Een nieuwe reconstructie van de kleding van de ‘Prinses van Zweeloo’ by Vons-Comis, S.Y., 1988

A new reconstruction of the garments of the

'Princess of Zweeloo'

 S. Y. VONS-COMIS

Translated by Susan Verberg, 2019.

For the PDF "Reconstructing the Wardrobe of the Princess of Zweeloo", click here:
https://www.academia.edu/38845292/Reconstructing_the_Wardrobe_of_the_Princess_of_Zweeloo
  

During the excavation of an early medieval burial ground east of Zweeloo, performed in 1952 by Prof. A.E. van Giffen, an important grave (no. 87) was found. Even in the field people were already convinced of the wealth of the grave. A large fibula and a row of beads were found from a belt. On the basis of these burial gifts it could be established that this was a female grave, dating from the middle of the 5th century. Considering the wealth, it was decided to take the skull and the torso of the 'princess' in order to continue excavation at a later date.

The later 'excavation' took place in 1971 at the Rijksdienst voor Oudheidkundig Bodemonderzoek (ROB) [currently the Rijksdienst voor Cultureel Erfgoed, RCE; Cultural Heritage Agency of the Netherlands] in Amersfoort, under the direction of J. Ypey. It was only at that moment that two strings of beads and two disk brooches of gilded iron were discovered in the coffin (fig. 1). Behind both fibulae and against a number of the bronze rings from the belt were found residues of different fabrics. On the basis of these remains, a reconstruction of the clothing was made by Ypey (fig. 2, after Van Es and Ypey 1977, Abb.13).

In connection with the restructuring of the pre- and protohistorian department of the Drents Museum in Assen, it was decided to have a responsible reconstruction of the clothing of the Princess of Zweeloo. At the beginning of 1987, I was asked to carry out an extensive literature study, possibly accompanied by renewed research into materials and techniques. This research was made possible by a financial contribution from the Drents Praehistorische Vereniging.

Research and questions.

During the study of Van Es and Ypey's publication, the researcher raised a large number of questions, some of these, as Van Es and Ypey rightly pointed to, p. 124, cannot be answered, but also to problems that can be solved thanks to new publications from after 1977.

These questions can be grouped as follows.

1.       Material analysis. To what extent are the material analyzed correctly?

2.       Textile analysis. Are there similarities between the textile residues at the bronze ring a (at the back of the body) and bronze rings b, c and h (front of the body)?

3.       What is the function of the textiles?

4.       Clothing reconstruction. The fabric remains on the back of the fibulae on the shoulders come from a different material (different garment?) than the fabric remains at the level of the waist. Did the woman wear multiple clothing pieces?

5.       Reconstruction of strings of beads. From the published drawing of the location of the beads one could conclude that both strings did not follow behind the neck but were attached to both round fibulae. This is quite common in that period (Bohme 1974, Abb 53). How long and how heavy are these strings of beads? Both the location of the fibulae and the strings of beads are important for the reconstruction of the clothing. 

FIG. 1. Zweeloo: the location of the grave gifts in the tomb of the princess (after Van Esen Ypey 1977, Abb.5, see for legend fig. 3).

FIG. 2. Zweeloo: original reconstruction of the princess's clothing (after Van Es and Epey 1977, Abb 13).

Some of these questions can be answered thanks to new literature. Thus, appeared in 1978 several important publications in this area (Christlein 1978, Hinz 1978 and Vierck 1978). But also, older publications such as those of Wild (1968) on (late) Roman clothing and Hagg (1974 and 1984) on early-medieval clothing should be studied in order to reflect the changes in women's clothing that must have taken place in the 4th-8th century.

In 1986 a publication was published in which archaeological and historical data are combined (Owen-Crocker 1986). The extensive bibliography, which accompanied this publication, seems to offer many opportunities for renewed research. The finds and the field drawings also have to be reviewed.

Results of the textile research as published in 1977

Thanks to careful preparation of the objects located on the upper body of the princess, not only most of the beads, but also the metal grave gifts could be lifted and preserved, carefully considering the presence of textile remains.

The fabric residues at the back of both disk brooches were left in situ and were preserved with the same agent as the metal objects and the oxide layer in which the textile was present. The textiles that have been preserved on / under or against the rings of the beaded belt were probably separate from the bronze objects. Unfortunately, the location of this textiles with respect to the rings could not be traced, since the field drawing, on which Ypey undoubtedly added notes about the original position, has not (yet) been found in his legacy. This essential data impedes the investigation of the reconstruction.

The bronze bracelet, found on the left hip of the woman, was located on textile that was pressed on a hard layer of organic mass, which most closely resembles leather. The textile appeared to have been well preserved over a length of 13 cm in a half arc.

The textile technical data such as Van Es and Ypey published (1977, 111 et al.) have been included for the sake of clarity (Table 1, printed not-bold).

Renewed research 1987

The renewed examination of the thread and fabric remains showed that the technical details are generally well described by Ypey. It is also known from other research that Ypey was particularly interested in textiles and had a good understanding of the research of various materials and techniques (eg Ypey 1962-1963). Because the textile remains in the 1977 publication only form a small part they will be discussed briefly. In this publication, the textiles of the princess will be discussed in more detail and will go deeper into the clothes that the princess may have worn.

Before discussing the questions on a point-by-point basis, first will be given a detailed account of the preserved textiles. In addition, the descriptions of the string of beads will be discussed in more detail, since they play an important role in the reconstruction of the clothing. The numbering of the objects corresponds to that stated in the original publication.

For practical reasons, the original alphabetical order in which the metal objects are discussed with the textile remains is not maintained. For the technical terms, reference is made to the appendix.

Table 1. Overview of the textiles from the grave of the 'Princes' van Zweeloo. Results of the research in 1977 (regular print) and 1987 (bold print).

FIG. 3. Zweeloo (grave gifts B and E up to and including J. B: disk brooch Bb with textile residues TB; E: bronze and bracelet with textile residues TE; F: two bronze keys; G: toiletry set of gilded silver; H: decorated bone fragment; silver ring; J: fragment of beaver tooth, probably attached to I (after Van Es and Ypey 1977, Abb.7, the distance indicated by the textile remains is 1 cm).

Textile finds

1. Left shoulder (disk brooch Bb (inventory number PMD 1952 / III 26-t2) with bronze back, bronze pin and iron spiral, with gilded bronze or gilded silver relief adornment on the front, fig 3: B).

At the back of the fibula is still the remainder of an iron spiral in situ. At the side of this spiral a crumpled fragment of a textile is visible in the oxide layer (We-1, size 0.7 x 0.7 cm). The technical data of this fabric can not be traced.

In a separate box is a lump of rust that after some fitting appears to come from the back of the spiral. In this lump a few textiles are visible besides the bronze pin. This textile has been well preserved.

On top there is a fine diamond twill, dim. 1.5 x 1.2 cm, 18 Z- x 18 Z-twisted threads per cm. A remnant of a card-woven trim can be seen at the side of the rust lump: 5 mm wide, woven with at least nine cards. No direct relationship is visible (anymore) between this card-woven trim and the aforementioned fabric. At the transition site, the needle of the fibula has been inserted so that the textile has made a sharp kink at this location. The fabric has not been preserved at the bend (fig. 4).

According to Ypey there was a small fragment of S-twisted thread behind this fibula. However, while it was missing in 1987, it is possibly identical to the thread that is now in the box of Ba.

2. Right shoulder (disk brooch with iron pin Ba (1952 / III 26-tl))

Only a relatively large rust clump with a part of the bronze bottom has been preserved from this disk brooch. Although neither spiral nor iron needle are visible, we are probably dealing with that part of the fibula where the (iron) spiral was attached (fig. 5).

On top of this rust clump embedded in the rust layer is a relatively large fragment of a fabric that is identical with the fabric at fibula Bb. The dimensions are 2.5 x 2.0 cm. On one side of the fabric fragment is a fragment of a 0.7 cm wide card-woven trim, woven with at least 12 cards.

The threads of this card weave fragment are threaded different through the cards, [text continues after illustrations]

FIG. 4. Zweeloo: disk brooch Bb with textile residue on the back.

1: front (see also fig. 3: B).

2: side view with textile residues at spiral.

dark stripe: linen fabric in [broken] diamond twill (We-1), dark dotted line: linen card-woven trim border (Kawe).

3: rear. Small dots: fabric in [broken] diamond twill (We-1), large dots: card-woven border (Kawe), dashed line: tear in fabric. The warp direction is indicated by an arrow.

FIG. 5. Zweeloo: remnant of disk brooch Ba with textile remains.

1: largest rust clump with spiral, part of needle and textile. Large dots: linen card-woven edge (Ka; e), small dots: linen fabric in [broken] diamond twill (We-1). In the latter fabric fragment, there is a hole in which a part of a underlying fabric with identical structure is visible. Dash line: tear in fabric. The warp direction is indicated with an arrow.

