Literacy and Luxury in the Early Empire: A Papyrus-Roll Winder from Pompeii

Author(s): Susan Wood

Source: Memoirs of the American Academy in Rome , 2001, Vol. 46 (2001), pp. 23-40
Published by: University of Michigan Press for the American Academy in Rome

Stable URL: https://www.jstor.org/stable/4238779

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 [email protected].

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at
https://about.jstor.org/terms

American Academy in Rome and University of Michigan Press are collaborating with JSTOR to
digitize, preserve and extend access to Memoirs of the American Academy in Rome

This content downloaded from

79.230.157.136 on Wed, 18 Oct 2023 10:35:29 +00:00
All use subject to https://about.jstor.org/terms
 LITERACY AND LUXURY IN THE EARLY EMPIRE:

A PAPYRUS-ROLL WINDER FROM POMPEII

Susan Wood, Oakland University

Q n 27 June 1873, excavators in Regio I, Insula 2 of Pompeii discovered wh

to be the remains of "un piccolo armadio di legno," a small wooden cabinet.' The frag-
ments included two fairly complete ivory plaques, each carved on both sides with mythologi-
cal scenes. Both the iconography and the function of these objects have presented modern
scholars with enigmas, but to date, the former problems have received far more attention
than the latter. In an earlier publication, I discussed my own theories about the subject mat-
ter; here I examine the function of the two plaques and reconsider their iconography in light
of that function.
Each plaque is approximately rectangular, but with rounded corners and ornamentally
scalloped sides. Each has four rounded and two square perforations along the lower border,
and each opening is neatly framed with a molding. Both plaques have two small perforations
along the upper edge as well. Two large circular perforations appear in the lower corners of
the plaques. Slightly higher above the bottom edge, and set in slightly from each corner, are
two smaller circular holes, and between these two is an incised groove that dips downward
toward the bottom center of the plaque. A beaded molding wraps around these perforations
and follows the dipping line of the incised groove. On plaque A (figs. 1, 2), the ivory below
the groove has broken away, but the other, slightly better preserved plaque B (figs. 3, 4) re-
veals that the bottom edges were straight, with leaflike patterns filling the spandrels below
the curving line and the lower border. Directly above the smaller rounded perforations that
mark the ends of this curving groove are the two square perforations.
In the earliest excavation report of this discovery, A. Sogliano speculated that these plaques
might have formed the fronts of two drawers. The concave surface of each plaque, as he
noted, showed greater signs of damage than the convex side, suggesting that small objects
routinely scuffed against it. H. Graeven, however, disputed this theory. Why place a rather
expensive work of craftsmanship where the inner side would seldom be visible? He con-
ceded, however, that the plaques might have been reused for such a purpose at some point

I would like to express my sincere thanks to Dr. J. 1 Sogliano 1874. The plaques are now housed in the
Michael Padgett of Princeton University for calling myNational Museum of Archaeology in Naples, accession
attention to the existence of the Cambridge book winder;
nos. 109905 A (figs. 1 and 2 here) and 109905 B (figs. 3
to Dr. Lucilla Burn, Keeper of Antiquities of the and 4 here). Precise measurements are unfortunately
Fitzwilliam Museum, Cambridge, and to the librarians not available. Major publications: Graeven 1903, 39-
of the Papyrology Collection, Harlan Hatcher Library, 46, nos. 25-28; Spinazzola 1928, xxxiii, no. 224; Rob-
University of Michigan at Ann Arbor for their generousert 1897-1919, 3:453-454; LIMC 4, s.v. "Demeter/
assistance with my research. The two anonymous read- Ceres" (Stefano de Angeli), 901, no. 135; Kampen 1981,
ers for this journal offered valuable bibliographical sug-
70, fig. 60; Kampen 1993, 124; Musso 1989-1990,14-
gestions and constructive criticism for which I am very
16, fig. 11 with full earlier literature; Wood 2000, 84-
grateful. 88.

MAAR 46, 2001

This content downloaded from

79.230.157.136 on Wed, 18 Oct 2023 10:35:29 +00:00
All use subject to https://about.jstor.org/terms
 24 SUSAN WOOD

Fig. 1. Ivory plaque from

Pompeii, Naples, _
National Archaeological
Museum, 109905 A
obverse, birth of
Meleager? (photo
Deutsches N:
Archdologisches Institut,
Rome, neg. no. KS.
66.1838).

Fig. 2. Reverse offig 1,

death of Meleager?
(photo Deutsches
Archdologisches Institut,
Rome, neg. no.4
74.1337)

