Introduction
Archaeologists often study social stratification in past communities through the inference of attitudes towards the deceased, manifested in the splendour of the funeral ceremony. This approach has various limitations, however, including the under-representation of organic materials, which are vulnerable to decay in various depositional conditions. In the cemetery of the Wielbark culture at Czarnówko, northern Poland, the state of preservation of human skeletal material is poor, and this observation also applies to other organic finds, including wood (Stępnik Reference Stępnik and Schuster2018; Skóra & Skrzyński Reference Skóra and Skrzyński2025) and textiles (Maik Reference Maik and Schuster2018). The durability of organic materials depends on their potential contact with metal objects and the subsequent conserving effect of metal salts migrating into the organic matter.
Located on the southern coast of the Baltic Sea, the Czarnówko cemetery contains more than 2000 graves (Figure 1). It was established in the younger pre-Roman period (mid-second century BC) and was in continuous use until the younger Roman period (first half of the third century AD). The cemetery is located in the northern zone of the Wielbark culture (currently northern Poland), which developed in this area from about the mid-first century AD to the middle of the third century AD, and in the lower Vistula River Valley until the beginning of the fifth century AD. In the second century AD and later, migration of the Wielbark culture population saw its expansion to areas east of the Vistula River, eventually covering vast areas of modern-day north-western Ukraine (Cieśliński Reference Cieśliński and Rzeszotarska-Nowakiewicz2016).
The location of the Czarnówko cemetery (red dot) within the Wielbark culture area (extent within Ukraine not shown): 1) early Roman period extent; 2) late Roman period extent (figure by J. Schuster).
Both inhumation and cremation burials are observed at Czarnówko with varying quality of grave furnishing. Assessment of this aspect of funerary rituals is hampered by the practice of reopening graves during antiquity (e.g. Skóra Reference Skóra2023). The numerous graves with very limited furnishings may therefore reflect a practice of interring limited goods or the goods may have been secondarily removed or lost as a result of practices of post-depositional manipulation. For this reason, burials of local elites, distinguished by their rich grave goods, are of particular importance in this context (Schuster Reference Schuster2018).
Concentrations of plant remains were identified in four graves (R263, R289, R300, R430; Figure 2) during excavations at Czarnówko. These deposits have a compact structure, with one previously identified as a ‘moss pillow’ (Schuster Reference Schuster2018: 11, pl. V; Figure 3A & B), and are currently stored at the Museum in Lębork, Pomerania. Given the scarcity of comparable finds from this period, these deposits provide a rare opportunity to broaden our understanding of the role of organic materials in burial contexts, which is an aspect of funerary behaviour that is otherwise largely invisible in the archaeological record. This article presents the results of their archaeological and archaeobotanical evaluation and interpretation within the context of the burial customs of the Central European Barbaricum (the area beyond the geographic limits of the Roman Empire) during the Roman period (first–fourth centuries AD).
Plan of the cemetery in Czarnówko showing the locations of the four graves containing cattail remains (Typha sp.) (figure by J. Schuster & M. Marczewski).
Plans of graves R263 (1) and R300 (2) at Czarnówko, the greenish colour indicates the location of cattail remains within each grave; A & B) photographs of the cattail ‘pillow’ in grave R300 showing a round imprint (A) and demonstrating that the silver cup foot recovered from the same grave fits the imprint (B) (figure by J. Schuster; images A & B after Schuster Reference Schuster2018: pl. V).
Materials and methods
The four graves that contained plant macro-remains featured inhumation burials with log coffins, dating from the first two centuries AD (subphases B2b–c to B2/C1a of the Roman period; cf. Schuster Reference Schuster2018: 11–12, 19–21, pl. I–VIII, XLVIII–LXIV). Additional furnishings suggest that graves R300 and R430 contained burials of members of the elite (Schuster Reference Schuster2018), while graves R263 and R289 present an average model of grave equipment. Bones have either not preserved or survive only as small fragments; osteological identification of sex and determination of age of the deceased is therefore not possible.
