18 Pages
English

The Palaeobotany of Tell Bouqras, Eastern Syria - article ; n°2 ; vol.11, pg 131-147

-

Gain access to the library to view online
Learn more

Description

Paléorient - Année 1985 - Volume 11 - Numéro 2 - Pages 131-147
Tell Bouqras is a Neolithic site (conventional radiocarbon years : 8350-7850 BP) in eastern Syria, on the edge of the Euphrates valley. Carbonized plant remains were recovered by means of water flotation. The charred seed concentration is generally low. Cultivated plants include emmer wheat, einkorn wheat, free-threshing wheat, hulled and naked barley, lentil and field pea. At the time when the tell was occupied rain-fed agriculture must have been possible on a fairly large scale. Species from the plateau steppe as well as from the valley-floor vegetation are represented in the Bouqras plant record. In addition to the wood of poplar and tamarisk from the riverine forest, woody perennials from the steppe vegetation and probably animal dung were used as fuel. The possible relation between seed contents and archaeological feature (oven, bin) is discussed.
Le site néolithique de Bouqras (8350-7850 В.P. selon la datation С14 conventionnelle) se trouve en Syrie orientale, en bordure de la vallée de l'Euphrate. Des restes de plantes carbonisées y ont été obtenus par flottation. La concentration des graines carbonisées est faible : blé amidonnier, engrain, blé nu, orge vêtue et nue, lentille et pois constituent les plantes cultivées. Pendant l'occupation du tell, l'agriculture sèche devait être pratiquée sur une assez large échelle. Des espèces du plateau steppique et de la vallée sont également présentes. Outre le peuplier et le tamaris de la forêt-tunnel, des espèces ligneuses vivaces propres à la steppe et des excréments animaux étaient utilisés comme combustible. La relation possible entre graines et structures archéologiques (four-silo) est discutée.
17 pages
Source : Persée ; Ministère de la jeunesse, de l’éducation nationale et de la recherche, Direction de l’enseignement supérieur, Sous-direction des bibliothèques et de la documentation.

Informations

Published by
Published 01 January 1985
Reads 127
Language English
Document size 2 MB

Wilhem Van Zeist
Willemina Waterbolk-Van-
Rooijen
The Palaeobotany of Tell Bouqras, Eastern Syria
In: Paléorient. 1985, Vol. 11 N°2. pp. 131-147.
Résumé
Le site néolithique de Bouqras (8350-7850 В.P. selon la datation С14 conventionnelle) se trouve en Syrie orientale, en bordure
de la vallée de l'Euphrate. Des restes de plantes carbonisées y ont été obtenus par flottation. La concentration des graines
carbonisées est faible : blé amidonnier, engrain, blé nu, orge vêtue et nue, lentille et pois constituent les plantes cultivées.
Pendant l'occupation du tell, l'agriculture sèche devait être pratiquée sur une assez large échelle. Des espèces du plateau
steppique et de la vallée sont également présentes. Outre le peuplier et le tamaris de la forêt-tunnel, des ligneuses
vivaces propres à la steppe et des excréments animaux étaient utilisés comme combustible. La relation possible entre graines et
structures archéologiques (four-silo) est discutée.
Abstract
Tell Bouqras is a Neolithic site (conventional radiocarbon years : 8350-7850 BP) in eastern Syria, on the edge of the Euphrates
valley. Carbonized plant remains were recovered by means of water flotation. The charred seed concentration is generally low.
Cultivated plants include emmer wheat, einkorn wheat, free-threshing wheat, hulled and naked barley, lentil and field pea. At the
time when the tell was occupied rain-fed agriculture must have been possible on a fairly large scale. Species from the plateau
steppe as well as from the valley-floor vegetation are represented in the Bouqras plant record. In addition to the wood of poplar
and tamarisk from the riverine forest, woody perennials from the steppe vegetation and probably animal dung were used as fuel.
The possible relation between seed contents and archaeological feature (oven, bin) is discussed.
