Archaeological Institute of America

The Chronology and Function of Ceramic Unguentaria
Author(s): Virginia R. Anderson-Stojanović

Source: American Journal of Archaeology, Vol. 91, No. 1 (Jan., 1987), pp. 105-122
Published by: Archaeological Institute of America
Stable URL: http://www.jstor.org/stable/505460 .
Accessed: 17/01/2015 06:20
Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at .
http://www.jstor.org/page/info/about/policies/terms.jsp
.
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of
content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms
of scholarship. For more information about JSTOR, please contact support@jstor.org.
Archaeological Institute of America is collaborating with JSTOR to digitize, preserve and extend access to
American Journal of Archaeology.
http://www.jstor.org
This content downloaded from 130.191.17.38 on Sat, 17 Jan 2015 06:20:38 AM

All use subject to JSTOR Terms and Conditions
The Chronologyand Function of CeramicUnguentaria
VIRGINIA R. ANDERSON-STOJANOVIC
Abstract 3) the functionof unguentariafound in burials was close-

Ceramic unguentaria are among the most common ly connectedto funeraryritual and not necessarilyrelated
to the commercialtrade in perfumedoils and unguents.
grave offerings of the Hellenistic and early Roman peri-
ods. The function of these vessels in funerary contexts,
however, and their chronology have not yet been firmly Small ceramic bottles, or unguentaria, sometimes
established.A preliminarystudy of the unguentariafrom referred to as lacrimaria or balsamaria, are familiar
the site of Stobi in Yugoslavian Macedonia suggests new artifacts at Hellenistic and early Roman sites, and are
directionsof investigation.Informationabout the context,
especially common in cemeteries throughout the
appearance, capacity, frequency, and chronologyof un- Mediterranean area from Palestine to Spain.1 Collec-
guentaria at Stobi and elsewhere indicates that 1) un-
tions of unguentaria have been published from many
guentaria were producedin many centers rather than in
just a few, 2) any proposed chronology for unguentaria of these cemeteries but comparatively little work has
must take into accountthis regionalor local variation,and been devoted to establishing their function.2
1 No comprehensivestudy of unguentaria has been pub- D. Mano-Zissi eds., Stobi II (Beograd 1975) 143-62. See
lished. For the Early Hellenistic period see L. Forti, "Gli also I. Mikulkii, "The West Cemetery: Excavations in
unguentari del primo periodo elenistico," RendNap 37 1965," Studies in the Antiquities of Stobi I (Beograd 1973)
(1962) 143-57. Forti provides a summary of the evidence 61-97. Publicationsof unguentariafrom Greece includethe
for the dates of Early Hellenistic examples and a typologyof following: P. Bruneau, "Tombesd'Argos,"BCH 94 (1970)
forms. Her drawings do not show the inner profile. Most 437-531; for Athens, see C.G. Boulter, "Gravesin Lenor-
published commentson this vessel type have been related to mant Street, Athens," Hesperia 32 (1963) 113-37; B.
chronologyand/or shape. The fullest treatmentof the sub- Schlorb-Vierneisel, "Eridanos-Nekropole,"AM 81 (1966)
ject as of 1965 was by P. Hellstrom, in Labraunda2.1. Pot- 1-111, and U. Knigge, Kerameikos9. Der Siidhiigel(Berlin
tery of the Classical and Later Periods, Lamps, and Glass 1976) 60, 158-68, 184-89; for Corinth, see C.W. Blegen,
(Lund 1965) 23-27. For more recentcomments,see I. Hagg H. Palmer, and R. Young, CorinthXIII. The North Ceme-
and J.M. Fossey, The Hellenistic Nekropolis and Later tery (Princeton 1964), and E.G. Pemberton,"Ten Hellenis-
Structures on the Middle Slopes. 1973-1977, in I. H~gg and tic Graves in Ancient Corinth,"Hesperia 54 (1985) 271-
R. Hdigg eds., Excavations in the Barbouna Area at Asine 4 307; on Cyprus, O. Vessberg and A. Westholm,
(Acta Universitatis Upsaliensis, Boreas 4:4, Uppsala 1980) SwCyprusExp 4.3. The Hellenistic and Roman Periods in
29-30, 49-53, 66-88, 72-74, 86, 123; and N.I. Khairy, Cyprus (Lund 1956) 60-61, 68, 74, 79, 80; G. McFadden,
"Natatean Piriform Unguentaria," BASOR 240 (1980) "A Tomb of the Necropolis of Ayios Ermoyenis at Kou-
85-91. rion,"AJA 50 (1946) 449-89, A. Oliver, "Tomb 12 at Epis-
Many of the ideas in this article formed the basis for a kopi," RDAC 1983, 245-56, and I. Nikolaou, "A Hellen-
paper presented at the 83rd General Meeting of the Ar- istic and Roman Tomb at Eurychou-Phoenikas,"RDAC
chaeological Institute of America in San Francisco in De- 1984, 234-57; for Samos, see C. Tsakos, "HellenistikoiLa-
cember of 1981 (AJA 86 [1982] 252-53). I should like to cheutoi Taphoi ste Samo,"ArchDelt32 (1982) 344-420; for
express my thanks to the following individuals who have Macedonia, see S. Drougou and G. Touratsoglou, "Hellen-
read drafts of this article and contributed helpful sugges- istikoi Lacheutoi Taphoi Beroias," ArchDelt 28 (1980)
tions: E. Mott Davis, John Hayes, Carolyn Koehler, Susan 123-29; for Yugoslav Macedonia, see I. Mikul'i', Pelago-
Rotroff, Malcolm Wallace, Al Wesolowsky, and James nia (Diss. Beograd 1966) pl. XVI, and K. Kepeski, "The
Wiseman. My thanks also to Dragan Stojanovi' for the Sondage Excavationsof a Part of the Hellenistic Necropolis
drawings, and to Margaret Vanderau for help with various under the Markove Kule in the SettlementVaro'," (in Ser-
aspects of my research. bo-Croatian with English summary) MacedoniaeActa Ar-
For permission to illustrate unguentaria and measure- chaeologica1 (1975) 155-64, pl. 1.2; for Yugoslaviain gen-
ments of capacities from the Excavations of the Athenian eral, see AleksanderJovanovid,Forms of Burial in the Ter-
Agora, I should like to thank T. Leslie Shear, Jr., Director, ritory of Yugoslavia in the Time of the Roman Empire (in
and Susan Rotroff. Serbo-Croatian with English summary [Beograd 1984]),
2 The cemeteriesat Stobi were excavatedunder the direc- and D. Basler, "Nekropolana Velim Ledinama u Gostilju
tion of Al B. Wesolowsky, who is preparing the final publi- (Donja Zeta)," Glasnik Zemaljskog Muzeja Bosne i Herce-
cation. For his preliminary comments,see "BurialCustoms govine NS 24 (Sarajevo 1969). I have examined unpub-
in the West Cemetery," in J.R. Wiseman and D. Mano- lished material in museums at Heraclea, Bitola, Stip, and
Zissi eds., Stobi I (Beograd 1973) 97-137, and "The Patho- Ohrid in Yugoslavian Macedonia and in Montenegro, at
logy of Human Remains from Stobi,"in J.R. Wiseman and Budva, Cetinje, and Titograd. For Asia Minor, see C.H.
American Journal of Archaeology 91 (1987) 105

106 VIRGINIA R. ANDERSON-STOJANOVIC [AJA 91
In the course of studying the unguentaria from bur- vary in size from miniature examples 4 or 5 cm. in
ials at the site of ancient Stobi, certain characteristics of height to "giants" 20-30 cm. high.5 Most of these ves-
these vessels and their context encouraged a reapprai- sels, however, fall within the range of 8-20 cm. in
sal of their function and chronology. The results of the height.
study of these vessels at Stobi and the review of pub- The term lacrimarium or lacrimatorium has been
lished groups of unguentaria from Greece, Asia Mi- used for these vessels in the belief that they were used
nor, Palestine, Italy, and Spain, together with sugges- to collect the tears of mourners, an assumption dis-
tions for future research, are presented in this article. missed as highly unlikely by most scholars.6 Balsama-
Most unguentaria are found in burials and are rium has also been used as a descriptive term by ar-
often intact or only partly damaged. Other examples chaeologists and, like unguentarium, was chosen to
are found in public and residential areas as vessels for indicate the probable contents of these small bottles.
everyday use and as offerings in religious sanctuaries.3 All of these terms, however, are of recent origin and,
The quality of unguentaria varies from well-made ex- in fact, we do not know that in antiquity any one term
amples with fine clay bodies and black gloss exteriors was used to describe the vessel that we now call un-
to others of extremely careless manufacture. Most of guentarium. Both alabastron and lekythos are terms
the unguentaria found in burials fall into the latter that may have been used, but no literary sources con-
group. Those found in settlement contexts, however, tain explicit references to the unguentarium.7
are also frequently of poor quality; hence, it is not pos-
sible to distinguish the purpose of the vessel on the FORMS AND CHRONOLOGY
criterion of appearance alone. Therefore, my com- The unguentarium was produced in two basic
ments, although based on an analysis of unguentaria shapes: 1) the fusiform or spindle shape with body
from a mortuary context, may also apply to vessels of varying from rounded to slender, and a foot usually set
similar manufacture from other contexts. off clearly from the body (figs. la-c, g; 2-4, 6), and
The ceramic unguentarium seems to succeed the 2) the bulbous shape with round or pear-shaped body
lekythos of Classical times as a popular grave offering and flat base (figs. Id-f, h-i; 5, 7, 8).8 The primary
throughout the Mediterranean world.4 Unguentaria difference between the two shapes is the presence or
Greenewalt, jr., "The Sardis Campaign of 1977," BASOR from the Near East, see F. Christensen and M. Johansen,
233 (1979) 12-19; for Palestine, P.W. Lapp, Palestinian Hama. Fouilles et Recherches 1931-1938 3.2 (Copenhagen
Ceramic Chronology200 B.C.-A.D. 70 (New Haven 1961) 1971) 45-50; for Italy, M. Maltese, "Gli unguentari seli-
79, 197-99, and P. Kahane, "PotteryTypes from the Jew- nuntini," SicArch 8:27 (April 1975) 25-35, and E. Fabbri-
ish Ossuary Tombs Around Jerusalem," IEJ 2 (1952) cotti, "Basilicatascavi in localita Fontana Bona 1972,"NSc
131-39, 176-82. For Italy, M. Fortunati Zuccala, "Gropel- 33 (1979) 347-413, passim, and esp. 409. For material in
lo Cairoli (PAVIA), La necropoli romana,"NSc 33 (1979) Austria, see comments by E. Schindler-Kaudelka, Die
1-88; for Spain, M. Almagro, Las necrcipolisde Ampurias diinnwandige Gebrauchskeramik von Magdelensberg
I-II (MonografiasAmpuritanas 3, Barcelona 1952, 1955), (Karntner Museumsschriften 58, Klagenfurt 1975) 219-22,
and E. Cuadrado, "Ungtientariosceramicos en el mundo pl. 33.
ib6rico. Aportaci6n Cronologia,"ArchEspArq 1977-1978, 4 Forti (supra n. 1) 143; Boulter
(supra n. 2) 115; Gold-
389-404; also of interest from Russia, M. Parovi&-Pevikan, man (supra n. 3) 171, and n. 65; Hellstrom (supra n. 1) 25;
Nekropoli Olvii Ellinisticeskogo Vremeni (Kiev 1974) and D. Kurtz and J. Boardman, Greek Burial Customs
108-13; and Bulgaria, P. Balabanov, "S~pulturetumulaire (Ithaca 1971) 164, 165. For commentson the functionof the
d'une Thrace riche d'Anchiale," Arkheologiia 21 (1979) lekythos, see C.H.E. Haspels, Attic Black Figured Lekythoi
23-32, esp. 25-27 (in Bulgarian). (Paris 1936) 92, 125-29, 176.
3 Publications of unguentaria from non-burial contexts 1One of the
"giants"is illustratedin J. Wiseman and D.
include: in Athens, from the Agora, H.A. Thompson, "Two Mano-Zissi, "Stobi:A City of Ancient Macedonia,"JFA 3
Centuries of Hellenistic Pottery," Hesperia 3 (1934) (1976) 280, fig. 11.
472-74; on Aegina, H. Walter ed., Alt-Agine 2.1 (Mainz 6
Thompson (supra n. 3) 472, and Hellstrom (supra n. 1)
am Rhein 1982), "Spditklassischeund hellenistische Kera- 23. The authority for this suggestion comes from Psalms
mik,"by R. Smetana-Scherrer,88-91, pls. 55-56; at Myce- 56.8, "putmy tears into your bottle."
nae, W. Rudolph, "Hellenistic Fine Ware Pottery and 7 On the derivation of the unguentarium, see Hellstrom
Lamps from above the House with the Idols at Mycenae," (supra n. 1) 24, 25. Another Latin term, olfactariolum,is
BSA 73 (1978) 223, 231; on Delos, P. Bruneau et al., Delos cited in J. de Alargco, Fouilles de Conimbriga5. La cera-
XXVII (Paris 1970) 254-55; in Asia Minor, K. Nohlen mique commun, locale, et regionale (Paris 1975) 36. See also
and W. Radt, AvP XII. Kapikaya bei Pergamon (Berlin W. Hilgers, Lateinische Geftissnamen (Dusseldorf 1969)
1978) pl. 17; H. Goldman ed., Excavationsat GiizliaKule, 233, no. 265, and 298, no. 376.
Tarsus I. The Hellenistic and Roman Periods, Section 6: For a good illustration of the shapes, see D. Thompson,
8
F.F. Jones, "The Pottery"(Princeton 1950) 171, 172, 200; AgoraPicBk 12 (Princeton 1971) ill. 37.

