LOCAL POTTERY OF THE PERSIAN PERIOD. Ephraim Stern

Transcription

1 CHAPTER TWO LOCAL POTTERY OF THE PERSIAN PERIOD Ephraim Stern INTRODUCTION A surprisingly meager quantity of local pottery of the Persian period was discovered in Areas A and C of Tel Dor. Still more surprising was the lack of diversity in the ceramic repertoire, even if we take into account the relatively small extent of the excavated areas. The low reliability of conclusions drawn from such sparse finds can be demonstrated by our excavations. Almost all the types absent from Areas A and C were encountered in the other excavated areas: B, D, E, F, and G. It therefore appears that the ceramic repertoire of Areas A and Cis a result of chance, or perhaps of the limited size of the excavated area or its special character of small units of residential buildings. Generally speaking, none of the pottery vessels found in these areas can be considered new types; all belong to types dealt with by the present writer in 198 in his typological discussion of the pottery of the Persian period (Stern 198). For this reason the present discussion will follow the outlines of the previous study. Since the publication of the above-mentioned study, many more large-scale excavations have been conducted and their material published, including a number from the area of Dor. Since this material did not appear in the 198 publication, we have taken the opportunity to include their results here and to discuss their contribution to pottery typology in general. The newly excavated sites in the vicinity oftel Dorwhose finds will be discussed here are Tell Keisan, Yoqne'am, Tell Qiri, Tel Michal, Qadum, and Shechem. It can be stated with confidence that our study of 198 is still valid and that, with the exception of a number of new types unknown at that time, all our conclusions regarding typology and chronology have proved correct. Moreover, the new finds from recent excavations, including those of Dor, have in fact strengthened our previous conclusions. Our observations as to the regional character of several of the ceramic types have also withstood the test oftime, and some of them have been finally proven. We shall mention here only two examples. The Phoenician jars oftype F, which were attributed to the Sharon and Carmel coast, were uncovered at Tel Michal in a kiln in the process of manufacture (Stern 198: , Fig. 14; Herzog 1989: , Fig. 8.14, Pis. 30, 3: 1-). The other example is that of jugs of Type H which were attributed to the same region (Stern 198: 11, Fig. 1 ); though these were found in large concentrations at Qadum near Samaria, they also occur at Dor and Tel Michal, but are completely absent from other regions. The following typological discussion differs from our previous summary in two respects. The imported East Greek and Attic vessels have been omitted and are dealt with in separate chapters (see below, Chapters 3-4). We have therefore also excluded all the undecorated vessels of Greek origin, such as wine jars and amphorae, which were dealt with in the 198 study. We have, however, included those vessels which we consider to have been of foreign origin - in our view Rhodian, but in the opinion of other scholars Cypriot - but in the course oftime were absorbed into the local repertoire and were locally produced in great quantities. Such vessels include the well-known 'Persian bowls' and 'basket jars.' Our discussion of the pottery will proceed in the usual order, beginning with the open vessels and continuing with the closed. In addition to the typological discussion, the reader will find at the end of the chapter the material from selected loci arranged by area and phase, in which the other finds are presented together with the pottery. This presentation also illustrates the gradual development of the pottery from phase to phase which was noted in the field. BOWLS (Fig..1) It is already evident that ordinary bowls are very poorly represented at Dor, both in number and in variety of types. We have already noted a similar phenomenon at the neighboring Tel Mevorakh (Stern 198: 30); this is also the case at the recent excavations at Tell Qiri (Ben-Tor and Portugali 198: 18) and to an even more marked degree at Tel Michal (Herzog 1989: 139). The reason is difficult to determine. In the final report of the Tel Mevorakh excavations, we offered the following explanation: 'The sparsity of the bowl types seems to stem from the fact that all the Persian vessels at the site belong to the later part of the period' (Stern 198: 30). Their absence at Dor, however, cannot be similarly explained. It is possible that the difficulty is a technical one, namely that we did not succeed in collecting all the small sherds of these bowls and thus could not estimate their numbers accurately. Alternatively, the reason may be connected with the coastal or regional location of the site. Whatever the case may be, as at Tel Mevorakh, four classes of bowls can be distinguished: A. Large and small bowls with rounded walls and inverted rims (Fig..1:1-, 8). B. Large and small bowls with ledge rims (Fig..1 :3-). C. Large and small bowls with rounded walls (Fig..1 :). D. Carinated bowls (Fig..1 :9). 1

2 ~1 I 3 4 \~ I J CM Fig..1. Bowls. No. Reg. No. Locus Area Phase I. 4/ /. 119/1. 430/. 430/ /4 48! C AI AI AI AI Cl Cl AI AI b /8 /8 /9 +topsoil +topsoil a?/c? (b'!/c?) Light red clay (.YR /8). Large and small white and shiny grit~ Red-brown clay (.YR /). Many white grits. Red-brown clay (.YR /). White and gray grits. Red-brown clay (.YR /). Many minute white grits. Yellowish red clay (YR /8). White and gray grits. Light red clay (.YR /8). Many minute white grits. Reddish yellow clay (.YR 8/). Smooth white surface. Very fe, white grits. Brownish yellow clay (I OYR /). Many white and gray grits. This classification is also valid for the finds at Tel Dor, with the addition of Type E, a new type which may be a local imitation of the Attic fish plate (Fig..1 :). Type A Flat bowl with disk base, rounded wall, and inverted rim (Fig..1: l-) (Stern 198: 9-9, Fig. 118, Type B l; Stern and Magen 1984: 11, Fig. : I). This type was common in all parts of the country throughout the Persian period. Our examples come from the earlier part of the period. Fig..1:8 is a small bowl with high to low disk base, rounded wall, and sharply inverted rim (Stern 198: 94, Fig. 11, Type AI). Examples have also been uncovered in the new excavations at Tell Qiri (Ben-Tor and Portugali 198: 18, Fig. :3), Tell Keisan (Briend and Humbert!980: Pl. 0:), Tel Michal (Herzog 1989: 141, Fig ), and Yoqne'am (Ben-Tor et a/. 1983: 4, Fig. 8.4). The bowl resembles its larger predecessor, but most examples occur in contexts dating from the mid-th to the end of the 4th centuries BCE. The smaller type is undoubtedly a copy of an Attic prototype. TypeB Bowl with ledge rim (Fig..1 :3-) (Stern 198: 94, Fig. 14, Type A3). It is also found in the recent excavations at Tell Qiri and Tell Keisan (Ben-Tor and Portugali 198: 18, Fig :4-; Briend and Humbert 1980: Pl. 0:), as well as a Yoqne'am (Ben-Tor eta/.!983: 4, Fig. 8:); it dates fron the th-4th centuries BCE. TypeC Bowl with high to low disk base, rounded wall, and rounc upright rim (Fig..1 :) (Stern 198: Type A). It dates fron the th-4th centuries BCE, and was also found at Yoqne'an (Ben-Tor eta/. 1983: 4, Fig. 8:). TypeD Bowl of a well-known type, whose characteristic feature is ' fairly sharp carination of the wall (Fig..1 :9). The bow appears in a wide range of variations in the base, which i' sometimes convex (as in our case) and sometimes flat or' concave disk. Parallels from other sites indicate that it wa' widely distributed throughout Palestine during the entin Persian period (Stern 198: 9, Fig. 11, Type A-). Thi' bowl is well represented at Qadum, where it is ascribed tc the second half of the th century BCE (Stern and Mager 1984: 1, Fig. :4-), and at Tell Qiri (Ben-Tor anc Portugali!98: 18, Fig. :), where it is dated to the sam( period. Our bowl comes from Area AI, phase, which i' attributed to the 4th century BCE.

