Understanding late medieval farming practices: an interdisciplinary study on byre remains from the historical centre of Brussels (Belgium)

Yannick Devos; Cristiano Nicosia; Luc Vrydaghs; Jan M A van der Valk; Lien Speleers; Elena Marinova; Mona Court-Picon; Terry B Ball; Christine Pümpin; Hugues Doutrelepont; Britt Claes

doi:10.1007/s12520-025-02248-w

. 2025 Jun 25;17(7):154. doi: 10.1007/s12520-025-02248-w

Understanding late medieval farming practices: an interdisciplinary study on byre remains from the historical centre of Brussels (Belgium)

Yannick Devos ^1,^✉, Cristiano Nicosia ², Luc Vrydaghs ¹, Jan M A van der Valk ³, Lien Speleers ³, Elena Marinova ^3,⁴, Mona Court-Picon ³, Terry B Ball ⁶, Christine Pümpin ⁷, Hugues Doutrelepont ⁵, Britt Claes ⁸

PMCID: PMC12198283 PMID: 40585909

Abstract

During the preventive archaeological excavation in the cellar of the site of petite Rue des Bouchers in the historical centre of Brussels (Belgium), the waterlogged remains of a thirteenth century AD sunken byre (potstal) were discovered. The exceptional preservation instigated a multiproxy approach, including micro-archaeology, micromorphology, phytolith and endoparasite egg analysis on thin sections, palynology and the study of plant macroremains on the fill of the structure. Beyond permitting to detail the content and the multiple origins of the fill (including excremental waste and urines, fodder and bedding material, plaggen and/or soil sods, household and construction waste), this study also provides detailed insights into the foddering customs, hygienic conditions within the stable, and the health status of the animals kept. On a more general scale this study documented late medieval farming practices in Brussels, more specifically the need to collect substantial amounts of manure to add as fertilizer onto the cultivated poor sandy soils.

Supplementary Information

The online version contains supplementary material available at 10.1007/s12520-025-02248-w.

Keywords: Stable, Manure, Micromorphology, Archaeobotany, Palaeoparasitology, Medieval

Introduction

Stables and stable manure are complex entities involving the accumulation of materials from different contexts, as well as different processes – both during the use of the structure, as well as post-depositional (Kenward & Hall 1997; Hall & Kenward 1998). Beyond informing us on the kind of animals that were kept, they can provide details on foddering practices, diet and the health status of the animals, waste disposal strategies, the environment of the site, etc. (see Dejmal et al. 2014; Banerjea et al. 2020b; Nicosia et al. 2022). On a more general scale they can also provide insights into farming practices, for instance the collection of manure and its application as a fertilizer. In order to come to a better understanding of such contexts, a holistic approach therefore seems appropriate. However, such studies – although very promising (see Albert et al. 2008; Dejmal et al. 2014; Macphail & Goldberg 2018; 354–368; Garcia-Suarez et al. 2018; Banerjea et al. 2020a & 2020b; Portillo et al. 2020; Alonso-Eguiluz et al. 2024) – remain rare. Most studies still tend to focus on one discipline or treat only one aspect. Soil physical and chemical analyses mostly deal with the detection of ancient stable structures (see Grabowski and Linderholm 2022 and references therein) and the manuring of fields (see Entwistle et al. 2000; Dercon et al. 2005), while botanical studies (for an overview, see Derremaux 2005), for instance, tend to focus on foddering practices and the study of excrements. The growing amount of micromorphological studies mostly deal with issues of structure identification and taphonomy (see Shahack-Gross 2017 and references therein). Interestingly, the study of individual coprolites often gives rise to a more interdisciplinary approach (for an overview on the latter see Linseele et al. 2013; Shillito et al. 2020), and especially promising in this respect are waterlogged sites, due to their excellent organic preservation (Kühn et al. 2013; Kenward and Hall 2012).

During the preventive archaeological excavation on the site of Petite Rue des Bouchers (BR229), in the historical centre of Brussels (Belgium), well-preserved remains of a late medieval stable fill were discovered (Claes 2018). An interdisciplinary study, including micro-archaeology, micromorphology, palynology and the study of plant macroremains, was performed in order to gain a better understanding of the structure, its content and use, the kind of animals kept, their feeding and health, and on a broader scale on late medieval farming practices. To fully exploit the potential of the thin sections, specific studies of phytoliths and endoparasite eggs were conducted. Whereas the study of phytoliths in thin sections has become a well-established method (see for instance Alonso-Eguiluz et al. 2025; Borderie et al. 2020; Devos and Vrydaghs 2023; Kovacs et al. 2020; Matthews 2010, and references therein), the study of endoparasite eggs in soil micromorphological thin sections is a quite young research field (see for instance Pichler et al. 2014; Pümpin et al. 2017), with many aspects that are still to be recognized and recorded systematically. One of the key advantages of this method – comparable to the analysis of the phytoliths – is the ability to gain insight into taphonomical processes and the repartition of the eggs within the sediment – the latter cannot be obtained through traditional palaeoparasitological approaches based on bulk sample analysis.