1987] THE CHRONOLOGY AND FUNCTION OF CERAMIC UNGUENTARIA 107
.d
b
a
OI 2 3 4 5cm.
Irirrg
f
Fig. 1. Profiles of Stobi unguentaria. a C-72-44, Grave 102; b C-74-374, Grave 295; c C-74-109, Grave 273; d C-70-71,
Grave 18; e C-74-114, Grave 274; f C-75-105, Grave 265; g C-74-102, Grave 265; h C-75-12, Grave 283; i C-74-363,
Grave 285. (Drawing by Dragan Stojanovii)

absence of the foot. The fusiform unguentarium, or
Shape 1, was in use for several centuries and naturally
displays greater variation in profile than the bulbous
shape popular for less than 100 years.9
Within the fusiform group are included the
rounded forms as well as very slender profiles that
tend to be later. The earliest examples of fusiform un-
guentaria with very globular bodies and small feet are
similar to the Classical amphoriskos but have neither
the vertical handles nor the convex mouth typical of
that shape.10 Some examples of this profile from the
cemeteries of Olbia in southern Russia have small
horizontal handles at the shoulder which appear to be
more decorative than utilitarian." Most examples of
this shape, however, have no handles. This form ap-
pears in Spain during the early to mid-fifth century
B.C. but is not found in the Eastern Mediterranean
until the late fourth century B.C. (fig. 2).12
Variations of the fusiform shape characterized by a
tall foot and smaller body cavity appear in Spain and
the Western Mediterranean during the fourth century
B.C.13 By the late fourth or early third century in
Greece one finds examples with the tall foot or stem.14
Different profiles seem to have a limited chronological Fig. 2. Shape 1, globular fusiformunguentarium,Athenian
significance in certain geographical areas." Some ex- Agora P636. Ht. 0.079 m. (CourtesyAgora Excavations)
amples retain a rounded profile, others are distin-
guished by a more slender body. A few examples of this low.17 Many variations in the fusiform shape occur,
variety also have small basket handles at the shoul- some of which are illustrated in this paper.
der.'6 There is a tendency over time toward lessening For example, at Stobi a variety of fusiform shapes
the amount of interior space, or capacity, of these ves- are found in graves of the second and first centuries
sels by making them increasingly slender or shorter in B.C., but only one (fig. la, Type A) is of the early
body, and by making the tall foot solid rather than hol- variety without a foot.18 The remaining six types (e.g.,
9This variation in form is evident in the typology and Grave 140, p. 83, "latein the 3rd quarterof the 4th c. B.C."
chronology of fusiform unguentaria proposed by L. Forti 13 Cuadrado(supra n. 2) 399, Type B, dated 310-50 B.C.,
(supra n. 1). and M. Vegas, Cerdmicacomuinromanadel Mediterrdneo

10For illustrations of the amphoriskos,see B.A. Sparkes occidental(Barcelona 1973) 173.
and L. Talcott, Agora XII. Black and Plain Pottery of the 14 FusifQrmexamples appear in Thompson's (supra n. 3)
6th, 5th, and 4th CenturiesB.C. (Princeton 1970) Pt. 2, pl. Deposit B (B6) and Deposit A (A64), now datedto the first
39, nos. 1147, 1150, 1159. In a paper deliveredat the 85th quarter of the third century B.C (Rotroff [supra n. 12]
General Meeting of the ArchaeologicalInstituteof America 107-108). In the Greek burials fromthe North Cemeteryat
in Decemberof 1983, S. Rotroff presentedresults of her re- Corinth (Blegen, Palmer, and Young [supra n. 2] 166),
cent research on the origin of the, characteristicAthenian however,where the latest gravesare datedto the late fourth
gray unguentarium("The Origins and Chronologyof Hel- and early third centuriesB.C., only examplesof the globular
lenistic Gray Unguentaria,"AJA 88 [1984] 258). It would type of unguentariumwere found (pls. 76-78).
appear that the type was derivedfrom Cypriot BichromeV '5As published,for example, by Forti (supra n. 1) passim;
amphoras. Higg and Fossey (supra n. 1) 29, 72; Goldman (supra n. 3)
' Parovig-Pegikan(supra n. 2) 109, fig. 93.1-2. 171-72; and Almagro (supra n. 2) 397.
12 For examples from Spain, see Forti
(supra n. 1) 146, n. 16 Forti (supra n. 1) 151; Almagro (supra n. 2) 397, nos.
13, and Cuadrado (supra n. 2) Group A, 390, 399; for the 21, 22, 35, 36; A.J.B. Wace and G. Dickins, "Excavationsat
Eastern Mediterranean,see Forti (supra n. 1) 116; Thomp- Sparta I, 1907, The Tombs,"BSA 13 (1908) 162, fig. 7; and
son (supra n. 3) B44 in Deposit B, now datedby S.I. Rotroff M. Andronikos,"HellenistikosTaphos Beroias,"AE 1955,
in AgoraXXII. Hellenistic Pottery:Athenianand Imported 34, fig. 15.
Moldmade Bowls (Princeton 1982) 108, to the years be- 17 Lapp (supra n. 2) 62, n. 41, disagrees on the basis of
tween 320 and 240 B.C.; Parovi&-Pe'ikan(supra n. 2) Form evidencefrom Palestinian material.
1, 108, 109, fig. 93.1-2; Schlorb-Vierneisel (supra n. 2) 18 A
typology of the Stobi unguentaria was presented in

fig. Ib, Type B; fig. Ic, g, Type G) are characterized other areas has a shorter body cavity, round or even
by either plump or angular bodies, and solid stemmed biconical. Examples of the short, globular body were
feet of varying height. found in Athens, and on Samos, and of the more angu-
In Greece, one of the latest varieties of the fusiform lar shape at Tarsus and Hama in Asia Minor, and in
shape has a long, slender profile (fig. 3).19 In his com- Macedonia (figs. Ic; 4).21
ments on unguentaria from Hellenistic tombs at It is not unusual, in fact, to find several different
Asine, Fossey suggests that the rise in the ratio of ves- forms of the fusiform unguentarium in contemporary
sel height to belly diameter during the course of the use. This is clearly seen at Ampurias where Types B
third and second centuries B.C. documents the devel- VI and VII, with slender bodies, are contemporary
opment from globular to slender body profile.20 with Types B III, IV, and V, with very round bodies.
Another variant common in Greece and also in Furthermore, the earliest forms of Type B, B I and B
II, popular over a long period of time, have a com-
pletely solid stem, elsewhere considered a late charac-
teristic.22 On Samos and in Macedonia as well, the
early variety of fusiform unguentarium with globular
body and small foot is found in contexts of the second
century B.C. (fig. la).23
Bruneau has noted that on Delos fusiform unguen-
taria with different fabrics change in profile at differ-
ent rates.24 Such a phenomenon has been observed
elsewhere as well.25 Types have often been formulated
on the basis of the single attribute of shape rather than
on an overall consideration of fabric, decoration and
shape. It seems clear that a uniform chronology for
Hellenistic fusiform unguentaria using the criterion
of shape alone is not meaningful.26
The apparent decrease in the absolute capacity of
the fusiform unguentarium over time has not been
verified through measurement, and is based on visual
inspection of vessels and profiles only. One result of
the reduced capacity of this vessel might have been to
save the consumer money or, even more likely, the
Fig. 3. Shape 1, slender fusiform unguentarium,Athenian manufacturer, since a smaller amount of oil or per-
Agora P628. Ht. 0.135 m. (CourtesyAgora Excavations) fume could be purchased.27 The lekythoi of earlier
my Ph.D. dissertation,Pottery of the Hellenistic and Early examples of fusiform unguentariafrom contexts of the sec-
Roman Periods at Stobi (University of Texas at Austin ond and first centuries B.C. at Stobi for which the ratio of
1977) 253-61. height to belly diameterwas calculated,no particular pat-
19For Athens, see Thompson (supra n. 3) D78, E138; tern was evident.
Schl6rb-Vierneisel(supra n. 2) Graves 199-201; for Argos, 21 Samos:Taskos (supra n. 2) 350, fig. 2, nos. 2154, 2193,
Bruneau (supra n. 2) 501, 516, 517; Haiggand Fossey (su- and 374, fig. 11; Tarsus: Goldman(supra n. 3) 172, pl. 187,
pra n. 1) 51. Note also the commentsof Drougou and Tou- and pl. 255; Hama: Christensenand Johansen (supra n. 3)
ratsoglou (supra n. 2) 123. 201-208, fig. 23; Macedonia: I have seen unpublishedex-
20
Hiaggand Fossey (supra n. 1) 29, 30, 72, 141, 142, and amples in the ArchaeologicalMuseum in Thessaloniki, and
Tables 3 and 5. Although such a trend toward increasing we have a numberof examplesat Stobi classifiedas Type G
slendernessmay be generally observablein some areas, it is (fig. lc).
not strictlyborne out by Fossey's figures for Athens and Ar- 22 Cuadrado(supra n. 2) 399, 401, fig. 1, dated ca. 200-75
gos in Appendix 3 (pp. 141-42) or for Asine in Tables 3 and B.C.