3 TypeE 'his type (Fig.. I :) is clearly an imitation of the Attic fish late, which is usually considered to belong to the Hellenistic eriod (Herzog 1989: 139, Fig ), At Dor, however, ~is vessel seems to make its appearance at the end of the 4th entury BCE. MORT ARIA (Fig..) Ve have already discussed at length the chronology and oriin of these heavy bowls (Stern 198: 9-98, Figs , 'ype B-). In the meantime, however, many additional xamples of this type have been found in all recent excavaions of sites containing strata of the Persian period. The ites in the vicinity of Dor at which these bowls appear are 'ell Keisan (Briend and Humbert 1980: Pl. 0:1-19, 31), ~ell Qiri (Ben-Tor and Portugali 198: Fig. 3:1-1), {oqne'am (Ben-Tor et a/. 1983: 4-4, Figs. 8:9-10, 1:1-14), Qadum (Stern and Magen 1984: 1, Fig. :9-1), rei Michal (Herzog 1989: 139, 141, Fig ), and Tell ~asile (Mazar 198: 18, Fig. 8:3-4). None of the new finds have altered the general picture vhich emerged from our previous discussion. The mortaria 1t Tell Keisan are attributed to Stratum 4 and those at Tell ~asile to Stratum VII. Both of these strata are dated to the :nd of the Iron Age (end of the th and beginning of the th :enturies BCE). At these sites only the flat-based type, of ight greenish material with rim attached to the wall, occurs. rhis type is now well known from late Iron Age sites in the :outhern part of the country, such as Ashdod, Tel Masos. Beersheba, Arad, and 'En Gedi (Aharoni and Aharoni 19 ). It is now also certain that most of these early vessels are either East Greek imports (in the opinion of the present writer) or Cypriot (according to other scholars). Most of the mortaria discovered in Stratum 4 at Tell Keisan are identical, though a few variations occur, such as everted rims, which are also found at Dor (see below). The mortaria from Tell Keisan differ from the type common in the th-4th centuries BCE, as is evident from the majority of the recently excavated sites containing strata from this period, where mortaria appear with both flat bases and ring bases. Against this background, it is strange that at Tel Michal, which is a coastal site, not a single bowl of the flat-based type has been reported, and the only finds are the regular late bowls with ring base which were dated to the late th century BCE. This sequence, however, contradicts the evidence from all the other Palestinian sites. A possible explanation for this is that the majority of the finds consisted of rims and walls, many of which may have belonged to the flat-based type, though their presence was not reflected in the report. The only flat-based type included in the Tel Michal report belongs to a relatively rare type which is characterized by a ledge, or more precisely a knob handle (Herzog 1989: Fig. 9.1.), for which analogies are known only from coastal sites such as Shiqmona (Stern 198: 9, Fig. 10, Type Ba), Tel Megadim (Stern 198: Fig. 4:1 ), Tel Mevorakh (Stern 198: Fig. 4:1), and a Persian period fortress near Ashdod (Porath 194: Fig. 4:3, Pl. XII:). Even more unacceptable is the excavators' assertion that mortaria ceased to exist at Tel Michal in Stratum VI ( BCE). This too must be Fig... Mortaria. Vo. Reg. No. Locus Area Phase I. 448/1 43 C b(l') Yellow pink clay ( 1 OYR 8/). Pale yellow surface (.Y 8/4). Many white grits.. 118/3 43 AO +8 Reddish yellow clay (.YR /). Pale yellow surface (.Y 8/4). Gray grits /9 143 AI 8?+9? Pale yellow clay (.Y 8/4). Tiny brown grits /1 13 AI (+8?) Light gray clay (.Y N). Very light olive-white surface(.y 8/). Minute gray and mica grits.. 48/4 4 C b Pink clay (.YR /4). Pale yellow-olive surface (.Y 8/4). Many white, gray, and pink grits / 141 AI Pale yellow clay (.Y 8/4). Very few mica and gray grits. Horizontal burnish outside.. 410/4 4 C / Reddish yellow clay (.YR /). White horizontal wash inside. Many white and beige grits C a Dark yellowish red clay (YR!). White horizontal wash inside. Few minute gray and beige grits C b Red clay (.YR /8). Light pink surface. Very few gray grits AO a Red clay (.YR /8). Minute white grits. II Cl 4b Reddish yellow clay (.YR /). Many large and small white, brown, and dark gray grits / 3 co (/?) Reddish yellow clay (YR /). Minute mica and red grits /1 48 C b Reddish yellow clay (.YR /). Smooth light pink surface. Many white grits AO a Pale yellow-olive clay (.Y /4). Gray and many small white grits AO (b?) Reddish yellow clay (.YR /). Pale yellow-olive surface (.Y /4). Many small white and gray grits C b Reddish yellow clay (YR /4). Many large white and gray grits /3 48 C b Pale yellow clay (.Y /4). Many small white and few gray grits /4 33 AO a Pale yellow clay (.Y /4). Many small while and few gray grits /1 481 Cl Later than 4 Reddish gray clay (YR /). Few white grits. 3

