHORIAMBIC.
6. IONIC.
IV. These metres have received their respective names
from the frequent recurrence in them of some particular
foot ; and it has been supposed that each species was ori-
ginally made up of those feet only from which it is denom-
inated, but that others, equal in time, were afterward ad-
mitted under certain restrictions.
V. Metres, however, are often called also after the name
of some celebrated poet, who used a particular species of
verse ; as, Sapphic, Anacreontic, Hipponactic, &c.
VI. Metres are likewise classified according to the num-
ber of feet or measures which they contain. Thus, a verse
of seven metres is called Heptameter.
" six " " Hexameter,
"five " " Pentameter.
METRE. 141
of four metres is called Tetrameter.
" three " Trimeter.
" two " " Dimeter.
" one " " Manometer.
VII. In Anapaestic, Iambic, and Trochaic verses, a mefre
consists of ;0 feet ; but in the other kinds of verse a sm-
gle foot constitutes a metre.
VIII. Thus, in Anapaestic, Iambic, and Trochaic versifica-
tion,
a Manometer contains two feet.
Dimeter " four "
Trimeter " six "
Tetrameter " eight "
Pentameter " ten "
Hexameter " twelve "
Heptameter 1 " fourteen "
On the other hand, in Dactylic, Choriambic, and a# oZAer
measures except the three just mentioned,
a Manometer contains one foot.
Dimeter " two feet.
Trimeter " three "
Tetrameter " ybwr "
Pentameter " _/ive "
Hexameter " *za; "
Heptameter " seven "
IX. A combination of two consecutive feet is sometimes
termed a dipodia (617706 ia) or syzygy (pv^vyia). Most
usually, however, the combination of two dissyllabic feet
is called a dipodia, and that of two trisyllabic, or a dissyl-
labic and trisyllabic, a syzygy.
X. The conjunction of two feet is often likewise termed
c base?
1. Some of these, of course, do not occur. They are given, however,
to illustrate the rule laid down.
2. Diorned.,?. 501. Mar. Viet., p. 2489.
142 ISOCHRONOUS FEET.
SECTION XXIX.
OF ISOCHRONOUS FEET.
I. By ISOCHRONOUS (or even-timed 1 } feet are meant those
that are interchangeable in metre.
II. In order to ascertain what feet are thus interchange-
able, recourse must be had to the arsis and thesis.
HI. As has already been observed, 2 that part of the foot
which receives the ictus, the stress of the voice, or beat of
the time, is called arsis, or " elevation." The rest of the
foot is termed thesis, or " depression."
IV. The natural place of the arsis is the long syllable of
the foot, and hence in the iambus it falls on the second syl-
lable ; in the trochee, on the first ; while the spondee and
tribrach leave its place alike uncertain.
V. The fundamental foot of a verse, however, deter-
mines the arsis for the other feet, and hence the spondee,
in iambic and anapaestic verse, has the arsis on the second
syllable, but in trochaic and dactylic on the first.
VI. So, again, the tribrach, when it stands for the iam-
bus, is to be pronounced ~ ~ ~ ; but when it stands for
the trochee, ^ ~ ~.
VII. Now the ancients considered those feet only iso-
chronous which were capable of being divided into parts
that were equal in time ; so that a long syllable should
have either a corresponding long syllable or two short
ones. 3
VIII. The following scheme will exemplify this more
clearly, the place of the arsis being denoted, as above (VI.),
by a mark resembling that of the acute accent.
( Iambus .
) Tribrach
Trochee
Tribrach
1. 'IffOf, "equal," and ^povof, "time."
2. Page 106, note 1.
3 Daiocs, Misc. Crit., p. 62, scqq., ed. Kidd,
ISOCHRONOUS FEET. 143
V Dactyl . . I ~ ^ ( Anapaest
/ Spondee . | ( Spondee
IX. By this we perceive that the iambus and the trochee
are each interchangeable with the tribrach, and that the
dactyl, spondee, and anapaest are interchangeable with each
other.
X. In like manner it will appear that the iambus and the
trochee are not interchangeable, and that an iambus never
admits a trochee into iambic verse, nor a trochee an iambun
into trochaic verse. Thus,
Iambus
Trochee ....
Here it will be perceived that the long syllable of the iam-
bus has neither a correspondent long syllable in the tro-
chee, nor two short ones in the same. And the case is
precisely similar with the long syllable of the trochee,
which has nothing to correspond to it in the iambus.
Hence the two feet are not interchangeable or isochronous. 1
XI. Again, it may be shown in the same way that the
spondee and amphibrach are also not interchangeable.
Thus,
Spondee .
Amphibrach . . ^ -
N
Here, in whatsoever way the amphibrach be divided, each
division contains either more or less than the correspond-
ent part, of the spondee. Hence the two feet are not iso-
chronous ; and, for the same reason, the amphibrach is not
isochronous with the dactyl or anapaest.
1. For this reason the scholiast on Hephaestion calls the iambic and
trochaic measures avrnradovvra fterpa, and Tricha remarks, avrnrafiris
tiol olov VTTEvavTiOf T& id/udu 6 rpoxaiof. (Schol. ad Hcphatt., p. 76.
Trich., de Melr., p. 9, ed. Herm. Compare Dawey, Misc. Grit., p.
103, ed. Kidd.)
144 VERSE.
SECTION XXIX.
OF VERSE.
I. A Verse is a certain number of feet disposed in a reg-
ular order, and forming a line of poetry.
II. A Hemistich' is, properly speaking, a half verse ; the
name, however, is commonly applied to either portion of a
hexameter line divided at the pcnthemimeris ; as,
^Ere ciere viros, || Marlemque accendere cantu.
III. Scanning is the dividing of a verse into the feet of
which it is composed, and the assigning of their proper
quantity to the respective syllables of each foot.
IV. Verses are denominated Acatalectic, Catalectic,
Brachycatalectic, ffypercatalectic (or Hypermetcr), and
Acephalous.
V. An Acatalectic verse is one that is complete in all its
parts, and comes to a full termination ; as the following,
which is Iambic Dimeter Acatalectic. 2
Musas | Joins || suntfll\tae. ||
VI. A Catalectic verse is one that wants a syllable at the
end to complete the measure ; as the following, which is
Iambic Dimeter Catalectic. 3
Muses | Jovem |1 cane\bant. ||
VII. A Brachycatalectic verse is that which wants two
syllables at the end to complete the measure ; as the fol-
lowing, which is Iambic Dimeter Brachycatalectic.*
Musee | Jovis I! gnatae \ -- ||
1. From fjfj.1, " half," and crt^of, " a verse."
2- Acatalectic, from uKarahijicTiitoc; (a and Karakfiyu, to stop), i. e.,
not having an abrupt termination. Compare Hephestion (c. 4, p. 24,
cd. Gaisf.): 'Aicarafo/jcra *a?.rat fie~pa baa rbv TE?.EV-CUOV iroda oAo-
3. Catalectic^ from KaraA^Jtruroc, whiqfr is itself a derivative from
KarciMiyu, denotes verses that stop before they reach their full ending.
Compare the language of Hcphaestion (/. c.) : KoraAjjicrwca <Je 6aa fu-
4. ~BpaxvKara'/i]KTa 6e Ka&eiTai, aaa axb diirodias
-at. ( Hepkaut. , I. c.)
DACTYLIC MEASURES. 145
VIII. A Hypcrcatalectic or Hypermeter verse is that which
has a syllable at the end beyond the complete measure ; as
the following, which is Iambic Dimeter Hypercatalectic or
Hypermeter. *
Musae | soror\\es sunt \ Minervlla:.
IX. An Acephalous verse is that which wants a syllable
at the beginning ; as the following, which is Acephalous
Iambic Dimeter. 2
Non | ebur || neque ~aur\eum. H
SECTION XXX.
DACTYLIC MEASURES.
I. The only feet admissible in dactylic verses are the
dactyl and spondee..
II. Originally all the feet were dactyls, but the spondee
was afterward admitted, as being an isochronous foot, and
in order to impart more gravity and dignity to the line.
III. It will be remembered that in dactylic verses a sin-
gle foot constitutes a metre ; and hence a dactylic hexam-
eter contains six feet, a dactylic pentameter jive, a dactylic
tetrameter four, &c.
IV. In treating of the dactylic, as of other measures, we
will commence with -the simplest in use, and end with those
of the greatest length.
1. DACTYLIC DIMETER ACATALECTIC, or ADONIC.
I. This measure consists of two feet, a dactyl and spon-
dee ; as,
Terruit \ urbem.
Visere- \ mantes.
II. The dactylic dimeter, or Adonic, is usually subjoined
1. "fTrepKaTa^.TjKTa 6e oaa Trpdf TU T&KM Trpoaekade
(Hcphast., I. c.) Some call it Hypcracatalcctic
e., going beyond acatalecfir..
9. Acephalous, from d/t'(?a/lof, " without a head" (u and
N
146 DACTYLIC MEASURES,
to three Sapphic lines, thus forming what is called the Sap-
phic stanza, of which an account will be given elsewhere.
III. In tragic choruses, the Adonic is arbitrarily added
to any number of Sapphics, without regard to uniformity. 1
IV. This measure was called Adonic (Adonium metrum
'Adomov fterpov) because employed originally by the
Greeks in the lamentations for Adonis. 2
V. Sapplio is said to have composed entire poems in this
measure, now lost. Terentianus Maurus, from whom we
derive this information, has also left us a short piece of this
kind, 3 and another of thirty-one successive Adonics occurs
in Boethius. 4
2. DACTYLIC TRIMETER CATALECTIC.
I. This measure, also called Archilochian Penthemime-
ris, is an heroic penthemimeris, or the first five half-feet of
a dactylic hexameter. In other words, it consists of two
feet and a half.
II. Horace, who uses this verse in one of his odes (4, 7)
alternately with the dactylic hexameter, uniformly has two
dactyls and a semifoot ; as,
Horat. Arbon\busque com\ee.
Id. Flumind \ pra:tere\unt.
Id. Ducere \ nuda chor\os.
Id. Quiz rapit \ hora dt\em, &c.
III. Ausonius, however, who has a poem of fifty-seven
lines, 5 all in this measure, sometimes makes the first foot a
spondee, and in two instances uses a spondee also in the
1. Scnec., (Ed., 110, seqq. Id., Troad., 1010, seqq. Id., Thyest.,
646, seqq., &c.
2. Mar. Viet , p. 2564. Plotius, p. 2640.
3. Terent. MJLUT., v. 2160. Some commentators think that the words
? Q rbv "\6uvtv, cited by Plotius (p. 26-10), belonged to the Adonic poem
of Sappho mentioned by Terentiauus. (Santcn. ad Ter. Maur., I. c.)
4. De Cons. Phil., 1, raetr. 7.
6. Auaon., Prof., 10.
DACTYLIC MEASURES. 147
second place. But the spondee, in either case, is a dispar-
agement to the verse, particularly in the latter.
Auson. Doctrl\na exigu\us.
Id. et ti\berti\na.
Id. et tu | Concor\di.
3. DACTYLIC TRIMETER ACATALECTIC.
I. The Dactylic Trimeter Acatalectic consists of the last
three feet of a dactylic hexameter ; that is, the first foot
may be either a dactyl or spondee, but the second must be
a dactyl, and the third a spondee ; as,
Horat. Grato \ Pyrrha sub \ dntro.
II. But the lines which are usually thus denominated
are with greater propriety included in the class of choriam-
bics, and ranked with Pherecratics.
4. DACTYLIC TRIMETER WITH A BASE.
I. This measure consists of two dactyls, followed by a
spondee, with a base (that is, one long, or two short sylla-
bles) prefixed to the beginning of the line. Ausonius will
afford an example. 1 Thus,
Modu\lammc \ rmma \ tristt,
Tacit\um sine hon\ore re\linquat,
Super | indole \ cujus ad \ ulli,
Mag\n<K bona \ copia \ laudis.
5. DACTYLIC TRIMETER HYPERCATALECTIC.
I. This measure consists of three dactyls, and a syllable
over. The following example of its use is taken from Au-
sonius. 2
Pdrva ei\amfmt \ Idatt\a,
Nomine \ pradita \ qua Paphi\ji,
1. Parent., 17, 1.
2. Ib., 28.
148 DACTYLIC MEASURES.
Et speci\em meru\it Vener\is,
Qua: genit\a est mthi \ pane so\ror ,
Fitia | namfuit \ lieec ami/!,
Quam cele\brdt sub hon\ore pi\o,
Noenia \ carmine \funerc\o.
II A variety of this measure is found in Boethius (3,
mett 6), which admits of a spondee" in the first two places.
Thus,
Hlc claus\lt membr\ts anim\os.
Onus e\riim re\rum pater \ est.
Ni viti\ts pej\orafov\ens.
III. The Dactylic Trimeter Hypercatalectic is called by
some Tetrameter Catalectic.
6. DACTYLIC TETRAMETER MEHJRUS, or FALISCAN.'
I. This measure consists of the last four feet of an ordi-
nary hexameter, except that the concluding foot is an iam-
bus instead of a spondee. Thus,
Sept. Ser. Vitis et \ ulmus ut\l simul j eant.
Id. Liberal \ arva pri\us fruti\cibus.
Boeth. Fdlce rub\os fili\cemque re\secdt.
II. The dactyl was preferred in the first three places,
though the spondee was nevertheless admissible into the
first and second.
1. The name Meiurus is from the Greek : UTTO rov pfiovadai TIJV
oiipdv, quod vtluti cauda minor et mutilus sit." (Forcelt., Lex.) Dio-
medes calls such verses " Ecaudes" (3, p. 499), and Marios Victorious
styles them Ttliambi (p. 2512). The origin of the name Faliscan has
been much disputed. The form Phaliscus, given in some editions of
Terentianus Maurus, as well as Paliscus (Avgustin.. dc Mus., 4, p. 83),
are erroneous. It is probable that the appellation Faliscan was given
to this species of verse from the poet Annianus, who frequently employ-
ed it, and wha was of the nation of the Falisci. (Santen. ail Tcrent.
Afflur, 1998.) Victorinus states that this measure was called by the
Greeks Calabrion (p. 2f>78). Servius, the grammarian, stylos Fnliscan
poems " docta," the reason for which appellation is unknown. (Scrv.,
in Centimetr., p. 1824.)
DACTYLIC MEASURES. 149
7. DACTYLIC TETRAMETER A POSTERIORE, or SPONDAIC
TETRAMETER.
I. This measure consists of the last four feet of a dac-
tylic hexameter, and hence it is called Tetrameter a poste-
riore, because supposed to be taken from the latter part of
an hexameter line (a posteriore parte versus hexametri).
II. In other words, the first two feet of this measure may
be either dactyls or spondees, the third is generally a dac-
tyl, rarely a spondee, but the fourth is invariably a spondee.
Thus,
Horat. Ibirnus, \ o soci\l cormt\esque.
Certus e\nim prd\mlsit Ap\dllo.
Me nee \ tarn pad\ens Ldce\dcemon.
O Jort\es pej\ordque \ passi.
III. In the following line a spondee is found in the third
foot, preceded by a dactyl, answering to the spondaic line
of the hexameter ; and in such a case as this, the prece-
ding foot ought always to be a dactyl, or the line will be
too heavy. Thus,
Horat. Mensor\em cohi\bent Arch\ytd.
8. DACTYLIC TETRAMETER A PRIORE, or ALCMANIAN.*
I. This measure consists of the first four feet of a dac-
tylic hexameter, with merely this difference, that the fourth
or last foot is always a dactyl. And it is called a priore
because taken from theirs* part of an hexameter (a priore
parte versus hexametri). Thus,
Boe'th. Lummt\busque pn\or redi\it vigor.
Id. Desuper | in terr\dm nox \ fund/itur.
Auson. Dice\bas in | me ma\tertera.
II. The Dactylic Tetrameter a priore is also called the
Alcmanian Tetrameter, from the Greek poet Alcman, by
whom it was frequently employed. 1
III. This metre was much used in tragic choruses.
1. Serv., p. 1820, ed. Putsch.
N 2
150 DACTYLIC MEASURES.
9. PHAL.ECIAN DACTYLIC PENTAMETER.
I. This measure, which is called Phalaecian on the au-
thority of Terentianus, 1 consists of a dactylic penthemime-
ris (two feet and a half), followed by a dactylic dimeter or
Adonic ; as,
Boe'th. Vlse\bdt getidfe || slderd \ brurricK.
Id. Heu qudm \ prascipit\l [\ jnersd pro\fundo.
II. Like the ^Eolic Pentameter, this species of Phalae-
lian verse admits a trochee in the first place ; as,
Senec. Arvd \ mutdn\tes || quasque Mai\dtis.
