^{page 124 note 1} The Inscriptions of Roman Tripolitania, edited by Reynolds, J. M. and Perkins, J. B. Ward, Rome, 1951Google Scholar; cited hereafter as IRT.

^{page 124 note 2} Published: Brogan, Olwen and Oates, David, PBSR xxi (1953) 74 ffGoogle Scholar.

Oates, David, PBSR xxi (1953) 113 ffGoogle Scholar.

Oates, David, PBSR xxii (1954) 113 ffGoogle Scholar.

Goodchild, R. G., Quaderni iii (1954) 91 ffGoogle Scholar.

Forthcoming: R. G. Goodchild—Latino-Libyan texts.

E. Vergara-Caffarelli—a group of Latin funerary inscriptions found in 1952/3 in a catacomb in the garden of the Officers' Mess at Horns, and other texts found in 1955 during excavations at Sabratha and Lepcis, to be published in Reports and Monographs, iii.

^{page 124 note 3} Most of the new texts from the coastal cities were found by members of the expedition of the British School at Rome to Lepcis in 1953, and those from the interior by Mrs. Olwen Brogan, Mr. R. G. Goodchild, and Mr. David Oates. I am indebted to all of them for information; as I am also to Signor N. de Liberali and Mr. Donald Strong for photographs; to Miss C. R. Attwood, Mrs. G. U. S. Corbett, Mr. G. Clarke and Miss M. C. Tredennick for drawings; to Mrs. Brogan, Mr. Goodchild, and Mr. Strong for checking many points; to Mr. E. Birley for advice on the cohort mentioned in S. 5; and to Mr. John Morris for advice on IRT 522; and to Mr. Ward Perkins on S. 11.

^{page 124 note 4} See IRT, introduction.

^{page 125 note 1} A welcome addition to the small number of early inscriptions on stuccoed sandstone from Sabratha; see IRT p. 25.

^{page 125 note 2} Presumably from municipium or munus: if the former, the text would suggest that the Punic constitution of Sabratha was remodelled in the late I cent., as was that of Lepcis Magna (see IRT p. 80), both cities perhaps becoming municipia, presumably with Latin Rights.

At this point the text may have read patron]um mun[icipi.

^{page 125 note 1} See introduction, p. 124.

^{page 125 note 2} For Heracles at Lepcis Magna, where he was Genius Coloniae, see also IRT 286–9, all in Latin. For another Tripolitanian dedication to him in Greek, see IRT 848.

^{page 125 note 1} The column resembles a milestone; and its present location, which is almost certainly near its findspot, is suited to a stone marking the caput uiae of either of two stretches of the coast road (Lepcis Magna–Oea and Lepcis Magna–Tubactis) or of the Eastern Djebel road (see also IRT 930). If it is a milestone, it is the only known evidence of work on the Tripolitanian road-system between the reigns of Tiberius and Caracalla (see Goodchild, Roman Roads, P. 7).

^{page 125 note 2} From 84.

^{page 126 note 1} Marcus Aurelius, Caesar from 139 to 16.

^{page 126 note 2} Lib(ertus) or lib(rarius). Liberti are sometimes described also as uernae, see e.g. L' Ann. Ep. 1941, 161 uernae et liberto incomparabili: but in view of the order of words here and of the context a term descriptive of the man's function might be preferable. For slave librarii see CIL VIII, 12165–9.

^{page 126 note 3} Written A. X. M: there are identical sigla in CIL XIV, 4319 at Ostia (a dedication to the numen domus Augusti by Victor et Hedistus uern(ae) disp(ensatores) cum Traiano Aug(usti) lib(erto) A.X.M), except that in the Ostian text there is a bar above the X. The lettering of the Ostian text and the name Traianus suggest that it was cut in middle to late II cent., and is in fact roughly contemporary with the Lepcitanian one. (I am indebted to Prof. G. Barbieri for a photograph of the Ostian text.)