4 \ ~I \ "( II "' ~~ / I ~ 1' 3 4 c::; ~ II 8 g ~- I ~ )- I ~ 1 c '- ~ "" 1 1 c E- ~- ~ I~ 1" Fig CM 4

5 onsidered a mere coincidence since it contradicts all the ther recent evidence. Moreover, there can be no doubt 1at the mortaria continue (in a different form) into the Helmistic and Roman periods (see discussions in Chapter ). At Tel Dorwe have uncovered all the known types, includ- 1g the earlier imported mortaria of light greenish clay, the Jcally produced heavy mortaria with flat base (Fig..:9, 1-1) or ring base (Fig..:14-1), as well as the version 1ith knob handle (Fig..: 13). The sequence here is in com,jete conformity with the development reconstructed by the resent writer in the previous treatments mentioned above. KRATERS (Fig..3)!lost of the kraters shown in Fig..3 were found as quite mall fragments which prevented their secure attribution to specific type. It is evident, however, on the basis of the rim hapes alone that the number of types was quite large. In.en era!, we can assume that they belong to the usual kraters fthe period, which are found in large numbers at other Palstinian sites (Stern 198: , Figs. 13-1). It.ppears that the majority are local types with vertical rather han horizontal handles, though some may also belong to the alter type. Most of them are made of light-colored, wellevigated clay. In addition to the above group, two other types were i:mnd. One (Fig..3: I 0) has heavy walls with grooves >eneath the rim and a pair of handles on the upper part. This esse! has analogies at other sites. The second type (Fig.!.3: II) is a holemouth krater with a short wide neck, everted ing rim, typical balloon-like body and round base. This crater is known so far only from sites located in the vicinity >f Dor: Shiqmona (Elgavish 198: Pl. 3:3, ), Tel v!evorakh (Stern 198: Fig. :4, Pl. 3:), and recently also )amaria (evidently from Wadi ed-daliyeh; Lapp and Lapp. 94: Fig. I 0:). The latter site also provides us with a fairly ejiable date (3-33 BCE). The kraters from Dor differ somewhat from the usual type n having an everted rim and concave disk base rather than he usual convex one, but they nonetheless clearly belong to he same family. At Dor, as at the other sites, the ;tratigraphic contexts point to a date in the later part of the 'ersian period. In recent years large numbers ofkraters have come to light n Persian period contexts, some of them ofknown types and Jthers new. They come from Tell Keisan (Briend and Clumbert 1980: Pl. 0:), Yoqne'am and Tell Qiri (Ben-Tor md Portugali 1983: 4, Fig. 9:; Ben-Tor and Portugali 198: 18-19, 1, Photo 4, Fig. 3:13), Tel Mevorakh (Stern 198: 31-3, Fig. : 1-), Tel Michal (Herzog 1989: 10, Fig. l..3; 130, Fig ), and Qadum (Stern and Magen 1984: 19, Fig. ). known from Phoenicia. The chronological contexts indicate that they were especially popular in the th and 4th centuries BCE. It should be noted that the most common cooking pot was of medium size. The cooking pots of this type discovered prior to 198 have previously been discussed by the present writer (Stern 198: , Figs. 19-3, Types B-D). The finds from Dor corroborate our conclusions regarding their distribution and date. Indeed, since then many other similar cooking pots have been uncovered at other sites, such as Qadum (Stern and Magen 1984: 19, Fig. : 1-13), Tell Qiri (Ben-Tor and Portugali 198:19, Fig.4: 1-13), and Tel Michal (Herzog 1989: Fig. 9.1.). In summing up the cooking pots at Tel Michal (which were exclusively of this type), the excavators arrived at the conclusion that during the Persian period the general profile remains unchanged, though the rim which was originally triangular in section later becomes straighter and thinner. This conclusion is not supported by the evidence from Dor. The small cooking pot with handles extending from rim to shoulder, carinated body, and pointed convex base (Fig..4: 1) is a rare type (Stern 198: , Fig. 134, Type F). So far it has been uncovered in only three other sites, all of them close to Dor on the Carmel coast: 'Atlit, Tel Megadim, and Tel Mevorakh (Stern 198: , Fig. 134, Type F). In the shape of the body and the handles, it shows affinities with vessels of the Hellenistic period (though it lacks the characteristic ribbing). At all the sites this pot appeared in contexts of the th and 4th centuries BCE. The example from Dor comes from Area C, phases -, dated to the th-th centuries BCE. It seems that a date within the th century would be appropriate. The medium-sized cooking pot (Fig..4: 1) is distinguished from the others by the absence of handles rather than the shape of the body. Though it is frequently encountered in the Palestinian assemblages of the period from the interior of the country (Stern 198: 101, Fig. 13; 10, Type G, with additional parallels), Doris the first site on the coast at which its appearance has been reported. This pot differs from the others in the dark red color of its clay, which is characteristic of the cooking pots of the coast, and also in its thin walls. At other sites it dates from the th-4th centuries BCE. Its context at Dor indicates a date in the early part of this period. The vessel shown in Figure.4:1 might be considered a carinated bowl of the usual type imitating a Persian metal bowl were it not for its typical cooking pot material (Stern 198: 9, Fig. 11, Type A-). As such, however, it has no parallels. Its discovery in Area C, phase b, points to a date in the 4th century BCE; it may however be a redeposition, as most of the carinated bowls are somewhat earlier. COOKING POTS (Fig..4:-1) fhe most common types of cooking pots of the Persian period, shown in Fig..4:-14, are characterized by wide necks md rims that are either upright or angled straight outwards :'shelf rims'). The handles are rounded and somewhat raised, and the bodies very round and swollen. The base is on vex to flat. These cooking pots are extremely common in both southern and northern Palestine, and parallels are also LIDS (Fig..4:1-4) Four lids of cooking pots from Persian period levels at Dor are shown in Fig..4: 1-4. A survey of the cooking pots discovered at Palestinian sites reveals that some were equipped with lids. Complete pots often have a small depression around the rim to receive the lid. Cooking pots with lids are known from Samaria, Gezer, Tell en-nasbeh, and 'En Gedi. At all these sites, the lids had the same dome-like shape