I ~~~~~~~~~F
'.

~~~~~~~~~~~~~~~~~~~

after their manufacture.2 In my earlier publication, I suggested that the plaques might have
been decorative handles that projected above the top of the cabinet to which they belonged,
and that the holes might have facilitated their attachment.3
Since then, however, Dr. J. Michael Padgett of Princeton University has kindly called my
attention to the existence of a closely parallel artifact that survives somewhat more intact
(figs. 5, 6). This object forms part of a small cache of ivories that were probably a burial
assemblage, perhaps of the type that Etruscan and Latin cistae sometimes contain.4 The ivory

2 Graeven 1903, 39-46, nos. 25-28. dowels 6.2 cm. Square dowels: length 36 cm, width at
thick ends 1 cm, at narrow ends 0.4 cm. Round dow-
3Wood 2000, 84-88. els: length 32 cm, thickness at thick end 0.8 cm, at
narrow end 0.2 cm. Published: Vassilika 1998, 118,
4 Fitzwilliam Museum, Cambridge, GR 25-1980. Mea- no. 57, photo p. 119. For an example of a bronze cista
surements, plaques: width at widest point 12 cm, that contained such an assemblage, E. Vassilika cites
height 8 cm, thickness 0.6 cm, width across square Bordenache Battaglia and Emiliozzi 1979, no. 10, 66-

This content downloaded from

79.230.157.136 on Wed, 18 Oct 2023 10:35:29 +00:00
All use subject to https://about.jstor.org/terms
 A PAPYRUS-ROLL WINDER FROM POMPEII 25

Fig. 3. Ivory plaque

Pompeii, Naples,
National Archaeological
Museum, 109905 B,
obverse, rape of
T'i, _ : , ; L , ,,, ,:- Persephone (photo

; a,N,,, Wbi,:.... . 66.1837).

/ Deutsches
Archdologisches Institut,
Rome, neg. no.

.r.... ..... ... ... -

-. .~' f Fig. 4. Reverse offig. 3:

Demeter, Art
. - - ~~~~~~~Athena pursue the
chariot of Hades
(drawing reproduced
rA.f ., ~~~~~~~~~~~~~~~from Robert 1897-1919,
3:454; used h
permission of the Harl
Hatcher Library Rare

University of Michigan,

Ann Arbor).

( '/

cache,
a small
smooth
except
scallop
higher
engrav
the bo
curve,
virtual
perfor
long iv

67. Unfortunately, the archaeological provenance of from a London dealer, but appear to form a homoge-
the Fitzwilliam ivories is unknown; they were acquired neous group.

This content downloaded from

79.230.157.136 on Wed, 18 Oct 2023 10:35:29 +00:00
All use subject to https://about.jstor.org/terms
 26 SUSAN WOOD

Fig. 5. Assemblage of * . ... ........................

ivory objects, f Ioi .. . ... .. . .
Fitzwiliam Museum,
Cambridge, England,
GR 2-2 7, 1980 (photo : . _ .:
Museum).

.................. ~~~~ ~ ~ ~ ~~~~~~~~~~~~~~... .... ......

.:: ....... | ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

Fig. 6. Ivory book

winder, Fitzwilliam
Museum, CR 25.1980 . . . ..... ....

(photo Museum).

throu
tions
The c
The a
repor
sugg
proba
papy
avail
slip t
squar
I R. V. Nicholls, former Keeper of Antiquities at the 6 CTP 2:225, I.ii.24. House 24 of that insula is identified
Fitzwilliam Museum, first identified this object as a bookas the "Officina libraria of Acilius Cedrus, L. Aelius
roll winder, and recognized several other objects, includ-Cydinus, Appuleius Adiutor, P. Instuleius Nedymus, C.
ing the Pompeii and Nimes ivories, as having similar Nonius Lorica." Graffiti identify the inhabitants as
functions. Unfortunately, however, he did not publish "librari"; Della Corte 1954, 229-230, nos. 557-561.
his research. I am indebted to Dr. Nicholls and Dr.
Lucilla Burn, the present Keeper, for this information.

This content downloaded from

79.230.157.136 on Wed, 18 Oct 2023 10:35:29 +00:00
All use subject to https://about.jstor.org/terms
 A PAPYRUS-ROLL WINDER FROM POMPEII 27

Fig. 7. Diagram sh
the threading patt
the Pompeii, Osti
Fitzwilliam book
winders (drawing
author).

Fig. 8. Replica of the

Pompeian hook winder,
constructed hy Mohamed
and Louisa Ngote (photo
Louisa Ngote).

take roll," so to speak, on her left. (The use here of the feminine pronoun is dictated not.by
. _ ; '5 ~.. '. .....

i__s: j~ ~ ~ ~ ~ ~ ~ . . . . . .:._
f - s. . _, s..t.................~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~.....

tak roll" so to spak on he et Teuehr ftefMinnU_ nu sittdntb

political correctness, but by the feminine objects of toilette among the Fitzwilliam ivories that
suggest a woman's burial.) The papyrus would probably tend to arch slightly upward rather
than lie flat, but the text would remain legible. After reading each column, the reader would
roll the papyrus to advance to the next column. The dowels themselves would not be used to
turn the papyrus, since they were not firmly attached to it, and in any case were too thin to turn
the rolls efficiently. Nor would the rolls be looped completely around the dowels, or threaded
on them, but simply passed under them to hold the exposed area in place. The rolled section
of papyrus would lie on each side of the winding device.
Some book rolls, though by no means a majority, were anchored to wooden rods, or "m
bilici" of their own.' When reading one of these deluxe editions, the reader could turn the
umbilicus with her right hand, and thus minimize handling of the papyrus, although she would
have had to use her left hand to wind up the other end. When she had finished the volumen
she could easily rewind it simply by rotating the umbilicus counterclockwise with her right

7 Capasso 1991, pl. 15 a and b; Capasso 1995, 73-98. Herculaneum book rolls, only a minority has perma-
Capasso notes that even among the fairly intact nently attached umbilici.

This content downloaded from

79.230.157.136 on Wed, 18 Oct 2023 10:35:29 +00:00
All use subject to https://about.jstor.org/terms
 28 SUSAN WOOD

hand, without needing to

els would allow the papyrus to slide smoothly under them, but the square dowels, which
could not rotate, would prevent the section that the reader was perusing from slipping. Ev-
ery extant example of these objects has square perforations for the inner dowels, a form more
difficult to carve than a circular opening, which can be drilled. We may assume, then, that
these dowels were intended to hold something stable, and that their role was critical to the
function of these devices.
Although the exposed section of papyrus would be fairly narrow, it would probably be ad-
equate for most single columns of writing.8 The width from outer edge to outer edge of the square
dowels of the Fitzwilliam book winder is 6.2 cm; the area of papyrus exposed across the middle
area between the two cylindrical rods could be as broad as 12 to 13 cm, depending on the degree
of curvature the reader was willing to tolerate (fig. 8). For book rolls with wider columns, or of a
height from upper to lower edge of 30 cm or more, this device would be useless. Surviving papyri
demonstrate, however, that well-written, legible columns of prose text could be as narrow as 3 cm