The explanation of the Ostian text given in CIL loc. cit.—a(nno) (decimo) m(agistr)o (sc. of a collegium in whose scola it is supposed to have been dedicated)—cannot stand now that another instance of the abbreviation has been found: nor does an earlier suggestion made by Dessau, in Eph. Ep. IX, 437Google Scholar—a(eris) (sc. stipendii) (decimi) mo(?) seem to help. It would be reasonable to suppose that the letters refer to a branch of the imperial financial service and perhaps specifically one concerned with financial administration arising in ports. For this the precise findspots of the two texts offer some confirmation—the Ostian text was found in the Piazzale delle Corporazioni, and the Lepcitanian text on the seashore near IRT 302, which mentions a seruus in the office of the IIII p. A. at Lepcis and suggests the proximity of that office. But I am unable to offer any satisfactory expansion of the letters. M is a standard abbreviation for modius.

^{page 126 note 1} See introduction, p. 124.

^{page 126 note 2} AVGG. The lettering is of the late II or early III cent.; the emperors might therefore be M. Aurelius and Verus, M. Aurelius and Commodus, Septimius Severus and Caracalla, Caracalla and Geta; but if Commodus or Geta were involved the second G of AVGG would probably have been erased. On grounds of lettering, Septimius Severus and Caracalla seem the most likely pair.

^{page 126 note 3} Otherwise unknown. Presumably Mon(tanorum).

^{page 126 note 4} It seems possible that the inscription records the erection of a monument (see fec[?erunt] in l. 4) by two men, each of whom had followed careers in the imperial services. The first, whose name is entirely lost, would appear to have reached a rank such as procurator Augustorum. The second, perhaps named Caecilia[nus], had passed through the militia equestris and possibly proceeded to an office, presumably as procurator, in a province whose name ends -atia (e.g. Galatia or Dalmatia).

^{page 127 note 1} Material and lettering closely resemble those of IRT 651, but there appear to be slight differences in the moulding of the two fragments.

^{page 127 note 2} See introduction, p. 124.

^{page 127 note 3} PRAEFF; the upper part only of the sixth letter survives, but praefe[ctorum … is improbable in view of the likely line-length calculated from l. 2. Since this uicarius is u.p. not u.c. the date must be early in the IV cent., see IRT introduction, p. 8.

^{page 127 note 4} Innocentiam or integritatem would fit the space available; but a phrase such as ob insignia meritorum …, running on to the next line, is also possible.

^{page 127 note 5} For P it would be possible to read B or R; for A, N or V.

^{page 127 note 1} See introduction, p. 124.

^{page 127 note 2} Perhaps amat]ori patriae, see also IRT 95, 275, 347, 567, 603, and Neo-Punic 32 ( = IRT 318).

^{page 128 note 1} A rare order for the words of this formula; see also, however, CIL VIII, 3828.

^{page 128 note 2} Both cognomina appear to be of Libyan origin. For Stiddin see also IRT 219, 236, 875 and, probably, 892, and S. 16, below; Ladas is not otherwise known. The initial letters of the name read as Stiddin are cut as S. This letter-form has been found in a number of Latino–Libyan inscriptions in Tripolitania and has sometimes been read as F, more recently as Z (see Goodchild, R. G., ‘The Latino–Libyan Inscriptions of Tripolitania,’ The Antiquaries Journal, xxx (1950) 137Google Scholar). Here S is clearly distinguished from F; and since there can be little doubt that the name intended is Stiddin (as also in IRT 892, where the initial letters are also written as S) it would seem reasonable to suppose that the form regularly represents ST in ligature.

^{page 128 note 1} I am indebted to the Dottoressa Giulia Fogolari, who obtained the photographs for me. On the museum-label the urn is dated in the I cent. The style of lettering suggests, in a Lepcitanian context, the second half of the II cent. or the very early III cent.