6 r t ((3F ( ))fz \ II ~~--- 3 '/ ~ \ ' i / --p: ~, ~ ~- ' 8 \ "" 4 '' ' CM Fig..3. Kraters. No. Reg. No. Locus Area Phase I. 40/10 4 C b Reddish yellow clay (YR /0). Beige and some shiny grits.. 4/3 48 C b Yellowish red clay (YR /8). Tiny gray and shiny grits /4 1 AI Reddish yellow clay (YR /). Many gray and white grits. 4, AI Dark brown clay (IOYR 3/3). Many large and small white grits., 1183/3 10 AI (b?/c?) Reddish yellowclay(yr /). Many small white and shiny grits AI b+c Reddish yellow clay (YR /). Light yellow surface. Red and gray grits AI a?/c? Reddish yellow clay (YR /). Few red-brown grits / 41 C a Yellowish red clay (YR /). Tiny shiny and gray grits. 9, C (b?)(+?) Pinkish gray clay (YR /). Light surface. Many white grits AI c Gray clay (loyr /1). Light reddish brown surface. Many large gray and white grits. II Cl 4b Light gray clay (YR /1). Light brown surface. Many large and small white grits.

7 .~ 1 ~-J, I --=) - -Q) CM Fig..4. Lids (1-4) and cooking pots (-1). Vo. Reg. No. Locus Area Phase I Cl b Red-brown clay (.YR 4/). Many white grits / 1198 AI or later Red-brown clay (.YR /8). Some white grits /4 30 co Dark red-brown clay (.YR 3/). Some' beige and shiny grits Cl 4b Reddish yellow clay (SYR /). Some white grits AI or later Red-orange clay (.YR /8). Large brown and small gray and shiny grits AI or later Red-brown clay(.yr 4/). Large brown and small gray and shiny grits AO b Dark red-brown clay (.YR 3/). Many small gray and shiny grits. 8. 4/4 48 C b Red-orange clay (. YR /8). Gray and shiny grits AI b+c Red-orange clay (.YR /8). Large brown and small white grits / 33 AO a Dark red-brown clay (.YR 3/). Many gray grits. II. 119/ 1183 AI b?(+a?/c?) Reddish yellow clay (YR /). Many small white and shiny grits co Red-brown clay (.YR 4/). Many small white and mica grits / AI b?l' Very dark gray clay (YR 3/1). Many white and shiny grits /1 49 co (b') Reddish brown clay (.YR 4/4). Many white grits /4 48 C b Red-brown clay (.YR 4/). Many small gray and white grits C (+?) Pale yellow clay (.Y 8/4). Beige and gray grits / 13 AI (+8') Dark reddish brown clay (.YR 3/4). Some white and gray grits. Much straw mixed with clay.

8 (Stern 198: , Fig. 133, Type E; Stern 198: 33, Fig. : 13). A lid very similar to two lids from Dor (Nos. 1-), which is also conical, has a knob handle, and dates from the 4th century BCE, was recently found at Tel Michal (Herzog 1989: 13-3, Fig ). Cooking pots with lids were quite rare prior to the Persian period; their popularity began in the Hellenistic age. Our lids can be considered early prototypes of the later Hellenistic and Roman vessels, an assumption strengthened by the finds at Dor, since all the lids shown here were uncovered in a 4th-century BCE context..jars Holemouth Jars (Fig..:1-3) Under this category we may have combined two separate types of vessels which share a common grooved rim. Strictly speaking, only No. I should be considered ajar. This type is also known from other sites (Stern 198: 103, Fig.!38, Type C). One was recently found at Tel Mevorakh (Stern 198: 3, Fig. 8:1-19, Pl. :1-4); it also has a grooved rim but differs in details of the body. Of the other two vessels (Nos. -3), only their rims have survived. Though they belong to the holemouth group, they are in fact kraters. This vessel generally has two to four vertical handles extending from the rim to the body, or on the wall, and swollen, rounded bodies. Typical features of this type are grooved handles and rims. This vessel is known so far from two other sites: Tel Mevorakh and a Persian period fortress situated north of Ashdod (Stern 198: 99, Type b; Stern 198:3, Fig. :-, Pl. 3:; Porath 194: Fig. 4:13, Pl. XII:), both dating from the th-4th centuries BCE. Others have recently been found at Qadum in a similar context (Stern and Magen 1984: 19; Fig. :4-). Jar with Cylindrical Neck, Rounded Body, and Pointed Base (Fig..) This jar was found in Area AI, phase Sa, and should thus be dated to the later part of the 4th century BCE. The example shown here was reconstructed, but many others (not shown here) were also found. Our examples have a short cylindrical neck, slanted shoulders, and a base varying from very convex to sharp. The handles extend from the shoulders to the walls, which slant at various angles and are often wavy. An examination of the distribution of this type indicates that most examples come from northern and central Israel and the Sharon coast. Many others have been salvaged from the sea. At all these sites they are dated from the end of the th to the 4th centuries BCE. These jars were dealt with in the present writer's previous summary (Stern 198: , Fig. 14, Type Fl); since its publication many more examples have come to light from the chronological and regional contexts discussed there. Especially large numbers have been published from Tel Michal, where they served as both storage and burial jars throughout the Persian period (Herzog 1989: 139-4, Fig , Pis. 1-3). Indeed, the assumption made in the above study that these vessels were produced in the Sharon coastal area received its final confirmation at this site: 'Since five complete specimens of the type were found in the pc tery kiln of the lower city it is without doubt a local prodw (Herzog 1989: , Fig. 8.14, Pis. 30-3, 3:1-). Other newly excavated sites in which large numbers ofth jar type were uncovered are located in the vicinity of De such as Qadum (Stern and Magen 1984: 19, Fig. 8:; Yoqne'am and Tell Qiri (Ben-Tor eta/. 198:, Fig. 9: Ben-Tor eta/. 1983: 4-4, Fig. 8:; Ben-Tor and Portugi 198:1, Photo 4, Fig. 4:9), and Tell Keisan. However, Tell Keisan the dates are rather anomalous: one grot (Briend and Humbert 1980: Pl. 8) is attributed to Stratu:, i.e., from 380 BCE to the Hellenistic period, and the oth1 to Stratum 4 (Briend and Humbert 1980: Pl. :, 9), date to the th-th centuries BCE; there is undoubtedly sorr error here. It can be observed that in this jar type, as in all other typ< of the entire Persian period, the shape of the body undergo< a change from a pointed base and straight walls to a sacl shaped body. Flat-shouldered Jars (Figs..-8) This is a large family of jars frequently encountered ' almost all sites containing Persian period material in th Galilee and the coastal plain. They are also common alon the shores of the eastern Mediterranean, from norther Syria to Egypt, as well as in Cyprus and Rhodes. Jars of th1 type were also uncovered in the Punic settlements in th western Mediterranean and evidently served as the predorr inant container in the widespread Phoenician marine trade. The main features of these jars are a broad flat shoulde set at a sharp angle to the wall, a wide mouth, a very sh01 neck, a short straight rim, sometimes serving as both rim an' neck, and a pair of handles carelessly attached from th shoulder to the wall and often twisted (Artzy 1980). These features are common to all the jars of this famil; but eight to ten subtypes exhibiting minor variations in th shape of the bases can be distinguished. The above description holds for-the entire jar family tha was designated Family H in our previous summary of th pottery vessels of the Persian period in Palestine (Sten 198: I 0-10, Figs. 14 -). At Dor sherds of many dozens if not hundreds, of jars of this family were recovered fron Areas A and C. As at the other coastal sites, this was withou doubt the most common type of jar in this period. Man: more were also later found in the other excavated areas a Dor. Only a few of these could be restored, since many o them were preserved only as sherds of bases, but the large number of shoulder and rim sherds made it possible t< divide them into two main types: Type H- (Stern 198: I 08 hg. 1), with biconical body, and Type H-8, with sack shaped body (Stern 198: I 08, Fig.!4). In recent years many more jars of this family have beer uncovered in newly excavated Galilean and coastal sites Among these sites were Tell Keisan (Briend and Humber 1980: Pis.!8: 1-11, :1-8, -), Tell Qiri and Yoqne'arr (Ben-Tor and Portugali 198: Fig. 4:10-14; Ben-Tor eta/ 198: photo on p. 8), a shipwreck off the beach at Ma'agar Mikhael (Linder 1991: 43), Tel Michal (Herzog 1989: 141 Fig ), as well as in new excavations along the southerr coast, such as Ashkelon, Tell el-hesi, Tel Sera', Tel Haror and Tel Qatif, which are still unpublished. Large numben 8

