A STRATIGRAPHIC SUMMARY OF THE POTTERY FROM SITE 11
This brief report is a very selective preliminary presentation of the pottery of Site 11 on the Via Gabina excavation38. The intent of the report is to present examples of the pottery which characterize and date the important stratigraphic contexts which have been identified by the excavation supervisors.
Summary of Site History by Pottery Evidence.
A great deal of datable pottery is associated with the major periods of construction and occupation. The majority of this material dates from the third, second and first centuries B.C. (Period 1), from the first, second and early third centuries A.D. (Period 2) and from a short period of squatter occupation in the early third century A. D. The range of black-glaze, thinwalled ware and Italian terra sigillata indicates that occupation of the site was continuous from the beginning of Period 1 to the end of Period 2.
Occupation of the site begins at some time in the third century B.C. or earlier; a few unstratified sherds of the fourth century B.C. were found. The last phase of construction in the villa's early period (Phase 1C) is associated with the earliest stratified dating evidence for the history of the site.
There are two major construction phases in the second period of the site's history; the first has a terminus post quem at the end of the first century B.C. and the second early in the second century A.D.
The history of the site becomes more problematic after the 2B construction phase. Pottery types indicate continuous occupation of the site through the second and into the early third century A.D. There is, however, a lack of imported table wares specifically from the Severan period. This suggests a change in the function of the site from a combination of residential and industrial occupation to primarily industrial occupation around 180 A.D. or even slightly earlier.
Pottery evidence on the site after the early third century A.D. is very scant. This date certainly represents the end of any ordinary occupation of the site. Odd bits of later pottery, dating from the fourth to possibly as late as the eighth century A.D., and also in slightly greater quantity from the seventeenth and eighteenth centuries A.D., are indication only of a transient population in the area.
Stratigraphic Contexts Dated by Pottery.
The stratigraphic contexts which are of interest for this report are as follows:
1. Layers which are associated with the last phase of construction in Period I (Phase 1C). The pottery indicates that these layers date to the beginning of the second century B. C.
2. Layers which are associated with the first phase of the second major construction period on the site (Phase 2A). The pottery indicates a date for these layers in the last two decades of the first century B. C. (c. 20-1 B. C.).
3. Layers which are associated with the second phase of the second construction period on the site (Phase 2B). The pottery indicates a date in the early second century A.D.
The latest ceramic material in these three groups, for each of which there was an appreciable amount of pottery evidence, provides a terminus post quem for the associated construction.
Besides these important construction phases, three limited stratigraphic contexts were studied to clarify the sequence of events on the site in the second century A.D. These contexts are as follows:
4. Material from layers of Phase 2B occupation undisturbed by squatter clean-up. The pottery extends only to the last quarter of the second century A.D.
5. Layers which are associated with the arrival of squatters at the end of Phase 2C occupation. The material involved was dumped by the squatters when they cleaned the site. The pottery types are approximately contemporary with those associated with 2B occupation. The material therefore probably represents the continuing decay of the residential quarters of the villa.
6. Destruction layers of Phase 2C and squatter occupation. The pottery indicates a date in the late second or more probably the early third century A.D.
Context 1 (Phase 1 C Construction).
The earliest stratigraphically defined group of pottery pre-dates the last phase of construction and occupation in Period 1. The dominant fine ware in this context is Campana black-glaze. The characteristic form is the small rounded bowl, Lamboglia Form 27B; a typical example (P34/41) is illustrated in fig. 36, 139. Bases of Lamboglia 27 from this context are decorated with small elegant palmette or rosette stamps, as fig. 36, 2 and 3 (P273; P272) illustrate. This type is identical in fabric, dimensions and decoration with the class which Morel ascribes to a workshop of the third century B.C. near Rome40. A small Lamboglia 36 patera (P277), (fig. 36, 4) was found in the same context. The fabric of P277 is comparable to that of the examples of Lamboglia 27B from Site 11, but the color of the glaze is a shiny black, while the glaze of the 27B bowls tends to olive-green. Morel identifies a black-glaze patera with a profile very similar to the Via Gabina find as being from the same workshop which produced the Lamboglia 27 bowls with small stamp decoration; the type is possibly contemporary41.
The dominant coarse ware type in this context is a jar with a half-almond rim above a short neck set on the shoulder of the jar at a slight angle, in several variants, fig. 36, 5 and 6 (P843; P802). The Via Gabina examples from the 1C context generally have a narrow band of creamy buff slip over the rim, both inside and out. Examples of this jar type date as early as the fifth century B.C. and related forms continue to the end of the Republican period 42.