In grave R263, a compact layer of organic material was identified beneath two brooches and the remains of the skull (Figure 3). A fragment measuring 47 × 45mm and up to 16mm thick is preserved in the museum collection, along with numerous smaller fragments (Online supplementary material (OSM) Table S1). In grave R289, the plant layer survived in the form of concretions measuring 57 × 35mm and up to 20mm thick, as well as several small fragments. Preserved in this material is the imprint of a metal artefact, possibly a brooch. In addition, archaeologists documented here the presence of wool remains (fabric). Three fragments of organic material were found in grave R300. One fragment (90 × 75mm, with a thickness of up to 6mm) preserves the imprint of a round, metal artefact and small fragments of textiles. The diameter of the imprint is 32mm, which matches exactly the diameter of the foot of a silver drinking cup recovered from the same grave (Figure 3A & B).
A substantial quantity of organic remains, in the form of ‘pillow’-like concentrations, was recovered from grave R430 (Table S1). The thickness of these concentrations varies from very thin (2–8mm), through medium (10–15mm) to relatively thick layers (25mm) and microscopic analysis has revealed traces of textile within one of the concentrations. The grave was reopened sometime after the burial and most of the grave goods were found in the fill of the reopening pit. Based on the excavation drawings, some of the organic concentrations seem to have also been located in this secondary pit, next to metal objects, while others were found in situ in the northern, undisturbed part of the grave pit under metal vessels and next to a wooden artefact with copper-alloy nails described in the documentation as a ‘small box’ (although this interpretation is not certain; Schuster Reference Schuster2018: 22–25, 106).
Examination of 23 organic samples using a Keyence VHX 7000 microscope (20–200× magnification) revealed that 13 contained identifiable archaeobotanical remains thanks to their morphological characters.
Three samples also contained textile fragments (no. 6 from grave R430, no. 18 from grave R289 and no. 19 from grave R300). For each, yarn (twist direction and thickness) and fabric structure (weave and thread density per centimetre) were examined under an Olympus SZX7 optical microscope at magnifications of 10 to 30×. After separating the threads into individual fibres, a microscopic image (300× magnification) was taken and compared with photographs of natural raw material fibres published in the Polish Standard (Polish Standard PN-72/P-04604).
Results
Archaeobotanical analysis
The results of the analysis are presented in Table S1. Seeds of cattail (Typha sp.) were identified in all 13 organic samples (the common name ‘seed’ is used here for plant diaspora, including fruit, remnants of gynophore and perianth). They are approximately 1mm long and 0.3mm wide, shiny and range in colour from light yellow to dark brown. Their arrangement is usually chaotic. These seeds belong to the broadleaf or narrowleaf cattail (Typha latifolia/angustifolia), which share the same fruit morphology (Tobolski Reference Tobolski2000: 447–48). Fragments of vegetative tissue, including small pieces of charcoal and wood, were also found in the samples (Figure 4). The soft tissues of the fruit (the fluff) do not appear to have preserved in any macroscopically recognisable form. None of the samples contained material identifiable as deriving from any species of moss.
Organic remains preserved in inhumation graves from Czarnówko: A–D) mineralised seeds of cattail (Typha sp.) from graves R300 (sample 19), R263 (sample 3), R430 (sample 4) and R289 (sample 18); E) organic structure partly resembling cattail remains from grave 395 (sample 23); F) fragment of wood charcoal preserved within the matrix of mineralised cattail seeds from grave R263 (sample 3). Scale bars = 1mm (photograph by A. Mueller-Bieniek).
Analysis of textile remains
Sample no. 6 from grave 430 is identified as poorly preserved remnants of unfelted woollen fabric made of right-twisted (ZZ) yarn in the warp and weft. The thickness of the threads in both the warp and the weft ranges between 0.30 and 0.40mm. Looking at the preserved interlocks, the fabric is either a plain tabby (Figure 5A) or 2/2 twill weave (Figure 5B). According to the classification by J. Maik (Reference Maik1988), which is based on the weave of the fabric, the direction of twist of its yarn in the warp and weft and the presence or absence of felting, this sample would be a type 1 or 7 wool fabric. Microscopic analysis has shown that it was made with core-free wool, a type of wool best suited for spinning (cf. Maik Reference Maik2001: 311, Reference Maik2012: 66, fig. 95).