Site and structure

Brussels is situated in the middle of the Belgian loess belt. However, in the historical town centre, these Aeolian deposits are rare, as along the steep slope and on the Brusselian plateau most soils developed directly on poor decalcified tertiary sands (see Schroyen 2003; Devos 2019), requiring substantial soil enrichment to obtain reasonable yields.

The site of Petite Rue des Bouchers (GPS (Lat/Long) = 50,8477 4,3540) is situated within the alluvial valley of the Senne river at an altitude of approximately 20 m above sea level. In the cellar of the house n°29, remains of a stable fill were observed in profile P2B.4.a, P2B.4.b, P2B.1 in trench 2B and profile P2C.5 in trench 2 C (Figs. 1a-g)(Claes 2018). The structure, dated to the thirteenth century AD, has been dug within dark earth deposits dated to the eleventh-twelfth century AD. These dark earth deposits were identified as being of colluvial origin, possibly with some input of alluvial sediments (Devos et al. 2017). The upper part of the structure is truncated and covered with dark earth deposits dated to the fourteenth-fifteenth century AD. Due to the limited extent of the excavation trench within the cellar, the size of the structure could not be determined. Its fill is composed of a succession of thin layers of due to the prevailing waterlogged conditions rather well to well-preserved organic remains, and therefore particularly well-suited for a multi-proxy approach (Table 1).

Fig. 1 — The studied site. a location of the site of Petite Rue des Bouchers (BR229). The site is indicated in blue, the cellars are indicated in orange; b detail of the cellar with indication of the excavated trenches and the location of the profiles were the fill of the sunken byre was observed; c photo of profile P2B.1; d drawing of profile P2B.1 (the micromorphological samples are indicated in blue); e photo of profile P2C.5; f drawing of profile P2C.5 (the micromorphological samples are indicated in blue); g photo of the bulk lifting of SU 23

Table 1.

Overview of the analyses conducted on the different stratigraphical units (SU) of the fill of the structure. (X = studied)

SU	61	21	62/23	63	36	50	182	183	184	185	186
Micromorphology	X	X	X	X	X	X	X	X	X	X	X
Phytoliths	X	X	X	X	X	X
Parasites	X	X	X		X	X
Micro-archaeology			X
Macrobotany						X
Pollen			X

Open in a new tab

Material and methods

Micromorphology (including the study of phytoliths and endoparasite eggs)

Samples for micromorphology have been taken from profiles P2B.1 and P2C.5 (Figs. 1d & 1f). One subsample of a coprolite has been taken from the bloc lifted from SU 23 (Fig. 1g). Thin sections (60 × 90 mm and 90 × 120 mm) were manufactured by the laboratory for Mineralogy and Petrography, Ghent University (Belgium), according to the guidelines of Benyarku and Stoops (2005). The thin sections were scanned with a flatbed scanner under normal light (NL), plane polarised light (PPL) and under crossed polarisers (XPL) following Arpin et al. (2002). Observations were made with petrological microscopes under plane polarised light (PPL), under crossed polarizers (XPL), and with oblique incident light (OIL) at magnifications of 25x, 100x, 200x, 400x, 500x, and 800x. Additionally, the thin sections were equally studied under UV and blue fluorescence at 200x, 400 × and 500 × magnification (Van Vliet-Lanoë 1991; Stoops 2017). The thin sections were described following the international nomenclature of Stoops (2021). The concept of “Soil Microfabric Type” (hereafter SMT) was adopted (see Macphail & Cruise 2001; Goldberg & Macphail 2006). Each SMT corresponds to a set of common micromorphological characteristics that can occur within different thin sections. They are labelled with progressive numbers. Variations within the same SMT are indicated by decimal numbers (for example SMT 1, SMT 1.1, SMT 1.2). Descriptions are recorded in I-GEOARCHRec (Lo Russo et al. 2023).