and rarely more than 9 cm, with the most common widths ranging between 4.5 and 6.5 cm.9
Lines of poetry in some verses, notably dactylic hexameter, would necessarily be wider, a format
enforced by the genre, but some lyric poetry could be written in very narrow columns.10 The
reader would need a little extra leeway at each side of the text, since scribes typically gave their
columns a leftward slant from top to bottom.11 The slope of a column would probably not add
more than a centimeter to the space required to accommodate it, however, and the reader could
of course shift the papyrus slightly to her right as she read down the column.
The length of the dowels would impose another limitation on these devices. The better-
preserved cylindrical dowel of the Fitzwilliam specimen is 32 cm long, that of the square
dowels slightly longer, 36 cm. A book roll higher than about 28 cm, then, could not easily fit
onto this book winder. How often might its owner encounter such a dilemma? Complete
pages, unfortunately, rarely survive, but one fragment from a manuscript of Aeschylus's
Dictyulci preserves the entire height of at least one column; the column was 15 cm long, and
the papyrus sheet, although unfortunately somewhat tattered at the lower edge, appears to
have been no more than 27 cm in height.12 Another fragment, from Demosthenes, in similar
condition, measures 26.5 cm.13 Books could sometimes be considerably smaller, however, as
another specimen of 12 cm demonstrates, and this miniature format appears to have been
popular for poetry in the first and second centuries A.D.14 Roman paintings and reliefs often

8 On the format and construction of book rolls, see Avi-1987, 44-45, no. 16, Alcman, Partheneia, Mus6e du Lou-
Yonah 1973, 15-27. vre, E. 3320, first century A.D.

9 Johnson 1994, 425: frequency tables demonstrate that

"Johnson 1993.
the widths of columns for all prose genres most com-
monly have widths of between 4.5 and 6.5 cm. He records 12 Turner 1987, 54-55, no. 24; Egypt Exploration Soci-
no columns of prose wider than 9 cm. ety, from Oxyrhynchus.

10 Two examples of Homeric verse: Turner 1987, 38-3 9, 13 Turner 1987, 112-113, no. 67, Demosthenes, De Falsa
no. 13, Iliad 2, Bodleian MS Gr. Class. a 1 (P), middle Legatione, 53-57, Cairo Museum inv. 47446, first-sec-
second century A.D.; Turner 1987, 40-41, no. 14, Iliad ond2 century A.D.
(the Harris Homer), British Museum, Pap. 126, later
third century A.D. The columns of the former are about14 Turner 1987, 72-73, no. 39, Herodas, Mimes, British
18 cm wide, those of the latter about 13 cm. Both fac- Museum, Pap. 135 (Pap. Egerton 1). The total height of
similes are reproduced at 1:1 scale with the originals in the fragment, which is preserved with its upper and lower
Turner 1987, vii. An example of lyric poetry: Turner margin, is only 12 cm.

This content downloaded from

79.230.157.136 on Wed, 18 Oct 2023 10:35:29 +00:00
All use subject to https://about.jstor.org/terms
 A PAPYRUS-ROLL WINDER FROM POMPEII 29

represent figures holding or reading from book rolls, a

us the precise dimensions of the books, the rolls seldom appear very large in proportion to
the hands holding them. One might reasonably estimate a height of less than 30 cm for most
of them. An example both chronologically and physically close to the Pompeian ivories is the
well-known double portrait of a man and woman from the house of Terentius Neo in Pompeii.15
The man grips a scroll by its lower end; the height of the roll appears to be about two and
one half times the width of his hand. If a large man's clenched hand is about 11 cm across the
knuckles, the roll would be 27.5 cm high at most.'6 Equally interesting is the fact that this
painting proudly advertises the literacy not only of the man but also of his wife. Another
literate woman appears in an equally well-known portrait, also from Pompeii, holding writ-
ing tablets and tapping her lip with a stylus, the same attributes and mannerisms as Terentius
Neo's wife.17
By raising the open section up across the square rods, and bracing the rolls against the
outer dowels, the book winder would prevent the vexation of letting the volumen acciden-
tally roll back over the text. The user could keep a book open to a passage she wished to
consult, rather than being forced to hunt constantly for it.'8 If she were reading aloud to her
children, for example, whose early education was one of her responsibilities, she could look
up from the text and gesture toward her listeners. Roman declamation involved a formal vo-
cabulary of gestures as well as of speech. She could also, if she wished, spread the book roll
outward on each side, so as to read a longer passage than that exposed across the square
dowels; she would have to hold the rolls with both hands in order to do so, but if she re-
leased the rolls, she would not lose her place. The tapering shape of the rods would allow her
to move the plaques closer together or spread them further apart depending on the width of
the volume she was reading, while the small knobs along the upper edges would facilitate
adjustment. As for the two smaller round holes below the square ones, and the V-shaped
groove between them, their function is a little more difficult to explain. They might, how-
ever, have accommodated smaller supports that could have held the device steady while in
use, or possibly additional knobs for adjustment of the plaques.
At least nine similar artifacts survive from various locations in the Roman Empire, includ-
ing Ostia, Etruria, Gaul, and Romania.19 The example from Ostia, like the Fitzwilliam book
winder, comes from a burial assemblage, along with many ivory and bone ornaments. Associ-
ated pottery fragments appear to indicate a date in the second or first century B.C. Although
much more fragmentary than the Pompeii or Fitzwilliam examples, the plaques from this find
had recognizably the same shape, with scalloped sides, circular holes in the lower corners, square
holes set higher and inward from them, and small round perforations directly under the square
holes. Like the Pompeian plaques, these examples were also decorated not only with moldings
along the edges and around the holes but with relief sculpture on both sides, and here too the

15 Naples, National Archaeological Museum 9058, from fig. 169; Maiuri 1953, 99-103, photo p. 100.
Pompeii VII 2,6. Publications: Ling 1991, 158-159, 160
fig. 170, with earlier references on 230; Maiuri 1958, pl. 18 Avi-Yonah 1973, 24-27.
XXIII; Maiuri 1953, 102-103.
19 Beal 1984, 112 n. 2. Other examples from Gaul:
16 This admittedly somewhat unscientific figure wasBarruol
de- 1971, 400, on a Roman burial at St. Gilles con-
termined by measuring the hand of my colleague Prof.
taining such an object, described as a "petit metier a