Id. Allu\it gent\es \\frlgldd \fluctu.-
Id. Quasque \ despect\dt || verfice \ summo.
III. Besides the trochee, Boethius uses the iambus in the
first and second places ; as,
Htc e\mm ca|sa,s || cernere \ promptum est.
HKc | ldten\tes || pectord \ turbant.
Cunctd | qua rar\a || provehlt \ atas.
Stupet | cum subi\tls 11 mobile \ vulgus.
IV. A Phalaecian Dactylic Pentameter may be formed
from an hexameter, by striking out the fourth foot and the
latter half of the third. Thus,
Hex. At re\glna grdv\ijdm [ dudum \ saucid \ curd.
" Consed\ere du\ces et \ vulgl \ stdnte cor\ona.
" Sdnguine\dque mdn\u crepi\tdntid \ concutit \ drma.
Phalaec. At re\gind grdv\l \\ saucid \ curd. \
" Consed\ere du\ces || stdnte cor\ond. \
" Sdnguine\dque mdn\u || concutit \ drma.
1. " Namque hie de genere est Phalccciorum." (Ter. Maur., r.
1946.) The name is derived from that of the poet Phalaecus, who wab
said to have been the inventor of this species of verse, along with the
others that went under this general appellation. The true form is Pha-
kecian, not Phaleucian. (Santen. ad '"cr. Maur., v. 2545.)
DACTYLIC MEASURES.
151
V. Those Phalaecian Ifties in which neither the trochee
nor iambus occur, may be scanned as Choriambic Tetram-
eters Catalectic. Thus,
Vtse\bat geltdce \ sidera brumfe
Jam nunc \ bldnda melds \ carpe Dto\ne.
10. ^EOLIC PENTAMETER.
I. The JEolic Pentameter, so called from its native dia-
lect, was invented, it is said, by Sappho, a native of the
^Eolic island of Lesbos. 1
II. It consists of four dactyls, preceded by a spondee, a
trochee, or an iambus. In other words, it is a Dactylic
Tetrameter Acatalectic with a base. 2 Thus,
Terent. Maur. Cordt \ quando fu\tsse sl\bi cantt \ dtthida
Id. edt\dit tuba \ terribi\lem soriit \um procul.
11. DACTYLIC PENTAMETER.
I. The Dactylic Pentameter was so called in consequence
of the manner in which it was scanned by some of the old
grammarians, who viewed it as consisting of five continu-
ous feet, namely, two dactyls or spondees, followed by a
spondee and two anapaests, according to the following scale :
1
2
3
4
5
^ ^
^ *^
^x < _
wx ^
Frigidi\us glaa\e pect\us amantfis erdt.
Nil mtht | rescrlb\ds dt\tamen tps\e verii.
Lassa\ret vidu\ds pend\ula tel\a manus.
Flebdm \ success\u pos\se cdre\re dolos.
1. Terent. Maur., v. 2148. Compare Mar. Viet., p. 2559. Plot.,
p. 2636. Scrv., p. 1824. The 29th Ictyl of Theocritus is in this
measure: Oivoc, | w <pft \ ircii, 7ieyl\Tai not u\^.u8tu, \ K.T.%. Foroth
er examples, consult Gaisford ad Hephast., p. 275.
2. Terent. Maur., v. 1722. Diomed., p. 502. Qumtil., 9, 4.
152
DACTYLIC MEASURES.
II. Hephsestion, however, who has been followed by al-
most all modern scholars, regards it as composed of two
dactylic penihemimers. In other words, the first two feet
may be either dactyls or spondees ; then comes a long syl-
lable, to which succeed two dactyls, followed by another
long syllable. 1 Thus,
1
2
3 n 4
5
6
V -^
_ -^
-II
^ -,
ldci\e \\ pectus dm\dniis er\dt.
Nil mihi \ rescrib\ds fl altamen \ ipse venjt.
Ldssd\ret vidu\ds |j penduld \ tela mdn\us
Flebdm | success\u \\ posse cdr\ere dol\os.
III. That this last is the proper view to take of the struc-
ture of the pentameter seems certain from the fact, that a
division of the verse takes place invariably at the end of
the fifth half-foot, as well in the Greek as in the Latin wri-
ters.
IV. Ovid is considered the model of this species of verse
among the Romans, and the wonderful smoothness and
melody of his compositions are the result of close atten-
tion to a number of minute observances, which were alto-
gether neglected by the Greeks, and by their imitators, Ca-
tullus, Tibullus, and Propertius. 2 These rules will now
be given.
Rules for the Structure of Dactylic Pentameters.
I. The pentameter must always be so constructed as to
have the caesural pause after the penthemimeris, and thus
1. Tot; <5c daicrvXiKOv irevdqfitpepovf 6tf ?.a[i6avoph>ov ytverai TO
fXf}'tov, AC. T. A.. (Hcphtzst., p. 92, cd. Gaisf.) Compare the language
of the scholiast (p. 186): BcArtov rfe OVTU fterpelv hrel xai ctf dvo diy-
prtrai rrevdjjuiuepij, K. T. X.
2. Ramsay's Lat. Pros., p. 184. Tote, Remarks on Dactylic Pen-
tameters, in the Classical Journal.
DACTYLIC MEASURES. 153
oe divisible into two equal portions of two feet and a half
each. And the pause ought always to be strictly the last
syllable in a word, and not fall on any syllable before this. 1
Hence the following line is faulty :
Inter nostros gentilis oberrat equus.
2. It is also deemed a blemish for the caesura to fall af-
ter a syllable which has become the last by elision ; as in
the following lines from Catullus :
Quam veniens una atque || altera rursus hyems,
Troia virum et virtutum || omnium acerba cinis.
Nee desistere amare || omnia si facias.
3. If the first caesural syllable be a monosyllable, which
ought not to happen frequently, it must be preceded by a
long monosyllable, or by a word of the same time, i. e., a
word consisting of two short syllables ; as,
Ovid. Et mihi si non vis || par cere, parce meis.
Id. Tu dominuS) tu vir, || tu mihi f rater eras.
Id. Nulla tibi sine me || gaudiafacta neges.
Id. ProRierito magis est || iste pudendus amor.
An exception to this rule, however, is made when the
monosyllable is est, and the word before it suffers elision.
Such lines as the following are not uncommon :
Ovid. Litteraqite invisa est, || hac mea parte tibi.
Id. Quo nisi consilio est || usa puella tuo.
But such as the following are very rare in Ovid :
Sed sic inter nos || tit latuisse velint.
Quod licet inter vos || nomen habete meum.
Justaque quamvis est, || sit minor ira dei.
Quaere suburbana hie II sit. mihi terra locum.
4. The last word of a dactylic pentameter is, in the
great majority of instances, a dissyllable in Ovid. We oc-
1. There is no exception to this, even in Greek, except in a proper
name ; as,
'leou vvv 6f biooKOvpitieu yeverj. (Callim., frag, cxcii.)
154 DACTYLIC MEASURES.
casionally find est in this place, preceded by a dissyllable
which suffers elision ; as,
Hie est cujus amans hospita capta dolo est.
Nee repetor ; cessas, iraque lenta tua est.
More rarely two monosyllables ; as,
Pr&mia si studio consequor ista, sat est.
But such a line as the following must be considered al-
together unworthy of imitation ;
Omnis an in magnos culpa deos scelus est. 1
. 5. The trisyllabic ending, although very common in the
Greek poets, in Catullus, &c., may be said to be altogeth-
er excluded from the Ovidian pentameter. We find one
example only in his earlier works, and five others in the
Epistles from Pontus, which, together with the Tristia,
were composed while the poet was plunged in the deepest
despondency, and bear tokens of less accurate revision
than his other productions.
Qua tamen externis dandaforent generis. 2
Ne non peccarim, mors quoque non faciet. 8
Quolibet ut saltern rurefrui liceat. 4
Aut quod saspe soles, exigis ut recitent. 5
Vix excusari posse miki videor. 6
Spectarem, qualis purpura te tegeret. 7
6. The quadrisyllable ending is likewise very uncommon,
except in the Tristia and Epistles from Pontus. We have,
however, two or three examples in his other works.
Undo simul miserum vitaque deseruit. 8 ,
Et circumfusis invia fluminibus. 9
Cantabat m&stis tibia funeribus. 19
1. Odd, Ep. ex Pont., 1, 6, 26. 2. Ovid, Heroid., 14, 62.
3. Id., Ep. ex Pont., 1, 1, 66. 4. Id. Ep. ex Pont., 1,8,40.
5. Id. ib., 3, 5, 40. 6. Id. ib., 3, 6, 46.
7. Id. ib., 4, 9, 26. 8. Heroid., 19, 202.
9. Fasti, 5, 582.
10. 76., 6. 660 Other examples are f ; ven in Ramsay, p. 184.
DACTYLIC MEASURES. 155
7. The quiuquesyllabic ending is still more rare than the
quadrisyllable.
Lis est cum forma magna pudicitise. 1
Nee sedeo duris torva superciliis. 2
8. As to the kind of words that conclude the line, they
ought to possess some emphasis. They are usually nouns,
the personal and possessive pronouns, or verbs. Adjec-
tives do not often occur in this place, adverbs still more
rarely, and less frequently than either, the present participle
active.
9. Elisions should be resorted to sparingly, especially in
the second half of the verse, where they are by no means
harmonious. They may be allowed in the first of the two
dactyls ; as,
Ultimus est aliqua decipere arte labor.
Incipis, incipiet desinere esse mea.
But when they fall on the second dactyl, the melody of
the line is destroyed ; as,
Quis scit an haec sasvas tigridas insula habet ? 3
10. At the beginning of the verse it is better to have a
dactyl followed by a spondee than the reverse. Thus,
Vix Priamus tanti totaque Trojafuit.
Res est solKciti plena timoris amor.
V. Dactylic pentameters are never found in a system by
themselves, in the classic writers (unless seven lines in
Ausonius can be taken as an exception), but always in com-
bination with hexameters.
VI. Hexameters and pentameters, placed alternately,
constitute what is termed the Elegiac Distich. 4
1. Heroid., 16,288.
2. 16., 17, 16. For other instances, consult Ramsay, p. 184.
3. 16., 10, 86.
4. The Greeks employed this combination of the two measures in
war-songs, hymns, and epigrams or inscriptions ; the Romans in epi-
grams, epistles, and amv',ory poetry.
156 DACTYLIC MEASURES.
VII. The name of Elegy (TEAeyof) was first applied to
the alternating hexameter and pentameter in the time of the
Greek poet Simonidea ; whether it was that he himself in-
troduced the name, or whether the mournful and plaintive
nature of his themes justified this appellation from others. 1
VIII. Ovid, in some of his elegiac pieces, employs oc-
casionally a species of metrical echo, if we may so term it,
the second half of the pentameter being a repetition of the
commencement of the preceding hexameter, either pre-
cisely the same words being used, or else a slight change
being made in them. Thus,
Militat omnis amans, et habet sua castra Cupido :
Attice crede mthi, militat omnis amans. 2 '
Graia juvenca venit, qu<B te patriarnque domumque
Perdat. lo prohibe ! Graia juvenca venit?
Invida vestis eras, qua tarn bona crura tegebas ;
Quoque magis species, invida vestis eras. 4
Ars tua, Tiphy,jacet, si non sit in aequore fluctuf
Si valeant homines, ars tua Phoebe jacet. s
Auspiciis animisque patris, puer, arma movebis,
Et vinces animis auspiciisque patris. 6
IX. Martial also, in one of his epigrams (9, 98), has six
consecutive distichs, each commencing and ending with the
words rumpitur invidia?
1. This combination of hexameters and pentameters is said to have
been invented by Callinus, and applied by him to martial themes. It
was not called lAcyof at first, but I of, the latter of which terms was
afterward confined to heroic verse, when Simonides brought in the name
Iteyof, and along with it the handling of plaintive themes in this species
of measure.
2. Am., 1. 9, 1 3. Her., 5, 117.
4 Am, 3, 2, 27. ' 5 Tnst., 4, 3, 77.
6. A. A, 1, 191.
7 This species of play upon words gave rise, in later ages, when
lasti- had become thoroughly corrupted, to entire poems. Verses of thi
DACTYLIC MEASURES. 157
Rumpitur invidia quidam, carissime Juli,
Quod me Roma legit, rumpilur invidia.
Rumpitur invidia, quod turba semper in omni,
Monstramur digito, rumpitur invidia, &c.
12. DACTYLIC HEXAMETER.
I. The Dactylic or Heroic* Hexameter was considered
to be the most ancient as well as the most dignified species
of verse, and was said to have been invented by Phemonoe,
the first priestess of the Delphian Apollo, who, when in-
spired by the god, was wont to chant his oracles in this
measure. 1
II. The origin of dactylic versification, however, is to be
traced back to the earliest periods of the Greek language,
and connects itself with the peculiar mode of intonation
that characterized the JEolic tribes. 2
kind were called " Versus Lyrici, Ophites, and Serpcnlini." Some of
these pieces may be found collected in the Miscell. Obs. Nov. (vol. 5, p.
475, scqq.). Wernsdorff gives in his collection (Poet. Lat. Min., vol.
3, p. 268) a poem of this kind by Pentadius, " De Adventu Veris," the
commencement of which is as follows :
Sentio fugit hyems, Zephyrisque moventibus orbem
Jam tepet Eurus aquis ; sentio fugit hyems^
Parturit omnis ager, prasentit terra calorem,
Germinibusque novis parturit omnis ager.
Lata vireta tument, foliis sese induit arbor,
Vallibus apricis Iceta vireta tument, &c.
1. Schott. ad Prod., p. 18. Voss., Inst. Poll., 3, 3, 2. Fdbnc.,
Bibl. Gr., vol. 1, p. 154. Plin., H. N., 7, 5G.Pausan., 10, 5.
Schol. ad Eurip., Orest., 1093.
2. Goltling's Greek Accentuation, 2, seq. The tendency in the
./Eolic pronunciation was to throw the accent back on the root, or as
near to it as possible, so that the dactylic rhythm, with the arsis on the
first syllable of the foot, could easily and naturally arise from this kind
of pronunciation. As the JEolic dialect Was spoken at Delphi, the na-
tive city of Phemonoe, the two accounts just given may easily be recon-
ciled. The most ancient hexameters known are those which Herodotus
informs us (5, 59) that he himself saw at Thebes, in the temple of the
Ismenian Apollo, inscribed on certain tripods, consecrated by Amphitry-
on, and by two other princes of the 13th or 14th century before our era.
It is more than probable, however, that the historian was here imposed
upon by the priests. (R'dhrad Herod., I. c. Wolf, Prolcgom. ad Horn ,
p. Iv.)
o
158 DACTYLIC MEASURES.
III. The dactylic hexameter was introduced into Latium
by the poet Ennius, who first discarded the rude Saturnian
measure of his predecessors. 1
IV. This species of verse is also called the Heroic, from
its having been selected by both the Greeks and Romans
as the proper medium for heroic or epic themes. It was
also employed in didactic and satiric compositions.
V. The Dactylic Hexameter consists, as its name im-
ports, of six feet, whereof the fifth is usually a dactyl, and
the sixth always a spondee, while each of the other four
feet may be either a dactyl or a spondee, at the pleasure
of the writer ; as,
Virg. Sic abe\unl rede\untque me\t vqri\antque tim\ores.
Catull. Et quam\vis te\cum mult\d con\jungerer \ usu.
Virg. Talis et \ ipse ju\bam cer\vice eff\udit e\quina.
Id. Collect\umque frem\ens volv\it sub \ naribus \ Ignem.
VI. Sometimes, however, in a solemn, majestic, or mourn-
ful description, or in expressing astonishment, consternation,
vastness of size, &c., a sporidee is admitted in the fifth foot,
and the line is thence termed Spondaic ; as,
Virg. Card deurri soboles, magnum Jovis \ lncre\mentum.
Id. Constitit,atque^>culisPhrygiq,agmina \ circum\spexit.
Cat. jEquoreas monstrum Nereides \ admir\antes.
Man. Scorpius ingentemperterru.it \ On\ona.
VII. In spondaic lines the fourth foot is usually a dac-
tyl ; not uniformly, however. Thus,
Virg. Saxa per et scopulos et \ depresses con\valles.
Id. Aut leves ocreas lent\o du\tunt ar\gento.
VIII. The t)lder poets do not scruple to use lines con-
taining spondees alone ; as,
Enn. Olli \ respond\et rex \ Alba\i long\ai.
1. Lua ;t., 1, 118, seqq. Sil. Ltd., 12, 410, seq. Column., Ennn
Vit., p. 6 "-For an account of the Sa'urnian, vid. page 199.
DACTYLTC MEASURES. f 159
Enn. Gives \ Roma\ni tune [ facti \ sunt Cam\pdni.
