Fennoscandia archaeologica XXXVII (2020)
Tuija Kirkinen, Jaana Riikonen, Carla Dove & Juha Ruohonen
THE IDENTIFICATION AND USE OF FUR AND FEATHERS  
EXCAVATED FROM THE LATE IRON AGE AND EARLY MEDIEVAL (12TH–13TH 
CENTURIES) RAVATTULA RISTIMÄKI CEMETERY IN KAARINA,  
SOUTHWEST FINLAND
Abstract
The Ristimäki (‘Cross Hill’) inhumation cemetery in Ravattula village in Kaarina municipality, 
Southwest Finland, dates to the Late Iron Age and Early Medieval times (12th–13th centuries). In 
addition to a large cemetery, this site is documented as having remains of the earliest known 
church in the country. This study presents the descriptions of fur garments, fur-lined artefacts and 
grave furnishings that were excavated in Ravattula Ristimäki in 2010–16. Animal hairs and feath-
ers were preserved in eight out of the 61 burials excavated in the research project. We identified 
eight species of mammals and one family of birds (Anatidae) from the burials that contained 
organic material. Human scalp hair was found in several female burials, adding information on the 
hairstyles of the period. Together with blankets filled with feathers, these items indicate highly 
versatile and varied burial customs within the cemetery. This study provides new evidence of the 
use of fur garments and accessories, as well as grave furnishings, at the dawn of Christianity in 
Northern Europe.
Keywords: animal hair identification, Christianization, feathers, furs, inhumations, Middle Ages
Tuija Kirkinen, Dept. of Cultures, P.O. Box 59, FI-00014 University of Helsinki, Finland: tuija.kirkinen@
helsinki.fi; Jaana Riikonen, Dept. of Archaeology, Akatemiankatu 1, FI-20014 University of Turku, 
Finland: jaana.riikonen@kaioksanen.fi; Carla Dove, Feather Identification Lab, National Museum 
of Natural History, Smithsonian Institution, NHB E-600, MRC 116, P.O. Box 37012, Washington, DC 
20013, U.S.A.: dovec@si.edu; Juha Ruohonen, Dept. of Archaeology, Akatemiankatu 1, FI-20014 
University of Turku, Finland: jukaru@utu.fi.
Received: 15 March 2020; Revised: 22 June 2020; Accepted: 7 September 2020.
INTRODUCTION metres from the village of Ravattula. As part of a 
comprehensive study of this site (e.g. Ruohonen 
The Ristimäki cemetery in the village of 2017; 2019) by the Department of Archaeology 
Ravattula, in the municipality of Kaarina in at the University of Turku, a stone foundation of 
Southwest Finland (Fig. 1), is well known be- the small wooden church, a large cemetery and 
cause it is located on the site of the earliest the remains of a wall surrounding the area were 
known church in the country. The well-pre- partly excavated in 2010–16.
served church foundations are approximately 4 A church with a narrow choir (Fig. 2) was 
km from the medieval Cathedral of Turku, near built on the site in the second half of the 12th cen-
the Aurajoki River, which flows to the Baltic tury and used well into the first half of the 13th 
Sea. The site is a low moraine hillock of for- century (Ruohonen 2019: 113) by local villages. 
est called Ristimäki (‘Cross Hill’) surrounded Ravattula church is the location of the oldest 
by clay fields and situated approximately 250 hitherto identified remains of an ecclesiastical 
45
Figure 1. Location of Ristimäki cemetery in Ravattula, Southwest Finland. 
(Drawing: Juha Ruohonen.)
Figure 2. Stone footings of Ristimäki church during the excavations. (Photo: Juha Ruohonen.) 
46
RAVATTULA RISTIMÄKI CEMETERY
The hill of Ristimäki was used as an inhuma-
tion cemetery from the first half of the 12th 
century, several decades before the church was 
built, while the latest burials probably date to 
the middle of the 13th century. This time period 
is of great interest because it sheds light on the 
Christianization process in Southwest Finland. 
All burials in Ravattula Ristimäki (hence-
forth called ‘Ristimäki’) cemetery are classi-
fied as Christian, or at least deeply Christian-
influenced, on the basis of an almost total lack 
of tools and weaponry, which were common 
grave gifts in many other early Late Iron Age 
sites (e.g. Hiekkanen 2007: 13; 2010: 340–1). 
Additionally, certain individuals were buried in 
full dress, including metal items such as pen-
annular brooches, belt buckles, knives, knife 
sheaths, and bronze spirals, while most of the 
excavated burials did not contain artefacts other 
than iron nails from the coffins. Most of these 
findless graves are probably from the last usage 
period of the site in the 13th century. The absolute 
Figure 3. The distribution of inhumation burials chronology of late inhumation graves is difficult 
around the church. (Drawing: Siiri Tuomenoja.) to determine due to the lack of datable material 
(see Hiekkanen 2010: 343).
In Finland, the preservation of organic mate-
building in Finland and, as far as is known, the rial such as bones and plant fibres is poor due to 
only one dated to the period of missionary ac- the acidity of soils (Arponen 2008; Hurcombe 
tivity in the area. The church buildings that fol- 2014: 93). The preservation of animal fibres is 
lowed date to the 1230s or later and are from the much better, and especially in contact with cop-
known period of established parishes (Ruohonen per-bearing items even large pieces of woollen 
2016: 235). textiles may have been preserved (see Solazzo et 
Evidence of inhumation burials (Fig. 3) is al. 2014). This is the situation in Ristimäki, too, 
numerous at Ristimäki, especially around the where woollen and even some linen textile frag-
church site. In 2010–16, 61 inhumations of the ments were preserved in 17 of the total of 61 ex-
several hundred graves were excavated as part cavated graves, although most of the pieces were 
of the ongoing research project (Ruohonen small and badly preserved. Regardless, garments 
2019: 114). This study of fur and pelt remains, and accessories were identified in 12 burials.