John Cameron, who is a tall and large-boned man. tisser," and Benoit 1962, 149, 151 fig. 5, for fragmentary
wooden plaques of the same type, identified as "chevalets
17 Naples Archaeological Museum 9084. Ling 1991, 158,
de lyre a huit cords et plectre."

This content downloaded from

79.230.157.136 on Wed, 18 Oct 2023 10:35:29 +00:00
All use subject to https://about.jstor.org/terms
 30 SUSAN WOOD

AP~~~~~~~'

l | | l | l . .
* ~ ~~~~ ~ ~ ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ .... . .. ..w

o |

... ....... .

Fig. 9. Ivory book winder, Nimes, Musee Archeologique, Fig. 10. Ivory book winder, Nimes, Mus:. e A rcb6ologique,
386-387 (photo Museum). 388-389 (photo Museum).

subjects appear to have been mythological. The condition of the fragments does not allow a
complete reading of the iconography, but a seated winged figure, apparently an adolescent
Eros, appears on one side of the largest piece, and a centaur and dancing maenad on the
opposite surface. The drawings in D. Vaglieri's excavation report seem to indicate that these
plaques, like the ones from Pompeii, suffered more abrasion on one side than on the other.20
Two especially well preserved book winders, in the Musee Archeologique at Nimes,
show the same basic structure as the Pompeii and Cambridge artifacts, but with some dif-
ferences of design (figs. 9, 10).21 One of these was excavated from a Gallo-Roman tomb,
and here too, as with the Cambridge ivories, many of the possessions indicate a woman's

20 Vaglieri 1912, 97-98, fig. 5; Squarciapino 1955, 14-

designed to hold the thicker ends of the tapering dow-
15, 19, 54 n. 19. els. Published: Beal 1984, 110-112, nos. 386-387; Vigne
1899, 13-24 on excavation of the Gallo-Roman tomb in
21 A. Musee Archeologique de Nimes, 906.10 Nimes 1, which the ivory object was found; 17 no. 2 for smaller
1895: two crescent-shaped plaques, two surviving flat round dowels (identified as "knitting needles"), 17 no.
dowels, and two round dowels. Measurements of 3 for larger round dowel; 18-19, no. 4, pl. 7 for the
plaques: width 15 cm, height 8 cm, thickness 1 cm. plaques and flat dowels.
Length of smaller round dowels 28 cm, of larger round B. Musee Archeologique de Nimes, 352.1 and 2,
dowel 22 cm, thickness of larger round dowel 1 cm; Nimes 2: approximately trapezoidal plaques, widest side
length of flat dowels 30 cm, maximum width 1 cm, spaceat top. Width of plaques 10.3 cm and 11.0 cm; height
between flat dowels 4 cm on one plaque, 5 cm on the 7.4 cm, thickness 0.4 cm maximum. One of the two
other. Width across tops of dowels when in place, about plaques has an octagonal opening to the left (viewed
7 cm. The plaque with the larger apertures is evidently from the outside face) of its center. Both have four small

This content downloaded from

79.230.157.136 on Wed, 18 Oct 2023 10:35:29 +00:00
All use subject to https://about.jstor.org/terms
 A PAPYRUS-ROLL WINDER FROM POMPEII 31

burial.22 In these cases, the brackets that held the c

nonrotating rods, which in these examples are broad
these devices (fig. 9) consists of crescent-shaped, rat
corners of the crescent take the form of birds' heads th
supports for round rods, one of which survives and
vide. Two smaller and thinner dowels from the same
ting needles," also probably belonged to this device,
form the birds' "eyes." Their function would probabl
in use and possibly also to hold the papyrus roll firm
The other book winder (fig. 10) has rectangular pla
Roman specimen in that the brackets for the round dow
this time looping outward and downward. On this ex
perforations, two along the upper edges and two dire
the plaque's height. These might have accommodated
Fitzwilliam specimen, or additional dowels. In the la
braced the roll of papyrus from above and below, pr
the reader's view of the open page. The larger octag
however, defies easy explanation. Both plaques are d
Pompeian and Ostian specimens; they have ornament
per borders are dentilated. In both of the Nimes boo
mistakably indicate the orientation of the object; the
of the plaques when the object is in use.
Both the initial excavation report of the Gallo-Rom
logue confess puzzlement as to the functions of the
however, recognized that these devices could be used
it flat, and both therefore suggested that they had som
cations of the more fragmentary discoveries at Ostia
belonged to musical instruments, but the more com
appear to disprove that speculation.24 R.V. Nicholls'
by far the most plausible, since the items are luxury ob
rated, and therefore probably destined for a prestig
an elite woman's hobby, and the discovery of at lea
might support the interpretation that they served s
ever, to allow a weaver or embroiderer to work on anyt
ric. It is also difficult to picture how they could have h
dery, or how a needleworker could have used them w
each end got in the way of her hands.
If the two items in Nimes were in fact book winders,
from the type that the Pompeii, Cambridge, and Ost

perforations, two along the upper edge,

this tomb two
(14). If below
so, the deceased them,
was a woman named
near the middle of the plaque, Cornelia.
probably to support slen-
der dowels. Published: Beal 1984, 110-112, nos. 388-
389, pl. 22; Vigne 1899, 19. 23 Vigne 1899, 19; B6al 1984, 111: "On ignore tout du
r6le de ces objets."
22 Vigne 1899, 13-15, who suggests that an inscription
in the wall of a nearby house might have belonged to 24 Vaglieri 1912, 97; Benoit 1962, 149.

This content downloaded from

79.230.157.136 on Wed, 18 Oct 2023 10:35:29 +00:00
All use subject to https://about.jstor.org/terms
 32 SUSAN WOOD

cylindrical dowels must h

the exposed area allowed t
reason, the cylindrical dowels can be removed from their brackets without dismantling the
entire device. A reader would have found this feature convenient when she wished to remove
one volumen from the winder and replace it with another. The lower dowels would support
the page, keeping the text somewhat level, but would not raise it or hold it taut. This design
might present some disadvantages, in that the text would have been slightly further from the
eyes, and the shadows both of the rolls and of the frame would inevitably have fallen on it.
On the other hand, the lower placement of the flat dowels would have avoided tension on
the delicate material and helped to prevent tearing, a desirable precaution for an old or par-
ticularly valuable volumen. The raised position of the cylindrical dowels, and the flat sup-
ports below the open area, would also have prevented the papyrus from rubbing against the
table on which the reader had placed the book and the unwinding device. Here again the
tapering shape of the dowels allows for adjustment to the height of the volumen. The design
of these objects, however, presents a disadvantage of its own. They could not have been used