5. Furthermore,his dates for the unguentaria from Grave 23 Supra n. 21.
B42 do not reflectthe reviseddating of the ThompsonAgora 24 Bruneau
(supra n. 2) 254.
Groups A-E, necessitatedby the Koroni evidence. For the 25 Tsakos (supra n. 3) 345-47.
revisedchronology,see Rotroff (supra n. 12) 107-10. 26As noted also by Hellstrom
(supra n. 1) 26, and Tsakos
Since Fossey's comments do not describe vessels of the (supra n. 2) 409.
late second or first centuries B.C., the ratios from Stobi ex- 27 Christensen and
Johansen (supra n. 3) 45, also com-
amples may not be expectedto follow his scheme.For the 38 ment upon the advantageto the perfumebusiness.

times, with smaller interior chambers, provide an to a bulbous profile with flat base did not significant-
analogy.28 It would be interesting to compare the ca- ly affect the amount of interior space in the vessel
pacities of lekythoi to those of fusiform unguentaria. (cf. fig. 1c and e).
The matter of unguentarium capacity is discussed When and why did this change in shape from fusi-
below but only insofar as it relates to function. What form to bulbous occur? The date the bulbous unguen-
is needed are measurements of a series of fusiform un- tarium first appeared will probably never be known,
guentaria of the same type, the dates of which extend but the change in shape from fusiform to bulbous is
over several centuries, before a decrease in volume can documented at Argos, Corinth, and Stobi where the
be substantiated. two shapes are found together in single graves (fig. Ic
Shape 2, the bulbous unguentarium with flat rest- and e, found together in Grave 273 at Stobi and fig. If
ing surface, becomes popular during the first century and g, found in Grave 265; figs. 4-7).30 The dates as-
B.C., and continues in use in most areas until the signed to these graves fall within the second half of the
latter part of the first century A.C. (fig. Id, e, h, i).29 first century B.C.3' The bulbous shape appears at
The change from the fusiform shape with solid stem Corinth in Roman graves dated to shortly after 44
28 See drawing of lekythos with false interior chamber in

Tx-2505, was dated to 2137+94 years B.P., equal to
Boulter (supra n. 2) p. 123, fig. 4, and referenceswith Fl; 281-93 B.C. Neither of the Stobi graves contained addi-
Haspels (supra n. 4) 129, 176; Kurtz and Boardman(supra tional diagnosticmaterial to provideany other indicationof
n. 4) 103. date.
29Thompson (supra n. 3) 473; H.S. Robinson, Agora V. Analysis of charcoalfrom Graves 273 and 265 is part of a
Pottery of the Roman Period (Princeton 1959) 15, F50; Al- more extensive projectinvolving46 samples from a series of
magro (supra n. 2) 181; Vegas (supra n. 13) 153-54; Khairy graves in the West Cemetery. See the preliminaryreport in
(supra n. 1) 85-86; J.W. Hayes, Roman Pottery in the S. Valastro, Jr., and A.B. Wesolowsky, "The Carbon-14
Royal OntarioMuseum (Toronto 1976) 12, no. 47; 33, 34, Dating Programof the West Cemetery,"in B. Aleksovaand
nos. 156-58, 161. For examples of the bulbousformdated to J.R. Wiseman eds., Stobi III (Titov Veles 1981) 301-10.
the late first century A.C., see A. von Saldern, Ancient and See also the earlier study, E.M. Davis, D. Srdoi?,and S. Va-
Byzantine Glass from Sardis (Cambridge 1980) 25, where lastro,Jr., "RadiocarbonDates from Stobi: 1971 Season,"in
one example was found with a coin dated A.D. 54-68 in Wiseman and Mano-Zissi (supra n. 2, Stobi I) 23-24.
Grave 67.32, and another in Grave 67.34 with an Eastern In a vaultedchambertomb (Tomb 1) at Beroia in Mace-
Sigillata B II bowl dated A.D. 75-150. donia, both fusiformand possiblytwo early examplesof bul-
30 For Argos, see Bruneau
(supra n. 2) 481, Grave 61, 483. bous unguentaria (nos. 1261-1262) were found. The tomb
This grave which containedtwo fusiform (61.13, 61.14) and and its contents are dated by the authors to a period just
one bulbous (61.15) unguentariawas apparentlyused twice after the middle of the second century B.C. (Drougou and
since it containedone complete skeleton at the east, and, at Touratsoglou [supra n. 2] 125). The authors refer to the
the west, the skull of another. Each was accompanied by bulbousvessels as small lekythoi and do not cite parallels for
different offerings. The offerings with the skull at the west the shape. Grave 300 in the West Cemeteryat Stobi, with a
date that burial to the end of the second or beginning of the '4C estimated range of 241-87 B.C., contained a vessel
first centuries B.C. The offeringswith the completeskeleton (C-74-335) similar to Beroia no. 1262. Perhaps also related
on the east side include the three unguentaria;on the basis of to the bulbous shape are Beroia nos. 1460, a black gloss ex-
the one bulbous example and the "datation tardive" of a ample fromTomb 2, and 2135, fromTomb 3. The latterex-
lagynos (61.9) and a skyphos (61.10), this burial is dated to ample, with point of maximum diameterclose to the bottom
the second half of the first century B.C. The evidence for of the vessel, is very much like a gold-bandglass bottle illus-
Corinth is found in G.R. Edwards, Corinth VII, Pt. 3: Co- trated in S. Goldstein, Pre-Roman and Early Roman Glass
rinthian Hellenistic Pottery (Princeton 1975) 216, Deposit in the Corning Museum of Glass (Corning, N.Y. 1979) 31,
77, with three Early Roman "onion"unguentaria and an no. 556, dated to the middle of the first century B.C., or the
imported unguentarium of Hellenistic form, suggesting a early first centuryA.C. (p. 203). This shape has been classi-
date in the last quarter of the first century B.C. For Stobi, fied by Isings (C. Isings, Roman Glass from Dated Finds
see J.R. Wiseman and D. Mano-Zissi, "Excavations at
[Groningen/Djakarta 1957]) as Form 7 (p. 23), a shape she
Stobi 1973-1974,"JFA 1 (1974) 135-37, figs. 17-18. Grave notes was also made in colored glass. One of the datable
273 in the West Cemetery contained three bulbous exam-
specimens she cites comes from a second century B.C. con-
ples, C-74-114, 115, 116, and eight fusiform examples, text. Her date for the piece, however,is the Nero-Vespasian
C-74-109 to C-74-113, and C-74-189, 218, and 219. Grave period.
273 retained charcoal from the grave fire; according to
32The dates for the gravesat Argos and Corinth, however,
Wesolowsky, a sample (Tx-2510) was dated-using the as well as those for many other gravesat sites that have been
5730 half-life, a dendrochronologicalcorrectionaccordingto
mentioned,are not secure,often having been determinedby
the Arizona calibration,and a 12C/13C ratio correction-to referenceto the presenceof the bulbous shape, datedat most
2183194 B.P., or 327-139 B.C. Grave 265 contained two sites, as we have stated earlier, to the second half of the first
fusiform unguentaria, C-74-102, 104, and two bulbous ex-
century B.C. and first centuryA.C.!
amples, C-74-103, 105. A charcoalsample from Grave 265,

It has been suggested that the discovery of glass-

blowing around 50 B.C. may have provided an im-
petus for the change to the bulbous shape in imitation
of new blown-glass flasks.34 Some of the earliest ce-
ramic bulbous unguentaria are small (6-10 cm. in
height) with a slim, short (or, less frequently, long)
neck, and a low point of maximum diameter, similar
to the shape of early glass unguentaria.35 Three ex-
Fig. 4. Shape 1, biconical fusiform unguentarium, Stobi

C-74-109 from Grave 273. Ht. 0.210 m.
B.C., and evidence from Israel and Jordan indicates a

date within the same period.32 At Tarsus a bulbous
unguentarium was found in the Hellenistic-Roman Fig. 5. Shape 2, bulbous unguentarium, Stobi C-74-114
unit, dated ca. 150-50 B.C.33 from Grave 273. Ht. 0.076 m.
32 For Corinth, see Blegen, Palmer, and Young (supra n. and alabasterare cited by Hellstrom (supra n. 1) 24, n. 2,
2) 167. The absence of bulbous unguentaria in graves at and Thompson(supran. 3) 474, n. 2. There are glass exam-
Corinthdatedto beforethe destructionby Mummius in 146 ples as well (Goldstein[supra n. 30] 176-77, pl. 21, and p.
B.C. does not mean that the shape could not have appeared 39, fig. 11). The horizontalstriped decorationon some ce-
in the interval.Bulbousunguentariado not appearin depos- ramic fusiformunguentariais reminiscentof decorationon
its of the Athenian Agora beforethe late first century B.C. glass bottles such as the amphoriskosand alabastron(Gold-
(Robinson,supra n. 29). Potteryfromthe first centuryB.C. stein [supra n. 30] 124-32 and pls. 15, 17, for "traildecora-
deposits has not been published.There are no bulbous un- tion"on alabastra,aryballoi,and amphoriskoi).See also su-
guentaria in Thompson's Group E (supra n. 3), dated by pra n. 29.
Rotroff(supran. 12) to the secondhalf of the secondcentury 35 For examples of the early shape in blown glass bottles,
B.C. For Israel and Jordan, see commentsin Khairy (supra see Isings (supran. 30) 22, 23, Form 6, and 40, 41, Form26;
n. 1) 85-86. FortunatiZuccala,NSc 33 (1979) 17, Grave IV, 20; Grave
33Goldman(supra n. 3) 200, no. 730, 270. See Oliver (su- VI, 4, 10, 30, 31; GraveXIV, 4, 35; GraveXVIII, 5, 6, 7; P.
pra n. 2) 250, on the date of this context. Agallopoulos, "Xronika-Achaia-Patrai,"ArchDelt (1973-
34Vessbergand Westholm (supra n. 2) 68, 80; McFadden 1974) 369, pl. 234, with early glass unguentaria.A denarius
(supra n. 2) 480, no. 73; Schlirb-Vierneisel (supra n. 2) of Augustus was found with six pear-shapedglass unguen-
108, n. 11. On the inventionof blown glass, D.B. Harden, taria in the AntikenmuseumBerlin from a tomb at Priene
"AncientGlass, II: Roman,"ArchJ 126 (1969) 47. Exam- (Gertrud Platz-Horster, "Zu Erfindung und Verbreitung
ples of fusiformunguentariain other materialsuch as silver der Glasmacherpfeife,"JGS 21 [1979] 27-31). Additional