Citer ce document / Cite this document :
Van Zeist Wilhem, Waterbolk-Van-Rooijen Willemina. The Palaeobotany of Tell Bouqras, Eastern Syria. In: Paléorient. 1985,
Vol. 11 N°2. pp. 131-147.
doi : 10.3406/paleo.1985.4371
http://www.persee.fr/web/revues/home/prescript/article/paleo_0153-9345_1985_num_11_2_4371:
vol. 11/2 - 1985 PALÉORIENT,
THE PALAEOBOTANY OF TELL BOUQRAS,
EASTERN SYRIA
W. VAN ZEIST and W. WATERBOLK-VAN ROOUEN
SUMMARY. — Tell Bouqras is a Neolithic site (conventional radiocarbon years 8350-7850 BP) in eastern Syria, on the edge of
the Euphrates valley. Carbonized plant remains were recovered by means of water flotation. The charred seed concentration is
generally low. Cultivated plants include emmer wheat, einkorn wheat, free-threshing wheat, hulled and naked barley, lentil and
field pea. At the time when the tell was occupied rain-fed agriculture must have been possible on a fairly large scale. Species from
the plateau steppe as well as from the valley-floor vegetation are represented in the Bouqras plant record. In addition to the wood
of poplar and tamarisk from the riverine forest, woody perennials from the steppe vegetation and probably animal dung were used
as fuel. The possible relation between seed contents and archaeological feature (oven, bin) is discussed.
RESUME. — Le site néolithique de Bouqras (8350-7850 В. P. selon la datation С14 conventionnelle) se trouve en Syrie orientale, en
bordure de la vallée de l'Euphrate. Des restes de plantes carbonisées y ont été obtenus par flottation. La concentration des graines
carbonisées est faible : blé amidonnier, engrain, blé nu, orge vêtue et nue, lentille et pois constituent les plantes cultivées. Pendant
l'occupation du tell, l'agriculture sèche devait être pratiquée sur une assez large échelle. Des espèces du plateau steppique et de la
vallée sont également présentes. Outre le peuplier et le tamaris de la forêt-tunnel, des espèces ligneuses vivaces propres à la steppe
et des excréments animaux étaient utilisés comme combustible. La relation possible entre graines et structures archéologiques (four
- silo) est discutée.
1.2. The grid system 1. THE SITE
During the excavations the registration was done
1.1. Introduction by squares of 9 x 9 m. The squares are identified by
their position on the x and y axes of the grid system
The tell site of Bouqras is situated on the western that had been laid out over the tell and which is
edge of the Euphrates valley, с 35 kilometres shown on the surface plan of Bouqras (fig. 2). For
southeast of Deir ez-Zor (fig. 1). The oval-shaped example, square 15/13 is the one with its southwest
mound measures с 2.75 ha (length с 250 m, greatest corner at point of the grid. Mention is made
width с 100 m) and in the centre of the tell the here of the grid system because of the sample
deposits of occupational debris are about 5 m thick. designations. In these sample designations the first
The information on Bouqras presented in this sec four figures are the grid notation, followed by the
tion is taken from the papers by P.A. Akkermans et field number, e.g. 1613/0507.
al.{\), to which the reader is referred for further
details.
1.3. Architecture and stratigraphy The site of Bouqras was discovered by
Dr. WJ. van Liere, at the time F.A.O. soil scientist
A conspicuous feature at Bouqras were the traces at Damascus. In 1965, two adjacent test pits, each
of mudbrick walls which under favourable condimeasuring 4 x 4 m, were excavated by H. de
tions showed up at the surface of the tell. This Contenson and W.J. van Liere (2). In 1976, 1977 and
phenomenon allowed the mapping of the architec1978, excavations at the site were carried out by a
ture at the surface, the result of which is presented team from the Universities of Groningen and
in figure 2. Amsterdam under the direction of Professors
H.T. Waterbolk and M.N. van Loon. The test pits of H. de Contenson and van Liere
had been carried down to the virgin soil. During the
1976-1978 campaigns five 9 by 9 m squares, num
bered 15/13 to 19/13, were excavated to the (1) AKKERMANS et ai, 1981, 1983. bed-rock. On the southwestern slope of the tell, over (2) CONTENSON and Van LIERE, 1966; CONTENSON,
a rather large area the surface layers were excavated 1985.