The examples of Lamboglia 27B, including the small elegant stamps of P38 and P39, and the coarse ware types are very comparable to pieces from the «Capitolium Fill» deposit at Cosa43. The Cosa deposit is dated from the early third century to the second quarter of the second century B.C.44. The pottery in the Republican Sutri deposit also offers many comparisons; the Sutri deposit should probably be dated to the mid-second century B.C.45.
A date c. 200 B. C. would seem likely for this Via Gabina context based on the Lamboglia 27B bowls, the Lamboglia 36 patera, the coarse ware types, and the find of a Republican uncia of c. 217-215 B.C. (76.25)46. However, a very fine sherd from an Italo-Megarian bowl, fig. 36, 7 (P298), in fabric and decorative motif («jewel-ribbed nymphaea caerulaeau» in Moevs' designation ) very similar to examples from Cosa47 suggests a date not earlier than the first quarter of the second century B.C.
Context 2 (Phase 2A Construction).
Layers which date Phase 2A construction are characterized by black-glaze and thin-walled wares and a minimal amount of early Italian terra sigillata.
Black glaze includes residual pieces; for instance, examples of Lamboglia 27B similar to those in Context 1 and examples of Sutri Form 7 in a fabric similar to that of the Lamboglia 27B pieces. An example of Sutri Form 7 is illustrated in fig. 36, 8 (P128). The type is a black-glaze bowl with ribbon band rim; P128 may be later than most of the examples illustrated in the Sutri report48. Variants of Lamboglia Form 28 also appear, including the thick-walled example illustrated fig. 36, 9 (P134). The glaze on this piece is a thin matte red-brown mottled to black and the paste is soft, fine, and light orange in color, which suggests a poor quality local imitation of black glaze. Variants of this form in a similar fabric are common in Phase 1C occupation contexts at Site 11; the type belongs to the first century B. C.49.
One of the most interesting pieces from this context is a moulded bowl in black glaze, fig. 36, 10 (P81). The piece may be an Eastern import. It recalls «Megarian» and related moulded bowls of the Hellenistic period; similar techniques and stamps continue to be used in the production of relief-moulded Arretine. The fabric of the piece is fine, hard and pale buff in color with a worn matte black glaze. Bowls of «Megarian» type with floral and vegetal motifs are generally dated from the late third into the mid first century B. C. The relatively thick walls of the Via Gabina piece suggest a date as late as the early empire50.
Only a very few early pieces of Arretine are represented in this context, including an example of the Augustan form Goudineau 17 (Haltern 1), fig. 36, 11 (P154) 51. Thin-walled beakers also appear. Most examples are unslipped and very thin-walled, as fig. 36, 12 (P265). The form of P265 belongs early in Moevs' sequence and is perhaps closest to Moevs Form III; she suggests a date from the mid second to the mid first century B.C. for this type52. One or two beakers have a rose-brown iridescent glaze, and P299 (fig. 36, 13) also has rouletted decoration. The complete profile would be most similar to Moevs Form LI; Moevs sets the early date for glazed beakers of this. form with rouletted decoration to the reign of Claudius53. At Site 11, the unslipped beakers seem to belong to the first century B.C., while the P299 type may be Augustan or slightly later.
Coarse ware pots related to types found in the I C context continue to appear. The characteristic form is the more evolved type shown in fig. 36, 14 (P808). At Cosa, Dyson shows related pieces in all his Republican contexts, which date from the third century B.C. to the mid first century B.C.54. A globular pot with sloping wall is a new type in this context, fig. 36, 15 (P164). P164 is related to Form 27 in the deposit of the first century A.D. at Sutri55.
The pottery range in this context is quite mixed. The fine wares date from the second and first centuries B.C., but some of the coarse ware and a few thin-walled pieces are comparable to the coarse ware of the first century A.D. deposit at Sutri. Since Dyson has demonstrated the long life of regional coarse ware types, a late date based on the suggested date for the Sutri coarse wares would be a mistake. The absence of a range of Italian terra sigillata, which becomes very common on the site in Period 2 occupation layers, suggests a terminus post quern for Phase 2A construction at some time within the last two decades of the first century B.C. and makes a later date very unlikely.
Context 3 (Phase 2B Construction).