Lucret. An cwl\um no\bls na\tura ult\ro cor\ruptum.
Catull. Quis te \ lent\rem no\bis, neu | conar\cre.
IX. Spondaic lines are much more common in the Greek
than in the best Latin poets. There are, for instance, some
twenty-eight of this description in Virgil, while in a single
piece of Catullus" (containing 409 lines), who formed his
verses upon the Greek model, we find a greater number. 8
CtEsura in Dactylic Hexameters.
X. The beauty and harmony of hexameter verse depend
in a very great degree upon the proper management of the
caesura. In its application to single feet, the caesura has
already been explained ; it only remains to consider it with
reference to whole verses, in which acceptation it may be
more correctly termed the Cassural Pause.
XI. The caesural pause most approved of in heroic po-
etry, and which, above all others, tended to give smooth-
ness and rhythm to the line, was that which took place
after the penthemimeris. This was particularly distin-
guished as THE Heroic Cassural Pause. Thus,
Virg. At dom^s \ interior, || regali splendida luxu.
Id. Julius \ a magn\o.\\ de'missum nomen liilo.
XII. Sometimes the penthemimeral pause is found com-
bined with others ; as,
Virg. Ad nos vix tennis \\ fama \\ perlabitur aura.
Id. Insignis || reserat || stridentia limina consuk
Id. Sunt gemincR || belli || porta, \\ sis nomine dicunt.
XIII. Instead, however, of the caesural pause at the ex-
act penthemimeris, a different division was equally admitted
as heroic, which took place after a trochee in the third foot ;
as,
Virg. Effigi\em statu\ere, \\ mfas qua triste piaret.
Id. Tecta met\u peti\ere, || ruunt de montibus amnes.
1. Catull, 64. 2. Ramsay's Lot. Pro*., p. 173.
160 DACTYLIC MEASURES.
XIV. Next in merit to the penthemimeral was the heph-
themimeral pause. Thus,
Virg. Arbori\busque sat\isque Not\us, \\ pccorique sinister.
Id. Haud mora \ prosilu\ere su\is : \\ ferit athera clamor.
XV. In some instances we find lines where the penthe-
mimeral is combined with the triemnneral ; as,
Tibull. Di patrii || purgamus agros, \\ purgamus agrestes.
Virg. Prima tenet, || plausuque volat, \\ fremituque secundo.
XVI. The caesural pause the least approved of in heroic
poetry was that which divided the verse exactly into halves,
since it gave the line an undignified air, and degraded it to
a Priapean. 1 Thus,
Virg. Cut non \ dictus Hy\las puer || et La\tonia \ Delos ?
Id. Exple\ri ment\em nequit || ardes\cilque tu\endo.
XVII. The caesural pause between the fourth and fifth
feet was considered by grammarians as peculiarly adapted
to pastoral poetry, more especially when the fourth foot
was a dactyl ; and it was termed, from this circumstance,
the Bucolic caesural pause ; 2 as,
Calpurn. Idas \ laniger\i domi\nus gregis, || Astacus horti.
Auson. Commu\nis Paphi\e dea \ slderis || et deafloris.
13. PRIAPEAN.
I. The Priapean measure was so called because origi-
nally employed on themes having relation to Priapus, the
god of gardens.
II. This measure is generally regarded as a species 01
hexameter, the first, as likewise the fourth, foot of whicL
was most commonly a trochee, often, however, a spondee,
1. The Priapean will be considered after the hexameter.
2. Valckenaer was the first to mark the bucolic ciesura in Theocritus.
The first seven Idyls, with the tenth and eleventh, contain 927 lines, ol
which not less than 71 1 have this casnra. Virgil's Eclogues consist ol
830 lines, bui of (,hese only 2.T2 conform to the bucolic model ( Wai
ton. de Vo'/is Bucul. Thcocril .. vo! 1. |>. xxxvi )
ANAPJESTIC MEASURES.
161
but rarely a dactyl ; the second almost always a dactyl ;
the third, though sometimes a dactyl, much more frequently
an amphimacer. The scale is as follows :
1
2
3
4
5
6
_ ~ _
:L
-~~
Catull. O col\oma \ qua cupls || ponte \ ludere \ longo.
Id. lnfdss\a Llgu\ri jdcet \ supper\nata se\curi.
III. A preferable mode of scanning, however, is to make
the first hepiistich a Glyconic, and the second a Pherecratic
verse, and thus to consider the line, not as forming one
dactylic verse, but as composed of two choriambics.
14. HEXAMETER MEIURUS.
I. The Hexameter Meiurus is a defective species of hex-
ameter, having an iambus in the sixth place instead of a
spondee ; as,
Liv. Andron. Dirige odorisequos ad certa cubilia canes.
II. It is to be considered, however, rather as a vicious
and defective hexameter, than as forming a separate spe-
cies of verse ; though Livius Andronicus designedly wrote
such lines as these, which he alternately mingled with per-
fect hexameters. They have all, however, perished except
two.
SECTION XXXI.
ANAPAESTIC MEASURES.
I. In Anapaestic verse the feet admissible without restric-
tion are, the anapaest, the spondee, and the dactyl.
II. Dactyls, however, ought to be employed sparingly in
Latin anapaestics. When introduced, they ought to be the
first foot in the dipode, and ought to be followed bj a spon
dee in preference to an anapaest.
02
162 ANAP.ESTIC MEASURES.
III. Anapaestic verse is scanned by metres of two feet
each ; thus, an anapaestic raonometer contains two feet ; a
dimeter, Tour ; a trimeter, six, &c.
IV. No specimen of anapaestic verse is extant in the pu-
rer Latin writers. Those that have reached us belong to
a later age.
V. No Latin poet ever wrote anapaestics necessarily con-
sisting of four anapaests (with the exception of a few in
Seneca and Ausonius) ; but they all appear to have intend-
ed their anapaestics for single measures or monometers,
leaving the reader to connect or disjoin, as the sense might
require or his own judgment dictate. Convenience in
printing, however, is answered by the division into dime-
ters, and hence they are generally exhibited in this form
in editions of ancient authors.
VI. Anapaestic verses arranged in monometers or dime-
ters, and thus forming a succession of many lines, are call-
ed Systems.
VII. The species of anapaestics most in use among the
Greek tragic writers was the Anapaestic Dimeter Acatalec-
tic, which is frequently found in systems interspersed with
the Manometer Acatalectic.
VIII. These systems of anapaestic dimeters are usually
closed, among the Greeks, by an Anapasstic Dimeter Cata-
lectic, otherwise called a Parcemiac, it having been a favour-
ite vehicle for proverbs (Hapoipiai).
IX. The Latin writers do not, as often as the Greeks,
close a system of dimeters by a Paroemiacv Seneca never
does.
X. It must be borne in mind that anapaestic systems pro-
ceed on the principle of continuous scansion. Hence the
last syllable of every verse is not common, but is subject to
the ordinary rules of prosody, unless at the end of a sen-
tence, or any considerable pause in the sense. 1
1. Consult remarks under the article Synapkeia, page 130.
ANAPAESTIC MEAStniES.
163
XL The only exception to the rule laid down in the pre-
ceding paragraph is in the case of the Parremiac, the last
syllable of which is common. An anapaestic system, there-
fore, is, in fact, one long line broken up into several lines,
the end of which long line is marked by the Paroemiac.
XII. Hence we see why the last syllable of the ParoEmi-
ac is the only one in the entire system that is common, it
being in reality the concluding syllable of a long line, which
line is supposed to commence with the first verse of the
system.
1. ANAPAESTIC MONOMETER ACATALECTIC, or ANAP^STIC
I. The Anapaestic Manometer consists of two feet, which
may be either anapaests, spondees, or dactyls.
II. The scale, therefore, of this measure is as follows :
Senec. ululass\e canes. \\
Id. Fundlte \fletig ||
Edite | planctus ||
Fmgite | luctus \\
Resonet \ trlsti ||
Clamor\e forum. \\
2. ANAPAESTIC DIMETER ACATALECTIC.
I. The Anapaestic Dimeter Acatalectic consists of four
feet, which may be either anapaests, spondees, or dactyls.
II. The scale is as follows :
164
ANAPAESTIC MEASURES.
Senec. Indus \ gclidum || potat Ar\axem, ||
Albim | Persx, \\ Rhenum\que bibunf H
Venient \ anriis || sceculd \ sens \\
Quibus O\cednus || vincula \ return || '
Ldxet et \ ingens || pateat \ tellus ||
Tethys\que novos \\ detegat \ orbes ||
Nee sit | terns \\ ultima \ Thule. \\
3. ANAPJESTIC DIMETER CATALECTIC, or PARCEMIAC.
I. This measure consists of three feet, followed by a
catalectic syllable.
II. The third foot of a^Paroemiac must always be an an-
apaest. The dactyl is less frequently used in the first and
second places than the spondee.
III. The following is the scale.
Prudent. Venient \ cito scBC\\ula cum [ jam
Socius | color oss\\ a revis\it,
Ariimat\aque sang\\ume viv\o
Hdbitac\ula pnst\\ind gcst\et.
IV. Prudentius, from whom the above lines are taken,
does not admit a dactyl, and uses a spondee in the first
place only. Boethius, however, allows himself greater lat-
itude, as may be seen by the following :
Qul se | volet ess\\e potent\em.
Jlnlmos \ domct ill(\e fero\ces,
Nee vict\d libl\\dine coll\a
Faidts | submitt\\at hdbe\nis.
Etcnim | licet ind\\ica long\e
Tellus \ tiiajur\\a tremis\\cat,
Et servfiat ultima T/tii\le,
Tdmen atr\as pell\\ere cur\ds t
'
IAMBIC MEASURES. 165
Miseras\que Juga\\re quere\las
Non poss\e, potent\\ia non | est.
SECTION XXXII.
IAMBIC MEASURES.
- . I. Iambic verses are scanned by measures of two feet,
it having been usual, in reciting them, to make a short
pause at the end of every second foot, with an emphasis.
II. This kind of verse derived its name from the iambus,
of which foot it was originally composed, to the exclusion
of all others.
III. Afterward, in order to vary the rhythm, and dimm-
ish the labour of the poet, a spondee was allowed in the
odd places of the verse, the iambus still occupying the
even places.
IV. In the even places, the long syllable of the iambus
was resolved into two short ones, and thus the tribrach,
which is isochronous with the iambus, gained admission.
V. In the odd places, by resolving the first long syllable
of the spondee, an anapaest was formed ; and, by resolving
the second syllable, a dactyl.
VI. Thus eventually all these feet were employed in
iambic measures, subject, however, to certain restrictions,
which will now be specified. 1
1. An iambus is admitted into everyplace, which maybe
resolved into a tribrach in every place except the last,
where there must be invariably an iambus.
2. The spondee may be used in the uneven places ; that
is, in the first and third of the dimeter, and in the first,
third, and fifth of the trimeter.
3. The dactyl may be used in the uneven places, like
the spondee ; but its appearance in the fifth place of
the trimeter is very rare.
1. The rules here given are much less strict, as regards the dactyl
and anapaest, than those which regulate the Greek iambic trimeter.
166 IAMBIC MEASURES.
4. The anapaest also is admitted into the uneven places,
and in the fifth place seems to have been particularly
affected by the Roman tragedians.
5. A proceleusmaticus is sometimes found in the first
place of the iambic trimeter.
1. IAMBIC DIMETER CATALECTIC.
I. This measure consists of three feet, properly all iam-
bi, and a catalectic syllable.
II. It admits, however, the tribrach, spondee, and ana-
paest into the first place, but suffers no variation in the third
foot.
Petron. anus \ recoct\\a v1n\o
trement\tbus II labellfis.
Prudent. Lex h<r.c \ data 1st \ cadu\ds
Deo | jubent\\e membr\ts
ut temp\eret || labor\em
medica\bilis || volupt\as.
2. IAMBIC DIMETER ACEPHALOUS.'
I. This measure is an iambic dimeter, wanting the first
syllable.
II. Horace and Prudentius made no variations, but uni-
formly employed the iambus in the few lines they have
left us of this metre.
Horat. Non \ ebur \\ neque aur\eum.
Prud. Do\nd con\\scien\t'i(e.
III. Some consider such lines as catalectic trochaic dim
eters, and scan them as follows-:
Non eb\ur ne\\que aure\iim.
Dona | consci\\entife.
1. Acephalous, i. e., without a head (or initial syllable), from a, priv
and KBjxi'/Jr, " a head."
IAMBIC MEASURES.
167
3. IAMBIC DIMETER HYPERCATALECTIC.
I. This measure, likewise called Archilochian, and form-
ing the third line in the Alcaic stanza, is an iambic dimeter
with an additional syllable at the end.
II. According to the usage of Horace, the first foot may
be either a spondee or an iambus, but is generally a spon-
dee ; the second foot is an iambus ; the third is invariably a
spondee, 1 and the fourth an iambus. The Horatian scale,
therefore, is as follows :
Horat. Sylvas \ ldbor\\dntes \ gelu\\que.
Id. Depro\me quadr\\imum \ Sabin\\a.
Id. Puer | quis ex \\ auld \ cdplll\is.
HI. The practice of Horace differs from that of Alcaeus
as regards the third place, the latter having uniformly an
iambus in this part of the line.
IV. This measure is called by some the Alcaic Ennea
syllabic?
4. IAMBIC DIMETER ACATALECTIC.
I. This measure consists of four feet or two metres.
II. Propeily speaking, all the feet are iambi. It admits,
however, the variations that are usual in iambic verse, and
its scale is as follows :
1. The line of Horace, which occurs Od., 2, 19, 15, "Disjecta non
led rttina," has been corrected by Bentley, from MSS., as follows*
" Disjecta non lenl ruina."
2. Consult remarks on the Alcaic measure.
168 IAMBIC MEASURES.
I
Horat. tndrs\lt a'St\\uos\ms. II
Id. Vel h(ed\us c\\rcptus \ lupo. \\
Id. imbres \ ntves^que com\pdrdt. II
Id. dst ego | vidss\\lm ris\ero. II
Id. Vtder\e proper\\dntes \ domum. H
Prudent. Jam mcll\d de \\ scopulls \ fluunt. II
Hadr. drnmu\ld vagu\\ld bldnd\uld, \\
Hospes | cdmes\\que corp\orls, ||
QUCB nunc \ dbl'i\bls In | locd, ||
PdlUdu\ld rtgl\\duld nu\duld ? ||
Nee ut j 50/65 || ddbls \ jocos. tl
III. Although Horace has not used this metre except in
conjunction with verses of a different kind, other authors
wrote entire poems in it, as Prudentius, St. Ambrose, Pope
Damasus, Sedulius, &c.
IV. The Liturgy of the Church of Rome has several
hymns in this metre. The following lines form the com-
mencement of one of them, and will also furnish an instance
of rhyming or Leonine versification.
Salutis humana sator,
Jesu voluptas cordium,
Orbis redempti conditor
Et casta lux amantium :
Qua victus es dementia
Ut nostraferres crimina? &c.
5. IAMBIC TRIMETER CATALECTIC.
I. This measure is the iambic trimeter, wanting the last
syllable. In other words, it consists of five feet, properly
all iambi, followed by a catalectic syllable ; as,
Horat. Vocd\tus dt\\que non \ vocd\\tus aud\it.
Prudent. Plus \ fidc\\Ks ln\nocens \\ pudt\cus. *
II. Like the trimeter, however, it admits the spondee
IAMBIC MEASURES.
169
into thirst and third places, but not into the fifth, which
would render the verse too heavy and prosaic.
Horat. Trdhunt\que stc\\cds md\chin~<B \\ cart\nas.
Prudent. Nonnull\d querc\u sunt \ cdvd\\ta et ulm\d.
III. Terentianus prefers scanning this kind of verse as
part of an iambic trimeter, with three trochees following.
Thus,
Trdhunt\que slc\cds \\ mdchi\naB cdr\lnds.
6. IAMBIC TRIMETER ACATALECTIC.
I. This measure consists of three metres, or six feet,
originally all iambi, and when a line of this kind still oc-
curs, it is called a pure iambic line.
II. The other feet that were subsequently allowed to en-
ter, and the places proper to each, have been considered
elsewhere. 1
III. When other feet besides the iambus enter into the
line, it is called a mixed iambic.
IV. Another name for this measure is the Senarius, from
the six feet of which the line is composed.
V. The caesural pause most commonly falls after the
fifth semifoot ; as,
Phdse\lus 1ll\e || quern \ vide\fis hosp\ttes.
VI. The scale of the mixed iambic trimeter is as follows:
Pure Iambics.
Horat. Comes \ minor\\e sum \ futur\\us in | metu. ||
Catull. Es im\pudi\\cus et ' vorax || et dl\co. \\
1. Consult introductory remarks on iambic verse, paragraphs III..
IV., V., VI.
