University of Virginia Library

I. 13

OMISSIONS AND OVERSIGHTS

I.13.1

INTENT OR INADVERTENCY

It would be incongruous to expect that a plan comprising
forty buildings and other installations, such as a cemetery
and gardens, would be free of omissions or oversights. A
few of these do indeed exist, but for absolute errors one
looks in vain. The majority of missing features, which
might appropriately be termed omissions, appear, however,
to have been left out by intent rather than by neglect or
inadvertence.

Most conspicuous among these are: the lack of consistent
attention to stairs and privies; the absence of any suggestion
of waterways to operate the monastery's water-driven
machinery and to dispose of the monastery's waste; and,
perhaps, the absence of a peripheral wall enclosure.

I.13.2

STAIRS

The traditional assertion that "stairs are omitted altogether"
on the Plan[262] is incorrect. Where stairs are vital to the
liturgical service or are of an extraordinary construction,
they are delineated with the greatest care. Two flights of
seven steps (septem gradus, similiter) lead from the crossing
of the Church to the forechoir (fig. 99). The altars in the
transept rise from platforms that are raised by two steps
over the pavement of the transept arms (fig. 99). The apse
of St. Peter at the western end of the church is raised by
one step over the contiguous pavement of the nave (fig.
84); the same condition is found in the two apses of the
church of the novices and the sick. Finally, the altars of St.
Michael and St. Gabriel at the top of the two circular
towers are made accessible by a circular stair, the winding
course of which is delineated by meticulously drawn spirals
(fig. 84).

On the other hand, one observes with some surprise that
the large double-storied buildings of the monks, which
surround the cloister, contain not the slightest suggestion
of stairs. We do not know at which point and by what
means the monks entered the Vestiary, above the Refectory
(fig. 211), or the Larder, which lies above the Cellar (fig.
225). The structure which houses Dormitory (above) and
Warming Room (below) has four exits. One of them, a
passage leading to Bathhouse and Laundry, is designated by
its title as issuing from the Warming Room. The others
may refer either to ground level or upper story—or to doors
located one above the other on two levels. The Privy was
probably on Dormitory level with cesspool and running
water beneath, but might also have been accessible from
ground level by stairs connecting Warming Room and
Dormitory internally (for suggestions how this might

[ILLUSTRATION]

ROME. THEATER OF MARCELLUS

51.B [after Calza-Bini, 1953, facing p. 15]

Plan showing substructure and two lower tiers of seating. The memorial was built by Augustus, and dedicated to Marcellus in 13 B.C.

51.C [after Calza-Bini, 1953, p. 14]

Plan showing two upper tiers of seating and gallery.

[ILLUSTRATION]

51.A FORMA URBIS ROMAE

ROME. ANTIQUARIUM COMMUNALE DEL CELIO

51.A [after Carretoni, 1960, vol. 11, pl. XXIX]

Group of fragments (no. 13) shows the semicircular rows of seats in the Theater of Marcellus, and the vaulted ramps and stairs by which they
are made accessible, as though they were lying on the same plane, interweaving parts that in the building itself belong to several different levels.

have been accomplished see below pp. 253 and 261; also
cf. fig. 192).

Likewise, in the Abbot's House (fig. 251) we are not told
from what point and by what means the Abbot reached the
solar and other rooms located in the upper level of his
residence. These details the designing architect left to the
ingenuity of the builder—perhaps to protect the Plan from
being overloaded with particulars. It is to policy decisions
of this kind, bringing relief to areas that required minute
attention to other details, that the Plan owes its extraordinary
over-all clarity.

I.13.3

DOORS AND WINDOWS

The location of all exits and entrances is treated with the
greatest care, not only the opening by which each individual
structure is made accessible from the outside, but also all
the doors giving access, internally, from one room to
another. There are altogether some 290 doors shown on the
various buildings of the Plan, and only five true oversights
that I can find: two doors in the House of the Gardener
(fig. 426)[263] and three in the Monks' Bake and Brew House
(fig. 462)[264] are not shown.