9 r 1 1 / CM Fig... Holemouth jars. No. Reg. No. Locus Area Phase I AI or later Reddish gray day (YR /). Light gray surface. Many white and gray grits.. 433/ 4 C +a Yellowish red clay (YR /). Many white and gray grits Cl Later than 4 Yellowish red clay (YR /). Light reddish brown surface. Many gray grits. '----~ (~! I CM Fig... Jar with cylindrical neck, rounded body, and pointed base. No. Reg. No. Locus Area Phase I. 1149/1 119 AI Sa Light reddish brown clay (SYR /3). White, brown, and red grits. 9

10 3 J r C~ Fig... Straight-shouldered jars. No. Reg. No. Locus Area Phase I AI Red clay (.YR /). White and brown grits AO? Light red clay (.YR /). Tiny white and red grits AI (b?ic') Reddish yellow clay (YR /). Few white grits C a Reddish yellow clay (YR /). Many red-brown grits.. 110/ 3 AO (a?/b?) Light red clay (.YR /). Few beige and brown grits.. 110/8 3 AO (a?/b?) Light red clay (.YR /). Few white grits.. 4/8 48 C b Reddish yellow clay (YR /). Sandy grits. 0

11 'b ' ~ 8 ~~ E" ,-=) &;;; o\:m.,_.;"s==='l i,',:SCM Fig..8. Straight-shouldered jars. No. Reg. No. Locus Area Phase L 431/ 431 Cl 4?/a? Very pale brown clay (loyr /3). Many gray and white grits /3 40 AO +8 Very pale brown clay (loyr 8/3). Many gray and white grits /3 4S AO +8 (+b?) Very pale brown clay (1 OYR /3). Many gray and white grits C b Very pale brown clay (loyr 8/4). White and gray grits. s. 100/1 4S AO +8 (+b?) Light reddish brown clay (YR /4). Few gray, brown, and white grits.. IIS0/I 111 AI Sa Reddish yellow clay (YR /). Red and beige grits.. 431/1 431 Cl 4?/a? Reddish yellow clay (YR /8). Large and small beige grits /4 10 AI (b'ic?) Reddish yellow clay (YR /). Many large gray and few brown grits AI or later Red clay (.YR /). Many gray and shiny grits /3 18 AO s Reddish yellow clay (.YR /}. Many small gray and few brown grits. II AI Reddish yellow clay (.YR /). Large beige and some tiny red grits /11 48 C b Yellowish red clay (YR /). Light reddish yellow surface (.YR /). Brown, white, and shiny grits / AO (b?) Reddish yellow clay (YR /8). Few small white grits. 14. S4S11 4 Cl 4b? Reddish yellow clay (YR /). Small red-brown and some large white grits S/3 4 C b Red clay (.YR /8). Many large and small white grits /13 48 C b Red clay (.YR /8). Many white grits co 3 Red clay (.YR /8). White and gray grits S AI b+(a'/sc?) Reddish yellow clay (YR /). Few red and gray grits /3 AO S(b?) Red clay (.YR /8). Very large and small white grits / IIS3 AI Sa Red clay (.YR S/). Light reddish yellow surface (.SYR /). Many white and beige grits / 408 co Sa Reddish yellow clay (.YR /). Many white, beige, and red grits. 1