and feathers suggesting textiles stuffed with In Ristimäki, female burials were main-
down, provides valuable information regarding ly clothed in peplos-type dress, well known 
the ways the graves were furnished and the man- from other Crusade Period (c. 1025–1200 AD) 
ner in which the deceased were dressed at the cemeteries in southwestern Finland (see, e.g. 
dawn of Christianity in Finland. We also ask if Lehtosalo-Hilander 1984; 2001; Luoma 2003). 
some of the findings can be linked to the process The dress was complemented with an apron dec-
in which Christianity was adopted. Moreover, orated with bronze spirals and an undergarment 
the finds illustrate hunting and/or trading of furs made of linen. The deceased were covered with 
in an era during which furs were highly valued. a shawl, and sometimes they were equipped with 
headgear and mittens in nålebinding (Vajanto 
2014). In one female grave (41/2016) dated to 
47
around 1200, the deceased was clothed in a gar- 3). The cross-sections of the fibres were made 
ment sewn of several pieces to give a moderate after Greaves & Saville (1995: 39–40). The rest 
width at the hem. This outer garment and the of the samples were archived in small Eppendorf 
sewn stockings show the influence of medieval tubes. The material was studied with light and 
fashion, which is unique in the Finnish material polarised microscopy, using a Leica DM 2000 
of this era (Riikonen 2019: 174–5, 180). LED microscope with 200x and 400x magnifi-
Due to the acidic soil, the human remains cation, and documented with a Leica ICC50 E 
were not preserved and most often there was camera.
only tooth enamel or a coloured soil layer left In order to study possible fur remains from 
from the deceased. Unfortunately, this prevented contexts in which the preservation of organic 
the determination of gender, which could only materials was relatively low, soil samples of ca. 
be presumed on the basis of the material found 0.5 dl in size were floated in water. The float-
in the graves. We determined that 13 of the exca- ing material was studied with a microscope (see 
vated burials were female graves, two were male above) and documented by a camera only. No 
graves, and one was a burial of a young girl. Of permanent slides were prepared.
these, both male graves and five female burials Fibres were identified by the morphology 
are included in this study. of their longitudinal shafts and by their cross-
sections. The key features for identification were 
MATERIAL AND METHODS the diameter and length of the hair, the structures 
of medulla and cuticular scales, the shape of root 
Laborative excavation and tip sections, and pigmentation (Goodway 
1987). The identification of fibres was based on 
The burials in which textiles or artefacts were the identification keys of Tóth (2017), Teerink 
detected were excavated in a conservation labo- (2003), and Rast-Eicher (2016). The samples 
ratory at the Museum Centre of Turku. Intact soil were compared with the reference material at the 
blocks were X-rayed and restored in a refrigera- Zoological Museum of the Finnish Museum of 
tor before micro-stratigraphic excavation under Natural History, University of Helsinki.
a stereo microscope. The study of the blocks was The morphology of human scalp hairs was 
documented by photographing and by 1:1-scale studied with light microscopy, following the 
drawings on plastic sheets. All organic materi- same protocol as with animal hairs. The classifi-
als (e.g. human hair) were sampled in alumini- cation of the hairs followed Ogle & Fox (1999). 
um folio or Eppendorf tubes for identification. In contrast to animal hairs, the morphologi-
In addition to these samples, the excavated soil cal features of human scalp hairs were largely 
from the block, as well as soil samples taken in decomposed.
the field, were analysed for pollen, macrofossils, 
and minuscule fibres. Feathers
A sample of matted feather material (from burial 
Samples 8/2015) containing plumulaceous (downy) feath-
ers was loaned to the Smithsonian Institution’s 
Hair Feather Identification Lab in Washington D.C., 
The fibres were separated from each other and USA for identification. The feather material was 
washed by gentle stroking with a soft brush. The not cleaned prior to examination due to the frag-
most clayed samples were purified by hydrogen ile condition of the material. Microslide prepara-
fluoride (HF), which effectively destroys sedi- tion of the feather fragments followed Dove and 
ment residues (e.g. LacCore SOP Collection). Peurach (2002). Microslides were viewed using 
The samples were prepared for microscopic a Leica® DM750 (Leica Microsystems, Wetzlar, 
examination by mounting them in Entellan Neo Germany) comparison light microscope. 
after Greaves & Saville (1995: 7). Additionally, Photomicrographs of diagnostic feather char-
the scale structures were studied by preparing acteristics were taken with a Leica® DFC290 
longitudinal negative casts with transparent nail HD camera (Leica Microsystems, Wetzlar, 
polish (after Kirk et al. 1949; Tridico et al. 2014: Germany) using the Leica Application Suite® 
48
(version 4.12.0, Leica Microsystems, Wetzlar, evidently from textiles. Additionally, one bea-
Germany). Twelve microslides were made from ver (Castor fiber) hair was detected from a soil 
various areas of the sub-sample that was loaned sample. The diagnostic features of the hairs, the 
for identification. Feather analysis focused on context of the finds, and the identifications are 
the plumulaceous feather types within the burial listed in Table 1 (see also Fig. 4).