2: reconstruction of side view with the rust clump depicted at 5: 1. At the needle holder there is a piece of oxidized bronze with remnant of an iron needle stuck through part of the card woven edge.

This edge originates from the back of the peplos in relation to the card-woven trim at the spiral. The latter fragment is located at the front of the peplos.

FIG. 6. Card weaving. A warp thread is passed through each hole of a card. These threads are knotted at both ends and then tightly stretched. By tilting the cards a quarter turn each time the warp threads twist together. By passing a weft thread between the warp threads, these twisted warp threads are held together and a card-woven trim is formed. Depending on the manner of entry into the cards, patterns or figures may arise.

resulting in a zigzag pattern during the weaving (fig. 6). The warp threads are Z-twisted, the weft threads are no longer visible. Such card woven trims were mostly used as a set-up edge or as reinforcement of the sides of a fabric. This is a technique that has been used from the Iron Age until the late Middle Ages and is almost always associated with the warp weighted loom (Schlabow 1978).

The fabric in the [broken] diamond twill is damaged in one place. An underlying identical fabric is visible in the hole. This means that the needle of the fibula has been pushed through the fabric once before it passes through the card-woven trim (fig. 5).

At the same spot in the grave a small, slightly rectangular piece of bronze has been found with a remnant of an iron needle sticking out. This needle is inserted through a card-woven trim, a fragment of which has been preserved in the rust layer. In view of the location of the needle with respect to the bronze one might have to do with the holder of the needle. The bronze cannot come from the fibula itself because of its shape. The card-woven trim is only 6 mm wide here, the number of cards is at least ten pieces. The orientation of the warp threads of this card-woven trim and thus also of the accompanying cloth is at right angles to the bronze. If the needle is inserted through the card-woven trim in the same way as the other piece, then this fabric could be the edge of the other fabric (Fig. 4b).

3. Torso ('gleicharmige Kerbschnittfibel', gilt silver with bronze needle and iron spiral axis A (1952 / III 26b)).

According to Van Es and Ypey (1977, 111), a fabric fragment was present during the original drawing, made during the excavation in 1952. During the study for the 1977 publication this fabric appears to have disappeared. However, a part of an S-twisted linen thread consisting of two Z-twisted single threads would still be there. This thread was no longer in situ in 1987.

According to Van Es and Ypey in 1971, there were thread residues and a fragment of a twill weave with 16 Z-x 14 S-twisted threads per cm where the fibula was located in the coffin.

In 1987, in a box with the inventory number of this fibula, some linen thread remnants were found and fragments of woolen fabric. Both linen thread fragments are Z-twisted but the thickness is different (Ga-1 and 2). The thread thickness of Ga-1 varies from 0.6-1.0 mm. The other thread fragment is only 0.3 mm thick and 0.3 cm long. Both threads show no undulations, in other words they are not threads from a linen fabric.

Of the woolen fabric, six fragments have been preserved, ranging in size from 0.3 x 0.5 cm to 0.8 x 0.4 cm. The textiles are woven into 2/2 twill or a twill that has been derived from this (broken diamond twill?). The fragments are warped a little, so that the number of threads per cm varies greatly from 15 to 16 threads per cm in the warp direction and 13 to 14 per cm for the weft direction.

 

4. Waist (string of beads around waist D).

At the bronze ring a (1952 / III 26th), four fragments of a fabric in [broken] diamond twill were preserved below the body (fig. 7: left). The fabric had remained intact thanks to the preservative effect of the copper oxide of the ring. In view of the twist and the number of threads per cm, this [broken] diamond twill corresponds to fabric We-2, which is found at the equal arm brooch (15 Z-x 15 S-twisted threads per cm).

The shape and orientation of the fragments indicate that part of the bronze ring has come to lie 'free' after the adjustment. About a quarter of the flat side was covered with textile. Whether this textile was originally located under or above the ring is unknown. The dimensions of the semicircular fabric remnants vary from 0.5 x 1.2 cm to 2.0 x 0.5 cm. The remaining textile residues found at this ring cannot be analyzed as clearly. This includes three adjacent S-twisted woolen threads (Ga-3, fig. 7: right); length 1 cm, thickness 1.5-2.0 mm. From another fragment, dim. 1.0 x 0.5 cm, it is not possible to find out whether it is part of coarse woolen fabric or whether it is a bundle of twisted threads (We? -3). At the adjacent bronze rings b and c (1952 / III 26f and g) two fragments of the wool fabric We-2 were found. The fragments, 1.0 x 1.0 and 1.0 x 0.5 cm, are too small to find out whether they are woven in 2/2 twill or the derived [broken] diamond twill. The number of threads is about 15 x about 15 per cm.

The warp and weft threads are resp. Z- and S-twisted. These rings are found on the left side of the abdomen. Originally, the rings were seated [text continues after illustrations]

FIG. 7. Zweeloo: textile remains found at the bronze ring a. Left: fragment of woolen fabric in [broken] diamond twill (We-2), right: S-twisted woolen threads (Ga-3).

FIG. 8. Zweeloo: fragment of coarse woolen twill weave (We-5), found at bronze rings f and g.

[Originally, the rings were seated] together. According to the text on a note accompanying the finds, there were also leather residues found here. Given the structure, this is not leather but wood bark remnants, possibly coming from the coffin. On a piece of wood, one can see wood bark with textile on it. On top of this textile is copper oxide from a bronze ring. This sequence of ring-oxide-textile-wood bark indicates that the textile was located on top of the string of beads.

According to Van Es and Ypey, remnants of S-twined threads were also present at these rings. In 1987 only pieces of a relatively thick Z-twisted thread are present (Ga-4). The woolen thread is too thick, about 1.1 mm, to come from the fabric. At ring d (1952 / III 26h), located on the left hip, two fragments of a wool fabric (?) We-4 were found. The dimensions are 1.3 x 1.0 and 0.5 x 0.5 cm. Through a mineralization process, the weave (?) is hardened in such a way that study is nearly impossible. In both directions Z-spun threads are visible that vary in thickness from 0.8-1.0 mm. The fabric density cannot be traced. Given the thread thickness, the fabric density in both directions would be about ten threads per cm. In any case, one has to deal with a coarser weave (?) than the previously discussed fabrics.

The textile fragment found on the belly with the rings (1952 / III 26j and k) is also derived from a coarsely woolen weave (We-5). This is a fragment, 0.5 x 1.3 cm in size, woven in 2/2 twill or the [broken] diamond twill derived from it, as analyzed by Ypey (fig. 8). Because the fragment has fallen apart in the course of time, the original weave can no longer be determined. In this fragment, for example, there are still about eight loose Z-twisted

FIG. 9. Zweeloo: textile fragments found at bronze ring i. Left: fragment of woolen fabric (We-2), right: part of bundle S-twisted woolen threads (Ga-3).

threads, which, given the dimensions, must come from this fabric fragment. The number of Z-twisted warp threads varies from nine to ten per cm whereas in the weft direction only six Z-twisted threads per cm are present. Microscopic examination has shown that these are thin and thick wool fibers, with or without medullary canal. At ring h, a disk-shaped, bronze disk with a central piercing with at the back (?) side three eyes (1952 / III 261), found on the right side of the abdomen, an intact fabric fragment was found. It is woolen fabric We-2, in [broken] diamond twill with 16/17 Z-twisted warp threads per cm and 15/17 S-twisted weft threads per cm. The fabric corresponds to the fabrics found at rings a, b and c.

In the area of ??the bronze ring i (1952 / III 26m), near the right hip, five fragments of the woolen fabric were found in [broken] diamond twill we-2. The technical data can be retrieved from only two fragments: approx. 15 Z- x 14/17 S-twisted wires per cm (fig. 9: left).

Furthermore, there are four bundles of possible S-twisted threads (fig. 9: right). The thickness of the bundles is 0.3-0.5 cm, the thickness of the individual threads is 0.5-1.0 / l, 4 mm. The length varies from 1.0-1.5 cm. Just like the bundles of yarn at ring a, according to microscopic research here they are wool fibers as well. Some dark brown fragments indicate, according to Van Es and Ypey, the presence of leather.

5. Left hip (bronze bracelet E (1952 / III 26c), fig 3: E).

The bracelet, which has a diameter of approx. 8 cm, has left deep impressions

FIG. 10. Zweeloo: textile fragments found under bronze bracelet E. The remains of wool fabric We-2 are embedded in an organic mass, most likely leather. The impression of the bracelet is clearly visible. The original location of the fragments is unknown.

FIG. 11. Zweeloo: largest fabric fragment located under bracelet E (see Fig. 10).

[The bracelet, which has a diameter of approx. 8 cm, has left deep impressions] in a solid organic mass (leather?). Seven to eight fragments of textile have been preserved between the oxidized bracelet and this hard mass (fig. 10). Here too we are dealing with We-2, the woolen fabric in diamond twill with 15/16 Z-x 15 S-twisted wires per cm (fig. 11).