^{page 128 note 2} Presumably these letters are the initials of the dead man, which appear to indicate that he was a Roman citizen C(aius) L( ? ) M( ? ) A( ? ). This provides another argument in favour of a date later than the I cent, for the burial, since the urn is that of a man of modest means and station—not such as would be likely to have obtained Roman citizenship before this was extended to the generality of Lepcitanians by Trajan (see IRT p. 81).

^{page 128 note 1} For a description of this church, which seems to have been built in the mid-VI cent., see Perkins, J. B. Ward and Goodchild, R. G., ‘The Christian Antiquities of Tripolitania’, Archaeologia, xcv (1953) 29 ffGoogle Scholar. The type of tomb concerned here is in the second and later group distinguished by these authors.

^{page 129 note 1} The photograph was taken when the surface was in a much better condition than it is now. The text is read from the photograph.

^{page 129 note 2} For this church see S. 10, n. 1 above.

^{page 129 note 1} For this church see S. 10, n. 1 above.

^{page 129 note 2} Κ(ὐρι)ε βο(ήθει); see also IRT 829, from Church III.

^{page 129 note 1} See introduction, p. 124. The text given in IRT was read from a drawing and is wrong.

^{page 129 note 2} L. Ceionius Commodus, Caesar from 136 to 138.

^{page 130 note 1} The block described here was seen and associated with that published in IRT by Romanelli (Leptis Magna, 132), but his reference was wrongly interpreted in the footnote to IRT 489.

^{page 130 note 2} There is a long space after IMP. Since the spacing of the other words is uneven, an interval here, with a figure cut upon the next stone, is not impossible; or space may have been left for a figure to be added when the cutter had ascertained the correct number of imperial salutations (see IRT 914). If no figure was intended because the emperor concerned had only received the initial salutation, the subject might be Gaius (March 38–March 39), Trajan (Dec. 97–Dec. 98), Hadrian (Dec. 117–Dec. 118) or Antoninus Pius (Feb. 139–Feb. 140)—lettering and material make a later date improbable: the balance is in favour of a I cent, emperor, since, in the II cent., lists of imperial titles do not normally include imperial salutations until the second has been received.

^{page 130 note 1} See introduction, p. 124.

^{page 130 note 2} One line is probably missing above the die.

^{page 130 note 3} Not otherwise known. The date should be before the creation of prouincia Tripolitana, i.e. before c. 312 at the latest, when the proconsul of Africa ceased to be responsible for this region. But letter-style, tone, and relation to IRT 610 (see below) suggest rather a IV cent. date. It is possible that l. 2 should be restored [ex] proconsule (as in IRT 526, ll. 2, 3); the position in which Claudius A] … earned the gratitude of Lepcis for his qualities as an administrator might then have been analogous to that of Decimius Hesperius (see IRT 526 and PW VIII, 1249, 1).

^{page 131 note 1} Flauius Victorianus was comes Africae between 375 and 378, see CIL VIII, 10937. The activities for which he is honoured in this inscription are clearly part of the aftermath of the Austurian invasion of Tripolitania c. 363–5, see Amm. Marc. XXVIII 6 and especially IRT 475, 526, 571, of approximately the same date as this text.

^{page 131 note 2} It would be possible to read com(iti) et or comit(i) at the beginning of the line. The word which follows is difficult to read—dioecesis would give sense, but the first four letters would have to be very cramped to fit into the space available.

^{page 131 note 1} See introduction, p. 124.

^{page 131 note 2} The new fragments show that the text relates to a praeses prouinciae Tripolitanae (unidentified) and is therefore of the IV cent. This text is closely related to IRT 522 (see above), which is cut on the left-hand side of the same base; and from the markedly superior quality of its lettering, should be the earlier of the two.

^{page 132 note 1} Presumably Lepcis. Fronto has not been identified; but appears to be the earliest known Senator of Lepcitanian origin.

^{page 132 note 2} Perhaps hunc diuus Traianus lato clauo donauit. If the donor had been Domitian, it is unlikely that his name would have been given; if Vespasian, we should have to suppose a very considerable gap in Fronto's early career, since the space vacant before diui Traiani Parthici in l. 4 cannot be long enough to hold the record of much activity.