The architect does not lavish quite the same degree of
attention on the designation of gates that give access from
court to court through enclosing fences. Here, as in the
case of the privies, he appears to discriminate between the
higher and lower levels of monastic polity. A gate in the
passage that connects the Abbot's House with the Church
(fig. 251) permits the Abbot to inspect the buildings lying
east of the Church, and at the same time admits the novices
and the sick to the Church on the days of the great religious
festivals.[265] A gate in the fence that separates the grounds of
the House of the Physicians from those of the House for
Bloodletting insures that the physicians have free access to
the structures that come within their professional care.
Gates in appropriate places of the enclosure of the Outer
School (fig. 407) permit the headmaster to communicate
with his own quarters (addorsed to the northern aisle of the
Church) and allow the students to attend the divine services
by passing through the quarters of the visiting monks into
the northern transept of the Church.

Yet one looks in vain for gates in any of the fences that
enclose the various installations of the large service yard in
the western tract of the monastery site, with its stables and
houses for the emperor's staff. Here, again, I believe we
cannot speak of these omissions as oversights. The draftsman
was eager to make it clear that these installations
should be surrounded by walls or fences, but the builder,
as he adapted the elements to the terrain, would have to
determine exactly where these enclosures should be made
accessible by gates.

He used the same discretion in the designation of
windows. Arcaded openings are delineated with the greatest
care (Monks' Cloister, cloisters in the Infirmary and the
Novitiate, porches in the Abbot's House);[266] and to make
unmistakably clear what he had in mind, he switched from
vertical to horizontal projection. In all other instances,
windows are omitted—with one exception, the Scriptorium
(fig. 99), where to neglect the appropriate conditions for
lighting would have had disastrous consequences.[267] Openings
for ventilation are indicated in the Monks' Privy (fig.
497), again to stress an important functional need. Had
windows been shown in such buildings as the Dormitory,
the Refectory, and the Cellar, they would have impaired
the clarity of the internal layout of these structures. In the
majority of the other houses, windows could not even be
expected, because these houses belonged to a building type
that had no windows, as we shall show later.[268]

I.13.4

FIREPLACES AND LOUVERS

The primary device for heating the guest and service
buildings of the Plan of St. Gall, as will be shown later,[269] is
an open hearth (locus foci) located in the exact center of the
house; above it in the ridge is a lantern or louver (testu)
for the escape of smoke and for light and air. In most of the
larger houses these two devices are entered very carefully;
but in some they are omitted, most conspicuously so in the
workshops of the wheelwrights and coopers, and the workshops
of the goldsmiths, blacksmiths, and fullers. Again, I
do not think that these are oversights. Rather they stem
from the draftsman's desire to avoid endless reiteration of a
feature that had been clearly established in all of the truly
important houses of the Plan, and could, therefore, be
taken for granted, the more so in installations where fire
was needed not only for warmth, but also for professional
reasons.

I.13.5

WATERWAYS

The availability of a good water supply was a prime condition
for the proper functioning of a monastic settlement.
This was expressed in unmistakable terms by St. Benedict[270]
and can be inferred from countless later accounts of the
selection of suitable sites for new monastic settlements.

Most monasteries were built in the immediate vicinity of
a stream. When, toward the middle of the sixth century,
Cassiodorus the Senator founded the monastery of
Vivarium near his ancestral home of Scyllacium, in
Calabria, Italy, he established it on the river Pellena,
deflected its flow so that it brought drink to the brothers,
serviced the monastery's garden and mills, and filled the
ponds (vivaria) for the stocking and breeding of fish.[271] In
like manner, during the reign of King Pepin (751-768),
when Count Wilbertus and Countess Ada searched for an
appropriate site for the new monastery of Lièssies, they
gave primary consideration to the availability of "water for
the running of the mill, the serving of the bakery, kitchen,

garden, and the other monastic workshops."[272] Even the
hermits were dependent on a good supply of water. St. Gall,
in 612, established himself with full deliberation at the side
of a pool which nature had carved beneath a waterfall of
the river Steinach, in Switzerland, and which he had found
to abound in fish. And a century later when this cell of the
Irish missionary was converted into a cenobitic monastery
by Abbot Otmar (719-759) it was—again deliberately—
erected at the side of this stream.[273] Elaborate waterworks
are known to have been installed by Sturmi (744-799) in
the monastery of Fulda to provide the brothers with drinking
water and to create the required slope for the sluices
which carried the water to the mills.[274]