An excavation photo from 1952 shows that the bracelet is uncovered on top (excavated for the photo?). In other words, the textile that is still present, was originally under the bracelet and on top of the leather (?). Moreover, this bracelet with matching textiles is located on the left hip and not above the waist as shown in the reconstruction of Ypey.

6. String of beads (three strings of beads Ca, Cb and D (1952 / III 26h, 26v and 26w)).

In the grave of the princess three strings were found: two at the height of the chest (Ca and Cb) and one at the belly at the level of the waist (D). Necklace Ca consists of 101 amber beads (see for description Van Es and Ypey 1977, 113 ff.). The investigation in 1987 showed that the string is 90 cm long. The weight is 139 grams. This includes the unknown weight of a modern twisted metal wire on which the beads are now laced.

The glass beads of string Cb were in very poor condition during the excavation. Originally, at least 130 beads were present, of which 110 could be recovered. In 1987 only 107 beads were intact. The length of this incomplete string is 60 cm. The necklace now weighs approx. 35 grams including the thread. Originally, the string would have been about 66 cm long and about 41 grams heavy.

The string of beads D, which was found at the height of the waist, consists of 30 large glass beads, 1 amber bead, 1 bronze disk and 10 bronze rings. The length of this cord is 55 cm. The weight of the beads, including nylon threaded thread, is 542 grams. The weight of the bronze disk and rings treated with PVC is 92 grams. The original weight of the entire cord must have been between 542 and 634 grams.

Summary of the results of the technical research 1987 (questions 1, 2 and 5).

If one now lists all available data, one concludes that all material analyzes as carried out by Ypey are correct (question 1). With renewed microscopic examination, linen has again been detected in the fabric behind the disk brooch. The knots and stripes that are characteristic of flax fibers, of which linen is made, are clearly visible under a polarization microscope.

Such fine fabrics in [broken] diamond twill are known mainly from wool fabrics, among others, excavated in the ‘terpen’ areas in the northern Netherlands (Schlabow 1974) and northern Germany (Hundt 1981).

Linen fabrics in plain weave are known from Germany. A linen fabric in 2/2 twill has been excavated in Sievern. However, this discovery dates back to the 8th century (Hundt 1980). In Munningen a comparable linen fabric in diamond twill with 13 Z-x 13/14 Z-twisted wires per cm was found, but this also dates from a later period (Hundt 1976, 77).

Only with coarse wool fabric, We-5, found with rings f and g, was it possible to study intact wool fibers under the microscope. These are thin and thick wool fibers, with or without medulla [core].

The determination of the other organic components is generally correct.

Only with the rings b and c is probably not leather but wood bark preserved. The organic remains, located under the bronze bracelet, come from leather (bag, pouch?).

The fabric analyzes as described by Van Es and Ypey are admittedly good but not complete. Incidentally, data on dimensions and material type are lacking in the description. Furthermore, Ypey did not recognize the card-woven trims behind the two disk brooches. At first glance it seems to be a fabric in diamond twill that was extremely distorted at that location, giving the same effect. But after a long study it turned out that it must be a card-woven trim. In one place, the transition between the normal fabric and the card-woven trim attached to it is visible. The occurrence of such a card-woven edge at the back of the disk brooches indicates a certain type of garment (see further on reconstruction clothing).

Furthermore, the descriptions of the location of the fabric remains with respect to the metal objects are not complete. This is understandable in a publication all about the finds from a grave, but for a responsible reconstruction of the clothing, it is necessary to mention all details. The number of threads remains is slightly different from that published by Van Es and Ypey. This may be related to the fact that the originally plied threads disintegrated in separate single ply threads after the investigation in 1977 or completely pulverized.

In fabric research, one has to deal with the problem that it is not always clear which fragments come from which fabric. For example, Ypey mentions that he doubted whether the fabric at the equal arm brooch matches the fabrics found behind the disk brooches. The microscopic examination has clearly shown that these are two different weaves.

On the other hand, although the number of threads per cm varies greatly with the fine wool fabric fragments in [broken] diamond twill, found around the waist, these fabric remains come from the same fabric as found at the height of the equal arm brooch and the bracelet. Since both spinning and weaving are done by hand, variations in thread thickness can affect the number of threads per cm. Moreover, the fragments are so small that one cannot see at what stage of textile use the fragments are hardened: if the fabric had already been stretched by use, the number of threads per cm will be smaller than in a part of the textile that has not been under tension.

The extent to which the fragments of the fabrics We-4 and 5 originate from the same source is difficult to say. It is true that the twist of the threads of both fabrics corresponds, but the weft (?) of the threads of We-4 appear to be thinner than those of We-5.

In summary, it can be said that at least three and possibly five different fabrics and four or five yarns (bundles) were found in the grave (Table 2).

Table 2. Overview of all fabrics and yarns from the grave of the 'princess' of Zweeloo.

The technical data can be compared with those of other finds from the Netherlands, for example. In the Netherlands, 5th-century textile finds are known from cemeteries in Maastricht (verbally from J. Ypey, see, among others, Ypey 1955, textiles on figs. 90 and 91), Wageningen (material by the author) and Rhenen. It has been found in sub-publications that a twill weave, according to Ypey dated circa 400, was found in this latter location at the top of a belt fitting in grave 839 (Ypey 1969, 98 and Abb.5). A 2/2 twill weave was found in the male grave 775. Both in the warp and in the weft direction were about 10 threads per cm. The warp (?) threads are Z-twisted whereas the weft (?) threads are predominantly S-twisted, with a Z-twisted thread at irregular distance. This grave dates from the first three quarters of the 6th century (Ypey 1983, 470-471).

Late Roman woolen fabrics from Deurne are too small in number for a good comparison. This concerns fragments of a trim and a fabric in herringbone (Leene 1973).

Other textile finds from the Merovingian period can be found in the unfinished collection of finds from the burial ground of Alphen (N. -Br.; Ypey 1955, in fig. 100 the remains of textile is visible). Later textile finds are known from Leersum (7th-8th century, Ypey 1965-1966), Huinerveld (Putten, Ypey 1962-1963) and Aalden (7th-8th century, Schlabow 1974).

Textile finds from settlements are mainly known from terp abvations in the north of the Netherlands (for an overview see Schlabow 1974). The problem with these findings, however, is that they are not or very roughly dated, for example: Westeremden 500-1000 after Chr., or Ezinge 200 BC. - 300 AD.

The only textile finds that may be useful for the reconstruction are some fragments of a large woolen garment, wrapped around a bog body that was found in 1897 near Yde (Gem Vries). This is probably a cloak worn by a woman. This find is dated between 200-500 AD. (Van Zeist 1956, 206-207).

Five fragments of the original rectangular mantle have been preserved, measuring approximately 130 x 150 cm (Schlabow 1974, 208-209). The fabric is characterized by a typical set-up edge. The fabric is woven in 2/2

twill with approx. 10 Z- x 9 Z-twisted threads per cm. Furthermore, a number of transverse bands have been woven into this mantle. For this dark brown weft thread was used. In addition, not one but two threads were inserted between the warps at each turn, so that a unusual effect was obtained. Although there is no indication of a fringe, Schlabow still has the idea that, given the other findings from northern Germany, it should have been present. The woolen band, made in sprang-technique, which is around the edge of the woman's neck, is currently being studied. In 1985, Mrs. A. Nienhuis made a replica of the mantle, which is now in the Drents Museum in Assen.

Women's clothing through the centuries

Before being able to reconstruct the clothes of the princess, it is necessary to give a short overview of clothing from Roman times up to and including the early Middle Ages. As mentioned in the introduction, relatively much is known about women's clothing as worn in Roman times and the Viking era. From the intervening period, in which the clothes of the princess dated however, little is known.

Roman time: 1st-4th century

Although a relatively large number of fabrics from Roman times have been excavated in NW Europe, only a few recognizable parts of clothing are known (for an overview, see Wild 1970, 1977). Although some garments of bog bodies are known to the north of the Limes, this mainly concerns men's clothing and the dating is not always clear (Schlabow 1976).

Until now the only women's clothing was found in Yde (Netherlands, Schlabow 1974, 208-209: cloak), Damendorf (Germany, Schlabow 1976, 95-96: two skirts), Huldremose (Denmark, Hald 1980, 359 etc.: skirt and peplos), Denmark (without location or date, Hald 1980, 359, 361: peplos) and within the Roman Empire at Les Martres-de-Veyre (Audollent 1923: shirt and stockings).

For further research into women's clothing one is dependent on sculptures, murals, mosaics and funerary monuments. In addition, there are historical sources, but it has been found that the names used herein cannot or hardly be linked to the garments (Wild 1968, 1985).

As long as we have little or no access to the garments themselves, we have to work with the aforementioned sources, which can give a false picture of the daily clothing, since one does not know whether the clothing matches the clothing that was in fashion at the time or because it concerns, for example, special clothing for goddesses or 'matrones'. Furthermore, for example, murals and mosaics can only be found in rich villas.