170 IAMBIC MEASURES.
Spondee in 1 and 3.
Catull. Per cons\uldltum peyerdt \\ Vdtin^us. fl
Spondee in 1, 3, 5.
Horat. Gnxe\rc ma\\tres /j/i dd\diclum \ feris. ||
Tribrach in 1, and Spondee in 5.
Horat. Sed oTtjus ardljor awi | /me//||5J candjufce.
Tribrach in 2 und 4, Spondee in 1, 3 and 5.
Horat. Vec^aj&or Au7ner||ts ^nc | ego ini\\mids \ eques,
Tribrach in 3, and Spondee in 5.
Hort. Li&e | jace\\re modo \ sub an^fiqua ll\ice. fl
Dactyl in 1, Spondee in 3 and 5.
Horat. Aut amtlte le\vl rar\a tend\\it re\(ia. ||
Dactyl in 3, Spondee in I and 5.
Horat. Quo quo \ scelest^l rui\tis out 11 cur dex\teris. D
Anapaest in 1 , Spondee in 3 and 5.
Horat. Positds\que vern\\ds di\tls ex\\amen \ domus. |
Anap&st in 1 and 5, Tribrach in 2.
Horat. Pawdum|yue lepo\\rem et dd\vendm | laqueo \ gruem.^
Dactyl in 1, Tribrach in 3 and 4.
Horat. Altti\bus dt\que cani\bus homt\\cldam Hect\orem. II
VII. The prevalent caesural pause in iambic triir,cters is,
as we have already remarked, the penthemimeral. If this
be wanting, the line must then have the hephthemimeral
pause. Thus,
Catull. Ait fuisse || navium cehrrimus.
Id. Rhodumve nobilem \\ horridamve Thraciam.
Horat. Defixa cado || devocare sidera.
Id. Cave ! cave namque \\ in malos asperrimus.
IAMBIC MEASURES. 171
Catull. Neque ullius natantis || impetum trabis.
Id. Proponiida trucemve \\ Ponticum sinum.
VIII. There is no instance in Catullus of the total omis-
sion of the caesural pause, and only two in Horace, namely
(Epode, 1, 19) Ut assidens implumibus pullis avis.
(Epode, 11, 15) Quod si meis inasstuat pracordiis.
Porsonian Pause.
IX. Porson, in his celebrated preface to the Hecuba, as
serted that the following rule was always observed by tho
Greek tragedians :
" When the iambic trimeter has, after a word of more
than one syllable, the cretic termination ( ^ ), either in-
cluded in one word, or consisting of ^ and a syllable, or
of a monosyllable and -^ , then the fifth foot must be an
iambus."
X. There is no exception to this law in Catullus, whose
iambic trimeters are almost all pure ; but it is constantly
violated by Horace, in those odes in which iambic trime-
ters are combined with other kinds of verse ; whereas in
Epode 17, where these form a system by themselves, it is
but once neglected :
Alitibus atque canibus homicidam Hectorem. 1
7. SCAZON, or CHOLIAMBTJS.
I. This measure, called also Hipponaclean, is a variety
of the Senarius. It differs from it in this, however, that
while the iambic trimeter has invariably an iambus in the
sixth place, the scazon has invariably a spondee in the
sixth place, and an iambus in the fifth.
II. In all other respects the scazon is the same as the
trimeter.
III. An iambus is necessary in the fifth place, in order
that the line may not be too lame and heavy ; as in spon-
1. Ramsay's Lot. Pros., p. 208.
172 IAMBIC MEASURES.
dale hexameters, we usually find the fourth foot a dactyl
for a similar reason.
JV. The terms Scazon (Luafav, " limping" or " halting")
and Choliambic (XwAta^fii/cd^, " lame iambic") have refer-
ence to the peculiar characteristic of this measure, namely,
its lame and halting gait, occasioned by the spondee in the
sixth place, in opposition to the brisk and lively ending of
the regular trimeter.
V. The name Hipponactean is derived from that of the
virulent poet Hipponax, who, according to some, invented
this measure. After his example it was much employed
in invectives and in sarcastic composition generally.
VI. The following may be taken as a specimen of this
measure.
Catull. Suffe\nus ist\\e Vdr\e quern \\ probe \ nosti, ||
Homo est | venust\\us et \ dlcax H et urb\dnus, \\
\dem\que long\\e plur\imds \\fdcit \ versus. \\
Puto ess\e ego lll\\i mttl\ia out \\ decem out \ plurd [|
Perscript\a nee || sic, ut \ fit, In || pdlim\psesld ||
Reld\td : chdrt\\<B rcg^iz, || novi \ libn, \\
Novi umb\ili\\ci, lbr\d rubr\\d, membr\dnd ||
Direcl\d plumb\\o, et pu\mice omn^ia <E\qudta. ||
8. COMIC IAMBIC TRIMETER.
I. In comedy, satire, and fable, the poets indulged in
very great licenses as regarded the structure and scansion
of the trimeter.
II. They admitted the spondee and its equivalents, the
dactyl and anapaest, into the second and fourth places, not
confining themselves to the iambus except in the sixth. 1
Thus,
1. " Cum nonsolum Tercntius, sed eliam Plautus, Ennius, Accius-
que et Ncevius atque Pacumus Turpiliusque, et omnr.s tarn tra.gce.diat
quam comaedift vetcris Latins scriptores codem metri modo lambici sunt
usi, ut omnibus in locis indifferenter poncrcnt qitinque pcdes ; id cst
iambum, vd tribrachyn, vel ana pas turn, vel dactyhtm, vel spondceum,
vbsquc postrcmo loco, in quo vd iambum vd pyrrkicfiium omnino poau-
IAMBIC MEASURES. 173
Petron. An ut \ mdtro\\na ornd\td phdle\\ris peld\gns. ||
Id. Two | pdld\\to cldu\sus pdv\\o pdsc\ltur. ||
Id. ~JE,quum est \ indue^re nupt\dm ven\(tum textfilfp. ||
Phaedr. Peri\culd\\sdm fe\cit medi\\andm \ lupo. ||
Id. Est dr\deUo\num qua\ddm Ro\\m(B nG\lio. \\
Id. Rex urb\is e\\jus ex\perien\\dl grd\tm. ||
Id. lgno\tos fdllfct no\tis est || den\sui. ||
Terent. Fide et \ tddtur\\nitd\te. Exspect\\o quid \ veils. [1
Id. Cur simu\lds if^-i|Uwr rem om\ncm dprin\\cipio au\dies. |
Id. Quod ple\riquc om^nes fdti\unt ddu\\lescen\tuli. \\
Id. Alere dTit \ canes II dd ven\dndum out ]| dd phtlo\so-
phos. II
Id. Nunqudm \ preEpon\\ens se ill\is ltd || fdcill\ume. ||
Id. Age\bdt ldn)\a dc te\ld vicC^um qu&\ritdns. \\
Id. Egomet | continued me\cum cer\\te cdpt\us est. ||
Id. Die so\des quts her\\l Chrysfidem hdbu^ll nam An]
dna. ||
Id. Qu(ere\bdm cdm^peru\bdm riihil || dd Pdm\philum. \\
9. IAMBIC TETRAMETER ACATALECTIC, or OCTONARIUS.
I. This species of verse consists of four measures or
eight feet, properly all iambi, but subject to the same vari-
ations as the iambic trimeter ; so that, by prefixing one
measure to a common iambic trimeter, we convert it into
an Octonarius.
II. This metre is often used by the Latin comic writers.
Terent. Sane \ pol ls\\td te [ mulent\\a est mutt\er et \\ tem-
erdr\id. \\
Id. Nunc htc \ dies \\ dlidm \ vitam a~u\\fert dli\os mo-\\
res post\uldt. \\
Id. Pdtere\tur : ndm \\ quern ferr\et si || pdrent\em non \\
ferret \ suum ? ||
issc inveniuntur ; miror quosdam vel abnegare esse in Tcrentii comacdiis
mctra, vel ca quasi arcana qu<edam, et ab omnibus doctis semota, sibi
salts esse cognita, confirmare." (Priscian, de Vers. Com. vol. 2, p.
403, ed. Krehl.)
P2
174 IAMBIC MEASURES.
Terent. Lena \ sum fatc\\or per\nlcles \\ cbmmu\ms ddo\\les-
cent\lum. \\
Id. Nequld \ proptcr \\ tudm \ fldem \\ decept\d pdte-\\
tetur | mail. ||
Id. Cujus | nunc mlse\\r<e spes \ opes\\que sunt \ In te
H\\no bmries \ slice. \\
Plaut. lllbs | qui ddnt \\ eos \ derl\\des; qui \ delu\\dunt de-\
perls. ||
10. IAMBIC TETRAMETER CATALECTIC.
I. This measure, called likewise Hipponactic, from its
inventor, Hipponax, is the tetrameter or Octonarius depri
ved of its final syllable.
II. The same variations are admissible here as in the
case of the trimeter and tetrameter, and the comic writers,
who sometimes used this species of verse, took as great
liberties with it as with the trimeter ; always observing,
however, to make the seventh foot an iambus.
Catull. Remltt\e pall\\lum \ mlhl \\ meum \ quod ln\\volds\ti.
Id. Dcprens\d nau||zfs In \ marl || ue*ajnzen||^e venl\o.
Id. Quum de \ via \\ muller \ aves || ostend\lt osc\\ltant\es.
Terent. Non poss\um saf? || narrdr\e quos || ludos \ prabue-H
rls lnt\us. ^
Id. Nostrd\pte cul\\pd facl\mus ut |) maids \ expedi\\at
ess\e.
III. In this measure there is uniformly a division of the
verse at the end of the fourth foot. Thus,
Remitte pallium mihi || meum quod involasti
Deprensa navis in mari \\ vesaniente vento.
IV. This species of verse is a great favourite with the
Greek comic poet Aristophanes, and is also found in many
passages of the Latin comedians. The only specimen of
it, however, in a pure state in the Latin classics, is a short
poem of Catullus'. 1
1. This measure comes to our ears with a very lively and graceful
cadence in those lines where accent and quantity do not clash. Thus,
IAMBIC MEASURES.
175
11. GALLIAMBUS.
I. This measure derives the first part of its name from
the Galli, or priests of Cybele, by whom it was employed
in their wild orgies.
II. The only specimen of this verse extant is the poem
of Catullus on the legend of Atys, consisting of 93 lines,
and remarkable for the wild dithyrambic spirit that per-
vades it.
III. From the scanty remains that we possess of this
measure, it becomes a difficult task to determine its struc-
ture with any great degree of precision. The following,
however, is the scheme given by Vulpius, an eminent com-
mentator on the Roman poet :
1. The first foot of the six into which he divides the
measure is generally an anapaest, but sometimes a
spondee or a tribrach.
2. The second is generally an iambus, rarely an anapaest,
a tribrach, or a dactyl.
3. The third is generally an iambus, rarely a spondee.
4. The fourth is a dactyl or spondee.
5. The fifth is often a dactyl, sometimes a cretic or
spondee.
6. The sixth is an anapaest, and sometimes an iambus,
preceded by a cretic.
According to this view of the Galliambic measure, the
scale is as follows :
Idemque Tholle turbida rapacior procclla ;
like the English ballad, " And thus we daily dance and sing, and cast
all care behind us "
TROCHAIC MEASURES.
IV. The following lines will afibrd an idea of the meas-
ure :
Super dl\td vect\us At^s ccler\l rate \ mdrid.
Ubi cdp\ild Md,\nddes j mjdci\unt hede\ngerte.
Viridem [ cltus dd\tt \\ddm proper\dnte pe\de chorus
Ahero | ford \ pdla:sl\rd stddi\o et gymn\dsus.
V. Some prosodians, however, make the Galliambic
measure consist of an iambic dimeter catalectic (the first
foot of which is generally a spondee or an anapaest), fol-
lowed by another such dimeter wanting the last syllable.
Hence they give the scale with its variations as follows:
1
2
3
4
5
6
7
^
~ -
~-
^
~^~^
ft- 1
i i
,
w W | *,_
Ubi cdpi\td M<cn\ddes \ vi \\ jdciunl \ hederi\gerce.
Viridem \ cilus dd\it l\ddm || properdnt\e pcde \ chorus.
SECTION XXXIII.
TROCHAIC MEASURES.
I. Trochaic verse derives its name from the foot which
prevails in it, namely, the trochee.
II. Originally the trochee was the only foot allowed to
enter into the line ;~ but variations were afterward introflu-
ced, as in the case of iambic verse.
II f. The trochee, like the iambus, is convertible into a
tribrach- Hence this last-mentioned foot is allowed to en-
ter, and so are the spondee and anapaest. The dactyl,
however, is in general not admitted, except in the case of
a proper name.
IV. A difference, deserving of careful notice, exists be-
tween the iambic and trochaic measures, in that the former
TROCHAIC MEASURES. 177
admits the spondee and anapaest into the uneven places
but the trochaic into the even only.
1. TROCHAIC DIMETER CATALECTIC.
I. This measure consists of three feet, properly all tro-
chees, and a catalectic syllable ; as,
Horat. Non e\bur ne\\que <&re\um.
Prud. Dona \ cdnsci\entl\<B.
II. In the second place it admits the spondee, the dac
tyl, and likewise the anapaest.
Senec. Lents \ dc modt\\cumjlu\ens,
Aura | nee ver\\gens ldt\us
Ducat | lnlrepV\ddm rdt\em
Tutd | me medi\\d ve\hdt
Vita | decurr\\ens vi\a.
III. The trochaic dimeter catalectic is otherwise scanned
as an Iambic Dimeter Acephalous.
2. TROCHAIC DIMETER ACATALECTIC.
I. The trochaic dimeter consists of four feet, properly
all trochees ; as,
Boeth. Nonfd\ctt quod \\ optat \ ipse.\\
II. The spondee, however,, and its equivalents in quan-
tity, the anapaest and dactyl, are admitted into the second
place.
Buchanan. Inco\lae terr\\drum db \ or/, ||
Solis | uUl[\mum dd cu\blle ||
Ejd | Domino || jubi\ldt e. \\
Consci\os sceler\\is ne\fdndi. \\
3. TROCHAIC TETRAMETER CATALECTIC.
I. This measure consists of seven feet and a catalectic
syllable.
II. Originally all the feet were trochees, but variations
178
TROCHAIC MEASURES
having been subsequently introduced, the following result
was finally obtained :
1. The trochee may in every place be resolved into a
tribrach.
2. In the even places, that is, the second, fourth, and
sixth, in addition to the trochee, a spondee is admis-
sible, which may be resolved into an anapaest.
3. A dactyl is aRnissible, in the case of a proper name,
in any place except the fourth and seventh.
III. The csesura uniformly takes place after the fourth
foot, thus dividing the verse into a trochaic dimeter acata-
lectic and a trochaic dimeter catalectic.
IV. The scale, according to what has just been laid
down, is as follows :
1
2
3
4
5
6
7
--
Cras am\et qui || nunquam am\amt || qutque am\avlt \\ eras
am\et.
Ver nov\um, ver || jam can\orum || vere j natus || orois \ est.
Vere \ concord\\ant am\ores, || vere \ nubunt | atit\es,
t nem\us com\\am re\solvit ][ de mar\itis \\ lmbn\bus. 1
V. The following lines are given without the marks of
scansion, to show the place of the caesura.
Quando ponebam novellas [| arbores mali et piri
Cortici summ<R notavi \\ nomen ardoris mei.
Nullafit exinde finis || vel quies cupidinis :
Crescit ardor, gliscit arbor, || ramus implet literas?
VI. This metre was much used in hymns, for whicH,
1. These lines are taken from the Pcrvigilium Veneru.
2. Burmann, Anthoi. Lot., vol. 1, p. 687.
CHORIAMBlC MEASURES. 179
from its grave and sonorous character, it is well adapted.
The division made in the line by the caesural pause would
suit, no doubt, the convenience of the chorus, one portion
of their number singing the complete dimeter, the other the
catalectic. Thus,
Prudent. Macte judex mortuorum, ||
Made rex viventjum.
M. Cap. Scande cceli templa Virgo, ||
Digna tanto feeder e.
VII. The comic writers took equal liberties with this as
with the iambic measure, introducing the spondee and its
equivalents, the anapaest and dactyl, into the trochaic places.
Terent. Quot mod\ts con\\temtus | spretus? \\fdcta \ trans-
act\\a omrii\a hem.
Id. Tantdm J rem tarn || necle\genter \\ agere? \ prceteri-\\
ens mod\o.
Id. Obsipu\i cen|jsea' me \ verbum || potulss\e ullum ||
prolo\qui out.
Id. Tot me lm\pediunt || euros \ qua me^um ariimum \
divorllsce tra\kunt.