In general the water required for the sustenance of the
community and the operation of its water-driven works was
diverted from this stream at the upper side of the monastery,
conveyed to the monastic workshops through a carefully
constructed system of flues, and then directed back to
the bed of the stream at a lower level, carrying with it all
of the monastery's waste. In many English abbeys of the
twelfth and thirteenth centuries, where the buildings themselves
have disappeared, the course of the waterways is now
completely exposed, and can be studied under ideal conditions.[275]
When a stream of running water was not available
nearby, the supply had to be brought in from a distant
source by means of an aqueduct.[276]

Such a system of aqueducts existed at the Canterbury
monastery and is depicted on two large sheets of parchment,
now inserted (with somewhat trimmed margins) into the
famous Canterbury Psalter of the Library of Trinity
College, Cambridge.[277] Drawn around 1165, probably by
Wibert (d. 1167) who engineered the system, these drawings
(one of which is shown in fig. 52.A) trace the course of
the water from its source in the surrounding countryside
through five settling tanks—located in cornfields, vineyards,
and orchards—to a circular conduit house; thence,
through a passage in the city walls into the precinct of the
monastery itself. There is branches out into several separate
subterranean systems serving the monastic houses and
workshops, and finally it empties into the large sewers from
which the waste is carried into the town ditch.[278]

A literary parallel to this depiction of a medieval monastic
water system is to be found in Book II of the Vita prima
sancti Bernardi, written in 1153 by Arnold of Benneval,
who refers to the reconstruction of the monastery of Clairvaux
after St. Bernard's return from Rome in 1133 and the
construction of its waterworks as follows:

The Plan of St. Gall nowhere suggests the existence of
any waterways. But it would be incorrect to infer from this
that the availability of water and its distribution throughout
the various monastic shops and houses was not a factor of
first importance in establishing their sites. The majority of
the privies are so placed that wastes can be sluiced through
straight channels, and the water-driven mills and mortars
are located at the monastery's edge, where water of an
adjacent stream was apt to be within easy reach. All other
shops and houses are placed in such a manner as to tie them
without difficulty into a logical and simple water system.

Figure 53 shows how easily a well-planned system of
waterways could be superimposed upon the Plan of St.
Gall.

[ILLUSTRATION]

52.A PLAN OF A WATERWORKS:

CANTERBURY, CHRISTCHURCH MONASTERY

[by courtesy of the Trustees of Trinity Library, Cambridge]

This plan of Christchurch waterworks, together with a supplementary and unfinished
plan of the extra-mural parts of the same waterworks, is inserted as a foreign leaf in
the famous Canterbury Psalter (Cambridge, Trinity College Library, ms. 110, fols. 284b
and 285). The plan dates around 1165 and was probably drawn by Wibert (d. 1167).
It is reproduced here slightly reduced from its original size of 11⅝″ × 16⅝.

For a detailed description and a brilliant analysis of the principles of delineation used in
making this extraordinary drawing, see Willis, 1868, 158ff and 176ff. Additional
literature is cited in James, 1935, 53.

[ILLUSTRATION]

52.B PLAN OF A WATERWORKS: AN INTERPRETATION

CANTERBURY, CHRISTCHURCH MONASTERY

[analysis by Willis, 1868, modified by Wysuph, Horn, Born, 1975]

[ILLUSTRATION]

52.C DESCENT BY GRAVITY FROM SUPPLY SOURCE
AT A HIGHER LEVEL TO TERMINAL
DEBOUCHEMENT

The delineation of Christchurch Monastery tells more as pictorial representation, and
of architectural appearance, than it reveals of functional building planning: waterways
shown are schematic. The document shows a water source on high ground, east of the
Monastery, flowing through five (settling?) tanks in cornfields, vineyards, and orchards,
through the monastery wall to Laver I (east cloister), thence to Laver II (Great
Cloister), thence returning on the east to Laver III. This waterway, with Lavers I,
II, III, may be taken as the primary supply system (solid blue line in Plan and
Diagram). Three secondary branches (segmented blue line) are designated on Plan
and Diagram as 1, 2, 3.

Branch 1 leaves the main line between Lavers I and II, flows southward to a
cemetery fountain, then on to debouche in the Piscina. Branch 2 flows northward
from Laver II to a point south of the Brewery where it turns abruptly eastward to
serve the monk's bathhouse, then flows southward to a tank or catchbasin (M) on the
drainage line (solid red line). Branch 3, departing where Branch 2 flows eastward,
serves the Brewery. A short eastward leg serves the Bakery, a short westward leg, the
Abbot's House. From Laver III, at the end of a short extension eastward, the
primary line terminates, draining into the Piscina (blue dotted line).