for volumes greater in diameter than about 12 cm. The height of the cylindrical dowels does
not allow clearance for a roll with a radius of greater than 6 cm at most. Again, artistic repre-
sentations of books, at least of the sort that a private owner might possess, indicate that they
were not very bulky. The book roll that Terentius Neo holds in the painting cited above is
narrow enough to be gripped comfortably in his closed hand. The book rolls discovered at
the Villa of the Papyri in Herculaneum varied in diameter from 6.5 cm to 13 cm.25 The Nimes
book winders would have worked very well with those of the former size, less easily with
those of the latter. But all of these book winders did have certain limitations and could not
be used with every book.
Their function, then, would explain why the reliefs of the Pompeian and Ostian
plaques are unevenly worn. The more abraded concave surfaces must have been the in-
ner sides of the plaques, which came in contact with the edges of papyri. The outer face
of Pompeian Plaque B, then, is the surface that represents the rape of Persephone, as
Hades carries his struggling victim away in his chariot (fig. 3). Zeus looks on from above
while Hermes leads the chariot toward the underworld. A partially preserved reclining
female below Hermes is probably Tellus or some such place-personification. Aside from
a few surprising early identifications, this scene has presented little difficulty to most
scholars.26 The inner, and more damaged, side of the plaque shows Demeter, Artemis,
and Athena pursuing the chariot in a futile attempt to foil the abduction (fig. 4). In a
grotto behind Demeter, a smaller fourth figure appears, of indeterminate sex, who raises
its right hand to grasp a billowing veil. This may be Hekate who, according to the Homeric
Hymn to Demeter, hears but does not see the rape of Persephone, and later tells Demeter
what she heard.27 If male, the figure may be Helios, the other witness. Ovid also describes
how Cyane, the eponymous nymph of the lake in Sicily where the chariot plunged into
the underworld, attempted to reason with Hades, who turned her into a pool of water

25 Capasso 1995, 60. rape of Persephone. Spinazzola 1928, xxxiii identified

the chariot scene as the rape of the Leucippidae, although
26 Sogliano 1874, 12-14 identified the three pursuing fig-only one aggressor and one victim appear.
ures on the reverse of Plaque B as Amazons, although
he recognized the chariot scene on the obverse as the 27 Hymn. Hom. Cer. 24-25, 52-58.

This content downloaded from

79.230.157.136 on Wed, 18 Oct 2023 10:35:29 +00:00
All use subject to https://about.jstor.org/terms
 A PAPYRUS-ROLL WINDER FROM POMPEII 33

for her pains.28 Another nymph, Arethusa, who travels through the underworld on her
journey to Sicily, later tells Demeter of Persephone's whereabouts.29 The grotto-like frame
around the figure seems to correspond best to Hekate as described in the Homeric Hymn,
who listens from a cave. Whoever she may be, the figure watches and evidently sympa-
thizes with the pursuers, but does not take part in the chase. Both compositions on Plaque
B are very familiar from many second- and third-century Roman sarcophagi. Although the
ivory necessarily abbreviates the scene, the figures and their arrangement recur so predict-
ably that they probably all reflect some well-known original. A celebrated painting by
Nicomachos that hung in the Capitoline Temple until its destruction in the fire of A.D. 64
or 69 probably supplied the prototype.30
On Plaque A, the outer face would have been the one that depicts a seated figure
attended by three others (fig. 1). The inner and more damaged side shows a familiar group:
a dead hero carried by two companions, while a grief-stricken old man stands behind the
body, lifting the man's limp arm with his hand (fig. 2). Depending on how the reader
held the device, he would have seen either the obverse of Plaque A and the reverse of
Plaque B at the same time, or else the obverse of B and reverse of A. We can, therefore,
consider the rape of Persephone as a pendant to the scene of the warriors carrying the
young hero's body. This association would have been very appropriate, since both scenes
deal with the untimely deaths of young people, and have an elegiac tone. Conversely, the
scene of the seated figure with three companions forms the pendant to the three pursu-
ing goddesses.
The identification of the figures on Plaque A, however, presents far more vexing prob-
lems than for Plaque B. Indeed, even the sex of the figures on the outer face of the plaque is
disputed. Opinions are approximately evenly divided between those who interpret this scene
as a physician treating a wounded warrior, or as a midwife attending a woman. P. Lehmann
identified the patient as Adonis, who has just suffered his fatal wound in a boar hunt; a doc-
tor attempts to treat him while Venus, the female figure at the left, "looks on, a heavily draped
figure extending both hands in helpless anguish."'31 Adonis, as Lehmann points out, was the
object of a mystery cult like that of Persephone at Eleusis, and he too miraculously returns
from the dead for part of every year. The earliest publisher of these plaques, on the other
hand, considered but rejected such an interpretation, concluding that the figure seated on
the higher chair had female breasts and must be a woman. Rather than attempt to identify
her, Sogliano declined to interpret the scene at all. The dead hero on the other side, as he
rightly observed, could be any of several mythological characters; the deaths of Antilochos,
Achilles, Patroclos, and Hector all appear in similar format on other monuments.32 Graeven
suggested that the seated woman on the obverse might be Venus, who in Homer's Iliad re-
ceives a wound while intervening in a battle to rescue her son Aeneas.33 N. Kampen, however,

28 Ovid, Met. 5.409-437. 31 Lehmann 1953, 57-58. Spinazzola 1928, xxxiii also
identified the patient on the obverse of this plaque as
29 Ovid, Met. 5.487-508. "Adone ferito."

30 Oakley 1986; Lindner 1984, 11; Pliny, HN 35.108: 32 Sogliano 1874, 15: "A prima vista la figura principale
Nicomachus . .. pinxit raptam Proserpinae, quae tabula pu6 sembrare di un giovane; ma dopo un attento esame
fuit in Capitolio in Minervae delubro supra aediculam si riconsce per muliebre." On the scene of companions
Iuventutis. See also Cohen 1996 on the fourth-century carrying a dead hero, see also Musso 1989-1990.
B.C. fresco at Vergina and its possible relationship to the
painting by Nicomachos. 33 Graeven 1903, 43-45; Hom. II. 5.336.

This content downloaded from

79.230.157.136 on Wed, 18 Oct 2023 10:35:29 +00:00
All use subject to https://about.jstor.org/terms
 34 SUSAN WOOD

Fig. 11. Terra-cotta Ls"_

plaque from the tomb of . 7
Scribonia Attice, Isola _, _
Sacra of Ostia, depicting