Id. Attquid | facerem, ut || hoc ne \ fdcerem || sed nunc \
quid pri^mum exse\quar?
SECTION XXXIV.
CHORIAMBlC MEASURES.
I. Choriambie verses are so denominated from the foot
(or measure) which predominates in them, namely, the
choriambus, compounded of a choree (or trochee) and an
iambus ; as, Tantalidae,
II. The structure of choriambic verses is extremely sim-
ple, the first foot (with the exception of the dimeter) being
generally a spondee, sometimes a trochee or iambus, the
last an iambus, while one, two, or three choriamb! are in-
terposed.
160 CHORIAMBIC MEASURES.
1. CHORIAMBIC DIMETER.
I. The Choriambic Dimeter consists of a choriambus
and a bacchius ; as,
Horat. Lydta die \ per dmnes.
Perdcre cur \ apricum.
Cur neque mi\Utaris.
Temperat o\ra freriis.
II. This measure occurs once in Horace, in conjunction
with another species of choriambic verse^
2. CHORIAMBIC TRIMETER ACATALECTIC, or GLYCONIC.
I. This measure, called Glyconic from the poet Glycon,
its inventor, consists, as it appears in Horace, of a spondee,
a choriambus, and an iambus ; as,
Sic te | Diva potens \ Cypri.
Vento\rumquc regal \ pater.
Ndvis \ qua tibi cre\ditum.
II. This species of choriambic verse is not used in *
system by itself in the works of either Horace or Catullus,
but in combination with other species of choriambic meas-
ures.
III. Horace, who was very fond of the Glyconic, invari-
ably adheres to the spondee in the first place, except in the
two following instances, where a trochee occurs :
Teucer et Sthenelus sciens. (Orf., 1, 15, 24.)
Ignis Iliacas domos. (Ib., 1, 15, 36.)
The best editions, however, now read,
Teucer, te Sthenelus sciens.
Ignis Pergameas domos.
IV. Catullus, however, frequently has a trochee in the
first place ; as,
Rustfya agricbla: \ bbnis.
Cinge | tempora flbr^ibus.
Flamme\um cape l<K\tus hue.
CHORIAMBIC MEASURES. 181
Nupti\alia con\cinens.
Voce | carmina inn\ula.
V. The same poet also occasionally uses an iambus in
the first place ; as,
Puell\(R et pueri ln\tegn.
VI. In the following line Horace lengthens a short final
syllable by the force of the caesural pause :
Siji\gtt adamant\inos.
3. CHORIAMBIC TRIMETER CATALECTIC, or PHERECRATIC.
I. The Pherecratic verse, so called from the poet Pher-
ecrates, is the Glyconic (which we have just been consid-
ering) deprived of its final syllable. It consists of a spon-
dee, a choriambus, and a catalectic syllable ; as,
Horat. Gralo \ Pyrrha sub an\tro.
Id. Nigris \ Hzquora vent\is.
Id. Sperdt \ nescius ourfe.
II. This species of verse, like the preceding, is not found
in a system by itself in the works of either Horace or Ca-
tullus.
III. In Horace, the first foot is invariably a spondee. In
Catullus, although a spondee is sometimes employed in *he
first foot, a trochee is far more common ; as,
Amrii\umque sonant\um.
Dicta \ famine Lun\a.
Tecta | frugibus ex\ples.
IV. An iambus also occurs in Catullus, but rarely ; as,
Puell\(zque cana\mus.
Hymen \ Hymence\e.
V. In one instance in Catullus, a long syllable is found
supplying the place of the two short ones that contribute to
form the choriambus ; as,
Nutri\unt humor\e.
Q
182 CHORIAMBIC MEASURES.
VI. The Pherecratic, as it appears in Horace, with a
spondee in the first place, is scanned by some as a dactylic
trimeter acatalectic. Thus,
Grata \ Pyrrha sub \ dntro.
Nigris | aequord \ ventis.
Sperdt | nescius \ aura.
4. CHORIAMBIC ASCLEPIADIC TETRAMETER.
I.This measure (sometimes called the Minor Aschpia-
dic) consists of a spondee, two choriambi, and an iambus.
II. The name Asclepiadic is derived from that of the
poet Asclepiades, who is said to have been the inventor of
the measure.
III. The following are specimens of it :
Horat. M<ece\nds atams || edile re\gibus.
Senec. Non ill\um poterdnt |I figere cuspfides.
Prud. Hostis | dirus ddest || cum duce per\fido.
IV. The caesural pause always falls after the first chori-
ambus, as marked in the lines just given ; and in the fol-
lowing verses this pause lengthens a short syllable.
Horat. Quam si quidquid ardt |] impiger Appulus.
Certa sede manet, \\ humor et in genas.
V. The position of the caesural pause after the first chori-
ambus, facilitates the scansion of this measure as a dactylic
pentameter catalectic. Thus,
Mazcc\nds atd\vts \\ edite \ regibus.
Non lll\um pote\rdnt \\ftgere \ cuspides.
Hostis | dirus dd\est^ cum duce \ perfido.
VT. This latter mode of scanning was, as we learn from
Terentianus Maurus, adopted by many of his contempora-
ries. He himself, however, condemns the practice. 1
1 Sunt qui tradidtrint, ultima nersui
Tanquam pentametro syllaba dempta sit,
Qxam si rcstituas, pentametrum fore, &c. (Ter. Maur., 2650.)
CHORIAMBIC MEASURES. 183
VII. Sometimes, though very rarely, the first foot of the
Asclepiadic was made a dactyl ; as,
Senec. E,ffugi\um, et ttdseros libera rnors vocet.
M. Capell. Omnige\num genitor regna movens Deum.
5. CHORIAMBIC TETRAMETER ACATALECTIC.
I. This species of verse consists of three choriambi and
a bacchius (-^ ) ; as,
Sept. Ser. Jane pater, \ Jane tuens \ dive biceps \ btformis.
Auson. Tu bene si j quid facias | non meminiss\efasest.
Claud. Omne nemus, \ cum fluvus, \ omne canal \ pro-
fundum.
II. It admits, however, of variations, each of the three
choriambi being changeable to other feet of equal time ; as,
Seren. Cui resera\t.d mugiunt \ aurea daust\ra mundi.
Id. Tlbi vetus ar\d caluit &bo\rigineo \ sacello.
6. CHORIAMBIC PENTAMETER ACATALECTIC.
I. This measure consists of a spondee, three choriambi,
and an iambus ; as,
Horat. Tu ne \ qu&sieris, \ scire neffc, \ quern mihi, quern \
tibi.
Id. Nullam \ Vare sacra \ vite prius \ sevens arb\orem.
Catull. Alphe\ne immemor, dt\que unammis \falsesoda\Hbus.
II. This species of verse is sometimes called the Great-
er Asclepiadic.
EPICHORIAMBIC VERSE.
I. By Epichoriambic verse is meant a species of measure
wb'ch admits some feet that do not properly belong to cho-
And again :
Quod jam pentametri nan patitur modus ;
Nam tic tres videas essc pedes datos. (Id., 2663.)
184 CHORIAMBIC MEASURES.
riambic measure, but which are, as it were, superadded
thereto. 1
II. The most important varieties of this species of meas-
ure are the two following :
1. EPICHORIAMBIC TRIMETER CATALECTIC, or SAPPHIC.
I. This measure is a variety of the choriambic trimeter
catalectic, and is composed of a second epitrit, a choriam-
bus, and a bacchius ; as,
Horat. Jam satis ter\ris ritvis at\que diras.
Catull. Casaris vts\ens mdnument\a magni.
II. In practice, however, it is more convenient to con
sider it as composed of a trochee, a spondee, a dactyl, and
two trochees ; as,
Jam satfis terras riivls \ atque \ dlr$.
Casarfit vis\ens monu\menta \ magni.
III. Horace invariably has a spondee in the second
place ; but Catullus, imitating the example of the Greeks,
admits a trochee ; as,
Seu Sac\ds sag\itttfer\dsque \ Parthos.
IV. Horace generally makes the first syllable of the dac-
tyl caeeural ; as, .^
Pindarum quisquis \\ studet CEmulare.
Sanguinem, per quos || cecidere justa.
Integer vitas || scelerisque purus.
V. More rarely the first two syllables of the dactyl close
a word, thus forming a species of trochaic caesura ; 2 as,
1 . The term Epichoriambic is from kxi, " in addition <o," and xopt-
2. Horace, however, seems to have changed his opinion with regard
to this pause. In the first three books of the Odes it occurs but seldom
(as, for example, 1, 10; 1, 12; 1, 25; 2, 30; 2, 6, &c.), while in
book fourth it happens eleven times in odes second and sixth, four timea
in ode eleventh, and twelve times in the Carmen Saculare. The form
Nuntivm curcecque lyra parcntcm,
where the enclitic que is the second *yl!al!c of the dactyl, occurs twice
CHORIAMBIC MEASURES. 185
Laurea donandus \\ Apollinari.
Pinus out impulsa \\ cupressus Eu f o.
VI. In one instance, Horace lengthens a short syllable
in the caesura ; as,
Angulus ridel || ubi non Hymetto. (Od., 2, 6, 14.)
VII. Catullus, following the Greeks, neglects this caesu-
ra altogether ; as,
Sen Sacas sagittiferosque Parthos. (11, 6.)
Ultimi flos prostereunte postquam. (11, 22.)
Sapphic Stanza.
I. This stanza, so called from the two celebrated frag-
ments of the gifted Sappho that have reached our times,
consists of three Sapphic lines, such as have just been de-
scribed, followed by an Adonic, or Dactylic Dimeter Acat-
alectic.
II. Taking Horace for our model, the scale of the Latin
Sapphic stanza will be as follows :
Jam satis terris \\ nivis atque dirts
Grandinis misit \\ pater, et rubente
Dextera sacras \\ jaculatus arces
Terruit urbem.
III. There is one feature prominently conspicuous in the
Sapphic stanza, namely, a close connexion between the
third and fourth lines, and hence Horace four times divides
a word between them :
only in the first three books, namely, Od., 1, 10, 6, and 18, while in the
fourth book it is found four times in ode second, once in ode sixth, and
seven times in the Carmen Saculare. (Ramsay's Lat. Pros., p. 195.)
Q 2
186 CHORIAMBIC MEASURES.
Labitur ripa, Jove non probante, ux-
-orius amnis. (1, 2, 19.)
Tkracio bacchante magis sub inter-
-lunia vento. (I, 25, 11.)
Grosphe non gemmis ncque purpura ve-
-nale nee auro. (2, 16, 7.)
Pendulum zona bene te secuta e-
-lidere collurn. (3, 27, 59.)
IV. We have a similar instance in Catullus (11, 11) :
Gallicum Rhenum horribilisque ultim-
-osque Britannos.
V. This division of a word is confined, it will be -remem-
bered, to the third and fourth verse; no example being
found of such a division at the end of the first, second, or
fourth. Hence it has been conjectured, and the supposi-
tion is a very probable one, that neither Sappho, nor Catul-
lus, nor Horace ever intended the stanza to consist of four
separate verses, but wrote it as three, namely, two five foot
Sapphics, and one of seven feet (the fifth foot of the long
verse being indiscriminately either a spondee or trochee) ;
thus: 1
Jam satis terris nivis atque dirce
Grandinis misit pater, et rubente
Dextera sacras jaculatus arces, terruit urbem.
dum se nimium querenti
Jactat ultorem, vagus et sinistra
Labitur ripa, Jove non probante, uxorius amnis.
Otium bello furiosa Thrace,
Otium Medi pharelra decori,
Grosphe, non gemmis, neque purpura venale nee auro.
VI. Elision sometimes takes place between the second
I. Monthly Review, January, 179? p. 45.
CHORIAMBIC MEASURES. 187
and third, and the third and fourth lines. Thus, in 'Hor-
ace,
I 2. Dissidens plebi numero beator(um)
I 3. Eximit virtus, &c. (2, 2, 18.)
2. Mugiunt vacca, libi tollit hinnit(\im)
3. Apia quadrigis equa, &c. (2, 16, 34.)
2. Plorat, et vires animumque moresfaue)
3. Aureos educit in astro, nigro(que)
4. Invidct Oreo. (4, 2, 22.)
:3. Romulas genti date remque prolemfaue)
4. Etdecusomne. (C. S., 47.)
VII. Elisions of this kind, however, are not necessary
in their nature. Thus we find an hiatus between the third
and fourth lines in the following, from Horace :
Neve te nostris vitiis iniquum
. Odor aura. (1,2, 47.)
Between the first and second, from the same :
Sive mutatajuvenem figura
Ales in terris, &c. (1, 2, 41.)
And between the second and third :
Aut super Pindo gelidove in Haemo
Unde vocalem temere insecutas
Orphea sylvas. (I, 12, 6.)
2. EPICHORIAMBIC TETRAMETER CATALECTIC, or GREATER
SAPPHIC.
I. This measure consists of a second epirrit, two chor-
iambi, and a bacchius ; as follows :
Horat. Te 'deos or\o Sybarin\\cur pr operas \ amandd.
188 IONIC VERSES.
II. It is, in fact, the ordinary Sapphic, with the addition
of a choriambus in the third place.
III. The first syllable of the first choriambus ought to be
caesural, and there ought to be a division of the verse after
the first choriambus. 1
SECTION XXXV.
IONIC VERSES.
Ionic verses are of two kinds, the Ionic a majore and
Ionic a minore, which are so denominated from the feet or
measures of which they are respectively composed.
1. IONIC A MAJORE VERSES.
Of these the most celebrated is the
Ionic a Majore Tetrameter Brachycatalectic.
I. This measure is otherwise called the Sotadean, from
Sotddes, a Thracian, who lampooned Ptolemy Philadelphus.
II. In its pure state it consists of three Ionic a majore
feet, followed by a spondee, according to the following
scheme :
Tula mans \ iras vtdet j e littore J nauta.
III. Several of these Sotadean verses are to be found ia
the remains of the Greek poets, and have been carefully
analyzed by Hermann. In Latin, a short fragment of En-
nius, and a few irregular lines in Martial and Petronius
Arbiter, are the only specimens of the measure, except
such as are met with in Plautus. (AuL, 2, 1, 30 ; 3, 2.)
IV. The Ionics a majore of Martial, and these are but
two lines, have the proper foot in the first two places, and
a ditrochaeus in the third, followed by a spondee.
Has cum gemin\d compede \ dedicdt catenas
Sdturne tib\l Zbilus \ dnnulos pri\ores. 2
1. Hermann, D. M. E., 3, 16. Ramsay's Lat. Pros., p. 199.
2. Ep., 3, 30.
IONIC VERSES. 189
V. This change of the third measure to .a ditrochaeus
eems to have been a favourite with the writers in this
species of verse, as tending to giv.e greater softness and
harmony to the otherwise stiff and monotonous line.
VI. By a farther variation, either of the long syllables
in each of the three Ionic measures might be resolved into
two short, which resolution was regarded as an improve-
ment ;' but it does not appear that both the long syllables
were ever thus resolved at the same time.
Petron. Pede tendite, | cursum addite, convolate planta.
Cafcillus er|i consimilis pedis figura?
Solet integer \ anapaestus et | in fine locari.
Hunc ejfici\et Minucms | ut quis vocitetur.
Catalexis enim dicitur \ ea clausula | versus.
Petron. Ferrum timui, quod trepi\do male dabat [ uswn.
2. IONIC A MINORE VERSES.
I. The Ionic a minore verse is entirely composed of that
foot or measure called the Ionic a minore.
II. We have one specimen of this kind of verse in Hor-
ace (Od., 3, 12), which is differently arranged by different
editors, but is usually considered as a system of Tetramc'
ters Acatalectic.
III. Ionic a minore verses, in fact, are not confined to
any particular number of feet or measures, but may, like
anapaestics, be extended to any length, provided only, that,
with due attention to synapheia, the final syllable in each
measure be either naturally long, or be made long by the
concourse of consonants ; and also that each sentence or
period terminate with a complete measure, having the
spondee or two long syllables of the Ionic foot for its close.
1. " Nam, quofuerint crebrius hipedes minuti,
Vibrare sonum versiculos magis videmus."
(Terent. Maur.,v. 2054, seq.)
2. This line, together with the* ihree that follow, are taken from Te-
rentianus.
190
LOGAOSDIC VEFxSES.
IV. The specimen from Horace above alluded to is as
follows :
Mlserdrum est \ neque dmbri \ dare ludum, \ neque dulcl
Mala vino \ Idvere ; out ex\dnimdri \ metuentes
Pdtruce verb\erd lingua. \ Tibi qudlum \ Cylhereas
Puer ales, \ tibi telds, \ operdsce\que Minerve
Studium laufert, \ Neobule, \ Lipdrei \ nitor Hebri, &c.