In addition to the potable supply system a scheme of drainage (red), more or less
polluted, is discernible. Originating in the Great Cloister, it descends southward and
terminates beyond the walls on the north.

The interpretation (figs. 52.B, 52.C) assumes that the drainage line (red), descending in
a short arc from Vestiarium to abut the roof line on the infirmary complex, continues
directly northward through or under the structure to join the drainage system (from
Piscina and tank M) at or near the Infirmary toilets; thus, non-potable water never
comes in contact with the Piscina.

I.13.6

PERIPHERAL ENCLOSURE WALL

Whether built of wood or stone, or simply in the form of a
hedge, the outer monastery wall is an intrinsic expression of
the concept of monastic seclusion. In 320, when St.
Pachomius founded the earliest Christian coenobium in
Tabennessi near Dendera in the Upper Nile Valley, he
surrounded it with a wall,[280] perhaps not so much for
defensive purposes as for insulating the monastic enclosure
from the noise and impurities of the secular world. The
wall became the symbol of monkish self-determination and
collective integrity.[281] Medieval texts distinguish time and
again between that which is "within" (infra, intrinsecus) or
"without" (extra, extrinsecus, foris, subtus, juxta), which
presupposes a separating enclosure.[282] In Irish and Anglo-Saxon
monasteries this enclosure often consisted of earthen
ramparts with a moat or ditch in front and a wooden
palisade above,[283] or even more simply, just a hedge of thorn
(sepes magna spinea, quae totum monasterium circumcingebat),
as at the monastery of Oundle, a foundation of Wilfrid.[284] A
wooden palisade enclosure existed at the monastery of
Lobbes as late as the twelfth century,[285] but from the end of
the eighth century, monastery walls were with increasing
frequency built of stone;[286] and from the end of the ninth
century, many of these walls assume a distinctly defensive
function (murus in modum castri).[287]

In view of these facts, the absence of a peripheral wall
enclosure on the Plan of St. Gall presents a puzzle. Was it a
feature so self-evident to the inventor of the scheme that he
did not bother to include it? Ot should we presume that it
existed on the original, but was omitted in the copy?

There are two reasons why I believe that it existed on the
original. First, the fences that separate the grounds of the
houses to the north and the south of the Church were useless
unless they connected with a peripheral wall enclosure.[288]
Second, our analysis of the scale and construction
methods used in the Plan will show that the location of the
axis of the Church as well as the major site divisions of the
monastery are related to a system of framing lines (fig. 62),
which on the original would only have meaning if they
defined an outer wall enclosure.[289]

The copyist might have dropped this feature for various
reasons: for one, simply because his sheet of parchment
was not large enough to include it; for another, because a
rectangular wall perimeter may have been meaningless on
the reconstruction of the monastery of St. Gall for which
the copy was to be used. The monastery of St. Gall was
wedged into an irregular area shaped by the capricious
course of the Steinach, whose steep embankments may
have served as an acceptable substitute for masonry walls
for a considerable stretch along the southern and eastern
boundaries of the monastery site. No such natural defenses
existed to the west and to the north, where the terrain is
flat. Yet even here there must have been a clear demarkation,
either architectural or topographical, between the
grounds of the monastery and the grounds of the secular
settlement that had begun to rise around it. There is
documentary evidence for the existence, in Carolingian
times, of a protective perimeter of masonry walls. When the
monastery was attacked by the Magyars in 926 the monks
found themselves compelled to throw up a temporary system
of heavy defense (castellum fortissimum) on the spur of a
nearby mountain.[290] This has been interpreted to mean that
the monastery was not sufficiently fortified to block their
advance.[291] It was doubtlessly in response to this alarming
event that Abbot Arno, in 953-954, decided to construct a
masonry wall with thirteen towers (muros . . . cum turris
tredecim) that encompassed not only the monastery, but
with it the entire city (urbs) of St. Gall.[292]

I.13.7

PRIVIES

With regard to privies either the designer of the original
scheme or the copyist was guilty of an oversight. The
apartment of the master of the Infirmary and the adjacent
room for those who are critically ill (fig. 236) lack this
facility. It is obvious that privies of the same design as
those provided for the master of the novices and for the
sick room of the novices were needed in the Monks'
Infirmary in the portion that corresponds exactly to the
Novitiate.