Fabric weave patterns: left) plain weave; right) twill weave (after Březinová Reference Březinová2007: 144, pl. 7).
Sample no. 18 (from grave R289) is another fragment of an unfelted woollen product, woven in a 2/2 twill weave with right-twisted (ZZ) threads in both the warp and the weft. Yarn with a thickness of approximately 0.40mm was used in the warp and weft. The fabric density in the warp is 16–18 threads per centimetre, while in the weft it is 12 threads per centimetre. These parameters allow the fabric to be classified as type 7 and grade 2 (Figure 5) of woollen fabric according to the classification of J. Kamińska and A. Nahlik (Reference Kamińska and Nahlik1958), in which fabrics are divided based on the density of the product in the warp and weft.
Remains of unfelted woollen fabric were also discovered in grave R300 (sample no. 19, preserved in the imprint left by the foot of a silver cup). These are again the remnants of a product woven in a 2/2 twill weave from ZZ twisted yarn (Figure 5B). The thickness of the warp threads is 0.30–0.40mm (average 0.38mm), while the weft threads range from 0.30 to 0.50mm (average 0.40mm). Again, it was made with a core-free wool. Due to the poor state of preservation, the density of the fabric was not calculated, but it likely represents type 7 textile according to Maik’s classification (Reference Maik1988).
Discussion
Typha in the funeral rites of Central European Barbaricum
Both Typha latifolia and T. angustifolia are common species within the landscape of present-day Poland, occupying similar habitats in wetlands, on the banks of water reservoirs and rivers and in eutrophic, non-acidic waters, though T. angustifolia can also grow in mesotrophic and saline conditions (Zarzycki et al. Reference Zarzycki2002). As they typically form dense rushes, the collection of infructances (the flowering heads or ‘fluff’, see Figure 6) from either species could be achieved easily throughout the whole year but especially during winter when the shallow water around the plants froze over.
Cattail (Typha sp.) seed fluff, historically used as a natural filling material for pillows and bedding (photograph by K. Skóra).
Plant macro-remains of Typha sp. have also been found in sepulchral contexts in four other Roman period burial grounds directly south of the Baltic Sea. A large collection of Typha sp. remains comes from the Wielbark culture cemetery in Grzybnica. In this case, like in Czarnówko, it was not possible to assess whether the remains were those of Typha latifolia or Typha angustifolia. A concentration of macro-remains was recorded in four inhumation graves from subphase B2a and stage B2/C1, dating from the end of the first century till the end of the second century AD (graves 19B, 29, 50, 68). The remains were recorded in various contexts—next to fabric and a brooch (grave 19B), next to a skull with a fragment of wood, next to teeth, and next to a metal ring (grave 50; Polcyn Reference Polcyn, Hahuła and Wołągiewicz2001: 85)—suggesting that the flowering heads were either placed at the bottom of grave pits or log coffins (Polcyn Reference Polcyn, Hahuła and Wołągiewicz2001: 85) or served as pillows (Hahuła & Wołągiewicz Reference Hahuła and Wołągiewicz2001: 58).
In burial mound no. 5 at Osiek, Jarocin County (Greater Poland), which is dated to the beginning of the first century AD (subphase B1a), brooches and metal fittings for drinking horns were discovered at the bottom of the grave, placed on a deposit of decayed remains of Typha sp. (20–30mm thick, 200 × 300mm in size) atop a layer of stones (Pawlak Reference Pawlak, Jaszewska and Michalak2015: 171). This could indicate that the bottom of the stone burial chamber was lined with the green parts of the plant, not just with the infructescences with fluff, serving as a lining on which the deceased was laid to rest along with grave goods (Pawlak Reference Pawlak, Jaszewska and Michalak2015: 176–77).