The third stratigraphic context consists of the layers associated with the renovations of the site in Period 2 (Phase 2B). The latest datable fine wares in these layers include only a very little African Red Slip; P84 (fig. 36, 16) is the only example identifiable by form (Hayes Form 7B). Hayes dates the type from the early to mid second century A.D.56.
Late moulded Italian terra sigillata bowls of Dragendorff 27 type also appear; it is typical of the ware that the figure types of P140/P38 (fig. 37, 17) seem hardly related to each other spatially or iconographically57. The inverted arcades and lunate beading of P44 (fig. 36, 18) are common motifs in the ware58. The presence of the type is indicative of the late first and early second centuries A.D.59, just before African Red Slip fine wares take over the Italian market.
A great deal of plain Italian terra sigillata was present in these layers, much of it residual. The latest pieces are examples of Goudineau 43, as fig. 36, 19 (P562), the form dominant in the latest stage of the ware toward the end of the first century A.D.60; another piece which is quite late is illustrated in fig. 36, 20 (P126). The form is similar to Dragendorff 7, which Lamboglia dated 85-115 A.D.61. A great variety of thin-walled ware was also found in this context; no doubt most of it is residual.
In coarse ware, no forms were clearly dominant; several identifiable pieces are illustrated: a wide-mouthed jar with horizontal rim, fig. 37, 21 (P187)62; a «cucurbitula», fig. 37, 23 (P121)63 and a lid with asymmetrical knob, fig. 37, 22 (P91)64. These coarse ware types are characteristic of, but not necessarily restricted to, the late first century A.D.
The date of this context is determined mainly by the introduction of African Red Slip. A terminus post quem in the reign of Trajan (98-117 A.D.) suits the pottery evidence best65.
Context 4 (Phase 2B Occupation).
The fourth context to be considered is a limited group of occupation material from the residential area of the villa. African Red Slip of the second century A.D. (terra sigillata chiara A) becomes the dominant fine ware in this context. Examples of Hayes Form 8A, as fig. 37, 24 (P635), are numerous66. Other ARS includes a nearly complete example of a variant Hayes Form 6B, fig. 37, 25 (P634)67.
Among other fine wares, there is an example of a dish in Eastern Sigillata B2, fig. 37, 26 (P691). The form dates to the second half of the first century A.D.68. A large cup with applied «pignola» decoration and a mottled red to black glaze, fig. 37, 27 (P425), may date to the late first or early second century A. D.69.
There were few identifiable coarse ware types in this context; an example of the African Red Slip casserole Form 197 suggests the second century A.D.; fig. 37, 28 (P636)70. A complete coarse ware lid of the same form as fig. 37, 22 above (from the 2B construction context, see note 27) was found. A nearly complete one-handled pitcher, P639 (fig. 37, 29), is an example of a type very common at Ostia in layers dated to the reigns of Domitian and Trajan71.
The material in this context is of Trajanic to Antonine date (98-180 A.D.). The absence of African Red Slip fine ware bowls specifically of the Severan period may mean that the residential occupation of the villa ended before the last two decades of the second century A.D.
Context 5 (Squatter Clean-Up).
The fifth context is again a limited one; the material in this group was dumped into the piscina of the hortus and into a smaller basin in the industrial area by squatters as they cleaned the site of debris from the collapsing western section of the villa (the main residential quarters).
The context is characterized by a quantity of amphora discards, including examples of Dressel Form 2-4 (P66, et al.) and one example of Pelichet Form 47 (P240). The latter form is common on Site 11 in contexts of Phase 2B and later.
The latest identifiable type of fine ware is a dish in Eastern Sigillata B2, fig. 37, 30 (P118)72. The particular piece is dated c. 80-120 A.D. by J.W. Hayes, who saw it during a visit to the Via Gabina site in the summer of 1977. Thin-walled ware and Italian terra sigillata in the context are residual and do not provide clues to its latest date. Identifiable coarse ware types include several examples of the wide-mouthed jar with horizontal rim of the type illustrated in fig. 37, 21 (see note 62), which appears in contexts of the late first to the late second century A.D. on Site 11. All other material is roughly contemporary with that in the Phase 2B occupation context.
This may provide further evidence that the residential quarters were abandoned as early as the mid-second century A. D. ; or it may be that whenever the occupants left the site, they took with them everything of contemporary use and value.
Context 6 (Phase 2C and Squatter Occupation).
The sixth stratigraphic context comprises villa destruction layers in areas occupied by the squatters. The identifiable material is generally African Red Slip coarse ware; there is practically no fine ware, which suggests that the site served only an industrial function in Phase 2C.