^{page 133 note 1} The name of the testator, e.g. Vitalis.

^{page 133 note 2} Another block containing, e.g., the words uxor (coniunx) et might intervene before heres.

^{page 133 note 3} Presumably the description of the building. The site has not been excavated and its precise character is obscure: walls enclosing an open courtyard (cf. the area? strata of the text) are the only certain features. A term with the sense of peristylia might be suggested, although this is in fact too long for the space available.

^{page 133 note 4} An ivy leaf after Ɖ indicates that the figure is complete; but the beam-holes seem to show that the block containing de suo did not follow immediately.

^{page 133 note 1} Presumably a reference to work undertaken at the expense of Severus and Julia Domna but finished by Caracalla, see also IRT 427, 428 from the Severan basilica. A reconstruction might be sought on the lines of templum et substructio a parentibus coepta.

^{page 133 note 2} Perhaps … c]onco[rdia …

^{page 133 note 1} The upper part only of the letters survives: the dotted letters could in each case be B, P, or R.

^{page 135 note 1} This inscription is to be published more fully by ProfessorVergara, E. in Reports and Monographs, IIIGoogle Scholar.

^{page 135 note 2} DI ligatured.

^{page 135 note 3} DE ligatured.

^{page 135 note 4} Perhaps [fili]u[s].

^{page 135 note 5} If the inscription is, as the letter-forms suggest, of the III cent., the presence of this family of soldiers in the interior of Tripolitania might be interpreted as the result of veteran settlements of the type presumed at this time by Goodchild, R. G. and Perkins, J. B. Ward, ‘The Limes Tripolitanus in the Light of Recent Discoveries,’ JRS, xxxix (1949) 93Google Scholar.

^{page 136 note 1} The type of cross with forked terminals ending in a small circle is certainly not earlier than the VI cent.

^{page 136 note 2} See also IRT 861, and for a description of the church on this site, Perkins, J. B. Ward and Goodchild, R. G., ‘The Christian Antiquities of Tripolitania’, Archaeologia, xcv (1953) 35 ffGoogle Scholar. and 58. The church seems to have been built before the Byzantine reconquest of Africa; but a baptistery of Byzantine type was added later. The inscription, which is certainly later than the Byzantine reconquest, affords valuable evidence of contact between the coast and the interior in this period. It is the only Greek inscription so far found in the interior of Tripolitania.

^{page 136 note 3} The formula Κὐριε βοήθει is found also at Lepcis, see IRT 829 and S. 12 above.

^{page 136 note 4} Perhaps another name.

^{page 136 note 5} At this point the arrangement of letters in straight lines fails; in the present state of preservation of the letters it is difficult, if not impossible, to decide which are to be taken together.

^{page 137 note 1} Of the two sites, one is a roughly rectangular ditched gasr with the foundation of a gate-tower projecting from its S.W. side: the surviving masonry consists of large blocks carefully cut and laid, but may have been a foundation of re-used material carrying rubble masonry: it is surrounded on all sides by rubble huts forming a small village of the type found at Gasr Hamed (see Oates, David, ‘Ancient Settlement in the Tripolitanian Gebel, II’, PBSR, xxii (1954) 96Google Scholar). The other site, 200 m. from the first, survives only as a mound, c. 18 × 14, surrounded by a ditch 25 m. wide, which probably represents the remains of a rubble gasr. Olive-press uprights and a stub of concrete walling beyond the S.W. lip of the ditch suggest that there was an earlier olive farm here. Both gsur would, on Oates' dating of these structures, at the earliest belong to the end of the IV cent. For the spread of Christianity into the interior of Tripolitania in the late IV and V cents., and for the incidence and significance of the use of the monogram cross, see Perkins, J. B. Ward and Goodchild, R. G., ‘The Christian Antiquities of Tripolitania’, Archaeologia, xcv (1953) 72 ffGoogle Scholar.