SECTION XXXVI.
LOGACEDIC VERSES.
I. Loga&dic verses are those which, are formed by add-
ing any number of trochees to any dactylic verse.
II. They receive their name from Adyo^, " discourse"
and aoidrj, "song" because dactylic verse is the lofty
language of poetry, whereas the trochaic approaches more
nearly to ordinary discourse. 1
III. Of logaoedic verses tlie most important are the fol-
lowing :
1. ARCHILOCHIAN HEPTAMETEE.
I. This measure is composed of a Dactylic Tetrameter a
priore, followed by a pure Trochaic Dimeter Brachycata-
lectic.
II. The first three feet may be either dactyls or spon-
dees ; the fourth is always a dactyl ; the last three are tro-
chees. Thus,
1
2
3
4 II 5
6
7
1
Horat. Solvilur \ acns hy\ems grd\td vice || verts \ et Fav\orii.
III. The first syllable of the third foot ought to be caesu
ral, and the fourth foot ought to end with a word.
1. The iambus is the true foot for discourse, and the trochee for dan-
cing. Thus Aristotle remarks, /idAiora ^EKTIKOV ruv ptrpuv TO mu-
tetov tan. (A. P., 4.)
LOGACBDIC VERSES. 191
IV. Horace uses this species of verse once in Od., 1, 4,
where it is placed alternately with an Iambic Trimeter
Catalectic.
2. ALCAIC DECASYLLABIC, or MINOR ALCAIC.
I. This measure consists of a pure Dactylic Dimete-
Acatalectic, followed by a pure Trochaic Manometer Acat
alectic ; as,
Flumina \ constiter\lnt ac\uto.
II. This forms the fourth line of the celebrated Alcaic or
Horatian stanza, Avhich we are presently to consider.
3. PHAL^ECIAN HF.NDECASYLLABIC.
I. This measure, termed Phalaecian from the poet Pha-
laecus, and Hendecasyllabic because consisting of eleven syl-
lables (IvSena aivl/lafiat), is composed of five feet, a spon-
dee, a dactyl, and three trochees ; as, .
Mart. Non est \ vlvere \ sed val\ere \ vita.
Catull. Quoi don\o lepld\um nov\um Kb\ettum.
II. In other words, it consists of a Dactylic Dimeter
Acatalectic, followed by a Trochaic Dimeter Brachycatalectic.
III. Catullus, with whom this is a favourite measure,
uses a trochee not unfrequently in the first place, and some-
times an iambus ; as,
Aridja modo pumice expolitum. (1, 2.)
Tota | millia me decem poposcit. (41, 2.)
Aml\cos medicosque convocate. (41, 6.)
Meas | esse aliquid putare nugas. (1, 4.)
IV. This liberty, however, was rarely taken by the po-
ets subsequent to Catullus. 1
V. Catullus has in some instances marred the elegance
1. In Statius, for instance, not a single example of the kind occurs in
upward of 450 lines; in Prudentius, not one in above 260 ; not one in
Ausonius, who has more than 2000 verses in this measure ; while Sido-
nius Apollinaris, in upward of 1200 Phalascians, has not above two that
can bo proved, and those are proper names. (Carey's Lot. Pros., p. 282.)
192 COMPOUND MEASURES.
and harmony of this measure, by introducing a heavy spon-
dee into the second place ; l as,
Te camp\o (\\\<es>\ivimus minore. (55, 3.)
Et f mult\ls \ang\uoribus peresus. (55, 31.)
VI. The same poet, in one line of a very irregular piece,
has a tribrach in the first place, a license, however, which
appears authorized by the difficulty of otherwise employing
a proper name.
Cameri]um mihi pessima puclla. (55, 10.)
VII. The name Hcndecasyllabic does not exclusively be-
long to Phalaecian verse, since there are other measures to
which it is equally applicable. For instance, the Sapphic
and a variety of the Alcaic not only contain the like num-
ber of syllables, but also in Iflte proportion of long to short,
so that the same words sometimes may, in different posi-
tions, become either a Phalaecian, a Sapphic, or an Alcaic.
Thu
(Phal.) Summum \ nee metu\as di\em nee \ optes.
(Sapph.) Nee di\em summ\um metu\ds nee \ optes.
(Ale.) Summum \ nee opt\es || nee metu\as diem.
SECTION XXXVII.
COMPOUND MEASURES.
1. DACTYLICO-IAMBIC.
I. This measure occurs in the eleventh epode of Horace
being used there alternately with the Iambic Senarius, and
consists of a Dactylic Trimeter Catalectic, followed by an
Iambic Dimeter Acatalectic ; as,
Scribere \ versicu\los \\ amor\c per\\cussum \ gram. ||
II. This measure properly falls under the head of Asy-
nartete* verses, that is, the component parts are not subject
1. This is made by some a separate measure, and called Pseudo-Pha-
leecian.
2. From a, privative, and avvapruu, " to join together ;" hence daw
oprj/rof , " not closely joined together"
COMPOUND MEASURES. 193
to the ordinary laws of prosody and versification, since the
last syllable of the first member of the verse may be either
long or short, just as if it were the final syllable of a separ-
ate line ; and, moreover, an hiatus may take place between
the two members of the measure. Thus,
Inachia furere, || silvis konorem decutit.
Arguit et latere |1 petitus imo spiritus.
Libera consilia j| nee contumelies graves
Fervidiore mero || arcana promdrat loco.
Vincere mollitia fl amor Lycisci me tenet.
These lines all occur in the eleventh epode of Horace. In
the first, second, and third, the short final syllables in fu-
rere, latere, and consilia are considered long, by virtue of
their position at the end of the dactylic trimeter catalectic ;
while in the fourth and fifth there is an hiatus between the
two members of the verse (mero arcana and mollitia amor).
2. IAMBICO-DACTYLIC, or ELEGIAMBIC.
I. This measure is directly the reverse of the preceding,
that is, it consists of an Iambic Dimeter, followed by a
Dactylic Trimeter Catalectic. Thus,
Horat. Tu vi\na Torq\\uato \ move \\ consuls \ pressa me\o.
Id. Redu\cet in \\ sedem \ vice. \\ Nunc et A\cha:meni\o.
Id. Levdr\e dlr\\ls pect\ora \\ solKci\tudin1\bus.
Id. Findunt ) Scamand\\ri flum\ina || lubncus \ et Simo\is.
II. This measure, like the preceding one, belongs prop-
erly to the class of Asynartete verses ; and hence, in the
second, third, and fourth lines just cited, the short final syl-
lables in vice, pectora, and Jlumina are considered long, by
virtue of their position at the end of the iambic dimeter.
HI. There are in all nine lines belonging to this species
of verse in Horace. It is not used in a system by itself,
but is placed alternately with the heroic hexameter in
Epode 13.
B
194 COMPOUND MEASURES.
3. ALCAIC HENDECASYLLABIC, or GREATEK ALCAIC.
I. This measure is compounded of an Iambic Monome-
ter Hypercatalectic and a pure Dactylic Dimeter Acatalectic ;
as,
Horat. Vides \ ut alt\a II stet riive \ candidum.
Claud. Venus \ revers\um || spernat A\dontdem.
II. But the first foot of the iambic portion is, of course,
alterable to a spondee ; as,
Horat. O ma\tre pul\chra || filid | pulchrior.
Claud. Victum \ fate\lur || Delos A\poltin<xn.
III. Horace much more frequently has a spondee than
an iambus in the first place, and Prudentius always a spon-
dee.
IV. The Alcaic is sometimes scanned with a choriambus
and an iambus in the latter colon or member ; as,
Vides | ut alt\a [| stet nive cdnd\idum.
Venus | revers\um |j spernat Jldo\ritdem.
V. Although Horace, who has made greater use of this
measure in his lyric compositions than any other, never
employed it except in conjunction with two other species
of verse (see Alcaic Stanza, below), other write.-s have
composed entire poems in it alone, as Prudentius, who h.ae
a long piece entirely consisting of unmixed Alcaics (Perif
teph., 14), and Claudian a shorter p'oduction (In Nup
Hon. Aug. et Mar.).
VI. Claudian's piece begins as follows :
Princeps corusco sidere pulchrior,
Partkis sagittis tendere certior,
Eques Gelonis imperiosior,
QuaK digna mentis laus erit ard.ua ?
Qua digna forma laus erit igneat ? &c.
THE ALCAIC STANZA.
I. This consists of four lines : the first two are Alcaic
COMPOUND MEASURES.
195
Hendecasyllabics, or Greater Alcaics ; the third is an lam-
bic Dimeter Hypermeter, or Alcaic Enncasyllabic ; and the
fourth is an Alcaic Decasyllabic, or Minor Alcaic.
II. The scheme of the Alcaic stanza is therefore as fol-
lows :
Lines 1
and 2.
Line 3.
1
2
3
4
5
^
v-'
_ s_x ^
V_x V.
Line 4. I
Vides ut alta stet nive candidum
Soracte nee jam sustineant onus
Sylvai laborantes geluque
Flumina constiterint acuto ?
III. According to the scheme which has just been gi^'en,
it will be seen that the first foot in each of the first two
lines may be either an iambus or a spondee. Horace,
however, as we have already remarked, gives a decided
preference to the spondaic commencement. Out of 634
Alcaic Hendecasyllabics extant in his works, 18 only have
an iambus in the first place ; that is, about one in thirty-five.
IV. Once only do we find in the same poet two lines in
succession beginning with an iambus ; as,
Metu deorum continuit? quibus
Pepercit arts? O utinam nova. (1, 35, 37.)
V. The fifth syllable in the first and second lines ought
always to be caesural ; as,
Non si trecen[is \ quotquot eunt dies.
VI. Horace, however, directly violates this rule twice. 1
1. Special rules for the structure of the Alcaic stanza may be found
in Ramsay, Lat. Pros., p. 220, seqq.
196 MISCELLANEOUS MEASURES.
Mentemquc lympTiatam Mareotico. (I, 37, 14.),
S&ectandus in certamme Marlio. (4, 14, 17.)
SECTION XXXVIII.
MEASURES NOT INCLUDED UNDER THE PREVIOUS
CLASSIFICATION.
1. CRETIC VERSE.
I. Cretic numbers belong in strictness to the trochaic,
and are nothing else b'ut a Catalectic ' Trbchaic Dipodia,
which consists of arsis, thesis, and arsis again.
II. Since this order is periodic, it is plain that the thesis
cannot be doubtful, but consists always and necessarily of
one short syllable only, but that each arsis may be resolv-
ed ; whence it comes to pass that both the first and fourth
paeon, and, moreover, even five short syllables, may be put
for the ere tic. Thus,
III. It must also be remarked, that when several cretic
feet are conjoined in one verse, no one coheres with an-
other in a periodic order ; and the last syllable of the last
foot, as every final syllable, cannot be resolved except in
systems in which, since the numbers are continued in one
unbroken tenour, the last foot of the verses, unless it is at
the same time the last foot of the whole system, is subject
to the same law as each intermediate foot.
IV. Cretics are much used by the Roman tragedians and
comedians, and with the same license as to prosody as the
rest of the metres. Whence, if they ever put a molossus
for a cretic, they do it in such a manner as to conceal the
faultiness of the measure under the ambiguity of a familiar
pronunciation. 1 Thus,
Plaut. Aut solu\tos smdt, \ qubs argcnt\o emerlt. 2
1. Bciitlcy ad Cic., Tusc., 3, 19. Id. ad Ter. Adelph., 4, 4, 2.
Hermann, D. E. M., 2, 19.
2. Captiv., 2, 1, 11.
MISCELLANEOUS MEASURES.. 197
Plain. Qua ne eject\ae c mdri dml>^a> sumus, \ te obsecro. 1
Id. Ot tuo | reclpids \ tecto, serv\esque nos.*
V. As they commonly use the tetrameter, they often
made the verse, divided into two equal parts, asynartete.'
Thus, Ennius in the Andromacha :
Quid pct.dm \ praesldi out \ exscqudr, \ quove nunc
Aut auxlK\o exsili \ out fuga \fretd simf
VI. Plautus has not only dimeters sometimes, but still
oftener catalectic tetrameters, and that, too, with the third
foot having the last syllable doubtful, and the fourth admit-
ting a resolution of the arsis. Thus, in the Trinummus
(2, 1, I7,seqq.):
Da mlhi hoc, \ mel meum, \ si me amas, \ si audts :
Ibi pendent\em ferit : \jarn dmplius \ drat.
Non satis id \ est mall, \ ni dmpUus't \ etiam, &c.
2. BACCHIAC VERSE. 4
I. The ancient metricians referred bacchiac numbers to
the paeonic kind, as having arisen from the contraction of
the second or fourth paeon.
II. Modern scholars, 4 however, on account of the iambic
anacrusis, have joined them with trochaic numbers, although
they are in reality spondaic with an iambic anacrusis.
III. The numbers of the amphibrach (~ ), if repeat-
ed, were with reason displeasing to the ancients, on account
of their too great weakness. Wherefore, to give them
strength, they changed the trochee into a spondee, and
thus produced the bacchius ( ).
IV. The Roman tragedians and comedians made great
use of bacchiac verses, joining also, for the most part, many
. of them together.
1. Rud., 1, 5, 15. 2. Ibid., 1, 5, 19.
3. Vid. page 192.
4. Herm., Elem. Doctr. Metr., 2, 22.
R2
i
198 MISCELLANEOUS MEASURES.
V. The legitimate measure of a bacchius in the middle
of verses is this, ; but in the end of verses this,
The freer prosody of the Latins, however, tolerates both a
long anacrusis and a dissyllabic one. And a dissyllabic
one was usually admitted by Plautus in the first and third
foot of tetrameters, that is, in the beginning of each mem-
ber, which is commonly composed of two feet ; sometimes
in the second and fourth foot also. Thus, in the Auhda-
ia (2, 1, 4, seq.) :
Quamquam hand fals\a sum nos \ odidsas \ hdberi.
Nam multum \ loquaces \ merito omnes \ habemur.
In the Menaschm. (5, 6, 6) :
Merito hoc no\blsflt qui \ quidem hue ven\erimus.
In the Amphitryon (2, 1, 15) :
Tun 1 me verb\ero dudes \ herum lu\difican.
VI. Tetrameters having a caesura at the end of the sec-
ond foot are a kind very much in use. That caesura, how-
ever, is often neglected. Plautus, who delighted very
much in this measure, sometimes inserted a dimeter in the
midst of tetrameters. Sometimes he even coupled two
verses by means of an elision ; as in the Amphitryon (2, 2) :
Satin parv\a res est \ volupta\tum in vita aij(que)
In <Bta\te agunda, \ prUe quam quod \ molestum est.
VII. These tetrameters sometimes appear to have clau-
sulae of an iambic dimeter catalectic ; as in Terence, An-
drian, (3, 2, 4) :
Quod jussi ei\dari btber\e et quantum im\peravi || date max \
ego hue | revert\or.
VIII. Bacchiac verses sometimes appear to be continued
in systems, so that a doubtful syllable has no place in the
end of ti e verses, and words may be divided between two
MISCELLANEOUS MEASURES.
199
verses. Thus, we have the following from Varro (rrepl
'EfaywyT/f, ap. Non., p. 336) :
Quemnam te css\e dicam- \ fera qui \ manu cor-)
ports ferv\idos font\ium aperis \ lacus san-\
gumis, te\que vita \ levas ferr\eo ense. \
IX. Catalectic bacchiacs, having the last foot an iambus,
are remarkable in Plautus. Thus, we have the following
dimeters in the Pcrsa (2, 28, 30) :
Perge, ut copperas,
Hoc, lend \ tlbl
Delude, ut \ lubet,
Herus dum hinc \ abcst.
Videsne, tit \ tuis
Dictls par\eo ?
3. SATURNIAN VERSE.
I. The Saturnian verse, which some rank among the asyn-
artete measures, appears to have been the only one used by
the most ancient Roman poets.
II. In it both inscriptions and poems were written. Liv-
lus Andronicus translated the Odyssey into this measure,
and in it Naevius wrote his poem on the First Punk; War.
III. The Saturnian has the following scheme :
1
2
3
4
5
6
7
^
^
V- 1
^
v-x
v-x
Dabunt j maltim \ Melell\l \\ Nasvi\b po\etas.
IV. But the rude poets of this early age both disregard-
ed the caesura often, and used every kind of resolution, re-
solving even the doubtful syllable in the end of the first
member. After the manner of ancient language, too, they
allowed spondees in all the places. Nay, the most ancient
of the poets seem to have thought it sufficient if their verses
only bore some sort of resemblance to these numbers. The
i
200 UNION OF DIFFERENT KINDS OF VERSE.
verses of the inscription composed by Naevius on himself,
and preserved by Aulus Gellius (1, 24), are tolerable
enough :
Mbrta\lis im\morta\lis \\fiere \ siforletfas,
Flerent \ diva | Cam<e\n(2 \\ N(zvi\um po\etam.