sample to search for diagnostic characters spe- Feathers were found in large quantities in 
cific to Orders of birds. burials 1/2015 and 8/2015, and small barbs 
were also detected in burials 4/2014, 11/2014, 
RESULTS and 18/2016 (see Table 1). The feather sub-sam-
ple from burial 8/2015 that was studied at the 
Organic fibres such as textile, fur remains, and Smithsonian contained both pennaceous barbs 
feathers were detected in 20 burials in Ristimäki and plumulaceous barbs with diagnostic micro-
(see Table 1). Animal hairs were relatively well scopic characters for identification. Analysis of 
preserved, and most of them could be identified the downy material revealed many diagnostic 
to a species or family level. Fur remains were nodes characters of the avian family Anatidae 
found in close association with bronze items, (duck, goose, swan). Species-level identification 
where toxic copper alloys were absorbed into was not possible due to the degraded condition 
the fibres by preventing the activity of micro-or- of the whole feathers and the lack of access to 
ganisms (Janaway & Scott 1989; Solazzo et al. the entire sample. The microscopic structures 
2014). It is very likely that the thin clayed soil of plumulaeous (downy) barbs of feathers of 
surface on top of the finds improved the preser- Anatidae are diagnostic by triangular-shaped 
vation of the fibres. nodes located on the distal portion of the bar-
As an exception, the human scalp hairs mac- bules (Dove & Agreda 2007), which were found 
roscopically looked well preserved, but micro- on all microslides (Fig. 5). We did not find diag-
scopic examination showed their poor condi- nostic feather characteristics of any other avian 
tion (Fig. 4F). As a result, no further analysis order. However, because only a small sub-sam-
of the hairs could be done. Human scalp hairs ple of material was submitted for examination, 
were preserved in large quantities in seven fe- we do not know if other bird species were used 
male burials (1/2015, 2/2015, 6/2015, 8/2015, in the complete burial garments. Visual differ-
18/2016, 20/2016 and 37/2016). Additionally, ences were noted in the internode length (space 
single hairs were found from graves 4/2014, between nodes) in some of the barbules, indicat-
15/2015 and 38/2016. ing that possibly more than one sub-family of 
Animal hairs were found from six graves for waterfowl was used for the item examined in 
a total of 11 samples (Table 1). The hairs were this study.
mostly loose, and pieces of skin were found in- Plumulaceous barbule fragments diagnostic 
termixed with soil samples taken from a dark to waterfowl (Anatidae) was also identified in 
layer in graves 3/2015 and 18/2016. Most of the a sub-sample from burial 1/2015, from a pho-
samples that we examined included both coarse tomicrograph of the material. This material was 
and fine hairs, which assisted with the identifica- highly degraded and species-level identifica-
tion of species. Preserved hairs were relatively tions were not possible.
long, about 0.8–1.5 cm, except for horse tail or 
mane hairs, which were longer. In most cases, DISCUSSION
both the scale structure and the medulla of the 
hairs were identifiable. The Christianization process started gradually in 
From the hair, we identified mostly preda- the Turku district in the 11th century at the latest, 
tors, such as brown bear (Ursus arctos), stoat when the first inhumations were made in crema-
(Mustela erminea), Canidae/red fox (Vulpes tion cemeteries. In the next stage, farm or fam-
vulpes), and Felidae/lynx (Lynx lynx). Besides ily cemeteries were established in the vicinity 
these, we also identified domestic animal hairs of old cremation cemeteries. In these inhuma-
(i.e. horse mane or tail hair), dog (Canidae) tion cemeteries, the deceased were buried fully 
skin remains, and sheep (Ovis aries) hair most dressed, wearing their personal jewellery. From 
49
Table 1.  Description and identification of hair, fur, and feathers excavated from the Ravattula Ristimäki 
Cemetery in Kaarina, Finland, in 2010–6. The items listed here are archived in the archaeological 
collections at the University of Turku (TYA). 
Catalogue  Grave  Item and  Species  Diagnostic features  Identification
number (TYA) number sample origin identification of hair, fur and feathers references
Hairs on the 
top of a belt Small amount of 8-mm-long coarse (120 Teerink 
863:217 1/2010 buckle, found Stoat (Mustela 
µm) and fine hairs. Medulla cloisonné 2003; 
near the erminea) pattern, medullar margins scalloped. Tóth 2017; 
thighs of the Cuticular scale pattern in coarse hairs is Rast-Eicher 
deceased. rounded wave-like. 2016
Organic mate-
rial under the Unmedullated and non-continuous 
head of the amorphous medullated hair fragments. 
deceased, Brown bear Cuticular scale pattern is triangular-
912:523b 4/2014 between (Ursus arctos); shaped and matches best with brown 
Tóth 2017; 
a bronze unidentified bird bear fine/intermediate hairs. Width 56.6 
Rast-Eicher 
sheath and (Aves) µm, max length 11 mm.
2016
the bottom of Unidentified pennaceous feathers pre-
the coffin. sent in sample.
Coarse and possible fine hair fragments, 
Organic mate- Teerink 
rial inside Red fox? (Vulpes 
some root sections included. Coarse 
912:523d 4/2014 vulpes / Cani- hair medulla is of cloisonné type, cuticu-
2003; 
a bronze dae) lar scales are wave-patterned. Coarse 
Tóth 2017; 
sheath. hair width 112.9 µm, max length 6 mm. Rast-Eicher 
Best match is red fox. 2016
Hair fragments from two species. Fine 
and intermediate bear hairs (max width 
53.8 µm) were identified by their nar-
Organic Brown bear row unstructured medulla, triangular/
material (Ursus arctos); diagonal-shaped petal scale type and Tóth 2017; 
912:523f 4/2014 from the top domestic sheep oval cross-section. Rast-Eicher 
of a bronze (Ovis aries); Several additional fine (width about 17 2016
sheath. textile µm) unmedullated sheep hairs (probably 
domestic). One kemp hair (50.9 µm) 
with its net-like medulla was present in 
the sample. Length max 10 mm.
Coarse hair fragments, some with 
Organic root sections attached. The medulla is 
material from cloisonné-like with fine-grained transpar-
the top of a ent structure and spaces filled with air. 
belt buckle, Eurasian lynx? The medullar margins have partly fibril-
912:1631 11/2014 on the pelvis (Lynx lynx / Fe- like fringes. The cuticular scales are 
Tóth 2017; 
area of the lidae); unidenti- wavy-patterned. The colour is very light. Rast-Eicher 
corpse, hairy fied bird (Aves) Width 52.4 µm, max length 15 mm. 2016
material side Best match is lynx.
facing the The sample also included unidentifiable 
buckle. microscopic feather fragments, length 
250 µm.
50
Catalogue  Grave  Item and  Species  Diagnostic features  Identification
number (TYA) number sample origin identification of hair, fur and feathers references
Soil sample 
collected A coarse/intermediate hair fragment, Tóth 2017; 
near the European beaver black in colour. The medulla is irregular Teerink 
MN197/2014 11/2014 pelvic area (Castor fiber); unicellular. Width 40 µm, length 4 mm.  
of the corpse unidentified bird Most similar to beaver.