There is a nearly complete example of Hayes Form 23B, a shallow casserole, fig. 37, 31 (P337). Hayes dates this form to the late second and early third centuries A.D.73. A large bowl, also nearly complete, similar in form to Hayes Form 50, in an African Red Slip coarse ware related to the ware of the 23B casserole, is illustrated in fig. 37, 32 (P341); it is evidently contemporary with Form 23 B74. Examples of Hayes Forms 197 and 196, in coarse ware fabrics which may be imitations of African fabrics, also occur in this context. These pieces suggest a date for the destruction context in the early third century A.D.
This brief presentation of the pottery related to the major stratigraphic contexts on Via Gabina Site 11 has necessarily passed over the wide range of pottery associated with the occupation in Periods 1 and 2. A great deal of pottery remains to be discussed in the final report; the pottery chosen for presentation in this paper clarifies the major periods and phases of the site's history, for which I suggest the following tentative dates:
Phase 1C: Construction, c. 200 (/175?) B.C. Occupation, c. 200 (/175?)-20/1 B.C.
Phase 2A: Construction, c. 20/1 B.C.;
Occupation, c. 20/1 B.C.-c. 100/120 A.D. Phase 2B: Construction, c. 100/120 A.D.;
Occupation, c. 100/120-c. 150/180 A.D. Phase 2C: Occupation, c. 150/180-c. 220/230 A.D.
Squatter Occupation and Final Abandonment: c. 220/230 A.D.
-- JOANN FREED
38. The pottery from Via Gabina Site 11 was excavated and processed over the years from 1976 to 1980. I owe much to the recording and notes of previous pottery supervisors on the site, particularly to M. Aylwin Cotton and to Paul Arthur. The drawings used in the present report were done by a number of participants in the Via Gabina excavations over those years.
Drafts of this report have been read by M. Aylwin Cotton and J. W. Hayes. I thank them for their suggestions and help.
39. For the type series, N. LAMBOGLIA, Per una classificazione preliminare della ceramica campana, in Atti del I Congresso Internazionale di Studi Liguri, (Bordighera, 1952), pp. 139-206.
40. Jean-Paul Morel discusses the regional production in Études de céramique campanienne I: L'Atelier des petites éstampilles, in MEFR 81 (1969), pp. 59-117. While Morel dated Lamboglia 27B from the last two decades of the third to the mid-second century B.C. in Céramique a vernis noir du Forum Romain et du Palatin (Paris, 1965), pp. 239-240, he now establishes the date for the type in the third century B.C., MEFR 81 (1969), p. 106.
41. The piece Morel illustrates has a similar central rosette, but its interior decoration is more complicated and includes rouletting, pl. 10, no. 127 and pp. 72-73 in MOREL, Céramique à vernis noir du Forum Romain et du Palatin, (Paris, 1965), also MEFR 81 (1969), fig. 15, p. 84.
42. This jar is related to Dyson's pot classes 2-4 and 11 in the deposit «Capitolium Fill» in Cosa: The Utilitarian Pottery, in MAAR XXXIII (1976), Rome. For fig. 5, DYSON, fig. 3, CF29-CF32 (Class 4); for fig. 6, DYSON, fig. 2, CF19-CF28 (Classes 2-3) and fig. 4, CF42 (Class 11). For dating, DYSON, pp. 24-28 and p. 55.
43. Examples of Lamboglia 27B occur in Deposit A («Capitolium Fill»), pls XXII-XXIV in DORIS MAE TAYLOR, Cosa: Black-Glaze Pottery, in MAAR XXV (1957), pp. 65-193 and pls. I-XLIV; cp. rosettes and palmettes (A 21 B) in upper left-hand corner of pl. XXIII. For the comparable coarse ware types, see note 42 above.
44. For the revised dating, M.T. MARABINI MOEVS, p. 171 in Italo-Megarian Ware at Cosa, in MAAR XXIV (1980), Roma, pp. 157-227.
45. G.C. Duncan, black-glaze Sutri Form 6 (Lamboglia 27) and coarse ware Forms 38 and 39 in Roman Republican Pottery from the Vicinity of Sutri, in PBSR 33 (1965), pp. 134-176. Duncan suggests the deposit may be as late as the first century B.C., p. 139.
46. This coin was identified by Michael Crawford in the spring of 1977; he suggested that the type would ordinarily be out of circulation by c. 200 B.C.