amples of this shape in clay were foundat Stobi in the

same grave (273) with eight fusiformunguentariaof
angular profile and solid stems (figs. ic, e; 4, 5).36
Other examples with shorternecks were found in the
Kerameikoscemeteryin Athens, Sardis, Samos, Stobi
(fig. 1h), and Palestine.37
Fig. 7. Shape 2, bulbous unguentarium, Stobi C-74-105

from Grave 265. Ht. 0.164 m.
In general, the earliest examples share a low point

of maximum diameter, clear in unguentaria from Ar-
gos and Ampurias.38 Some taller examples (12-16
cm.) of this shape-which John Hayes refers to as a
"sagging profile"-have ovoid bodies and often appear
Fig. 6. Shape 1, fusiform unguentarium, Stobi C-74-102 to imitate the very thin vessel walls and smooth, shiny
from Grave 265. Ht. 0.185 m. surfaces of glass (figs. If; 6).3"
evidencefrom the Jewish quarterof Jerusalem (N. Avigad, 181, top row.
"QuartierJuif," RBibl 80 [1973] 579) suggests to Platz- 38For Argos, see supra n. 30, and for Ampurias, see Al-
Horster that free blown glass unguentaria may have been magro (supra n. 2) 118. Unguentariaclassifiedas Group 1
producedearlier than previouslythought. The glass bottles from the North Cemeteryin Corinth are thought to be the
may not have becomecommon,however,until the Augustan earliest;they have full, roundbodies,and do not fit our clas-
periodwith its favorableeconomicconditions. sification.The statementis made, however,that the "differ-
36 Supra n. 30. ence in shape may be the result of a differentcenterof man-
37Kerameikos,Schlbrb-Vierneisel(supra n. 2) 110, Grave ufacture ratherthan a chronologicalfeature"(Blegen, Pal-
202, Nos. 1 and 1.62, dated Late Hellenistic or earlier, and mer, and Young [supran. 2] 167). Severalbulbousunguen-
p. 110, n. 11; Sardis,Greenewalt(supran. 2) 14, fig. 16, top taria found in a well near the cavea of the Theater at Cor-
row P77.7, P77.43B, P77.43A found in close association inth and dated to the last quarter of the first century B.C.
with four fusiform unguentaria in chamber tomb 77.1. I (T.L. Shear, "Excavationsat Corinth in 1928," AJA 32
owe this informationaboutcontextto Prof. Greenewaltwho [1928] 485) may better fit our classification.For Palestine,
generouslyallowedme to see the pertinentinformationfrom see Kahane (supra n. 2) 176, 179, Variant B, pl. 12.
excavationnotebooks;Samos, Tsakos (supra n. 2) pl. 124, 3" Hayes (supra n. 29) 62, no. 341. In Grave 265 at Stobi
no. 629 in Grave a7 of the West Cemetery,and pl. 141, no. (supra n. 30), two very thin-walled bulbous vessels with
2033 in Grave Bakente VII; Stobi, C-74-128, from Grave smooth surfaces were found with two angular, solid-
283, with 14C estimate of 214-26 B.C.; Palestine, Hayes stemmed, fusiform examples. Parallels for both of these
(supra n. 29) 63, pl. 37, no. 346; and Almagro (supra n. 2) shapes exist at Tarsus. No. 730, (pp. 200, 270), a bulbous

Ceramic bulbous unguentaria are common in the presumably, less costly, pushing ceramic imitations
late first century B.C. and early decades of the first almost completelyout of the marketduring the course
century A.C., in contrast to glass unguentaria which of the first centuryA.C.42
are not found in quantity until the second half of the By the late first century and during the second and
first century A.C.40 The rarity of the blown-glass un- third centuries, in areas such as Thrace and Cyprus
guentaria at many sites may reflect their value. Even- where the ceramic bulbous unguentarium continued
tually, in graves of the first century A.C., ceramicand in use, the vessel shape had changed considerably.43
glass unguentaria of bulbous shape are frequently Characteristicof that period was a bell-shaped body
found in the same context, but I do not know of any with the point of maximum diameterat the base and a
instances of ceramic fusiform and blown glass un- very long neck, probably in imitation of the "candle-
guentaria in contemporary contexts. Thus the se- stick" unguentarium shape in glass, popular in the
quence in unguentaria-from ceramic fusiform to ce- late first and secondcenturiesA.C.44
ramic bulbous to glass examples-seems clear as far In the cemeteriesof settlements establishedduring
as grave goods are concerned,despite the chronologi- the early decadesof the first centuryA.C., ceramicun-
cal precedenceof the techniqueof glassblowing.41 The guentariaare rare;blown glass unguentaria,however,
blown glass vessels soon became quite common and, are found in quantity.45At sites dating to the period
unguentarium, is describedas having unusually fine fabric taria in the same context with glass bulbous unguentaria
and surface treatment, as do the Stobi examples. The fusi- suggest that ceramic bulbous unguentariawere: 1) not imi-
form examples find parallels in nos. 225, 239, 243, 247, and tating a new blown glass shape, but came into existence ear-
F, on fig. 187. Also see supra n. 21. A similar bulboustype is lier and independentlyas a result of unknown influences;2)
identified by Hayes (supra n. 29, p. 33) as Campanian in not imitating blown glass unguentaria (dated post-50 B.C.
origin. See also Kahane (supra n. 2) 1176, 1178, Variant a, by most), but an earliershape of glass bottle (like the arybal-
p. 12, but without thin walls and smooth surface. los) with a flat base rather than a round one, a result of its
40 Harden
(supra n. 34) 49; J.W. Hayes, Roman and Pre- manufactureon the wheel; or 3) imitating blown glass un-
Roman Glass in the Royal Ontario Museum (Toronto 1975) guentariaof a date earlier than is now known. Isings (supra
42, 43, 146; J. Carrington-Smith, "A Roman Chamber n. 30, p. 1) cites a blown glass unguentariumfrom an Etrus-
Tomb on Monasteraki Kephala," BSA 77 (1982) 274. At can tomb group from Toscanella dated to the secondor first
Stobi in graves of the West Cemetery, ceramicbulbous un- centuries B.C. See also Harden (supra n. 34) 47, n. 18. Few
guentaria appear to be earlier than glass unguentaria. scholars, however, would agree with suggestion3.
41 From the
preceding it may seem that I have presented 42 Robinson (supra n. 29) 15, under F50; Kahane (supra
contradictoryevidence regarding the chronologyof bulbous n. 2) 180, 181.
unguentaria. There is persuasive evidencethat bulbous un- 43 Thrace: P. Pandos, "Xronika-Thraki, Nea
Orestias,"
guentaria may appear as early as the second century B.C. ArchDelt 29 (1973-1974) 823, pl. 611; L. Getov, "Sepul-
There are, first of all, the four bottles from the second cen- tures tumulaires thraco-romainesde la region de Kazan-
tury B.C. vaulted chamber tombs at Beroia (supra n. 30) lik," (in Bulgarian with French summary)Arkheologiia11
and a similar example from Stobi. The three examples at (1969) 42, 43, 45, fig. 16, dated to the middleor secondhalf
Sardis (supra n. 37) called flasks, from chambertomb 77.1, of the secondcenturyA.C.; Cyprus: Hayes (supra n. 29) 35,
could belong with the other objects deposited in the second nos. 166, 167, and pl. 20.
century B.C. 44 Isings (supra n. 30) 97, Form 82.
Schlorb-Vierneisel (supra n. 2) 108, n. 11, supports a 45 For the most part these comprisethe non-Hellenized or
date somewhat before 50 B.C.; she assigns Grave 202, with non-Italianized areas added to the Roman Empire during
one example of the early form similar to those at Sardis, to a the Augustan or Tiberian period: Raetia, Noricum, Panno-
Late Hellenistic date or earlier (p. 119). Three graves at nia, Moesia, and Dalmatia (inland), excluding, of course,
Stobi-two of which contained both fusiform and bulbous Roman military camps. For example, at Emona, one un-
shapes (Graves 265 and 273) and one which contained a guentariumis illustratedin the volume by L. Plesniiar-Gec,
single bulbous unguentarium (Grave 283)-produced rath- Keramika Emonskih Nekropol (Ljubljana 1977) 59, 88, pl.
er early estimates for the age of the wood burnedin the cre- 9.5; and no examples appear in Severno Emonskih Grobizce
mation or grave fire (supra n. 30). The radiocarbonestimate
(Ljubljana 1972) by the same author. In a short but useful
may not, however, reflect a calendrical date contemporary publication entitled Roman Pottery in Slovenia (Ljubljana
with the finds. Still, the closest parallels for the fusiform 1973), publishedby the National Museum, unguentariaare
shapes in Graves 265 and 273 are found at Hama (Chris- not includedamong the many shapes illustrated.At Doclea,
tensen and Johansen [supra n. 3] 45, fig. 23), dated to the on the coast of Montenegro, in the earliest burials dated to
period between the mid-secondand mid-firstcenturiesB.C., the first decades of the first century A.C., no ceramic un-
and at Tarsus (Goldman [supra n. 3] fig. 187), dated to the
guentaria were found, but numerous glass examples were
same period. found. See A. Cermanovi&-Ku'manovi',D. Srejovic,O. Ve-
Furthermore,the fact that ceramic bulbous unguentaria The Roman Cemetery at Doclea (Cetinje
are commonin burials at an earlier date than glass unguen- limirovi&-iiic,
1975) 311, 313, and passim for examples of glass bottles.
taria, and that one does not find ceramic fusiform unguen-