The use of Typha sp. as burial linings also finds analogies in the discoveries from the cemetery in Lubowidz, where the presence of organic materials, including ‘moss’, may suggest the lining of coffin logs (graves 21, 89, 98, 277 and 280) or the use of headrests (grave 101) (Wołągiewicz Reference Wołągiewicz1995: 15, 21, 33–34). Assessment of the ‘moss’ was carried out macroscopically by an archaeologist during the excavation, so it would be useful to revisit these remains for microscopic botanical analysis, particularly since examination of potential ‘moss pillows’ from Czarnówko (above) and another Wielbark culture burial site in Gronowo (mound 10, grave 2; Machajewski Reference Machajewski2013: 23–24, 64, tab. XI) showed that the remains were of Typha sp.
At Weklice, while it is not possible to identify the specific plant species, the body of the deceased male in grave 64 was suggested to have been placed in a boat woven from reeds (Okulicz Reference Okulicz1992: 141). In a more recent reanalysis (Natuniewicz-Sekuła & Okulicz-Kozaryn Reference Natuniewicz-Sekuła and Okulicz-Kozaryn2011: 38–39, pl. XXVIII), it was concluded that the bottom and side walls of the grave ‘trough’ were covered with a thick and compact lining of the charred leaves and stems of herbaceous plants, possibly reeds. This arrangement could be interpreted as the remains of a mat used to line the grave and then burned on site.
The internal arrangement of burial pits
Information on the presence of organic furnishings in graves from the Roman period in the Barbaricum is limited. Sometimes only indirect references to decorations survive, such as nails used to attach textiles or fur to the wall in the Lübsow/Tunnehult 1 grave (now Lubieszewo; Schuster Reference Schuster2010: 178). An increasing number of mineralised textile fragments found on or inside bronze vessels may suggest cloth coverings or packaging, or even the ritual concealment of grave goods (see Schuster Reference Schuster2010: 270–72). Most archaeobotanical studies relate to wood remains, which often come from vessels, furniture or coffins (for example, Hellmund & Willerding Reference Hellmund, Willerding and Becker2010). Relatively little is known about the lining of burial pits or coffins with plant materials, making the evidence of Typha in some of the graves in Czarnówko all the more significant. The use of cattail in four graves from Czarnówko adds to the identification of Typha sp. remains at Grzybnica (where finds came mostly from the head area of the burials; Polcyn Reference Polcyn, Hahuła and Wołągiewicz2001: 85), Osiek (Pawlak Reference Pawlak, Jaszewska and Michalak2015) and Gronowo (Machajewski Reference Machajewski2013), and to the organic linings used at Lubowidz (Wołągiewicz Reference Wołągiewicz1995) and Weklice (Natuniewicz-Sekuła & Okulicz-Kozaryn Reference Natuniewicz-Sekuła and Okulicz-Kozaryn2011).
A better understanding of organic grave furnishings is found in later periods, for example in the ‘beds’ from the grave in Poprad, Slovakia (Lau et al. Reference Lau2022: 61–75), and in Högom, Sweden (Ramqvist Reference Ramqvist1992: 47–50, pl. 10, 12). The laying of corpses directly in grave pits, chambers or log coffins on mattresses and pillows filled with various organic materials has been confirmed in Central and Northern Europe for various periods, although finds of Typha fluff are rare. Feathers and fluff, bristles, animal hair or thickly felted wool fibres were used as fillings (Malmius Reference Malmius2020: 160). This list can be supplemented with seaweed, based on the discovery of a male buried in Alby on the island of Öland, Sweden, where the head of the deceased was resting on a ‘pillow’ of this aquatic plant measuring 0.5 × 0.3m and 0.1m thick (Helgesson & Königsson Reference Helgesson and Königsson1973: 87, fig. 3).