^{page 137 note 2} Perhaps [Laus] Deo.

^{page 137 note 3} A Libyan name, see also S. 8 n. 2 above.

^{page 137 note 4} Possibly … sa]ecolo[rum].

^{page 137 note 1} The gasr is constructed for the most part of undressed stones and is clearly later in date than the inscription. It is probable that a mausoleum was demolished and robbed to provide material for the gasr, and this mausoleum might be associated with the olive farms of which a number of traces survive in the area.

^{page 137 note 2} For the formula filio quieto see also IRT 692 at Lepcis.

^{page 137 note 1} The stone is said by Dr. C. Chiesa to come either from the Garian massif or from the area of Beni Ulid in the Orfella. Its findspot was probably not its original position—there were a few large stones near-by, but no sign of other dressed blocks.

^{page 137 note 2} For this name cf. Baline, at Cirta, CIL VIII, 7827Google Scholar.

^{page 137 note 3} TR ligatured. Structori presumably refers to the builder, C. Caecilius Lupus.

^{page 138 note 1} The gasr is likely to be, at earliest, of the late IV cent. Since a stone inscribed with a chi-rho monogram was found on the site, the inhabitants may well have been Christian. See also note 2 below.

^{page 138 note 2} At the beginning of the line Oates read humilitas, which is epigraphically possible, but the traces of letters also seem compatible with […] diu; at the end, the letter after in could be C—which suggests perhaps militabo in Christo. For the conjunction of militia and labor in a Christian context cf. Diehl 892, with note on l. 4 and quotation from Vulg. II Tim. 2. 3: labora sicut bonus miles Christi.

^{page 138 note 3} AE ligatured.

^{page 138 note 4} The reading at the end of l. 4 and the beginning of l. 5 is obscure, since the stone is both worn and pitted here. It would be just possible to read E for A and IT for E, making fuerit, but this is very uncertain.

^{page 138 note 5} For EIVS it would be easier to read CIVS, presumably the end of a name, Donatina[. ]cius, for which, however, I have not found a parallel.

In patrem, TR are apparently in ligature, but the T cannot be regarded as certain.

At the end of the line there is barely room for exuperan[t] and the traces of the N are very uncertain: exupera[t] is quite possible.

^{page 138 note 1} For a brief reference to this site see Goodchild, R. G., ‘The Limes Tripolitanus II’, JRS xl (1950) 33Google Scholar.

^{page 138 note 2} S =ST as in S. 8 above. For the probable meaning of Mynstysth = monument see IRT 906, n. 1, below.

^{page 138 note 3} The following seem to be names:

Μ[…] VСΗΑΝ(cf. Masauchan in IRT 906 below).

ΙΥḶVḶ (see also in S. 24 below).

ΜẠЅІRАΝVṗ(for the suffix -uy, see S. 24, n. 3 below).

^{page 138 note 4} The letters in this line appear to be much more widely spaced than in the remainder of the inscription.

^{page 139 note 1} MA ligatured; the first C could be read as G; K may be Y.

^{page 139 note 1} The primary inscription of this mausoleum (IRT 899, see above) is still in situ above its door. The new text, if it is to be associated with this, must be considered as secondary; but it is quite possible that it belonged to a mausoleum now destroyed.

^{page 139 note 2} Presumably the name of the family concerned. All the names in this text, in so far as they can be read, are otherwise unknown.

^{page 139 note 3} Thelower serif of a letter, probably an upright, is visible to the left of A. There is just room here for a P, giving parentaliorum (for this form of the genitive of parentalia see ILS 8370); the sense seems to require a noun descriptive of the type of victims, but I cannot find an instance of parentalia used in this way.

^{page 139 note 4} The unusually large number of victims is noteworthy, and demonstrates the great wealth of the settlement at Ghirza (see also S. 23 below). Since the inscription is unlikely to be earlier than the middle of the IV cent, it also demonstrates the vigour of the pagan tradition there.