131 1. — The location of Tell Bouqras FIG.
(redrawn from Akkermans et al., 1981 : fig. 1).
for a detailed examination of the architecture (fig. 3). made of bone (awls, tubes, pins). Cups, vases, plates
and dishes of limestone and harder material were In addition, several test trenches were opened up to
establish the extent of the habitation. recovered. The 1976-1978 campaigns yielded appre
ciable numbers of pottery sherds, indicating that the The majority of the houses excavated were built
original view of Bouqras as an aceramic site is according to the same principle, viz. rectangular to incorrect. In addition, fragments of white ware, square rooms bordering on an unroofed courtyard made of gypsum, were found. Figurines with human which usually had an oven in one of the corners. The and animal representations show a fair amount of lay-out of part of the houses near the surface of the variation. Other « art » objects include an alabaster tell was somewhat different. Here the groundplan vessel in the shape of a hedgehog and a wall- was tripartite (fig. 3, houses 16-20) and there was no
painting representing ostriches or cranes. courtyard. The mudbrick walls rested on wall foun
The biological material recovered at Bouqras dations likewise made of mudbricks. Walls as well
consists of human skeletal remains, animal bones as floors were usually plastered with gypsum.
and charred plant remains. The animal bones are of Ten major architectural levels, numbered 1 to 10
domestic animals (sheep, goat, cattle, dog and from top to bottom, are distinguished. Occupation
possibly pig) as well as of wild species (gazelle, debris between the floor of level 10 and the terrace
aurochs, wild sheep and deer) (4). The vegetable subsoil is indicated as virgin soil (vs). Radiocarbon remains will be the subject of this paper. A preldeterminations date the occupation of the site from iminary note on the palaeobotany of Bouqras has 8350 to 7850 BP in conventional radiocarbon years, been presented by W. van Zeist and W. Waterbolk- which corresponds to approximately 7250-6650 ВС van Rooijen (5). in calendar years (3).
Although some comparison with other sites will
be made, the discussion will be principally confined
1.4. The finds to the Bouqras plant remains. Thus, no attempts will
be made to place the Bouqras plant husbandry in a
The lithic materials include implements of flint broader context or to discuss the role of Bouqras in
the development and spread of Neolithic plant and obsidian, chopping and splitting tools, and
grinding and rubbing stones. Tools had also been
(4) CLASON, 1979/80; in AKKERMANS et al, 1983. (3) Cf. KROMER et ai, in press. (5) In AKKERMANS et al, 1983.
132 FIG. 2. — Contour map of Tell Bouqras showing traces
of architecture visible at the surface,
excavated areas (framed areas) and test trenches
(Akkermans et aL, 1981 : fig. 5)
cultivation. Discussions of this kind are certainly river valley. The border between plateau and valley
appealing but not very fruitful as long as no other is formed by a rather steep slope. The valley floor
comparative data are available. consists of alluvial deposits of Holocene age. To the
west of the valley the plateau has a gypsiferous
subsoil. The tell itself, on the edge of the plateau,
lies on the remnant of a Late Pleistocene terrace (6). 2. THE ENVIRONMENT
A former meander of the Euphrates comes close to
The landscape of the Bouqras area consists of two
components, viz. the high plateau and the Euphrates (6) BOERMA, 1979/80, in AKKERMANS et al, 1983.
133 :
:
but it could not be established whether this Bouqras, Populus euphratica (poplar), constituents of the
meander was in operation during the occupation of riverine forest. At present the river valley is under
the site (7). intensive irrigation agriculture and forest is no
longer found. Mean annual precipitation is approximately
125 mm. Precipitation falls mainly in the period
from October until March. Summers are hot and dry.