after the invention of glassblowing and the inaugura- the household. Perhaps they were placed against a
tion of the Pax Romana, we should expect more glass vertical surface, or even allowed to lie horizontally as
vessels to be available, transported from centers of in a burial. Some kind of stopper must have been nec-
production in Italy and the East, as part of the expan- essary to keep the contents from evaporating, or escap-
sion of commerce in the Augustan period. ing when the vessel was in a horizontal position. Stop-
APPEARANCE AND FUNCTION
pers for unguentaria have not been found but stands
and stoppers may have been made of perishable mate-
As stated earlier, unguentaria vary considerably in rial, like wax or cork.s0 Lead stoppers for glass vessels
appearance, some being of very high quality like the are relatively common."'
fusiform examples in Athenian or Italian black gloss How were fusiform unguentaria used? Unlike ala-
which may have been exported widely throughout the bastra, aryballoi, or flasks which cannot stand alone,
Mediterranean and other parts of the Roman Empire. most unguentaria, and especially the poorly made ex-
The bulbous unguentaria of the Roman period were amples, have no handles. It is possible that these ves-
not produced in any of the early Roman red gloss sels were suspended from the individual's wrist or
wares, such as Arretine or Eastern Sigillata B. Most from a hook by means of a strap or cord tied around
unguentaria, especially those from graves, were not the neck of the vessel, and held there by the projecting
carefully made; surfaces are frequently rough and un- rim, although the lip on many examples is quite small.
slipped, and profiles asymmetrical.46
Many fusiform unguentaria are decorated with a
slip applied in horizontal lines or diagonal stripes
reminiscent of glass examples, while others have slip
or gloss which covers the entire surface.47 On some
fusiform examples and most bulbous types, the neck is
decorated with a dark slip, usually fired brown or red,
contrasting in color with the fabric. The drip-marks
on the body of many unguentaria suggest that the neck
was dipped into a large container containing the slip
and then placed upright to dry (fig. 8).
The spreading ring foot of early examples of the
fusiform shape provided a firm base for the vessel.
Many fusiform unguentaria, however, were made
with a very small and uneven base and could not stand
without a tripod or other means of support.48 Al-
though no stands for unguentaria have been found,
several rings have been published in connection with
amphoriskoi.49
The absence of stands, and the fact that few of the
fusiform vessels would have been able to stand alone
(and would have rolled about if laid down), make it Fig. 8. Shape 2, bulbous unguentarium, Stobi C-71-168
difficult to imagine how unguentaria were stored in from Grave 95. Ht. 0.067 m.
46 See recent comments by Tsakos (supra n. 2) 409. In 1950," Hesperia 20 (1951) 50-51, pl. 25.2, illustrates a
published discussions of unguentaria, examples from vari- wine bottlefound with a cork,and cites referencesin ns. 4-8.
ous contexts and of various types tend to be considered 51 A lead stopperfor a glass aryballosis illustratedin Car-
together. rington-Smith(supra n. 40) 282, no. 77, and pl. 40a, b. See
47Among those with gloss are examples of Campanian the forthcomingarticle by C.G. Koehler, "The Handling of
ware in J.-P. Morel, CUramique campanienne: les formes
Transport Amphoras," Proceedings of the First Internation-
(Rome 1981). al Colloquium on Greek Transport Amphoras, Athens, Sep-
48 Khairy (supra n. 1) points out that the spindle shape tember, 1984 (BCH Suppl.). George Bass mentionsthe use
does not have the proper balance for standing. of driedfigs as stoppersfor flasks:"ABronzeAge Shipwreck
49 Sparkesand Talcott (supra n. 10) 43, 156, 180. at Ulu Burun (Kay): 1984 Campaign,"AJA 90 (1986) 285.
so H.A. Thompson, "Excavationsin the Athenian Agora,

Both fusiform and bulbous shapes, however, are most transported in bulk, in large vessels, and later trans-
suitable for holding in the hand.52 ferred to smaller containers.56
It has been suggested that unguentaria were pro- In order to establish a typology, analyses of clay
duced exclusively by a small number of workshops, composition are needed to determine provenience for
perhaps allied with perfume or unguent manufactur- different fabrics." It is probable that, as in the typical
ers who used the bottles as containers for shipping ceramic assemblage of any site, there is a mixture of
their products throughout the Mediterranean.53 Such imported and locally produced vessels. Some unguen-
a reconstruction may be applicable to some of the taria clearly were produced locally, such as those ex-
finer, better-made bottles, but most ceramic unguenta- cavated in kiln contexts at Gordion and other sites."s8
ria hardly seem worth transporting, nor are they fine In a potter's establishment at Stobi, nine fusiform ex-
or distinctive enough to serve as a trademark for any amples of the same fabric were found in destruction
manufacturer of such a costly substance.54 Further- debris dating to the mid-first century B.C.59"
more, the wide variety of fabrics and shapes present at Let us return to the question of the function of these
various sites does not support the idea of only a few small bottles. Further consideration of their appear-
centers of production.55 Aside from the general simi- ance, manufacture, size, and shape may provide infor-
larities of shape, the only characteristic shared among mation about their use.60 The closed shape and nar-
large groups of unguentaria is the horizontal striped row mouth of the unguentarium are more appropriate
decoration on fusiform examples and the slipped neck attributes for a container of liquids such as wine or
on bulbous types. Considering the cost of transport water, or viscous substances such as oil or honey, than
and the danger of the neck breaking off from the body for solid perfume or unguent which would have to be
of the unguentarium, it seems more likely that, in extracted with a small, long-handled spoon or, per-
most cases, the contents of the small bottles were haps, by breaking off the neck.6' The small opening
52 Perhaps unguentaria were used as candle holders. Note of Bronze Age Pylos large stirrup jars were used for the
the comments of R.J. Forbes, Studies in Ancient Technology transportof oil in bulk while small stirrupjars containedthe
6 (Leiden 1966) 140. perfumed oil producedin the palace workshops (The Per-
" In many of the published comments on unguentaria, fume Industry of Mycenaean Pylos [SIMA Pocketbook 34,
there is a tendency to simplify complex mechanisms and to Goteborg 1985] 145).
look for a few centersof productionand a single purpose for In the Late Roman period,there is evidencethat unguent
these vessels. See Thompson (supra n. 3) 472-73; Goldman was transportedin the familiar brown micaceousjars (Car-
(supra n. 3) 171; Edwards (supra n. 30) 99; and Khairy thage Late Roman Amphora 3). See M.C. Rothschild-Bo-
(supra n. 1) 88, who states that the Nabataeans exported ros, "The Determination of Amphora Contents," in G.
bottled perfume in their vessels to Europe, yet cites no ex- Barker and R. Hodges eds., Archaeology and Italian Society.
amples of such unguentaria actually found in Europe. Both Papers in Italian Archaeology 2 (Oxford 1981) 85-86. I
Hellstrom (supra n. 1) 24 and Bruneau (supra n. 2) 526, know of no unguentariarecoveredas cargo fromshipwrecks.
528, have recognizedthe complexity of the situation in their 57 Petrographic analysis of an unguentarium fragment
discussions. from Stobi has revealed a composition indistinguishable
54 An analogy might be the trademarkbottles used from that known for local products.
today
by perfume manufacturerssuch as Guerlain or Nina Ricci. 18For Gordion, see G.R. Edwards, "The Gordion Cam-
11In
general, see Kahane (supra n. 2) 137-39. The diffi- paign of 1958,"AJA 63 (1959) 267, pl. 68, fig. 21. Also see
culty of assigning a definite centerof productionfor any par- Edwards (supra n. 30) 98, 99; Bruneau (supra n. 2) 523;
ticular ware is evident in a considerationof the slender,gray
H~gg and Fossey (supran. 1) 86; ChristensenandJohansen
unguentaria found on many Greek sites: Thompson (supra (supra n. 3) 45; Tsakos (supra n. 2) 413; Smetana-Scherrer
n. 3) assigns them to the East, perhaps Syria, Edwards (supra n. 3) 89.
(supra n. 30, p. 99) calls them imported, and Hellstrom "9The vessels were found beneaththe narthex of the Cen-
(supra n. 1, p. 25) Athenian! Rotroff (supra n. 10) is now tral Basilica. See J.R. Wiseman and D. Mano-Zissi, "Exca-
engaged in a study of this problem. vations at Stobi, 1971," AJA 76 (1972) 409, 415, pl. 86, fig.
56 Hellstrom (supra n. 1) 24; Sparkes and Talcott (supra 2; and Anderson(supra n. 18) 25-31. The variationin pro-
n. 10) 191, n. 27; Goldman (supra n. 3) 171, n. 64; and E.L. files within this single series, with the same fabric, empha-
Will, "The Ancient CommercialAmphora,"Archaeology30 sizes the need for caution when establishing a chronology
(1977) 265. The wall painting of cupids in the House of the based on shape alone.
Vetii in Pompeii (A. Mau, Pompeii. Its Life and Art [Lon- 60 Vessel function is commonly derived from vessel
form,
don 1904] 333, 334, fig. 165) illustrates (presumably) per- since we seldom have more direct evidence.
fumed oil stored in large vessels, sold by weight, and then 6' See commentsof Kahane (supra n. 2) 131. In an account
transferredto a small container. of the annointingof Jesus at Bethanyin the New Testament
Cynthia Shelmerdinesuggests that in the palace economy (Mark 14:3), an alabastron is broken (o-vvrpl/ao-a) to re-