Does one start from the idea that fashionable clothing, as worn by the higher classes, was over time imitated by the lower positions (indicating that it usually concerns imitations performed in materials and techniques of lesser quality) then it would be possible to make statements about civilian clothing.

A second problem has to do with the influence of the Romans on the clothing of the vanquished peoples or peoples with whom the Romans were at war. In most cultures, clothing is the external characteristic of a community or people. In this way the members of a community indicate that they belong together because of family ties or faith (visible in regional costumes in the 18th and 19th centuries in the Netherlands). The position of the person within the community can also be determined on the basis of clothing (gender, age, marital status, status, etc.). Moreover, pre-Roman clothing from pre-Roman times, both in terms of manufacture and use, must be simple and practical for work at home and on land. The garments were woven in one piece into the desired model so that no material was lost. With the arrival of the Romans, the local population got a confusing picture of military clothing on the one hand and costly clothing of dignitaries who visited the area or settled there on the other.

Depending on traditions and solidarity within a local community and the relationship with the rulers, parts of clothing may or may not be assimilated from the ruler, either for practical reasons or for reasons of status.

For example, those who want to have status with the new rulers will relatively quickly wear the new clothes in order to distance themselves from the barbaric clothing of the indigenous people. Especially pants were in the eyes of the Romans an inferior piece of clothing.

The extent to which this can be traced on the basis of grave finds is difficult to say, as long as one does not know whether the dead were buried in special clothes that differed from daily clothing. For example, Roman glass, earthenware and metal clothing accessories such as buckles and fibulae are found in graves in Scandinavia, where it is not clear whether the found clothing accessories, such as gilded bronze fibulae, had been brought home / traded or robbed in order to increase the status of the deceased. Whether the function of the objects still corresponds to the original function, namely the keeping together of (Roman) garments, is difficult or impossible to trace. If we turn back to Roman clothing, it looks as though we can only use images on funerary monuments for the examination of civilian garments, because they probably give the best picture of the clothes that were in fashion at that time. (Wild 1968, 167).

It appears that there were similarities in the clothing of men and women: both wore a loose tunic with deep folds, with or without long sleeves, sometimes with fringes at the bottom (Wild 1968, 168). In the men this tunic reached below the knee, while that of the women covered the legs to the ankles. This corresponds to the finding of a short tunic of a man excavated in the peat of Reepsholt (Germany, date: 1st or 2nd century, Schlabow 1976, 73). According to Wild, this garment could have been imported from the Roman Empire (1968, 169). A female tunic from a tomb in Les Martres-de-Veyre (France) has a length of 1.25 and probably reached the ankles (Audollent 1923). Over this tunic only a so-called Gallic cape would be worn by the men, with or without a sewn-on hood. The women, on the other hand, probably only wore a cloak or scarf over their shoulders. The barbarians dressed in a tunic with long sleeves, cloak and trousers. Moreover, there are indications that a shoulder cape and leg coverings were worn in the countryside.

The woman's basic clothing also includes a cloak that was worn over the tunic. This consists of a long, rectangular piece of cloth, partially, or not, pulled over the head. This cloak can be worn in different ways (Wild 1968, 197-198). Occasionally an undergarment is visible at the neck and at the ankles. Wild calls it striking that the woman from Les Matres-de-Veyre did not wear any underwear. This undergarment could, however, be made of linen, a raw material that did not preserve under local conditions, which, incidentally, were beneficial for the preservation of animal wool fibers. The woman also wore stockings, made of a woolen fabric.

Wild's research has shown that 'fashion' in women's clothing hardly changed during the Roman period, although minor differences occurred. Only the hair dress is fashion-sensitive. After studying many images, Wild comes to a regional layout of clothing for women. For example, he distinguishes the so-called Menimane costume, which was worn in the Mainz area. This costume consists of the following parts: an undergarment with long sleeves, an overclothe on the shoulders held together with fibulae (= peplos), a cloak, scarf, hairnet and a hat (Wild 1985, 393-401). Characteristic of the Menimane costume is that the overgarment with a third, smaller fibula on the chest is attached to the undergarment. In a grave at shoulder height two large and in the middle of the chest a small fibula, one could be dealing with such a combination of garments.

The overgarment is of the so-called peplos-type, a garment that was worn before this era in Greece and Central Europe (Hagg 1967-1968). This garment is tubular. The front and back are held together on top of the shoulders by two mostly identical fibulae. Such clothing items are also known from Roman times. In the first centuries after Chr. the peplos was worn both inside and outside the Roman Empire. The best-known example is the tubular garment with a circumference of 264 cm, which was found on the body of a woman in Huldremose (Denmark, Hald 1980, 359 etc.). The great length of this garment, 168 cm, suggests that this garment can be worn in different ways. For example, the upper part can be draped over a tightly pulled belt at the level of the waist. Usually this garment of Huldremose is depicted with an edge turned outwards at the top, because the fabric (now?) is too thick or not flexible enough to hang over a belt (Hald 1980, Fig. 443a and b). Both methods of 'draping' allow a variable length and size: practical during pregnancies. A second, identical item of clothing comes from Denmark. The length of the peplos is only 137 cm and the diameter is smaller than that of the dress from Huldremose (only 240 cm). Unfortunately, its location, discovery conditions and dating are unknown (Hald 1980, 361).

The cloak, which belongs to the so-called Menimane costume, is probably rectangular, approx. 2 m long and approx. 1.40 m wide. This garment can be draped in various ways and, among other things (like Menimane), can be held together with a fibula on the right shoulder. According to Wild there are indications that this costume contains relics of local clothing from the late Iron Age. Until the 2nd century AD images of these garments are known mainly from rural areas (1985, 399).

The discussion of the clothes of the Ubi in the Rhineland is of less importance in this respect. A third regional costume, lastly, is inspired by the clothes of the goddess Nehalennia in the mouth of the Schelde. Characteristic of this costume is the shoulder cape, which, given the smooth appearance on the altars, could be made of leather (Wild 1985, 404-405). There is little we can say about the other garments from Nehalennia.

Early Middle Ages: 5th-6th century

After the fall of the Roman Empire, the problem of clothing reconstruction becomes greater. The fashions, as worn by the Romans, disappears. The question is whether the indigenous population continues to be dressed in clothing inspired by Roman fashion, or whether one returns to farm clothing from before Roman times. Furthermore, the number of finds from the Netherlands is too small to make clear statements about women's clothing in the 5th century. The Northern Netherlands is in the 5th century in a border area where on the one hand influences from the south (Franconia), and on the other hand influences from North-German country (Saxony) and Scandinavia play a role. From this period in North Germany, England and Scandinavia a lot of inhumed graves with numerous grave gifts are known, on the basis of which statements can be made about the function of the fibulae and clothing. With these burial finds, and the same applies to the grave of the princess of Zweeloo, one must also consider the fact that only the upper class could afford the luxury of metal (ornamental) objects, amber and weapons. Since the preservation of textile in graves is linked to the appearance of metal, one only gets an idea of the clothing of the elite, which was undoubtedly of better quality.

The finds from the tomb of the princess show that she must also have belonged to the elite, so that this context can be compared with that from other sites. However, due to the lack of sufficient comparative material, it is difficult to say whether the princess is buried in her daily summer or winter clothing or in special funerary clothing. Nor can one determine on the basis of the grave finds the age and its correct status within the community. One can only say that she is buried with precious jewelry. But: were young, or old women allowed to wear jewelry? Did the woman wear inherited family jewelry? Was she (just) married? Questions that need to be considered before any statements can be made about its social status (Gebühr 1976, 47 th., Vierck 1978d).

Unfortunately, there is no detailed publication about grave inventories in the Netherlands in which an overview is given of all burial gifts. The extensive exhibition catalog 'Sachsen und Angelsachsen' (1978) describes many inventories of rich and less rich graves from the 5th century, but no mention is made of textile finds for this period. 

Germany

Up to now, only a grave from North Germany is known whose inventory is almost identical to that of the tomb of the princess. This is a 5th-century grave excavated in 1948 in Bad Lippspringe, Kr. Paderborn (Lange 1959). The original drawings of the finds show many similarities with the objects from the tomb of the princess (see also page 64).

Thanks to extensive research of cemeteries we know much about grave inventories of the Alamen, living in southern Germany and Switzerland. In the 4th century, a typical combination was found in the women's graves: a pair of fibulae, hair pins, bracelets, rings, buckles of belts (Christlein 1978, Hundt 1976). Many of these objects indicate traditions in the Germanic world that are based on Roman examples. In the course of the 4th and 5th century the two disk brooches on the shoulders go out of fashion. These fibulae are replaced by others with a curved shape. There are also more strings, consisting of beads of glass and amber. Especially these bead strings are un-Roman.

It is unclear which data Christlein (1978, 78, Abb. 52) bases his reconstruction on the clothing of a woman from the second half of the 4th century. According to this reconstruction, she is dressed in a long-sleeved overcoat, the skirt only reaching below the knee. This reconstruction shows that the function of the shoulder fibulae has changed: the fibulae no longer serves to hold parts of a garment together but as an attachment for a leather thong off which hangs an amulet.