a midwife. Museo
Ostiense (photo Fototeca
Unione, American
Academy in Rome,
neg. 2388). -t

Fig. 12. Terra-cotta i

plaque from the tomb of
Scrihonia Attice, Isola Ar~~~~~~~r
Sacra Ostia, depicting a on many s
physician and patient.
Museo Ostiense
(photo Deutsches
Archiiologisches Institut,
Rome, neg. no. 65.278).

advanced the most plausible interpretation of this scene, as a mother in labor attended b
a midwife.4 p
Similar scenes appear on a terra-cotta plaque (fig. 1 1) from the tomb of Scribonia Attice,
an Ostian midwife, and on many sarcophagi that depict the life history of the deceased from
birth to death. The Ostian plaque offers the closest parallel in scale to the Pompeian book
winder, depicting crudely but clearly the elements of the birthing procedure that Soranus
dictates in his Gynaeceia. The mother sits on a chair with hand grips while an attendant,
standing behind her, braces her back to restrain her from writhing. The midwife, seated on a
lower stool in front of the parturient, reaches between her legs. The parturient's body is na-
ked, although she wears a veil over her head, perhaps a token mark of her respectable status
as a matron. Soranus recommends that three attendants should restrain the parturient, but
the plaque depicts only one, probably because of limited space." In other respects, however,
it corresponds precisely to his instructions, even to the detail that the midwife looks away

34 Kampen 1981, 70, fig. 60, and Kampen 1993, 124. 35 Kampen 1981, 69-70, 140, cat. I.6, fig. 58; Soranus,
Gynaeceia 2.5.

This content downloaded from

79.230.157.136 on Wed, 18 Oct 2023 10:35:29 +00:00
All use subject to https://about.jstor.org/terms
 A PAPYRUS-ROLL WINDER FROM POMPEII 35

from the parturient's genitals so as not to cause embarr

plications during the delivery. The poses, gestures, and placement of the three figures all
closely resemble those on the ivory.
The Ostian plaque has a companion piece, however, that also displays parallels to the
mythological scene (fig. 12). The second plaque commemorated the profession of Scribonia's
husband, a physician, whom we see here treating a male patient, in a pose that closely re-
sembles the midwife's.36 Once again the patient sits on a chair facing the physician, who
sits on a lower stool. On the Ostian plaque, the procedure involves washing a wound or
sore, because the patient places his right foot in a basin, and the physician holds a round
object in his hand that may be a roll of bandages or, more probably, a sponge. In the
Pompeian plaque, the woman sitting on the lower stool likewise holds some small, round
object in her hand, and reaches toward the patient's exposed left leg, rather than toward
her groin. On the Ostian plaque of the physician, there is no third figure behind the pa-
tient, who has more freedom of movement. He bends and gestures toward the doctor, per-
haps answering his questions or assisting with the procedure in some way. The woman on
the Pompeian ivory sits rigidly upright, braced by an attendant, to whom she reaches back-
ward for support with her left hand. The Pompeian plaque, therefore, conflates two stock
representations, that of the midwife and the physician. The sex of all figures in the scene,
however, seems unmistakably female, and points toward an interpretation as a birthing scene.
The sponge that the midwife holds is part of her standard equipment, as Soranus dictates.
One of her tasks prior to delivery is applying soothing oils to the parturient's abdomen and
labia.37
Kampen has also examined the iconography of nine sarcophagi that represent the mo-
ment just after childbirth: the bathing of the newborn infant.38 Once again, the new mother,
now modestly covered, sits on the birthing chair, while the old nurse sits or bends in front of
her. The nurse holds the baby by his waist over a basin in which she prepares to wash him.
The attendant behind the mother's chair may or may not still be present. At least four extant
sarcophagi, three of children and one of an adult whose birth appears on a short end, include
this figure.39 Others, however, omit her, since the delivery is complete and her services are no
longer necessary. On all these sarcophagi, other female figures are present, however, usually
three in number, with attributes that suggest Muses, or Fates, or a conflation of the two.
Sometimes they make a transition between the scene of birth and that of the child's educa-
tion, and therefore perform the functions of both sets of goddesses. Usually they include
Urania, who points to a globe, suggesting that she is displaying the child's astrological des-
tiny.40 These figures are borrowed from scenes of mythological births, in which nymphs or
goddesses may attend the child as kourotrophoi, and they lend these biographical scenes a

36 Kampen 1981, 70, 72, 143, cat. I.16, fig. 59. 151 n. 7; Valbruzzi 1993, 155-158, pls. 66.1 and 67.1-2.
C. Museo Chiaramonti, Vatican, inv. no. 1632. Marble,
37 Soranus, Gynaeceia 2.4. height 0.265 m, length 0.53. Valbruzzi 1993, 155, pl. 66.2;
Kampen 1981, 35, 39, fig. 5, cat. no. III.28; Huskinson
38 Kampen 1981, 33-44, text fig. 1; 38-39, figs. 1, 3, 5, 1996, 10-11, 23, cat. no. I.38, with earlier literature.
7-10. D. Florence, Galleria degli Uffizi, inv. no. 82. Marble,
height 0.94 m, length 2.43 m, depth 1.10 m. Kampen
39 A. Paris, Musee du Louvre, MA 319. Kampen 1981, 34, 1981, 38, 40, fig. 8, cat. no. III.24, with earlier litera-
37, 38, cat. no. II1.23, p. 147, with earlier literature, fig. ture.
3; Huskinson 1996, 10-11, 22, cat. no. 1.21, pl. 3 fig. 2.
B. Agrigento, sarcophagus of a child. Amedick 1993, 144,40 Amedick 1993, 145; Valbruzzi 1993, 155.

This content downloaded from

79.230.157.136 on Wed, 18 Oct 2023 10:35:29 +00:00
All use subject to https://about.jstor.org/terms
 36 SUSAN WOOD

certain mythological gra

three Fates, who predict
includes one female figure
back: the veiled woman a
ited space might not have
be one of the Fates.
Since the obverse and r
fact are two sections of
the scenes on Plaque A.
then, must be the same
where that this man is m
although Achilles is also
men's mothers attempte
to an early death which
Meleager's mother, Alth
her that her son would
wood, but after Meleage
she threw it back into t
sible identity for the fa
Roman art, that of the
myth of Meleager.43 Th
gree of popularity in fu
Indeed, H. Herdejiirgen
ion during the second c
to hand in the pattern b
is perhaps no coincidenc
parallels in later funera
for depicting mytholog
of the patron, indicatin
Greek and Latin literatu
ments than on luxury ob
One final question we m
the stories of Persephon
been familiar to most ed
in greatest detail the stor
her vengeful rage.46 The
closely to his version, s
Euripides who describes

41 Mosaic from Nea

44 Herdejiirgen 1989, 24-26. Paphos, C