allows the substance to be poured out very slowly. Yet red ochre, and perhaps used as a cosmetic.6" The con-
the soft and porous clay fabric and unslipped surfaces tents, therefore, of ceramic unguentaria might also
of many poorly made grave unguentaria would proba- have ranged from liquid to a powdered or granular
bly not have retained liquids for any length of time substance like incense. Pliny, in fact, refers to sprinkl-
without a lining of some kind. Both water and oil ing powders made of dried scents (HN 13.3.19). The
seeped through two well-made but unslipped exam- porosity of the clay fabric would only have been a fac-
ples from Stobi within eight hours.62 Resin or wax tor, of course, when liquids were involved.66
may have been used but I am not aware of any exam- The representations of unguentaria on grave reliefs
ples of unguentaria with a preserved lining. reflect only their function as offerings. The vessels are
Rotroff reports that while testing the porosity of most frequently shown on a shelf at the upper back of
unguentaria from the Agora Excavations, an unglazed the relief, behind the deceased, or, less frequently, are
gray unguentarium (probably ca. 300 B.C.) lost only held by a person other than the deceased, usually a
20% of its water in five days and only 5.5% of its con- child.67 Clear reference to the use of the unguenta-
tents in a week when tested with oil. In contrast, the rium cannot be identified in ancient literature.
black-glazed examples belonging to the same period Was, in fact, a ceramic unguentarium intended as a
were more porous, losing 36-100% of their water in grave offering and used for the storage of some costly
only two days. Of course, the porosity of any vessel substance? Were funerary unguentaria different from
would have been influenced by a number of factors those used for the needs of daily life?
including clay composition, manufacture, and firing In the publication of the Hellenistic necropolis at
temperature.63 Additional tests of intact vessels with Asine, Fossey suggested that the unguentaria in bur-
thick gloss, slip, and without slip, from various sites ials at Asine reflect some standardization in capacity.68
and cemeteries, should yield additional information. Are we to assume that such unguentaria represent
The contents of several glass unguentaria from Brit- commercial containers? Such an assumption also im-
ain have been preserved and after analysis were found plies that unguentaria for household use and for funer-
to consist of resinated wine in one case and frankin- ary purposes were manufactured in the same manner.
cense in another.64 A Roman chamber tomb at Knos- A close look at capacities of additional excavated
sos yielded a glass unguentarium with a pink powder groups of unguentaria should reveal the extent of
preserved on the inside; the powder was derived from a standardization.69 Such an opportunity will be limited
lease the perfume of nard which is then poured (Ka7TEXE'v) diver and C.G. Koehler.
onto Jesus' head. The account in Matthew (26:7) does not 64 See
J.P. Alcock, "ClassicalReligious Belief and Burial
mention the breaking of the alabastron. It is difficult to de- Practicein Roman Britain,"ArchJ 137 (1980) 62. Analysis
termine the shape of the vessel referredto in these accounts, of an oily material in a glass unguentariumfrom the Sardis
or in the third (John 12:3), where the oil was used for Jesus' region suggests that the original contents may have been
feet. In the latter account, no vessel is mentioned, but only olive oil. See C. Greenewalt, jr. et al., "The Sardis Cam-
the amount as being a litran (X-Mrpav) or one Italic kotyle paigns of 1981 and 1982,"BASOR Suppl. 23 (1985) 84-85,
(ca. 12 oz.). I agree with Hellstrom (supra n. 1) 24, n. 3, 25, with references to two other glass unguentaria with oily
that since the great value of the perfume is emphasized in contents.
the account in Mark, the alabastron may have been of ala-
65 See Carrington-Smith(supra n. 40) 293, who cites addi-
baster. Pliny (HN 13.3.19) notes that unguents keep best in tional examples. Also P. Astr6m,"Two Unguentaria and an
alabasterboxes and that they are commonlystoredin vessels Obol,"OpAth 7 (1964) 187-90, and by the same author, "A
of lead. Red Lustrous Wheelmade Spindle Bottle and its Contents,"
62 Both were bulbous examples, one tested with
water, the MedMusB 5 (1969) 16-21. I am grateful to Susan Rotroff
other with oil. for these two references.
63 On porosity studies see F.P. Matson, "PorosityStudies 66 Most vessel shapes would have been suitable for a vari-
of Ancient Pottery," Papers of the Michigan
Academy of Sci- ety of tasks and we need not assume that the unguentarium
ences, Arts, and Letters 26 (1940) 469-77. In her studies of had only one function.
Corinthian Type A and B amphoras, Koehler has discov- 67 E. Pfuhl and H.
ered that the differencein manufactureof the two types re- M6bius, Die bstgriechischenGrabre-
liefs (Mainz am Rhein 1977) nos. 384, 507, 538, 892, 909,
sulted in fabricsof differing porosity.See P.B. Vandiverand
924, 989, 922 all on a shelf; nos. 405, 406, 410, in the hand.
C.G. Koehler, "The Structure, Processing, Properties and 68 Supra n. 1, 72-73; infra n. 76.
Style of Corinthian Transport Amphoras,"in W.D. King- 69 Recent work on standard
vessel capacities is being car-
ery ed., The Technology and Style of Ceramics (Ceramics ried out by M.B. Wallace of the University of Toronto. See
and Civilization, Vol. 2, Columbus, Ohio 1986); and C.G. P.M. Wallace Matheson and M.B. Wallace, "Some Rho-
Koehler, Corinth VII, Pt. 5, Corinthian Transport Ampho- dian Amphora Capacities," Hesperia 51 (1982) 293-320.
ras (Princeton) forthcoming,with Appendix I, "The Manu- Also see V. Grace, "StandardPotteryContainersof the An-
facture of Corinthian Transport Amphoras,"by P.B. Van- cient Greek World" (Hesperia, Suppl. 8, Princeton 1949)

to vessels in burials inasmuch as most intact examples capacities among bottles that to the eye appear similar
come from cemeteries. Unguentaria of generally poor does not suggest any attempt at standardization.72
quality found in graves are less likely to have been The capacity of the Stobi fusiform shapes ranges
made according to a commercial standard than un- from 85 ml. (measured with rice) for an early example
guentaria of export quality. (C-74-343) to 10 ml. for one of the latest examples
In order to test this assumption, I measured the ca- (C-71-85).73
pacities of a series of fusiform and bulbous unguen- Of the six examples of bulbous Type A measured,
taria from the cemeteries at Stobi. Since many vessels, no groupings are apparent although three examples
although complete, are mended at the foot or neck, our are within a few milliliters of one another. In general,
sample of measurable examples was limited to 35. the bulbous examples show a somewhat greater range
Each vessel was filled to the top with rice. Then the than the fusiform with the smallest at 10 ml. and the
contents were poured into a plastic graduated cylinder largest close to 190 ml.74
and the amount noted.70 The same procedure was Another group of unguentaria of fusiform shape
then repeated, using water instead of rice. (Those ves- from one of a series of burials in Lenormant Street,
sels with cracks were not measured with water.) The Athens, has been measured by Rotroff and Wallace at
results of these measurements are given in Table 1. the Agora Excavations.75 It is difficult to see mean-
The figures for vessel capacity measured with rice are ingful groupings in the capacities of the 25 vessels
not actual measurements because of the dispropor- (Table 2) measured most recently by Rotroff, aside
tionate packing that occurs when a dry substance is from a possible group of nine with a capacity ranging
poured."7 These figures are cited here to give an idea from 19-22 ml.76 In the Hellenistic burials at Asine,
of relative size range where the sample would other- some of the fusiform unguentaria were measured with
wise have been very small. water-two from Burial 39 and nine from Burial
Eleven examples of Type G, a fusiform shape dated 42.77 They range from approximately 6-43 ml. in
to the first century B.C., show a range in capacity Burial 42, with an example at every few milliliters.78
from 31 to 58 ml. when filled with rice, and from 34 to Although one hesitates to base conclusions on such
60 ml. when filled with water. There is some cluster- a small sample, there does not seem to be evidence for
ing apparent at 34 ml. and at 40 ml. in the groups standardization of capacities in unguentaria intended
measured with water; less clustering is apparent for burial. The lack of standard capacity implies that
among those measured with rice. We are, of course, these bottles were probably not made by the potters as
dealing with a very small sample. Still, such a range in containers for a standard amount of oil or perfume,
175-89. D.J. Pullen has also worked with vessel capacities: 74 The ratio of belly diameter to height for bulbous un-
"ProgressReport on the Ceramic Vessel Volume Study at guentaria does not seem to follow any particularpattern, at
Halieis, Greece,"AJA 85 (1981) 212-13. least in the group of 72 examples at Stobi for which it was
70 These measurementswere taken by the author and calculated.
by
Malcolm Wallace using only those vessels that were without 75All are from Grave K, publishedby C. Boulter (supra n.
cracksand able to stand alone. When possible,each example 2). Measurements were taken by M. Wallace and S. Ro-
was measured by filling twice to the top with water and/or troff. Each unguentariumwas measuredfive times with wa-
twice with rice using a plastic graduatedcylinder. ter. The vessel was filled to maximum capacity with me-
71 Dry figurescan only be reliable as actual capacitiesafter niscus and the contentsthen poured into a graduatedcylin-
an equivalent for dry to wet has been established for each der. I would like to thank Drs. Rotroff and Wallace for
general vessel shape and size by remeasuringthe vase sever- sharing the results of their tests with me.
al times to determine the amount of rice equivalent to the 76 Rotroff, in fact, measured all the measurable unguen-
amount of water used to fill the vase. I am grateful to Caro- taria at the Agora (83 pieces, in additionto those from Le-
lyn Koehler for this information. normantStreet) with water in the methoddescribedin n. 75,
72 A kind of
template might have been used for controlling each a numberof times, but was able to see no patternin the
the exterior shape of the bottle. To reproduceconsistently resultingcapacities.
the same interior space, however, time after time on the 77Hagg and Fossey (supra n. 1) 147-48.
wheel, would have been more difficult. That unguentaria 78Fossey has seen a relationship between these measure-
were wheelmade is clear from the ridges usually left on the ments and the Athenian standardsof the kyathos and cubic
surface and the characteristicmarks at the base where the finger where each variation in unguentarium capacity is
vessel was removedfrom the wheel with string. seen as a proportionor multiple of these, thus implying a
73 C-74-343 is from Grave 295, West Cemetery, with 14C scheme of liquid capacity measures. His conclusions,how-
date ranging from 150-25 B.C., and C-71-85 is from the ever, are based on a rather small sample (Higg and Fossey,
destructiondebris (LOT 132) in the Potter'sWorkshopbe- supra n. 1, 72-73).
neath the narthex of the Central Basilica (supra n. 59).

Table 1. Stobi Unguentaria Capacities
Grave Inv. No. Date Type Form Capacity Ht. Belly Ratio
(ml.) (mm.) Diam. Ht.:Diam.
H20 Rice (mm.)
295 C-74-343 150-25 B.C. Fusiform B 85 159 061 2.603

288 C-74-349 .. D 30 154 042 3.695
84 C-74-19 " " D 50 170 046 3.665
S.B. C-71-85 ... F 10 134 039.5 3.390
265 C-74-102 281-93 B.C.* " G1 55/56 52/53 185 054 3.420
265 C-74-104 " G1 60 58 197 057 3.460
273 C-74-109 327-139 B.C.* " G2 34 30/31 210 052?
273 C-74-110 .. G1 41 40 188 055 3.412
273 C-74-111 ... G1 41 40 188 056 3.360
273 C-74-112 " " G2 34/35 31/32 210 056 3.745
273 C-74-113 ... G2 34/35 30/31 220 060 3.670
273 C-74-189 .. G2 40/41 38 200 055 3.645
273 C-74-218 ... G1 48 43/44 188 053 3.548
273 C-74-219 " " G1 41 38 185 050 3.695
303 C-75-92 25 B.C.-25 A.C. " G, 43/44 174 053 3.285
265 C-74-103 281-93 B.C.* Bulbous A 121 162 068 2.380
265 C-74-105 " " A 188 164 067 2.448
273 C-74-114 327-139 B.C.* " I 20 076.5 039 1.995
273 C-74-115 I 14 065 034 1.915
309 C-75-59 ? " A 185 180 145 070 2.070
277 C-74-168 60 B.C.-94 A.C.* " A 194 177 065 2.722
277 C-74-178 ... A 154 161 065 2.475
277 C-74-181 A 170 171 063 2.720
277 C-74-172 25 B.C.-25 A.C. " B1 90 139 061 2.381
277 C-74-174 ... B2 75 129 056 2.302
264 C-74-22 C 60 119 051 2.330
259 C-74-15 . F 30 125 043 2.908
94 C-71-162 F 53 126 049 2.570
18 C-70-65 " " G 25 102 041 2.484
18 C-70-78 ." G 45 114 048 2.375
76 C-71-96 .. L 25 091 054 1.685
95 C-71-171 .. N 10 063 042 1.500
95 C-71-174 . 0O 25 081 045 1.800
94 C-71-161 " O 45 085.5 050 1.708
21 C-70-59 50-100 A.C. P 30 072 055 1.308
* radiocarbonestimated
range
but were purchased empty to be filled as the need In some cases it may have been a commonpractice
arose."9 If it should ever become possible to measure to use vessels already present in the household, per-
the volumes of a sizeable sample of vessels from a non- haps the propertyof the deceasedor of one of the fam-
mortuary context, perhaps we might determine ily or friends, rather than to buy new unguentariaes-
whether standard volume was characteristic of any pecially for burial. Careful observationof the vessel
unguentaria.80 surface for marks of wear might give some indication
79See supra n. 56. I have not made any measurementsof ideal, as Koehler agrees. The thesis that most unguentaria
glass unguentariato determineif any standardexists. were locally produceddoes not offer much hope for such a
80 A shipwreck with
quantities of unguentaria would be find.