2003; 
beside metal (Aves) The sample also included unidentifiable 
Rast-Eicher 
2016
artefacts. microscopic feather fragments.
Human scalp 
914:1259:11 1/2015 hairs from a Human Badly decomposed human scalp hairs. Ogle & Fox 
bun. Width 39 µm. 1998
Organic mate-
rial under 
the right side Several badly decomposed bird feather Duck, goose, fragments. A barbule with diagnostic Dove & 
914:1259:4 1/2015 waist of the deceased, swan (Anseri- microscopic triangular-shaped node Agreda 
on the top of formes) characteristic of waterfowl. Species 2007
the wooden undetermined.    
coffin 
914:1269:9 2/2015 Human scalp Human Badly decomposed human scalp hairs. Ogle & Fox hairs Width 47 µm. 1998
Fur from the 15-mm-long hairs, in which the continu-
bottom of the ous medulla is about 1/3 of the width of 
grave, near the hair. The cuticular scales in coarse 
Teerink 
914:1276:2 3/2015 the lower Canidae hairs (max 90.7 µm) are triangular 
2003; 
right corner of petals and have the best match with Ca-
Rast-Eicher 
the coffin. nidae. The cuticular scales in fine hairs 
2016
(25 µm) are wavy petal-like.
Human scalp 
914:1316:6 6/2015 hairs formed Human Badly decomposed human scalp hairs. Ogle & Fox 
into a bun. Width 72 µm. 1998
914:1367:7 8/2015 Human scalp Human Badly decomposed human scalp hairs. Ogle & Fox hairs. Width 66 µm. 1998
Organic mate-
rial from the Several relatively well-preserved light or 
bottom of the white bird feathers with both penna-
grave, under Duck, goose, ceous and downy barbs. Barbs with di- Dove & 914:1367:11 8/2015 the right swan (Anseri- agnostic microscopic triangular-shaped Agreda 
side of the formes) nodes located distal on barbules 2007
deceased’s characteristic of waterfowl.  Species 
hip. undetermined.  
Organic 
material from 
914:1367:20 8/2015 the left side Unidentified bird Several badly decomposed bird feather 
bottom of the (Aves) fragments.
grave.
993:173:10b 18/2016 Hair inside Horse (Equus Tail hair.  The identification is based on Rast-Eicher headgear. caballus) the width of the hair. 2016
51
Catalogue  Grave  Item and  Species  Diagnostic features  Identification
number (TYA) number sample origin identification of hair, fur and feathers references
Coarse and fine/intermediate hairs of 
Small amount a bear. The coarse hairs have uniserial 
of fibres ladder medulla; no scales are visible 
under a Brown bear due to the decomposing of hairs. The 
993:173:10b 18/2016 headgear and (Ursus arctos); intermediate/fine hairs have amorphous 
Tóth 2017; 
an apron, on unidentified bird interrupted medulla and triangular- Rast-Eicher 
the stom- (Aves) shaped scales. Max width 108.9 µm 2016
ach of the (coarse), max length 4.5 mm. 
deceased. Small amount of feather barbs also 
found in the sample.
Lots of organic material, skin fragments 
Soil sample and hairs floated from the soil sample. 
993:173 18/2016 collected Brown bear Max width 96 µm (coarse) and 12–15 
Tóth 2017; 
from the (Ursus arctos) µm (fine). The width of unicellular/con- Rast-Eicher 
grave. tinuous medulla is 1/3 of the total width 2016
of the hair.
Human scalp 
993:214:17 20/2016 hairs from a Human Badly decomposed human scalp hairs. Ogle & Fox 
bun. Width 53 µm. 1998
A single 16-mm-long and 232-µm-wide 
hair possibly from the tail. The hair is 
dark brown and the width of the medulla 
Long hair is difficult to determine in longitudi-nal photos. No cuticular scales were 
993:214:16a 20/2016 from the Horse (Equus Rast-Eicher selvedge of a caballus) preserved. The cross-section is oval and 
headgear. it shows a spongy medulla the width of 
2016
which is about 1/3 of the total width of 
the hair. Possible identification is based 
on the cross-section and on the width 
of the hair.
the middle of the 12th century, roughly east-west- in the beginning of the 13th century. (Ruohonen 
oriented inhumations were made in cemeter- 2018; see also Hiekkanen 2010.)
ies maintained by several villages (Ruohonen One of the most important source materi-
2018). In Finnish archaeological periodization, als for the study of the community that buried 
this era has been called the Crusade Period their dead in the cemetery is the remains of dress 
(1025–1200 AD), which underlines the ongo- details and the accompanying bronze spirals, 
ing Christianization of the area (e.g. Hiekkanen brooches, and other jewellery. Most evidently, 
2010: 325–6). fur and feather remains suggest highly versatile 
At Ristimäki hill, the church remains and the and varied burial customs within the cemetery. 
cemetery around comprise a vital monument of We discuss in detail below the use of fur and 
this epoch. The cemetery was in use from the down-stuffed textiles as grave items.
Late Iron Age until the beginning of the Early 
Medieval period. It seems that Ravattula church, Grave furnishing
located at Ristimäki, was in use until the estab-
lishment of the first parishes in Aurajoki district In Ristimäki burials, animal skins and bird 
feathers were used for furnishing the grave. 
52
Figure 5. Plumulaceous (downy) microscopic 
feathers of the Ristimaki sample showing di-
agnostic waterfowl characters of triangular-
shaped nodes located distally on barbules. 
(Photo: Carla Dove.)
The height of the deceased was approximated to 
be under 140 cm, which indicates the young age 
Figure 4. Animal and human hairs recovered of the girl. She was clothed in a dress, which was 
from Ristimäki cemetery: A) Vulpes vulpes fastened with a penannular brooch, and an apron 
(H4/2014), B) Felidae/Lynx lynx (H11/2014), and a spiral-ornamented shawl. These garments, 
C) Mustela erminea (H1/2010), D) Ursus arc- as well as a string of pearls and headgear, point 
tos (H18/2016), E) Castor fiber (H11/2014), to a status of an adult woman. The probable quilt 
F) Human scalp hair (H6/2015). (Photo: Tuija found from this grave may also indicate the high 
Kirkinen.) status of the deceased.