The remains of a linen pillowcase stuffed with bird feathers were found in a grave from the Merovingian (c. AD 450–750) cemetery in Morken in the Rhineland (Böhner Reference Böhner1959: 23, 25, fig. 15). Inside the burial chamber of a mound excavated at Eberdingen-Hochdorf (Germany) and dating from 530/520 BC, a bronze bed featuring a mattress crafted from densely woven hemp fabric provides a particularly distinctive example of the use of organic material in grave furnishings. The mattress was filled with tufts of badger hair and elements from 20 plant species, including spruce, fir and juniper needles, herbaceous plant stems, blades of grass, moss, beech buds and a substantial volume of oak, birch, hazel and ash twigs, as well as seeds and fruits from trifid bur-marigold (Bidens tripartita), wood avens (Geum urbanum) and others (cf. Banck-Burgess Reference Banck-Burgess, Biel and Keefer2023: 119). Under the head, a pillow was also found, made of plain weave fabric and again filled with plant material and badger hair (Grömer Reference Grömer2010: 55–57, 275; Banck-Burgess Reference Banck-Burgess, Gleba and Mannering2012: 140, 142, 144, fig. 5.7). The choice of materials appears to have been carefully planned; the mattress was meant to be durable, breathable and ‘aromatic’.
Confirmed Typha finds are known from Central European burials at several early medieval cemeteries (c. AD 1000–1250). In an elite grave (no. 43) from Ciepłe (Pomerania), dated to the first half of the eleventh century, Typha sp. remains were found under a bronze bowl and next to the clavicle of the adult male buried in the grave. Either the bottom of the coffin was lined with rushes and sedge or the deceased and the bowl were placed on pillows or a mattress filled with these plants (Badura & Skrzyński Reference Badura, Skrzyński and Wadyl2019: 300, 301, tab. 4.15).
The seeds of Typha latifolia have been identified at numerous Merovingian sites, including Birka (Bj RAÄ 34) and in the boat graves at Vendel XII and Valsgärde 13 (Hansson & Bergström Reference Hansson and Bergström2002: 52). In grave XII at Vendel, Typha seeds were found in three instances in layers of organic material that included birch bark, wood, leather and rushes. It is assumed that these materials are the remains of a bed and a pillow or a mattress cover (Malmius Reference Malmius2020: 96–98, 100, 106, figs. 7.23 & 7.25, tab. 7.2). Similarly, a layer of Typha fluff discovered at the bottom of the Valsgärde 13 grave could derive from mattress filling (Malmius Reference Malmius2020: 106). At Birka, in a grave located under the rampart of a fortified settlement, one seed of Typha sp. was identified together with textiles, which may indicate the use of just the fluff to fill a pillow (Malmius Reference Malmius2020: 129). According to Malmius, these ephemeral remains of pillows or mattresses reveal changes in funeral rites, manifested in the discontinuation of the custom of laying the deceased on bear skins (Malmius Reference Malmius2020: 196).
In larger Merovingian cemeteries (e.g. in Krefeld-Gellep) fillings take the form of straw, moss and feathers, while ‘wool’ made of Typha fluff was found in a grave from the Frankish row cemetery in Niedernberg, Kr. Miltenberg (Pirling Reference Pirling1979: 215, tab. 120; Malmius Reference Malmius2020: 129). Typha is identified in samples from urban settlement deposits (household waste) in early medieval Kraków (Wasylikowa Reference Wasylikowa1978; Mueller-Bieniek Reference Mueller-Bieniek and Mueller-Bieniek2012: 94), where researchers drew attention to the possible use of the stems and fluff as raw materials (Kluk Reference Kluk1788; Pirożnikow & Szymański Reference Pirożnikow and Szymański2004; Mueller-Bieniek Reference Mueller-Bieniek and Mueller-Bieniek2012: 102; Pirożnikow Reference Pirożnikow2014: 28). The culinary significance of the rhizomes is known from several ethnobotanic sources (Pirożnikow & Szymański Reference Pirożnikow and Szymański2004; Pirożnikow Reference Pirożnikow2014: 28).