Itaque \ pdstquam est \ Orcl\nb || trad~t\tus thes\auro,
ObU\tl sunt | Roma \ loqui\\er Lat\ind \ lingua.
V. The last of the Romans who used this measure ap-
pears to have been Varro in his Satires. 1
SECTION XXXIX.
ON THE UNION OF DIFFERENT KINDS OF VERSE. 2
I. A poem receives the name of Monocolon, Dicolon, Tri-
colon, &c., according to the number of different species of
verse which it contains.
II. When a poem contains one species of verse only, it
is called Monocolon (from ^ovog, " alone" " single" and
KtoAov, " a limb? or " member''"). The Eclogues, Gcorgics,
and JEneid of Virgil, the Satires and Epistles of Horace,
the Metamorphoses of Ovid, are all examples of Carmina v .
Monocola, since they consist of hexameters alone. So also
the first ode of the first book of the Odes of Horace is a
Carmen Monocolon, since it is a system of choriambic As-
clepiadics, unbroken by any other species of verse ; and
so on.
III. When a poem contains two species of verse, it is
called Dicolon. The Fasti and Epistles of Ovid, the Ele-
gies of Tibullus and Propertius, which are composed of
dactylic hexameters and dactylic pentameters, placed alter-
nately, are Carmina Dicola. So also those odes which are
written in the Sapphic stanza : the third of the first book
of Horace, which contains two different species of chor-
iambic verse, and numerous others.
1. Herm., Doctr. El cm. Metr., 3, 9.
2. Ramsay's Lat. Pros., p. 230, seq.
UNION OF DIFFERENT KINDS OF VERSE. 201
IV. When a poem contains three different species of
?erse, it is called Tricolon. Of this we have an example
in the Alcaic stanza of Horace, which is composed of three
different kinds of verse.
V. Another series of terms has been "devised in order to
point out the intervals after which the first species of verse
used in any poem regularly recurs.
VI. When the first species of verse regularly recurs
after the second line, the poem receives the epithet of
Distrophon.
VII. Thus, poems composed in elegiac verse are called
Crmina Dicola Distropha. But a poem in the Sapphic
stanza, although Dicolon, is not Distrophon^ because the
first species does not recur regularly until after the fourth
line.
VIII. When the first species of verse recurs after the
third line, the poem receives the epithet Tristrophon ; after
the fourth line, Tetrastrophon ; and after the fifth line, Pen-
tastrophon.
IX. According to this system, a poem written in the
Sapphic stanza is termed Carmen Dicolon Tetrastrophon;
in the Alcaic stanza, Carmen Tricolon Tetrastrophon ; while
the Epithalamium of Julia and Manlius, in Catullus, is Di-
colon Pentastrophon.
X. This species of nomenclature, however, is by no
means perfect, as it does not point out the circumstances
under which the first species of verse is repeated. Thus, in
the Alcaic stanza, the first two lines are in the same species
of verse, the third and fourth are different from this and
from each other ; the grammarians, however, call a poem in
this stanza Tricolon Tetrastrophon. But if a stanza of four
lines is arranged in such a manner that the first line is one
species of verse, the second and the third different from the
first, but the same with each other, and the fourth different
from any of the preceding ; or if the first and second are
different from each other, the third and fourth different from
202 LATIN ACCENTUATION.
the two preceding, but the same with each other, then, in
either of these cases, the poem must be called Tricolon
Tetraslrophon. So a poem in the Sapphic stanza is called
Dicdlon Tetr a strop/ton ; but if a stanza were composed con-
taining one Sapphic line followed by three Adonics, the
poem would still bear the same appellation.
SECTION XL.
LATIN ACCENTUATION.'
In every word of more than one syllable, one is distin
guished by a peculiar stress or elevation of the voice, which
is called accent, of which those that precede or follow are
destitute. The syllable so distinguished is said to have
the acute accent, which is sometimes marked thus (') ; the
grave (^), which is seldom marked, is supposed to be placed
over those syllables which are pronounced without that
stress of the voice before spoken of. The circumflex, (*) or
(~), is supposed to be formed by a combination of the acute
and the grave, and hence is usually placed over contracted
syllables.
In modern languages, the accent, when it falls upon a
short syllable, has, in most cases, the same effect as if it
were long ; but in Latin and Greek, accent and quantity
were distinguished from each other ; and, by care and prac-
tice, this may be done in reading those languages.
Words of two syllables have in Latin the accent on the
first : if this is naturally long, as in Roma, mater, there is
no difficulty ; if short, as homo, pater, we must endeavour
to give the first syllable that percussion of the voice which
constitutes the accent, without lengthening the vowel, 01
yet doubling the following consonant.
The accent never falls on the last syllable of Latin words
except when words of the same letters, but different senses
are to be distinguished by it : e. g., pone, behind ; ergo, on
1. Zum.pt' s Latin Grammar, Kenrick's edition, p. 469, seqq.
LATIN ACCENTUATION. 203
account of; to distinguish them from p6ne (imper. of pono),
and rgo, therefore.
Words of three syllables or more ha.ve the accent on the
last syllable but one (penultima) when it is long, and on the
last but two (antepenultima\ when the penult is short ; as,
amasse, audisse, imperdtor, homines, Constantindpolis. No
accent is in Latin thrown farther back than the antepenult-
ima.
Some words, from their close connexion with those which
precede them, are pronounced as if they were the last syl-
lables of those -words ; e. g., prepositions when they are
placed after their cases, and ne, que, ve. They are called
enclitics; and the last syllable of the word to which they
are appended always has the acute accent ; as, pectoribus-
que.
As the system of accents in Latin is so simple, no ac-
centual marks are used except the circumflex, which is
placed over some contracted syllables, and over the abla-
tives of the first declension (musd, poetfi), to distinguish
them from the nominative. The Latins themselves do not
place the circumflex over the genitive ; and it is doubtful
if this form arose from contraction.
APPENDIX.
RELATIVE VALUE OF THE LATIN POETS AS METRICAL
AUTHORITIES.!
I. WE will first give a list of the Latin poets, with the
dates of their birth and death, where these particulars can
be ascertained, and then a statement of their relative value
as authorities in matters of a metrical nature.
Born. Flourished. Dud.
Lmus ANDRONICUS .B.C. . . 240 . . 220
NJEVIUS . . 235 . . 204
ENNIUS 239 . . . . 169
PLAUTUS 227 .. . . 184
C-ECILIUS . . 179 . . 168
PACUVICS 219 .. .. 130 (?)
TERENTIUS 194 . . . . 160
ATTIUS 170 .. 139 (alive 103)
LUCILIUS 149 . . 121 . . 103
AFRANIUS ...... . . 100 . .
LUCRETIUS 96 . . . . 52
CATULLUS ...... 87 .. . . 46
VIRGILIUS 70 .. . . 19
HORATIUS 65 .. . . 8
TIBULLUS 59 (!) . . . 20
PROPERTIES 54 (?) . . . 14
OVIDIUS 43 .. A.D. 17
Cornelius Gallus.
Pedo Albinovanus.
Publius Syrus.
Marcus Manilius.
Gralius Faliscus.
Aulus Sabinus.
CcEsar Germanicus.
1. Ramsay's Lat. Pros., p. vii., seqq.
S
206
APPENDIX.
Born. Flctibhed. Died.
PH.EDRUS .... A.D. . . 48 ..
SILIUS ITALICUS .... 25 . . . . 100
PERSIUS 34 . . . . 63
LUCANUS 38 . . . . 65
JUVEXALIS 40 .. . . 120
MARTIALIS 40 .. . . 101
PETROMUS ARBITER ... . . 61 . .
VALERIUS FLACCUS ... . . 69 . . 88
STATICS 61 .. . 96
SULPITIA . . 88 . .
Avianus . . 160 . .
Dionysius Cato .... . . 160
Serenus Sammonicus . . . . . . 212
Commodianus .... . . 265 . .
Nemesianus . . 280 . .
CALPURXIUS . . 284 . .
Porphyrius . . 326 . .
Juvencus . . 337 . .
ACSONICS 309 . . . . 394
Falconia . . 394 . .
Prudentius 348 . . 392 . .
CLAUDIANCS 365 (!) . 400 . .
Numatianus . . 416 . .
Paulinus 353 .. . . 431
Prosper Aquitanus ... . . . . 463
Sedulius . . 450 . .
Mamercus . . . . 474
Sidonius Apollinaris . . . 438 (!) . . . 484
Dracontius ' . . 456 . .
Martianus Capella ... . . 474 . .
Avitus . . 490 . .
Boethius 470 (?) . . . 524 (!)
Verrantius Fortunatus . . 530 . . . .
II. In the above list, some who precede Lucretius must be
thrown out of consideration altogether. We can attach no
importance, in controverted points, to these early bards, of
whom nothing has descended to us except short and mutila-
ted fragments. It is well known that these scraps are all
APPENDIX. 207
I
collected, at second hand, from the old grammarians and
others, who. cited them for the purpose of proving or illus-
trating particular points, which seldom have any reference
to quantity. The quotations, it would seem, were frequently
made from memory, and therefore subject to every kind of
change and corruption in the first instance, in addition to the
subsequent mutilations which they suffered in transcription,
arising from the strange and uncouth dialect in which many
of them were expressed.
III. The comic dramatists, Plautus and Terence, must also,
in strictness, be excluded. We are still comparatively igno-
rant of the laws by which their verse is regulated, notwith-
standing the labours of such men as Erasmus, Scaliger, Fa-
ber, Hare, Bentley, Hermann, and a host of others.
IV. Lucretius and Catullus, although inferior in genius to
none of their successors, scarcely occupy the first rank in
the estimation of the prosodian, because they may be said to
exhibit the language in its transition state, at a period when
much of the ancient roughness was removed, but when it had
not yet received the last brilliant polish.
V. Virgil, Horace, Tibullus, Propertius, and Ovid are our
great standards ; yet even among these slight differences may
be perceived. The first two never admit the double i in
the genitive oj^iouns of the second declension in ium and ius,
which is common in Ovid ; and the shortening of final o in
verbs, which was afterward extended to nouns and adverbs,
first begins to appear in the immediate successors of Virgil.
VI. Of the above, Propertius is the least valuable, on ac-
count of the small number and imperfections of the MSS.,
which have, in many passages, baffled the acuteness of the
most practised editors.
VII. Next follows a group of seven, all of little moment.
After these we come to Phaedrus, whose fables are now gen-
erally received as authentic ; but the text is derived from one
or two indifferent MSS., and is, consequently, in many places
confused and unsatisfactory.
VIII. With regard to those who come after, up to the end
of the first century, it may be laid down as a rule, that their
authority is admissible in points where we can obtain no
information from purer sources, but must never be placed
208
APPENDIX.
in competition with that of the great masters who went
before.
IX. All the successors of Statins must be regarded as of
little value for matters of prosody, except Calpurnius, Auso-
nius, and Claudian, the latter of whom is not more remark-
able for the purity of his diction than for the false glitter of
his style.
INDEX.
Fige
A final 66-68
A (i in numerals n. 3 on 68
A ending 1st member of com-
pound words 43
A in sing, increment of 1st
declension ...... 49
A in sing, increment of 3d de-
clension 50
A in plural increment of nouns 59
A. in increment of verbs . . 61
A the preposition in composi-
tion 40
A in Greek compounds n. on 43
A vocative of Greek nouns in
es and e ... n. 2 on 68
Ab, the preposition in compo-
sition 41
Abicit, abici 126
Abiegni as ab-yegni . . .120
Abiete as ab-yete . . . .119
Absridi and abscidi ... 33
Academia . . . n. 2 on 22
Acatalectic verses . . . .144
Accent., Latin 202
Acephalous verses .... 145
Achille .... n. 1 on 69
Ad, preposition in composition 41
Adicit (adjicit), adici, <fc. . 126
Adjectives ending in -acus,
-icus, cf-c., quantity of pe-
nult of 99
Adonic verse 146
Ador n. 1 on 57
Aeta n. on 25
^Egaeon 5-7
JEolic dialect, pronunciation
of n. 2 on 157
JEolic pentameter . . . .151
Afranius, metrical authority of 206
Agnltum 37, 40
Agri-cultura 45
Ai in genitive of 1st declension
explained . . . . n. on 17
Ai, quantity of 19
S2
Aio (trisyllabic and dissylla,
bic) 19
Aio explained . . n. 2 on 19
AL, nouns ending in, incre-
ment of 51 and n.
Alcaic verse (the greater) . .194
" (minor) . . .191
enneasyllabic . . .167
decasyllabic . . . .191
hendecasyllabic . . .194
stanza .... 194-5
Alcmanian tetrameter . . . 149
Alexandrea 21
Alius 18
Alterius .... n. 1 on 18
Ambe 23
Ambitus and ambitus
25, 35, n. on 40
AN, Greek accus. from nom.
a n. 2 on 85, 87
AN, Greek accus. from nom. as 85
Anap&st 162
Anapcestic verses . . . 1615
" dimeter . . . .163
" " catalectic . 164
" manometer . . . 163
Anas 90
Anceps, ancipes, <fc. n. 1 on 49
Antea, ^-c., formation of n. 1 on 66
Antithesis 125
Aperio, quantity of initial syl-
lable 127
Aph&resis 122
Apocope 124
Appendix, increment of
n. 1 on 56
AR, nouns in, increment of
56 and n.
Arabia, quantity of 1st syllable 128
Archaisms 121
Archilochian iambic dimeter
hypermetcr 167
Archilochian heptameter . . 190
Ariete as ar-yete . . . .119
210
INDEX.
Aristophanes, metre much used
by 174
As final 52, 89
As, Greek nouns in, increment
of 52
As, Greek nominative ... 90
As, " accusative plural . 90
Asclepiadic verse (minor) . . 182
AT in contracted preterites . 81
Atreides, Atrldes . n. 2 on 98
Attius, talus as metrical au-
thority 206
Ausonius, value as metrical
authority 208
Authority in quantity ... 15
B.
3 final 80
Bacchiac verse 197
Base in metre 141
Bebryx, increment of ... 56
Bibi, from bibo 32
Bldnum, 4-c 46 and n.
Bigae, 4-c n. 3 on 45
Bimus, 4-c- n. 6 on 45
Bobus and bubus . n. 1 on 48
Brachycatalectic verse . . . 144
Britto, increment of ... 57
C.
81
C final
Caecilius, value as metrical
authority 206
Ceesura 101
kinds of .... 102
" rules for .... 104
" power of . . 106-107
-Cams, trisyllabic .... 18
Callinus, inventor of elegy . 156
Calpurnius, value as metrical
authority 208
Catalectic verses .... 144
Catullus, value as metrical au-
thority 207
Causidicus 40
Cave, quantity of . n. 2 on 70
Cecldi and cecldi .... 31
Celtiber 50, 88
Censltum, censltor, 4-c.
n. 2 on 36
Choliambus 171
Chorea and chorea .... 20
Choriambic verses . . 17 188
Choriambic pentameter . . 183
" tetrameter . . . 183
" Asclepiadic te-
trameter 182
Choriambic tetrameter catalec-
tic 151
Choriambic trimeter acatalectic 180
" " cataicctic 181
" dimeter acatalectic 150
Ciris, value as metrical au-
thority n. on 7?
Citum, citum, and compounds 35
Claudian, value as metrical
authority 208
Cogmtum, 4-c 37, 40
Compendl-facio 45
Compound words, quantity of 39
" metres . . 192-196
Concitus 34-35
Conjicio 126
Conjugations, ancient form of
n. 4 on 23
Connubium 40
Conopeum and Conoplum
n. 1 on 22
Consonants . 13
Contracted syllables . . 23-24
Contraction of ea . . . .114
" " ei . . . . 115
" " eo . . . . 115
" " ia . . . . 118
" " ii . . . . 116
" " io . . . . 116
" " iu . . . . 116
" " oo .... 117
" " ee . . . . 117
Corcodilus and crocodilus . 126
Cretic verse 196
Cui (dissyllabic) .... 74
Cur n. 1 on 88
D.
D final 80
Dactylic verses . . . 145-161
" " casura in . 159
" versification, origin
of 157
Dactylic hexameter . . . .157
" " Meiurus .161
" Priapean .... 160
" pentameter . . .151
" " rules for 152
" " JGolic . 151
INDEX.