Grave 18/2016 contained small fragments 
of feathers under an apron and headgear, which 
Brown bear (Ursus arctos) pelts were identi- was placed above it, possibly indicating a down-
fied in two burials (i.e. in female graves 4/2014 stuffed item. In burial 4/2014, single feather re-
and 18/2016). In grave 4/2014, bear hairs were mains between the bear skin and the knife sheath 
found from the bottom of a coffin as well as from under the head may provide evidence of the use 
the top of a wide, bronze-plated knife sheath of a pillow filled with down. However, down and 
found under the deceased’s head (Fig. 6). In feathers were preserved only in limited areas, 
grave 18/2016, brown bear hairs were identified which makes interpretation of the finds difficult.
near the deceased’s waist and from a soil sample The furnishing of graves and the covering 
taken from the stomach area of the deceased. In or wrapping of the deceased in animal pelts is 
these burials, the corpse was evidently wrapped an age-old tradition, which was practised in 
in a bear pelt. Finland throughout the Iron Age. In Late Iron 
In five burials (see Table 1), feather fragments Age inhumations, the deceased were most of-
indicate the use of pillows or quilts filled with ten laid to rest on deer (Rangifer tarandus) and 
down. The large quantity of downy feathers un- European elk (Alces alces) skins, sometimes 
der the deceased’s pelvis in grave 8/2015 can be also on brown bear and cattle skins (Cleve 1978: 
hypothesized as a quilt, which was placed on 82; Lehtosalo-Hilander 1982: 68; Asplund & 
the bottom of the grave (Fig. 7). Also in burial Riikonen 2007: 25; Kirkinen 2019: 62–5). In 
1/2015, downy feathers were found on the bot- Ristimäki, the lack of Cervidae skins is excep-
tom of the grave near the waist of the deceased. tional compared to other Late Iron Age and 
This item has also been hypothesized as a quilt. Early Medieval cemeteries in Finland. Also, 
53
Figure 7. Downy feathers and woollen tex-
tile from female grave 8/2015. (Photo: Riikka 
Saarinen/Museum Centre of Turku.)
Figure 6. Bear hairs in grave 4/2014, found from 
the bottom of a coffin. (Photo: Jaana Riikonen.) Garments
Evidence of fur garments were found from two 
although brown bear skins were a central ele- male graves. In male grave 11/2014, Felidae/
ment in Merovingian Age (c. 550–800 AD) and lynx hairs were detected on a belt buckle found 
Viking Age (c. 800–1050 AD) cremation cem- near the pelvis, of which the skin side was against 
eteries, they have been found from more recent a thick woollen twill fabric (Fig. 8). Fragments 
inhumations only occasionally. of two different tabby fabrics preserved under 
Luxurious feather remains are rare in Finnish the belt buckle were maybe from a shirt and a 
Late Iron Age and Early Medieval contexts, as tunic girded with a leather belt. The deceased 
a small amount of down has been found in pre- was probably covered with a woollen, fur-lined 
vious research from only one burial in Luistari cloak. It is also possible that the lynx skin origi-
cemetery, Eura, Southwest Finland, possibly in- nated from a fur-lined rug. From the same grave, 
dicating the remains of a pillow (Kirkinen 2015: one beaver hair, black in colour, was found from 
107). However, this is partly due to the excava- a soil sample. The grave was found under the 
tion methods used in the past. This underlines foundation of the church, so it was older than 
the importance of high-resolution laboratory ex- the building.
cavations in producing new data about the use of Stoat (Mustela erminea) hairs were found on 
feathers. Also, microscopic examination of old top of a bronze belt buckle found near the thighs 
materials might produce new feather finds. of the deceased in male burial 1/2010. This find-
According to Dove & Wickler (2016: 33), ing, as well as the Felidae hairs presented above, 
the use of feathers in burials is often associ- most evidently indicate the lining of fabrics with 
ated with precious clothing and costly textiles. fur animal skins. A fragment of mineralized tab-
In Scandinavia, feather- and down-stuffed pil- by fabric was preserved under the belt buckle.
lows and quilts have been found in two Vendel Finally, horse tail hairs were found in two fe-
Period boat-graves in Valsjärde, Sweden, and in male burials. In grave 20/2016, they were inside 
a number of Viking Age high-status graves (see the tubular selvedge, and in grave 18/2016 in-
Gräslund 1980: 14 with references; Berglund side the headgear. In both cases, horse tail hairs 
2009; Rast-Eicher 2016: 291 with references). were used probably for hardening the structure 
Also in Ristimäki, feathers were found from bur- of the garment.
ials in which the dress, glass beads, metal items, 
and textiles might indicate the relative wealth of Accessories
the deceased.
In female grave 4/2014, Canidae/red fox skin 
remains were detected inside a knife sheath, 
54
Figure 8. Felidae/lynx hairs, recovered from a 
belt buckle from male grave 11/2014. (Photo: Figure 9. Canidae hairs and skin fragments from 
Riikka Saarinen/Museum Centre of Turku.) grave 3/2015. (Photo: Riikka Saarinen/Museum 
Centre of Turku.)
indicating the lining of the item with fur. These 
kinds of large bronze-plated knife sheaths lined Luistari cemetery, where dog bones have been 
with lynx (Luistari in Eura, grave 56), red squir- identified from several male burials (Lehtosalo-
rel (Kirkkomäki in Kaarina, grave 40/1992; Hilander 1982: 29–30). According to Anne-
Kirkinen 2015: Fig. 10), possibly Phocidae Sofie Gräslund (2010: 138), dogs are commonly 
(Kirkkomäki in Kaarina, grave 1/1950; present in Late Iron Age graves. On the basis of 
Riikonen 1990: 25–6) and Bovidae (Kirkkomäki Old Norse texts, she interprets dogs as facili-
in Kaarina, grave 23/1991; possibly calf [Lehto tators between the living and the underworld. 