47. Especially pl. 1, fig. 6 and pl. 2, fig. 29 and p. 172 in M.T. MARABINI MOEVS, Italo-Megarian Ware at Cosa, in MAAR XXXIV (1980). Moevs dates fig. 6 to 175 B.C., p. 191; she dates other pieces with this motif 150-75 B.C.; passim.
48. The Sutri report tentatively dates the type to the mid-second century B.C. or later: fig. 3 and p. 144, DUNCAN, in PBSR 33 (1965).
49. LAMBOGLIA, Ceramica campana (1952), p. 152 (see note 39).
50. P81 is a derivative version of thin-walled moulded bowls in a similar fabric and glaze from Corinth, G. ROGER EDWARDS, Moulded Relief Ware, pp. 151-187 and pls. 65-83 in Corinth, Vol. VII, Pt. III, Corinthian Hellenistic Pottery, (Princeton, 1975). J. Schäfer suggests that vegetal motifs become important in the second half of the second century B.C. in Hellenistische Keramik aus Pergamon, (Berlin, 1968), p. 19. For a late group, pp. 222-223 and pl. IX B in W. SCHWABACHER, Hellenistische Reliefkeramik im Kerameikos, in AJA 45 (1941), pp. 182-228 and pls. I-IX.
51. For the type series, CHARLES GOUDINEAU, La ceramique aretine lisse, Fouilles de l'Ecole Francaise de Rome a Bolsena (Poggio Moscini) 1962-1967, in MEFR Supplement 6, vol. IV (Rome, 1968). Goudineau sets the early date for this form c. 12 B.C., p. 376; also J.W. HAYES, table, p. 449 in Roman Pottery from the South Stoa at Corinth, in Hesperia 42 (1973), pp. 416-470.
52. M.T. MARABINI MOEVS, pp. 58-59 in The Thin Walled Pottery from Cosa, in MAAR XXXII (1973). It is equivalent to Republican Sutri Form 20, especially A14 and A16 (fig. 6) and pp. 150 and 160, DUNCAN, in PBSR 33 (1965).
53. MOEVS, in MAAR XXXII (1973), p. 193 and n. 18. The Via Gabina piece is also very similar to the Form I thin-walled beaker from the first century A.D. deposit at Sutri, fig. 7 and pp. 54-55 in G.C. DUNCAN, A Roman Pottery near Sutri, in PBSR 32 (1964), pp. 38-88; a date in the third quarter of the first century A.D. is suggested for the deposit,
54. DYSON, in MAAR XXXIII (1976), fig. 8, FG21; fig. 13, 161V24 and fig. 20, V-D25. Form 36b in the Republican deposit at Sutri is very similar; DUNCAN, in PBSR 33 (1965), fig. 11. Form 26 in the Sutri deposit of the first century A.D. is a continuation of the same basic form; DUNCAN, in PBSR 32 (1964), fig. 11, pp. 60-61.
55. DUNCAN, in PBSR 32 (1964), fig. 12 and p. 61.
56. J.W. HAYES, Late Roman Pottery (London, 1972) = LRP, fig. 3 and pp. 31 and 33.
57. Compare examples published by M.L. LAVIZZARI-PEDRAZZINI, in La terra sigillata tardo-italica decorata a rilievo nella collezione Pisani-Dossi (Milan, 1972), especially pl. XIII, fig. 64.
58. LAVIZZARI-PEDRAZZINI (1972), pl. VII, fig. 25, pl. XXX, fig. 253 and pl. XXVIII, figs. 193, 194.
59. Present opinion on the dating of late mould-decorated Italian terra sigillata is summarized by C. Pucci, in Le Terme del Nuotatore, Studi Miscellanei 21, Rome 1973 (2 vols.) = Ostia III, pp. 315-321; the Ostia team believes that it appears in the reign of Domitian (81-95 A.D.), p. 660; Pucci concludes that the type may continue as late as 120 A.D.
60. At Ostia in construction levels of the Terme del Nuotatore dated to 80-90 A.D., Goudineau Form 43 is by far the most common form in Italian terra sigillata, Ostia III (1973), table on p. 315.
61. N. Lamboglia, p. 177 in a review of Christoph Simonett's Tessiner Gräberfelder, 1941, in RStLig IX (1945), pp. 163-194. The form may also be related to Goudineau Form 21, which appears in the same Domitianic layers at Ostia in which Goudineau Form 43 is common (see note 60).