211
Fige
Dactylic Phi, .cecian pentameter 15C
" tetrameter a priore . 14S
" " aposteriore 149
" " Meiurus . 148
" " catalectic . 148
" trimeter .... 147
" with a base . 147
" ." hypercatalec-
tic 147
Dactylic trimeter catalectic . 146
" dimeter . . . . ]
Dactylico-iambic metre . .192
De, preposition in composition 40
Dedifrom do 32
Dejero 40
Deponent verbs . . n. 1 on 60
Derivatives, quantity of . . 37
" " " excep-
tions to . . . n. 3 on 37, 39
Desiderative verbs in -urio . 100
Di, preposition in composition 40
Diaeresis 121
Diana n. on 19
Diastole 128
Dlcare, dicere 38
Dicolon 200
Dionaeus, remarks on quantity
o/ae 25
Diphthongs 13
" quantity of . . 25
" erroneously rank-
ed as 26
Dipodia 141
Dirimo . . . . . n. on 40
Dirutum n. on 35
Dlsertus n. on 40
Distrophon 201
Diu n. 1 on 79
Do, increment of
61-62, and n. 1 on 61
Oonec n. 1 on 81
Double letters 13
Duo n. 1 on 47
E.
E final 68-72
E " of adverbs .... 70
E " " monosyllables . . 71
E in Greek neuter plural . . 70
E in increment of 3d declen-
sion 52-54
E in contracted gen. and dat.
of 5th declension . n. 1 on 69
Ihp
E in plural increment of nount 59
E in verbal increment . . . 62
E before R (in verbs) ... 62
E " RAM, RIM, RO . . . 63
E terminating 1st member of
compound words .... 44
E, preposition in composition 40
E long from Greek ei . . . 21
E ire imperative of verbs of 2d
conjugation . . n. 2 on 70
EA, Greek accusative from BUS 22
Ectasis 128
Ecthlipsis 112
Ego, quantity of final syllable
n. 1 on 78
Eheu n. 1 on 20
Ei of 5th declension . n. on 17
EL in Hebrew names . . 53, 83
Elegiambic measure . . .193
Elegiac distich 155
Elegy, origin o/(!tayof) . .156
Elision affinal vowels . . . 108
" of M final and its vow-
el sometimes neglected . .113
Elision, effect of in poetry . 108
" at the end of a verse 113
" of the vowel of ne 109
<( U
sometimes neglected . . .110
Elision of a short vmoel neg-
lected in Virgil . . . .112
Elision in compound words . 24
" of final s
n. 2 on 108, 107, 108
Elision, rules for . . . .113
Endo for in . . . n. 3 on 23
Ennchemimeris . . n. 1 on 102
Ennius, value as metrical au-
thority 206
Eos, Greek genitive from EVS 22
Epenthesis 123
Epichoriambic verse . . .183
" trimeter catalec-
tic 184
Epichoriambic tetrameter . . 187
ER, nouns in, increment of '53-54
Ergo n. 1 on 77
-erunt of preterite, shortening
of penult in . 63, n. 1 on 127
Srutum n. on 35
s final . . . 90, n. 2 on 90
Ss of Greek nouns .... 92
Ss Greek masc. singular . 53-54
212
E s from edo
INDEX.
n. 1 on 91
Es/rom sum, and compounds
n. 1 on 91
Ens, adjectives in .... 21
Eus, Greek proper names in . 26
" according to Latin de-
clension 26
Excitus 34-35
F.
Fac n. 3 on 81
Facio, verbs compounded with 44
Faliscan verse
n. 1 on 148
Fame, quantity of final syllable
n. 1 on 69
Far n. 1 on 88
Feet, metrical 131
" table of. 138-139
" simple .... 132-135
compound .
isochronous
135-138
142-143
Fere 71
Fidi from findo 32
FIGURES OF PROSODY 101-131
Final syllables 66
" " of a verse . 97
Fio, quantity of penult n. on 16
" verbs compounded of . . 44
Fluviorum as fluv-yorum . 120
Foris n. 1 on 94
Fortuitus 120
Fur n. 1 on 28
Fuvi, an old perfect . . . 123
G.
Galliambus 175
Gerunds in do, quantity of
final syllable . . n. 3 on 76
Glyconic verse 180
Gratis n. 1 on 94
H.
H a mere breathing . 14, 16, 27
H dropped between two vowels 24
Hannibal . . . n. 1 on 51, 82
Hemistich . 144
Hcphthcmimeris
n. 1 on 102
Heroic casural pause . . .159
" verse 158
Hexameters most ancient
n.. 2 on 157
Htxametcrs introduced into the
n by Ennius . . .156
I lie and hoc . . 82 and n. 1
Hipponactic tetrameter . . 174
" trimeter . . .172
Hodie 127
Horace, value as metrical au-
thority 207
Hymen 53
Hypercatalectic verses . . . 145
Hypcrmeter verses . . . . 143
I.
I as a consonant, and sounded
as initial y 119
I terminating 1st member of
compound words .... 45
I in singular increment of 3d
declension 54
I in plural increment of nouns 59
I in verbal increment ... 63
Ifinal ......... 72
I long from Greek ei . . . 21
I Greek' vocative of 3d declen-
sion 72
I final lengthened by arsis
n. 1 on 72
I in Zdfut. ind. and perf. subj.
65 and n. 1
I in Greek dative 3d declension 72
Iambic verses . . . 165-176
trimeter 169
" comic . . 172
" catalectic . 168
scazon .... 171
tetrameter ... 173
catalectic 174
' dimeter .... 167
" " acephalous 166
" " hypermeter 167
" " catalectic . 166
lambico-dactylic metre . 193
Iambus, origin of name . 133
Iber 50, 86
Ibi n. 2 on 73
Ibidem 46
Idem n. 2 on 45
IESUS 97
ILIS, adjectives in, quantity of
penult
In, preposition in composition
H
41
Increment of nouns .... 48
" " 1st and 2d
declension . . . 50 end n. 2
Increment of nouns, 3d declen-
sion . . 50-59
213
* "S c
Increment, plural of nouns 59-00
" ofrcrbs . . . 60-66
Indu .... 79 and n. 2, 23
Injicit (inicit) 126
Innuba 40
INCS, adjectives in . . . 99-100
lo and io .... 20 and n. 2
/owe verses .... 188-190
" a majors 188
*' a minore 189
Is m rcris of 4,th conjugation 93
Is in 2d fut. ind. and perf.
subj n. 2 on 94
IB final 92
Isochronous feet 142
IT in contracted preterites . . 81
Italia 128
Iter and Itiner . n. 1 on 48-49
ITUM, supines in . 36 and n. 2
las, genitive in . . . n. on 18
Ins, proper names in . n. on 18
Ivi in the preterite .... 64
Tx, increment of nouns in . 55
J.
. 14
. 27
27-28
. 126
J not a double consonant
J, vowel long before
Jacio, compounds of .
Jecur, increments of
n. 1 on 48-49
Tuvencus, value as metrical
authority . . . . n. on 47
Kdfj.avdpof
K.
L.
n. 2 on 28
L final 82
Labare, labi 38
"Lar, quantity of a in n. 1 on 89
Legare, legere 38
Leonine verses 168
Letters 13
Liquefacio 44
Liquids 13
Livius Andronicus, value as
metrical authority . . . 206
Logaccdic verses . . 190-192
Lucerna from luceo . n. on 38
liticilius, value as metrical au-
thority 206
Page
Lucretius, vclut as metrical
authority . . . . . . 207
Luorlfacio and lucrlfio . . 45
Ludi-magister 45
M.
M final short 84
M " elided 84
M " " (the reason)
n. 1 on 84
M " not elided .... 84
M " in the best writers eli-
ded except in compounds of
com (con) and circum . . 85
Maeotis, quantity of 1st sylla-
ble n. on 25
Maledicus 40
Mamercus, value as metrical
authority . . . . n. on 47
Maragdus (MupayJof) n. 2 on 28
Martianus CapeUa, value as
metrical authority n. 1 on 89
Mas, maris 51
Mastix, -igis, and mastix,
-ichis 55
Meiurus tetrameter . . . .148
Metathesis 125
Metre, definition of . . 140
" classification of 140-141
Mihi, sibi, tibi, <fc. . n. 2 on 73
Modo, quantity affinal syllable
n. 1 on 78
Molestus/rom moles
n. 3 on 37-38
Monocolon 200
Mutes 13
" and liquids . . . 29-30
N.
N find 85
Naevius, value as metrical au-
thority 206
Nasidienus as Nasld-yenus .- 120
Ne, elision in 109
Necesse, nefas, ec. n. 3 on 44
Nenu n. 2 on 79
Nequeo .... n. 3 on 44
Nequis, nequam, <fc. n. 3 on 44
Nescis . . . . n. 4 on 93, 100
O.
O final. . . .
in composition
74-79
. 47
214
INDEX.
O in Greek cases .... 56
O in " nominatives . . 75
O in singular increment of 3d
declension 56
. terminating 1st member of
compound Greek words . . 47
O terminating 1st member of
compound Latin words . . 47
i/i plural increment of nouns 59
O in the gerund . . n. 3 on 76
O in increment of verbs . . 66
O of adverbs ... n. 4 on 76
Ob, preposition in composition
41,42
Objicio (obicio) 126
Octo 78-79
Octonarius 173
Odyssey translated in Satur-
nian verse 199
Ohe 19 and n., 71
OLUS, diminutives in ... 100
Omitto 127
ON, Greek singular ... 87
Ox (QX), Greek gen. plural . 85
Ophites versus . . . n. on 157
OR, Greek nouns in ... 57
Orion .... 23 and n. 1, 57
Qsfinal 95
Ovid, value as metrical au-
thority 207
P.
Pacuvius, value as metrical
authority 206
Paeonius and Paeonius
n. 2 on 116
Palus 96 and n. 2
Par, paris, and compounds
51, n. 1 on 89
Paragoge 124
Parietibus as par-yetibus . 119
Paramiac 162, 164
Patefacio 44
Patronymics, quantity of pe-
nult of 98-99
Pejero 40
Peleldes and Pelides n. 2 on 98
Penes 92
Pentadius, de adventu veris .
n. 1 on 157
Pentastrophon 201
Penthemimeru . . n. 1 on 102
Penult ima ... 15
Pajt
Penultima of u* 'ds, quantity
of 98-100
Pepedi 31
Per, preposition in composition 41
Perfects, old, how formed . .123
Phaedrus, value as metrical
authority 207
Phalcccian dactylic pentameter
150, 191
Phal&cian hendecasyllabic . 11
P/terccratic verse . . . .181
Placare, placere .... 38
Platea 21
Plautus, value as metrical au-
thority 207
Pluvi, old perfect . . . .123
Poets, Latin, relative value as
metrical authorities . . . 205
Polypus ..I.... 97
Porsonian pause . . . .171
Position 27
Possls n. 4 on 93
Postea n. 1 on 67
Prae, preposition in composi-
tion before a vowel . 25 and n.
Prepositions in composition 40-43
Preterites, reduplicating . . 31
" of two syllables
32 and n.
" in ui anciently had
u long 124
Priamides 128
Priapean verse 160
Pro, the preposition in compo-
sition 42-43
Pronuba ........ 40
Propago 43
Propertius, value as metrical
authority 207
Propino 43
Prosody :B
Prosodial rules, origin of . 5
Prosthesis 122
Prudentius, value as metrical
authority ... n. 1 on 21
Puta n. 2 on 67
Putrefacio 44
Q.
Q, sound of 14
Quadriga; 45
Quando .... n. 1 on 47
Quandoque ...... 47
INDEX.
215
Page
QuandoquiJem 47
Quantity, definition of . . . 14
Quasi n. 1 on 72
Quatuor, <f-c 129
Quia, final syllable of n. 3 on 67
Quivis, cfc. . . . n. 1 on 46
Quomodo .... n. 3 on 78
Quotidianus . . . n. 4 on 46
Quotidie .... n. 4 on 46
R.
R final 87
Rare facio, tmesis in . . 125
RE in composition .... 41
RE, vocative from nominative
in ER 126
RE made long in religio, <c. 128
Recensitum . . . n. 2 on 36
Refert 41 and n.
Regula/nwn. rego . . n. on 38
Reperit, repulit, and retulit
128-129
Rutum and compounds . . 35
S.
S final 80
S " elided 108
Sal, salis . . . 51, n. 1 on 83
Sandix, -icis 56
Sanscrit analogies in Latin . 17,
18, 33, 45, n. 1 on 46, n. 2 on 83
Sapphic verses . . . 184-185
" " connexion be-
tween 3d and 4Jh line in . 185
Sapphic verses, elision in . .186
" greater 187
Sappho, inventress of 3Zolic
pentameter 151
Satin' n. 2 on 86
Saturnian verse . . . . .199
Sc, SP, SQ, ST, <$<:., initial . 28
Scamander and Camann
n. 2 on 28
Scanning 144
Scazon 171
Scidi from scindo .... 32
Scio and nescio . . f . 75
Se, preposition in composition 40
Sedare, sedere, sidere, sedes 38
Selibra 44
Semisopitus 40
Semivowels 13
Senarius ....... 169
Serpentini versus . . n. on 157
Sidon, increment of ... 57
Sis from sies . . . n. 3 on 93
SM initial 28
Smaragdus and, Maragdus
n. 2 on 28
Sol n. 2 on 83
Solvo, soluo . . . n. 3 on 35
Sotadic verse 188
Spondaic tetrameter . . . 149
Statlra 100
Statim n. 1 on 100
Statum, quantity of penult of 34
Steti and stiti 32
Sub, preposition in composition 41
Subjicio (subicio) . . . .126
Supellex, increment of
n. 1 on 48-49
Supines of two syllables 33 and n.
" " " short,
as citum, 6fC . . . . . 34
Supines, polysyllabic . . 35-37
" contracted n. 1 on 37
Syllable, definition of . . . 13
" length of .... 14
Synteresis 114
" in compounds of
semi ....... 119
Synalapha 108
" absorbs 2 syllables 108
" confounded by
Quintilian with syn&resis
and ecthlipsis . . n. 1 on 107
Synapheia 130
" with elision . .131
Syncope 122-123
Syphax, -acis ... 52 and n.
Systole 126
Syzygy 141
T.
T final 80
Tantidem ... n. 2 on 46
Tenuia as ten-via . . . .120
Tepefacio 44
Terence, value as metrical au-
thority 207
Tetrastrophon 201
Tiblcen .... n. 1 on 45
TIM. adverbs in 100
Tibullus, val-<-K <f metrical au-
thority 207
Tmesis . . 124, n. 1 on 125
216
r\rn;x,
Trans in composition ... 42
Trir.olon ...... 201
Triemimcris . . . n. 1 on 102
Tristr option 201
Trochaic verses . . . 176-179
" tetrameter catalectic 177
" dimeter .... 177
" " catalectic . 177
Tubicen .... n. 1 on 45
Tuli 32
U.
U sounded as W, and not form-
ing a diphthong with follow-
ing vowel 26
U sounded as V, and used as
a consonant 120
U in 1st member of compound
words 45
U in singular increment of
nouns 58-59
U in plural increment of nouns 59
U in increment of verbs . . 66
U final 79
U, contracted dative of &th de-
clension . . . n. 1 on 79
Ubi and compounds n. 2 on 73, 74
Ubicunque . . . n. 5 on 46
Ubique and ubivis . a 5 on 46
ULUS, diminutives in . . .100
Ulysse (vocative) . n. 1 on 69
URIO, quantity of verbs end-
ing in 39, 100
UEUS, future participle in . 66
Usfinal 95
Us, genitive of 4th declension,
how formed . . n. 3 on 96
Utf. and compounds . . . 73-74
Ui CM, supines in
36 and n. 2. n. 3 on 35
V.
V, use of . . . ... 14
Value of Latin poets as met-
rical authorities . . 205-208
Vemens, <fc- n. 2 on 118
Verse, union of different kinds
of 200
Verse, Latin, how denominated 144
Viden' . . 127, n. 2 on 86, 109
Vin' for visne 109
Virgil, value as metrical au-
thority 207
Volvo, voluo . . . n. 3 on 35
Vowels 13
Vowel before another vowel . 16
in
Greek words 20
Vowel before h 16
Vulte'ius .... n. 2 on 18
X.
X, double letter, equivalent to
what 13
X initial 28
Y.
Y final 72,80
Y terminating 1st member of
compound words .... 45
Y in increment of nouns . . 54
Yi as diphthong 26
Ysfinal 92
Yx, increment of nouns in . 55
Z.
Z, double letter, equivalent to
what . 13
Z initial 28
Z, reason for preceding rowel
remaining short before
n. 3 on 28
(0
BINDING SECT. NQV8 1978
PLEASE DO NOT REMOVE
CARDS OR SLIPS FROM THIS POCKET
UNIVERSITY OF TORONTO LIBRARY
LaL.Gr Anthon, Charles
A6285s A system of Latin
prosody and metre