1993: 33] or sheep [Kirkinen 2015: 119]) Alternatively, Canidae fur fragments might be 
have been excavated from other cemeteries in interpreted as remains of footwear or gaiters.
Southwest Finland, too.
CONCLUSIONS
Human scalp hairs
Before homogenized Christian burials, funerary 
The material collected from Ravattula gives us practices varied greatly. The rituals sought to 
valuable information about the hairstyle and equip the deceased for the afterlife and helped 
fashion of this time. Before Ristimäki, com- them to become a member of the ancestors. In 
plete early medieval hairdressings have been turn, in Christian burials, the deceased were re-
found only once, at Kirkkailanmäki cemetery in leased to God, and all members of the society 
Hollola (Hirviluoto 1985: 30). were buried in a dedicated churchyard (Gräslund 
No further information was obtained by the 2010: 134–5). Archaeologically, these graves 
morphological analysis of the hairs. We continue appear as “invisible” due to their findless nature, 
to study the human hair samples in hopes of ob- except for a couple nails or coffin wood. The de-
taining aDNA sequences through collaborations ceased was not clothed in full dress but wrapped 
with the University of Turku. or covered with a fabric. An example of a body 
being carried to the grave pit, covered only by 
Animal remains a shroud, can be seen on the Bayeux tapestry 
dated to the 11th century (Gräslund 1980: 15). 
In burial 3/2015, at the lower part of the grave No remains of coffins can be found when the 
on top of the coffin bottom planks, a decom- deceased started to be buried with shrouds only.
posed dark layer containing Canidae hairs and Ristimäki cemetery in Ravattula village is 
skin fragments was detected (Fig. 9). We inter- truly a unique site in Finland because of the lo-
pret this finding possibly as the remains of a dog, cation near the earliest known church. The bury-
which was placed at the foot of the deceased. ing of the dead started before the church was 
This tradition is known especially from the built and probably continued for some time after 
55
the building was abandoned. As such, the graves also with cattle skin (Asplund & Riikonen 2007: 
evidence the Christianization process of the so- 25; Kirkinen 2019: 219). Equally, there were no 
ciety in southwestern Finland. signs of bear pelts or bird feathers, although the 
This study describes and identifies the hair, preservation conditions in contact with bronze 
fur and feather material from the Ristimäki cem- items were relatively good. Neither were there 
etery and provides an interpretation of their ori- significant differences in wealth, as Kirkkomäki 
gins as grave furnishings, garments, accessories also had several very richly equipped graves.
and accompanying animals. The wide variety of To conclude, these nearby cemeteries dif-
fibres and feathers traced to the graves were ob- fer from each other in respect to the ways in 
tainable through high-scale laboratory excava- which the graves were furnished. This might 
tion and documentation. relate to the stage of Christianization of the 
Compared to other Crusade Period or Early sites. Kirkkomäki represents a cemetery for 
Medieval cemeteries in Finland, where Cervidae, a single house, being slightly older (11th–12th 
mountain hare (Lepus timidus) and fur animal centuries) than Ristimäki (12th–13th centuries). 
skins were used especially for mittens, seal The change from cremation burials to inhuma-
skins for trousers or gaiters, reindeer skins for tions provides evidence of the early influences 
outer garments and furs for linings and collars, of Christianity on the community, although 
Ristimäki fur garments are one of the most luxu- most of the deceased were still equipped with 
rious ones. Black beaver skins, which are known tools, weapons and meals (Asplund & Riikonen 
to have been one of the most valuable furs in the 2007). However, in grave 15 at Kirkkomäki 
past (Pylkkänen 1955: 90; Seppälä 2009: 164), the deceased was carried to the grave on a bier 
have been previously found in a male burial in and probably wrapped in a shroud. In grave 17, 
Kekomäki in Kaukola, Karelian Isthmus (nowa- there was a coffin and probably a ring head iron 
days Russia), and in Luistari cemetery in Eura, pin with mineralized tabby weave textile, pos-
southwestern Finland (Kirkinen 2019: 67–8, sibly from the shroud (Riikonen 2011: 212). In 
73). Also, lynx skins were of special value, and Ristimäki, the clear evidence of Christianity was 
they have been found from Finnish medieval in- the church, built in the second half of the 12th 
humation burials only once, from a burial of a century, at a site which was already in use as an 
baby girl in Luistari cemetery (Kirkinen 2015; inhumation cemetery. It was probably used by 
see Zachrisson & Krzewińska 2019). the local residents from the surrounding villages 
The resulting fur, hair, and feather material from the very beginning, without having a his-
indicates a gradual change in burial customs, tory of a single farm cemetery. As Kirkkomäki 
which can – at least partly – be interpreted by the is earlier than Ristimäki, it is possible that its 
Christianization process of the community. On cervid pelts represent an older Late Iron Age tra-
a broad scale, Ristimäki followed a long-lasting dition. In Ristimäki, a quilt stuffed with down 
tradition in furnishing the graves with pelts and replaced cervid pelts and was most probably a 
clothing the dead in furs, documented widely luxury product of its time. The change in fashion 
in the Late Iron Age cemeteries in Finland (see is seen also in grave 41/2016, dated to around the 
Kirkinen 2015; 2019). The differences between year 1200, in which the deceased was clothed in 
Ristimäki and the other Late Iron Age sites can an outer garment and sewn stockings, indicat-
be best formulated by comparing Ristimäki ing the medieval influence on fashion (Riikonen 
with Kirkkomäki (‘Church Hill’) cemetery 2019: 174–5, 180).
in Kaarina (Turku), which is located only two The reasons behind the observed differenc-
kilometres away from Ristimäki on the Aurajoki es can also be a result of local traditions, even 
River, and in which the excavation and docu- though the sites are located close to each oth-
mentation methods are nearly parallel in accu- er. In Ristimäki, the change in burial tradition 
racy. The results of this comparison show that in happened gradually; and the dead being buried 
Kirkkomäki, the species combination is differ- wrapped only in simple shrouds, with or without 
ent, especially in regard to the grave furnishing. coffins, took place only in the final phase (see 
In Kirkkomäki, the coffins were often furnished Hiekkanen 2007: 13–14). This last stage of the 
with Cervidae pelts or moss, and in one case 
56
cemetery is challenging to study, as the preser- Dove, C. J. & Peurach, S. C. 2002. Microscopic 
vation of organic material is poor. analysis of feather and hair fragments asso-
However, the use of furs and pelts in com- ciated with human mummified remains from 
memoration rituals was an element which the Kagamil Island, Alaska. In B. Frohlich, A. B. 