In all other cases where privies are missing, I believe
they are left out intentionally.[293] More will be said about this
in the chapter on monastic sanitation.[294]

I.13.8

KITCHEN FOR SERFS AND WORKMEN

The ensemble of buildings shown on the Plan of St. Gall
includes no fewer than six kitchens: the Monks' Kitchen[295]
(drawn with all of its furnishings), the Abbot's Kitchen,[296]
the Kitchen for the Novices,[297] the Kitchen for the Sick,[298]
the Kitchen for the Distinguished Guests,[299] and the
Kitchen for the Pilgrims and Paupers.[300] In view of the
meticulous attention given to the need for all of these
installations, we are surprised to note the absence of a
kitchen and dining hall for laymen. The aggregate number
of serfs, workmen, and servants living within the monastic
enclosure, as a rule, exceeded that of the monks.[301] Where
did they eat? Clearly, not in the Refectory. This would be
impossible not only because there is no entrance into the
Refectory for the serfs and servants; but also because a
regulation of the Second Synod of Aachen (817) prescribes
"that laymen should not be conducted into the refectory
for the sake of eating and drinking" (Ut laici causa manducandi
uel bibendi in refectorium non ducantur).[302]

Emil Lesne's contention[303] that the meals of the servants
were not prepared in the Monks' Kitchen is right, in my
opinion, if for no other reason than because of the differences

[ILLUSTRATION]

PLAN OF ST. GALL

WHAT SCHEME OF WATERWAYS?

The Plan invites speculation concerning the water supply for the needs of the monastic
population and the removal, beyond the confines of the site, of wastes and waters polluted
by man and beast. The Plan does illustrate, as a paradigmatic concept, the composition
of structures. But what conditions, topographic, hydrographic, climatic (winds, snow,
rain) governed considerations for site selection?

Essential to the operation and good health of the monastery were good drainage, an
adequate supply of potable water, and a well-functioning scheme of sanitation, all
conditions related to the type of terrain. The site illustrated in figure 53 is entirely
assumptive. Contours provide downslope to the south: a downslope to the west of
about 15 feet in the length of the Plan is shown, (roughly 0.25 inch per foot, or 25
inches per 100 feet). That of 83 privies shown on the Plan, 74 are on or near the
north and east boundaries, may or may not allude to planning for odors as related
to prevailing winds, or to a system by which, for most of the privies, excreta could
be removed by water-carriage for building sites on terrains where generous slope and
abundant water supply existed.

Besides solid wastes, sewage comprises water fouled from baths, kitchen use, lavatories,
and also surface drainage from rain or snow. These fluid wastes may have been
removed by surface channels, or by subsurface conduits, probably of terra cotta made
with nesting joints (such as present day "bell and spigot" clay pipe). Drinking
water is best distributed by conduit with tight or sealed joints to prevent leakage and
contamination from without. These two waterways, supply and waste, must be
distinct and separate at all points through their full length.

Bronze and lead, and sometimes wood, were used for piping by the Romans. Delivery
of drinking water by horse drawn tank or barrel was possible. Probably all of these
materials and modes of water supply and disposal of sewage were in use for a
monastery of the period and, no doubt, varied according to the availability of materials,
the skill of local craftsmen and the level of technology of the region.

On sites where no water was available for sluicing of sewage, surface or underground,
excreta was probably disposed in pits or middens, and treated with ashes or floor dust
and sweepings, then removed at intervals and applied as manure on neighboring
ground. From the agricultural view this was sound practice; from the view of hygiene,
not to be tolerated. Nevertheless, such practice existed and exists today in some
communities.

While the 50 privies along the north boundary suggest planning for a system of
sewage removal, the mills and mortars on the south boundary imply the presence of
a running stream as a power source for their operation.

It is notable that only those buildings of the Plan designated for occupancy by clerics
or by men of noble status are provided with privies. The various buildings for essential
services to the monastery, including services requiring high expertise (cooking, beer
and wine making, goldsmithing, etc) have no privies. These men, about 70 percent
of the monastery population, made shift for themselves according to the customs of
the time and the regulations of the establishment.