In the second half of the 5th century, the number of metal (ornamental) objects in the women's graves increases. The large shoulder fibulae are now found at hip height, under which a leather belt hangs with various beads and pierced coins. In the neck area there are two small, asymmetrically placed fibulae. According to the reconstruction, these fibulae would hardly have had a function. The woman wore an ornamental pin in her hair, earrings, necklace and bracelet, all made of silver. Furthermore, a golden ring and beads of glass and amber at the left forearm. On the right side of the body, a leather belt with a knife hangs from the waist. This costume would have come from the East or from the North (Christlein 1978, 80, Abb 54).

Belgium and France

Little is known from Belgium about textiles of the 5th century. On the other hand, in graves from the 6th and 7th centuries textiles have been found (among others Alenus 1963, textile residues on PI, IV, Ven VI).

The above also applies to France. From the early period, the aforementioned garments from Martres-de-Veyre are known (Audollent 1923). It is true that there are also rich female graves in this country, such as the grave from Fecamp, dated to around 400 AD., but here next to the many fibulae and other metal closures, no textile remains have been preserved (Werner 1962). The only female grave in France with well-documented textile finds dates back to the 6th century when fashion had changed again. It concerns the grave of probably Aregonde, wife of King Clotair I, who ruled from 558 to 561. She was dressed in exclusive robes, including a fine wool shirt that reached to the knee, a silk dress and a long cloak that had been finished with trim incorporating precious metal wire (France-Lanord 1979).

England

In England, in many female graves from the early Middle Ages two or three fibulae are found, which can also be associated with a peplos robe. Buckles and various attributes such as keys indicate a belt worn around the waist or on the hips. The extent to which the different fibula types per region are also related to differences in clothing is difficult to determine. Owen-Crocker assumes that a certain type of fibula originates from the mainland. Whether the clothing in question is also imported is unknown (Owen-Crocker 1986, 28).

During the winter months, the women in England, as in other parts of NW Europe, will have their arms covered. This can be done by turning a cloth or by pulling on a shirt with long sleeves under the peplos. This shirt would be closed at the neck opening with a third fibula. In both men's and women's graves occasionally, metal sleeve closures are found at the wrists, which probably belong to this type of shirt. These closures occur mainly in England and in Western Norway (Owen-Crocker 1986, 39-40).

Large fibulae on the chest would have served as a closure for a cloak, cape or shawl. This would correspond with textile residues at the top of buckles. In the latter case, however, one must also consider the fact that the dead person is buried in or under a cloth.

Scandinavia

In Scandinavia, much is known about the textiles found in graves from the 4th and 5th century AD. Research has shown that all preserved textiles are made of wool. It is known that linen threads and fabrics were imported later in the Middle Ages. Most textiles are woven into four-string twill with predominantly Z-twisted warp and weft threads. Fabric assembled with 2/2 twill is very rare, as are fabrics in plain weave (Bender Jørgensen 1986, 59-69). Focused research into the function of this textiles has not (yet) taken place.

In Denmark much attention has been paid to the jewelry, such as the bead strings (Hinz 1978). Characteristic of these strings is that they are always found on the chest, hanging from two fibulae on or in front of the shoulders. In Sweden there are no such strings, but they are known from later periods in Finland (Koch 1974).

From the detailed publication about the inventories of two Danish women's graves from the 7th century it appears that people there after 600 AD. wore multi-row beaded strings, attached to two hangers for the shoulders. It is interesting to note that in a grave at the height of the chest, apart from a bronze box with sewing and embroidery thread, remnants of a leather bag were found in which there was a toiletry set similar to that of the princess of Zweeloo (Becker 1953, 135)

For the area between Elbe and Loire, the fibulae belong to normal women's clothing, since 65% of the 187 women's graves with fibulae from the 4th and 5th centuries, one to two fibulae and 19% three fibulae are found.

In the 4th century, identical fibulae types are found throughout the coastal area. Small regional differences appear to occur in the next century, noting that very little is known from Belgium and northern France (Bohme 1974, 159-160). Bohme (1974), Vierck (1978b) and Owen-Crocker (1986) all agree on the basis of the archaeological data that in the area between Elbe and Loire and parts of England the women in the 5th century used to dress more or less the same. In the women's graves from the 5th century in these areas one finds a pair of disk brooches at the height of the shoulders, to which one or more bead strings are attached. In addition to bead necklaces, silver and gold pendants are also found. Beaded strings that are located at the waist are very rare. In several graves numerous beads, more or less randomly, are found in a broad band at the waist in the 6th century and later on. Schulze (1976, 157) concluded that these beads were originally sewn onto a belt band. Long hairpins would have served for attachment of hat, cloth or veil to the head. East of the Rhine, the women appear to have worn a leather belt with small objects.

In the course of the 6th and 7th centuries clear changes begin to show in clothing, both in North and South Germany and in England and Scandinavia (Christlein 1978, Zeller 1974, Vierck 1978c, Hagg 1974 and 1984). However, these developments will remain unspoken here.

Reconstruction of clothing on the basis of Zweeloo's finds and clothing- (reconstructions) from abroad (questions 3 and 4).

There is little information available for the reconstruction of women's clothing in the 5th century. Many researchers local and abroad have made grateful use of the reconstruction of Ypey after 1977. In almost all recent publications one comes across the drawing in a more or less modified form.

Thus, the drawing appeared not only in educational publications (Den Otter et al., 88 etc.) But also, in Past Land (Bloemers, Louwe Kooijmans and Sarfatij 1981, 128) and a cut-out sheet from the National Museum of Antiquities in Leiden (Peddemors z.j.). In the German literature the drawing has been published or mentioned in resp. Vierck (1978a, Abb.3) and Hinz (1978, 352, with critical comments on the position of the strings in note 19). The publication of Owen-Crocker also pictures a reconstruction, which is very similar to the drawing of the Princess of Zweeloo (1986, Fig. 30). With all these reconstructions, one always has to ask whether these are based on the recovered textile residues or whether one has only assumed the location of the metal objects in a grave. For example, certain other reconstruction drawings are based solely on the location of metal objects, such as, for example, in a Danish female grave that can be dated to around 400 AD. (Brinch Madsen 1975). In this grave 2 fibulae, 2 pins and 3 bead strings with a total of 170 beads were found. According to the researcher, two strings of beads would be attached to the two pins on the chest and not to the fibulae (fig. 12).

FIG. 12. Reconstruction of the clothes of the woman from Sønderholm (Denmark). Dating about 400 AD. (after Brinch Madsen 1975, fig. 7).

Unfortunately, the weight of the strings is not published, but given the weight of the two comparable string of the Princess of Zweeloo, this reconstruction seems unrealistic. The pins would be pulled down by the weight.

Another reconstruction was made on the basis of finds from earlier on p. 60 mentioned women's grave from Bad Lippspringe (Lange 1959). In the following list, (*) indicates which objects were found in both Bad Lippspringe and in Zweeloo.

In the German grave were found: 1 denarius, 1 brace fibula, 2 disk brooches *, 7 glass beads *, 1 pierced canine of a bear *, 1 bronze pin, remnants of a bone comb, a part of an iron knife, 2 almost identical keys *, bronze fittings, 1 oval iron ring, possibly from a belt, 9 bronze rings * and a scale *. It can be noted that the bone comb differs from the triangular type that was in fashion at that time. This would mean that the bone fragment, found on the right breast of the princess, could also come from a comb. Although according to Lange (1959) textile finds are missing and the location of all objects is unknown, Polenz (1985, Bild 36) later made a reconstruction that is clearly based on the reconstruction of the princess of Zweeloo (fig. 13).

Up to now, only a grave from the 5th century has sufficient known textile data to determine the possible function of the fabric remnants. The researchers, however, wisely did not make a reconstruction drawing. This is a rich male grave from Evebø / Eide in Norway (Magnus 1983, Raknes Peddersen 1983) dating from the second half of the 5th century. The deceased man was laid in the grave on a bear skin and covered with a woolen blanket. The clothing would consist of the following parts: a coat, two tunics and a pair of trousers (see table 3 for the technical data).

 

Table 3. Overview of the 5th-century textile finds from a male grave at Evebø / Eide (Norway), based on data from Raknes Peddersen 1983.

Before a new reconstruction drawing of the clothes of the Princess of Zweeloo can be made, the location of the textile remains in the grave first have to be studied. Through this research the function of the various fabrics and threads can be traced (= question 3). The fabrics and threads from table 2 will be discussed in turn.

We-1 linen fabric in [broken] diamond twill (18 Z-x 18 Z-twisted threads per cm) with card-woven fabric border.