3, p1. 56 no. 3, s.v. "Achilleus"
with full literature.
45 Robert 1897-1919,3:453-454 noted the close relation-
ship between Plaque B and the later sarcophagi that de-
42 Birth and death of
pict the myth of Persephone; Meleage
this object, indeed, pre-
LIMC 1:578, s.v. "Althaia" (E. Simon). sented the closest parallel to those works in media other
than funerary sculpture.
43 LIMC 6:428-429, pl. 224, s.v. "Meleagros" (S.
Woodford); Koch 1975, 28-38, 106-118 nos. 73-111. 46 Ovid, Met. 8.451-547.

This content downloaded from

79.230.157.136 on Wed, 18 Oct 2023 10:35:29 +00:00
All use subject to https://about.jstor.org/terms
 A PAPYRUS-ROLL WINDER FROM POMPEII 37

their search.47 But Persephone sarcophagi consistently inclu

probably because they appeared in the painting by Nicoma
no doubt guaranteed that later artists would follow the m
slightly different literary source might be uppermost in th
also share comparable themes, not only of untimely death
child. Demeter, the loyal mother, presents a striking cont
myths in combination, then, might have a didactic messag
one side, the reader would see two parallel death scenes: th
the limp, sagging body of Meleager. In each case, the figure
horizontal, but with the shoulders higher than the legs, so
off the center. In each case, too, a vertical figure stands b
cases, the figures transporting the dead or dying person m
tional parallels must be intentional. The equally striking dif
contrasting with Persephone's violent resistance, introduce
From the other direction, the two mothers would present themselves to the viewer in a
single glance, both shown at the moment when they first attempt to foil destiny. Demeter sets
out on her long quest for her missing daughter, while Althaia receives the prophecy about
the burning brand. The designer might rely on the viewer's knowledge of both stories to fill
in the rest: Demeter will be unswervingly devoted in her quest, and her search will at last be
rewarded when Zeus agrees to let Persephone return to the upper world for part of every
year. Althaia, on the other hand, will destroy both her son and herself. Here too there are
similarities of composition, although not as obvious as those of the other pair. In each case,
four figures appear, and in both, the mother of the protagonist bends her left knee at an
oblique angle while gripping a long, diagonal object. Demeter is running forward, holding
torches, while Althaia sits, gripping her scepter in her left hand, but there are similarities in
the nearly frontal presentation of their upper torsos and in the way that the drapery partly
exposes each figure's left leg. In each case, also, the mother receives assistance from other
female figures, Althaia from her midwife and attendant, and Demeter from Artemis and
Athena, the latter of whom repeats Demeter's aggressive, lunging movement. A fourth fig-
ure-"Hekate" on Plaque B and the Fate on Plaque A-looks on, playing no active role in
the scene, but in each story delivering an important message to the mother. Here again, as in
the "death" scenes, an active figure (in this case, Demeter), resembles but also contrasts with
a relatively passive one, and suggests a contrast of the steadfast virtues of the goddesses with
the weaknesses of the mortal Althaia. Such a characterization is consistent with Ovid's por-
trayal of these females: Althaia makes her fatal decision only after much vacillation, during
which Ovid compares her to a ship driven one way by the tide and another by the wind.48
Althaia, in other words, is controlled by rather than in control of her passions. Ceres, on the
other hand, he describes in terms that express decisive strength: she searches day and night
for Proserpina, without eating or drinking, and when Jupiter gives his judgment, she remains
determined to recover her daughter: "Dixerat, at Cereri certum est educere natam."49
All of the surviving book winders from datable contexts appear to belong to the first
centuries B.C. or A.D. We can perhaps infer that book-winding devices passed out of favor

47 Ovid, Met. 5.332-678; Eur. Hel. 1313-1318. 4 Ovid, Met. 5.438-445, 533-534.

48 Ovid, Met. 8.460-477.

This content downloaded from

79.230.157.136 on Wed, 18 Oct 2023 10:35:29 +00:00
All use subject to https://about.jstor.org/terms
 38 SUSAN WOOD

after the Julio-Claudian period, possibly because of their limitations of size. Many extant
sarcophagi of the second through fourth centuries represent the deceased reading a book,
but always simply holding it in his hands, without any sort of lectern.50 If the figure is sitting,
he usually spreads the scroll on his lap. In later times, however, the problems of holding
one's place in a text, especially while reading aloud, continued to present difficulties. Several
of these sarcophagi demonstrate a rough-and-ready solution to the problem, in which the
reader folds the open papyrus back on itself, grips both rolled ends in his left hand, and
gestures with his right.5" One could probably perfect this mannerism with practice, but it is
easy to imagine the book roll slipping from the reader's left hand as he declaimed, causing
him to lose both place and face.
Although their popularity was of limited duration, the survival of eight such objects im-
plies the original existence of many more. Most of the surviving examples are in ivory, a fairly
durable material, but similar devices could easily have been made from cheaper and more
perishable materials such as wood. Another striking fact about several specimens from known
contexts is that they apparently belonged to women. An elite Roman woman not only could
but perhaps should have been literate, since one of her responsibilities was the early educa-
tion of her sons. The owner of the Pompeii specimen might have been a scribe, and therefore
presumably a man if the device did indeed come from 1.2.24. One can understand why a
member of his profession would find such a device particularly useful for holding the texts
that he was copying. Whether the owner was a scribe or a private bibliophile, however, we
can be fairly certain that he, or she, savored the mythological meanings of the decoration
while putting the device to practical use.

Afterword

On 10 September 2001, I flew to Britain to examine the book winder in the collection of the
Fitzwilliam Museum in Cambridge. The following afternoon I learned of the catastrophic
terrorist attacks on the World Trade Center and the Pentagon. After the immediate reactions
of shock, disbelief, and then concern for friends and relatives at home, my feeling was one of
bitter irony. At a time of such a global crisis, could there be a more trivial or inconsequential
activity than my study of ancient luxury objects? As I watched the videos of the disaster over