62. The type first appears at Cosa in Deposit 2211, dated to the first half of the first century A.D., DYSON, in MAAR XXXIII (1976), figs. 42, 43 and p. 115. Similar types appear in the Late Shops deposit, which includes material from the late first to the early third centuries A.D., LS25-32, figs. 56-57, pp. 139-40. Form 23 in the first century A.D. deposit from Sutri is related, DUNCAN, in PBSR 32 (1964), fig. 10 and p. 59.
63. Emilio Rodriquez-Almeida discusses the type in «Sobre el use del anforisco, 'Cucurbitula'», pp. 813-818 in MEFR 86 (1974). It appears in a Flavian context at Albintimilium, p. 122, fig. 63, no. 39 in N. LAMBOGLIA, Gli Scavi di Albintimilium (Bordighera, 1950). At Ostia, the type appears in a level dated to the era of Domitian or Trajan; Ostia III (1973), pl. LV, fig. 463, and pp. 234, 447-448 and 655-656.
64. The coarse ware lid with asymmetrical knob and S-shaped rim profile appears at Cosa in a context of the first century B.C.; DYSON, in MAAR XXXIII (1976), fig. 35, PD88 and p. 87. It appears in the first century A.D. deposit at Sutri, DUNCAN, in PBSR 32 (1964), fig. 177, Form 48. Examples at Ostia were found in Domitianic contexts, pl. XXVIII, fig. 512, pp. 101 and 260 in Le Terme del Nuotatore, Studi Miscellanei 16, Rome, 1970 = Ostia II and pl. LIX, fig. 521 and pp. 253 and 654-655 in Ostia III (1973).
65. A reused tile with a stamp from the reign of Domitian (78.102) was built into the Phase 2B renovation. A number of brick-stamps of c. 115-126 A.D. (78.34, et al.) were also found on Site 11. These finds indicate construction on the site early in the reign of Hadrian, but their distribution suggests that they may be related to a minor modification.
66. HAYES, LRP (1972), fig. 4, esp. 8. 3, and pp. 33-35. Hayes originally dated Form 8A c. 80/90-160 + A.D. He now believes this form continues at least until 180 A.D., Supplement to Late Roman Pottery (London, 1980), p. 515.
67. For Form 6B, HAYES, LRP (1972), fig. 3 and pp. 29-31. Hayes dates this form «mid to end second century A.D.»: several of his examples appear in Hadrianic as well as Antonine excavation contexts.
68. Eastern Sigillata B2 refers to the later stage of this ware, in which the forms no longer resemble the forms of Arretine; the terminology is from Hayes, who dates ESB 2 c. 50/75-150 A.D., LRP (1972), p. 10. Hayes' type series will be published in a forthcoming supplement to Enciclopedia dell'Arte Antica. The ware first appears in contexts of 80-90 A.D. at Ostia; it is the most common of the eastern sigillatas in levels dated 90-155/160 A.D., and pieces in later layers are residual, Pucci, Ostia III (1973), p. 326.
69. The form is related to thin-walled hemispherical bowls. Two bowls published in Ostia III (1973) are similar in form and decoration: the piece shown on pl. LXXII, fig. 675 and p. 292, is without the heavy rim of the Via Gabina example, but has a similar «zone» demarcation. It is in a context dated to the third quarter of the first century A.D., pp. 654-655. Also Ostia III, pl. XX, fig. 90 and p. 80; the vase is unglazed and smaller than the Via Gabina example; the context is dated c. 155/160-190 A.D.
70. The profile is almost identical to fig. 9, pl. XIV, Ostia III, (1973), classed as «ceramica a patina cenerognola»; the context is dated 155/160-190 A.D., pp. 39 and 92.
71. Ostia III (1973), pl. XLIV, figs. 346-347; p. 203; table on p. 429 and p. 656.
72. See note 68, above.
73. HAYES, LRP (1972), fig. 7, 23.21 and pp. 45-48. The fabric is denoted A2 in the Ostia publications; Le Terme del Nuotatore, Studi Miscellanei 13, Rome, 1968 = Ostia I, p. 47; thin section, p. 38; p. 30 for fabric date.
74. The fabric is «a strisce» in the Ostia terminology, Ostia I (1968), p. 31; note that both P337 and P341 have the blackened rims which continue a long tradition in African coarse ware. Examples of the third and fourth century Form 50 listed by Hayes are in the very fine «terra sigillata chiara C» of Lamboglia, HAYES, LRP (1972), fig. 12 and pp. 69-73. Hayes suggests a date c. 200 A.D. for P341.