Table 2. Fusiform Unguentaria Capacities from Grave K, Lenormant Street, Athens, now in the
Athenian Agora. (Measurements taken by S.I. Rotroff)
Wet Measurements (ml.)
Inv. No. 1 2 3 4 5
P 10374 21.5 21 21 21 21.5

P 10378 20 21 21 21 21
P 10379 22 22 22 22 22
P 10380 28.5 29 28.5 28.5 28.5
P 10381 29 29 29 29 29
P 10382 24.5 24.5 24 24.5 24
P 10383 22 22 22 22 22
P 10384 22 22 22 22 21.5
P 10386 22 21.5 22 22 22
P 10387 21 21 21 21 21
P 10388 13 13 13 13 13
P 10390 10 10 10 10 10
P 10391 14 14 14 14 14
P 10392 11 11 11 11 11
P 10393 16 16 16 16 16
P 10394 16 16 16 16.5 16
P 10395 17 17 17 17 17
P 10396 19 19 19 19 19
P 10397 11.5 11.5 11 11 11.5
P 10398 20 20 20 20 20
P 10400 15 15 15 15 15
P 10401 10 10 10 10 10
P 10402 13 12.5 12.5 13 13
P 10404 7 7 7 7 7
P 10599 15 15 15 14.5 15
of long use. The fact that unguentaria often occur in certain personal possessions were buried with the
multiples of similar type makes it unlikely, however, deceased.83
that these bottles were normally favorite possessions Many ceramic vessels found in burials would have
transferred to the tomb. A comparative analysis of been unsuitable for daily use, either because of some
grave goods and skeletal material from many ceme- aspect of their manufacture, such as fragile painted
teries suggests that most grave goods did not originally decoration or a general lack of durability, or because
belong to the deceased.81 It is not my intention to begin of their unusual size.84 In Corinth during the fifth
a lengthy discussion of grave offerings and the classifi- century B.C., for example, a series of white-ground
cation systems suggested or criticized by various au- lekythoi was produced especially for use in burials.85
thors.82 It has generally been assumed, however, that In the necropolis at Myrina a number of graves were
81 D.M. Robinson, Olynthus XI. Necrolynthia (Baltimore bee, Death and Burial in the Roman World (New York
1943) 182-83. 1971) 50; Bruneau (supra n. 2) 526; Blegen, Palmer, and
82 For
example, Bruneau (supra n. 2) 528-31; and Robin- Young (supra n. 2) 81, n. 92.
son (supra n. 81) 174-89; G. Childe, "DirectionalChanges 84 The subject of pottery types associated with burial is
in Funerary Practices during 50,000 Years," Man 1945, taken up by S. Weinberg, "Ceramicsand the Supernatural:
13-19; P. Ucko, "Ethnographyand Archaeological Inter- Cult and Burial Evidencein the Aegean World,"in F. Mat-
pretation of Funerary Remains," World Archaeology 1 son ed., Ceramicsand Man (New York 1965) 187-89. See
(1969) 262-80; and R. Chapman, I. Kinnes, and K. Rands- also Blegen, Palmer, and Young (supra n. 2) 81; Robinson
borg eds., The Archaeology of Death (Cambridge 1981). (supra n. 81) 185; Boulter (supra n. 2) 125-26, Grave K.
83 For comments on personal possessions in the tomb, see 85 Haspels (supra n. 4) 93; Blegen, Palmer, and Young
Kurtz and Boardman (supra n. 4) 207-209; J.M.C. Toyn- (supra n. 2) 129; Kurtz and Boardman(supra n. 4) 212.

found to contain small alabastra and unguentaria that guentaria. Elsewhere, as well as at Stobi, graves with
were of solid clay and obviously of no practical use.86 unguentaria often contain several-even as many as
Perhaps many of the examples of the poorly manu- 15 or 20-in one burial.92 When many occur in one
factured unguentaria from burials under considera- grave they are usually of the same fabric and shape, or
tion in this paper represent a special class of unguen- they fall into a small number of different types.93
taria, inexpensive and mass-produced for grave use, There seems to have been no particular concern in
either not designed to retain their contents (if liquid) planning the funeral of the deceased, or among the
for any length of time, or not intended to be filled (see mourners, that each bottle be unique.94 Other vessels
infra).87 An analogy is evident in another class of ves- often occur as multiples in tombs but rarely to the ex-
sels that has been found in burials at Stobi-medium- tent that unguentaria do.95 Multiples of some vessel
sized amphoras of probable local manufacture. Frag- types may reflect the concern that the deceased be well
ments of similar vessels have been found in large supplied with whatever utensils and provisions might
quantities in pottery refuse deposited during the third be necessary in the underworld. On the other hand,
century A.C. in an early Imperial building on the the function of the cups, bowls, and other vases most
flood plain of the lower city.88 Almost all of these often found in burials has also been the subject of con-
pieces preserve a resin lining. But the three complete troversy, and is, by no means, resolved. It is likely that
examples of the same vessel type from the graves of the burial offerings are related in some way to the nature
late first and early second centuries A.C. in the West of the funeral.
Cemetery at Stobi have no lining.89 One of these am- Archaeological evidence from the burial context for
phoras served as the burial urn for the cremated re- the nature of the funeral ceremony is somewhat lim-
mains of one individual.90 The lack of inner coating ited. The fact that ceramic offerings are often broken
on the vessels used in the cemetery suggests that they has suggested to some a breakage ritual at the grave
were produced especially for funerary use rather than associated with the funeral.96 Broken unguentaria are
as storage jars for liquid. frequently found, but most show breaks only at the
What other information about the use of unguen- neck or foot, the naturally weak parts of the bottle,
taria is reflected in the excavation of a burial? Un- and can probably be explained as accidental." Very
guentaria are common in both inhumation and crema- few unguentaria have actually been broken into more
tion burials but by no means are they found in every than a few pieces, or give the appearance of being
grave, apparently not comprising an indispensable shattered, which is what one might expect in a ritual
part of every burial or the funeral ritual.91 Out of a breakage ceremony.98
total of 313 graves excavated in cemeteries at Stobi, In some inhumation burials there is a tendency to
only 68 had offerings, and of those only 29 had un- place certain vessels near particular parts of the
86
E. Pottier and S. Reinach, La necropole de Myrina I 7, 17, 18, 39; Incineraci6nRupert, nos. 1, 6, 8.
(Paris 1887) 106, nos. 221-23, and n. 1, p. 223. Robinson 94Boulter (supra n. 2) 115.
(supra n. 81) 187, also mentions vases found in Palestine 95 Other shapes that most frequentlyoccur in multiples are
with interiors partly filled with terracotta. plates, bowls, cups, pitchers, baby feeders, alabastra, and
87Kurtz and Boardman(supra n. 4) 165; Robinson (supra lekythoi.
n. 81). See the most recent commentsin Tsakos (supra n. 2) 96 For a discussionof breakageritual, see Blegen, Palmer,
418. and Young (supra n. 2) 82; Bruneau (supra n. 2) 531; Rob-
88Context Storage LOTS 1296-1307. inson (supra n. 81) 183; Kurtz and Boardman(supra n. 4)
89 C-71-127 in Grave 60; C-71-201 in Grave
45; 215-16; Alcock(supra n. 64) 62; Higg and Fossey (supra n.
C-74-270 in Grave 262. No vessels with resinous lining 1) 123-25; L.V. Grinsell, "The Breaking of Objects as a
have been recoveredfrom the West Cemetery. Funerary Rite," Folklore 72 (1961) 475-91, and by the
90
C-71-201 in Grave 45. same author, "The Breakingof Objectsas a Funerary Rite:
91Bruneau (supra n. 2) 522-31.
SupplementaryNotes," Folklore84 (1973) 111-14. A num-
92Twenty-three unguentaria were found outside Grave ber of other vessels found in the cremationburials at Stobi
188 at Argos and the remains of 87 inside! Grave K at Le- seem to have been broken before or as they were placed in
normant Street (Boulter, supra n. 2) contained31 unguen- the secondary grave fire because some of the sherds were
taria, and Grave f of the Kerameikos(Knigge, supra n. 2, p. burned and others not, a conditionquite apparentwhen the
187) contained43. vessels are mended.
93 This is true at Stobi; also at Myrina (Pottier and Rei- 97 Or, on the analogy of Mark 14:3, broken
intentionallyto
nach, supra n. 86); Bruneau (supra n. 2) passim; Boulter release the contents.
(supra n. 2) passim; Schl6rb-Vierneisel(supra n. 2) pl. 61; 98 Of the total of approximately250 unguentariaat Stobi,
Almagro (supra n. 2) vol. 2. Incineraci6nBallesta, nos. 4, 6, only 30 have been mendedfrom more than a few pieces.