12 have also been found along the coasts of Phoenicia (Tell Sukas) and Cyprus; they will not be discussed here. This vast body of material has strengthened our previous conclusions not only regarding the geographical distribution of these jars but especially regarding their chronology: a clear developmental tendency (which apparently also existed in many other closed vessels of the period) can be observed in the jars of this family. The earlier jars possess a biconical body and a relatively high neck (Type H-) (Stern 198: Fig. 1), as at Dor (Fig.. ), and date from the th-th centuries BCE, while the later types, with sackshaped body (Type H-8; Fig..8) (Stern 198: Fig. 14), date from the th-4th centuries BCE. This sequence seems to have been present also at Tel Dor, though most jars were preserved there only in small fragments. 'Torpedo' Jars (Fig..9) The only restorable jar of this type at Dor (Area A) had a single handle (Fig..9:1). Though apart from this only rim sherds of the jars of this type have survived (Fig..9:-4), as well as body sherds that could not be restored, the impression was received that it was quite common at Dor; later it also turned up in other areas. Many specimens have also been recovered at neighboring sites, such as Tel Mevorakh, Tel Megadim, and 'Atlit (Stern 198: 10-10, Fig. 14, Type G), as well as along the Galilean and Phoenician coasts (Raban 19: 49-0, Fig. 4:-9). It is therefore surprising that none have been reported from other nearby sites, such as Shiqmona and Tel Michal (unless we assume that the vessel shown in Herzog 1989: Fig represents the body of such a jar). However, the type does appear in sites in the Samaria region, such as Qadum (Stern and Magen 1984: 19-0, Fig. 8: I, 3, 4) and perhaps also at Shechem (Stern 1980: Fig. :9, differing in details). This jar was widespread in the Persian period. Its name derives from its elongated shape. The typical features of this family are slanted, conical shoulders, a wide mouth, thick everted rim, and a long, narrow cylindrical body. The wall is either straight or curved. The handle, or handles, extend from the lower edge of the shoulder to the wall. The base is usually pointed. This family can be subdivided according to two criteria. The first is the number of handles: there are jars without handles, with one handle (shown here in Fig..9: 1), and with two handles. The second is the shape of the neck: the jars have a narrow neck (Fig..9:1, 3-4) or a cylindrical, long and relatively wide neck. Of the latter type, we found only one sherd of part of a neck (Fig..9:). Storage jars of this type are known from Mesopotamia, Syria, Phoenicia, and especially Cyprus, as well as both the coastal and mountainous regions oflsrael. The contexts of these finds, both in Israel and in other countries, indica I that they were in use throughout the Persian period, fror the th to the 4th centuries BCE, with the main distributio dating to the period from the mid-th to the end of the 4t centuries BCE. Raban, who recovered many examples ol the coast ofphoenicia and Palestine, believes they were use for preserving fish (Raban 19: 0). Our finds, uncovered here in phases - in Areas A an C, do not alter the general picture presented above. Thes phases are generally dated to the th and first half of the 4t centuries BCE. Basket Handle Jars (Fig..10) The rims, body sherds, and handles presented in Fig..11 belong to the basket handle family of jars which is character ized by a round shoulder, short cylindrical neck, ovoid body and two basket handles that rise above the rim. The base i pointed or truncated. Like the flat-shouldered family dis Fll\?T\ 3 IT\ CM Fig..9. 'Torpedo' jars. No. Reg. No. Locus Area Phase I / / /4 10 AO C AI AI (a?/b?) b (b'?) (b?/c?) Very pale brown clay ( IOYR /3). Few white and gray grits. Reddish brown clay (.YR /). Many white, brown, and black grits. Reddish brown clay (.YR /). Many white grits. Reddish yellow clay (YR /8). Brown grits.

13 ussed above, the basket jar is confined in Palestine to the ialilee and the coastal plain, but it is also known on the eastrn shore of the Mediterranean, as well as in Cyprus and lhodes; it must have been in use in marine trade during the 'ersian period. Gjerstad was formerly of the opinion that hese jars originated in Cyprus, but it now seems that the arliest examples come from Rhodes and in fact belong to he plain East Greek wares (together with the flat-based nortaria discussed above). In Palestine these jars appear for the most part in two vari.nts: one with low rounded handles and a biconical body Stern 198: 110-1, Fig. 1, Type I I), and the other with aller handles with the widest point ofthe body at the shoullers, from which point it narrows towards the pointed or tump base (Stern 198: Fig. 1, Type I). A survey of hese jars from Palestine and other countries shows that the irst type dates to the th-th centuries BCE and the second o the th-4th centuries BCE. In recent years jars of this family have been found and mblished not only from Dor, but from many sites in Cyprus md all along the eastern coast of the Mediterranean, from cell Sukas (Riis 199: 0, Figs. 4-49) through Tell Keisan, cell Qiri, and Tel Michal (Briend and Humbert 1980: 3-41, Pls. 3-4, ; Ben-Tor and Portugali 198: 19, Fig. kl-19; Herzog 1989: 143, Fig. 9.1.). Large amounts 1ave also turned up in shipwrecks off the coast ofdor itself md off the beach at Ma'agan Mikhael, dating from the end lfthe th century BCE (Linder 1991: 43). Others comefrom ;outhern coastal sites and the nearby Shephelah, such as \shkelon, Tel Qatif, Tel Sera', and Tel Haror, which have 10t yet been published.* In general our chronological obserration regarding the Palestinian examples according to the :riterion of body shape is still valid. It is based largely on the :onclusions arrived at by Gjerstad on the basis of the Cyp iot finds, i.e., a transition from the earlier form with the Jiconical body and low handles to the later one character.zed by wide shoulders and body sharpened towards the Jase. The date of the earlier type is from the th to the middle Jr late th centuries BCE, and of the later, the th to 4th cen :uries BCE. The excavators of Tell Keisan have recently proposed an 1dditional chronological criterion for these jars, in which they distinguish three variations in the shape of the neck and the rim: I) a relatively long neck with rounded rim and neck ridge; ) a very short neck with everted rim and neck ridge; and 3) completely lacking a neck, the rounded rim rising directly from the shoulder (Briend and Humbert 1980: ; Herzog 1989: 143). This development, which in general is reflected also in the Dor finds, is a positive addition to our basic typology noted above, as Type II with the biconical body also possesses a long neck and Type 1, with wide shoulders and sharp body, lacksaneckandhasaflat roundedrim(fig..10: 1-9). There are also several transitional types between Types II and I that can be ascribed to the earlier or later part of the period according to their body shape. As for their distribution, these jars have been uncovered in recent excavations only in coastal sites. At Qadum, which is located somewhat further inland, not a single vessel of this type has been found, despite the presence there of a relatively rich pottery assemblage (Stern and Magen 1984). Regarding their origin and function, we should again mention the recent and very interesting finds from Ma'agan Mikhael. * In a shipwreck discovered only a few meters off the shore, a brand-new ship dating from the end of the th or beginning of the 4th century BCE was found. The hull of the ship was almost completely preserved in the shallow water. The ship's cargo consisted mainly of these jars, though there were also one or two jars of the flat-shouldered type. In the opinion of the present writer, all the other pottery vessels in this ship (jugs, juglets, and even one pithos that held part of the ship's fresh water supply) were of East Greek origin (Stern 198: Ill), with the exception of one or two Cypriot vessels. This shipwreck provides us with additi\)nal evidence that these basket handle jars were produced mainly on the shores of East Greece and that they were employed in the marine trade (Barag 191: PI. :; Zemer 19: Pl. 8:4). As for the origin of these vessels, it will be possible to reach a decision between an East Greek and a Cypriot source only after archeometrical analysis is carried out on the imported pottery. There is a general consensus that there are also local imitations, as decisively proved by the petrographic analysis of the Tell Sukas and Tel Michal material, and the Phoenician letters incised on some of the jars from Tell Keisan. As for their use, in addition to the maritime trade, it should be noted that it is now clear that in some of the newly excavated sites mentioned above (Salamis in Cyprus and Tel Michal) these vessels were also used as burial jars. JUGS (Fig..11:-13) Not a single complete jug was found in Areas A and C and our impression is that the number of sherds of this type was not large. Some examples of rim and neck sherds are shown here (Fig..11 :-11), as are a base and some body sherds (No. 1). It is of course difficult to identify these jugs with confidence; nevertheless most of them seem to be of Type H-G (Nos. -1 0), with globular or sack-shaped body (Stern 198: 11, Figs. I 1-). These are especially common along the northern coast and are found in large concentrations in Samaria and nearby Qadum (Stern and Magen 1984: 0, Figs. :9, 8:-9), but they are also known from 'Atlit, Shiqmona, and Tel Mevorakh (Stern 198: 11), and recently also from Tel Michal (Herzog 1989, Fig. 9.1.). Most of them were found in contexts of the th-4th centuries BCE, like our own examples. One sherd (No. II) belongs to a uniquejugwhichdisplays two handles instead ofthe usual one. Another sherd (No. 1) is apparently the base of a jug of Type I, which is 'a jug with short thick neck, pinched mouth, rounded handle extending from rim to shoulder, rounded body with either convex or ring-base which is especially common in the th-4th centuries BCE' (Stern 198: 11, Fig. 13). It may derive from the Phoenician repertoire of late Iron Age vessels. * The material was shown to the present writer by its excavators, E.L. Stager and E. Oren, whom I wish to thank. * This material was shown to the present writer by the excavator, E. Linder, to whom I am greatly indebted. 3