Under these climatic conditions the natural upland 3. THE MATERIAL vegetation is a steppe or desert-steppe. As a result
of overgrazing the vegetation has been seriously
affected. Of the plant species observed in the spring 3.1. Field and laboratory work of 1965 on the plateau in the vicinity of the site
(vegetation cover from 5 % to locally 30 %) are During the successive excavation campaigns mentioned here (identifications by Dr. Y.I. Barkou- (1965, 1976, 1977, 1978), samples were taken from dah, Damascus) : various provenances and features for botanical — Poa sinaica Steud. (common) examination. In the field, charred plant remains
— Carex stenophylla Wahlenb. (common) (seeds, fruits, charcoal) were recovered from the soil
by means of manual water flotation, except for some — Artemisia herba-alba Asso.
charred grain samples that were dry-sieved. For only — Helianthemum salicifolium (L.)Mill. some of the has the volume of soil floated
— Daphne angustifolia C.Koch been measured (and recorded), so that charred seed
— Suaeda asphaltica Boiss. concentrations per volume of soil could not be
determined (if one had wanted to do so). In this — Prosopis farcta (Banks et Sol.)Macbride
paper the term « seeds » will often be used in the — Ephedra spec. sense of « seeds and/or fruits ». The difference
— Astragalus spinosus (Forsk.)Muschl. between seeds and fruits is usually not evident to a
non-specialist in the field of plant anatomy. — Astragalus cruciatus Link
Only at two places were seeds observed with the — Peganum harmala L.
naked eye. In the fill of houses 12 and 25 (fig. 3) — Evax anatolica Boiss. et Heldr. fairly great numbers of cereal grains occurred. These
— Plantago cylindrica Forsk. charred grain concentrations must have been the
— Ziziphora tenuior L. remains of supplies stored in houses that were
destroyed by fire. The other samples are from — Silène conoidea L.
deposits in which no seeds were visible in the field. — Malcolmia crenulata (DC.)Boiss. Samples were secured from the fill of ovens with
— Iris sisirynchium L. clear signs of the burning of vegetable material
(charcoal, dark soil). Another type of deposit fr— Koelpinia linearis Pall.
equently sampled is the fill of rooms (including the — Calendula persica C.A.Mey indication « on floor »), with and without visible — Salvia de serti Decne. charcoal remains. Various samples are from the
— Matthiola bicornis (Sibth. et Sm.)DC. contents of bins. A few types of bin are distinguish
ed : rectangular bins of gypsum plates fixed against — Gagea reticulata (Pall.)Schult.
a wall; gypsum-plastered, bell-shaped bins dug into — Schismus arabicus Nees the floor; and silo-like bins inside the walls with a
— Adonis dentata Del. round opening in the front side (8).
— Scleropoa memphitica (Spreng.)Parl. Preliminary examination of samples was carried
— Roemeria hybrida (L.)DC. out in the field, but the definitive analyses were done
in the laboratory. The sorting of the samples and the — Silène damascena Boiss. et Gaill.
seed identifications were carried out under a — Ceratocephalus falcatus (L.)Pers. low-powered wild m-4 stereo-microscope, at magnif
ications of 6 to 50 times. For the charcoal identifica
Wadi bottoms will naturally be covered by more tions, which have been executed by Mr. R. Neef, a
luxuriant vegetation because of the accumulation of high-powered, incident-light microscope, at magnifi
run-off water. At present in the wadis cereals and cations of 40x to 400x, was used.
vegetables are grown with varying success. Of the material secured for botanical analysis
In contrast to the treeless uplands, forest covered only part could be examined within a reasonable
at least part of the periodically flooded valley floor. period of time. Many samples turned out to be
Tamarix (tamarisk), Fraxinus (ash) and perhaps a disappointingly poor in seeds and fruits and some
few more arboreal species were, in addition to of them did not yield identifiable seeds at all. The
(7) AKKERMANS et ai, 1981 494. (8) AKKERMANS et al, 1981 499.
134 9m J
FIG. 3. — Plan of buildings on the southwestern slope (Akkermans et ai, 1981 : fig. 7).