Church aimed to stop. From the 15th century Harper & R. Gilberg (eds.) To the Aleutians 
onwards, the Church tended to Christianize and Beyond. The anthropology of William 
bear skins by collecting them from parishioners S. Laughlin: 51–62. Publications of The 
and using them in front of the altar as carpets National Museum Ethnographical Series, 
(Korhonen 1982a; 1982b; see also Østergård Vol. 20. 
2009: 120–1). Later on, animal skins were re- Dove, C. J. & Agreda, A. 2007. Differences in 
placed by textiles, in burials especially by rugs plumulaceous feather characters of dabbling 
(pall-clothes), the tuft of which probably imi- and fiving ducks. The Condor 109(1): 192–9.
tated fur (Pylkkänen 1974: 27–31). Dove, C. J. & Wickler, S. 2016. Identification 
of bird species used to make a Viking Age 
ACKNOWLEDGEMENTS feather pillow. Arctic 69(1): 29–36.
Goodway, M. 1987. Fiber identification in prac-
We are grateful to Teija Alenius for her kind help tice. Journal of the American Institute for 
with laboratory analysis. The research has been Conservation 26: 27–44.
funded by the Alfred Kordelin Foundation. Greaves, P. H. & Saville, B. P. 1995. Microscopy 
of Textile Fibres. Microscopy Handbooks 32. 
REFERENCES London & New York: Taylor & Francis.
Gräslund, A.-S. 1980. Birka IV. The Burial 
Unpublished sources Customs. A Study of the Graves on Björkö. 
Stockholm: Kungl. Vitterhets Historie och 
Lehto, L. 1993. Ruumishaudasta varastoon. Antikvitets Akademien.
Esimerkki konservoinnista laboratiivisen Gräslund, A.-S. 2010. Religionsskiftet speg-
arkeologian osana: Turun (ent. Kaarina) lat i gravskicket: Ny svensk forskning kring 
Kirkkomäen haudan osan kaivaus ja konser- senvikingatida gravar och gravskick. In 
vointi. Thesis, Conservation dept., Vantaan B. Nilsson (ed.) Från hedniskt till kristet. 
käsi- ja taideteollisuusoppilaitos. Förändringar i begravningsbruk och gravs-
kick i Skandinavien c:a 800–1200: 131–164. 
Literature Sällskapet Runica et Mediaevalia, Opuscula 
14.
Arponen, A. 2008. Arkeologisen materiaalin Hiekkanen, M. 2007. Suomen keskiajan kivikir-
konservointi. In P. Halinen, V. Immonen, kot. Suomalaisen Kirjallisuuden Seuran toi-
M. Lavento, T. Mikkola, A. Siiriäinen & P. mituksia 1117.
Uino (eds.) Johdatus arkeologiaan: 225–35. Hiekkanen, M. 2010. Burial practices in Finland 
Helsinki: Gaudeamus. from Bronze Age to the Early Middle Ages. 
Asplund, H. & Riikonen, J. 2007. Kirkkomäki. In B. Nilsson (ed.) Från hedniskt till kristet. 
In M. Söderström (ed.) Arkeologisia kaivauk- Förändringar i begravningsbruk och gravs-
sia Turussa 1990-luvulla: 9–45. Turun maa- kick i Skandinavien c:a 800–1200: 271–379. 
kuntamuseo, Raportteja 20. Sällskapet Runica et Mediaevalia, Opuscula 
Berglund, B. 2009. Fugela Feđerum in archaeo- 14. 
logical perspective: Eider down as a trade Hirviluoto, A.-L. 1985. Hollolan vaiheet ennen 
commodity in prehistoric northern Europe. kristinuskon tuloa: Kivikaudesta keskiai-
Acta Borealia 26(2): 119–135. kaan. In H. Mantere (ed.) Hollolan kirkko. 
Cleve, N. 1978. Skelettgravfälten på Kjuloholm i Asutuksen, kirkon ja seurakunnan historiaa: 
Kjulo II. Vikingatid och korstågstid. Gravfältet 7–36. Hämeenlinna: Hollolan seurakunta.
C. Suomen Muinaismuistoyhdistyksen Hurcombe, L. 2014. Perishable Material Culture 
Aikakauskirja XLVI:2. in Prehistory. Investigating the Missing 
Majority. London & New York: Routledge.
57
Janaway, R. & Scott, B. (eds.) 1989. Evidence Pylkkänen, R. 1974. The Use and Traditions 
Preserved in Corrosion Products. New Fields of Medieval Rugs and Coverlets in Finland. 
in Artifact Studies. United Kingdom Institute Helsinki: Suomen Muinaismuistoyhdistys.
for Conservation of Historic and Artistic Rast-Eicher, A. 2016. Fibres. Microscopy of 
Works, Occasional Papers 8. London: United Archaeological Textiles and Furs. Budapest: 
Kingdom Institute for Conservation. Archaeolingua.
Kirk, P. L., Magagnose, S. & Salisbury, D. 1949. Riikonen, J. 1990. Naisenhauta Kaarinan 
Casting of hairs: Its technique and Application Kirkkomäessä. Karhunhammas 12.
to Species and Personal Identification. J. Riikonen, J. 2011. White linen: Cloth of luxu-
Crim. L. & Criminology 40 (2): 236–41. ry. In J. Harjula, M. Helamaa & J. Haarala 
Kirkinen, T. 2015. The role of wild animals in (eds.) Times, Things & Places: 36 Essays for 
death rituals: Furs and animal skins in the Late Jussi-Pekka Taavitsainen: 198–221. Raisio: 
Iron Age inhumation burials in Southeastern Newprint Oy.