^{page 140 note 1} This fragment from the Ghirza settlement is the only known monument found in the pre-desert area of Tripolitania which is not of local stone; and its significance lies in this fact. At least one settlement in this area was in close touch with the coast, sufficiently wealthy to afford to import a luxury therefrom, and sufficiently sophisticated to do so.

^{page 140 note 1} The lettering of this inscription is very much better than that of the inscriptions on other mausolea at Ghirza, and strongly suggests that this is the earliest tomb in the group nearest the habitations: this accords with its position in relation to the other mausolea.

^{page 140 note 2} M. is presumably for Marchius which appears in full in IRT 898. For the name Nimira, see also IRT 898: for Fydel, see IRT 900 below.

^{page 141 note 1} Letters printed in italics are those published in CIL, loc. cit., as read on the stone now in Istanbul.

^{page 141 note 2} The names are Libyan; for Fydel see also IRT 899 above, at Ghirza. For Thesylgum it would be possible to read Thefylgum, neither form being otherwise attested. M. and F. are presumably abbreviations for Marchius, as in IRT 898 and 899, and for Flauia.

^{page 141 note 3} The reference to folles shows that the date cannot be earlier than the reign of Diocletian. The missing word may perhaps be n[onaginta].

^{page 141 note 4} The text is closely related in form and content to IRT 898, and the ornament on the two stones is also very similar; but, since the workmanship of this one is of rather better quality, it is likely to be the earlier of the two. This accords also with the position of the mausoleum in the group.

^{page 141 note 1} Presumably for systan, with S = ST as in S. 8 above.

^{page 141 note 2} Fely probably = made, see IRT 906, n. 1 below.

^{page 141 note 3} The following are certainly names:

MASAVCHANVY

IYLLVL BVNE M[. ]MILTH[?.]?VY (it is not clear where this word ends)

ANOBAL BVNE M[… 1 or 2 …]CHAN

Perhaps also SYSTAN.

Bune = son of, see IRT 906, n. 3 below. Some of these names recur in IRT 906—Iyllul (Iylul), Anobal (Annobal), and Masauchan without the final letters -uy which appear here. Vy, apparently as a suffix to names, appears in several Libyan inscriptions, cf. Thychlethuy in IRT 906 below and Masiranuy in S. 20 above; it is possible that it may indicate feminine gender.

Presumably the persons recorded here are of the same families as those in the closely associated IRT 906, the principals here being apparently of the older generation, as is also suggested by the rather better lettering and more ambitious ornament on this stone. It seems possible to construct three family trees, but the relation between the three is obscure:

^{page 142 note 1} S here and in l. 3 should presumably be interpreted as ST ligatured as in S. 8 above. For the probability that felu mynstysth = ‘constructed the mausoleum’, see Goodchild, R. G., ‘The Latino-Libyan Inscriptions of Tripolitania’, The Antiquaries Journal, xxx (1950) 140–141Google Scholar.

^{page 142 note 2} Clearly a sum of money is given here—2100 denarii—presumably with reference to the cost of the mausoleum, as at Ghirza (IRT, 898, 900). The sign for 1000—a∞—recurs in l. 5.

^{page 142 note 3} The following are certainly names:

THANVBDAV BYNE NASIF

MASAVCHAN BYN IYLVL

THYCHLETHVY BVNY ANNOBAL

Perhaps also CHVLAM.

Byne, byn, and buny are presumably forms of the Semitic ben = son of, see Goodchild, loc. cit. 139. If -uy indicates feminine gender, buny presumably = daughter of.

The majority of the names are Libyan: Nasif appears also in IRT 886, 899, Iylul (lyllul) and Masauchan (with a suffix -uy) in the closely related S. 24 above (see S. 24, n. 3), Chulam in IRT 898. Annobal (Anobal) is a common NeoPunic name.

For the relation of this inscription to S. 24 see S. 24, n. 3 above.

Both in the nomenclature and in the language this inscription attests an amalgam of the native Libyan with elements of Punic and Roman cultures.