14 / CM Fig..10. Basket handle jars. No. Reg. No. Locus Area Phase I. 1100!!83 AI b?(+a?/c?) Red clay (.YR /). Gray and white grits..!194/ 10 AI (b'ic?) Light red clay (.YR /8). Gray and beige grits /l 414 co (a?) Red clay (.YR /8). Large brown and white grits /l 4184 co (b?) Light red clay (.YR /). Very pale brown surface. Many white and brown grits / 40 co (c?) Light reddish yellow clay (YR /8). Very pale brown surface (IOYR 8/4). White and shiny grits.. 439/l 41 C a Light red clay (.YR /). Shiny and white grits C b Light red clay (.YR /8). Pale yellow surface (.Y 8/4). Small shiny grits /3 409 co b+ Light red clay (.YR /). Small gray and shiny grits AO Red clay (.YR /8). Many white grits. 10.!00/ 4 AO +8 (+b?) Reddish brown clay (YR /4). Tiny gray and white grits. II. 4!/ 48 C b Light red gray clay (.YR /). Small white and shiny grits /l 48 C b Light yellowish brown clay(l OYR /4). Light yellow surface ( 1 OYR /). Small white and shiny grits. 13. II3/l 11 AI Light red clay (.YR /). Gray and white grits ! C a Light yellowish brown clay ( 1 OYR /4).Gray grits. 4

15 JUGLETS (Fig..11:1-) uglets too were relatively rare in Areas A and C at Dor. Of he dipper juglets. mainly flat bases have been preserved. rwo of these bases (Fig..11 :-3) are made of a distinctive ight-colored clay. By coincidence, they both come from >hases belonging to the 4th century BCE, but juglets from 'arlier phases are also present. Since the majority of sherds found were of flat bases, it is lifficult to determine to which of the two types of dipper uglets possessing this base they belong. They may be oftype!a, which has an elongated, cylindrical body, wide neck and mtward-flaring rim, flat base, and loop handle extending 'rom rim to shoulder, or of Type B with sack-shaped body Stern 198: , Figs. 18-9; see also Stern andmagen 1984: 0, Fig. 8:10; Herzog 1989: Figs. 9.4., 9..14). It ;eems, however, that both types are represented, with the najority being of Type B, whose distribution is limited nainly to the coastal region during the th-4th centuries BCE. 3 m B The other juglets present all belong to perfume juglets of well-known types. These include examples of Type SA (Nos. 1 and ) (Stern 198: 11-, Fig. 188), ajuglet with round body, handles extending from shoulder to rim, and a disk or low ring base. This is the most popular perfume juglet ofthe Persian period in all parts ofpalestine; it is similar to the jug which is also common in the th-4th centuries BCE (Stern 198: 11, Fig. 19, Type E; see also Herzog 1989: 13, Fig ). Another perfume juglet well known in the pottery repertoire of the th-4th centuries BCE (No. 4) is an example of Type 3 (Stern 198: 11-, Figs. 1-, Types 3-4), which has the early type of rounded body. It is certain, however, that the sherds also included examples oftype 4, which is sack-shaped and somewhat later in date (Stern 198; see also Herzog 1989: Figs. 9.1., 9..10). Perfume juglet No. belongs to Type in the general typology (Stern 198: 13, Fig. 18). This family of juglets has a long narrow neck and thick ring rim. The body shape varies from globular to spindle-shaped. Its distinctive fea- CD ~ 4 \ ( 9 ~ :,:--- ':.---_- "Y L._---j CM Fig..11. Juglets (1-) and jugs (-1). No. Reg. No. Locus Area Phase I Cl 4c Reddish yellow clay (.YR /). Light yellow surface. Red decoration AI a?/c? Reddish yellow clay (YR /). Some white grits /4 AO (b?) Reddish yellow clay (.YR /). Light yellow surface outside. Some white grits. 4. 1/3 4 AO +8 (+b?) Very pale brown clay (loyr 8/4). Few tiny white grits AI b?/? Reddish yellow clay (YR /). Tiny gray grits. Red-orange decoration.. 438/ 41 C a Reddish yellow clay (YR /). Tiny white and gray and some red grits C b Very pale brown clay (loyr 8/3). Tiny gray-black grits C b?/a? Yellowish red clay (YR /8). Large and small white and gray grits AI I Yellowish red clay (YR /). Small gray grits /1 100 AI Reddish yellow clay (.YR /). Pale yellow surface outside. Pale red paint on rim. II. 110/1 11 AI a(lb'') Very pale brown clay (!OYR /3). Many small gray grits AI?1' Pale yellow-olive clay (YR 8/3). Few gray grits. Vertical burnish.