Fennoscandia. Fennoscandia archaeologica Riikonen, J. 2019. Sukkia ja säärisiteitä: Jalan ja 
XXXII: 101–20. säären verhot myöhäisrautakauden Suomessa. 
Kirkinen, T. 2019. Between skins. Animal skins in In J. Harjula, V. Immonen & J. Ruohonen 
the Iron Age and historical burials in Eastern (eds.) Puukenkien kopinaa. Henrik Asplundin 
Fennoscandia. PhD thesis, Archaeology, juhlakirja: 155–187. Karhunhammas 19.
University of Helsinki. Helsinki: Unigrafia. Ruohonen, J. 2016. Built of wood and turned 
Korhonen, T. 1982a. Saaliseläimen talja kirkko- to soil: Perspectives of research history and 
uhrina. Suomen Museo 89: 45–68. new observations concerning Finland’s old-
Korhonen, T. 1982b. Saaliseläimen taljan ja est churches with reference to Ristimäki in 
kotieläimen vuodan käyttö uhreina. Suomen Ravattula. In J. Harjula, S. Hukantaival, V. 
antro pologi 3: 96–121. Immonen, A. Randla & T. Ratilainen (eds.) 
LacCore SOP Collection (dated 26/5/2016). New Sites, New Methods. Proceedings of the 
Pollen Preparation Procedure. <http://lrc.geo. Finnish-Russian Archaeological Symposium, 
umn.edu/laccore/assets/pdf/sops/pollen.pdf> Helsinki, 19–21 November, 2014: 229–245. 
Lehtosalo-Hilander, P.-L. 1982. Luistari Iskos 21.
III. A Burial Ground Reflecting the Ruohonen, J. 2017. Ristimäki in Ravattula: 
Finnish Viking Age Society. Suomen On the remains of the oldest known church 
Muinaismuistoyhdistyksen Aikakauskirja in Finland. In J. Harjula, S. Hukantaival, 
82(3). V. Immonen, A. Randla & T. Ratilainen 
Lehtosalo-Hilander, P.-L. 1984. Ancient Finnish (eds.) New Visits to Old Churches. Sacred 
Costumes. Helsinki: Suomen Arkeologinen Monuments and Practices in the Baltic Sea 
Seura ry. Region: 46–60. Cambridge: Cambridge 
Lehtosalo-Hilander, P.-L. 2001. Euran puku ja Scholars Publishing.
muut muinaisvaatteet. Vammala: Vammalan Ruohonen, J. 2018. Viimeisistä polttohau-
kirjapaino Oy. tauksista ensimmäisiin kirkkoihin: Aura- ja 
Luoma, H. (ed.) 2003. Sinihameet kultavyöt. Vähäjokilaaksojen kristillistyminen hau-
Suomalaisia muinaispukuja. Tampere: taustapojen ja kalmistojen perusteella. In 
Pirkanmaan käsi- ja taideteollisuus ry. J. Harjula, S. Hukantaival, V. Immonen, T. 
Ogle, R. & Fox, M. 1999. Atlas of Human Hair. Ratialainen & K. Salonen (eds.) Koroinen, 
Microscopic Characteristics. Boca Raton: Suomen ensimmäinen kirkollinen keskus: 
CRC Press LLC. 51–65. Turun Historiallinen Yhdistys, Turun 
Østergård, E. 2009. Woven into the Earth. Historiallinen Arkisto 71.
Textiles from Norse Greenland. 2nd edition. Ruohonen, J. 2019. Identifying a stone struc-
Aarhus: Aarhus University Press. ture in the Medieval church at Ravattula 
Pylkkänen, R. 1955. Säätyläispuku Suomessa in Finland: Reflections on free-standing 
vanhemmalla Vaasa-ajalla 1550–1620. crosses in Northern Europe. Zeitschrift für 
Suomen Muinaismuistoyhdistyksen Archäologie des Mittelalters 46: 113–135.
Aikakauskirja 55.
58
Seppälä, S. 2009. Viljana, nahkoina, kapaka- and the dead: Which ones are taphonomic? 
lana. Talonpoikien maksamat kruununverot Proceedings of the Royal Society B 281: 
Suomessa vuosina 1539–1609. Bibliotheca 20141755.
Historica 125. Suomalaisen Kirjallisuuden Vajanto, K. 2014. Nålbinding in prehistoric buri-
Seura. als: Reinterpreting Finnish 11th–14th-century 
Solazzo, C., Walton Rogers, P., Weber, L., AD Textile Fragments. In J. Ikäheimo, A.-K. 
Beaubien, H., Wilson, J. & Collins, M. 2014. Salmi & T. Äikäs (eds.) Sounds Like Theory. 
Species identification by peptide mass fin- XII Nordic Theoretical Archaeology Group 
gerprinting (PMF) in fibre products pre- Meeting in Oulu 25.–28.4.2012: 21–33. 
served by association with copper-alloy arte- Monographs of the Archaeological Society of 
facts. Journal of Archaeological Science 49: Finland 2. 
524–535. Zachrisson, T. T. & Krzewińska, M. 2019. The 
Teerink, B. J. 2003. Hair of West-European ‘Lynx Ladies’ : Burials furnished with lynx 
Mammals. Atlas and Identification Key. skins from the Migration and Merovingian 
Cambridge: Cambridge University Press. periods found in present-day Sweden. In M. 
Tóth, M. 2017. Hair and Fur Atlas of Central Augstein & M. Hardt (eds.) Sächsische Leute 
European Mammals. Hungary: Pars Ltd. und Länder. Benennung und Lokalisererung 
Tridico, S., Koch, S., Michaud, A., Thomson, von Gruppenidentitäten im ersten 
G., Kirkbride, K. P. & Bunce, M. 2014. Jahrtausend: 103–119. Wendeburg: Verlag 
Interpreting biological degradative process- Uwe Krebs.
es acting on mammalian hair in the living 
59