19871 THE CHRONOLOGY AND FUNCTION OF CERAMIC UNGUENTARIA 121
body.99 No particular pattern with regard to unguen- a short period of time and was used in the funeral rites
taria has emerged, with the exception that the head or at the graveside and then placed in the burial-in
and foot seem the most common location.100 At Stobi, fact, in much the same way that the lekythos seems to
where almost all of the funerary unguentaria come have been used.'15
from cremation burials, this observation is not perti- The contents of the unguentarium, depending upon
nent.101 Most of the cremations at Stobi involved two their nature, might have been used in a variety of ways
fires: the first, in an unknown location, was used to in the funeral rites. Oil, wine, or water are possibili-
cremate the corpse; the second fire, at the grave, may ties for short-term use. One of the first duties of the
have served ritual purposes of consecrating the site living was to annoint the body of the deceased with
and burning the offerings. The normal practice at oil.'06 With the practice of inhumation it was custom-
Stobi was to place the unguentaria in this second grave ary to make libations of oil or wine at the graveside.107
fire,102 thus accounting for the burning and discolora- After a body was cremated, the remains might have
tion visible on most ceramic and glass unguentaria as been drenched with wine in Homeric fashion and li-
well as on other pottery vessels found in graves at bations of oil made at the grave as well.0os Using un-
Stobi. Since both ceramic and glass bottles were placed guentaria to hold wine for a last toast at the tomb is an
in the grave fire (not the cremation fire), it is possible attractive notion.'09 To cite a modern parallel-at
that the vessels were filled with scented oil or incense, graveside ceremonies in Montenegro it was once cus-
providing a pleasant fragrance for the deceased on his tomary for the mourners to drink a last toast to the
journey to the underworld.103 The oil alone, if allowed deceased in small bottles (also used for drinking rakia)
to seep out of the bottle, would probably have intensi- called Tokanja or cokan'e that are similar in shape to
fied the fire. In any case, the burned surfaces and fre- bulbous unguentaria. In Serbia it was common until
quently distorted or melted bodies of ceramic and glass 15 or 20 years ago to pour out small amounts of oil
unguentaria indicate a very hot flame that would have and wine as part of the graveside ceremony."10
melted or consumed the contents. It is also conceivable that after use in the funeral
What then was the function of the unguentarium in rites, or even if not used, the empty unguentarium was
the inhumation burial? Should we expect a different placed in the grave as a symbol of the honor paid to the
ritual to accompany a different method of burial? It is dead at the time of burial."1
possible, as has been suggested, that these small bot- Some unguentaria may never have been intended to
tles, filled with oil, were placed in the tomb for the be used as containers of any sort, even for a short pe-
needs of the deceased. As noted earlier, however, the riod of time, and were produced for symbolic use only.
porosity of the clay meant that the contents might dis- Certainly their appearance gives every indication that
appear after a short time.104 It seems likely that if the they were mass-produced simple grave offerings. Two
unguentarium contained any substance, that it was for examples found at Stobi in the West Cemetery, one
99 See, for example, Blegen, Palmer, and Young (supra n. guent which has been broughtto the funeral before crema-
2)81. tion as a gift, but he says nothing of its function.
100 A tabulationof the
positionsof unguentariain Keramei- 103 See Kurtz and Boardman(supra n. 4) 215.
kos gravespublishedby Schl6rb-Vierneisel(supra n. 2) gave 104
Glass unguentaria would have solved the porosity
the following results: unguentaria at foot (10 graves); at problem.
head (6); at right side, along left leg, along right leg, at right 105 Robinson (supra n. 81) 186-87.
shoulder, at left arm, at thighs, along left thigh, along right 106
Kurtz and Boardman(supra n. 4) 144; Toynbee (supra
thigh, between thighs (1 grave each). From the graves pub- n. 83) 44; Cuadrado(supra n. 2) 390; and Mark 14:8.
lished by Knigge (supra n. 2) the results were as follows: at 107 Kurtz and
Boardman(supra n. 4) 145.
head, left side (1); at head, right side (1); at head and shoul- 108 For use of wine on the remains of the deceased, see
ders (21 in 2 graves); at right shoulder (2 in 1 grave); at Toynbee (supra n. 83) 50, n. 173. Also Hom. II. 24.791, and
shoulders (2 in 1 grave); in left hand (1 each in 2 graves); Prop. Elegies 4.7.33, 34.
between legs (1); between feet (7 in 4 graves);along left leg 109 There are
many parallels for this practice in Greece
(2 in 1 grave);along right leg (3 in 2 graves);at right knee (2 during the BronzeAge and the Dark Ages. See, for example,
in 1 grave);at left knee (1); at left foot (1 each in 2 graves);at Lord W. Taylour, The Mycenaeans, rev. ed. (London
feet (6 in 4 graves);left underarm (2 in 1 grave). 1983); G. Mylonas, Mycenae and the Mycenaean Age
101The exception at Stobi is C-72-44, from an inhumation (Princeton 1966) 112.
of a single individual of the early second century B.C. 110I am grateful to Dragan Stojanovidfor this information.
102Wesolowsky (supra n. 2) 108, 135. Propertius (Elegies I'I Robinson (supra n. 81) 187.
2.13.30) is presumably talking about perfumed oil or un-

122 VIRGINIAR. ANDERSON-STOJANOVIC [AJA91
fusiform and one bulbous, although otherwise ordi- of funerary unguentaria, one is tempted to answer in
nary in appearance,have the body cavity sealed off by the affirmative.
clay at the neck.112It seems quite possible, then, that It is clear that many of the questions posed above
some poorly made unguentariawere never intendedto cannot be answered conclusively by reference to the
hold anything, rather like the souvenir vases and existing published informationon unguentariaor the
knickknacksthat one buys in the potteriesof Amarou- data from Stobi. Excavatorsof Hellenistic and Roman
si outside Athens, none of which retain water for long. cemeteries would do well to provide a detailed de-
The presenceof many unguentariain one grave has scription of the fabric of each unguentarium found,
still not been explained. Further speculationabout the including color, the nature of the core (if present), in-
nature of the funeral ceremony may offer a possible clusions, surface character, and slip or decoration.
explanation. We know several people must have ac- The following questions might be considered when
companied the corpse to the burial place. Each indi- publishing collectionsof unguentaria:1) How porous
vidual might have made a libation or an offering of oil is the vessel fabric? How long does it take for sub-
at the graveside with a single unguentarium perhaps stances such as water, oil, or wine to seep through the
providedby the family of the deceased.This might ex- vessel wall? 2) Is the unguentariumbrokenand, if so,
plain why the unguentaria in any one grave are simi- where? 3) Is the vessel from a cremationand, if so, is it
lar, since it is possible that they were purchased as a burned and to what degree?4) If the vessel is from an
group from a single potter'sestablishment.Variations inhumation, what was the vessel's position in the
among unguentariamay indicate purchasefrom more grave?5) Was the unguentariumcarefullymade, or of
than one potter. On the other hand, if only a few shops poor quality, and is the neck sealed off with clay? 6) If
made unguentaria,several individualscould purchase a sizeable homogeneousgroup is available, do meas-
vessels which would be similar. The number of un- urements of capacity suggest any standards? 7) Is
guentaria in any grave might then representthe num- there any trace remaining of the original contents?'14
ber of persons present at the graveside ceremony of Careful study of ceramic funerary unguentaria and
burial. their contextcan add much to our knowledgeof burial
In his concludingarticle in Ceramicsand Man enti- rites and other rituals associated with death in the
tled "Ceramic Queries," in the section on rituals, Hellenistic and early Roman periods.
FrederickMatson asks if the occurrenceof large num- DEPARTMENTS OF CLASSICS AND FINE ARTS
bers of similar vessels in a deposit might indicate the WILSON COLLEGE
single use of each vessel in some ritual.113In the case CHAMBERSBURG, PENNSYLVANIA 17201
112 C-71-88andC-74-363in the StobiInventory. veal, throughcarefulbiochemicaltestingsuchas that used

113 F. Matson (supra n. 84) 282. for lipid residues,evidenceof the originalcontents.Roths-
114 Evenvesselsthathavebeenwashedwithwatermayre- child-Boros(supran. 56) 83.


Archaeological Institute of America

Uploaded by

Copyright:

Available Formats

Archaeological Institute of America

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Archaeological Institute of America

Uploaded by

Copyright:

Available Formats

The Chronology and Function of Ceramic Unguentaria

Author(s): Virginia R. Anderson-Stojanović

This content downloaded from 130.191.17.38 on Sat, 17 Jan 2015 06:20:38 AM

Abstract 3) the functionof unguentariafound in burials was close-

American Journal of Archaeology 91 (1987) 105

This content downloaded from 130.191.17.38 on Sat, 17 Jan 2015 06:20:38 AM

This content downloaded from 130.191.17.38 on Sat, 17 Jan 2015 06:20:38 AM

This content downloaded from 130.191.17.38 on Sat, 17 Jan 2015 06:20:38 AM

(supra n. 1). and M. Vegas, Cerdmicacomuinromanadel Mediterrdneo

This content downloaded from 130.191.17.38 on Sat, 17 Jan 2015 06:20:38 AM

gos in Appendix 3 (pp. 141-42) or for Asine in Tables 3 and B.C.

This content downloaded from 130.191.17.38 on Sat, 17 Jan 2015 06:20:38 AM

28 See drawing of lekythos with false interior chamber in

This content downloaded from 130.191.17.38 on Sat, 17 Jan 2015 06:20:38 AM

It has been suggested that the discovery of glass-

Fig. 4. Shape 1, biconical fusiform unguentarium, Stobi

B.C., and evidence from Israel and Jordan indicates a

This content downloaded from 130.191.17.38 on Sat, 17 Jan 2015 06:20:38 AM

amples of this shape in clay were foundat Stobi in the

Fig. 7. Shape 2, bulbous unguentarium, Stobi C-74-105

In general, the earliest examples share a low point

This content downloaded from 130.191.17.38 on Sat, 17 Jan 2015 06:20:38 AM

This content downloaded from 130.191.17.38 on Sat, 17 Jan 2015 06:20:38 AM

This content downloaded from 130.191.17.38 on Sat, 17 Jan 2015 06:20:38 AM

This content downloaded from 130.191.17.38 on Sat, 17 Jan 2015 06:20:38 AM

This content downloaded from 130.191.17.38 on Sat, 17 Jan 2015 06:20:38 AM

This content downloaded from 130.191.17.38 on Sat, 17 Jan 2015 06:20:38 AM

Table 1. Stobi Unguentaria Capacities

295 C-74-343 150-25 B.C. Fusiform B 85 159 061 2.603

This content downloaded from 130.191.17.38 on Sat, 17 Jan 2015 06:20:38 AM

Wet Measurements (ml.)

P 10374 21.5 21 21 21 21.5

This content downloaded from 130.191.17.38 on Sat, 17 Jan 2015 06:20:38 AM

This content downloaded from 130.191.17.38 on Sat, 17 Jan 2015 06:20:38 AM

This content downloaded from 130.191.17.38 on Sat, 17 Jan 2015 06:20:38 AM

112 C-71-88andC-74-363in the StobiInventory. veal, throughcarefulbiochemicaltestingsuchas that used

This content downloaded from 130.191.17.38 on Sat, 17 Jan 2015 06:20:38 AM

You might also like