Peplos, fibulae and strings of beads

Fragments of this weave were only found behind both disk brooches, at the level of the collar bone (?). If one looks at the form and the pleating of these weaves (fig. 5) it appears that the fibulae had a practical function. The needles are inserted through the two opposrwing card-woven edges, i.e. this is a peplos-like garment, the edges of which are held together on the shoulders. Although no remains of this fabric have been found at the other metal objects in the grave, one surely deals with a long garment here: a shirt with, for example, sleeves would have been stitched at the top. Moreover, there are no indications for sleeves: no metal closures were found at the wrist, for example at the grave from Evebø, Norway (Magnus 1983) and in graves in England (Owen-Crocker

1986, 39-43). It can be noted here that a possible sleeve closure may also have been made of perishable material, such as leather or bone.

Both iconographic sources and archaeological research have shown that the peplos may have been very long. In order not to stumble over this robe, a part was draped over a belt. Van Es and Ypey (1977, Abb.13) have the garment draped as well in their reconstruction drawing over an extra band [belt], hidden behind the beautiful cord of beads around the waist. This extra band would also be used for the suspension of the keys and any purse or bag. Given the weight of the beads, about 600 grams, this does not seem very likely. The cord would have weighted heavily on the draped part. Especially with crease-sensitive, linen fabrics this gives a particularly ugly result. The above arguments point to a smooth dress that reaches right down to the footwear.

As far as the bead strings are concerned, it should be noted that when carefully examining the drawing with the location of the artifacts it is immediately noticed that the strings on the chest do not continue behind the neck as suggested in the original reconstruction drawing (fig. with fig. 2). The bead string can only be followed from the fibulae to below and above the waist. This also corresponds better with the lengths of the strings of beads. It is true that string Ca is relatively long, 90 cm, but if this string is placed around the neck, it reaches at the front only to above and not below the waist (as found in the grave). The same applies to the other bead string, which is only 66 cm long. This necklace, according to the original reconstruction, should have reached halfway to the breast, but that is not possible if the necklace was hung around the neck.

Between the two disk brooches there is only one bead, no. 101, which may have been rolled away later. Bead No. 98 is located behind the left fibula and nos. 99 and 100 are found behind the right fibula, in other words: the strings were attached to the two fibulae at the ends.

The string of beads at the waist also deserves extra attention. It is unknown if one deals here with a cord around the middle or

FIG. 13. Reconstruction of the clothing on the basis of 5th-century finds from Bad Lippspringe, Kr. Paderborn (based on Polenz 1985, Bild 36).

with a long string of beads randomly placed into the grave. According to the excavation drawing, most beads lay at the level of the belly. Only a few beads were on the side or behind the body. The small length, 55 cm, indicates that the cord did not cover the entire waist. On the contrary, it looks as if the beads were strung on a rope or a leather cord and that a knot was added when no more beads were available. This prevented slipping of the beads. Another possibility is that beads made from (now decayed) bone or wood have sat behind the back. Since there is no indication of a buckle closure, the cord could have been plainly tied.

Ga-1 linen thread, Z-twisted: thickness 0.6-1.0 mm

Ga-2 linen thread, Z-twisted: thickness 0.3 mm

Basting threads of bead strings?

Two different linen threads have been preserved behind the equal arm brooch fibula (Ga-1 and Ga-2). Due to the lack of exact information about the location of these thread residues with respect to the fibula, no conclusions can be drawn about the function of these threads. From the description on p. it has been found that they are not threads of a fabric, since the characteristic undulations [twist?] are missing. This also means that the threads are not used as sewing thread. The only possibility is that they have served as beading threads for the bead strings. Especially linen threads are suitable for this because of the large fiber length. In addition, these threads stretch less than threads spun from wool.

We-2 wool fabric in diamond twill with approx. 15 Z-x approx. 15 S-twisted threads per cm Ga-3 woolen yarn (bundles), Z-twisted threads, twisted in S-direction or not; thickness of the individual wires: 0.5-1.5 mm, thickness of the bundles: 2.0-5.0 mm

Coat [mantle] or shawl with fringe?

Fragments of wool fabric We-2 were found to be present under the equal arm brooch on the chest, with a number of bronze rings at the waist and under the bracelet, at the level of the hip. From the fact that this fabric is found under the equal arm brooch, one can conclude that one is dealing here with a fragment of an overgarment or with a cloak / shawl. Fragments of identical fabric found at the waist and under the bracelet argue for an overcoat, over the peplos. Moreover, this would explain why no linen remains were found around the waist: only the fabric that was in direct contact with the bronze rings have been preserved. For the other possibility speaks the fact that at rings b and c were residues of organic material with textiles on them found; copper oxide of a ring was found on top of this textile. If it is assumed that this organic material originates from the wooden box, just like the other fragments of wood bark that were found, it would mean that the woolen fabric is situated over the string of beads, in other words: it comes from a cloak or a scarf / shawl. Unless one is dealing with a draped garment here.

The latter does not seem very likely because then the precious string of beads would no longer have been visible.

The fabric fragments, located under the bracelet, do indicate an overcoat, but one could also imagine that a point of a cloak or cloth had been wrapped around the wrist.

With some bronze rings at the level of the waist were found remnants of thick wool bundles, both behind and on the body (Ga-3). It is doubtful whether these threads have served as thread for the large cord of beads. Consideration can be given to remnants of a woolen fabric of which only one thread system has been preserved (compare these threads with the coarse fabric We?-3, at ring a). It is much more likely that the yarn bundles are remnants of a fringe belonging to a double folded cloak or shawl.

Ga-4 woolen thread, Z-twisted: thickness 1.1 mm

We?-3 woolen threads or remnants of fabric; relatively thick, coarse threads

Function unknown

At the bronze rings c and d, a remainder of a relatively thick, Z-twisted thread was found (Ga-4), which could also originate from a yarn bundle identical to or comparable to Ga-3.

Under the torso a piece of woolen fabric (?) was found at ring a (We?-3). At first sight it seems to come from a coarse fabric. Since neither spin-twist nor weave type are visible, little can be said about the function of this textile.

We-4 coarse wool fabric with Z-x Z-twisted threads.

We-5 coarse woolen fabric in four-legged twill or rhizome with 9/10 Z-x 6 Z twisted wires per cm

Coarse blankets and / or cloths

Somewhat to the left of the middle of the cord of beads around the middle, with bronze ring d two fragments of a coarse fabric have been preserved (We-4). The fabric weave cannot be traced. It could be deduced from the thickness of the Z-twisted threads that the fabric contained approximately 10 wires per cm in both directions. This makes it less likely that these fragments are identical to fabric fragment We-5, which has been found at rings f and g. One possibility is that these are remains of a repair piece of the mantle.

If one looks more closely at the location of the fabric fragments in the grave, it is striking that these fabric=s could only come into contact with the bronze rings at those places where the cord of beads on the woman's belly was not covered by a cloak or shawl. This means that the woman wore a cloak / cloth that was closed on the chest with the large equal arm brooch, but which had fallen open at the front of the waist, so that part of the string of beads was visible. The occurrence of one or two coarse fabrics at that location could then be explained by assuming that a coarse cloth / blanket - whether or not with a repair piece cut from a coarse material - was placed over the dead body.

Conclusion

On the basis of the textile remains and the location of these remnants in the grave and on the other hand data on clothing of contemporary bogs, it is possible to make a new reconstruction drawing (fig. 14).

The Princess of Zweeloo could have been dressed in a linen peplos with a woolen coat or cloth with fringe over it. The peplos is made of linen, woven in [broken] diamond twill with 18 Z- x 18 Z-twisted threads per cm. Although the number of threads per cm in both directions is equal, the warp direction could be determined on the basis of the twist. During microscopic research it appeared that the twist of the threads in the one thread system was greater; the fibers of those threads are twisted more strongly than in the other system. This could mean that the former thread system was the warp, as warp threads, unlike weft threads, are subjected to high stress during weaving. The garment is woven in diamond twill 20/18, that is to say, after every 20 warp and 18 weft threads the rhombic pattern is repeated.

If one starts from the previously assumed warp direction, indicated by an arrow in fig. 4, it appears that the grain is perpendicular to the pin of the disk brooch, in other words, the weave is worn in the direction of the width and the card-woven edge, which runs parallel to the grain, is a selvage. At the top of this garment there was a 0.7 cm wide part of a card-woven trim, which was woven with at least 12 cards. Although no textiles have been left at the bottom of the grave, it may be assumed that the other selvedge, which was on the underside of the peplos, was also woven with cards.

From locations in Denmark it is known that such garments were made from a rectangular piece which was about 250 cm long and of which the two short sides were woven together or sewn. The width of the fabric corresponded to the length of the garment. If one starts from a hanging garment, without over-bloom on the waist, and a body length of 165-170 cm (Van Es and Ypey 1977, 109), then the fabric should have been at least 130 cm and at most 150 cm wide. A tubular garment was formed by sewing together or weaving the short sides. Since the number of threads per cm is relatively high for the garment of the Princess of Zweeloo, this is a supple, easily foldable garment, which was very wrinkle-resistant (linen!).