16 ro.. 11} CM Fig..1. Amphoriskoi (1-) and flasks (3-4). No. Reg. No. Locus Area Phase I Cl 4a. 493/ 43 C (b?) (+?) /1 43 C (b?) (+?) 4. 38/1 9 Cl 3?!4? Reddish yellow clay (YR /). Red-orange decoration. Reddish yellow clay (YR /). Red-orange decoration. Some gra grits. Light red clay (.YR /8). Light outer surface. White and gra: grits. Light reddish brown clay (YR /4). Few brown grits. lure is the high truncated base. The handle (missing here), usually rounded and raised in a triangular shape, extends from the middle of the neck or from a ridge to the shoulder, and is decorated with a horizontal painted band in brown or red. This is a well-known type in levels of the th-4th centuries BCE, mainly along the coast (Stern 198: 13; and now also Herzog 1989: Figs , 9.., , 9.9., , , in many variations). Our example also comes from a 4th-century BCE context. AMPHORISKOI AND FLASKS (Fig..1) In Fig..1 two different types of vessels are displayed. Nos. 1- are small amphoriskoi with raised basket handles and Nos. 3-4 are flasks. The amphoriskoi have upright handles which are attached at right angles to the shoulder. The body is decorated with reddish painted bands on a yellowish background. A similar vessel was found at nearby Shiqmona (Elgavish 198: Pl. :18; see Stern 198: 114, Fig. 13). Amphoriskoi of this type were also uncovered in very large numbers in tombs of the th-4th centuries BCE in Cyprus, where they apparently originated (Gjerstad 193: Pis. I 08: I, top row middle; 108:, right). The vessels from Dor, however, seem to be local imitations. Except for Dor and Shiqmona, no parallels have beer found up to now in Palestinian sites, though it should b< noted that an amphoriskos of another type, larger and will similar basket handles, was uncovered at Tel Michal in Stra tum VIII and was shown by petrographic analysis to b< imported(herzog 1989: 14, Fig. 9..4). The other two vessels (Nos. 3-4) are flasks of which on!) the necks and small fragments of the handles attached tc them have survived. These flasks are also of a very weu. known type common in the Persian period in all parts ofth< country. Initially, we were of the opinion that they wen mainly typical of the southern part of the country (Stern 198: 114, Fig. 1, Type A). However, recent finds hav< shown that large numbers also come from the Galilee and the coastal region. In addition to those found at Dor, flasb have also been uncovered at Tel Michal (Herzog 1989: 13, Figs. 9..1, ). BASIN (Fig..13:1) The vessel shown in Fig..13: I has been defined as a basin, a fairly rare type of vessel; it is similar to basins found in other Persian period sites. One was uncovered recently at Tel Michal (Herzog 1989: 19, Fig. 9..8), where it was CM Fig..13. Basins. No. Reg. No. Locus Area I AI AI Phase or later a?/c? Light reddish brown clay (YR /4). Thick dark gray core. Large beige grits. Traces of straw in clay. Light red clay (.YR /). Very thick gray core. Thumbed decoration outside. Even buff burnish inside.

17 ttributed to Stratum VIII ( BCE). The basin from lor differs somewhat from the one found at Tel Michal..tudy of the rna terial from other sites will undoubtedly eveal additional examples. LAMPS (Fig..14) 'he lamps shown here are only two examples of dozens of imilar lamps discovered in all parts of the areas discussed 1ere; other lamps of this type from Dor have been treated lsewhere (Stern 1989: 11). These two lamps are both from "ea A. No. I belongs to phase -8 and No. to phase -; hey thus extend throughout the entire Persian period, as at 1ther sites. The lamps are both of the same type, i.e., the open, flat ype with straight base. The lip is flat and very wide and the ingle wick hole is pinched sharply. The lamps differ in that -.Io. I is unusually large, its surface is knife-shaved, and it tas a light burnish with a wide band. These features are mique to the coastal region (Stern 198: 1-9, Figs. ~0-03). Since the publication of our general summary, many more lamps of this type have been uncovered at other sites and published. They have appeared, for example, in Persian period levels at nearby sites on the coast and in the northern valleys such as Tell Keisan (Briend and Humbert 1980: 13, Pl. 1: 1-), Tel Michal (Herzog 1989: 143, Fig ), and Qadum (Stern and Magen 1984: Fig. 9.-4), as well as at numerous other sites. The many new finds have not altered the general picture presented in our earlier survey (Stern 198: 1-9, Figs. 0-03). Only at Tel Michal have the excavators claimed that these lamps ceased to occur in their Stratum VI ( BCE), a fact which in our opinion should be regarded merely as a coincidence. As for the closed lamps of the late Persian period which are usually local imitations of Greek lamps (Stern 198: 19, Fig. 04; and see now Rosenthal-Heginbottom 1988), these also appear in Palestine towards the end of the Persian period and are common mainly in loci dated to the transitional period between the Persian and Hellenistic periods. This seems to be the case at Dor as well. ' ' / ' ~ ~/ ~ ' ' ' ///~- -- ;/-/,: _ " ~ ' I ' I ' I! I,'I.'.'' ' /' ~ ~ '-, '. '\ \ ' ' I ' I I ' ' I I I ::~, Fig..14. Lamps CM No. Reg. No. Locus Area!. 1198/ 40 AO. 113/ 1193 AI Phase +8 b'l? Reddish yellow clay (YR /8). Minute white and brown grits. Burnished outside and on rim inside. Light reddish yellow clay (YR /). Very pale yellow-brown surface. Few white grits.