3.2. The presentation of the results examination was further handicapped by the rather
poor preservation of a great number of seeds (de
formations as a result of the carbonization, broken No full data obtained for the samples included in
specimens). It was felt that the analysis of a greater the discussion of this report are presented here. As
number of samples would not contribute substant an example, the full results of the examination of
ially to the charred seed record of Bouqras. Simi samples from square 16/13 are shown in table 2,
larly there was no sense in continuing the charcoal while table 6 presents the numbers of seeds and
identifications very much beyond the point at which fruits in samples from the charred grain deposits
the results had become stabilized. As for the samples (6.5.). The identifications of charcoal in samples
which yielded seeds and fruits, the results for 97 of from square 16/13 are shown in table 3.
these are included in the discussion of this report, For various groups of samples mean seed percenand in 75 samples (in part the same as the seed tages (Table 4) and mean charcoal volume percentasamples) charcoal was identified. ges 5) have been determined. Mean
The seed and fruit types attested for Bouqras are frequencies should facilitate the comparison be
listed in table 1. Illustrations of many of these types tween groups of samples. Differences between sam
are shown in other papers on Near Eastern archaeol ples from different provenances could be due to
ogical plant remains prepared in the Biologisch- different functions of the features involved. Possible
Archaeologisch Instituut. Reference to the published differences in the seed or charcoal contents of upper
illustrations is made in table 1. In addition, a few and lower levels could point to changes in the plant
of the Bouqras seed types are depicted in figure 4. husbandry and/or environmental conditions.
135 -
TABLE 1
List of seed and fruit types attested for Bouqras. Total number of samples is 97. For the wild plant taxa it is indicated in which paper(s) the
seed type concerned is depicted. The numbers after the site names are figure numbers.
sample illustrations frequency
-lordeum "nudum" 26 Hordeum distichum/vulgare 1 12 -lordeum spec. Hordeum internode( s) ( fragments) 19 Triticum durum/aestivum 16 durum/aest. internodes 10 Tritic чя dicoccum 10 Num monococcum 5 "■urn spikelet forks 11 Tritic glume bases 12 Tritic :um spec. 10 :um boeoticum-type 1 Ramad 11:1-3 _ens с 3 ulinaris 1 Disurn sativum
Aegilops 1 Ramad 11:7 29 es-Sinn 1 :4; Ramad 23:5 Aizoon hispanicum Erbaba 10:2; Mureybit 5:9 Alyssum-type 4 Androsace maxima 3 Ramad 30:8 Arnebia decumbens 40 22:1 ,2; Mureybit 6:1 Arnebia linearifolia 18 Ramad 22:3,4; Mureybit 6:2,3 28:7,8; es-Sinn 1 :7 Astragalus 36 3romus (sterilis-type) 4 Ramad 25:1-3 Camelina-type 1 "apparis 2 Ramad 24:1; Mureybit 6:10,11 Capsella-type 1 2 cf. Centaurea 1 Cephalaria Umdent. Cruciferae 1 Cyperus 3 this paper, fig. 4:2 Ramad 25:4; Mureybit 7:3,4 iremopyrum 10 icus 5 30:1 Galium 3 Umdent. Gramineae 23 jypsophila 1 2 Ramad 23:6 Helianthemum 16 23:7; es-Sinn 1 :3 Heliotropium Hordeum (wild) 22 Ramad 26:5 Jnident. teguminosae 6 _ i num (wild) 2 .. i thospermum tenuiflorum 18 Ramad 22:6,7 26:6 l ol i um 4 Malva 8 Ramad 24:11 ^edieago (minima-type) 2 Ramad 28:4 4icromeria* 2 Mureybit 5:13 Phalans 2 Ramad 25:6,7 cf. Phragmites (stem fragments) 3 Distacia 8 Ramad 21 :4
plantago (psyllium-type) 2 Rarnad 30:4 Doa-type 1 ^olygonum ( venantianum-type) 14 Ramad 30:7; Mureybit 5:8 3rosopis 7 this paper, fig. 4:1
^umex ( inner fruit) 1 Scirpus maritimus 42 Ramad 24:5 5etaria( -type ) 8 es-Sinn 1:1,2; Mureybit 7:7, Silène 1 Erbaba 10:4,5; Ramad 23:4 1 Solanum-type Suaeda 10 Mureybit 5:2-4 Ihymelaea 3 Erbaba 10:7; Ras Sharnra 7:12 Trigonella astroites-type 17 Ramad 28:2,3 Typha 1 Verbascum 2 this paper, fig. 4:3,4 Vicia 1
* A well preserved fruit from Bouqras got lost before it had been drawn.