The many uses of cattails are also detailed in non-archaeological sources. Handbooks from the mid-twentieth century describe the use of fibres obtained from leaves of Typha latifolia and T. angustifolia in producing ropes, thick fabrics (bags), carpets and as filling for mattresses and other upholstery, as well as the process of fibre extraction itself (Dowgielewicz Reference Dowgielewicz1954: 236–38). Typha fluff is suitable, among other things, for fireproofing, filling sleeping bags and for weaving into wool (Fischer Reference Fischer1854: 297; Ostapkowicz et al. Reference Ostapkowicz2001: 82; Dogan et al. Reference Dogan and Peter2008). Described in modern times as ‘poor man’s wool’, Typha fluff and other plants with long hairs on seeds and fruits—bay willow (Salix pentandra), fireweed (Epilobium angustifolium), hare’s tail cottongrass (Eriophorum vaginatum), narrow-leaved cottongrass (Eriophorum angustifolium) and creeping thistle (Cirsium arvense)—can be used for various purposes as a substitute for more expensive raw materials (Mattsson Reference Mattsson and Pettersson2001).
In the material from the Czarnówko cemetery, the seeds of cattail predominate, while the remains of the hair, the fluff, are hardly visible. The hairs improve wind-borne seed propagation (anemochory) and are built from several cells, with a typical length of 10–15mm, though this can reach 25–30mm, and a diameter of about 17 microns. The hairs are very brittle and easily break up into single cells (Dowgielewicz Reference Dowgielewicz1954: 237), which might explain their general absence in the studied samples (Figure 4).
Recent research conducted on the textile production of the Wielbark culture indicates that fabrics similar to those discovered in the graves at Czarnówko were used in Pomerania and northern Greater Poland in the period from around the beginning of the first century AD until the end of the third century AD (Maik Reference Maik2012: 95, Reference Maik and Schuster2018: 154; Antosik & Piotrowska Reference Antosik and Piotrowska2024: 70–71). Outside of Poland, type 7 fabrics have been found in Scandinavia, northern Germany and the northern provinces of the Roman Empire (Maik Reference Maik1988: 35; Bender Jørgensen Reference Bender Jørgensen1992: 126). The textiles used to make pillows or mattresses that were filled with Typha fluff were made of high-quality raw materials, as demonstrated by the dense fabric structure and the use of core-free yarn (cf. Maik Reference Maik2001: 311).
Conclusion
The identification of cattail (Typha sp.) remains and woven woollen fabric within concentrations of organic material found in four graves at the Wielbark culture cemetery in Czarnówko, northern Poland, suggests that at least some individuals were laid to rest with pillows or mattresses. Organic preservation at the cemetery is generally poor, with these rare instances reflecting fortuitous associations with metal artefacts. It is, therefore, not currently possible to determine whether this practice of grave furnishing was proscriptive or prevalent at the site, though examples from other sites suggest that similar practices were widespread in later periods. The graves at Czarnówko provide key insights into a community from the Central European Barbaricum during the Roman period, when written records for this area are lacking. Although it is possible that cattail held a ritual value within funerary contexts, burials may also reflect customs from everyday life; thus, the more mundane use of cattail-stuffed, woollen bedding to provide a comfortable night’s sleep within the Barbaricum might be inferred.
Acknowledgements
We express our sincere gratitude to Prof. Grzegorz Wolski (Faculty of Biology and Environmental Protection, University of Łódź) and the management of the National Museum in Szczecin for making available for re-analysis the organic material identified as ‘moss’ from grave 2 of burial mound 10 in Gronowo, and to the staff from the Museum in Lębork.
Funding statement
This research received no specific grant from any funding agency or from commercial and not-for-profit sectors.
Online supplementary material (OSM)
To view supplementary material for this article, please visit https://doi.org/10.15184/aqy.2026.10419 and select the supplementary materials tab.
Author contributions: CRediT categories
Kalina Skóra: Conceptualization-Lead, Data curation-Equal, Formal analysis-Equal, Methodology-Equal, Writing - original draft-Equal, Writing - review & editing-Equal. Aldona Mueller-Bieniek: Data curation-Equal, Formal analysis-Equal, Investigation-Equal, Methodology-Equal, Writing - original draft-Equal, Writing - review & editing-Equal. Jan Schuster: Conceptualization-Equal, Data curation-Equal, Formal analysis-Equal, Visualization-Equal, Writing - original draft-Equal. Łukasz Antosik: Data curation-Equal, Formal analysis-Supporting, Investigation-Supporting, Methodology-Supporting, Writing - original draft-Equal.
Open access