and over, the flames bursting from the tops of the towers and the clouds of smoke and rubble
pouring onto the city below, I was reminded of another historical disaster, the destruction of
Pompeii. What happened to the resident of Regio I, Insula 2, who owned that ivory device,
and had to abandon it when the disaster struck? Did the person survive, or, as is far more
likely, was he or she one of the thousands of victims? We may never know the individual's

50 For a survey of "philosopher sarcophagi," see Marrou declaming from an open papyrus roll: Huskinson 1996,
1964, passim. Well-known examples of the type include 16, 22, nos. 1.21, pl. 3.2 (Paris, Musee du Louvre, MA
the sarcophagus of L. Publius Peregrinus (Gasparri 1980, 319), no. 1.23, pl. 2.1 (Paris, Musee du Louvre, MA 659),
205, no. 424, with earlier literature) and the sarcopha- and 38-39, no. 5.5, pl. 10.2 (Vatican Museums, Galleria
gus of Junius Bassus in Rome, treasury of St. Peter's Ba- dei Candelabri 1.20, inv. 2422).
silica, in which the enthroned figure of Christ in the
upper central panel holds a partly opened papyrus roll 51 The child's sarcophagus in the Vatican Museums, inv.
in his left hand (Malbon 1990, passim, with earlier lit- 2422 (see previous note), depicts the deceased making
erature). Sarcophagi of children that depict the educa- this gesture twice, once in the relief on the lenos and
tion of the deceased child also sometimes show the child once in the reclining figure on the lid.

This content downloaded from

79.230.157.136 on Wed, 18 Oct 2023 10:35:29 +00:00
All use subject to https://about.jstor.org/terms
 A PAPYRUS-ROLL WINDER FROM POMPEII 39

name or fate, but indeed it does matter almost two millennia later that that person read lit-
erature, perhaps enjoyed Ovid's Metamorphoses, and certainly cherished precious books. May
the literature and art that brought pleasure and meaning into the lives of those who died on
September 11th matter to their descendants another two millennia from now. And for that
reason, I dedicate this essay to their memory.

Bibliography

ABBREVIATIONS

CTP Corpus Topographicum Pompeianum, part 2 (Austin 1983)

LIMC Lexicon Iconographicum Mythologiae Classicae, 8 vols. (Zurich 1981-1997)

WORKS CITED

Amedick, R., "Zur Ikonographie der Sarkophage mit Darstellungen aus der Vita Privata und dem Cur-
riculum Vitae eines Kindes," in Grabeskunst der romischen Kaiserzeit, ed. G. Koch (Mainz 1993),
143-153, pis. 63-67.
Avi-Yonah, M., Ancient Scrolls (London 1973).
Barruol, G., "Circonscription de Languedoc-Roussillon," Gallia 29, fasc. 2 (1971) 369-405.
Beal, J.-C., Les objets de tabletterie antique du Musee Arche'ologique de Nimes (Nimes 1984). Cahiers
des Musees et Monuments de Nimes 2.
Benoit, F., "Nouvelles epaves de Provence (III)," Gallia 20, fasc. 1 (1962) 147-176.
Bordenache Battaglia, G., and A. Emiliozzi, Le Ciste Prenestine (Rome 1979).
Capasso, M., Manuale di papirologia ercolanese (Lecce 1991).
- , Volumen. Aspetti della tipologia del rotolo librario antico (Naples 1995).
Cohen, A., "Portrayals of Abduction in Greek Art: Rape or Metaphor?" in Sexuality in Ancient Art, ed
Natalie Kampen (Cambridge 1996) 117-135.
Della Corte, M., Case ed abitanti di Pompei, 2nd ed. (Rome 1954).
Gasparri, C., Materiali per servire allo studio del Museo Torlonia di scultura antica (Rome 1980). Atti
della Accademia Nazionale dei Lincei, ser. 8, vol. 24, fasc. 2.
Graeven, H., Antike Schnitzereien aus Elfenbein und Knochen in photographischer Nachbildung (Hannover
1903).
Herdejiirgen, H., "Beobachtungen an den Liinettenreliefs Hadrianischer Girlandensarkophagen," Antike
Kunst 32 (1989) 17-26, pls. 5-8.
Huskinson, J., Roman Children's Sarcophagi: Their Decoration and Its Social Significance (Oxford an
New York 1996).
Johnson, W. A., "Column Layout in Oxyrhynchus Literary Papyri: Maas's Law, Ruling and Alignment
Dots," Zeitschriftfiir Papyrologie und Epigraphik 96 (1993) 211-215.
-, "Is Oratory Written on Narrower Columns? A Papyrological Rule of Thumb Reviewed," Pro-
ceedings of the 20th International Congress of Papyrologists, 23-29 August 1992 (Copenhagen 1994)
423-427.
Kampen, N., Image and Status: Roman Working Women in Ostia (Berlin 1981).
- , "Social Status and Gender in Roman Art: The Case of the Saleswoman," in Roman Art in
Context, ed. E. D'Ambra (Englewood Cliffs, N.J. 1993) 115-132; reprinted from Feminism and
Art History, ed. M. D. Garrard and N. Broude (New York 1982) 62-77.

This content downloaded from

79.230.157.136 on Wed, 18 Oct 2023 10:35:29 +00:00
All use subject to https://about.jstor.org/terms
 40 SUSAN WOOD

Koch, G., Die Mythologische Sarkophage, vol. 6. Meleager (Berlin 1975).

Lehmann, P., Roman Wall Paintings from Boscoreale in the Metropolitan Museum of Art (Cambridge,
Mass. 1953).
Lindner, R., Der Raub der Persephone in der antike Kunst (Wiirzburg 1984). Beitrige zur Archaolog
16.
Ling, R., Roman Painting (Cambridge 1991).
Maiuri, A., Roman Painting, trans. S. Gilbert (Geneva 1953).
, Paintings from Pompeii, Herculaneum and Stabia in the Museo Nazionale at Naples (Milan 1958).
Malbon, E., The Iconography of the Sarcophagus of Junius Bassus: Neofitus Iit Ad Deum (Princeton
1990).
Marrou, H., Mousikos Aner (Grenoble 1938; reprint, Rome 1964).
Musso, L., "Il trasporto funebre di Achille sul rilievo Colonna-Grottaferratta," Bullettino della
Commissione archeologica comunale di Roma 93 (1989-1990) 9-19.
Oakley, J., "Reflections of Nicomachus," Bulletin Antieke Beschaving 61 (1986) 71-75.
Robert, C., Die Antiken Sarkophagreliefs: Einzelmythen, 3 vols. (Berlin 1897-1919).
Sogliano, A., "Rilievi di Avorio," Giornale degli scavi di Pompeii n.s. 3 (1874) 12-16.
Spinazzola, V., Le arti decorative in Pompei e nel Museo Nazionale di Napoli (Milan 1928).
Squarciapino, M. F., ed., Scavi di Ostia III. Le Necropoli. Part 1, Le tombe di eta' Repubblicana e Augustea
(Rome 1955).
Strong, D., Roman Imperial Sculpture (London 1961).
Turner, E. G., Greek Manuscripts of the Ancient World, 2nd ed., revised and enlarged by P. J. Parsons
(London 1987). Bulletin of the Institute of Classical Studies, suppl. 46.
Vaglieri, D., "Ostia," Notizie degli scavi di antichita (1912) 95-101.
Valbruzzi, F., "Un sarcofago di bambino rinvenuto ad Agrigento," in Grabeskunst der r6mischen
Kaiserzeit, ed. G. Koch (Mainz 1993).
Vassilika, E., Greek and Roman Art: Fitzwilliam Museum Handbooks (Cambridge 1998).
Vigne, A., "Un coin de sol Nimois," Decouvertes arche'ologiques, extrait du Bulletin du Comite' de l'Art
Chretien 6 (1899) 1-29.
Wood, S., "Mortals, Empresses and Earth Goddesses: Demeter and Persephone in Public and Private
Apotheosis," in I Claudia II: Women in Roman Art and Society, ed. D. Kleiner and S. Matheson
(Austin 2000) 77-99.

This content downloaded from

79.230.157.136 on Wed, 18 Oct 2023 10:35:29 +00:00
All use subject to https://about.jstor.org/terms