es-Sinn: van Zeist 1979/80; Erbaba: van Zeist & Buitenhuis 1983; Ramad: van Zeist & Bakker-Heeres 1982Í19B5); Ras Shamra: van Zeist & Bakker-Heeres , in press; Mureybit: van Zeist & Bakker-Heeres, in press.
examined in part; estimates of total seed frequencies The mean frequencies have been calculated as
follows. For each of the seed types the numbers were made for these samples. Calculation of mean
found in the samples from one particular prove seed percent includes these converted figures. In the
nance group (and level) have been added up and calculation of the mean seed frequencies presented
thereupon the relative frequencies are expressed as in table 4, crop-plant seeds and fruits and unident
percentages of the sum of all seeds per group of ified types are not included. A few samples with an
samples. Upper levels include levels 1-5, and lower anomalously great number of seeds of one type have
levels are 6-vs (virgin soil). Some of the samples were been excluded from the calculation of the mean
136 .
TABLE 2
Numbers of seeds and fruits in samples from square 16/13.
Jrovenance
0026 0367 0424 0497 0507 0533 0536 0541 0544 0548 0555 0572 0602 0613 0615 Sample number (1613/...) Level 257778888B8 8-9 9VSVS Part of sample examined 1/1 1/1 1/1 1/5 1/2 1/1 2/5 1/1 1/1 3/5 1/4 1/1 1/1 1/9 1/1
Hordeum "nudum" 1 1 3
Hordeum spec. 2 internode(s) ( fragments) 1 3 3 2 8 3 Triticum durum/aestivum 1 durum/aest. internodes 1 (die.) spikelet forks 1 1 4 glume bases 2 6 Triticum spec. boeoticum-type Cereal grain fragments Lens
Aizoon 2 1 1 4 Alyssum-type 1 1 1 Androsace maxima 7 3 6 1 Arnebia decumbens 3 linearifolia 1 Astragalus 1 23 3 1 1 3 2 2 2 Bromus (sterilis-type) 1 4 Capsella-type 1 cf. Centaurea 2 Cephalaria Cyperus 1 1 Eremopyrum 1 1 1 Ficus 2 Galium
Unident. Gramineae 1 Helianthemum Heliotropium 1 5 Hordeum (wild) 4
Unident. Leguminosae 5 Lithospermum tenuiflorum Lolium 2 1 Malva 1 Medicago (minima-type)
Micromeria 1 Pistacia + Polygonům ( venantianum-type) 1 Prosopis 1
Rumex 1 1 Scirpus maritimus 103 37 3 5 У Q 2 Setaria(-type) Suaeda 1 1 Trigonella astroites-type 13 3 2 1 21 Unidentified 5 4 4 2
TABLE 3
Charcoal in samples from square 16/13. Volumes in ml.
Sample -i Q_ number E CO О 1613/.. Level Provenance Q_
0.2 4 0367 5 oven 2 0424 7 refuse pit 3 10 110 0497 7 oven 2 6 95 7 18 80 Ó.5 310 3 0.5 81 0507 7 oven 25 7 1 12 4 70 0516 8 bin 9 28 300 0520 area fill 7 24 120 8 0. 4 12 46 0522 8 oven 0.5 100 0533 8 9 12 doorway 0536 8 street fill 24 1 58 0537 8 52 8 260 oven 32 40 5 350 0538 8 10 0541 8 in between floors 3 0.5 ? 0542 8 6 14 55 0544 8 bowl 11 6 21 oven 0. 1 9 72 0548 8 98 8 oven 7 0555 8 10 41 300 8 oven 0. 5 9 90 ash pit 7 20 0.8 350 0562 8 underneath bin 18 30 120 0572 9 32 0.5 500 0613 vs oven 80 vs 0. 5 6 45 0615 vs oven 5 30 105
mainly very small fragments and charcoal dust TABLE 4
Mean percentages of seeds and fruits of wild plant species in various groups of samples. For explanation and discussion see 3.2., 4., 6.3.