Phytoliths

Based on the micromorphological observations, a phytolith study was performed on 5 thin sections (ie. SU 61–21, SU 21–62, SU 62–63-36, SU 50 & SU 23). Phytolith nomenclature and inventories follow ICPN 2.0 (ICPT (Neumann et al.) 2019a & b). Observations were systematically conducted under PPL and XPL at magnifications of 100 × and 400 × to gain general and detailed views of the objects submitted to analysis.

Within the SMTs defined during the micromorphological study 20 to 40 observation fields were scanned systematically under PPL and XPL at a magnification of 400x.

In present case, the phytolith analysis consists of a four-step approach:

a systematic recording of the distribution patterns of the phytoliths (for definitions see Vrydaghs et al. 2017);
the recording of the orientation of the phytoliths within the articulated patterns (side, planar, transverse sensu ICPT 2.0. (ICPT (Neumann et al.) 2019b);
The description of visibility, preservation and colour of the individual phytoliths following Vrydaghs & Devos (2020);
The counting of the phytoliths within each distribution pattern to develop a semi-numerical approach; three categories were defined: ( $\leq$ 5; 5 < x $\leq$ 10 and > 10).

Complementary, the thin section of a coprolite from stratigraphical unit (SU 23) was also extensively studied. This study primarily targeted articulated systems. The analysis of the systems was conducted following a four-step approach:

The counting of the phytoliths within each distribution pattern to develop a semi-numerical approach; three categories were defined: ( $\leq$ 5; 5 < x $\leq$ 10 and > 10);
Their inventory according to ICPN 2.0 (ICPT (Neumann et al.) 2019a & b);
The evaluation of the morphological diversity of the articulated systems (1, 2, 3 or more types);
Whenever possible, the wave patterns of Elongate dentate/dendritic phytoliths following Rosen (1992) were recorded. Sufficiently large and appropriately oriented articulated Elongate dentate/dendritic phytoliths were further analysed following Vrydaghs et al. (2016) and Ball et al. (2016) to provide botanical identifications.

Endoparasite eggs

For the study of endoparasite eggs in Petite Rue des Bouchers four thin sections (i.e. SU 21–62, SU 62–63-36, SU 36–50 and SU 23) were analysed. The thin sections were scanned from bottom to top using a screening pattern of 0.5 cm² in order to ensure a constant recording. Screening was carried out at magnifications of 100x, 200x, 400 × and 630 × with PPL and XPL. The identification relies on Thienpont et al. (1990) and Boch and Supperer (1983). To visualize the presence and distribution of helminth eggs within the sediment, their occurrence was mapped using Adobe Illustrator. Both identifiable and potential endoparasite eggs of each species were recorded using distinct color codes to represent them. A systematic morphometric analysis of the parasite eggs in the thin sections was not carried out, as the classic rehydration method (Bouchet et al. 2003) seemed more appropriate to obtain an accurate size ratio. However, some eggs were sporadically measured.

Study of bulk samples

Micro-archaeology

Two bloc samples from SU 23 impregnated in a solution of water and ethanol have been dissected with a pair of tweezers under a binocular in order to get a better understanding of the 3D organisation of the sediment.

Plant macroremains (including wood and charcoal remains)

A subsample of 1L of sediment from SU50 was measured according to the water displacement method (Antolín et al. 2015) and subsequently wet-sieved using mesh sizes of 2 mm, 1 mm and 0.5 mm. Sieving residues were sorted, and carpological remains were identified and counted under a binocular microscope at 6.5–50 × magnification. Identifications were made using the reference collection of the Royal Belgian Institute of Natural Sciences, identification keys and seed atlases (e.g. Cappers et al. 2006; 2009). The nomenclature of the plant taxa follows Verloove & Van Rossum (2023). The 2 and 1 mm fractions were studied completely, while a representative part (38%) of the 0.5 mm residue was analysed. The counts from this fraction were extrapolated to the entire subsample volume.

Due to the scanning of the dried residue (10,5L) prepared for the archaeozoological and anthracological study, some additional species with large fruit stones could be added to the taxa list. From the sieving residues 150 random charcoal and 100 subfossil wood fragments were selected and identified under a high magnification reflected light microscope.

Palynology

Three samples from SU 23 have been studied. Two different extraction procedures were applied:

1: one set was processed following Faegri & Iversen (1989). Clay contamination was removed by ultrasonic sieving (5 μm mesh opening), although some pollen grains (like those of Juniperus species) could be lost during this procedure. Further, the acetolysis will reduce the finds of non-pollen-palynomorphs (NPP) and most of the intestinal parasite eggs.