The nine privies in Building 4, the Monks' Toilets (adjacent to Building 3, the
Monks' Dormitory) attached to the core of the Plan, pose a problem for sewage
removal. A surface sewage ditch does not seem likely here in the midst of clean
water lines; removal from this central location in underground conduit by water
would be no small feat even if feasible. The proximity of these privies to the Monks'
Vegetable Garden X and the Monks' Orchard Y may indicate that an agricultural
destination for human wastes produced in Building 4 was intended.

Figure 53 suggests how, within the scheme of topography shown, supply lines could
penetrate the east boundary of the site. Interior branches we do not show.

The Plan of St. Gall, by the careful ordering of its structures, could adapt well
to an engineered system of water supply and waste drainage consonant with the state
of the art of the plumber and the sanitary knowledge of the time.

E. B.

SCALE: ⅛ ORIGINAL SIZE (1:1536)

in diet of monks and servants. The monks were forbidden
to eat meat; the servants were not.[304] To provide two
completely different menus in a kitchen not more than 30
feet square for about 300 people[305] would not have been
feasible.

The 370 pigs that the monastery of Corbie hung in its
larder annually were for the serfs, the guests, and the sick.[306]
Abbot Wala of Corbie (826-833), in his list of monastic
officials, mentions a cellerarius familiae, i.e., a cellarer for
the laymen, who is subordinate to the prior, and provides
the monastery's servants with their drink.[307] Hildemar (845850)
in his commentary to the Rule of St. Benedict refers
to a monk whose special task was to take care of the serfs'
needs for food and drink.[308] Yet nowhere in the vast body
of Carolingian consuetudinaries, or for that matter in any
other contemporaneous sources as far as I know,[309] is there
any evidence for the existence of a kitchen for laymen.

The Plan of St. Gall, I believe, enables us to solve this
puzzle. It suggests that the meals of the serfs and the
servants were cooked in their own houses, which differ
from those of the monks in that they were furnished with
hearths and in this way equipped for the cooking of meals.[310]
The serfs apparently continued within the monastic enclosure
to eat as they had before they entered the monastery's
service and to live in the same kind of houses.

Had the serfs and workmen eaten at a common table,
their considerable number would have required the monastery
to have been provided with a dining hall and kitchen
even larger than that provided for the monks. The absence
of such structures on the Plan cannot be interpreted as an
oversight or omission, because presumably the dining
arrangements were handled in quite a different way.

It would nevertheless be unreasonable to assume that
every layman cooked his own meals. In fact, this was probably
not even done by every group of servants living
together in a given house. It is more logical to assume that
the meals for the lay servants were prepared and eaten in
two or three of the larger service structures strategically
located so as to accommodate people living in different
parts of the enclosure. I feel this view is supported by an
unusual arrangement in the House for Horses and Oxen
and Their Keepers.[311] This house is provided with a large
central hall with benches all around its walls, offering sitting
space for forty-three people (fig. 54). The hearth is three
times as large as the hearths in the other houses and
has inscribed into it an

-shaped symbol, which
might well stand for the kind of boom or rigging that in the
Middle Ages was used to hang pots over open fires. It is
here, in my opinion, that the meals were cooked and eaten
by all the serfs and herdsmen who lived west of the Great
Collective Workshop.[312] The latter house, too, may have
served a similar purpose. It is occupied by men of higher
skills and its yard is separated from that of the herdsmen
by a conspicuous fence. A third center of this kind may
have existed in the area of the gardeners and the fowl-keepers.[313]

[ILLUSTRATION]

54. PLAN OF ST. GALL

HOUSE FOR HORSES AND OXEN (BUILDING 33, SEE PLAN PAGE XXIV)

The hall of this building (DOMUS BUBULCORUM ET EQUOS SERUANTIUM) may have served as kitchen and dining space for herdsmen of this
and other houses. Its unusually large hearth, and benches ranging around the walls, offering seating for over forty people, appear to attest
communal use of this space.

[ILLUSTRATION]

56. PLAN OF ST. GALL

Plan of Monastery Church as interpreted by
Ostendorf [after Ostendorf, 1922, 42, fig. 53]
(Illustrated above at about 1:600)

[ILLUSTRATION]

55. PLAN OF ST. GALL

Plan of Monastery Church, shown ½ original size,
scale 1:384