Room fill Oven fill Fill bin etc.
lower upper upper lower upper total total tota levels levels levels levels levels
Number of samples 18 5 23 4 12 16 6 9 of types 20 15 27 13 31 35 13 16
Aegilops 0.6 0.1 Aizoon 3.3 9.7 5.0 3.6 5.1 4.9 3.8 3.4 Alyssum-type 0.3 0.2 0.8 0.7 0.2 Androsace maxima 0.2 0.2 Arnebia decumbens 59.6 43.5 67.5 2.7 12.2 17.3 26.4 linearifclia 2.3 1.7 2.4 0.8 1.0 1 .1 1 .4 Astragalus 5.9 30.1 11.4 9.8 12.2 12.5 14.5 Bromus (sterilis-type) 0.9 0.2 0.6 0.5 Camelina-type 3.6 0.5 Capsella-type 0.4 0.3 cf. Centaures 0.8 0.7 Cephalaria 0.4 0.3 Unident. Cruciferae 0.7 0.5 Cyperus' 0.2 0.8 0.7 Eremopyrum 5.3 1 .4 1.2 1.9 1 .8 Ficus 0.8 0.7 Galium 0.9 0.2 Unident. Gramineae 1 .6 4.4 2.4 11.8 2.2 3.7 1 .5 1.5 Hel i anthemům 1.8 1 .6
Heliotropium 1.6 6.2 2.9 1 .2 4.3 3.8 1 .7 1.4 Hordeum (wild) 0.2 15.0 4.2 1.2 6.5 5.7 5.2 4.7 Unident. Leguminosae 0.7 0.5 3.7 3.1 25.6 21.3 Linum (wild) 8.0 2.2 0.4 0.3 Lithospermum tenuiflorum 1.3 1 .0 4.7 0.8 1 .4 Lolium 0.9 0.2 2.2 1 .9 Malva 0.3 0.2 1.2 0.4 0.5 Medicago (minima-type) 0.2 0.2 Micromeria 0.3 0.2 0.4 0.3 Pistacia + + Plantago (psyllium-type) 0.9 0.2 Poa 0.4 0.4 Polygonům (venantianum-type) 1.6 4.4 2.4 0.8 0.7 5.1 4.3 Prosopis 5.3 3.8 1.2 0.2 Rumex 0.4 0.3 Scirpus maritimus 7.9 7.1 7.7 24.9 21.3 0.4 2.7 0.7 0.5 5.3 4.5 0.7 Setaria(-type) Solanum-type 1.0 0.7 Suaeda 12.9 11.0 0.2 0.2 Thymelaea 0.7 0.5 Trigonella astroites-type 4.6 5.3 4.8 5.7 4.9 23.1 19.1 Typha 0.8 0.2 Verbascum + +
percentages. Mean charcoal volume percentages
(Table 5) have been determined in a similar way as
the mean seed frequencies.
Objections can be raised against the use of mean
percentages as it is done in this paper. One of them
is the fact that the samples of one group show a
rather great diversity in types as well as in numbers
of seeds, whereas mean frequencies easily suggest a
more uniform distribution. In fact, it turned out that
some differences between groups are mainly due to
the seed contents of one particular sample, which,
of course, has its consequences in evaluating the
data.
FIG. 4. — A few seeds and fruits from Bouqras. 1 : Prosopis; 2 :
Cyperus; 3 : Verbascum, part of seed capsule; 4 : Verbascum, seeds.
138 TABLE 5
Mean volume percentages of charcoal in various groups of samples. For explanation and discussion see 3.2. and 6.1.