2: one sample was processed according to the procedure applied at the Royal Belgian Institute for Natural Sciences for archaeological sediments. It is based on Bastin & Coûteaux (1966), Goeury & de Beaulieu (1979) and Nakagawa et al. (1998), involving dense media separation.

Observations were made using a transmitted light microscope at 400 × magnification. A minimum of 500 pollen grains was counted. Identification was based on identification keys (e.g. Moore et al. 1991; Beug 2004) and the reference collection of the Royal Belgian Institute of Natural Sciences. Nomenclature follows Lambinon et al. (1998). Non pollen palynomorphs (NPPs) were also recorded. Identification follows Van Geel (2001) and Van Geel et al. (2003). Additionally, for two of the three samples, the different Cerealia pollen were identified by application of phase contrast microscopy following Beug (2004). For this, 50 random Cerealia pollen grains were selected from 5 different slides per sample (10 grains per slide).

Results

Micro-archaeology:

The first sample originating from SU 23 is composed of intertwined flattened and compressed lenses, of which three main types have been recognised:

A: the most common category is predominantly composed of fragmented leaves of grasses that were horizontally deposited and show different orientations. The leaves are embedded within a brown organic matrix.
B: these lenses are dominated by grey silty sediments with humified organic material and clean sand sized quartz. Macroscopic organic remains are rare.
C: this type is dominated by randomly distributed and unoriented plant macro remains (including small seeds).

Further, pupae of Diptera and fragments of Coleoptera were observed, as well as some horse- or donkeyhair.

The second sample is an ovoid, dense coprolite, composed of fragmented leaves of grasses randomly distributed within a matrix of largely decomposed plant material.

Micromorphology

The different SMT’s are presented in Fig. 2. The main characteristics of each SMT are summarised and illustrated in Figs. 3, 4 and 5. A summary of the opal of biogenic origin that has been observed in the SMT’s can be found in Table 2. Soil micromorphological descriptions of the SMT’s can be found in the supplementary information.

Fig. 3 — Main micromorphological characteristics of SMT1

Fig. 4 — Main micromorphological characteristics of SMT2

Fig. 5 — Main micromorphological characteristics of SMT3

Table 2.

Opal of biological origin

	Opal of biological origin
	Phytoliths	Diatoms	Sponge spicules	Chrysophycean cysts
SMT 1	Very dominant	Rare		Rare
SMT 1.1	Very dominant	Rare
SMT 1.2	Very dominant	Few	Rare	Few
SMT 1.3	Very dominant	Rare	Rare	Rare
SMT 2	Very dominant	Few	Rare	Few
SMT 2.1	Very dominant	Rare	Rare	Few
SMT 3	Very dominant	Few		Few
SMT 3.2	Very dominant			Few

Open in a new tab

Three main groups of SMT’s were observed:

SMT 1: this microfabric type includes all microstratigraphic units characterised by the presence of horizontally layered vegetal tissue fragments with a more or less crenulated aspect. Subtypes of SMT comprise:

SMT 1.1: composed of strongly compacted herbivore dung.

SMT 1.2: contains abundant mineral matter (silt and very fine sand).

SMT 1.3: characterised by a very dense appearance.

SMT2: this microfabric has been assigned to all microstratigraphic units composed of a mixture of aggregates or fragments of vegetal matter and of mineral components, without traces of internal layering.

SMT2.1 differs from SMT2 only for a coarser texture of the mineral part, comprising more medium sands.

SMT 3: this microfabric includes all microstratigraphic units consisting predominantly of mineral material, mostly with a silt loam texture. Organic matter is still abundant, but not as much as in SMT1 and SMT2. Frequent reworked fragments of Bt horizons were observed.

SMT3.1: differs from SMT3 in being richer in medium sands.

SMT3.2: aggregates of very well sorted very fine sand with occasional clay illuviation traces.

Throughout the studied sequence root and mesofaunal galleries are rare.

Main observed pedofeatures include vivianite nodules/crystal intergrowths (SMT 2, 2.1 & 3), phosphatic crystal intergrowths (SMT 2), pyrite framboids (SMT 1 & 2) and limpid clay coatings (SMT 3.2).

The preservation status of the plant remains varies from well-preserved remains still containing cellulose, over brownified to blackened and humified remains. In SMT 1.1 an important part of the plant remains also shows traces of mineralisation. SMT 3 and 3.2 are exceptions as most of the plant remains are humified.