Considering the fact that linen threads have been used for this peplos, one can make statements about the color of this garment. Linen threads could not be dyed in pre- and protohistoric times with the then known vegetable dyes. This means that the garment originally had a light, natural-colored hue. By washing the linen often and allowing it to dry in the sun, the peplos can become white in color after a while.

This garment was held together on the shoulders by gilded disk brooches, the needles of which were inserted through the card-woven textile edges. Two string of beads hung from these fibulae, reaching to and below the waist. Off the left fibula, a small toiletry set with a thread of perishable material is confirmed. Indications for a leather bag are missing.

On the right breast an elongated bone fragment with line and hole decoration was found: a comb fragment (?), hanging off the right fibula or possibly belonging to the toilet trimmings?

Halfway up the chest, more or less in the middle, a tooth of a beaver, attached to a silver ring, has been found. Whether this amulet was attached to a cord around the neck or to the fibula, can no longer be determined.

On the chest, around the waist and on the left hip, woolen fabric fragments in [broken] diamond twill, with Z-x 15 S-twisted threads per cm, were found. This fabric has a diamond-shaped pattern which is repeated after every 20 warp threads and 18 weft threads as well. These fabric fragments could come from a cloak or a scarf / cloth with fringes. The presence of this garment under the bracelet, but on the leather of the bag or pouch, could be explained by assuming part of the cloak / shawl was placed under the hand and wrist with the bracelet. Whether the left hand and wrist were visible is not clear, because any remaining remnants of the cloak / shawl are not preserved at the top edge of the bracelet. This garment was closed at the level of the chest with a large equal arm brooch. The question is whether one is dealing here with a cloak or a shawl. The dimensions of coats excavated in, for example, Northern Germany are so large (about 270-300 cm long and about 150-170 cm wide, Schlabow 1976) that it is not possible to find fringe at the back of the body. This fringe could only be explained if one starts with a scarf with fringes whose dimensions are on average 140 x 50 cm (Hald 1980, 28 and 51, finds from Denmark). One option, of course, is that the cloak is folded in half and wrapped randomly as a scarf around the dead woman (fig. 15).

One or two coarse fabrics could, finally, come from one or two coarse cloths that had been laid over the dead.

Which other items of clothing and / or accessories the princess wore is unknown. Indications for an undergarment, such as a closure at the neck in the form of a small fibula or a metal closure at the wrist, are completely missing. However, it is not possible to exclude the possibility that the princess still had a shirt, but without metal closures, under the peplos. Such closures can in fact also be made of perishable materials.

Due to the lack of metal objects on or near the legs, no conclusions can be drawn about the presence of stockings or footwear. The woman probably wore low shoes, comparable to the footwear of several other sites from Drenthe (Groenman-van Waateringe 1970). These shoes belong to the 'Bundschuhe' type. In this footwear sole and upper leather consist of one piece. The front sheet could be cut out into strips. Through these strips a leather strap could be threaded so that the shoe could be pulled tight around the foot.

No metal pins were found in the grave near the skull. Such long pins are known from both Roman times and the early Middle Ages. They were used to hold a braid or knot in place, or to secure a veil,

hair netting or other head coverings (Laux 1983 and Moller 1976-1977). For this attachment you can also have wooden or leg pins. Any pins of organic material present at the burial site, however, have not been preserved in the Zweeloo soil.

Hair bands, made of textile, are only found if, for example, metal wire has been used in such a band [trim]. Some such bands have been found in female graves from the 6th century (Owen-Cracker 1986, 53, Schulze 1976, 158). Because it is unknown how the haircut of the princess originally looked, Van Es and Ypey (1977) decided to let the princess have loose hair. This hairstyle has been maintained in the new reconstruction.

At the left hip, keys and remnants of leather (?) have been found. The metal objects were hung off a string or (in this case more likely) the cord of beads. The leather remains undoubtedly originate from a bag or a pouch. How this looked is unknown. The preserved piece of leather is not folded. Folds are known from a money bag from the Bargercompascuum which has been excavated from peat (Schlabow 1956). In this pouch there were more than 300 coins, the youngest of which dates from the end of the 2nd century AD.

In the pocket or pouch of the princess, which is located under the bracelet on the left wrist, only objects of perishable material may have been stored. The keys were on top of the bracelet.

FIG. 14. Zweeloo: new reconstruction of the princess's clothing (based on watercolor by O. Goubitz, shown on cover).

FIG. 15. Zweeloo: possible reconstruction of cloth or double-build cloak of the princess.

Word of thanks

First of all, I would like to say thank you to Dr. V.T. van Vilsteren, Drents Museum in Assen. His request to make a design for the manufacture of one or more replicas of the clothes of the princess van Zweeloo, turned out not to be so simple.

Making a reconstruction drawing is easier than drawing up a concrete work plan for a weaver who actually has to make a replica. Fortunately, thanks to the stimulating discussions with Van Vilsteren, the image of a woman dressed in a linen peplos and adorned with precious jewelry slowly came together.

How the princess actually dressed could be visualized thanks to the great willingness of Mr. O. Goubitz, who works at the National Office for Ancient Soil Research in Amersfoort. Based on the new data, he was able to make a line drawing, which served as the basis for both fig. 14 and the watercolor painting on the cover of this New Drentse Volksalmanak. I would like to thank not only Mr. Goubitz but also Mr. F. Gijbels (Albert Egges of Giffen Institute for Pre- and Protohistory in Amsterdam) for the manufacture of resp. the watercolor and the photographs of the textile finds.

My thanks go out to Dr. W.A.B. van der Sanden, Drents Museum in Assen, for the critical reading of the text and the Drents Praehistorische Vereniging, for funding this research.

The drawings were made by the author.

BIBLIOGRAPHY

Alenus, J., 1961: Fouille mérovingienne à Folx-les-Caves, Bulletin de la Société Royale Belge d'Antropologie et de Préhistoire 72, 5-79.

Audollent, M.A., 1923: Les tombes gallo-romaines à inhumation des Martres-de-Veyre (Puy-de Dôme), Memoires de l'Institut national de France, Académie des inscriptions et belles-lettres, 275-328.

Becker, C.J., 1953: Zwei Frauengräber des 7. Jahrhunderts aus Nørre Sandegaard, Bornholm, Acta Archaeologia 24, 127-155.

Bender Jørgensen, L., 1986: Forhistoriske textile, i Skandinavien / Prehistoric Scandinavian textiles, København Nordisk Fortids minder, Serie B, 9).

Bloemers, J.H.F., L.P. Louwe Kooijmans en H. Sarfatij 1981: Verleden Land, Amsterdam.

Böhme, H. W., 1974: Germanische Grabfunde des 4. bis 5. Jahrhunderts zwischen unterer Elbe und Loire, München '(Münchner Beitrage zur Vor- und Frühgeschichte, 19).

Brinch Madsen, H., 1975: En nordjysk kvindegrav fra omkring 400 e. Kr., Hikuin 2, 123-130.

Christlein, R., 1978: Die Alamannen, Stuttgart.

Es, W. van en J. Ypey 1977: Das Grab der 'Prinzessin' von Zweeloo und seine Bedeutung im Rahmen des Gräberfeldes, Studien zur Sachsenforschung, 97-126.

France-Lanord, A., 1979: Les vêtements d'Arégonde, Dossiers de l'Archeologie 32.

Gebühr, M., 1976: Der Trachtschmuck der älteren römischen Kaiserzeit im Gebiet zwischen unterer Elbe und Oder und auf den westlichen dänischen lnseln, Neumünster (Göttinger Schriften zur Vor- und Frühgeschichte, 18).

Groenman-van Waateringe, W., 1970: Pre- en (proto)historisch schoeisel uit Drenthe, Nieuwe Drentse Volksalmanak 88, 241-262.

Hagg, I., 1967-1968: Some notes on the origin of the peplos-type dress in Scandinavia, Tor 12, 81-127.

Hagg, I., 1974: Kvinnodräkten i Birka, Uppsala (Archaeological Studies, Uppsala University Institute of North European Archaeology, 2).

Hagg, I., 1984: Die Textilfunde aus dem Hafen von Haithabu,  Neumünster (Berichte tiber die Ausgrabungen in Haithabu, 20).

Hald, M., 1980: Ancient Danish textiles from bogs and burials, København.

Hinz, H., 1978: Zur Frauentracht der Völkerwanderungs - und Vendelzeit im Norden, Bonner Jahrbücher 178, 347-365.

Hundt, H.-J., 1976: Textil- und Lederreste aus einem alamannischen Grab von Munningen, Kr. Donau-Ries, Saalburg Jahrbuch 33, 76-79.

Hundt, H.-J., 1980: Textilreste aus dem frühgeschichtlichen Kriegergrab von Sievern, Kr. Wesermünde, 1954, Studien zur Sachsenforschung 2, 151-160.

Hundt, H.-J., 1981: Die Textil- und Schnurreste aus der frühgeschichtlichen Wurt Elisenhof, Frankfurt am Main/Bern (Studien zur Ktistenarchaologie Schleswig-Holsteins, Serie A, Elisenhof, 4).