Upper levels (1-5) 30 22.2 73.6 2.2 0 .1 0.1 1.7 1313.1 Lower levels (6-vs) 45 33.3 65.3 0.4 0.8 0 .2 0.03 1691 .8
Room fill/on floor Upper levels _ 9 28.9 61 .4 9.2 0.2 0.3 294.7 Lower _ _ _ 6 3.0 95.9 1 .1 178.3 - Total 15 19.2 74.4 5.7 0.5 0.2 473.0
Oven
(23 from lower levels) 24 33.7 64.5 0.2 1.0 0 .4 0.1 1005.6
Bin/bowl _ - (6 from upper levels) 8 9.9 87.7 2.3 0.1 483.5
4. THE ORIGIN OF THE CHARRED PLANT Prosopis are inedible for man, but the pods are eaten
REMAINS by sheep and goat. Thus, Prosopis seeds could very
well have been present in dung fuel. The fruits of
Scirpus maritimus (sea clubrush) have the highest
sample frequency of all seed types attested for In the last few years the way in which seeds and
Bouqras (Table 1). This species formed part of the fruits may have arrived in the settlement and subse
vegetation of the river floodplain (see 6.1.) and one quently have become incorporated in the occupatio
nal deposits in charred condition has been the may assume that at least during part of the year
domestic animals grazed on the riverine vegetation. subject of renewed discussions. The hypothesis of
H. Helbaek(12) has made a point that the seeds of dung fuel as a potential source of charred seeds (9)
Astragalus and other wild leguminous species were has greatly contributed to our understanding of the
collected by the early farmer-gatherers for human origin of plant remains in settlement sites. Some of
consumption. On the other hand, the rather high the seeds and fruits eaten by domestic (and wild)
frequencies of Astragalus (milk vetch) seeds in some herbivores pass through the digestive tract undamag
of the samples could point to the use of woody ed (10). The burning of dung, either as such or
perennials as firewood (see 6.2.). made into dung cakes, may result in the carbonizat
ion of seeds and fruits that were present in the With regard to the high proportions of Arnebia
animal droppings. Although wood was plentifully decumbens the following should be remarked. Bou
available in the vicinity of the site (see 6.1.), animal qras is not an isolated case, but various other
dung may nevertheless have been used as fuel at archaeological sites yielded likewise considerable
Bouqras. Domestic goat, sheep and cattle are r quantities of seeds of Arnebia decumbens and/or of
epresented among the animal bones from the site (11). other Boraginaceae {Lithospermum tenuiflorum,
It is clear that the two charred grain deposits L. arvense). It is unlikely that these species played
constitute the remains of supplies stored in houses such a prominent role in the ancient vegetation, but
which burnt down. Most of the charred cereal grains they must be seriously over-represented in the
dispersed through the tell deposits were probably the archaeological seed record. This may in part be due
result of minor accidents during household activi to the fact that because of the silica skeleton also
ties, but some of them may have come from dung non-carbonized seeds remain preserved. On burning
the seeds of various Boraginaceae do not turn black, that was burnt. In addition to those of the cultivated
but they become whitish to yellow-grey. For boragi- plants, seeds and fruits of wild food plants, e.g.
naceous seeds in archaeological sites it is often Pistacia (pistachio) nuts, were brought to the settl
difficult to determine whether or not they have been ement for human consumption.
in contact with fire. In this respect charred remains It is difficult, if not impossible, to indicate with of the interior of the seeds are informative. A more confidence which seed types could particularly be serious problem is the uncertainty whether these due to the burning of dung, assuming that dung fuel seeds are of the same age as the deposit in which was already used at Bouqras. The large seeds of they were found or whether they are due to a later
intrusion, since modern and near-modern seeds of
Arnebia, Lithospermum and various other boragina- (9) MILLER, 1982; MILLER and SMART, 1984.
(10) Cf. BOTTEMA, 1984.
(11) CLASON, 1979/80; in AKKERMANS et al, 1983. (12) HELBAEK, 1969.
139