Koch, U., 1974: Mediterrane und fränkische Glasperlen des 6. und 7. Jahrhunderts aus Finnland, in: G. Kossack en G. Ulbert (Hrsg.), Studien zur vor-und frühgeschichtlichen Archäologie, München, (Festschrift für Joachim Wern er zum 65. Geburtstag, Teil 2), 495-520.

Lange, W.R., 1959: Zwei Körpergraber der frühen Völkerwanderungszeit aus Bad Lippspringe, Kr. Paderborn, Germania 37, 298-302.

Laux, F., 1983: Überlegungen zum Kopfputz der germanischen Damen im Niederelbegebiet und im übrigen freien Germanien, Studien zur Sachsenforschungen 4, 213-221.

Leene, J.E., 1973: Analyse der textilen Reste, in: H. Klumbach  (Hrsg.), Spätrömische Gardehelme, München (Münchner Beiträge zur Vor- und Frühgeschichte, 15), 80-83.

Magnus, B., 1983: How was he dressed? New light on the garments from the grave at Evebø/Eide in Gloppen, Norway Studien zur Sachsenforschung 4, 293-303.

Möller, J., 1976-1977 (1982): Zur Funktion der Nadel in der fränkisch-alamannischen Frauentracht, Jahrbuch des Römisch Germanischen Zentralmuseums Mainz 23-24 (Festschrift H.-J. Hundt zum 65. Geburtstag, Teil 3), 14-53.

Otter, J. den (red.), z.j.: Bij de tijd 2, geschiedenis voor de basisschool, Den Bosch.

Owen-Crocker, G.R., 1986: Dress in Anglo-Saxon England, Manchester.

Peddemors, A., z.j.: Nederland in de Vroege Middeleeuwen, Leiden (Knipblad Rijksmuseum van Oudheden).

Polenz, H., 1985: Römer und Germanen in Westfalen, Münster.

Raknes Pedersen, I., 1983: The textiles from the Migration period grave at Evebø/Eide, Gloppen, Sogn og Fjordane, Norway, Studien zur Sachsenforschung 4, 304-313.

Sachsen und Angelsachsen, 1978: Ausstellung des Helms-Museums. Hamburgisches Museum für Vor- und Frühgeschichte 18. November 1978 bis 28. Februar 1979, Hamburg.

Schlabow, K., 1956: De Romeinse muntschat van Bargercompascuum, III: De leder en buidel, Nieuwe Drentse Volks almanak 74, 219-221.

Schlabow, K., 1974: Var- und frühgeschichtliche Textilfunde aus den nördlichen Niederlanden, Palaeohistoria 16, 169-221.

Schlabow, K., 1976: Textilfunde der Eisenzeit in Norddeutschland, Neumünster.

Schlabow, K., 19784: Die Kunst des Brettchenwebens, Neumünster.

Schulze, M., 1976: Einflüsse byzantinischer Prunkgewänder auf die fränkische Frauentracht, Archäologisches Korrespondenzblatt 6, 149-161.

Vierck, H., 1978a: Trachtenkunde und Trachtgeschichte in der Sachsen-Forschung, ihre Quellen, Ziele und Methoden, in: Sachsen und Angelsachsen 1978, 231-243.

Vierck, H., 1978b: Die anglische Frauentracht, in: Sachsen und Angelsachsen 1978, 245-253.

Vierck, H., 1978c: Zur angelsächsischen Frauentracht, in: Sachsen und Angelsachsen, 1978, 255-262.

Vierck, H., 1978d: Religion, Rang und Herrschaft im Spiegel der Tracht, in: Sachsen und Angelsachsen, 1978, 271-283.

Werner, J., 1962: Ein reiches Laetengrabder Zeit um 400 n. Chr. aus Fécamp (Seine-Mariti me), in: Miscellanea archaeologica in honorem J. Breuer, Brussel (Archaeologia Belgica, 61), 145-154.

Wild, J.P., 1968: Clothing in the north-west provinces of the Roman Empire, Bonner Jahrbücher 168, 166-240.

Wild, J.P., 1970: Textile Manufacture in the northern Roman provinces, Cambridge.

Wild, J.P., 1977: The textiles from Vindolanda 1973-1975, Vindolanda.

Wild, J.P., 1985: The Clothing of Brittania, Gallia belgica and Germania inferior, in: H. Temporini en W. Haase (Hrsg.), Aufstieg und Niedergang der römischen Welt II, 12, Berlin, 362-422.

Ypey, J., 1955, Kunst en schoonheid uit de Vroege Middeleeuwen, Nijmegen/Amersfoort.

Ypey, J., 1962-1963: Die Funde aus dem frühmittelalterlichen Gräberfeld Huinerveld bei Putten im Museum Nairac in Barneveld, Berichten van de Rijksdienst voor her Oudheidkundig Bodemonderzoek 12-13, 99-152.

Ypey, J., 1965-1966: Das frühmittelalterliche Gräberfeld in Leersum, Prov. Utrecht, Berichten van de Rijksdienst voor het Oudheidkundig Bodemonderzoek 15-16, 145-168.

Ypey, J., 1969: Zur Tragweise frühfränkischer Gürtelgarnituren auf Grund niederlandisc her Befunde, Berichten van de Rijksdienst voor het Oudheidkundig Bodemonderzoek 19, 89-128.

Ypey, J., 1983: Ein Männergrab mit D-Brakteatenbeschlägen des frankischen Graberfeldes bei Rhenen, Provinz Utrecht, Niederlande, Frühmittelalterliche Studien 17, 460-478.

Zeller, F., 1974: Zum Wandel der Frauentracht vom 6. zum 7. Jahrhundert in Austrasien, in: G. Kossack en G. Ulbert (Hrsg.), Studien zur Vor- und frühgeschichtlichen Archäologie, München (Festschrift für Joachim Werner zum 65. Geburtstag, Teil 2), 381-385.

Zeist, W. van, 1956: Palynologisch onderzoek van enkele Drentse veenlijken, Nieuwe Drentse Volksalmanak 74, 199-209.

GLOSSARY

number of threads per cm: a measure of the fineness of a fabric.

2/2 twill: the weft thread always passes over one and under two warp threads. When the weft thread is turned, the binding point always shifts a place, forming an oblique line.

plain weave: the weft thread passes over one and under a warp thread. Here, too, the binding point shifts a place at each impact.

weft: see weaving.

card weaving: weaving narrow bands with the help of cards. There are at least 2 and at most 6 holes in each card. Through each hole a warp thread is inserted. The threads of all adjacent cards are stretched tightly. By Turning the cards a quarter of a turn, the warp threads rotate. This creates an opening through which the weft thread can be passed through (fig. 6). Card woven bands are usually used as a top edge or as a reinforcement of the sides of a fabric.

warp: see weaving.

Diamond twill: weave derived from the 2/2 twill, whereby the oblique line changes not only in the width but also in the longitudinal direction of the fabric at a regular distance from the direction, resulting in a diamond-shaped pattern

spinning: gradually pulling out a piece of fiber, which is given a certain twist (ply), whereby the fibers are rotated in turn. Depending on the direction of rotation one speaks of Z- or S-twisted single ply thread. This means that the orientation of the fibers in the spun thread runs parallel to the oblique line of the capital letter Z or S.

sprang: certain type of braid work.

standing weight loom: a loom, in which the warp threads, which are fastened at the top of a beam, are kept under tension. Hanging underneath, on the hanging warp threads, are the so-called weaving weights. The warp threads are attached to one or more beams, which can be pulled forward as desired, creating an opening [shed] where the weft thread can be guided.

twist (direction): see spinning.

ply: the twisting of two or more single spun threads. Twisting is usually done in the opposite direction to the spin direction.

2/2 twill: the weft thread is always about two and under two warp threads. The binding point shifts a place at each impact. This creates an oblique line, the so-called twill line.

herringbone pattern: a bond that is derived from the four-legged twill. With the herringbone pattern the direction of the twill line changes at regular intervals in the longitudinal direction of the fabric.

weaving: this assumes at least two wire systems, namely the string (in the longitudinal direction of a fabric) and the weft (in the width direction), which intersect at right angles. This crossing can take place in different ways, so that multiple bonds are formed (see plain weave, three-, four-binder, herringbone and diamond twill).

From:

Vons-Comis, S.Y., 1988. Een nieuwe reconstructie van de kleding van de ‘Prinses van Zweeloo’. Nieuwe Drentse

Volksalmanak 105, 151–187.
https://www.worldcat.org/title/van-rendierjager-tot-ontginner-nieuwe-oudheidkundige-ontdekkingen-in-drente-33/oclc/1014869443&referer=brief_results


For more on this project, check my paper "Reconstructing the Wardrobe of the Princess of Zweeloo" at Academia:
https://www.academia.edu/38845292/Reconstructing_the_Wardrobe_of_the_Princess_of_Zweeloo